I'm asking about the etymologies of two Japanese color terms.
First, <ki-iro> 'yellow'. This is transparently "<ki>-color", but
what is that element <ki>? The character used to write it appears to
be unique here: I can't find any other instances of its use.
Second, <midori> 'green'. It seems to be agreed that this term does
not belong to the oldest stratum of color terms in Japanese, but I
haven't been able to find out anything about its origin. Is anything
known about its etymology?
Thanks.
Larry Trask
lar...@sussex.ac.uk
The "ki" in 黄色 is the usual Chinese character for yellow.
Will wait for an expert to explain midori.
[LT]
> > First, <ki-iro> 'yellow'. This is transparently "<ki>-color", but
> > what is that element <ki>? The character used to write it appears to
> > be unique here: I can't find any other instances of its use.
> The "ki" in 黄色 is the usual Chinese character for yellow.
Very interesting. So, the suggestion appears to be that Japanese took
over the Chinese word for 'yellow' and then combined this with its own
word for 'color' to produce the name for this color. It's rather as
though we English-speakers, when we borrowed 'blue' from Norman
French, had converted it to 'blue-color' and used that as our ordinary
name.
Larry Trask
lar...@sussex.ac.uk
I've recently found this wonderful etymological database,
http://starling.rinet.ru/cgi-bin/main.cgi?flags=eygtnnl&root=config
according to which "ki" comes from proto-Japanese "kui",
which in turn comes from proto-Altaic "ki(o\jbu", meaning
'pale'. This protoform would be the ancestor of items
like Middle Mongolian "qubi" meaning 'yellow (horse)'
and Khalkha "xuvxai". The Turkic protoform would have
been "Kuba/Koba" meaning 'pale yellow/pale grey',
giving Kazakh and Uzbek "quw".
Querying for other basic Japanese colour names I've
found this:
*si(a:\yri (white, yellow)
-> *siarIg ---> OT sarIG ----> TR sarI (yellow)
-> *sira -----> WM sira -----> MN s^ar (yellow)
-> *si\rua\ --> OJ si\rwo\ --> JA shiro-i (white)
*ka\ru (black)
-> *Kara -----> OT qara -----> TR kara (black, strong)
-> *kara -----> WM qara -----> KH xar (black)
-> *ku\rua\ --> OJ ku\rwo\ --> JA kuro/-i (black)
*i(a:\ka (light, white)
-> *A:k ------> OT aq --------> TR ak (white, aristocratic)
-> *yagaGan --> WM yaGan -----> KH yagaan (pink)
-> *a/ka/ ----> OJ a/ka/ -----> JP a\ka-i (red)
*a(\jbo (grass)
-> *ebe-su" --> WM ebesu"(n) -> KH o"ws(o"n) (grass)
-> *a\w@\ ----> OJ a\wo\ -----> JP ao/-i (blue, green)
OT Old Turkic
WM Written Mongolian
OJ Old Japanese
TR Turkish
KH Khalkha
JP Tokyo Japanese
> Second, <midori> 'green'. It seems to be agreed that this term does
> not belong to the oldest stratum of color terms in Japanese, but I
> haven't been able to find out anything about its origin. Is anything
> known about its etymology?
I couldn't find anything about the etymology of midori
in the above database, but I know the word originally
(still?) meant "vegetation".
Cheers,
Javier
There is a rather interesting appendix in my Oobunsha kogojiten, listing the
classical terms for
colors, along with a color swatch to show that color was actually referred to.
There is actually no
color directly corresponding to yellow, the closest term is "kibako" using the
same kanji ki as in
kiiro, but it's a rather greenish yellow. Kibako obviously is derived from
"yellow leaves," just as
many other colors are derived from a common standard from nature, i.e. asagi
(sky color/blue).
A little poking around in Kojien reveals an interesting definition of yellow,
"pale color." This is fairly
typical of classical Japanese color names, it refers to a vague quality of the
color rather than an
actual color. One of my painting instructors told me "yellow is just like white
but with a little color"
and I completely agree with him.
>Second, <midori> 'green'. It seems to be agreed that this term does
>not belong to the oldest stratum of color terms in Japanese, but I
>haven't been able to find out anything about its origin. Is anything
>known about its etymology?
There were fewer names for colors in early Japanese, and from what I recall
reading, there were
fewer distinctions between colors. Some color ranges have finer discrimination,
there are quite a
few names for shades of red and peach, but hardly any for cooler colors. In
later times, colors like
midori were used to refer to a wide range of both blue and green. According to
Oobunsha again,
the classical color closest to green was "moegi" (bud color).
I know I've read more about the etymology of midori somewhere, but I haven't
found it, I'll keep
looking.
The Normans were the ones that stuck a French ending on the latin
'color' to give 'colour' too.
-------------------------------------------------------------
Dale Walker London Techno Events Saiko!
da...@sorted.org lon...@sorted.org sa...@sorted.org
London, UK london.sorted.org saiko.sorted.org
> There were fewer names for colors in early Japanese, and from what I recall
> reading, there were
> fewer distinctions between colors. Some color ranges have finer discrimination,
> there are quite a
> few names for shades of red and peach, but hardly any for cooler colors. In
> later times, colors like
> midori were used to refer to a wide range of both blue and green. According to
> Oobunsha again,
> the classical color closest to green was "moegi" (bud color).
Kind of like the way "aoi" refers to something blue or green.
> I know I've read more about the etymology of midori somewhere, but I haven't
> found it, I'll keep looking.
Now you've got me interested in how different languages developed color
definitions, and what their needs were. When I think of the colors in
English, I get very specific colors in mind. Is that only because in
the last 200 years or so, the English speaking world had mass-produced
pigments to refer to?
In other words, did I learn the shade for "green" from the Crayola
corporation, or was green always associated with that shade?
--
---------------------------------------------------------------------------
Travers Naran | Visit the SFTV Science Blunders
F/T Programmer,P/T Meddler In Time&Space | Hall of Infamy!
New Westminster, British Columbia, |
Canada, Earth, Milky Way, etc. | <www.geocities.com/naran500/>
"Stand Back! I'm a programmer!" |
---------------------------------------------------------------------------
> On 30 Dec 2003 09:52:42 -0800, lar...@sussex.ac.uk (Larry Trask)
> wrote:
>
>
>>Geoff <grw...@hotmail.com> wrote in message news:<Q2hIb.11910$lo3....@newsread2.news.pas.earthlink.net>...
>>
>>[LT]
>>
>>
>>>>First, <ki-iro> 'yellow'. This is transparently "<ki>-color", but
>>>>what is that element <ki>? The character used to write it appears to
>>>>be unique here: I can't find any other instances of its use.
>>
>>>The "ki" in 黄色 is the usual Chinese character for yellow.
>>
>>Very interesting. So, the suggestion appears to be that Japanese took
>>over the Chinese word for 'yellow' and then combined this with its own
>>word for 'color' to produce the name for this color. It's rather as
>>though we English-speakers, when we borrowed 'blue' from Norman
>>French, had converted it to 'blue-color' and used that as our ordinary
>>name.
>
> The Normans were the ones that stuck a French ending on the latin
> 'color' to give 'colour' too.
My Grade 12 English teacher, who was an Englishman, absolutely hated
spelling color with a 'u'.
Obunsha's Dictionary of Classical Japanese (Kogo Jiten) 8th ed. has the
following for ki:
Ki: (noun; nari-type adjectival verb) The color yellow ["ki-iro"].
"Kawakami no hou yori ~~ naru mono nagare kite" ["... a yellow object
flowed down the river ..."] (from the Sarashina Diary).
Yellow source ["ki-naru izumi"] - (Japanese reading of Chinese "kwausen"
["yellow+source" in classical Japanese pronunciation]) The world one
goes after dying. The land of the dead
["meido"]). "Omohoede ~~ ni kie kaheri" ["... (he) unexpectedly
disappeared, returning to the land of the dead ..."] (from the "Atemiya"
section of the Tale of Utsuho).
Yellow tears ["ki-naru namida"] - Tears shed in profound despair or
sadness. Often applied to animals. "Tears of blood" [Japanese expression
meaning "bitter tears"]. "(Shishi ha) ~~ wo nagashitsutsu" ["...
shedding bitter tears ("yellow tears"), the lion ..."] (from Taiheiki).
[ ] = my comments, translations of quotes, or readings of characters.
>>The "ki" in 黄色 is the usual Chinese character for yellow.
>
>
> Very interesting. So, the suggestion appears to be that Japanese took
> over the Chinese word for 'yellow' and then combined this with its own
> word for 'color' to produce the name for this color.
Not exactly. Japanese "ki" is Japanese. Geoff is saying that the
Japanese use the Chinese character for "yellow" (or, since I
don't speak Chinese, what I assume is the Chinese character for
"yellow," since it could have a different meaning and/or other meanings
as well) when writing "ki."
"Iro" is the native Japanese word for color.
> Second, <midori> 'green'. It seems to be agreed that this term does
> not belong to the oldest stratum of color terms in Japanese, but I
> haven't been able to find out anything about its origin. Is anything
> known about its etymology?
The Kojien dictionary, which is considered the standard reference for
modern Japanese, lists "mido" as the root for the word "midori," with a
possible link to "mizu" in old "mizumizushi" (modern "mizumizushii")
("fresh," "young and beautiful"). Note that "mizu" also means water,
although in this case the "mizu"s are written with different Chinese
characters (which could be irrelevant.)
I assume that you are calling "the oldest stratum of color terms" the
group of colors which can end in "-i," of which "midori" and "ki" do not
form a part.(1) Most colors, come to think of it, do not end in "-i,"
however, so maybe you're applying a different definition.
I won't translate the lengthy entry for "midori" from the classical
dictionary, but it looks like classical usage corresponds fairly closely
with modern usage, in which "midori" either means the color green or
"greenery" (vegetation, etc.) although this last meaning was originally
"new leaves, buds, etc." Interestingly, the earliest example quoted in
the entry is from the Manyoshu (compiled between 630 AD and 760 AD) and
is used in the form "midoriko/midorigo" (green + child), meaning "baby"
or "a child until about the age of three." "Green hair" was apparently
an expression which meant "shiny black hair," which is interesting
because until recently (not sure how recently), the word for blue, "ao,"
also did double duty for green. In fact, a green light (on a traffic
light) is still referred to as "ao-shingo" ("blue signal"), so "midori"
as a color might have referred to dark green, while lighter hues of
green might have fallen under "ao." Grass, in fact, can also be called
"ao" (slightly archaicly), as in English "bluegrass."
My non-specialist's guess would be that "ao" originally covered most
shades of green and blue, but later "midori" ("greenery, vegetation")
was borrowed to distinguish blue(r) from green(er).
(1) - I've heard children say "midorii" (never "kii" though), but I
think this is just a mistake children make when they over-apply a rule
they've learned, as in "I maked a snowman with daddy yesterday."
..
> I'm asking about the etymologies of two Japanese color terms.
> First, <ki-iro> 'yellow'. This is transparently "<ki>-color", but
> what is that element <ki>? The character used to write it appears
> to be unique here: I can't find any other instances of its use.
> Second, <midori> 'green'. It seems to be agreed that this term
> does not belong to the oldest stratum of color terms in Japanese,
> but I haven't been able to find out anything about its origin. Is
> anything known about its etymology?
I don't think anyone knows the etymology of either of these words.
(The many-volume Kokugo Daijiten_ undoubtedly has some guesses, but
if I ever checked in that dictionary, I don't remember.)
"Ki" is not from the Chinese, in which that character is "huang"
([o:] < [wau] in Japanese).
Best guess I've ever come up with for "yellow," and though one
usually likes one's own etymologies, this one isn't very satisfying,
is "wood (color)." Two other "ki" that are probably less likely yet
are the "ki" of "kiba" = "fang," where the "ba" part is presumably
"tooth" (are fangs yellow?), and an old word for sake (survives in
"miki" = "taboo/holy sake"). They spoke (or rather wrote) of
"kuroki" and "shiroki" = "black sake" and "white sake" respectively,
though.
As for "midori," the late Peter Boodberg--much better at Chinese
etymology than Japanese, in my opinion--wanted to make it "water
bird" --> "green" in parallel with the Chinese use of its word for
kingfisher as a color term. But there isn't the slightest hint that
any birds were ever called "midori," to the best of my knowledge.
My own guess--I like this better than my "wood color" for yellow--is
that it might be related to "(ao)midoro" = "green pond scum." At
least there is a hint that "midori" is a wet color in the expression
"midori shitataru" = "midori drips" = "is verdant green." But I have
no explanation as to why the final "o" of "midoro" should have
survived in that word. (Most cases where "o" > "i" have "o"
surviving only in compounds. E.g. "ho" = "fire" shows up that way in
"hokage" = "firelight," etc., but is now only "hi" as a single word.
Bart
> "Larry Trask" writes:
>
> ...
>
> Best guess I've ever come up with for "yellow," and though one
> usually likes one's own etymologies, this one isn't very satisfying,
> is "wood (color)."
> ...
> (Most cases where "o" > "i" have "o"
> surviving only in compounds. E.g. "ho" = "fire" shows up that way in
> "hokage" = "firelight," etc., but is now only "hi" as a single word.
Don't forget "ki" -> "kokage," "konoha." Is "ko-" for "yellow" attested?
> Bart Mathias wrote:
> > "Larry Trask" writes:
> > ...
Well, there's "kogane" = "gold."
Iwanami kogo says that was "kugane" in the Nara period, though.
"Ku" to "ki" is no problem, but I'm unaware of any "tree/wood" "ku-"
at the moment.
I like the idea of a "ko/ku" doublet. I've always wanted to see "mi"
in the "mo" of the word "mogak-" = "squirm" (i.e., "mo-" > "mi"; cf.
"agak-" = "paw the ground," from "a-" as "foot"), but the consensus
ties it in with "mu" as in "mukuro" = "cadaver."
Bart
According to Samuel Martin, _The Japanese Language Through Time_, Yale
1987 (o_ indicates an underlined o in the text):
"midori < myido_ri < *minto_ri. ... An etymology midu iro 'water
color' has been suggested, presumably by metathesis --> *mid[u]-ori
but that would lead to /dwo/ instead of the attested /do_/, unless
*miduiro > *midi[y]ro_ -> mido_ri is assumed. Also possible: *min[a]
to_ri 'water bird', with an extended meaning or perhaps as a
truncation 'water-bird [color]'." (477).
Mikael Thompson
[...]
> The Normans were the ones that stuck a French ending on the latin
> 'color' to give 'colour' too.
Not precisely, no. The Latin source is <colo:rem>, and Latin
long /o:/ gave in early Old French a very close vowel in the [o:]
~ [u:] range that was represented by both <o> and <u>. Thus, you
find in OFr both <colur> and <color>, among others. The spelling
<ou> appears in OFr in the 13th century, I believe, and was taken
from late Anglo-French into Middle English. According to the
OED, <colour> has been the normal English spelling since the 14th
century.
Brian
Bart Mathias wrote:
>
> Well, there's "kogane" = "gold."
>
> Iwanami kogo says that was "kugane" in the Nara period, though.
My 漢和辞典 under 黄 says :
*黄は、田(土地の意)が意符、コウ(光の古字)が音符で、
また、日光の色の意を表す。土地の色、黄色のこと。
*黄←これは俗字らしいです。
緑もなんか書いてあった。偏と作りにわけて、糸偏は意味、
作りは音を表すそうです。作りの方の漢字(りょく)は、
萌黄色(もえぎいろ)に染める草の名前だそうです。
□■ <:3 )~
■楽猫 <:3 )~
You s.l.j. folk might want to have a look at Berlin & Kay, *Basic Color
Terms* (MIT Press, 2nd ed. ca. 2000), which is about the (non-)diversity
of words for color in the world's languages; they distinguish between
words that don't denote anything but a color ("red," "green" in English)
and words for colors that are taken from names of things that are
typically that color (midori, moegi).
--
Peter T. Daniels gram...@att.net
And what about the possibility of an Altaic protoform
"ki(o/jbu" meaning 'pale', which would be the ancestor
of proto-Japanese "kui" (whence "ku" and "ki"), Written
Mongolian "qubi" meaning 'yellow (horse)' and proto-Turkic
"Kuba/Koba" meaning 'pale yellow/pale grey', a possibility
that looks plausible (as the mentioned database shows,
there seems to be correspondences between many other
roots, like those of other colour terms: ak/yagaan/aka,
kara/xar/kuro, sarI/s^ar/shiro, also "mura" in "murasaki"
could be related to Turkish "al" and Khalkha "o"l") but
that no-one so far has cared to comment on?
Someone has pointed that Kojien offers a definition of
"ki" as 'pale yellow' and remarked that ancient Japanese
colour terms didn't refer to specific hues but to other
more general qualities (in this case 'paleness'), which
I think is very plausible because this is also the case
with many other languages, like all those Indoeuropean
colour terms coming from words referring to luminosity,
brightness, flashing or burning, like English "blue",
"black" (from 'burned'), Russian "belyi" (white) and
Latin "flavus" (orange yellow), all from PIE *bhel and
only later specialized in a certain (different) hue.
This meaning 'pale' would only reinforce the connection
with Mongolian and Turkic words like "qubi" and "quw",
which in their overt form do already look suspiciously
similar to the Japanese form "ku/ki".
Cheers,
Javier
Has anything new been said about this topic in the last few years? I
have seen discussions of this topic in the old IE list where linguists
dismissed Berlin and Kay.
The problem seems to be that Berlin and Kay's methodology [showing the
usual color chips and asking for names] assumes that
the perceptual space is necessarily decomposed into a product of a
brightness axis and a 2-dim color region and in the same way in all
cultures. They seem to have made no attempt to verify this while
there were serious objections to making this assumption.
Nath Rao
I don't see how it assumes this. They needed some way to identify
a variety of color samples for their experiment in such a fashion
as to make their results reproducible. Why not do it the way
they chose to, even if it turns out human perceptual space is
organized in some different way? How could it make any difference
to their result?
--
Greg Lee <gr...@ling.lll.hawaii.edu>
This is precisely the thrust of Robert MacLaury's work. MacLaury has
identified systems of color terms which are not based upon the B & K
schema, but on other factors. Sorry I can't provide references while
I'm sitting at home, but he had a big article in Current Anthropology
a few years ago that lays out much of his thinking.
Note, however, that MacLaury is very much a universalist and a
proponent of the essential correctness of B & K's approach. He sees
himself as making necessary modifications to that approach, and not as
rejecting it.
Larry Trask
lar...@sussex.ac.uk
I have in mind the following four terms:
<kuro> 'black'
<shiro> 'white'
<aka> 'red'
<ao>, today 'blue' but earlier 'grue'
According to the information I've assembled from various sources,
these four terms are distinguished by five criteria:
1. They alone take the ending <-i>: <kuroi>, etc.
2. They alone take the intensive prefix <maC->: <makkuro> 'pitch
black', <masshiro> 'pure white', <makka> 'bright red' and <massao>
'sky blue'.
3. They alone are frequent in compounds.
4. They alone are frequent in surnames.
5. They alone are frequent in proverbs.
Such evidence suggests that these four terms belong to a more ancient
stratum than do the other color terms.
Larry Trask
lar...@sussex.ac.uk
> > "Ku" to "ki" is no problem, but I'm unaware of any "tree/wood"
> > "ku-" at the moment.
> And what about the possibility of an Altaic protoform "ki(o/jbu"
> meaning 'pale', which would be the ancestor of proto-Japanese "kui"
> (whence "ku" and "ki"), Written Mongolian "qubi" meaning 'yellow
> (horse)' and proto-Turkic "Kuba/Koba" meaning 'pale yellow/pale
> grey', a possibility that looks plausible (as the mentioned
> database shows, there seems to be correspondences between many
> other roots, like those of other colour terms: ak/yagaan/aka,
> kara/xar/kuro, sarI/s^ar/shiro, also "mura" in "murasaki" could be
> related to Turkish "al" and Khalkha "o"l") but that no-one so far
> has cared to comment on?
..
I'm hardly more prepared to speak of Japanese-protoAltaic than of
Japanese-Nostratic, but there are some problems with those.
"kui" > "ki" works, but "kui" > "ku" seems implausible. The "ku" has
"kui" > to come first, then "kui" by the addition of some suffix,
"kui" > perhaps. But how do we get any of that from "ki(o/jbu"? I'd
"kui" > want to see the sound-correspondence rules and the evidence
"kui" > for them (a la Martin 1966).
The protoform is glossed here as an adjective ('pale'). The Japanese
requires a noun.
What do "ak" and "yagaan" mean? Japanese "ak(a)" is basically
"bright; dawn" then "red." (A curious parallel is the Korean set of
"palk-" words.)
The connection between the "mura" (= "cluster") of "murasaki" and
colour terms escapes me.
Bart Mathias
And that's why they ruled out "orange" as BCT... uh... oh...
no, wait, that's why they kept "orange" as a basic English
colour term and loudly proclaimed it a UNIVERSAL, INNATE
human colour category. You know, being "orange" a universal,
innate human colour category, that clearly explains why
such an overwhelming majority of languages around the world
feature a specific, basic, non-derived name for it denoting
nothing but that colour...
You slightly misquoted me, Kojien says "pale color" not "pale yellow." Oobunsha
cites "pale leaves"
as the name for greenish yellow.
Think about what it must have been like before scientific knowledge of color and
light. There were
two systems of color, observed, and created. You could observe colors (those
autumn leaves are
pale green) or you could try to produce dyes or pigments with that color.
Painters often speak of
color variations in terms that seem vague, but are quite precise if you think
about it from mixing
pigments. We might refer to two distinct shades, one is "greenish yellow" and
one is "yellowish
green," the former would be made by adding a little more green tint to yellow,
the latter would be
by adding a bit of yellow tint to green. The "-ish" color is the minor additive
to the primary. This
sort of ad hoc color system would be familiar to anyone who used colored
pigments. Also consider
that many color names were taken directly from the natural material that
produced them, i.e.
indigo.
I am reminded of the tale of a modern "Wild Child" that grew to adolescence
abused, chained in a
closet, without ever learning to speak. One of the most notable experiences of
the researchers who
tried to teach her to speak was when they went to a dimestore and the girl saw a
color chart above
a row of spools of colored thread. The girl asked the name of every color, and
the researcher could
not explain them all. The researcher described them qualitatively, i.e. "this
one is a bit more red
than the next one," etc. It was reported that this horribly frustrated the
child, who could not grasp
that there were not distinct nouns for every quality of color.
"Murasaki" can be broken down into "mura" (cluster, group) and "saki"
(bloom). It's the name of a flower, and its use as the name of a color
originates in the practice of dying things with the root of the flower
(according to Koujien: "ムラサキの根で染めた色。" [murakaki no ne de
someta iro]).
> I've recently found this wonderful etymological database,
>
> http://starling.rinet.ru/cgi-bin/main.cgi?flags=eygtnnl&root=config
It's somewhat innacurate if the words you've given us are anything to
go by.
> according to which "ki" comes from proto-Japanese "kui",
> which in turn comes from proto-Altaic "ki(o\jbu", meaning
> 'pale'. This protoform would be the ancestor of items
> like Middle Mongolian "qubi" meaning 'yellow (horse)'
> and Khalkha "xuvxai".
No. For one thing, that's a confusion of two different words. First
there's kubi (often transcribed qubi) meaning "portion, share,
allotment," giving rise to a large number of derivatives, including
kubil- "to apportion; to be transformed," the root for the emperor
Qubilai / Kublai Khan's name (meaning "(miraculous) transformation").
Then there's kuba(n), meaning "amber" and by extension "pale yellow."
(Khalkha reflexes are xub' and xub, respectively.) The other word,
xubxai, means "dried, withered, wilted, bare, denuded," and comes from
MM kubakai (qubaqai), same meaning. In addition, there's kua (Kh xua)
"light yellow, chestnut, bay."
> The Turkic protoform would have
> been "Kuba/Koba" meaning 'pale yellow/pale grey',
> giving Kazakh and Uzbek "quw".
Dunno Uzbek, but Kazakh has much the same bundle of words as
Mongolian. First there's quw "dried up, withered, pale of
complexion," quba "pale yellow," and qubaqan "light-bay (of horse)."
(The corresponding words for changing or transformation only show up
for the root qubIl-.)
These roots might well be related, but if so, the most likely relation
is Tkc quba being borrowed into Mongolian (quite common with horse
terminology, for which there are more prestigious Turkic terms going
with less prestigious native Mongolic terms). There's also the fact
that Mongolian kuba is related to Mandarin hu3po4 "amber." Which way
the borrowing went is something I'll have to look into, but from what
I have on hand it could have been either way. (The fact that the
Mongolian has a hidden n doesn't determine it, since that was often
added to loanwords, as in MM conx-n "window," borrowedfrom (an earlier
form of) Mandarin chuanghu id.) In any case, there are a number of
borrowings between Kazakh and other Qipchaq Turkic languages and
Mongolian over many centuries in both directions, and it's quite
possible that what is involved is a convergence of two distinct words
with two different but easily relatable meanings, of yellowness and of
dryness.
Mikael Thompson
rose wrote:
>
> 緑もなんか書いてあった。偏と作りにわけて、糸偏は意味、
> 作りは音を表すそうです。作りの方の漢字(りょく)は、
> 萌黄色(もえぎいろ)に染める草の名前だそうです。
なので、糸をりょくで染めたら緑色になる。
。。と書き忘れました。(^^;ゞ
□■ <:3 )~
■楽猫 <:3 )~
>Phil Healey <com.hotmail@psa_healey> wrote in message news:<vgoIb.33478$Vs3....@twister.socal.rr.com>...
>> I assume that you are calling "the oldest stratum of color terms" the
>> group of colors which can end in "-i," of which "midori" and "ki" do not
>> form a part.(1) Most colors, come to think of it, do not end in "-i,"
>> however, so maybe you're applying a different definition.
>I have in mind the following four terms:
><kuro> 'black'
><shiro> 'white'
><aka> 'red'
><ao>, today 'blue' but earlier 'grue'
I wonder if underlying the color names isn't something more generalized,
like <kuro/kura> 'dark', <shiro/a> 'pale', <aka> 'bright, and <ao>
'dim/subdued'? It's New Year's Eve here still, so I just indulging in
idle thoughts.
[Snip evidence]
>Such evidence suggests that these four terms belong to a more ancient
>stratum than do the other color terms.
Which might be consistent with my musings.
--
Don
Old age is when you start saying "I wish I knew now what I knew then."
> > they distinguish between words that don't denote anything but a color
I'm not defending B & K's conclusions here, but I think this comment
is a little unfair. B & K in no way predict that a basic color term
'orange' should occur in most languages -- in fact, if anything, they
predict that such a term should be rare.
The interesting question, I think, is this: how often do we find a
term for 'orange' in languages whose speakers are unacquainted with
oranges?
Larry Trask
lar...@sussex.ac.uk
Gold was considered yellow (e.g. rendered with yellow pigment in
depictions) until oranges were introduced. But no kindergartener now
would use the same crayon to color the sun and jewelry.
Strangely back on-topic, I was struck by the fact that Japanese
schoolchildren draw the moon yellow and the sun orange/red.
Well, on their flag -- the Rising Sun-- it's that way too, isn't it?
Maybe it isn't just the Japanese kids who do the sun orange/red.
Joe Murphy
Boy Linguist
Gold was commonly rendered with powdered gold pigment.
> Phil Healey <com.hotmail@psa_healey> wrote:
>>Peter T. Daniels wrote:
>>>Larry Trask wrote:
>>>>uaxu...@hotmail.com (Javier BF) wrote in message news:<b0461a9.03123...@posting.google.com>...
>>>[...]
>>Strangely back on-topic, I was struck by the fact that Japanese
>>schoolchildren draw the moon yellow and the sun orange/red.
>
> Well, on their flag -- the Rising Sun-- it's that way too, isn't it?
> Maybe it isn't just the Japanese kids who do the sun orange/red.
Well, there ya go.
But the Rising (and Setting) Sun _is_ red. (You landlocked Hoosiers --
you've never seen a proper sunrise or -set!)
Define "commonly."
Explain the technique of "powdered gold" pigment.
I could tell you how gilding was and is done, but you can look it up.
No powder.
And only in very sumptuous mss., not in more ordinary ones, or e.g.,
stained glass windows or tapestries as at Bayeux.
As in, a technique known to artists worldwide, in widespread use. For example,
http://optometry.berkeley.edu/~fiorillo/texts/topictexts/faq/faq_metallics.html
>Explain the technique of "powdered gold" pigment.
You grind up gold into tiny flakes, sometimes add a little brass or mica powder
for a nice sparkly
effect, add a binder like gum arabic for watercolors or linseed oil for oil
painting. Apply with a
brush or via transfer on a printing plate.
>I could tell you how gilding was and is done, but you can look it up.
Unlike you who merely read about it, I've done it. Gold leaf is an incredible
pain to apply. There are
several ways to apply gold, gilding is just one of them. It's far easier to use
gold powdered
pigments. You DO realize, you're disputing someone who has been using gold
pigments for about
30 years, and has researched their historical use while working on his BFA in
Painting?
>No powder.
I have a very old bottle of powdered gold pigment in suspension in gum arabic
sitting on my shelf,
for use on special paintings. But if you say there's no such thing, I guess I
better inform the
manufacturer that they're selling an imaginary product. They will be quite
disappointed since they
take great pride in describing how this product is manufactured according to an
age-old traditional
formula.
>And only in very sumptuous mss., not in more ordinary ones, or e.g.,
>stained glass windows or tapestries as at Bayeux.
Apparently you are a complete idiot and know nothing about art. Until the
invention of aniline
pigments in the 1850s, nearly all bright pigments were expensive and precious,
used only
sparingly, and even then, only when a rich patron commissioned an expensive
work. Look at Old
Master paintings like Rembrandt, you'll notice they are almost entirely executed
in tones of brown
and yellow, but masterfully use simultaneous contrast to make the color spectrum
of the paints
look wider than it actually is. Even pure white was expensive, and typically
only used for small
highlights. Rembrandt was a cheapass S.O.B. Michelangelo was even cheaper, you
ought to read his
notebooks, he constantly complains about the expense of pigments.
> Define "commonly."
>
> Explain the technique of "powdered gold" pigment.
>
> I could tell you how gilding was and is done, but you can look it up.
>
> No powder.
I heard that Henry Bessemer made his (first) fortune from devising a
method to make gold (coloured) powder for pigments from brass. It was
obviously in demand somewhere. This income then subsidised the rest of
his inventing career.
James
My point was that they claimed the 11 basic colours of
English - those 11 and no other than those 11 - to be all
and the only basic ones. That is, that even given that
orange is a fairly rare colour among languages worldwide
and that not even English has a term that exclusively
refers to it, they still had no inconvenience to proclaim
it one of the universal, basic, innate colour categories
that supposedly are hardwired into every human being.
While colours such as azure (basic at least in Russian),
crimson (basic at least in Hungarian) or glaucous (basic
at least in Cantonese, Latin and Greek) were completely
left out, they didn't predict anything about their 'order
of appearance by evolution' - not even that they would
be rare like orange - but by excluding them of their
privileged set, they claimed them not to be universal,
basic, innate colour categories of all human beings
like supposedly orange, but seemingly kind of weirdoes
invented by some exotic peoples who seemingly liked to
mess with the hardwiring of their brains.
Cheers,
Javier
>Joseph W. Murphy wrote:
>>
>> Phil Healey <com.hotmail@psa_healey> wrote:
>>
>> >Peter T. Daniels wrote:
>> >> Larry Trask wrote:
>> >>
>> >>>uaxu...@hotmail.com (Javier BF) wrote in message news:<b0461a9.03123...@posting.google.com>...
>> >>>
>> >>>
>> >> [..]
>> >>
>> >> Gold was considered yellow (e.g. rendered with yellow pigment in
>> >> depictions) until oranges were introduced. But no kindergartener now
>> >> would use the same crayon to color the sun and jewelry.
>> >
>> >Strangely back on-topic, I was struck by the fact that Japanese
>> >schoolchildren draw the moon yellow and the sun orange/red.
>> >
>>
>> Well, on their flag -- the Rising Sun-- it's that way too, isn't it?
>> Maybe it isn't just the Japanese kids who do the sun orange/red.
>
>Peter T. Daniels wrote:
>But the Rising (and Setting) Sun _is_ red. (You landlocked Hoosiers --
>you've never seen a proper sunrise or -set!)
>--
:-)))
Out here in the endless prairie, we Hoosiers like to think it's sorta
"yaller".
Joe Murphy
Boy Hoosier
P.S. "Yaller" is a basic color term in Indiana.
Because what we choose to measure often affects what we find. If the
color terminology is a result of how pigments are mixed, or how color is
produced on a monitor, it has nothing to do with linguistics, but
only technology.
Nath Rao
> Joseph W. Murphy wrote:
>>
>> Phil Healey <com.hotmail@psa_healey> wrote:
>>
>>>Peter T. Daniels wrote:
>>>> Gold was considered yellow (e.g. rendered with yellow pigment in
>>>> depictions) until oranges were introduced. But no kindergartener now
>>>> would use the same crayon to color the sun and jewelry.
What exactly are you trying to say here? Nowadays there are "gold"
crayons (in my time, we had gold watercolor), but given a choice of
yellow and orange, I would always choose yellow for gold. If available,
a darker yellow than for the sun.
>>>Strangely back on-topic, I was struck by the fact that Japanese
>>>schoolchildren draw the moon yellow and the sun orange/red.
>>
>> Well, on their flag -- the Rising Sun-- it's that way too, isn't it?
>> Maybe it isn't just the Japanese kids who do the sun orange/red.
>
> But the Rising (and Setting) Sun _is_ red.
But the Japanese children paint the sun red always, high in the sky,
too.
Oliver C.
> The interesting question, I think, is this: how often do we find a
> term for 'orange' in languages whose speakers are unacquainted with
> oranges?
I (German) would call the color of carrots "orange", but one traditional
German term is /Gelbe/ /Rüben/ (/yellow/ /roots/). Thus, it should be
possible to call oranges yellow, or red, too. On the other hand, my son
at 23 months refused to call other things than oranges "orange".
Oliver C.
That's a good one. If, as you claim, you've done gilding, you have to be
aware that gold is the most malleable of metals. "Powdering" gold is
physically impossible.
> sometimes add a little brass or mica powder
"Brass powder" was invented in the 19th century (by one of the metals
engineers -- Bessemer, who figured out how to make steel, or the
aluminum guy whose name I don't recall -- and the family fortune was
maintained not from steel/aluminum, but from brass powder, a process
that wasn't patented because it was a luxury and in high demand).
> for a nice sparkly
> effect, add a binder like gum arabic for watercolors or linseed oil for oil
> painting. Apply with a
> brush or via transfer on a printing plate.
Gold. Au. Yeah, right.
> >I could tell you how gilding was and is done, but you can look it up.
>
> Unlike you who merely read about it,
Sorry, I learned to do it in calligraphy class almost 30 years ago.
> I've done it. Gold leaf is an incredible
> pain to apply.
Applying gold leaf isn't particularly painful. It's everything that goes
before that takes forever and has to be very precise.
> There are
> several ways to apply gold, gilding is just one of them. It's far easier to use
> gold powdered
> pigments. You DO realize, you're disputing someone who has been using gold
> pigments for about
> 30 years, and has researched their historical use while working on his BFA in
> Painting?
And you think they contain the element gold? What do they sell for?
> >No powder.
>
> I have a very old bottle of powdered gold pigment in suspension in gum arabic
> sitting on my shelf,
> for use on special paintings. But if you say there's no such thing, I guess I
> better inform the
> manufacturer that they're selling an imaginary product. They will be quite
> disappointed since they
> take great pride in describing how this product is manufactured according to an
> age-old traditional
> formula.
From what?
> >And only in very sumptuous mss., not in more ordinary ones, or e.g.,
> >stained glass windows or tapestries as at Bayeux.
>
> Apparently you are a complete idiot and know nothing about art. Until the
> invention of aniline
> pigments in the 1850s, nearly all bright pigments were expensive and precious,
> used only
> sparingly, and even then, only when a rich patron commissioned an expensive
> work. Look at Old
> Master paintings like Rembrandt,
Do you see anything about "paintings" in what I wrote? What Rembrandt
did wasn't even invented until centuries after the period we're talking
about, viz., when "orange" entered the English language.
> you'll notice they are almost entirely executed
> in tones of brown
> and yellow, but masterfully use simultaneous contrast to make the color spectrum
> of the paints
> look wider than it actually is. Even pure white was expensive, and typically
> only used for small
> highlights. Rembrandt was a cheapass S.O.B. Michelangelo was even cheaper, you
> ought to read his
> notebooks, he constantly complains about the expense of pigments.
Look a little lower and smaller. Look at the Book of Kells, look at the
various (Très) Riches Heures available for admiration.
Why, after all these months or years, do you continue to lie about what
they claimed?
> While colours such as azure (basic at least in Russian),
> crimson (basic at least in Hungarian) or glaucous (basic
> at least in Cantonese, Latin and Greek) were completely
> left out, they didn't predict anything about their 'order
> of appearance by evolution' - not even that they would
> be rare like orange - but by excluding them of their
> privileged set, they claimed them not to be universal,
> basic, innate colour categories of all human beings
> like supposedly orange, but seemingly kind of weirdoes
> invented by some exotic peoples who seemingly liked to
> mess with the hardwiring of their brains.
Color terminology is a little bit older than pigment-mixing or TV sets.
That's the case with Hungarian: carrots are "sárgarépa"
(yellowbeet) and orange is "narancssárga" (orangeyellow).
In Chinese orange is a kind of yellow too: "chénghuáng".
Modern Basque has mimicked the colour semantics of the
surrounding prestige languages Spanish and French, but
anciently "gorri" (red) had a wider meaning that included
brown and orange. This can be seen e.g. in the traditional
practice of nicknaming a brown cow "Gorri" or in the name
("esnegorri", "onddo gorri", "ziza gorri") of an edible
mushroom quite well-known in Spain (its scientific name
is 'Lactarius deliciosus', its most standard Spanish
name is 'níscalo') that features a characteristic orange
- definitely, unmistakably orange - colour, as you can
see here:
http://www.valdorba.org/micovaldorba2/setas/lactdeli.html
http://www.mykoweb.com/CAF/species/Lactarius_deliciosus.html
Cheers,
Javier
Javier BF wrote:
>> I (German) would call the color of carrots "orange", but one
>> traditional German term is /Gelbe/ /Rüben/ (/yellow/ /roots/).
>> Thus, it should be possible to call oranges yellow, or red,
>> too.
> That's the case with Hungarian: carrots are "sárgarépa"
> (yellowbeet)
Ditto Danish: <gulerod> 'yellow root'. Then there is the striking
resemblance between Danish <gul> [gu:?l] 'yellow' and <guld>
[gul] 'gold' (ON <gulr> and <gull> respectively). Both akin to
<galde> ["gal@] 'bile', 'gall', a fluid I wouldn't call yellow in
Danish.
--
Torsten
First you make gold leaf, then you break it up into flakes. That's one way to do
it. Ever panned for
gold? I have, you collect mostly tiny flakes of gold powder. Perhaps you ought
to investigate the
history of lacquerware, there are many techniques involving powdered gold
applied in the lacquer.
>> sometimes add a little brass or mica powder
>
>"Brass powder" was invented in the 19th century (by one of the metals
>engineers -- Bessemer, who figured out how to make steel, or the
>aluminum guy whose name I don't recall -- and the family fortune was
>maintained not from steel/aluminum, but from brass powder, a process
>that wasn't patented because it was a luxury and in high demand).
You didn't read the article. The author of that page is also the co-author with
David Hockney of a
recent book about ancient art techniques. He subjected a print to XRay
spectroscopy and found
brass in the pigments. The article also obliquely refers to obscure aspects of
Sumptuary Laws in
Japan, which prohibited the use of Gold in artworks, which is why they didn't
find any gold in that
particular object. I cited the article as an example of the techniques of
applying metallic pigments.
>> for a nice sparkly
>> effect, add a binder like gum arabic for watercolors or linseed oil for oil
>> painting. Apply with a
>> brush or via transfer on a printing plate.
>
>Gold. Au. Yeah, right.
>
>> >I could tell you how gilding was and is done, but you can look it up.
>>
>> Unlike you who merely read about it,
>
>Sorry, I learned to do it in calligraphy class almost 30 years ago.
>
>> I've done it. Gold leaf is an incredible
>> pain to apply.
>
>Applying gold leaf isn't particularly painful. It's everything that goes
>before that takes forever and has to be very precise.
>
>> There are
>>several ways to apply gold, gilding is just one of them. It's far easier to use
>> gold powdered
>> pigments. You DO realize, you're disputing someone who has been using gold
>> pigments for about
>> 30 years, and has researched their historical use while working on his BFA in
>> Painting?
>
>And you think they contain the element gold? What do they sell for?
Yep, flakes of gold. I don't know what it cost, but I remember it being very
very expensive, even
back when I bought it around 1970, when Gold was still under $50/oz. I use it in
photographic
printing, in a process that reacts badly with almost everything except pure
elemental gold.
>
>> >No powder.
>>
>> I have a very old bottle of powdered gold pigment in suspension in gum arabic
>> sitting on my shelf,
>> for use on special paintings. But if you say there's no such thing, I guess I
>> better inform the
>> manufacturer that they're selling an imaginary product. They will be quite
>> disappointed since they
>>take great pride in describing how this product is manufactured according to an
>> age-old traditional
>> formula.
>
>From what?
From gold. You could settle the issue quite easily. I'll sell you one of my gold
prints, and you can
submit it for Xray spectroscopy. I have a nice print that is entirely printed in
powdered gold
pigment, I'll part with it at a huge discount, only $2000, far below market
prices that usually start
around $4000. As for me, I'll just believe what's written on the label.
>
>> >And only in very sumptuous mss., not in more ordinary ones, or e.g.,
>> >stained glass windows or tapestries as at Bayeux.
>>
>> Apparently you are a complete idiot and know nothing about art. Until the
>> invention of aniline
>>pigments in the 1850s, nearly all bright pigments were expensive and precious,
>> used only
>> sparingly, and even then, only when a rich patron commissioned an expensive
>> work. Look at Old
>> Master paintings like Rembrandt,
>
>Do you see anything about "paintings" in what I wrote? What Rembrandt
>did wasn't even invented until centuries after the period we're talking
>about, viz., when "orange" entered the English language.
I don't know what the hell you're talking about, you blathered on about applying
gold to stained
glass and tapestries, which makes no sense whatsoever. Applying gold to glass
would make it
opaque, rendering it useless as stained glass. Tapestries were woven with gold
threads. What WE
are talking about in this thread is the etymology of color terms in the Japanese
language. What did
YOU think this thread was about?
>> you'll notice they are almost entirely executed
>> in tones of brown
>>and yellow, but masterfully use simultaneous contrast to make the color spectrum
>> of the paints
>> look wider than it actually is. Even pure white was expensive, and typically
>> only used for small
>>highlights. Rembrandt was a cheapass S.O.B. Michelangelo was even cheaper, you
>> ought to read his
>> notebooks, he constantly complains about the expense of pigments.
>
>Look a little lower and smaller. Look at the Book of Kells, look at the
>various (Très) Riches Heures available for admiration.
Since we're not talking about English etymology in this thread, that would be
irrelevant. I only cited
world artists as an example other techniques used to AVOID expensive pigments.
Perhaps you
might investigate relevant Japanese artworks, like this ukiyo-e made in the 17th
century, using
both gold leaf and gold inks:
http://www.nla.gov.au/worldtreasures/html/theme-literature-8-genji.html
Or perhaps you can look up the original Heian era genji emaki, which are even
older. Other ancient
lacquerware techniques predating your British calligraphy examples would be
useful to your
research.
>Applying gold to glass would make it opaque, rendering it useless as stained glass.
Whilst mostly agreeing with you on this one, gold is often applied to
glass whilst still being transparent. One example that springs to mind
is the visors on Astronauts space suits. I can't see why gold 'one way
glass' can't be made in the same way as conventional one. No idea how
this could be done way back when but it's certainly possible today
using electrolysis.
I'm sure a fine spray of gold pigment could achieve some sort of
transparent result but that would then tend to prove your theory
correct, not Peters.
-------------------------------------------------------------
Dale Walker London Techno Events Saiko!
da...@sorted.org lon...@sorted.org sa...@sorted.org
London, UK london.sorted.org saiko.sorted.org
That technique usually involves vapor deposition, a technique developed in the
semiconductor
industry. It has to be done in an almost complete vaccuum under an intense
electric field. I don't
think that even von Guericke's vacuum pump could have been adapted to a
manufacturing process.
BTW, I should plug one of my favorite contemporary artists, who uses vapor
deposition of metal in
his artworks. He crumples up sheets of gold mylar and other metal foils, lays
them on a sheet of
paper or canvas in a vacuum chamber, runs high voltage through the chamber, and
the metals
evaporate from the mylar and foil, and deposit on the paper. That's the most
convoluted way I
know to apply metal to a painting, but the results are really incredible.
http://larrybell.com/available/mirage.html
>I'm sure a fine spray of gold pigment could achieve some sort of
>transparent result but that would then tend to prove your theory
>correct, not Peters.
To get the transparent effect, the gold has to be applied in monoatomic layers,
the final result is
only a few nanometers thick. It requires an extremely high level of technology
that did not exist
prior to the late 1960s.
Even ancient glassblowers have been known to roll hot glass in gold flake, but
it merely makes
opaque flecks of gold inside the glass. That doesn't sound very likely to be
used in stained glass
windows, the flecks of gold would only be visible upon close inspection. Metals
can be
transparently incorporated into the glass itself, i.e. lead glass. Gold glass is
known as ruby glass,
it's a transparent red or purple color. That isn't likely to be used to
represent gold in a stained
glass window.
But none of this has anything to do with application of gold pigments to
surfaces by painting or
printing, which is the topic under dispute.
I'm not lying. They claimed to have found there exist 11
universal colour foci defining 11 basic colour categories
and a universal evolutionay scheme that dictates the order
of their lexicalization. And among those 11 supposedly
universal foci are the ones for brown and orange but not
the ones for azure or glaucous. So a language with a basic
glaucous but lacking a basic brown, like Cantonese(*), is
supposed to be breaking in two the basic universal
category of green by adding a non-basic focus for glaucous
and _heretically_ turning it into a basic focus, something
that, if ever, should occur only well after stage VII when
all the 11 'real' basic universal categories have been
lexicalized. But that's not the case here because the
'universal' category of brown isn't yet recognized as
basic in Cantonese, being it merely a dark kind of yellow.
So what stage is Cantonese supposed to be in, since by
lacking brown it clearly hasn't yet accomplished stage VI
while OTOH seems to be already in stage VII with its grey,
orange and purple and way further into the unknown with
its 'anomalous' basic glaucous? How can Berlin & Kay
explain that Cantonese uses a glaucous focus as basic
when it's not one of those '11 universal basic foci'
they found, are Cantonese people from Mars?
(*) The basic colours of Cantonese are:
stage I: baahk (WHITE), haak (BLACK)
stage II: huhng (RED, + PINK)
stage III: wohng (YELLOW, + BROWN)
stage IV: luhk (GREEN, - GLAUCOUS)
stage V: laahm (BLUE)
stage VI: -
stage VII: fui (GREY), chaang (ORANGE), ji (PURPLE), -
stage Martian: cheng/ching (GLAUCOUS)
A paper on Cantonese colours can be found here:
http://www2.soas.ac.uk/Linguistics/papers/Muisoaswp1999.pdf
>> I (German) would call the color of carrots "orange", but one traditional
>> German term is /Gelbe/ /Rüben/ (/yellow/ /roots/). Thus, it should be
>> possible to call oranges yellow, or red, too.
>
> Modern Basque has mimicked the colour semantics of the
> surrounding prestige languages Spanish and French, but
> anciently "gorri" (red) had a wider meaning that included
> brown and orange. This can be seen e.g. in the traditional
> practice of nicknaming a brown cow "Gorri"
The most common color of cats beside black, white, and grey is called
/rot/ (red) in German, I have never heard it otherwise, though I
wondered if /orange/ wouldn't be more exact.
Oliver C (not Spam)
PS F'up set, please regard.
I don't click links.
> The author of that page is also the co-author with
> David Hockney of a
> recent book about ancient art techniques.
Hockney's assertions about everyone using a camera lucida to lay out
their paintings hasn't found wide acceptance in the art history
community, it doesn't seem.
One Christmas I gave my calligraphy teacher some pure gold wire (sold,
IIRC, as a dental supply) of a thickness to go into a mechanical
pencil-holder, so that he could do goldpoint sketches along with the
silverpoints he was rather good at. (I don't care to part with the nudes
he did of me in the medium.) (It cost about $20 for about an inch,
around 1977.) He may have been the first artist to work in goldpoint
since the Renaissance.
> >> >And only in very sumptuous mss., not in more ordinary ones, or e.g.,
> >> >stained glass windows or tapestries as at Bayeux.
> >>
> >> Apparently you are a complete idiot and know nothing about art. Until the
> >> invention of aniline
> >> pigments in the 1850s, nearly all bright pigments were expensive and precious,
> >> used only
> >> sparingly, and even then, only when a rich patron commissioned an expensive
> >> work. Look at Old
> >> Master paintings like Rembrandt,
> >
> >Do you see anything about "paintings" in what I wrote? What Rembrandt
> >did wasn't even invented until centuries after the period we're talking
> >about, viz., when "orange" entered the English language.
>
> I don't know what the hell you're talking about, you blathered on about applying
> gold to stained
> glass and tapestries, which makes no sense whatsoever.
No, I didn't. I pointed out that in those two media, gold was depicted
with yellow rather than orange, as later.
> Applying gold to glass
> would make it
> opaque, rendering it useless as stained glass. Tapestries were woven with gold
> threads.
Once again, your sense of chronology is failing you.
> What WE
What you mean WE, kimosabe?
> are talking about in this thread is the etymology of color terms in the Japanese
> language. What did
> YOU think this thread was about?
Basic color terms. And, more recently, when "orange" entered the English
language, and how orange items were depicted before then.
> >> you'll notice they are almost entirely executed
> >> in tones of brown
> >>and yellow, but masterfully use simultaneous contrast to make the color spectrum
> >> of the paints
> >> look wider than it actually is. Even pure white was expensive, and typically
> >> only used for small
> >>highlights. Rembrandt was a cheapass S.O.B. Michelangelo was even cheaper, you
> >> ought to read his
> >> notebooks, he constantly complains about the expense of pigments.
> >
> >Look a little lower and smaller. Look at the Book of Kells, look at the
> >various (Très) Riches Heures available for admiration.
>
> Since we're not talking about English etymology in this thread,
No, we're talking about basic color terms.
> that would be
> irrelevant. I only cited
> world artists as an example other techniques used to AVOID expensive pigments.
> Perhaps you
> might investigate relevant Japanese artworks, like this ukiyo-e made in the 17th
> century, using
> both gold leaf and gold inks:
> http://www.nla.gov.au/worldtreasures/html/theme-literature-8-genji.html
> Or perhaps you can look up the original Heian era genji emaki, which are even
> older. Other ancient
> lacquerware techniques predating your British calligraphy examples would be
> useful to your
> research.
British?
Where'd you get "hardwired into every human being"? Algross is going to
start taking off after _you_ with talk like that.
Ah, then you have no idea how your shaky theory was blasted to shreds. Go read
it.
>
>> The author of that page is also the co-author with
>> David Hockney of a
>> recent book about ancient art techniques.
>
>Hockney's assertions about everyone using a camera lucida to lay out
>their paintings hasn't found wide acceptance in the art history
>community, it doesn't seem.
Speaking as a member of the art history community, I assure you that many of
Hockney's theories
are accepted as indisputable. And besides, we're not talking about optics, we're
talking about XRay
Spectroscopy tests Hockney's buddy from Berkeley performed. Hockney's research
on materials is
impeccable, and widely accepted. It's not theory, it's scientific fact.
Nope. I was doing goldpoint and silverpoint well before 1977. My teachers all
did gold and
silverpoint, they taught it to me. It is a very common technique. But then, you
seem to think a lot of
common techniques are uncommon.
And I don't know what the hell ANY of that has to do with your claim that gold
can't be powdered,
that gold can't be painted, gold was never represented with gold pigments, etc
etc. You seem to
like to divert attention from your smackdown with irrelevant tangents. Your
tactic isn't working,
maybe you should quit while you're so far behind.
>> >> >And only in very sumptuous mss., not in more ordinary ones, or e.g.,
>> >> >stained glass windows or tapestries as at Bayeux.
>> >>
>> >> Apparently you are a complete idiot and know nothing about art. Until the
>> >> invention of aniline
>>>> pigments in the 1850s, nearly all bright pigments were expensive and
>>precious,
>> >> used only
>>>> sparingly, and even then, only when a rich patron commissioned an expensive
>> >> work. Look at Old
>> >> Master paintings like Rembrandt,
>> >
>> >Do you see anything about "paintings" in what I wrote? What Rembrandt
>> >did wasn't even invented until centuries after the period we're talking
>> >about, viz., when "orange" entered the English language.
>>
>>I don't know what the hell you're talking about, you blathered on about applying
>> gold to stained
>> glass and tapestries, which makes no sense whatsoever.
>
>No, I didn't. I pointed out that in those two media, gold was depicted
>with yellow rather than orange, as later.
You pointed out a lot of incoherent things unrelated to the topic of this
thread, and provided no
evidence whatsoever. I merely pointed out that gold was often depicted IN GOLD,
which you
claimed was impossible. Considering your abysmal track record here, I think you
ought to stop
making obviously incorrect claims.
>> Applying gold to glass
>> would make it
>>opaque, rendering it useless as stained glass. Tapestries were woven with gold
>> threads.
>
>Once again, your sense of chronology is failing you.
OK, I guess that with your comprehensive knowledge of art history, you can
declare I made a
chronological error, therefore nobody ever wove tapestries with gold thread, and
gold applied to
stained glass isn't opaque. You're an idiot, you know that?
>> What WE
>
>What you mean WE, kimosabe?
For starts, the ORIGINATOR OF THIS THREAD, or perhaps you didn't notice the
subject is "Two
Japanese color terms?"
>>are talking about in this thread is the etymology of color terms in the Japanese
>> language. What did
>> YOU think this thread was about?
>
>Basic color terms. And, more recently, when "orange" entered the English
>language, and how orange items were depicted before then.
I don't know what the hell you're going on about, the thread was about the
etymology of midori
and kiiroi, until you sci.lang pedants hijacked it into a dispute about who
could stuff the biggest
orange up their asshole.
>
>> >> you'll notice they are almost entirely executed
>> >> in tones of brown
>>>>and yellow, but masterfully use simultaneous contrast to make the color
>>spectrum
>> >> of the paints
>>>> look wider than it actually is. Even pure white was expensive, and typically
>> >> only used for small
>>>>highlights. Rembrandt was a cheapass S.O.B. Michelangelo was even cheaper, you
>> >> ought to read his
>> >> notebooks, he constantly complains about the expense of pigments.
>> >
>> >Look a little lower and smaller. Look at the Book of Kells, look at the
>> >various (Très) Riches Heures available for admiration.
>>
>> Since we're not talking about English etymology in this thread,
>
>No, we're talking about basic color terms.
Right, like kiiroi and midori? I must have missed your contribution. You seem to
be primarily
talking about things you have no knowledge of.
>> that would be
>> irrelevant. I only cited
>>world artists as an example other techniques used to AVOID expensive pigments.
>> Perhaps you
>>might investigate relevant Japanese artworks, like this ukiyo-e made in the 17th
>> century, using
>> both gold leaf and gold inks:
>> http://www.nla.gov.au/worldtreasures/html/theme-literature-8-genji.html
>> Or perhaps you can look up the original Heian era genji emaki, which are even
>> older. Other ancient
>> lacquerware techniques predating your British calligraphy examples would be
>> useful to your
>> research.
>
>British?
Oh great, another tangent you can bludgeon everyone with. Perhaps you can now
republish your
long-ignored dissertation on why Celtic isn't British.
Let me explain something to you, before I give up entirely. There is a huge gap
between theory and
practice. You are all theory, and no practice. You are lecturing someone with
many decades of
DAILY PRACTICAL EXPERIENCE in the very art production techniques that you claim
are impossible.
You should learn not to dispute art techniques with an artist, and in
particular, you should not
dispute metallic pigment usage with an artist who specializes in metallic
pigments.
Who on Earth told you THAT? Every sun I've ever seen a child draw here
(Shizuoka, Japan)has been yellow, without exception. (Well, there are
some creative exceptions. We ARE talking children here.)
>>> But the Rising (and Setting) Sun _is_ red.
>>
>> But the Japanese children paint the sun red always, high in the sky,
>> too.
>
> Who on Earth told you THAT? Every sun I've ever seen a child draw here
> (Shizuoka, Japan)has been yellow, without exception. (Well, there are
> some creative exceptions. We ARE talking children here.)
Well, my "always" wasn't a scientific use, but your contrary "every" is
surprising to me. It is a set issue of intercultural comparison talk,
and I have seen more examples confirming the theory than otherwise,
though my experience with children's drawings are limited.
A little browsing through <http://www.kids-space.org/indexJ.html> gave
me the following statistics for kids from Japan:
red 7
yellow 3
orange 1
pink 1
yellow with red border 1
violet 1 (sun?)
The other set issue about color is the prototypical color of apples.
When I saw the apple on a Japanese yoghurt package, I intuitively
expected cherry flavor.
Oliver C.
It might amuse you to hear of a psychology study I read. The researchers had
jellybeans
manufactured with the "wrong" colors, i.e. red was lime, blue was cherry, etc.
They gave people the
jellybeans and asked them what flavor they were. Over 75% reported a flavor that
corresponded to
the color, not the actual flavor.
That's bizarre. Kids in Shizuoka must differ from the rest of the
country then.
I was in Toda Syoten this afternoon looking for books for my daughter,
and one of them was a book on colours. It has flip up panels on it. It
shows a colour, and when you flip it up it shows something with that
colour (ie red has a picture of an aple underneath). Under yellow was
a picture of a sun.
I've seen violet suns and pink suns and black(!) suns, but I don't
recall any red ones.
> >> What WE
> >
> >What you mean WE, kimosabe?
>
> For starts, the ORIGINATOR OF THIS THREAD, or perhaps you didn't notice the
> subject is "Two
> Japanese color terms?"
>
> >>are talking about in this thread is the etymology of color terms in the Japanese
> >> language. What did
> >> YOU think this thread was about?
> >
> >Basic color terms. And, more recently, when "orange" entered the English
> >language, and how orange items were depicted before then.
>
> I don't know what the hell you're going on about, the thread was about the
> etymology of midori
> and kiiroi, until you sci.lang pedants hijacked it into a dispute about who
> could stuff the biggest
> orange up their asshole.
s.l.j. is just a nom de plume for soc.culture.japan? How sad.
> You should learn not to dispute art techniques with an artist, and in
> particular, you should not
> dispute metallic pigment usage with an artist who specializes in metallic
> pigments.
Or maybe "an artist" shouldn't try to dispute linguistics.
> Today I went to the Met Museum to see the El Grecos (which are leaving
> after Sunday), and happily there were lots of other special shows that
> were much more interesting. One of them was of Italian manuscript
> paintings in the Lehman Collection (14th-century, mostly), so I had a
> serendipitous opportunity to check for orange paint. Sho' 'nuff, none!
> Except for one piece, the last one in the sequence, which from across
> the way I could see had orange in it. And whadya know, it was a late
> 19th century imitation of early Italian manuscript painting!
>
> There was even a rainbow, an oval around a picture of a Creator God. It
> used just three colors of paint, red, yellow, and blue, and five
> apparent colored bands were made by streaking red and yellow together
> (not mixing them on the palette, but side by side) between the red and
> yellow bands, and similarly streaking yellow and blue together. (BTW,
> Newton supposedly put "indigo" in the rainbow list because he wanted
> there to be seven color names, for mystical reasons.)
How _else_ can a universal language tendency be
explained if not because it is in some way innate?
The logical corolary to claiming that a certain language
feature is universal is that it is an innate feature
hardwired into the mind of every human being, because
there's no other possible explanation to its regular
appearance regardless of language, culture and location
than because it is an innate feature of the human mind.
Unless, of course, you attempt to explain it as a kind
of divine enforcement or as the result of a superior
alien intelligence secretly controlling human minds.
> It might amuse you to hear of a psychology study I read. The researchers had
> jellybeans
> manufactured with the "wrong" colors, i.e. red was lime, blue was cherry, etc.
> They gave people the
> jellybeans and asked them what flavor they were. Over 75% reported a flavor that
> corresponded to
> the color, not the actual flavor.
I believe this offhand. I had the same experience with (transparent)
"Crystal Pepsi" - tastes of lemon, and only lemon, to me.
Oliver C.
Apparently you failed to read my remarks about how Old Masters avoided the use
of the most
expensive pigments by using simultaneous contrast effects. But that doesn't
surprise me since
you're mostly interested in listening to yourself.
Maybe, unless they have academic qualifications in that area (check) or are
disputing some idiot
trying to make an invalid linguistic point from inaccurate artistic evidence
(doublecheck).
In my case, I drew a white moon and a yellow sun on the board once when
I was a JET, and all the kids laughed. The teacher told me that in Japan
it was yellow and red, respectively.
I guess people in Shizuoka are just weirdos.
Oh, come on, Peter. Don't raise the white flag like that! This was
getting so interesting! I was learning so much about the topic from
Charles.
It's great how you were playing foil to his knowledge and facts. You
seem to know as much about this topic as gasoline prices! (I'm still
looking for a gas pump that shows the price to three decimal places...)
Well of course. Every school child knows the basic primary color spectrum:
Pink Hearts
Orange Stars
Yellow Moons
Blue Diamonds
Green Clovers
It's magically delicious!
>
> Well of course. Every school child knows the basic primary color spectrum:
>
> Pink Hearts
> Orange Stars
> Yellow Moons
> Blue Diamonds
> Green Clovers
>
> It's magically delicious!
>
But do they make the green out of powdered leprechauns?
KWW
"Innate" and "hard-wired" are not the same thing. All that's
"hard-wired" is a particular batch of enzymes & other proteins that
somehow sort themselves into organisms, and one property of the human
brain is the re-creation of at least one language in every individual,
and every individual does certain linguistic things the same way. Does
that mean that, e.g., subject/predicate structure is "hard-wired"?
> The logical corolary to claiming that a certain language
> feature is universal is that it is an innate feature
> hardwired into the mind of every human being, because
> there's no other possible explanation to its regular
> appearance regardless of language, culture and location
> than because it is an innate feature of the human mind.
>
> Unless, of course, you attempt to explain it as a kind
> of divine enforcement or as the result of a superior
> alien intelligence secretly controlling human minds.
You're being loose with language, is all.
Have you no sense of time? At long last, "an artist," have you no sense
of chronology?
These "Old Masters" of yours are considerably younger than 14th-century
manuscripts and are utterly irrelevant to the absence of orange in
Medieval art.
Conventionally they begin with Giotto, no? who was working about the
same time his buddies were painting manuscripts? Now I don't know
whether Giotto used orange or not (I've never seen his work in the
flesh, since frescoes probably don't travel very well), but I'd doubt
it.
I also wouldn't care to find out by looking at reproductions. No matter
how much I like an art exhibit, I can never buy the catalog on sale
outside the display, because after having just seen the originals, I
also see how poorly the printed images reproduce them. This was
especially the case in the show of Hudson River School painting, where
the label for one work rhapsodizes about the red of the undersides of
the clouds in a sunset -- and in the book they're sort of a weak peach.
I certainly do. Every time I read one of your misinformed statements, I feel as
if I've aged 10 years.
>These "Old Masters" of yours are considerably younger than 14th-century
>manuscripts and are utterly irrelevant to the absence of orange in
>Medieval art.
You keep changing your point. First you were arguing that in ancient artworks,
gold objects were
depicted with orange. Now you're arguing that they didn't use gold OR orange,
despite evidence
that you disbelieve because you refuse to look at it.
So I wil go to the laborious effort of transcribing text from the authoritative
treatise on art
materials throughout history, The Artist's Handbook by Ralph Mayer. And since
this thread is about
the etymology of Japanese color terms, I will focus on that topic.
Introductory notes, p. 17.
"As with the earliest European painters, permanence was regarded by the Chinese
and Japanese as
an essential requirement in a work of art. The Japanese have carried on this
tradition and today
there are still some traditional painters that use the same twenty or so
pigments that were used
cenuries ago (see page 137). Although most colors on the list are of natural
mineral origin, one
(indigo) is derived from a plant and another (cochineal) from an insect."
pp. 137-8
"White:
Ground quartz crystal (sui sho matsu)
Ground calcite (hokai matsu)
Shell white (go fun)
Mica (unmo)
The whites have various textural and reflective characteristics.
Blue:
Powdered Azurite (gunjo)
Very finely powdered Azurite (byankungun)
Japanese Indigo (ai)
Green:
Malachite (byaku roku)
Bluish Green:
Azurite and malachite mixture
Malachite rich (shin sha)
Burnt azurite and malachite
Azurite rich (yakigunroku)
Red:
Pure vermillion
Cochineal red (enji)
Cochineal crimson
Red lean (tan)
Red earth color (benigara)
Yellow earth (ado)"
Note that yellow earth is listed under red, it is more ochre than a proper
yellow. Note that all these
materials are derived from unrefined minerals. You can't mix orange from yellow
earth and any red,
so they just didn't try. There was no way to produce orange pigments in ancient
Japan, not even
from primary red and yellow, because there was no primary yellow, just a murky
tan/ochre
yellowish pigment.
THIS is why I described the Old Masters. They were well versed in painting
techniques of their
predecessors. Most painting, even well into the renaissance, were executed in
dark tones, plenty of
deep reds and blacks, but mostly in ochres and umbers and other earth tones,
because these
pigments were dirt cheap. They were MADE from dirt dug from the ground. Pigments
like Iron
Yellow were unstable, Yellow Ochre wasn't intense enough, AFAIK it wasn't until
nearly modern
times that properly stable, intense yellow pigments made from cadmium, barium,
chromium etc.
were commonly available. Prior to the availability of modern pigments, artists
faked it. Pigments
were usually applied in flat colored areas, with no color mixing, they made weak
yellow ochre look
bright yellow by contrasting it against a background of a darker, complementary
color that would
make it appear brighter. Colors were not blended in smooth graduations, they
were applied flat
with hard edges. This is one of the hardest things to impress upon painting
students, they want to
blend everything, but that's just not the way to get the best effects, and tends
to cause weak,
murky colors.
I could go on and on, but talking about painting is like talking about music,
it's futile unless you
can get someone to experience the real thing, something you refuse to do. So let
me take one
more crack at answering one of your ever-shifting propositions:
There are two reasons why you do not observe orange in ancient paintings:
1. There IS orange, but it's not intense enough for you to call it orange by
your modern standards.
Viewers contemporary to the work would have recognized it as orange.
2. There ISN'T anything you'd call orange, because bright yellow and orange
pigments hadn't been
discovered yet. They faked it with simultaneous contrast effects.
So, are you satisfied now? Or shall I continue to pummel you?
>Conventionally they begin with Giotto, no? who was working about the
>same time his buddies were painting manuscripts?
Actually, my Modern Art History classes started with Giotto. I was rather
surprised since
traditionally, Modern Art classes start with Jacques-Louis David and the French
Neoclassicists.
Nonetheless, I shall accept your restrictive definition of Old Masters as a
small group of artists that
postdate the works you are talking about. Which implies that these artists
studied and knew the
techniques of prior artists, since artists of that time worked in
apprenticeships. And guess what?
Giotto used GOLD to depict stars in the sky, halos, etc. He also used a weak
solid orange. Giotto
had extremely rich patrons who could afford expensive materials and gave him an
almost unlimited
budget, he is most known for his extensive use of powdered lapis lazuli for dark
blue skies.
>Now I don't know
>whether Giotto used orange or not (I've never seen his work in the
>flesh, since frescoes probably don't travel very well), but I'd doubt
>it.
Yes, he did use orange (albeit weak intensities), gold, and many other precious
pigments in his
work.
>I also wouldn't care to find out by looking at reproductions.
Photographs are sufficient for art historians and students, but not you?
Apparently you cannot see
beyond the literal appearance of things. Sometimes photographs can capture what
the eye cannot.
For example, one of my favorite art illustrations is of a Rembrandt portrait, it
shows the overall
view of the painting reduced to about 25% of actual size, but in a large, full
page color photograph
that is a quite adequate for the original, considering you'd never really be
allowed to get very close
to the original. The sitter is wearing an ornate necklace of flat gold pieces,
they appear to be
extremely intricately painted, and seem to reveal an immense amount of detail.
But there is an
extreme closeup photo of one of the necklace pieces on the next page, and it is
astonishing.
Rembrandt just took a dry round brush, dipped it into ochre, and went SPLAT into
the canvas, there
IS no detail, it is all an illusion. You would never be able to see this by
viewing the painting
firsthand. I know, I saw the painting and was chased away by the guard for
getting too close.
I cannot put it to you more forcefully: artists are FAKERS. We spend our lives
trying to trick your
eye. We see things you cannot, because we know how people see and you do not. We
make the
easy look hard, and the hard look easy, just to impress you. Trying to deduce
linguistics from the
work of artists is like trying to learn surgery from the magician who saws a
lady in half.
>No matter
>how much I like an art exhibit, I can never buy the catalog on sale
>outside the display, because after having just seen the originals, I
>also see how poorly the printed images reproduce them.
If you want to stick your head in the sand, and refuse to even click on a web
link with TEXT
descriptions that devastate your crackpot theories, then I doubt you'd even be
convinced if I
dropped the Scrovegni Chapel on your head. In any case, the Chapel is currently
being restored and
only accessible to qualified scholars, which means you aren't invited.
>This was
>especially the case in the show of Hudson River School painting, where
>the label for one work rhapsodizes about the red of the undersides of
>the clouds in a sunset -- and in the book they're sort of a weak peach.
You have terrible taste in art.
You have the reading comprehension, and/or the memory, of a toad. From
the very beginning, I have been saying that gold objects were NOT
depicted with orange.
> So I wil go to the laborious effort of transcribing text from the authoritative
> treatise on art
> materials throughout history, The Artist's Handbook by Ralph Mayer. And since
> this thread is about
> the etymology of Japanese color terms, I will focus on that topic.
This thread is not about the etymology of Japanese color terms.
> so they just didn't try. There was no way to produce orange pigments in ancient
> Japan,
Finally mentioning something relevant.
> I could go on and on, but talking about painting is like talking about music,
> it's futile unless you
> can get someone to experience the real thing, something you refuse to do. So let
> me take one
> more crack at answering one of your ever-shifting propositions:
I'm the one who went to the Museum on Friday. (It was mobbed.)
> There are two reasons why you do not observe orange in ancient paintings:
>
> 1. There IS orange, but it's not intense enough for you to call it orange by
> your modern standards.
> Viewers contemporary to the work would have recognized it as orange.
>
> 2. There ISN'T anything you'd call orange, because bright yellow and orange
> pigments hadn't been
> discovered yet. They faked it with simultaneous contrast effects.
>
> So, are you satisfied now? Or shall I continue to pummel you?
3. The conceptual category "orange" didn't exist yet.
> I cannot put it to you more forcefully: artists are FAKERS.
(Speak for yourself.)
> We spend our lives
> trying to trick your
> eye. We see things you cannot, because we know how people see and you do not. We
> make the
> easy look hard, and the hard look easy, just to impress you. Trying to deduce
> linguistics from the
> work of artists is like trying to learn surgery from the magician who saws a
> lady in half.
So don't be surprised when, when you try the same tricks outside
painting, you get called on them.
> >No matter
> >how much I like an art exhibit, I can never buy the catalog on sale
> >outside the display, because after having just seen the originals, I
> >also see how poorly the printed images reproduce them.
>
> If you want to stick your head in the sand, and refuse to even click on a web
> link with TEXT
> descriptions that devastate your crackpot theories, then I doubt you'd even be
> convinced if I
> dropped the Scrovegni Chapel on your head. In any case, the Chapel is currently
> being restored and
> only accessible to qualified scholars, which means you aren't invited.
Are you claiming that there are printed reproductions of the Scrovengi
Chapel that reproduce the original more accurately than the $50 catalogs
prepared by the Metropolitan Musuem's staff?
> >This was
> >especially the case in the show of Hudson River School painting, where
> >the label for one work rhapsodizes about the red of the undersides of
> >the clouds in a sunset -- and in the book they're sort of a weak peach.
>
> You have terrible taste in art.
It's tasteless to want a reproduction to accurately reproduce an
original?
Larry Trask
lar...@sussex.ac.uk
I'm not sure what qualifies as derivation, but just to make sure there's
no misunderstanding, <midori> *is* the word for vegetation. The two
meanings can only be distinguished through context (unless
<midoriiro(i)> is used, in which case it is only "green").
Interesting how the thread morphed into a mudslinging match between two
eggheads. I started a thread about the spread of tanuki in eastern Europe
which quickly became a thread about weight loss and Japanese airline workers
with French accents or something.
--
http://www.soundclick.com/bands/0/seanhollandmusic.htm
pantssea...@telus.pants.net Remove pants to email me.
OK, I'm going to start way back in this thread, since you refuse to stick to one
point and keep
shifting around to other dubious premises of obvious incorrectness. Here I will
disprove you with
your own words.
>> >> >> There was even a rainbow, an oval around a picture of a Creator God. It
>> >> >> used just three colors of paint, red, yellow, and blue, and five
>> >> >> apparent colored bands were made by streaking red and yellow together
>> >> >> (not mixing them on the palette, but side by side) between the red and
>> >> >> yellow bands, and similarly streaking yellow and blue together. (BTW,
>> >> >> Newton supposedly put "indigo" in the rainbow list because he wanted
>> >> >> there to be seven color names, for mystical reasons.)
Despite your deliberate obtuseness and refusal to click on links, here you will
be forced to do so,
or else you will be proven not only ignorant, but willfully ignorant, and proven
so in front of anyone
who can click a link (which means everyone).
The following link is from the Bodelian Library collection of ancient
manuscripts.
http://www.bodley.ox.ac.uk/dept/scwmss/wmss/medieval/mss/lat/liturg/d/042.htm
Notice the date of the manuscript, 1307. Now click on a higher resolution image,
like:
http://www.bodley.ox.ac.uk/dept/scwmss/wmss/medieval/jpegs/lat/liturg/d/042/1000/
04200752.jpg
Gosh, is that ORANGE? It surely isn't red because there are other colors in the
illustration that
clearly are red, and the orange shade is significantly different. It isn't gold,
because there is gold in
other parts of the illustration.
Now how is it that you can claim that there was no such concept as orange in
this time period? You
clearly stated so later in your message. I'll requote you here:
>3. The conceptual category "orange" didn't exist yet.
In your answer, perhaps you can explain how the English used orange pigment, but
the French and
the Japanese didn't. Dont' bother to switch dates again, you clearly cited works
without orange like
Tres Riche Heures as your evidence, yet that work postdates my citation by a
hundred years.
>>So I wil go to the laborious effort of transcribing text from the authoritative
>> treatise on art
>> materials throughout history, The Artist's Handbook by Ralph Mayer. And since
>> this thread is about
>> the etymology of Japanese color terms, I will focus on that topic.
>
>This thread is not about the etymology of Japanese color terms.
I guess you better explain that to the thread's originator, who just today,
wrote this:
In article <48c7f19.04010...@posting.google.com>, Larry Trask says...
]
]I would like to thank everyone who wrote in with etymological
]information on those two color terms. The information provided was
]fascinating, but I guess the conclusion is that no cut-and-dried
]etymology exists for either term, except that <midori>, if I've
]understood correctly, does appear to derive *directly* from a term for
]'vegetation'.
]
]Larry Trask
]lar...@sussex.ac.uk
>
>>so they just didn't try. There was no way to produce orange pigments in ancient
>> Japan,
>
>Finally mentioning something relevant.
Which is why you snipped the entire text I cited? Note that I gave you a couple
of gaping holes to
drive through, but you didn't recognize them. So I made one of the holes a
little more explicit in
this message. Go back to the beginning of this msg and read it again. And then
go back to my
original text that you snipped, where I showed you how Giotto used orange
pigments. You cut
those remarks because you cannot respond to evidence that destroys your
theories. Giotto used
orange, you claim he didn't. You cannot prove your point by merely cutting the
evidence that
disproves your point.
Here's a good example:
http://www.mystudios.com/gallery/giotto/03a.html
Note that the mild orange in the kneeling figure is different from the red in
the chest to the right,
and the yellow in the figure on the left. That one image pretty much shows the
whole spectrum of
warm colors, a color distinction you claim did not exist at that time.
>> I could go on and on, but talking about painting is like talking about music,
>> it's futile unless you
>>can get someone to experience the real thing, something you refuse to do. So let
>> me take one
>> more crack at answering one of your ever-shifting propositions:
>
>I'm the one who went to the Museum on Friday. (It was mobbed.)
Goody for you. You DO realize that most museums only have about 1% of their
collection on
display, and that for proper research you must consult their catalogs with (oh
the horror!)
PHOTOGRAPHS of their collection?
>> There are two reasons why you do not observe orange in ancient paintings:
>>
>> 1. There IS orange, but it's not intense enough for you to call it orange by
>> your modern standards.
>> Viewers contemporary to the work would have recognized it as orange.
>>
>> 2. There ISN'T anything you'd call orange, because bright yellow and orange
>> pigments hadn't been
>> discovered yet. They faked it with simultaneous contrast effects.
>>
>> So, are you satisfied now? Or shall I continue to pummel you?
>
>3. The conceptual category "orange" didn't exist yet.
You have clearly been proven wrong. In case you missed it:
http://www.bodley.ox.ac.uk/dept/scwmss/wmss/medieval/jpegs/lat/liturg/d/042/1000/
04200752.jpg
The English had orange pigment, the Japanese did not, merely because there were
no locally
available material to make orange pigment. Minerals are not evenly distributed
upon this planet.
This has nothing to do with perception. If the Japanese had orange minerals, or
even sufficently
intense yellow pigments to make orange, they would have used them. They didn't,
so they used
ochre and gold, the same as everywhere else that orange pigments were not
available.
>> I cannot put it to you more forcefully: artists are FAKERS.
>
>(Speak for yourself.)
Now that is just stupid. Every art object involves fakery, every art object is
an abstraction. The
history of Art is the history of evolution of little tricks artists used to fake
out the eye and brain.
But I guess you are incapable of seeing this, since you cannot see beyond the
literal appearance of
things.
>> We spend our lives
>> trying to trick your
>>eye. We see things you cannot, because we know how people see and you do not. We
>> make the
>> easy look hard, and the hard look easy, just to impress you. Trying to deduce
>> linguistics from the
>> work of artists is like trying to learn surgery from the magician who saws a
>> lady in half.
>
>So don't be surprised when, when you try the same tricks outside
>painting, you get called on them.
And what tricks would those be? Tricks like FACTS? Be specific. I've cited
dozens of incontrovertible
facts, you have responded with only weak theories.
>> >No matter
>> >how much I like an art exhibit, I can never buy the catalog on sale
>> >outside the display, because after having just seen the originals, I
>> >also see how poorly the printed images reproduce them.
>>
>> If you want to stick your head in the sand, and refuse to even click on a web
>> link with TEXT
>>descriptions that devastate your crackpot theories, then I doubt you'd even be
>> convinced if I
>>dropped the Scrovegni Chapel on your head. In any case, the Chapel is currently
>> being restored and
>> only accessible to qualified scholars, which means you aren't invited.
>
>Are you claiming that there are printed reproductions of the Scrovengi
>Chapel that reproduce the original more accurately than the $50 catalogs
>prepared by the Metropolitan Musuem's staff?
Absolutely. I've seen them. Not only that, I've MADE them, in past times I have
been a professional
color separator at a prepress bureau, and it was my job to ensure color accuracy
in printed works,
which included art catalogs. The things you merely speculate about are my
profession.
But most art scholars work from slides when doing their critical analyses, not
printed works, since
the color gamut of film is greater than the gamut of CMYK printing. We also work
from images in
infrared, UV, XRay spectroscopy, positron/neutron emission tomography, etc etc.
>> >This was
>> >especially the case in the show of Hudson River School painting, where
>> >the label for one work rhapsodizes about the red of the undersides of
>> >the clouds in a sunset -- and in the book they're sort of a weak peach.
>>
>> You have terrible taste in art.
>
>It's tasteless to want a reproduction to accurately reproduce an
>original?
No, if you went to a show of the Hudson River School, you have terrible taste.
They were a minor
dead-end in the art-historical world, most art schools don't even bother to
teach about them, they
are irrelevant. But that's fairly typical of your interests, you went down quite
a few irrelevant dead-
ends in this discussion, and refused to return to reality despite the
preponderance of evidence.
I keep trying to return the subject to Japanese color etymology, but to no
avail. At least I provided
some data that was directly relevant to the original question.
> Interesting how the thread morphed into a mudslinging match between two
> eggheads. I started a thread about the spread of tanuki in eastern Europe
> which quickly became a thread about weight loss and Japanese airline workers
> with French accents or something.
You are supposed to thank me for continuing your thread for that long.
>In article <a8abvv87qrdf14qik...@4ax.com>, Dale Walker says...
>>
>>On 2 Jan 2004 08:47:43 -0800, Charles Eicher <cei...@inav.net> wrote:
>>
>>
>>>Applying gold to glass would make it opaque, rendering it useless as stained
>>>glass.
>>
>>Whilst mostly agreeing with you on this one, gold is often applied to
>>glass whilst still being transparent. One example that springs to mind
>>is the visors on Astronauts space suits. I can't see why gold 'one way
>>glass' can't be made in the same way as conventional one. No idea how
>>this could be done way back when but it's certainly possible today
>>using electrolysis.
>
>That technique usually involves vapor deposition, a technique developed in the
>semiconductor
>industry. It has to be done in an almost complete vaccuum under an intense
>electric field. I don't
>think that even von Guericke's vacuum pump could have been adapted to a
>manufacturing process.
>
>BTW, I should plug one of my favorite contemporary artists, who uses vapor
>deposition of metal in
>his artworks. He crumples up sheets of gold mylar and other metal foils, lays
>them on a sheet of
>paper or canvas in a vacuum chamber, runs high voltage through the chamber, and
>the metals
>evaporate from the mylar and foil, and deposit on the paper. That's the most
>convoluted way I
>know to apply metal to a painting, but the results are really incredible.
>
>http://larrybell.com/available/mirage.html
>
>>I'm sure a fine spray of gold pigment could achieve some sort of
>>transparent result but that would then tend to prove your theory
>>correct, not Peters.
>
>To get the transparent effect, the gold has to be applied in monoatomic layers,
Or sprayed. 50% coverage would result in effectively 50% transparency.
Obviously, the more you spray, the less transparent it becomes.
>the final result is
>only a few nanometers thick. It requires an extremely high level of technology
>that did not exist
>prior to the late 1960s.
>
>Even ancient glassblowers have been known to roll hot glass in gold flake, but
>it merely makes
>opaque flecks of gold inside the glass. That doesn't sound very likely to be
>used in stained glass
>windows, the flecks of gold would only be visible upon close inspection. Metals
>can be
>transparently incorporated into the glass itself, i.e. lead glass. Gold glass is
>known as ruby glass,
>it's a transparent red or purple color. That isn't likely to be used to
>represent gold in a stained
>glass window.
Correct me if I'm wrong but doesn't silver incorporated into the glass
itself give a gold colour?
-------------------------------------------------------------
Dale Walker London Techno Events Saiko!
da...@sorted.org lon...@sorted.org sa...@sorted.org
London, UK london.sorted.org saiko.sorted.org
I hope I can be pardoned for not believing that.
The closest thing I know that could be called *the* (pure Japanese)
word for vegetation is "kusaki."
I am willing to believe that vegetation can be referred to
metaphorically as "midori."
Bart Mathias
Oh, is it a contest to see how long a thread continues? Gee. I naively
thought I was just providing some interesting and diverting information.
Well, thank you, I guess.
If you wanted that effect, you wouldn't need gold, you could just sputter
droplets of opaque black
paint on 50% of the glass, that would cut down transmission of everything. But
this coating is
designed to reduce UV and not visual wavelengths, so it has to be extremely
thin, like near the
wavelength of light. Anyway, the electrochemical deposition technique is pretty
widely known and
used, it's just an extremely high tech version of electroplating except that you
can lay it down on
nonconductive surfaces like glass, due to the vacuum. You can read a little
about it from NASA
http://www.sti.nasa.gov/tto/spinoff1997/hm2.html
>>the final result is
>>only a few nanometers thick. It requires an extremely high level of technology
>>that did not exist
>>prior to the late 1960s.
>>
>>Even ancient glassblowers have been known to roll hot glass in gold flake, but
>>it merely makes
>>opaque flecks of gold inside the glass. That doesn't sound very likely to be
>>used in stained glass
>>windows, the flecks of gold would only be visible upon close inspection. Metals
>>can be
>>transparently incorporated into the glass itself, i.e. lead glass. Gold glass is
>>known as ruby glass,
>>it's a transparent red or purple color. That isn't likely to be used to
>>represent gold in a stained
>>glass window.
>
>Correct me if I'm wrong but doesn't silver incorporated into the glass
>itself give a gold colour?
I never heard of that, some online research yields only a reference to a modern
process for silver
nitrate glass, it's amber. That doesn't sound like a medieval technology.
You may not be pardoned, thank you very much.
> The closest thing I know that could be called *the* (pure Japanese)
> word for vegetation is "kusaki."
I see a tricky qualification in there ("pure Japanese"?). Also those
asterisks... I didn't mean it was the *only* word for vegetation in
Japanese. I'm sure there's a whole gaggle of synonyms.
> I am willing to believe that vegetation can be referred to
> metaphorically as "midori."
Or maybe the trap lies in the use of "vegetation."
The Green Goddess has "green; verdure" for midori. Verdure. Vegetation.
(Cf. *"The verdure of that sweater is just beautiful.")
広辞苑 has
みどり【緑・翠】
(ミドが語根で、「瑞々みずみずし」のミヅと関係があるか)
① 草木の新芽。また、初夏の若葉。広く、植物一般。(季語:夏)。日葡辞書
「ミドリガタツ」。「―したたる山々」「地球の―を護る」
② 青と黄との間色。草木の葉のような色。みどりいろ。万葉集10「春は萌え夏
は―に紅の綵色しみいろに見ゆる秋の山かも」。「―の大地」
③ 深い藍色。天草本伊曾保物語「海の―のなごやかなを見れば」
Note the 「広く、植物一般。」 in definition ①.
So what do you have to say about that?
> "Phil Healey" writes:
>>I'm not sure what qualifies as derivation, but just to make sure
>>there's no misunderstanding, <midori> *is* the word for
>>vegetation. The two meanings can only be distinguished through
>>context (unless <midoriiro(i)> is used, in which case it is only
>>"green").
As you have already said in this thread, "midori" meant "young
shoots, sprouts, buds"(but I don't know "midori" "originally"
meant it) before. That fact makes me guess that "midori-iro"
used to mean rather 黄緑 or 若草色 than 深緑. And I guess that
the sense of "vegetation" of the word have recently developed.
>
>
> I hope I can be pardoned for not believing that.
>
> The closest thing I know that could be called *the* (pure Japanese)
> word for vegetation is "kusaki."
>
> I am willing to believe that vegetation can be referred to
> metaphorically as "midori."
Yes, but "midori" virtually means "vegetation" without any
contexts today. Phrases like the following, words other than
"midori" are rarely used if it is not a technical report or
the like, I think.
・今度、引っ越した先は都会にもかかわらず緑が多い
・緑の乏しい環境
・緑を増やす活動
In other words, colloquially, green plants in general are
referred to as "midori" today and there is no other word which
is more prefered or appropriate.
--
dareka dar...@inter7NS.jp
> ・今度、引っ越した先は都会にもかかわらず緑が多い
> ・緑の乏しい環境
> ・緑を増やす活動
>
> In other words, colloquially, green plants in general are
> referred to as "midori" today and there is no other word which
> is more prefered or appropriate.
That would be kind of like how teh Maltese word haxix refers to
everything from grass to carrots to er, hashish.
--
--
Fabian
Visit my website often and for long periods!
http://www.lajzar.co.uk
>> I (German) would call the color of carrots "orange", but one
>> traditional German term is /Gelbe/ /Rüben/ (/yellow/
>> /roots/). Thus, it should be possible to call oranges yellow,
>> or red, too.
Javier> That's the case with Hungarian: carrots are "sárgarépa"
Javier> (yellowbeet) and orange is "narancssárga" (orangeyellow).
Javier> In Chinese orange is a kind of yellow too: "chénghuáng".
To me, a Hongkonger growing up in the late 20th century speaking
Cantonese natively, "orange" (cheng3) is not a shade of yellow, and
"cheng3huang2" is a shade between orange and yellow (much like what
"yellowgreen" means in English).
--
Lee Sau Dan 李守敦(Big5) ~{@nJX6X~}(HZ)
E-mail: dan...@informatik.uni-freiburg.de
Home page: http://www.informatik.uni-freiburg.de/~danlee
OK, I stand corrected: In Chinese orange _was_
a kind of yellow.
No surprise that nowadays it's become an independent
colour. But how old is that separation traceable to?
My Spanish-Chinese dictionary (Shanghai International
Studies University, 1991) lists <cheng2huang2> only as
plain "color naranja" (not as "amarillo anaranjado")
and <ju2huang2> as "color naranja, anaranjado", although
<ju2hong2> is listed also as "(color) naranja". But
online Lin Yutang's Chinese-English Dictionary of
Modern Usage lists the first again as simply "orange"
(not as "orange yellow") while the last as specifically
"orange red". Tang Ken-de's online dictionary says
<cheng2> can be used alone or followed by <huang2>
or <se0>, again not specifying there to be a new shade
of meaning when followed by the former.
Lin Yutang's
http://humanum.arts.cuhk.edu.hk/Lexis/Lindict/
Tang Ken-de's
http://www.geocities.com/Tokyo/Pagoda/7951/o/orange.html#A
Cheers,
Javier
Javier> I'm not lying. They claimed to have found there exist 11
Javier> universal colour foci defining 11 basic colour categories
Javier> and a universal evolutionay scheme that dictates the order
Javier> of their lexicalization.
From some "discovery"-type programs I watched here in Germany, I
learnt that the visual system of some organisms has up to 12 primary
colours. i.e. their colour perception is a 12-dimensional vector
space.
Javier> And among those 11 supposedly universal foci are the ones
Javier> for brown and orange but not the ones for azure or
Javier> glaucous. So a language with a basic glaucous but lacking
Javier> a basic brown, like Cantonese(*), is supposed to be
Javier> breaking in two the basic universal category of green by
Javier> adding a non-basic focus for glaucous and _heretically_
Javier> turning it into a basic focus, something that, if ever,
What? Cantonese lacks a basic brown? How so?
The colour word 啡 [fE55], which comes from "coffee" ([ka33 fE 55])
means brown. This is because we normally have coffee blended with
cream, which give a brownish colour.
Before we learnt about coffee, there was an older colour term 棕
[tsUN55] which means brown. In writing, the old word 褐 [hOt3] is
used. I often hear both of these used in older people's speech and in
old Cantonese films.
Javier> should occur only well after stage VII when all the 11
Javier> 'real' basic universal categories have been
Javier> lexicalized. But that's not the case here because the
Javier> 'universal' category of brown isn't yet recognized as
Javier> basic in Cantonese, being it merely a dark kind of yellow.
Definitely not. As a native speaker, I won't call the colour [fE55] a
kind of [wON21] (yellow).
Javier> (*) The basic colours of Cantonese are:
Javier> stage I: baahk (WHITE), haak (BLACK) stage II: huhng (RED,
Javier> PINK) stage III: wohng (YELLOW, BROWN) stage IV: luhk
Javier> (GREEN, - GLAUCOUS) stage V: laahm (BLUE) stage VI: -
Javier> stage VII: fui (GREY), chaang (ORANGE), ji (PURPLE), -
Javier> stage Martian: cheng/ching (GLAUCOUS)
[fE55] is missing here. Ask a HKer what colour leather or wood or
shit typically is, and they'll say [fE55 sIk5]. The older versions
[tsUN55] and literary version [hOt3] are also missing from that list.
Javier> A paper on Cantonese colours can be found here:
Javier> http://www2.soas.ac.uk/Linguistics/papers/Muisoaswp1999.pdf
Interesting. But I have some questions...
1) The definition of "basic colour" excludes "golden". Why? Why then
is "orange" considered basic for English in Figure 1? Or "pink"
(which, AFAI can remember from some past posts in sci.lang, is some
flower)?
2) Figure 1 is also unclear. The hierarchy is not clearly shown. In
that figure, "pink" lies to the left of "orange". Then, according
to the descriptions above the figure, any language having a term
for "orange" as primary colour must also have "pink" as primary
colour? Sorry Cantonese has "orange" but not "pink". (Then, you
can argue that like English, the Cantonese term "orange" is the
same as the fruit. Since this colour name in Cantonese is written
with a character with the 'tree' radical, I'd assume that the
colour is named after the fruit, not the other way around.)
Or is that "pink" an outlyer, not belonging to the hierarchy?
Sorry, the diagram is not draw in a clear and ambiguous manner.
And I suspect that most languages have "grey" as a primary colour
term, but not necessarily "pink".
3) I'm so disappointed that the list of "basic colour terms" at the
begining of section is just some *traditional* enumeration order
that we often use in speech for our colour terms. That list isn't
meant to be complete nor scientific (=linguistically). It lacks
the new (may just 100 years?) term [fE55]. In some older films, I
hear [tsUN55] more often. Both of these mean "brown". It is so
disappointing that the author takes this list and treats it
_prescriptively_ as the set of basic colour words in Cantonese.
BTW, I consider [ts'EN55] ("light-green" in the paper) a shade of
[lUk2] "green". Should it be considered a basic colour term, then?
Indeed, [ts'EN55] is not a well-defined colour. It can also mean
"white" or "pale". e.g. when some looks sick, we say "your face
looks [ts'EN55]".
Descriptively, I'd exclude [ts'EN55] from the list and add [fE55].
4) The translation of [ts'EN55] as "light-green" is inappropriate. It
should be "yellow-green" instead. As mentioned above, the hue of
[ts'EN55] varies a lot depending on context. However, out of
context, I'd assign "yellow-green" to [ts'EN55].
5) For non-basic colour terms, yes, we do use modifiers in addition to
a basic colour term to "generate" ranges of colours terms with
different degrees as mentioned in the paper. The list given is,
however, not complete. There is also [pou35] meaning "deep".
Now, another argument why [ts'EN55] is not a basic colour term. I
won't use these modifiers with [ts'EN55]. [ts'in35 ts'EN55]
("light yellow-green") or [s@m55 ts'EN55] ("dark yellow-green")
both sound nonsense to me. The "basic colour terms" in the list in
the paper where I won't apply these modifiers are: [ts'EN55]
(yellow-green), [ts'a:N35] (orange), [h@k5] (black) and [pa:k2]
(white). The modifiers don't make any sense on these colours.
6) Too bad that I have a strong feel that these researchers still need
some more studies on chromatology. They can't even accurately use
terms such as "hue", "saturation", etc., which I think would be
helpful in such studies. It'd be interesting, too, to position the
so called "focal colours" into the colour spaces to see whether
they're sort of "optimal".
e.g. the Cantonese modifier [f@n35] is described in the paper as
"light in a sense of 'to whitewash'". What a clumsy and imprecise
description! Why not say it means "low saturation" in the HSB
colour model? Then, it becomes obvious that [f@n35 hUN21] means a
shade of red with low saturation, i.e. pink.
7) The table in Fig. 2 also invites questions:
Given the non-productivity of the [m@k2] row, I wonder why the
author would consider it a modifier of the same kind of others.
[m@k2] means "ink", in particular: the kind of Chinese black ink
that we use with brush-writing. So, [m@k2 lUk2] is a blending of
"black" and "green", given "dark green".
Also note the special column for [ts'EN55]. Shouldn't the author
wonder why most modifiers don't like to pair with [ts'EN55]?
8) Concerning the colour-related idioms, I find a big error:
Section 4.2 (3) (ii) Ching- wohng- bat-jip
light-green -yellow-not-connect
Meaning: There is a confusion between two stages.
That's not the meaning that I understand under this idiom ([ts'IN55
wON21 p@t5 tsi:p33], lit. translation correct except for
"light-green"). I think the idiom means: a gap in the passing of
skills in a certain area of speciality from one generation to the
next, often caused by a lack of interest hence a lack of people who
want to learn those skills. As a result of such a gap, many skills
and knowledge are going to be lost as the skilled people age and
die. e.g. Many traditional Chinese arts and skills such as
Cantonese opera and ivory sculpture are suffering from [ts'EN55
wON21 p@t5 tsi:p33].
9) Section 4.2 (3) (iii) Ching - lihn
light-green -year
Meaning: Youth.
The transcript is wrong. The second word/syllable is [nin21], not
[lin21].
10) Section 4.2 (6) (i) huhng-fung-baau
red-cover-packet
This is a Mandarin/literary term retrofitted into Cantonese.
Cantonese speakers won't use this term orally, because we prefer
the shorter term [lei22 si22] (variant: [l@i22 si22]). No child
would spontaneous say [hUN21 fUN55 pao55] when he's not asked to
write (in literary Chinese).
Some elderly people would also sometimes say [hUN21 tsi35] (lit:
red paper). But definitely not the Mandarin term.
11) Section 4.2 (8) (i)baahk-faat-chong-chong
white-hair-hoary Meaning:
Not a mistake. But I'd point out that [ts'ON55] is a colour term
like [ts'EN55], in the sense that it is pretty vague. It can mean
"pale" or "green". [ts'ON55] as a colour term is often used in
[ts'ON55 ba:k2] (pale white) to describe one's face when one's
anxious, frightened or sick. [ts'ON55 tin55] meaning "blue sky"
is another place where this word is often encountered. [ts'ON55
sa:n55] means "green mountain". So, like [ts'EN55], it is a very
vague colour term, covering a very very wide range of colours.
Ah! You know the so called "cold colours" that artists use?
[ts'ON55] covers those "cold colours".
12) Section 4.2 (8) (i)
Reason: People's hair becomes white in colour when they get old.
Well... the author isn't that familiar with English and other
western languages. In those languages, they say the hair colour
of old people is "gray", not "white".
(Similarly, after sun-bathing, your skin becomes "darker". In
Chinese we say the skin gets "blackened" after sun-bathing.)
> in article 48c7f19.04010...@posting.google.com, Larry Trask at
> lar...@sussex.ac.uk wrote on 1/04/04 9:02 AM:
>
>
>>I would like to thank everyone who wrote in with etymological
>>information on those two color terms. The information provided was
>>fascinating, but I guess the conclusion is that no cut-and-dried
>>etymology exists for either term, except that <midori>, if I've
>>understood correctly, does appear to derive *directly* from a term for
>>'vegetation'.
>>
>>Larry Trask
>>lar...@sussex.ac.uk
>
>
> Interesting how the thread morphed into a mudslinging match between two
> eggheads. I started a thread about the spread of tanuki in eastern Europe
> which quickly became a thread about weight loss and Japanese airline workers
> with French accents or something.
It'd be interesting to have a metric for topic drift; or maybe I'm just
bored again.
Dan
>In article <1cghvv45nbj8mnh91...@4ax.com>, Dale Walker says...
>>
>>On 2 Jan 2004 11:20:05 -0800, Charles Eicher <cei...@inav.net> wrote:
>>
>>>In article <a8abvv87qrdf14qik...@4ax.com>, Dale Walker says...
>>>>
>>>
>>>>I'm sure a fine spray of gold pigment could achieve some sort of
>>>>transparent result but that would then tend to prove your theory
>>>>correct, not Peters.
>>>
>>>To get the transparent effect, the gold has to be applied in monoatomic layers,
>>
>>Or sprayed. 50% coverage would result in effectively 50% transparency.
>>Obviously, the more you spray, the less transparent it becomes.
>
>If you wanted that effect, you wouldn't need gold, you could just sputter
>droplets of opaque black
>paint on 50% of the glass, that would cut down transmission of everything.
I was thinking more along the lines that the gold being reflective
would be more effective looking at it from the outside whereas the
'stained' part would be better from the inside. Are all stained glass
windows only designed to be looked at from the inside?
>>Correct me if I'm wrong but doesn't silver incorporated into the glass
>>itself give a gold colour?
>
>I never heard of that, some online research yields only a reference to a modern
>process for silver
>nitrate glass, it's amber. That doesn't sound like a medieval technology.
http://www.stainedglass.info/stainedglass_056.htm suggests it's pre
medieval, and
http://www.ariadne.org/studio/michelli/sgmedieval.html has a few
examples of silver stained glass dating from the early 1300's.
Ah, I was thinking more about space helmet visors, or the more common
equivalent, those ugly
Oakley sunglasses.
>>>Correct me if I'm wrong but doesn't silver incorporated into the glass
>>>itself give a gold colour?
>>
>>I never heard of that, some online research yields only a reference to a modern
>>process for silver
>>nitrate glass, it's amber. That doesn't sound like a medieval technology.
>
>http://www.stainedglass.info/stainedglass_056.htm suggests it's pre
>medieval, and
>http://www.ariadne.org/studio/michelli/sgmedieval.html has a few
>examples of silver stained glass dating from the early 1300's.
Interesting. Looks like the evidence is mounting up that there were color
distinctions between
yellows, oranges, golds, etc. in the medieval period, like I've been saying all
along.
I am not sure if one can draw such a conclusion from the
mere presence of the word huang2 (yellow) in the compound
word cheng2huang2 (orange-yellow). For example, while
"scarlet" in English is usually not written in a compound
"scarlet red", are we to conclude that scarlet is not a
kind of red? Certainly this kind of morphological-semantic
convention varies across languages and diachronical
within a language. Since words can be borrowed from other
languages and different words are formed at different times
in history, without detailed etymological information I am
not sure one can draw any conclusion.
(This reminds me the the often repeated myth circulated
among Chinese that the suffix -ese in English is derogatory.)
I think the only way to ascertain whether a color is
considered "a kind of" another color is to query the
informant(s) directly and systematically. It follows that
for most languages we don't really know what the conceptual
catergories were more than a century or so ago.
Some datapoints. In Guangzhou Cantonese, up to the sixties,
the term for the color orange is "gam55 yu21 wong35"
(goldfish yellow) but AFAIK no one consider the color to
be a kind of yellow. Note the wong21 (yellow) is sandhi'ed
to wong35 in this compound, a sign that the compound is
used frequently. Also, pink in Cantonese is "fan35 hung21"
(chalk/white-wash red) but again no one consider pink a kind
of red.
Tak
--
----------------------------------------------------------------+-----
Tak To ta...@alum.mit.eduxx
--------------------------------------------------------------------^^
[taode takto ~{LU5B~}] NB: trim the xx to get my real email addr
> Bart Mathias wrote:
> > I hope I can be pardoned for not believing that ["midori" is
> etymologically equivalent to "vegetation."]
> You may not be pardoned, thank you very much.
Damn! Excuse me, I mean Darn!
> > The closest thing I know that could be called *the* (pure
> > Japanese) word for vegetation is "kusaki."
> I see a tricky qualification in there ("pure Japanese"?). ...
It seemed to me that one would come up with shokubutsu ahead of
kusaki.
> > I am willing to believe that vegetation can be referred to
> > metaphorically as "midori."
> Or maybe the trap lies in the use of "vegetation."
> The Green Goddess has "green; verdure" for midori. Verdure.
> Vegetation.
Maybe my GG is too old? It has "green; verdant; emeral" for
"midorino," so I think the word originally referred to greenish
rocks.
If the definition had been "vegetation; verdure; green," my disbelief
might have been tempered.
> (Cf. *"The verdure of that sweater is just beautiful.")
Is "sono seetaano midoriha hontouni kireidawayo" just as bad? If
not, then you have wounded the "midori=verdure" formula.
> [Kojien] has
> ...
> ([Perhaps "mido" is the root, and it is related to the "midu" of
> "mizumizushii."]) [1. New buds of plants. Young foliage of early
> summer. Broadly, vegetation in general.]
> [...]
> [2. A color between ao and ki. ...] [... (Is there a typo in the
> Isoho mg. quote?)]
> Note the [Broadly, vegetation in general] in definition [1].
Noted, and previously believed. But dictionaries often use "broadly"
or "hiroi imide" to indicate that the meaning given is probably an
extension of an original narrower meaning.
Iwanami Kogojiten suggests that "new buds" was the original meaning
and the color sense was an extension of that. (Hey, "shinme" must be
another synonym for "kusaki"!)
I'm more inclined to go with Shinmura; it seems obvious that midori
includes a sense of wetness or luster, and unanalyzable
three-syllable words are rather rare in Japanese, so it is very
likely that the "mid-" or the "mi-" part has to do with "water."
Granted, on can say it IS water only when one can come up with an
equally plausible account of the "-ori" or "-dori" that is left over,
and I can't.
> So what do you have to say about that?
See above.
Bart
Lee Sau Dan wrote:
LSD> To me, a Hongkonger growing up in the late 20th century speaking
LSD> Cantonese natively, "orange" (cheng3) is not a shade of yellow, and
LSD> "cheng3huang2" is a shade between orange and yellow (much like what
LSD> "yellowgreen" means in English).
JBF> OK, I stand corrected: In Chinese orange _was_
JBF> a kind of yellow.
I am not sure you can conclude that. See my other post.
Btw, Sau Dan was talking about Cantonese, whereas you were
talking about an unspecified period of "Chinese". The
morphological-semantic conventions are not necessarily the
same.
JBF> No surprise that nowadays it's become an independent
JBF> colour. But how old is that separation traceable to?
JBF>
JBF> My Spanish-Chinese dictionary (Shanghai International
JBF> Studies University, 1991) lists <cheng2huang2> only as
JBF> plain "color naranja" (not as "amarillo anaranjado")
JBF> and <ju2huang2> as "color naranja, anaranjado", although
JBF> <ju2hong2> is listed also as "(color) naranja". But
JBF> online Lin Yutang's Chinese-English Dictionary of
JBF> Modern Usage lists the first again as simply "orange"
JBF> (not as "orange yellow") while the last as specifically
JBF> "orange red". Tang Ken-de's online dictionary says
JBF> <cheng2> can be used alone or followed by <huang2>
JBF> or <se0>, again not specifying there to be a new shade
JBF> of meaning when followed by the former.
Morohashi has entries for <cheng2 huang2>, <cheng2 huang2 se>
as well as <cheng2 se>. No citation for any of them though;
probably because their usage are fairly recent. However,
there is a citation of a 12th century poem that contains the
line <zheng4 shi4 cheng2 huang2 ju2 lyu4 shi4> (right-be-orange
-yellow-tangarine-green-time, i.e., "right when the oranges
are yellow and tangarines are green"). Go figure.
I have also heard and used <dzE35 (sik5)> 赭 (u+8D6D, zhe3 in Mandarin),
but it is definitely not widely used.
I guess those 12 primaries are grouped into 6
channels, thus rendering a 6-dimensional space.
Human vision uses(*) 6 primaries grouped into 3
channels: a scalar luminosity channel (white-black)
and two polar hue channels (red/green, blue/yellow),
thus a 3-dimensional space.
BTW, do you remember which ones are those
colourful-sighted animals with 12 primaries?
If they could speak, I wonder how they would
manage to understand human colour terminology.
Or how we would manage to understand their
colour terminology. I guess that would be as
impossible as understanding the perceptual
world of cetaceans, or as trying to explain
colours to a colour-blind or a blind from birth.
--
(*) As the final data reaching the brain after
the raw optical information is processed by the
retinal neurons combining the information from
the three kinds of cones into two hue channels.
> What? Cantonese lacks a basic brown? How so?
>
> The colour word +VWE- [fE55], which comes from "coffee" ([ka33 fE 55])
> means brown. This is because we normally have coffee blended with
> cream, which give a brownish colour.
>
> Before we learnt about coffee, there was an older colour term +aNU-
> [tsUN55] which means brown. In writing, the old word +iRA- [hOt3] is
> used. I often hear both of these used in older people's speech and in
> old Cantonese films.
[...]
> [fE55] is missing here. Ask a HKer what colour leather or wood or
> shit typically is, and they'll say [fE55 sIk5]. The older versions
> [tsUN55] and literary version [hOt3] are also missing from that list.
Rather than asking an urban HKer, I'd ask a
Cantonese speaker from the countryside of the
Canton province, where Western influence hasn't
been so intense as in what was a British colony
for quite a long time. If you ask a young urban
Basque, they will hardly give you a clue as to
the traditional Basque colour system, since their
conceptual colour space is nowadays shaped just
like the pan-Western 11-colour system, while if
you venture into a remote hamlet in the mountains
and ask an elder who has never spoken anything else
than Basque and hasn't been exposed to Western
models on Spanish or French TV, you'll have more
chances to find what the native conceptual colour
space of Basque actually was.
Besides, the author of the paper did explain why
those terms you mention were rejected as basic.
> Javier> should occur only well after stage VII when all the 11
> Javier> 'real' basic universal categories have been
> Javier> lexicalized. But that's not the case here because the
> Javier> 'universal' category of brown isn't yet recognized as
> Javier> basic in Cantonese, being it merely a dark kind of yellow.
>
> Definitely not. As a native speaker, I won't call the colour [fE55] a
> kind of [wON21] (yellow).
Maybe not nowadays anymore, but traditionally
<huang2>/<wong4> covered a range from faded ivory
to the yellowish browns. This is reflected in terms
like <niu2huang2>, meaning "ox gallstone", which
certainly isn't of a colour that nowadays I would
refer to as "yellow".
> 1) The definition of "basic colour" excludes "golden". Why? Why then
> is "orange" considered basic for English in Figure 1? Or "pink"
> (which, AFAI can remember from some past posts in sci.lang, is some
> flower)?
Ask Berlin & Kay.
I would say that beige is properly neither a kind
of white, nor brown, nor yellow, but the intermediate
colour (like orange is properly neither a kind of
red nor yellow but the intermediate colour). And
I've always used and heard used <beis> as an
everyday word. The same for some others like <dorado>
and <granate>. The more reddish shades of <granate>
I would accept to call a dark kind of red, but not
the more blackish or purplish kinds. And in my Spanish
dialect <morado> -purple- is not basic (the basic one
is <violeta>), which means red-purple shades are not
included. So, dark red-purple shades, too bluish to
fit as <marrón>, aren't for me a proper kind of any
other colour than <granate> (just like vivid or light
red-purple shades, e.g. fuchsia and magenta, are for
me proper kinds only of <rosa>, see below). Does that
suffice to make it a basic Spanish colour (at least
in my dialect)?
Also, would you say turquoise is properly a kind of
green or of blue? If of neither, is it then <turquoise>
a basic colour term (someone pointed that this seems
to be the case for at least some English speakers)?
> 2) Figure 1 is also unclear. The hierarchy is not clearly shown. In
> that figure, "pink" lies to the left of "orange". Then, according
> to the descriptions above the figure, any language having a term
> for "orange" as primary colour must also have "pink" as primary
> colour? Sorry Cantonese has "orange" but not "pink". (Then, you
> can argue that like English, the Cantonese term "orange" is the
> same as the fruit. Since this colour name in Cantonese is written
> with a character with the 'tree' radical, I'd assume that the
> colour is named after the fruit, not the other way around.)
>
> Or is that "pink" an outlyer, not belonging to the hierarchy?
> Sorry, the diagram is not draw in a clear and ambiguous manner.
That's a transcription error or something. Pink
should be on the same column as grey/purple/orange.
And yes, B&K proclaim it one of their 11 universal
basic colour categories, so it's not an 'outlier'
like Russian <goluboj>.
> And I suspect that most languages have "grey" as a primary colour
> term, but not necessarily "pink".
I think most Westerners won't agree with you in
thinking of pink as a less basic colour than grey.
I, at least, definitely wouldn't, because there
are shades, like magenta and fuchsia (mixes of
red with white and blue), which for me do not fit
in any other category than <rosa>. I do not feel
comfortable describing them as <rojo claro> (like
I would do with the non-bluish part of pink), so
if I need to think of a shade that is purely <rosa>
and nothing else than <rosa>, I think of those.
Also, the overtone associations of <rojo> and
<rosa> are quite different: <rosa> I associate
with "feminine", "gay" and sometimes "artificial"
(because magenta and especially those hurtingly
vivid shades of fuchsia look awfully 'unnatural'
to me - even though they are actually the natural
colour of some kind of flower), while <rojo>
I associate with "danger", "blood" and "passion"
and I'd say it's the most 'natural-looking' colour.
For one, Japanese has borrowed <pinku>, but still
refers to grey as <haiiro> 'ash-colour' (I think
it's also referred to as 'mouse-colour') if not
simply as <ao>.
Maybe the fact that pink is still a kind of red
in Cantonese and Mandarin is influencing your
perception of it as a less basic colour.
> 3) I'm so disappointed that the list of "basic colour terms" at the
> begining of section is just some *traditional* enumeration order
> that we often use in speech for our colour terms. That list isn't
> meant to be complete nor scientific (=linguistically). It lacks
> the new (may just 100 years?) term [fE55]. In some older films, I
> hear [tsUN55] more often. Both of these mean "brown". It is so
> disappointing that the author takes this list and treats it
> _prescriptively_ as the set of basic colour words in Cantonese.
You admit those colours are traditional while <fe1>
is a recent loan for 'coffee'. <Jung1> and <hot3>
also have the inconvenience that they are the names
of a tree and a cloth, so they're rather like <gold>.
> BTW, I consider [ts'EN55] ("light-green" in the paper) a shade of
> [lUk2] "green". Should it be considered a basic colour term, then?
Would you say that *anything* you would name
<ceng1> is also a kind of <luk6>?
> Indeed, [ts'EN55] is not a well-defined colour. It can also mean
> "white" or "pale". e.g. when some looks sick, we say "your face
> looks [ts'EN55]".
I don't see it as a more well-defined colour than
brown. If you think of a yellowish brown like ochre,
and then of a blackish brown like umber and of an
orangish brown like sienna, and place them side by
side, how much similar do they really look to fit
as varieties of one same basic colour category?
I'd say <brown> is a very vague colour concept(*)
whose range could be described as "any dark shade
between red and yellow". Which isn't much more
specific than "any light cold shade" that
describes the broad range covered by <ceng1>.
--
(*) Probably the vaguest of all, since I'm unable
to find any single variety that I could settle on
as the only "brownest brown" - right offhand I can
think of at least two "brownest browns" or 'focal'
browns: one more reddish and one more yellowish
(French <marron> and <brun>), and none looks to
me any more purely or paradigmatically brown than
the other.
> Descriptively, I'd exclude [ts'EN55] from the list and add [fE55].
The author shows the term <ceng1> is used in
idioms (that's a usual feature of basic colour
terms), it only and primarily the colour and,
as you pointed, it is traditionally counted.
While <fe1> is a recent loan and is derived
from the word for "coffee", both features
strongly arguing against its basicness.
> 4) The translation of [ts'EN55] as "light-green" is inappropriate. It
> should be "yellow-green" instead. As mentioned above, the hue of
> [ts'EN55] varies a lot depending on context. However, out of
> context, I'd assign "yellow-green" to [ts'EN55].
Since <ceng1> is placed between green and blue
in the rainbow (not between yellow and green),
it seems kind of odd that you first think of it
as yellow-green. Although yellow is the inherently
lightest hue, thus yellowish green and light
green look related (both are "light colour with
a green component"), so it's not surprising
a shift from light blue-green (Greek <glaukos>)
to yellow-green (Greek <chloros>).
> 5) For non-basic colour terms, yes, we do use modifiers in addition to
> a basic colour term to "generate" ranges of colours terms with
> different degrees as mentioned in the paper. The list given is,
> however, not complete. There is also [pou35] meaning "deep".
>
> Now, another argument why [ts'EN55] is not a basic colour term. I
> won't use these modifiers with [ts'EN55]. [ts'in35 ts'EN55]
> ("light yellow-green") or [s@m55 ts'EN55] ("dark yellow-green")
> both sound nonsense to me. The "basic colour terms" in the list in
> the paper where I won't apply these modifiers are: [ts'EN55]
> (yellow-green), [ts'a:N35] (orange), [h@k5] (black) and [pa:k2]
> (white). The modifiers don't make any sense on these colours.
<Light> and <dark> with <white> or <black> do not
make much sense in English either, AFAIK (except
maybe <light white> as opposed to <off-white>).
Besides, your argument doesn't argue against
<ceng1> any more than it argues against <caang2>.
> 6) Too bad that I have a strong feel that these researchers still need
> some more studies on chromatology. They can't even accurately use
> terms such as "hue", "saturation", etc., which I think would be
> helpful in such studies. It'd be interesting, too, to position the
> so called "focal colours" into the colour spaces to see whether
> they're sort of "optimal".
>
> e.g. the Cantonese modifier [f@n35] is described in the paper as
> "light in a sense of 'to whitewash'". What a clumsy and imprecise
> description! Why not say it means "low saturation" in the HSB
> colour model? Then, it becomes obvious that [f@n35 hUN21] means a
> shade of red with low saturation, i.e. pink.
No, in the paper it is described as "whitened",
which I think is quite an appropriate and
descriptive definition, and more accurate
than yours since "low saturation" can refer
also to 'blackened' or 'greyened'.
> Also note the special column for [ts'EN55]. Shouldn't the author
> wonder why most modifiers don't like to pair with [ts'EN55]?
Not many seem to pair with <caang2> either.
> 11) Section 4.2 (8) (i)baahk-faat-chong-chong
> white-hair-hoary Meaning:
>
> Not a mistake. But I'd point out that [ts'ON55] is a colour term
> like [ts'EN55], in the sense that it is pretty vague. It can mean
> "pale" or "green". [ts'ON55] as a colour term is often used in
> [ts'ON55 ba:k2] (pale white) to describe one's face when one's
> anxious, frightened or sick. [ts'ON55 tin55] meaning "blue sky"
> is another place where this word is often encountered. [ts'ON55
> sa:n55] means "green mountain". So, like [ts'EN55], it is a very
> vague colour term, covering a very very wide range of colours.
Being pretty vague is not an argument against
basicness. Just think of how vague the meanings
of basic colours are in languages which only
feature the trio 'white', 'black', 'red'.
> Ah! You know the so called "cold colours" that artists use?
> [ts'ON55] covers those "cold colours".
Also the dark ones like indigo?
> 12) Section 4.2 (8) (i)
> Reason: People's hair becomes white in colour when they get old.
>
> Well... the author isn't that familiar with English and other
> western languages. In those languages, they say the hair colour
> of old people is "gray", not "white".
In Spanish at least, both "pelo blanco" and "pelo
gris" are used (along with "pelo canoso") and they
aren't interchangeable, since the former refers
to that hair that looks all white, while the latter
to greyish or salt&pepper hair. 40-ish people may
feature "pelo gris", while "pelo blanco" makes
one look at least late-50-ish or 60-ish.
Cheers,
Javier