NetNews Usenet Archive 1993 #1/NN_1993_1.iso/spool/sci/philosop/tech/4771

Text File | 1993-01-12 | 6.0 KB | 118 lines

Newsgroups: sci.philosophy.tech
Path: sparky!uunet!psinntp!panix!mls
From: [e-mail] (Michael Siemon)
Subject: Re: Colors, Schemes, Theories
Message-ID: <[e-mail]>
Summary: Whorf redivivus
Organization: Panix Public Access Internet & Unix, NYC
References: <[e-mail]> <[e-mail]> <[e-mail]>
Date: Wed, 13 Jan 1993 00:57:38 GMT
Lines: 106

In article <[e-mail]> [e-mail] (Jamie) writes:
>>I mean exactly what is contained in Berlin and Kay (1969?/1991?).
>
>Can you give a more complete citation? I would like to look at the
>article.

It's a book _Basic Color Terms_, unfortunately I don't have a full citation
on hand.  But it is a classic in the linguistic discussion of "language
universals" and discussions of whether anything sensible and nonvacuous
can be said about the influence of language on thought.

There may have been a recent revisitation of the material, but a date of
1969 seems to me about right for the original.  Its relevance to this
context is a bit tangential and complex.  The limit of Mr. Green's usage
of it is, however, reasonable.

To be a bit circumstantial: most languages have a lot of ways to express
"colors" (whatever THOSE are -- the assumption that colors map in any
easy way, in *any* language, onto a physicists' spectrum is a bit awkward
-- among other things, human color vision has colors that DO NOT appear
in the spectrum.  See the CIEE color standards; Schroedinger did some basic
work in this field, BTW.) but many such expressions are used as qualifiers
or inessential to the general patterns of word usages ("mauve" for example
is NOT a basic color term in English :-)).  Typically, secondary color
terms derive from co-opted descriptions of something *else* as in the
English term "orange" (which has apparently over the centuries come to be
more basic, just to add the diachronic dimension to further confuse things).
Basic color terms are the *mininal* set that a speaker will use to cate-
gorize responses to an experimenter eliciting names for color swatches.
Some languages have as few as two or three such terms, English has at
least half a dozen.  Colors as "distinct" to us as gray and green would
have been names _glaukos_ in classical Greek.  Languages (and people who
speak them) divide up "color space" (which is at least arguably the "same"
for non-colorblind humans) differently.

In English, we differentiate SOME visual appearance ("metallic" for example)
from colors -- i.e. we recognize different colors of metals, and we also
use the same color names, sort of, for metals as for nonmetals.

Color naming, by ordinary people who don't *think* they are doing anything
but "simple observation" is -- when looked at carefully -- very structured,
very complex, and has some interesting biological and cultural structure
that is not at all evident on the surface (and only partly related to the
physics of the situation.)  This may not be "theory laden" but it is close
enough to such a condition for *this* observer not to make much of the
distinction.

The Berlin/Kay work is also critical in *rejecting* the "strong" Whorf
hypothesis, which would claim something like "ancient Greeks couldn't
observe orange."  In fact, people can usually learn to make distinctions
of color that any other people make, whether or not they have names to
use for the distinctions.

The reaction of most scientists to the claim that facts/observations are
theory-laden is an over-reaction (I think) to a misunderstood claim --
or if you find people who *make* the claim, it is a reasonable reaction,
but you shouldn't assume this without evidence -- to say that our facts
are drenched in theory does not make them any the less "facts."   But it
does say that "fact" has a different semantic content than we (common-
sensically) like to pretend that it has.  The practice of science is
sufficiently contextualized by IMITATION of older scientists that a
"shorthand" from some standardized observation to a (possibly very
sophisticated and highly developed) theoretical construct like "mass"
is perfectly fine.  Yes (Mr. Turpin :-)), such theory does emerge over
time from ambiguous popular usages -- but exactly ZERO of the original
senses are really maintained in even the most casual scientific usage
of once "ordinary" terms (try "force" for an example!)   The ideal of
scientific publication is such as to imagine that an "isolated" observer
(say, millenia from now) could infer the observational context and hence
see the same "facts."  This is part of the ideology of modern empirical
science.  It is even reasonable, up to a point.  But we can (sometimes!)
figure out the "facts" of medieval alchemy, too -- by essentially dis-
counting what they say they observed whenever it disagrees with our
theories.  Only the theoretical structures SHARED by scientists enable
them to HAVE a shorthand and to understand each other as using it.

And in such cases as have inchoate or undeveloped or controversial
theoretical understanding, the status of what is and what is NOT fact
is not uncontroversial until theory can *exclude* a lot of "mistaken
observation."

>
>>That we all have the same visual apparatus, and therefore make the
>>same similarity judgments is not at all to the point.  I mean simply
>>that different cultures divide up the spectrum (a theoretical claim,
>>by the way) in different ways. E.g. they call two shades species of the
>>same color where we call them different colors. No controversy here.
>
>I see. Thank you for explaining this.
>
>Here is how I would gloss the phenomenon:
>by "color" those people mean something different from what we mean.
>
>Is that how you would gloss the phenomenon, or were you thinking of
>some other point?
>
>And, what does it have to do with scientific realism?
>
>Jamie


--
Michael L. Siemon    "We honour founders of these starving cities
[e-mail]         Whose honour is the image of our sorrow ...
             They built by rivers and at night the water
             Running past the windows comforted their sorrow."