Re: Survey(?) of ConLangs' Calendars and Colors and Kinterms

--- In conlang@yahoogroups.com, caotope <johnvertical@H...> wrote:
> [snip]
> I am contemplating a system of the following kind:
> *3 basic "hueless" colors (^= black, gray, white)
> *7 basic "primary hued" colors (^= red, orange, yellow, green,
> turquose, blue, purple)
> *7 basic "intermediate hued" colors (^= tangerine, gold, lime, **,
> azure, indigo, magenta)
> *7 basic "pastel" colors (^=maroon, brown, beige, **, teal, **,
lilac)

That looks good.

Looking at it, I think it is the same "sort" of system that mine is,
with some variations:
1) My hueless subsystem -- my gray scale -- has the 1/4 and 3/4
points as well as the endpoints and the 1/2 - way point.
2) My fully-saturated hues are 4 "primary" -- red, yellow, green,
blue -- and 4 "intermediate" -- orange, "chartreuse" (?), cyan
(aqua?), violet/magenta/purple (?).
3) I went ahead and created 3, rather than just one, "pastel"-type
modification of each fully-saturated hue; one half-hue-plus-half-
white, one half-hue-plus-half-black, and one half-hue-plus-half-gray.
4) My "pastels" are based on all 8 of my saturated hues, not just the
"primary" ones.

So, your system has considerably more detail than mine in the fully-
saturated hues -- 14 instead of 8; considerably less in
the "pastels" -- 7 instead of 24; and slightly less in the grays -- 3
instead of 5.

-----

In natlangs, there are generally at most 12 monomorphemic color terms.
There are particular color-terms in my conlang that are never
monomorphemic in natlangs.

AFAIK no natlang has more than three gray-scale monomorphemic terms --
 in essence, black or white or gray -- but that is not the most
common omission.

Also, the colors I've called "chartreuse"(=Green+Yellow)
and "cyan=aqua=cerulean"(=Green+Blue) don't appear to be
monomorphemic color terms in any natlang -- but that, also, doesn't
seem to be the most common "omission".

The most common omissions seem to be that, for each fully-saturated
hue, at most one "pastel" is monomorphemic cross-linguistically.

Red+White=Pink is monomorphemic in many natlangs,
but Red+Black and Red+Grey never are, AFAIK.

Yellow+Black=Brown is monomorphemic in many natlangs,
but AFAIK Yellow+White never is; and Yellow+Gray(="beige"?) is
either "polymorphemic", or labeled by a term from another semantic
field, or covered as a non-core or non-central meaning of "Brown".

Orange+Black is also covered by the extended meaning of "Brown"; and
Orange+White ("creamsicle"?) is not a monomorphemic colorterm in any
natlang I've heard of.

(BTW English's "orange", and many other Western European natlangs'
equivalent terms, started off etymologically as a label from a
different semantic field; but in some natlangs it is represented by
its own monomorphemic color term.)

Green, "Chartreuse", and "Cyan" (or "Aqua") don't seem to have off-
white or off-black or off-grey versions that are monomorphemic in any
natlang I've heard of.

Blue+White, "galuboi" in Russian IIRC, is monomorphemic in many
natlangs. ("cielo" in Spanish is not considered a monomorphemic color
term; it is, instead, a label borrowed from a different semantic
field, since it also means "sky".) But Blue+Black and Blue+Grey are
not monomorphemic color terms in any natlang I've heard of.

I do not know whether "Purple"+White="Lavender" is a monomorphemic
color term in any natlang; I suspect it might be, but I seriously
doubt "Purple"+Black and/or "Purple"+Gray(="Mauve"?) are.

-----

Look at
"Date: 23 Sep 91 14:07:00 EDT
From: "61510::GILSON" <gilson%61510.decnet@...>
Subject: color terminology
To: "conlang" <conlang@...>"

and the thread it started to see an archived conlang discussion of
color (or colour) terms.

The idea of using Munsell chips or something like them as the
semantic centers or core meanings of conlangs' color terms is
included in
"To: conlang@buphy.bu.edu
Subject: defining colour terms in conlangs
Date: Mon, 23 Sep 91 20:01:05 +0100
From: And Rosta <ucleaar@...>

Bruce Gilson asks how conlangs are to define colour terms. My proposed
solution is to state which Munsell Colour [Color] Chips correspond to
the senses of the colour terms. Munsell chips are an international
standard."

A later post says:
"One problem is that there are several different systems for giving
color standards: CIE, Munsell, Hickethier, etc. Each has its
advantages and disadvantages."

The archived thread also discusses the advantages of concentrating on
the semantic center or semantic core of each colorterm, rather than
on the boundaries between them; and, also has some argument about why
even that may not be sufficiently unambiguous.

> altho some of these may end up non-monomorphemic. I'm particularily
> sceptic of "lime", "gold", "indigo", "pastel-green" and "pastel-
blue"
>
> The difference between all the hued colors is more varied than may
> seem, however. For example, the difference between "yellow"
and "gold"
> (or "beige" and "brown") is that of lightness/darkness at least as
> much as it is that of hue. Some other similar pairs include
> azure&blue; lilac&indigo; and greenish-turquose&teal.
>
> ("^=" is used for the "corresponds" symbol. ** do not have good
> English translations available, at least ones I'd know of.)
Thanks, John.

> No decided kinterm system yet.
There are, IIRC, six major types of systems.  Let's see if I can
recall them.

"Diagonal" systems (2 types) can be either patriline-oriented or
matriline-oriented.

In the matriline-oriented diagonal system, whatever kinterm is used
for the relationship of a man to any given female relative, is also
used for the relationship to that relative's mother, and to any of
that relative's daughters. In the patriline-oriented diagonal system,
whatever kinterm is used for the relationship of a woman to any given
male relative, is also used for the relationship to that relative's
father, and to any of that relative's sons.

In a matriline-oriented diagonal system, (such as the Navajo),
for example, a man uses one kinship term to refer to his mother, his
mother's mother, his mother's mother's mother, etc. ..., and all of
their daughters, daughters' daughters, daughters' daughters'
daughters, etc.
So, a mother and a sister and one grandmother and some aunts and some
nieces and so on, are all covered by one term.
A different kinterm applies to his wife, his wife's mother, his
wife's mother's mother, ... and all of their daughters (including his
own daughters), their daughters' daughters (including some of his
sisters-in-law), etc.
There may also be a kinterm to apply to his father's mother and
father's sisters and all of their mothers and all of their daughters;
and other kinterms for other relatives, if any, follow a similar
pattern.

In a patriline-oriented diagonal system (such as, IIRC, the Crow),
OTOH, a woman uses one kinship term to refer to her father, her
father's father, her father's father's father, etc. ..., and all of
their sons, sons' sons, sons' sons' sons, etc.
So, a father and a brother and one grandfather and some uncles and
some nephews and so on, are all covered by one term.
A different kinterm applies to her husband, her husband's father, her
husband's father's father, ... and all of their sons (including her
own sons), their sons' sons (including some of her brothers-in-law),
etc.
There may also be a kinterm to apply to her mother's father and
mother's brothers and all of their fathers and all of their sons;
and other kinterms for other relatives, if any, follow a similar
pattern.

"Classificatory" kinship systems are based on calling any male
relative's brother by the same kinterm as that male relative, and any
female relative's sister by the same kinterm as that female
relative.  In particular, a father's brother is a "father" and a
mother's sister is a "mother".
I am not sure whether or not this is counted as a separate type
within "classificatory"; but, in the Hawaiian system, a father's
sister is also a "mother" and a mother's brother is also a "father".
In these systems, the only meaning of a kinship term is, how many
generations older or younger than Ego the kin-referent is.

The Sudanese system (and other systems of the same type) fully
distinguishes between different kinds of relationships that are not
distinguished by any other system.
In particular, a father's father and a mother's father are not called
by the same term (as English "grandfather"); nor are a father's
mother and a mother's mother (as English "grandmother").
Father's brother, mother's brother, father's sister's husband, and
mother's sister's husband, are four separate kinterms in systems of
this type (English calls them all "uncle"); likewise father's sister,
mother's sister, father's brother's wife, and mother's brother's
wife, are four separate terms, unlike English "aunt".
Continuing on, father's brother's daughter, father's sister's
daughter, mother's brother's daughter, and mother's sister's
daughter, are four separate terms, unlike French "cousine"; and
sister's husband, wife's brother, and spouse's sister's husband, are
three separate terms, unlike English "brother-in-law" (I don't think
English calls a brother's wife's brother a "brother-in-law").

I believe, if I recall correctly, that the Classical Latin system
exemplifies the fifth type, and the Modern English system exemplifies
the sixth type.

-----

When I was trying to learn the Chinese (specifically the Mandarin)
kinship system from an informant; often when I'd ask "what do you
call such-and-such a relative", she would reply "well, which one is
older?"
In Mandarin, the kinterm for younger brother and the kinterm for
older brother are distinct from each other; as are the kinterms for
older sister and for younger sister.
In other ways it is a lot like the Sudanese system; there are two
terms for grandfather and two for grandmother, since father's father
is not the same as mother's father, etc.
So our English word "uncle" has eight distinct translations in
Mandarin:
Father's Older Brother
Mother's Older Brother
Father's Younger Brother
Mother's Younger Brother
Father's Older Sister's Husband
Mother's Older Sister's Husband
Father's Younger Sister's Husband
Mother's Younger Sister's Husband

On the other hand, like the Classificatory Kinship Systems, Mandarin
does not distinguish between Mother and Step-Mother:
nor between Father and Step-Father:
nor between any two of; full-brother, agnate half-brother, uterine
half-brother, and step-brother.

> Calendar will likely be the Roman
> calendar (since this is a personal rather than concultural artlang)
Since I have yet to complete even one conlang, but have at least two
different uses for a conlang, I wonder if I should go for two or
three conlangs instead of just one.

One would be for a Terrestrial-originated human society with a long
history or prehistory on Earth;
the other would be for an interstellar humans-and-many-other-species-
as-well-as-Artifical-Intelligences society with a long history of
multi-sapient-species living-together.

The "lunisolar" calendar I talked about is very obviously based on
Earth's day, Earth's year, and Earth's Moon's month.

-----

Thanks, John.

Tom H.C. in MI