Dans un courrier dat=E9 du 01/06/99 23:57:39 , Ed a =E9crit :
> > Nowadays, we know that a noun cannot "stand by itself" and must fit into =
> > predicate-argument pair however implicit it is. And we know that an
> > "adjective" is only the attributive form of a concept while a "noun" is=20
> > substantive form.
> "Who's 'we', white man?"
hugh :-) i'm not quite light-skinned and i often get checked by french=20
policemen in the parisian underground in summer but i agree i'm definitely=20
not a "peau-rouge" (as we say over here).
> If I do know this, I have forgotten it and need to be reminded of it
> a la Plato.
> Could you perhaps explain what this means and why we know it,
> patiently, as if to a child? I sense that this will be something
> (like Noli) difficult for me to grasp.
i really can't. this is beyond my capacity. i learned it in my grammar schoo=
book years ago. ask linguists on the list. i try below to sum it up somehow=20
but i'm afraid it's not exact :
even "wolf" as a lonely noun is part of an implied predicate-argument clause=20
via a presentative for instance : "(here/there is the/a) wolf" (? -=20
"pr=E9sentatif" or "existentiel" in french). this does not mean that "wolf" =
either a predicate or an argument.
beauty is a noun and beautiful is an attributed quality, both pointing a sam=
long is an attributed quality and length is a noun, both pointing a same=20
concept. (just to remind that both may derive from either and none prevails=20
on the other).
in white languages (;-) many attributed qualities are expressed with=20
adjectives and therefore usually substuff : you say "pretty girl" or "she is=20
pretty", you can't say "she pretty". but you may say "il rougit" (he=20
"reddens"). so attribution is usually regarded as subPoS and quality is ofte=
mixed up with subordination and "state", which i regard as the main flaw of=20
eurocentred auxlangs. attribution does not have anything to do with PoS.=20
personal verbs or nomen agentis are also attributive whatever PoS they are.=20
many conlangers do not regard attributed qualities as subPoS and declare tha=
they rather derive adjectives from "quality verbs" such as "to-be-pretty".=20
this is a common pattern of many natlangs. japanese "state verbs" are also=20
called in french "verbal adjectives" as opposed to "nominal adjectives" (i'm=20
sorry to give japanese examples below, it's not for showing off, only to=20
explain what i mean so japfans please do not post 'why "ga" not "wa"' : it's=20
just because i parallel with subclause, ok ?) :
verbal root "taka-" : height
taka-sa/taka-mi =3D height/loftiness
taka-i =3D to be high.
yama ga takai. =3D the mountain is high.
takai yama =3D high mountain
nominal root "kirei" : nice/clean stuff (you would expect that ambivalency=20
from japs, wouldn't you ? :-)
kirei da =3D to be nice/clean ("da" =3D it is < "de aru" =3D to be as)
ie ga kirei da =3D the house is nice/clean
kirei na ie =3D the nice/clean house ("na" < "da")
to me "takai" is really a quality verb. there is a "stative" verb "takaru"=20
with quite a different meaning "to be many" ("ari ga satouni takaru" =3D the=
are many ants in the sugar). many quality "adjectives" have state twins like=20
nagai (to be long) / nagareru (to flow).
maybe due to japanese i have half a dozen specific verbs to say in my conlan=
"to liken (a cicle), "to shape as (a circle)", "to move as (a circle)", "to=20
wend according to the pattern of (a circle)", "to posture according to the=20
pattern of (a circle)". i can speak my conlang better like that but i doubt=20
this is fit for an auxlang.
now i'm sure i'll be flamed by linguists because all that is very simplistic=20
- and thus wrong. but at least i will have done my earnest best.
> Ed Heil ------ [log in to unmask]
> --- http://purl.org/net/edheil ---