Re : Re: Re : Re: Sawilan Constructions

Dans un courrier dat=E9 du 17/08/99 21:40:03  , Ed a =E9crit :


>  I will.  Any particular aspects of Danoven I should look at?
the role mapping of nouns taken as predicates.

>  =20
>  > (ii) many-stuff is wrong. it's still pairing concepts, and pairing
>  > different concepts many times in a clause doesn't change that fact.
> =20
>  I'm not sure I understand this criticism at all.
"yacht" refers to richness and boat and crew, etc.
but only one of these references is used at one time
with another argument.

>  =20
>  > (iii) in "jail bait", "jail" is no "result". jail is a social
>  > facility, as  bait is an individual one and both refer to specific,
>  > basic relations and/or processes.
> =20
>  If we take "jail" to be shorthand for "the condition of ending up in
>  jail as a result of conviction for statutory rape," then it is indeed
>  a result.
but it's not a condition. it's a place of social retaliation.
using this instrument refers to "social retaliation", not
"condition". this is very important.

act
social
rule
break
retaliation
allow not
move out


But then, I'm not sure I understand what you mean by a
>  "facility," whether "social" or "individual," so I'm not sure I fully
>  understand this criticism.
> =20
cognitive fields pertain to different "experiences" that you may
classified as "social" or "individual", then "intellectual",
"physical", "emotional", etc.

>  > funny how these guys use fundamentals and despise them at
>  > the same time.
> =20
>  I'm not sure I understand this criticism either.
> =20
they pinpoint a few examples and don't even try
to list them to see how many them come up with.

>  > cognitive fields of activities imply a limited
>  > number of processes implying a limited number of roles as
>  > basic as the one consuming, the one carrying, the one covering, etc.
> =20
>  Unless I misunderstand what you mean when you say "cognitive fields
>  of activities," which is very possible, I don't see this implication
>  as correct, or at least relevant.
> =20
>  > of course, these roles refer to "vehicle", "container", "trap",
>  > "stem", etc. and that sounds less smart than "multi-referential
>  > cognitive" stuff.
> =20
>  Ah.  Are you trying to say that Fauconnier and Turner are just using
>  unnecessarily big words?
>=20
=20
they are using unnecessary arrows and circles.

>  > take that example (i guess Charles doesn't mind) :
>  >=20
>  > the moon eats the sun
>  >=20
>  > "eat =3D consume" in "food" field.
>  > if "consume =3D hide" in "astronomy" field, then you understand :
>  >
>  > the moon hides the sun
> =20
>  Again, I am at a bit of a loss to understand the import of this
>  example, divorced from context.
> =20
cognitive fields are lists of contexts.

>  > it's very easy to encode and decode such metaphora
>  > because cognition fields are limited in number
>  > and basic processes like "consume" or "hide" are likely limited.
> =20
>  "Limited" is a fairly vague word.  By "a limited number" do you mean
>  ten, twenty, a hundred, five hundred, ten thousand?
>  =20
is it for me to say that ?
it depends whether you want to make sense or
to make a translation.
it depends whether you want to encode a precise
cognitive field such as "motorized grass mowing"
or a vague one such as "cutting".


>  > the correlations between basic roles such as "consumptible" and
>  > "cache" are cultural. for instance, even if you're not japanese,
>  > you can understand that "cow-ear" refers to "leading someone"
>  > once you have noticed that cows obey when you grasp their ear.
>  > i believe this experience pertains to a cognition field listing
>  > the many roles re-acting to the execution of a will. and mind you,
>  > there are very few such roles.
> =20
>  Ah, I think I am starting to understand.  You are, I think, saying
>  that Turner and Fauconnier's analysis is too open-ended, and ignores
>  the idea of a fairly small core of abstract action-patterns, with
>  their activities and roles, which undergird all verbs?  Is that
>  correct, or close to it?
>  =20
yes.

>  > the only condition to map this out is to look and dream and stop
>  > intellectualize what billions of people understand instinctively.
> =20
>  I think trying to attain a clear description of what it is that
>  people do instinctively is precisely what they want to do.  Asking
>  them not to describe it is a little beside the point -- like asking a
>  painter not to paint the flowers, but instead to stop and smell them.=20
>  A useful thing, but a painter wants to paint, no matter how nice they
>  smell.
> =20
they don't describe anything.
they just pinpoint obvious possible semantic roles
of vocabulary taken as predicate and pertaining
to different cognitive fields.
any all-noun-for-fun conlanger can do that.

mathias