Re: Re : Re: Artyom Kouzminykh: Answers & proposal

Dans un courrier dat=E9 du 23/08/99 12:02:42  , Christophe a =E9crit :

> Lars Henrik Mathiesen wrote:
>  >
>  > > Date:         Mon, 23 Aug 1999 11:09:34 +0200
>  > > From: Christophe Grandsire <grandsir@...>
>  >
>  > >         I wonder if it is true. This makes perfect sense for
>  > > English, but it is something impossible for French for example:
>  > > "J'ai cuit de la viande" vs. "J'ai de la viande cuite". But maybe at
>  > > an earlier stage French had the adjective before the noun (and
>  > > before its article).
>  >
>  > Well, I cheated. The 'have'-construction did not arise in any of the
>  > modern European languages, but sometime back in Proto-Romance and
>  > Common Germanic times, or perhaps a bit later. (Neither family had
>  > strict word order or mandatory articles back then). It just happens to
>  > work in Modern English too.
>  >
>  > And it did used to have the participle agreeing with the object noun
>  > phrase in gender and number, unlike the periphrastic perfect with 'is'
>  > where it agreed with the subject. (Agreement getting lost as part of
>  > grammaticalization is expected --- but what may be interesting is that
>  > the 'have'-form generally lost it much earlier than the 'is'-form).
>  >
> =20
>          We still have it in French, but only when the object takes place=20
> before
>  the verb, when it's a personal object pronoun for example, or when it's
>  a relative pronoun. I think in older times agreement was more important
>  (I can't check it by now, but I will).

this french attributive form has disappeared :

je suis mang=E9e =3D *je suis (une chose) mang=E9e
je l'ai mang=E9e =3D *j'ai (une chose) mang=E9e (qui est elle)
but we can't say anymore :
*j'ai viande mang=E9e =3D *j'ai (une) viande mang=E9e
but the "archaic" form is still possible :
"j'ai viande mang=E9 et r=F4t d=E9gust=E9 puis endormi me suis" ;-)

the problem is that the attributive function of "be" and "have"
(viande mang=E9e) are mixed up with other functions like aspect
(=EAtre endormi) and even tense (s'=EAtre endormi) !
or in other words, attributive "be" and "have" have turned into aspect,
which in turn has turned into tense and now who can really say
what is what ?

english still has pairs like "sunk/sunken", "proved/proven" etc.
that show the distinction.

> =20
>  > Lars Mathiesen (U of Copenhagen CS Dep) <thorinn@...> (Humour NOT=20
> marked)
> =20
>  --
>          Christophe Grandsire
> =20
>          Philips Research Laboratories --  Building WB 145
>          Prof. Holstlaan 4
>          5656 AA Eindhoven
>          The Netherlands
> =20
>          Phone:  +31-40-27-45006
>          E-mail: grandsir@natlab.research.philips.com
mathias