Re: Antigenetive case?

En réponse à Christian Thalmann <cinga@...>:

>
> How do you mark grammatical case on this phrase?  If you were to
> say "His cat sees me" or "I see his cat" or "I give the cat some
> food", you'd have to mark nominative, accusative and dative cases
> on the phrase "his cat".  But the head noun (cat) already has a
> case, your antigenitive.
>
> Would you mark the global case on the describing word (in this example
> "he")?  That sounds quite couterintuitive to me.  You could add
> *both* case suffixes to the head noun, which is something I wouldn't
> like either, but it might work for you.  Some real-life languages
> stack cases IIRC.
>
Well, languages with a construct state usually don't have case. There again,
Semitic languages show that it's not always the case, as Classical Arabic has
both construct state and case, which are compatible, as they are marked by
different means (case with the vowel on the last consonant of the noun,
construct state by not putting the article and sometimes having a "shorter"
form). In the case of a construct *case* though, indeed there is a problem. My
language O which has a construct case answers it by stacking both construct
case and the other case on the noun, à la Basque. But you could also just have
the noun in the construct case and not any other. So to know the function of a
completed noun you'd have to rely on word order and context (and the cases of
other words. If you already have a subject in the nominative and no object in
the accusative, there's a big chance that your noun in the construct case would
be the object :)) ). Plenty of languages work with even more ambiguity than
that, so there's no problem using such an idea. Add verbal agreement for the
core functions and common use of adpositions for the oblique ones and ambiguity
is gone! And all are pretty natural features.

> However, if you use the classical genitive construction, you can
> leave the head noun in its grammatical case and just tag the
> describing word with the genitive: "I give cat:DAT he:GEN some food".
>
But it was the very point of this idea to *not* use the genitive case ;))) .

> As for the name of antigenitive, it suggested something else to me
> at first sight...  I would expect cat:AGN man to mean "the man who
> possesses the cat" rather than "the cat possessed by the man".
>
Me too. At first that's what I thought when I saw the title. And it's also
something natural languages often have, although it's mostly marked by using
convolutions or stretching the meaning of some case. Basically it's the other
end of the opposition ownership-belonging. The genitive case marks ownership
(it marks the owner) while the "antigenitive" would mark belonging (it marks
the thing owned). In Latin, the "antigenitive" was marked with the ablative
case: vir magno animo: a man who has a great spirit. But in Latin, like in many
other languages, those constructions often evolve into adjectives, since
this "antigenitive" is mostly used for qualification. So "vir magno animo"
became "vir magnanimus". Hence French "magnanime" and English "magnanimous".

It's also an interesting case to have in my opinion ;)) .

Christophe.

http://rainbow.conlang.free.fr

Take your life as a movie: do not let anybody else play the leading role.

Reply