Nik Taylor wrote:

> J Matthew Pearson wrote:
> > Incidentally, how do you know that Arabic treats the copula as transitive?
> > Couldn't it be that accusative is the unmarked case in Arabic?  (Cf. English
> > "It is me" or French "C'est moi".)
> Well, with the French, _moi_ is simply the free form of the pronoun,
> contrasting with the clitic _je_ and _me_, and with English, arguably,
> "me" is the same way.

Arguably, yes.  But my point was more general.  In languages with rich
morphological case-marking systems, one case is normally identifiable as the
'default' case--viz., the case form which is used in situations where the nominal
is not acting as an argument, and hence does not receive case (in the syntactic
sense).  Included here would be situations in which the nominal appears in
isolation, as well as situations where it's used as a predicate.

As far as I know, the 'default' case in this sense is almost always the case which
is least morphologically marked, viz., nominative in languages with
nominative-accusative case systems, and absolutive in languages with
ergative-absolutive systems.

> Classic Arabic has 3 cases, nominative -u,
> accusative -a, genitive -i.  My source doesn't say much about how
> they're used.  Someone with more knowledge on Arabic will have to answer
> that.

If all three cases are morphologically marked, then one would need to look at
their distribution/use to determine which one is the 'default', in my sense.  Not
knowing anything about Arabic, I can't answer this question either.

> Still, if it's possible for the accusative to be the unmarked case,
> could it be possible for the ergative to also be an unmarked case?

No, apparently this is impossible.  From what I know, the ergative is *never* the
unmarked case.  (Assuming this is true, it's interesting speculate on *why* it
should be so...)