On Tue, 13 Jan 2004 12:40:22  0000 Peter Bleackley
<[log in to unmask]> wrote.
>There exists one case which expresses the subject of an intransitive
>sentence, the object of an transitive sentence, and the indirect object of
>a ditransitive sentence. Another case expresses the subject of a transitive
>sentence or the object of a ditransitive sentence. A third case expresses
>the subject of a ditransitive sentence.


S(case 1) V
A(case 2) V P(case 1)
A(case 3) V DO(case 2) IO(case 1)

I can't see it surviving very long in the wild. My guess is that it isn't
found in nature, but I'd be very happy to be proven wrong.

On further inspection, it *could* be the form manifested if the phrase was
natively understood as a kind of causative construction, with inherently
passive verbs like "is given to" instead of "gives (to)".

A(case 3) ( DO(case 2) V IO(case 1) )

Even then, I suspect IO would take case 2, and DO would take case 1.

OTOH, I can see the following surviving...

S(case 1) V
A(case 2) V P(case 1)
A(case 2) V DO(case 3) IO(case 1)

This is a combination dechticaetiative and ergative pattern, which I think
relies on animacy for default case marking. I suspect there'd be some kind
of "antiditransitive" voice for monotransitive sentences with low-animacy