Re: inverse constructions

Dirk Elzinga skrev:

> Hmmmm. I've always associated inverse marking with languages
> with a hierarchical argument structure (Georgian notwith-
> standing), and you're proposing just such a hierarchy (1 > 2 > 3
> > 3'). Languages to look at for ideas are Algonkian, Yuman, and
> Navajo.=20
Hi, hi... That's exactly the langs I've been taking ideas from.
(That is, Cree, Kutenai (Br. Columbia, Montana, Idaho),=20
Wayampi (Brazil), Panare (Venezuela) and of course Yuman
(Papua New Guinea)).

> Again, it seems to me that inverse is incompatible with case
> marking of any stripe. (I wish I knew more about Georgian!)
> Maybe you could make it work, though!
Yes, I realize that it's incompatible. Of course I could make
it work anyway ;) but that would lead to a very strange lang.
Cebuano has an ergative system and uses constituent order
that can be thought of as a kind of inverse voice. But that
might be stretching it a little.
Also Nocte (Naga) has a mixed erg/acc system, but that works
like this:

nga-ma ate hetho-ang   'I will teach him'
1-ERG  3   teach-1sg

ate-ma nga-nang hetho-h-ang    'He will teach me'
3-ERG  1-ACC    teach-INV-1sg

The last example still has verbal agreement with 1sg,
but inserts a 'h' to indicate that 1sg is patient
instead of agent.

I could of course decide that my cases don't mark semantic
roles when the inverse is used, but I think I'd better
make up a new lang instead. :)

In fact I've been asked to make up like five-six conlangs
for a conworld that some guys here in Sweden has made up.
They found my website and asked me about it. I guess they
will be somewhat sketchy... :)

To see my Rinya case system, go to=20
http://hem1.passagen.se/noldo/roles.html

/ Daniel Andreasson