Re: But where does *-mi come from?

--- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@...>
wrote:

> > > PIE nominative is part endingless, part has -s, most likely

from

> > the
> > > deictic *so. The PIE nominative might therefore have been a

pre-

> > PIE
> > > absolutive.
> >
> > That is non sequitur.
>
> Erh, how so?

Nothing is clarified if the nom.sg. marker *-s is identified with
the initial consonant of one of the allomorphs of the stem variation
*so-/*to-. And even if so: *so is nominative, *-s is also
nominative; how can that be construed as an indication that the IE
nominative was earlier something else? We need a positive argument,
not just a smokescreen.

> > If *-m marks the object in Uralic and Indo-European, what is it
> most
> > likely to have marked the day they split from each other?
> Something
> > completely different?
>
> If the language was ergative at that time, the question makes no
> sense. Objects, along with subjects of intransive sentences, would
> of course be 'marked' with nothing, since the absolutive was
> endingless. And both languages would have an allative (or

illative)

> in *-m.

The observation that *-m marks the object in a non-ergative syntax
in IE and Uralic alike is not a very good basis for supposing the
languages were any different when they separated.

> > I do not see a PIE ergative system lying around in ruins. I do

not

> > see an ergative system in the deep structure of the other

families

> > either. Where I do see an ergative system, as in Eskimo-Aleut,

it

> is
> > plainly secondary.
> >
> OK.

"OK"? How nice. My point is really that I can derive the ergative
use of the *-m case in Esk.-Al. from a genitive; and I can also
derive its use as an object marker in IE/Uralic from an older
genitive. I have myself been soft on ergative, but the more I have
tried to find support for it, the more it has been losing ground.

Jens