Re: [tied] Nominative: A hybrid view

--- In cybalist@yahoogroups.com, Jens Elmegård Rasmussen <jer@...>
wrote:

>
> Are you being attacked by yourself? Again I believe I can figure

out

> a way to agree with both of you.
>
> The Eskimo way ("ergative") uses a possessive syntax implemented on
> a passive participle: "The man has seen the seal" implies that the
> man *has* something, to wit, a seen seal. That is precisely the way
> it is expressed, Greenlandic anguti-p puiSi taku-va-a. The third
> word is properly a pass.ptc. *taku-paR- 'seen', inflected with the
> 3sg possessive ending *-a, so that it means "his seen (thing), what
> he has seen". The sentence is construed, then, as "of the man the
> seal is his seen (thing)". This is exactly comparable to our
> periphrases with 'have': what you have, is yours, and the one to
> whom something belongs is in the genitive; therefore, the ergative
> makes perfectly fine structural sense as an old genitive.
>
> I know far too little about comparative Uralic, and my personal
> collection of handbooks belongs in a museum, so I'll have to

reserve

> my final verdict until I can get better informed. Accepting your
> information I see two possibilities of interpretation: 1. The fact
> that the extra plural marking only takes place in some language if
> the *object* is a plural word, may indicate that the verbal stem
> here used is in reality a passive participle just as in Eskimo. -

2.

> The same fact may also have come about secondarily in that speakers
> only felt the need for the extra plural marker if there was a
> special reference for it. It all began with a simple-minded line of
> reasoning on my part: In the days when the forms were posited as
> 1.sg. *luke-m 'I am reading' and 1.pl. *luke-k-me-k 'we are
> reading', I found it strange that the interior *-k- was identified
> with the present marker. Doesn't the corresponding singular forms
> need it too? And why does it just happen to be identical with the
> plural marker seen in the plural forms? If it means "I am a

reader",

> then 'we two read' should be expressed as "we-two are two-readers"
> with number marking on subject and participle alike. And that

looked

> so nicely like what I could see that I went on and subjected the IE
> forms to a comparable analysis. I know I am way out in the shady
> prehistory where we can't trust a single observation, so I don't
> really insist upon it. I would, however, like the easiest solution
> to make some sense, and perhaps than can be delivered.
>
> The two structures, the "possessive" one with the passive
> participle, and the "essive" one with the active participle, need
> not be mutually exclusive. We have active and passive expressions
> which may even be stylistic variants used to express what is in
> essence the same message. So perhaps both existed side by side.

That

> would immediately explain that some languages have one structure

and

> others the other (provided that is so), for both were there, and
> they could just generalize one or the other. It could also
> potentially offer an explanation for events of confusion between

the

> two systems. Once the original intention of the verbal stem as the
> expression of an active participle has been lost sight of, other
> sets of inflectional endings with different rules of concord could
> exert their influence. The result could be a total clash of the old
> double system leaving only a single set of inflectional endings
> which, seen in relation to the original intentions they were
> designed to convey, may even in part have inconsistent forms.
>
> We would still have *-m as the old genitive marking the one
> that "has" found himself an object of a transitive action. And *-m
> would also mark the old objective genitive expressing that "of
> which" the actor is mentioned in the nominative. I see no problem

in

> assuming the simultaneous existence of both structures, especially
> if the language is to have a chance to switch from one to the other.
>
> I repeat: I will have a closer look at it all when I find the time
> for it.
>
> Jens

I think another (if not a simpler) possibility, given the typological
evidence, is that originally there was a postposition *ma which had
some kind of locative meaning. It subsequently became enclitic and
grammaticalized in many languages. Then it was re-analyzed as an
accusative (in PIE and PU) or as an ergative/genitive (in PK and
PEA). The fact that Eskimo-Aleut has -m as both an ergative and as a
genitive is very interesting, and gives credence to my theory that
ergatives often form from genitives. Given this, I posit that the
Eskimo-Aleut *-m(a) was a genitive first, which arose from the
following development: "the dog at me" = "my dog".

Also, the Old Georgian ergative ("narrative") suffix was -man, so the
*-ma element was only one formant of the future ergative.

- Rob