Re: R: Re: erg/abs; verbs.

At 07:13 PM 3/16/2000 +0100, Mangiat wrote:
>Tim wrote:
>> Not too long ago, I read Winfred P. Lehmann's recent book on Indo-European
>> (I can't find it right now, so I can't give you a citation), and he argued
>> that at some stage in its history, Proto-IE was an active language and
>> _did_ mark case on nouns (-s for agent and -m for patient).
>
>*Every* Indoeuropean language uses those endings:
>Latin has NOM lupus ACC lupum
>Greek has NOM kalòs ACC kalòn < *kalòm
>Sanskrit the same and so on.
>
>Luca Mangiat
Yes, but his point was that they were used differently then.  During the
active stage of the protolanguage, the -m ending would have been used not
only for direct objects, but also for subjects of intransitive stative
verbs, and the -s ending would have been restricted to subjects of
transitive verbs and of dynamic (non-stative) intransitives.  It was only
with the transition from an active language to a nominative-accusative one
that they took on their current functions.

For what it's worth, I had rather mixed reactions to the book, although it
certainly was interesting.  I found Lehmann much more convincing in casting
doubt on other people's theories than in promoting his own.  Thus the
overall impression that the book left me with was of the fragility and
tentativeness of _all_ attempts at reconstructing protolanguages.  And if
they're fragile and tentative for Indo-European, which has written records
going back thousands of years to serve as a reality check, how much _more_
fragile and tentative must they be for most other language families!

- Tim