Re: [tied] Ergative construction in Middle Indo-Aryan and Iranian

The Sanskrit construction is just a kind of periphrastic passive, although of course it underlies the modern IA ergative construction historically. Ergative patterns (of disputed origin) occur in Kurdish as well (but not generally in NW Iranian or in Iranian at large). Ergativity in Pashto (and Balochi) is probably an areal thing, given their closeness to the Indian Sprachbund. Maybe the following online articles (and the references therein) will interest you:

http://www.cogsci.ed.ac.uk/~siamakr/Kurdish/KURDICA/2001/3/ia-erg.html

http://www.cogsci.ed.ac.uk/~siamakr/Kurdish/KURDICA/2001/3/ergativity.html

There's also a small discussion of the rise of the ergative construction in MIA in H.H. Hock's (1986) _Principles of Historical Linguistics)_ (pp. 431-432).

Piotr

----- Original Message -----
From: Harald Hammarstrom

To: cybalist@yahoogroups.com

Sent: Saturday, March 23, 2002 6:19 PM

Subject: [tied] Ergative construction in Middle Indo-Aryan and Iranian

Since we have some experts here on IA and II developments, what is the status of this common feature ? As far back as sanskrit I've seen the agent in the instr., patient. in the nom. and past ptcpl. in -ta- congruent with the patient - completely ancestral to the case of Modern Hindi. In Indo-Iranian, for instance Sogdian or Pashto it doesn't look the same morphologically, but completely the same structurally i.e in transitive past sentences. Are there are theories on how it developed? Independently in II and IA or joint? Due to some substratum? Why only in the preterite-ish tense?