Re: [tied] RUKI-rule in Indo-Iranian

From: Piotr Gasiorowski
Message: 5373
Date: 2001-01-09

Hi, Anna,
 
I'll try to help you but I need a little time. Here are some basic facts just in case. Hope I'm not insulting you by giving them. If you've got it all at your fingertips, it may still be of use to other interested Cybalist members.
 
The "RUKI" rule consists in the retraction (palatalisation or retroflexion) of *s and *z, i.e. the voiced allophone of *s, after etymological *r, *k and *u, *i (short and long, also occurring as final elements of diphthongs). In Old Indic the RUKIed *s changes into [s.] (a retroflex fricative), causing also the retroflexion of any following t, d, n:
 
*ek^wo-isu 'horse-Loc.pl.' > *as'wais.u
> Skt. as'ves.u
*snus-a: 'daughter-in-law' > Skt. snus.a:
*do:-tr-su 'giver-Loc.pl.' > Skt. da:tRs.u (R = syllabic r)
*dweis-ti 'hate-3sg.' > *dwais.t.i > Skt. dves.t.i
 
RUKIed *z is eventually lost, causing the compensatory lengthening of the preceding vowel and the assimilatory retroflection of any following coronal, e.g.
 
*nisdo- [nizdo-] > *niz.d.a- > Skt. ni:d.a
 
The place-of-articulation contrasts in sibilants were generally eliminated in Middle Indic, and are lost in most of Indo-Aryan.
 
In Iranian, RUKIed *s, *z > *s^, *z^. Adjacent coronals are not affected.
 
*misdHo- [mizdHo-] 'wages' > Avestan miz^da- (cf. Skt. mi:d.Ha-).
 
Later developments are rather complex, but distinct reflexes of *s^, *z^ generally survive in Iranian.
 
Piotr
 
 
----- Original Message -----
From: Anna Shparberg
To: cybalist@egroups.com
Sent: Monday, January 08, 2001 11:06 PM
Subject: [tied] RUKI-rule in Indo-Iranian

Hello,

I am interested in the manifestations of the RUKI-rule in Indo-Iranian. 
Could anyone help me with locating any literature on the subject?

Thanks in advance!

Anna Shparberg