On 2008-10-11 15:51, afyangh wrote:
> As regards the Ruki law *s > s^ in Balto-Slavic and Indo-Iranian when
> s is preceded by any of r u k i,
>
> Do we have cases like r u k i + T/D + s ? Does the phonotactical
> "affricate" ts become s^ ? or does it remain -s- ? In other words, is
> there a different development when some intermediary dental stop was
> there ?
Yes, RUKI was blocked by an intervening stop. Medial *-Ts- was possible
only if there was a morphological boundary between the consonants, and
in suffixes beginning with *s there is no shift to *s^ after *-iT-,
*-uT- or *-r(.)T- (I can't think of any examples involving *-KT-s-).
Clear examples are provided e.g. by such Sanskrit forms as sigmatic
aorists, desideratives and futures, locatives in <-su> and the like:
<abHutsi, bubHutsati, bHotsyati> from *bHeudH-, <vartsyati> from *wert-,
<hr.tsu, trivr.tsu> from *k^erd- 'heart' and *tri-wr.t- 'threefold'.
Piotr