--- In
cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>
> On Mon, 31 Oct 2005 10:21:02 +0000, tgpedersen
> <tgpedersen@...> wrote:
>
> >It could accounted for with these rules (I've added an
intermediate
> >stage pre-ProtoSlavic out of convenvenience of rule ordering)
> >
> >PSatem > p-PS
> >*i > *ji
>
> No. *i gives /I/, *ji gives /ji/ (at least in OCS).
>
> >alternatively, one could skip the kj-stage and let the front
vowels
> >and j palatalise the k's directly
>
> Yes. That's the first palatalization.
>
> When *ai was fronted and monophthongized to the long vowel
> */E:/, /kai/ underwent the second palatalization, resulting
> in /cê/. That's all.
Not quite.
The interesting part here is that you can derive the whole thing
from a state where PIE traditional *kW has allophones kW and k. That
means it may be possible to derive of eg. Indo-Iranian, which also
has secondary palatalisation, in the same way. Secondary
palatalisation might thus not be einzelsprachlich.
Here's a version with even fewer rules:
PIE > pp-PS
*ai, *oi > *e
pp-PS > p-PS
*V(:) > *jV(:), /V front
p-PS > PS
*k > *c
*kW > *k
PS >
*cj > *c^
PIE pp-PS p-PS PS
*kWa/*kWo > *kWa > *kWa > *ka > ko
*ke > *ke > *kje > *cje > c^e
*ki > *ki > *kji > *cji > c^I
*kWu > *kWu > *kWu > *ku > kU
*kWa:/*kWo: > *kWa > *kWa > *ka > ka
*ke: > *ke: > *kje: > *cje: > c^a
*ki: > *ki: > *kji: > *cji: > c^i
*kWu: > *kWu: > *kWu: > *ku: > ky
*kWai/*kWoi > *kWe > *ke > *ce > ce
*kWau/*kWou > *kWau > *kWau > *kau > ku
*kei > *kei > *kjei > *cjei > c^i
*keu > *keu > *kjeu > *cjeu > c^u
*kja/o > *kja > *kja > *cja > c^e
*kju > *kji > *kji > *cji > c^I
*kja:/o: > *kja: > *kja: > *cja: > c^a
*kju: > *kji: > *kji: > *cji: > c^i
*kjai/oi > *kji: > *kji: > *cji: > c^i
*kjau/ou > *kjau > *kjau > *cjau > c^u
Torsten