--- In
cybalist@yahoogroups.com, "stlatos" <stlatos@...> wrote:
> --- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@> wrote:
> > as in *dje:us. Secure examples with *-e:h2(s) or *-h2e:C-s may be
> > difficult to find. What was definitely not coloured was "Narten"
length,
> > as in Hitt. hekur 'summit' < *h2e:k^-wr.
>
> There is no reason for lengthening in this position. The change
of stressed a>e when
> kY>k before w caused this (as in ekw- / akw- 'drink').
> All of this is even more complicated; I'm sticking close to
standard theory.
>
The timing of kYw>kw and the nature of the change can also be seen
by taking into account some other rules:
n-N > s-N
*enterno- > L internus, H istarna 'in the middle', istarnya- 'inside'
*xWn,gWHno- > Celtic *xaNWgW(i)no-, H saNkWa:(i)- 'fingernail'
H eshanwant- / eshaskant- 'bloody'
*xWn,gW-, *xWn,gWn, > L unguen 'salve', H isk- 'annoint', (met.?)
sakni- 'oil'
This leaves the question of why eshanwant- / eshaskant- has a change
sw > sk. This is probably related to a regular rule:
sw- > st-
*sworgheye 'lie ill' > H istarkya- 'fall ill'
This also affects kYL- which became kYw- after the change kYw>kw above:
kYw- > c^w-? > sw-
*kYLu- > L clue:re, H istuwa- 'become famous'
*kYLewos > *swowos >> H istamas- 'hear, learn'
The change of a special velar to a fricative before w is probably
also seen in:
*gWixWwo- > *xWigWwo- > *xWidzwo- > *xWizw(a)- > H huisu- 'alive'
The order in standard theory would become:
n-N-N > l-N-N
n-N > s-N
kYw > kw
L>w / kY_
uw > um
w-w > w-m (varies by env.)
kYw- > sw-
sw- > st-
sw > sk
0>e / #_sC
e>i / #_
Another possibility is that all sw>st and *eshastant- > eshaskant-
by t-t dissim.; the timing of gWw changes depends on the development
of gYL and kWw, which probably didn't exist or leave traces.