Re: ka and k^a [was: [tied] *kW- "?"]

Miguel Carrasquer wrote:

> As I have stated before, I believe that PIE */a/, where it
> does not come from colouring by *h2 (*h2e > *a) or by
> *k/*g/*gh (**qa > *ka), must come from a (pre-)PIE nasalized
> long vowel */a:N/ before C (e.g. **(H)na:Ns > *(H)nas-,
> **g^ha:Ns- > *g^ha(n)s, **dha:Nbh- > *dhabh-, etc.)

Then perhaps *k^as-, if from **k^a:N-s-, has something to do with
*k^ank- ~ *k^a:k- 'rush, spring forth, leap', as in Gk. ke:kio:
(ka:kio:) 'gush', Lith. s^ókti 'leap', and Gmc. *xanxistaz/*xangistaz
'steed'.

> The existence of a PIE /ai/-diphthong would seem to be an
> embarassment, but the word *laiwos gave me an idea. What if
> the original shape of the word was **la:ñ-wos (out of
> earlier ***la:niwos or ***la:miwos), where the palatalized
> /ñ/ nasalizes the preceding vowel, and itself gets reduced
> to /j/? That would allow tying in the word *laiwos with the
> root *lem-/*lom- "weak, lame", which makes an excellent
> semantic fit.

Interesting, though I suspect that the "adjectives of defect" with
*a-cocalism form some kind of group, if only because of secondary
contamination. If we admit k-colouring in *skaiwo- and *kaiko-, the
vocalism of *laiwo- may have been the victim of "peer pressure".

Piotr