Re: [tied] *kW- "?"

--- In cybalist@yahoogroups.com, "Rob" <magwich78@...> wrote:

> --- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@...>

wrote:

>
> > But if *e occurs only as an allophone in the set /e/, /o/, zero,

of

> > the phoneme *a, then 'palatalised velars' are palatalised

only

> > before the allophone /e/. In other words, the phoneme

sequence

> > *-k^a- (which is realised as *-ka- in proto-IE) has the

allophones

> > *-c^e-, *-ko-, *-k- in PIE.
>
> That is certainly true from an articulatory point of view. Velars

are

> much more likely to be palatalized next to front vowels than any

other

> kind of vowel. It seems to me, then, that the centum-satem split

did

> not occur until after the separation of the qualitative Ablaut
> allophones into independent phonemes.

Ah, thank you. It has seemed that way to me too now for some time.
Finally someone gets it. Now bring on the citicism.

>Then the IE velar series (at
> least in the dialects that would later become the satem languages)
> would have had palatalized allophones next to front vowels and non-
> palatalized allophones otherwise.

Most likely in them all. I believe depalatalisation by paradigm
regularisation is possible, followed by propagation of the "right"
allophone into uninflected envronments as well (by 'shibboleth'
effect).


Torsten