Re: [tied] PIE *&>u?

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

>
> On 2007-09-07 21:53, stlatos wrote:
>
> > So, you believe that PIE & > o in Latin (homo:);
>
> Originally <hemo:>, as in Old Latin. Some people prefer the notation
> *CR.RV- or *CR.V- to my *C&RV-. My preference for the latter is due to
> the fact that there are occasional discrepancies between the behaviour
> of "true" *R. and the epenthetic vowel before a prevocalic resonant,
> like here:
>
> > what about your
> > previous description of Cr,H- > Car-,
>
> This _is_ exceptional (with /a/ rather than /o/), but the Lindeman
> treatment of *CrV- is nevertheless the same as the reflex of *Cr.HV- in
> Latin, both yielding CarV-.
>
> > or supposed *t&thlo+ > tabula?
> > What are the changes and order of H,>a in comparison to this?
>
> It's the "laryngeal schwa", which behaves differently from the
> epenthetic vowel.

You said that wasn't from a laryngeal last time (but l>& after
non-syl. h dropped causing t>tH). How many schwas are there? Do you
say there's only one "laryngeal schwa"?

If there are two types (at least) why does x() > u or i in some
environments in Gmc.? The change C+pal $ x,>+pal>i seems necessary;
how can that happen if they were schwas not C+syl. in PIE? Why would
Indic r>&r be changed to ur in some env. but not n>&n (and not
x()>&(2-4?)>u)? What possible characteristic of schwas would prevent
this if they existed at the same time?

> > Why bul- if from *bh&l-? Why 0>& in such a common
> > initial cluster and not in other words (or for equally common clusters
> > of the same type)?
>
> "Lindeman's Law" appears (optionally) in words that would otherwise be
> monosyllabic, like *dje:m ~ *die:m, *k^wo:n ~ *k^uo:n, possibly

*wre:n ~

> *w&re:n, (Gk. hré:n ~ [w]aré:n), etc. The word *bHle:n meets this
> description.

What about Slavic verbs like mInEti, osedUlati, or Skt. rudhira-?
Are these supposed to be explained by PIE schwas instead of specific
rules in each branch?