--- Miguel Carrasquer <
mcv@...> wrote:
> On Wed, 21 Jun 2006 22:17:41 -0700 (PDT), Sean
> Whalen
> <stlatos@...> wrote:
> > Most of these examples of o-derounding are found
> in
> >Andrew Sihler's _New Comparative Grammar of Greek
> and
> >Latin" and more can be found there (though I don't
> >think he gets everything concerning this correct).
>
> Well, none of the examples below are given by Sihler
> as
> examples of o-derounding, which is why I asked.
Check in Section 71.1 for
> >L mare < *mori
The problem is he appears to assume the rule in
individual entries but doesn't describe it under the
section on PIE *o, H3e. I was sure I remembered it
being detailed between examples of wo>we and ow>aw but
maybe that was a later paper.
> Here Celtic points to *mori, Latin to *mari.
Since they're obviously from the same word and
there's no question o>a sometimes in Latin I'd say
*mori.
> >o>a after a rounded or labial C
> >
> >L quattuor *(kWe->kWo->kWa->kwa-)
>
> That's rather from *kWtwó:r, like Grk. pisures or
> (West-)Slavic c^Ityre.
Check in Section 389.4 and elsewhere for the
necessity of *kWé-.
There's no such form as *kWtwó:r in PIE; *twor-/tur-
is original and *-kWé (prefixed to mean "and _") was
added by false morpheme division (also in "five").
In Greek *e>i in some environments; I'll begin
describing some if you want.
I thought unstressed *e in an open syllable was
commonly assumed to become *i>I? I know some
languages show variation between c^e- and c^I- in some
words.
> >L faber < *dhobhro-s
>
> From *dhabhros, cf. Arm. darbin.
In Arm. o>u in some environments, then o>a. None of
the cognates for *dhobhro-s, etc., is able to
definitely show PIE *o or *a due to their specific
changes.
> >L pars < *porti-s
>
> From *pr.Htí-, cf. Skt. pu:rtí-. *port- gives port-
> as in
> portare, porta, portus, etc.
I'd say that *pr.tí-s > *porti-s > pars. If
*pr.Htí- then *pra:ti- in L. This stem and *pel- (or
my *p^èl-) have forms plain or with various affixes.
> >L maneo: *moneye-
>
> From *mn.-éh1- "I stay". *mon-éye- gives moneo: "I
> warn".
I gave examples enough to show that an individual
form may have either o or a (fo-/faveo:); with the
sporadic nature of the changes this is no
counterexample. Also see 100.c for counterarguments
to your derivation and n.>an, etc.
Since *o>a in so many languages it's sometimes hard
to find a cognate to a Latin word with a to show it
came from *o but I think there's plenty of examples to
support this change of *o (and r.>or, etc., which I
didn't specifically mention before).
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com