O&S has root #231, *bariK-, 'shine, be bright' (K being under-dotted k; O&S
also use q and top-dotted q)
O&S have recaptured the earliest vowels of this word, I believe.
This word is seen in Arabic b-r-q and Egyptian b3q, 'bright'. Even the
Sumerian cognate is known: pirig-2/3, 'bright' (from *parig through vowel
assimilation).
The point I will make is related to but not meant to replace Piotr's
anticipated answer to your question in terms of standard PIE theory. But it
is such a neat demonstration of one of the points I was making that I cannot
resist.
The earliest theoretical form (without the effects from stress-accent) of
the PIE is **bhere:g^-. With the natural tendency of long vowels to attract
the stress-accent, it produced the form *bhré:g^- with zero-degree root
vowel.
When the stress-accent was either retained on the root syllable, or
transferred back from *bhré:g^-, the result is *bhér6g^-, with *6 (schwa)
representing the formerly stress-accented long vowel.
Piotr (and, of course, many others) restore *H1 _after_ the *r to account
for the length of *e: and the *6 which is the result of a stress-unaccented
long vowel.
*H1 is not supposed to 'color' vowels; and since the vowel we expect in a
stress-accented syllable (according to the rules of Ablaut) is *é, the
addition of *H1 lengthens it to *e: _without_ coloring it. If *H2 were put
in its place, the anticipation by standard theory would be *bhra:g^-; and
with *H3, *bhro:g&-. Both of these, however, would become *bhér6g^- ejust as
from *bhre:g^-.
But, I believe this is one of the words in which an aspirated 'r' /rH/
lengthens the vowel that follows it, so retaining its quality through all
the previous stages to PIE.
Since the original (PPPIE) syllable was *rhi, the resulting lowering to *re:
can be reconstructed whenever the stress-accent allows its full expression
in subsequent stages of PIE.
I think it may provoke a thought or two by some that we could be talking
about virtually the same phenomenon: the presence of *H before or after a
vowel lengthening it (*rhi can be rewritten as *rHi). This would mean, of
course, that all speculation concerning the phonetic nature of the various
*H's: *H1, *H2, and *H3 is an exercise in futility.
This means that *H(no numeral) was simply [h].
Now, just to round off the analysis: the _root_ of this word is PPPIE
*bharHi-/*bhari:-.
It became in PPIE *bhAre:-; and in PIE *bhére:-; to this _root_, *g^ (from
PPPIE *[n]gi; PPIE *[n]gYA-; PIE *[n]g^[A]) was added.
Normally the closed syllable would attract the stress-accent, so the first
outcome of this addition is **bho/Øré:g^-; the subsequent general shift of
stress-accent to the root syllable produced *bhér6g^-.
That is to say that I disbelieve in "laryngeals" per se.
Some roots of the form *Ce:- are the result of earlier *CiH-; e.g. *pe:-,
'damage' (it has the HS cognate *pa?-, 'dig, bury'; the vowel discrepancy is
due to early vowel leveling: <- **pi?).
I am sure that Piotr has gien you an swer that will satisfy almost all
PIEists on the list.
Patrick
----- Original Message -----
From: "fournet.arnaud" <fournet.arnaud@...>
To: <cybalist@yahoogroups.com>
Sent: Monday, February 11, 2008 3:06 AM
Subject: Re: [tied] Re: The meaning of life: PIE. *gWiH3w-
>
> *bHr.h1g^o-
>
> What are the reasons for H1 in this word ?
>
> I prefer H2.
>
> Arnaud
> =============
>
>