From: elmeras2000
Message: 33334
Date: 2004-06-30
--- In cybalist@yahoogroups.com, enlil@... wrote:
[I will have to reply to this in a number of sections. My computer
may have a virus that makes things disappear; as if the discussion
were not unplesant enough.]
> Jens:
> > [...] and the thematic vowel is unaccented in gignó:skein,
> > gignó:sko:n.
>
> Forgot about that recent *-ske- suffix. My bad.
>
>
> > When such an accent is retracted from an athematic stem the
> > reduplication shows -e-, when it is thematic it shows -i-:
>
> Since *bHer- shows both *bHebHor- and *bHibHer-, and since we know
that
> *bHer- is irrefutably a thematic stem, you assert yet another
baseless
> rule that can't possibly exist.
The reduplicated present of *bher- is irrefutably athematic: Ved.
bíbharti/bibhárti, Gk. piphránai. Its alternants were *bhi-bhér- (or
*bhi-bhór- in case those who claim "o-grade" for reduplicated
presents parallel to the perfect, the intensive, and the
reduplicated aorist, are right) and *bhé-bhr-.
> >> The original accent in the *i-reduplicated stems is on the
second
> >> syllable.
> >
> > Sure thing, nice to be understood.
>
> I don't understand you, but it's nice to agree nonetheless.
>
>
> >> We then don't need an imaginary *i-infix here because we can
> >> unceremoniously explain *i in these stems as the reflex
> >> of phonetic schwa in an open syllable before accent.
> >
> > No, that gives zero: gen.sg. *H2n.r-ós (stem *H2ner- 'man').
>
> No, you're confusing MIE schwa with eLIE schwa.
>
> This *xnrós is of the same pattern as other athematic stems such as
> *kwo:ns/*kunós that have already been explained. We see zeroing in
> the genitive because of the loss of unaccented MIE *a. So in MIE,
> we have a nominative *hanéra-sa with a normal genitive *hanar-ása
> (with accented *e being reduced to unaccented *a), just as *kawána-
sa
> has genitive *kawan-ása. This all becomes *xné:rs/*xnrás after
Syncope
> just as we have *kwa:ns/*kunás. From there, Nominative Loss and
Vowel
> Shift create what is actually attested.
>
> What we see here is _MIE unaccented *a_ (which disappears after
Syncope),
> It's not eLIE unaccented *a, which is from MIE *e. That later
schwa can
> indeed become *i pretonically. You're not following properly.
Is the chronological order MIE - eLIE?
> > Wait, I can't see the accent mark on agricola, and if I could it
> > would not amount to much since it's Latin. What is the basis of a
> > stem variant of Greek agrós, Ved. ájra- with -i- as its stem
vowel?
> > I don't know such a form at all.
>
> Well, true enough, I guess. I'd presume that a compound like this
would
> have accent on the initial in the latest stage of IE since it's the
> acrostatic crap again. It would be exceedingly hard to identify a
> compound built with a thematic stem plus athematic stem, which is
> what we need to see the pattern clearly.
No, that is quite common: Vedic vr.tra-hán-, ratha-yúj-, veda-víd-,
r.ta-stúbh-, ghr.ta-dúh-, etc. For the accent, however, the ablaut
and the evidence of Greek homó-zuks, khér-nips, etc. agree to show
that it was originally on the first part. The shift to the second
part in Indic (after ablaut) may serve to indicate the adjectival
function (high animacy).
> However, we can use the
> Acrostatic Regularization rule, which is real enough, to understand
> what has happened.
I very much doubt there is any such rule, although I don't have the
faintest idea what you are talking about. We already know what has
happened.
> > Well, we do, -u- is the unaccented reduplication vowel with roots
> > containing -u-.
>
> Example? Evidence in multiple branches?
Right: Vedic pf. babhú:va, babhu:vúr vs. Avestan /buba:va/,
3pl /ba:buvar/ show -u- before full grade, *-e- before -u-. The same
alternation is seen in Italic, Osc. 3pl.pf. fufens, Umb. 3pl.fut.ex.
fefure. There is also agreement between Ved. s'us'rá:va and OIr.
ro:cúalae (Welsh cigleu from *ku:-klow-e).
Jens