From: Piotr Gasiorowski
Message: 31181
Date: 2004-02-19
----- Original Message -----
From: "elmeras2000" <jer@...>
To: <cybalist@yahoogroups.com>
Sent: Tuesday, February 17, 2004 4:53 PM
Subject: [tied] Re: Eggs from birds and swift horses (was: the palatal sham)
> I don't like to compromise the infix theory by basing it on weaker
etymologies than the ones I have used.
There's no danger of that. Since the infix theory is already firmly based on
secure etymologies, it stands on its own (at least as far as I'm concerned;
I'm surprised that there's been so little resonance among the IEists). My
hypothesis concerns only a subregularity that I think can be observed in a
certain specific environment. It could easily be integrated with your
findings, since it doesn't contradict them in any way, but it can be
rejected or accepted on its own, without affecting the validity of your more
general theory. In fact, it _depends_ on its validity, since it _assumes_
that your analysis is fundamentally correct. It predicts this and only this:
For phonotactic reasons, roots of the shape *CLeh will yield *CoL(h) when
*O-infixed ; they will also characteristically fail to attract stress, just
like *Ceh roots, but unlike *CeLh roots.
This means that from the root type in question I'd expect to get the
following infixed forms:
*CoLh-o- (not *CLoh-o-)
*CoL-m/n-�-, *CoL-m/n-�h2 (not *C�L-m/n-o/ah2 or *CLoh-m/n-�/�h2)
*CoLh-�je- (not *CLoh-�je-)
The above may also be true of *Cjeh- roots, cf. e.g. *gWojh3-�je- (Slavic
*gojiti 'heal') from */gWjeh3(-w)-/. I have already presented some forms
consistent with this analysis. To _falsify_ it one should offer clear
counterexamples. Individually, the forms may be capable of other
interpretations, but what I offer instead is a minor generalisation that
rounds them up as a (sub)type. I also think it may help to clarify some
cases of apparent schwebeablaut.
Whether Gk. ple:ma and ble:ma (either or both) reflect original full grades
or analogical nil grades is of little importance. There's enough (or shall I
say a plethora of) evidence of Greek words being derived from *pleh1-
(rather than *pelh1-, which I would argue is a secondary variant, like
*perk^- from *prek^-), presupposing pre-Gk. *pl�h1-mn. whatever its later
fate; and *gWl�h1-mn. is supported by Welsh blif 'catapult'.
Piotr