Re: Scientist's etymology vs. scientific etymology

> > but once you get quatio: separated from the rest of Latin with a
> > respectable family elsewhere, which is new, it is very tempting to
> > join quattuor, tri-quetrus etc to that family.
>
> LIV explains <quatio:> etc. as a neo-weak grade of a root
> reconstructed as *(s)kweh1t- (root aorist *kwe:t-, cf. Gk. pe: -->
> present *kwat-je/o- like <facio> from *dHeh1-). Such structures,
> however, are notoriously difficult to analyse. Cf. LIV *kwath2-
> 'bubble, ferment' -- shouldn't it really be *kwah2t- ~ *kwatH- (via
> Olsen's preaspiration)? This would at least account for the Slavic
> alternation *kvas-/*kys-, not to mention
> the *a in the root.
>
> > But I was wondering if that schwa secundum could also somehow
> > explain that weird wa/u 'ablaut' in Latin?
>
> I suppose it could, but which particular words do you mean?

Let's try this:

PIE *ku- "beat",
ppp. *kW-tó- -> *kW&-tó- -> *kWa-tó- -> stem *kWát-
but with preverb
ppp. *´-ku-to- -> stem *´-kut-

similarly
PIE *dhegh- "put, do" (trad. *dheh1-, but I don't like h1's)
ppp. dhx-tó- -> *dh&x-tó -> *fax-tó- -> *fak-tó- -> stem *fak-
but with preverb
ppp. *´-dhex-to- -> *´-fex-to- -> *´-fek-to-

The reason I don't like h1's:
Dutch stems rijd-/rij- "drive"
eg. rijden, ik reed "I drove", but ik rij "I drive"
Sw. bry(da) sig "care", but pret. bröd "cared"
Da. bryde sig om "like", pret. brød etc
in other words double stem verbs of the type C*VC-/C*V-,
if something similar happened in PIE (with the same motive, saving one
syllable in the verb suffix), but with compensatory lengthening of the
stem vowel, we'd get an artificial h1 into the analysis. That's what I
think happened with PIE *dhegh-.


Torsten