As I said I don't quite understand it
myself. However, this Gen.sg. *dem-s is quite well substantiated (Avestan da:N
'of the house'). Nasal (and heteroclitic) stems in general show similar
alternations at least in some branches (*-o:n/-r/-l, *-en-s), cf. also Avestan
xWa:N 'of the sun' < *s(h2)wen-s. PIE *do:m is not necessarily inanimate
(gramatically, that is) -- an asigmatic nominative is to be expected in this
case.
The -o-/-e- alternation originally
reflected pre-PIE stress contrasts, but by PIE times it had acquired
morphological functions of its own, quite independent of stress patterns. The
form *dem-s is "late" in the sense that it must have been formed after the
period of stress-based vowel reductions but "early" in the sense that it
reflects an old consonantal stem (irregular already in PIE) and is more archaic
than thematised *dom-o-. As for thematisation itself, it hardly needs accounting
for, being the simplest and most productive method of regularising
morphologically "difficult" nouns.
Piotr
----- Original Message -----
Sent: Wednesday, July 04, 2001 10:11 AM
Subject: Re: [tied] Neptune, Poseidon, Danu, etc.
> One IE reconstruction of the "house" word is a consonantal
stem declined like this: Nom.sg. *do:m, Gen.sg. dem-s (the o-stem form *domos is
supposed to be younger).
Is there a plausible (or at least consistent)
explanation of the ablaut in that specific case? Stress seems not to help,
laryngeals as well. Again, does the consonantal stem indicate a former inactive
class? If so, why thematized and converted to active ->
masculine?