--- In
cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:
>
> On 2008-12-19 12:40, tgpedersen wrote:
>
> > Why so few? Why so many a's?
> > And a good deal of them seem to be NWEurope or Germanic only.
>
> Few? There are _a good_ few, with some more material mentioned in
> passim, though not discussed in separate sections. Since the book
> deals only with Gmc. words, no wonder that some of the are areally
> restricted.
There are three explanations for 'areally restricted'
1) loan from substrate
2) spontaneous creation
3) sole survivor
I tend to prefer 1), as Brian noted. I think there are good reasons
for that.
> Quite number of them, though, are widely distributed IE words with
> impeccable PIE etymologies.
Like *Bar(u)xa- : *BaruGa- "verschnittener Eber", which looks
suspiciously like *pork^-, but then it's a loan from somewhere else.
And then there's my old friends from the transport department
*þre:xila- (*þraxila-?) : *þreGila- "Diener, Knecht"
*þranxu- : *þranGu-(?) "gedrängt"
http://tech.groups.yahoo.com/group/cybalist/message/61626
> Why so many a's? Well, perhaps because most of them reflect PIE *o
> PGmc. *a.
The alternations going the way they do in these glosses, that would
mean an alternation between stressed o-grade and pretonic o-grade.
That is not possible. The alternative explanation, schwa secundum,
presupposes that that has been generalized from the zero-grade form.
But why, then, does that not happen in verbs (except possibly the
class VI ones)? Why can't the nouns behave properly like the verbs?
Why have those few nouns generalized away the vowel-grade
alternation, but kept the effects of it (Verner)? And this must have
happened when the alternation was still paradigmatic in the language,
because there is no trace of a connection between the chosen vowel
grade and the Verner variant in existing forms in the descendant
languages?
> > There are only two s-neuters on the list beside *xrinþaz-, one is
> > *kelþaz, of which *kendaz might be suspected of being a variant,
> > although I can't see what kind of root *k- would represent.
>
> *gel- 'curl up' (vel sim., see Pokorny), hence some words
> for 'fetus' (--> child) and related meanings (Goth. kilþei 'womb',
> Skt. jartu- 'vulva').
I don't think the semantics fits what those speakers would see as a
defining characteristic of fetuses or children. Anayway, I was
wondering what kind of verb *k- in *k-ent- would be.
Torsten