Re: Diphthong Distributions

From: tgpedersen
Message: 46275
Date: 2006-10-04

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:
>
> On 2006-10-04 10:57, tgpedersen wrote:
>
> > Tocharian has a lot of affrication of t's. If *st was once an
> > independent phoneme and not a composition with *t, it doesn't
> > have to follow *t to become *sts or *stj vel sim.. The problem is:
> > if 2sg *-si is from *-ti, where's 3g *-ti from?
>
> Perhaps **-Nt-, as suggested by Olsen/Rasmussen. You should like
>it :)

I noticed you stopped objecting? Sometimes I think I'm right. ;-)

I do, actually. I read up on Beekes' "Roots with Nasal Infix in
Pokorny": turns out they end in occlusive, laryngeal, glide
(-y, -w) or -s. I toyed with the idea of letting the choice of
root-final occlusive/laryngeal be dependent on succeeding consonant
or vowel (athematic/thematic), -s comes in handy there as stand-in
for a missing laryngeal.


> >>> 4) Roots in *gW(H)eu-/*kWeu-/k'weu- are nonexistent
> >> How about the 'cow' word?
> >
> > I think you mean *gWogW-?
>
> No, I mean *gWou- (you can check it in that PDF I sent you). It's
> true, though, that the underlying root *gWeu- may have been more
> complex at some pre-stage of PIE (I leave that possibility open
> in the article).

Actually the choice depends on whether you assume root-'normalization'
(Grassmann-style) to have taken place before or after ablauting. The
fact that there's no root of the shape *KWew-, assuming that's what
'illegal' roots of type *KWeKW- were made into, means the cow/mobile
thing is unique in PIE, either because *gWeH- was the only verbal
root in *DeC- they ever thought of reduplicating, or because it's a
loan.

That means the OC *wàân,? "go" / *wàG- "go to"
http://tech.groups.yahoo.com/group/cybalist/message/45335
might be the same root, 'normalized' at the other end (and not
PST = PPIE *w > PIE *gW, as Pulleyblank proposed). But then, the
corresponding PIE roots (Lat. vacca, OHG wanchon) make a distinctly
NWEuropean, substrate impression (root vowel -a-, "mot populaire").

Hope I'm not revealing too much. How's the article coming along?


Torsten