Re: Indo-European question

On Tue, Jun 19, 2001 at 10:37:35PM -0000, Lars Henrik Mathiesen wrote:
> > Date:         Tue, 19 Jun 2001 15:32:20 -0500
> > From: Eric Christopherson <rakko@...>
> >
> > On Tue, Jun 19, 2001 at 11:02:38AM -0000, Lars Henrik Mathiesen wrote:
> > [snip very interesting specululization on PIE phonetics]
> > >                         PIE             //
> > >         short           *Ø i u          /a i u/ > /@ i u/
> > >         stressed        *e ei eu        /"a "i "u/ > /a ai au/
> > >         long            *o oi ou        /a: i: u:/ > /A Ai Au/
> > >                         *e:C            /a:CX/ > /ahC/
> > >                         *o:C            /a:CX/ > /AhC/
> > >
> > > where /C/ is any conoid and /X/ is some unknown sound that marked
> > > nom.sg. for thematic nominals (since that's the main occurrence of the
> > > lengthened ablaut grade).
> >
> > I'm confused here. What would be an example of this /X/? Also, how can *e:C
> > and *o:C both represent /a:CX/, but end up with two different outcomes?
>
> *e:C should be /"aCX/, sorry. (I should perhaps mention that the *e
> grade seems to occur where the stress reconstructed for PIE fell --
> unless there was an *o there already. If those *o's are to be
> explained by an older stress pattern, as I've seen suggested, the
> ordering of diphthongization, lengthening and backing becomes a bit
> complex).
Also adding complexity is the presence of a few accented syllabic
consonants (which "shouldn't" occur), such as *sept"m= instead of *"septm=.

> I'm not a phonologist --- but /X/ could be /h/, for instance, if your
> model doesn't assign that to any of the laryngeals. Looking in Beekes
> again, in his version of Pre-PIE-was-ergative this lengthened grade
> occurs in the original ergative singular of animates (unlike the -s of
> most later nominatives, which was borrowed from a genitive). Beekes
> assumes that the marker of this ergative case was zero, but doesn't
> explain (anywhere I can find) why the vowel gets lengthened.
Ah, fascinating stuff. I know a little about conventional PIE theories, and
next to nothing about the ergative or active theories. Is there anything of
Beekes on the web?

From the little I've read, nominative singulars which once had **-s but then
dropped it experienced vowel lengthening to compensate for the lost *-s.
E.g.:

**ph_2ters "father" (nom.) > *ph_2te:r (perhaps via **ph_2terr)

where ** means a form which has been reconstructed based on inference from
other forms present in the same level of development the language, as
opposed to *, meaning a form reconstructed from later forms

So anyway, is *ph_2te:r the kind of lenghtened grade word you're talking
about?

--
Eric Christopherson | Rakko

Reply