Re: IS PIE * DERU EXCLUSIVELY INDO-EUROPEAN ?

--- In cybalist@yahoogroups.com, "Patrick Ryan" <proto-language@...>
wrote:

>
> In my opinion, the changes you attribute to a possible *d-prefix are

better

> explained as the result of phonological processes.

Then why are they so irregular?

> /n/ is the apical nasal corresponding to the voiced apical stop /d/.
>
> Denasalize /n/, perhaps for dissimilation, and you get /d/.
>
> Denasalize an alveolar /n/, and you are liable to get an /l/.

Are you suggesting that PIE had phonemic apical and alveolar nasals.

> There are absolutely no infixes in PIE. The one reputed example is the
> suffix -*nV, which, in certain cases, is metathesized to a position

before

> the final consonant of the root.

Possibly. My guess is that the *w and *y prefixes would have undergone
metathesis, just as *d does in a number of Greek words.

> Any word that shows an initial vowel in PIE must be reconstructed

with a

> preceding laryngal: *H. If there is a true variation between initial

*kV and

> *ØV, it may be a case of a dialectal hardening of the laryngal,

namely /h/.

But why the irregularity?

> In other cases, we simply coincidentally have words of _similar_

meanings

> beginning with *d- and *n-.

Postulating a *d prefix solves all of these problems in one fell
swoop. This prefix wins by Occam's Razor.

> For prefix status, I should have added that we be able to isolate the
> meaning of the prefix: any ideas or what these purported prefixes mean?

This isn't really necessary. Sino-Tibetan has several prefixes with no
discernible meaning. Yet we know they existed. PIE and Proto-Semitic
have numerous suffixes (root determinatives) with no discernible meaning.

I'm willing to hazard a guess though. I suggest that these prefixes
are all remnants of noun classes (verbs would agree in class with
their head noun). That's very speculative, I admit. However, if I can
stray into the world of Nostraticism for a moment I would point to the
PAA affix *t which indicates feminine gender. PAA *t normally
coincides with PIE *(a)t and PAA *t' with PIE *d, however, according
to Bomhard in PAA affixes lose their ejective feature. I'm tempted,
therefore, to propose a Nostratic *t' noun class marker (it needn't
necessarily indicate feminine gender). I'm really only at the
beginning stages of my research in this area, but if I'm right there
should be evidence for this noun class system in other Nostratic
languages. However, that's a discussion best left for Nostratic
oriented groups.