Re: IS PIE * DERU EXCLUSIVELY INDO-EUROPEAN ?

On 2008-01-27 09:30, fournet.arnaud wrote:

> dakru = akru

OK, that's one of them, "Western" *dak^ru ~ "Eastern" *h2ak^ru-.

> arbor = doru

Any particular reason for connecting these two, other than the meaning?

> arse = dorsum

The 'arse' word is to be reconstructed as *h1orsos. But medial *-rs-
gives Lat. -rr-, so <dorsum>, whatever its etymology, reflects *-rt-t-
rather than *-rs-.

> eigh = deigh (to stick, to sting)

What is "eigh"?

> nebh = dneph (cloud)

If you mean variation in Baltic, it's an inner business of that branch
and involves a few items with /d/ for *n (Lith. debesi`s 'cloud') and
/n/ for *d (Lith. na~mas 'house'). The numeral '9' has /d-/ also in
Slavic, but this may be due to the influence of *dek^m.t- in serial
counting. Miguel Carrasquer Vidal had a theory about the peculiarities
of '9' in Greek, Armenian and Balto-Slavic, explaining them all with one
fell swoop as something like *h1(e)dH-néwn., *h1edH- meaning,
hypothetically, 'one' (cf. Slavic *(j)ed-inU), but even if there's any
truth in it, *h1edH- is not *d-, and it's the first member of a compound
rather than a prefix.

> etc

What other words show it? I can only think of Skt. ahar/n- vs. Gmc.
*DaGaz, but in the latter the initial goes back to *dH, not *d.

> And there are a mountain of examples
> when it comes to comparing PIE with other languages.

That leads you into a morass: a "prefix" not supported by IE data is
used to justify long-distance comparisons. Mark my words: it can't end
well ;)

> Like word "day" :
> d-yew "day"
> Semitic *yaw-m "day" "to-day"
> As a matter of fact, Ju-piter has not #d-
> when dies has !!

It's *di(j)- (from the Lindeman form *di[j]e:m) vs *dj-. Old Latin still
shows traces of [dj] where Classical Latin has [j].

Piotr