Re: Celtic Tanarus and Gmc Thunaraz

On 2009-03-25 13:18, A. wrote:

> Thus Matasovic follows Pokorny and Watkins in saying that PIE *(s)ten-r-
> is the origin of Thunder/Donner ; and yet Matasovic disagrees with
> Pokorny by staing that the Celtic forms do not derive from a distinct
> PIE root (*tor-) but rather from a metathesis of *tonaro- > *torano-.
>
> On the other hand, the IE database of Nikolaev and Starostin, gives
> Proto-IE: *taron- <PIH *-rH-> as the origin of BOTH the Celtic Taran and
> the Gmc Thunr/Thunraz; as well as the Hittite Tarhun and other derivatives.
>
> Any ideas on whether Taran and Thunder share a common IE root
> (whether that root be *taron- [per Nikolaev and Starostin] or (s)tene-
> [per Matasovic] )?

The PCl. form is reconstructed as *torano-(~ *-i-, *-u-, *-a:-), with
the Gallo-Brittonic assimilation or *-oRa- > -aRa-, a process that seems
to have remained productive for some time, cf. Lat. monachus --> W.
manach. The underlying root may be *terh1- 'pierce, bore' (the usual
derivation), but Ranko's *tonaro- > *torano- looks good to me,
especially as regards its semantics. The root in question is *(s)tenh2-.
Germanic *þunra- is from *tn.h2-ró-, which cannot be the source of the
Celtic words. The vocalism *ton[h2]-ro- looks aberrant and one would
not expect a vocalised laryngeal (*&2 > *a) in it. The Gaulish and
Brittonic forms could best be analysed as *ten&2-ro- > *tanaro-
(Joseph's Law) > *tarano-, but that doesn't account for OIr. torann. I
don't know if there is a possible explanation of it within Irish.

Piotr