> There is also Theo Vennemann's hypothesis about the Vanir >being
>what he calls "Atlantiker" -- Afroasiatic seafarers and >megalith-builders
>[...] this Atlantean hypothesis has been >circulated in serious journals
>and post-conference publications and >has acquired some ardent enthusiasts.
I suppose it wouldn't have anything to do with Basque numerals "sei" and
"zazpi" for instance, which despite Vasconicist Larry Trask's disapproval on
the Nostratic List (Linguist.org), seems to be quite Semitic in origin? Of
course, Larry Trask disapproved of such a Semitic influence on Basque mostly
because of the time period involved and went for... (please, public, do not
gasp at this frightful idea since he is often a reasonable and conservative
man)... he went for awkwardly analysing "zazpi" as coming from *botzaz-pi
thus meaning "two from five" which still has grammatical problems and
doesn't speak for the origin of accompanying "sei". These words would need
to have been borrowed in prehistoric times, before Latin influence took hold
because of the native form of the words. The same numerals however are also
found in Egyptian which is closely related to Semitic. I would propose that
the forms "sei" and "zazpi" would come from slightly earlier forms *s^es^i
and *saspi (that is, just before the Romans?)
> I think at least some of the Vanir (Freyja/Freyr, Nj�rd) >have
>names which may well be IE. Could anything beginning with Fr- >be
>Finno-Ugric anyway? Is there a single convincing Uralic etymology >for any
>of the Vanir?
This is probably a rhetorical question on Piotr's part but I'll be witness
to the fact that there are certainly no *fr- or *pr- clusters in Finno-Ugric
and what's worse, the accent was _initial_ as far back as Uralic. This means
that even perceiving such consonant clusters in borrowed PFU words would be
quite unlikely for IE-speakers.
An outside origin of these names, if justified, could only be validly sought
in NWC, NEC, Kartvelian where such "initial stop + resonant" clusters are
possible. However NWC tends to have very small words and is thus unlikely,
leaving only NEC and Kartvelian as pseudo-viable options.
There, that theory shouldn't be causing anyone anymore problems now.
- gLeN
______________________________________________________