On 2006-08-15 03:09, tgpedersen wrote:
> It's interesting BTW that ON taka belongs to
> class VI since a stem *T-T in Germanic can't be
> directly from PIE (where it would be **D-D,
> violating root constraints).
Germanic *tak-/*te:k- (Goth. reduplicating tekan, taítok, tekans
'touch') is probably somehow related to Lat. tango: (tetigi:), PToch.
*täk- 'touch' and the isolated Gk. aor.part. tetagó:n 'having seized'.
See Kortlandt's analysis of the set:
https://openaccess.leidenuniv.nl/bitstream/1887/1939/1/344_123.pdf
... with which I only partly agree. I'd reconstruct the root as *teg-,
with the *g colouring the vowel to *tag- and producing the minor ablaut
alternation zero ~ *a ~ *e:. The initial *t in Germanic may be due to
voicing assimilation in reduplicated forms: *te-tg- > *tedg- > *dedg-
(like Lat. bibo:) -- here I agree with Kortlandt. I also agree with his
explanation of OE þaccian 'pat' as a related form produced by Kluge's
Law; *tag-ná:- > *þakko:-.
Piotr