Res: [tied] Latinus geminus

On Sat, 23 Feb 2008 23:50:28 -0000, "alexandru_mg3"
<alexandru_mg3@ yahoo.com> wrote:

>But if I add here the Baltic form reflected in Lettish jumis 'a
>double ear of corn' (and Jumis - the God) => following the same path
>that : "semantics usually takes preference over phonetics" , I think
>that the phonetic puzzle will be complete
>
>In this case I need to put away : Lettish jumis as from *yeuh- 'to
>join , to mingle' => but the semantism of 'the duality' somehow will
>dissapear ...

Latvian jumis (with short /u/) would be from *yeu-. There
are other Baltic forms which *are* based on the root *yeuH-
(as well as on the root *yeug-).

>And once oing this why not to put together regularly : Latin geminus
>with Skt. yam'a- as from the root

g^em- ?

That means exchanging a Latin irregularity (gem- for
expected *iem-) for a Sakt. irregularity (yam- for expected
*jam-).

>Doing all these : only Celtic *yem-o-no would remain isolated.
>
>Viewing this: What is your 'preference' here ?
>
>You would try to put 'all together' or to keep outside some?

I really haven't got an opinion at this point. Pokorny lists
some other forms (Germanic, and others) under *yem-. I've
always liked the possibility to connect *yem- with *nem- and
*em- "to take".

============ ========= ==
Miguel Carrasquer Vidal
miguelc@... nl