Twenty.

Piotr:

>The form *dwi- is of uncertain origin -- either analogical to *tri- >or
>produced by some archaic alternation like *du-o-/*du-i- (cf. >kW-o-/*kW-i-
>in pronouns).

I choose "archaic alternation". The alternation would simply be caused by
stressed and non-stressed forms existing side-by-side. These alternations
exist in the pronouns, demonstratives and the simple numerals. In Early IE,
it would be *t:We (unstressed) versus *t:Wei (stressed) for "two" and
similarly *kWe versus *kWei for "who".

>*wi(:)k^mti: possibly resulted from a conflation of two types of >decad
>formation coexisting in PIE: the noun phrase "two tens" >*du-o-ih1
>d(e)k^mt-ih1 (*-ih1 is the dual ending of PIE neuters) and >the compound
>"2-decad" *wi-(d)k^omt- (with a neuter collective as the >second element).
>Similarly for other decads: *tri-h2 d(e)k^mt-ah2 or >*tri-(d)k^omt-, with a
>variety of mixed forms.

I agree... except I'd prefer *dwoiH1 k^ntiH1 versus *wi-k^ont-, and *trix
k^ntax versus tri-k^ont-, without the yucky unattested *-d-'s if you don't
mind.

- gLeN



_________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com