Petusek:
> Well, the semantic motivation might be opposite. Do you think, we
> could start with *dek^- , "to reach" [...]
It's a 'reach', alright. This etymologizing is doomed for the very
same reason that we can't deny the Semitic origin of *septm. There is
no suffix *-m except for the accusative and this just doesn't make
sense on a numeral at all, especially when undeclined. I think we'll
just have to accept that *dekm can't be broken down in IE itself...
Oh-oh, but then that would cause people to consider that mesolithic
bands weren't so dumb afterall and actually did have a word for '10'
many millenia beforehand.
= gLeN