Re: Number nine

On 2008-01-16 21:00, afyangh wrote:

> I have looked at North-caucasian
> and the root for "nine" displays some oddities :
> - it should be a verb,
> - it should be *-w_H2-
>
> Because it is a verb
> it can "generate"
> - n_w_H2- : Cf. PIE now-m
> - d_w_H2- (Cf. Slavic) *devjantj

How does a verb generate such things? What evidence have you got for the
*h2 in any IE language? In Slavic and Baltic, the d- is due to the
pairwise assimilation of numerals in serial counting (*deveNtI,
*deseNtI). The final *-m is not warranted either; Baltic points to *-n.

> The lost of H2 in *now-m might be
> a kind of "Saussure-Hirt" law.

There was no PIE "*nowm."; an *o appears only in those languages where
*e is regularly rounded by a following *w.

> but
> When one looks at some words in IE,
> some words are "strange" :
> Celtic : forms with -a- in Breton nao and Welsh

This is an inner Brittonic development: *newn. > Celt. *nowan > Britt.
*nawan by assimilation.

> Greek : enn- with two -n- from *enH2- ?

There's no such change in Greek. The origin of <en-> is unclear. It
looks like some obscured prefixed element. The oft-posited *h1newn.
doesn't explain the geminate.

> I 'd like some insight upon these forms.
> Do we have Mycenian Greek for "9" ?

Yes, <e-ne-wo>, as expected

> What kind of "verb" could become "9" ?

There may be a connection (but a very uncertain one) with the adverb *nu
and the adjective *newo-, if 'nine' was a "new" number after two groups
of four ('eight' clearly has a dual ending).

Piotr