From: Gerry Reinhart-Waller
Message: 1857
Date: 2000-03-13
>Gerry here: Yes, Arutiunov concurs with you that there is no
> Gerry:
> >Then what is the Hyperborean family? Is it not real?
> >And about Gilyak -- where does it fit if it is not Boreal?
>
> I don't think that Hyperborean could be real but then I'm only conjecturing.
> Based on the scanty stuff I know on Gilyak, I would still say that it is
> either more closely related to Altaic or that it is part of a seperate
> branch all its own, perhaps even the most ancient branch of Steppe.
>
> According to Zompist (http://www.zompist.com/asia.htm), the Gilyak numbers
> are the following (together with my Steppe reconstruction):
>
> Gilyak Steppe... or Pre-Pre-Pre-IE :)
> 1 ny(i)- *t:u
> 2 m(i/e)- *t:ui
> 3 c(e)- *gul
> 4 n(@)- *nil
> 5 tho- *ut:u
> 6 ngax- *ru
> 7 ngamg *ral
> 8 minr *munri (?)
> 9 nyenyben **t:u-bi (kum-ta)
> 10 nyandorng *t:u-kum
>
> First off, I am certain I saw /mxo/ for "ten" before but then again, I could
> very well be losing my mind... (get the ants off me!)
>
> The word for "four" has good connections with Eurasiatic *nil "four" (Uralic
> *nelja, Altaic *nu"lu" (<-my Korean-included reconstruction), Dravidian
> *na:l). As well, we find a version of Steppe *ut:u "five" in Gilyak tho-
> (Uralic *witte, Altaic *utu > Old Japanese itu-tu).
>
> What gets weird about "eight" is that it looks eerily like my reconstruction
> for DeneCaucasian *mnrit (Basque bederatzi "9", SinoT *bryat)... Now the
> problem is that although I think that Nostratic is a DC language, I can't
> recover "eight" from Steppe because all the languages in Steppe other than
> Gilyak like to create compound words, usually something like "(five) and
> three" or "two from (ten)". The form wouldn't be from SinoDene or its
> daughter languages because we should find *mRy@:t with the distinctive
> syllabic contraction.
>
> Is it possible that *mnrit has survived in Steppe as *munri? Is it connected
> with forms in AfroAsiatic with a prefixed *s- (Coptic s^moun) or Dravidian's
> lack-of-initial-*m form? One would imagine that *munri was an ancient Steppe
> numeral which had been later lost through use of derivative forms like IE's
> corrupted dual form of "four". This would justify Gilyak as being a most
> ancient branch of Steppe and hence very seperate from Boreal or
> ChuckchiKamchatkan.
>
> Just a thought. Carry on.
>
> - gLeN
>