Lynx (was Re: Uralic and IE)

[Glen Gordon]

ROBERT WHITING:
  Indeed it was meant as an example, but it is actually the only
  possible example.  Flash Gordon said that there were no IE stems
  in -(n)k and Miguel replied "Not sure.  There is in Greek
  (lynx)."  So to prove the point there has to be an example in
  Greek that (1) is inherited from PIE and (2) preserves an -nk
  stem that other IE languages have lost.

I didn't mean that there were no STEMS that ended in *-nk-, just to
clarify. I said that such stems always have a suffix after them so
that there is no final *-nk ever in IE. In the case of *lunk-, it ends
in *-s in the nominative, hence *lunks, or *-m in the accusative, etc.
but there's never a complete word like **lunk with a final *-k.

Of course, if other IE languages have lost this *-k ending a la
Miguel, it'll certainly be difficult to prove its existance - Flash
Gordon empathizes. If even Greek has -x (/-ks/) then still, how does
this connect with Miguel's neuter **-k? Miguel has alot to figure out
just yet.

ROBERT WHITING:
  So getting back to the two criteria that could prove a survival of
  PIE -nk stems in Greek, there is little doubt that lynx is a PIE
  word as it occurs in practically all of the main IE branches (see
  Gamkrelidze and Ivanov 1995, p. 431, 2.1.5.1).  However, it is also
  generally connected with the PIE root *leuk[h]/*luk[h] (ibid.) and
  I don't see any way to get a consonantal n into this root.

The n-affix. Thus from a verb *lu:- (or *leu-, if you wish) we get
*lu-n-k- where *-k too is an affix built on the core verb *lu:- but
I'm unsure of the etymology of *lunks myself. I mean, if it's actually
built on *leuk- "to shine", that makes no sense to me. Yet, *leu-
means "to lose", no? Still doesn't make sense. At any rate, I doubt
that -nx is meant to be "expressive", in this case at least, and Greek
<lynx> is hardly an example of anything in re of this **-k.

--------------------------------------------
Glen Gordon
glengordon01 at hotmail.com