Re: [tied] Re: -st

On 04-11-05 13:43, tgpedersen wrote:

> I think there's a distinction here that has slipped by you: Kuhn
> distinguished between two languages: a non-IE Nordwestblock (also
> called the ar-/ur- language) and an IE one which replaced for a short
> time only before replaced itself by Germanic; actually only the
> latter one deserves the "Nordwestblock" label, the former extended,
> based on placenames far out to the east where traces become sparse.

Well, here we have a transparent IE formation, *ni- 'under' + *-sd-o-,
with the same pattern of compositional reduction as in, say, Skt.
upa-bda- (< *upo- + -pd-o-). The distribution of the word covers _most_
of the known branches of IE, east and west and south and north. There's
nothing aberrant in *nisdo- from the point of view of IE phonology and
morphology. The word refers to something all humans are familiar with,
not a new invention or a local phenomenon. Why should anyone want to
derive it from an essentially enigmatic non-IE source? You really seem
to be fond of explanations "per obscurius".

Piotr