On 2008-04-15 10:26, fournet.arnaud wrote:
> > And why should this Celto-Germanic *marko- have to be connected with
> > <mori> etc.? A vague similarity involving short words with commonly
> > occurring consonants is hardly compelling.
> > > Piotr
> ================
> I suppose this statement applies to 80% of IE cognates.
>
> Arnaud
> ==============
That's why they would mean nothing *as isolated similarities*. Their
evidential value depends on the existence of systematic correspondences
pervading the lexicon, derivational morphology and inflectional system.
It isn't a priori impossible that a Mongolian word spread westwards,
passed along by the Scythians, Thracians etc., and ending up in
Proto-Germanic and Celtic as a word for a 'saddle-horse'. But as an
etymology, it's just a shot in the dark without any supporting evidence.
Where is the attestation of *mar(ko)- in Iranian or Thracian? Who, when
and why extended it with a suffix?
Piotr