>as "the running animal", maybe applied to some kind of new horse, faster than the original Germanic horses, *exwaz
> *xorsa- "horse" <*kers- "to run", so CGermanic can be explained
>I'm afraid you've made two unwarranted assumptions. The first one is that the original meaning of *kors-
> > was 'to run' and not 'horse', andthe second one is that the original Germanic word for 'horse' was *exwa-z. > > IMHO none of them is correct.
>I think late IE *h1ek´w-o- 'horse' is a Caucasian loanword reflecting the domestication of the horse in the Pontic-Caspian steppes. However, pre-Germanic had already *kurs-, which originally must have designated the wild horse, hence its Vasco-Caucasian cognates (of course, I don't think the Spanish and Portuguese 'roe deer' comes from Latin).
> Proto-Germanic *exwaz has many IE cognates with the same meaning, so it's plausible that PIE was *ek^wos, while *xorsa- has many cognates, but none with the exact meaning of "horse" (cf. Latin currere, Celtic *carros). The relationship to Portuguese corço "roe deer" and corça "hind" (<*curtius, *curcius) sounds likely, but its relation to *kers- is obscure. Could we assume that *xorsa- was the wild horse and *exwa- the tamed one, or the opposite?