Re: HORSA vs. EXWA

--- In cybalist@yahoogroups.com, "Tavi" <oalexandre@...> wrote:

>
> I'm afraid there's no such "thorny cluster" here. As I've explained in
> another post, the geminate velar of Hittite hartagga- must be a suffix
> like the one found in the Turkic cognate qarsaq 'steppe fox'. With the
> help of external data, I reconstruct an affricate sibilant (possibly
> alveolo-palatal) in the internal cluster *rC.
>

My own reconstruction is *XrC-o-, with *C being an affricate sibilant.

> > If an item has Anatolian and numerous extra-Anatolian cognates, it can
> > safely be labelled as PIE (= the latest common ancestor of the
> > family). As simple as that. There is no earthly reason to classify the word as
> > 'recent'.
>
> Sorry, but I disagree with the traditional model. I must insist that not
> all the words labelled as "PIE" are equally older not they belong to the
> same (proto-)language.
>

You might redefine "PIE" to designate the most recent lexicon layer in the IE family, but then you should create new names for the older strata.

> > One does not even have to insist that *h1ek^wos referred originally to
> > domesticated horses. Wild horses were very common throughout Eurasia
> > and they may have kept their name after domestication.
>
> Only that there's no actual evidence this domestication was done by IE
> speakers.
>

My theory is that *h1ek´w-o- specifically designated the domesticated horses from the Pontic-Caspian steppes. As in the case of other domesticated animals, this is a loanword which originated in the language of the domesticators.

Of course, as horses were also domesticated in more than just one place, there're also other 'horse' words. For example, I've already suggested that Germanic *xurs-a-/*xruss-a- must have coined by people familiar with the wild animal.