Re: Indo-European family tree (was Re: Celtic and Afro-Asiatic?)

On Thu, 22 Sep 2005 16:15:24 -0400, Paul Bennett <paul-bennett@...>
wrote:

>"Some of the ethnic groups were not native to the area they inhabited",
>which seems self-evident to me. How many generations of continuous
>inhabitation are required to establish nativeness, or must one merely be
>first to populate an area?
I think a fair rule of thumb would be to assume that the first group
documented in a location, whether by archaeological or historical
evidence, constitutes the autocthonous population.

I'm not sure what linguistic, archaeological, or historical evidence
exists to suggest that the Urartians, the Hurrians, or the Hattic speakers
were not autocthonous to Anatolia.  There is, however, evidence that the
Hittites, and later the Armenians, were relative newcomers.  Furthermore,
FWIW, a substantial portion of the placenames in western and south-central
Anatolia are unanalyzable as Indo-European.  Mallory compares Parnassos
(the root of which is allegedly parna, "house") with Hurrian purni.
Furthermore, the Hittites and the Luwians borrowed a number of terms
related to the material culture of the region, and other terms which have
been reconstructed for PIE (such as the aforementioned wheeled vehicles
and horses) were not introduced to Anatolia until the fourth millennium or
even later.

Renfrew was trying to propose a simple model of Indo-European migration,
which could be observed through the archaeological record.  If we accept
his theory, that Indo-Europeans were indigenous to Anatolia, and also
accept (with most linguists) that the only attested Indo-European
languages spoken there, such as Hittite, Luwian, and Armenian, were
latecomers to Anatolia, we must assume that they had migrated into
formerly Indo-European territories from which speakers of otherwise
unattested Indo-European languages had been displaced, which seriously
complicates his model.  I'm not sure what justification we have for
adopting such a complicated model.