IE, Genetic Data, Languages of Anatolia

Miguel Carrasquer Vidal mcv at
Fri Feb 5 17:56:09 UTC 1999

"Glen Gordon" <glengordon01 at> wrote:

>Yes, of course. My brain froze and I couldn't recall the Anatolian side
>of the cognate at all. Thanx. Anyways then, we can say with some degree
>of certainty that IE (Indo-Anatolian) speakers spoke of horses. I guess
>the contention Miguel has is whether these horses were domesticated or

The problem is whether Luw. asuwa is native or a borrowing from
Indo-Iranian (think of Kikkuli).

>I don't see why the agriculturalists that moved into Europe and who show
>up in genetic data of what's-his-name-Sforza have to be speaking
>Indo-European of all things. Indo-European just can't have been in the
>Balkans or Anatolia at such an early date. I reason that the most likely
>language candidate of these budding agriculturalists would be a language
>closely related to North-East Caucasian. As Indo-European spread across
>Europe later on, it would have wiped out almost all traces of the
>earlier non-Indo-European languages.

That's funny.  I tend think of North Caucasian (NEC/NWC) as the
primary candidate for the original language of the steppe lands.
The Northern Caucasus is a "residual zone", in Johanna Nichols'
terminology.  It contains the linguistic residue of the peoples
that were once dominant in the neighbouring "spread zone" (the
steppe).  At the outer layer we have Russian, then Mongol
(Kalmyk) and Turkic (Nogai, Karachai, Balkar etc.), then Iranian
(Ossetian), and the inner layer is formed by NWC and NEC.  This
suggests that before IE, the steppe was peopled by North
Caucasians.  And if there's indeed a genetic link between North
Caucasian, Yeniseian and Sino-Tibetan (the Sino-Caucasian
hypothesis), that's indeed what we would expect.

Likewise comparing the linguistic, archaeological and genetic
maps of Europe shows a large amount of overlap between
Indo-European, the initial agricultural expansion from Anatolia
to Holland, and the main genetic component.  Of course Renfrew
and Cavalli-Sforza are excessively naive in simplifying that to
complete equivalence.  On linguistic grounds, Proto-Indo-European
cannot be placed in Anatolia in the 8th millennium BC [although
some language ancestral to PIE (and Etruscan) can be], and on
archaeological grounds there surely was a movement from the
steppe into the Balkans around 3500 BC.  But the "steppe" or
"Kurgan" model cannot adequately explain all the archaeological
facts (cultural change, but no evidence for invasions in Northern
and Western Europe) or all the linguistic facts (why the gap
between Anatolian and the rest of IE, why Germanic).
Furthermore, history shows that steppe invadors have never
penetrated linguistically beyond Hungary and the Balkans.  Even
where steppe invasion or infiltration is the only possible
solution, as in the case of India, Indo-Aryan did not succeed in
wiping out all traces of the languages of the earlier Neolithic
population, far from it.

Miguel Carrasquer Vidal
mcv at

More information about the Indo-european mailing list