Testing Hypotheses of Language Replacement in the Caucasus with Genetic  Evidence 

Ivan Nasidze and Mark Stoneking

Max Planck Institute for Evolutionary Anthropology, Leipzig, Germany

email: nasidze@eva.mpg.de

     The Caucasus is a region of extraordinary diversity; the language
diversity in the region equals or exceeds that of the rest of Europe,
with three or four linguistic families (Turkic, Indo-European and
Caucasian; the last one is sometimes split into two independent
families) and several dozen languages are present in the region. 

     We analyzed eleven bi-allelic Y-chromosome markers and sequenced
the first hypervariable segment of the mtDNA control region in different
populations from the Caucasus and neighboring populations, representing
all major linguistic groups in the region.  Analysis of these two
genetic systems showed that the genetic relationships among Caucasus
populations primarily reflect geographic rather than linguistic
relationships; in particular, the Indo-European speaking Armenians and
Turkic-speaking Azerbaijanians are most closely related to their nearest
geographic neighbors in the Caucasus, not their linguistic neighbors
(i.e., other Indo-European/Turkic populations).  The mtDNA and
Y-chromosome evidence thus suggests that the Armenian and Azerbaijanian
languages represent instances of language replacement, possibly via
elite dominance, that had little impact on the mtDNA and Y chromosome
gene pools.  Moreover, the genetic distance based on mtDNA between pairs
of populations is significantly correlated with geographic distance,
which not only emphasizes the lack of correlation between genetic and
linguistic relationships, but also indicates that the Caucasus Mountains
have not been a significant barrier to gene flow.