On 2008-01-26 07:25, mkelkar2003 wrote:
> The comparative method cannot be used as evidence for population movements.
As has been emphasised here many times, the purpose of the comparative
method is to demonstrate genetic relationship between two or more
languages and to enable a partial reconstruction of their common
ancestor(s). Full stop/period. It has practically nothing to do with
populations and their movements but only with language-internal
phenomena. Sometimes it can be used _in combination with other evidence_
to evaluate alternative scenarios of spread and dispersal (since it
stands to reason that related languages were initially spoken close to
each other, and that e.g. massive borrowings and areal convergences
point to a particular geographical configuration). Lyle Campbell is
absolutely right -- but what he criticises is not the comparative method
itself (he himself is one of its foremost practitioners in the fields of
American Indian and Uralic linguistics), only its illegitimate use.
Piotr