bob keeter

There was a very interesting article in this mornings Boston Globe
(apparently not on line yet or I would have included a URL) covering the
linguistic ****yses of Russell Gray, a biologist turned linguist. He was
using some of the same "correlation algorithms" and other mathematical tools
used in credible DNA research to trace the expansion of the "mother
language" across Europe and central Asia. Guess the language tree seems to
spread in consonance with the spread of organized farming (as opposed to the
Kirgun invasion).

A couple of interesting points:

1. This research points to a major influx of language and farming out of
Anatolia somewhere between 7800 and 9800 years ago.

2. The new, rigorously mathematical, techniques apparently accomodated the
time-variant rate of change in the language development that were evident
from Swadesh's attempts at "glottochronology", even when butted against the
kinds of historical and archaeological evidence that SOME self-proclaimed
DNA experts would prefer to ignore! 8-)

The original paper is published in the current "Nature" journal:

http://www.nature.com/nsu/031124/031124-6.html.

Regards
bk

cai

Facetiously Swadesh's equivalent of the molecular clock is called
'glottoclock' while in their Nature paper Gray & Atkinson tried to base
the tree topology on a harder ground using a bayesian ****yses =!-)

Michael

bob keeter

Ain't it just amazing how such an obscure, sparce-matrix logic process
should just keep showing up in these kinds of things (and how some would
choose to rail so loudly against it! ;-) )

Mathematics seems to be such a demanding domain for certain kinds of
"science"!

;-)


Take care
bk