Re: HORSA vs. EXWA

--- In cybalist@yahoogroups.com, Piotr Gasiorowski <gpiotr@...> wrote:

>
> > Interestingly enough, one of the authors is Q.D. Atkinson, co-author
> > with R.D. Gray of the famous glottochronological tree of the IE

family

> > first published in 2003.
>
> They did the best they could using this kind of Bayesian phylogenetic
> analysis. At least they avoided many of the usual pitfalls and did not
> assume constant rates of replacement, so their method is *not*
> glottochronology in the traditional sense.
>

But it's **still** glottochronology as they put absolute dates to tree
clades. But I don't think absolute dating can be done without the help
of extra-linguistic data (archaeology, paleoclimatology).

> > Unfortunately, language evolution implies SEMANTIC SHIFT
> > <http://en.wikipedia.org/wiki/Semantic_change>, which can't be

measured

> > by this kind of studies, because the fixed variables are the

meanings

> > themselves, not words.
>
> A straw man argument. You should read the article first.
>

No, thanks. Besides, it's on a pay-per-view basis. Instead, I'd
recommend you these FREE articles on measuring semantic shift:

http://www.aclweb.org/anthology/W/W11/W11-0134.pdf

http://edoc.hu-berlin.de/conferences/qitl-4/sanchez-marco-cristina-79/PD\
F/sanchez-marco.pdf