Re: Automatic clustering of languages

From: Francesco Brighenti
Message: 48227
Date: 2007-04-04

--- In cybalist@yahoogroups.com, "mkelkar2003" <swatimkelkar@...>
wrote:

> That [i.e., the schemes of "transliterated" words presented by
> Batagelj, Kerz^ic^ & Pisanski in their study at
> http://tinyurl.com/3exccb%5d *is* the correct way to classify with
> complete objectivity.

No, it isn't. Complete objectivity would require words being
scientifically transliterated with the use of the symbols of the
International Phonetic Alphabet (IPA). *That* would be the only way
to compare words taken from so diverse language families. On the
contrary, the Slovenians were only able to present dozens of flawed
transliterations, which choice undermines a priori the results of
their study as well as those of any future study based on this same
methodology, whose adoption "on a much greater scale" by comparative
linguists they recommend (!) in their paper.

Regards,
Francesco