sauvage noble

Digital Medievalist alerts us to a message posted by Joseph Eska, an Indo-Europeanist specilizing in Celtic, to the Old Irish listserv:

     Those of you who have been on the Continental Celtic or Old Irish list for more than 15 months or so will recall that much excitement was generated early last summer by an article in the Proceedings of the National Academy of Sciences by Peter Forster and Alfred Toth on the genetic relationships of the Celtic languages and the absolute dating of the Celtic family tree. I mentioned back then that Don Ringe and I were composing a reply to their article. For those of you who are interested in the subject, I'd like to inform you that our reply has now appeared in the journal _Language_, vol. 80, no. 3, 2004, pgg.569–582.

There was indeed quite a bit of buzz in the biz on LinguistList, sparked off by a now-archived-and-to-be paid-for New York Times article, unfortunately titled “Celtic Found to Have Ancient Roots” (duh!). I collected the relevant messages:

• 14.1825
  Tue Jul 1 2003
  Media: NYT: Celtic Found to Have Ancient Roots
  Anthony Aristar
  —first announcement of the NYT article to the list
• 14.1876
  Mon Jul 7 2003
  Disc: New: Re 'Celtic Found to Have Ancient Roots'
  Larry Trask
  —a lively and devastating analysis of the PNAS article
• 14.1899
  Wed Jul 9 2003
  Disc: Re 'Celtic Found to Have Ancient Roots'
  Peter T. Daniels
  —references to earlier disagreement regarding Tocharian
• 14.1908
  Thu Jul 10 2003
  Steve Hewitt*; William J. Poser†
  *—reference to a now-archived article on the PNAS paper in Le Figaro
  †—reasons why Forster/Toth was published where it was
• 14.2012
  Fri Jul 25 2003
  Paul Purdom*; Peter Forster†
  *—playing devil’s advocate
  †—Forster himself responds
• 14.2016
  Sat Jul 26 2003
  Pier Marco Bertinetto
  —what we have here is a failure to communicate: mastering the linguistic data
• 14.2022
  Sun Jul 27 2003
  Sally Thomason
  —more on mastering the linguistic data
• 14.2034
  Tue Jul 29 2003
  Larry Trask
  —fights back
• 14.2040
  Wed Jul 30 2003
  Larry Trask
  —and back
• 14.2050
  Thu Jul 31 2003
  Robert Orr
  —more on Forster/Toth playing fast and loose with the linguistic data
• 14.3055
  Mon Nov 10 2003
  Media: New Scientist Review of Forster and Toth
  Marc Liberman
  —reference to “a badly-informed and credulous” review of Forster/Toth in New Scientist (cf. LanguageLog with further links)
• 14.3096
  Wed Nov 12 2003
  Media: Modified Issue 14.3055: Forster/Toth Review
  Marc Liberman
  —correction: American Scientist

It all boils down to this: classification of languages using methods from other sciences is to be welcomed, but it must take the linguistic data into account. I’d be interested to read what Joe and Don have to say.