Re: [tied] Reconstructing a future language

From: P&G
Message: 18396
Date: 2003-02-03

>why does PIE falsely share these innovations wiith
> the Romance languages?

Is it all possible that the languages included are skewed? If we include
Portuguese Spanish Catalan French Provencal Occitan Italian Sardinian
Sicilian Romanian Rhaetish and German, might there not be a slight skew
towards the Romance langauges?

If we use Icelandic, Old Iclandic, Norse, Old Norse, Swedish, Frisian,
Yiddish, Plattdeutsch Schwiezertuetsch ... oh I'm bored with this. You get
the picture. An algorithm which works entirely by numerical analysis has no
way of assessing the variety, independence, and quality of its input. And
besides, which version of PIE were they using? A Hittite-inclusive model, or
a Sanksrit-Greek model?