Re: Feasibility of Long-Range Reconstructions (was: PIE-Arabic Corre

----- Original Message -----

From: Richard Wordingham

To: cybalist@yahoogroups.com

Sent: Sunday, January 20, 2008 11:43 PM

Subject: [Courrier indésirable] [tied] Feasibility of Long-Range Reconstructions (was: PIE-Arabic Correspondences)

--- In cybalist@... s.com, "fournet.arnaud" <fournet.arnaud@ ...>
wrote:
> Brian M. Scott wrote:

I would add that it is clear that evidence of linguistic relationships will eventually be swamped by the noise introduced by random changes. And on the evidence to date, this noise accumulates more than fast enough to make any attempt to reconstruct a 'proto-world' language an exercise in crackpottery. 

Even just securely identifying a few odd traces of one is highly unlikely: even if such traces still  exist, odds are that it's impossible to distinguish them
>> from false positives.
=========
> The "can't be done" dogma is **useless** Arnaud

========

I wouldn't deduce that 'it can't be done' from Brian's argument.

Richard

=========

I think it is explicitly stated that *it's impossible* and that *any attempt is an exercise in crackpottery*.

Nothing needs be deduced.

(I disagree)

Arnaud

==========