Re: [tied] Re: searching for common words for all today's languages

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

From: tgpedersen

To: cybalist@yahoogroups.com

Sent: Monday, February 06, 2006 5:09 AM

Subject: [tied] Re: searching for common words for all today's languages

 

<snip>

> Ruhlen and Greenberg use the method of "mass comparison" which is a
> load of crap. It consists of looking at lists of words from various
> languages and picking out lookalikes of vaguely similar meaning. It
> has two (well, at least two) fatal flaws. First, there's no
mechanism
> to distinguish inherited words from borrowings. Second, it's easy
to
> miss seeing cognate words that don't look alike (like English I and
> Latin ego).


Which means the method is partially crap.


Torsten

***

Patrick:

What is "crap" is when someone is unwilling to find out and accept at face value what someone else has explicitly announced as his objective — and then insists on RE-DEFINING the objective, and then faulting the that "someone" for not accomplishing the RE-DEFINED objective.

 

Torsten, if you had ever even read this chapter in Ruhlen's book, you would know that they acknowledge the deficiencies of mass comparison that you mention above. They are unimportant for the objective they have set for themselves, which is to provide a provisional framework for classification, through which putative language families can be preliminarily identified and constituent languages grouped, preparatory to the rigorous application of historical linguistic procedures.

 

No different from the _first_ steps taken on the path to PIE reconstruction!!!

 

"Latin 'pater'; German 'Vater'; hmmm! kinda similar. Gadzooks, maybe they worth investigating as 'related' in some organized way." 

***