6.90 Comparative Method

[The Linguist List]

----------------------------------------------------------------------
LINGUIST List:  Vol-6-90. Sun 22 Jan 1995. ISSN: 1068-4875. Lines: 77
 
Subject: 6.90 Comparative Method
 
Moderators: Anthony Rodrigues Aristar: Texas A&M U. <aristar at tam2000.tamu.edu>
            Helen Dry: Eastern Michigan U. <hdry at emunix.emich.edu>
 
Asst. Editors: Ron Reck <rreck at emunix.emich.edu>
               Ann Dizdar <dizdar at tam2000.tamu.edu>
               Ljuba Veselinova <lveselin at emunix.emich.edu>
 
-------------------------Directory-------------------------------------
1)
Date: Fri, 20 Jan 1995 20:43:10 +0100
From: KNAPPEN at VKPMZD.kph.Uni-Mainz.DE
Subject: Re: 6.83 Comparative Method
 
2)
Date: Fri, 20 Jan 95 22:11:56 EST
From: amr at ares.cs.wayne.edu
Subject: Comparative: N-ary vs. Binary Comparison
 
-------------------------Messages--------------------------------------
1)
Date: Fri, 20 Jan 1995 20:43:10 +0100
From: KNAPPEN at VKPMZD.kph.Uni-Mainz.DE
Subject: Re: 6.83 Comparative Method
 
Should we introduce a sharper terminology? I think, a typical Greenbergian
`family' is quite different from the indoeuropean `family' or the germanic
`family' in time depth and the possibility of constructing a
proto-language.
   Biologists have developped for a long time a whole set of notions, from
species and genus  to families, classes and orders. It seems, that such
kind of differentiated terminology would do something good to historical
linguistics, too.
 
 --J"org Knappen.
 
--------------------------------------------------------------------------
2)
Date: Fri, 20 Jan 95 22:11:56 EST
From: amr at ares.cs.wayne.edu
Subject: Comparative: N-ary vs. Binary Comparison
 
I was really surprised to read Jacques Guy's declaration
that n-ary comparison is just repeated binary comparison,
and not to read any objections from anybody else.
 
There are many important differences, which mostly
indicate that n-ary comparison is a much superior
strategy for doing comparative linguistics.
 
The bigger the n, the bigger the chance that we will
recover more of the proto-language (e.g., more of
the vocabulary).
 
The bigger the n, the less the chance that we will be
misled by a spurious set of correspondences.
 
In some situations, a smallish value for n may not
offer these advantages (so that in certain special
cirucmstances ternary comparison, say, may be worse
than binary), but in general n-ary is better.
 
Nor surprising then that in work on such families
as Indo-European, Uto-Aztecan, Afro-Asiatic, and so on,
no one to my knowledge has ever proceeded on binary
basis, comparing for example every pair of IE
languages.
 
Alexis MR
 
--------------------------------------------------------------------------
LINGUIST List: Vol-6-90.