Concerning WALS

[Salinas17 at aol.com]

In a message dated 11/10/08 9:22:26 AM, dryer at buffalo.edu writes:
> Nor should we accept the generative linguists' view of work in typology 
> that
> characterizes what languages are like as atheoretical.  Too many typologists 
> and
> functionalists seem to accept that view.
> 
Typology classically is a matter of categorizing.   There are many different 
ways to categorize any complex set of raw data.   Implicit in most choices is 
theory.

For the moment, let's hypothesize that the main "function" of language is 
common reference. From there, we would theorize that all structure -- from 
phonology to lexicon to syntax -- serves to seek common reference -- to disambiguate 
between what the speaker is referring to and what the listener understands.  
Then, the World Atlas of Language Structures becomes a catalog of approaches 
to that objective, disambiguation in communication -- attempts towards a more 
accurate common reference between speakers and listeners.   The research would 
be valuable in assessing why those attempts differed.

This is quite a different view than generativist theory would have of that 
Atlas.   It's also a different view than say a traditional Indo_europeanist 
would have -- something that Greenberg was certainly conscious of. 

Admitting that describing "what languages are like" involves theory from the 
start helps and does not hurt the process.   Otherwise we have hidden 
assumptions in the descriptions that must be pryed out rather than given up front.


Steve Long
 



















**************
AOL Search: Your one stop for directions, recipes and all other 
Holiday needs. Search Now. 
(http://pr.atwola.com/promoclk/100000075x1212792382x1200798498/aol?redir=http://searchblog.aol.com/2008/11/04/happy-holidays-from
-aol-search/?ncid=emlcntussear00000001)