Universals and the Evans/Levinson Paper

[Salinas17 at aol.com]

To the Funklist:
I was very interested in Stephin Levinson's response to Tom's 
middle-of-the-road approach to universals.   (I'm not sure that response was intended for 
the list to see, and so I hope it's okay to refer to it here.)

Levinson wrote: "I think you are right about development as key in biology, 
and also about exceptions. But the question is can we list the strong 
tendencies?"

Of course, "strong tendencies" are not quite "universals".   Is it true 
that "once we honestly confront the diversity offered to us by the world's 
6-8000 languages" (as the abstract says) that all we can hope to find are strong 
tendencies?

And is it true that "Linguistic diversity then becomes the crucial datum 
for cognitive science: we are the only species with a communication system 
which is fundamentally variable at all levels" (as the abstract says)?   

The abstract allows for "stable engineering solutions satisfying multiple 
design constraints, reflecting both cultural-historical factors and the 
constraints of human cognition" -- but I take it these don't qualify as 
"universals".

I am extremely sympathetic to the basic points made in the paper, but I'd 
question whether these statements are going too far.   

One of the great difficulties that cognitive studies have run into is 
describing cogitive processes in a way that also describes what should be going 
on in the brain.   Forget human cognition or language.   I previously quoted 
Heinrich and Bugnyar on the paper wasp: ""By some process that still remains 
one of the great unsolved mysteries of biology, exquisitely precise 
behaviors can be genetically programmed in animals with brains no larger than a 
pinhead..."   If we have problems understanding how a tiny brain can process an 
insect's programmed behavior, it should not be a surprise that we have 
difficulty with a much larger brain and   behavior as bewilderingly complex as 
human language.

Cognitivists need to know that human language is diverse and that they are 
not looking for a discrete syntactic template that guides all human 
language.   And biologists should know that therefore simple equivalencies will not 
be likely in the brain.

But there are some things that may be "universal" or "strong tendencies" in 
the "the world's 6-8000 languages" the abstract mentions.   And they may be 
very useful to keep in mind when we talk about "cognition".

Number one, all of these languages would appear to be communal.   At least, 
I presume there is not one instance of a language spoken by a single 
individual and no one else.   Chomsky's "self-expression" function of language has 
not yielded any instances of private languages that I am aware of and I 
presume none of the thousands of languages the authors studied fell into this 
category.

So, no, it's not complete diversity.   It's not "every snowflake is unique" 
diversity.   It's not a hundred million language diversity or even close to 
the genetic diversity of a typical human population.   

This is very important because it should signal to the cognitivist that the 
nature of language may not   be found in only studying individuals or 
individual "cognitions."   If language is shaped communally, then the processes 
are communal.   I'm not speaking of extralinguistic "social" factors,.   I'm 
talking about the structure of language itself, which certainly does not 
exist as a whole in any one individual.

If you are looking for an analogy, look to the latest developments in 
computer networking, where such concepts as "cloud" are emerging.

Here's a stab at another universal -- the abstract mentions "meaning" as an 
example of diversity.   But meaning cannot be that diverse, can it?   There 
has to be some sense that speakers of these languages were not quite free 
to use words completely as they choose, no matter how grammatical or 
syntactical.   

The "strong tendency" here would be COMMON REFERENCE.   No matter what 
language we look at, we can expect all its speakers to strive to refer to the 
same things or circumstances with the same sounds (or symbols) in order to be 
understood.

This isn't trivial.   What constraint could be stronger than a need for a 
common meaning.   Tense, case, word order -- all aspect of language in 
whatever form or morphology, they are all forms of references -- and they all most 
basically depend on the speaker and listener having a common sense of what 
is being spoken about.   All vocabulary and grammatical variations of any 
kind make reference to an object, an action a process, a relationship in place 
and time with greater or lesser specificity.   If my reference is different 
than your reference, we do not understand what we are saying.   Which of 
the 6-800 languages consisted of speakers who did not understand one another?

Esa's example of Tamil's diachronic changes in endings tells us not that 
Tamil suddenly became incomprehensible to its users, but instead that it 
changed to stay comprehensible to its users.   

If this makes any sense, then I don't see why we can't say that the need 
for common reference is a "strong tendency." As close as we can get to a 
universal.   And one that should be a great help in telling linguistics as well 
as the cognitive science why languages are structured as they are.

Regards,
steve long




















**************
Big savings on Dell’s most popular laptops. Now starting at 
$449! 
(http://pr.atwola.com/promoclk/100126575x1221827510x1201399090/aol?redir=http:%2F%2Fad.doubleclick.net%2Fclk%3B214663377%3B36502382%3Bh)