Eliminating bias in experiments

[dkp at EMAIL.ARIZONA.EDU]

Thankyou A Katz,

Species-centrism strikes me as a counterproductive (but widely accepted)
bias in linguistics.

It is not clear to me how bias helps linguistics as a science...and it seems
to make it difficult for many biologists, psychologists and neurologists
to value the other claims, observations and techniques of linguistics...sad,
really.  Many of the techniques of linguistics are very clever and many observations
very telling...

Language: Language is very special...and what humans manage to do is quite
incredible. But, demonstrating that language shares some features with other
communication systems should not surprise us...we are animals too. Vervet
monkeys have different cries for eagles and jaguars.  Birds and whales produce
complex sequences composed of recombinable parts. Many animals can learn
to identify and respond appropriately to a small vocabulary of human words.
Parrots can learn to produce words, identify objects and follow commands...and
they spontaneously recombine parts of words in what appears to be a sort
of play behavior....so reference, in and of itself, doesn't seem unique to
language. And recombination, in and of itself, is not unique to language.
 But, the language ability displayed by a typical adult human being is greater
than the some of such parts....call it an "emergent" phenomenon...something
spectacular that happens when you have the right confluence of capabilities.


I have studied linguistics and animal communication systems (I got my PhD
studying Alex the parrot)...I've never met an animal that has a hope of understanding
this email...but I've met a lot of biased people on both sides of the animal
"language" issue that did not understand the people on the other side.

Theory of Mind: It is interesting that when Chomsky dismissed Skinner's explanation
of language, that somehow, (some) linguists decided that the claims of behaviorism...the
dismissal of an internal life...continued to apply to
all non-human animals. Read Donald R Griffin for a radically different view
on the notion of animal minds.

That's my 2-cents...time to go back to work...but have a great debate.

Dianne Patterson
Cognition and NeuroImaging Labs
University of Arizona


>-- Original Message --
>Date:         Thu, 5 Dec 2002 11:09:02 -0800
>Reply-To: "A. Katz" <amnfn at WELL.COM>
>From: "A. Katz" <amnfn at WELL.COM>
>Subject: Eliminating bias in experiments
>To: FUNKNET at LISTSERV.RICE.EDU
>
>
>>Date: Thu, 5 Dec 2002 17:55:26 +0000 (GMT)
>>From: Dan Everett <Dan.Everett at man.ac.uk>
>>
>>Subject: Re: Separating language from biology
>>
>>> If a child says: "I want a banana" we all assume that he knows what he's
>>> talking about. If a chimpanzee says the same thing, people ask: "Yes,
>but
>>> does he have a theory of mind?"
>>
>>Why would anyone want a 'level playing field'? I am not
>>democratic with respect to chimps.
>
>I'm sorry if my metaphor evoked images of democracy. Democracy is
>about voting, and I am not suggesting that we vote on this.
>In a scientific invesitigation, the outcome is not open to
>voting. It is determined by adhering to careful testing
>procedures. Bias must be eliminated in order for the result of
>the test to be valid. (It's kind of a shame that the idea of bias
>has been so politicized.)
>
>You don't need to have any feelings about chimpanzees one way or
>another to wish to minimize bias. If you were studying lunar
>dust, the same rules would apply. You don't want a circular
>definition, because it will not help you to investigate the
>facts.
>
>> I know that children have a
>>semantics, even if  I cannot follow or misinterpret in specific
>>cases.
>
>How do you know this? Could you prove it? What test for the
>presence of semantics is applicable that would give the correct
>result regardless of what entity it was applied to?
>
>> The child has earned its right to a charitable
>>interpretation. The chimp has not. Nor has any other species.
>
>How has the child earned this right? Are you talking about a
>specific child or all children? Are you aware that some
>anatomically normal, uninjured children can't speak or think?
>Does this include them?
>
>>A test simpler than a Turing Test is just this: do the members of the
>>species talk to one another with anything remotely showing properties of
>>the type that Hockett argued for? And do they do this 'in the wild'.
>
>What do humans do when placed in the wild from infancy? Do they
>come up with language?
>
>We can't give credit to individual members of the species for
>achievements of the species as a whole. Mozart may have achieved
>a great deal as a composer, and he was a human being,
>but you can't assume that a human chosen at random has any
>musical ability. You have to test them individually.
>
>>There is no reason for assuming a level playing field and many reasons
for
>>not doing so.
>
>We shouldn't assume anything. We should find out.
>
>
>     --Aya