[ExI] Meaningless Symbols.

[Eric Messick]

Damien writes:
>On 1/15/2010 1:38 PM, Eric Messick wrote:
>
>> Sounds like you've got a problem with behaviorist descriptions.  Can
>> you explain?
>
>I don't have to. Chomsky did it in 1959 when he killed Skinner with a 
>single review.
>
>[reprinted http://www.chomsky.info/articles/1967----.htm ]

I haven't seen the book Chomsky is reviewing, and I only skimmed the
review, but I saw no reason to disagree with Chomsky.

Chomsky points out that Skinner is trying to stretch the results of
simple scientific experiments on rats and pigeons to cover linguistic
activity of humans, and that Skinner looses specificity by doing this.

Chomsky also gives the impression that Skinner is ignoring much of the
internal state of a complex organism, again resulting in a vague lack
of specificity:

Chomsky: 
      One would naturally expect that prediction of the behavior of
      a complex organism (or machine) would require, in addition to
      information about external stimulation, knowledge of the
      internal structure of the organism, the ways in which it
      processes input information and organizes its own behavior.

I'm not sure how my discussion of a CPU adder understanding the
difference between zeroness and oneness prompted your remark about
behaviorism.  I'd say I was a behaviorist only in the weak sense that
our access to the internal state of a brain is only currently
available through observing behavior.

I can't imagine that Skinner would claim that the previous life
history of a complex organism had no bearing on current complex
behavior of that organism, but Chomsky seems to be implying that.  In
any case, I think that the internal state built up by life experience
is crucial to complex behavior, even if simple behavior can be
successfully molded by simple conditioning, as Skinner's experiments
show.

So:

The meaning of a symbol in a brain is encoded in the interconnections
of the neurons which activate when that symbol is active.  We can
probe the meaning of a symbol by observing the behavior of the
processor.

Processing elements in a CPU are connected such that certain symbols
mean zero and one.  We can probe that meaning by observing how the CPU
adds numbers.  Behavior at this level is simple enough that we're
still in the realm where Chomsky wouldn't be criticizing Skinner about
specificity.

-eric