Re: The Behaviorist Mantra

On Fri, 28 Nov 2003 21:39:26 +0100, "Rolf Marvin Bøe Lindgren"
<[EMAIL PROTECTED]> in comp.ai.philosophy wrote:

>[Lester Zick]
>
>| Well you seem like a decent conversationalist so I'd rather not risk
>| being unnecessarily contentious on this subject. However my basic idea
>| is that the universal implications and bona fides of scientific
>| methodologies have to be established in plain terms rather than in
>| terms of technical jargon and specialized terminology. This is what
>| Newton did for physics with his laws and this is what I would expect
>| of behaviorism and don't find in any significant degree.
>
>I'd think you'd have to be a physicist in order to believe that Newton
>stated anything in plain terms.  the plain terms that Newton's physics
>is expressed in today comes from years of tradition and refinement -
>and Newton drew upon an already established but incomplete science. 

Sure. But I think both sides of the discussion have been comparably
improved in terms of understanding and expression.
>
>Skinner tried to establish Psychology from scratch, from his
>perspective.  this may sound funny now, but as late as 1930 there was no
>universal agreement about what Psychology was nor any great body of
>knowledge like what we have today.  cognitive psychology can be dated to
>1967 if you want to be really fussy. I'd say it'd be unfair at the time
>to demand of psychology to be like physics in any trivial way and
>perhaps still is.

Except that in terms of methodology behaviorism simply seems to have
drawn on empiricism for its view of behavioral science. And I don't
regard that as a psychological or behavioral issue requiring arcane
terminology. 
>
>| By and large there just seems to be a presumption that science can
>| only be conducted in experimental terms.
>
>Skinner wrote a lot about that.  he said there'd be a time for theories
>of learning, but not until we have much more data.  he said that even
>in 1980 after over a decade of cognitive revolution and restated it a
>decade later:  cognitive psychology is premature but doesn't admit it.

It may be premature. I'm not an adherent of either school. But the
empiricist approach to psychological analysis defines its science
through circumstantial reference to particular characteristics not
universally evident or applicable. And what we need are not more
experiments but more insights.
>
>| And that the conduct of experiments on behavior ispso facto makes the
>| experimental analysis of behavior a science in universal terms. I
>| disagree and contend that behaviorism actually represents a particular
>| rather than universal science analogous to the physical sciences.
>
>what makes the physical sciences universal sciences?  they were first
>and get to define the terms?
>
Actually they aren't universal sciences in methodological terms. But
they deal with universal particulars in the properties associated with
matter and material interactions which is why conclusions of these
sciences have universal implications.

Only analytical comprehension can yield universal science in general
terms - geometry, math, and whatever can be known of physical and
behavioral sciences in analytical terms. But that doesn't mean that
experimental analysis of behavioral particulars yields anything like
the universal implications characteristic of the physical sciences. In
other words we have a particular science in methodological terms
experimenting on particular behavioral circumstances without universal
implications on either count.

And this isn't an experimental issue. Newton had Kepler's conclusions
to draw on in celestial mechanics and Galileo's in mechanics and yet
he was able to extrapolate several material properties of universal
character and import. There is nothing like this kind of insight in
the realm of behavior. Yet behaviorism and other schools routinely
continue to promise the kind of insight and results they can't and
never have been able to deliver.

Glen Sizemore maintains that experimental behavioral analysis can
control and predict behavior through behavioral contingencies. So what
else is new? I can do the same thing by tying a chain to a rat's ankle
nailing it to the floor and predicting the exact radius of motion for
the rat.

I maintain that experimental analysis of behavior does not yield
universal results on behavior at all. What it actually identifies are
environmental constraints on behavior some of which may be very subtle
in nature. My own take on the differential origin and nature of
cognition in fact leads me to suggest that no experimental analysis
will ever be able to yield universal results for very specific
reasons.


Regards - Lester