Path: utzoo!attcan!uunet!cs.utexas.edu!tut.cis.ohio-state.edu!ucbvax!decwrl!shelby!labrea!sri-unix!orawest!ejs
Newsgroups: comp.ai
Subject: Re: Free will and responsibility.
Summary: Behaviorism is just a theory
Keywords: Behaviorism, materialism, dogma, science
Message-ID: <528@orawest.UUCP>
Date: 15 May 89 20:29:09 GMT
References: <10333@ihlpb.ATT.COM> <3850@uhccux.uhcc.hawaii.edu> <52019@linus.UUCP> <1309@lzfme.att.com> <1966@ucsfcca.ucsf.edu>
Reply-To: orawest!ejs@unix.sri.com (e john sebes)
Organization: Odyssey Research Associates, West; Menlo Park, CA
Lines: 125
Return-Path: <ejs@ord>
To: sri-news@mailhost

In article <1966@ucsfcca.ucsf.edu> brianc@daedalus.UUCP (Brian Colfer) writes:
>One of the classic arguments against Behaviorism is that it is
>essentially a fascistic philosophy.
>...
>The fundamental basis of Behaviorism is that we are material beings.
>There is no evidence that there are any other causes for our behavior
>other than genetics, experience (biological and psychological), and 
>current environment.  That's it! ... nothing else!  

I have a number of corrections to Mr. Colfer's rhetoric.

Behaviorism is not a philosophy, it is psychological theory or method.
The *philosophical assumptions* of behaviorism are what is generally
  called materialistic determinism.
  As an assumption, this is scientifically acceptable; as a 
    philosophical dogma, it is like all such, unacceptable.
The *criticism of* (not "arguments against") some behaviorists' rhetoric
  (not "Behaviorism") is that it is dogmatic. That is, in addition
  to *assuming* materialistic determinism, some behaviorists assert it,
  making essentially metaphysical claims that are out of the bounds
  of science; this reduces the credibility of them as scientists,
  and hence clouds the value of their work, which is as important
  (to me, at least) as any theory of human actions.
Such observations about behaviorists should not be construed as
  logical arguments against the psychological theory.
This is not to deny that many people are concerned about
  the totalitarian possiblities of the application of techniques
  that behaviorists claim are efficient.
  This concern even predates behaviorism as such! Just check out any
    of the classic dystopic novels, such as 1984 and Brave New World.
  I would like to point out, however, that this concern is
    *amplified* as a result of the apparent dogmatism of some behaviorists.
Finally, the point of such concern isn't that the practitioners of
  such techniques will be immune to them, but that they will
  have the knowledge of the use of the techniques--
  in other words, POWER over others.

Mr. Colfer's posting is classic example of a naive form this dogmatism.

For his benefit, I will attempt to answer some of his questions
in an attempt to illustrate some of my remarks.

>Why should our ablility to describe the behavior relationship for 
>things in a human skin be different than anything else?

The fundamental answer is: nobody knows. Still, a lot of people
have the belief that this is the case, both from observation and
from unverifiable suspicion. The key word is unverifiable--
that the human mind is not completely deterministic is
a metapysical claim that can't be scientifically addressed,
just like the existence of God.

A lot of people (yes, Virginia, even Scientists) think that it might
be the case. Such folks (responsible ones at least), recognize that
there is nothing to do about it but take an agnostic stance, and
get on with work.

The key point here is that you can't base respectable scientific
work on an assertion *either way* about an unverifiable question;
you just assume that since you can't prove it, it is not important,
and forget about it, except for bull sessions over <your favorite beverage>.

The best we can do, it seems, is do enough observation and develop our
predictive powers to such a degree that we can say that we don't
care at all metaphysics, that we are plenty satisfied that we can
treat humans as deterministic systems, and that *practically* that is
all that matters.
... Unfortunately, this is just a tautological answer to the question:
when we can "describe the behavior relationship", etc., then we
will know that we can.
In other words, get on with investigation, and we'll see.


The rest of my answers are it bit more brief.

>My question is why would any one think that 
>1) we are more than materialistic beings ... what valid evidence is there? 
>Valid in the sense of observable, reliable and well constructed measures.
No evidence, *either way*. That doesn't stop people from supposing
that it might be so, in a way that is beyond our ability to either
a) to verify (i.e. theoretically beyond our ability), or
b) to collect convincing empirical evidence about, at present,
   (i.e. beyond our present practical ability)

>2) is the Behaviorist explanation of behavior accurate... that is, it works 
>so often that we can call it accurate.
Again, this a question resolved by lots of observation, theorization,
testing, etc, the whole load of scientific endeavor.
(The current consensus, from where I sit, is that nobody was convinced
 then of the ultimate wonderful of behaviorism, and few people are
 pushing it these days as anything but one of several interesting
 approaches to the study of human actions. However, poeple (like Colfer)
 who are already of a materialistic bent, are continually rediscovering
 behaviorism, and getting keen on it-- evidence that the theory does
 some essential value, at least in the world of ideas.)

>3) if 1 and 2 are true is it immoral to not apply these explanations
>in ways that can help people?   e.g. Teach people lots of info in non-
>coercive environments, help the mentally ill... help people work
>better ... airline maintenance, air traffic controllers etc.
Yes, it might be immoral. Even if you are not "coercive", you
can still manipulate people in ways that they are not aware of,
or even worse, in ways that they are aware of, cannot control,
and which go against their own wishes.

That is where the concern over behaviorism arises. To a strict
determinist, this is not a problem: the controlling people are
behaving deterministically in their deterministic control over
others, so the controllers are no different; developing
behaviorist control techniques is the whole deterministic system
developing new ways of changing.
However, there are few real determinists so dyed in the wool
that they would like to let other people use such techniques
on them without any caution, just because it is all determined.

Practically speaking, however, it seems likely to me that
the influence of the work and ideas of the behaviorists
will (or has) helped in the development of such benign
techniques as Mr. Colfer describes, which can easily be
applied in ways that raise few moral qualms.
I seems equally likely that unbenign, moral-qualm-raising
applications (or at least the practical possiblity thereof)
will arise as well.

E. John Sebes