Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!bloom-beacon!ai-lab!miken
From: miken@wheaties.ai.mit.edu (Michael N. Nitabach)
Newsgroups: comp.ai
Subject: Re: Free will and responsibility.
Summary: The hobgoblin of behaviorism
Keywords: Behaviorism, materialism, dogma, science
Message-ID: <2492@rice-chex.ai.mit.edu>
Date: 18 May 89 15:21:59 GMT
References: <10333@ihlpb.ATT.COM> <3850@uhccux.uhcc.hawaii.edu> <52019@linus.UUCP> <1979@ucsfcca.ucsf.edu>
Organization: MIT AI Lab, Cambridge, MA
Lines: 57

I have several comments regarding the recent discussion about free will
and behaviorism.
First, it seems that many of the participants have assumed that a
materialist stance somehow requires one to accept the validity of the
radical behaviorist agenda.  This is a misconstrual of the nature of
the basic behaviorist claim, which is that regularities in animal
(including human) behavior stem solely from environmental regularities,
and *not* from any structure inherent in the mind of the animal (except
for the primitive associative capacity).  This position neither implies
nor is implied by a materialist metaphysics.  Radical behaviorism *is*
compatible with a mentalist metaphysics; the "environment" in this case
is taken to be a milieu of mental events, rather than physical events.
The important point for behaviorists is not the ontological status of
the events that shape behavior, but rather the idea that it is solely
regularities in events which arise *externally* to the agent which
determine the structure of its behavior.  Furthermore, it is possible
to hold a materialist point of view, yet not accept the tenets of
behaviorism.  This is embodied in the view, e.g., that the structure
of behavior stems not only from external environmental regularities,
but also from regularities in *physically instantiated*, yet internal,
mental processes.
Second, it appears to me that the notions of free will being discussed
in this group all miss the mark.  The decision as to whether an agent
has free will does not hinge on whether that agent is bound to obey
the laws of physics, deterministic or probabilistic, either in its
overt behavior or in the internal physiological processes that "make
it go".  Nor does this decision depend on whether the determinants
of behavior are physical or mental events.  Rather, free will, or its
absence, derives from the source of the determinants of behavior.  If
the behavior of an agent can be fully explained by events occuring
*external* to the agent, then that agent should be said to lack free
will.  If, on the other hand,  the regularities in this agent's
behavior can only be explained by taking into consideration factors
*internal* to the agent, such as its beliefs and desires, then the
agent must be said to possess free will.  Note, that this concept of
free will is continuous, in the sense that an agent can possess more
or less free will, to the extent that its behavior is internally and
externally caused.  It also appears that a notion of free will similar
to this is regularly applied in social determinations of moral
responsibility.
Third, I don't understand the undercurrent of fear regarding the
potential sinister uses of behaviorist techniques for purposes of
social control.  This concern is founded on a belief that behaviorist
techniques *are*, in fact, successful in drastically altering the
social and private behaviors of human beings, including unwilling and
unknowing objects of these methods.  I submit that there is no evidence
to support this supposition, and that this fear is unfounded.

I am new to this newsgroup, so I apologize if I am repeating arguments
and perspectives which have already appeared here.

Michael Nitabach                       e-mail: miken@wheaties.ai.mit.edu
Dept. of Brain and Cognitive Sciences
Massachusetts Institute of Technology
E25-534
Cambridge, MA 02139
(617) 253-0771