NetNews Usenet Archive 1992 #31/NN_1992_31.iso/spool/bit/listserv/csgl/2028

Text File | 1993-01-02 | 7.7 KB | 179 lines

Comments: Gated by [e-mail]
Path: sparky!uunet!gatech!paladin.american.edu!auvm!VAXF.COLORADO.EDU!POWERS_W
X-Envelope-to: [e-mail]
X-VMS-To: @CSG
MIME-version: 1.0
Content-transfer-encoding: 7BIT
Message-ID: <[e-mail]>
Newsgroups: bit.listserv.csg-l
Date:         Sat, 2 Jan 1993 11:39:32 -0700
Sender:       "Control Systems Group Network (CSGnet)" <[e-mail]>
From:         "William T. Powers" <[e-mail]>
Subject:      firm facts; almost facts; evolution of system concepts
Lines: 164

[From Bill Powers (930102.0930)]

Dick Robertson (930101.1600) --

The SS number you want is 361-16-3506.

Some more solid facts:

1. William James' observation: organisms achieve repeatable ends
by variable means.

2. The PCT version of this observation: outcomes, not outputs,
are stabilized against disturbance.

3. The concept of hierarchy in behavior (there must be lots of
references to this): behaviors organized at one level are means
of and part of accomplishing behaviors organized at a higher
level.

4. The concept of reorganization (change in the properties of
routine behavior).

Another category might be called "Near, but yet so far."

1. Thorndyke's Law of Effect.

2. Control BY consequences.

3. Skinner's Operant Conditioning.

4. Tolman's Purposive Behaviorism

5. S-R chaining.

6. Knowledge of Results.

7. Naming behaviors for their effects (scratching an itch,
writing a letter, uttering a word).

This post was full of great ideas, Dick. The idea of doing an
end-run around the psychological establishment has great appeal.
The difficult with that is avoiding getting lumped with all the
other ideas that have gone in that direction with much less
justification.

>Bless us all in csg

, every one.
--------------------------------------------------------------
Gary Cziko (930102.0341 GMT) --

That was a whole load of useful ideas from Don Campbell.

I agree that beliefs -- as system concepts -- must have some
functional reason for existence. They must represent in some way  an improvement
 on the ability of organisms to control what
happens to themselves, in comparison with organisms lacking that
level.

(Belief as imagination is not as hard to understand; we have to
be able to continue control actions at least for brief periods
when not all the usual perceptions are available. I don't have
any problem with pushing on a door because you "believe" it's
unlocked. That's not the interesting kind of belief.)

Campbell wants to explain how beliefs can get people to do things
against their own immediate interests. His explanation brings in
evolution. That's an interesting track, but there's another. I
think that belief systems have to do at least in part with a need
to explain things -- that is, to put experiences into a framework
where they can fit comfortably with other experiences. In trying
to explain belief, aren't we doing that? I think there may be an
interesting progression of system concepts, starting way back
with animism.

Animism, it seems to me, is simply an attempt to explain natural
phenomena as if everything that happens is controlled by
somebody. In prescientific and prephilosophical times, the
purposive nature of one's own behavior must have been self-
evident. If I want a rock to be on top of another rock, I just
put it there. All I need to do, basically, is want it to be
there, and it's there in a jiffy. That's not much different from
understanding that to move my hand, all I need to do is want to
see it in a different place, or see it moving. So wants and
intentions and goals must have been obvious, if mysterious,
aspects of everyday life. I can see how this sort of experience
would lead to the idea that wanting and intending, by themselves,
must have causal effects on the world. People would naturally
wonder whether, just by wanting or intending, they might be able
to cause _anything_ to happen by "force of will": stare at the
sticks of wood long enough, willing them to become a fire, and
they ought to burst into flame. Stare at the back of another
person's head long enough, willing that person to look at you,
and that person will eventually turn and look in your direction
(which is probably true, but not for the reason assumed).

Looking at other people, you see that they, too, can
spontaneously make things happen just as you can. So there's
plenty of evidence that control exists, that things are caused to
occur by someone willing that they occur. All it takes is a
straightforward generalization to conclude that everything that
happens works that way. If a rock comes tumbling down the
mountain, somebody intended for it to do that. If rain comes and
puts out your fire, somebody is mad at you. Animism is just the
logical outcome of noticing purposive behavior: your own, that of
other people, and finally, through imagination, that of invisible
beings. The ultimate generalization was to conclude that
everything, in the final analysis, existed because someone willed
that it exist.
  The next step in the progression I mentioned is what happened
with the advent of science. Now certain things, like the swinging
of pendulums and the rolling of balls down ramps, occurred not
because of anyone's will but because of the action of blind
natural forces. This was a new system concept, replacing the one
that said everything that happens is willed by someone. As more
and more natural phenomena were explained in terms of the new
system concept, the same urge to generalize led to trying to
apply this system concept to everything. Now NOTHING was caused
by someone willing it to happen. EVERYTHING happened because of
the operation of blind natural laws. Even the apparent
purposiveness of the behavior of animals and people was actually
just the outcome of these blind natural forces at work. So we end
up with anti-animism applied in the same overenthusiastic way
that animism was applied.

The third step is, of course, the discovery of control theory and
the realization that SOME systems made of matter are purposive,
and the rest are not. We're still sorting this out.

So I think that a lot can be encompassed within the idea that
people simply want to create system concepts that bring as much
of experience as possible into a common framework, a common
model. The evolutionary arguments would apply to acquisition of
this ability to create models, but not (at least not in terms of
surviving to the age of reproduction) to any particular model.
The particular models grow out of experience in what seems to me
a simple, even simple-minded, way. We get an idea, and try it out
on everything to see how it works.

Considering the time-scale involved, I don't see how we could
blame evolution for the way we follow out the logical
consequences of particular system concepts -- consequences such
as heroism and martyrdom. Such things affect only a minute sliver
of the human population. I don't see any philosophical difficulty
in the idea that some people would rather die than violate their
system concepts. Nobody knows what dying is, anyhow --
personally, that is. In the course of a few weeks' torture by the
Gestapo, no person is going to be able to reorganized a complex
system concept and all the lower organizations that depend on it
and support it. A lot of people reorganize enough to spill the
secrets -- most of them, probably -- but they generally end up
being shot anyway, so there's no great evolutionary advantage
involved one way or the other.

I don't think we can blame evolution for any basic capacity of
the mind arising out of events over any time span shorter than,
say, recorded human history. I prefer a simpler explanation for
the way people can be led by their own models of reality to act
against their personal interests.
--------------------------------------------------------------
Best to all,


Bill P.