Basic lessons in PCT

[From Bill Powers (940923.1740 MDT)]

Hans Blom (940923) --

Your tasks in this assignment are twofold:

1. to show that the term "irrelevant side effects of control" is,
within the theory of B:CP, selfcontradictory.


a. irrelevant side effects can be perceived (from the above sentence);
b. irrelevant side effects are not controlled (by definition);
c. B:CP (axiom).

MY behavior is not the control of YOUR perceptions. An irrelevant side
effect of MY behavior is some effect of my outputs that YOU notice and
that is interesting to YOU, but an effect that is not related to one of
MY controlled variables. If I stand up at a football game to see the
action better, doing so may block your view. That is an irrelevant (to
me) side effect of my behavior; I can control only those effects of my
actions that I can perceive.

You may bring that side-effect into the realms of relevance to me if you
take some action to get me to move, such as yelling "down in front!" If
in fact something I care about is disturbed by being told to sit down, I
may comply with your request (if I have the goal of avoiding disturbing
other people and now perceive that I have done this to you) or I may
turn around and punch you in the nose (if I have an even stronger goal
of not being bugged by other people).

On the other hand, you may suffer in silence from my lack of
consideration; in that case, my effects on you are entirely irrelevant
to my own processes of control.

Another case, which should take care of the other meaning of relevance
you have in mind, "objective" relevance (you see, I do remember what you
have said, even if you don't remember what I have said). Suppose you
stand up to get a better view and drive the top of your head into a low
supporting girder for the balcony above. As far as the control system
for getting a better view is concerned, cracking your head on the girder
is an irrelevant side-effect, provided that you do in fact get a better
view. If you are prevented from standing up far enough to get a better
view, then the physical limitation of your movement amounts to a
disturbance of the control system in question, and unless some higher-
level system has been knocked out of action, you will alter the goal of
standing up before that control system can drive your head into the
girder more than two or three times. The pain is irrelevant to this
control system, because all it controls for is a free field of vision.
But not getting a better view is relevant, because it affects the
control process that is in progress.

The pain from hitting the girder is, of course, a disturbance of a
different control system, one that avoids painful contact with
stationary objects. That is a different control system; what matters to
it is not relevant to the better-view system, not is seeing the football
game relevant to the obstacle-avoiding system. Normally, the avoiding-
obstacles system can keep its own perceptions under control while the
getting-a-better-view system does the same for its own perceptions. In
this case, the avoiding-obstacles system did not work because it didn't
see or know about the girder.

I don't see why I have to explain this to someone who has been around
the net as long as you have.



Bill P.