[From Bill Powers (981016.0345 MDT)]
Fred Nickols (981015) --
re:Hank Folson 981013.0800 --
What I think Carver calls 'behavior' is the output in a PCT model. If so,
Carver says the output of a PCT loop is the 'goal' of the system. Anyone
care to explain how that works?
Nickols:
I think I understand what you say above and I think I agree with
it. I think the answer to your question is that it doesn't work;
you've put your finger on an internal inconsistency in whatever
C&S are saying.
When they were about to publish their first book on this subject, they sent
me chapters to critique, as they were then frankly acknowledging that they
were using my ideas. I objected to "control of output" then, to little
avail. They didn't get it, although (instructively for us) they could
reproduce a convincing introduction to how control systems work (mostly
hand-fed to them by me). They used a new language but they didn't give up
their old ideas.
To say that observable actions are goal directed and to extend
that to say that those observable actions are what is being
controlled is also true (but not the whole truth). I do, for
example, sometimes wish to move my arm and have it wind up in
a certain place and, at the same time, wish for it to follow
a prescribed path in getting there. In this case, I am (or so
I think) controlling for a perception of eventual arm position
and the path taken in getting there.
Correct. You can control at many levels. You can even tense a muscle to a
specific degree on purpose.
Thus, I have no quarrel with those who argue that we control
our observable behavior patterns (and thus produce patterned
behaviors). However, I take issue with the notion that that
is all there is to it. We also act to control the outcomes of
our actions (our perceptions of reference conditions if I
grasp PCT at all).
That is to say, we can control more than one level of outcome.
In my poor layman's lexicon, I believe we human beings act to
control the ends we seek as well as the means we employ.
But not at the same time. If you decide to steer your car so it stays on
the road, you must give up independent control of the steering wheel angle.
You have to let disturbances control the steering wheel. If the car's
parked by the side of the road, you (or your child) can move the steering
wheel any way desired, with no bad effects.
On
many an occasion, the means used are our own observable actions,
what many people call "behavior" and what PCTers seem to prefer
calling "output" (reserving the term "behavior" for referring
to the operation of the entire loop).
Since they don't know the difference, most people jump back and forth
between meaning actions and meaning outcomes when they say "behavior."
"He's balancing his checkbook" or "he's sharpening a pencil" refer to
outcomes without describing the means. "He's pushing on the car to move it
off the road" refers to both the means and the end. It's hard to find
examples of strict reference to means (actions), because when the means are
described, they're usually the end in the particular circumstances (The
ballerina raised her arm gracefully. But who would say "by means of tensing
her muscles"?).
Perhaps there's ground to be gained in the relationship between
ends and means (a couple of impostors if ever there were any).
"Actions" and "consequences", "results", or "outcomes" describe what's
happening a lot more clearly.
Best,
Bill P.
