NetNews Usenet Archive 1992 #20/NN_1992_20.iso/spool/bit/listserv/csgl/938

Text File | 1992-09-08 | 8.2 KB | 187 lines

Comments: Gated by [e-mail]
Path: sparky!uunet!paladin.american.edu!auvm!AERO.ORG!MARKEN
X-Delivery-Notice:  SMTP MAIL FROM does not correspond to sender.
Posted-Date: Sun, 06 Sep 92 11:17:26 PDT
Message-ID: <[e-mail]>
Newsgroups: bit.listserv.csg-l
Date:         Sun, 6 Sep 1992 11:17:26 PDT
Sender:       "Control Systems Group Network (CSGnet)" <[e-mail]>
From:         [e-mail]
Subject:      Misc topics
Lines: 174

[From Rick Marken (920906.1030)]

Good morning CSGNet!


Bill Powers (920905.1400) says:

>Rick Marken (920905) --

>A great great post, Rick.

I'm kvelling. Thank you so much, Bill. What a wonderful
weekend this is turning out to be.

Hank Folson (920905) says:

>The Flat Earth Society has much in common with the position
>CSGNet and PCT are in now. They are our brothers.

Not from my perspective. It's great to be skeptical but I think
that the Flat Earthers are just being silly. PCTers don't reject
linear causal models (SR, cognitive, etc) because we don't like
them. It's because there is NO EVIDENCE for them. There is no
data that supports these models -- period. The data presented
in the journals is meaningless; I was just reading an article last
night (in the Human Factors journal) where an r sqaure of .36
was call "strong". The PCT model is based on data where the
relationships between variables is ALWAYS in the range of r
squared = .98.  The data in favor of PCT is overwhelming (since
there is virtually no data in favor of the other views). I would
say that, based on the data,  PCT is as solidly the model
of choice for living systems as is "spherical" the model of
choice for the shape of the earth -- maybe more so. The Flat
Earthers seem to me much more like the SR and cognitive
psychology types -- trying desperately to make a false model fit
the available data.

Ed Ford (920905:2335) says --

>Rick Marken (9209040) Seriously, I would very much appreciate those of
>you who have raised/are raising children would pass on your tips on
>"aligning their control structures" with those of civilization.

Actually, I didn't ask -- it was Curt McNamara. My kids are
raised (at least, from their point of view).

>All this means the parent must set standards and rules which
>reflect the parent's own values and beliefs (systems concepts
level)

The parent has no choice, right?  You don't have to tell a control
system (the parent in this case) to control.

>and follow through with the natural consequences of not
>following the rules.

They are "natural consequences" only if the kid's behavior is
part of what the parent is controlling; the parent can't help but
"naturally" take action (ANY action) in an effort to move the
kid's behavior back to the parent's reference for it.

>I'm not speaking of hurting children (punishment) nor do I
>believe in being permissive.  All children eventually must
>learn to respect the rules of the culture in which they live or
>they will be in conflict with that culture.

Skinner also was big on non-punitive control. How do you "non-
punitively" get a kid to produce a perception for themselves
(and, incidentally, for you) that they have no intention of
producing?

This amazing new approach to child rearing based on PCT
sounds pretty much like the old version; set limits, teach
children to follow the rules; show that their actions have
"natural" consequences (a cop out that tries to make us believe
that the environment is trying to control us so we have to learn
to deal with it), etc. It sounds to me like you are saying what
Greg appears to be saying -- a little control is good for the kid. I
claim that there is no reason (based on the PCT model) to think
this is the case. The kind of control you are suggesting, Ed, is
fine for the PARENT, but it is not necessarily great for the KID.

Confirming the above hypothesis, Greg Williams (920906) says:

>Both Pat and I think that Ed's comments are wise. His
>techniques are important, in our opinion.

>As outlined by Ed, his suggested interactions with children
>include instantiations of purposefully influencing them -- what
>I have been calling "manipulation."

And to the extent that that is the case I think that Ed's
recommendations, if carried out seriously (REALLY enforce your
references for how the kid should behave -- ie manipulate the
kid's behavior) are prescriptions for conflict and pain (for the
kid, at least). In practice, I believe that Ed does not treat kids as
he suggests; he might be verbally firm and he surely tries to
influence people (by talking to people, getting them to ask
questions, explaining possible consequences, etc). But I'm sure
that Ed would never pursue the inevitable violence that would result
from attempting to get a person to "behave according to my rules and
standards, OR ELSE"

I think there is an idea lurking in the background here. The idea is
that those who recommend against "manipulation" based approaches to
child rearing are therefore in favor of allowing kids "complete
license" to do whatever the hell they want. But that is a false
implication. When kids do things that disturb the variables you
care about then you do something to stop it. The alternative to
manipulative child rearing is not to become a doormat. It's to
become a TEACHER. The goal is to help the kid learn to control
mode skillfully -- so that it's activities don't have unfortunate
side effects and so that it doesn't run into conflicts with other
control systems. But you don't get a kid to be skillful by
manipulating it into this state -- the control model implies that
this just cannot be done. Educate -- don't control. We have to
learn to educate better, not to control better. And don't worry,
unless parents adopt the "let the kids walk all over you" reference
levels of "summerhill" fame, they can be trusted to take care of
themselves just fine. You don't have to tell a control system to
control; you do have to tell a control system what to expect
when it's dealing with other control systems. I'm just trying
to do the latter; it sounds to me like Greg and Ed are trying
to do the former.

>Also, Bill and Rick: You both said in posts of yesterday that "education is
>not control." Please explain what education is, as related to PCT notions. Can
>the concept of education even be approached via PCT ideas?

Education takes advantage of aspects of the control hierarchy
that we don't discuss that much on the net -- imagination,
memory -- and also aspects that nobody understands too well --
how we turn word perceptions into the imagined perceptions
that those words represent. Education involves teaching a
control system to control. Words can be used to try to describe
the kinds of perceptions that can be controlled in order to
control a higher order perception. Like in teaching tennis -- the
"teachee" usually comes in with a good idea of the higher level
perception to be produced (a great tennis game) but doesn;t
know which sensations, configurations, transitions etc to control
in order to get this to happen. The coach tries to communicate
what s/he thinks are good perceptions to control and, possibly,
good reference level to get these perceptions too.  The teacher
(a good one) doesn't  MAKE the teachee exhibit a partiular
behavior; the teacher tries to help the person learn to control
perceptions that will allow the person to control.  Education is
not control exerted by the teacher -- it is a process where the
teacher uses the teachee's ability to understand, imagine,
remember, and reorganize in order to help the teachee be able
to control.

Bill Powers (920906.0600) on the imagination connection:

>Anyway, this ought to be a lot easier to put into that
>spreadsheet model, Rick.

I think this is the way I had it set up in the first place (error
goes directly back into perceptual signal) but I changed based
on the discussion we had some time ago with Martin where he
proposed that the imagination connection go through the
perceptual input. We liked his idea better at the time. I'll go
take a look.

Best regards

Rick

     **************************************************************

Richard S. Marken                   USMail: 10459 Holman Ave
The Aerospace Corporation                   Los Angeles, CA 90024
E-mail: [e-mail]
[phone] (day)
[phone] (evening)