NetNews Usenet Archive 1992 #26/NN_1992_26.iso/spool/bit/listserv/csgl/1412

Internet Message Format | 1992-11-09 | 6.0 KB

Path: sparky!uunet!stanford.edu!bcm!convex!news.oc.com!eff!sol.ctr.columbia.edu!spool.mu.edu!hri.com!noc.near.net!mars.caps.maine.edu!maine.maine.edu!cunyvm!psuvm!auvm!AERO.ORG!MARKEN
From: [e-mail]
Newsgroups: bit.listserv.csg-l
Subject: PCT, Politics;learning
Message-ID: <[e-mail]>
Date: 9 Nov 92 17:16:31 GMT
Sender: "Control Systems Group Network (CSGnet)" <[e-mail]>
Lines: 109
Comments: Gated by [e-mail]
X-Delivery-Notice:  SMTP MAIL FROM does not correspond to sender.
Posted-Date: Mon, 09 Nov 92 09:16:31 PST

[From Rick Marken (921109.0900)]

While walking with me wife this weekend she challenged me to
come up with some surprising conclusion derived from PCT that
might grab the interest of a layperson. Here's what I came up with.

"You can't tell what people are doing by watching what they are doing".

It's not often that I say stuff about PCT that grabs her attention --
but she really liked this one. So I offer it to the net for comment.

Bill Powers (921107.1100) --

>Rick Marken (921106) --
> You have hit a productive note on the political scene.

I would like to talk more about the system concepts of the
right and left; including what I think are the implicit models
of human nature that are a part of these concepts. But there was
an article in the Sunday Times that made me realize that there
is an implication of PCT for politics (or policy) that we have not
really considered yet. I'll say what it is -- then describe the
article that motivated it: PCT suggests that it is the GOAL,
not the MEANS, that should be important in politics. What we
should do is try to articulate the kind of society we want, NOT
how we should get there. The latter "depends" on the current "context"
of society (ie. disturbances and constraints). So there can't be one
"right" way to produce a particular societal result -- just as there cannot
be one "right" way to move the handle to keep the cursor on target.
The means MUST vary to produce the intended result. Social ideologies
are generally descriptions of policies which are supposed to
achieve some vague goal -- like a better society. But it is clear
from looking at the world that the results of various policies have
always been different. Sometimes socilistic practices "work" -- sometimes
they fail. Sometimes free market practices "work" -- sometimes they
fail. PCT suggests that this is to be expected -- it's not that we
cannot know what to do to produce a particular social result (though
this is close to being true); more important, continuous contextual
changes in society (and the environment) mean that a particular means
of producing a particular result CANNOT POSSIBLY be expected to work
all the time; the means must vary in order to produce consistent results.

The article that got me thinking about this was in the Business
section (which I rarely read). It was about how "history may judge
Reagonomics very harshly". I, of course, read and enjoyed the article
since it pointed out some of the cruel results of the Reagan era.
Now one can try to blame the Reagan disaster (my kids and grandkids
will be paying for it for years) on congress or communists or whatnot.
But the fact of the matter is that the main agenda of the Reagan
era was passed happily by congress -- massive tax cuts, especially
for the upper tax brackets. The effect was supposed to be increased revenue
due to all that economic growth -- ie. trickle down. This didn't happen
(to a disasterous extent) -- thus, a massive debt was born.

What struck me was that the "trickle down" idea is not necessarily
"intrinsically" wrong. It might have worked -- maybe it did work some-
where at some other time. The problem was that when it
became obvious that it was NOT working there was no effort to
take any steps that MIGHT remedy the situation. Their was  a BELIEF
about how the MEANS to an end SHOULD work -- beliefs are just
references that are not altered in order to achieve higher level
goals.

It may be (as I suggested in an earlier post) that the Reagan/Bush
crowd HAD NO higher level goals; tax cuts WERE the highest level
for them (they seem to be the highest level for most people).
I hope that we can move politics to a level where we are trying
to articulate system concepts rather than yelling at each other
saying "this is the RIGHT way to do things". I believe PCT shows
that this is just wrong-- there can be no RIGHT way to achieve
a particular social goal -- no matter WHAT that goal might be.

That's why I like Clinton -- he seems to understand that it's the
vision that matters, not the means (he basically said that in the
debates). You have to be willing to try various things and see
whether the result is moving you toward or away from the goal. But
without a goal, there is no way to judge whether or not you are
"doing the right thing". I think the "vision thing" is the whole
ball game; the vision thing means "having a reference level for
a particular system concept". That's what's important. What you do
to achieve that vision MUST BE determined by "context" -- not belief.

Bill Powers to Jeff Hunter --

> So the difference in control with the disturbance present is not due
> to a change in loop gain, but to using disturbances too large to
> resist. The loop gain is zero for average disturbances greater than +
> or - 2. This shows the pitfalls of estimating apparent loop gain
> without looking at the details of behavior -- in fact, there was no
> control at all, but the data averaged over 10,000 iterations
> concealed that fact.

Thanks for doing the work, Bill.

Jeff. Again, I think it's great that you are exploring the control
model through simulation. But do try to read some of the literature
that I suggested; it can save you from some of these blind alleys.

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)