[From Bill Powers (9909030.1024 MDT)]
Marc Abrams (990929.1319)--
Thanks for your continued research into basic PCT ideas. That was a
beautiful post by Rick, an antidote to much of what has been going on on
CSGnet recently, a moment of clarity and sanity. And thanks to Tom Bourbon
for some of the underlying ideas. If we could somehow hew closer to this
line, PCT would be better off.
Best,
Bill P.
from [ Marc Abrams (990929.1319) ]
Bruce Abbott. This one is for you :-). In lieu of the recent thread on Proof
and PCT, here is an oldie but goodie from Rick. How does the Paul Simon song
go? "Still (crazy)
true after all these years..." :-).
Marc
So here is a new thread!!
I finally received my CSG Newsletter and was moved almost to tears by
Tom's presidential address. It was so good I read it aloud to my wife.
I think he makes a point that we should all be reminded of and reflect on
regularly; we are not dealing with absolute truth but with testable
models of truth. This is a point that Jacob Bronowski (sp?) made in
"The Ascent of Man". The beauty of science is that its "truths" are
tentative; they are always open to test. The worst horrors of humanity
occur when people are certain; when they know that they have the "truth".
This is the nightmare of religion and ideology. As Tom said so well,
this nightmare occurs when people believe that certain "ideas
are so beneficial and appealing that their truth and beauty must be
evident to everyone". Control theory does appear to be beneficial and
appealing. But that is not the test of its truth. Control theory is,
as Tom said, "just an idea", but a "true" one as long as it stands
up to continuous, rigorous and fair testing. It is it's ability to
explain a phenomenon -- purposive behavior -- that gives it its
value; not its intrinsic beauty (and it does, indeed, have intrinsic
beauty as well).
Tom is again right when he says that modelers -- the people who are
actually testing the control theory idea -- have run into places where
the model does not seem to work. I have had this experience in my
work on hierarchical control. Phenomena that seemed to require an
explanation in terms of hierarchical control actually did not. It was
not so much that the basic idea of control theory was wrong-- but a
mistake in how I saw the model being mapped into behavior. The correct
model (not the "true" one, but the one that worked) was a non-hierarchical
model that controlled a different variable than I had originally guessed.
I want to add something to what Tom said. I believe that there is a
misconception about what it means when a model does fail a test. People
who look at models in the same way as they look at religious ideas think
the model is either TRUE or it is not. The model is seen as "testable" but
what is being tested, according to these types, is its TRUTH. Thus, when
the model fails (as, I believe, the passive, Darwinian evolutionary
model fails) then the conclusion is that the model is FALSE and a radical
alternative is accepted (such as creationism). The fallacy here is related
to what Tom pointed out in his talk; we don't test the TRUTH of a model; we
test its explanatory power. Tom is right; all models are false. That's a
good way to start. Testing does not evaluate TRUTH; it evaluates how well
the model explains what we experience. Some models ARE BETTER than others;
the ones that explain the phenomena and survive the tests. The models that
fail in this regard can be considered WRONG: but they are not necessarily
useless. I'd say that Alchemical models in chemistry can now be
considered WRONG; moreover, they are also not useful compared
to the atomic model. The Newtonian model of the universe is also
demonstrably WRONG -- but, since it is still useful, I think that it is
less WRONG than the Alchemical model. Input-output models of behavior
are demonstrably WRONG and, I think, useless in the same way as the
Alchemical model(to the extent that that is even a model). This is because
the input-output model is actively misleading -- in ways I won't go into
here because they have been documented rather fully in Bill's books
and just about everything written by members of CSG.
While control theory is not TRUE, it is currently less WRONG than input-
output type models of behavior. Eventually, there will probably be a
better model to replace control theory, but I can guarantee one thing,
that model will not be an input-output model. It will also be a model that
can behave purposively, just like a control model. At the end of his 1973
book, Powers himself acknowledged that the details of the model he described
may be wrong (some of these details are already being tested and the model
augmented) but he said (and I confidently concur) that he would be surprised
if the basic organizing principle of purposive systems -- that they are
organized around the control of perception-- turned out to be completely
off base. Again, in some distant future there may be some super-model that
goes beyond control theory in some fundamental way (because failure of some
tests of the control model demanded a new approach) but it is almost
certainly not going to be a model that says that perception guides behavior.
There is one other little point I would like to make. Although models
should not be considered TRUE, even when they have passed all tests to date,
they can be considered our best shot at understanding some aspect of our
experience. People do care about models because they are part of our
system level understanding of our experience. We should not be dogmatic
about them and enforce belief in them; that is the job of religion and
ideology. But we do care about them. The understanding of human nature that
I get from the control model is important to me; it makes me feel satisfied
and enabled. It is important to me to show why this model is a better one
than input-output models. It is important to me to try to show other
people the fundamentally different perspective on "how people work" that
we get from the control model. But I just think of the control model
as a model -- a tentative step toward trying to understand the cause
of one aspect of my experience -- the experience I have of other people
and myself. But I understand that the model is tentative; it is an
approach; it is testable and I expect revisions. Control theory is an
extremely satisfying model but we must always remind ourselves (as
Tom did in his speech) that it is a model, not revealed TRUTH.
On a different, but related topic: I am going to give a talk in about
a month on the value of theory in the practice of human engineering. I
wrote a brief article about this once. One of the topics I would
like to discuss is "The difference between descriptive and working
models of behavior". Does anyone have a nice, succinct way of describing
the difference and explaining why descriptive models are stupid. The
reason I would like to find a simple way to explain this is because, to
the extent that models are used in human engineering (human factors) they
are descriptive models. That is, they model a person doing some task
in terms of a bunch of boxes that break down the task into components
that take some time to do with some probability of error. So the model
describes, say, typing behavior but the model can't type.
Any suggestions or comments on this would be greatly appreciated.
Best wishes
Rick
Date: Thu, 18 Oct 90 10:12:07 -0700
From: Rick Marken
**************************************************************
from [Kenny Kitzke (991024.1200EDT) ]
<Marc Abrams (990929.1319)>
Thanks for posting informative stuff from the CSGNet archives like the
following:
Date: Thu, 18 Oct 90 10:12:07 -0700
From: Rick Marken
<I finally received my CSG Newsletter and was moved almost to tears by
Tom's presidential address. It was so good I read it aloud to my wife.
I think he makes a point that we should all be reminded of and reflect on
regularly; we are not dealing with absolute truth but with testable
models of truth.>
I can't guess what about such a conclusion would move Rick to tears. For me,
it reveals the biggest strength and biggest weakness of science.
Science gives us enhanced understanding of how things we experience in the
world *may* actually work based on models of the actual things. If the
models are highly representative, and if the modeler's demonstrations are not
biased, scientific knowledge can be reproduced and tested for correlation and
accuracy of our understanding of the actual things. That is a strength.
But, many issues in the experience of human beings can't be adequately
modeled by scientists. And, many of these things are more important to
humans than what science is able to model accurately.
Science, including PCT science, gives no answers for what one *should* do.
It can't tell us whether the things science shows us how to do are *good* or
*bad* to do in any absolute sense. Yet these are serious issues faced every
day by people.
For example, science has been of no help in selecting which person one
*should* marry. Science is of no help in determining whether one *should*
divorce a spouse. Yet marriage and divorce issues affect most people. They
are the source of much anger and tears with over half the marriages in the
USA ending in divorce. But, only a few people [like Rick] are moved to tears
by a revelation that science cannot provide absolute proof of truth. Science
can only approximate truth with models.
Many mature people don't expect science to provide them with absolute truths
about things that matter most in their lives. They simply, and without
tears, turn to other disciplines to find truths and virtues in living. Those
disciplines may be philosophy, history, theology, etc.,
History proves that people lived and survived for thousands of years without
any knowledge of science at all, or the testable models it could devise, or
the conditional, relative truth it could reveal. Humans and their nature are
amazing; far more amazing and inspiring than the discipline of science, which
they created. For the scientist is greater than his science. Science has
its place in our world. Wise are those who know how limited that place is.
[From Rick Marken (991030.1030)]
Me:
I think he [Tom B.] makes a point that we should all be
reminded of and reflect on regularly; we are not dealing
with absolute truth but with testable models of truth.
Kenny Kitzke (991024.1200EDT) --
I can't guess what about such a conclusion would move
Rick to tears.
What was moving was the attitude of skepticism and humility that
was expressed in Tom's post. It is the scientific attitude, which
requires one to doubt and be willing to subject one's ideas to
experimental test (be humble before the observations). It is the
opposite of the religious attitude, which is an attitude of
certainly and arrogance. The religious attitude requires blind
faith and an unwillingness to subject one's ideas to experimental
test.
But, many issues in the experience of human beings can't be
adequately modeled by scientists.
Yet.
Science, including PCT science, gives no answers for what
one *should* do.
Actually, PCT science shows that nothing -- not science, not
religion, not mathematics, not history, nothing -- can tell a
person what s/he _should_ do. People can only figure out what
they _themselves_ should do (what perceptions they should control)
in the context of their own wants (existing perceptual control
hierarchy) and needs (intrinsic control systems). What one _should_
do depends on what other perceptions one is controlling _and_ on
the prevailing state of the external environment (disturbances) in
which one is controlling.
For example, science has been of no help in selecting which
person one *should* marry.
Nothing can tell you whether you should or should not marry a
person; only _you_ can "tell yourself". PCT can help, though.
If you can't decide whether or not you should marry a particular
person then you are clearly in conflict; you want to marry _and_
you don't want to marry this person. PCT shows that all "answers"
to this conflict are _wrong_. If the bones (the divining method
used by biblical prophets) say "marry her" then the part of you
that doesn't want to marry her experiences increased error; if
the bones say "no" then the part of you that wants to marry
will experience increased error. One way to solve conflicts
like this is to run away from them so there is no choice. A
better way to solve such conflicts -- really solve them -- is
to change the systems that are creating the conflict. PCT suggests
that this can be done by "going up a level" and revising the goals
of the systems creating the conflict or waiting until the
persistent error created by the conflict leads to a
reorganization (like going into the priesthood) that solves it.
Many mature people don't expect science to provide them with
absolute truths about things that matter most in their lives.
They simply, and without tears, turn to other disciplines to
find truths and virtues in living. Those disciplines may be
philosophy, history, theology, etc.
People turn to all kinds of things to find answers to their
problems. Most turn to the equivalent of throwing bones; they
turn to the bible, astrology, tarot, seances, etc. When people
have problems (conflicts) they want a nice, simple answer: "do
X". These answers are sometimes just what people want to hear;
the bible's "answers", for example, seem to work for you. PCT
offers the promise of a more efficient and systemactic approach
to helping people who are in conflict help themselves.
History proves that people lived and survived for thousands
of years without any knowledge of science at all
Yes. But they lived pretty miserably. Do you really want to
live in a world without high speed dentistry?
Humans and their nature are amazing; far more amazing and
inspiring than the discipline of science, which they created.
Agreed. And they are infinitly more amazing than the religions
and theologies they've created. But the fact is that people
created (controlled for) all these systems of beliefs because
they have brains that are capable of perceiving such systems
of beliefs. It's all perception.
Best
Rick
--
Richard S. Marken Phone or Fax: 310 474-0313
Life Learning Associates e-mail: rmarken@earthlink.net
http://home.earthlink.net/~rmarken/
[From Kenny Kitzke (991030.1930EDT)]
<Rick Marken (991030.1030)>
<It is the
opposite of the religious attitude, which is an attitude of
certainly and arrogance. The religious attitude requires blind
faith and an unwillingness to subject one's ideas to experimental
test.>
Others have tried to explain to you that you do not perceive what they have
perceived and therefore you can't understand them. I should be as wise as
they.
Sure, there are some religious people with an attitude of certainty and
arrogance. And, when contrary evidence from the Bible is presented and is
contrary, they look like fools. And, the same is true for certain and
arrogant scientists for their pet theories---like behaviorists who believe
the illusions of their own observations.
<Actually, PCT science shows that nothing -- not science, not
religion, not mathematics, not history, nothing -- can tell a
person what s/he _should_ do.>
I suspect what you mean is that PCT shows that humans have free will to
determine for themselves what they should believe or do. That is precisely
the fallen nature of human beings as described in the Bible--in total
agreement with PCT. I find no disagreement on this pretense between PCT and
the Bible.
I don't think PCT provides a clear understanding of where a person's
reference perceptions come from or for which variables they choose to
control. Got any demos on your web site for that?
<Nothing can tell you whether you should or should not marry a
person; only _you_ can "tell yourself".>
This is a perception of yours. Tell yourself is an oxymoron.
<PCT can help, though.>
And, so can the Bible.
<A
better way to solve such conflicts -- really solve them -- is
to change the systems that are creating the conflict. PCT suggests
that this can be done by "going up a level" and revising the goals
of the systems creating the conflict or waiting until the
persistent error created by the conflict leads to a
reorganization (like going into the priesthood) that solves it.>
Yep. That is exactly what people led by their human spirit do. They rise
above such system concepts of science and logic to find real solutions that
science cannot render.
<the bible's "answers", for example, seem to work for you.>
Have not failed me so far.
<PCT
offers the promise of a more efficient and systemactic approach
to helping people who are in conflict help themselves.>
Sure. To move up a level where they can grasp what to do. Just the opposite
of what you claimed earlier, "Actually, PCT science shows that nothing -- not
science, not religion, not mathematics, not history, nothing -- can tell a
person what s/he _should_ do."
<Yes. But they lived pretty miserably. Do you really want to
live in a world without high speed dentistry?>
This is pure speculation on your part. People may have gotten along just
fine without modern dentistry. Ever check out old skeletons for signs of
tooth decay?
<Agreed. And they are infinitly more amazing than the religions
and theologies they've created. But the fact is that people
created (controlled for) all these systems of beliefs because
they have brains that are capable of perceiving such systems
of beliefs. It's all perception.>
And, agreed, if man created the Bible. But, if the Creator created the Bible
through man, then one would draw a different conclusion about whether the
created or the Creator was indeed the more amazing being. It is all
perception.
Thanks for discussing human perceptions with me civilly, even though ours are
quite different at the highest levels. Its fun to exercise our unique human
nature.