[from Mary Powers (9802.14)]
(but finished 9802.16)
Because of being off the net for a few days, our CSG mail for the beginning
of Feb. finally arrived in a gigantic lump of some 150 double-column pages.
Reading the whole thing at once was a depressing experience. Practically
everyone was complaining about something or somebody.
The main message that I got was this: "PCT is a wonderful theory. However,
it is too (simple-minded/undeveloped/narrow/you-name-it) to apply to the
field I happen to have (invested my life in/care a lot about/did my
dissertation on/have been teaching my students, etc.)"
Thus we found, just in the first 10 days of Feb., 1998, that PCT can't say
anything that's as useful as observations of social relationships; that
tracking experiments are merely about tracking; that correlations of .95+
aren't any more interesting than correlations in the .60's; that there is
nothing questionable about EAB experiments for PCT to call into question;
that there is nothing questionable about psychophysics experiments for PCT
to call into question; and that one can take a fuzzy term like prediction,
define it several different ways, claim that PCT can't explain it, and
criticize PCT because the concept loses its fizz in the context of a control
system model.
The from the sidelines we get remarks about PCT being this little religion
with fanatical adherents who ought to lighten up.
Here is my take on all this:
CSGnet was established 7 1/2 years ago as a forum for people interested in
the PCT model. The idea was to enable PCTers around the world to stay in
touch and share their ideas and work. Anyone new coming onto the net was
and is very welcome. PCTers are happy to explain, answer questions, etc.
All we ask is a little homework -- such as reading the web page, doing
demos, reading publications -- and an open mind.
What we keep getting, however, are individuals who think the PCT model is
interesting, but who continue to cling to the idea that it can be integrated
with other research (especially research that each individual has been
involved with, such as psychophysics and EAB) and other approaches in
general. I don't want to keep repeating history but apparently it's not
getting through! PCT has been there and done that - or rather it has been
done to PCT. By Carver and Scheier. By William Glasser. By R. Lord and
his various co-authors. By Todd Nelson. Etc., etc., etc. For over 20
years. All these people, in a well-meaning way, tried to explain control
theory in terms of other theories, or other theories in terms of PCT.
Without exception, the loser in these encounters was PCT. The crucial
features of control systems - HOW they control and especially WHAT they
control - always gets lost in the translation. Because all the other
theories of the organization of living systems that have been proposed so
far, whether the model and principles employed are implicit or explicit, do
not take into account the fundamental and rather unfamiliar properties of
control systems.
All these people thought they were doing control theory a big favor; that it
would help spread appreciation of control theory to show folks from
different orientations how it works in their terms, in their language, with
their models. The consequence has been only that some of the language has
been adopted, as trendy jargon, or in some cases a feedback loop is added
here and there.
And the moment anyone runs into a little snag, usually based on their pet
concern, then they blame the model's inadequacy - since to them a PCT block
diagram is just a conceptual sketch - instead of questioning their
understanding of control theory. Thus we had, a few years ago, R. Lord
drawing a gratuitous arrow up from the output to the reference signal,
labelled "decision-making", because that just had to be a discrete
phenomenon (it had a name, didn't it?), that was mysteriously or stupidly
"left out" of the standard control diagram. I think Jeff Vancouver's
attachment to "prediction" is a somewhat similar case. I hope not.
A few years back, an extremely intemperate attack on control theory came
out: "The Emperor Is Naked" by Edwin Locke. Locke attacked all forms of
control theory in psychology pretty indiscriminately, and in fact was
heavily sarcastic about the idea that there were a couple of flavors of it.
But his basic idea was that control theory, as adapted by Carver, Lord, etc.
offered nothing to contemporary, traditional psychology except jargon, and
in that assessment he was absolutely right. Unfortunately he didn't care to
entertain the idea that PCT had something different to offer (and curtly cut
off my attempts to discuss this with him).
The reason I'm bringing all this stuff up is that we have now been informed
at some length (and not for the first time) that we are a little cult trying
to keep our religion pure, and I want to get across what has happened and
what we expect will continue to happen when people think they see how to
connect control theory to the rest of the scientific world and oh so
generously set out to do so -- as if it wouldn't have been done years ago if
it were possible!!!
Control theory is its own self -- a cuckoo in the nest. It isn't going to
grow up to be an EAB chickadee or a psychophysical bluebird. Chickadee and
bluebird fanciers have got to quit hoping that this nestling is a brother to
the others. It isn't. In fact, some day it's going to throw those others
right out of the nest 
Mary P.
