[From Rick Marken (930406.1200)]
Martin Taylor (930406 12:00), on the importance of the PCT - IT debate,
says:
They [living systems] must control, which means that they must actively
oppose the effects of the external influences. To do this, they must have
an energy source that allows them to act strongly enough to oppose the
potentially damaging influences that they are likely to encounter,
Agreed so far.
and
they must have a way of getting information that allows them to direct
the action appropriately to oppose these influences.
This is the heart of the problem. To the extent that this means that
the perception of temporal variations in external influences on
controlled variables (distrubances) is used to direct system outputs
that oppose these influences, then this is precisely, unequivocably
and demonstrably wrong. And it is wrong in precisely the way needed
to preserve an SR (perception causes output) view of the behavior
of living systems. THAT's why this debate is going on; it's why
it's important.
Rick and Bill have been casting the question in terms of WHAT information
is acquired, whereas I keep trying to bring the discussion back to
HOW MUCH is needed and HOW MUCH is possible under different conditions.
If you will just admit that perception of external influences on
controlled variables does not guide output, then I have no dispute
with whatever you want to say about "HOW MUCH information is
needed or is possible under different conditions".
We have, so far as I can see, no difference of opinion on how the control
system is structured or in the signal flows. No one has claimed that
knowledge of the amount of information available (a scalar quantity) is
sufficient to reconstruct a waveform (a vector quantity). The only
relevant claim is that if the vector quantity can be reconstructed, that
is prima facie evidence that information about it was available.
Not quite. It's not just "if the vector quantity can be reconstructed"
but "if the vector quantity can be reconstructed BY THE CONTROL SYSTEM
ITSELF". There is no dispute that, if p = g(o)+ d, you can
reconstruct d given p and g(o). That's just algebra. The point is
that the control system only has access to p. We were throwing in
o = O(e) for free. I showed that even giving o was no help; you still
cannot reliably reconstruct d because o is not a mirror of d when
g( ) is not a constant = 1.0.
(Bill Powers 930405.1330 MDT) --
Under the best of circumstances, people who believe in
X learn about control theory, and say "Wow, sure, PCT is the most
revolutionary thing since sliced bread, but you know that we Xers
have been saying the same things all along, and when you get
right down to the science of it, X is fundamental. We're glad to
see that PCT fully supports X."
Martin Taylor replies --
How can anyone refute that? I don't believe it of myself. but it could
be true. What I think is that the last sentence should read "I'm glad
to see that X fully supports PCT and can be used to deepen our understanding
of PCT."
This is also why this debate is important to me. I am far more
interested in having people understand PCT than in having them agree
with me. I know that Martin is a PCT fan and if I dropped this
"information about the disturbance in perception" debate we
could walk off into the sunset nodding in agreement. But what I
am hearing Martin et al saying about PCT is fundementally wrong --
and it is wrong in just the way needed to preserve an SR conception
of the operation of a negative feedback system. So if what I am
hearing Martin et al say about control systems is correct (and
everything said and done in this debate so far convinces me that
I am correct) then I feel compelled to try to straighten it out.
I hope I don't alienate Martin et al and drive them away from PCT;
but I, personally, would rather have just a few people doing PCT
who understand it correctly than lots of people doing it who
have it all wrong (like Carver and Scheier, Bill Glasser, etc).
Maybe Martin does understand PCT correctly; but if that's the
case then why do I keep getting the impression that he is
saying something like:
Perception of temporal variations in external influences on
controlled variables (distrubances) is used to direct system
outputs, ie. o = f(p).
This is just not true! And it is IMPORTANT that it is not true.
Because the fact that it is not true of control systems (as
demonstrated consistently by PCT) conflicts with the most
basic assumption of conventional behavioral science. I you
can't see that o = f(p) is not true, then you can't see why
PCT is revolutionary.
If IT "fully supports" PCT than it should support some of
the basic facts of PCT. There are two main facts of PCT that are
of fundemental importance to understanding the behavior of living
systems:
1) p = r
2) o = -g(d)
These relationships hold for relatively high gain, negative feedback
control system. The first equation says that the perceptual input
to a control system depends on the value of the reference signal.
In a control system, controlled perceptual input, p, is the dependent
variable, reference input,r, is the independent variable. The
second equation says that output (the proximal, environmental
effect of the system) depends on the disturbance (d) and the nature
of the relationship is determined by the feedback function, g(),
that relates output to perceptual input. Note that p NEVER APPEARS
IN EQUATION 2); output does not depend on perception at all;
the relationship is directly from d to o. The relationship between
o and d is a result of the properties of the negative feedback
loop (which we have reviewed in grisly detail in this debate); it
does NOT result from any dependence of o on p. This lack of
dependence of o on p is what I was trying to get at in the "no
information about the disturbance in perception" debate. The debate
should not really have been necessary because my point is made by
equation 2).
If, as Martin implies above, IT "fully supports" PCT than IT should
"fully support" this fact about control systems, which is described
in equation 2), namely, that outputs depend on the disturbance, not the
perceptual input, p (no matter how illogical this may seem to him).
But Martin has been arguing that equation 2) holds because o = f(p) --
a sensible intuition but, unfortunately, not a correct description
of how control systems work. [NB. that in the two equations above,
p is already busy being a dependent variable; it can't be
an independent variable, too. In other words, in a control loop,
r tells p what to do; p can't, at the same time, be telling o
what to do].
So please don't tell me that other theories support or are
consistent with PCT before you know that it is really PCT that
they support or with which they are consistent.
Best
Rick