[From Bruce Nevin (2003.0201.1607 EST)]
I need some help with wording.
Consider this statement: "Behavior is the control of perception(s)."
[...]
Help me out here. Is there a way to say that behavior is the control of
perception without saying that?
"Behavior" is generally understood as observable activity. It's even used in this sense when speaking of things with no (discernible) volition: "the behavior of a ping-pong ball under these aerodynamic conditions". This confounds control actions with the action and reaction of physics - Bateson's creatura and pleroma.
The canonical formula "behavior is the control of perception" limits the reference of "behavior" to control actions. The behavior of a ping-pong ball is not control actions no matter what the aerodynamic conditions.
If you toss a cat out of a window, its twistings in the air are control actions, but its falling, like the falling of a ping-pong ball, is not a control action. On this view, the twistings about are behavior, but the falling is not.
This nicety of distinction is not shared when people are accustomed to expressions like "Variable moments and changing magnetic behavior of thin-film FeNi alloys", "Asymptotic behavior of variable-coefficient Toeplitz determinants", "FakeHash.pm is a Perl module that simulates the behavior of a Perl hash variable", and so on.
So we need to stipulate that we are talking not just about observable "activity" in the sense that someone might talk of the activity of leaves blowing around or wave action or ping-pong balls moving through the air, but rather about activity on purpose. "Purposeful activity is the [means of] control of perception."
We could argue that all observable actions of living organisms are control actions. To sustain this view, we would have to deny that the falling of the cat was an action, just as we would deny that changes in the moments and magnetic properties of alloys were actions.
We could argue that phenomena described as behavior of alloys, behavior of Toeplitz determinants, behavior of a Perl hash variable, and the like, are not behaviors at all. But these common uses of the word are metaphors, analogic extensions of the basic meaning of "behavior". In common usage, the objection that marks them as metaphoric is that these things are not living; for us, it is because they are not purposeful. For us to speak of the behavior of a thermostatic heating system or cruise control is not metaphoric, but in common usage, it is.
Going back to the common sense of the word "behavior" as "observable activity", there's an interesting consequence for the word "observable". In control theory, we shift attention from observing activity to observing a variable that resists disturbance - an absence of "activity" in the variable where, if it were a ping-pong ball, you could reliably expect "activity" to reflect the disturbance. Observing resistance of a variable to disturbance, we infer control even when no control actions have been observed. We can even infer control actions when they are not observable: imagine a situation where observation would require invasive procedures that would disrupt control or even kill the organism.
This is the crucial shift of attention that makes all the difference in communicating what control theory is about.
Those "observable activities" that have the effect of resisting disturbances to controlled variables are obviously control actions. "Observable activities" that are intended but fail to resist disturbances, such as the running motions of an animal when it is picked up, are also understood to be control actions and therefore "behavior". By hypothesis, all activities (other than those of the ping-pong sort) are control actions, either resisting disturbances to CVs or attempting to do so, though to prove that or even to prove the distinction would be difficult indeed.
The slogan "behavior is the control of perception" steps squarely in the middle of this in a rather demanding way. Small wonder that we have to explain ourselves. If you are prepared to, that may not be a bad thing.
If not, then an alternative to saying "behavior is the control of perception" is to say "behavior maintains perceptions at reference levels, that's what it's for" and just persist in careful PCT usage without comment or explanation, focusing instead on the "nuts and bolts" of how it's done (Isaac Kurtzer 2003.01.31.0930). You can even smile and nod when they say that organisms "control their actions" etc., agreeing that the reason they seem to do so is by maintaining certain perceptions at reference levels (Bruce Gregory 2003.0130.2024).
/Bruce Nevin
···
At 07:08 PM 1/31/2003, Fred Nickols wrote: