[From Bill Powers (2000.01.16.1453 MDT)]
Mary has been trying to reply to Jeff Vancouver, who seems to think that
Jeff Pressing's paper, "The referential dynamics of cognition and action,"
is something to take seriously. I have avoided commenting on it because
even trying to do so overwhelms me with the futility of what I've been
trying to do for 47 years. Old is bad enough. Old and depressed is MUCH
worse.
Consider the very first two sentences in the paper:
"Some human behavior, such as the knee-jerk reflex, is automatic and
reflexive. Once triggered, it unfolds according to a preset muscular scheme."
That statement is stupid and completely false; the knee-jerk reflex is a
control action which "unfolds" according to the way the disturbed negative
feedback control system is organized. It is not "triggered" and it does not
proceed according to a preset scheme of muscle contractions. So what do I
do with the next statements, the truth of which depends on the truth of
this one? And if the next statements are also hogwash, what about the
statements that follow them, and so on to the end of the paper?
Consider the equation on p. 715:
(1) y(t) = S[u(t)]
where y is the vector output of the human system and u is the input vector.
Starting with this equation, Pressing basically says that feedback from y
to u doesn't matter if you just leave it out of the "general" equation.
Even though he thinks he manages to re-introduce it in his Discrete Control
Equation, he fails to see that any general description of the human system
_must_ include the dependence of input on output as well as the dependence
of output on input. That dependency is not an option for the modeler to
take or not to take. It is a fact of nature, and if your model doesn't
include it, you have no model but only a fantasy. If "feedback is too
slow," the organism will simply have to suffer the consequences, because
the feedback is there whether it is too slow or not. Of course it is not
too slow. It is exactly fast enough to account for the behavior we see.
That's just the tip of the iceberg, but I'm already sinking from just
brushing against it. I really can't go on. It's obvious that Pressing is
persuasive enough to convince the editors of Psych Rev that he has
something worth communicating; I guess that shows me how much the
publication of the "Spadework" paper in the same journal actually meant,
20-odd years ago. The last illusions die the hardest.
Best,
Bill P.