Empirical PCT

[Tom Bourbon 941014.1736]

A few thoughts inspired by watching recent activity on csg-l.

PCT appeared in published form over 30 years ago. CSG-L appeared
a few years ago. Many thousands of people have read about PCT in
one or the other of those media. Several hundred people have
spoken with us, or corresponded with us, about PCT (by "us" I
mean mainly Bill Powers, but also some of the few other people
who do PCT research and modeling). Many who have spoken or
corresponded expressed positive opinions of PCT. All well and
good, but something is lacking.

What is missing in this superficially rosy picture is an increase
in the number of people doing quantitative PCT research and
modeling. Rick Marken has written often (and recently) on this
subject. From time to time I have joined him in his lamentations
and in pleading that others take up some of the load, but all to
no avail.

Some supporters of PCT profess a lack of skills for modeling.
But as a rule in life, each person gladly acquires the skills necessary to
do the things he or she most deeply wants to do.

Many advocates of PCT protest that they believe profoundly in the
importance of PCT, but they simply do not have the time to add
PCT research to their already busy schedules. But one of the
best measures of the importance a person attaches to any subject
is the time the person allots to that subject.

Many supporters of PCT claim that PCT research and modeling only
deal with the lower two or three levels of the hierarchy, but
that they are more interested in "the higher levels." But their
characterization of previous work is not true, and even if it
were, the best remedy would be for the protestors to do the
additional work themselves. Otherwise, the handful of people
doing empirical work might never get around to the topic of
greatest interest to the passive supporter.

Many advocates of PCT claim that the traditional sciences contain
rich stores of data from research in which experimenters were
"really doing PCT" but simply didn't know it, and that, to win the world
to their cause, PCT supporters need only reinterpret those
data and show the original researchers that PCT answers all of
their significant questions. But traditional experimenters
almost never study their subjects as controllers and their data
are almost always of a sort that does not allow us to identify a
subject's controlled variables or disturbances.

Many supporters of PCT offer up logical arguments and proofs to
demonstrate the futility of empirical studies: there are too
many degrees of freedom in behavior and in teh environment, consequently
empirical studies of control are futile; any successful model of control
is only one of an infinite number of equally successful models,
hence no successful model is significant to the development of
PCT; and so on -- PCT ought to be an exercise in pure logic.
What can I say?

The list goes on. There are many excuses, all offered as reasons for
not engaging in PCT research or modeling. If its advocates and
supporters do not participate actively in quantitative research
or modeling or both, then PCT will not survive for long as a
science and the many individual excuses for avoiding active
participation will no longer matter.

Later,

Tom