[From Rick Marken (920802.1400)]
CHUCK TUCKER (920902 12:23:30) --
The sentence he did not cite was: "His behavior (i.e., "when a man
controls himself") in so doing is a proper object of analysis, and eventually
it must be accounted for with variables lying outside the individual himself."
(pp.228-229)
This makes Greg's point (that Skinner used 'control' in a non-PCT sense) quite
clear (to me at least)
It doesn't make it clear to me. Do the "variables lying outside the
individual" control behavior or influence it or what? All we have above
is a new description of what these variables do ("account for") and a new
description of the behavior that these variables account for ("controlling
oneself"). Do these variables influence the behavior (is "controlling oneself"
just what happended to result THIS TIME from the presence of these
variables) or do they control the behavior (is "controlling oneself" the
only result that could have occurred -- the variables would have somehow
compensated for any deviation from that behavior).
Skinner just didn't really know the meaning of "control" so he used the word
in whatever way made sense in context. If he knew what control was and
tried to figure out how it could be done, all the verbal confusion would
have disappeared -- he would have discovered PCT.
I would suggest that the conversation will be clear if you restrict the term
'control' to "self-control" that can only be performed by "a negative feedback
purposeful process" which is either part of a particular organism or
restricted to a single organism
I don't like "self control" as a replacement for control. There may very
well be a perceptual variable that could be called "self" that is
controlled; but that is only a small aspect of control, and an unconfirmed
one at that. I think the word control works just fine in the discussion of
behavior control. Behavior refers to variable results of action. Control
refers to variables that vary far less than would be expected given the
variance of influences that are know to affect them. Thus, some behaviors are
controlled and some are not. PCT says that controlled behaviors (controlled
results of action) are controlled by systems inside the behaving
system itself. If another system tries to control the results that are
controlled by the behaving system (ie. if the other system is trying
to control behavior), then the two systems will most likely be in conflict
(if they have even slightly different references for the controlled results
of action).
penni sibun (920902.1100) --
it looks to me like all this proves is:
1) you're more inclined to believe bill than avery
Not really. But Bill does have an awfully good track record. I've
tested nearly all Bill's claims about PCT (with experiments and
computer modelling) and he's right over 99% of the time (I have
caught him a couple times, but only when he's speculating off the top
of his head).
2) you will happily believe things w/o going to the primary source to
check for yourself
Not only do I go to the primary sources (when I deem it worth the effort based
on what I've heard about them or based on excerpts that I have read from them
-- as with the Chapman book), I also TEST claims. I don't just believe what
I read (at least, when it is testable); I try to demonstrate it to myself and
see if something is not what it seems. I believed Bill without going to
the primary source (Chapman) because I've worked in the field of perception
for 20 years and if someone were really doing something any better, in terms
of modelling perception, than we've been doing than I would have been told
about it already. If I had to read every piece of, er, stuff, that people
thought of as a great breakthrough I wouldn't have time to read anything of
real value -- like Jane Austen. So, indeed, I filter -- through the fine
strainer of PCT.
And that stuff about being right every time -- it was sarcasm. I was
kidding (sort of).
Regards
Rick
ยทยทยท
**************************************************************
Richard S. Marken USMail: 10459 Holman Ave
The Aerospace Corporation Los Angeles, CA 90024
E-mail: marken@aero.org
(310) 336-6214 (day)
(310) 474-0313 (evening)