[From Bill Powers (920524.1230)]
Greg Williams (920524) --
I omitted one point that flitted through the space between my ears and out
again.
Your main point is that the PCT model doesn't contain any basis for
assuming a self with a free will. I think I can dispute this point. You
say,
PCT models imply that an individual's current control structure is the
result of past interactions between the (changing-over-time) control
structure and its niche, and NOT the result SOLELY of an autonomous
dynamics of the control structure (which would mean that when
reorganization stopped, in any particular case, would be independent of
the niche, denying the very reason for reorganization).
The implication here seems to be that the current control structure is some
REGULAR function of past interactions with a niche. But reorganization
occurs precisely when interactions with the niche lead to loss of control
-- that is, when the current regularities in the interactions are
insufficient to preserve control.
The reorganizing system is effective because the changes it institutes do
NOT depend in any regular way on the current organization or the current
niche. The whole point is to break out of the conflict or the circle or the
failure -- the local minimum -- by trying something NEW. So the idea of
tracing the current organization backward, while all right in a general
sense, is wrong if it implies any predictable course of development.
Reorganization breaks the cause-effect chain.
Reorganization -- that is, the actual output effect of the process -- is
independent (save for the frequency with which reorganizations occur) of
any prior causes. The outcome of reorganization, to be sure, has to be such
that intrinsic error is corrected; if it's not, reorganization simply
continues. But there are uncountable ways of reorganizing that would result
in correcting intrinsic error, so that result is not a constraint on any
particular act of reorganization. In fact, one episode of reorganization is
just as likely to make matters worse as it is to make them better (unless,
of course, there is an unsuspected systematic component in it). The
statistics of reorganization are very different from the statistics of
stochastic -- but on the average systematic -- casuation. Reorganization
will work even when the changes it produces show no trend at all in any
direction.
Also, we mustn't forget that what makes reorganization effective are not
the individual reorganizational events, but the selection effects that
terminate reorganization. All that is required is the existence of
something that can say "There! That feels better." Or, of course, something
that says "Oh, no! Reorganize!" In fact, we could accept a mechanical
randomness generator that actually does the reorganizing acts, and limit
free will to the single act of triggering a reorganization. The "awareness"
part of free will would then superimpose its judgments of what is
acceptable and what is not on auytomatic judgments about such things as
body temperature and state of nourishment. Thus free will could select for
outcomes acceptable to it simply by causing reorganizations until the
result is acceptable. The grounds for acceptability need have nothing to do
with the niche.
By the way, for free will to exist, it isn't necessary for it to be the
SOLE cause of anything. If it's simply A cause of SOME things, that's
enough to throw the monkey-wrench into the causal gears.
···
---------------------------------------------------------------
Best,
Bill P.
[FROM MARY POWERS, RE MEETING]
Here is a repeat of the CSG conference announcement and
registration form.
A SERIOUS GLITCH has developed regarding transportation. All
airlines are fully booked on Sunday, Aug. 2 from Durango to
Denver. One alternative is to stay over until Monday (but book
NOW). I am going to check into flights from Durango to
Albuquerque. Another possibility is to fly out of Farmington, New
Mexico, which is about 40 miles away, but if you fly in and out
of Farmington you'll probably have to rent a car. Ditto if you
fly to Albuquerque - about 200 miles. I will find out if Saturday
is still open from Durango if you must leave then. More on all of
this in a couple of days.
CONFERENCE ANNOUNCEMENT
The 8th annual meeting of
THE CONTROL SYSTEMS GROUP
July 29 - Aug 2
Fort Lewis College
Durango, Colorado
This is an interdisciplinary conference on the application of
Control Theory to the behavioral, social, and life sciences.
The conference is small and informal. There will be 7 plenary
sessions (Wed. evening, morning and evening Thurs., Fri., Sat.).
Afternoons and Sunday morning are unscheduled, and are available
for more specialized and technical one-on-one and small group
meetings as interest dictates, or for rest and recreation.
While all participants share the common language of Control
Theory and therefore communicate unusually well across
disciplines, it is probably a good idea to keep in mind the wide
diversity of fields represented when choosing discussion subjects
for the plenary meetings.
Our best sessions take the form of a brief presentation of
research, problems, insights, future plans, etc., followed (and
frequently interrupted) by questions and open discussion. We
encourage participants to bring papers for distribution (25-30
copies) which can be read by participants before any presentation
in order to have an informed discussion. We discourage the
reading of a long paper or a faithful reproduction of
presentations to one's classes, clients, or dissertation
committees! For those who need academic brownie points, a
distributed paper is considered as presented. No one has to give
a talk, and a number prefer simply to listen, learn, and comment.
For those with computer demos, there will be an AT and (we hope)
a MAC. There will be an overhead projector with (again, we hope)
a computer projection plate. There will also be a VCR and
monitor.
Durango is in southwestern Colorado at an altitude of 6500 feet.
It is a major vacation area, with the San Juan Mountains to the
north, Anasazi Indian ruins, a narrow-gauge steam railroad, river
rafting, etc. July-August weather is usually hot and dry, with
afternoon rain common, and cool nights. Participants and guests
can stay over at the college for up to three nights FOLLOWING the
meeting. No space, unfortunately, is available before the meeting.
Registration
The cost of the meeting is $220 for a single room, $180 for a
double. This includes the $40 membership in CSG which supports
the publication of Closed Loop, plus all other expenses including
meals from Wednesday dinner through Sunday breakfast. Sunday
lunch is extra, and so is the airport taxi ($12 each way). The
student rates are $185 and $145. Student aid is now exhausted.
The fee for a guest sharing a room with a participant and
including all meals is $125. The charge for a guest without meals
is $60, and meals can be bought on campus (breakfast $4.50, lunch
$5.50, and dinner $6.25) or off-campus. The no-meals fee does
include dinner Friday night - the banquet/business meeting. The
charge for staying over after the meeting is $20 per night for a
single person in a room, $12.50 per person sharing a room. No
meals included. For more than one guest, or any other special
arrangements such as tourist information, day care for children,
etc. please contact Mary Powers (address below, or 303-247-7986
or at powers_w%flc@vaxf.colorado.edu (the symbol following powers
is an underscore and the symbol following w is a percent sign).
For those specifying in advance that they are attending half or
less of the meeting, any refund from the full cost will be
returned after the meeting.
Registration is $50 of the total ($15 for students), and MUST be
received by July 13. Registrations after this date CANNOT be
accepted. There will be no refunds of registrations except for
students applying for assistance (but membership will be paid).
THE BALANCE IS DUE BEFORE JULY 27. PLEASE DO NOT PLAN TO PAY
AT THE MEETING.
Make checks out to The Control Systems Group and send to
Mary A. Powers
73 Ridge Place
Durango CO 81301-8136
The Control Systems Group
8th Annual Meeting
July 29-August 2, 1992
NAME_____________________________________________________________
ADDRESS__________________________________________________________
PHONE (evening)__________________________________________________
will attend (circle) WED(pm) THURS FRI SAT SUN(am)
will also stay (circle) SUN MON TUES nights
room: SINGLE DOUBLE (share with_____________________________)
GUEST(S) name___________________________________________________
________________________________________________________________
guest meals (check) ALL__________ BANQUET ONLY___________
Banquet drink preference (iced tea provided - main course will be
roast beef and vegetarian lasagna):
WINE red_____ white_____ BEER_____ POP reg_____ diet_____
Coffee break preference (majority choices to prevail)
morning: COFFEE_____ TEA hot_____ iced_____ FRUIT JUICE_____
evening: COFFEE reg_____ decaf_____ TEA hot_____ iced_____
LEMONADE_____ FRUIT JUICE_____
EQUIPMENT NEEDS:
AT computer_____ MAC ______ Overhead Projector_____________
Projection Plate___________ VCR/Monitor__________
OTHER (specify)_____________________________________________
(All three waivers for student fees have been claimed. One may become
available again, so enquire a week or two before the deadline of
July 27th).