[From: Bruce Nevin (Fri 920807 08:07:45)]
(Eric Harnden (Thu, 6 Aug 1992 14:19:35 EDT) ) ---
what appears
to me to unify the field is the question 'how can organization be seen as
a phenomenon?' put differently, how can order arise from disordered elements,
without explicitly planning for it?
the question most important to alife [is] 'how can anything organize'
Just before everybody disappeared for the annual Durango bash, I made
some proposals about the relation between control systems of different
orders of structural complexity. Let me restate these in this new
context as follows:
Structure of order n at least sometimes "emerges from" (is a
colligative property of) the behavioral outputs of control systems of
order n-1, each controlling its individual perceptions. The "gather"
program illustrates this nicely.
Control systems may perceive aspects of such structure, and may come
to control those perceptions. (Is there anything interesting going
on? There's something. I wonder why they're all gathering around
her? Must be some reason to pay attention to her. I'll think I'll
try to get as close as I can.) They may create and may come to depend
upon conventions and institutions that reflect these perceptions of
structure (e.g. build amphitheaters).
Control systems of order n (e.g. multicellular) can evolve from
structures emergent from control by control systems of order n-1 (e.g.
cells). There seems to be a vast evolutionary "distance" between e.g.
fungae vs. bacterial colonies (it seems likely that fungae are control
systems?) on the one hand and neural ECSs vs. neurons on the other
hand. The differences seem to reside in interlocking cellular
specializations, each constituting part of an environment on whose
reliable structure the other cells depend, and for the maintenance of
which they each in their several ways control.
My focus previously was on "communication" downward from order n to
order n-1: reorganization, I suggested, is the result of CSs of order
n-1 controlling to reduce environmental "toxicity", while concurrently
CSs of order n, so long as they continue to experience conflict (and
consequent chronic error), produce intrasomal effects that happen to be
"toxic" to their constituent elements of order n-1, more or less
localized around the order-n CSs in conflict and needing reorganization.
The focus now is on "communication" upward from order n-1 to order n,
and on the question of how the later may first evolve from and then be
continuously recreated and sustained by the former.
I know that others here (Bill, Greg, Gary, Francis Heylighen, others)
have been involved in the relation between control theory and evolution
and know a heck of a lot more about it than I. Do these hunches seem
feasible?
Bruce
bn@bbn.com