[From Bill Powers (2002.09.23.0545 MDT)]
Dan Palmer (2002.09.23 11:40 Melbourne Time)--
> ... a very short definition of self-correction
(i.e., negative feedback) Bateson once gave in an interview, where
self-correction entailed that "differences can be used to stimulate that
which will make them not different." The example he then gave
incorporated your rightly-emphasized comparator-clause - "The difference
between the way I want to go and the way I am going can be used as
information to touch off events which allow me to correct."
This is OK, but it tends to slow down and fragment the picture of the loop
to the point where the way most of them operate is unrecognizable. You get
the idea that first I see a difference, then I use this information as a
trigger that sets off automatic processes that correct the error, and then
it's all done. Most real processes of control involve errors that are
driving actions that are bringing perceptions closer to their desired state
which is resulting in continuously diminishing error, all parts of the
process overlapping in time. This process continues while the perception
approaches the reference condition and the error gets smaller and the
action changes according to disturbances or the demands of environmental
properties or both, thus causing the perception to approach the reference
condition .... Everything in the loop is changing at the same time, rather
than a sequence of events taking place. Hard to talk about, but well worth
understanding.
I always liked the example of the difference in color between a
chameleon's body and background contributing to the elimination of that
difference via a change in body color. The chameleon's job is to avoid
becoming information - and thereby making various differences to some
predator's processes of digestion ;-).
That would be a very good example of a control system if we could prove
that the chameleon is capable of sensing the difference in color between
its body and the background. I've never thought of that before, but
obviously if the chameleon's color comes to a close match with different
backgrounds of varying shades, _something_ has to be detecting the
difference. I would suspect the chameleon's own eye, unless its skin is
photosensitive.
The alternative to a control system would be a stimulus-response system in
which each different shade of background caused a corresponding skin color
to be produced, which is roughly what biologists think now, as far as I
know. That's possible, of course, but then we'd be looking at a very slow
evolutionary contriol process for adjusting the amount of the response
until a match, or an effecty of a match, were achieved. How about arranging
to place a chameleon on a series of backgrounds of varying colors or shades
and using a color photometer to record its skin color relative to the
background? Has anyone ever done that? Has anyone ever done that while
presenting _other_ colors to the chameleon's eyes?
The interesting thing, of course, is that ideally the contrast should be
eliminated in the eye of the predator, not the chameleon's eye, so if the
eyes have different spectral responses we would see a match in terms of
the predator's color vision rather than the chameleon's or the meter's. I
rather suspect that it's the chameleon's color vision that determines the
result and that we have a true control system. On that hypothesis I would
guess that a chameleon's color vision has evolved to be similar to that of
its predators.
Like you, I've found a lot of sense in Bateson's proposal that mind-body
dualism is resolved if mental organization (Bateson's creatura) is
viewed as _immanent_ in matter (his pleroma). The same point has been
made by Maturana and Varela in terms of organization and structure and
by John Dewey in terms of life-activity and environmental medium.
Interestingly, all three sets of scholars ended up arguing that
organism, as ongoing organization, literally extends beyond sensory and
motor surfaces (i.e., the skin).
I don't see why we have to resolve it. Who said it's a bad thing in the
first place? I detect a rather pronounced difference between my mind and my
body -- I've never known my stomach to think, or my fingernails to have
goals, or my muscles to decide where I want to walk. I really think the
whole argument comes from scientific nervousness about anything that looks
like mysticism or religion.
Also, I think I can distinguish rather easily between my body and its
environment. After all, I move the same body around through all sorts of
different environments, and when I do, it is the environment that changes,
not my body. You have to retreat into the upper reaches of abstraction
before you have any problem with that, I should think. If you get abstract
enough, of course, you can no longer tell the difference between an
elephant and an elevator, but that would be viewed as a pathological state,
I believe, like the man who mistook his wife for a hat.
For similar reasons, I don't like thinking of mind as a property of matter.
If it were, we would see signs of all matter having goals, perceptions,
error signals, computational methods, and so on (all the things I think of
as having to do with mind), and that is far from true of most matter. It's
not true even of most matter inside a living organism (the water, for
example). To say it's a property of a piece of matter, you must show that
you can take a piece of matter and demonstrate that it has this property,
and of course you won't find mind in _any_ "piece" of an organism. It
exists only in the _intact arrangement_ of parts relative to each other,
and that, of course, is not a property of any of the parts by itself. You
can alter the arrangement of the parts without (necessarily) altering the
parts, given fine enough dissection techniques, but what's left will not
necessarily be alive. It certainly won't be the same organism, or mind.
As far as I can see, organization is an immaterial aspect of every
organized thing, living or not. Some organizations amount to minds, but
most do not. As in all categorical definitions, when you look at them too
closely they develop fuzzy edges -- does my lawn mower have a tiny bit of
mind because it contains a speed control system? I wouldn't say so.What
about my pocket calculator? No, it doesn't contain any control systems --
except the voltage regulator, but that doesn't affect the computations
being carried out. B. F. Skinner? Yes, of course, but if Skinner worked as
he claimed that organisms worked, he would have to say he had no mind.
It's fun to speculate, but I've never been turned on by that sort of stuff,
or rather, my liking for it is satisfied by a very small amount of it. It's
like a Westerner eating Chinese food -- two hours later, all the unsatified
reference signals start becoming noticeable.
Best,
Bill P.