|
In looking through several of my father's books, I see aspects that
pertain to modelling all over the place. The concept of "feedback", the concept
of "optimality", the concept of "side effects" (related to "deadly infinite
regress"!), the discussions on anticipatory behavior, and his observation of the
fact that organisms unconsciously employ modelling relations within their very
organization (an "internal predictive model")... There are clues everywhere
as to what would help us figure out how to "build a better
model".
The book, "Anticipatory Systems" is devoted almost entirely to
discussions about modelling relations and what the qualities he was describing
in biological systems (anticipatory behavior) meant to the issue of modelling.
He says, "In it's most general terms, we found that a modelling relation between
systems is established through an encoding of qualities pertaining to one of
them into corresponding qualities of the other, in such a way that the linkages
between those qualities are preserved." Later, he says, "The raw material for
the construction of modelling relations is, and must be, the result of
observation."
My father talked a lot to me about the need for observation without
preconceived notions getting in the way of the truth. Again, that generated the
analogy that I've mentioned before, of what explanation the ancient Greeks
came up with for answering the question "Why does the sun rise in the east every
morning, and set in the west every evening?" Their answer was not quite correct.
It would be a few thousand years before someone would figure out what was really
going on. It takes a very open mind to see the true explanation in the
phenomena! Very often, these things turn out not to be what they "look like".
It's useful, in observation, to come up with as many different explanations for
some phenomenon as possible, such that, if you were studying the behavior of the
sun in relation to the Earth, you would come up with the explanation that
describes the true mechanics of our solar system. Frankly, I don't know if my
father would have done that, had he lived in ancient Greece with Aristotle. The
fact that Copernicus and Galileo both came up with pieces of the puzzle taught
my father something about observation. The actual truth sounds totally
ridiculous when you're coming from a mindset of the Earth being the center of
the universe, the Earth being flat, etc. Similarly, preconceived notions about
how organisms "work", that are not based on organization, are
going to make discussions of "anticipatory" behavior and "internal predictive
models" sound insane as well. But the truth is what it is, regardless of
whether we see it or not.
The anticipatory nature of biological systems is something that has
never been studied the way he said it needed to be studied. The organization of
complex systems in general, and organisms in particular, including the aspects
that have always heretofore been discarded as unimportant (the relationships
between the parts, for example)... is another thing that hasn't been studied the
way he felt it needed to be studied. All through "Anticipatory
Systems" he sketches various suggestions for further study and the reasons
why he believed they would be helpful. He also concedes, "However, here too, the
necessary ideas have only just begun to be formulated and are themselves still
in an essentially embryonic state."
He says, in summation, "A deep understanding of anticipatory
systems in general, and the character of the modelling relations which direct
them, will be central [to applications such as modelling]. The ubiquitous
character of anticipatory mechanisms in biology, and the emergence through
selection mechanisms, provides for us a vast encyclopedia for how to solve
complex problems (and also, equally usefully, of how not to solve them). This
encyclopedia represents a natural biological resource to be harvested; a
resource perhaps ultimately more important to our survival than the more
tangible resources of food and energy."
Ultimately, my instincts are that the hallmarks of biological
systems need to be built in to models of them. The relational aspect of living
systems has to be preserved in the models. The concept of function needs to be
built in. If complex (and anticipatory) systems are studied in the ways he
suggested, and a better understanding of their nature is derived from that
study, those aspects can also be built in to models of biological systems in
general. The laws of organization of complex systems seem to hold, whether that
system is an atom, an organism, a weather system, or a social system. What are
those laws? That's what he was saying we need to study if we want to create more
accurate models of these systems.
Judith
|