|
The reason I seperate the definition of information in a natural
system from the definition of information in artificial systems (those created
by humans, where the internet is an example) is because humans are
short-circuiting (pardon the pun!) the evolutionary process and creating systems
that are very much grafted together in bits and pieces from all over the
evolutionary scale as far as complexity is concerned.
I don't know if it is actually possible to learn about
human-created systems by studying them directly, especially when we don't
understand the naturally occurring ones as yet... But I think my father was on
to something when he said we can learn about the human-created ones by studying
the natural ones. The natural ones make the "natural laws" clearer. Natural laws
are universal, supposedly, so they will be present in the behavior of human
created informational systems-- just harder to see through all the static of
human creation surrounding them.
Judith
----- Original Message -----
Sent: Tuesday, February 03, 2004 6:57
PM
Subject: Re: [ROSEN] Information and
Volition
Judith,
see below
Judith Rosen wrote:
I've been thinking about this while I slept, apparently... It's
still on my mind this morning! The point at which complexity or complex
systems seem to make the change where natural phenomena become
information seems to be the same threshold at which complexity
creates life. Certainly, I think. As mentioned
before (so I won't belabor the point again) I personally speculate that we
won't be able to fundamentally distinguish complexity and life, except when
discussing life form, i.e., its more elaborate realizations above the quantum
level. This is obvioiusly not agreed by all on this list or stated by Rosen,
but I remind you of this interpretation because it relates to why I keep the
information concept throughout nature, and also to continue to fish for a
valid argument against this idea. The mechanists, of course, would dismiss it
out of hand, but I am curious that even the Rosenites find it uncomfortable.
My guess is that it is a habit of thought that prevents its acceptance, and
that no clear distinction will be found in principle.
However, it seems to me that this is only true for organically
derived, evolved, complex systems. Again, my
question would by why? What principle creates that distinction? Or are
suggesting just a convention for using the term? In that case the more
primative systems would be exchanging something else, what? Interaction? How
do they follow natural law without information being transfered about what
those laws are? it is a very fundamental question.
The systems human beings create, like a computer internet for
example.... seem to have a whole 'nother relationship with phenomena, data,
and "information".
In organisms, the stability of the organization of that level
of complexity (ie; a living organism's ability to
maintain its internal environment within necessary
parameters) seems to be what also makes possible the
generation of relationships that turn outside phenomena into
information. In other words, this is when percepts begin to have
indirect causal effects on organismal processes. It looks to
me (not being an ecologist or biologist, just an observer) that the
difference between an automatic process and a process where
information has been created and utilized is the level of complexity of
the system. As in my earlier post, I'd say it is
not the level of complexity, but the scale at which the information applies.
Physics allows only universal information to be causal, i.e., natural "law."
Rosen suggests that information relating to local systems can also be causal.
That seems to me to be the operative distinction.
The higher the level of complexity in an organism,
the more capacity for relationships there is, and information plays a bigger
and bigger role up the scale. So, yea, the role of
information would increase as systems build on systems and relate
hierarchically, holarchically, and in modeling relations. In humans, arguably
a "more" complex system, information takes on an almost dominant role. But I
see this as a continuum with no lower or upper thresholds.
One question I have about smaller complex systems, like the
atom (which bears on the "three-body problem"): What are the
similarities and differences in the relationships formed between an atom and
its environment compared with the relationships between an organism and
ITS environment? I'm being the succor here and
taking the bait. Treat it as a WAG (Wild-Assed-Guess).
I'd say the
relationships are fundamentally the same, but operationally different. They
share the property that any two system interact via functions. What one system
does in another system is a function in that system. If ecosystem produce
food, that is a function with respect to the human system, where it is defined
as food and has that use or purpose. Otherwise it is just plant or animal
matter. When a particle takes on a particular physical state in the structure
of an element, say iron, it also can be said to have a function in that
system, which is merely to make iron. In both cases the function is then used
to inform the sub-system of what state it should be in. Iron itself, made up
of its component sub-systems, provides the physical constraints on quantum
uncertainty that limit the states to those suppporting iron. Otherwise they
would be fully uncertain. In the same way, while we can't calculate it
dynamically, the system of three orbital bodies informs each of the thee
components what its constraints are as such a system, and that maintains a
classical system of three bodies in motion. Biochemistry, cellular biology,
organisms, ecosystems, etc. all do this same thing, but seem to isolate the
phenomena to their own systems, so that the laws involved can be local, not
just the universally shared ones.
Nice story anyway.
The stability of complex systems even below the threshold for
life is inherent in atomic structure-- (which my father probably would
prefer to call "atomic organization"; structure being too static a word
to apply to a complex system, really)-- That stability allows atoms to adapt
and evolve in certain ways similar to the way organisms adapt and evolve.
For example, an atom can absorb particles and continue to keep its basic
organization, with certain changes to relationships within the atom
(adaptation/evolution?). I don't see that the relationship between the
atom and its environment could be characterized as a
relationship where phenomena
becomes "information". Why not? It is only the
existence of some structure that can provide the conditions by which the state
of quantum particles become constrained. Those constraints then help define
the larger structure. The structure and the uncertain particle states are in
communication and are mutually reinforcing. Why can't one call the interaction
one of information transfer?
Thus, it looks to me that complexity below a threshold has
"automatic" interaction with its environment. Complexity above that
threshold has both automatic and "volitional" interaction with anything
outside itself and within itself... Volitional implies
the metamorphosis of phenomena to information.
We have a lot of baggage associated with the idea
of volition. Keeping our human orientation, one must draw a line as you are
attempting, in order for reasonable people to agree. However, if we consider
that human definition as a very complex and evolved form of something much
more elemental, then is is possible to think of a continuum. What is
"automatic" is merely what occurs within a well-connected system. It is
automatic to the extent that its laws are shared. The universe looks automatic
to the physicist because he/she assumes all the laws are shared.
It seems to me that this is a key feature of biological
systems. Is it a defining feature? Do all organisms possess the ability to
turn phenomena into information? Does this partly explain the "internal
predictive model" that my father discussed in Anticipatory Systems?
Life emerges at the same threshold at which anticipatory
behavior emerges, so it would make sense. I think
this is on a good track, but I would suggest that the "key feature" is not
genericity of information as such, which may be considered a universal
phenomenon. Instead, would it work to think of it as the use of "local
information" as opposed to "universally shared" information?
--
© 2004 John J. Kineman
all rights reserved
|