----- Original Message -----
Sent: Wednesday, November 03, 2004 1:22
PM
Subject: Re: [ROSEN] Maximally
constrained
It must be that
organisms - as physical systems - have constraints...and some of them are
holonomic and some are non-holonomic. It is required so that organisms
are "held together" by forces rather than merely being blobs of statistically
congregated particles in some area of space, like a gas cloud. Whether the
non-holonomic constraints that are present are such in number that
all velocity variables of all the structural 'particles' of an organism are
determined by configuration alone is in question.
When I wrote my
post this morning, it seemed compelling to me that an organism could be a
maximally constrained system. Howard in his reply disagrees, and I thought
about it some more and maybe it is not so compelling. And the
reason I am reconsidering has to do with what Rosen called
"structure-function complementarity" in the paper.
If I adopt the view that an organism is not a maximally
constrained system, then what comes to mind for me is: if an organism's
pattern (for lack of a better word) of
behavior is not dependent upon forcing all velocity variables to be
determined by configuration, then what allows its pattern of behavior to
remain invariant despite the removal of some information (i.e., the removal
of momenta due to freezing) from the system? [Here I am using 'information' in a general sense,
not specifically referring to Howard's 'informational constraint'.]
My conjecture would be: those momenta (of the unconstrained 'particles')
must not therefore constitute
information relevant for the
system's pattern of behavior. In that case, it would appear to follow that the dynamics of
structure are not the
relevant criteria. What I am now wondering
is if the invariance of behavior to the removal of all dynamics indicates that
(at least some) of the non-holonomic constraints that are
present constitute a nexus of such constraints which serves to keep
the functional organization invariant. This would be very difficult to
describe in strictly structural terms, i.e. in the usual formal representation
for constraints in analytical mechanics, because we don't have a good way to
first represent biological "functions" in structural terms onto which these
constraints could then be imposed.
The result would
be that an organism is not maximally constrained structurally, but
is maximally constrained functionally.
That's my
thoughts for the moment anyway,
Tim
Hallelujah! Thank you, Tim!
I think, however, it might be worth discussing whether
organisms are also "maximally non-holonomically constrained" systems or not.
The more I consider the nature of organization and how it
creates context, the more I see that non-holonomic constraints are
actually a possibility engine.
When this thread first came up, it was presented as if my
father was actually saying that, and I knew the only way he could have
said it, which would be consistent with his belief
structure, was if there was some aspect to it that created more
possibilities than it limited. Hence, it seemed quite reasonable to
interpret the language such that "maximal" is to "maximized" what "optimal"
is to "optimized", and that may still be the case-- he used the word that
way in conversation. It became apparent, as more of the paper was included,
that the argument was a moot point; he was talking about how a machine could
mimic an organism.
The more I have researched the range of what can be
defined as a "non-holonomic constraint", though, the more I realize that the
whole phrase represents a situation in a system where the potentials are the
most highly developed (maximized). I regard anything that interacts with a
system in any way that causes change as "a constraint". Certainly, from
that vantage point, non-holonomic constraints in system
organization generate more potential than no constraints at
all.
That fascinates me because it is, as my father would say;
"counterintuitive".