Re: info & life

from jamesbarham@supernet.com:
------------------
Dear Gyorgyi and Norbert:

Thank you very much for your comments. Just a couple of points of
clarification:

I think that perhaps a big part of our disagreement is purely semantic
(no pun intended!). As Karl Popper liked to say, one should not quarrel
about the meanings of words. On the other hand, if we can't agree on
what we mean by words, then we run the risk of misunderstanding each
other and talking past one another. So let me give my reasons again for
preferring to use "information" in the way that I do.

First, I think it is crucial that "information" be understood as having
a semantic interpretation. Without any need for meaning to enter the
picture, I just don't see why we wouldn't use a purely physical term,
whether "structure", or "process," or "dynamics," or something else. But
if that is so, then we should limit the word "information" to cases
where there is clearly a cognitive agent involved.

When I speak of "objective" information, I just mean information that is
independent of the human observer---that is, information that would
still exist whether human beings existed or not. Gyorgyi wrote that what
I am calling "intrinsic, observer-independent information" is precisely
what he wants to call "subjective." I think I see what he means. If I am
saying that every instance of information use in the true sense depends
upon some cognitive agent, then I guess there is a sense in which it
must be "subjective," in the sense of being FOR a subject. But even so,
we still need to distinguish information FOR organism A as its own
subject from information ABOUT organism A as an object for
me-as-observer.

When I say that information exists "objectively," I mean that it exists
in the first sense, for other organisms as subjects in their own right,
whether I or any other human being observes them as objects or not. This
is kind of confusing, because of course I am simultaneously an
objectifying observer and also a subjective cognitive agent in my own
right with intrinsic information dynamics. So I am also like organism A
and have my own objective information that exists whether I am observed
from the outside or not. It was to avoid this confusion in the terms
"subjective" and "objective" that I introduced the terms "allotelic"
(information ABOUT organism A FOR organism B) and "autotelic"
(information FOR organism A itself), but whatever we choose to call it,
I still feel that it is an important distinction, which we ignore at the
peril of great confusion.

Of course, as Gyorgyi says, "dynamical interactions characterise
physical processes as well." But I believe that there is a sui generis
dynamics of the living state which characterises information use in the
true sense. Probably no one else will agree with me here, so perhaps
this is the source of most of the confusion. I see all pre-living
dynamics as simply minimizing energy, whereas living dynamics is doing
WORK, i.e., not merely minimizing energy but guiding it in a
goal-directed way. More specifically, I think that information is the
means by which cognitive agents do work (attain their goals). More
specifically still, according to my dynamical model, information is a
low-energy trigger that correlates the higher-energy oscillation of a
nonlinear oscillator with the external conditions that will tend to
preserve its dynamical stability. Of course, explaining how all of this
is possible in terms of some recognizable laws of physics is a tall
order, mainly for the future (although Giuseppe Vitiello and others have
some interesting ideas). But however that may be, the issue with respect
to information is not dynamics per se, but a very special sort of
dynamics characteristic of the living state alone and constitutive of
information. Or so I believe.

Finally, Norbert says that "matter, energy and information are bound in
inseparable unity," but that is clearly only true in the allotelic sense
of information, not the autotelic sense---at least if one is willing to
grant me that autotelic information emerged into existence only with the
advent of life and its sui generis dynamics as mentioned in the previous
paragraph. Before there was life and such a special dynamics, there was
simply no information at all in the autotelic sense. Sure, we can
interpret anything whatsoever as information in the allotelic sense, but
I do not see what advantage there is to reinterpreting physics in terms
of information theory, when the old-fashioned dynamical approach seems
to work just as well, with much less confusion.

On second thought, perhaps it is not so much confusion that is the
issue, as a deep metaphysical disagreement. My chief concern is to
understand how the universe gave rise to life and mind. If we build
information into the foundations of our physics, as Wheeler and Stonier
and Norbert and I believe John Collier and others wish to do, then we
are saying that the human being is at the center of existence. By
putting the human being logically before the cosmic evolutionary
process, we make it impossible
to understand the human being as an outcome of that process, and so make
it impossible to understand the human being in naturalistic terms at
all. In short, I see pan-informationism as either a gratuitious and
confusing misuse of terminology, or (what is worse, in my view) as a
form of idealism.

Best regards,

James
Received on Wed May 15 10:13:30 2002