[Date Prev][Date Next][Thread Prev][Thread Next]   [Date Index] [Thread Index] [Author Index] 

Re: What is closure?

Tim and list,

A final note (unless discussion warrants more) on this question of
relating new and old physics, thermodynamically open/closed systems with
other forms of causal closure ...

The ideas of bringing forcings into a system, and thereby interalizing
"gravitational" or causal entities solves the problem of open systems by
providing a way to close them to their causes; however while each
interalized system (say enzyme or DNA) helps close the original system
(to its causes), each newly internalized system (and new set of system
variables) must also be open to the outside. So R says we get bigger
open systems, not the result desired in the "old physics" of a finally
closed system. However, he then says one can get that practical result
if the nth stage needing internalization is itself produced by one of
the earlier stages. Then, for the practical purposes of those
interacting sub-systems, we have achieved a form of system closure. The
cost is to introduce causal loops (nth stage explained by a previous
one).  This he equates with efficient closure in the following paragraph
(from Ch. 1 of Essays, pg. 24):

"Breaking off such an infinite regress does not come for free. For it to
happen, the graphs to which we have drawn attention [modeling
relations], and which arise in successively more complicated forms at
each step of the process, must fold back on each other in unprecedented
ways. In the process, we create (among other things) closed loops of
efficient causation."

So we have gone from the old physics to the new physics and have bridged
from the discussion of thermodynamic closure to one of efficient
closure. In this chapter I think he has accomplished quite a feat!

JK

Continuing

John Kineman wrote:

Tim,
Maybe this helps resolve our respective points; i.e., yes they are
different and yes they can be related. In Essays, Chapter 1; pg. 15:

"The language of constraints as manifestations of order [this relates to
efficient cause as a constraint manifesting order] can be made
compatible with the language of entropy coming from thermodynamics, but
the two are by no means equivalent. Schrodinger took great pains to
distinguish them, associating the latter with old physics, embodied in
what he called "order from disorder," marking a tranasition to
equilibrium in a closed system. By speaking of order in terms of
constraints, he opened the door to radically new possibilities."

The old physics was called "order from disorder", the "new physics" was
called "order from order." And later, just before the section titled "An
Open System" he wrote:

"In addition to the principle of order from order that Schrodinger
introduced to get from genotype to phynotype, and the aperiodic solid
that he viewed as constituting the genetic end of the process, and the
idea of a cryptographic relation between holonomic constraints in
genotype and the nonholonomic ones characterizing phenotype, Schrodinger
introduced one more essential feature: the idea of feeding (on "negative
entropy," he said, but for our purposes it does not matter what we call
the food). This was not just a gratuitous observation on his part. He
was saying that, for the entire process of order from order to work at
all, the system exhibiting it has to be open in some critical sense."
[emphasis from original text]

JK



John Kineman wrote:

JohnM & list,

John, this is to reinforce what you say in your post, which I find no
fault with.

In the first chapter of Essays, which is about Schroedinger's idea,
Rosen discussed openness and closure in a very constructive sense which
finally got through to me. He pointed out (after Schroedinger) that 2nd
thermodynamics applies to closed systems, of which there are no natural
examples of an absolute nature - as JM states here also. Open systems
can, of course, exist within theoretically closed systems and have the
property that they dissipate energy to an "environment" - thus
establishing an inside and an outside. The inside can become ordered
with respect to the outside, which must then become more disordered. It
is also clear that relatively closed systems can exist within a
relatively open one - in other words if we have systems that behave
mechanically they will conform to the 2nd law of increasing entropy,
conservation of energy, and so forth.

--- As an aside, there is nothing in science itself that can tell us
whether the "ultimate" system is open or closed. Hence the universality
of 2nd law thermo vs its negation is not something that can be
determined. What we know is that both occur in nature. ---

Now, he discusses the difference between "gravitation" and "inertia" as
a metaphore for "genotype" and "phenotype" claiming that this dichotomy
is epistemologically identical. It represents a split between a source
of action and an object of it.  The essence of efficient closure is then
described as  whatever means a system (organism) has for bringing the
source of action into the object of it. It is an embedding principle.
Whereas in more simple system concepts the force is separate and outside
the object it forces (the force representing genotypical gravitation,
the object representing phenotypical inertia); in complex and living
systems certain "forcings" have been incorporated into the system. It
thus remains conceptually the same as a system being forced from the
outside, except that parts of the "outside" are now inside. This creates
a very different kind of system than can be analyzed by separating
outside from inside. How do organisms achieve this inclusion of outside
forcings? One way is enzymes, another is DNA, and there are presumably
more.

We can see from this picture that "closure" means not that the whole
thing - i.e., every aspect --  is closed to efficient cause, but that
certain efficient causes have become incorporated such that
self-directing properties appear as characteristics of the system. How
much of such closure phenomena is required for an organism to be stable,
evolutionary, etc. is an open question, but it certainly is not abolute
closure that is required. Furthermore, identifying the importance of
such closures does not exhaust all questions about how such a thing can
occur, or will occur, or to what extent or result, and hence it cannot
be counted as the only criteria for organisms or life - much more
research is needed.

John K.

John M wrote:

Judith:

"Closed to efficient causation" means that everything about the
system that
involves "efficient cause" is entailed by something else about the
system.
It means that there is no one outside the system actively creating the
"effecient cause" aspects of the system-- the system is
self-sustaining."

Thanks for the definition, I can see the logical language-link, which
was
obscure so far in my feeble English. Problem: it sounds like being
cut off
from "outside the system" - which in my views is a nono. I can
understand
this as a chosen aspect of a closed (limited) model (=system?) unless
the
"system" is boundariless - unlimited.
As you explain it - a "self sustaining system" - means a cut-off view
of a
model, extracted from the total interconnectedness (interefficiency).

JR:
"All complex systems manifest closed loops of entailment; this is one
of the
definitions of complex organization... "

Conform with your additional defining: this is a necessary, but by
far not
sufficent description. ("one of")  - IMO complex systems (any, as
everything
else) have open connections with 'the world outside our model', do we
recognize them, or not. We cannot list all sufficient causes, unless we
restrict the topical view to be explained. I still doubt the 'working'
closed loops without (un)closed triggering, maybe indirectly. "A
system does
not DECIDE by itself."
As you wrote: " It certainly doesn't mean "closed" in general; quite
the
opposite."

Finally - hurting my past chemist-self:
JR:


Atoms (which are complex but not alive) are not "closed systems",
either.
Atoms give and take electrons all the time as they interact with one
another, etc., but still maintain their organizational stability.



When 'atoms' give or take electrons they change their organization from
'atom' to 'ion'. Their stability becomes different (to information they
receive) from the same shown as 'atoms'.
(If I condone the term atom, which washed away lately in my thinking,
just
as molecules, giving place to 'limited effects depicted in a way of
mathematically formalizable models of conventional (old?) observation
and
its (maybe obsolete) explanation".)
In my reductionist chemistry, however, I can condone a "repair" of an
atom: - from its anion-form, ('plus' electron) by discharging that
electron
for certain stability -reconfirmation into the more stable (?)
(neutral)
atomic format. In certain cases it goes spontaneously as rearrangement.
(Indeed a stability-induced "closed to efficient causation"). However
the
opposite process: to take 'in' an electron from the 'environment'
into its
'body' is metabolism at its best.
Restoration is repair, just as "imbibing" from the environment and
building
it into its body is metabolism. Should we call an atom "alive"? I may
do it.
Maybe you can paste other 'necessary' conditions into the satate of
'being
alive' to exclude atoms.
It all would go into the "sufficient" to assure YOUR choice of
selection
into YOUR theory. (The YOUR does not mean Judith Rosen of course).

John M


----- Original Message -----
From: "Judith Rosen" <***>
To: <***>
Sent: Tuesday, September 14, 2004 12:21 PM
Subject: Re: Could you give me your analysis of this?

Skipped