Tim,
Thanks. That helps a lot. I still need to work through the
logic more carefully with some examples, but it seems to make sense to me, all
except...
The "infinite regress" indeed goes away if all the
entailments and contingencies are internal, but I'm starting to see that as
unrealistic - the closure is not absolute but relative it seems; perhaps a
threshold phenomenon - i.e., enough closure and you get an
organism.
Well, in order
to create a closed loop of efficient causation the loop must actually be
closed: either it is predicative (non-looping entailments)
or impredicative (has a loop of entailment) - there is no
middle ground.
But this closed
loop of efficient causation is only applied to a very specific aspect of the
organism: the efficient causes of its functional organization. It does
not apply to its structural organization.
Making this caveat also
resolves any contradiction with the idea of "larger systems."
I am not sure I
follow. Is it that by being closed to eff. causation it appears that the
organism cannot then engage in functional relations with other entities in
some larger system (e.g., social)? If so, I think there is no
contradiction. The closed loop of entailment in the (M,R) system does not
prevent two organisms (say, two people) from interacting socially. Indeed, it
would be extremely weird if people interacting socially (e.g., being efficient
causes upon some aspect(s) of one another) and thereby forming new
functions of a social type between them as a result, were doing so by actually
combining their respective internal metabolisms and repair/replication
functions. Instead, the respective metabolism and repair/replication of the
(M,R) systems of each individual remain invariant and functionally intact
whether they are interacting with other persons or not. If these internal
functions were open to efficient causation, then they would be
context-dependent (contingent) upon some external efficient cause in
order to maintain them. (Certainly, they are open to material cause from
outside, since they require a compatible ecosystem to supply nutrients,
heat, etc. in which to thrive.)
But perhaps this is why he
made the distinction between entailments and contingencies - perhaps the
entailments refer to the closure, while it is the contingencies that can
extend via larger and smaller systems.. again, hmmm.
A function is
contingent upon, or exists only within the context of, a specific
organizational structure of entailments. So, breaking a system (say, an
(M,R) system) apart, either alters or destroys those functions that were
contingent upon its original whole structure. Conversely, allowing
interactions between two systems that were previously isolated will
potentially create new functions that previously did not exist in either of
those two separate original systems.
JJK
Tim Gwinn
wrote:
Hi John
K,
The reason for
the focus on efficient causation in the functional organization is that it
relates to where final cause for the functions in that organization are
located. As Rosen says (regarding a machine view of
life):
"Specifically, suppose that f represents a component;
let us ask the question "why f?" and seek an answer in terms of efficient
causation: because Phi. As we have seen, if we are dealing with a machine,
we cannot generally provide such an answer within the system;
therefore, any such answer must pertain to the environment of the
system, something outside. In effect, we are thereby ascribing a
function to the environment, namely, to answer the question
"why f?" Accordingly, we can provide an answer to the question "why Phi?"
in terms of a category of final causation: "because Phi entails
f." We thereby put not only efficient cause (of f) but also final cause
(of Phi) into the environment.
In short, efficient
causation of something inside the system is tied to final cause
of something outside the system." [LI p.
246]
So, the
machine view does not dispense with final cause, as the Newtonian ideal
would suggest; instead, it merely displaces it from inside the system and
places it into the environment. Final cause is what a thing entails; in
other words: its "function". [LI sec 5I] So, the machine view
inevitably incorporates notions of function and final cause, despite its
desire to avoid them. And further, these functions (final causes) are strewn
across the system/environment boundary, and also end up in an incipient
infinite regress precisely because there are no closed loops of efficient
causation.
On the other
hand, having the efficient causation of all the functional components be
entailed from within the organism has the effect of placing the related
final causes (functions) of the components inside the organism, independent
of entailments from the environment, and avoids the infinite
regress.
Regards,
Tim
(I am
forwarding this on behalf of John K. It came up with an error for
some reason because the Lsoft server found email addresses embedded in the
body of the msg. I snipped them out. Not sure why the server considered
this a problem. - Tim)
Hi Tim,
Thanks. I cross posted your reply to the Life list,
where this question arose.
It is an interesting discussion. On the
one hand I am suspecting that no system can be entirely closed to any of
Aristotles causes.
But if we take it as
relatively closed - a threashold of closure sufficient to get a
self-perpetuating system, then it could work. Any given organisms may be a
combination of self-perpetuating components and acquistions.
Why
questions can be answered by any of the A-causes, so we are to focus on
efficeint answers.
Efficient cause has to do with origin or production
of the component, so:
-Why is there a heart, other organs - efficient
answer is production mech. by the organism - that works fine. If we answer
with functional answers, i.e., Heart - to pump blood, that is formal
cause. If we answer "to maintain life", that is final cause. These also
work.
-Why are there mitochondria in cells? The efficient answer
at one time was an invading organism. Later they combined functions. You
can say that mitochondria, before becoming a component, was not part of
the system; but then we're introducing a tautology. Component is defined
as something meeting the same criteria as organism.
Also,
functions can be components, can they not?? so we can ask:
Why memory?
The efficient answer would have to address the production of the memory,
which is not entirely internal. The formal answer would be more diagnostic
in this case.
Hmm.
Tim Gwinn wrote:
John K et al,
-----Original Message-----
From: ROSEN Forum [ -snip- ]On Behalf Of John
Kineman
Sent: Wednesday, September 10, 2003 11:05 AM
To: -snip-
Subject: Re: [life] Machines and wholeness - a speculative hierarcy
Hi Jon & list,
This makes a lot of sense to me. I think there is an inconsistency in
asserting efficient closure except as a relative matter. It would be
interesting to get reactions from the Rosen list, so again I'm
cross-posting this to both lists.
-snip-
I do not perceive "closed to efficient causation" as presenting an
inconsistency. However, Rosen certainly does not make it terribly clear what
he is talking about in that sentence, and it can appear to lead to
inconsistencies. Certainly, how could an organism be "closed" to, or cut off
from, or fail to be susceptible to, impressed (efficient) forces from
outside?
Including the next sentence after the "closed to efficient causation"
sentence helps some:
"The answer we propose is now this: a material system is an organism if, and
only if, it is closed to efficient causation. That is, if f is any component
of such a system, the question "why f?" has an answer within the system,
which corresponds to the category of efficient cause of f." [LI p. 244]
And, in regards to the (M,R) system and the above criteria:
"Clearly, the diagram [10C.6] satisfies the condition we have laid down at
the outset of this chapter, namely, that everything in it is entailed in the
sense of efficient cause entirely within the diagram. Any material system
possessing such a graph or relational model (i.e., which realizes that
graph) is accordingly an organism." [LI p. 251-252]
The discussion in that chapter, and these quotes specifically, are based
around characteristics of functional organization. In the first quote,
"component" refers to Rosen's use of the term as referring to a functional
element - a "particle" of functional organization. And, the (M,R) system is,
of course, a functional model of an organism.
So "closed to efficient causation" means that the functional organization of
the system possesses the characteristic of having answers to the question
"why f?" be answerable, in terms of efficient causation, from within that
functional organizational structure, and not from some efficient cause from
outside that organization. The latter (requiring efficient cause from
outside that organization) was what led to the incipient infinite regresses
in machines and mechanisms in ch. 9.
On the other hand, from a perspective of structural organization (its
configuration of atoms, molecules, organs, etc.), the organism is certainly
"open" to efficient causes from outside (and from inside, too, most likely)
forces.
When the question of self-organization is tacitly framed entirely in terms
of structural organization, the question of organism is something
approaching disbelief:
How can an organism possibly maintain its self-organization in the face of
all the forces and influences to which it is subjected??
Which is more fully rendered as:
How can an organism possibly maintain its [structural] self-organization in
the face of all the forces and influences to which it [its structural
organization] is subjected??
The solution is to be found in breaking apart the term "self-organization"
into the two distinct and incommensurable categories: "structural
organization" and "functional organization". (Other than Rosen, it is
unfortunately still widely assumed that "functional organization is
reducible to "structural organization".)
This allows us to restate the question thusly:
How can an organism possibly maintain its functional organization in the
face of all the forces and influences to which its structural organization
is subjected??
This question is clearly not so disconcerting, since the two types of
organization are completely different, and what constitutes an efficient
cause for one type of organization does not in any way imply that it is also
an efficient cause for the other type of organization. It is in this way
that an organism can be "closed to efficient causation" in its functional
organization, and simultaneously open to efficient cause in its structural
organization. Thus an organism can maintain an invariance of functional
organization while suffering various external efficient causes to its
structural organization. Seen this way, I feel it is reasonably clear.
Perhaps the greatest difficulty is in attempting to conceptualize or
visualize in the mind these two incommensurable modes of organization being
realized simultaneously in one material system without trying to reduce one
to the other. That is a bit of a mind-bender. :)
This also makes it clear why structural approaches to studying
self-organization (e.g., self-organizing criticality, cellular automata,
etc.) are not very relevant to understanding what makes an organism
organized, and it also makes clear why studies such as molecular biology are
likewise generally not terribly fruitful in that direction.
Finally, just to note, at some later point Rosen appeared to have
reconsidered "closed to efficient causation" as a sufficient condition, and
in Essays (p.28), it seems that this has become only a necessary condition.
Regards,
Tim