[Date Prev][Date Next][Thread Prev][Thread Next]
[Date Index]
[Thread Index]
[Author Index]
Re: Function and functional organization
- From: "Tim Gwinn" <***>
- Date: Fri, 29 Aug 2003 22:13:34 -0500
John,
Geez, there may
actually be some hope yet. :) I'll try to be brief, since your mentioned
your time is limited.
We were each
apparently using "material" in very different senses that led to
some differences.
To me, "material
reality" refers to what exists in (to use Rosen's term) the "external
world". In this sense, it is not Newtonian or Rosennean, it is rather:
what is actually "out there". This material reality is what we attempt to
learn about via modeling relations (and can learn about only via modeling
relations).
In this sense, since "material" refers to what exists in the
external world, and "material reality" encompasses all that exists in the
external world, then to use a term like "non-material" is to propose something
beyond that which exists in the external world. The latter is not your
claim, I believe. (Although it is what I had previously been thinking you might
have been proposing.)
I believe Rosen uses the term "material reality" in the same
sense as I (or, more correctly, I am using it in the sense I perceive him to use
it), and that quotes from LI p. 119 is an instance of the use of that
meaning of "material reality".
If I understand
you correctly, when you say "material in the Newtonian sense", 'material' refers
to "things" such as atoms, molecules, toasters and planets - all the "objects"
which are idealized in Newtonian mechanics as 'particles' or 'mass
points' or so on. These particles are then pushed around the universe by
Newtonian forces in accordance with Newtonian laws.
In this sense of
"material", the term "non-material" may have many possible meanings (since
the term only tells us what it is not, rather than what it is). I suspect by
"non-material" you mean (at the least) something that can't be
represented in Newtonian mechanics as particles or mass points (or as
Newtonian laws, for that matter).
If I understand
this correctly, then I suspect I roughly agree with this statement
from below, and then we are generally in overall agreement, I
think:
This supports both of my main assertions in this discussion, that
functions are not material in the Newtonian sense, and that the relational
theory linking functions to material states is itself general to biology AND
physics, thus giving a new definition to matter as well as
organisms.
My hesitation is
that I am still unclear exactly what "non-material" means to
you, when phrased in terms of what it is, rather than what it is not.
My own view of
'functions' and 'functional organization' is
that they exist in material reality (using "material reality" in my, and I
believe, Rosen's sense) ; that is, they exist in the external
world. We can model them via relational models. And many natural
systems (biological and non-biological) have functional organization(s) and are
thus amenable to relational modeling. Not every natural system has
functional organization, since a natural system can be a simple system (when
defined with the proper set of observables and encodings/decodings), and simple
systems allow no noncomputable models and no functional
descriptions. It is important, though, to distinguish "natural
system" - which is some subjectively selected collection of
observables of material reality that we decide to call a "system"
- from the complete, undifferentiated material reality. Function and
functional organization may be a pervasive quality of material reality, or
it may merely be prevalent; I leave it as an open question.
Are functions
and functional organization "objects" or "things" in the Newtonian sense?
On the face of it, apparently not. But more importantly, in the enlarged
Rosennean view, I am not certain if that Newtonian question/distinction is
even relevant or meaningful. Because: 1) for me that is a question that
arises from the very feebleness of
the Newtonian paradigm, and 2) when the Rosennean paradigm is
restricted to the Newtonian subset of it - where the question of
thing/non-thing necessarily arises because of the restrictions in
the formalism - functions and functional organization disappear from the
formalism (that is, they cannot exist within that Newtonian subset)
anyway. So, I personally do not worry much whether functions and
functional organization are Newtonian "things" or not. I suspect the
question lacks meaning in the Rosennean paradigm.
Are we getting
close now...at least in terminology?
Regards,
Tim
Tim,
These statements are not
the contradicton you claim they are, although we do have to navigate the
semantics carefully according to their context. The context is how
scientific the theory can be considered and whether it is less valid or
rigorous than the traditional formulations. A quick look at the section this
is in and the surrounding paragraphs can easily confirm that. Also, it was an
important point to make because traditionalists would have criticized the
relational view on the grounds that it is epistemologically weak compared to
traditional formulations from the "old school" thinking. But here, he claims
it is equally rigorous, equally valid as physical theory. For example, just
after the phrase you bolded, he writes "in the sense of "ideal" physics."
We should not overlook that contextual clarification. He is thus expanding
physics to something more ideal, but is not thereby producing something that
is any less valid as a material theory -- it is a better one, he says. If he
did not say this, his entire analysis could be rejected on epistemological
grounds as having any importance to physics. He is in fact explaining what
material is actually composed of better than the Newtonian model did, which
incorrectly predicted it would be rigid particles all the way down, and when
it was proved not to be, that it must then be seen as fundamentally uncertain.
Rosen says there's another alternative for basing physical reality - the
relational model. Knowing that context, one can then attribute the appropriate
meaning.
As an aside, this issue reminds me of a quote from a
particular philosopher, Alan Watts, who turned the semantics around very
similarly. He said many people criticise the West for being materialistic, but
we are not true materialists because we seem to have so little regard for
material, so little respect or reverance for its intricacy and mystery. We
imagine it as a dull static thing to be manipulated at our whim and nothing
more. We don't appreciate its amazing origins or accord it any intrinsic
value. Here Rosen is echoing this sentiment, moving us to a truer materialism
where we might have some respect for what actually goes into material - the
incredible complexity of functional relations that at no point devolve into
the traditional concept of a thing.
Recall our earlier conversation
where you questioned my assertion that the theory applied to physics. Now you
are criticizing my view by making the claim that it IS physics. Which
criticism should I answer? Rosen is saying it goes beyond traditional physics,
but it is valid enough to be considered a more ideal physics. Precisely my
view as well (except that I would be less placating to the physicists). This
also supports my earlier arguments that the theory can be taken and was meant
to be general, and the process is as I said, he worked out the theory for
organisms, then applied it to physics and all of reality.
Look closely
at the statements. He does not say there are no unphysical or non-material
components, if you read carefully. This quote refers directly to a
comparison of conventional analysis with relational analysis,
not components of reality as discussed in the body of the book. He said there
is nothing in the "relational strategy" that is unphysical. The
strategy is the analysis. It is a physical analysis, he says - a new physical
view and theory if we wish to stretch the definition of physics beyond its
traditional Newtonian foundations (which many physics would agree to, if they
liked the theory, because they want to ensure that physics potentially
includes everything.)
Contrast this statement with his other statements
about physics when he was considering it in its traditional sense. He said we
"keep the organization and throw away the phyiscs" (I think that's an
exact quote, but I didn't look up the page ref). Obviously he was then talking
about traditional physics, not his expanded physics which includes the
organization. In fact he dealsa rather destructive blow to the traditional
formulation of physics, without suggesting that the traditional view has its
value in a limited domain. But he shows how to modify it, and now claims that
the modified view is just as valid and rigorous an analysis as any prior
theory in physics. Many, of course, are not convinced, but I am. He was
correct in this assertion, as I see nothing epistemologically invalid in his
construction. It is good science in the best tradition, and I think that is
what this section is saying.
Also, in the very saying that the
(functional) organization is "as much a part of its material reality as
the specific particles that constitute it" he has clearly distinguished the
two. His theory distinguishes them and places them in relationship, unlike
Newtonian theory which distinguishes them but keeps them separate and not
interacting. So, once again, functions are distingished from the Newtonian
definition of material, although both can be said to comprise a new concept of
material, or a new material theory. This supports both of my main
assertions in this discussion, that functions are not material in the
Newtonian sense, and that the relational theory linking functions to material
states is itself general to biology AND physics, thus giving a new definition
to matter as well as organisms.
Note that Newtonian theory also depends
on the unquestioned existence of "non-material" elements -- natural laws
themselves. The only difference is that it does not allow any genericity of
those laws from within the system being studied. So, in this sense, Rosen's
relational approach is more physical than Newtonian physics (as in the
quote), because Newtonian physics placed the "non-material" elements out in
the absolute Platonic realm - the domain of God. Rosen's approach makes it
part of the system, and thus subject to analysis.
All of what I am
saying should be really clear from his last sentence, which I'll repeat,
bolding it all to retain the context: "the resulting formalisms have at least as much
right to be called images of material reality as any
reductionistic model based on states and dynamical laws."
Images of material reality are not the material reality
itself, they are models, theories, concepts. And he is clearly talking about
the right to call his theory a scientific formalism.
Finally, let me
offer one more opportunity to agree:
I would
suggest that you are arguing this from a Newtonian-type paradigm, where
(roughly) 'Newtonian = all material things', and thus, 'non-Newtonian =
non-material'. A notable point of Rosen's overall arguments was that
'Newtonian = all material things' is
false.
Yes, what I have also said, that there
must be something more than the Newtonian concept, but you are playing
semantic games here. The statement is false if by "material things" you mean
Rosen's view, in which "thing" becomes a different concept that most are used
to. If by "material thing" we mean what most people would assume from
traditional Newtonian view, then the statement is true. The Newtonian view
sees Newtoning "things" and writes laws about nothing more than that.
In
the Newtonian view of the material world, functions and
context-dependence are disallowed.
Yes,
my statements exactly.
In
the Rosennean view of the material world, functions and context-dependence
are part of the material world
Only in
an expanded view of material, which explicitly includes functions as a
component of the system. This is by no means common parlance, so it has to be
clearly spelled out if you want to refer to the relational definition of
material. You have to be in the context of discussing the expanded theory for
this to be understood or obvious without saying it.
and, so too, they should be part of the
enlarged physics that describes that material world.
Yes
This is why there is nothing unphysical about the
relational strategy.
Yes, again allowing that
we are expanding the traditional meaning of the concept "physicsl." For a
critic who did not allow anyone to change the meaning of words, it would be
entirely unphyiscal, and mysterious. So Rosen's construction demystifies the
relationship between our usual concept of material, and our poor treatement or
understanding of contingent functions.
Thus, as the quote at top indicates, there is
no need to posit that function is "non-material".
Don't presume the need that others may have
in discussing how the new theory relates to the old. Rosen's view is not the
common one. It is essential for communication to relate the semantics, and
aside from defending the theory as Rosen does in this section of the book, I
would suggest that it would be very confusing to shift semantics this way
without telling people that you are doing it. Functions are decidedly not
Newtonian-material, but are intimiately and fundamentally embedded with all
material in Rosen's view. This is one of the quotes I would use to support the
idea that his theory is meant to be general and that this principle of
relationship must be taken all the way down with the turtles. You seem to
agree, but not to agree.
Instead, functions are "as much a part of material
reality" as any other material thing.
It
can be said, if you clarify the implied semantic differences between the first
reference to "material" and the second, even in this sentence. By "other
material thing" do you mean in the Newtonian view? If so, why wasn't it stated
for clarity. And if not, is it not a tautology, i.e., "functions are part of
Rosen-things as much as any other Rosen-thing is." So we are both relating old
and new meanings here, and we have to be careful when doing so.
I think my view
has been clearly stated by now, BUT guessing that you will find this
unconvincing, perhaps it would be better for me to ask what your point of view
is. Maybe give an example of how you are applying Rosen's theory in which I
can see how your are interpreting the concepts. What in fact is your
interpretation? I would like to write less and listen more at this point. I
think we are in full agreement that Rosen provides an expanded meaning for
"material" and "physical" thereby making the theory generally applicable to
reality itself.
John K.