We can legitimately speak from different paradigms. As you know a paradigm
begins from certain assumptions which are essentially untestable (except
according to their usefulness in spawning theories that work for a given
purpose). Don, arguing from Rosen's perspective, speaks often from a
paradigm that defines "functional" elements as real, then constructs theory
from that. It is more traditional to speak from a mechanical paradigm which
defines material elements as real, then builds theory from that. There is
no legitimate debate as to which is more correct in general, only that both
are valid, as Louis correctly states. There is a legitimate debate as to
which is more useful for explaining life, and which is more useful for
explaining non-life, as Don has been saying all along. It is also true, I
believe, that each view by itself is incomplete. It is very important for
us to realize that and ADD the perspectives together for a more complete
view. That may not be entirely possible in one model, but the essential
assumption of one view does apparently appear in its opposite. For example
semantics appear as impredicativities or uncertainty in the mechanical
view, and pure mechanism appears as an exception in the functional view. To
my understanding they are not in conflict, they are complimentary and this
complimentary nature is what we need to grapple with in discussing life
(which seems to involve both paradigms). Now I'm not taking this to the
point of resolving the duality -- I don't know how to do that; Some claim
that Rosen's approach may lead in that direction. I don't know. It may
proove to still be just complimentary; but it is AT LEAST complimentary,
i.e, both necessary for explnation and yet not equatable to each other,
like wave-particle duality only at a much more fundamental level.
At 08:54 PM 3/13/99 +0100, you wrote:
>Choice is an anthropomorphic notion, which I would not apply to describe
>either Darwinian evolution [or] self-organization.
I agree with the comments about anthropomorphism on both sides - The
absence of choice is also an anthropomorphic notion. At some level, every
human idea is anthropomorphic. However the real point is not
anthropomorphism, it is that these concepts are metaphysical, underlying
our theory structure. Invoking the criticism of anthropomorphism is usually
a veiled excuse for dismissing someone else's metaphysics on the grounds
that it makes too much sense (i.e., to humans). A legitemate discussion
would evaluate the metaphysics and skip the pejorative labels.
Cognitive "choice" is part of a functional view of reality. I think it is
analogous to mechanism in the material perspective. A mechanical view of
evolution (neo-Darwinism) does not involve choice because it does not
recognize function as a causal parameter in evolution theory. Thus the
question "what role does function have in evolution" cannot be asked in
this paradigm (because function is formulated as a result not a cause in
this view). However, IF one takes the view that function is causal, and
formulates theories from that basis, then the question makes sense and so
do the terms of the answer. Choice, involving some notion of purpose,
becomes an essential element of theory. As an example, it is common to
refer to purposful choice in ecology. When pressed, because of the tacit
agreement that we should all think mechanistically, most biologists will
back off and say the use of this language was metaphorical. That's just a
way of paying homage to the BELIEF system that says, ultimately, we will be
able to explain "choice" in some mechanical and deterministic way. There is
no proof for that BELIEF, as Francis says, and plenty of evidence that such
a conclusion will never arrive. Hence, realistically and honestly, both
paradigms are equally valid in biology, even though we have the strong
emphasis of the mechanical because of its success in the material
disciplines. In truth, one can explain the existence of materiality in
terms of life just as validly (some say more so) as we currently believe
the mechanical view can be used to explain the existence of life. There is
absolutely no epistemological difference between these views as far as
which is good science.
Now "choice" or "selection," NOT implying any cognition, purpose,
intention, or functional end, is strictly part of the deterministic and
mechanical view. There is no arguing this, it is definitional. So Darwinian
selection is even more basic than Louis implied. It doesn't even involve
fitness in its definition - fitness is defined in terms of selection and it
is because selection itself is a tautology that they can be thought of in
terms of each other ("survival of the fittest").. Natural selection is a
metaphysical perspective. That does not invalidate it, that means simply
that it is part of the metaphysics underlying the formulation of mechanisms
in Darwinian evolution. Selection is the obvious fact of this paradigm,
where there is only one manifest reality which is defined by what we find
in it as a function of time. Selection means that what exists today is by
definition "adaptive," or else it wouldn't exist. Differential selection,
on the other hand, extends this to the idea of fitness, saying that more
viable survivors of one type are more fit because of their greater viable
numbers (the only measure), again by definition. In this paradigm, the
process is entirely passive and definitional - there are no agents of
selection. A criticism of this is that it is not testable as a theory. But
I dismiss that complaint because NO metaphysical assumptions underlying
theory are directly testable. It is simply a statement of the paradigm from
which one is theorizing. Like Louis, however, I am not advocating this view
(in vact I advocate the opposite) , but I do suggest recognizing it as a
theory structure.
Now the really interesting thing about scientific paradigms is that they
eventually fail - all of them so far. If applied generally and thoroughly
we find their limits. Natural selection fails when we ask how cognitive
choice affects selection. If you defend the mechanical/deterministic
paradigm, you must dismiss the question as unscientific. The paradigm
presumes that all such questions are eventually explainable by the
mechanism of passive selection (of material form) and hence there is no
cognitive choice. This, unfortunately, doesn't help us understand the human
case (and by analogy, other species) and causes serious problems with
explaining the psyche, responsibility, and why we would get worked up over
choices we seem to have. As Don constantly reminds us, it does no good
whatsoever to try to model "choice" as a mechanical process because that
will never escape the paradigm. If one wants to deal with choice in a
robust way that will apply to our own experience, we must invoke additional
assumptions. I see no way around that. Don, of course, has been arguing
this latter point on Rosen's behalf in all of his posts.
-----------------------------------------------
John J. Kineman, Physical Scientist/Ecologist
National Geophysical Data Center
325 Broadway E/GC1 (3100 Marine St. Rm: A-152)
Boulder, Colorado 80303 USA
(303) 497-6900 (phone)
(303) 497-6513 (fax)
jjk@ngdc.noaa.gov (email)