At 12:26 AM 12/31/97 -0600, Linas wrote:
>
>Oh, I forgot to drive the point home.
>See below ...
>
Ack, you gave away all my stuff ... ;-)
>It's been rumoured that linas@linas.org said:
>>
>> It's been rumoured that Rick Becker said:
>> >
>> > Cross-snip:
>> >
>> > any feedback?
>> >
>> > Return-Path: <owner-evolution@udomo.calvin.edu>
>> > X-Sender: bharper@pop.service.ohio-state.edu
>> > Date: Tue, 30 Dec 1997 21:23:02 -0500
>> > To: evolution@calvin.edu
>> > From: Brian D Harper <bharper@postbox.acs.ohio-state.edu>
>> > Subject: RE: ABCD... Fibbonacci and gold
>> > Sender: owner-evolution@udomo.calvin.edu
>> > X-UIDL: a7773610dac96e190af997000c45b1cc
>> >
>> > At 09:37 AM 12/30/97 -0800, Greg wrote:
>
This was me saying this not Greg
>> > Come to think of it, I saw a little math show for
>> > kids with my daughter several years ago. The
>> > history of the Golden Section was discussed in
>> > some detail with many examples from ancient Greek
>> > architecture. They then showed a multitude of examples
>> > where the pattern emerges in biological forms. The
>> > leaf pattern was one example but there were several
>> > others that I can't recall now. Is the Golden Ratio
>> > an example of the Archetype that Richard Owen searched
>> > for?
>
>?? I thought archtypes as an intellectual concept, had
>died in the last century.
>
About the time Owen died I imagine ;-).
I obtained the information for my previous post from the
writings of Brian Goodwin (primarily <How the Leopard
Changed its Spots>), a structural biologist. Interestingly,
Goodwin frequently uses the term "natural kind" which
tends to make some peoples knees jerk and their eyes
twitch :). Goodwin is sometimes accused by his critics of
being a Platonist, which is very unfair. For Goodwin,
evolution is first and foremost a nonlinear dynamical
process with "natural kinds" being the stable or near
stable attractors of the process. The role of natural
selection for Goodwin is minimal, primarlily that of
stabilizing the attractors.
>> > Or is the arrangement beneficial to the plant
>> > in some way so that one could imagine it being selected
>> > for some time in the past?
>
>No, it would not have to be selected for; it occurs quite
>naturally as the solution of certain reaction-diffusion
>differntial eq's. The diffeq's themselves arise from
>the flow of hormones accross cell boundaries, so the
>question is, did this diffeq provide something beneficial
>to the plant? The answers can sometimes be guessed:
>less energy consumption, greter resistnace mold/mildew,
>minimized surface area, etc.
>
You stole my ammunition before someone took the bait and
tried to find a just so story to account for the pattern :).
Actually, there were a couple of points behind the question.
One often finds an argument going along these lines:
"If it can't be explained by natural selection, then
it must have been designed". This is an argument from the
false alternative, it fails to recognize that there may be
other alternatives to NS/Design.
Another point was that a proposed just so story would offer
a natural way of getting across the difference between
explanations a la Goodwin and historical explanations.
Goodwin showed that the pattern is generated naturally
from what he calls a "morhogenetic field" which one could
think of as a "law of growth". There's nothing to be selected
for. So, the spiral pattern would be a "natural kind".
Goodwin has many examples of this type. He'll first describe
a particular phenomenom and then he'll present a "historical
explanation" or just so story, quite often the just so
stories were actually proposed in the literature as an
"explanation". Following this he'll give his own explanation.
Here's a rather lengthy example from one of his papers.
Sorry about the length, but I think its really a much
better and more interesting example than the leaf patterns:
====begin quote================================
THE EVOLUTION OF GASTRULATION
A basic conundrum in biology is why organisms ever become
more complex than the unicellulars that populated the
Precambrian oceans, whose 'success' can be measured by the
teeming millions of micro-organisms that continue to pursue
similar life histories to those of their ancient progenitors.
One of the most significant steps in the emergence of
organismic complexity was the evolutionary origin of
gastrulation. The transformation of a hollow ball of cells
into a multilayered structure, with the consequent
combinatorial potential for reciprocal inductive interactions
leading to diverse patterns of cell differentiation, stands
out as a major event in the evolution of the metazoa. How
are we to understand the origin of this process? We have
on offer two rather different types of answer. [...]
[...]
NATURAL SELECTION AND THE GASTRULA
Buss starts with the observation that the cells of a metazoan
can be either ciliated or they can divide, not both. [...]
Imagine, then, protists with these characteristics aggregated
to form a hollow ball of cells. Buss suggests that this ball
faces a problem: it needs to be ciliated so that it can
undergo dispersion and act as an efficient propagator;
and it needs to continue cell division if it is to develop
into a more complex organism. But these states are
incompatible. How does the ball 'solve the problem'? It
forms a gastrula. The cells on the surface remain ciliated,
while some move to the interior where they lose their cilia
and so can divide. 'Animal gastrulation is the solution to
the requirement of simultaneous development and movement'
(p. 44).
What kind of an explanation is this, and what are its
predictions? Buss says that 'the demand for simultaneous
ciliation and continued development gave rise to
gastrulation'(p. 65). This 'solution to a problem' scenario,
which is symptomatic to the cognitive style of explanations
by natural selection, is frankly and explicitly teleological
and needs to be recast to conform to accepted notions of
causality. This requires that the 'correct solution', the
ciliated gastrula, has a selective advantage over other
forms and so will increase as a result of differential
survival. This is what is meant by solving the problem.
A simple prediction is that the forms that failed to
find the solution should have been eliminated. Sponges
failed: they form an unciliated ball which then gastrulates,
but they seem to be surviving well. Colonial organisms
failed: they form hollow ciliated balls that do not
gastrulate or develop further, but they, too, continue
to survive. Many protists are not constrained by the
either/or limitation: they can be ciliated and divide at
the same time. Here Buss makes an interesting remark: in
ciliates 'macronucleus division is a complex and inefficient
process'. Given the success of the ciliates, it is not
clear what objective criterion is being used here for
efficiency.
Looked at from the perspective of explanation in the exact
sciences, these arguments are exceedingly bizarre. There
are three major difficulties with 'explanations' by
natural selection. First, the selective scenario proposed
is untestable because the conditions that prevailed cannot
be reconstructed. Second, the simple predictions that appear
to follow but are violated can all be explained away by
inventing other selective scenarios for surviving forms
that did not solve the problem, and so the hypothetical
explanation is not falsifiable in these terms. Third, and
most seriously, no account is given of what causal process
may have actually generated the gastrula. Yet this is the
only type of explanation that would be accepted in an exact
science. And this is what developmental biology is all
about. [...] If there is to be a unification of development
and evolution, it must be in terms of an analysis that
attempts to articulate the evolution of developmental
processes as consequences of clearly-defined dynamic causes.
Natural selection itself explains nothing about generative
dynamics. Differential survival is a consequence of this
dynamics, not a cause. Furthermore, the 'units of selection'
are also consequences of the dynamics, the (relatively)
stable attractors in the evolutionary process (stable life
histories). They cannot be defined initially. In order to
understand the evolutionary origins of the gastrula, we must
proceed with the analysis in a different way. The basic
structure of this analysis was provided several years ago
by Willmer (1960). Oddly enough, Buss gives no reference
to this very interesting work.
THE DYNAMIC BASIS OF GASTRULATION
In the course of his cytological studies, Willmer observed
that a number of species of protists have the interesting
property that the undergo dramatic, reversible changes of
state, from flagellate to amoeboid, depending on whether
they are in a low or high ionic strength medium, respectively.
Suppose that protists with these properties aggregated to
form a hollow ball of cells in the Precambrian oceans, as
assumed by Buss. Such a form is a minimal energy configuration
for such an aggregate, making reasonable assumptions about
cell affinities, so this assumption is dynamically well-
founded. Due to the osmoregulatory properties of cells,
which pump ions across their boundaries to maintain
physiological ratios of ions, the confined internal space
of the hollow ball will have a higher ionic strength than
the external medium. These observations therefore suggest
that cells on the outside of the ball would tend to be
flagellated or ciliated, while any cells that moved into the
interior as a result of the balance of forces between cell-
cell adhesion, compression of the ball, and spontaneous
movement, would tend to become amoeboid, in which state they
can divide. This is the essence of Willmer's proposals, which
he developed in relation to the different patterns of
gastrulation that occur in different species.
Hence we have a perfectly acceptable hypothesis about the
origins of the gastrula in terms of known properties of
protists and the balance of forces acting within and on
a spherical cell aggregate. Furthermore, a whole series
of experiments are possible to explore this hypothesis
further. One process that is not understood is the
flagellate-amoeboid transition under a change of ionic
strength. Some might suggest that this property is a
consequence of natural selection. But this amounts to
the suggestion that there must be some natural causal
explanation of the phenomenom, and that it must be
consistent with the dynamic stability of certain protist
life-cycles in the context of particular environments.
This adds nothing to a research programme attempting to
understand generative causes. [...]
THE EVOLUTION OF GENERIC FORMS
Within this perspective, a distinctly different dimension
is added to the functionalist selective scenario. What
becomes evident from Willmer's proposals about the origin
of the gastrula is that this state of spatial organization,
the ball of ciliated cells with dividing amoeboid cells
inside, may be a robust and inevitable consequence of the
forces acting on and within a cellular aggregate. But it is
not the only one. Other states may be equally robust and
inevitable, given other paramenter values; for example the
non-ciliated gastrula of the sponge, or the ciliated
non-gastrula of colonial organisms, may be equally natural
forms. These are simply different possible states, different
attractors for different parameter values in a self-
organizing dynamic system. Each of these forms must, of
course, be able to survive and reproduce in arder to persist.
This is what is meant by dynamic stability. [...]
-- Brian Goodwin, "Evolution and the Generative Order,"
in <theoretical Biology>, Goodwin and Saunders, eds.,
Johns Hopkins University Press, 1989.
=======end quote=========
Brian Harper
Associate Professor
Applied Mechanics
The Ohio State University
"... we have learned from much experience that all
philosophical intuitions about what nature is going
to do fail." -- Richard Feynman