re: pure chance

brian d harper (harper@hcgl1.eng.ohio-state.edu)
Sat, 14 Dec 1996 22:15:34 -0500 (EST)

At 12:08 PM 12/11/96 EST, Jim Bell wrote:

>Brian, I think you've done an excellent job of sorting out what is and is not
>being said about "pure chance." Thanks.
>
>Given all of this, there are two points that continue to surface for me.
>
>First, echoing Denton, there is an intuitive sense that mutationa plus
>selection plus time cannot achieve the degree of complexity we observe. This
>is why Behe's work resonates with me. [I'm very interested in the Kauffman
>stuff, which I understand only on a surface level. How is complexity theory
>viewed in the scientific community at large? Is it still too new to have a
>strong opinion about?]
>

Actually, I think most of the complexity folks who are trying to apply
their ideas in biology are probably motivated to a large extent by the
intuition you describe.

I really don't know how it's viewed by the "scientific community at large".
In fact, given the specialization we have today I suspect most
practicing scientists know very little about it.

One problem with assessing complexology in some general way is
that so many different topics and "names" fall under this title.

I think Kauffman has done some interesting things but the common
complaint with him is that his "systems" are all abstract models
which don't make contact with reality. In Kauffmans defense I
would say that he has done this deliberately because he is looking
for general principles that apply to all complex systems. But somewhere
along the way he'll have to show the relation between his models
and reality.

Bak is another interesting case. A severe blow to his sand pile model
of self-organized criticality were the early experiments showing that
real sand piles do not display self-organized criticality. Horgan had
a good time with this. "If the model can't even describe real sand piles
what good is it ..." Subsequent analysis and new and more careful
experiments have rescued Bak, but I doubt we'll ever hear about it
from Horgan :).

A common criticism is that the field relies too heavily on computer
simulations without making contact with the real world. Maynard
Smith's "fact free" science. This is not, however, true universally.
Goodwin, for example, uses real, established, physical models and
established numerical methods (Finite Element Method) in his
work. Some of his models of morhogenesis are really amazing.

Well, I'm starting to ramble. Overall, my feeling is its much too
early to pass judgement. But the nature of the field is such as to
produce strong opinions, people seem to either love it or hate it.

>Second, we need to keep in mind the negative side of mutational change.
>Natural selection theory, I believe, depends too much on the assumption of an
>ever upward dynamic. ReMine calls this "naive natural selection," and I think
>he has a good point.
>

I don't really know what to say about this.

Brian