Re: [asa] ID/Miracles/Design (Behe vs. Behe)

From: Preston Garrison <>
Date: Tue Apr 28 2009 - 01:44:21 EDT

>Preston Garrison wrote:
>>This isn't the case. There are effectively an infinite number of
>>possible mutations.
>Hi Preston,
>I snipped this because of length, but let me say that I don't
>_strongly_ disagree with anything you wrote as you offer what is, I
>think, the "orthodox" analysis of the problem at hand. For reasons
>to be explained, however, I depart from "the orthodox analysis" and
>thus deny that the above excerpt from your post properly represents
>the situation. In consequence, I'll argue that my original point
>about chance and predictability of outcome stands undefeated.
>Let me start by saying that I'm greatly influenced here by the work
>of Simon Conway Morris and his notion of convergence - the idea that
>despite our sense that there are, as you describe, an theoretical
>near-infinite number of evolutionary options available, it is
>demonstrably the case that, in practice, evolution has thrown up the
>same 'solution' to a particular 'need' on a repeated basis. To take
>one remarkable example, Conway Morris can elucidate at least six
>instances of independent evolution of the camera eye (Life's
>Solution, Cambridge, 2006, 151-158) - and that's just one of a
>wealth of actual examples upon which SCM draws.
>One of his conclusions is that whilst there are, in theory, an
>immense number of possibilities, what we actually OBSERVE is a
>repeated tendency for outcomes to reoccur in a most curious manner.
>This should - as Cameron has been arguing - really drive us to ask
>just how much "chance" exists in the process of evolution. If the
>camera eye has evolved on at least six occasions, then it implies
>that there is a whole lot more going on than SIMPLY the pointless
>activity of a Blind Watchmaker.
>What strikes me, however, is that the only mechanisms we OBSERVE
>(there's that word again) for the modification of DNA are apparently
>random, then we are left with the difficult situation in which
>apparently random mutations of DNA lead to apparently common
>morphological characteristics.
>Against that background, I can only surmise that the one false
>assumption in all of this (and it IS an assumption) is that there
>are, in effect, an infinite number of possibilities. In short, I
>disagree with the below;
>But when you consider all the possible insertions or deletions,
>inversions of large or small segments, rearrangements, transposition
>of the whole sequence or a fragment of any of the thousands of
>transposable elements to any position in the genome, sometimes
>carrying pieces of flanking DNA with them, insertion of virus DNA,
>occassional insertion of bits of other foreign DNA, etc., etc, etc,
>I would say that there is no doubt that nothing approaching
>saturation of "all possible mutations" occurs.
>I don't disagree that this is "in principle" correct - that if one
>was to perform a mathematical calculation regarding prior
>probabilities taking into account all potential point mutations,
>recombinations, etc, etc, then we would be dealing with enormously
>improbable events and that your closing assessment in the above is
>spot on.
>HOWEVER, observation of what actually happens - that mutations are
>random, yet outcomes are strangely repeatable - suggests that, in
>practice, something truly remarkable is happening.
>On that basis, I shrink from affirming that the concept of "random
>mutation and natural selection" is quite so inept a mechanism for
>producing evolutionary change as Cameron seems to have suggested -
>but that's only secondary to the main point.
>The main point is this; if we restrict our comments to that which we
>have OBSERVED, then it is apparent that although we observe an
>apparently random process of DNA mutation as the "engine" of
>evolution, we also observe a repeated set of outcomes.

In order to get any idea why convergent evolution happens, we would
need to know what sets of genes are involved in each case, that is to
say, what genes present in the precursor species changed to give rise
to the structure in each case.

If the genes involved were completely unrelated to start with in each
case, but converged to something similar during the evolution of the
similar new structures, then extensive mutagenesis would be implied.
But if the genes are similar to start with in the parallel cases,
then there is no reason to suppose that saturation of mutation
possibilities is necessary.

Eukaryotic cells have chararteristic systems of genes which are
present in primitive organisms and in more elaborated forms in
higher organisms. Genes which are present as large sets of paralogs
in higher organisms are often present in many fewer copies with much
simpler regulation in lower organisms.

Given a certain set of genes to start with, I would guess there are
only so many ways to do the basic things that have to be done to form
a new structure. Since various organisms existing without the
structure would have overlapping sets of genes (and complex variants
of those genes,) it wouldn't be surprising that, if a particular
structure could arise in one species, something similar in outline
could develop in a more or less distant species. This doesn't imply
saturation of all possible mutations, only overlapping sets of gene
systems in the precursor organisms.

You may know all of this already - I don't know what your field is.
I'm just thinking about it the way a geneticist does. If you want to
know what happened, find out what genes are involved and how they
changed. If there were a bunch of related genes between two systems,
and eukaryotes have a buch of common gene systems, it's not
surprising that the systems can evolve in similar ways.


>And as repeatability seems antithetical to chance, I think it
>follows that the inference "mutations occur randomly therefore
>outcomes are a matter of chance" is unnecessary and unjustified. We
>should rather be asserting "mutations occur randomly HOWEVER
>outcomes are largely predictable" - which assertion is grounded in a
>posteriori observations as to what actually happens in biological
>systems rather than a priori theorizing about what should be the
>To unsubscribe, send a message to with
>"unsubscribe asa" (no quotes) as the body of the message.

To unsubscribe, send a message to with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Tue Apr 28 01:44:46 2009

This archive was generated by hypermail 2.1.8 : Tue Apr 28 2009 - 01:44:46 EDT