Re: Genes and Development Conference

Kevin O'Brien (
Thu, 1 Apr 1999 17:56:25 -0700

>At 08:17 PM 3/30/99 -0700, Kevin wrote:
>>>there seems to be an inconsistency in approach especially during
>>>discussions regarding information and/or complexity. Let me give
>>>an example. In discussions of spontaneous increase in complexity
>>>one often sees the example given of a seed growing into a tree
>>>without any apparent violation of natural law or intervention from
>>>an intelligent agent. The inevitable rejoinder is that this does not
>>>involve an increase in complexity (information) since the information
>>>required to direct the growth is contained in the DNA. Thus the
>>>information (complexity) remains constant. But another way of saying
>>>the same thing is to say that the growth is directed by a genetic
>>>programme. So, my question is whether this is a fundamental
>>>inconsistency or am I missing something?
>>For my part I'm not sure I understand the question. It's probably just
>>but I do not see what is suppose to be inconsistent.
>Well, perhaps my thinking is muddled, let's try again. The "rejoinder"
>I mention above is a perfectly legitimate (as far as I can tell)
>response if one is using the definition of complexity from
>algorithmic information theory. Here complexity has to do with
>the length of an algorithm and not its output. No matter how complex
>the Mandelbrot set may appear, it is actually very simple since
>it comes from a short algorithm. For this reason, I never use
>ice crystals or snow flakes as examples of spontaneous increases
>in complexity since actually the opposite is the case. A much
>better example is melting of a snow flake :).
>Back to the case of growth, this would not be an increase in
>complexity since this growth is the output of an algorithm,
>the genetic programme. But, if one rejects genetic reductionism
>then one can no longer say that growth from a seed is not an
>example of spontaneous increase in complexity consistent with
>natural laws and without intervention of intelligence. IOW,
>it is inconsistent to base ones argument on a principle that
>you have rejected.


So are you saying that IDers are doing this, basing their arguments on a
concept they have rejected? That for example they would deny genetic
reductionism, but would still claim that seed growth is not an example of
spontaneous increase in complexity, which requires genetic reductionism to
be true?

>>>OK, so one of the lofty goals (yet to be attained of course)
>>>of Webster, Goodwin and many others belonging to "that tradition
>>>in biology" is to develop a rational theory of form which would
>>>(among other things) explain homology in terms of fundamental
>>>theory, independent of history. My question then is, supposing
>>>for the moment that they are wildly successful, what effect would
>>>such a theory have upon the theory of common ancestry?
>>I guess for me it would make evolution predictable in the same way that
>>chemistry is predictable: knowing the starting conditions, we can say for
>>certain what the results will be. We wouldn't need to dig up
>>because we could accurately predict how a new species would evolve from an
>>old species. However, I'm not sure that will ever be possible.
>Your answer reminded me of a comment made during the discussion
>session following one of Goodwin's papers:
>"I find your conjecture very attractive, because it holds out
>the prospect of learning a hell of a lot of biology by just
>studying a little math." -- Epstein

"I like that, that's neat." Peter O'Toole as Henry II, _Becket_

>Yes, I'm sure there are biologists tearing their robes and
>gnashing their teeth after reading that ;-).
>But I don't think your answer is quite to the point I was
>trying to make. How could we predict ancestry ["how a new species
>would evolve from an old species"] from a theory which decouples
>form from ancestry?

If I understand your concept correctly, you decouple it from contigency, not
ancestry. You remove the random chance elements imposed by the environment
and leave the genetic changes, being predicted by laws and not
probabilities. You still have to know what the starting conditions
(ancestral species) are.

Kevin L. O'Brien