Re: Complexity and Information Theory

Chris Cogan (
Tue, 12 Oct 1999 19:48:55 -0700

> Conclusion
> Contrary to creationist contentions, evolution does not violate the Second
> Law of Thermodynamics or information theory.

Maybe not, but it violates the sensibilities of the creationists.

> The evolution of organisms does
> not violate the Second Law of Thermodynamics any more than the growth of
> individual organisms violates the Second Law. The creationist contention
> that intelligent information in DNA somehow gets around the Second Law is
> erroneous. The only requirement for localized decreases in thermodynamic
> entropy that accompany protein synthesis or organism growth is the
> requirement for an open system. Organisms are open thermodynamic systems
> long as they eat and breathe.
> The real connection between entropy and evolution comes from looking at
> information theory. The kind of entropy that is important to evolution is
> informational entropy. Like thermodynamic entropy on a universal scale,
> informational entropy tends to increase over time. Since an increase in
> informational entropy means the complexity of a message increases, the
> message transmitted by DNA over generations increases in complexity.

Not necessarily, but often. Overall, as long as informational complexity
*can* increase to the benefit of the genes, they will. Some organisms and
their genes may (or at least can) remain simple, but evolution means that
variations will keep cropping up that tend to fill available "niches," and
as long as there is a large open "space" of niches, this will mean
increasing informational entropy (complexity) in the genes (and often the
organisms that cart the genes around).

> The
> organisms specified by the message will be more complex as a result.


> Evolution thus seems to be an inevitable consequence of the properties of
> information. Selection provides a filter that determines which of the more
> complex messages survive.

If all variations survived, evolution (of a sort) would actually proceed fas
ter than it does (there'd be more chances for more variations and for those
occasional highly-unlikely variations). But, then, if they all survived, the
galaxy would be filled *solid* with life forms all emanating from Earth.

I'd like to add here my standard view of evolution:

All evolution by variation and culling is the evolution of information
"about" how information can survive given the media it is stored in and the
environment it is to survive in. It doesn't make any difference whether the
evolution is of genes, or societies, or systems of ideas, or the computer
"organisms" in Tom Ray's Tierra computer program, or computer viruses. This
same principle applies to them all.

This implies pretty much everything that the author of the piece said,
because, assuming that complex information-storing structures are possible,
evolution will keep "trying" variations (many of which will be increases in
complexity), and thus keep "pushing" into new territory wherever it is
possible to do so, gradually building complexity.

Further, using this approach, we can begin a mathematical treatment of the
nature of evolution and the conditions under which it can occur. The
information storage mechanism must be flexible enough to allow for
variations, but not so volatile that it doesn't actually store information.
It must have some means whereby it is reproduced with variations. Selection
is secondary, though it is critical in determining *which* packets of
information get transmitted to the future, and therefore in determining
*which* packets of information represent the best current "theories"
available about how information can survive in that environment.

We can also predict on this basis that there may be chunks of information
that "come along for the ride" buried among the active pieces of
information. Why? Because that is one method for that information to
survive, as long as it "picks" messages to ride along with that are
themselves survivors. The information is the thing; the phenotype is a
means, so it does not matter to a piece of information whether it
contributes to the phenotype or not, as long as the phenotype is able to see
to it that the information gets reproduced. Ultimately, of course, this may
amount to parasitism, if the hitch-hiking information is costly to
reproduce. On the other hand, it may, at some point, provide materials for
new *active* information, information that *does* contribute to the
preservation of the overall message, so it's not necessarily correct to say
that a piece of information that is currently just along for the ride is
ultimately useless.

Given the fundamental similarities to all forms of evolution by variation
and culling, we can apply this to the debate over evolution itself, and
observe that there is a typical evolutionary "arms race" going on between
the creationists and the evolutionists. Thus, in a way, the evolution of the
debate itself is a disconfirmation of the creationist's claims, since there
is no good reason to accept evolution in the case of the development of
human ideas and systems of ideas while rejecting it in the case of
biological development.