A Resolution, if not the desired one

0 views
Skip to first unread message

Tristan

unread,
Jul 28, 2008, 6:24:03 PM7/28/08
to Open-Ended Evolutionary Innovation / Quarantined Syst.
I believe this issue has a trivial resolution.

First, as a previous poster has said, the Earth most likely is a QS,
and even if it is not, there is SOME QS out there. The universe
(barring some fringe science being true) is a QS, and since life has
emerged in it, clearly complex life can emerge in a QS.

Even from a more practical point of view, let's say that evolution
either began or was sped up with the influx of some ET material. This
material came from somewhere, where it either evolved or again was
added from somewhere else. Since the universe is finitely old, this
regress cannot go back infinitely; there must be an origin, or
origins, of living forms - DNA, amino acids, however basic you want to
get. These origins could, in theory, easily all be on one planet.
Whether or not they were is an interesting question, but since you are
asking the much broader question of "what is possible," we can easily
settle on this: it is possible that all this life began on Earth.
After all, it had to begin somewhere, and Earth is the best candidate
we know of so far.

In general, if the Law of Conservation of Mass-Energy holds true (even
in the statistical form that Quantum Mechanics requires), there is an
immediate proof that this universe is a QS, and as such, your question
is resolved.

Second: if you are looking for a demonstration of this to supplement
this line of reasoning, I would say a few things. First, such
supplementation is not necessary from a purely scientific point of
view. The default position, based on the available evidence, must be
that panspermia is not the case. Evidence to the contrary will always
be considered, of course, but until such evidence is found, there is
no reason to suppose a theory that "multiplies entities beyond
necessity."

Additionally, and this is more important - I think your concept of
evolution is critically flawed in a way that affects whether or not
you consider a solution valid. Evolution by no means "Open-Ended" in
the way you describe. It does not create organisms with limitless
complexity. In fact, there is a perpetual evolutionary pressure
towards reduced complexity.

The evolutionary reason why in Drake's equation we can't consider the
[probability of intelligent life given the existence of life] to be
close to 1 -- the evolutionary reason why we're the only intelligent
species on the planet -- is because intelligence comes at an enormous
resource cost. True intelligence requires remarkable adaptation to
one's environment, which requires a long learning period, which
requires a long stage of youth. This means there's a long period
where we're defenseless and where we require parental investment,
compared to virtually all other mammals. Additionally, once we've
learned and become smart adults, we now have big brains which use a
ridiculous amount of nutrients to run at full capacity. Most if not
all animals are so successfully adapted to their environmental niche
via instinct, physical ability, and other attributes, that to move
closer to intelligence requires an objective loss of fitness, and so
those mutations get quickly culled.

So intelligence, or at least a move towards it, can often be
detrimental for a species. But there is a much more general reason
why complexity is limited. If a species' environment is static and
that species has a bounded variance in genetic expression via
mutation, evolution will guide that species to a stable local maximum
*and then leave it there*. Think of the Ginkgo tree, or alligators,
or so many species of bacteria. While they may have produced
evolutionary offshoots that became new species, those species
themselves still remain. This indicates that they continue to occupy
a niche at which they are functioning more or less optimally. (Of
course, "optimally" doesn't mean that nothing could do better; it
simply suggests that nothing within that species' normal variance of
mutation could do better.) Accordingly, the reason that so many
species "continue" to evolve is not because evolution is making them
arbitrarily complex; rather it's because their environment is in a
continual state of change. In fact, we've seen remarkable evidence of
a sudden environmental change that spawns a quick evolutionary change
in a species that has been relatively static for a few centuries; and
then the removal of that environmental change four decades later
causes the species in question to drift right back to the set of
traits they had originally.

Natural selection doesn't push species towards arbitrarily levels of
complexity, because complexity carries a high price: more nutrients
required to function, to say nothing of requirements spawned by the
specific sort of complexity. Natural selection pushes a species
towards local maxima of fitness, where fitness is specialized to the
niche that species currently occupies, and the local maximum is
defined by the highest maximum the species can attain via its variance
in genetic mutation. If a species doesn't have sporadic enough
mutations, they'll never get over the local hump to the next maximum
point - they'll just hover at a stage that's better than all around
it, but not necessarily (and in fact almost certainly not) the global
maximum.

Now, I don't know anything about the Avida project. But the upshot of
your misunderstanding about evolution is that, based on what people on
this board have said, Avida may be exactly the solution you're looking
for. It sounds to me like Avida creates a static environment (not
realistic, but a very nice simplifying step, and acceptable for
showing general principles in my opinion), and evolves digital
creatures or a chunk of code towards a local maximum in that
environment. It also sounds like it reaches that maximum in real
time, so that in the end you're left with a finished product, and not
something that continues to change. This is a necessary consequence
of having a static environment, however, and not something that should
disqualify it from consideration. The response, that "if you insist
on Open-Ended, I can simply slow it down to a crawl" is in fact
EXACTLY correct. There is no "Open-Ended" in an evolutionary sense,
except insofar as a perpetually changing environment creates the
illusion of one - and I don't think that's a useful point of
fixation. The illusion of Open-Ended that comes about due to
environmental change can be simulated simply by slowing down the rate
of adaptation to a fixed environment; it's really a less complicated
version of the same phenomenon.

In conclusion:

1) If you're looking for an abstract demonstration of OEEI/QS, the
universe should suffice, along with the logical point that any life or
organic matter *comes from somewhere*, and there's no logical reason
why it couldn't have all come from the same place.

2) You seem to have a misunderstanding of evolution - "Open Ended"
doesn't exist in evolution, except insofar as environments are almost
never static.

3) Based on (2), from what little I've heard, the Avida project seems
like a resolution of your concerns.

I'll add this:

4) If the motivation for this project is to provide an alternative
explanation to satisfy IDers - don't bother. Those people are not
taking rational points and looking for the truth, they're trying to
justify their own mythology. If you're looking for scientific truth,
ignore them. If you're trying to socially make the world a better
place, argue against them as Dawkins and Christopher Hitchens have. I
don't think a half-assed middle ground will do any good in the end.

-Tristan
Reply all
Reply to author
Forward
0 new messages