First of all, let me say that I agree with the overall conclusions
reached by Glenn and Bill, i.e. that the probability argument is
bad both for scientific and apologetic reasons. I differ only on
*why* its bad. Before starting, I also want to protest somewhat
on labeling this as a creationist argument. I don't want to offend
anyone, but the truth of the matter is that I have never read anything
written by a Scientific Creationist. Nevertheless, I have in my files
numerous probability calculations and arguments. Off the top of my
head, I can think of probability calculations being given by
Eddington, Hoyle, Shapiro [not the lawyer ;-)], Morrowitz, Quastler
and, of course, Yockey. In addition to this, I would bet that it
would be very difficult to find a recent book or article on the
origin of life which does not at least mention the improbability
of life arising purely by chance in a primordial soup of prebiotic
chemicals. I recall this point being made by both Richard Dawkins
and Daniel Dennett, also by Manfred Eigen if memory serves. The
probability argument is part of the scientific literature and had
a major role in the direction of abiogenesis research.
Creationists go wrong in the interpretation of the probability
calculation. This calculation is relevant to only one scenario
for the origin of life, i.e. the formation of life purely by
chance in a primeval soup. This idea is so foreign to the current
direction of research in abiogenesis that many on talk.origins
flatly claimed that no one had ever proposed that life could
arise purely by chance and that this was a strawman invented
by creationists. In fact, this was probably the number one
explanation around 30 years ago, here's one example:
_The creative accident_. Living things are highly
ordered systems--much more elaborately ordered than
any known nonliving things. Even the simplest living
things show great complexity. In terms of ordered
structure the distance between a bacterium and a man
is much less than between a bacterium and, say, a
giant electronic brain. In the general course of events
order tends to give way to less order, and not the
other way around; hence it is not easy to see how life
could have arisen from nonliving precursers. Several
explanations have been proposed, some of which are not
subject to scientific inquiry. Of those that are, the
most attractive is the proposition that nonliving
components have assumed a configuration compatible
with life through some lucky accident.
To be compatible with life, a configuration must combine
metabolic activity with stability and even adaptability,
and it must be able to reproduce itself from available
components. This is a big order. To assess its implications
we will investigate what seems to be a critical step in
the emergence of life: given a "rich, hot, salty sea"
containing organic molecules such as amino acids, sugars,
polyphosphates, pyrimidines, purines, and so on, what is
the probability of an occasion at which a random
configuration taken by such molecules is compatible with
life? To estimate this probability, two numbers are needed:
the probability that any given molecular configuration is
compatible with life, and the number of occasions available
for forming configurations.
-- Henry Quastler, _The Emergence of Biological Organization_,
Yale University Press, 1964.
The probability calculations performed by numerous highly respectable
individuals and published in the mainstream scientific literature
have had a very significant influence upon abiogenesis research.
The probabilty that life could arise purely by chance is so small
as to be practically indistinguishable from a miracle, so people have
discarded this notion and moved on to other ideas involving catalysis
on mineral surfaces, hypercycles and self-organization, nonlinear
dynamics, cumulative selection starting with simple self-replicators
etc. So, creationists go wrong by saying "wow, this is so improbable
that it couldn't have happened" instead of "wow, this is so improbable
that it couldn't have happened purely by chance".
Bill wrote:
"Together with cumulative selection and nonlinear dynamics, I
believe this argument pretty well demolishes the creationist
probability arguments."
My first comment to this is that Glenn's argument is not needed to
demolish the creationists argument from improbabilty. All one needs
is cumulative selection, nonlinear dynamics or any other scenario
that doesn't rely solely on chance. I hope this is clear. Try to
imagine for a moment that you are running a computer simulation
and that your screen represents some phase space on which the
trajectories around a strange attractor are being plotted. You are
sitting there mesmerized by the beauty of the pattern that is unfolding.
Now, ask yourself: are all points on your screen equally accessible?
The answer is clearly no, in fact some points on the screen will be
impossible to reach. One can say beyond any reasonable doubt that the
pattern that arises on the screen is not the result of some program that
selects points on the screen purely by chance. Why? It's too improbable
that random selection would produce a pattern.
Now the falacy of the creationist argument should be clear. Hopefully
no one will compute the probability of getting some pattern on the
screen and then conclude, since the probabilty is so low, that such
a pattern is impossible :-).
But this example also brings to mind a falacy that evolutionists are
often guilty of, one that occurs very often on talk.origins. Suppose
now that you have another program that selects points at random on the
screen and changes the color from black to white or white to black,
depending on what the initial color is when the point is selected.
You let this program run for awhile until some random array of white
dots is generated. Any *specific* arrangement of dots that result
has exactly the same probability of occurance as any other specific
arrangement, including the specific arrangement generated by the
strange attractor. Now suppose you are in a large computer lab with
one hundred computers. 99 of these have the random program running
while one has the strange attractor. You are told absolutely nothing
about the programs being used to generate the 100 images that you
see. Would you be surprised at all when you notice the one with the
strange attractor? According to the evolutionist probabilty argument,
you should not be surprised at all, since the specific arrangement on
that screen has exactly the same probability as that on any one of the
other 99 screens. So, you shrug your shoulders, say "so what", go home
and watch Rosanne on the tube ;-).
Well, there is something odd going on here. There is a falacy here but
it is somewhat subtle. I used a similar example to this one on
talk.origins awhile back using binary strings instead of computer
screens:
Suppose you tossed a fair coin 64 times, for each throw recording
a 1 for heads and a 0 for tails. Would either of the following two
sequences be more surprising?
(A) 1010101010101010101010101010101010101010101010101010101010101010
(B) 1110010001010001110000000000011101110101101110100101000111111010
For greater impact, I should have generated 99 sequences by tossing
a fair coin, put sequence A in somewhere and then said: "only one of
the following sequences was not generated by tossing a coin, can you
find it?
The person I was talking to was honest enough to admit that he would be
very surprised to see something like (A), but went on to say that,
given what he knows about probability, he shouldn't be surprised.
So close but yet so far, the reason that one could pick (A) out of
99 random sequences and could find the one computer terminal out of
a hundred (or a thousand or a million or a trillion) that has the
strange attractor is because of the pattern. Random processes are
expected to give random results. Very very occasionally they may not,
but this is so improbable that it can be discounted. This is well
known and can be proven, but I'm going to skip those details here.
I dare say that very few natural laws would have ever been discovered
if scientists actually employed the falacy mentioned above in practice.
For example, we observe that planets follow an elliptical path around
the sun. What do we conclude? Well, there are many many ways that
a planet could travel around the sun with any specific orbit being
equally probable with any other specific orbit. The ellipse just
happens to be the one selected at random from a number of possibilities.
It's really no big deal. No, the regular, orderly, predictable motion
of the planets should be enough in and of itself to convince us that
planets don't move about the sun at random and we conclude that there
must be some underlying law responsible for their motion. We would
conclude this even if we had no idea whatsoever what that law might
be.
To summarize: Non-randomness is something special. One doesn't
expect non-random results from a random process. Given this, the
observation of non-random results strongly suggests an underlying
law-like mechanism.
I should add here that I don't think Glenn is guilty of committing
the falacy described above, however, there is a real problem with
his proposal that is closely related to what is discussed above.
Since I've gone on too long here already, I'll save this for another
post.
===
Brian Harper:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=
"I believe there are 15,747,724,136,275,002,577,605,653,961,181,555,468,
044,717,914,527,116,709,366,231,425,076,185,631,031,296 protons in the
Universe and the same number of electrons." Arthur Stanley Eddington
:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=:=