[...]
JB:==
>You did not, however, respond to the lengthy Denton quote I posted. I see
>nothing wrong with it. Instead, you focus on "pure chance" as if it some sort
>of phantom. It's also a convenient way to get off the issue. So forget that
>phrase for a moment and respond to the quote that I so painstakingly typed
>out!
>
Ho Ho, your good Jim :). Trying to evade the issue yourself and then accusing me
of same. Let's remember that my complaint was regarding the quote given by
Randy, not your quotes. The whole issue I was responding to was Denton's
characterization of Darwinism as pure chance. Pure chance is not a phantom,
it was the central issue of my reply to Randy.
Now, as to your diligently typed quotes ;-), let me say first that from my
response to Randy one should not conclude that I'm some sort of neo-
Darwinian wacko myself. I don't find classical ultra-gradualistic
Dawkinsonian type Darwinism to be very convincing and this is
essentially what Denton is criticizing. I am myself always on the
lookout for good arguments against this extremist view of evolution
so I'm certainly not going to try to defend it. But the arguments need
to be good arguments and mischaracterizing Darwinism as pure chance
is not a good argument.
Well, I'm beating around the bush somewhat, on to your quotes.
About the only thing I can agree with is that showing something
is not impossible is not the same as showing its probable. I have
the same complaint against Dawkins sometimes. He goes on and
on for pages on something and then his final conclusion is
something like " ... and so we find that the evolution of such and
such is possible". And I say to myself, so its possible, so what?
So I agree with you (and Denton) on this point. And your saying
to yourself so what? Harper's usually not this agreeable, there
must be a catch :).
Well, the catch is that going on and on about how improbable
something is not the same as showing that its improbable.
Denton is making a fundamental mistake in probability. He's
assuming (1) that there are many many many multitudinously
many paths that could be taken, only an exceedingly small
number of which lead to the desired result (an eye say) and
(2) that the random search involves a random selection among
all these possibilities each of which occurs *with equal probability*.
It is the highlighted part which is the most serious error. As
part of my (long ago) posts on algorithmic information theory
I worked out several simple example problems, one of which
was intended to show the types of errors that might be produced
when making the "with equal probability" assumption. Again
these were really simple examples involving tossing a fair and
a weighted coin, yet the error introduced in making the erroneous
assumption was, if I remember correctly, about 20 orders of
magnitude. This is why I could feel fairly confident in allowing
you a leeway of 10 orders :).
Now lets consider another problem that will be sure to give Glenn
a thrill. Lets consider a problem from nonlinear dynamics, say the
double pendulum. There are various ways of "watching" the dynamics
of this problem unfold with time. One way is to look at the two angles
that the two pendulum arms make with the horizontal by plotting
one angle versus the other. The best way to get this example is to
actually watch this plot unfold. Anyway, suppose we were to partition
the screen into a bunch of equally sized squares on 200x200 grid
and then ask the question "are all the squares visited by the system
with equal probability". The answer is obviously no. In fact, some of
the squares will never be visited because their coordinates correspond
to a situation which violates some fundamental law like the conservation
of energy.
Well, I could go on and on this. But I think I have established the
point well enough. In a complicated process such as evolution, exceedingly
more complex than my simple double pendulum example, one simply
cannot assume that there are multitudinous paths with equal probability.
Actually, what Goodwin and others are trying to show is that practically all
the "possible" paths are not really possible after all and are excluded
for some fundamental reason so that evolution is driven in certain
directions (generic forms for example) by the nonlinear dynamics with
practically no role being played by natural selection.
BH
><<Jim, what is the PROBABILITY of evolution?
>
JB
>Lower than a gnat's ankle.
>
BH
><<I mean, just roughly, to within say 10 orders of magnitude. Please show your
>work.>>
>
BH
>Shall I apply the same "method" you used to measure and conclude that my
>arguments apply at least "10 times" more to ID? ;-)
>
Clever reply. Okay, I'll spot you the factor of ten then and you only have
to give me the probability to within eleven orders of magnitude instead
of ten.
Brian Harper | "If you don't understand
Associate Professor | something and want to
Applied Mechanics | sound profound, use the
The Ohio State University | word 'entropy'"
| -- Morrowitz
Bastion for the naturalistic |
rulers of science |