> I can say our nerves are slow, as this is a chemical
> process and I still react *fast* if I burn myself.
> It is "slow" AND "fast" -- depends on what you
> want to convey. There is no controversy.
Uh-huh.
> That's a different book, David! Obviously he left
> those statements for *your* book about Evolution.
> When you write it there will be lots of smart-asses
> who will tell you what they think you should have
> written. Don't pay any attention to them.
I never said Dawkins should have wrote what I wrote.
I'd be rather shocked if he did.
Since you've to have forgotton way Dawkins is an idiot,
allow me to repeat the list:
I've just read "Climbing Mount Improbable" by Dawkins
and I must say, it sucked. It was out of touch with
modern evolutionary theory and contained numerous
inconsistancies. I've outlined a few of these below.
The format is a quote from the book, followed by the
problems with this quote. For those who respect
Dawkins (I used to be one), I ask that you keep an
open mind when reading this, and notice that I use
facts, not opinions, to discount what Dawkins says.
Often, I use Dawkins own statements.
------------------
* From the cover flap.
"The only way to explain evolution is by slow,
gradual evolution."
* Most evolutionary changes occured in a 5-10 million
year period just before the beginning of the Cambrian
era. This period represents 0.125% - 0.25% of the
time evolution has been in effect on the planet, this
is lightning fast, an evolutionary "blink of an eye".
* Dawkins himself later describes the evolution of the
eye as "instantaneous".
* "Much of evolution's novelty arises from the
actualization of such latent potentials, not from the
slow and explicit improvement of an unchanged function
by natural selection."
- Steven J. Gould; Discover Magazine - 1996
* Preadaptaions are the use of structures that have
first emerged for function X for a new function Y.
Exaptations in the proper sense are the use of
structures that have first emerged for adaptively
neutral reasons for some function X. Both types of
exaptations appear to occur in our populations of
artificial organisms.
- Henrik Lund, Domenico Parisi; Preadaptation in
Populations of Neural Networks Evolving in a Changing
Environment; Artifical Life Volume 2, Number 2; MIT
Press
------------------
* From the chapter "Facing Mount Rushmore", pg 36.
"One of the main things that makes biomorphs so unamenable
to natural selection is that they are built of florescent
pixels on a two-dimensional screen. This two-dimensional
world does not lend itself to physics of real life in most
respects. Quantities like sharpness of teeth in predators
and strength of armor in prey [...] do not emerge naturally
in a world of two-dimensional pixels."
* Computer programs are n-dimensional. On a 64-bit machine
with maximum memory, dimensions of one byte can be added at
the rate of 100 per millisecond for 5000 years. Dozens of
such parameters can be displayed on the screen using standard
GUI tools available on all popular platforms. All of them
can be manipulated by code in the program. Computer games do
exactly this. This quote by Dawkins reveals his lack of
understanding of the tools he uses, as he states explicitly
that to get past two dimensions we must abandon the computer.
These tools, by the way, are _far_ from the state of the art.
------------------
* Dawkin's "Blind Watchmaker" program is obsolete. For
example, it does not use parasites, a decade-old (ten years!)
concept known to quickly and dramatically increase the
efficency of such programs. Nor does it take advantage of
new techniqes such as multiple fitness functions.
------------------
* From the chapter "Getting Off The Ground"
"I've stressed that going downhill is not allowed."
* Dawkins then shows three ways it _is_ allowed.
* Experiements in artifical evolution show that selecting
"less fit" individuals is often useful. One such method is
called "Tournament Selection", there are several others. In
fact, to make selection work as Dawkin's explains it one
must use a special type of selection known as "Elitism".
This type of selection often keeps the same individuals alive
one generation after another, a process that is not and cannot
be duplicated in nature.
------------------
* From the chapter "Getting Off The Ground"
Dawkin's implies that the Flounder fish is at a non-optimal
peak in evolution, giving the fact that its eye are on one
side of its body as proof.
* Dawkin's provides no reason _why_ such an arrangement is
non-optimal.
* If such an arrangement _were_ non-optimal, then according
to evolution, some more-fit creature would have displaced
the Flounder. In other words, if the Flounder were non-
optimal and unable to move out of this non-optimal
configuration, there would be no Flounder.
------------------
* From the chapter "Getting Off The Ground"
Dawkins gives the whale breathing air as an example of how
evolution cannot "go back" in the direction it came from.
* Dawkins then goes on to show that the whale has no need
to "go back" to breathing water.
* Dawkins fails to notice that the whale's limbs, once
similar to a wolf's, _have_ "gone back" to fins.
------------------
* From the chapter "Getting Off The Ground"
Dawkins states that his theory that evolution can't "go
back" is based on neo-Darwinism.
* Neo-Darwinism is a discredited theory from the 1920's
(seventy-five years ago! (three-quarters of a century!)).
------------------
* From the chapter "Forty-Fold Path To Enlightenment"
In his description of how to simulate evolution on a
computer, Dawkins says "Going upwards means mutating,
one small step at a time, and only accepting mutations
that improve [...] performance."
* While evolution can duplicated the effects of this
process, we know from computer models that this is
_not_ how evolution works. In fact, it is very
difficult to "force" evolution to keep only improvements.
Techniques which do not occur in nature, such as Elitism,
must be used.
------------------
* The basic premise of Dawkins book assumes that creatures
are at a single location on the fitness landscape.
* Creatures occupy _many_ locations on the fitness
landscape simultaniously. For example, a given creature
has a fitnes X at movement, a fitness Y at survival rate
of offspring, and a fitness Z at intellect. Deficientcies
in one area of often best made up for by strengths in
another area. For example, humans naturally have a
fitness of zero at flying, but rather than evolve wings,
they use their strong fitness at intelligence to build
airplanes.
-
David McF has convinced me to remove the "Blind Watchmaker"
program is obsolete reason from the list. It _is_ obsolete,
but it may be that dawkins wants it that way.
To the list we add the fact that Dawkins spends much time
on mutation when describing evolution and little on
cross-over, even though cross-over is more effective.
Dawkins' picture of Evolution is underpowered and in some
cases even wrong. The fact that he is your personal hero
does not change the fact that he's an idiot. :-p
-- David Leeper dleeper@gte.net Homo Deus http://home1.gte.net/dleeper/index.htm 1 + 1 != 2 http://home1.gte.net/dleeper/CMath.html