Reflector
On Mon, 23 Oct 2000 02:09:42 +0100, Richard Wein wrote:
[...]
>SJ>If Richard can show that I (or Dembski) is wrong, I (and I am sure
>>Dembski), would *thank* Richard.
RW>This time I'm keeping my resolution not to waste time in futile arguments
>with Stephen (unless one of our newly-found uncommitted lurkers asks to see
>a rebuttal of his post).
I am happy either way. I am trying to wind down my posts on the Calvin
Reflector, and I will probably from now on only reply to responses to my
posts.
RW>However, I've just finished the first draft of my article rebutting
>Dembski's Design Inference in detail, and will be posting it soon. In the
>meantime, here's a copy of my Amazon review of "The Design Inference".
>
>--------------------
>Pseudo-scientific mumbo-jumbo, June 3, 2000
The Design Inference was based on Dembski's Ph.D in Philosophy at the
University of Illinois:
"The monograph itself is a revised version of my philosophy
dissertation from the University of Illinois at Chicago (1996)."
(Dembski W.A., "The Design Inference: Eliminating Chance
Through Small Probabilities," Cambridge University Press:
Cambridge UK, 1998, p.xv)
and was published by Cambridge University Press.
So Richard is here claiming that: 1) the Philosophy department of the
University of Chicago awarded Dembski a Ph.D on the basis of; and 2)
Cambridge University Press allowed a book to be published under their
name which was; "Pseudo-scientific mumbojumbo"!
I think most people would conclude that it is more likely that Richard
simply doesn't understand it.
RW>Reviewer: Richard Wein (see more about me) from Bristol, England
>I'm hoping to find the time to write a web page analyzing this book in
>detail, but here's a brief summary.
>
>1. Dembski's Law of Small Probability relies on the ability to specify the
>relevant pattern AFTER observing an event. Unfortunately, there is no
>objective way to do this--in general we can fit many different patterns to a
>given event. Dembski fails to note this problem, instead giving the false
>impression that he is providing a definitive method for formulating
>specifications. (His "tractability" requirement is worthless.)
Richard contradicts himself. If it is only "*in general* we can fit many
different patterns to a given event" then that does not mean that there is
"*no* objective way to do this."
Also, it does not necessarily follow that "many different patterns" are not
"objective".
And Richard does not say how many is "many". It may be that there really
are only a few that truly "fit ... a given event".
Moreover, even if there were a great "many different patterns to a given
event" they *all* may be consistent with design.
Richard seems to be getting confused with a "pattern" and a number of
possible "causal stories" to explain that pattern (see TDI, p.xii)
Besides, if this were the case, then the police and SETI might as well give
up! But the fact is that we can infer design from a "pattern". Dembski gives
the example of an electoral official who 40 times out of 41 put his preferred
political party at the top of the ballot paper.
Here in Western Australia a person won several prizes on a radio show
which were awarded to the first person who rang in after the announcer
gave a signal. The company refused to award him any more prizes when it
found out he worked for the telephone company. They suspected fraud
based on the small probability of one person being first to call in so many
times and the pattern that he may have been in a position to ensure only his
call got through. He was later charged and convicted of fraud when an
investigation revealed he had in fact intercepted calls to the radio station so
only his went through.
RW>2. Dembski's Design Inference amounts to no more than this: once we've
>eliminated all the natural explanations we can think of, we must conclude
>design, and the way we eliminate chance explanations is by using the Law of
>Small Probability. However, he dresses up this simple idea with a lot of
>pseudo-scientific mumbo-jumbo (almost the whole of Chapter 2 fits into this
>category). In so doing, he obscures some important points. For example,
>while the formalized statement of the Design Inference reveals that we must
>consider and reject ALL possible chance hypotheses before concluding design,
No. All *known* "chance" and law "hypotheses before concluding design".
That is in fact how SETI works.
ID is falsifiable in that naturalists can always come up with a naturalistic
law-chance explanation which trumps design.
RW>the less formal description of the Explanatory Filter fails to mention this
>vital point, and almost all the examples in the book deal with only a single
>chance hypothesis (the only exception is a minor example relegated to a
>footnote). Also, Dembski fails to note that there may be explanations of
>which we're not aware, and so any inference of design must be considered
>provisional. (In a separate article, Dembski falsely claims that "the
>Explanatory Filter successfully avoids false positives".)
Of *course* "there may be explanations of which we're not aware"! But
what Richard, as a committed atheist, wants is for naturalistic explanations
to *always* be preferred over design, even if the naturalistic explanations
are unknown.
This would mean that even if design was *real*, Richard would prevent it
being admissible within science. In that case "science" would be protecting
*falsehood* in the name of a naturalistic *philosophy*
RW>3. In this book, Dembski doesn't actually apply the Design Inference to the
>question of design in nature (despite the reviewer below who writes:
>"Dembski stands by his calculation"!). Elsewhere, he does claim to have
>detected design in nature, but I've never yet seen him provide the
>probability calculation needed to substantiate this claim.
This is just an argument from ignorance on Richard's part. Richard might
add that TDI is the *only* book by Dembski that he has read! Before
Richard makes such sweeping claims, it is incumbent on him to thoroughly
research all Dembski's known writings.
RW>"The Design
>Inference", Dembski writes: "Do the calculation. Take the numbers seriously.
>See if the underlying probabilities really are small enough to yield
>design." But he never heeds his own advice! In reality, the calculation of a
>probability for an event in nature is far too complicated to be practical.
>(I've seen attempts to apply Dembski's methods to nature, but they're always
>based on totally unrealistic assumptions, which make the results worthless.)
The Design Inference was Dembski's Ph.D thesis at the University of
Chicago, so it was meant primarily as developing a methodology by which
design in general could be inferred:
"Who will want to read this monograph? Certainly anyone
interested in the logic of probabilistic inferences. This includes
logicians, epistemologists, philosophers of science, probabilists,
statisticians, and computational complexity theorists. Nevertheless,
a much broader audience has a vital stake in the results of this
monograph. Indeed, anyone who employs small- probability
chance-elimination arguments for a living will want to know the
results of this monograph. The broader audience of this work
therefore includes forensic scientists, SETI researchers, insurance
fraud investigators, debunkers of psychic phenomena, origins-of-life
researchers, intellectual property attorneys, investigators of data
falsification, cryptographers, parapsychology researchers, and
programmers of (pseudo-) random number generators." (Dembski
W.A., "The Design Inference," 1998, p.xii)
Dembski has developed the application of "the Design Inference to the
question of design in nature" in his later book, "Intelligent Design".
RW>In short, this book says nothing of any value. However it is being used by
>Dembski and his supporters to provide a false aura of legitimacy for their
>claims to have detected intelligent design in nature.
Richard only *assumes* it is "false" because his metaphysical *assumption*
is atheism. Hence if there really was detectable design in nature, Richard
would always deny it and assume some naturalistic explanation.
[...]
RW>"Do the calculation. Take the numbers seriously. See if the underlying
>probabilities really are small enough to yield design."
> -- W. A. Dembski, who has never presented any calculation to back up his
>claim to have detected Intelligent Design in life.
This is misleading, as I have pointed out:
1) in the context of "The Design Inference" Dembski is not necessarily
referring to "Intelligent Design in life" and he does present a "calculation"
in detecting design in the case of electoral fraud mentioned above;
2) There is no special "calculation" required - Dembski's explanatory filter
uses existing probability calculations, as he makes clear on the same page;
3) Dembski in his later book "Intelligent Design" gives an example of the
bacterial flagellum of having a CSI value of greater than 500 bits, which
presumably is easily calculated from standard information theory formulae
(e.g. in Yockey's writings), and would not be disputed even by Dawkins.
Steve
--------------------------------------------------------------------------
"Unfortunately, the fact that scientists have devoted their lives to the study
of Darwinism does not automatically mean that the theory is necessarily
scientific. The alchemists in the Middle Ages spent their time and energy
trying to convert base metals into gold and, of course, failed. We can now
see that the theories underlying the alchemists' efforts were fundamentally
mistaken, and although they would undoubtedly have considered
themselves 'scientists', we would hesitate today to call their experiments
scientific." (Leith B., "The Descent of Darwin: A Handbook of Doubts
about Darwinism," Collins: London, 1982, pp.26-27)
Stephen E. Jones | Ph. +61 8 9448 7439 | http://www.iinet.net.au/~sejones
--------------------------------------------------------------------------
This archive was generated by hypermail 2b29 : Tue Oct 24 2000 - 19:07:39 EDT