Reflectorites
On Mon, 23 Oct 2000 12:31:06 +0100, Richard Wein wrote:
[...]
>>SJ>I would appreciate Ivar quoting where in "The Design Inference" Dembski
>>>says that: "one calculates ... all possible (non-design) hypotheses."
RW>The fact that Stephen was unaware that Dembski requires us to
>consider all relevant chance hypotheses,
No. Ivar said "all possible (non-design) hypotheses", not "all relevant
chance hypotheses".
RW>a fact which is fundamental to the
>design inference, indicates that any claims he has made up until now that
>the design inference has actually been applied have been made in ignorance
>of what the method of the design inference actually is!
No. I was trying to clarify what Ivar was getting at.
I was well aware that Dembski's explanatory filter aims to eliminate chance,
which means "all relevant chance hypotheses".
I regard this as meaning "all" *known* and "relevant chance hypotheses".
I don't regard this as the same as "all possible (non-design) hypotheses"
because that would require omniscience.
RW>However I don't blame Stephen for this. I blame Dembski, who has been
>equivocal and obfuscatory in his descriptions of the design inference.
The problem (as usual) is not in those who Richard criticises, but in
Richard himself. Richard seems to have such a strong need to prove himself
right and his opponent wrong (and not merely wrong but permanently
beyond consideration="irrational", "nonsense", "absurd", etc) that he
assumes his opponents must be saying what Richard *wants* them to say,
instead of him taking the trouble to find out what they actually *are*
saying!
A little reality check might help Richard (and Ivar too BTW). Dembski has
Ph.D's in psychology, philosophy, mathematics, physics and theology (not
an easy subject BTW). Before Richard assumes that such a person has
made an obvious and glaring error (which was missed by his Ph.D
supervisors and peer-reviewers too) Richard might just consider that
maybe he (Richard) has got it wrong?
RW>Many
>other people (both critics and supporters of Dembski) have made the same
>mistake as Stephen.
See above. Richard has not shown there was a "mistake" on the part of
Dembski in the first place.
>IY>Page 222:
>>
>>"..., we obtain the following definitive statement of the design
>>inference (cf. Section 2.2):
>>
>>"The Design Inference. Suppose a subject S has identified all the
>>relevant chance hypotheses H that could be responsible for some event
>>E. Suppose further that S has identified (1) a probability measure
>>P that estimates likelihoods with respect to the chance hypotheses
>>in H...."
Yes - "all the relevant chance hypotheses". There is nothing strange about
this. It is what science does all the time. In particular those sciences and
occupations which depend on identifying design, such as:
"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)
>IY>The paragraph goes on for half a page and uses lots of mathematical
>>symbols that I cannot reproduce in a post like this.
>
>I recommend that the casual reader refer to the equations on pp 50-51, which
>present a simplified version of the design inference. In particular, Premise
>3 at the top of page 51 says (substituting text for symbols): for all
>relevant chance hypotheses that could be responsible for the event, the
>probability of the event is small. The full version of the equation (P2 on
>page 222) says essentially the same thing, except that it clarifies that the
>event must match a specification, i.e. be a specified event.
I don't understand the symbols, but I do understand the words: "all relevant
chance hypotheses that could be responsible for the event".
The "relevant" qualifies it as *known* - "RELEVANT implies a traceable,
significant, logical connection":
------------------------------------------------------------------------
http://m-w.com/cgi-bin/dictionary?book=Dictionary&va=relevant Main
Entry: relevant ... 1 a : having significant and demonstrable bearing on the
matter at hand b : affording evidence tending to prove or disprove the
matter at issue or under discussion <relevant testimony> c : having social
relevance 2 : PROPORTIONAL, RELATIVE - relevantly adverb
synonyms RELEVANT, GERMANE, MATERIAL, PERTINENT,
APPOSITE, APPLICABLE, APROPOS mean relating to or bearing upon
the matter in hand. RELEVANT implies a traceable, significant, logical
connection <found material relevant to her case>. GERMANE may
additionally imply a fitness for or appropriateness to the situation or
occasion <a point not germane to the discussion>. MATERIAL implies so
close a relationship that it cannot be dispensed with without serious
alteration of the case <facts material to the investigation>. PERTINENT
stresses a clear and decisive relevance <a pertinent observation>.
APPOSITE suggests a felicitous relevance <add an apposite quotation to
the definition>. APPLICABLE suggests the fitness of bringing a general
rule or principle to bear upon a particular case <the rule is not applicable in
this case>. APROPOS suggests being both relevant and opportune <the
quip was apropos>.
------------------------------------------------------------------------
RW>Dembski also writes:
>
>"In case still more chance hypotheses are operating, design follows only if
>each of these additional chance hypotheses gets eliminated as well, which
>means that the event has to be an SP event with respect to all the relevant
>chance hypotheses and in each case be specified as well." (p 44, footnote)
>
>"Since an event has to have small probability to eliminate chance, and since
>the design inference infers design by eliminating all relevant chance
>hypotheses, SP(E;H) has to be satisfied for all H in [curly H]." (p 52)
>
>"Given a chance hypothesis H that could conceivably explain E, S [a subject]
>is obliged to retain H as a live possibility until a positive warrant is
>found for rejecting it." (p 220)
Yes - "by eliminating all relevant chance hypotheses".
>IY>I substituted
>>the word "non-design" for the word "chance" because I thought it
>>would be clearer for people who had not read Dembski's book.
On such matters one should stick to what Dembski actually said. I cannot
see how "all possible (non-design) hypotheses" is clearer than "all relevant
chance hypotheses" and in fact Ivar's substitution is: a) wrong, substituting
"all possible" for "all relevant"; and 2) without the "possible" and
reinserting "relevant" it is then just a meaningless tautology. *Of course*
one detects design by eliminating "all [relevant] (non-design) hypotheses".
>IY>Dembski defines design on page 36 as the "set-theoretic complement"
>>of "regularity" and "chance" and later points out that regular
>>hypotheses can be treated as special cases of chance hypotheses in
>>which single outcomes have high probability.
And?
RW>This is another example of Dembski's obfuscation. Since regularity
>hypotheses are simply chance hypotheses which yield a high probability, why
>complicate matters by dividing our non-design hypotheses into these two
>separate categories,
Richard needs to show that "regularity hypotheses are simply chance
hypotheses which yield a high probability."
Is Richard saying that natural selection is the same thing as random
mutation?
RW>especially as Dembski never bothers to specify a
>boundary value to enable us to distinguish between a high probability and an
>intermediate one?
From memory he does, but he points out that this varies between
disciplines, so there is no hard and fast "boundary value". It is up to each
scientific discipline and occupation using the design inference to agree on
their own boundary values.
I am re-reading TDI again, so if Richard demands I find it, he might have
to wait a few weeks!
>SJ>Essentially, Dembski's claim is that if all natural (i.e., non-design)
>>explanations can be shown to be extremely unlikely, then we can safely
>>conclude design.
There is a qualifier here: and also fit an independent pattern, i.e. a
specification. An arrow hitting any spot on a wall could be random. An
arrow hitting a bull's-eye previously painted on the wall is most probably
design.
>>>IY>If you or others want to propose the hypothesis that complex things
>>>>such as the genetic code are always assembled by intelligent beings,
>>>>that is fine with me.
>>SJ>This is in fact our universal experience that codes "are always assembled
>>>by intelligent beings." SETI is based on this.
>>>IY>But I don't see how you plan to collect evidence
>>>>to confirm your hypothesis.
>>SJ>See above. I am not sure that it *is* the "hypothesis".
>>>IY>You seem to be stuck with a sample size of one (man).
>>SJ>I am not sure what Ivar means. The hypothesis is not about Dembski so he
>>>is not the "sample".
>IY>By man, I meant mankind. Man seems eager to believe that all gods
The issue of "gods" or God is a separate issue and if I get into it, it will be
manipulated into another "four-alarm" mess by those who want to confuse
ID with creationism.
>IY>and all things that generate complexity, especially those in outer
>>space, must be man-like.
I understand that all SETI is assuming is at least a human level of
*intelligence*.
IY>But why should this be true?
There is nothing to say it "should...be true". In the case of SETI this is an
*assumption*.
IY>If there is a
>>Creator, is He so limited that He ran out of ideas when He created man?
Pass. See above on confusing ID with creationism.
[...]
>>>IY>Further, this is not Dembski's approach. Dembski says that he is not
>>>>proposing "design" as an hypothesis; rather, he is proposing to deduce
>>>>design by eliminating all alternative hypotheses to design.
>>SJ>I would like appreciate Ivar quoting where "Dembski says that he is not
>>>proposing `design' as an hypothesis."
>>
>>Page 68:
>>
>>"The design inference is in the business of eliminating hypotheses,
>>not confirming them. .... Because the design inference is eliminative,
>>there is no "design hypothesis" against which the relevant chance
>>hypotheses compete...."
OK. Note that "design hypothesis" is in quotes. In the context Dembski is
saying that there is no "design hypothesis" as such competing with 1.
chance and 2. law. Design is only inferred after chance and law fail as
explanations. Design never *competes* with chance and law. They win
without a contest if they can explain the event(s) in question.
RW>Yvar, I think what Dembski means here is just that the design hypothesis
>doesn't compete with other hypotheses in the design inference. Although it
>may sound like he's saying that there's no design hypothesis at all, he
>obviously can't mean that. The design hypothesis is what Dembski is trying
>to establish. The important point is that he establishes it purely by
>eliminating all non-design hypotheses, not by considering the merits of the
>design hypothesis.
Agreed. I wrote what I did before I read what Richard said.
RW>So I think your statement that "Dembski says that he is not proposing
>`design' as an hypothesis" is misleading, and is just introducing an
>unnecessary element of confusion.
Agreed. It is rather *amazing* that Ivar could read it that way.
>>>IY>His approach is better characterized as philosophical rather than
>>>>scientific.
>>SJ>Not really. The same methodology of inferring intelligent cause by
>>>eliminating unintelligent natural causes underlies the *sciences* of
>>>archaeology and SETI.
>IY>The scientific approach is to propose hypotheses and then seek to
>>confirm or refute them. Dembski isn't doing this. Rather, he is
>>attempting to show that an intelligent being, hitherto unknown to
>>science, is logically necessary. He is making a kind of ontological
>>argument for God.
No. He is setting up a robust methodology to detect design. Methodologies
are part of science. In our Biology labs we are taught about setting up the
Null Hypothesis and then testing the result against it, by comparison with
statistical standards Chi-square and Student's t values.
RW>I don't agree with you. I think that in any inference of design where we
>don't have positive proof (like having actually seen the object being made),
>we must use a process of elimination, and, beyond a certain point, we must
>simply say "We consider it inconceivable that any natural process could have
>created this object, so it must be designed." Suppose, for example, that
>SETI receives an extraterrestrial message like the one in the book Contact.
>We cannot prove that the message was not produced by some unknown natural
>process, i.e. in accordance with some unknown chance hypothesis.
>Nevertheless, because we can't conceive of any natural process producing
>such an effect, we reject all such unknown hypotheses.
Not bad. But I don't agree with the subjective "inconceivable". Dembski
substitutes an objective measure, e.g. a probability bound of 10^-150, for
subjective judgments on what is " inconceivable".
RW>Effectively this is what has happened in the past. Before Darwin, a natural
>explanation for the complexity of life seemed inconceivable, so most people
>did accept the existence of a designer. Even Dawkins says that he could not
>have been an intellectually fulfilled atheist before Darwin. But, now that
>we have a natural explanation, the complexity of life cannot be offered as
>scientific evidence for a designer without ruling out that explanation.
Two points: 1) Even if Darwin did completely explain "the complexity of
life" that would not establish the non-"existence of a designer". Paley's in
his Natural Theology was aware of evolutionary theories like Lamarck's
and Erasmus Darwin and does not base his watchmaker argument on how
the design was realised. Paley explicitly says that a machine produced by
another machine, is still designed (see tagline). As even Dawkins says, the
"blind forces of physics" have to be "deployed in a very special way" for
Darwinism to work:
"All appearances to the contrary, the only watchmaker in nature is
the blind forces of physics, albeit deployed in a very special way.
(Dawkins R., "The Blind Watchmaker," 1991, p.5)
Darwin's argument effectively destroyed extreme creationist arguments that
each *species* was created as is, where is. But there is no evidence that
this was a *Biblical* creationist position. As the historian Gale points out,
Darwin set up a straw man creationist position:
"Yet even here, where Darwin's arguments are strongest, nagging
questions remain. For example, a reader of the Origin might be
justified in wondering what Creationist view Darwin is referring to.
Perhaps this is a problem more for the present-day reader. Darwin's
contemporaries may have known exactly what he meant, though I
doubt it. Often the Creationist position seems merely a straw man-
set up only to be knocked down." (Gale B.G., "Evolution Without
Evidence," 1982, p.139)
2) Few biologists these days would claim that Darwinism has (or even can)
explain "the complexity of life".
RW>If Dembski was able to rule out the theory of evolution, that wouldn't prove
>that life was designed, but it would strongly support the case.
I note how Richard switches from "Darwin" to "the theory of evolution".
Evolution has been defined as "a change in gene frequencies in a
population" so who could "rule out" a "theory" which proposed that!
RW>Even Dawkins
>has said something to the effect that Darwinian evolution is the only
>possible natural explanation for the complexity of life. I think some other
>people (like Kaufmann) would disagree.
Indeed he has and indeed they do!
RW>Anyway, it would certainly give a massive boost to ID.
What? "If Dembski was able to rule out the theory of evolution"? It sure
would! :-)
RW>However it's important to bear in mind that the theory
>of evolution does not yield a single chance hypothesis, but a whole family
>of them. There are many unknowns in the theory, and Dembski would have to
>rule out all possible variations.
Dembski does not have to "rule out all possible variations" of "the theory
of evolution". What he (and ID) aims to do is demonstrate that there is
detectable evidence of intelligent causation in the history of life, that cannot
reasonably be explained by any combination of known law and chance
unintelligent causes.
RW>Given the difficulty of applying the design inference to living organisms,
>perhaps IDers would prefer to apply it to the origin of the first
>self-replicating entity.
Actually it is not all that much of a "difficulty". Mike Behe is "applying" his
IC version of "the design inference to living organisms", i.e. biomolecular
subsystems of "living organisms".
RW>(I think this is what DNAUnion would like.)
It is not a case of either/or. The "design inference" can be applied to
*both* "living organisms" and "the first self-replicating entity", i.e. the
origin of life.
RW>But it
>can only be applied to certain simplistic chance hypotheses, such as a
>complex molecule being formed by the random accumulation of amino acids. But
>we do not know that this is the only possibility. Cairns-Smith for
>example has proposed that the first self-replicating entities were crystals.
>Other OOL researchers are working on other possibilities.
They tend to be going around in circles now, recycling old "possibilities".
That shows that *none* of them work. Richard seems to think that having
a lot of mutually contradictory "possibilities" is a good thing.
RW>DNAUnion argues
>that, because we currently have no viable natural explanations, we must
>accept ID.
I am unaware of what DNAUnion's position is. But no one AFAIK among
the ID leadership would say that anyone "must accept ID".
What they are saying is that ID should not continue to be ruled out apriori.
RW>But ID researchers do not find the possibility of a natural
>explanation to be inconceivable, and frankly I'll put their judgement ahead
>of DNAUnion's (or the small minority of scientists who share DNAUnion's
>opinion).
On Mon, 23 Oct 2000 19:34:49 +0100, Richard Wein wrote:
RW>Oops, I wrote...
[...]
RW>Of course, that should have read "But OOL researchers..." ;-)
It was true either way. "ID researchers do not find the possibility of a
natural explanation" for the origin of life "to be inconceivable" they just
want to *see* that "natural explanation"!
And "ID researchers" do not assume that there *must be* a "natural
explanation", no matter what the evidence is pointing to. That is
philosophical naturalism.
>IY>Archaeologists know that man exists. They are proposing hypotheses
>>that certain objects were made or modified by man.
No. Archaeologists are looking for intelligently designed objects, no matter
which species made them. The Golan Venus I posted the other day is ~ 200
kya, i.e. ~ 100 kyr before "man" (i.e. anatomically modern Homo sapiens)
appears in the fossil record.
If geologists found a spaceship buried in strata 100 mya, they would call in
archaeologists.
Similarly SETI is looking for intelligently designed objects (i.e. messages)
the defining quality of which is that they were *not* "made or modified by
man"
RW>But suppose they found relics of a pre-Cambrian alien civilization. They
>might be able to recognize those as designed objects, i.e. they might
>conclude that there is no conceivable natural explanation for such objects.
I really should read on ahead! But then Richard and I would not show such
rare agreement! :-)
[...]
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.
Richard really should know that: 1) Dembski's book, "The Design
Inference" from which this is quoted (p.228), is Dembski's Ph.D thesis in
philosophy at the University of Chicago, and is about a methodology to
detect intelligent design in general, not "life" in particular; 2) in that book,
Dembski does in fact present a "calculation" to detect a case of electoral
fraud; and 3) there is no special "calculation" that Dembski is proposing,
just the ordinary probability calculations that statisticians and information
theorists make.
In his later book, "Intelligent Design" Dembski refers to the bacterial
flagellum having a complex specified information (CSI) value of greater
500 bits. Dembski does not give a calculation and in fact he may not have
personally calculated this. But it would not be particularly hard, using
standard information theory calculations that Yockey (for example) gives in
his various writings.
What Richard has not demonstrated is whether anyone of any note actually
question this. Dawkins, for example, would probably claim that 500 bits
was an *underestimate*!
Steve
--------------------------------------------------------------------------
"SUPPOSE, in the next place, that the person, who found the watch,
should, after some time, discover, that, in addition to all the properties
which he had hitherto observed in it, it possessed the unexpected property
of producing, in the course of its movement, another watch like itself (the
thing is conceivable;) that it contained within it a mechanism, a system of
parts, a mould for instance, or a complex adjustment of laths, files, and
other tools, evidently add separately calculated for this purpose; let us
inquire, what effect ought such a discovery to have upon his former
conclusion? I. The first effect would be to increase his admiration of the
contrivance, and his conviction of the consummate skill of the contriver.
Whether he regarded the object of the contrivance, the distinct apparatus,
the intricate, yet in many parts intelligible, mechanism by which it was
carried on, he would perceive, in this new observation, nothing but an
additional reason for doing what he had already done; for referring the
construction of the watch to design, and to supreme art. (Paley W.,
"Natural Theology: or, Evidences of the Existence and Attributes of the
Deity, Collected from the Appearances of Nature," [1802], St. Thomas
Press: Houston, TX, 1972, reprint, p.7)
Stephen E. Jones | Ph. +61 8 9448 7439 | http://www.iinet.net.au/~sejones
--------------------------------------------------------------------------
This archive was generated by hypermail 2b29 : Sun Oct 29 2000 - 17:35:40 EST