Re: [asa] The Myth of the Rejected ID Paper (science stoppers and OOL)

From: Rich Blinne <>
Date: Sat Jul 05 2008 - 20:49:08 EDT

On Jul 5, 2008, at 5:31 PM, Murray Hogg wrote:

> And if you think about the above for long enough, you might see why
> I'm rather adamant that ID is not a "science stopper." In turns out
> that unless one allows that Dembski in particular HAS gone further
> than a claim of ignorance, THEN my own critique of ID theory
> actually becomes groundless! One can't say that ID theorists have
> FAILED to convince in regards to the presence of irreducibly complex
> systems UNLESS one acknowledges that they have at least made the
> attempt!

They are not science stoppers in the sense from both specified and
irreducible complexity you can derive testable propositions from them.
It is an unfair characterization of ID's hypotheses as hypotheses are
not science since they can be falsified. But, ID did not themselves
try to falsify them. So, no, they did not make the attempt. Rather,
others did it for them. In the case of Behe, the falsifying
information comes from all the studies that he ignored on the witness
stand at Dover. Dembski claims that information cannot increase as
the result of evolution. That's a testable proposition, too, and it
also has been falsified. See the following from Miller's Only a Theory:

> If Dembski’s claim that the mechanism of evolution cannot produce
> new information was correct, we’d be able to verify it by using
> computer programs that mimic evolution. There are quite a few of
> these around, but one of the most straightforward has been written
> by Thomas Schneider of the National Institutes of Health (figure
> 3.4). He calls his program ev, for evolution, and it starts with a
> set of completely random sequences of 256 bases, sequences written
> in the four-letter language of the DNA bases (A, T, G, and C). At
> each generation, it evaluates how well a protein binds to
> (recognizes) a collection of “binding sites,” places where a small
> protein encoded by the same DNA sequence might bind. At the
> beginning of a test run, of course, they don’t bind very well.
> Nonetheless some of these random sequences do bind a little better
> than others, so Schneider’s program allows half of the sequences,
> the half that bind best, to form the next generation. [25]
> That succeeding generation is formed by introducing unpredictable
> errors (mutations) into the successful sequences, and then testing
> their binding ability again. Schneider’s program automatically
> calculates the information content of the sequence and shows how it
> changes as this process of selection, reproduction, mutation, and
> selection continues. The results are striking.
> The information content of the sequence rises steadily throughout
> the selection process and gradually levels off at a point where the
> fit between protein and binding sequence is nearly perfect. The
> strength of Schneider’s simulation is that the actual content of
> information can be measured mathematically in terms of bits of
> information. His measurement is taken according to methods
> developed by Claude Shannon, the father of modern information
> theory, and unequivocally shows that this process leads to an
> objective and quantifiable gain in information. [RDB Note: I use
> information theory in my work. While Dembski abuses Shannon,
> Schneider does not. He uses information theory in a straightforward
> fashion.] To find out how this happens, Schneider allows the
> simulation to proceed as before, but removes the selective step in
> which only the best-binding sequences are allowed into the next
> generation. The information content of the sequences quickly drops
> back to zero, showing that high information content is directly
> dependent upon a continuing process of selection.
> What’s needed to drive this increase? Just three things: selection,
> replication, and mutation. The best sequences are selected for
> replication into the next generation, then they are mutated, and
> then they are selected again. It’s no coincidence that the same
> three things are required for evolution, since what we are observing
> is nothing less than evolution on a small, observable scale.
> Figure 3.4: Natural selection is responsible for an increase in
> informational complexity. A computer simulation of evolution by
> natural selection shows that randomized information increases as
> the result of selection applied to an evolving digital “organism.”
> The ev program, written by Thomas Schneider of the National
> Institutes of Health, simulates a protein binding site and
> demonstrates how the pressure of natural selection increases the
> information content of the site. Note how the information content
> drops when selection is stopped. (Dr. Thomas Schneider, National
> Cancer Institute, National Institutes of Health.)
> Where’s the new information coming from? Perhaps the investigator is
> sneaking it into the system at the start? No chance of that, since
> the starting sequences are completely randomized. Maybe there’s
> hidden information in the program itself? Not likely. Schneider has
> made the source code of his program open for inspection, and there
> isn’t even a hint of such nonsense. Did Schneider rig the
> parameters of the program to get the result he wanted? Not at all.
> In fact, changing the program in just about any way still results
> in an increase in measurable information, so long as we keep those
> three elements—selection, replication, and mutation—intact. Where
> the information “comes from” is, in fact, from the selective process
> itself.
> Does this mean that evolution gives us a “free lunch,” that we get
> something for nothing in terms of information content? No. In fact,
> when you think about it, a very high price is paid to produce that
> information and then to keep it. That price is the cost of
> replication and selection. Schneider’s model requires an
> extravagant degree of waste, since half of the “organisms” (his DNA
> sequences) must be thrown away in each generation, and
> considerable energy supplied to replicate the surviving organisms,
> which then undergo a round of mutations to generate further
> variation. The information in the system is generated and preserved
> by this costly process of selection and replication.
> [25] Dr. Schneider has made summaries of his results, as well as the
> ev program itself, publicly available at
> [RDB Note: Figure 3.4 in Only a Theory is the graph on this web
> site.]


Murray continues:

> And one of my suspicions, I have to say, is that because ID
> theorists HAVE attempted to go beyond claims of ignorance, to
> actually raise ID to the status of a testable hypothesis, THEN it
> actually bolsters the perception of persecution when it is claimed
> that ID is ONLY an argument from ignorance. Rather than claim that
> ID is "not science", or even that it is "bad science", I think it
> should be argued that it is wrong. So, in that respect, I think that
> Pim's response needs the sort of qualification I have been offering.
> I don't, if Pim will forgive me for saying, think that Pim's
> position is VERY wrong, just wrong enough to need tweaking to avoid
> what seems to me a dangerous misrepresentation.

Hopefully, I just did that kind of qualification above.

> At the end of the day I personally think that an objection to ID as
> an argument from ignorance should be retired as manifestly false and
> positively harmful. False because ID theorists HAVE attempted to
> show that the issue ISN'T merely ignorance. Harmful, because it
> perpetuates the myth (?) of persecution. Instead I think that it
> should be argued that - just as Johnson and Nelson have acknowledged
> - even when taken on its own terms ID theory seems not to have
> successfully demonstrated its case.
> Your thoughts?

I would add that arguments that I personally am more fond of, e.g.
fine tuning and that evolution as of yet does a poor job of explaining
moral behavior, are just as much arguments from ignorance. If
evolution does end up explaining fine tuning, or moral behavior or
OOL, etc., leaning too hard on these arguments will produce just as
much damaging backpedaling. This whole exercise should not produce any
sense of superiority on our part but careful introspection to make
sure that we are not likewise using "dangerous" apologetic arguments.
Or, at the very least provide sufficient qualifications on them.

Rich Blinne
Member ASA

To unsubscribe, send a message to with
"unsubscribe asa" (no quotes) as the body of the message.
Received on Sat Jul 5 20:49:38 2008

This archive was generated by hypermail 2.1.8 : Sat Jul 05 2008 - 20:49:38 EDT