Another Uninformed Critique of WIDF Fails

In 2009, Lita Cosner of Creation Ministries International posted a review of the 2004 book “Why Intelligent Design Fails”, claiming in summary that there was nothing in the book to cause discomfort to an “informed creationist”, and that actual problems for “intelligent design” creationism were rare in the book. Here in 2013, let’s have a look at Cosner’s specific critique of the chapter that Jeff Shallit and I contributed.

“Playing Games with Probability”

Shallit and Elsberry, in their essay “Playing Games with Probability”, claim that intelligent design theorists misuse probability in the same way that Bible code pseudo-scientists do. They more specifically argue that Dembski’s use of probability is unjustified and inconsistent (p. 130), and gives “wildly differing results” depending on which method is used (p. 132).

They, along with many of the other authors, use genetic algorithms, and specifically Dawkins’ “weasel” algorithm, to prove that natural selection and mutation are sufficient for evolution to occur. But, as even the authors acknowledge, they do not accurately represent biological evolution, so it is dishonest to use the algorithms as proof for evolution.11–13

It is common for evolutionists to use situations (as Shallit and Elsberry do), such as flipping a coin 50 times to come to an improbable sequence, to demonstrate how improbable things can happen. But in that example, there must be some outcome. But it was not necessary that the universe be suitable for life, or that life would occur even in a universe suitable for it. Since there are many more ways for the universe to be unsuitable for life than suitable for it, and for something to be dead than to be alive, this warrants a sufficient explanation.14

Let’s toss in Cosner’s associated footnotes:

11. Batten, D., Genetic algorithms—do they show that evolution works? , 2008.

12. Truman, R., The weasel returns: Truman replies to Curtis, Journal of Creation 15(2):55– 58, 2001.

13. Abel, D.L., The Capabilities of Chaos and Complexity, International Journal of Molecular Sciences 10:247–291, 9 January 2009 doi:10.3390/ijms10010247 (emphasis added): All too many evolutionary computationists fail to realize the purely formal nature of GA [genetic alogrithims] procedures. GAs are not dealing with physicodynamic cause-and-effect chains. First, what is being optimized is a formal representation of meaning and function. A representation of any kind cannot be reduced to inanimate physicality. Second, “potential solutions” are formal, not merely physical entities. Third, at each iteration (generation) a certain portion of the population of potential solutions is deliberately selected by the agent experimenter (artificial selection) to “breed” a new generation. The optimized solution was purposefully pursued at each iteration. The overall process was entirely goaldirected (formal). Real evolution has no goal. Fourth, a formal fitness function is used to define and measure the fittest solutions thus far to a certain formal problem. The act of defining and measuring, along with just about everything else in the GA procedure, is altogether formal, not physical.

14. Sarfati, By Design, ch. 16.

The first paragraph shows that Cosner read material that makes the second paragraph completely unsupportable. We certainly did note that Dembski’s use of probability was inconsistent, but we stated that this was apparent in an example of something Dembski himself critiqued, and that something was the Dawkins’ “weasel” program. Cosner’s second paragraph claims things that Cosner might wish that we said, but did not. Specifically, we never stated any such thing as genetic algorithms being “proof of evolution”. In dealing with Dembski’s critiques, one has no need of such claims. It is entirely sufficient to note that Dembski is misinformed about, overlooks, or misrepresents the properties of evolutionary computation systems that he criticizes or dismisses out of hand, or even (as we did) demonstrate inconsistencies within the critiques he makes of a single system. Genetic algorithms and evolutionary computation more broadly are useful rejoinders to certain broad claims religious antievolutionists make, but they do not directly bear on the question of a history of evolution and biological common descent as Cosner erroneously asserts we claimed. (Abel’s gobbledygook from footnote 13 is portentous but irrelevant; Dembski remains inconsistent in assigning probabilities, whether one calls the system Dembski critiqued “formal”, “informal”, or “stunning in a mauve chiffon dress”.)

In the third paragraph, Cosner asserts, “It is common for evolutionists to use situations (as Shallit and Elsberry do), such as flipping a coin 50 times to come to an improbable sequence, to demonstrate how improbable things can happen.” This only makes sense if we were attempting to make some sort of apologia for small probabilities. Cosner rather thoroughly misses our point. That’s the charitable reading of Cosner’s critique. On page 122, we noted the central problem that for a given observation of a result such as that of a sequence of coin flips, the likelihood of any possible such sequence is exactly equal to any other such sequence, but we would like to distinguish between those that can be ascribed to chance and those which cannot be ascribed to chance. This is exactly the thing that Dembski claims to have halfway solved, but which we demonstrate he is not even close to doing. Our critique is aimed at showing that Dembski has not sufficiently justified his claims.

Let’s see Cosner’s dismissive conclusion:

To refute all the arguments in this book would require a book in and of itself. Indeed, most of the arguments are addressed in Jonathan Sarfati’s By Design, though it was not written to refute this book specifically. Where Why Intelligent Design Fails does point out a valid weakness in ID theory (which is rare), it is one of the weaknesses which results from ID’s refusal to name the Creator, or from the acceptance of theistic evolution by some ID theorists. These are weaknesses that biblical creation does not share with ID theory, so there is nothing in this book that should challenge an informed creationist.

Our conclusion is entirely overlooked by Cosner:

The bottom line is that Dembski’s specified complexity or complex specified information is an incoherent concept. It is unworkable, is not well-defined, and does not have the properties he claims for it. Even Dembski himself, in attempting to calculate the specified complexity of various events, uses an inconsistent methodology. Most important, specified complexity does not provide a way to distinguish designed objects from undesigned objects.

Biochemist Russell F. Doolittle (1983) once remarked, “The next time you hear creationists railing about the ‘impossibility’ of making a particular protein, whether hemoglobin or ribonuclease or cytochrome-c, you can smile wryly and know that they are nowhere near a consideration of the real issues” (261). That same wry smile might be useful to keep handy when reading Dembski’s claims.

The faults we note for Dembski’s ideas are not due to “failure to name the Creator”, and certainly have nothing to do with Dembski’s thoroughgoing rejection of theistic evolution (the actual case, to be distinguished from Cosner’s inexplicable assertion to the opposite effect). That also means the faults remain the same for those ideas whatever the belief status of the promulgator of them.

“Informed creationists” should know that Dembski’s ideas and more generally those of “intelligent design” creationism didn’t appear as a novelty. No less a personage than Henry Morris of the Institute for Creation Research pointed that out fairly bluntly in “The Design Revelation”:

Our other hesitation to get on this bandwagon is their use of the same arguments and evidences we Biblical creationists have used for years, while simultaneously trying to distance themselves from us. Our adherence to Biblical literalism is ridiculed by evolutionists, and the ID advocates would be embarrassed to be tarred with the same brush.

Morris specifically (!!) noted concordance between ideas Dembski writes about and ideas Morris wrote about long before Dembski appeared on the scene.

Dembski uses the term “specified complexity” as the main criterion for recognizing design. This has essentially the same meaning as “organized complexity,” which is more meaningful and which I have often used myself. He refers to the Borel number (1 in 1050) as what he calls a “universal probability bound,” below which chance is precluded. He himself calculates the total conceivable number of specified events throughout cosmic history to be 10150 with one chance out of that number as being the limit of chance. In a book4 written a quarter of a century ago, I had estimated this number to be 10110, and had also referred to the Borel number for comparison. His treatment did add the term “universal probability bound” to the rhetoric.

The notion that “informed creationists” could rest easy, that “Why Intelligent Design Fails” wasn’t about anything they have an interest in, is complete balderdash. “Intellgent design” creationism is just “creation science” with more Wite-Out, just as “creation science” is simply “scientific creationism”, and both of those are “biblical creationism” with Wite-Out. Dembski is just among the latest to mouth and embellish improbability arguments for religious antievolution. As Morris noted, Dembski modified the rhetoric, not the argument.