15 July 2006

Evolution Number Nine

Oh, dear, oh, dear. The Inscrutable Design folks are back to their usual misattributions of authority and dubious logic. Casey Luskin of the so-called "Discovery Institute" may be an "attorney," but he's not a very good writer. As quoted on Jason Rosenhouse's Evolutionblog, Luskin said:

After all, Darwin's theory of evolution by natural selection is fundamentally based upon an algorithm which uses a mathematically describable trial and error process to attempt to produce complexity. (emphasis added)

No. Natural processes are not "based upon" an algorithm. Perhaps some explanations of natural processes are "based upon" an algorithm; but the processes themselves are not, because an algorithm is an artificial construct. Further, "complexity" has nothing to do with evolution. Evolution is about changing from initial states; "complexity" is at most a descriptive byproduct.1 In some circumstances, that could be a reduction in complexity; for example, my hazy recollection of the genetic mechanism of sickle-cell anemia indicates that the disease results from omission (or, at least, blocking) of one or more genes.

Then, too, there's the "monolithic explanation" problem: Luskin et al. assume that there must be some kind of Unified Biological Theory (cognate to the Unified Field Theory that neither Einstein nor his successors have found) that explains everything without fail or exception, and for which all parameters are easy to describe. Quiz question: What does the Planck Constant really represent?

If Luskin were being intellectually honest—or even just a moderately good and accurate writer—he would have said something like this:

After all, Darwin's theory of evolution by natural selection can be represented as a set of algorithms which uses a mathematically describable trial and error process to attempt to produce a result distinct from the initial state.

Note that this more-accurate statement allows for the scientific method to operate: "Disproving" a given set of algorithms does not "disprove" the theory, but requires modification of the set of algorithms. Only once one proves that no set of algorithms could describe the theory can one move on to question the underlying theory itself solely upon that basis.2 I don't agree with the "corrected" Luskin restatement, either, but it is at least an accurate-enough statement that one can support (or refute) it without reference to nonlogical authority… "And God said: 'Let there be light.' And there was light."3

The whole "mathematics matters too!" argument reminds me a great deal of the intellectual dishonesty of Johnson et al. Those of us who have even a passing understanding of the law of evidence quickly spotted what he was up to: He was establishing a burden of proof in favor of one "side" while defining the other "side" in different ways depending upon his preferred outcome. (The reality is that he inverted the burdens of proof from what logic requires: "innocent until proven guilty" can logically apply only to evolution, not to theism.) Johnson's works are show trials that make the King of Hearts look a positively dispassionate, contemplative, and reasonable jurist. Welcome to that dubious club, Mr Luskin.

  1. A reminder for those who don't get thermodynamics (which clearly includes the DI folks): The Second and Third Laws apply only to closed systems that do not have external inputs of energy. If one actually does the math and examines the boundary conditions for discontinuities and inversions, it's pretty obvious. One certainly can have, and even create, local differences in energy; for example, that's how your air conditioner works.
  2. This is a distressingly common problem when translating mathematically described concepts to reality: Disproof of a particular mathematical operation in a given instance is taken as disproof of the entire framework of reality. I observed far too many applications of that lemma while in government service to accept its validity.
  3. Genesis 1:3. Curiously, the passage never mentions who paid the power bill. Or how the concommitant heat dissipated in time for the next step. On the other hand, it is poetry (of a sort), so strict conformance to observable and verifiable reality may be an unrealistic expectation.