The favorite argumentation method of Inscrutable Design proponents is the creation and misuse of false dilemmas. A false dilemma reduces a multivalued argument to a binary result: If not evolution, then necessarily and sufficiently Inscrutable Design is an obvious example. At the metaphoric level, this is a "if not black, then white" type of logic. Sometimes binary logic is the correct way to analyze a puzzle. "Have you stopped beating your dog?" can logically have only a "yes" or "no" answer. The usual form of that inquiry, though, is "When did you stop beating your dog?"—which does not have a necessary and sufficient binary answer. Possible answers include "I haven't stopped"; "I never did"; "last Tuesday at 11am"; "sometime last year"; "I don't have a dog"; and so on.
In any event, Professor Leiter has posted an entertaining dissection of Van Dyke violating the Second Law of Thermodynamics—the three laws are often referred to as (1) you can't win; (2) you can't break even, except at absolute zero; (3) you can't get to absolute zero, and therefore can't quit the game—to which I'd like to add a nonbinary response in support. I apologize for the long quotation.
First, Beckwith notes that Laudan, like every other major philosopher of science now alive, thinks that the "demarcation problem" that exercised mid-20th-century philosophy of science—how do we demarcate science from non-science, or genuinely cognitive domains from nonsense—cannot be solved. This now banal piece of philosophical wisdom goes no distance, obviously, towards showing that ID and creationism aren't bad science, with nothing to commend them as research programs—which Laudan clearly believes, as Beckwith correctly notes. Has VanDyke read Beckwith's book?
If so, he might have also noted that Beckwith quotes Laudan [at 25] noting that ID "is inconsistent with methodological naturalism and ontological materialism… [b]ut that fact has no bearing whatsoever on the plausbility of the arguments for ID." Why does Laudan say that? Because methodological naturalism is an a posteriori doctrine, which means if ID generated any empirical results incompatible with it—it has not, of course—then so much the worse for MN. The problem is purely a posteriori: ID has no research program and no empirical support, so it presents no challenge at all to the reliance on naturalistical explanatory mechanisms. Laudan thinks talk of "pseudo-science" is misleading in the absence of a solution to the demarcation problem; Laudan has no reservations about talk about "good" and "bad" science as measured by their results and the evidence on behalf of their claims.
The fallacy here is Beckwith/Van Dyke/ID's assumption that ID's claim to scientific status is a close-enough question that the exact countour of the "boundary" between "science" and "not-science" matters. In a criminal-law context, they argue as if the definition of "justification" for a homicide is so nebulous that we can't even determine that there was a homicide. (Sorry, but he's dead, Jim.) This "false dilemma by restatement" is a relative of the inductive fallacy; but it is not quite the same thing. As Professor Leiter points out, in different words, ID is so far from any consensus definition of "science" that we need not know the exact boundary; the charcoal is so dark that we can clearly perceive it as "not-white" with no legitimate objection.
If there is one thing that should be apparent, it is that use of the results (or their absence) of scientific inquiry in an argument does not make that argument scientific. Otherwise, we would end up with something like "The simple nuclear fission yield of a standard-design weapon with x kilograms of plutonium is f(x); therefore, such a weapon is a good military choice for its destructive force" as a scientific argument. I will leave the fifty-seven logical problems with such an argument as an exercise for the student. I will not leave it as an exercise for ID proponents, because they have made it clear that they are not students—they already know all the answers.