A Field Guide to Design Detection

Via John Wilkins, a law student’s dissection of his professor’s insistence that the universe displays evidence of design. It’s an excellent analysis of the emptiness of the intelligent design movement’s argument for detecting design in the world. An extract:

I said that in order to infer intelligence from something, you would need an analytical framework. For example, “These particular factors, present in a given phenomenon, are indicative of intelligence for these reasons. Etc. Those factors are present in this phenomenon, therefore we can conclude that this phenomenon is the result of intelligence.” It seems like a simple framework; no more than instructions on how to recognize something, a sort of “Field Guide to Discerning Intelligence in the World.”

Some people, whose intellectual honesty is as questionable as my professor’s, have actually tried to posit “particular factors” that should be indicative of intelligence. Popular methods include Michael Behe’s “irreducible complexity” and William Dembski’s “specified complexity.” Neither is satisfactory. Behe’s idea has been shown wrong by experiment and Dembski’s idea assumes that we can know the probability of the occurrence of any phenomena. (“Specified complexity” is supposed to be anything that is both highly complex and highly unlikely. Except how do you know if it is unlikely? What is the probability of trees? Impossible to say.)

Nobody has yet come up with a convincing “Field Guide to Discerning Intelligence in the World,” but that did not stop my professor from insisting that I have no basis for failing to see intelligence in “natural” phenomena. Apparently it did not occur to him that since he (via Cicero, or vice versa) was making the proposition that “Intelligence is evident in natural phenomena,” it was up to him to explain why exactly that proposition should be accepted, not up to me to demonstrate why it is incorrect.

Just so. ID creationists endlessly assert that they have a methodology for detecting design in biological things, but when push comes to shove, they never ever actually apply it. Has anyone ever seen systematically gathered validation or reliability data on any of their design detection methods? I haven’t. So why is the methodology so difficult to apply? Because it rests firmly (and solely) on the claim “I know it when I see it”. And who is doing the “seeing” is the main variable, not the “it”. It’s an entirely subjective notion.