[O]ne needs to relax Darwin’s criterion from this: “If it could be demonstrated that any complex organ existed which could not possibly have been formed by numerous, successive, slight modifications, my theory would absolutely break down.” to something like this:
If a complex organ exists which seems very unlikely to have been produced by numerous, successive, slight modifications, and if no experiments have shown that it or comparable structures can be so produced, then maybe we are barking up the wrong tree. So, LET’S BREAK SOME RULES!
Of course people will differ on the point at which they decide to break rules. But at least with the realistic criterion there could be evidence against the unfalsifiable. At least then people like Doolittle and Miller would run a risk when they cite an experiment that shows the opposite of what they had thought. At least then science would have a way to escape from the rut of unfalsifiability and think new thoughts.
(Michael Behe, “Answering Scientific Criticisms of Intelligent Design,” Science and Evidence for Design in the Universe, Proceedings of the Wethersfield Institute, Vol 9:146-147 (Ignatius Press, 2000))
Therein lies the key to much Darwinism today. Because Darwin must be right, any observation or information that can be gerrymandered to support his successors’ views trumps good evidence that Darwinism was not at work. And then the contrary evidence disappears down the memory hole. When the peacock’s tail did not turn out to support sexual selection, the fact disappeared from view so completely (along with the fact that natural selection does not explain the Monarch-Viceroy mimicry complex) that it is common to see them cited both by ardent Darwin believers as unqualified examples of Darwinism at work.