Steven Pinker in Edge.org:
In his defense of social psychology as it is currently practiced, Timothy Wilson repeats the canard that evolutionary explanations of traits are exercises in “storytelling” which can “explain anything.” He boasts, for example, that he can make up a story in which the redness of blood is an adaptation:
“What if in our very early mammalian history, blood was more brown, but there was a mutation that made it more red, and that turned out to have survival value because if an animals were bleeding, those with red blood would be more likely to notice it, and then they'd lick it. Because licking has healing properties, this conveyed a survival advantage, and so red blood was selected for, and blood became red. Am I right? Or is Steve [Pinker] right, that the color of blood is not an adaptation? Who knows.”
What this shows is that Timothy Wilson can think up a ludicrous evolutionary hypothesis. It does not show that all evolutionary hypotheses are similarly ludicrous. In fact we do know who's right about blood. Chemists tell us that the redness of blood is a necessary physical property of oxygenated hemoglobin, necessary for gas exchange in virtually all vertebrates. This immediately implies that any adaptive hypothesis is otiose. Adaptive hypotheses are needed to explain traits that are improbable given the biologically and physically possible variation in organisms. (This is a basic principle of theoretical biology, most clearly articulated by George Williams in 1966.) The redness of blood is not improbable among mammals; its probability is 1. Moreover, molecular phylogeny has traced the history of hemoglobin hundreds of millions of years, and we know that there was never a stage of mammalian evolution in which oxygenated blood was any other color but red.
Even if, in defiance of biology and common sense, one were to take the wound-licking hypothesis seriously, it would be easy to test it empirically.