Two Sides of the Same CoynePublished 10 January, 2011
In the new paper, the authors show that fruit fly wings have really pretty patterns when viewed against a black – but not a white – background. As Coyne says, the work is described well over at Discover; it’s worth checking out, if for no other reason than the patters of color really are striking.
My interest is in Coyne’s remarks. Discussing the finding, Coyne asks, “What are these patterns for?” He goes on to discuss evidence regarding the function of these patterns.
But that simple question is interesting and caught my attention. In it lies the implicit assumption that the patterns are for something. Coyne seems to be assuming that these traits are adaptations. (He ventures some guesses, in part based on the fact that the trait is sexually dimorphic, having to do with mating and species recognition.)
Now, the thing is that this reminded me of some stark criticisms leveled by a well known biologist, and author of Why Evolution is True, Jerry Coyne. In his book, he talked about how traits should emphatically not just be assumed to “reflect genetically based adaptations.” Now, true, in that quote there he was talking about behaviors, not physical traits, but the argument he makes does not, as far as I can tell, as a logical matter apply only to behavior. Indeed, he cautions about rushing to conclusions about a physical trait — interestingly enough, one that is, just like the wing patterns, sexual dimorphic – writing that, for one particular species – readers may guess which one – you can’t infer that because males are larger, this is necessarily due to male-male competition; there could be another reason for the difference, given what we know about the natural history of the species in question.
That said, it’s clear that he thinks that one can, in principle, identify adaptation in the context of behavior, more generally. He discusses lots of behavioral adaptations in the book, including, for instance, the “adaptive behavior” (p. 112) of honeybee defense against hornets. (As an aside, he refers to the hornet as “designed for mass slaughter,” p.112, his italics.) (By the way, this is really, really cool. When a hornet comes into the nest, the bees swarm the intruder, vibrate, and cook the hornet to death. See Coyne’s description for a bit more.)
In any case, Coyne (author of the book) says that the problem is that it’s just too easy to make up “just so stories” like the ones that Coyne (blogger) makes up in his post. Discussing some functional hypotheses, Coyne (author) reiterates that these ideas “come down to untested—and probably untestable-speculations,” adding that “it’s almost impossible to reconstruct how these features evolved (or even if they are evolved genetic traits) and whether they are direct adaptations or…merely by-products…” (p. 250 his emphasis). In short, he cautions: “We should be deeply suspicious of speculations that come unaccompanied by hard evidence.”
So my worry here is that Coyne (blogger) in his very nice blog post has made the sort of error that Coyne (author) would criticize, assuming the traits are functional and speculating about what the function might be. I hope that the one doesn’t become aware of the other. It might set off quite a fight.
It’s probably obvious, but in case it is not, I agree with the blogger not the author. Yes, one ought not simply assume any observed trait is an adaptation, but well specified guesses regarding adaptive function can be tested. Coyne (author) actually endorses this inference, saying: “Every time we see an obvious adaptation, like the camel’s hump or the lion’s fangs, we clearly see evidence for selection.” (p. 135). So he is saying that one can infer the function (rending) from form (fangs), given the right sort of evidence. And in his blog post, he discusses some relevant evidence in terms of the fly wing patterns. The question is, why is it that when discussing humans, the logic of identifying adaptations changes, such that the problem of the ease of generating ideas of function means that developing a hypothesis about function in non-human animal behavior is just fine, but a hypothesis about function in humans is “not science”? (p. 228).
So, in all seriousness, I want to say that I only just discovered this blog, and so far I have enjoyed it. I might also note that with the exception of a few pages, I actually liked the book, too, though I haven’t read all of it cover to cover.
But I do believe that it’s worth thinking about why Coyne thinks honeybee behavior can be characterized using the framework of adaptationism but worries about applying these principles to human behavior, so much so that hypotheses of function in humans cease to be science. Why is the claim that humans have, say, cheater detection devices different from saying that bees have hornet-defense systems? Why is he an enthusiastic adaptationist about fly wing pattern coloration while condemning hypotheses about the function of human traits as merely a “parlor game?”
In short, shouldn’t we do “species-neutral biology,” to Coyne a phrase?
(Note that Liddle & Shackleford address some of the ideas here in their review of the book, as well as Coyne’s (incorrect) reading of the field as endorsing genetic determinism.)