Everything You Know About Natural Selection Is Wrong
It’s called “a fresh theoretical framework” but it undermines the popular conception of natural selection. It’s called a “dense and deep work on the foundations of evolutionary biology” but it criticizes as simplistic and false the ideas of Richard Dawkins, one of the most outspoken proponents of natural selection as “the greatest show on earth.” It produces a new scheme for how natural selection works, but raises more questions than it answers. What is it? It’s a new book by Harvard philosopher Peter Godfrey-Smith, Darwinian Populations and Natural Selection (Oxford, 2009), reviewed mostly positively by Jay Odenbaugh in Science.1
Odenbaugh is in the philosophy department of Lewis and Clark College, Oregon. Get ready to jettison your “classical” concepts of fitness, selection and reproductive success. Unload your simplistic ideas of gene selection, individual selection and group selection. Prepare to see Richard Dawkins demoted from his status as a leading spokesman for modern Darwinism. In his first paragraph, Odenbaugh clears the deck to get ready for the “fresh” ideas of Godfrey-Smith:
Peter Godfrey-Smith’s Darwinian Populations and Natural Selection is a dense and deep work on the foundations of evolutionary biology. Evolutionary biologists tell us that evolution by natural selection occurs when a few ingredients are present—specifically, when there is variation with respect to a trait, those variants differ in the numbers of offspring produced, and this variation is heritable to some degree. Unfortunately, as Godfrey-Smith argues, this recipe is far too simple, and even more complicated versions such as the replicator approach offered by Richard Dawkins suffer serious flaws. This “classical recipe,” for example, ignores the fact that for some organisms numbers of offspring don’t necessarily determine reproductive success (“fitness”) whereas rates of population growth, age structure, or variation in expected numbers of offspring do. Likewise, natural selection and patterns of heredity can “cancel” each other out, leaving no evolutionary change. The concept of Dawkins’s replicators—those entities that interact with like entities and of which copies are made—presupposes that there can be no reproduction without replication, which is false when we have continuously varying traits evolving by natural selection. Thus, our standard models for understanding what evolution by natural selection is are just too simple.
Wow. If you have survived that devastating paragraph, you realize that Godfrey-Smith had better replace all the simplistic notions with some profound and testable alternatives quickly before the creationists latch onto what Odenbaugh just admitted. Unfortunately, Godfrey-Smith replaces it with a scheme that is more ethereal than empirical. He envisions three parameters, H (reliability of inheritance), C (relation of traits to fitness), and S (dependence of reproductive differences on intrinsic traits). Then he graphs them in “population space.” Odenbaugh tries to give this scheme respect: “Godfrey-Smith then uses this spatial framework (along with others concerning reproduction) to understand controversies concerning the nature of random genetic drift, levels of selection, major transitions in evolution (such as the appearance of multicellular organisms), and cultural evolution.” Then he starts the clock: “Let’s consider what light Godfrey-Smith’s framework shines on some of these topics.” OK; we have just been promised light on the Cambrian explosion, the units of selection (genes, individuals, or groups), and whether natural selection produced the university itself (cultural evolution).
Odenbaugh delves into some examples to illustrate the new framework (e.g., a twin struck by lightning can’t reproduce, whether or not the other twin is less fit). He explains Godfrey-Smith’s view that neutral drift is not a “force” or label for ignorance; “rather it concerns where one is in the space of Darwinian populations” (got that?). Regarding units of selection, we hear more that Godfrey-Smith rejects group selection than offers a plausible replacement: “For example, in cases where selection occurs in neighborhoods, there are no causally cohesive groups for selection to operate on.” Well, then, what does natural selection operate on? If Godfrey-Smith has an answer, Odenbaugh did not reveal it.
Let’s see if the book has an answer for the controversy of where great transformations and innovations come from (the classic case being the Cambrian explosion). “With regard to evolutionary transitions, he notes that often the formation of new biological individuals involves marginal Darwinian populations moving to paradigmatic ones and the parts of such populations (that is, the lower-level entities) moving from paradigmatic ones to marginal ones—a process he terms ‘de-Darwinizing.’” The casual reader might have to re-read that sentence a few times. Did he just say that members of a population move, by some unexplained force, into a new paradigm? Like from a sponge into a trilobite or something? And that others move out of the paradigm into the margins? It is difficult to see how any of this wording explains the origin of complex biological information such as eyes, wings, and new body plans. And how appropriate is it to introduce a new concept like “de-Darwinizing” right now, right at the pending 150th anniversary of Darwin’s book on natural selection, what E. O. Wilson calls “the greatest idea anyone ever had”?
The next paragraph involves debating distinctions about reproducers – whether they can be described as collective, simple, or “scaffolded” (i.e., parts of reproducing entities that get reproduced, such as a gene in a mammal giving birth). Here’s where Dawkins gets another sucker punch:
These distinctions are skillfully employed. For example, contrary to Richard Dawkins, many instances of genic selection are instances of scaffolded reproduction of genes by cells, and evolutionary models are ultimately representing selection of organisms via their genetic properties. Often (though not always), when we treat genes as evolutionary units we imbue evolutionary biology with an “agential” framework involving agents, goals, strategies, and purposes that can corrupt the foundations of evolutionary biology.
So we certainly must have none of that. No teleology allowed. Dawkins’s “selfish genes” have just been criticized as imbued with the concept of agency or strategy or purpose. Dawkins is corrupting the foundations of evolutionary biology, Odenbaugh and Godfrey-Smith said. One can only imagine his reaction at such a charge from fellow evolutionists.
The last paragraph of the book review arrives. The Cat in the Hat had better show up in the nick of time to clean up this mess. No; now, philosopher Odenbaugh turns on philosopher Godfrey-Smith and accuses him of hypocrisy and obfuscation:
Darwinian Populations and Natural Selection raises difficult questions as well. Godfrey-Smith and others have argued that there is a role in evolutionary biology for “functional” notions. For example, they hold that it makes sense to claim that the heart in humans has the function of circulating blood. However, given the author’s criticism of the “agential” framework and the teleology behind it, is this new work compatible with the old? In addition, although spatial frameworks or state spaces can be exceedingly useful for understanding evolutionary processes, one can ask if they also conceal much of importance. Their use is critically dependent on which dimensions are included (and which omitted) and on whether one can “score” those dimensions in plausible ways. Sometimes one wonders whether too much is being omitted and worries that variables like S cannot be scored in any object sense.
Not to leave any bad feelings, he finds something to praise: “Nevertheless, Godfrey-Smith’s book fruitfully forces us to think in new ways about evolution and natural selection.”
1. Jay Odenbaugh,“Evolution: A Fresh Theoretical Framework,” Science, 16 October 2009: Vol. 326. no. 5951, pp. 368 – 369, DOI: 10.1126/science.1176940.
Folks, you have just watched the undoing of Darwinism and natural selection. Pray tell, what remains after this gentle demolition derby? Everything you have been taught about natural selection is wrong. Is there any concept left on which you can hang your hat and say, “this is natural selection in action”? No; now you have to worry whether the reproducer was simple, collective, or scaffolded. Now you have to worry whether selection acts on the gene, the individual, the group, or the population. Now you need to draw meaningless graphs of arbitrary parameters that might omit key concepts, without knowing how to score them objectively. You need to be able to talk out both sides of your mouth: demonizing teleology on one side, using “functional notions” on the other. All the while, you need to keep the Great Cover-Up covered up. You need to hide the elephant in the room, the “emergence” of specified complexity (such as entirely new body plans in the Cambrian explosion) in rhetorical blankets like “marginal Darwinian populations moving to paradigmatic ones.”
This brief book review has all but destroyed the basis for the big Darwin party next month: the 150th anniversary of the Origin. Oh, the party will go on. Speeches will extol The Great Man and his Greatest Show on Earth (watch Colbert tweak Dawkins over that line on Uncommon Descent). Bells will ring and partygoers will get lubricated with copious quantities of free Dar-wine. The pagan festival (see 10/10/2009) will be loud and long and full of hoopla. Party goers will be like drunken celebrants in the gondola of a hot-air balloon, unaware it has lost its flame and is on its way down. They will revel in their luxury Pullman cars, oblivious to the fact the engine has died and the train is rolling down the wrong track into a desert in the dark of night. Let the fireworks play, leaving burnt pieces of debris; let the brambles crackle in the fire, leaving ashes. In the morning will come time to face reality – with a major hangover and a lot to regret.