Sunday, October 19, 2014

Razib Khan defends old-fashioned evolution theory

Razib Khan writes at Gene Expression. He's a big fan of personal genetics and genome sequencing and, in the past, has been a defender of the Modern Synthesis version of evolutionary theory. In light of the recent Nature discussion on "Does evolutionary theory need a rethink?" (Laland et al. 2014), Razib thought he would re-state his position [Evolution Ever Evolves].

I laid out my position in: Rethinking evolutionary theory. I don't think any of the new ideas like epigenetics, plasticity, facilitated variation etc. are about to change evolutionary theory significantly. However, I do think that the standard version of the 1940s Modern Synthesis was far too rigid and that a modern emphasis on population genetics (including Neutral Theory and more emphasis on random genetic drift) have significantly changed evolutionary theory—something close to a "revolution." The problem is that many scientists, and even many evolutionary biologists, haven't really integrated this change into their way of thinking. This resistance was very well described in a paper by Stephen J. Gould and Richard Lewontin over 45 year ago (Gould and Lewontin, 1978) [What Does San Marco Basilica Have to do with Evolution?]

I think there's already been a "revolution" but most people didn't notice and are still stuck in the 1940s adhering to an old-fashioned version of evolutionary theory that emphasizes adaptation.

Razib Khan doesn't like Gould and doesn't like new-fangled ideas like "neutralism" and "random genetic drift". Let's see what he thinks of the latest kerfluffle.
It seems that rather regularly there is a debate within evolutionary biology, or at least in public about evolutionary biology, where something new and bright and shiny is going to revolutionize the field. In general this does not pan out. I would argue there hasn’t been a true revolution in evolutionary biology since Mendelian genetics and classical Darwinism were fused in the 1920s and 1930s during the period when population genetics as a field was developed, and the famous "synthesis" developed out of the interaction of the geneticists with other domains of evolutionary relevance. This does not mean that there have not been pretenders to the throne. Richard Goldschmidt put forward his "hopeful monsters," neutralism reared its head in the 1970s, and evo-devo was all the rage in the 2000s. Developments that bore scientific fruit, such as neutralism, were integrated seamlessly into evolutionary biology, while those that did not, such as Goldschmidt’s saltationism fell by the wayside. This is how normal science works.
The main point here is whether Neutral Theory and an increased emphasis on random genetic drift "were integrated seamlessly" into the Modern Synthesis view that was popular in the 1960s. Is it true that the way modern population geneticists look at evolution is just a little bit different from the way evolutionary biologists thought about evolution in the 1920s, 1930s, and 1940s? I don't think so. I think there's been a significant shift—so much so that we can no longer refer to the "Modern Synthesis" as the most modern version of evolutionary theory.

Unlike Razib Khan, I am not convinced that most evolutionary biologists have made the shift. At my university, for example, the students must take a first-year course on evolution taught by members of the Dept. of Evolution & Ecology. I see these students in subsequent years and they don't understand the basics of population genetics. Nor do they appreciate the role of neutral alleles and random genetic drift. They are being taught the evolutionary theory of the Modern Synthesis (circa 1960).

Also the debates we are having over junk DNA suggests strongly that most scientists are not familiar with modern population genetics and Neutral Theory.
But every now and then you have a self-declared tribune of the plebs declaring that the revolution is nigh. For decades the late Stephen Jay Gould played this role to the hilt, decrying "ultra-Darwinism," and frankly misrepresenting the state of evolutionary theory to the masses from his perch as a great popularizer. More recently you have had more muted and conventional revisionists, such as Sean Carroll, who promote a variant of evo-devo that acclimates rather well to the climes of conventional evolutionary biology.
I do not believe that Gould misrepresented evolutionary theory to the masses. I believe that Richard Dawkins misrepresented evolutionary theory to the masses.

Like Razib, I'm not a big fan of evo-devo and I don't think it contributes much to fundamental evolutionary theory.
Nature now has a piece out which seems to herald the launching of another salvo in this forever war, Does evolutionary theory need a rethink? It’s written in the form of opposing dialogues. I’m very much in the camp of those believe that there’s no reason to overturn old terms and expectations. Evolutionary biology is advancing slowly but surely into new territory. There’s no problem to solve. The one major issue where I might have to make a stand is that it focusing on genetics is critical to understanding evolution, and dethroning inheritance from the center of the story would eviscerate the major thread driving the plot. The fact that evolutionary biologists have the conceptual and concrete gene as a discrete unit of information and inheritance which they can inspect is the critical fact which distinguishes them from fields which employ similar formalisms but have never made comparable advances (such as economics).
I agree with Razib Khan that genetics (population genetics) is the key to understanding evolutionary theory at the population level. I think we disagree on exactly what version of population genetics we support and on the importance of adaptation.

Gould, S.J. and Lewontin, R.C. (1979) The Spandrels of San Marco and the Panglossian Paradigm: A Critique of the Adaptationist Programme. Proceedings of the Royal Society of London. Series B, Biological Sciences, Vol. 205, No. 1161, The Evolution of Adaptation by Natural Selection (Sep. 21, 1979), pp. 581-598. [AAAS reprint] [printable version]

Laland, K., Uller, T., Feldman, M., Sterelny, K., Müller, G. B., Moczek, A., Jablonka, E., Odling-Smee, J., Wray, G. A., Hoekstra, H. E., Futuyma, D. J., Lenski, R. E., Mackay, T. F. C., Schluter, D. and Strassmann, J. E. (2014) Does evolutionary theory need a rethink? Nature 514, 163-165. [PDF]


  1. Is it really true that the proponents of a reorganization in current understanding of evolutionary theory advocate "dethroning inheritance?" As I read him, not even Dobbs in his universally condemned article called for this.

    Is it true that evolutionary biologists have "the conceptual and concrete gene as a discrete unit of information and inheritance which they can inspect?" Coupling information (a deceptively familiar word in my reading of science popularizations) with inheritance assigns genes determinative powers so far as I can tell, while rendering the notion of a gene amazingly indeterminate.

    If students don't appreciate the roles of neutral mutations and random genetic drift, and even many evolutionary biologists have not really integrated them into their thinking, and you can even say most people are still stuck in the early Sixties' version of the Modern Synthesis, how can we say that the neutralists/drift "revolution" has occurred?

    My best judgment, based on wide reading, on the situation in popular discussions of evolution is this. There is only lip service acknowledgement of neutral mutation and genetic drift and they are almost never even considered as explanations. In the predominant "selfish" gene/gene selectionist presentation, junk DNA is the product of natural selection for the genes that are more successful in reproducing, even to the detriment of the organism. It's analogous to sexual selection, or possibly even the genetic equivalent. Also, in popular discussions, gene selection is the opposite of drift. The emphatic assertion of gene selection is the implicit dismissal of genetic drift in this context. Since natural selection acts on genes, by extension phenotypic traits are selected, that is, adaptive. And when evolutionary psychology offers explanations based on hypothetical entities like neural modules or propensities, the justification for them is ultimately the inescapable effect of natural selection: Genes compete by differential reproduction.which means they must have some mechanism to effect themselves. Many (maybe most?) EP popularizers like to emphasize that gene selection has also resulted in nice things like morality and vaguely associate conflicting strategies in gene competition with a resulting kind of freedom on the part of people. Human nature is a product of genes but people's natures are their choices. They decline to discuss differences in human populations other than to insist on the biological reality of races, since genes matter. Criticisms of these views is either falsification in pursuit of personal benefit, or left-wing ideological commitment. (Right-wing ideological commitments died with the Nazis.)

  2. Do you think the modern synthesis is analogous to Newtonian mechanics? Well, it does leave one with an incomplete and unsatisfactory (in some contexts) understanding of evolution, but it does provide one with a preponderance evidence that evolution via natural selection through comment descent has occurred and a theoretical framework to understand its mechanisms.

  3. Pardon the interruption, but this leaves obscured the role of natural selection in maintaining species form. Evolution in the form of changes in the gene distribution in a population always goes on, because of neutralism and random genetic drift, yet natural selection usually conserves the morphology. It is undoubted that natural selection plays a role in multiplication of species (in their common descent,) particularly in the appearance of novel adaptations. But I don't think it's at all clear that novel adaptation is the main mechanism of new species.

  4. I think my comment got lost. I have blogged about this here:

    I note that the "all is well" side concedes the legitimacy of every non-traditional kind of causation invoked by the EES (extended evol synthesis) side, while at the same time undercutting the EES position by claiming that none of the ideas are new (just new names for old concepts) and that the ideas are not well substantiated.

    I go on to argue that the debate is misframed and suffers from some common flaws in evolutionary debates. Comments are welcome.