I owe Martin Nowak an apology

So, if you’re an Evolutionary Biologist, you’re already familiar with the dust-up prompted by a Nature paper published in 2010 by Martin Nowak, Corina Tarnita, and E. O. Wilson.  If not, I wrote about the paper, and the response from the community, here and here.

Briefly, the article attacked one class of approaches to modeling the evolution of traits affecting social interaction: models based on kin selection and inclusive fitness. The authors made strong claims about the effectiveness of such models, claiming that they were useless or even wrong for thinking about eusociality (e.g., in species of bees and ants). The paper prompted a number of written responses, in blogs and in letters to Nature, one co-authored by 137 prominent biologists, refuting many of the claims of the paper.

The paper comes with a weighty appendix, which contains a lot of calculations. Those calculations are not problematic. Rather, it is the main text (the only part most people will read) that triggered the vocal response. The main text made a bunch of unsupported (and wrong) claims, knocking down a straw-man caricature of kin-selection models. It was this straw-man caricature that people found so offensive, along with the failure to cite a huge body of literature (which would have undermined that straw man).

The disconnect between the careful, meticulous appendix and the swaggering, irresponsible main text led most readers to assume that we were looking at a frankenpaper, the imperfectly integrated product of multiple authors. In this sort of circumstance, the impulse is to partition blame among the authors.

My sense was that most people held Tarnita, a postdoc with Nowak at the time, blameless, a talented junior scientist in the wrong place at the wrong time.

The blame, in most people’s eyes, fell primarily on Nowak, for a complex set of reasons that I tried to untangle here.  In particular, Nowak has a reputation for not being generous in attribution of credit to other scientists.

Wilson was not blamed. He is, after all, a living legend among evolutionary biologists. If anything, the discussion about Wilson was along the lines of, “Why is Wilson keeping such bad company?” Some people even speculated that he was perhaps being taken advantage of, that he had been roped into putting his name on the paper.

It now appears that I, along with all the other rumor-mongering evolutionary biologists, owe Nowak an apology.

Over the past year, Wilson has been on the warpath, giving various interviews in which he reiterates the major arguments presented in the paper. The most recent just appeared here in the Atlantic. This article, I think, makes it clear that Wilson was the ideological driving force behind all of the misrepresentation in the original Nature article. It also seems to indicate that the disingenuous argument will be expanded to book length in Wilson’s forthcoming The Social Conquest of Earth.

The richest part of the Atlantic article comes in Wilson’s trashing of Stephen Jay Gould. Trashing Gould is, of course, a popular pastime among evolutionary biologists.

“I believe Gould was a charlatan,” [Wilson] told me. “I believe that he was … seeking reputation and credibility as a scientist and writer, and he did it consistently by distorting what other scientists were saying and devising arguments based upon that distortion.” 

This is a valid enough criticism of Gould. It is also a dead-on description of what was wrong with the Nowak et al. paper. I suspect that the irony is lost on Wilson.

Nowak, M., Tarnita, C., & Wilson, E. (2010). The evolution of eusociality Nature, 466 (7310), 1057-1062 DOI: 10.1038/nature09205

10 thoughts on “I owe Martin Nowak an apology”

  1. The figure I was most curious about in this whole kerfuffle was Tarnita, who I know next to nothing about. The quote from her in that Atlantic article was interesting, and certainly *suggests* that she’s … on board? with Nowak and Wilson. She also comes across as thinking that she’s right because she’s a mathematician (not the first time I’ve run across *that*), and I hope that that’s just an out of context misinterpretation on my part.

  2. @behavecology, I agree. My sense it that people don’t blame her because, well, being on board with your senior collaborators is sort of your job as a postdoc.

    Also, my guess is that she is responsible for all the math in the appendix, which is just fine. It is the contextualization of that math that is the problem.

    As for the “I’m right because I’m a mathematician” thing, that is certainly a common phenotype, and it is an argument that people often use to shield themselves from criticism. I read a number of internet comments by people who are defending Nowak et al. by dismissing critics of the paper as innumerate biologists who resent having mathematical rigor brought into the field. The thing that is so infuriating about that is that most of the people who wrote in criticizing the paper are, themselves, mathematical biologists.

  3. @Neuroskeptic,

    Good point. Although my instinct is to lay that blame at the doorstep of D. S. Wilson. E. O. Wilson wrote a few papers with him a few years back, and he has long held an ideological grudge against kin-selection / inclusive-fitness models.

  4. OK – I was just thinking, humans-evolved-to-be-nice-through-group-selection could be made to fit rather well with a Catholic / Templeton-ian viewpoint.

    Although in the Atlantic interview it says that Wilson is critical of “organized religion” so maybe not… just a thought.

  5. Yes! Almost every big mathematical brain *I* can think off of the top of my head was a signatory to that Nature letter, and most of them are people that I would happily throw into a mathematical cage match with Nowak / Tarnita any day.

    On a completely unrelated note, I’m off to arrange mathematical cage matches. I’m going to make millions on the pay-per-view.

  6. Re: Tarnita. I was at the MMEE 2011 conference in Groningen where she was an invited speaker, talking about the paper. I am not qualified to address the scientific issues, but the talk itself gave a poor impression. She bashed inclusive fitness theory rather than promoted her alternative, and the whole thing was more of a lecture than a talk, if you know what I mean. She’s certainly on board with Nowak&Wilson.

  7. Maybe, just maybe, you owe Martin Nowak an apology because he (and they) were basically correct. It is certainly the case that kin selection and multilevel selection are interconvertable, however, there is much reason to be suspicious of kin selection. Basically it hides population structure with a mathematical conceit and then pretends that it doesn’t exist. The net result is that it takes great discipline to use kin selection constructively. A great deal of pain over the evolution of complex traits could have been avoided had we used a multilevel selection perspective rather than a kin selection perspective.

  8. Anon, “inclusive fitness” and “multilevel selection” seem to be the same theory. Kin selection seems to be a bit different. As W. D. Hamilton once said: “kinship should be considered just one way of getting positive regression on genotype in the recipient”.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.