Social Evolution Forum
FIND sef:
Mopping up Final Opposition to Group Selection

In 2009, the distinguished evolutionary biologist Jerry Coyne published a book for a general audience titled Why Evolution is True. The in-your-face title drew criticism for hubris unbecoming to a scientist. Shouldn’t scientists cultivate a more humble attitude toward their theories du jour? Doesn’t history provide many examples of the pride of scientists, broken in the dust again?

why-evolution-coyneAll true (Oops! Perhaps I shouldn’t even use the t-word this way!) but Coyne was making a valid point with his assertive title. Science can’t move anywhere, forward or back, without accepting some hypotheses and rejecting others. If some people obstinately refuse to accept a hypothesis, no matter how much evidence piles up in its favor, then they have brought the scientific process to a halt. It is important to move on when enough evidence has accumulated, even knowing that the future might reveal new weaknesses and unforeseen alternatives. This is why creationism is not and should not be taught in schools as a legitimate scientific alternative to evolution, even though its exclusion seems to offend sensibilities about letting everyone have their opinion and say.

Does-Altruism-Exist-e1421085230885Now I am drawing fire for hubris in my new book, Does Altruism Exist? In it, I claim to offer a “post-resolution” account of group selection, which explains the evolution of altruistic traits at face value. By this I mean that there is no need to re-define the traits as selfish after showing how they evolve. The chapter of my book titled “How Altruism Evolves” could have been titled “Why Group Selection is True.” I think that I can justify my immodesty in the same way that Coyne justifies his. How ironic that Coyne is among those accusing me of hubris.

Sign up for our newsletters

I wish to receive updates from:

My strong claim about group selection cannot be understood or evaluated without also understanding a concept called equivalence. Imagine that Joan is talking to Bob and makes a statement based on certain definitions. Bob gets upset and blurts “Well that’s not true according to my definitions!” That is a stupid move on Bob’s part, which puts an end to productive communication. Of course any given statement depends on the definitions underlying the statement. Productive communication requires Bob to evaluate Joan’s statement according to her definitions and then translate her statement into an equivalent statement according to his definitions, like translating a statement from one language to another.

I wish I could say otherwise, but Coyne is acting like Bob in his recent blog post on my book. Steve Pinker also acts like Bob in his essay titled “The False Allure of Group Selection” published on with commentaries (be sure to read the commentaries), which is the only source that Coyne provides to his readers to learn more. Both are saying in effect: “Wilson is wrong! Altruism doesn’t evolve by group selection according to my definitions!” Why don’t they try evaluating my claims according to my own definitional framework (multilevel selection theory)?

What is the optimal number of definitional frameworks for the study of a given subject? Part of the answer to this question is “more than one” because different frameworks can offer different perspectives on a given topic, like viewing a mountain from different angles. Another part of the answer is “not too many” because of the burden of translating among frameworks. Also, a definitional framework is like a mutation; for every one that works well, hundreds work poorly. In productive scientific and scholarly inquiry, there is a manageable number of definitional frameworks that pull their weight in some sense, and the community of scientists/scholars is capable of recognizing and operating within each framework.

The task of the scientist/scholar is made even more complex when we take history into account. Definitional frameworks aren’t static. They change as inquiry proceeds. The definition of the coefficient of relatedness (r) in inclusive fitness theory (also called kin selection theory) provides a good example. Originally interpreted as a measure of genealogical relatedness, it was later generalized to become a measure of genetic correlation for any reason. In one of the most recent reviews of inclusive fitness theory, Birch and Okasha1 provide three definitions of r that are useful for different purposes, illustrating the principle of “more than one but not too many.” Multiple definitions and changes in definitions over time are not arbitrary. They are driven by scientific/scholarly inquiry, as reported in peer-reviewed books and journals. Scientists/scholars have a responsibility to keep all of these distinctions and their rationales alive in collective memory.

When someone like Coyne or Pinker declares a theoretical framework such as multilevel selection confusing, good for nothing, and changing arbitrarily over time, there is a good chance that they are acting like Bob. While I’m calling out luminaries, let me add Martin Nowak and E.O. Wilson on the subject of inclusive fitness theory. When Nowak2 wrote, “Casting a problem in terms of inclusive fitness is like having to undergo elaborate and time-consuming initiations to join a club, only to end up with nothing in the way of privileges,” he sounded like Bob. It was appropriate for Nowak and Wilson to be held to a higher standard and criticized for their dismissal of inclusive fitness theory as a useful theoretical framework in the infamous letter published in Nature magazine,3 signed by nearly 150 scientists. Now let’s apply the same standard to Coyne and Pinker for their dismissal of multilevel selection theory.


Water striders, a species that includes males with either selfish or altruistic mating strategies. Experimental groups that include only altruistic males produce nearly double the offspring of groups with only selfish males4.

When you consult the work of scientists who think hard enough about multilevel selection to publish on the subject in peer-reviewed journals, you find widespread acceptance of the fact that traits evolve by virtue of benefiting groups, despite being selectively disadvantageous within the groups; this is the central claim of multilevel selection theory (for some empirical examples, see references 4-6). Even scientists who prefer to use other frameworks, such as David Queller7 and Stuart West,8 acknowledge this much. To summarize: The earth revolves around the sun. Continents drift. Evolution is true. And altruism evolves by group selection as “altruism” and “group selection” are defined within multilevel selection theory.

Does Altruism Exist? covers this ground in more detail for the general reader. It has been praised by reviewers for its clear account of multilevel selection and equivalence (go here and here). Even H. Allen Orr, who also accuses me of hubris in his New York Review of Books review, commends the chapters on multilevel selection and equivalence.

To learn more online, please consult my essay titled “Richard Dawkins, E.O. Wilson, and the Consensus of the Many,” where I call out both Dawkins and Wilson for their Bob-like behavior. Please read the comments in addition to the article to see how others are working with the ideas. My more recent essay titled “Challenge to Kin Selectionists—Explain This!” explores the interesting possibility that two theoretical frameworks can be equivalent in some but not all respects. My challenge to kin selectionists is friendly and I will be happy if they can meet it. Bob-like rants against kin selection or group selection are not productive. It is time to move on.


1. Birch, J., & Okasha, S. (2014). Kin Selection and Its Critics. BioScience, 65(1), 22–32.

2. Nowak, M., & Highfield, R. (2011). SuperCooperators: Altruism, Evolution, and Why We Need Each Other to Succeed. New York: Free Press.

3. Abbott, P., Abe, J., Alcock, J., & al., et. (2010). Inclusive fitness theory and eusociality. Nature, 471, E1–E4.

4. Eldakar, O. T., Wilson, D. S., Dlugos, M. J., & Pepper, J. W. (2010). The role of multilevel selection in the evolution of sexual conflict in the water strider aquarius remigis. Evolution; International Journal of Organic Evolution, 64(11), 3183–9.

5. Kerr, B., Neuhauser, C., Bohannan, B. J. M., & Dean, A. M. (2006). Local migration promotes competitive restraint in a host-pathogen “tragedy of the commons.” Nature, 442, 75–78.

6. Pruitt, J. N., & Goodnight, C. J. (2014). Site-specific group selection drives locally adapted group compositions. Nature, 514(7522), 359–362.

7. Queller, D. C. (1991). Group selection and kin selection. Trends Ecol. Ecol., 6(2), 64.

8. West, S. A., Griffin, A. S., & Gardner, A. (2007). Social semantics : altruism, cooperation , mutualism , strong reciprocity and group selection, 20, 415–432.




Join the discussion


  1. Daniel Hoyer says:

    I am one of the ‘general audience’ members (having no training in the evolutionary sciences) who read “Does Alturism Exist?” I found it extremely informative, persuasive, enjoyable, and actually very measured in its claims. I am thus surprised that it is receiving such harsh criticism for scientific immodesty. (Indeed, one of the few criticisms I had of the work was that Prof. Wilson spent a bit too much space contextualizing and almost apologizing for some of the propositions put forward, though I guess the reaction to it shows that such caution was justified!).

    I read through Pinker’s essay in Edge linked to this post, and find the ‘pro group selection’ comments much more convincing than Pinker’s arguments or the other anti-group selection comments, though again from my naive perspective. I must say, though, as an historian, I did find Pinker’s very dismissive and ill-informed stance against historical arguments in favour of group-selection quite frustrating (Pinker casually claims that “States and empires are the epitome of large-scale coordinated behavior and are often touted as examples of naturally selected groups. Yet the first complex states depended not on spontaneous cooperation but on brutal coercion” – he cites slavery and harsh penal laws as evidence of this ‘coercion,’ which is a gross distortion of the historical reality of early states. I can think of no ancient state that fielded armies made up largely of enslaved soldiers, nor did soldiers typically need to be ‘brutally coerced’; rather, large armies of conscript farmer-soldiers who risked their lives in a semi-voluntary (but certainly generally accepted) manner for the good of the group as well as themselves (i.e. the safety of their property and families) were the norm. The central point seems to be that societies with fewer draft-dodgers, for example Rome, out-competed less cooperative enemies such as loose confederations of Etruscan or Samnite states. Peter Turchin’s work in particular shows the historical utility of the multi-level selection approach [e.g. link to

    The entire controversy to me is reminiscent of a long-standing modernism/primitivism debate among ancient historians, which essentially revolves around whether one believes that ancient people had the same fundamental economic inclinations and activities as we do today (the modernists), even if on a much smaller scale than modern economic activity, or if they had qualitatively different attitudes and expectations about economic life (the primitivists). The debate, which proved entirely useless for improving our understanding of any facet of ancient society, has begun to subside in recent years thanks to an increasing understanding among specialists that both sides essentially agreed over almost all details, and the differences were mainly in how these details were defined — much like the way that Prof. Wilson describes the situation among evolutionary theorists. The field of ancient economic history is now moving forward, with most scholars devoting their energy not to picking a camp and pointing out flaws in the arguments of the other side, but focussing on the details, exploring the evidence, and trying to construct theories that can explain all the material without qualifying claims about ‘how like us’ were the historical societies under study.

    It is a shame when so many great minds spend their energies squabbling over such issues and speaking past each other, when it is likely much more fruitful and beneficial for everyone to focus instead on where the different theories overlap; as Wilson puts it, exploring “the interesting possibility that two theoretical frameworks can be equivalent in some but not all respects”.

    • David Sloan Wilson says:

      Thanks for this helpful post and I’m glad that you are already familiar with Peter Turchin’s work. Increasingly, people who are newly encountering the kin vs. group selection controversy tell me that is sounds to them like some arcane family feud.

  2. […] selection” argument is right, and has just published a note on his website called “Mopping up final opposition to group selection.”  As the battle winds down, D. S. Wilson has declared victory for the wrong […]

  3. […] and scientifically unwarranted note of triumphalism in posts on his website such as “Mopping up final opposition to group selection,” and “The tide of opinion on group selection has turned.” Barash rejects the cry […]