Social Evolution Forum
FIND sef:
The Tide of Opinion on Group Selection has Turned

Science is not a popularity contest. A fact such as “continents drift” is true regardless of whether it is heretical or universally accepted. Nevertheless, it’s gratifying to know that science is enough of a truth-finding process that facts once branded as heretical can become widely accepted with time.

A survey of evolutionary anthropologists on the subject of group selection1 provides evidence that the tide of opinion has turned on that once heretical subject. William Yaworsky, Mark Horowitz, and Kenneth Kickham surveyed 175 evolutionary anthropologists (including biological/physical anthropologists and behavioral ecologists) from departments with graduate programs. The respondents spanned the range from assistant to full professors, and 28.7% were women. The authors also assessed political orientation; 74.5% of the respondents identified as liberal. Although women and conservatives were in the minority, they were sufficiently represented in the sample to test for differences in attitudes toward group selection.

The results of the survey demonstrate a big generational effect. When asked to assess their own training as graduate students around the best explanation for human sociality, 57.5% reported that their mentors were “strongly” or “leaning” toward kin selection, 34.2% of their mentors were “neutral,” and 8.2% were “strongly” or “leaning” toward group selection. Among the respondents themselves, 55% regarded multilevel selection as superior to kin selection as an explanation of human sociality. 80.7% disagreed with the statement that group selection misidentifies replicators and vehicles. 64.9% disagreed with the statement that group selection has no utility. 78.7% agreed that cultural group selection is important.

Sign up for our newsletters

I wish to receive updates from:

Critics of group selection sometimes accuse proponents of wishful thinking—of wanting to see the world through rose-colored glasses (even though a proper of understanding of between-group selection is anything but rosy!). The results of the survey demonstrate statistically significant effects of gender and political orientation, although in a complicated way. Liberals were more likely to prefer multilevel selection as an explanation but were less likely to regard tribal conflict as a principle selective force. It seems that liberals and conservatives and men and women differed most on matters such as the frequency of homicide, tribalism, and tribal conflict, which in turn influenced their views on group selection vs. kin selection. Liberals and conservatives did not differ in their belief in the possibility of an egalitarian future or genuine altruism.

An important development in the history of thinking on kin and group selection is called equivalence—the possibility that the two theories do not invoke different causal processes and are inter-translatable, thereby deserving to co-exist rather than one replacing the other. This is in contrast to the original formulation of kin selection by both W.D. Hamilton and John Maynard Smith as a theory that does not invoke group selection. Hamilton’s reformulation of his theory in terms of the Price equation, which he published in 1975,2 marked the beginning of awareness about equivalence. It is gratifying that most evolutionary anthropologists have absorbed its meaning. Yaworsky et al. state:

Again and again in our survey respondents took the time to point out that mathematically, kin selection and group selection amount to the same thing. Instead, they are in disagreement about ontology: that is, which selective pressures in the environment matter, and what are their effects upon genes, individuals, genetic relatives, and groups.

In addition to their survey, Yaworsky et al. present a nice synopsis of the group selection controversy and the polemics of the current “debate.” One thing that stands out is that most of the evolutionary scientists known and revered by the general public, including Richard Dawkins, Steven Pinker, and Jerry Coyne as critics of group selection and E.O. Wilson and Martin Nowak as proponents, are out-of-touch with their own scientific discipline—as represented by the survey results. I include Wilson and Nowak for their failure to absorb the concept of equivalence by continuing to treat kin selection vs. group selection in an either/or fashion. The posturing and hyperbole of these major figures (sample: Pinker calls group selection a “scientific dust bunny” that “has no useful role to play in psychology or social science”) does not reflect well upon them, compared to the more reasoned attitude of the average evolutionary anthropologist.

This year marks the 40th anniversary of my first article on group selection3—and I entered the field when most experts had declared the controversy already over! As one illustrious professor counseled a graduate student in the 1980’s, “There are three ideas that you do not invoke in biology—Lamarkism, the phlogiston theory, and group selection.”4 It’s gratifying to see hard evidence that the tide of opinion on group selection has turned, at least among evolutionary anthropologists in departments with graduate programs. I would like to end this post by relating the survey results to some of my own recent writing on the subject.

In my article Richard Dawkins, Edward O. Wilson, and the Consensus of the Many in This View of Life, I pointed out that these two icons are out of touch with their own field of evolutionary biology in their failure to absorb the concept of equivalence. I based my argument on dozens of scientists who publish on the subject and might have a preference for one theory over the other but at least recognize their equivalence. The nearly 150 scientists who co-signed a letter criticizing the article by Nowak et al. in Nature (see my TVOL article for details) were complaining about Nowak et al.’s failure to recognize the legitimacy of inclusive fitness theory, not their use of group selection as their own preferred framework. It’s nice to know from the survey that the average evolutionary anthropologist is in tune with the current scientific literature, the polemics of Wilson and Dawkins notwithstanding.

In Mopping Up Final Opposition to Group Selection, I explained the concept of equivalence with the following story:

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.

It’s nice to know through the survey that the average evolutionary anthropologist doesn’t act like Bob on the topic of group selection. Now if only our most prominent public spokesmen for evolution would just follow suit.

In Challenge to Kin Selectionists: Explain This! I first affirm the concept of equivalence and then raise the interesting possibility that two theoretical frameworks can be equivalent in some but not all respects. I then present three major examples (equilibrium selection, community-level selection, and human cultural evolution) that can be straightforwardly understood in terms of multilevel selection but not kin selection theory—or so I claim. My challenge to kin selectionists is friendly and I will be delighted to be proven wrong.

In my newest book, Does Altruism Exist? Culture, Genes, and the Welfare of Others, I say that the group selection controversy is over. Just as with previous controversies that took decades to resolve, such as the controversy over continental drift, people will look back and wonder what all the fuss was about. This enables me to provide what I call a “post-resolution” account of how altruism evolves—one that anyone can understand. I also devote an entire chapter to the concept of equivalence. Again, my assessment that the group selection controversy is over is based on the published literature. It’s hard to find a recent article on kin selection or group selection in the recent, peer-reviewed literature that doesn’t acknowledge the concept of equivalence. It’s gratifying to know that the message has been received by the average evolutionary anthropologist.

Science is a truth-finding process, but as the philosopher Thomas Kuhn pointed out with his concept of paradigms,5 scientists can take a long time to arrive at the facts of the matter, taking many twists and turns along the way. The acceptance of group selection has all the earmarks of a paradigm shift. Decades were required and some illustrious scientists have chosen to go down with their ships rather than change their minds. The average evolutionary anthropologist in a department with a graduate program has made the transition, but the general public receives such mixed signals on the topic of group selection that it is impossible to know what to think. Hopefully, the survey by Yaworsky et al. will be followed by others and the consensus about group selection among the experts can begin to be shared with the general public.

  1. Yaworsky, W., Horowitz, M., & Kickham, K. (2014). Gender and Politics Among Anthropologists in the Units of Selection Debate. Biological Theory.
  1. Hamilton, W. D. (1975). Innate social aptitudes in man, an approach from evolutionary genetics. In R. Fox (Ed.), Biosocial anthropology (pp. 133–155). London: Malaby Press.
  1. Wilson, D. S. (1975). A theory of group selection. Proceedings of the National Academy of Sciences, 72, 143–146.
  1. Recounted on p. 40 of Sober, E., & Wilson, D. S. (1998). Unto Others: The Evolution and Psychology of Unselfish Behavior. Cambridge, MA: Harvard University Press.
  1. Kuhn, T. S. (1970). The structure of scientific revolutions (2nd ed.). Chicago : University of Chicago Press.

Join the discussion


  1. Brittany Sears says:

    As I’m sure all readers of Does Altruism Exist? want to know, I’m curious how Yaworsky et al. defined “genuine altruism.”

    And a side note on the three things one “does not invoke in biology,” I daresay that epigenetics is pulling Lamarkism back into the fold! Is phlogiston theory next?

    • David Sloan Wilson says:

      Right! Evolution in Four Dimensions by Eva Jablonka and Marion Lamb provides a great discussion of Lamarkism past and present (including Darwin as a Lamarkist). As for phlogiston…maybe we should leave that one in its grave!

  2. Justin E. Lane says:

    Although the statement is that this isn’t a popularity contest, it seems that all of the data reported here outlines that it is a popularity contest. Quite interesting particularly given the fact that the sample comes from non-biologist affiliates (a subset of anthropologists).
    Why not do the same with biologists?

    • David Sloan Wilson says:

      We should do the same for biologists! Let the surveys continue! I don’t regard evolutionary anthropologists as an especially avant garde group, however. One issue concerns the importance of genetic vs. cultural group selection. One meme I have been fighting is that the received wisdom about genetic group selection is correct and that only cultural group selection is important. Bosh!

  3. Charles Goodnight says:

    I agree, group and kin selection are mathematically vanishingly similar, however there are two mathematically trivial points that make them very different. First, kin selection uses dw/dg, the change in fitness as a function of genotype, whereas MLS uses dw/dz, the change in fitness as a function of phenotype. This is not a trivial difference, since kin selection assumes a much clearer and more knowable connection between genotype and fitness then does MLS.

    The second difference solves for dw/dg = 0, the optimum, whereas MLS solves for dw/dz at the current population conditions. In an additive world this would be a trivial point. In a nonadditive world it is anything but trivial. With gene interactions you are essentially changing the rules as the population conditions change. It would be like using a game theoretic approach where the rules randomly changed from time to time.

    It is time we moved beyond saying they are mathematically the same, since their underlying assumptions are very different.

    • David Sloan Wilson says:

      For me, multilevel selection theory is a causal hypothesis based on the following biological assumptions: 1) natural selection is based on relative fitness; 2) traits that maximize relative fitness within groups are often not good for the group; 3) The evolution of traits that are for the good of the group and selectively neutral or disadvantageous within groups requires a process of between-group selection. The substantive biological question is whether between–group is a significant force in nature. The merits of any particular method for modeling MLS are important, as long as they don’t overshadow the substantive biological issues. I agree with Charles that equivalence can be taken too far. Two theoretical frameworks can be equivalent in some respects but not in others, which was the point of my “Challenge to kin selections” article. Typically the overlap between methods is for relatively simple situations such as additive fitness effects.

    • Steve Roth says:

      @Charles Goodnight: “First, kin selection uses dw/dg, the change in fitness as a function of genotype, whereas MLS uses dw/dz, the change in fitness as a function of phenotype.”

      This statement’s clarity and simplicity makes it something of an “aha” for me. That simplicity may of course mask underlying complex realities, but still I’d like to see this thinking explored and explained more at the higher levels of the “narrative chain” where mere mortals like myself can engage in the thinking.

      (Technical notation like “dw,” “dg,” and “dz” could be eschewed at that level.)

  4. Andrew Atkinson says:

    I agree with Justin to some extent. So, “ok” a survey of a select group of scholars from a select few disciplines goes in favour of group selection as being plausible. I think group selection works for certain cultural trait groups – hence why it might be of interest more to anthropologists than biologists in the strict sense. Additionally, let’s remember that the enterprise of group selection was one that sought to originally to explain the problem of altruism by appeal to selection further up the biological hierarchy. We don’t have to begin with that problem to look at the Darwinian arena in which groups operate. We can just have group selection there to think about regardless of the original problem of altruism. It seems to be a matter of different scholars talking at crossed purposes and with differing foci. Preferential bias over the evidence is bound to differ too – as are the phenomena to be explained. And so what if this article appears to relay a popularity contest – that only goes to show that the matter is of interest to many people in some areas of research – and that’s a good thing. Perhaps the differing fields will come together one day. What David, and others have done, is raise interest in the idea of group selection – it just seems that, with respect to their specific training, some have climbed different ladders in order to model the theory, or show that it has been at work in evolution generally. Also, a great deal depends on what one means by group section, and I suspect, that many mean different things.

    • David Sloan Wilson says:

      There is a problem of not seeing the forest for the trees! The reason that Adaptation and Natural Selection was an important book is that Williams was able to talk about the major issues in plain English, without the need for math. He established that higher-level adaptations require a process of higher-level selection. Too bad that he erroneously concluded that group-level adaptations don’t, in fact exist. Everyone who thinks about evolution, biological or cultural, needs to have a Williams-style account that concludes that group-level adaptations do exist and can often be ascertained without much difficulty. With respect to the end of Andrew’s comment, equivalence is a mixed blessing. In my “mopping up final opposition…” article, I note that the optimal number of equivalent frameworks is between “more than one” and “not too many” And the community of scientists need to be multi-lingual or else science turns into a tower of babel.

      • Andrew Atkinson says:

        An interesting article here – and thanks for your reply, David, I wish there were remit here to comment on the litany of orthdoxy in some special sciences that examine group selection. I suppose the key to my own argument is simply pointing out that many, according to their respective disciplines have climbed diffferent ladders in order to ascribe credulity to group selection, and fallen off them – each of their own accord and respective heights gained. These have been mathematical, genetic, philosopical, academic in competition, and all coupled with a rigid adherence to their being a dogmatic problem of altruism and one that faviours individual selection, that might prevent talk about multi-level, evoliutionary, explantions of a given sort of groups’ evolutionary success – as with many other cultural phenoma and invidual actions. As said, the original appeal to group selection was to explain altruism as an adaptation selected for at the group level – ergo by moving up the bilogical hierarchy to consider selectiion for costly altrusim at the level of the group over selfish individual interest – the adaptation (if their is one to be had) is that altruism favours all concerned. Altruism, might ultimately turn out to be selfish in any form – not just within a species, but over other species as well (a bombastic claim indeed; but one to which there might be a point) that being “altruistic” itself – might in turn out to be quite sycophantic in the grand evolutionary scheme of things. There has been a history of debate whereby models of evolutionary altruism has been simply acquainted with selfish reciprocal altruism, or similarly kin-ship selection. But, getting back to my comment made on the article, the key is that the principles of a Darwinian interplay between groups remains the case in every instance where there are competitors for life, and therefore we need not have, nor have to be bothered with the concept of altruism in order to think of many phenomena as having group/individual fitness augmenting qualities that might make a given group successful. One might wonder how culturally augmenting qualities are retained, not genetically, but through cultural transmission, and the cognitive competence for it – and the genes that endow us with cognition from the outset – they are interlinked via selection at various levels.

  5. David Sloan Wilson says:

    Thanks for these comments so far. One shouldn’t make too much of the survey–that’s why I begin my post by saying that science is not a popularity contest. At the same time, it is welcome news that a consensus that by my estimation already exists in the peer-reviewed literature has penetrated to a particular academic sector. We need more surveys of other sectors and the general public. Next I’ll try to respond to individual comments.

  6. Oliver Scott Curry says:

    The multiple meanings of ‘group selection’ make it difficult to interpret the results of this survey. As the original article documents, the term ‘group selection’ is used to refer to a number of different theories and processes, including ‘naive’ group selection, new group selection (trait-group, multilevel), cultural group selection, as well as simple mutualism. So when 35% agree (and 65% disagree) that ‘group selection has no utility’, we don’t know which theory/theories they were talking about. The finding that 79% think cultural group selection ‘important’, whereas 63% think trait-group selection only ‘plausible’, suggest most respondents were thinking about the effects of culture.

  7. […] 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 of […]

  8. […] a scientific controversy (group selection) that took over half a century to resolve itself. (Yes, it is resolved.) Even so, the ability of the scientific process to arrive at the facts of the matter is […]