This View of Life Anything and everything from an evolutionary perspective.
FIND tvol:
Challenge To Kin Selectionists. Explain This!
David Sloan Wilson
David Sloan Wilson
is the SUNY Distinguished Professor of Biology and Anthropology at Binghamton University and Arne Næss Chair in Global Justice and the Environment at the University of Oslo

Major controversies in science have a way of appearing obvious in retrospect. We find it hard to understand why smart people took so long to agree that the earth revolves around the sun, that glaciers once covered the northern latitudes, that the continents drift, and that species are derived from other species.

So it is with group selection, a theory that was declared dead in the 1960’s, only to come to life as an essential tool for understanding animal and human societies. Group selection theory employs the following assumptions.

1) Natural selection is based on relative fitness.

Sign up for our newsletters

I wish to receive updates from:

2) Selection among individuals within groups tends to favor traits that are called selfish in human terms; that is, traits that benefit individuals at the expense of other members of the group and the group as a whole.

3) The evolution of group-advantageous traits typically requires a process of selection among groups in a multi-group population.

4) Groups are defined as the individuals who influence each other’s fitness with respect to the evolving trait.

Or, as Edward O. Wilson and I put it in a 2007 article[i], “Selfishness beats altruism within groups. Altruistic groups beat selfish groups. Everything else is commentary.”

Assumption (1) is one of the bedrock assumptions of evolutionary theory. Assumption (2) is a basic matter of tradeoffs. Imagine modifying the game of monopoly so that the goal is to collectively purchase and develop the properties. That game would call for a completely different set of strategies than when the goal is to beat the other players. Most social endeavors are like the game of monopoly in this respect. Assumption (3) seems to follow inevitably from (1) and (2). If fitness differences within groups weigh in favor of selfishness, then what else can weigh in favor of group-advantageous traits but fitness differences between groups in a multi-group population? Assumption (4) is a common-sense definition of groups that is invariably employed in mathematical models of social evolution to identify the individuals that influence the fitness of any given focal individual. Taken together, these assumptions have the same kind of obviousness in retrospect that makes the three basic ingredients of natural selection (variation, selection, and heritability) so obvious and caused Huxley to exclaim “How stupid of me not to have thought of that!”

Against this background, it seems remarkable that George C. Williams could declare in 1966 that “group-level adaptations do not, in fact, exist” [ii]and that every other major theory of social evolution was self-consciously developed as a way to explain seemingly group-advantageous behaviors without invoking group selection. As an example, Richard Dawkins began his book The Selfish Gene [iii]by describing how a group of altruists invaded by a selfish individual must inevitably convert to a selfish group and how the process of between-group selection, while possible in principle, was invariably too weak to oppose the within-group advantage of selfishness. By his own account, the major import of selfish gene theory was to provide another explanation for how selfish genes can cooperate in the bodies of individuals and, occasionally, groups.

Today, the four assumptions listed above can be accepted at face value and many examples have been documented of traits that evolve on the strength of between-group selection, despite being selectively disadvantageous within groups. Between-group selection can even dominate within-group selection for multiple traits, turning groups into organisms in their own right, a phenomenon known as major evolutionary transitions. But the revival of group selection theory doesn’t mean that the alternative theories are wrong. All of the major theories of social evolution can qualify as right and might deserve to coexist, as strange as this might sound. A concept of equivalence has emerged from the controversy over group selection that rivals the concept of paradigms in importance when it comes to understanding the scientific process.

To introduce the concept of equivalence, imagine an Englishman who is so pompous that he regards his language as superior to every other language. In his most insufferable moments, he claims that other languages aren’t capable of expressing concepts that can be expressed in English. In his saner moments, be begrudgingly admits that other languages can express the same concepts but not nearly as well as English. To him, it is only reasonable for everyone to speak English.

The concept of equivalence points out that some scientific theories are like languages. They examine the same causal processes but in ways that require a degree of fluency to translate from one to the other. Scientists who are well versed in only one theory can mistakenly assume that the other theories are just plain wrong or at best provide a clumsy way to describe what their own theory describes more elegantly. Scientists who don’t get the concept of equivalence are like pompous Englishmen.

Equivalent theories are different from alternative hypotheses or the concept of paradigms that was first articulated by the philosopher Thomas Kuhn in the 1970’s. Alternative hypotheses invoke different causal processes. In the game of Clue, I might hypothesize that the murder took place in the library, with the revolver, by Miss Scarlet. There is a right and wrong to the matter and more than one hypothesis can’t be correct. Kuhn observed that scientists sometimes get stuck viewing a topic a certain way. Their particular configuration of ideas is capable of a limited degree of change through hypothesis formation and testing, but cannot escape from their own assumptions in other respects, which makes the replacement of one configuration of ideas (a paradigm) by another configuration a messy and uncertain process. Still, in Kuhn’s rendering, one paradigm eventually does replace its rival. Scientists don’t talk about pre-Copernican views of the solar system any more.

The concept of equivalence therefore adds to the more familiar concepts of alternative hypotheses and paradigms in the philosophy of science, not just for the group selection controversy but all scientific controversies. The great danger is to mistake an equivalent theory, which invokes the same causal processes, for an alternative hypothesis or paradigm that invokes different causal processes. If equivalent theories are treated as if one deserves to replace another, nothing but confusion can result.

Precisely this confusion has plagued evolutionary theories of social behavior for over half a century. The replicators and vehicles of selfish gene theory, the coefficient of relatedness in kin selection theory, and the N-person groups of evolutionary game theory are all so many ways of describing evolution in multi-group populations that include the four assumptions listed above. They are inter-translatable with each other and with group selection theory. This means that they can all be right in predicting what trait evolves in a given situation. But they were wrong when they claimed to explain the evolution of these traits without invoking between-group selection, since they invoke everything but the name within their own frameworks.

Once the concept of equivalence is understood, the group selection controversy can be declared over. I provide a post-resolution account in my newest book Does Altruism Exist? Culture, Genes, and the Welfare of Others [iv]. In keeping with the obviousness-in-retrospect of resolved controversies, I describe how altruism evolves and the concept of equivalence in terms that can be understood by any interested reader without requiring specialized knowledge. If you doubt my own assessment, then consider a recently published article in the Annual Review of Psychology titled “The Evolution of Altruism in Humans” by Robert Kurzban, Maxwell N. Burton-Chellew, and Stuart A. West[v], who is one of the main current critics of group selection theory. They state: “The new group selection (multilevel selection) and kin selection are just different ways of conceptualizing the same evolutionary dynamics…In all cases, where both methods have been used to look at the same problem, they yield identical results (p 10.5).” Or consider a recent article published in BioScience titled “Kin Selection and Its Critics” by Jonathan Birch and Samir Okasha[vi], who is widely regarded as one of the foremost authorities on theories of social evolution. They state: “In earlier debates, biologists tended to regard kin and multilevel selection as rival empirical hypotheses, but many contemporary biologists regard them as ultimately equivalent, on the grounds that gene frequency change can be correctly computed using either approach. Although dissenters from this equivalence claim can be found, the majority of social evolutionists appear to endorse it (p 28).” That sounds like a resolution to me. Anyone who argues the major theories of social evolution against each other runs the risk of sounding like a pompous Englishman who insists that everyone else should speak English.

Some terminological points are in order to fully make sense of the passages quoted above. Group selection has always been a bi-level theory by partitioning evolutionary change into within- and between-group components. It becomes multi-level when it is extended downward (between genes within individual organisms) and upward in a multi-tier hierarchy of units. Some authors think that multilevel selection theory has changed enough over the years to distinguish between an old and new version. The same can be said of kin selection, which no longer is restricted to genealogical relatedness, for example. Birch and Okasha distinguish between three current versions of kin selection in their article. Theory development is to be expected and is not a sign of weakness or inconsistency.

With equivalence firmly in mind, a new set of issues comes to the fore. An equivalent theory must pull its weight in some way to remain in use, presumably by offering insights that are less forthcoming from the other theories by virtue of their different perspectives. Also, two theories might be equivalent in some but not all respects. For the rest of this article I will provide three examples of group selection that might not be translatable into kin selection. I challenge my kin selectionist colleagues to prove me wrong.

Kin selection theory attempts to state the criterion for the evolution of a trait in the form rb>c, where c represents the effect of the trait on the fitness of the actor (which is negative for an altruistic trait), b represents the effect of the trait on the fitness of the recipient (which is positive for an altruistic trait), and r is a coefficient of relatedness between the actor and the recipient, which can be based on genealogical relatedness or any other process that causes like to associate with like. Birch and Okasha provide an up-to-date review of the ins and outs of these terms. Here are three examples of group selection that are hard and perhaps even impossible express in this form.

Equilibrium selection: Complex social interactions often result in multiple locally stable equilibria, which means that a given trait can be favored within some groups but not others. Locally stable equilibria can vary widely in their group-level properties and between-group selection favors those that collectively survive and reproduce better than others[vii]. It is difficult to see how this process, which is straightforward to describe in terms of multilevel selection, can be expressed in the form rb>c.

Community-level selection: In the 1980’s, Charles Goodnight raised two species of flour beetles (call them A and B) in vials of flour, with each vial containing both species[viii]. After several weeks, vials with the highest density of one of the species (say, A) were selected to initiate a new set of vials. In other words, two-species communities were selected on the basis of a phenotypic trait, which was the density of one of the species. There was a response to selection, such that the density of species A increased over a number of generations. What genes were selected to produce this effect? It turns out that genes were selected in both species that interacted with each other to increase the density of species A. In essence, the species were like chromosomes inside a single vehicle of selection. How can this be expressed in the form rb>c at the level of individual beetles?

William Swenson and I continued this line of research with more complex communities[ix]. In one experiment, we grew plants in sterilized soil to which we added six grams of unsterilized soil from a well mixed slurry. In separate lines, the soil from underneath the largest and smallest plants was selected to inoculate a new generation of pots to grow plants from the same seed source as the previous generation. In other words, we were selecting soil ecosystems rather than plant genes. In another set of experiments, we selected aquatic microbial ecosystems for their ability to change the acidity of their media or to degrade a toxic compound. The communities that initiated each generation included millions of individuals and hundreds of species. The initial variation among communities at the beginning of each generation, based on sampling error, was negligible. Nevertheless, complex interactions among the species created variation among communities in the phenotypic trait under selection and there was a response to selection, which is proof of heritability at the ecosystem level. The experiment was recently replicated for soil ecosystems using the phenotypic trait of plant flowering time[x] and the method holds promise for selecting ecosystems with useful properties, such as plant growth or soil remediation. It can be easily understood as an example of higher-level selection but how can it be expressed in the form rb>c from the perspective of individual organisms within the ecosystems?

Human Cultural Evolution: Multi-level selection theory is proving to be indispensible for the study of human cultural evolution. Human groups vary widely in their practices in ways that have little to do with genetic relatedness, genealogical or otherwise. Phenotypic variation occurs at all spatial scales, from hunter-gatherer groups to nations. A given cultural trait can spread at the expense of other traits within the group or by virtue of giving the group a competitive advantage over other group. Peter Turchin uses multi-level cultural evolution to explain the rise and fall of empires. Extreme between-group conflict acts as a crucible for the selection of cooperative societies. The most cooperative expand into empires, but then disruptive cultural evolution takes place within the societies, causing their collapse. In a TVOL article titled “Blueprint for the Global Village”, Dag Olav Hessen and I show how multilevel selection can be used to formulate public policy, using Norway’s pension fund as an example. I find it difficult to see how cultural group selection operating at large spatial scales over long time periods can be translated into kin selection theory.

There is a good reason why multilevel selection theory and kin selection theory might not be equivalent in all respects. Multilevel selection theory is not a modeling method. It is a causal hypothesis based on the four assumptions listed at the beginning of this article, which can be modeled in any number of ways (e.g., analytical models, computer simulations). In contrast, kin selection theory is a modeling method for traits that can be measured in individual organisms, which requires estimation of the b, c, and r terms. This makes kin selection theory unsuitable for cases in which individual-level traits interact with other individual-level traits in a complex fashion to produce group level phenotypes. As we have seen, when groups and multi-species communities are selected in the laboratory, the entire experiment can take place by measuring variation, selection, and heritability at the level of higher-level units without requiring any knowledge of individual-level traits, just as artificial selection experiments at the level of individual organisms can take place without any knowledge of the genes and their interactions that result in variation and heritability at the individual level.

It is not my purpose to disparage kin selection theory, however. As someone who appreciates equivalence, I acknowledge the usefulness of kin selection theory in some cases. My challenge to kin selectionists to explain these three cases is friendly and I will applaud them if they succeed.

[i] Wilson, D. S., & Wilson, E. O. (2007). Rethinking the theoretical foundation of sociobiology. Quarterly Review of Biology, 82, 327–348.

[ii] Williams, G. C. (1966). Adaptation and Natural Selection: a critique of some current evolutionary thought. Princeton: Princeton University Press.

[iii] Dawkins, R. (1976). The Selfish gene (1st ed.). Oxford: Oxford University Press.

[iv] Wilson, D. S. (2015). Does Altruism Exist? Culture, Genes, and the Welfare of Others. New Haven: Yale University Press.

[v] Kurzban, R., Burton-Chellew, M. N., & West, S. A. (2014). The Evolution of Altruism in Humans. Annual Review of Psychology. doi:10.1146/annurev-psych-010814-015355

[vi] Birch, J., & Okasha, S. (2014). Kin Selection and Its Critics. BioScience, 65(1), 22–32. doi:10.1093/biosci/biu196

[vii] Samuelson, L. (1997). Evolutionary games and equilibrium selection. Cambridge, MA: MIT Press.

[viii] Goodnight, C. J. (1990). Experimental studies of community evolution I: The response to selection at the community level. Evolution, 44, 1614–1624.

Goodnight, C. J. (1990). Experimental studies of community evolution II: The ecological basis of the response to community selection. Evolution, 44, 1625–1636.

[ix] Swenson, W., Wilson, D. S., & Elias, R. (2000). Artificial Ecosystem Selection. Proceedings of the National Academy of Sciences, 97, 9110–9114.

Swenson, W., Arendt, J., & Wilson, D. S. (2000). Artificial selection of microbial ecosystems for 3-chloroaniline biodegradation. Environmental Microbiology, 2, 564–571.

[x] Panke-Buisse, K., Poole, A. C., Goodrich, J. K., Ley, R. E., & Kao-Kniffin, J. (2014). Selection on soil microbiomes reveals reproducible impacts on plant function. The ISME Journal. doi:10.1038/ismej.2014.196


Join the discussion


  1. Eric Falkenstein says:

    The models I’ve seen in this area seem to be steady states, but intuitively the dynamic of individuals gaining via selfish traits and groups gaining via altruistic traits seems like an oscillating equilibrium that doesn’t have a nice closed analytic form. Is there a canonical multilevel selection model?

    • David Sloan Wilson says:

      No, and there are good reasons why this is so. Statistical partitioning methods such as the Price equation and contextual analysis correctly classify some cases of multilevel selection but misclassify others–unsurprising, because correlation does not imply causation. Different population structures, such as groups fixed in space with migration among adjacent groups vs. groups with periodic complete mixing, call for different models. Some cases with complex nonlinear interactions require computer simulation. There is no canonical complex system model and no canonical multilevel selection model. Thanks for your comment.

  2. David Whitlock says:

    The problem with “kin selection” is that it cannot be genetic. Sensory systems are unable to differentiate kin from non-kin at birth. Sensory systems have to self-modify (i.e. learn to recognize kin) before kin can be recognized. For example ants learn the home nest odor following eclosion, humans learn to recognize faces.

    The idea of the “selfish gene” cannot be correct other than for viruses. As far as I have been able to determine from the literature there are no examples of “selfish genes” in any extant organism, except for Homing Endonuclease Genes (HEGs). HEGs are extremely selfish. My hypothesis is that it is the necessity of protecting species from extinction due to accumulation of HEGs that limits the fidelity of DNA replication. Limited DNA replication fidelity is a “feature”, not a “bug”.

    The problem with Selfish Gene Theory is that organisms can only detect phenotypes, not genes. There is no way for an organism with a Green-Beard to know if a conspecific with a green beard shares the same Green-Beard-Gene, or has a green beard via green beard mimic gene(s) (or via green beard dye).

    A phenotype that expresses a green beard while not expressing the phenotype favoring Green-Beards will out compete all Green-Beards who also favor Green-Beards. There are many more ways to have a green beard without favoring Green-Beards than there are ways to have a Green-Beard while also favoring Green-Beards. I am pretty sure this is why there are no observations of selfish genes of the Green-Beard-type, they are all out-competed by Green-Beard-Mimics. Viruses don’t have any replication mechanisms to be hijacked, so viruses are immune to selfish genes and their mimics.

    • Eric Falkenstein says:

      But isn’t the multilevel solution that a Green Beard society with lots of selfish cheaters will be exterminated by a Red Beard society with few cheaters, because a group with greater cooperation will be more numerous and efficient?

    • Chris says:

      Sorry David, you have really not understood Kin Selection. Rather than go over each error, I suggest you read Dawkins’ ’12 Misunderstandings of Kin Selection’ as you have made at least two of them. By the way your Green Beard point is right and Dawkins mentions precisely the reason you state as the reason we do not know of any Green-Beard type altruism.

      • Chris says:

        Oops I should have read further – I see DSW has beaten me to it (by months -D’oh). Nonetheless, still recommend the Dawkins article. And of course defer to DSW if he says Green Beard altruism has been observed.

  3. David Sloan Wilson says:

    Thanks for these interesting comments but I take issue with some of them. There are examples of fish and tadpoles that can recognize kin even when separated before birth, presumably using odor cues. There are green beard models that work. The mere fact that some individuals can sport the green beard without behaving altruistically does not fatally undermine the process, any more than does the selfish individual invading the group of altruists in Dawkins’s scenario? And what if the green beard is the observed altruistic act? Finally, your definition of selfish genes is different than the Dawkins version, which is a popularization of the population genetics concept of average effects–the fitness of genes averaged across all genotypic and social contexts. It is a useful concept, as long as it isn’t used to argue against group selection!

  4. David Sloan Wilson says:

    In the standard green beard model, one gene codes for altruism and another gene codes for a visible trait. If the two traits are correlated with each other, then like can sort with like and altruism can evolve. The critique points out that all four combinations can exist and if the traits become uncorrelated, then the visible trait no longer becomes a proxy for the altruistic trait. Fair enough, but there are scenarios in which the traits remain correlated. The most plausible is when the visible trait is the expression of altruism. This is essentially what a reputation is; a visible marker of altruism that can’t be easily faked because it is based on actual behavior. Notice that there is no red beard society in this scenario. If there was, then whatever undermine the Green Beard society could presumably undermine the Red Beard society. However, something called Lineage Selection corresponds to Eric’s scenario, as Leonard Nunney has conjectured for the evolution of sexual reproduction. Imaging one sexual lineages. One gives rise to asexual forms more easily than the other. Every time a sexual form arises, then it quickly drives the sexual form extinct and then itself goes extinct for lack of variability. This could result in the lineage that rarely gives rise to asexual forms replacing the lineage that frequently gives rise to asexual forms.

    I’m still waiting for a kin selectionist to explain my three examples!

    • David Whitlock says:

      I think I am in complete agreement with you about multi-level and group selection. I think that group selection is the more fundamental process, and kin selection is a special case of group selection (and is of relatively minor importance).

      One of the things that bothers me (in general), is the fetishization of “genetic causation”, without any actual data that differential genetics is responsible for the differential behaviors and not differential learning, or differential development due to environmental effects (a particular case of “learning).

      This is my objection to the green-beard idea and also the kin-favoring behaviors. If the behaviors are “learned” (as in ants imprinting on the home-nest-odor), then it really isn’t “kin-favoring-behavior”, it is “favoring of those who share a common home-nest-odor”.

      Where this is especially important is in human interactions. Racism and human bigotry is learned. Humans cannot identify adults of different phenotype at birth. Humans cannot learn to hate a certain phenotype until after that phenotype can be identified.

      • K.R. Pledger says:

        The particular home-nest odor they imprint on may be learned, but they still start out with a genetic predisposition to imprint on whatever odor they happen to be exposed to.

        “If the behaviors are ‘learned’ (as in ants imprinting on the home-nest-odor), then it really isn’t ‘kin-favoring-behavior’, it is ‘favoring of those who share a common home-nest-odor’.

        You seem to be confusing proximate and ultimate explanations of behavior here. Proximate explanations refer to the immediate psychological and/or neurobiological mechanisms of behavior. The theory of kin selection is an ultimate explanation, not a proximate explanation. It doesn’t matter *how* an ant comes to interact with genetically related individuals; as long as the individuals it interacts with are, in fact, genetically related, kin selection can account for the presence and propagation of the cooperative or altruistic trait. If sharing a home-nest odor is correlated with sharing genes, it doesn’t matter (on the “ultimate” level) that the ants are responding to the former instead of the latter.

        • David Whitlock says:

          It does matter if it is ultimate or proximate.

          Genetic “kin selection” does not explain how some ant species can parasitize other ant species by stealing larva and hatching them in the new nest. Learned “group selection” based on a learned home nest smell does.

          If imprinting on a specific “home nest smell” was mediated genetically, then there would be no ability for ants to exclude conspecifics because they would all imprint on the same home-nest-smell (the one coded for genetically).

          The home-nest-smell, the home-nest-smell-detector, and the home-nest-smell-response-behavior, very likely cannot be mediated genetically because ants require genetic diversity in worker genomes (why queens need to mate with multiple drones) to improve nest stability and infection resistance; and the dispersion in home-nest-smell detectors due to the necessary genetic diversity would not allow nests to be distinguished.

          If home-nest-smell imprinting was not “learned”, then ants would be unable to distinguish one ant nest from another. Foreign ants could not be excluded.

          If we consider how home-nest-smell-imprinting evolved, it must have first required a home-nest. Only when nests became attractive enough reservoirs of biomass (ants, ant larva, and stored food), so that excluding parasites and other “cheaters” became important, would the behavior to exclude those not bearing the home-nest-smell evolve (when the benefits of excluding parasites and cheaters exceeds the costs of excluding parasites and cheaters).

          The mechanism of how to exclude parasites and cheaters had to evolve as quickly as it could in order to “protect” the expanding biomass of the nest (as eusocial ants became more successful). A system whereby recognizing conspecifics is learned (from the learned home-nest-smell), is a lot easier to evolve (requires fewer DNA changes, and can occur step-wise from simple to complex) than a system of genetic identification and genetic recognition (which essentially needs to appear de novo with high fidelity).

          The adaptive immune system is learned. The ability to distinguish self from non-self is learned. That self-from-non-self is learned makes it a lot easier for organisms to evolve new tissue types. As organisms evolved to be multicellular, they needed to evolve an adaptive immune system to protect their accumulated biomass from predation by bacteria. That adaptive immune system needed to evolve as quickly as it could. An adaptive immune system mediated by learning can evolve faster than an immune system that is “hard-wired” genetically.

          The adaptive immune system doesn’t protect “kin”, it protects cells that are a member of the “group” that comprises the organism that the adaptive immune system “learned” to protect. Chimeric sheep/goats are easy to produce by combining undifferentiated cells into a single embryo. The adaptive immune system of the chimeric sheep/goat is able to recognize and protect all the cells, sheep derive and goat derived, even though sheep and goats have not been “kin” for millions of years and cannot interbreed. It is not “kin” recognition that allows goat and sheep chimera. Sheep and goats are not “kin”.

          • K.R. Pledger says:

            David Whitlock: “It does matter if it is ultimate or proximate.”

            As I was saying, “ultimate” and “proximate” refer to different levels of explanation; they aren’t mutually exclusive. The mechanisms by which ants interact with one another is a matter of proximate explanation. Immediate mechanisms of behavior are _always_ a matter of proximate explanation; there’s no such thing as an “ultimate” immediate mechanism.

            You seem to think “ultimate” refers to a type of mechanism — one that consists of a genetic predisposition with no component of learning or feedback. But that is not what “ultimate” refers to; it refers to the general, long-term consequences or practical significance of traits within particular ecological contexts. Ultimate explanations are on an entirely different level from proximate explanations; they pertain to separate things. The theory of kin selection is an ultimate explanation.

            As David Sloan Wilson said in one his comments, the existence of exceptions — e.g., cheaters, your human-engineered chimera example, etc. — does not negate the “average effects” of the trait or behavior. The average effects are what are of essential relevance to ultimate explanations.

            For example, geese are genetically predisposed to “imprint” on the first caretaker they’re exposed to (a mechanism of behavior consisting of a predisposition with a component of learning). The fact that Konrad Lorenz once took advantage of this mechanism and got some baby geese to imprint on him doesn’t change the fact that, most of the time, the first caretaker they’re exposed to is their biological mother. The ultimate significance of the behavior (the reason a trait becomes more and more common across generations) comes from its general consequences; exceptional consequences in exceptional circumstances are just that — exceptional.

            I’m not presently concerned about the nuanced conflict between the theories of kin selection and group selection — only about the fatal flaw you believe exists in kin-selection theory, which many prominent scientists would have to have somehow managed to overlook. The distinction between ultimate and proximate levels of explanation is fundamental to evolutionary theory and a common source of confusion and misunderstanding. Your criticism of kin selection here appears to be based, at least in part, on this misunderstanding.

          • K.R. Pledger says:

            David Whitlock: “If the behaviors are ‘learned’ (as in ants imprinting on the home-nest-odor), then it really isn’t ‘kin-favoring-behavior’, it is ‘favoring of those who share a common home-nest-odor’.”

            You could say the Westermarck effect isn’t really an “anti-incest” predisposition but rather an “anti-being-sexually-attracted-to-people-you-grow-up-around” predisposition, but there wouldn’t actually be anything contradictory about these two descriptions, because the first is an ultimate description and the second is a proximate description.

            We don’t need to be born knowing who exactly our siblings will be, or be able to recognize genetic relatedness per se. As long as the people we grow up around _tend_ to be our biological relatives, it will reduce the incidence of incest and thus be adaptive for that reason. And it will be meaningful to describe it as an anti-incest/anti-kin-mating trait.

            The immediate mechanism of the behavior may sometimes lead to sexual repulsion between unrelated individuals raised together, or sexual attraction between related individuals raised apart. But as long as related individuals are raised together most of the time, the ultimate function of the predisposition can be said to be anti-incest.

    • David Whitlock says:

      I finally got a chance to read your book (although quickly and incompletely). I like your invocation of Kuhn.

      I think the “problem” of disagreement in the field is mostly one of perspective and results from human hyperactive agency detection; the human cognitive compulsion to impute top-down agency, “causation”. The problem is that before development happens, there is no “top”. There can be no top-down “agent” that can “control” what is going on during the course of development (or in any cell-to-cell interactions). I am still in the process of translating between our conceptualizations. I am coming at this more from the perspective of control theory and physiology, rather than from genetics.

      The problem with a gene-centered view of evolution is that genes can’t work in isolation, either by themselves (without the cooperation of other genes), or as a genome (without the cooperation of other cell components), or as a cell without the cooperation (interaction) of the environment. Genes can’t compel cooperation with anything. However genes can be part of a system that self-modifies so as to cooperate with the environment, i.e. optimize itself for the conditions it encounters (within limits).

      The easiest way to describe this is that in the “conventional view,” the “locus of control” is gene-centered. That is, the “genes” control the development and the eventual properties of the phenotype. I am pretty sure that the motivation for the gene-centered view is human hyperactive agency detection. Human cognition has the default of imputing “agency”, due to the evolution of sensory systems to detect and trigger escape from predators. False-positive detection of agents that are not there is much better than false-negative non-detection of agents that are there.

      The locus of control cannot be gene-centered, because “genes” can’t “sense”, or “know” what the environment is, and how to modulate themselves or the development trajectory without “signals” from the environment mediated through physiology. The “genes” don’t “know” what to do, they only “do” what the “signals” tell them to do (e.g. via transcription factors). How “the signals” differentially regulate differential transcription, repression/expression remains essentially completely unknown.

      My formulation is consistent with group selection. My formulation is incommensurate with a gene-centered view of evolution. I think that what happens is that “control” compels the genes and molecules to “work together”, “in sync” and so instantiate the physiology of the organism which instantiates that control system. The primary locus of that control system has to be “bottom-up” (because without a control system that spans the whole organism there is no “top”). That bottom-up emergent control system then emulates a “top-down” control system that keeps the whole of physiology “in sync”.

      Physiology has arranged itself as a self-organized criticality. That is how an organism “optimize itself for the conditions it encounters (within limits).” The “signals” are at the percolation threshold, where the length scale of effects diverges across the whole organism. In the near critical percolation threshold, the properties of the system are exponentially sensitive to differential changes in the connectivity and magnitude of those parameters. At the critical point there is true divergence, the system becomes infinitely sensitive. That degree of sensitivity is not stable, so living organisms have to operate somewhat below, or beyond the critical point. These interactions are exponentials. They are inherently non-linear. A linear approximation cannot be used to determine where the non-linear system “breaks” and becomes unable to maintain viability.

      Ecosystems self-organize as a self-organizing criticality. The genome has arranged itself as a self-organizing criticality. Neural networks do to. I think that all systems that dissipate free energy and maintain stability over time do so by instantiating a self-organizing criticality. That self-organizing criticality is self-regulated by interactions between the active components at the shortest length scales, from the “bottom-up”. Those interactions between cells must be (real-time) “tuned” so as to maintain the self-organizing criticality. Only then can emergent behaviors occur, resulting from whole tissue compartments operating “in sync”. Thinking is an emergent behavior of many millions of brain cells working together “in sync”. Each cell doing something different, but the cells working “together” to instantiate the collective behaviors.

  5. Mark Sloan says:

    It is great to see the two referenced papers arguing that the generally accepted modern “kin selection theory” and multi-level selection theory are equivalent. That is wonderful progress!

    I hope your challenge to kin selectionists is answered. Assuming kin selection theory can do the job at all, comparing the two approaches side by side should be highly revealing.

    To me, the big advantage in many areas for the group selection perspective over kin selection theory is group selection’s intuitive and simple “in practice” explanatory power which I expect such a comparison would make obvious.

    I also had a question about:

    “The replicators and vehicles of selfish gene theory, the coefficient of relatedness in kin selection theory, and the N-person groups of evolutionary game theory are all so many ways of describing evolution in multi-group populations that include the four assumptions listed above.”

    But winning cooperation strategies for N-person groups in evolutionary game theory (such as direct and indirect reciprocity) appear to contradict the second assumption:

    “2) Selection among individuals within groups tends to favor traits that are called selfish in human terms …”

    Specifically, selfishness can reduce fitness if it is punished. In species where reputations are important, a bad reputation for selfishness could cause everyone, selfish and unselfish alike, to refuse to cooperate with the selfish person, thereby preventing them from accessing the benefits of cooperation. Winning cooperation strategies for N-person groups (which all include punishment of exploitation of altruism) are in act favored in “groups” as defined by assumption 4).

    Are these winning ‘altruistic’ strategies just exceptions to the assumptions? If so, they seem too important to not be mentioned.

    After all, our first ancestors who were motivated to behave altruistically toward non-kin were doubtless pre-equipped with the emotional machinery to be motivated to punish, at minimum by non-cooperation, anyone who exploited their altruism. Biology based altruism toward non-kin that does not come as a packaged set with punishment of exploitation of that altruism is rare for good reasons.

  6. Carmi Turchick says:

    A couple issues I have that I believe both sides generally miss. First, I know you are familiar with the work of Eldakar since you were his adviser and co-authored work with him, but it seems to me that even he has not fully appreciated the implications of really recognizing that selfish are selfish. The common assumption has been that selfish will invade and overrun a group of altruists, and this is claimed to be “obvious.” But selfish and altruists will co-evolve: the ability to target altruists for exploitation evolved after altruists evolved and then altruists evolved to defend against exploitation and so on. For adapted selfish then a group of altruists is a fitness limiting resource, and being selfish it is not one they should like to share. This results in the selfish punishment of selfish that Eldakar finds. It also, I believe, results in selfish seeking to convince altruists to get rid of other selfish in the group, a process that essentially underlies what we know as politics today. So it is adaptive for both the selfish and the altruists in a group to have lower numbers of selfish; selfishness, as Eldakar notes, is a self-limiting strategy.

    But after a good decade of accepting MLS it was convincingly shown to me that really not matter how you wrap it, groups do not replicate. And no matter what, selection requires replication. One might then reject MLS, but instead what I realized is that both sides in this enormous intellectual struggle have been making the same error, just in opposite ways.
    Both sides conflate two distinct questions of selection. They both conflate the question of how something is selected with the question of what is selected, and these are two quite different questions. The MLS side says, look, the only way how these various traits like altruism can be selected is at the group level, therefore there is MLS. This is claiming that the answer to how is the answer to what, and it is not. On the other side those who reject MLS say that groups cannot be selected, therefore there is no “real” altruism and so on. This is claiming that the answer to what is the answer to how. It is not.
    Pretend for example that one was in grad school in evolutionary biology and was asked how flight evolved in birds. “Because of selfish genes” or “At the gene level” or “birds do not actually fly” would be the equivalent pathetic answer given by the anti-MLS folks. These are not answers for the question of how for birds or for the evolution of altruism, they are answers for what. What is actually not that hard a question these days.

    So how did altruism evolve? By force of selection operating at the group level, which is different from group selection. If a group competes for resources or reproduction at the group level then there must be force of selection, the how of selection, at the group level, or the theory of evolution itself is wrong. This does not mean that the group itself is the what of selection, but the results in terms of genes and behavior is the same AS IF the group was what was selected. Genes respond to the level of the force of selection, not to the what of selection.

    If we stop confusing and conflating how and what when we use the term selection then this whole conflict vanishes and we can move ahead in our understanding with some basic costly errors behind us.

    • David Whitlock says:

      Groups do not need to replicate for MLS to be effective. Groups may not “replicate”, but organisms that do better in groups where the organism contributed to increasing the inclusive fitness of the group do replicate and do become more prevalent.

      I think the “problem” is in premise #2;

      “2) Selection among individuals within groups tends to favor traits that are called selfish in human terms; that is, traits that benefit individuals at the expense of other members of the group and the group as a whole.”

      This is not necessarily true, and in some cases it is clearly false.

      In the example of the group consisting of plant plus bacteria, if a particular strain of bacteria can produce a particular substrate that improves photosynthetic generation of substrates (for example a siderophore), and increased iron allows the plant to release increased amounts of substrates to the rhizosphere (for example amino acids), then both the plant and the bacteria do better.

      The division of metabolic labor and increased productivity is (I think), a major driving force for the evolution of cooperation.

      The bacteria that generates the right siderophore can cooperate with any plant that can use that siderophore and that also releases substrates that the bacteria can use. Once substrates are released into a complex ecosystem, many things happen, and the interactions become complex. The bacteria producing the siderophore can be sustained by metabolic products generated by other bacteria in the rhizosphere, not necessarily directly by substrates released by the plant. The rhizosphere can then become an ecosystem, where there is dynamic interaction between numerous organisms.

      If the bacteria tries to “cheat” and not release the siderophore, then the plant can’t get more iron and can’t produce more substrates to release and nourish the bacteria. If the plant doesn’t release substrates, then the bacteria can’t generate siderophore. The plant and bacteria don’t need to share genes for their respective metabolic activity facilitate the increased growth of both. The plant and bacteria do need to share “signals”, the signal of the plant that it needs more siderophore (so the plant releases substrates that bacteria can turn into siderophores), and the bacteria “signals” that it needs substrates by releasing siderophores.

      Bacteria that cheat, and act as parasites on the group (consisting of the rhizosphere), may persist long term, but then there is a niche for bacteria that are parasitic on the cheaters. Bacteria that suppress the cheaters, contribute to the positive growth of the group (the rhizosphere) and MLS will favor groups (and all the individuals within groups) that have agents that “police” the cooperation of the members of the group.

      In David’s book, he reiterates Ostrom’s point of the importance of Proportional Equivalence Between Benefits and Costs is the key behind successful groups. If there is proportional equivalence, then the group will be successful. If there isn’t, then the group will not be successful. Shared genes are not necessary, just the “shared goal” of the group. If the shared goal of the group is to increase biomass of the group, then Proportional Equivalence divides the Benefits among the group members.

      I see that as the “invisible hand” of the market. Money is the substrate used to communicate/signal the “value” of “goods-and-services” being traded, and also is the exchange medium for transferring substrates used to generate those “goods-and-services”. Without that “signaling”, there cannot be division of metabolic labor between plant and bacteria (or between cells) (or within cells).

      • Carmi Turchick says:

        “Groups do not need to replicate for MLS to be effective. Groups may not “replicate”, but organisms that do better in groups where the organism contributed to increasing the inclusive fitness of the group do replicate and do become more prevalent.”

        We essentially agree on what is actually happening, but I have to say that in my opinion the MLS idea is conceptually confused and leads to a lot of objections and bad complex unnecessary calculations and definitions. Forgive me, Mr. Wilson, as I have not read your latest book yet. But your earlier efforts have included rather ornate definitions of groups and complex calculations of migration between them and all of this seems to be simply the product of confusing the how and the what question of selection.

        We can say that answering the question of how X was selected for in social territorial species can require examining the force of selection at the gene competition level, the individual level, and the group level. So there is are Multiple Force of Selection Levels, MFSLs. But what is actually selected is just genes, arguably individuals but I do not see that making a difference.

        As for the rest of your post, I feel I need to examine it more closely before responding.

        • Catherine Claxton-Dong says:

          I realize that the current paradigm accepts that groups do not replicate. But what makes you so sure? How can you tell that the group that you see today is not the descendent of what appeared to be the same group several decades previously?

          • David Whitlock says:

            What does it mean for a “group to replicate” as a necessary “feature” of group selection leading to “altruism” (a term which I do not like as it imputes an anthropomorphic motivation)?

            I presume it means that individual descendents of the original group constitute members of the later group. In other words, that there is a one-to-one correspondence between members of the original group and members of the later group, and that one-to-one correspondence is mediated through descent (the later group members are descendants of the initial group members).

            This type of group replication is not necessary for groups (and individuals in groups) to prosper.

            In the plant-bacteria example; when the interaction is mediated through siderophore production in exchange for carbon substrates, any bacteria that produces any siderophore that any plant can use to generate any carbon substrate will prosper, so long as it is with any plant that produces carbon substrates in response to siderophore release. The “group” that is prospering can be dynamic and generated from unrelated organisms whenever they find themselves in the proper niche.

            It isn’t “altruism” for a bacteria to convert carbon substrates into siderophores. It is making the niche it finds itself in a better habitat for itself and its descendants over the long term. The plant isn’t being altruistic when it releases substrates, it is fostering an environment where bacteria generate siderophores.

            I am pretty sure that the experiment could be replicated with a different plant, and the microbiome adapted to increase growth in the first plant would (very likely) facilitate growth in the second plant, even if the plants were unrelated (so long as they used the same siderophore that the microbiome generated).

  7. Jorge Juarez says:

    How is the article by Goodnight a proof of community selection? If he selected for vials with higher density of either species, it simply means that he was selecting for that species’ population in which genes were fixated that gave it a certain fitness. Genes that first appeared at the individual level. Making the fitness measure of a population to be the “community trait” is bogus. Also, in his between species correlated traits, you could explain it even without genetic change in the emigrating species. If species A is increasing in density, species’ B emigration rate may be simply a case of phenotipic plasticity (they are being pushed away by increasing numbers of species A). And even if species B was indeed evolving through selection, it would be evolution within species B. Species A is part of B’s environment.