My new book, This View of Life: Completing the Darwinian Revolution, claims that the Darwinian revolution won’t be complete until it makes sense of everything associated with the words “human”, “culture”, and “policy” in addition to the word “biology”. I rely heavily on Multilevel Selection (MLS) theory, and since this theory is famously controversial, it might seem to be a weak link in my argument. But this inference is mistaken. Instead, the major claims of my book are supported by all theories of social evolution based on what they share in common, as I will now show.

The first thing that all theories of social evolution share in common can be called the fundamental problem of social life. Think of any behavior that can be called prosocial, which means oriented toward the welfare of others or one’s group as a whole, and that behavior is very likely to be vulnerable to exploitation by behaviors that would be regarding as self-serving in human terms. Altruism is vulnerable to selfishness. Honesty is vulnerable to deceit. Bravery is vulnerable to cowardice, Generosity is vulnerable to hoarding one’s own wealth. Pitching in is vulnerable to free-riding. There might be some cases where maximizing one’s relative advantage within a group also benefits others and one’s group as a whole, but there can be no doubt about the rule.

This basic fact about social life posed a dilemma for Darwin. If prosocial behaviors are vulnerable to more self-serving behaviors in every group where both types of behaviors occur, then how can they evolve by natural selection? This dilemma only gradually dawned upon Darwin and is reflected in the revisions that he made in successive editions of On The Origin Of Species (go here for more). It was a problem of the first rank. If his theory of natural selection could not explain the evolution of prosocial behaviors, then it could not explain all aspects of design that had been attributed to a creator.

The dilemma faced by Darwin is inescapable and therefore faced by all modern theories of social evolution. It is simply a fact of life that most prosocial behaviors are not selectively advantageous within the groups where the social interactions are taking place. Darwin’s solution was that the evolutionary advantage of prosocial behaviors takes place at a larger scale. If the evolving population consists of many groups of socially interacting individuals, then more prosocial groups will robustly outcompete less prosocial groups, even if more prosocial individuals lose to less prosocial individuals within groups. Competition between groups can be direct or indirect, just as for competition between individuals. Darwin became increasingly clear about the need for between-group selection to explain the evolution of prosocial behaviors in the Descent of Man and later editions of On the Origin of Species. In short: Group-level selection was an amendment that he needed to add to complete his theory of natural selection.

All theories of social evolution must also reach the same conclusion. If a given prosocial trait is selectively disadvantageous within every group of a multi-group population, then how else can it evolve except with a positive selective differential at a larger scale?

Remarkably, all theories of social evolution must also reach the same conclusion about what counts as a group. For non-social traits, the fitness of an individual can be calculated based on the properties of that individual. For social traits, calculating the fitness of an individual requires knowing its properties and the properties of the other individuals with whom it interacts. This is not arbitrary and must be determined on a case by case basis. If individuals interact in groups of 5 individuals, that will result in different outcomes than if individuals interact in groups of ten individuals. There are other details that must be determined on a case by case basis, such as how the groups form and dissolve, which are loosely termed the “population structure” for the evolving trait. For example, if members of a group disperse as individuals, that will result in a different outcome than if they bud off from the group as subgroups. Any model of social evolution must get these biological details right on a case by case basis, just to calculate the fitness of individuals (and their genes in a genetic model).

What I just described clearly deserves the label “Two-level selection”, which can be generalized to Multilevel selection by frameshifting both downward (e.g., selection among genes within individual organisms) and upward (e.g., selection among groups-of-groups or multispecies ecosystems such as microbiomes). Its logic is represented in Darwin’s writing, the three pioneers of population genetics theory (Ronald Fisher, J.B.S. Haldane, and Sewall Wright), and G.C. Williams, who wrote in his 1966 book Adaptation and Natural Selection (p 92): “It is universally conceded by those who have seriously concerned themselves with this problem that…group related adaptations must be attributed to the natural selection of groups of individuals and that the natural selection of alternative alleles within populations will be opposed to this development.” The great puzzle that historians of science will need to answer is why the central logic appeared to be rejected in the 1960’s and replaced by a parade of other constructs, such as inclusive fitness theory, selfish gene theory, evolutionary game theory, and social selection theory, which claimed to explain “seemingly” prosocial behaviors in more individualistic terms (go here for more).

In the interest of keeping this essay short, suffice it to say that all of these constructs are like the old song Let’s Call the Whole Thing Off with its lyric “You say Tom-A-to, I say Tom-AH-to”. Look under the hood of any theory of social evolution and you will find the central logic of MLS theory. The evolving population consists of multiple groups. The groups are defined in terms of the individuals who are influencing each other’s fitness with their social interactions. The behaviors labeled “altruistic” or “cooperative” are selectively disadvantageous within the socially interacting groups. And a positive fitness differential at the group level is required for the prosocial behaviors to evolve. If a theory of social evolution does not include the central logic of MLS theory, it isn’t representing the biological facts of life and can’t even get started. See my previous book Does Altruism Exist? for a longer but still concise defense of this claim.

The bottom line is that my new book does not rely on a controversial theory in its claim that an evolutionary worldview can make sense of everything associated with the words “human”, “culture”, and “policy”. It relies on assumptions that are as robust as the assumptions underlying the theory of natural selection, which all theories of social evolution share in common.

Published On: March 6, 2019

David Sloan Wilson

David Sloan Wilson

David Sloan Wilson is SUNY Distinguished Professor of Biology and Anthropology at Binghamton University. He applies evolutionary theory to all aspects of humanity in addition to the rest of life, both in his own research and as director of EvoS, a unique campus-wide evolutionary studies program that recently received NSF funding to expand into a nationwide consortium. His books include Darwin’s Cathedral: Evolution, Religion, and the Nature of Society, Evolution for Everyone: How Darwin’s Theory Can Change the Way We Think About Our Lives, and The Neighborhood Project: Using Evolution to Improve My City, One Block at a Time and Does Altruism Exist? Culture, Genes, and the Welfare of Others. .

6 Comments

  • Steve Davis says:

    David, you said, “Think of any behavior that can be called prosocial, which means oriented toward the welfare of others or one’s group as a whole, and that behavior is very likely to be vulnerable to exploitation by behaviors that would be regarding as self-serving in human terms. Altruism is vulnerable to selfishness. Honesty is vulnerable to deceit. Bravery is vulnerable to cowardice, Generosity is vulnerable to hoarding one’s own wealth. Pitching in is vulnerable to free-riding. There might be some cases where maximizing one’s relative advantage within a group also benefits others and one’s group as a whole, but there can be no doubt about the rule.”

    I believe the examples you gave of “the rule” are far too general to be complying with a rule. In fact, I think the examples are only occasionally observed.

    While I admire your dedication and passion for the TVOL project, it seems to me that the debate has been diverted from a search for truth by the Hamiltonians with their obsession with altruism and cheating, into which you have been drawn with the quote above.

    Their flawed analysis of altruism, which has become the dominant feature of current evolutionary thinking, has pushed to the margins the true feature not only of evolutionary biology but of life itself; cooperation. The same with their focus on cheating, with the more delusional among them even pulling spite into the discussion!

    In a social group cheating and free riding can only rarely survive, and can certainly not flourish because they are punished. Even the most successful examples of cheaters – con-men and thieves, cannot make cheating their main activity. Most of their social interaction consists of cooperation. Which is why “the Family” is so important in mafia culture.

    Cooperation is the principal feature of social groups, because until cooperation begins for a particular purpose, all that exists is a collection of disparate individuals. Cooperation defines sociality, and so it is cooperation that should be the focus of evolutionary studies. I believe it was you David, who years ago wrote something along the lines of cooperation being the ocean we swim in.

    Punishment flowed from the origin of morality; rules to preserve the group. For example, if subsistence farmers form a group to build a dam on a creek, then that “behavior that can be called prosocial, which means oriented toward the welfare of others or one’s group” as you put it, is not going to be as you said “very likely to be vulnerable to exploitation by behaviors that would be regarding as self-serving” because cheating or slacking would be punished to preserve the group, which in turn would preserve the project. From this we see that cheating and slacking are only sustainable in modern large populations, and even then only to a small degree.

    Punishment, by the way, is not confined to human societies; even sparrows make use of it.

    The bottom line is that cheating and slacking, despite the excitement they provide for biologists of sociopathic bent, (not you David) because they are only recently prominent, provide little insight into social evolution. They are marginal issues that divert attention from the real story of evolution – the overwhelming significance of cooperation.

    • David Sloan Wilson says:

      Steve–You don’t address the fact that punishing is itself a public good subject to free-riding. If we consider two evolving traits, (a) cooperation vs non-cooperation and (b) punishment vs. non-punishment, then cooperators can become more fit than non-cooperators if there are enough punishers, but punishers remain less fit than non-punishers within the same group. The problem of group selection hasn’t been solved, just shifted from the cooperation trait to the punishment trait. Elliott Sober and i discuss this in Unto Others and numerous others have also.

      • Steve Davis says:

        David, I agree in general with this take on punishers and non-punishers, but it’s perhaps a little too simplified.

        People, and presumably other social animals, are not rigid or programmed in their actions. Just as I can be selfish today and altruistic tomorrow, so I can be a punisher today and tomorrow could not be bothered. And if I am typical of the group I’m in, then at some point we will all become suspicious of non-punishers doing well out of the system without contributing. Punishment could result.

        This is why I believe that cheating and free-riding cannot survive in small groups, they can only survive in modern large populations where there are more possibilities for concealment or evasion. If that view is correct, then as a late development these influences had little or no input into the development of the societies we see today, because when all groups were small groups, punishment was swift.

        Cheating and slacking might be fascinating for sociopaths, (for “sociopaths” read “gene-centrics”) but in evolutionary terms they are irrelevant compared to cooperation.

  • Steve Davis says:

    David, at the risk of sounding like a harping critic with too much time on his hands, I feel obliged to comment on another statement in the article.

    You said “The first thing that all theories of social evolution share in common can be called the fundamental problem of social life”, that being various forms of cheating.
    I can think of at least two arguments as to why that should not be so.

    The first is that it should be taken as given that there is nothing of value to be found in gene-centrism. The gene-centric view of social evolution is an ideology posing as science while based on assumptions for which no evidence is given, for example, the kinship requirement for altruism. It has no redeeming features, so to seek common ground with these fanatics is futile. I use the term “fanatics” deliberately. The near-hysterical response to the Wilson Nowak Tarnita paper some years back was that of a priesthood attacking heresy. One prominent critic even stated that the paper should not have been published. What’s next – book burning?

    We can see that there is no fundamental problem with social life because of its spectacular success in evolutionary history. A phenomenon so enduring, so awe-inspiring, so essential to development and progress cannot have a fundamental problem or it would have been selected out aeons ago. The problem the ideologues have with social life is that their false interpretation of evolution cannot explain the extraordinary preponderance of cooperation that we see in the natural world. They even refer to it, with uncharacteristic honesty, simply as the problem of cooperation.

    Which brings me to my second argument.

    The “fundamental problem of social life” from the gene-centric perspective is not cheating, which they see as quite natural, it’s social life itself with its dependence on cooperation, cooperation not fitting comfortably with their ideology of individualism.

    The fundamental mistake made by those who see cooperation as problem is that they agonise over the means by which cooperation could have come into existence in a natural world that in their view is constantly evolving through ruthless competition and natural selection. The simple answer to that is that cooperation preceded evolution.

    It is generally accepted that the first life forms came into existence when complex molecules combined into forms that could survive and replicate. When these even more complex forms began to perform the functions we consider to be life functions – metabolism homeostasis and replication, those functions were the result of cooperation among their constituent parts. Even the coming together of molecules to form a new entity can be seen as cooperation. And at this point evolution as we know it had not begun, as natural selection requires reproduction with variation on which it can act.

    Now it is true that cooperation underwent changes through evolution. It appeared in different forms, the most notable being the emergence of altruism, but the fact remains that cooperation is not a problem for sociality and cannot be a problem for sociality, as it is in fact the definition of sociality. The problem is in the eye of the beholder.

    David, you have the intelligence and courage to take on the fanatics, as you have demonstrated, but you will not win by fighting within a framework of their creation. Ditch their fascination with marginal issues that distract from the main game – evolution, and life itself, are all about cooperation.

  • David Sloan Wilson says:

    Steve–I’m with you that cooperation extends to the origin of life itself as groups of cooperative molecular interactions, but that’s not an argument against the fundamental problem of social life. It is still the case that proto-cells had to exist and it was selection between the cells that favored the cooperative interactions. I sometimes use the example of a no-cost public good to illustrate the need for between-group selection. Imagine a trait that benefits everyone in a group, including the actor, and at no cost to the actor. There is no way for this type to increase in frequency within the group, because the no-cost public good does not generate the fitness DIFFERENCES required for selection to operate. To find fitness differences, it is necessary to go to the group level. In the case of a no-cost public good, any variation among groups is sufficient because the traits selectively neutral within groups. The bottom line: You can’t use the pervasiveness of cooperation to argue against group selection. Cooperation is pervasive because group selection is pervasive. But, since cooperation isn’t everywhere, there are also many cases where within-group selection prevails against between-group selection.

    • Steve Davis says:

      David, you said; “To find fitness differences, it is necessary to go to the group level. In the case of a no-cost public good, any variation among groups is sufficient because the traits (are) selectively neutral within groups.”

      This topic was handled well by Robert Ardrey in The Social Contract. So well that Dawkins was forced to write The Selfish Gene in an unsuccessful attempt to discredit group selection. Most people who read Dawkins assume that he was successful in this, but even he admitted in the book that he was merely presenting an alternative view.

      Ardrey explained, (and this seems logical to me) that successful societies are those that strike the right balance between the conformity necessary for social cohesion, and the internal diversity necessary for the group to be adaptable to a changing environment. Because individuals have different capabilities, groups that make use of this through division of labour for example, become successful groups.

      The internal diversity of successful groups is also relevant to the discussion above about punishers and non-punishers.

      Getting back to Ardrey, The Social Contract was well researched and well written, but at the end he used his conclusions to indulge his personal political tastes, which was a great pity.

Leave a Reply

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