Social Evolution Forum
FIND sef:
Human Cultures are Primarily Adaptive at the Group Level

The question of whether a given trait qualifies as an adaptation must be answered on a case-by-case basis. Nevertheless, a strong case can be made for species as primarily adapted to their environments. A similar argument can be made for human cultures as primarily adapted to their environments at the group level. The reason that human cultures are primarily adaptive at the group level is because the capacity for culture is itself a group-level adaptation. Establishing a consensus on human cultures as primarily adapted at the group level will enable human cultural diversity to be studied in the same way as biological diversity.

A recent conference on cultural evolution sponsored by the Ernst Strungmann Foundation affords an opportunity to assess the state of the art for this subject. Forty-five scientists were present, representing a melting pot of disciplines and nationalities, making the event as close to a world congress on cultural evolution as one can get.

The participants were divided into four groups that met over a period of five days to address the topics of “Cultural Evolution of the Structure of Human Groups”, “Cultural Evolution of Technology and Science”, “Cultural Evolution of Language”, and “Cultural Evolution of Religion”. The results will appear as a volume published by MIT press, but one conclusion emerged loud and clear: Human cultures are primarily adaptive at the group level.

This conclusion might seem shocking to some, given the long history of controversy over the topics of cultural evolution and group selection, separately and in conjunction with each other.  It might also seem suspect coming from me, a lifelong proponent of group selection. Nevertheless, not only did it represent the consensus view of all four groups—a claim that the participants are free to dispute—but it can also be justified by analogy with species as biological entities that are primarily adapted to their environments.  In this article, I will first outline the case for biological species and then make the parallel case for human cultures.

If I am correct, then the consensus that emerged at the Ernst Strungmann forum represents a watershed event for the study of cultural evolution as a whole. In the past, there has been a profound lack of agreement on whether cultural change counts as an evolutionary process, the degree to which cultural traits are adaptive, the processes by which they become adaptive, or the unit(s) of adaptation, which might be individuals, groups, or the cultural traits themselves. If a consensus can be reached on human cultures as primarily adaptive at the group level, then the future study of cultural evolution can start to build upon this foundation, rather than questioning whether it can serve as a foundation at all.

Why Biological Species are Primarily Adapted to Their Environments

When evolutionary biologists study any given trait in any given species, they must test a number of major hypotheses. Is the trait an adaptation that evolved by natural selection? If so, did it evolve by virtue of increasing the fitness of genes relative to other genes within the same organism, individuals relative to other individuals within groups, or groups relative to other groups relative to the total population? If it is not an adaptation, then why does it persist in the population? Was it adaptive in the past but not the present? Is it a byproduct of another adaptation, in the same way that a spandrel is a byproduct of an arch (Gould and Lewontin 1979)? Did it increase in frequency by virtue of being located close to a beneficial mutation (Hedrick 1982)? Or is it a neutral trait that drifted into the population without having any effect on survival and reproduction?

In the analysis of single traits, there is nothing privileged about an adaptationist hypothesis; all of the hypotheses must be evaluated on a case-by-case basis. Nevertheless, we can still robustly say that species are primarily adapted to their environments. Pick any species, study the traits that are required for it to survive and reproduce in its environment, and the list of traits will be very impressive indeed. We can begin with the processes required for life to exist in any species. Then we can proceed to the long list of traits required to survive and reproduce in any particular environment.

Any species could serve as an example, but I will choose the hemipteran insect Aquarius remigis, commonly called the water strider. It is adapted to live on the surface of water, where it scavenges and preys on other invertebrates. The vast majority of insects do not have the traits required to move on the surface of water, which is why they become prey for water striders when they inadvertently fall in. Water strider feet are so hydrophobic that they can support the weight of 16 water striders and their ultra-structure is being studied to create waterproof fabrics for human use (Feng et al. 2007). Water striders use ripples on the surface of water as a source of information, in the same way that we use sound waves traveling through the air. They also use ripples as a medium of communication with each other (Wilcox 1979).  A short list of adaptations required to survive and reproduce include the avoidance of predators, overwintering, dispersal, competition with members of the same sex, and mating with members of the opposite sex (e.g., Preziosi and Fairbairn 1996, 2000) . It becomes mind-boggling that so many adaptations can be packed into such a small creature, as I recount for a general audience in a chapter of my book The Neighborhood Project  (Wilson 2011) titled “The Parable of the Strider”, where details and additional references can be found.

Every species has a similar story to tell, right down to microbes and viruses. Surviving and reproducing is not a simple matter–the minimal set of required adaptations is very extensive indeed. Thus, the study of any given biological species can begin with the knowledge that it is primarily well adapted to its environment; otherwise it would not be there. This does not mean that each and every trait is adaptive, or that the adaptations are easy to identify. There will be examples of drift, hitchhiking, and environmental mismatch. Most adaptations produce byproducts. A given adaptation hypothesis might be rejected in favor of another adaptation hypothesis, in addition to non-adaptation hypotheses. That is why all of the major hypotheses must be evaluated for any given trait on a case-by-case basis.  Nevertheless, the overall inquiry is guided by the search for functional organization, which is justified by the fact that the species must be impressively adapted to its environment to even be there.

Biological species are primarily adaptive, but not necessarily at the group level. The water strider provides an example of a species that is primarily adaptive at the individual level. Most of its adaptations, such as its hydrophobic feet, evolve by virtue of causing individuals to survive and reproduce better than other individuals in their immediate vicinity. Considerations of “for the good of the group” or “for the good of the species” need not be invoked.  An important exception concerns the mating behavior of males, which differ greatly in their aggressiveness toward females. In a series of experiments, Omar Eldakar and his colleagues showed that aggressive males outcompete docile males in their immediate vicinity, but that groups with docile males are more productive than groups with aggressive males—the classic group selection scenario. Conditional movement creates enough variation among groups for docile males to be maintained in the population by between-group selection, despite their selective disadvantage within groups (Eldakar et al. 2010). This example underscores the fact that the level(s) of selection for any particular trait must be determined on a case-by-case basis for any given species. Water striders are primarily adapted to their environment at an individual level, with some exceptions.

Other species are primarily adaptive at the group level, notably eusocial insect species such as wasps, ants, bees, and termites (Hollbobler and Wilson 2008) . In these species, most traits evolve by virtue of increasing the fitness of colonies relative to other colonies, not by virtue of increasing the fitness of individuals relative to other individuals within the colonies. As a result, the colonies become highly cooperative units, even qualifying as “superorganisms”. Some traits do evolve by within-colony selection in the eusocial insects, but these tend to disrupt the functional organization of the hive (Ratnieks et al., 2006). A chapter of The Neighborhood Project titled “The Parable of the Wasp” serves as a companion to “The Parable of the Strider” to stress the comparison between levels of selection for a general audience.

It is important to stress that my multilevel account can also be framed in terms of inclusive fitness theory and other theoretical frameworks such as selfish gene theory. These frameworks should be regarded as different “languages”, “perspectives”, or “accounting methods” that are inter-translatable, a principle called “equivalence” that I have described in a previous Social Evolution Forum (Wilson 2012). Once we appreciate the concept of equivalence and cultivate an ability to translate between frameworks, much of the controversy that seems to surround the topic of multilevel selection evaporates.

To summarize, virtually all research on biological species takes an impressive degree of overall functional organization for granted, even if any given trait must be evaluated on a case-by-case basis. This is what I mean by saying that species are primarily adapted to their environments.

The Parallel Case for Human Cultures

The statement “Surviving and reproducing is not a simple matter–the minimal set of required adaptations is very extensive indeed” applies as forcefully to human cultures as to biological species. Robert Boyd and Peter Richerson have emphasized this point for cultures that live in challenging climates such as the arctic (Richerson and Boyd 2005, Boyd et al., 2011). The knowledge required to build a sea kayak or clothing protective against frigid weather, when all of the tools must be made in addition to the objects made with the tools, is mind boggling when one pauses to consider it. Yet, these are only two items in a long list that is required to survive and reproduce in the arctic; others include building a shelter, hunting and gathering techniques, navigating long distances, social conventions within the group, and conduct toward other groups.  The adaptedness of any given belief or practice must be determined on a case-by-case basis, but the basic adaptedness of the culture as a whole can be taken for granted; otherwise it would not be there.

This point can be made as strongly for human cultures as for biological species on the basis of existing information. It is best appreciated by stepping back to consider the raw fact that while we are genetically a single species, our cultural diversity enables us to occupy all climatic zone and to occupy hundreds of ecological niches–an adaptive radiation comparable to the dinosaurs, birds, and mammals (Pagel and Mace 2004).

What’s new, going beyond the raw fact of cultures adapted to their environments, is our emerging knowledge of how we evolved our capacity for cultural evolution. It all revolves around cooperation. In other primate species, members of the same group cooperate to a degree but they are also each other’s chief rivals for status, resources, and mates. The lack of cooperation and trust limits the ability of group members to learn from each other in a cumulative fashion and to transmit their learned behaviors across generations.

Small human groups are much more cooperative than primate groups, thanks largely to the their ability to suppress bullying and other self-serving behaviors, a kind of social organization that evolutionary anthropologist Christopher Boehm calls “reverse dominance” (Boehm 1993, 1999, 2011).  If members of a group can’t succeed at the expense of each other, then their main avenue of success is to succeed as a group, compared to less cooperative groups. The competition might be direct, as in warfare, or indirect, as in producing more offspring that emigrate to other groups or form new groups. In the language of multilevel selection theory, reverse dominance suppresses within-group selection and magnifies variation among groups (especially through the establishment and enforcement of norms), making between-group selection the dominant evolutionary force.

Hunter-gatherers cooperate to raise their children, to hunt and gather, to defend themselves from predators, to regulate their social interactions within groups, to trade with other groups, and to raid and defend themselves against other groups. We are familiar with this list of physical activities. More novel is to think of the mental activities required for cumulative cultural evolution as forms of cooperation, starting with an awareness of the interests of others that is far more highly developed in humans than in our closest primate relatives (Tomasello 2009). Even something as simple (for humans) as a shared symbol is a communal activity. In short, the entire package of traits that are so distinctively human, such as cooperative physical activities, cumulative cultural evolution, and our capacity for symbolic thought, are all examples of cooperation that evolved due to the shift in the balance between levels of selection. Something close to Christopher Boehm’s reverse dominance scenario came first, and everything else followed.

Symbolic thought is an especially important capacity for open-ended cultural evolution. There is a nearly infinite variety of imaginary worlds, just as there is a nearly infinite variety of genotypes in sexually reproducing species. Each imaginary world motivates a suite of behaviors, creating a “symbotype”-phenotype relationship similar to the familiar genotype-phenotype relationship.  In short, symbolic thought provides a cultural inheritance system that rivals the genetic inheritance system in its combinatorial diversity (Jablonka and Lamb 2006, Wilson et al., 2013). This goes a long way toward explaining how our ancestors were able to expand their range to include the entire planet, adapting to all climates and hundreds of ecological niches, in just a few tens of thousands of years. Emile Durkheim (1912/1995 p 233) wrote that “at every moment of its history, social life is only possible thanks to a vast symbolism.” Modern evolutionary science is proving him right.

The origin of agriculture resulted in a positive feedback loop between the production of resources and the scale of society that continues to this day.  The increasing scale of society is also a matter of multilevel selection. Cultures vary widely in their ability to function at a large scale. The most cooperative spread by conquest and/or productive superiority. They are also copied on the basis of their success. Yet, all societies are vulnerable to factionalism and self-serving strategies that disrupt cooperation at the societal level. Recorded history provides a detailed fossil record of multilevel cultural evolution, as documented by books such as War and Peace and War, by Peter Turchin (2005).

Perspectives on culture are incredibly diverse, from Durkheim’s functionalism to Derrida’s postmodernism. Even when we restrict our attention to modern evolutionary perspectives, evolutionary psychology pays scant attention to transmitted culture (Wilson 2010) and the concept of memes portrays cultural traits as free agents, as likely to harm as to help their human hosts (Blackmore 1999). It is therefore remarkable that the 45 scientists at the Strungmann Forum were in such agreement on the account of human cultural evolution that I have outlined above (audio interviews with members of the four groups are available at  The best way to demonstrate their consensus is to let them speak for themselves. Here are some excerpts from group one’s preliminary report, which was circulated and discussed on the last day.

What we call small-scale societies are still huge cooperative endeavors compared to the scale of cooperation in other vertebrates.

The identification of a minimal set or sets of predispositions necessary for small-scale societies to arise then gives us building blocks necessary for thinking about the cultural evolution of large-scale societies.

We now turn to three requirements/necessary conditions that…could produce SSS (Small-Scale Society) cooperation. These are (1) Increasing returns to scale in group size; (2) Control of defectors and (3) Cultural group selection/assortativity.

Increasing returns to scale is a prerequisite for cooperation to evolve, but essentially, all this means is that there should be some benefits to cooperation…The hard problem is how does cooperation evolve, given that exploiters will appropriate these benefits causing the cooperation to dissolve.

The products of human cooperation have to be protected by monitoring and enforcement systems. These systems start with norms, and the importance of norms is recurrent throughout this chapter.

At this point we are armed with some idea of the behavioral predispositions that are necessary for the evolution of small-scale society and cooperation. The cultural evolutionary perspective then allows us to hypothesize how those features can be exploited in the transition from small- to large-scale societies.

We can identify two evolutionary “engines” in (1) cultural group selection and (2) within-group processes which will produce change in social structure and allow transitions between levels of complexity.

Cultural Group Selection (CGS) where between-group competition favours societies that are more effective in production and/or warfare. Larger and more complex societies with more efficient social institutions can arise through such between-group selection. Between-group selection can result through warfare…,through differential population growth…, through immigration into more successful societies…, and through adopting the social institutions of successful groups.

Social change can also arise endogenously, from within-group processes that generate variation. Endogenous change can result from prosocial preferences like a regard for equitable or fair or parochial outcomes that has resulted from a longer history of cultural group selection. Such preferences, combined with abilities for persuasion, leadership, or deliberation—can allow societies to adopt norms that are consistent with these preferences. Democracies, or jury systems, may be the result of preferences shaped by cultural group selection (like fairness and peer sanctioning, respectively). It is important to note that on longer time-scales, these institutions will persist only if they lead groups that have adopted these social arrangements to fare better than other groups. However, on shorter time scales, some of the change that we see in human societies can be the result of people tinkering with their social institutions in accordance with their preferences, rather than due to between-group selection itself.

Notice that even the within-group processes discussed in the last passage are largely the result of psychological processes shaped by past between-group selection and are ultimately winnowed by ongoing between-group selection over the long term.  

The Case of Religion

The artifacts that we associate with technology, from the first hand tools to hand computers, are straightforwardly utilitarian. Many artifacts are used by individuals for their personal benefit, but a whole culture is required for their evolution. Language is likewise eminently utilitarian and a group-level process, even when it is used by individuals for their personal gain. Thus, it is not surprising that the second and third groups largely agreed with the group-level functional account of the first group.

Religion is another matter. It fascinates the scientific imagination precisely because it does not seem utilitarian. How can all that hocus-pocus increase the fitness of either individuals or groups? Durkheim thought otherwise. His definition of religion (“a unified system of beliefs and practices relative to sacred things…which unite in one single moral community called a Church, all those who adhere to them”) emphasized its great “secular utility”, as he put it (Wilson 2002). Yet, other theories abound that portray religions as primarily dysfunctional, such as sociologist Rodney Stark’s economics-inspired theory that religions involve the invention of Gods with whom we bargain for that which we can’t have (e.g, Stark and Bainbridge 1987).

The modern study of religion from an evolutionary perspective began at the turn of the 21st century with books such as Religion Explained (Boyer 2001), In Gods We Trust (Atran 2002), Breaking the Spell (Dennett 2006), The God Delusion (Dawkins 2006), and my own Darwin’s Cathedral  (2002). The authors agreed on the basic hypotheses that needed to be tested about religion, but differed sharply on the ones that they favored. Some interpreted religions as primarily adaptive for human individuals and groups, while others interpreted them as primarily byproducts or as parasitic memes that are as likely to be harmful as helpful for their human hosts.

Research on religion from an evolutionary perspective has burgeoned since the appearance of these books and the Ernst Strungmann Forum provided an opportunity to assess progress. There was a strong consensus among members of the religion group that the major elements associated with religion, such as conterfactual beliefs (including but not restricted to supernatural agents), a sense of the sacred, and costly activities (including but not restricted to rituals), are typically functional at the level of the religious community, despite first appearances. Essentially, Durkheim was right in basic outline, even though the modern study of functionalism differs from the earlier tradition of functionalism in many respects.

As with the study of biological species, all of the major hypotheses are still relevant and must be evaluated on a case-by-case basis for any given element of a religious system, but the overall inquiry can be guided by the search for functional organization. The study of byproducts provides a good example. Just as many biological adaptations are derived from previous adaptations or byproducts (e.g., the bones in our ears are derived from reptile skull bones), many elements of religion are derived from biological adaptations that evolved without reference to religion (e.g., kinship terminology or the tendency to infer agency from events).  They quality as byproducts as far as genetic evolution is concerned, but the question of whether they qualify as adaptations vs. byproducts in cultural evolution requires a separate determination. Enough empirical research has been conducted, using a variety of cross-validating methods, to conclude that the adaptations and byproducts of the past have been woven into religious systems that are impressively functional at the group level (Atran and Henrich 2010).  The hypothesis that religion writ large is a functionless byproduct has been authoritatively rejected.

The study of cultural parasites provides another example. It is theoretically possible for a cultural trait to spread, disease-like, without benefiting either human individuals or groups. It is a legitimate hypothesis that must be evaluated on a case-by-case basis. A religion called Millerism provides a candidate example (Numbers and Butler 1993), as I describe for a general audience in a chapter of The Neighborhood Project titled “The Natural History of the Afterlife”. Miller was a Baptist farmer in New York State during the 19th century who became convinced that the return of Jesus was imminent.  The movement he started would have remained localized and unknown to history, were it not for the advent of print media such as newspapers and pamphlets, which caused Miller’s “meme” to spread viral-like around the world. Thousands of people made decisions in preparation for the second coming that were not useful for them, either as individuals or groups. The movement quickly collapsed in boom-and-bust fashion, which is typical of disease epidemics. Millerism therefore qualities as a good candidate for an empirically documented cultural parasite.

The story does not end there, however. After the collapse of Millerism, a number of splinter groups clung to the belief that Christ’s return was imminent and modified some of their other beliefs to accommodate the fact that he did not return on the day predicted by the Millerites. One of these splinter groups became 7th Day Adventism, one of the fastest growing religions in the world (Numbers 1976). Unlike Millerism, 7th Day Adventism harnesses the psychologically motivating belief in Christ’s imminent return to sustainable practices on earth, at both the individual (e.g., healthful eating habits) and group (e.g., schools and hospitals in addition to churches) levels.

This example illustrates how history provides a detailed fossil record of cultural evolution that allows the major hypotheses to be evaluated on a case-by-case basis, similar to the branch of the biological sciences known as paleontology. By now, many historical examples of religion have been analyzed from an evolutionary perspective, using quantitative in addition to qualitative methods (e.g. Voland and Schiefenhövel 2009, Wilson 2005). The evidence for the group-level functionality of most enduring religions is very strong, with dysfunctional examples such as Millerism constituting a minority.

Multilevel selection theory predicts an ongoing tension between beliefs and practices that enable the group to function as a whole with beliefs and practices that benefit some members of the group at the expense of others. This tension has a temporal component; groups that start cooperative can fall apart as self-serving strategies take root and spread from within. The conflict between levels of selection is amply preserved in the historical record for religions, as for non-religious social institutions (Wilson 2002). The conclusion that most enduring religions are functional at the group level is therefore not a claim that within-group selection never happens, but rather that between-group selection is a strong force in religious cultural evolution, thanks in large part to mechanisms that effectively suppress factionalism and self-serving strategies within the group.

The historical evidence is supported by evidence from different sources. A growing psychological literature demonstrates that when people are primed with religious cues (e.g. unscrambling words such as “church” compared to unscrambling non-religious words), they become more cooperative (Sharriff and Norenzayan 2007).  Field and laboratory studies show that some of the most costly aspects of religion, such as time-consuming and pain-inflicting rituals, can be understood as credibility-enhancing displays (Bulbulia and Sosis 2011,Henrich 2009).

Another important conclusion that emerged from the religion group was that most of the major elements associated with religion also extend to other cultural systems. A theory is needed to explain why we are so prone to accept counter-factual beliefs of all sorts, not just belief in supernatural agents. Likewise for a sense of the sacred and costly credibility-enhancing displays. A well-known Atlantic Monthly article by Harvey Cox titled “The Market as God” (Cox 1999) demonstrates how a cultural system that few people associate with religion—free market capitalism—has all the trappings of religion.

In short, group 4 (on religion) affirmed the conclusions of group 1 (on the structure of human groups) as strongly as group 2 (on technology and science) and group 3 (on language). Human cultures are primarily adaptive at the group level.

Why a Consensus Matters

A consensus, by definition, involves leaving some alternatives behind and concentrating on a reduced set of possibilities. A consensus is never final. Group selection provides an outstanding example of a possibility that was excluded by a consensus that formed in the 1960’s, only to force its way back into consideration, as the Ernst Strungmann Forum made abundantly clear.

Even though a consensus is never final, it is still important—even essential—for scientific progress.  A lack of a consensus, when one is warranted, is a kind of paralysis that prevents a field of inquiry from moving forward. Consider the diversity of opinion about culture that currently exists across academic disciplines:

• A dominant position throughout the 20th century, which is still common in some quarters, is to regard the study of culture as outside the orbit of evolutionary theory altogether.

• The study of cultural evolution has a complex history in anthropology, leading to many different conceptions in the minds of different anthropologists. The tradition of functionalism, for example, was seldom associated with evolution, as strange as that might seem against the background of modern evolutionary theory.

• Evolutionary theory in the biological sciences is highly gene-centric. The idea that evolution requires an inheritance system, and that other inheritance systems exist in addition to the genetic inheritance system, needs to become more prominent within mainstream evolutionary biology (Jablonka and Lamb 2006).

• When we restrict our attention to contemporary evolutionists interested in human evolution, the version of evolutionary psychology associated Cosmides and Tooby (1992) pays scant attention to transmitted culture (as opposed to what they call evoked culture), as previously mentioned.

• When we restrict our attention to contemporary evolutionists interested in cultural evolution, the consensus reached at the Ernst Strungmann forum is an important advance over other conceptions of cultural evolution, such as memes as primarily atomistic free agents, religion writ large as primarily maladaptive, and so on.

The study of culture becomes paralyzed when it means so many different things to different people. Reaching a consensus that human cultures are primarily adaptive at the group level will enable human cultural diversity to be studied in the same way as biological diversity—and will result in the same kind of scientific progress.

Literature Cited

Atran, S. (2002). In  Gods we Trust: The evolutionary landscape of religion. Oxford: Oxford University Press.

Atran, S., & Henrich, J. (2010). The Evolution of Religion: How cognitive by-products, adaptive learning heuristics, ritual displays, and group competition generate deep commitments to prosocial religions. . Biological Theory: Integrating Development, Evolution, and Cognition, 5, 18–30.

Blackmore, S. (1999). The meme machine. Oxford,UK: Oxford University Press.

Boehm, C. (1993). Egalitarian society and reverse dominance hierarchy. Current Anthropology, 34, 227–254.

Boehm, C. (1999). Hierarchy in the Forest: Egalitarianism and the Evolution of Human Altruism. Cambridge, Mass: Harvard University Press.

Boehm, C. (2011). Moral Origins: The Evolution of Virtue, Altruism, and Shame. New York: Basic Books.

Boyd, R., Richerson, P. J., & Henrich, J. (2011). The cultural niche: why social learning is essential for human adaptation. Proceedings of the National Academy of Sciences, 108, 10918–10925.

Boyer, P. (2001). Religion Explained. New York: Basic Books.

Bulbulia, J., & Sosis, R. (2011). Signalling theory and the evolution of religious cooperation. Religion, 41(3), 363–388. Retrieved from

Cox, H. (1999). The Market as God. Atlantic Monthly, March 1999.

Dawkins, R. (2006). The God Delusion. Boston: Houghton Mifflin.

Dennett, D. C. (2006). Breaking the spell: religion as a natural phenomenon. New York: Viking.

Durkheim, E., & Fields, K. E. (n.d.). The elementary forms of religious life. New York: The Free Press.

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

Feng, X. Q., Gao, X. F., Wu, Z. N., Jiang, L., & Zheng, Q. S. (2007). Superior water repellency of water strider legs with hierarchical structures: Experiments and analysis. Langmuir, 23, 4892–4896.

Gould, S. J., & Lewontin, R. C. (1979). The spandrels of San Marco and the panglossian paradigm: A critique of the adaptationist program. Proceedings of the Royal Society of London, B205, 581–598.

Hedrick, P. W. (1982). Genetic Hitchhiking: A New Factor in Evolution? Bioscience, 32, 845–853.

Henrich, J. (2009). The evolution of costly displays, cooperation and religion. Evolution and Human Behavior, 30(4), 244–260. Retrieved from

Holldobler, B., & Wilson, E. O. (2008). The Superorganisms. New York: Norton.

Jablonka, E., & Lamb, M. J. (2006). Evolution in Four Dimension: Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of Life. Cambridge, MA: MIT Press.

Numbers, R. L. (1976). Prophetess of Health: A Study of Ellen G. White. New York: Harper & Row.

Numbers, R. L., & Butler, J. M. (Eds.). (1993). The Disappointed: Millerism and Millerarianism in the Nineteenth Century (p. 280). Knoxville, TN: Univ. of Tennessee Press. Retrieved from

Pagel, M., & Mace, R. (2004). The cultural wealth of nations. Nature, 428, 275–278.

Preziosi, R. F., & Fairbairn, D. J. (1996). Sexual size dimorphism and selection in the wild in teh waterstrider Aquarius remigis: body size, components of body size,and mating success. Journal of Evolutionay Biology, 9, 317–336.

Preziosi, R. F., & Fairbairn, D. J. (2000). Lifetime selection on adult body size and components of body size in a water strider: opposing selection and maintenance of sexual size dimorphism. Evolution, 54, 558–566.

Ratnieks, F. L. W., Foster, K. R., & Wenseleers, T. (2006). Conflict resolution in insect societies. Annual Review of Entomology, 51, 581–608.

Richerson, P. J., & Boyd, R. (2005). Not by genes alone: how culture transformed human evolution. Chicago: University of Chicago Press.

Shariff, A. F., & Norenzayan, A. (2007). God is watching you: supernatural agent concepts increase prosocial behavior in an anonymous economic game. Psychological Science, in press.

Stark, R., & Bainbridge, W. S. (1987). A theory of religion. New Brunswick, N.J.: Rutgers University Press.

Tomasello, M. (2009). Why We Cooperate. Boston: MIT Press.

Tooby, J., & Cosmides, L. (1992). The psychological foundations of culture. In J. H. Barkow, L. Cosmides, & J. Tooby (Eds.), The adapted mind: evolutionary psychology and the generation of culture (pp. 19–136). Oxford: Oxford University Press.

Turchin, P. (2005). War and Peace and War. Upper Saddle River, NJ: Pi Press.

Wilcox, R. S. (1979). Sex Discrimination in Gerris remigis: Role of a Surface Wave Signal. Science, 206, 1325–1327.

Wilson, D. S. (2002). Darwin’s Cathedral: Evolution, Religion and the Nature of Society. Chicago: University of Chicago Press.

Wilson, D. S. (2005). Testing Major Evolutionary Random Sample, 16(4), 419–446.

Wilson, D. S. (2010). Learning from the Immune System about Evolutionary Psychology. In A. Andrews & J. Carroll (Eds.), Evolutionary Review. Albany: SUNY Press.

Wilson, D.S. (2012). Clash of Paradigms: Why Proponents of Multilevel Selection Theory and Inclusive Fitness Theory Sometimes (But Not Always) Misunderstand Each Other. Social Evolution Forum, July 13, 2012.

Wilson, D. S. (2011). The Neighborhood Project: Using Evolution to Improve My City, One Block at a Time. New York: Little, Brown.

Wilson, D. S., Hayes, S. C., Biglan, A., & Embry, D. (2012). Evolving the Future: Toward a Science of Intentional Change. Behavioral and Brain Sciences (under review).

David Sloan Wilson is President, Evolution Institute ( )
SUNY Distinguished Professor, Departments of Biology and Anthropology, Binghamton University, Binghamton, New York, 13902


Join the discussion


  1. Tim Tyler says:

    Haidt once claimed: “human beings are 90 percent chimp and 10 percent bee”. That’s probably an enthusiastic over estimate. Come back in a few decades to see what a real cultural superorganism looks like.

  2. John Lilburne says:

    Another facsinating topic is the co- evolution of culture and genes . A good example is the evolution of dairy farming and lactose tolerance amongst northern europeans. Harry harpending and Cochran points out that rather than evolution slowing (on a genetic level) since the advent of agriculture it has accelerated.

    A second thought is how religion adapts and leads to environment. Calvinism in the early USA was highly authoritarian. All markets were to highly regulated so that the ingroup inequality would be low (and outgroups.. who cared) However in the 17th century the theology completely reversed and became radically pro market. This formed the basis of the radical free markets dominant in the USA. So the relationship between markets and religion is not so strange.

  3. Roy Niles says:

    All organisms belong to a group, generally a family within a larger culture. All organisms communicate precisely because they need to learn their species’ survival lessons from the group with lessons preserved, one way or another, by the culture. Learning is what spreads the benefits of individual experiences. All evolution is the proximate result of the entity involved reacting strategically to its experience. The fitness theories that you continue to come up with are outdated. The genetic bases you attribute to traits such as altruism or selfishness are outdated; these traits are essentially learned. Organisms that are the most cooperative within their groups will generally be the most competitive when separate groups at any level need to compete for resources. The strategies involved are different for each species and each group to some extent within that species. They are learned instincts supplemented by learned cultures and day to day learning from experience.
    There’s much more to the story and to the way these strategies have had to develop, but in short, you are still tilting at windmills with these group selection hypotheses. Group competitions, etc., offer different lessons, but their multilevel effects are also much more varied within and across species than your models have influenced you to allow for.
    You will of course ignore all of this as you are stuck within the limits of your academic rules and shibboleths. Too bad.

    • Tim Tyler says:

      Culturally-transmitted and individually-learned traits have fitnesses too. Meme fitnesses depend on competing memes and the social environment. Individually learned ideas have fitnesses too – which depend on their competition with other ideas within a mind. Learning-based theories don’t compete with fitness-based theories. Individual and social learning are both themselves intrinsically evolutionary – and have their own associated conceptions of fitness.

  4. Mato Nagel says:

    Well, you are approaching fauceir theory 🙂

  5. Mato Nagel says:

    My comment is published it on my website.
    (link to

  6. Juan Alfonso del Busto says:

    We are living an amazing paradox: Dawkins, the first proponent of evolutionary cultural evolution based on cultural replicators (memes), has a personal dislike for religion as a cultural phenomenon and therefore he doesn´t want to admit that his theory leads to “religion as an adaptation” at whatever level. On the other hand we have D. Wilson and others who have fully grasped the relevance and reach of this concept (specially regarding religion), but they are reluctant to use the term coined by their oponnent: “meme”.

    It is understandable but this issue is slowing the pace a bit. Perhaps memes are a metaphor or perhaps they are not. But this is not the point. The point is that cultural evolution, including the evolution of religions, show the dynamics of some kind of selfish replicators analogous to the genes. In other words: Cultural evolution (namely evolution of religions) show the landmarks of the selfish meme.

    As Wilson points out there are many examples of religions who were clearly maladaptive, and which underwent an epidemiological dynamics of growth and decine. These examples are the evidence of the selfish memes as parasites that are transmited horizontally, and that don´t require to be adaptive to their hosts. On the other hand we have mutualistic memes that are transmited vertically an which are much more likely to be adaptive for the hosts (individuals or groups). As a matter of fact, religions seem to start as maladaptive horizontal selfish memes and slowly evolve into more adaptive vertical selfish memes. This is what happens with parasites that coevolve with the host and evolve to become less harmful and eventually even beneficial. This is the only gene-meme analogy that must be taken into account: the selfishness replicator model. The rest is non-relevant. It is not important if memes ara atomistic or are stored in the brains or in artifacts.

    “Meme”, more than Boyd &Richerson “cultural variant”, is the right word for adressing cultural traits that are transmitted.

    • tmtyler says:

      Amish families average around seven children. The effect of some religions on human fertility is obvious. I don’t think I ever heard Dawkins or Dennet deny this. Both Dennet and Dawkins regard religion as an adaptive phenomenon in the cultural realm. Dawkins gives his adaptive account of religion in “Virus Of The Mind” (available in essay and video formats). He acknowledges that, possibly, religions might be adaptive to their hosts (comparing them to the effect of placebos on stress relief). However, he says he doesn’t believe genetic theories, are “big” enough to explain the widespread nature of religion – and that a successful explanation needs to invoke memetics.

    • Tim Tyler says:

      It is true that memes are confined to culture. However, we already have a science of heredity: genetics. If you also want a more general science of copied information, the best way is surely generalising from genetics – e.g. see my “Generalised genetics” article.

      Complaining that memes are not hierarchical is like complaining that computer memory is not hierarchical. It’s true – but it’s not a limitation. Memetic linkage is probably the most natural way of exploring the geometry of the resulting spaces. They go beyond hierarchies and include webs and partially-overlapping entities – so you don’t really want an explicitly hierarchical model.

  7. tmtyler says:

    This article quotes nine conference participants. None of the quotes seems to me to offer much support for the thesis that: “Human Cultures are Primarily Adaptive at the Group Level”. If the participants reached a consensus, perhaps it was about something else.

    • Peter Turchin says:

      My take on the consensus at the conference is in this post:

      link to

      In one sentence,

      there was a great degree of consensus among the participants that cultural group selection is the only general mechanism for explaining how human sociality evolved that really has the logical coherence and empirical support.

  8. David Sloan Wilson says:

    Thanks to everyone for their comments. Some quick replies, to supplement my general reply to the commentaries that will be posted soon:

    To John Lilburne: Calvinism in its early stages was not authoritarian. The church was subordinate to the civic government, Calvin was a leader among equals, and the elders of the church had to be approved by the people they governed. See my chapter on Calvinism in Darwin’s Cathedral.

    To Roy Niles: Multilevel selection applies to both genetic and cultural evolution and cultural evolution includes learning. Where did you acquire the impression that I restrict group selection to genetic evolution? Regardless of whether a trait is learned or innate, it is possible to determine where the fitness differences exist in a nested hierarchy of units.

    To Tim Tyler: For once I agree with you about your comments following Roy Niles!

    To Juan Alfonso del Busto: The term “meme” has acquired multiple meanings, so we need to be mindful of equivalence (see my SEF essay titled “Clash of Paradigms” and my general reply to commentaries of this essay). You are using a broad definition that is synonymous with “any cultural trait”. According to this definition, memes can evolve without benefitting their hosts in any sense, they can evolve by benefiting their individual hosts compared to other individuals within the same group, or by benefiting entire groups, compared to other groups. Distinguishing among these categories recreates multilevel selection theory within meme theory.

    To Peter Turchin: We agree that “a sea change” is occurring. One point that you make in your blog is that another conference (the NIMBios group) reflected the same sea change, even though there was little overlap among the participants of the two conferences.