Beyond Demonic Memes

Why Richard Dawkins is Wrong About Religion

By David Sloan Wilson
Skeptic, July 4, 2007

Edited by Andy Ross

Richard Dawkins and I agree that evolutionary theory provides a powerful framework for studying religion. Evolutionists employ a number of hypotheses to study any trait. Is it an adaptation that evolved by natural selection? If so, did it evolve by benefiting whole groups, compared to other groups, or individuals compared to other individuals within groups? With cultural evolution, since cultural traits pass from person to person, perhaps they evolve to enhance their own transmission without benefiting human individuals or groups.

If the trait is not an adaptation, then it can nevertheless persist in the population for a variety of reasons. Perhaps it was adaptive in the past but not the present. Perhaps the trait is a byproduct of another adaptation. Finally, the trait might be selectively neutral and persist in the population by genetic or cultural drift.

Dawkins and I agree that these major hypotheses provide an excellent framework for organizing the study of religion. We also agree that the hypotheses are not mutually exclusive. Evolution is a complicated process, and all of the hypotheses might be relevant to some degree. Nevertheless, real progress requires determining which hypotheses are most important for the evolution of particular traits.

Few experiences are more thrilling for a biologist than to discover a complex adaptation. Myriad details that previously defied explanation become interpretable as an interlocking system with a purpose. Non-adaptive traits can also be complex, but the functional nature of a complex adaptation guides its analysis from beginning to end. Failing to recognize complex adaptations when they exist is as big a mistake as seeing them where they don't exist. Only hard empirical work can settle the issue.

Dawkins would probably agree with everything I have said so far. For religion, however, he argues primarily on behalf of non-adaptation. Perhaps religious impulses were adapted to the tiny social groups of our ancestral past, but not the mega-societies of the present. If current religious beliefs are adaptive at all, it is only for the beliefs themselves as cultural parasites on their human hosts. That is why Dawkins calls God a delusion.

To understand Dawkins' skepticism about the group-level benefits of religion, it is necessary to trace the history of "for the good of the group" thinking in evolutionary theory. Groups can be adaptive only if their members perform services for each other, yet these services are often vulnerable to exploitation by more self-serving individuals within the same group. Fortunately, groups of individuals who practice mutual aid can out-compete groups whose members do not.

According to this reasoning, traits that are "for the good of the group" require a process of between-group selection to evolve and tend to be undermined by selection within groups. Darwin was the first person to reason this way about the evolution of human morality and self-sacrificial traits in other animals. Unfortunately, many biologists during the first half of the 20th century uncritically assumed that adaptations evolve at all levels of the biological hierarchy without requiring a corresponding process of natural selection at each level. This Age of Na´ve Groupism ended thanks largely to two books: George C. Williams' 1966 Adaptation and Natural Selection and Richard Dawkins' 1976 The Selfish Gene.

In Adaptation and Natural Selection, Williams affirmed the logic of multi-level selection but then added an empirical claim: Even though between-group selection is theoretically possible, in the real world it is invariably trumped by within-group selection. Virtually all adaptations evolve at the individual level and even examples of apparent altruism must be explained in terms of self-interest.

Another theme developed by Williams was the concept of the gene as the fundamental unit of selection. In sexually reproducing species, an individual is a unique collection of genes that will never occur again. Individuals therefore lack the permanence to be acted upon by natural selection over multiple generations.

The concept of the gene as the fundamental unit of selection, for example, is identical to the concept of average effects in population genetics theory, which averages the fitness of alternative genes across all of the individual genotypes and environmental contexts experienced by the genes. A decade later, Dawkins played the role of interpreter for an even broader audience. Average effects became selfish genes and individuals became lumbering robots controlled by their genes.

In retrospect, it is hard to fathom the zeal with which evolutionists such as Williams and Dawkins rejected group selection and developed a view of evolution as based entirely on self-interest. In the first place, calling genes "replicators" and "the fundamental unit of selection" is no argument at all against group selection. The question has always been whether genes can evolve by virtue of benefiting whole groups and despite being selectively disadvantageous within groups.

Na´ve groupism is still a mistake that needs to be avoided, but between-group selection can no longer be categorically rejected. Claims for group selection must be evaluated on a case-by-case basis. Group selection can sometimes even be the dominating evolutionary force. A major transition occurs when selection within groups is suppressed, making it difficult for selfish elements to evolve at the expense of other members of their own groups. Selection among groups becomes a dominating evolutionary force, turning the groups into super-organisms.

Dawkins fully accepts the concept of major transitions, but he pretends that it doesn't require a revision in his ideas about group selection. Most important, he doesn't pose the question that is most relevant to the study of religion: Is it possible that human genetic and cultural evolution represents the newest example of a major transition, converting human groups into the equivalent of bodies and beehives?

Dawkins coined the term "meme" to think about cultural evolution. Consider genetic evolution by itself. When a new mutation arises, the total population consists of one group with a single mutant and many groups with no mutants. There is not much variation among groups in this scenario for group selection to act upon. Now imagine a species that has the ability to socially transmit information. A new cultural mutation can rapidly spread to everyone in the same group. Culture can radically shift the balance between levels of selection in favor of group selection.

In this context, the human major transition probably began early in the evolution of our lineage, resulting in a genetically evolved psychological architecture that enables us to spontaneously cooperate in small face-to-face groups. Defining, motivating, coordinating, and policing groups requires an elaborate system of proximate mechanisms, something akin to the physiological mechanisms of an individual organism. Might the elements of religion be part of the "social physiology" of the human group organism?

One of my projects is a collaboration with the psychologist Mihaly Csikszentmihalyi, who pioneered the Experience Sampling Method (ESM) which involves signaling people at random times during the day, prompting them to record their external and internal experience. I teamed up with Csikszentmihalyi to analyze some of his past studies from an evolutionary perspective.

These studies were performed on such a massive scale and with so much background information that we can compare the psychological experience of religious believers vs. nonbelievers on a moment-by-moment basis. We can even compare members of conservative vs. liberal protestant denominations, when they are alone vs. in the company of other people. On average, religious believers are more prosocial than non-believers, feel better about themselves, use their time more constructively, and engage in long-term planning rather than gratifying their impulsive desires. On a moment-by-moment basis, they report being more happy, active, sociable, involved and excited.

Hypothesis testing does not always require quantification and the other trappings of modern science. Darwin established his entire theory on the basis of descriptive information carefully gathered by the naturalists of his day. This kind of information exists in abundance for religions around the world and throughout history. It should be possible to use this information to evaluate the major evolutionary hypotheses.

In Darwin's Cathedral, I initiated a survey of religions drawn at random from the 16-volume Encyclopedia of World Religions, edited by the great religious scholar Mircia Eliade. By my assessment, the majority of religions in the sample are centered on practical concerns. New religious movements usually form when a constituency is not being well served by current social organizations in practical terms and is better served by the new movement. The seemingly irrational and otherworldly elements of religions in the sample usually make excellent practical sense when judged by what they cause the religious believers to do.

Jainism is one of the oldest and most ascetic of all the eastern religions and is practiced by approximately three percent of the Indian population. Jain ascetics filter the air they breathe, the water they drink, and sweep the path in front of them to avoid killing any creature. They are homeless, without possessions, and sometimes even fast themselves to death. How could such a religion benefit either individuals or groups in a practical sense?

It turns out that Jain ascetics comprise a tiny fraction of the religion, whose lay members are among the wealthiest merchants in India. Throughout their long history, Jains have filled an economic niche similar to the Jews in Western Europe, Chinese in Southeast Asia, and other merchant societies. The ascetics obtain their food by begging but their religion includes so many food restrictions that they can only accept food from the most pious lay Jain households. When they enter a house, they inspect the premises and subject the occupants to sharp questions about their moral purity before accepting their food. It is a mark of great honor to be visited but of great shame if the ascetics leave without food. In effect, the food begging system of the ascetics functions as a policing mechanism for the community. Jains make up one of the most conspicuously successful communities in India.

Jainism appears obviously dysfunctional based on a little information, such as the sight of an emaciated ascetic or beliefs that appear bizarre when taken out of context. The same religion becomes obviously functional based on more information. This is the kind of "natural history" information that enabled Darwin to build such a strong case for his theory of evolution, and it can be used to build an equally strong case for most of the enduring religions of the world.

Explaining religions as primarily group-level adaptations does not make them benign in every respect. The most that group selection can do is to turn groups into super-organisms. Like organisms, super-organisms compete, prey upon each other, coexist without interacting, or engage in mutualistic interactions. Sometimes they form cooperative federations that work so well that super-super-organisms emerge at an even larger spatial scale.

I share Dawkins' concern about other aspects of religions. Religions can be ruthless in the way that they enforce conformity within groups. Most alarming for a scientist, religions can be wanton about distorting facts about the real world on their way toward motivating behaviors that are adaptive in the real world. The problem with Dawkins' analysis is that he doesn't get the facts about religion right. At the moment, he is just another angry atheist, trading on his reputation as an evolutionist and spokesperson for science to vent his personal opinions about religion.

AR  This is a powerful critique by a man who knows what he's talking about. I think his case against Dawkins' polemic is valid.