I’m not sure that Boyer, Atran and Wilson regard their explanations as complementary. I have talked to all three of them about it. My feeling is that while all three of them understand that the explanations might be complementary, they prefer to believe that all the work is done by their preferred model. It’s not clear to me how one could decide this point in principle.
He refers to Scott Atran (In Gods We Trust), Pascal Boyer (Religion Explained) and David S. Wilson (Darwin’s Cathedral). Atran & Boyer are of very similar intellectual outlook, they’re cognitive anthropologists who synthesize psychological and evolutionary paradigms and apply them to anthropological questions. They were both part of Dan Sperber’s salon during the early 1980s in Paris, and work within a ‘naturalistic’ tradition of their discipline which attempts to decompose & reduce anthropological phenomena to make them analytically tractable. In contrast, David S. Wilson in Darwin’s Cathedral is by training an evolutionary biologist who ventures frequently into anthropological territory. All three overlap insofar as they employ methods from a range of fields to explore human social and anthropological questions, though Atran and Boyer emphasize psychology far more than Wilson. In my comments & posts on this weblog I’ve contended that these two models actually are complementary and analyze different levels of religion as a natural phenomenon.
Though Atran and Boyer have their differences (e.g., Boyer tends to place greater weight from what I can see in a conventional evolutionary psychological model of modular cognition) their outlook and thesis is roughly the same: religious belief emerges due to the structural nature of the human mind and how it processes information. The classic example is the tendency for humans to impart agency to inanimate objects (e.g., faces in clouds). This bias is something we all know about, and it likely serves an evolutionary purpose (i.e., a false positive is a lot less deadly than a false negative when it comes to identifying strange humans or animals). We can’t always constrain a mental utility to its proper domain, and in their most extreme manifestations they become pathological (e.g., schizophrenia). Nevertheless, the hypothesis is that the various cognitive functions which characterize the modal human entail a strong bias toward particular non-functional phenomena, a suite of habits, and outlooks which didn’t arise specifically to increase reproductive fitness. Consider for example our attitudes toward stories. This seems a human universal, and the reality is that no matter how original a storyteller is common themes, motifs and plot elements reoccur in all cultures. One might wonder often if stories are nothing more than the same tale with names shifted to give us a sense of novelty. My point is that though one can posit functional rationales behind the universality of stories (e.g., binding the tribe with a common mythos?) it seems to me that our attraction to particular forms of narration, those involving people in circumstances with which “we can relate,” which are memorable and arouse interest and fascination, derives from the propensities of our various cognitive competencies. We love gossip, and stories are an embellished form of gossip, and gossip is an activity which enables use to gain social information and smooth our relations with conspecifics. Similarly, though one can posit functional roles that religion plays within a community, the universal nature of most religions suggests that they are under strong cognitive constraint, so their origin might be due to the nature of human cognition, not the necessary behavioral morphs which increase fitness.
When I say the “universal nature of most religions” some of you might balk. After all, Islam and Hinduism are very different, no? Though ultimately I think Atran and Boyer have a hard time explaining religious variation because their evolutionary psychological model is predicated on modal cognitive architectures, they do have an answer proximately. In Theological Incorrectness D. Jason Slone makes the argument that Atran and Boyer allude to in their books: that on a fundamental cognitive level religious belief is basically the same, the god of the Hindus and Muslims is the same mental conception. So why the difference in formulation? Elite expositions of religious belief often have little relationship to the day to day mental conceptualization of the believers. Religious formula, such as the Nicene Creed, are easy to memorize, but their cognitive impact is minimal in relation to the “implicit mind” which scaffolds and frames many of our perceptions, interactions and rationalizations. The universality of devotionalist sects within the “great” religions attests to this common need to personalize the god. Religious elites amongst Christians and Muslims accept a personal God, while Hindus differ in their opinions, and Buddhists reject the idea. But the empirical reality is that most believers operationally act and worship as if a personal god of some sort exists. While monotheists tend to philosophize that their gods have very abstract and difficult to comprehend characteristics, the believers on the ground tend to focus their religious energies upon entities with whom a personal relationship is attainable, just as Hindus have their avatars who live amongst men, so Jesus is god in the flesh and Mary is the intercessor.
The god concept is not the sum totality of religion, but it seems to me to be a necessary precondition for the emergence of religion as we understand it, a supernatural set of beliefs compelling to individuals coupled with ritual practices and spanning communities and generations. Some “secular” movements clearly draw from the same cognitive wells, for example, Communism in its totalitarian phase. The miracles attendant upon the birth of Kim Jong Il (flowers blooming in winter?) are to me evidence that the North Korean dictatorship, though atheist, has some similarities to the god-kingships of the ancient past. Similarly, the cult of Stalin exhibited similar emotional valences to religious movements. During the Nazi period in Germany the SS officer corps began to shift toward a Neopagan religious sensibility where Adolf Hitler became a central savior figure. But in the end the North Korean dictatorship will likely fall, and Stalin & Hitler are dead. Non-supernatural movements which overlap with religions in some of their characteristics differ insofar as their gods do die, and so generation to generation passing of the belief is less likely as new quasi-gods arise to draw from the same emotional impulses.
But in any case, religion is more than belief in a god, even if that is necessary. Both Atran and Boyer address the importance of ritual and communal bonding. Though they put a reductionistic and cognitive slant on it, and can make a good case that the nature of mass religious events are constrained and shaped by our psychological biases, I think they basically have to admit that their understanding of the interplay between supernatural actors as mediators within a community bound by ritual has to give some nod to functionalism. David S. Wilson in his work argues strongly that the ultimate ends of religious practice and belief are functional, insofar as they exist to increase group and individual fitness. Since Wilson believes that between group competition irreducible to the level of the indivdiual plays an important role in human evolutionary history it stands to reason that he sees the existence of gods and other supernatural motifs as ancillary and almost incidental to the cohesive affect of religion. It is rather difficult for me to tease apart the factors here, for while I think both Atran and Boyer imply some level of functionalism, I don’t see that Wilson has explained why religion is as it is. In Evolution for Everyone Wilson asserts that Buddhism offers the hope that a non-theistic (i.e., non-supernatural) religion can be created which has all the positive group level effects. I have argued before that this misunderstands Buddhism, the success of the Buddha dharma is to some extent a function of its transformation into a theistic religion from a reformist philosophy (though the reformist philosophy always can and does emerge from the matrix of Buddhist ideas and religion). Scott Atran argues that Wilson’s conception of religion is “mind blind,” and I think he is right. Wilson believes in the power of natural selection to shape human societies toward an adaptive optimum, but he neglects the cognitive constraints imposed by the multi-purpose nature of our mental utilities. Religious ideas are tested not just on the adaptive landscape of reproductive fitness (e.g., religions which impose obligate celibacy, like the Shakers, obviously must exist parasitically upon the general population), but against the landscape of our minds.
“Religion” as a suite of characteristics has multiple levels:
1) There is the individual level predicated on psychological biases & needs. All religions on the level of the typical believer exhibit reoccurring supernatural motifs (e.g., personal gods of great power but constrained scope). Additionally, the non-theistic elements of a religion, such as as mass ritual, tend also satisfy psychological impusles (e.g., images and pageantry which induce sensory arousal).
2) These basic motifs are reshaped to form cultural markers in the form of giving communal names and peculiar characteristics to the god concepts. Rituals obtain their own interpretations and novel flairs. These are not functional per se, but rather serve as a way to mark ingroups vs. outgroups. Just as body art is not necessarily useful as camouflage, so propitiating the gods is not necessarily functional because the gods do exist. Rather, they serve as easy to identify group markers (just like linguistic dialect).
3) As human culture elaborated and mass societies expanded the nature of religion changed. Just as tribes became reconceptualized as universal imperiums, so tribal religious prejudices and superstitions were conflated toward universal theological truths. Just as natural philosophy arose to offer an alternative model of the universe to the supernaturalism, so religious professionals with a philosophical bent took the theistic raw material and transformed it through quasi-propositional word games. The rise of priest-kings in Sumer and god-kings in Egypt shows how early on the tribal supernaturalism was a useful natural lever which the rising urban centered polities could utilize to control their subject population. The subjects of Ur were bound to their devotion to Innana, while the residents of the Nile Valley were both worshipers and subjects of the god-king pharoh.
4) The specialization of human life in generally has resulted in the diversification & stratification of religion vertically, from the monk & priest down to the illiterate peasant. Though the structural nature of the human mind means that the peasant and the priest can comprehend the same god-image in their mind’s eye, the latter has access to literate abstractions (e.g., Greek philosophy in the case of Christianity) which can overlay the basic god-concept with a stylistic exterior. Though the peasant may not truly understand the word games which the theologian plays, like body art or a particular style of dress, he can accede to a pared down catechism or profession of faith. This is an extension of the differentiation of tribal level religions based purely on names as opposed to substance (e.g., “My god of the river is better than your god of the river!”).
5) Just as there was vertical differentiation between elite believers and the masses, there was horizontal differentiation between the “great” religions (which really recapitulates the differentiation of tribal beliefs). The worship of Allah, Shiva, etc., became associated with different rituals, priestly superstructures and variant motifs. Though the god concept is fundamentally the same (the omni-God of the Abrahamics & the pantheistic divinity of the Hindus are both out of the reach of gestalt comprehension), the names and quasi-propositional elucidation differ a great deal. There are also differences in the rule sets which define how and why the believers should behave. On a fundamental level I believe this is relevant to the elites only. I suspect that the relative advantage that monotheism has over non-monotheism is due primarily to the turnover in the elites, because the masses continue to satisfy the same psychological urges under a different name. Bouchard’s twin studies suggest moderate heritability in religious zeal, but none in specific denomination.
One practical conclusion is that even though religious sentiment of some sort is highly likely within a society, one religious outlook is not. Religions can adapt to the cultures in which they reside (e.g., Muslims in the southern Indian state of Kerala did not necessarily abandon matrilineal inheritance practices, Christianity picked up the Greco-Roman pagan bias against polygyny). When religion changes a culture when it is introduced, it is often a function of the robusticity of the two cultures in question. Christianity induced some changes in northern Europe because of the higher prestige of Roman culture as a whole, but when it spread with the Empire it tended to adapt and make peace with many aspects of the pagan Mediterranean matrix.
Related: “Must reads” are bolded….
The nature of religion and Breaking the Spell
Modes of religion
Who Dan Dennett think he be foolin’?
An evolutionary anthropology of religion
God lives, deal with it!
Belief & belief in belief
Logical consistency is irreligious
God & morality
Are people naturally religious? Yes….
The round-eyed Buddha
Nerds are nuts
Atheism, Heresy and Hesychasm
The God Delusion – Amongst the unbelievers
Innate atheism & variation across societies
“Hard-wired” for God
Buddhism, a religion or not?
Why do people believe in God?
Is religion an adaptation?
Theological incorrectness – when people behave how they shouldn’t….sort of
The gods of the cognitive scientists