Many attempts to provide a naturalistic explanation for religion start from the assumption that it must have had survival benefits for our hunter-gatherer ancestors. But Atran does not believe that religions are adaptations which enhance genetic fitness or that they have evolutionary functions as such. This does not mean that he neglects evolution, however; quite the opposite. Rather, he thinks that religion needs to be seen in the wider context of how the mind itself has evolved.
As his subtitle implies, he uses the metaphor of a mountain/valley landscape to explain the forms that religions take.
This landscape is shaped by natural selection. It is ancestrally defined by specific sets of affective, social, and cognitive features (different mountain ridges). Each mountain ridge in this landscape has a distinct contour, with various peaks whose heights reflect evolutionary time.All religions are supposed to follow the same structural contours, so the shape of the landscape determines the kinds of religion that can arise. The constraining forces that shape the landscape are folk-mechanics, folk-biology, and folk-psychology. All of these have in turn been shaped by natural selection over long periods; indeed, Atran suggests that folk-mechanics may have its ultimate origin in the reptilian brain.
Probably the most characteristic feature of religions is that they presuppose the existence of invisible beings such as gods, demons, and spirits that influence human life. How to explain this is a major problem. Atran's solution is fairly similar to that of a fellow anthropologist, Pascal Boyer, being based on the idea that belief in these invisible entities results from the operation of the same mental processes as are involved in the formation of ordinary beliefs.
Atran postulates the existence of an "innate releasing mechanism" in the mind or brain. This mechanism has been selected for during evolution because it allowed our ancestors to detect hostile animals and humans in time to take avoiding action. It would be advantageous for the mechanism to have hair-trigger sensitivity; better to respond to a threat that wasn't there than to miss one that was.
This hair-trigger response, however, caused the mechanism to respond not only to "real" threats but also to inanimate phenomena that mimic the appearance of living agents: apparent voices in the wind or running water, faces in the clouds and so on. And eventually the response became still further removed from actual phenomena in the outer world to encompass imaginary beings derived from dreams and visions.
While this is probably as plausible an account of the origin of these counter-intuitive beliefs as any, it requires supporting evidence if it is to amount to more than armchair speculation. This is what Atran seeks to provide, with ample citation of anthropological material and other sources.
The book has four parts. Part 1 looks at the innate releasing mechanism theory in detail. Part 2 is about the cognitive structures and social commitments that make up belief in the supernatural and promote stability in societies. Part 3 deals with ritual practice and religious experience. Part 4 contrasts the approach used in the book with rival theories such as sociobiology, group theory, and meme theory which have been suggested by people who think that religion is best explained in terms of selection and adaptation.
Atran finds all of these rival ideas to be wanting in various ways. I thought his demolition of "neurotheology" was particularly telling ("A 'God Module' in the Temporal Lobe? Not Likely"). Still, there do seem to be some intriguing connections between brain function and belief: Atran cites studies which show that exposure to descriptions of death and suffering increases strength of belief in God. "[E]motional stress associated with death-related scenes seems a stronger motivator for religiosity than mere exposure to emotionally unstressful religious scenes, such as praying." Catholic rituals such as the Stations of the Cross take on a new light in this context.
In a note on p.286 Atran suggests that the Assassins of mediaeval Islam closely resembled the Indian sect of Thugs. This seems misleading to me. The Assassins used the murder of eminent opponents as a political weapon, because they were a relatively powerless minority; this is quite different from the Thugs, who murdered travellers as a tribute to the goddess Kali. Atran takes Marco Polo's unreliable account of the Assassins as drug-crazed terrorists more or less at face value, but the Assassins (or Nizaris, as they are properly known), were not drugged, nor were their motives remotely comparable with those of the Thugs, who were devotees of Kali and robbed their victims. The Assassins used murder as a political expedient in a way that made good sense in the circumstances at the time but these were in no sense ritual killings.
The book does not constitute easy reading. As one would expect, there are extensive notes and references (though the indexing is a little erratic). The language, in Atran's word, is often "dense", which seems to mean not devoid of jargon; it is also sometimes repetitious. There are large amounts of detail, which at times tend to obscure the argument as much as to illuminate it. But it is chock-full of interesting ideas. The casual reader might care to start with the introduction and then go straight to the final chapter, which provides a useful summary of the book as a whole.
Atran's conclusion is that religion is not going to fade away any time soon.
As long as people share hope beyond reason, religion will persevere. For better or worse, religious belief in the supernatural seems here to stay. With it comes trust in deities good and bad, songs of fellowship and drums of war, promises to allay our worst fears and achieve our most fervent hopes, and heartfelt communion in costly homage to the absurd. This loss and gain persist as the abiding measure of humanity. No other seems able to compete for very long. And so spirituality looms as humankind's provisional evolutionary destiny.This seems to me to be essentially correct, whether or not one fully agrees with Atran's account of how religion originated.
20 January 2007