Breaking the Spell

by Daniel C. Dennett

  These are short periods of time, biologically speaking. They are not even long compared with the ages of other features of human culture. Writing is more than five thousand years old, agriculture is more than ten thousand years old, and language is—who knows?—maybe “only” forty thousand years old and maybe ten or twenty times older than that. It’s a contentious research topic, and since it’s widely agreed that fully articulated natural languages must have developed out of some kind of proto-languages (which may have evolved over hundreds of thousands of years), there is no consensus about what would even count as the birthdate of language. Is language older than religion? However we date its beginnings, language is much, much older than any existing religion, or even any religion of which we have any historical or archeological knowledge. The earliest impressive archeological evidence of religion is the elaborate Cro-Magnon burial sites in the Czech Republic, and they are about twenty-five thousand years old.2 It is hard to tell, but something like religion may well have existed from the early days of language, however, or even before that. What were our ancestors like before there was anything like religion? Were they like bands of chimpanzees? What, if anything, did they talk about, aside from food and predators and the mating game? The weather? Gossip? What was the psychological and cultural soil in which religion first took root?

  We can tentatively work backward, extrapolating under the guidance of our fundamental biological constraint: each innovative step had to “pay for itself” somehow, in the existing environment in which it first occurred, independently of whatever its role might become in later environments. What, then, could explain both the diversity and the similarities in the religious ideas we observe around the world? Are the similarities due to the fact that all religious ideas spring from a common ancestor idea, passed on over the generations as people spread around the globe, or are such ideas independently rediscovered by just about every culture because they are simply the truth, and obvious enough to occur to people in due course? These are obviously naïve oversimplifications, but at least they are attempts to ask and answer explicit questions often left unexamined by people who lose interest once they have found a purpose or function for religion that strikes them as plausible: responding to a suitably grand “human need” to account for the manifest outlay of time and energy that religion requires. The three favorite purposes or raisons d’être for religion are

  to comfort us in our suffering and allay our fear of death

  to explain things we can’t otherwise explain

  to encourage group cooperation in the face of trials and enemies

  Thousands of books and articles have been written defending these claims, and such compelling and familiar ideas are probably at least partly right, but if you settle for one of them, or even all three taken together, you succumb to a disorder often encountered in the humanities and social sciences: premature curiosity satisfaction. There is so much more to ask about, and so much more to understand. Why would these ideas comfort people? (And why are they comforting, exactly? Might there be better, more comforting ideas to be found?) Why would these ideas appeal to people as explanations of baffling events? (And how could they have arisen? Did some would-be proto-scientist hit upon a supernatural theory and enthusiastically proselytize her neighbors?) How do these ideas actually manage to enhance cooperation in the face of suspicion and defection? (And once more, how could they have arisen? Did some wise tribal leader invent religion to give her tribe a teamwork edge over the rival tribes?)

  Some people suppose that we can never do better than such simple speculations about these processes and outcomes from the remote past. Some insist on it, in fact, and their vehemence betrays the fact that they are afraid they are wrong. They are. Today, thanks to advances in a variety of sciences, we can sharpen the questions and begin to answer them. In this and the next four chapters, I will try to tell the best current version of the story science can tell about how religions have become what they are. I am not at all claiming that this is what science has already established about religion. The main point of this book is to insist that we don’t yet know—but we can discover—the answers to these important questions if we make a concerted effort. Probably some of the features of the story I tell will prove in due course to be mistaken. Maybe many of them are wrong. The purpose of trying to sketch a whole story now is to get something on the table that is both testable and worth testing. It is usually easier to fix something that has flaws than to build something from scratch. Trying to bridge the gaps in our knowledge forces us to frame questions we haven’t framed before, and puts the issues in a perspective that enables further questions to be posed and answered. And that in itself can undercut the defeatist proclamation that these are mysteries beyond all human comprehension. Many people may wish that these were unanswerable questions. Let’s see what happens when we defy their defensive pessimism and give it a try.

  2 The raw materials of religion

  We may conclude, therefore, that, in all nations, which have embraced polytheism, the first ideas of religion arose not from a contemplation of the works of nature, but from a concern with regard to the events of life, and from the incessant hopes and fears, which actuate the human mind.

  —David Hume, The Natural History of Religion

  My guides are the pioneering scientists who have begun to tackle these questions with both imagination and discipline. An evolutionary biologist or a psychologist who knows only one religion at all well and has a smattering of (mis) information about the others (like most of us) is almost certain to overgeneralize from idiosyncratic familiarity when it comes to framing questions. A social historian or an anthropologist who knows a great deal about the beliefs and practices of people all around the world but is naïve about evolution is equally unlikely to frame the issues well. Fortunately, a few well-informed researchers have recently begun to pull these distant perspectives together, with tantalizing results. Their books and articles are well worth reading in their entirety, as I hope I will convince you by introducing a few highlights.

  Jared Diamond’s Guns, Germs, and Steel (1997) is an eye-opening exploration of very specific effects of geography and biology on the early development of agriculture in different parts of the world at different times. When the first agriculturalists domesticated animals, they naturally began living in close proximity to them, and this enhanced the likelihood of species-jumping by the animals’ parasites. The most serious infectious diseases known to humanity, such as smallpox and influenza, all derive from domesticated animals, and our farming ancestors lived through a horrific pruning in which untold millions succumbed to early versions of these diseases, leaving only those fortunate enough to have some natural immunity to reproduce. Many generations of this evolutionary bottleneck guaranteed that their later descendants would be relatively immune to, or have a high tolerance for, the descendants of those virulent strains of parasite. When these grand-offspring, living mainly in Europe, developed the technology to cross the oceans, they brought their germs with them, and it was the germs, more than the guns and steel, that wiped out large fractions of the indigenous populations they encountered. The role of agriculture in spawning infectious diseases, and the relative immunity to them that had evolved among the peoples who had lived through the ravages of the early days of agriculture, can be studied with some precision now that we can extrapolate backward from the genomes of existing species of plants, animals, and germs. Accidents of geography gave European nations a head start that goes a long way to explain why they were the colonizers rather than the colonized in later centuries.

  Diamond’s Pulitzer Prize–winning book is deservedly well known, but not alone. There is a new generation of interdisciplinary researchers working to put together the biology with the evidence gleaned by centuries of work by historians, anthropologists, and archeologists. Pascal Boyer and Scott Atran are anthropologists who have done extensive fieldwork i
n Africa and Asia but who are also trained in evolutionary theory and cognitive psychology. Their recent books, Religion Explained: The Evolutionary Origins of Religious Thought (Boyer, 2001) and In Gods We Trust (Atran, 2002), develop largely harmonious accounts of the major steps into the swamp that they and others have been taking. Then there is David Sloan Wilson, an evolutionary biologist who has been devoting himself in recent years to analyses that systematically exploit the Human Relations Area File, a database of all the world’s cultures compiled by anthropologists. His recent book Darwin’s Cathedral: Evolution, Religion, and the Nature of Society (2002) makes the best case to date for the hypothesis that religion is a social phenomenon designed (by evolution) to improve cooperation within (not among!) human groups. According to Wilson, religion emerged by a process of group selection, a controversial wrinkle in evolutionary theory that is dismissed by many evolutionary theorists as at best a marginal process whose conditions for success are unlikely to arise and persist for long. There are deep reasons to be skeptical about group selection, especially in our species, and precisely because Wilson’s thesis—religion as a cooperation-enhancer—is deeply attractive to many people, we need to brace ourselves to avoid wishful thinking. It is quite generally agreed among his critics that he has not (yet ) succeeded in making the case for his radical thesis of group selection, but even a roundly refuted scientific theory can make a major contribution to the steady accumulation of scientific understanding if the evidence marshaled for and against it has been scrupulously gathered. (For more on this point, see appendix B.) Here I will introduce the main points of agreement, as well as acknowledging the continuing points of contention, packing off most of the controversial details into the endnotes and appendixes, where those with a taste for them can (begin to) pursue their own deeper consideration of them.

  Both Boyer and Atran present the work of a small but growing community of researchers in relatively accessible terms.3 Their central thesis is that in order to explain the hold that various religious ideas and practices have on people, we need to understand the evolution of the human mind. For many centuries, most philosophers and theologians contended that the human mind (or soul) was an immaterial, incorporeal thing, what René Descartes called a rescogitans (thinking thing). It was in some sense infinite, immortal, and utterly inexplicable by material means. We now understand that the mind is not, as Descartes confusedly supposed, in communication with the brain in some miraculous way; it is the brain, or, more specifically, a system or organization within the brain that has evolved in much the way our immune system or respiratory system or digestive system has evolved. Like many other natural wonders, the human mind is something of a bag of tricks, cobbled together over the eons by the foresightless process of evolution by natural selection. Driven by the demands of a dangerous world, it is deeply biased in favor of noticing the things that mattered most to the reproductive success of our ancestors.4

  Some of the features of our minds are endowments we share with much simpler creatures, and others are specific to our lineage, and hence much more recently evolved. These features sometimes overshoot, sometimes have curious by-products, sometimes are ripe for exploitation by other replicators. Of all the quirky effects generated by the whole bag of tricks—our set of “gadgets,” as Boyer calls them—a few happen to interact with one another in mutually reinforcing ways, creating patterns observable in all cultures, with interesting variations. Some of these patterns look rather like religions, or pseudo-religions, or proto-religions. The by-products of the various gadgets are what Boyer calls concepts:

  Some concepts happen to connect with inference systems in the brain in a way that makes recall and communication easy. Some concepts happen to trigger our emotional programs in particular ways. Some concepts happen to connect to our social mind. Some of them are represented in such a way that they soon become plausible and direct behavior. The ones that do all this are the religious ones we actually observe in human societies. [p. 50]

  Boyer lists more than half a dozen distinct cognitive systems that feed effects into this recipe for religion—an agent-detector, a memory-manager, a cheater-detector, a moral-intuition-generator, a sweet tooth for stories and storytelling, various alarm systems, and what I call the intentional stance. Any mind with this particular set of thinking tools and biases is bound to harbor something like a religion sooner or later, he claims. Atran and others offer largely concurring accounts, and the details are well worth exploring, but I am just going to sketch some of the big picture so that we can see the overall shape of the theory, not (yet) assess it for truth. It will take decades of research to secure any of this theory, but right now we can get a sense of what the possibilities are, and hence what questions we ought to be trying to answer.

  3 How Nature deals with the problem of other minds

  We find human faces in the moon, armies in the clouds; and by a natural propensity, if not corrected by experience and reflection, ascribe malice and good-will to every thing, that hurts or pleases us.

  —David Hume, The Natural History of Religion

  “I saw you take his kiss!” “’Tis true”

  “Oh Modesty!” “’Twas strictly kept:

  He thought me asleep; at least I knew

  He thought I thought he thought I slept”

  —Coventry Patmore, “The Kiss”

  The first thing we have to understand about human minds as suitable homes for religion is how our minds understand other minds! Everything that moves needs something like a mind, to keep it out of harm’s way and help it find the good things; even a lowly clam, which tends to stay in one place, has one of the key features of a mind—a harm-avoiding retreat of its feeding “foot” into its shell when something alarming is detected. Any vibration or bump is apt to set it off, and probably most of these are harmless, but better safe than sorry is the clam’s motto (the free-floating rationale of the clam’s alarm system). More mobile animals have evolved more discriminating methods; in particular, they tend to have the ability to divide detected motion into the banal (the rustling of the leaves, the swaying of the seaweed) and the potentially vital: the “animate motion” (or “biological motion”) of another agent, another animal with a mind, who might be a predator, or a prey, or a mate, or a rival conspecific. This makes economic sense, of course. If you startle at every motion you detect, you’ll never find supper, and if you don’t startle at the dangerous motions, you’ll soon be somebody else’s supper. This is another Good Trick, an evolutionary innovation—like eyesight itself, or flight—that is so useful to so many different ways of life that it evolves over and over again in many different species. Sometimes this Good Trick can be too much of a good thing; then we have what Justin Barrett (2000) calls a hyperactive agent detection device, or HADD. This overshooting is not restricted to human beings. When your dog leaps up and growls when some snow falls off the eaves with a thud that rouses him from his nap, he is manifesting a “false positive” orienting response triggered by his HADD.

  Recent research on animal intelligence (Whiten and Byrne, 1988, 1997; Hauser, 2000; Sterelny, 2003; see also Dennett, 1996) has shown that some mammals and birds, and perhaps some other creatures as well, carry these agent-discriminations into more sophisticated territory. Evidence shows that they not only distinguish the animate movers from the rest, but draw distinctions between the likely sorts of motions to anticipate from the animate ones: will it attack me or flee, will it move left or right, will it back down if I threaten, does it see me yet, does it want to eat me or would it prefer to go after my neighbor? These cleverer animal minds have discovered the further Good Trick of adopting the intentional stance (Dennett, 1971, 1983, 1987): they treat some other things in the world as

  agents with

  limited beliefs about the world,

  specific desires, and

  enough common sense
to do the rational thing given those beliefs and desires.

  Once animals began adopting the intentional stance, something of an arms race ensued, with ploy and counterploy, deceptive move and intelligent detection of deceptive move, carrying animal minds to greater subtlety and power. If you have ever tried to catch or trap a wild animal, you have some appreciation of the wiliness that has evolved. (Clam-digging, in contrast, is child’s play. Clams have not evolved the intentional stance, though they do have simple hair-trigger HADDs.)

  The utility of the intentional stance in describing and predicting animal behavior is undeniable, but that doesn’t mean that the animals themselves are clued in about what they are doing. When a low-nesting bird leads the predator away from her nestlings by doing a distraction display, she is making a convincing sham of a broken wing, creating the tempting illusion of an easy supper for the observing predator, but she need not understand this clever ruse. She does need to understand the conditions of likely success, so that she can adjust her behavior the better to fit the variations encountered, but she no more needs to be aware of the deeper rationale for her actions than does the fledgling cuckoo when it pushes the rival eggs out of the nest in order to maximize the food it will get from its foster parents.