Breaking the Spell

by Daniel C. Dennett

  Some aspects of this theory sketch are pretty well established, but getting down to specifics and generating further testable hypotheses is work for the future. I wanted to give readers a good idea of what a testable theory would be like, what sorts of questions it would raise, and what sorts of explanatory principles it could invoke. My theory sketch may well be false in many regards, but if so, this will be shown by confirming some alternative theory of the same sort. In science, the tactic is to put forward something that can be either fixed or refuted by something better. A century ago, it was just a theory that powered fixed-wing flight was possible; now it is fact. A few decades ago, it was just a theory that the cause of AIDS was a virus, but the reality of HIV is not just a theory today.

  Since my proto-theory is not yet established and may prove to be wrong, it shouldn’t be used yet to guide our policies. Having insisted at the outset that we need to do much more research so that we can make well-informed decisions, I would be contradicting myself if I now proceeded to prescribe courses of action on the basis of my initial foray. Recall, from chapter 3, the moral that Taubes drew in his history of the misguided activism that led us on the low-fat crusade: “It’s a story of what can happen when the demands of public health policy—and the demands of the public for simple advice—run up against the confusing ambiguity of real science.” There is pressure on us all to act decisively today, on the basis of the little we already (think we) know, but I am counseling patience. The current situation is scary—one religious fanaticism or another could produce a global catastrophe, after all—but we should resist rash “remedies” and other overreactions. It is possible, however, to discuss options today, and to think hypothetically of what the sound policies would be if something like my account of religion is correct. Such a consideration of possible policies can help motivate the further research, giving us pressing reasons for finding out which hypotheses are really true.

  If somebody wants to put a sticker in this book, saying that it presents a theory, not a fact, I would happily concur. Caution! it should say. Assuming that these propositions are true without further research could lead to calamitous results. But I would insist that we also put the stickers on any books or articles that maintain or presuppose that religion is the lifeboat of the world, which we dare not upset. The proposition that God exists is not even a theory, as we saw in chapter 8. That assertion is so prodigiously ambiguous that it expresses, at best, an unorganized set of dozens or hundreds—or billions—of quite different possible theories, most of them disqualified as theories in any case, because they are systematically immune to confirmation or disconfirmation. The refutable versions of the claim that God exists have life cycles like mayflies, being born and dying within a matter of weeks, if not minutes, as predictions fail to come true. (Every athlete who prays to God for victory in the big game and then wins is happy to thank God for taking his side, and chalks up some “evidence” in favor of his theory of God—but quietly revises his theory of God whenever he loses in spite of his prayers.) Even the secular and nonpartisan proposition that religion in general does more good than harm, either to the individual believer or to society as a whole, has hardly begun to be properly tested, as we saw in chapters 9 and 10.

  So here is the only prescription I will make categorically and without reservation: Do more research. There is an alternative, and I am sure it is still hugely appealing to many people: Let’s just close our eyes, trust to tradition, and wing it. Let’s just take it on faith that religion is the key—or one of the keys—to our salvation. How can I quarrel with faith (for heaven’s sake)? Blind faith? Please. Think. This is where we began. My task was to demonstrate that there was enough reason to question the tradition of faith so that you could not in good conscience turn your back on the available or discoverable relevant facts. I am quite prepared to roll up my sleeves and get down to examining the evidence and considering alternative scientific theories of religion, but I think I have already made my case that it would be indefensibly reckless not to do this research.

  My survey has highlighted a small fraction of the work that has already been done, using it to tell one of the possible stories of how religion became what it is today, leaving other stories unmentioned. I told what I think is the best current version, but perhaps I have overlooked some contributions that will eventually be recognized retrospectively to be more important. This is a risk that a project like mine takes: if, by drawing attention to one avenue of research, it helps bury some better avenue in oblivion, I will have done a disservice. I am acutely aware of this prospect, so I have shared drafts of this book with researchers who have their own vision of how to make progress in the field. My network of informants inevitably has its own bias, however, and I would like nothing better than for this book to provoke a challenge—a reasoned and evidence-rich scientific challenge—from researchers with opposing viewpoints.1

  I anticipate that one of the challenges will come from those in academia who are unmoved by my discussion of the “academic smoke screen” in chapter 9, and who firmly believe that the only researchers qualified to do the research are those who enter into an exploration of religion with a “proper respect” for the sacred, with a deep commitment to hallowing the traditions if not converting to them. They will want to maintain that the sort of empirical, biology-based inquiries I have championed, what with their mathematical models and use of statistics and the rest, are bound to be woefully superficial, insensitive, and uncomprehending.

  Recent history shows that this is a concern to take seriously. A few decades ago, the field of “science studies” was born, when historians of science and philosophers of science were joined by sociologists and anthropologists who decided to apply their techniques, honed on the exploration of tribal cultures isolated in distant jungles and archipelagoes, to science itself, such as the subcultures of particle physicists, or molecular biologists, or mathematicians. Some of the early attempts by well-intentioned teams of social scientists to study these phenomena “in the wild” (of the laboratory and seminar room) led to the publication of studies that were met with—and deserved—the derision of the scientists who were the topic of the research. However sophisticated the researchers may have been as anthropologists, they were still naïve observers, largely clueless about the technicalities of the science they were witnessing, so they often came up with comically bad interpretations of what they had observed. If you don’t understand in some detail the enterprise of the people you are studying, you have scant chance of understanding their interactions and reactions at the human level. The same maxim should apply to the study of religious discourse and practices.

  People in science studies have had to work hard to overcome the bad reputation the field garnered in its early days, and there are still many scientists who do not bother suppressing their contempt for it, but the misguided work has by now been more than balanced by deeply informed and comprehending work that has actually managed to open scientists’ eyes to patterns and foibles in their own practice. The key to this more recent success is simple: do your homework. Anybody hoping to make sense of any highly sophisticated and difficult field of human effort needs to become a near-expert in that field in addition to having the training of his or her home field. Applied to the study of religion, the prescription is clear: scientists intent on explaining religious phenomena are going to have to delve deeply and conscientiously into the lore and practices, the texts and contexts, the daily lives and problems of the people they are studying.

  How could this be guaranteed? Religious experts—priests, imams, rabbis, ministers, theologians, historians of religion—who are skeptical of the qualifications of those scientists who would study them could create and administer an entrance examination! Anybody who could not pass the entrance exam that they devised would be quite appropriately judged not sufficiently knowledgeable to comprehend the phenomena
under investigation, and could be denied access and cooperation. Let the experts make the entrance examination as demanding as they like, and give them total authority on grading it, but require some of their own experts to take the exam as well, and require that the examination be blind-graded, so the graders couldn’t know the identity of the candidates. That would give the religious experts a way of confirming their mutual esteem while weeding out the clueless from their own ranks and certifying any qualified investigators.2

  2 Some avenues to explore: how can we home in on religious conviction?

  Thou shalt not answer questionnaires

  Or quizzes upon World-Affairs,

  Nor with compliance

  Take any test. Thou shalt not sit

  With statisticians nor commit

  A social science.

  —W. H. Auden, “A Reactionary Tract for the Times”

  What research is needed? Consider some of the unanswered empirical questions already raised by me so far in this book:

  Chapter 4: 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? Do the burial practices of Neanderthals show that they must have had fully articulate language?

  Chapter 5: Could an ape (without language) concoct the counterintuitive combination of a walking tree or an invisible banana? Why don’t other species have art? Why do we human beings so consistently focus our fantasies on our ancestors? Does impromptu hypnosis work as effectively when the hypnotist is not the parent? How well have nonliterate cultures preserved their rituals and creeds over the generations? How did healing rituals arise? Does there have to be someone to prime the pump? (What is the role of charismatic innovators in the origin of religious groups?)

  Chapter 6: For how long could folk religion be carried along by our ancestors before reflection began to transform it? How and why did folk religions metamorphose into organized religions?

  Chapter 7: Why do people join groups? Is the robustness of a religion like the robustness of an ant colony or a corporation? Is religion the product of blind evolutionary instinct or rational choice? Or is there some other possibility? Are Stark and Finke right about the principal reason for the precipitous decline after Vatican II in Catholics seeking a vocation in the church?

  Chapter 8: Of all the people who believe in belief in God, what percentage (roughly) also actually believe in God? At first it looks as if we could simply give people a questionnaire with a multiple-choice question on it:

  I believe in God: _____ Yes _____ No _____ I don’t know

  Or should the question be:

  God exists: _____ Yes _____ No _____ I don’t know

  Would it make any difference how we framed the questions?

  You will notice that hardly any of these questions deal even indirectly with either brains or genes. Why not? Because having religious convictions is not very much like having either epileptic seizures or blue eyes. We can already be quite sure there isn’t going to be a “God gene,” or even a “spirituality” gene, and there isn’t going to be a Catholicism center in the brain of Catholics, or even a “religious experience” center. Yes, certainly, whenever you think of Jesus some parts of your brain are going to be more active than others, but whenever you think of anything this is going to be true. Before we start coloring in your particular brain-maps for thinking about jesting and Jet Skis and jewels (and Jews), we should note the evidence that suggests that such hot spots are both mobile and multiple, heavily dependent on context—and of course not arrayed in alphabetical order across the cortex! In fact, the likelihood that the places that light up today when you think about Jesus are the same places that will light up next week when you think about Jesus is not very high. It is still possible that we will find dedicated neural mechanisms for some aspects of religious experience and conviction, but the early forays into such research have not been persuasive.3

  Until we develop better general theories of cognitive architecture for the representation of content in the brain, using neuro-imaging to study religious beliefs is almost as hapless as using a voltmeter to study a chess-playing computer. In due course, we should be able to relate everything we discover by other means to what is going on among the billions of neurons in our brains, but the more fruitful paths emphasize the methods of psychology and the other social sciences.4

  As for genes, compare the story I have told in the earlier chapters with this simplified version, from Time magazine’s recent cover article “Is God in our Genes?”:

  Humans who developed a spiritual sense thrived and bequeathed that trait to their offspring. Those who didn’t risked dying out in chaos and killing. The evolutionary equation is a simple but powerful one. [Kluger, 2004, p. 65]

  The idea that lurks in this bold passage is that religion is “good for you” because it was endorsed by evolution. This is just the sort of simpleminded Darwinism that rightly gives the subtle scholars and theorists of religion the heebie-jeebies. Actually, as we have seen, it isn’t that simple, and there are more powerful evolutionary “equations.” The hypothesis that there is a (genetically) heritable “spiritual sense” that boosts human genetic fitness is one of the less likely and less interesting of the evolutionary possibilities. In place of a single spiritual sense we have considered a convergence of several different overactive dispositions, sensitivities, and other co-opted adaptations that have nothing to do with God or religion. We did consider one of the relatively straightforward genetic possibilities, a gene for heightened hypnotizability. This might have provided major health benefits in earlier times, and would be one way of taking Hamer’s “God gene” hypothesis seriously. Or we could put it together with William James’s old speculation that there are two kinds of people, those who require “acute” religion and those whose needs are “chronic” and milder. We can try to discover if there really are substantial organic differences between those who are highly religious and those whose enthusiasm for religion is moderate to nonexistent.

  Suppose we struck paydirt and found just such a pattern. What would be the implications—if any—for policy? We could consider the parallel with the genetic differences that help to account for some Asians’ and some Native Americans’ difficulty with alcohol. As with variation in lactose tolerance, there is genetically transmitted variation in the ability to metabolize alcohol, due to variation in the presence of enzymes, mainly alcohol dehydrogenase and aldehyde dehydrogenase.5 Needless to say, since, through no fault of their own, alcohol is poisonous to people with these genes—or it turns them into alcoholics—they are well advised to forgo alcohol. A different parallel is with the genetically transmitted distaste for broccoli and cauliflower and cilantro that many people discover in themselves; they have no difficulty metabolizing these foods, but find them unpalatable, because of identifiable differences in the many genes that code for olfactory receptors. They don’t have to be advised to avoid these foods. Might there be either “spiritual-experience intolerance” or “spiritual-experience distaste”? There might be. There might be psychological features with genetic bases that are made manifest in different reactions by people to religious stimuli (however we find it useful to classify these). William James offers informal observations that give us some reason to suspect this. Some people seem impervious to religious ritual and all other manifestations of religion, whereas others—like me—are deeply moved by the ceremonies, the music, and the art—but utterly unpersuaded by the doctrines. It may be that still others hunger for these stimuli, and feel a deep need to integrate them into their lives, but would be well advised to steer clear of them, since they can’t “metabolize” them the way other people can. (They become manic and out of control, or depressed, or hysterical, or confused, or addicted.)

  These are hypo
theses that are definitely worth formulating in detail and testing if we can identify patterns of individual variation, whether or not they are genetic (they might be culturally transmitted, after all). To take a fanciful example, it could turn out that people whose native language was Finnish (whatever their genetic heritage) were well advised to moderate their intake of religion!

  A “spiritual sense” (whatever that is) might prove to be a genetic adaptation in the simplest sense, but more specific hypotheses about patterns in human tendencies to respond to religion are apt to be more plausible, more readily tested, and more likely to prove useful in disentangling some of the vexing policy questions that we have to face. For instance, it would be particularly useful to know more about how secular beliefs differ from religious beliefs (and as we saw in chapter 8, “belief” is a misnomer here; we might better call them religious convictions to mark the difference). How do religious convictions differ from secular beliefs in the manner of their acquisition, persistence, and extinction, and in the roles they play in people’s motivation and behavior? There has been a substantial research industry devoted to conducting surveys on all aspects of religious attitude.6 We regularly see the highlights of the latest results in the media, but the theoretical underpinnings and enabling assumptions of the survey methodologies are in need of careful analysis. Alan Wolfe (2003, p. 152), for one, thinks that the surveys are unreliable: “The results are inconsistent and puzzling, depending, as is often the case with such research, on the wording of the questions in surveys or the samples chosen for analysis.” But is Wolfe right? This should not just be a matter of personal opinion. We need to find out.