Opiate of the Philosophers.

Breaking the Spell

Religion as a Natural Phenomenon

Daniel C. Dennett

Viking, $25.95, 448 pp.

The concept or, if one prefers, the "phenomenon" of religion is evidently somewhat indeterminate. It has clear instances: Judaism, Christianity, Islam, Hinduism, and so on; but, and this is why "and so on" is not very helpful, there are less clear cases, such as Taoism, and progressively more doubtful ones, for example Falun Gong, the Aquarian Foundation, and Scientology.

The Web site adherents.com lists more than four thousand "religions, churches, denominations, religious bodies, faith groups, tribes, cultures, movements, ultimate concerns, etc."--notice again the terminal drift into vagueness marked by "etc." Even restricting oneself to more or less clear cases, though, there are vast numbers of religious believers throughout the world. Indeed, of the 6.5 billion humans on the globe, about 80 percent belong to recognized religions, about 33 percent are Christians, holding allegiance to teachings only formulated quite recently in the history of humankind, and half of the latter are Roman Catholics. Arguably nothing compares with religion as a domain of commitment, and no other extensive commitment so unifies humanity, even when it divides it. What then explains the origins of religion and its power to have and to hold the allegiance of so many?

One answer is that religion gives expression to a universal need to acknowledge and respond to a sense of cosmic order and what theologians often call human "creatureliness." In writing that "ever since the creation of the world [God's] invisible nature, namely his eternal power and deity, has been clearly perceived in the things that have been made" (Romans), St. Paul was expressing an idea familiar to him before his conversion. A century earlier, Cicero wrote "what can be so obvious and clear, as we gaze up at the sky and observe the heavenly bodies, as that there is some divine power of surpassing intelligence by which they are ordered" (On the Nature of the Gods). Four centuries before that, Empedocles observed that matter is inanimate and hence must be moved by another, from which he concluded that ultimately there must be some intrinsic principles of attraction and repulsion--which he termed "Love" and "Strive." Earlier still and in other cultures the idea of a transcendent source of existence, order, and movement seems to have been prevalent, a reflective response to experience.

Another possibility is that religion originates in an innate idea. Experiences of wonder may elicit it, but it is not a conclusion derived from them; rather we have built within us a notion of a supreme other and an attitude of awe or piety toward the world as the work of that "Other." Initially the idea might be embryonic and ill-defined, but given time it could grow into natural religion. And the god(ish) idea might, as Anselm and Descartes supposed, have been put there by God himself--in order that we should have a good chance of coming to know and to love him.

A third answer might be developed by drawing on something else that Empedocles is reported to have said, namely, that whenever things turned out as they would have done if they were happening for a purpose, then creatures survived, but when this did not happen, they perished. Suppose, then, that religion does indeed derive from an idea in our minds but that this idea has not been arrived at by observing nature or ourselves, or been put there for a purpose, but instead is simply the result of an ancient accident of circumstance that has survived because it brings certain advantages. Suppose, in other words, that religion is a product of blind evolution.

Since the thinkers of classical antiquity had the intellectual resources to advance this suggestion, one may wonder why they did not favor it. The answer may be that they realized that it is a bad explanation because it fails to address the nature of religious beliefs as beliefs. What needs to be accounted for is why people actually hold to certain ideas and engage in particular practices, and part of that explanation will involve their beliefs about the point and value of those religious notions and practices. The fact that ancestors behaving in related ways enjoyed certain reproductive benefits in consequence, hardly touches the issue. Or so one might think.

Daniel Dennett, who teaches at Tufts University, is a striking advocate of evolutionism: a highly talented, highly voluble professor of mind, reason, and science, determined to expose religion as a product of purely natural and entirely unpurposeful forces; and then to suggest that whatever benefits religion might once have conferred, the phenomenon as we have it is best dispensed with--and the sooner the better, for it is a cause of needless misery.

Dennett has been a significant presence on the philosophical scene since the publication of his first book (Content and Consciousness) in 1969, and in the intervening years he has developed highly distinctive lines of thought about the nature of mind and consciousness. Given the subtlety of his work, any brief characterization risks oversimplification, so let me just leap in with a two-word summary: Dennett is an iconoclastic demytholigizer. Let any of the great themes of metaphysics and religion pass before his gaze: consciousness, rationality, free will, the soul, God, and he will be at them tearing away until no vestige of the metaphysical remains. For Dennett, all that there (really) is, is matter in motion--though as he would no doubt say, that is no small show.

Materialism can be maintained as a quiet intellectual conviction, shared among like-minded scholars, but Dennett is an evangelist, and having recognized what he believes to be the truth he wants to shout it on the highways and byways, and will not rest content until no one will be able to persist in one's error with the excuse that one has not heard the good news.

And the gospel is as follows:

1. "Religions are social systems whose participants avow belief in a supernatural agent or agents whose approval is to be sought."

2. "Religion is a human phenomenon composed of events, organisms, objects, structures, patterns, and the like that all obey the laws of physics or biology, and hence do not involve miracles."

3. "If we want to know why we value the things we love, we need to delve into the evolutionary history of the planet, uncovering the forces and constraints that have generated the glorious array of things we treasure. Religion is not exempt from this."

4. "Religion is a human phenomenon, it is a hugely costly endeavor, and evolutionary biology shows that nothing so costly just happens ... the ultimate measure of evolutionary 'value' is fitness, the capacity to replicate more successfully than the competition."

5. "Cultural transmission can sometimes mimic genetic transmission, permitting competing variants to be copied at different rates, resulting in gradual revisions [that] have no deliberate foresighted author."

6. "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."

7. "The memorable nymphs and fairies and goblins and demons that crowd the mythologies of every people are the imaginative offspring of a hyperactive habit of finding agency wherever anything puzzles or frightens us. ... The [myths] that get shared and remembered are the souped-up winners of billions of competitions for rehearsal time in the brains of our ancestors."

8. "Innate curiosity, stimulated by music and rhythmic dancing and other forms of 'sensory pageantry' could probably account for the initial motivation to join the chorus."

9. "[Religion] is a finely attuned amalgam of brilliant plays and stratagems, capable of holding people enthralled and loyal for their entire lives."

10. "Free-floating rationales that are blindly sculpted by earlier competitions can come to be augmented or even replaced by represented rationales ... [ones that are] used--argued over, reasoned about, agreed upon."

Having developed this series of steps, Dennett then moves to consider two questions: first, whether folk religions, and the organized religions they have become, have conferred fitness benefits on their practitioners; and second, whether, religions are morally deserving of allegiance and service. Since the announced purpose of the book is to "break the spell" of religion, which has hitherto protected it from serious scientific analysis and critical evaluation, and to expose it to empirical testing, Dennett makes a point of saying that the answers to these questions now remain to be determined. But that unassuming nicety observed, it is pretty clear that he has come to a conclusion and that he feels that open-minded readers will share it: good motives and occasional benefits allowed for, religion is a bad thing, and the good sometimes associated with it could be got otherwise.

The book is a brick (450 pages) possibly suitable for breaking church windows--but was it designed as such? In dealing with artifacts one needs to distinguish between intended purposes and incidental effects. I assume that Dennett intended his book to be read not thrown; but it can be thrown and its use as a missile might confer reproductive benefits on those who so deploy it. The logic of evolutionary explanations requires, though, that one discriminate between heritable adaptations that have been selected for per se because of the advantages they confer, and incidental byproducts selected per accidents. It may have been the case that objects of the rough size, shape, and weight of Dennett's book were used by his ancestors and are used by his contemporaries as missiles, but that doesn't settle the question of the actual point of the book. Is its being a book, or a potential weapon, an intended or incidental feature of its manufacture? I shall assume that the answer is clear.

Likewise with religion. Whatever benefits may have attached, or do attach, to engaging in certain forms of behavior, the question we need to address in understanding religion is the meaning and value of that behavior, and its associated beliefs and values, as religious expressions, not as evolutionary adaptations. Readers should already have a sense that there is (as there must be) much of pure conjecture in Dennett's speculations about our ancestral past, but even when these make for interesting reading they are beside the point so far as concerns the obvious issues raised by religious claims, namely, what do they mean and are they true?

Dennett is too bright simply to have missed this, but he is driven by the conviction that religious claims are not true, and are even of dubious coherence; therefore, confronted by the fact that they are so widely voiced, he assumes that they must have some other kind of explanation. Consider by analogy the situation of an adult who reports each morning that a golden fairy visited her in the night and told her stories about the building of a fairy queen's silken castle. To engage in discussion with the woman about the details of the conversations would be at most a calming move preparatory to seeking a clinical diagnosis. Most people would not even consider the possibility that the narrative was factual--and anyone who did would themselves be subject to the presumption of disorder.

Dennett does consider whether religious beliefs may be true, but the brevity, barely six pages, and the dismissive character of the discussion speak volumes. The stuff of supernaturalism is to him no more credible than that of fairyism, indeed it is probably less so since he might observe that unlike fairyism, Godism is not experimentally testable. This is not the occasion to respond, yet it is important to assure readers who may feel the weight of Dennett's prose pressing down upon them that the pressure is eminently resistible. Often witty, sometimes jovial, and occasionally tender, Dennett is also somewhat domineering, hardly allowing the reader pause to take breath or to develop responses.

He writes in the preface that the book is addressed primarily to American readers, "the curious and conscientious citizens of my native land--as many as possible, not just the academics." Later he writes of "religious folk" as if speaking to members of a rural congregation; but he can't really imagine that the folk will read this, or that if philosophically trained believers do so, they will be unsettled by it. My guess is that the intended purpose of the book is to say a lot out loud, with the hoped-for incidental effect that others may listen. In truth, though, this is a personal testament unlikely to win many converts.

So, we end where we began with the fact of billions of believers and with the ancient suggestion that religion is a natural response to the universal sense of being in a world created and governed--by what and to what end one does not quite know. Whether that sense of creation and creatureliness is warranted is indeed a question for investigation, but it calls for attention to the content of religious claims and requires some assessment of their plausibility. The spell of religion is the sense that those claims, however problematic, register something of the truth about the nature of human beings and their place in the cosmos. However evolution proceeds, so long as humans continue to experience and reflect, then religion will remain a pervasive aspect of their existence.

John Haldane is professor of philosophy at the University of St. Andrews. Among his books is Faithful Reason: Essays Catholic and Philosophical (Routledge). Funding for this review was provided by a grant from the Henry Luce Foundation.