You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.
Skip Navigation

The Celestial Teapot

Letter to a Christian Nation

By Sam Harris

(Alfred A. Knopf, 96 pp.,



I have not believed in God since I was fifteen, and now, at forty, I suspect that I am too late to change. But the velocity of that flight from belief has not been constant: there have been hesitations, interruptions, acute nostalgias. Like many raised in a religious household, I often find myself caught in a painful, if comic, paradox, whereby I am involved in an angry relationship with the very God whose existence I am supposed to deny. There is the joke of the atheist out fishing with a believing friend. Theatheist casts his net and draws up a stone on which is carved: “I do not exist. Signed: God.” And the atheist exclaims: “What did I tell you!” Contradictory this kind of atheism might at times be, but those contradictions feed, perhaps constitute, its brand of militancy; it is because God cannot be entirely banished that one is forced to keep on complaining rather than merely finalize one’s elegies.

This is the kind of atheism that Dostoevsky was interested in, the kind that stands on the ladder just one rung below belief. For me,though, the gap between those top two rungs was always large, and is now ever wider. I was brought up in a Christian environment that had retained more than a memory of nineteenth-century evangelicalism. It was not fundamentalist, nor was it literalist: my father was a professor of zoology, a rationalist, a very good scientist. But it was scriptural, with a great burden of meaning and import placed on the Gospels, and on Christ’s revolutionary challenge to Nicodemus: one must be born anew, of the spirit rather than the flesh.

Ordinary language was saturated in religiosity. A happy occurrence was a “blessing” or was “providential”; an unhappy one might well have been the product of “sinful” or “unedifying” behavior, andthis behavior was almost certainly “unscriptural.” An untidy bedroom was evidence of “poor stewardship.” I was encouraged not to wish people “good luck,” this being rather secular. The word I heard most often, of course, was “faith,” since none of the other words could have functioned without it. I was fascinated when from time to time my parents would discuss, in hushed tones, an acquaintance who had “lost his faith. “ The phrase, so solemnly unsheathed, seemed to point to unimaginable wildernesses.

I did not lose my faith so much as hand it on, since I never really possessed much of one. This, in an admittedly rather circular way, is the first reason that I am not a believing Christian: the roots of that belief were always relatively frail, and gave way easily enough when pressure was applied. Unlike others in my family, I had no moment of great conversion or revelation. In addition, a number of other powerful disincentives were presenting themselves to me. I vividly remember the day I sat down with a piece of paper and drew a line down the middle: on one side I would compile my reasons to believe and on the other the reasons not to. Perhaps this was rigged—anyone who does something like this has already lost his faith, well before the pretended ratiocination. It would certainly be hard to imagine anyone led to believe in God by such a method.

On the debit side was: God’s failure to answer prayers; the worldwide varieties of religious experiences and traditions, and a feeling that they could not be compatible; and, overwhelmingly, the difficulty of reconciling God with the reality of evil. The standard atheist package, alas. My childhood, which was a happy one, had been marked by my witnessing several deaths from cancer, friends of my parents who were members of their church. Great efforts were made, in the usual charismatic or evangelical way, to save these people. Prayers were raised and raised. Hands were laid on ailing heads, oil was poured. But when it was time for these people to die, they died. After years of hearing thousands of petitions offered to the Lord, I cannot recall a single answered prayer.

How would you know, asks the believer, since God’s ways are inscrutable to us? But prayer is one of those cases where an inscrutability argument will not work, because one knows what one has oneself requested, and therefore what has been denied. If you pray for a member of your congregation to get better and she dies, your prayer was not answered. To retort that God’s mysterious way of answering your prayer—“but God needed her by his side in heaven, that’s why he let her die”—might involve not really answering your prayer at all is essentially to nullify prayer, to kill it. I knew that at fifteen. Years later I read Samuel Butler’s The Way of All Flesh, with its extraordinary image of the futility of prayer: a bee, inside a room, mistaking the floral wallpaper for the real thing and briefly attempting to extract its illusory pollen.

More devastating still to belief was the apprehension that there was a grotesque amount of suffering in the world, and that I could not reconcile this with any of the powers or qualities usually ascribed to God. Even Cardinal Newman, in that terrible, beautiful book Apologia Pro Vita Sua, acknowledges this crisis:

I look out of myself into the world of men, and there I see a sight which fills me with unspeakable distress. The world seems simply to give the lie to that great truth, of which my whole being is so full; and the effect upon me is, in consequence, as a matter of necessity, as confusing as if it denied that I am in existence myself. If I looked into a mirror, and did not see my face, I should have the sort of feeling which actually comes upon me, when I look into this living busy world, and see no reflexion of its Creator. This is, to me, one of those great difficulties of this absolute primary truth, to which I referred just now. Were it not for this voice, speaking so clearly in my conscience and my heart,I should be an atheist, or a pantheist, or a polytheist when I looked into the world.

Without the succor of Newman’s “voice, speaking so clearly in my conscience and in my heart,” I can see only a world without a Creator, and a set of intractable cruxes. They are familiar to anyone who has read any theodicy. The existence of great pain and evil in the world limits God’s power or qualifies his goodness. If he allows enormous suffering, then he is not good, and if he cannot stop this suffering, then he is not very powerful.

Orthodox theology admits that it cannot “solve” this dilemma. Still,over the centuries all sorts of arguments, some of them eloquent and some merely glib, have been advanced to attempt such a solution. Ancient commentators and believers were tempted to relieve God of some of his power rather than his goodness, by positing a rival lesser God who is responsible for the world’s woes—a Satan, or a false God, a demiurge. This is what Newman meansby “polytheism.” The Old Testament, despite Yahweh’s obsession with monotheism, dangles this possibility from time to time, most famously at the beginning of Job, where God arranges with Satan to test their poor victim. Gnosticism and Manichaeanism are the best-known versions of this recourse to dualism.

Theology’s two most reliable offerings at the bazaar of theodicy are the Soul-Making Theory and the Free Will Defense. The former argues that suffering is some kind of mysterious training, inseparable from worldly existence, and finally a good in its own right. Inheaven, anyway, God will “wipe away all tears from their eyes,” as the Book of Revelation has it. Dostoevsky seems to have believed this (he was obsessed with that phrase), and Simone Weil, in her essay on affliction, seems to argue that pain is a kind of apprenticeship. When an apprentice hurt himself on the job, she writes, it used to be a saying among workmen that “it is the trade entering his body.” But it is not clear why there is so much suffering in the world (this is sometimes called “dysteleological suffering”); and the cost of accepting this argument seems to be that God must seem, however lovingly corrective, unbearably cruel.Pierre Bayle, the late seventeenth-century skeptic, likens such a God to a father who lets his son break his leg so that he could have the educative experience of mending it.

The Free Will Defense suggests that for us to act as moral agents we must have free will, and that as soon as we have free will we will abuse it. A world without such freedom would be unimaginable,robotic; moreover, in order for our relations with our Creator to be morally meaningful, we must have the freedom freely to grow and contract in that relationship. We must live in a dappled world, and we can only really conceive of existence in terms of such contrasts—light and shade, good and evil, obedience and disobedience. Free will, to such defenders, is the highest good,higher than human happiness. Again, though, the objection can surely be made that we seem to have so much freedom, more than we need. Origen argued that Judas’s evil made possible the good of Jesus’s crucifixion, and from here it is but a step to the proposition that Hitler’s evil allowed various righteous gentiles to act righteously, and from here it is but a step to the proposition that Hitler’s evil is itself a good thing because it allowed Hitler himself the exercise of free will—an argument made in our day by Richard Swinburne, an Oxford theologian. It is the kind of “move” that gives academics a bad name.

What if we had been so constituted that our freedom to act wickedly existed along a narrower spectrum of possibility? Thieves, bullies,horrid husbands, liars, nasty little onanistic boys, corrupt politicians, stealers of apples and pears and knowledge, defilers of Eden—but no Hitlers, no Stalins, no murderers, no child-killers? Obviously, we cannot imagine what this limited freedom would look like: it does not, in fact, look much like our idea of freedom. But if God had made such a world, we would never have had any experience of our more murderous freedoms, and we would not lament their lack. Such a world would not be without any pain; death would still have its sting. But suffering would have had much of its sting pulled. There would not be so much simply monstrous suffering in the world. Not possible, you say? But as Schopenhauer retorts to Leibniz, even if the world we currently inhabit were the best of all possible worlds, God created both the world “and also the possibility itself, accordingly he should have arranged this with a view to its admitting of a better world.”

Actually, the more complete confounder of the Free Will Defense is the concept of heaven (less important to Judaism than to Christianity and Islam). In heaven, it seems, all tears will be wiped away and we will be free of pain and suffering. We will also be free of freedom—necessarily so, because if freedom were to exist in heaven, we would merely replicate our lives on earth and start doing terrible things to each other again. Heaven, as an intellectual category as much as an “actual place,” depends on the idea that the highest form of happiness—to be face to face with our Maker, and so on—is a state without freedom, or with severely curtailed freedom. But if this is the ideal state, the state tha tour Creator longs to have us in, then why was heaven not instituted on earth? Since heaven was not created on earth, we must conclude that our lives here are more or less painful experiments, and that the world is a training ground for heaven.

Yet it is a rigged experiment, since the experiment already knows its own answer. Not just because God, being omniscient, must know what will become of each of us (the Catholic church tied itself up in knots over this issue, and eventually had to repudiate its own doctrine of “double predestination”), but also because a real experiment would put the existence of heaven itself in doubt. Arigged experiment simply puts our going to heaven in doubt. Yet if heaven must exist, if there is no doubt that heaven exists, then we know that we are being trained here on earth to exercise a free will that will not be needed in heaven, a free will the exercise of which causes immense pain to many people, but a pain that will be miraculously eased in heaven. This is nothing less than a definition of torture. (Though presumably the likes of Richard Swinburne would argue that seventy years of torture versus an infinity of heavenly bliss is a “reasonable” experiment.) Heaven is not and never has been the solution to theodicy; heaven is the very problem.

As a teenager, of course, I did not have the philosophical geometry set to map out this horror with the proper lucidity. But I blundered my way to these arguments, and every way I turned I walked into a dark wall. I could not continue to describe this Godin the ways my tradition insisted I should, and Jesus’s softening of Yahweh’s austere awfulness in some ways only aggravated the problem. The Old Testament God, at least, shrinks from attributes,and a vast gulf must separate his unspeakability from our understanding. But Jesus, through his incarnation of God, takes on the very qualities that cannot be attributed to his father:merciful, loving, wise. Above all, he intervenes: he performs miracles, he raises the dead, he is himself raised, and he promises intercession—his own, and the Holy Spirit’s. Christ is nothing if not describable. This is the central joy for Christians. Yet the Creator he incarnates is not describable in terms that would make any sense of his providential creation and control of the world.Christ, to me, seems orphaned of the very patrilineage that constitutes his bold appeal. He incarnates what cannot be incarnated. What use is his sonship if his father is lost? To be blunt: to worship Christ, it seemed to me, was to worship the bastard child (in the strict sense of the word) of an absolute bastard (in the vernacular sense of the word). And never forget that Christ came and nothing changed: salvation had to be pushed on again, infinitely deferred to a Second Coming. “Silence, that only Voice of our God,” writes Melville in Pierre, and “how can a man get a Voice out of Silence.”

So what, though, if God cannot easily be described as omnipotent,loving, and omniscient? Maimonides felt that it was only our human limitations that led us into trying to pin attributes onto God. We cannot comprehend his essence, and merely hurl human construction sat him. We are like the men, he says, who praise an earthly king with lots of gold coin for owning millions of silver coin. It is more becoming to define God by what he is not, and to reflect in silence rather than to utter our incomprehensions. “Silence is praise to Thee,” Maimonides approvingly says, quoting Psalm 65, andas if replying to Melville centuries before.

Yet for all the magnificence of its reasoning, this Aristotelian coolness, which Aquinas to some extent shared, leaves me cold.Suspended between Maimonides and Melville—a fine place to be suspended in any other context—we forlornly watch a describable God vanishing into a horizon of negatives. This God is silent, and does not speak to us; and we are silent too, enjoined not to speak to him. He is just the God of the new physics. Above all, though we might come to respect and certainly fear and obey this God, why would we ever worship him? If we know him only by his non-ness,then perhaps non-worship, along with non-love and non-recognition,is the appropriate response? Wittgenstein’s little phrase, from a different context, comes to mind: “a nothing will serve just as well as a something about which nothing can be said.” And what do wedo with our traditions, which insist on telling stories about,insist on describing, this God about which nothing can be said? Of one thing we can be absolutely certain: if God exists, then he—it,she—really is nothing like the various representations of him—it,her—that our sacred books contain. If God really exists, he is beyond love, beyond worship, beyond reach. A silence beckons to a silence. But if that is the case, our attachment to our religious traditions and descriptions, however noble their approximations,traditions over which we have spilled and continue to spill so much blood, is sentimental tribalism and one of the greatest tragedies in history.


We are in the midst of that tragedy, and America is drowning in God’s attributes. The Lord will increase your salary, teach your children, raise your self-esteem, boost your career, be a lifelong friend, and take you into his heart if you only take him into your heart. He is love, and gentleness, and charity, unless he is for bidding homosexuality or stem-cell research or punishing New York with September 11 for its high proportion of gays, lesbians,and degenerates. He greatly dislikes evolutionists, largely because he created the world six thousand years ago. He certainly dislikes Nancy Pelosi—and now, alas, Pastor Ted Haggard. The Bible is his inerrant word. According to recent polls, 53 percent of Americans are creationists, and 87 percent—or 260 million people—claim to”never doubt the existence of God.” An avowed atheist cannot be elected president. And so on. You know the stupefying recital. Many millions across the world are absolutely sure they know what God is like, and what he likes. Heine’s unbelieving joke, reported by the Goncourt brothers, rises up: on his deathbed, while his wife was praying that God might forgive him, he interrupted her to say,”Have no fear, my darling. He will forgive: that’s his profession.”

The rise of evangelicalism, and the menace of fundamentalism, alongwith developments in physics, and in theories of evolution and cosmogony, has encouraged a certain style of public atheistic critique. Many of these names are well-known: Richard Dawkins,Daniel Dennett, Sam Harris. The events of September 11 were the direct spur for Harris to write his best-selling book The End of Faith, which vibrated with an admirable anger. It has a suggestive thesis, too, which is that America cannot possibly fight fundamentalist Islam while it is itself gripped by Christian fundamentalism. This symmetry of fundamentalisms means that America will not stoop to defeat the religious content—and dangerous idiocy—of its foes. I am not sure if this is exactly provable.Britain, for instance, almost 40 percent of whose citizens professnot to believe in God, has not yet mobilized its secularism in victorious ways (though Harris would doubtless point to Tony Blair’s strong Christianity). But it is not his job to win the so-called war on terror, and the essential intellectual approach seems right: attack all the troops of irrational religiosity at once.

The End of Faith starts well and then becomes a bit predictable,because it begins to follow the rules of its rather thin genre.Letter to a Christian Nation, which is an open letter to the many Christians who wrote to Harris in complaint, is even thinner. I have an almost infinite capacity for the consumption of atheistic texts, but there is a limit to how many times one can stub one’s toe on the thick idiocy of some mullah or pastor. There is a limitto the number of times one can be told that the Bible is a shaky text, and that Leviticus and Deuteronomy are full of really nasty things. Ratio vincit omnia, but the page-by-page demonstration of this rationalist conquering can become wearisome. This may be no especial insult to Harris so much as to his family; Bertrand Russell’s Why I Am Not a Christian made a great initial impact on me when I was a teenager—it was like seeing someone in the nude,for the first time—until I began to get bored with its self-exposure. Russell complaining that Jesus was not a moral teacher, that he was really rather a bad example because he threw the money lenders out of the temples and cursed the fig tree,seemed somehow a little undignified. Russell is reliably at his least philosophical when he is at his most atheistical.

The genre tends to proceed thus: the atheist must first remove all possible respect from religious belief. The tone is a little perky,and lively thought experiments bloom. They go a bit like this: if I told you that President Bush prays every day to his vacuum cleaner,you would judge him insane. But why is there any evidence that the God he prays to exists? It is fun, knockabout. Harris likes to compare belief in God with belief in Wotan or Zeus: “Can you prove that Zeus does not exist? Of course not. And yet, just imagine if we lived in a society where people spent tens of billions of dollars of their personal income each year propitiating the gods of Mount Olympus.”

The model is Bertrand Russell’s “celestial teapot,” gleefully quoted by Richard Dawkins in The God Delusion. If, says Russell, I told you that a celestial teapot was orbiting the sun but that you could not see it, nobody would be able to disprove me; “but if I were to go on to say that, since my assertion cannot be disproved, it is intolerable presumption on the part of human reason to doubt it, I should rightly be thought to be talking nonsense.” God is like the teapot, we are supposed to infer. Dawkins uses Russell to argue that we cannot prove God’s non-existence, but then we cannot prove anything’s non-existence. “What matters,” writes Dawkins, “is not whether God is disprovable (he isn’t), but whether his existence is probable.... Some undisprovable things are sensibly judged far less probable than other undisprovable things.”

I agree with Dawkins’s conclusion, and consider God highly improbable, but I dislike the way he gets there. It seems to occur neither to him nor to Russell that belief in God is not like belief in a teapot. The referent—the content of the belief—matters here.God may be just as undisprovable as the teapot, but belief in God is a good deal more reasonable than belief in the teapot, precisely because God cannot be reified, cannot be turned into a mere thing,and thus entices our approximations. There is a reason, after all,that no one has ever worshiped a teapot: it does not allow enough room to pour the fluid of our incomprehension into it.

Interestingly, Dawkins himself seems to agree with this complaint.In a recent conversation in Time with the geneticist Francis Collins (who is a believing Christian), a conversation in which both men spoke eloquently, Dawkins was pushed by Collins to admit that, in Dawkins’s words, “there could be something incredibly grand and incomprehensible beyond our understanding.” That’s God,said Collins. Yes, but it could be any of billions of Gods, replied Dawkins: “the chance of its being a particular God, Yahweh, the God of Jesus, is vanishingly small.” In other words, the God of a particular scripture and tradition is a parochial and inherently improbable notion. But the idea of some kind of creator, said Dawkins, “does seem to be a worthy idea. Refutable—but nevertheless grand and big enough to be worthy of respect.” To which one should add: by definition, then, this “grand and big” idea is not analogically disproved by referring to celestial teapots or vacuum cleaners, which lack the necessary bigness and grandeur.

My inner atheist nevertheless enjoys the “naughtiness” of this disrespect, even if a little of it goes a long way. And all these writers are correct to argue that religion is unfairly protected by a cordon sanitaire of “respect.” In America, all you need to do isintone the word “faith” and your opponent will start backing away from you in terror, like a vampire before a crucifix. In these books the vampire bites back, and Harris has an Orwellian robustness and a good journalistic way with his one-liners. To the creationists who believe that the world is six thousand years old,he says: “This is, incidentally, about a thousand years after the Sumerians invented glue.” The principal concern of American Christians “appears to be that the creator of the universe will take offense at something people do while naked.” Twenty percent of all recognized pregnancies end in miscarriage, he writes: “if God exists, He is the most prolific abortionist of all.”

Having set fire to religion’s firewall of respect, the genre moves on to point out that millions of people believe that their religious traditions are the right ones, and that they cannot all be right. Then it is time to dismantle the texts, usually by pointing to historical inconsistencies and moral outrages, such as the Bible’s silence on slavery, or God’s instruction that heretics,adulterers, and homosexuals should be stoned. From here it is a merestep to arguing that we do not need divine laws—especially given their historically specific nature—in order to be good. Atheists are not wicked, and several studies suggest that they may even be a bit less wicked than religionists. By the way, though Stalin was an atheist, Hitler probably was not. Nazi anti- Semitism was a Christian monster, at bottom, and received plentiful support from the churches. Dawkins and Harris say almost exactly the same thing here. Harris makes the perfectly good point that

Auschwitz, the Soviet gulags, and the killing fields of Cambodia are not examples of what happens to people when they become too reasonable. To the contrary, these horrors testify to the dangers of political and racial dogmatism. It is time that Christians ...stop pretending that a rational rejection of your faith entails the blind embrace of atheism as a dogma.... The problem with religion—as with Nazism, Stalinism, or any other totalitarian mythology—is the problem of dogma itself. I know of no society inhuman history that ever suffered because its people became too desirous of evidence in support of their core beliefs.

This may be a little self-serving, in that “reasonable” is herebeing defined by what it excludes: it cannot be reasonable to be aNazi, and therefore Nazism is by definition unreasonable. Butcommunism, of course, put great stock in scientific materialism, in the reason of historical progress. The argument against dogma is quite just, though, and communism can certainly be seen as a rival religion, with messianic tendencies.


This brand of public atheism is very good at the necessary disrespecting of religion, and it has a properly hygienic function.But how worthy of respect is it itself? The problem is that its bright certainty about the utter silliness of religion leads very quickly away from philosophy and argument. There is a dismaying gap, in these books, between the righteous anger of the critique of the many absurdities of religious belief and the attempts to account for why people have believed this apparent nonsense for so many centuries. I would rather that these writers refrained from speculation altogether than plunge into their flimsy anthropological kit bag. It is peculiar indeed to read Dawkins’s eloquent pages on evolution, and on how evolution may in the end solve the question of who created us, and then to find that very evolutionary theory being applied in the most hypothetical,rampantly unscientific ways to the question of why we have believed in God for so long.

For Dawkins, it may all be explained by our evolutionary need to fall in love, or perhaps by our childish need to have a big friend.At the same time, we have also evolved a HADD, a “hyperactive agent detection device”: “we hyperactively detect agents where there are none, and this makes us suspect malice or benignity where, in fact,nature is only indifferent.” (Daniel Dennett is also fond of the argument from HADD.) Dawkins’s example of this tendency is a moment in Fawlty Towers when John Cleese’s car breaks down. Cleese, drunk with HADD, one supposes, starts thrashing his car to death. Dawkinstruly appears to think that this high-table guffawing will do as an explanation of why thousands of generations have been drawn to believe in God. And mystical experience of the divine does not detain him, either. We have evolved superb “simulation software in the brain,” which is “especially adept at constructing faces and voices.... It is well capable of constructing `visions’ and`visitations’ of the utmost veridical power. To simulate a ghost oran angel or a Virgin Mary would be child’s play to software of this sophistication.” And he concludes: “That is really all that needs to be said about personal `experiences’ of gods or other religious phenomena.” Evolutionary biologists never seem happier than when they are talking about humans as crafty but malfunctioning computers, with “toolkits” and “menus” and “software.” The possibility that this might itself be a mad “vision,” an example of a highly evolved Oxonian computer on the blink, does not occur to Dawkins’s own simulation software.

The emphasis on evidence, on provability and probability, is an inevitable part of the cleansing rationalism of these books, and is not always unwelcome. Dawkins is careful to talk about the improbability of God’s existence, not his impossibility, though Harris rather too easily lapses into talk about “insufficient evidence”: “It is undeniable that people of faith make heroic sacrifices to relieve the suffering of other human beings. But is it necessary to believe anything on insufficient evidence in order to behave this way?” Yet he surely knows that we believe all kinds of things on insufficient evidence. Or rather, what might be sufficient evidence to him could well be insufficient to someone else and vice versa. Hume was right, says the French philosopher Alain, “to mock the King of Siam who believed that ice wasimpossible because he had never seen it.” Harris might say that he has sufficient evidence in order to believe in the laws of physics because they have never failed him, but that level of evidence-satisfaction would seem pretty low to a physicist. On what evidence, for instance, does he believe that the universe is expanding?

We have to acknowledge that most religious language cannot be tested for its provability by a philosophical rationalism that anyway defines the terms of that provability. Religious language, as Wittgenstein never tired of pointing out, is a practice, not an experiment; its referents are defined by how it is used. There are grammatical differences between the use of religious language and ordinary language. One might finally disagree with Wittgenstein,and certainly some of his more irritating followers have wielded his idea of language games as a way of refusing any demand to justify belief in God. I think that Wittgenstein’s notion seems an excellent way of defending a way of life, an embedded habit of religious practice—say, kissing an icon—that already exists; but it is hard to see how any novice, who had never believed, would beled to adopt a religious practice by Wittgenstein’s language. In his world, one seems always to be born into such practices.Conversely, people do indeed lose their faith because they cease to believe in certain propositions (I did); Wittgenstein always seems to treat as lunatic anyone who would want to stop kissing anicon—for him this would be tantamount to wanting to leave one’s family and change one’s surname—because he cannot imagine a propositional content to religion.

These are not easy questions, then, but the jauntily unphilosophical way in which most popular atheistic writing simply ignores the Wittgensteinian dilemmas is disappointing, and explains why its explanations of the sources of religious belief are so jejune. Is there not a kind of insult to language in so comprehensively banning the incomprehensible? Shouldn’t a physicist—a friend of mine—be able to say that, for her, Coltrane’s A Love Supreme “is God,” without atheism busily correcting her lexical lapse into the unprovable? Dawkins says of James Frazer’s The Golden Bough, “Read such books and marvel at the richness of human gullibility,” as if that solves that. Sam Harris gets himself into a telling knot in The End of Faith, when he attempts to float a kind of vaguely Eastern, vaguely New Agey form of meditation. (The dirty secret of that book is that Harris turns out to be a Buddhist.) He dislikes having to use words like “spirituality” and “mysticism” because they have “unfortunate associations.” But use them he does,explaining that mystical meditation makes us happy and is good forus, and suggesting that we should do it from time to time. Of course, he thus falls into the very consequentialism that he dislikes in some religious discourse (the kind that says that you should believe in order not to be sinful). Perhaps realizing this,he explains that his kind of mysticism “is a rational enterprise. Religion is not.” He continues:

The mystic has recognized something about the nature of consciousness prior to thought, and this recognition is susceptible to rational discussion. The mystic has reasons for what he believes, and these reasons are empirical. The roiling mystery of the world can be analyzed with concepts (this is science), or it can be experienced free of concepts (this is mysticism). Religion is nothing more than bad concepts held in place of good ones for all time. It is the denial—at once full of hope and full of fear—of the vastitude of human ignorance.

But this rational mysticism seems a pretty poor substitute for the grandeurs of religious mysticism, however one judges the latter’s empirical content. Harris is welcome to sit on his floor and get off on his Buddhism; I’ll go and sit in a cathedral.

And what does it mean to say that the mystery of the world can be analyzed with scientific concepts? Harris knows as well as anyone else that it is precisely the ultimate mysteriousness of the universe that science has so far failed to explain. The Big Bangtheory has gone far in explaining—no, in proving—how the universe first began to expand from an infinitesimal compacting of matter, a theory foreshadowed, in fact, in the speculations of some of the very irrationalists disdained by atheism: ancient and medieval philosophy and theology boldly discusses the idea that space and time were created together. But about the very moment before this first expansion, and the conditions that made it possible, there is only speculation. About this moment, we are all in the dark, and silence calls to silence.