Hard Questions · Answered Honestly

Where Did the Idea of God Come From? C.S. Lewis, the Numinous, and Why Fear Cannot Explain Religion

The standard secular story says early humans invented God for two reasons: fear of a dangerous world, and the wish for a comforting one. C.S. Lewis thought the actual evidence of religious experience refutes both explanations at once — and he built his case on a distinction as old as the human race: the difference between fearing a tiger and dreading a ghost.

The Objection, Stated Fairly

As always, we begin by giving the skeptical position its full strength. The claim that religion originates in fear is ancient — the Roman poet Lucretius argued in De Rerum Natura that terror of nature’s violence first put gods into human minds, and the maxim primus in orbe deos fecit timor (“fear first made gods in the world”) circulated in antiquity. Its modern forms are more sophisticated. On the danger hypothesis, early humans facing predators, storms, disease, and death personified the forces that threatened them, imagining agents behind the lightning who might be appeased. Cognitive scientists of religion refine this further: human minds evolved a hypersensitive tendency to detect agency — better to mistake a rustling bush for a predator than a predator for a bush — and gods, on this account, are false positives of that survival circuitry. On the wish-fulfillment hypothesis, most famously developed by Ludwig Feuerbach and then Sigmund Freud in The Future of an Illusion (1927), religion is the projection of human longing: helpless before nature and haunted by death, we dream a cosmic Father who protects us, and call the dream God.

Both explanations share a structure: the idea of God is a natural byproduct of the mind’s machinery, requiring no supernatural input, and understanding its origin dissolves its authority. It is a serious position held by serious people. And C.S. Lewis — who as an atheist had held it himself — came to believe it fails at precisely the point where it should be strongest: it cannot account for what religious experience, at its root, actually is.

The Tiger and the Ghost: Lewis on the Numinous

Lewis opens The Problem of Pain (1940) with the observation on which everything else turns. Consider what happens in the mind, he says, under three announcements. Told there is a tiger in the next room, you feel fear — ordinary, biological, well-founded fear of a creature that can maim and kill. The emotion is entirely explained by the danger. Told instead there is a ghost in the next room, you feel something categorically different. You may believe the ghost can do far less to your body than the tiger — and yet the feeling is, if anything, more disturbing. It is dread: a shrinking, a sense of inadequacy, an unease that has nothing to do with claws or teeth, because it is not a response to what the thing might do but to what the thing is. And told, finally, that a mighty spirit — a presence — is in the room, the feeling deepens into something for which “fear” is the wrong word entirely: awe, wonder mixed with trembling, the profound sense of being small before something wholly other. This third experience is what Lewis, borrowing the term, calls the numinous.

The word itself comes from the German theologian and philosopher of religion Rudolf Otto, whose 1917 study Das Heilige (translated as The Idea of the Holy) remains the classic analysis. Otto named the object of this experience the mysterium tremendum et fascinans — the mystery before which one trembles and toward which one is nonetheless drawn. He argued that the experience carries a distinctive signature: what he called creature-consciousness, the felt smallness of the finite self in the presence of that which is ganz andere — “wholly other,” not merely a bigger or stronger item within the world, but something of a different order than the world. And Otto’s central claim, the one Lewis takes up, is this: the numinous is sui generis. It is a primary, irreducible category of experience. It cannot be constructed out of anything else — least of all out of ordinary fear.

Why the Danger Hypothesis Fails: The Unbridgeable Jump

Here is the argument’s logical core. The danger hypothesis requires a smooth progression: early man feared the tiger, then feared the storm, then imagined an agent behind the storm, and the fear of that agent became religion. But examine each step. Fear of the tiger and fear of the storm are the same kind of state — aversion to physical harm — and the imagined storm-agent, if he is merely a bigger tiger in the sky, should evoke merely more of the same state. What he cannot generate, by any amount of magnification, is dread of the wholly other. A being who is simply very dangerous is still an object of prudential fear; one hides from him as one hides from the tiger. The numinous is not prudential. The ghost, Lewis’s middle case, proves it: the dread it evokes is disproportionate to and independent of any danger it poses. Something else entirely has entered the experience.

This is why Otto maintained that one cannot logically progress from everyday fear of physical dangers to supernatural awe, and why Lewis described the move from the one to the other as a “jump” — a leap to something not contained in, and not derivable from, anything observed in the natural, visible world. Perceived danger explains the emotion of fear completely; nothing in the visible world explains the emotion of the numinous at all. Its object is not given in nature. And yet the experience is among the most widely attested in the human record — in the earliest religious texts, in the terror of Jacob at Bethel (“How dreadful is this place!” — Genesis 28:17), in Isaiah undone before the throne (Isaiah 6:5), in Job silenced out of the whirlwind, in the disciples more terrified by the calmed sea than the storm (Mark 4:41), and in the anthropology of peoples on every continent. Either the human race has always been subject to a systematic hallucination with no natural object, or the experience is what it presents itself as being: a perception — however dim, however distorted — of a reality genuinely there.

Lewis put the dilemma plainly: the numinous is either a mere twist in the human mind, corresponding to nothing and serving no biological purpose, or it is a direct experience of the really supernatural, to which the name Revelation properly belongs. What it cannot be is an outgrowth of ordinary fear — because it is not, in its essential character, fear at all.

Why Wish-Fulfillment Fails: Nobody Wishes for the Numinous

The Freudian alternative fares no better, and for a reason that becomes obvious the moment the numinous is described accurately. Wish-fulfillment predicts a religion of comfort: a grandfatherly protector, a cosmic guarantee that all shall be well, a projection shaped by what frightened creatures want. But the God actually encountered at the root of religion is the mysterium tremendum — the presence before which Isaiah cries that he is undone, before which Job puts his hand over his mouth, before which Peter begs, “Depart from me; for I am a sinful man” (Luke 5:8). Dread, shrinking, exposure, the collapse of self-sufficiency: this is not the content of anyone’s wish. If early man were inventing a god to soothe his anxieties, he was spectacularly bad at it — he invented instead a Holy One who made his anxieties look trivial and his guilt unbearable. The experience Freud’s theory predicts is precisely the experience the historical record does not show at religion’s foundations; the experience it does show is precisely the one the theory cannot generate. As an explanation, wish-fulfillment explains the wrong religion.

Two further points deserve mention. First, the wish-fulfillment argument commits, in its popular form, the genetic fallacy: even if a belief’s psychological origin were fully explained, nothing would follow about its truth. The desire for water does not prove there is no water; if anything, as Lewis argued elsewhere, innate desires ordinarily correspond to real objects. Second, the argument is a sword with no handle — it cuts its wielder equally. If theism can be dismissed as the wish for a Father, atheism can be dismissed as the wish for no Judge: the desire for autonomy, for a universe with no witness to one’s life, is at least as humanly powerful as the desire for protection. The philosopher Thomas Nagel — an atheist — has candidly acknowledged hoping there is no God and being made uneasy by the alternative. Once wishes are admitted as evidence, they testify in both directions, and the question returns to where it always was: not what humans want, but what is real.

The Numinous and the Moral Law: Where All Developed Religion Begins

Lewis did not treat the numinous as an isolated curiosity. In his account, developed religion everywhere is woven from identifiable strands, and the numinous is the first and universal one — the raw awareness of the Holy that appears at the root of every tradition the human race has produced. The second strand is the moral law: the equally universal experience of an obligation that binds from beyond us, a standard we did not invent and cannot silence, which accuses us when we break it. (Lewis develops this second datum at length in the opening of Mere Christianity.) These two experiences arise independently — awe of the Other, and the pressure of the Ought — and the pivotal religious development in human history, Lewis argued, occurred when they were identified as one: when the numinous Presence and the Lawgiver were recognized as the same Being. That identification is the theological achievement of Israel — “Be ye holy, for I am holy” (Leviticus 11:44) fuses the two strands in a single sentence — and its completion, Lewis held, is the Incarnation, where the Holy One and the Righteous One steps into history as a man. The point, for our purposes, is this: the skeptical hypotheses do not merely fail to explain one odd emotion. They fail to explain the two universal foundations on which every developed religion actually stands — neither of which is reducible to danger, and neither of which is anything anyone would wish.

And the persistence matters. The numinous is not a phase the race outgrew with the lightning-god. It has survived every scientific revolution, every disenchantment, every century that declared it obsolete — surfacing in the foxhole and the cathedral, in the atheist ambushed by awe he cannot name, in the researcher startled by the intelligibility of the universe. A biologically useless, object-less, universal, and ineradicable experience is a strange thing for a purely material process to produce and preserve. Lewis’s conclusion was that its persistence throughout human history is best explained as the mark of a genuine supernatural reality pressing on human consciousness — not a meaningless quirk of the mind, but the fingerprint of the One who made it. The soul — the nephesh — trembles before the Holy because there is a Holy One before whom trembling is the sane response.

The Conclusion of the Matter

The two standard secular stories about the origin of the idea of God each fail on the evidence of the experience they claim to explain. The danger hypothesis cannot cross the jump: no magnification of tiger-fear yields dread of the wholly other, because the numinous is not a degree of fear but a different kind of awareness, with no object in the visible world. The wish-fulfillment hypothesis predicts a comfortable projection and finds instead the mysterium tremendum — a Holy presence no frightened creature would ever have wished into being. What remains is the datum itself: a universal, persistent, irreducible human encounter with something wholly other, standing at the fountainhead of every developed religion alongside the equally stubborn moral law. Explanations that explain it away have to ignore what it is. The explanation that takes it seriously is the old one: in the beginning, God — and man has been trembling, rightly, ever since.