The Argument from Religious Experience: Does Encountering God Count as Evidence?

In the autumn of 1654, the mathematician Blaise Pascal sewed a scrap of parchment into the lining of his coat, and carried it there until he died. On it he had scrawled a date, an hour, and a single ecstatic word repeated: Fire. "Certitude, certitude, feeling, joy, peace. God of Abraham, God of Isaac, God of Jacob, not of the philosophers and scholars." Whatever happened to Pascal between half past ten and half past midnight that night, he was sure it was not an inference. It was an encounter.

Pascal's "Memorial" is one data point in something vast. Across every documented culture, in every century, enormous numbers of ordinary people report not merely that they believe in God but that they have, in some sense, met him: been addressed, comforted, convicted, overwhelmed, loved. The question this essay takes up is whether that immense body of testimony counts as evidence—and if so, evidence for whom, and how strong. The argument from religious experience says that it does count. The honest version of the argument also says exactly where it runs out.

The claim, stated carefully

The argument is easy to state crudely and hard to state well. Crudely: "People have experiences of God, therefore God exists." Stated that way it is plainly too quick, since people also have experiences of pink elephants. The serious versions are more disciplined. They do not claim that religious experience proves God the way a syllogism proves a conclusion. They claim that an apparent experience of God carries genuine evidential weight—a defeasible presumption in favor of its own truthfulness—much as my apparent experience of a tree in front of me carries weight unless I have reason to think I am dreaming or drugged.

The lineage is long. Medieval and early-modern Christian writers treated the soul's experiential knowledge of God as real cognition, not mere feeling. But the modern conversation begins with the Harvard psychologist and philosopher William James, whose Varieties of Religious Experience—the Gifford Lectures delivered at Edinburgh in 1901–1902 and published in 1902—took the reports seriously as data rather than dismissing them as pathology.¹ James catalogued conversions, mystical raptures, and quiet assurances, and proposed that mystical states share four marks: they are ineffable (resistant to words), noetic (they feel like genuine knowledge, not just emotion), transient, and passive (they happen to the subject).² The noetic mark matters most for our purposes: the experiencer does not feel he is having a mood. He feels he is learning something true.

Two modern defenses: Swinburne and Alston

The argument's sharpest contemporary form comes from the Oxford philosopher Richard Swinburne. In The Existence of God (1979; revised 2004) he proposes what he calls the Principle of Credulity: roughly, "if it seems (epistemically) to a subject that x is present, then probably x is present—in the absence of special considerations" defeating that appearance.³ This is not a special pleading for religion. It is the principle by which we run ordinary perceptual life. If I treated every apparent perception as guilty until proven innocent, I could never get started—I would need independent evidence for the reliability of my eyes, but that evidence would itself come through my eyes. Rationality requires trusting appearances by default. Swinburne's move is simply to apply the same default to apparent experiences of God: they too should be taken as veridical unless we have positive reason to doubt them. He pairs this with a Principle of Testimony—that we should normally believe what others report—so that the experiences of others count too. Importantly, the Principle of Credulity is asymmetric: it says that if you seem to experience God, that counts for God; it does not say that if you fail to experience God, that counts against him.⁴

The American philosopher William Alston developed a deeper, more structural defense in Perceiving God: The Epistemology of Religious Experience (1991). Alston's central idea is the doxastic practice—a socially established, self-monitoring way of forming beliefs, complete with its own checks and overriders. Sense perception is one such practice; memory is another; reasoning from premises is another. None of these can be proven reliable without circularity, since any proof would have to use one of the practices. Yet it is rational to engage in them, because we have no independent court of appeal above our well-established practices. Alston then argues that what he calls Christian Mystical Practice (CMP)—the practice of forming beliefs about God on the basis of apparent direct awareness of him—is, structurally, just another such practice. It is socially established, it has internal standards (a "spirit" that contradicts Scripture and the saints is suspect), and it cannot be shown to be unreliable without arguments that, applied evenhandedly, would also discredit sense perception.⁵ His aim, as he put it, was to show that mystical perception "plays much the same role in justifying beliefs about God that sensory perception plays in justifying beliefs about the external world."⁶

Behind both philosophers stands the sheer scale of the phenomenon. This is not a handful of eccentrics. Whatever one makes of the surveys, large fractions of people across radically different societies report experiences they describe as contact with a transcendent reality. The argument asks: is it reasonable to treat all of that testimony as worthless by default? Or does it, like any large body of apparent perception, deserve a presumption of innocence?

The objections, taken at full strength

Three objections deserve to be stated in their strongest form, not knocked down as straw.

1. The neuroscientific explanation

The first says: religious experience has a known cause, and the cause is the brain, not God. The neuroscientist Andrew Newberg popularized the term "neurotheology" and, with Eugene d'Aquili, scanned the brains of Tibetan Buddhist meditators and Franciscan nuns at the peak of contemplation. Using SPECT imaging, they reported increased activity in the frontal lobes (the attention centers) and a sharp decrease in the parietal lobes responsible for orienting the self in space and time—a plausible neural correlate of the mystic's sense of boundless unity with reality.⁷ More aggressively, the Canadian psychologist Michael Persinger claimed that weak magnetic fields applied to the temporal lobes via a modified helmet—dubbed the "God helmet"—could induce a "sensed presence" in ordinary subjects. The implied conclusion: if you can manufacture the experience with a magnet, the experience tells you nothing about a real God.

This is a serious challenge and should not be brushed aside. But two replies have real force. First, the logic: showing that an experience has a neural correlate shows nothing about whether it is veridical. Every experience has a neural correlate—seeing a tree lights up the visual cortex, but that does not make the tree a hallucination. To find the brain-state of perceiving God is no more debunking than finding the brain-state of perceiving your mother. The inference "it happens in the brain, therefore it is only the brain" is a reductionist fallacy. Second, the empirics on Persinger have not held. In 2005 a Swedish team led by Pehr Granqvist of Uppsala University attempted a double-blind replication and published it in Neuroscience Letters. They found that the magnetic field had no relationship to reported experience; what predicted "sensed presence" was the subject's suggestibility and personality, not the helmet.⁸ Persinger objected that the Swedes had not reproduced his precise magnetic-field timing and waveform parameters; Granqvist replied that the equipment and field-generating software had been supplied by Persinger's own laboratory.⁹ Honesty requires the concession: the God-helmet result, the most cited piece of hard "debunking," failed independent replication. Newberg's findings, by contrast, are real—but they are correlations, and their interpreter, Newberg himself, has consistently declined to read them as proof that God is illusory.

2. Conflicting claims and religious diversity

The second objection is, in my judgment, the strongest. It runs: the world's traditions report experiences of different and incompatible objects. The Advaita Vedantin experiences an impersonal Brahman in which all distinction dissolves; the Theravada Buddhist experiences no self and no God at all; the Christian experiences a personal, loving, triune God who addresses her as "thou." These cannot all be veridical descriptions of the same reality, because they contradict one another. If apparent experience generates a presumption of truth, then religious experience generates a contradiction—and a method that yields contradictions cannot be reliable. The philosopher J. L. Mackie pressed versions of this, treating the diversity of religious experience as evidence that such experiences track culture and psychology rather than any divine object.¹⁰

Several replies have force. The conflict is over descriptions and interpretations, not necessarily over whether something transcendent was encountered; that more minimal common claim survives the diversity. Alston grants the problem and offers a frank, limited response: it does reduce the rational confidence an outsider can take from any one tradition's reports, but it does not show that engaging in one's own well-established practice is irrational, any more than the existence of rival scientific paradigms makes doing science irrational.¹¹ John Hick took a bolder route, proposing that the traditions are culturally conditioned responses to a single ineffable "Real," refracted through different conceptual lenses.¹² Hick's pluralism, though, comes at a steep cost: it requires saying that no tradition perceives the Real as it is, which most believers—and most experiencers—would reject as a description of what they take themselves to have met.

Here the objection must be put at full strength. The diversity objector can say: "You have conceded that the experiences, taken at face value, contradict each other. A perceptual practice that systematically delivers incompatible outputs across its practitioners is exactly what an unreliable practice looks like. You cannot quarantine your own tradition's reports from the discrediting force of the others without special pleading." The point has real force. My reply is that it limits the argument rather than destroying it. It shows that religious experience cannot, by itself, settle which God is real—it under-determines theology. It does not show that there is nothing transcendent being encountered, and it leaves the experiencer's own warrant largely intact while sharply reducing the experience's force as public proof to an outsider. That is a real concession, and the argument is healthier for making it.

3. Projection and wish-fulfilment

The third objection is Freud's. In The Future of an Illusion (1927) Sigmund Freud argued that religious belief is an "illusion" in his technical sense—a belief whose driving force is wish rather than evidence. Faced with a terrifying and indifferent nature, the helpless human psyche projects a cosmic Father who promises protection, justice, and an afterlife.¹³ Religious experience, on this reading, is the felt residue of infantile longing, not perception of anything real.

The steelman is strong: many religious experiences do arrive exactly when people most want comfort—in grief, fear, and crisis—and that fit is suspicious. But two replies temper it. First, Freud's argument is itself a genetic fallacy if pressed to a conclusion about truth: showing why someone wants a belief to be true tells us nothing about whether it is true. People who long for water in a desert sometimes long because there really is an oasis. The wish does not refute the object. Second, the projection thesis fits some religious experience badly. A great deal of it is unwelcome: the experience of conviction of sin, the terrifying holiness that flattened Isaiah ("Woe is me, for I am undone"), the call that wrecks a comfortable life. Freud's theory predicts a consoling Father; it strains to explain a God experienced as a consuming fire one would rather avoid. As a partial sociological account it has explanatory value; as a universal solvent it overreaches.

The honest state of the question

Where does this leave us? Several things should be said plainly.

First, the philosophical defenses are taken seriously in the academy; Swinburne and Alston are major figures, and Perceiving God is widely regarded as a landmark. But neither has won the field. Critics—including sympathetic ones such as the epistemologists who worry that Alston's parity argument proves too much—note that if every self-supporting doxastic practice is innocent until proven guilty, then astrology and crystal-gazing might claim the same shelter. Alston's defenders answer that CMP, unlike astrology, has internal overriders and a coherent object; the debate continues, and it is genuinely unsettled.

Second, the contested points should be labeled as contested. Whether religious experience counts as perception in a strict sense, rather than a different kind of cognition, is disputed. Whether the Principle of Credulity is too permissive is disputed. The diversity objection's force is disputed in degree, though almost no one denies it has some force.

Third, and most importantly, we should be clear about what the argument does and does not establish. It does not establish, on its own, that the Christian God in particular exists; the diversity of reports blocks that inference from experience alone. It does not function as a public proof that should compel a determined skeptic, because the skeptic can always invoke a defeater—neurology, projection, or the conflict of traditions—and the experiencer cannot hand over the experience for inspection the way one hands over a fossil. What the argument does establish is more modest and more durable: that for the person who has had such an experience, it can supply real, rational warrant for belief—warrant a third party is not entitled to dismiss as obviously irrational—and that the vast, cross-cultural weight of such testimony is a genuine data point that a complete worldview must explain rather than ignore. It is best understood not as a freestanding proof but as one strand in a cumulative case, where it joins cosmological, design, and moral considerations and is strengthened by them.

Pascal's parchment will not convince a skeptic, and it was never meant to. But it records something the skeptic must reckon with: that across the centuries, sober and brilliant people, in numbers beyond counting, have reported not that they reasoned their way to God but that he found them. The argument from religious experience asks only that we not pretend such testimony is nothing. On that modest demand, it stands. "Sanctify the Lord God in your hearts: and be ready always to give an answer to every man that asketh you a reason of the hope that is in you, with meekness and fear" (1 Peter 3:15)—including, it turns out, a reason for taking the witnesses seriously.