The claim under examination in this essay is specific, and it deserves to be stated plainly before anything else is said. Mainstream monotheism asserts that God is a personal being who desires a relationship with every human creature, who possesses the omnipotence necessary to reveal himself in any manner he chooses, and who has motivated the production of scriptures, the founding of institutions, and the sacrifice of lives precisely because that relationship matters to him infinitely. This is not a vague cultural assertion or a poetic metaphor about the numinous; it is a proposition about a real agent with real intentions and real capacities. And the question that follows from it is equally plain: if all of that is true, why does the evidence for this being’s existence look exactly like the evidence we would expect to find if he did not exist at all?
The argument from divine hiddenness is not, despite frequent mischaracterisation, simply the complaint of a sceptic who has not looked hard enough. It is a philosophical challenge to the internal coherence of theistic claims, and it was given its most rigorous modern formulation by the philosopher J.L. Schellenberg, though its spirit runs through Hume, Russell, and a great deal of the Enlightenment critique of religion. The argument does not demand that God produce a burning bush on demand or write his name in the stars. It demands something far more modest: that the evidence for the most important being in the universe be, at minimum, unambiguous. Instead, what we find is evidence that is geographically distributed, culturally specific, historically contingent, and so thoroughly entangled with human psychology that it is, by any honest measure, indistinguishable from the evidence we would produce ourselves if we were inventing a religion. That is not a small problem. That is a devastating one.
This essay examines that problem in detail. It tests the theistic claim that God is both willing and able to reveal himself, holds that claim against the actual distribution of religious belief across the world and across history, and then examines the standard defences offered by sophisticated theologians and ordinary believers alike. Those defences, it will become clear, are not explanations of divine hiddenness. They are rationalisations of it, and rationalisations so elastic that they would excuse any absence whatsoever, which is precisely what makes them philosophically worthless.
1. The Claim and Its Consequences
Let us begin with the strongest version of the theistic position, because intellectual honesty demands it. The God of classical monotheism is not merely a first cause or an impersonal ground of being. He is described, across Christianity, Islam, and Judaism, as a being who loves humanity, who seeks its flourishing, who grieves over sin, who intervenes in history, who answered prayers in the ancient world with fire from heaven and parted seas, and who desires, above all other things, that human beings know him and choose to be in relationship with him. The stakes, in the theistic picture, could not be higher: eternal salvation or damnation frequently hangs on whether a person comes to know and accept this being. That is not a casual arrangement. That is the highest-stakes epistemic situation conceivable, and it deserves to be treated with corresponding seriousness.
Given those stakes, we should ask what we would expect the evidence for such a being to look like. If a surgeon who has the knowledge, the instruments, and the physical access to save a patient’s life simply does not operate, we do not describe that as a mystery of medicine. We call it a failure, or perhaps something considerably worse. The analogy is imperfect, as all analogies are, but the logical structure holds: a being who has the motivation, the capacity, and the opportunity to make himself known, and who does not, requires a genuine explanation. The theist owes us one. The standard response is that God has indeed made himself known, through scripture, through creation, through the testimony of believers, through personal religious experience. The problem is that each of these putative revelations fails in precisely the same way: it reaches only those already positioned by birth and culture to receive it, and it produces not consensus but centuries of violent disagreement even among those who claim to have received it.
Scripture, to take the most obvious case, is not a universal communication. It is a series of documents produced in specific times and places, written in languages that most of humanity has never read, preserved by communities with their own institutional interests, translated with varying degrees of accuracy and fidelity, and interpreted in so many contradictory ways that the very communities who share a scripture have gone to war with one another over its meaning. If the God of scripture intended it as his primary mode of self-revelation, he chose a remarkably inefficient instrument for communicating with seven billion people across one hundred thousand years of human existence, the overwhelming majority of whom lived and died before a single word of it was written.
The argument from creation is no more convincing. Natural theology, from Aquinas through Paley to the modern intelligent design movement, has always tried to read God’s signature in the structure of the universe. But the universe does not read that way to those without a prior commitment to reading it that way. The cosmological and teleological arguments have been subjected to sustained philosophical critique since at least Hume’s Dialogues Concerning Natural Religion, and the honest verdict is that they establish, at most, that the universe had some cause or causes, not that those causes were personal, singular, loving, or interested in the moral lives of primates on one particular planet. The jump from “the universe exists and is complex” to “therefore a being matching the description in the New Testament is real” is not a philosophical inference; it is a confessional one dressed in philosophical clothing, and the disguise does not survive close examination.
2. The Geography of Belief
Perhaps the single most devastating piece of evidence against the claim that God has successfully made himself known is the geographical and historical distribution of religious belief. This is not a new observation, but it deserves to be stated with full force because its implications are rarely followed through to their conclusion.
If you were born in rural Alabama in 1970, you are overwhelmingly likely to be a Southern Baptist Christian. If you were born in Riyadh in the same year, you are overwhelmingly likely to be a Sunni Muslim. If you were born in Varanasi, you are overwhelmingly likely to be a Hindu. If you were born in rural Tibet, you are overwhelmingly likely to be a Tibetan Buddhist. These are not small statistical variations. They are near-total determinants of religious identity across the global population. The correlation between birthplace and religious belief is, in the social science literature, one of the strongest predictors of any belief system we have ever measured. Pew Research data consistently demonstrates that the single greatest predictor of a person’s religious affiliation is the religion of their parents and the dominant faith of their community. The variation within regions is vastly smaller than the variation between them.
Now consider what this means for the theistic claim. If the Christian God is real and desires to be known by all people, why does belief in him correlate almost perfectly with having been born in a historically Christian region rather than with, say, the quality of the evidence encountered? If the Islamic revelation is genuine and universal, why does it reach almost exclusively those already within the cultural orbit of the Arabian peninsula and its historical expansion, rather than spreading spontaneously through rational conviction? The theist will sometimes respond that missionaries carry the faith to new populations, but this only sharpens the problem: a god who requires human intermediaries operating under historical and political constraints to communicate his existence to his own creatures has, by any reasonable standard, outsourced his most important work to the most fallible possible agents, and the results reflect that outsourcing faithfully.
The geography of belief does not look like the distribution of a truth that is being discovered. It looks like the distribution of a cultural inheritance that is being transmitted. There is a difference, and it matters enormously. Mathematical truths, physical constants, biological mechanisms, historical facts that can be independently verified: these tend toward convergence as investigation proceeds and as investigators from different backgrounds compare notes. Religious truth claims, by contrast, diverge persistently and dramatically. There is no movement toward consensus. After three thousand years of theology and religious experience and claimed revelation, the world’s religious landscape is not converging on a single, agreed account of the divine. Every decade produces new denominations, new sects, new schisms, new prophets with new revelations that are incompatible with the previous ones, and the fragmentation continues to accelerate as communications technology removes the geographic isolation that once forced regional religious homogeneity. This is not what the discovery of a real, objectively existing being looks like. This is what the proliferation of human invention looks like, and the difference is not subtle.
Jerry Coyne, in Faith Versus Fact, puts the epistemological problem with characteristic precision: “In the end, religious investigations of ‘truth,’ unlike those of science, are deeply dependent on confirmation bias. You start with what you were taught to believe, or what you want to believe, and then accept only those facts that support your prejudices. This is the basis for the theological practice of ‘apologetics,’ designed to defend religion against counterarguments and disconfirming evidence. In contrast, science has no apologetics, for we test our conclusions by trying to find counterevidence.” The geography of belief is the most visible symptom of this structural problem. People do not generally arrive at their religious beliefs by examining the full range of competing claims and assessing the evidence for each; they arrive at them by being raised inside a tradition and then constructing post-hoc defences of what they already believe. The god who is supposedly reaching out to them happens, with extraordinary regularity, to be the god of their grandparents.
It is worth pressing this point in its most uncomfortable form. Consider the billions of human beings who lived and died before any of the current major monotheistic traditions existed, or who lived in regions those traditions never reached. The indigenous populations of the Americas before European contact, the peoples of sub-Saharan Africa before the spread of Islam and Christianity, the populations of East Asia who developed their own sophisticated traditions entirely independently: what account does the theist give of their epistemic situation? The standard Christian answer, where one is offered at all, tends to be extraordinarily unsatisfying: either God revealed himself to them in ways we cannot access, or they are judged by the light they had, or, in the more honest Calvinist tradition, they were simply not among the elect. None of these responses constitutes a genuine engagement with the problem. They are emergency exits from an argument that has been lost, and they are chosen precisely because any honest engagement with the question produces conclusions the theist would prefer not to reach.
3. The Certainty Demanded and the Silence Delivered
There is a profound asymmetry at the heart of religious epistemology that does not receive the attention it deserves. On one side of the asymmetry stands what is demanded of the believer: total commitment, absolute trust, the willingness to subordinate every other value to the demands of faith, to organise one’s entire life around the existence and will of a being whose existence cannot be directly demonstrated. Christianity does not ask for mild provisional assent pending further evidence. It demands that a person be “born again,” surrender their autonomy to a divine will, accept doctrines that strain every instinct of rational scepticism, and stake their eternal destiny on the correctness of a decision made largely on the basis of cultural inheritance. Islam demands submission, which is in fact what the word means, not tentative hypothesis-formation. These are not epistemically modest requests, and they deserve not to be treated as such.
On the other side of the asymmetry stands what is actually delivered in return for this commitment: the silence. Not an absence of religious experience, since religious experience is widespread and well-documented, but an absence of the kind of clear, unambiguous, universally accessible evidence that the magnitude of the demand would seem to require. The experiences people report are invariably private, culturally specific, neurologically explicable in terms of documented brain states, and consistent with precisely what we would expect if religious experience were a product of human psychology rather than divine communication. The mystic in medieval Castile encounters the Virgin Mary. The Sufi mystic in Persia encounters the divine through ecstatic union with Allah. The Hindu mystic in Benares encounters Brahman. These experiences are described in the categories available to those who have them, which is to say in the categories of the tradition they already inhabit. The god who reveals himself in religious experience has the remarkable property of always confirming what the recipient already believes, which is a property we associate with mirror images rather than with independent external realities.
David Hume set the standard for proportionality in evidence over two and a half centuries ago: “A wise man, therefore, proportions his belief to the evidence.” The demand that a person stake everything on a conviction they cannot adequately demonstrate to a sceptic, and that this demand be met with ambiguous private experience and culturally contingent tradition, does not satisfy Hume’s principle. The relationship between the magnitude of the claim and the quality of the evidence offered for it runs precisely in the wrong direction: the larger the claim, the thinner the evidence, and this inversion is not a theological puzzle awaiting resolution. It is a structural feature of faith as a cognitive practice, one that has persisted precisely because faith, by design, insulates itself from the standard demands of evidence. That insulation is not a sign of the claim’s profundity. It is a sign of its precariousness.
We should note what would constitute adequate evidence, so that the objection cannot be dismissed as setting an impossibly high bar. We are not demanding that God appear in physical form in every city simultaneously, though for an omnipotent being this would presumably present no difficulty. We are not demanding miracles of a spectacular, cinematic variety. We are simply demanding what any other knowledge claim of comparable importance would require: evidence that is public rather than private, consistent rather than contradictory, universal rather than geographically restricted, and responsive to investigation rather than retreating before it. The heliocentric solar system was, once, a minority hypothesis contested by powerful institutions. It is now established fact because the evidence for it is available to anyone with the instruments to examine it, regardless of their cultural background. The fact that religious claims have not converged toward consensus after millennia of examination is not evidence of their profundity. It is evidence of their nature, and their nature is the problem.
4. The Standard Defences and Why They Fail
Theologians and philosophers of religion are not unaware of the hiddenness problem, and they have produced a range of responses to it. These responses deserve careful examination, because the question is not whether intelligent people have tried to answer the challenge but whether their answers succeed. They do not, and the reason they do not is both interesting and revealing: every defence of divine hiddenness that has been constructed shares the same fatal property. It would excuse any degree of absence whatsoever, up to and including the complete non-existence of the being in question. A defence with that property is not a defence at all; it is a blank cheque written against an account that may well be empty.
The most common defence is the free will argument. God does not make himself unambiguously known, the argument runs, because to do so would overwhelm human freedom. If God’s existence were obvious, we would be compelled to believe, and compelled belief is not genuine faith. Genuine faith requires the possibility of disbelief, which in turn requires that the evidence be ambiguous enough to allow a reasonable person to doubt. This argument has a certain surface plausibility, but it dissolves under moderate pressure. The first difficulty is that the argument assumes, without justification, that clear evidence of God’s existence would compel belief in the way that, say, a bullet wound compels pain. But this is not how human cognition works. People routinely encounter clear evidence of inconvenient propositions and choose to disbelieve them anyway. The evidence for anthropogenic climate change is not ambiguous, yet hundreds of millions of people reject it. The evidence for evolution is not ambiguous, yet substantial proportions of several populations deny it. Clear evidence does not eliminate the possibility of free rejection; it simply makes rejection less intellectually defensible, which is not the same thing, and the conflation of the two is the engine driving the free will argument forward.
The second difficulty is more fundamental. Even granting, for the sake of argument, that clear evidence of God’s existence would compel belief in a way that undermines free will, it does not follow that the current level of hiddenness is necessary to preserve that freedom. There is an enormous space between “God appears in the sky over every city simultaneously” and “God leaves ambiguous cultural traditions that happen to match local geography.” The claim that the current, geographically specific, culturally contingent, historically variable distribution of religious belief is precisely and necessarily calibrated to maximise human freedom requires a level of specificity that no theologian has managed to provide. And if God is omniscient, he already knows who will freely choose to believe and who will freely choose to disbelieve, regardless of how much or how little evidence is available. The free will defence of hiddenness is not a precise argument; it is a rough gesture in the direction of plausibility, and gestures do not constitute philosophical explanations.
The third and most serious difficulty is that the free will argument, taken seriously, proves far too much. It would justify any level of hiddenness, including complete absence. If God’s non-appearance in human history were total, if no tradition had ever formed, no scripture ever written, no religious experience ever reported, the free will argument could still be deployed: God is absent precisely so that humans can freely choose to believe in him without any coercive evidence. The fact that an argument would justify both the current situation and a maximally different one is not evidence that it explains the current situation. It is evidence that it explains nothing, and an argument that explains everything explains nothing.
The second major defence is what we might call the soul-building argument, associated in its classical form with the theologian John Hick and drawing on Irenaeus’s account of human moral development. The idea is that hiddenness is spiritually productive: faith exercised in the absence of certainty builds character, deepens virtue, and produces the kind of mature, freely chosen relationship with God that is more valuable than one produced by obvious evidence. Doubt, on this account, is not a problem to be solved but a gift to be cultivated. The soul is strengthened precisely by having to hold on to belief in the face of difficulty. There is something almost admirable about the psychological sophistication of this argument, but its logical structure is equally empty.
The soul-building argument assumes that hiddenness produces the relevant virtues, when the evidence suggests otherwise. If centuries of faith practised in the face of divine silence were producing measurable moral improvement across humanity, we might have reason to take the argument seriously. What we actually observe is that populations with high levels of religious faith show no consistent moral advantage over populations with lower levels, and in several measurable dimensions, including rates of violent crime, social cohesion, and trust in institutions, highly secular societies frequently outperform highly religious ones. The Scandinavian countries, among the most secular in the world, consistently rank at or near the top of global indices measuring happiness, social trust, low corruption, and human wellbeing. This is not an argument that secularism causes virtue, but it is certainly evidence against the claim that divine hiddenness is producing the soul-building benefits attributed to it, and evidence of that kind cannot simply be waved away by pointing to individual examples of faith-inspired virtue.
More fundamentally, the soul-building argument, like the free will argument, would excuse any degree of absence. If the current level of hiddenness is justified because it builds souls, then a greater level of hiddenness would presumably build souls even more effectively, and the souls built by complete absence of evidence would be the most admirably faithful of all. We are back to the blank cheque: an argument that justifies any situation up to and including God’s non-existence cannot be used as evidence that God exists and has chosen, for good reasons, to be hidden. The argument proves too much, and arguments that prove too much prove nothing.
A third defence, more philosophically sophisticated than the preceding two, involves the concept of epistemic distance. The idea, developed at length by Hick and others, is that God maintains a certain distance from human perception precisely so that humans are not overwhelmed by his presence. Finite creatures encountering an infinite being directly would be annihilated, or at least so dominated by the encounter that free response would be impossible. God therefore filters his presence through the natural order, through religious experience, through the indirection of symbol and tradition, in order to make genuine encounter possible. The world is set at just the right epistemic distance from its creator: close enough for faith, far enough for freedom. This argument has the virtue of being internally coherent, but it has a very obvious problem: it is unfalsifiable in exactly the way a bad explanation is unfalsifiable. Whatever the actual distribution of religious belief and religious experience turned out to be, the epistemic distance argument could be adjusted to explain it. The argument does not constrain what we should expect to observe; it merely re-describes whatever we do observe as consistent with a hidden god, and that is not an explanation of the evidence but a redescription of it in theological language. This pattern of infinitely adjustable explanation is one of the defining features of claims that have no genuine evidential content.
5. The Schellenberg Argument and Its Full Force
J.L. Schellenberg’s formulation of the hiddenness argument, first published in his 1993 book Divine Hiddenness and Human Reason and refined in subsequent work, deserves extended attention because it is the most precise philosophical statement of the problem and because the responses it has generated, while numerous, have not succeeded in defeating it. The argument runs roughly as follows: if God exists and is perfectly loving, God would want all people to be in a relationship with him; a personal relationship requires, at minimum, that both parties are aware of the other’s existence; God could ensure that all people who are non-resistantly open to believing in him actually do believe; therefore, if God exists, all such people would believe; but some people are non-resistantly in a state of non-belief, meaning they are not resisting belief but simply without it; therefore, the perfectly loving God of classical monotheism does not exist. The logical structure of this argument is considerably harder to escape than its critics have generally acknowledged.
The crucial concept here is non-resistant non-belief. Schellenberg is not talking about people who have examined the evidence and chosen to reject God, nor about people who are living in a state of motivated rejection. He is talking about the many people who have approached the question genuinely, examined what evidence was available to them, and found themselves simply unable to believe, not because they wanted to disbelieve but because the evidence was insufficient to produce belief. These people are not the village atheist defiantly shaking his fist at the heavens. They are the honest enquirer who finds, after careful examination, that the door to belief simply does not open for them. If God is loving and desires a relationship with these people, and if it is within God’s power to produce belief without coercing it, since the free will defence itself implies that God deliberately chose not to produce such belief rather than being unable to do so, then the existence of non-resistant non-believers is, on the theistic picture, a genuine puzzle requiring a genuine answer rather than a theological shrug.
Theistic responses to Schellenberg generally take one of two forms. The first is to challenge the existence of genuinely non-resistant non-believers, arguing that all apparent non-belief conceals some form of resistance, pride, sin, or deliberate turning away. This response is unfalsifiable and morally objectionable in equal measure. It is unfalsifiable because there is no state of non-belief that could not, in principle, be characterised as concealed resistance by someone determined to characterise it that way. And it is morally objectionable because it requires imputing hidden bad faith to millions of honest enquirers without any evidence beyond the bare fact of their non-belief. The argument also carries the unpleasant implication that the billions of people born into non-Christian or non-Islamic traditions who never encountered a compelling case for those faiths were somehow secretly resisting a god they had never been meaningfully introduced to, which is not an explanation but a moral failing dressed as theology.
The second form of response is to challenge the premise that a perfectly loving God would necessarily ensure that all non-resistant people believe in him. Perhaps, the argument goes, God has reasons for permitting non-resistant non-belief that we cannot fully comprehend, reasons that are consistent with his perfect love but that transcend our limited understanding. This is the “greater goods” response, and it shares the fundamental defect of all the other defences: it is infinitely adjustable. Any absence of God, however total, could be explained by positing that God has greater goods in view that we cannot comprehend. An argument with this structure does not explain the evidence; it explains it away by replacing genuine engagement with an appeal to mystery, and appeals to mystery, however grandly theological their formulation, are not arguments.
There is something revealing about the pattern we keep encountering. Every major defence of divine hiddenness has the same structure: it posits a reason why God might choose not to reveal himself, and that reason is specified vaguely enough that it would justify any degree of hiddenness whatsoever. The free will defence, the soul-building defence, the epistemic distance defence, the greater goods defence: each of them is consistent with a world in which God is completely absent and in which the evidence for his existence is precisely what we would expect if he did not exist. When the hypothesis “God exists but chooses hiddenness for good reasons” and the hypothesis “God does not exist” produce identical predictions about the observable world, the principle of parsimony tells us clearly which hypothesis we should prefer. The thought experiment that follows from this is uncomfortable but necessary: would a world without God look any different from the world we actually inhabit?
6. The Psychological and Neurological Evidence
One of the theist’s most reliable resources in discussions of divine hiddenness is the appeal to religious experience. Billions of people across history have reported experiences of the divine that feel, to those who have them, utterly and undeniably real. These experiences take many forms: the sense of a divine presence, visions, voices, moments of overwhelming love or certainty, mystical union, answered prayer, the feeling of being guided or protected by a force beyond oneself. Surely, the argument runs, this vast accumulation of testimony constitutes evidence of something real. Surely the sheer scale of human religious experience cannot be explained away as mere psychology.
But the appeal to religious experience runs into several serious difficulties that have been exhaustively documented in the philosophy and neuroscience of religion. The first is the problem of cultural specificity, which we have already noted in a different context. Religious experiences are not culturally neutral. The content of religious experience tracks almost perfectly with the tradition of the person having it. Christians have experiences of Christ and the Holy Spirit; Muslims have experiences of Allah’s nearness; Hindus have experiences of Brahman or of specific deities; shamanic practitioners have experiences consistent with their animist worldview. If religious experience were genuine perception of an objectively existing divine reality, we would expect its content to converge across traditions, just as the perceptual experiences of astronomers from different cultures converge on the same description of Jupiter. Instead, the content of religious experience diverges in exactly the way we would expect if it were a product of cognitive processes operating within culturally specific frameworks, and this divergence is not a minor footnote to the evidence but one of its most significant features.
The second difficulty is the neuroscientific evidence for the mechanisms that produce religious experience. Research on temporal lobe stimulation, the neurological profiles of mystical experience, the role of hypoxia and hyperventilation in producing states of altered consciousness, the documented ability of psychedelic substances to produce experiences subjectively indistinguishable from “genuine” mystical ones: all of this points toward a straightforward naturalistic account of religious experience. The brain produces these states; the specific content is filled in by cultural expectation and prior belief; the conviction of reality that accompanies them is a feature of the cognitive mechanism, not evidence that the experience tracks an external reality. This does not mean the experiences are not real in the sense of being genuinely felt; it means that the inference from “I had an overwhelming experience of the divine presence” to “therefore the divine presence exists” is not warranted, for the same reason that the inference from “I had an overwhelming dream about being pursued by wolves” to “therefore wolves were pursuing me” is not warranted. The vividness of an experience is not a reliable guide to its external cause.
The third difficulty is the problem of reliability. Religious experience, as a source of knowledge, is extraordinarily unreliable by any standard measure. It produces contradictory results across traditions, it is subject to manipulation by fasting, sleep deprivation, rhythmic chanting, and other techniques known to alter cognitive states, and it is not reproducible in the controlled conditions that would allow us to distinguish genuine perception from hallucination. The epistemological problem here is not that religious experiences do not occur; they clearly do, and their occurrence is well-attested and psychologically interesting. The problem is that there is no method by which a person having such an experience can determine whether it constitutes genuine perception of an external reality or an internally generated state. And in the absence of such a method, the appeal to religious experience as evidence for divine existence is an appeal to a source whose reliability has not been established and whose track record, judged by the standards we apply to every other form of knowledge, is poor enough to warrant considerable caution.
7. Prayer and the Problem of Selective Response
One specific form of divine communication deserves its own treatment, because it is one of the most widely practised and most empirically testable of all religious claims. Prayer is not merely a devotional practice; it is, in most theistic traditions, a form of communication with a personal being who hears, considers, and responds to requests. The existence of prayer as a practice implies the existence of a being who is listening, which is a testable claim of some significance, and it has in fact been tested.
The empirical literature on intercessory prayer is not extensive, but it is instructive. The most methodologically rigorous study of intercessory prayer, the Study of the Therapeutic Effects of Intercessory Prayer, published in the American Heart Journal in 2006 and involving over 1,800 cardiac bypass patients, found that intercessory prayer had no significant effect on patient outcomes. Indeed, the only statistically significant finding was that patients who knew they were being prayed for did slightly worse than those who did not, a result most plausibly explained by the anxiety-inducing nature of being told your condition is serious enough to require organised prayer. Smaller studies have produced similarly null results when controlled adequately for the confounds that plague research of this kind. The failure of prayer to produce demonstrable effects under controlled conditions is not by itself decisive: one can always argue that God does not submit to experimental testing. But that argument is precisely what is at issue. A God who refuses to allow his interactions with the world to be examined empirically, who adjusts his behaviour precisely to avoid leaving any measurable trace, is a God whose existence is indistinguishable from non-existence by the only methods we have for reliably distinguishing real from imagined causes.
The anecdotal evidence for answered prayer, which is genuinely vast, suffers from well-documented cognitive biases. Humans are extraordinarily good at noticing coincidences that confirm existing beliefs and forgetting those that do not. The cancer patient who prayed and recovered remembers the prayer. The cancer patient who prayed and died is not available to report the failure. The prayers that appear to be answered receive attribution to God; the prayers that are not answered receive a range of post-hoc explanations, “God said no,” “it wasn’t his will,” “he had better plans,” all of which render the hypothesis permanently immune to disconfirmation. A hypothesis that can absorb any result without modification is not describing a real mechanism. It is describing a theological framework flexible enough to accommodate any world whatsoever, including a world in which no god exists, and the flexibility is a warning rather than a virtue.
8. The Honest Comparison: What Evidence for a Real Being Would Look Like
It is sometimes argued that the hiddenness argument sets an unfairly high standard for divine evidence, that we are demanding from God a kind of proof that we do not demand from other real but non-obvious entities. We cannot directly observe the interior states of other minds, yet we are confident other minds exist. We cannot directly observe quarks or black holes, yet we are confident they are real. Why should the evidence for God be held to a higher standard than these other genuine but unobservable entities?
The comparison fails for several reasons that are worth spelling out carefully. We infer the existence of other minds not from ambiguous cultural traditions that happen to match our geography, but from a consistent, cross-culturally universal pattern of behavioural evidence that converges on the same inference regardless of the observer’s cultural background. We infer the existence of quarks and black holes not from personal experiences filtered through culturally specific frameworks, but from mathematical predictions that have been independently verified by scientists from dozens of different countries and cultural backgrounds working in different laboratories on different continents. The convergence of independent lines of evidence is what distinguishes genuine inference about reality from culturally transmitted belief. Religious claims lack this convergence, not because the standard is unfairly high, but because convergent, independent evidence is precisely what we would expect to find if the object of inquiry were real, and its absence is therefore informative.
Consider what it would look like if a specific theistic claim were true. If the God of orthodox Christianity exists and desires to be known by all people, we would expect, at minimum, that the core theological claims of Christianity converge across traditions of independent inquiry in the way that the core claims of chemistry or physics converge. Instead, what we find is that Christian theology has produced approximately forty thousand distinct denominations, differing on questions of soteriology, Christology, eschatology, sacramental theology, and biblical interpretation, and that the disagreements have shown no tendency to resolve over time as the evidence accumulates. This is not the signature of a being who has communicated clearly. It is the signature of a human institution constructing meaning in the absence of authoritative external input, and doing so with all the creativity and disagreement that human institutions characteristically display. The human tendency to construct religious meaning is itself a fascinating area of enquiry, but it is an explanation of religion, not a vindication of any particular religious claim.
We should also note the curious asymmetry in the way theists respond to evidence from competing traditions. The Muslim who finds the Christian evidence for the resurrection unconvincing, the Christian who finds the Islamic evidence for Mohammed’s prophethood unpersuasive, the Hindu who is unmoved by the arguments of both: each of these believers is applying, to a tradition not their own, a scepticism they do not apply to their own. The Christian who dismisses the miracle claims of Islam on the grounds that mere testimony is insufficient evidence, and then defends the miracle claims of Christianity on the same grounds of testimony, has adopted a standard that is not principled but convenient. The honest application of a consistent epistemological standard across all religious traditions produces the same result: the evidence for any particular tradition is insufficient to justify the certainty demanded by that tradition. This does not make all traditions equally valid; it makes all of them equally unestablished by the standards of evidence that govern every other domain of serious inquiry.
9. The Moral Weight of the Silence
There is a dimension of the hiddenness argument that is less often discussed but that deserves explicit treatment, because it has direct moral consequences that are not merely theoretical. If the God of classical monotheism exists, is omniscient, and is watching human history, then he is watching the consequences of his hiddenness in real time. He is watching the religious wars that have been fought between communities who could not agree on his nature: the Thirty Years War, the Crusades, the sectarian conflicts of Northern Ireland, the ongoing violence between Sunni and Shia communities across the Middle East, the centuries of persecution of heretics and apostates by religious majorities confident of their authority to act in God’s name. He is watching the members of minority religious traditions being persecuted by majority ones. He is watching the children born into high-demand religious groups who will spend their adolescence in terror of eternal damnation for sins of thought and doubt. He is watching all of this, and he is choosing to maintain his hiddenness, presumably for the greater goods we cannot comprehend.
The moral character of this choice deserves honest scrutiny. If a human being possessed knowledge that, when shared, would prevent widespread suffering and conflict, and chose to withhold that knowledge for reasons they declined to explain, we would not describe that person as loving. We would describe them as culpable. The theist’s standard response is that God’s ways are not our ways, that divine morality transcends human understanding, that we are not in a position to judge God’s choices by human moral standards. But this response abandons the very ground on which the theist wishes to stand: if God’s moral character cannot be assessed by human moral concepts, then the claim that God is “perfectly loving” is either empty or inaccessible to us. We cannot have it both ways. Either God’s love is love in a sense we can understand and apply, in which case the consequences of his hiddenness require genuine moral justification, or God’s love is something utterly different from what we mean by the word, in which case the claim that God loves us is not a comfort but a warning about the meaning of theological language.
The history of religious violence is not, on this argument, an indictment of God’s character in the way a prosecutor might indict a person. It is evidence about the quality of the alleged communication. If the most important being in the universe intended to communicate a moral framework to humanity, and the result of that communication is millennia of violent disagreement even among those who received it, then the communication has failed, and the failure requires explanation. The explanation that immediately presents itself, the one that requires no additional assumptions about inscrutability or transcendence, is that the communication failed because there was no communicator. The disagreement looks exactly like what we would expect if competing groups of humans were constructing religious systems independently, in the absence of any external authority to adjudicate between them. Because that, in all probability, is what happened, and the probability is not small.
10. What Hiddenness Actually Indicates
The argument from divine hiddenness does not require us to claim certainty about the non-existence of God. Philosophical arguments rarely deliver certainty, and intellectual honesty requires acknowledging that we are dealing in probabilities and inference rather than demonstration. But what the argument does establish, with considerable force, is that the evidence we actually have, the geographical distribution of belief, the culturally specific nature of religious experience, the failure of prayer under controlled conditions, the internal contradictions of scripture, the absence of convergence across traditions of independent inquiry, all of this evidence is far more consistent with the hypothesis that religion is a human construction than with the hypothesis that a personal God has been attempting, with the full resources of omnipotence, to make himself known. When we lay the two hypotheses side by side and ask which one better accounts for the world as we actually find it, the answer is not close.
The concept of divine hiddenness, as deployed by sophisticated theologians, has the intellectual form of an explanation but lacks its substance. An explanation earns its place by making specific predictions that differ from what we would observe in the absence of the explanatory factor. The various defences of divine hiddenness fail this test completely. Every one of them is consistent with a world in which God exists and has chosen, for various good reasons, to conceal himself. Every one of them is equally consistent with a world in which God does not exist. When two hypotheses are equally consistent with all available evidence, the simplest one is preferred, and the hypothesis that does not require the existence of an unobservable omnipotent agent who has arranged things to be indistinguishable from his own absence is considerably simpler. The theologian who insists otherwise is not offering an explanation. He is offering a preference.
The charge of divine hiddenness, in the end, is not primarily a philosophical point, though the philosophy is rigorous and has not been successfully answered. It is a moral and phenomenological observation: the world does not look like a place in which a loving, all-powerful being is attempting to reach every human creature with a life-or-death message. It looks like a place in which human beings, across diverse cultures and historical periods, have been constructing stories about the divine in response to the profound challenges of mortality, meaning, and community. Those stories are not without interest or beauty or psychological function. But interest, beauty, and psychological function are not evidence of literal truth, and they are not a substitute for the genuine evidence that the claim of a personal God actually requires and has not produced.
The silence, honestly interpreted, is not a mystery waiting for a theological key. It is precisely the kind of silence we would expect if the question of divine communication were being asked into an empty room. The question we should be asking is not why God is hidden but why, given an absence of evidence that on any other topic would settle the matter, we continue to extend the kind of benefit of the doubt that no other claim of comparable magnitude would ever receive. That question is partly psychological, partly sociological, and partly a matter of the enormous institutional investment that has accumulated around religious belief over millennia. But it is not, on the evidence, a question with a theological answer, and the distinction between those two categories is one that the argument from hiddenness forces us to confront with unusual directness.
The claim that God exists, desires to be known, and has the power to make himself known is not a claim that deserves the permanent benefit of the doubt that millennia of religious authority have accustomed us to extending to it. It is a claim that, examined with the same rigour we bring to any other claim about the nature of reality, fails to meet its own burden. The absence of God in the evidence is not evidence of God’s inscrutability. It is evidence of God’s absence, and those two propositions are not interchangeable, however often the one is substituted for the other in theological discourse. Bertrand Russell’s demand that we proportion our beliefs to the evidence was not offered as a counsel of sterile scepticism but as the foundation of intellectual integrity and, he argued, of a more honest and therefore more humane civilisation. That foundation, applied honestly to the question of divine hiddenness, leads to a conclusion that is uncomfortable for the theist but unavoidable for the honest enquirer: the silence is not inscrutable, and the most parsimonious account of it is also the most accurate one.
References
Schellenberg, J.L., Divine Hiddenness and Human Reason, 1993
Schellenberg, J.L., The Hiddenness Argument: Philosophy’s New Challenge to Belief in God, 2015
Hume, David, Dialogues Concerning Natural Religion, 1779
Hume, David, An Enquiry Concerning Human Understanding, 1748
Russell, Bertrand, Why I Am Not a Christian, 1927
Coyne, Jerry A., Faith Versus Fact: Why Science and Religion Are Incompatible, 2015
Bering, Jesse, The Belief Instinct, 2011
Benson, Herbert, et al., “Study of the Therapeutic Effects of Intercessory Prayer,” American Heart Journal, 2006
Pew Research Center, The Global Religious Landscape, 2012
Hick, John, Evil and the God of Love, 1966
Bloom, Paul, Just Babies: The Origins of Good and Evil, 2013