Religion from First Principles

There is a prior question before any question about which religion is true, or what God is like, or whether particular doctrines are coherent. That prior question is: why does religion exist at all? Why does every human society, regardless of geography, historical period, or technological development, generate supernatural belief systems, ritual practices, sacred spaces, and religious specialists? This is not a theological question. It is an empirical one, and its empirical answer turns out to be one of the most illuminating things we can know about how human minds work.

The question of religion's universality is genuinely extraordinary. Archaeologists find evidence of ritual behaviour at Shanidar Cave in northern Iraq dating to approximately 65,000 years ago: Neanderthal burial practices suggesting belief in some form of afterlife. The sculptures of the Venus of Willendorf, the cave paintings of Lascaux and Altamira, the deliberate orientations of megalithic monuments like Newgrange toward solstice sunrise, all suggest that symbolic, ritualistic, cosmologically organised belief is among the oldest attested features of human cognitive life. Not agriculture. Not metallurgy. Not writing. Religion is older than all of them.

Anthropologists have documented religious belief in every society they have studied. The content varies enormously, the specific deities, narratives, rituals, and moral codes differ so radically across traditions as to make comparison superficially absurd. But the structural features persist. Every tradition posits beings or forces that are invisible, powerful, and interested in human behaviour. Every tradition includes rituals, prescribed, repeated, meaningful behaviours that enact or maintain relationship with those beings or forces. Every tradition demarcates the world into the sacred and the profane, and regulates passage between them. And every tradition offers answers to the questions that generate the deepest and most persistent human anxiety: why are we here, how should we live, and what happens when we die.

Three broad theoretical frameworks have organised serious scholarly inquiry into religion's origins and persistence. The first treats religion as a cognitive by-product (the side-effect of cognitive mechanisms that evolved for other purposes. The second treats religion as a functional adaptation) a set of beliefs and practices that enhanced the survival and reproduction of religious groups. The third (not incompatible with either) treats religion as the primary vehicle through which cultures achieve large-scale social coordination. These frameworks conflict at points, converge at others, and together constitute the most serious attempt yet made to answer the prior question.

In 1993, the anthropologist Stewart Guthrie published Faces in the Clouds, a systematic argument that religion arises from a specific feature of human perception: the systematic over-attribution of agency and intentionality to natural phenomena. Guthrie's key observation was that humans routinely perceive faces, voices, footsteps, and purposeful behaviour where none exists, in cloud formations, ambient noise, the wind in trees, random statistical distributions. This is not a pathology of credulous individuals. It is a baseline feature of normal human cognition.

The evolutionary logic is straightforward. Consider the asymmetry of detection errors in ancestral environments. Falsely detecting an agent (hearing a predator that isn't there) costs you unnecessary fear and wasted energy. Failing to detect an agent that is there (not hearing the predator) costs you your life. In a world where predators, rivals, and enemies were genuine threats, natural selection would have strongly favoured a cognitive system calibrated toward false positives over false negatives in agency detection. Better to be the organism that spins around at the rustling sound and finds only wind than to be the organism that ignores the rustling sound and finds a leopard.

Theory of Mind (the cognitive capacity to attribute mental states (beliefs, desires, intentions) to others and use those attributions to predict and explain behaviour) is the other critical piece of cognitive hardware for understanding religion. Theory of Mind underlies all social cognition; without it, social life as we know it is impossible. But it comes with a notable feature: it runs on any sufficiently agent-like input, including inputs that are not actually agents. Humans readily attribute mental states to moving geometric shapes in Heider and Simmel's famous 1944 experiments (a large triangle "bullying" a small triangle), to cars that have been "rude" in traffic, and to natural phenomena that display apparent purposiveness.

When HADD detects an apparent agent in a natural phenomenon, Theory of Mind immediately begins operating on it, attributing beliefs, desires, and intentions to the entity. This is not a conscious process. It is automatic, fast, and largely below deliberate awareness. The thunder doesn't just seem loud; it seems angry. The unusually good harvest doesn't just seem fortunate; it seems intended. The Theory of Mind system, operating on HADD outputs, reliably generates the proto-religious attribution: something with intentions did this, and those intentions were directed at us.

Pascal Boyer's Religion Explained (2001) is one of the most systematic attempts to account for the specific content of religious belief, not merely why humans generate supernatural beliefs in general, but why the particular kinds of supernatural beings that populate religious traditions have the specific properties they do. Boyer's framework draws on cognitive science, anthropological fieldwork, and evolutionary psychology to construct what he calls a "natural history" of supernatural concepts.

Boyer's central theoretical contribution is the concept of minimal counterintuitiveness. He begins with the observation that cultural transmission is selective: not every idea that is generated persists and spreads. Some ideas die with the individual who had them; others replicate through thousands of minds across centuries. What determines transmission success? Boyer argues that ideas are optimised for transmission by a specific cognitive property: they must violate just enough of our intuitive expectations about a domain to be memorable and attention-arresting, while conforming to enough of those expectations to be cognitively tractable and easy to reason about.

This framework explains something anthropologists had noted but not fully accounted for: the specific clustering of supernatural belief around certain types of beings. Across radically different cultures, we find the same basic types: ancestors who persist after death and monitor the living (persons minus mortality), deities who are invisible but omniscient and powerful (persons with vastly expanded cognitive and physical capacities), spirits who can move through physical barriers (persons minus solidity), and demons who spread disease (persons with malign intent and unusual causal powers). This is not cultural diffusion, many of these traditions developed in isolation. It is convergent output of the same cognitive machinery operating on the same set of intuitive categories.

Boyer is careful to distinguish between the generation of supernatural concepts and their systematic organisation into religion. Minimal counterintuitiveness explains why certain types of supernatural entities are cognitively "sticky", memorable, transmittable, easy to reason about. But organized religion involves much more than a catalogue of supernatural beings. It involves ritual, moral codes, specialist practitioners, institutions, sacred texts, and elaborate cosmologies. Boyer's framework accounts for the basic building blocks; explaining their elaboration into full religious systems requires additional mechanisms, which the following sections develop.

What makes Boyer's framework genuinely explanatory (rather than merely descriptive) is that it generates specific predictions that can be tested against the anthropological record and against experimental memory data. Where those predictions hold, we have evidence not merely of a clever theory but of the actual mechanisms driving the cross-cultural patterns. The minimal counterintuitiveness effect has been replicated across multiple cultures, in both religious and secular contexts, and in both recall and recognition memory paradigms. The framework is falsifiable in Popper's sense, and it has largely survived the tests.

Émile Durkheim was a French sociologist who, in 1912, published The Elementary Forms of Religious Life, a study of Australian Aboriginal totemism that became one of the most influential analyses of religion ever produced. Durkheim was not interested primarily in the psychological question of why individuals generate religious beliefs. He was interested in the sociological question: what function does religion serve for the groups that practise it? His answer was precise and has proven robust across a century of subsequent research: religion is the primary mechanism by which human groups constitute and sustain themselves as groups.

The core of Durkheim's argument begins with a distinction between the sacred and the profane, the two fundamental categories through which religious thought organises the world. The profane is the ordinary, the mundane, the world of everyday utility and individual life. The sacred is set apart, invested with exceptional significance, approached only through prescribed rituals of preparation and transition. This is not merely a subjective classification; the sacred/profane distinction is enforced socially, through taboos, rituals, and the strong emotional reactions that violations provoke. What makes something sacred is precisely that it is collectively designated as such, and the collective designation both reflects and constitutes the group's shared identity.

Durkheim's analysis of ritual is particularly rich. He observed that the primary function of religious ritual is not cognitive (it is not primarily about transmitting beliefs or propositions) but social and emotional. Rituals bring people together in synchronised, emotionally charged, collectively meaningful activity. This synchronisation generates what Durkheim called collective effervescence: a heightened emotional state that arises from collective participation in shared activity, a sense of being part of something larger than the individual self, of the boundaries between self and group becoming permeable.

Collective effervescence is not metaphorical. Contemporary research in social psychology confirms that synchronised movement, shared music, collective vocalisations, and other features of ritual activity produce measurable increases in oxytocin levels, pain tolerance, feelings of trust and bonding, and self-sacrifice for group members. The bodies of ritual participants are physiologically altered in ways that increase their commitment to the group and to one another. Religion achieves this through precisely the activities (shared liturgy, communal singing, coordinated prayer, collective feast) that research shows to be most effective at generating the neurochemistry of social bonding.

Durkheim's framework has been extended in significant ways by subsequent researchers. Jonathan Haidt's moral foundations theory (specifically the sanctity/degradation and loyalty/betrayal foundations) can be read as the psychological mechanisms through which Durkheimian social bonding operates. What is designated as sacred by a group generates strong purity intuitions; violations of those designations generate moral disgust. This disgust reaction enforces the boundary between in-group members (who respect what is sacred) and potential out-group members or transgressors (who don't). Religion is, among other things, a system for marking and enforcing the moral boundaries of the moral community.

The sociologist Christian Smith has argued that this means religion functions as an identity-conferring system, it answers the question "who are we?" by reference to a shared sacred narrative and a set of practices that perform that identity. This is why religious conversion and religious apostasy are so socially consequential: they are not merely changes of opinion but renegotiations of identity and community membership. The convert doesn't just adopt new beliefs; she adopts a new social world, a new set of practices, a new language of self-description. The apostate doesn't just abandon beliefs; he exits a community and often finds himself without a language for identity that others in his former community can recognise.

Harvey Whitehouse, an anthropologist at Oxford, has developed one of the most empirically productive frameworks in the cognitive science of religion: the theory of modes of religiosity. The framework arose from Whitehouse's fieldwork in Papua New Guinea, where he observed two radically different types of religious practice operating simultaneously, and noticed that they correlated with sharply different social and cognitive structures.

Whitehouse identified two fundamental modes in which religious practice is transmitted and experienced, and he named them according to their characteristic arousal profiles and memory mechanisms.

The modes framework is significant for several reasons. First, it explains variation within and across religious traditions that other theories treat as mere superficial diversity. Why do some religious communities practise intense, physically demanding initiation rites while others rely on gentle weekly worship? They are not simply different tastes, they are different cognitive and social technologies, each with characteristic strengths and weaknesses. The imagistic mode binds small groups with extraordinary ferocity; the doctrinal mode coordinates large populations across space and time. Many traditions incorporate elements of both, strategically deploying intensity for key transitions (initiations, pilgrimage, revivalist meetings) while maintaining regular doctrinal practice for ordinary cohesion.

The modes framework intersects productively with the sociology of religious conversion and religious violence. Religious conversions most often occur through imagistic mode experiences (the Pauline vision on the road to Damascus, the evangelical "born again" experience, the life-transforming spiritual crisis) rather than through careful doctrinal evaluation. The experience comes first; the doctrinal framework is subsequently used to interpret and integrate it. This reverses the apparent sequence: it looks like the doctrine caused the conversion, but the conversion was triggered by an imagistic episode that the doctrine then colonised.

Religious violence most commonly involves either the mobilisation of imagistic mode bonding (small, intensely bonded groups willing to die for one another) or the dehumanisation of out-group members via sacred/profane distinctions that render their deaths cosmically necessary rather than merely strategically useful. Both mechanisms are predicted by the modes framework and confirmed by anthropological and historical evidence.

Carl Jung broke from Freud in 1912, in significant part over the question of religion. Freud's account of religion, developed in Totem and Taboo (1913) and The Future of an Illusion (1927), was essentially dismissive: religion is a projection of infantile wishes onto the cosmos, the collective neurosis of humanity, the persistence of childhood dependency on an omnipotent father figure into adult life. The cure was psychoanalysis and the maturity to accept a godless universe. Jung found this account not merely incomplete but intellectually dishonest, a failure to take seriously the psychological reality of what religious experience involves.

Jung's alternative begins with an expansion of the unconscious. Freud's unconscious was personal, the repository of repressed individual experience, infantile wishes, and traumatic memories. Jung proposed that beneath the personal unconscious lies a collective unconscious: a layer of the psyche that is not formed by individual experience but is inherited, the psychological residue of ancestral experience encoded in the structure of the psyche itself. The collective unconscious is not a literal store of memories but a set of predispositions: recurring patterns of perception, emotion, and imagery that appear spontaneously in dreams, visions, myths, and religious symbols across cultures and throughout history.

Jung's framework has significant explanatory power in domains where purely cognitive accounts struggle. The Jungian reading of why the same narrative structures (the hero's journey, the dying and rising god, the great flood, the world tree) appear independently across unconnected cultures is not that these narratives diffuse across populations (though sometimes they do), but that they arise from the same structural dispositions of the collective unconscious. They are not copied; they are independently generated by the same psychic template.

The scientific status of the collective unconscious is genuinely contested and deserves an honest assessment. The specific mechanism Jung proposed (inherited psychological memory) is not plausible in standard evolutionary terms. But the general claim that humans share inherited structural tendencies in cognition, emotion, and narrative processing is not controversial; it is the claim of cognitive science and evolutionary psychology, articulated in different language. The deep structural similarities in religious and mythological content across cultures can be explained by shared cognitive architecture (the same mechanisms producing similar outputs) without requiring the specific Jungian metaphysics. What Jung's framework adds that purely cognitive accounts tend to miss is the psychological weight and function of this material: its felt significance, its resistance to purely intellectual engagement, its tendency to return in dreams and fantasies when denied conscious expression.

The relationship between Jungian depth psychology and contemporary neuroscience is an active area of research. Mark Solms, a neuropsychoanalyst working at the intersection of psychoanalysis and neuroscience, has argued that Freudian and Jungian concepts (the unconscious, the pleasure principle, the archetype) can be mapped onto specific neural systems, and that the correspondence between psychoanalytic concepts developed through clinical observation and neural systems identified through neuroscience is too close to be accidental. Whether this mapping will ultimately vindicate, transform, or dissolve the Jungian framework remains genuinely open.

Long before the cognitive science of religion was a discipline, David Hume attempted what we would now call a naturalistic account of religious belief, an explanation of religion that proceeds from human psychology and historical observation rather than theological presupposition. His Natural History of Religion, published in 1757, is one of the founding documents of the secular study of religion, and it anticipates with remarkable precision several of the theoretical moves that later cognitive scientists would formalise.

Hume's starting question was not "is religion true?" but "how did religion begin?" His answer was psychological and anthropological. Primitive religion, Hume argued, arose not from philosophical contemplation of the cosmos but from fear (specifically, from the fear and uncertainty generated by an uncontrollable natural environment. Unknown causes of important events) drought, flood, disease, military defeat, generate anxiety; anxiety generates the need for explanation; the natural human tendency (what we would now call HADD) is to find explanation in intentional agency. Early religion is therefore polytheistic: a proliferation of specific agents responsible for specific domains of powerful natural phenomenon.

Hume traced the historical progression from polytheism to monotheism not as a progression toward truth but as a progression driven by the logic of divine flattery. Once gods are conceived as powerful agents, the competition among their followers to demonstrate greatest piety incentivises escalation of divine attributes (greater power, greater wisdom, greater goodness) until a single, maximally great being emerges as the logical terminus of the competitive elevation of divine qualities. Monotheism, on this account, is not a revelation but a social equilibrium: the stable endpoint of competitive divine inflation.

What makes Hume's account significant beyond its historical priority is the epistemological framework he brings to it. His separate work on miracles, in the Enquiry Concerning Human Understanding, presents what remains the most rigorous philosophical argument for scepticism about miraculous claims: a testimony to a miracle, however apparently reliable, is always less credible than the assumption that some error in observation or testimony has occurred, because the laws of nature (established by the most consistent and universal human experience) constitute stronger evidence than any testimony can overcome. The prior probability of a miracle, given the weight of experience, is always very low; the posterior probability, even after testimony, should remain low unless the testimony is overwhelming and cross-verified. Hume's argument is not a proof that miracles cannot occur; it is a Bayesian argument about the appropriate credence to assign them.

Human cooperation is anomalous in the animal kingdom. Other social animals cooperate within small groups of genetic relatives or stable reciprocal partnerships. Humans cooperate at scales (cities, nations, international institutions) that vastly exceed what kin selection or direct reciprocity can explain. How did this cooperation emerge? And what role did religion play in enabling it?

Ara Norenzayan, a psychologist at the University of British Columbia, has developed the most comprehensive empirical account of this question in his book Big Gods (2013). Norenzayan's core thesis is that the specific features of moralising supernatural beings (gods who observe human behaviour, who care about how humans treat each other, and who punish moral transgressions) are not incidental ornaments of religion but functional mechanisms that enabled the transition from small-band to large-scale human cooperation.

Norenzayan's thesis has been challenged on several fronts, which he engages with seriously. The most significant challenge is the direction of causation: do moralising gods enable large-scale cooperation, or do large-scale social organisations (once established by other means) generate moralising god concepts to legitimate and stabilise their authority? The evidence from the database of religious history suggests the relationship is reciprocal: moralising beliefs can precede and facilitate scale-up, but they also intensify once scale is achieved and a ruling class has incentive to deploy supernatural sanction.

Peter Richerson and Robert Boyd's framework of cultural group selection provides the broader evolutionary context for Norenzayan's thesis. Richerson and Boyd argue that cultural evolution can operate at the level of groups, groups with beliefs and practices that promote cooperation and effective collective action will outcompete groups without such beliefs and practices. Religious beliefs that enforce cooperation are not individually adaptive in all circumstances, they impose real costs on individual members. But they are group-adaptive: groups whose members cooperate outperform groups whose members defect, even when cooperation is individually costly. This means that cultural group selection can maintain religious beliefs that are individually costly but collectively beneficial, a pattern that fits the empirical profile of committed religious practice in many traditions.

All the frameworks discussed so far explain why humans generate religious beliefs, transmit them, organise them socially, and use them for group coordination. None of them has yet addressed what may be the most important data point: the first-person phenomenology of religious experience itself. What it is actually like, for the person having it, to encounter what they describe as the sacred, the divine, or the transcendent.

William James (psychologist, philosopher, and founding figure of American pragmatism) gave the first serious academic treatment of this data in The Varieties of Religious Experience (1902), delivered originally as the Gifford Lectures at Edinburgh. James's approach was empirical in a specific sense: he took seriously the reports of religious experience as data, examined them with the same rigorous attention he brought to other psychological phenomena, and attempted to identify their structural features and their psychological consequences.

James's pragmatist criterion for evaluating religious experience is subtle and often misunderstood. He did not ask "is the experience caused by what the experiencer believes causes it?" He asked "what are the consequences of the experience for the life of the person who has it?" If a religious experience reliably produces greater compassion, increased resilience, reduced fear of death, greater ethical seriousness, and enhanced sense of meaning (and James's evidence showed that it frequently does) then it is, by the pragmatist criterion, a real and significant psychological event, whatever its ultimate metaphysical cause.

The contemporary neuroscience of religious and mystical experience has produced a substantial literature that James would have found highly interesting. Andrew Newberg's neuroimaging studies of meditating Tibetan Buddhist monks and praying Franciscan nuns show characteristic changes in parietal lobe activity (specifically, reduced activation of the area responsible for generating the sense of self-other boundary) during intense religious practice. The felt dissolution of the boundary between self and universe that is central to mystical experience in almost every tradition appears to have a specific neural correlate. This is not, as is sometimes claimed, a debunking of the experience, explaining the neural mechanism of an experience does not answer the question of what, if anything, the experience reveals. But it does specify the mechanism, which is the beginning of genuine understanding.

Psilocybin research conducted at Johns Hopkins and NYU (building on the earlier work of Walter Pahnke's "Good Friday Experiment" in 1962) has shown that pharmacologically induced mystical experiences share the structural features James described and produce long-term changes in personality, wellbeing, and the fear of death that are comparable to the changes James documented in naturally occurring religious conversion and mystical experience. If the same experiential structure produces the same psychological consequences regardless of the mechanism that triggered it (religious practice or pharmacological intervention) this suggests that the psychological significance of religious experience is real and separable from the question of its supernatural causation.

The frameworks assembled across this artifact converge on a complex but coherent picture. Religion is not a single thing. It is a family of phenomena (cognitive, social, experiential, and institutional) that interact and reinforce each other in ways that produce one of the most persistent and universal features of human life. To explain religion from first principles is to see these layers simultaneously: the cognitive hardware that generates supernatural attribution, the transmission dynamics that select cognitively optimal supernatural concepts, the social functions that organise individuals into moral communities, the evolutionary pressures that have favoured prosocial belief, and the experiential core that gives the whole system its felt weight and personal significance.

What emerges from this analysis is an appreciation for religion that is, in an important sense, more serious than either the devout believer's or the dismissive sceptic's. The believer's account (that religion is true and therefore important) leaves its importance contingent on the truth of its metaphysical claims. The sceptic's account (that religion is false and therefore merely a remnant of ignorance) fails to explain its extraordinary persistence, its universality, and the genuine psychological and social work it performs. The first-principles account says: this is among the most significant productions of the human cognitive and social apparatus, performing functions that nothing else performs with equivalent effectiveness, arising from mechanisms that are deeply embedded in human nature, and generating experiences that are among the most profound available to human beings. Understanding it this way makes both the case for religion and the case against it considerably more interesting.

The question of what religion does that nothing else does can be answered at each of the levels we have traversed. Cognitively: it provides frameworks for causal explanation of events that exceed the ordinary explanatory reach, the mechanisms of death, the causes of misfortune, the ultimate source of moral authority. Emotionally: it manages the terror generated by consciousness of mortality more effectively than any competing system, by situating finite individual existence within an infinite or at least persisting narrative. Socially: it generates moral communities with shared identity, mutual obligation, and the emotional bonding of collective effervescence. Institutionally: it coordinates behaviour at scale through supernatural monitoring and sanction. Experientially: it provides access to states of awareness that are among the most transformative and significant available to human consciousness.

The secular alternatives to each of these functions exist, science for causal explanation, philosophy and therapy for mortality management, civil institutions for social coordination, law for behavioural regulation, art and meditation for transformative experience. But no secular system currently integrates all these functions with the holistic effectiveness of a mature religious tradition, transmitted through embodied community, sustained by costly commitment signals, validated by collective effervescence, and providing a complete narrative from cosmological origin to individual death.

This does not settle the question of whether any particular religion's supernatural claims are true. It specifies what would be lost if they are not, and what would be required to replace what is lost. That is a more honest framing of the stakes than either religious apologists or secularist critics tend to offer.

The next artifact in this series turns to the deep story structures that underlie religious and cultural belief across civilisations, the mythology and the hero's journey, the narrative archetypes that Joseph Campbell spent a lifetime mapping. Religion generates the community and the metaphysics; mythology generates the story through which those metaphysics are felt, dramatised, and transmitted across generations. They are not separate phenomena. They are two layers of the same machinery.