Why Do People Believe in Gods?

Gary M. Bakker


Why do people believe in gods? And ghosts, angels, demons, fairies, goblins, and other imagined conspiracies?

Rationalists have found the ubiquity and persistence of belief in the supernatural hard to satisfactorily explain. Recent psychological research uncovering a universal, innate, adaptive tendency toward purpose-based intuitions in explaining the world may have largely resolved the question.

Much of the literature and discussion among the skeptical, agnostic, humanist, and rationalist communities is concerned with what people believe, the evidence supporting or refuting these beliefs, and the consequences of these positions. This is a perennial endeavor, and it will continue as long as variety and change occur in people’s beliefs.

However, an issue addressed almost as much—why so many people believe “weird” or unsubstantiated things—presents a different sort of question. It has an empirical, scientific, psychological answer that, when substantially formulated, should require no more than occasional tinkering or elaboration as research and knowledge on human cognitive function gradually progresses.

Numerous surveys have shown us that the most widespread, persistent category of such beliefs is in supernatural beings—in gods, ghosts, and suchlike. A sufficient explanation for this phenomenon is critical for several reasons. It is necessary to counter the “So many believe, therefore there must be something in it” argument. It can guide us strategically in our quest to enable more people to hold beliefs that are evidence-based and are therefore more likely to correspond to the realities that the Enlightenment and the scientific process have revealed to us.

Many explanations for other weird beliefs, usually in the form of cognitive errors, have been formulated. But continuing speculation over why intelligent, educated, stable, happy people are drawn in overwhelming majorities in most countries to unsubstantiated, arbitrary, paradigm-shattering beliefs in supernatural agents indicates that a further form of powerful, almost universal, cognitive error remains to be identified.

Fortunately, recent psychological research uncovering a universal human tendency toward “promiscuous teleological intuitions”—strong purpose-
based intuitions (Kelemen et al. 2012)—has brought us to the point of a substantial explanation for widespread belief in supernatural agents despite a paucity of evidence for them, such that we may now be able to move on from speculation about multiple casual factors and intuitive suggestions about subsequent tentative responses to assertive policy development and coordinated social action.

Scientific and Philosophical Arguments Rarely Work

Some will argue that the answer to the question “Why do people believe in gods?” is a straightforward one. “It’s due to the evidence,” they say. This answer is adequate for questions such as “Why do people believe in the Sun?” or “Why do people believe the Sun is powered by a fusion process?” because the relevant evidence has been assembled and a consensus of opinion/belief has formed. But it cannot explain the ubiquitous belief in gods across almost every culture on Earth because no consensus has developed supporting belief in any of them even after several thousand years of research and debate.

This is starkly illustrated by the existence of a substantial, intelligent, educated atheist/agnostic community, and by the hundreds of mutually contradictory opinions about the gods believed in (i.e., religions). The evidence for any supernatural beings is clearly inadequate to explain the widespread and profound belief in so many particular ones. Scientific and philosophical argument has, it seems, convinced no one and changed very few minds on the matter.

Why People Believe Other Weird Things

Similar questions are asked about people believing in other evidence-sparse phenomena such as alternative medicine, diagnostic auras, flying saucers, Yeti, and so on.

A common denominator among these examples is a lack of understanding of what constitutes real evidence. In theory at least, if a believer were to be presented with sufficient evidence one way or the other and had been educated to understand what counts as real evidence and why, then they could be persuaded not to bother with, let alone rely on, their homeopathic cancer cure or the latest pyramid scheme. Most explanations for the above beliefs involve cognitive errors such as confirmation bias, pareidolia, frequency illusion, illusory correlation, availability heuristic, bandwagon effect, and so on. The scientific method was largely developed to help us counter these universal human tendencies toward making errors in our thinking.

Man with tinfoil on his head

However, as has been described, when it comes to religious beliefs, scientific and philosophical arguments rarely work. People do not stumble into a misguided belief in a god through deductive error. They seek out, assume, or create de novo for themselves such a belief. It seems to be Homo sapiens’ default position.

The most similar nontheistic widespread beliefs would be in the many conspiracy theories attracting the interest of up to half of our population (Oliver and Wood 2012). These also involve an attraction to evidence-sparse assumptions about powerful purposeful agents behind what the rest of us assume is more likely to be the naive, messy behaviors of very ordinary people.

There Must Be a Cognitive Urge or Instinct

So it is not enough to propose to a god-believer or a goblin-believer, “You don’t have enough evidence, so you are probably wrong.” They “just know” they are right; there is no clinching evidence against their beliefs, and most other people share a similar intuition. These features, they feel, place the onus on the skeptic (no matter how unfair to William of Occam this may seem) to explain how the intuitions of so many intelligent people can be so consistently wrong in roughly the same direction.

If logic and evidence do not explain the near-universal belief in some higher power, then it must be in the nature of human beings to want to believe in it. There must be some drive or urge or innate cognitive tendency toward such belief. Almost every culture ever studied has included a—often dominant—metaphysical element despite a lack of any substantial evidence. Hence, the question “Why do people believe in gods?” is a psychological one about drives, instincts, or cognitive tendencies, which can therefore be answered psychologically and scientifically.

Past Psychological Explanations

Several explanations for this psychological phenomenon have been proposed over the years. But the question still arises and is addressed in books, articles, and at every skeptical conference I have attended. Clearly, the explanations tendered to date have not been comprehensive, explanatory, satisfying, or instructive enough for a working consensus to form on this issue, allowing us to thus move on using what we now better understand.

Listed below are a few of the many interesting, perhaps useful—but incomplete—explanations for ubiquitous god-belief that have been hypothesized in the past.

The Fear of Death

The price of consciousness is an awareness of one’s own inevitable death. The emotion of fear has motivated the behavioral avoidance of death in all higher organisms. In humans, therefore, this fear is instinctive, and people have a massive incentive to self-delude and to assume (despite a dearth of evidence) that an afterlife awaits. This carries the bonus of promising a reunion with deceased loved ones.

However, this explanation for ubiquitous beliefs in gods, angels, leprechauns, and suchlike is not sufficient. It does not explain hundreds of common extraneous features beyond the promise of an afterlife that are found in religions around the world, such as notions of angels and demons, sin and ritual, hell and sacrifice, goblins and ghouls, prayers for help, and rules to live by in this life. It only explains why many people choose to assume some form of consciousness after death. A creator god or a judgment day god or an interactive benevolent god is not necessary for this.

Further, religious people still seem to fear and mourn death just as much as atheists do. As a sop to this fear, the afterlife assumption does not really work very well, except perhaps to motivate suicide bombers. So the fear of death explains only a tiny fraction of people’s supernatural beliefs.

Desire to Control People

Friedrich Nietzsche, though an atheist, approved of religion as a means for the elite minority of humanity (the “supermen”) to control the gullible unimportant majority. Religions have clearly been used in this way throughout history, both as a means of internal state social control and as rhetoric to justify and excuse wars actually instigated by motives of power, wealth, or land acquisition.

But ruling elites would be unlikely to successfully delude the masses, and maintain the delusion for centuries, with a narrative that did not come naturally to the people. The people must, at root, want to see the emperor’s new clothes. The desire to control can explain the construction of religions but not their eager adoption. For example, Socrates proposed that the strata of society be justified by disseminating to its ordinary citizens the “myth of the metals” in which it is contended that we are all born with gold, silver, bronze, or iron mixed in our souls. Only the first of these can make us fit to rule.

This “noble lie” did not take off anywhere near as well as the tale that some of us are God’s chosen and some are unworthy infidels. The “divine right”of kings to rule grew from this—more popular—noble lie. What is the critical difference between these alternate narratives, explaining their contrasting popularities? Could it be that bronze in your soul is only a cause of your mediocrity, whereas God’s allocation is done to a purpose? I will revisit this critical difference.

Direct Religious Experience

Some people have chosen to interpret their mystical experiences, drug-induced hallucinations, hyperventilation symptoms, psychotic episodes, hypoglycemic effects, near-death experiences, or dreams as direct evidence of a spiritual world and of a god. However, this path to faith is a minority route. It does not explain god-belief’s ubiquity. Many more people believe in God than have seen him.

Nor is it an inevitable deduction from such phenomena. Neurological and psychological explanations have been developed for all these experiences. We must still explain why so many people prefer to blunt Occam’s razor by assuming that elaborate fantastical supernatural events are occurring—rather than concerning but ultimately mundane ones.

An Evolutionary Advantage

Several theorists, such as psychologist Jesse Bering (2012) in The Belief Instinct, have claimed that a tendency to believe in a higher being that oversees and punishes us has the evolutionary advantage of making people more cooperative, conformist, patient, and restrained and that societies with such individuals and with a religious culture will more likely thrive or at least survive.

However, this hypothesis has been hotly debated. It better describes the processes of cultural natural selection and evolution among the memes of competing societies than the processes of biological natural selection among the genes of competing individuals. There are clearer examples of cooperative societies flourishing than of submissive individuals procreating profusely. Genghis Khan is a stark illustration of a highly nonconformist, uncooperative, impatient, and unrestrained individual whose multitudinous progeny subsequently dominated an entire gene pool.

Furthermore, we are not trying to understand a ubiquitous belief in a certain god (a meme). We are trying to understand a cognitive trait, skill, or tendency (influenced by a gene). Having been uncertain to date what this tendency is, we cannot assume it is overall desirable and evolutionarily advantageous. Is it gullibility, religiosity, suggestibility, conscientiousness or “consciencefulness,” superstitiousness, or submissiveness? Which of these confer survival advantage for the individual and why? The evolutionary advantage hypothesis for god-belief is as yet incomplete.

The Brain’s “God Spot”

A related line of explanation has arisen since the proliferation of fairly precise functional brain-scan studies using fMRI, SPECT, PET, and other technologies. Claims have been made (e.g., d’Mayberry 2014) that when people are experiencing “transcendence”—meditating or praying—that certain regions of the average brain (people’s brain maps do vary) light up.

This very promising technology has unfortunately been taken, especially in the mass media, as a new source of explanation for a raft of familiar mental phenomena. For example, when alcoholics have a drink, certain brain regions react. The public and even some scientists have concluded from these findings that problems such as alcoholism, obsessive-compulsive disorder, or depression are somehow more “real” now that we have found a neurological site or correlate.
The logic here is bizarre. Has any scientist ever asserted, or even imagined, that there is no neurological substrate or correlate to all mental activity? And the fact that brain bit X fires up explains nothing. Nor does it, yet, tell us what to do about any of these problems beyond what we have discovered to be useful at a psychological level fifty years ago (Satel and Lilienfeld 2013).
So if there is a God spot, why is there a God spot and what exactly does it do?

A Need for Meaning

The meaning of a person’s life is a cognitive concept. It only exists if viewed or held by a conscious entity, just as purpose only exists if there is a “purposer.”

The meaning or purpose in an individual’s life can be derived internally or externally. Common internally derived sources of purpose (developed or appreciated by the person him- or herself) include achievement, procreation, “happiness,” or contribution to mankind or to fellow travelers.

But this source appears to be insufficient for many people. Externally imposed purposes abound, such as God’s grand plan, avoiding Hell, pleasing God, fear of God, seeking sainthood, etc. These are relatively simplistic and immature motivations for doing good or striving hard, just as they provide simplistic, stifling, and immature explanations for natural phenomena. A child may be satisfied when asking “Why should I behave?” to be told “Because I say so” or “Because it’s good” or “Because God says so.” But an adult may need a more sophisticated explanation in terms of effect on others or on society’s cohesive functioning. In the same way a reference to Thor when explaining lightning is a simplistic and stifling response. It stunts further inquiry but has satisfied a surprising number of otherwise intelligent people throughout history.

So the apparent human desire for meaning or purpose does in part explain the almost universal phenomenon of god-belief. But the question remains: Why are so many people focused upon externally derived over internally derived purpose? Similarly, humans clearly thirst for knowledge and explanation. But why do so many leap to “because God made it so” explanations for natural phenomena? As Tim Minchin asked in his beat poem and short film Storm:

Isn’t this enough?
Just this world?
Just this beautiful, complex,
Wonderful unfathomable
Natural world?
How does it so fail to hold our attention
That we have to diminish it with the invention
Of cheap man-made myths and monsters?

Promiscuous Teleological Intuition

The work of psychologist Deborah Kelemen and her research colleagues may enable us to now fill the explanatory gaps I have described. Drawing on her own and others’ research programs, Kelemen, director of the Child Cognition Laboratory at Boston University, has found that children around the world “evidence a general bias to treat objects and behaviors as existing for a purpose” (Kelemen 2004, 295). There is now overwhelming evidence that children are innately prone to “promiscuous teleological intuitions,” preferring teleological, purpose-based rather than physical-causal explanations of living and nonliving natural objects (Kelemen et al. 2013).

For example, young children do not see raining as merely what a cloud does but as what it is “made for.” If asked why prehistoric rocks are pointy, children will greatly prefer “so that animals could scratch on them when they got itchy” over “bits of stuff piled up for a long time.”

Early parenting or explaining makes little difference to this strong tendency. It appears to be modifiable only from around ten years of age. For example, the children of both religious fundamentalist and non-fundamentalist parents, when asked why a certain animal exists, favor “God made it” or “a person made it” over “it evolved” or “it appeared.” This tendency declines only after eleven years of age and only in the children of non-fundamentalist parents.

Much research supporting and developing this hypothesis has been undertaken. From infancy, humans are excellent “agency detectors,” sensitive to others’ mental states. Even twelve-month-olds will follow the gaze direction of symbolic faces. Children’s complex imaginary companions, like supernatural agents, occur cross-culturally (Taylor 1999).

Kelemen’s explanatory hypothesis is that this generalized default view, that entities are intentionally caused by someone for a purpose, is a side effect of a socially intelligent mind that privileges intentional explanations.

It is not difficult to posit the evolutionary advantages of such “attribution of agency” among infants and children. An infant’s entire world comprises an intentional agent—its parent. The sooner and more thoroughly an infant can develop a “theory of mind” and respond accordingly, the better for it. It must attach. The parent must bond. It must anticipate and manipulate its world on the assumption of purposeful agency occurring all around it. An absence of such is starkly illustrated by the autistic child, to whom its parents are just another set of shapes in its visual field. No attachment occurs, and in less affluent, protected, aware times than we have now, such children rarely survived. They could not control their (almost entirely interpersonal) environment and starved, ate poison, or just wandered away.

The tendency to attribution of agency extends beyond the world of man-made artifacts to the natural world. Children intuitively identify people as the designing agent of artifacts and God as the designing agent of nature (Kelemen 2004, 299). “All known folk religions involve nonnatural agents and intentional causation—the substrate of intuitive theism” (Kelemen 2004, 297).

Reasoning about all aspects of nature in non-teleological physical-reductionist terms is a relatively recent development in the history of human thought (Kelemen 1999a), and contemporary adults are still surprisingly bad at it. For example, evolution is generally misconstrued as a quasi-intentional needs-responsive designing force (Kelemen 2004). But our brains did not evolve “to enable consciousness.” Our brains evolved, and consciousness resulted. No purpose, just causes.

Aristotle distinguished “efficient” causes (the antecedent sources of objects and events) from “final” causes (the ends, goals, functions, or purposes of objects or events). Unfortunately, his focus was on the latter. He applied teleological explanations to all living and nonliving natural phenomena. For example, leaves exist on a plant to provide shade, and water exists on Earth to sustain life (Kelemen et al. 2013, 1074). Since the Renaissance, physical scientists have overtly rejected teleological explanation in search of physical-causal, “efficient” explanations. However, under psychological stress even physical scientists will tend to revert to default teleological explanations such as endorsement of “trees produce oxygen so that animals can breathe” (Kelemen et al. 2013).
So promiscuous teleological explanation is almost universal among children but is also a developmentally persistent cognitive default position. For example, the tendency returns in strength if a person develops Alzheimer’s disease later in life (Lombrozo et al. 2007).

Kelemen et al. (2013) concluded that the teleological tendency is robust, resilient, developmentally enduring, arises early, and becomes masked with cognitive maturity and education but is not replaced. Hence religious belief is cognitively natural and culturally resilient. “Notions of purpose are central underpinnings of the world’s religions” (Kelemen et al. 2013, 1081).

The rise and persistence of the intelligent design argument for God in the wake of the demolition of creationism’s simplistic claims is an illustration of this resilience (Kelemen and Rosset 2009). Children naturally see lions, mountains, and icebergs as “made for something” (Kelemen 1999b), irrespective of parental explanations (Kelemen et al. 2005) or ambient cultural religiosity (Kelemen 2003). Later in life they will still tend to assume that “Mother Nature” or “Gaia” is a goal-directed, self-preserving organism (Kelemen and Rosset 2009)—another “attribution of agency.” “As evidenced by many religions, artifact design represents a powerful analogical base that children and adults use to understand the natural world” (Kelemen et al. 2012, 440).

“Promiscuous teleological intuition” and excessive “attribution of agency” do not explain all indiscriminate, non-skeptical, or paranormal beliefs. Other cognitive errors, such as confirmation bias, must be invoked to understand why so many think that acupuncture will cure migraine or that a plesiosaur lurks in Loch Ness.

However, a tendency toward conspiracism may be explained in exactly the same way as ubiquitous god-belief. The dictum “If it’s a choice between a conspiracy and a cock-up . . .” contrasts purposeful, powerful, inferred, possibly malevolent agency with natural, caused-but-not-directed, “random” events controlled by only the laws of nature and mathematics. Some people clearly have a strong cognitive trait tendency to assume conscious purposeful agents are behind most events in society. If not God or Satan, then powerful people.

If the similarity is meaningful, we would expect that people with a strong cognitive bias toward promiscuous teleological intuition would also tend toward both god-belief and conspiracism. This is exactly what we find. For example, Oliver and Wood (2012) found that a supernatural belief scale strongly predicted support for conspiracism. Both correlated with a measure of magical thinking. To illustrate, they found an extremely high correlation between conspiracism and belief that “We are currently living in End Times as foretold by Biblical Prophecy.”

The bizarre and surprising observation that, even in the face of profound advances in the scientific physical-reductionist understanding of nature, most of humanity insists on clinging to evidence-sparse, redundant, arbitrary notions of gods, demons, goblins, nature spirits, or the recent vacuous fallback “feeling” that “there just must be something there,” has not been adequately explained until now. This has left psychologists unsatisfied, allowed theists to argue that the cross-cultural profusion of metaphysical belief is evidence that there must be something to it, and atheists and agnostics with some uncertainty as to how to correct this widespread “dangerous delusion.”

A recently developed, well-supported explanatory hypothesis suggests that the ubiquity of human belief in supernatural agents in the face of a paucity of evidence derives from an innate cognitive tendency to attribute agency to all active entities in the world and to therefore assume purposive motivations within those entities, and indeed to the world itself. This tendency is highly adaptive for infants and young children and will therefore be favored in evolutionary terms.

That it is innate is supported by its universality among children and its cross-cultural dominance among adults. Kelemen (2004) has described a scientific education as suppressing rather than replacing teleological explanatory tendencies, citing the finding that scientifically uneducated Romanian Gypsy adults have promiscuous teleological intuitions much like scientifically naïve British and American elementary school children (Casler and Kelemen 2008).

However, such findings do not imply a dramatically new solution. We have long known that education can make people less superstitious and less religious. Less educated and politically knowledgeable people exhibit higher levels of conspiracism (Oliver and Wood 2012). Interestingly, Kelemen et al.’s (2013) physical scientists (who became more teleological under pressure) performed no better than equivalently schooled humanities scholars, and both held their line better than undergraduates. So a specialized scientific training and knowledge base is not necessary. Any further education seems to help one outgrow teleological promiscuity.

There is no perfect negative correlation between intelligence and teleological delusion. Great minds have applied themselves to the arbitrary theological intricacies of the gods’ purposes for us. The ability to see past or see through our unjustified attributions of agency and purpose in the world may have more to do with imagination (Bakker 2013) once an alternative physical-causal worldview has been introduced, by education, as a sufficient explanatory option. The question “What if…?” has led to more insights than has brute intellectual power.

The question “What if the world is really big and round and there is no absolute ‘up’ and ‘down’?” overtook the flat-Earth theory. “What if the world is not 6,000 years old but is billions of years old?” has nearly supplanted creationism. “What if I am my brain? How would it look to me?” has dented the need for a soul or immaterial mind to explain consciousness (Bakker 2013).
So the best cures for teleological delusion are those we already know of: A high level of education; encouragement to think independently, flexibly, and with insatiable curiosity; and having an input as soon as people are capable of abstract conceptual thought.

The major contribution of this new evidence-based conceptualization of the problem is that we now know more clearly what we are fighting. Promiscuous teleological intuition is powerful, innate, and adaptive for the young. But it is not adaptive for the adolescent or adult leaving the human environment of the family hut for the natural world of the forest, where food does not present in the hands of intentional beings but on the branches of naturally occurring trees. Similarly, humanity may now use its growing knowledge and insight to escape its dependent teleological worldview in favor of a more mature one.

Ironically, the motto thus reinforced may be a biblically sourced one:

“When I was a child, I spoke like a child, I thought like a child, I reasoned like a child. When I became a man, I gave up childish ways.” (1 Corinthians 13:11)


Bakker, Gary M. 2013. God: A Psychological Assessment. Boca Raton, FL: Universal-Publishers.

Bering, J. 2012. The Belief Instinct: The Psychology of Souls, Destiny, and the Meaning of Life. New York: W.W. Norton & Co.
Casler, K., and D. Kelemen. 2008. Developmental continuity in teleo-functional explanation: Reasoning about nature among Romanian Romani adults. Journal of Cognition and Development 9: 340–362.

d’Mayberry, D. 2014. The psychology of faith: The human quest for meaning. The Australian Atheist 43(1): 15–18.

Kelemen, D. 1999a. Beliefs about purpose: On the origins of teleological thought. In M. Corballis and S. Lea (Eds.), The Descent of Mind: Psychological Perspectives on Hominid Evolution (pp. 278–294). Oxford, England: Oxford University Press.

———. 1999b. The scope of teleological thinking in preschool children. Cognition 70: 241–272.

———. 2003. British and American children’s preferences for teleo-functional explanations of the natural world. Cognition 88: 201–221.

———. 2004. Are children “intuitive theists”? Reasoning about purpose and design in nature. Psychological Science 15(5): 295–301.

Kelemen, D., M. Callanan, K. Casler, et al. 2005. Why things happen: Teleological explanation in parent-child conversations. Developmental Psychology 41: 251–164.

Kelemen, D., and E. Rosset. 2009. The human function compunction: Teleological explanation in adults. Cognition 111: 138–143.

Kelemen, D., J. Rottman, and R. Seston. 2013. Professional physical scientists display tenacious teleological tendencies: Purpose-based reasoning as a cognitive default. Journal of Experimental Psychology: General 142(4): 1074–1083.

Kelemen, D., R. Seston, and L. Saint Georges. 2012. The designing mind: Children’s reasoning about intended function and artifact structure. Journal of Cognition and Development 13(4): 439–453.

Lombrozo, T., D. Kelemen, and D. Zaitchik. 2007. Inferring design: Evidence of a preference for teleological explanations in patients with Alzheimer’s disease. Psychological Science 18(11): 999–1006.

Oliver, J.E., and T.J. Wood. 2012. Conspiracy theories, magical thinking, and the paranoid style(s) of mass opinion. University of Chicago Department of Political Science Working Paper Series. Online at http://political-science.uchicago.edu/faculty-workingpapers/.

Satel, S.L., and S.O. Lilienfeld. 2013. Brainwashed: The Seductive Appeal of Mindless Neuroscience. New York: Basic Civitas Books.

Taylor, M. 1999. Imaginary Companions and the Children Who Create Them. New York: Oxford University Press.

Gary M. Bakker

Gary M. Bakker is a practicing clinical psychologist and clinical lecturer at the University of Tasmania, Australia, who has published in both clinical (Practical CBT) and skeptical (God: A Psychological Assessment) fields.