The Psychology of Religion, Chapter 20: Religious Abuse

Abuse is tragically common. It affects every kind of community and family. Over the years I have seen numerous cases in which religious texts, or elements of religious faith, were used as instruments to abuse innocent children. (To protect privacy, identifying details have been altered, and some of the following examples are composites.)

Among them is the worst case of emotional abuse I have encountered in my career.

A teenager—gentle, intelligent, altruistic, living in a wealthy household—was subjected to forced "family sessions" late at night. She would be made to sit for hours in her bedroom while various family members recited Bible passages in a harsh, accusatory tone, orchestrated by a brutal, controlling father. The purpose was not moral guidance; it was humiliation and intimidation.

The teenager was, in fact, actively engaged in altruistic leadership at a church. But the family accused her of hypocrisy and of being a "false disciple," citing passages such as Matthew 7:21–23 and Matthew 23:13–20, telling her repeatedly that "God has abandoned you," and threatening her with hell. The irony, of course, is that those very passages condemn hypocritical, performative, self-aggrandizing religiosity.

The family would then pivot to the Old Testament, including Deuteronomy 21:18–21, which prescribes that a "stubborn and rebellious son" be stoned to death by the community. (That the passage concerns a son seems not to have troubled them.) It is worth noting what the tradition itself did with this text: in the Talmud, the rabbis hedged the law of the rebellious son with so many conditions that they pronounced it had "never happened and never will happen," preserved in scripture only so that one "may expound it and receive reward"—a teaching exercise, never a licence to kill. But this family took precisely the opposite attitude, brandishing a commandment their own tradition had disarmed two millennia earlier. Because the girl was devout herself, the experience was not merely frightening; it was torturous, and permanently traumatizing—especially in combination with the family's other abuse and neglect.

These episodes were interspersed with the family's evangelical outreach in the community, "to spread the word." As is so often the case, the parents were regarded by many others as pious and respectable. Abusive behaviour has complex causes, and in the absence of religion these parents might well have found another vehicle for cruelty. But in this family the abuse intensified as religious involvement intensified. Congregants who knew what was happening were horrified—yet very few intervened beyond offering prayer.

In a second case, the children of a deeply religious mother endured profound daily neglect and emotional abuse for years. Once again, members of the religious community did little to change the situation other than pray. When one of these children later went to live with the other, non-religious parent, her quality of life improved dramatically. She grew into an intelligent, kind, and accomplished young woman—though she still carries post-traumatic symptoms from that earlier phase of life.

Clinicians have begun to give this kind of injury a name: what some now call religious trauma, or spiritual abuse. Neither term is a formal diagnosis, and whether it names a distinct condition is still debated—but the symptoms often overlap with those of post-traumatic stress disorder.

In a third case, a family that had once been happy and well-integrated with its extended kin seemed to change in personality as it became absorbed in extreme fundamentalist religion. They grew dark, angry, and suspicious, and eventually estranged themselves from everyone else. Threatening posters appeared on their property bearing scriptural warnings about hell. Attempts to reach out with kindness were met with scolding condemnations about religious difference. A particular low point was an angry, rambling religious tirade delivered during the funeral of a family elder. These changes tracked precisely with the family's deepening insularity and its growing commitment to extreme belief and practice. To this day I feel badly for the children who had to grow up in that home.

I have seen many estrangements: religious parents ostracizing, shaming, or shunning their children over differences of lifestyle or belief—sometimes with the encouragement and applause of the religious community. In other cases, religious adults shunned their own ageing parents, depriving them of access to grandchildren, always with some pious justification.

It would be a mistake, however, to read these cases as a verdict on religion in general. A clinician sees the casualties, not the far larger number of families for whom religious life is quiet, ordinary, and unremarkable, and who therefore never come to clinical attention. Most religious parenting is ordinary or better. An authoritarian layer within it tilts toward harsher, more physical discipline, often with negative effects, and only a small, extreme minority shades into genuine abuse. So the honest picture is neither that religion harms children nor that it protects them: it is that religion is mostly beside the point, except where it becomes a vehicle for something else.

It is sometimes argued that religion is positively good for children, since some studies—Bartkowski and colleagues, for instance—report an association between parental religiousness and desirable child outcomes. But such studies are correlational, confounded by nearly everything that inclines people toward religion in the first place: conscientiousness, marital stability, social class. Where religious life does help, it helps through nonspecific factors—community, routine, social support, a shared moral vocabulary—not through the supernatural content of belief. The most secular societies on earth raise children at least as well as the most devout.

This cuts both ways. If religious belief is not the cause in families that flourish, neither is it usually the root cause in families that harm—an abuser deprived of scripture would likely have been abusive in some other way. But an authoritarian, fear-based religion can be a fearsome weapon in the hands of a tyrant: the authority of God behind the father's will, the infinite stakes of hell behind every threat, scripture as a ready-made prosecutorial script, and a congregation to approve the verdict. The psychologist Bob Altemeyer, who spent a career studying authoritarianism, described exactly this cluster—submission to authority, aggression toward the nonconformist, and rigid conventionalism—and found it strongly correlated with religious fundamentalism. What I had watched play out in a handful of families turns out to be a documented and predictable psychological type. And the girl's own devotion deepened the wound, because a threat of damnation lands hardest on someone who genuinely believes in hell.

One phrase I have heard from abusive religious parents is "turn or burn." I find it typical of a conviction that lurks in the background: if you don't adopt my belief, you deserve to be tortured forever. Such a phrase can be offered as an invitation, but it functions as a threat. Surely, if a way of life is divinely inspired, it ought to be compelling because it is beautiful and ethically coherent—not because it terrifies people into compliance.

It can be helpful to listen to those who have escaped abusive religious communities. Megan Phelps-Roper was raised in an extreme religious sect and was once among its most fervent young members. Her most useful insight is not a clever argument against dogma but a relational one: what changed her mind was sustained contact with outsiders who treated her with patience and respect—people willing to build a human connection before attempting to debate her beliefs. This is not merely an inspiring story; it aligns with some of the best-supported findings in the study of persuasion. The way to reach the faithful is not to defeat them in argument but to remain in relationship with them.

If this is familiar from the inside, it is worth knowing that people do leave, that the symptoms can ease with time and good support, and that help is available from clinicians and from communities of others who have gone through similar things.

References

Altemeyer, B. (1996). The authoritarian specter. Harvard University Press.

Altemeyer's research on right-wing authoritarianism—the cluster of authoritarian submission, authoritarian aggression, and conventionalism. He reports that the trait correlates strongly with religious fundamentalism and predicts punitiveness, the policing of nonconformity, and hostility toward outsiders. With Bruce Hunsberger, Altemeyer also studied why people enter and abandon fundamentalist faith.

Bartkowski, J. P., Xu, X., & Levin, M. L. (2008). Religion and child development: Evidence from the Early Childhood Longitudinal Study. Social Science Research, 37(1), 18–36. https://doi.org/10.1016/j.ssresearch.2007.02.001

A large national study reporting an association between greater parental religiousness and a range of desirable outcomes in young children, on both parental and teacher report. The association is correlational and heavily confounded—by conscientiousness, family structure, social class, and community—and so does not establish that religious belief itself does the causal work. It represents the strongest available form of the claim that religion benefits children, and, read carefully, the clearest caution against overstating it.

Broockman, D. E., & Kalla, J. L. (2016). Durably reducing transphobia: A field experiment on door-to-door canvassing. Science, 352(6282), 220–224. https://doi.org/10.1126/science.aad9713

A landmark field experiment showing that a single brief, non-judgmental, perspective-taking conversation produced a measurable reduction in prejudice that persisted for months and withstood subsequent counterargument. It is among the strongest pieces of causal evidence that respectful, relational contact changes minds more durably than confrontation does.

Phelps-Roper, M. (2019). Unfollow: A memoir of loving and leaving the Westboro Baptist Church. Farrar, Straus and Giroux.

A memoir of growing up inside, and eventually leaving, the Westboro Baptist Church. Phelps-Roper attributes her change of heart not to anti-religious argument but to the patient, respectful engagement of strangers—several of whom she first encountered while running the church's Twitter account. Her 2017 TED talk makes the same case in brief.

Winell, M. (2011). Religious trauma syndrome [Three-part series]. Cognitive Behavioural Therapy Today (British Association for Behavioural and Cognitive Psychotherapies), 39(2–4).

A clinical psychologist's coinage of the term "Religious Trauma Syndrome" for the harm sustained by those leaving authoritarian, dogmatic religion. The label is not a formal DSM or ICD diagnosis—it is generally subsumed under post-traumatic stress—but it has clinical currency, and Winell is careful to confine it to authoritarian and fundamentalist forms of religion. Lisa Oakley's work on "spiritual abuse" offers a complementary, empirically grounded framework.

A note on scriptural references: Matthew 7:21–23 and Matthew 23:13–20 record Jesus's denunciations of hypocritical and self-aggrandizing piety; Deuteronomy 21:18–21 sets out the law of the "stubborn and rebellious son." On the rabbinic neutralisation of the latter, see the Babylonian Talmud, Sanhedrin 71a, which declares that the case "never happened and never will happen," and was given only "that you may expound it and receive reward."
 

← Chapter 19 | Contents | Chapter 21 →

The Psychology of Religion, Chapter 19: Object Relations

Humans have a far more richly developed capacity for imagination than other animals, and one of its quietest uses is that we can hold inside the mind a representation of the people who matter to us. This is not childish fantasy; it is a normal and indispensable developmental achievement—the capacity to keep another person in mind when they are not physically present. Think of how an adult, long after a parent has died, can still hear that parent's voice—its counsel, its disapproval, its reassurance—and be genuinely steadied by it. The person is gone; something of them remains, and it remains inside. This is one of the foundations of object relations theory, the branch of psychoanalysis—running from Fairbairn and Klein to Winnicott—that takes the inner world to be populated not by drives alone but by internalized images of others.

Developmentally, we are at first comforted by a literal parent who is physically there. Over time we also come to carry in memory an internalized representation of that parent—an inner sense of their presence, their values, their voice—which can console us even when we are alone. The developmental psychoanalyst Margaret Mahler called the mature form of this achievement object constancy: the gradual consolidation of a stable inner image of the caregiver, one that survives the caregiver's absence. It is what allows a small child to tolerate separation without panic, and it is the same inner resource we draw on much later to endure loss and, eventually, grief. The person is gone; the internal representation remains.

For many people, religious life includes this type of internalized relationship, with an idealized figure they call God. Ana-María Rizzuto, in her study The Birth of the Living God, showed that each believer creates a personal "God representation" out of the same materials as the rest of the inner world—parental images, early longings, the imprint of how love and authority were first encountered. In much Western Christian imagery, and in many people's mental pictures, this figure is imagined in human form: often a bearded white man, despite the Middle Eastern setting of the biblical "Holy Land" and the obvious diversity of human appearance. (Other traditions refuse the picture entirely—Islam and Judaism are largely aniconic—which is itself a revealing difference.) Some people experience this inner figure as gentle, fatherly, wise, consistent, almost coach-like or therapist-like; others internalize a God who feels stern and frightening, poised to punish. These images tend to track what a person first learned to associate with safety, love, and authority in their own family and community.

The psychologist Lee Kirkpatrick took this a step further, arguing that the bond with God behaves like an attachment relationship in John Bowlby's sense: a figure turned to for comfort in distress, a "secure base" from which to face the world, and a "safe haven" to return to when afraid. It is why people so often pray hardest in fear and grief—the same conditions under which a frightened child reaches for a parent.

Now, is such a figure "real"? Here the most useful idea in this book comes from Donald Winnicott. Winnicott studied the child's first treasured possession—the blanket or the soft toy that must not be washed and must not be lost—and called it a transitional object. Its importance lies in a peculiar dual status: the blanket is not simply found in the world, like a stone, nor simply invented, like a daydream. It lives in between. And Winnicott's crucial move was to insist that we must never ask the child the question—"Did you find this or make it up?"—because the question itself misunderstands the kind of thing it is. He came to believe that a whole intermediate area of human life works this way: art, culture, play, and religious experience all inhabit the space that is neither purely inner nor purely outer. This reframes the question the whole book keeps circling. To demand of the believer, "Is your God real or imaginary?" may be, in Winnicott's terms, the wrong question—the one that misses what sort of thing a God representation is, and what it is for. That the figure is fashioned inwardly does not make it a lie; the inner mother who steadies the grieving adult is fashioned inwardly too, and no one calls her a delusion.

After all that, I'll confess something. I have often found this kind of psychoanalytic theorizing tiresome—it is the impression I have had all through my career, and one reason I compared psychoanalysis to religious doctrine earlier in this book. Insightful at times, yes, but burdened with jargon and a generous dose of overconfidence on the theorist's part, and sometimes needing hundreds of pages to convey a few simple ideas. Here is the whole thing condensed, at least as I see it: we form imaginary friends, many of them crafted from past experience, present longing, and our own personality. God, in a sense, is the ultimate imaginary friend.
But as with any very important friendship, we can be intensely loyal, and we can defend it fiercely. Sometimes people go to extremes, including a willingness to suffer or die in service of what they experience as sacred. And because the relationship is felt as profoundly real, it is no surprise that believers often respond with anger or grief when someone calls it "imaginary," a mere construct rather than an external reality. They are not defending a proposition; they are defending a bond.

There is a darker side to this. Many traditions personify not only the good but also the bad—a concept of ultimate evil, cast in devil-like terms. This is an example of what the psychoanalyst Melanie Klein called splitting: the mind's tendency, under strain, to divide experience into an all-good idealized figure and an all-bad persecutory one. The loving God and the malevolent Devil are a textbook split pair, with every scrap of goodness gathered onto the one and every scrap of badness projected onto the other.

Psychologically, a personified evil makes moral struggle vivid and coherent: it recasts temptation, cruelty, or one's own worst conduct as a battle against an outside adversary, rather than something to be faced in oneself. In a tight-knit community this can even ease reintegration. If a wrong can be blamed on the Devil rather than the wrongdoer's character—particularly once a ritual of repentance, prayer, or "deliverance" has been performed—the community may find forgiveness easier.

But to locate evil outside oneself is to blunt accountability; and once evil has a face, it is too easily seen in people we simply dislike—in outsiders, dissenters, scapegoats. The historian Elaine Pagels followed this in early Christianity: the figure of Satan, she argued, became the name for the human "other," the device by which a community demonized its rivals. The result, down the centuries, was fear, prejudice, and recurrent moral panic—the same dynamic I described earlier in connection with the witch hunts, in which a community's anxieties were given a satanic shape and then hunted down among its own neighbours. The inner world, it turns out, does not stay inner. We build our gods and devils from the materials of early life, and then we live—sometimes gently, sometimes violently—as though they had built us.

References

Kirkpatrick, L. A. (2005). Attachment, evolution, and the psychology of religion. Guilford Press.

Extends John Bowlby's attachment theory to religion, treating a believer's felt relationship with God as an attachment bond—a source of comfort in distress, a secure base, and a safe haven in fear. Kirkpatrick marshals evidence that people turn to God under the same conditions that lead a child to seek a parent, and that a person's style of attachment to God partly mirrors their attachment history with early caregivers. 

Mahler, M. S., Pine, F., & Bergman, A. (1975). The psychological birth of the human infant: Symbiosis and individuation.Basic Books.

The classic statement of the separation–individuation process in early childhood. Its culminating phase is the consolidation of emotional object constancy—a stable inner representation of the caregiver that endures through absence and underwrites the later capacity to tolerate separation and loss. 

Pagels, E. (1995). The origin of Satan. Random House.

Traces how the figure of Satan, in early Judaism and Christianity, became a means of demonizing one's opponents—rival Jews, then pagans, then heretics—transforming ordinary human conflict into cosmic warfare. Pagels shows that the Devil is most often invoked not against distant strangers but against intimate enemies: members of one's own broader community who have come to be seen as traitors. A personified evil slides readily into the persecution of the human "other."

Rizzuto, A.-M. (1979). The birth of the living God: A psychoanalytic study. University of Chicago Press.

The foundational object-relations study of the personal "God representation." Drawing on Freud, Erikson, Fairbairn, and Winnicott, Rizzuto shows that each individual constructs a private image of God from parental figures and early experience—one that functions as a living element of the inner world rather than an abstract doctrine. The work effectively launched the empirical study of God-images.

Winnicott, D. W. (1953). Transitional objects and transitional phenomena: A study of the first not-me possession. International Journal of Psycho-Analysis, 34, 89–97.

Introduces the "transitional object"—the child's blanket or toy—and the "intermediate area" of experience, neither wholly inner nor wholly outer, of which Winnicott insisted one must never ask whether it was created or found. The idea was revised as the opening chapter of Playing and Reality (1971), where Winnicott extended it to art, culture, and religion. 

← Chapter 18 | Contents | Chapter 20 →

The Psychology of Religion, Chapter 18: Prayer

Prayer may mean different things to different people. For many, it is essentially a meditative act: a kind of philosophical reflection on existential themes, a relaxation practice, a "grounding" moment, a way to name fears and hopes, to confess or grieve, to feel less alone. Some who pray believe they are holding a conversation with God, though the manner in which God is understood to "speak back" is usually figurative—a new idea, an inclination, a redoubling of confidence, a wave of emotion that feels like guidance. Others do not expect God to speak back at all; they are content simply to vent, confess, grieve, or reflect within a reverent frame. In this it resembles classical psychoanalysis, where the listener is largely silent and the act of speaking—slowly, honestly, repeatedly—becomes the mechanism.

In fact, understood this way, prayer is close to the purest instance of the mechanism I set out earlier in this book. Recall Jerome Frank's four elements shared by all forms of healing: a confiding relationship with a helper, a setting set apart from ordinary life, a rationale the sufferer believes, and a ritual that enacts it—with faith supplying the expectation that it will help. Prayer has every one of them. God is the perfectly patient, perfectly available listener; the sanctuary, or simply the bowed head and the closed eyes, is the setting; the theology supplies the rationale; the words themselves are the ritual; and faith consolidates the whole. Whatever good prayer does for a person, in other words, it does through the same channels that make psychotherapy work—which is precisely why it can steady and console even if no one is listening on the other end.

But many people also pray for things—for an outcome to change, for an illness to heal, for a surgery to go well, for a war to end, for a relationship to mend. That kind of prayer is different. If it were literally effective, it would mean that events in the physical world were being altered—that something in the ordinary chain of causation was being nudged off course on request. And if that were happening consistently and repeatably, you would expect to see clear clusters of unusually good outcomes wherever people pray the most, or wherever the "right" kind of prayer is most common. You would expect the world to look, in its more devout regions, as though the rules of physics were being bent to order. I am not aware of any such pattern.

When researchers have tried to test this carefully—especially with intercessory prayer, praying for someone else—the results have produced no solid, repeatable signal. The best-known example is the STEP trial in cardiac bypass patients: people were randomly assigned to receive or not receive intercessory prayer, with a third group told with certainty that they were being prayed for. Overall, prayer did not reduce medical complications. The one statistically reliable effect ran the wrong way—and it was not about whether people were prayed for but whether they knew it. Among patients who all received prayer, complications struck 59% of those told with certainty it was happening, against 52% of those left uncertain. One plausible reading is psychological: being told "people are praying for you" can quietly raise the stakes. What if I don't get better? What would that mean—about me, about my faith, about God? For someone already frightened and vulnerable, that extra layer of expectation and scrutiny, the sense that a spiritual test is under way, may add stress rather than comfort.

The same logic can be run at the scale of whole societies. If prayer—or divine favour more broadly—were an instrumental force that altered physical reality, then the most devout regions ought to enjoy a measurable dividend: higher rates of recovery, fewer disasters, lower mortality, all else being equal. No such supernatural dividend appears. If anything, when prosperous democracies are compared, the more secular ones tend to score at least as well as the more religious ones—often better—on the standard markers of societal health.

I want to be careful about what this does and does not show, because the serious scholarship points in a particular direction. Pippa Norris and Ronald Inglehart, comparing societies worldwide, argue that the causal arrow runs largely backward: it is existential insecurity—poverty, danger, fragile institutions—that drives populations toward religion, and the security and prosperity that reduce human suffering that erode the felt need for it. On that reading, secular societies are not healthy because they are secular; they are secular because they had already become safe. The cross-national comparisons are in any case correlational and run at the level of whole nations, so they cannot by themselves establish what causes what. (The most-cited of them, by Gregory Paul, was offered explicitly as "a first look," and was sharply criticized for exactly this.) But notice that none of this rescues the supernatural claim—it buries it. If devout populations enjoyed real divine protection, the secular societies should be visibly worse off, and they are not, on any reading of the arrow. Whatever genuine benefits religion confers—and they are real: comfort, cohesion, a shared story for making sense of loss—appear to travel through ordinary social and psychological channels rather than through any bending of the physical world.

Spatial Language

One small feature of religious behaviour, inherited from ancient practice, is its spatial language: "God above." People sometimes literally look upward when they pray. But "up" points in different directions depending on where you stand on Earth, and it changes minute by minute as the planet rotates, orbits the sun, and travels through the galaxy. A person in Australia looking up toward Heaven is facing the same direction as a person in North America looking down into the ground. It is a pre-Copernican metaphor, entangled with the older intuition that up is good and down is bad—one of the most basic ways the human mind maps value onto space.

Of course, looking upward is often figurative—but many people do take it quite literally, and if one were going to, it would be just as "valid" to look downward, or inward into one's own body. If God is omnipresent, shouldn't God be as present in the depths of the planet, or in our own cells, as in the sky? The gesture tells us less about the geography of a deity than about the structure of the human imagination.

A related embodied gesture appears in some fundamentalist worship: people in an entranced state reach forward with their hands during songs or prayer—eyes half-closed, rocking, repeating sacred phrases, the emotional intensity magnified by the synchrony of the crowd. This is a normal human ecstatic capacity, present in all cultures with or without religion. But the gesture still implies a location for God—reaching out to take God's warmth in one's hands, as though God were physically just ahead, perhaps at the front of the building. Again, the scene tells us a great deal about embodied human longing and very little about the whereabouts of a deity.

Prayer and Empathy

The moral structure of prayer often mirrors the moral structure of empathy. Many prayers are genuinely compassionate: people think of struggling friends, of relatives in trouble, of terrible events in the news, and ask for comfort, protection, and healing. But if prayer is believed to cause divine comfort to arrive, an uncomfortable counterfactual follows: had the prayer not occurred, would the comfort have been withheld? Shouldn't a loving deity comfort the suffering regardless of whether anyone happened to pray—especially since some of the worst suffering on earth happens in isolation, unwitnessed, with no one even aware enough to pray? It implies a troubling arrangement in which help depends not on who suffers most, but on who is fortunate enough to be noticed.

This is where it helps to recall Paul Bloom's critique of empathy, which I have reviewed at length elsewhere. Empathy, Bloom argues—provocatively, and against much received opinion—works like a spotlight: it is biased, and therefore often unjust. It is pulled toward those who resemble us, toward vivid stories and dramatic suffering, while it neglects the quiet, the distant, the stigmatized, and the statistically larger tragedies that arrive without a single tear-streaked face. Prayer tends to inherit the same distortion. We pray intensely for the salient and the familiar, and far less for abstract fairness, or for the invisible victims who never enter our field of attention.

Many prayers, of course, are not about others at all; they are wishes for oneself. There are battlefield prayers, prayers before surgery, prayers for money or a job or the return of an ex-partner, prayers for relief from pain, prayers over the outcome of a baseball pitch. As a meditative act this is entirely understandable. But it can set up a reinforcement loop of exactly the kind I described earlier in connection with other unfounded but comforting beliefs. If the prayer is followed by a good outcome, it naturally feels as though it worked, and the person is bolstered to pray again. If the outcome is bad, the failure is easily absorbed: perhaps the prayer was not sincere enough, or not faithful enough; perhaps God said no, or was testing them. Either way the practice is insulated from disconfirmation—validated whatever happens.

That is why prayer can "work" psychologically even when its supernatural claims are false. As meditation and reflection, it genuinely calms and organizes the mind. As a lever on the physical world it shows no measurable effect—yet it remains self-confirming, and that combination carries a hidden cost. It quietly places the burden on the believer, fostering the illusion that one's own spiritual effort is the decisive factor in how events unfold. We glimpsed the sharp end of this in the prayer trial, where the patients most certain of being prayed for fared slightly worse: once an outcome is felt to ride on the sufficiency of one's faith, comfort and pressure become difficult to tell apart.

References

Benson, H., Dusek, J. A., Sherwood, J. B., Lam, P., Bethea, C. F., Carpenter, W., Levitsky, S., Hill, P. C., Clem, D. W., Jr., Jain, M. K., Drumel, D., Kopecky, S. L., Mueller, P. S., Marek, D., Rollins, S., & Hibberd, P. L. (2006). Study of the Therapeutic Effects of Intercessory Prayer (STEP) in cardiac bypass patients: A multicenter randomized trial of uncertainty and certainty of receiving intercessory prayer. American Heart Journal, 151(4), 934–942. https://doi.org/10.1016/j.ahj.2005.05.028

The largest and most rigorous trial of intercessory prayer, following more than 1,800 cardiac bypass patients across six hospitals. Being prayed for made no difference to complication-free recovery; the only statistically reliable effect was that patients who were certain they were being prayed for had slightly more complications (59%) than those who were prayed for but left uncertain (52%). The study was funded by the John Templeton Foundation, which supports research at the intersection of science and religion.

Bloom, P. (2016). Against empathy: The case for rational compassion. Ecco.

Argues that empathy—feeling what another feels—operates like a biased spotlight, favouring the near, the vivid, and the similar while neglecting the distant and the statistical, so that it frequently misguides moral judgment. Bloom proposes "rational compassion" in its place. 

Lakoff, G., & Johnson, M. (1980). Metaphors we live by. University of Chicago Press.

The founding work of conceptual-metaphor theory, arguing that abstract thought is structured by bodily and spatial metaphors rather than floating free of them. Their "orientational metaphors"—GOOD IS UP, MORE IS UP, and their downward opposites—map value onto space in ways so basic we rarely notice we are doing it. 

Norris, P., & Inglehart, R. (2004). Sacred and secular: Religion and politics worldwide. Cambridge University Press.

Develops the influential "existential security" thesis through comparison of societies worldwide: populations facing insecurity and hardship tend to be more religious, while security and prosperity drive secularization. On this account the correlation between secularism and societal health reflects causation running largely backward—safe, prosperous societies tend to become secular, rather than secularism making them safe. 

Paul, G. S. (2005). Cross-national correlations of quantifiable societal health with popular religiosity and secularism in the prosperous democracies: A first look. Journal of Religion & Society, 7. (Open access)

A cross-national comparison reporting that the more secular prosperous democracies tend to have lower rates of societal dysfunction than the more religious ones—the source of the chapter's "no supernatural dividend" observation. As the author's own subtitle concedes, it is "a first look": the analysis is ecological and correlational, and was sharply criticised (notably by Rodney Stark) for the ecological fallacy and for selective measures. 

← Chapter 17 | Contents | Chapter 19 →

The Psychology of Religion, Chapter 17: Shepherding

A related religious metaphor—gentler on its surface than sacrifice, but, as we will see, not unrelated to it—is shepherding. Jesus is called the "Good Shepherd," and there are many other biblical passages that liken God to a shepherd. It is a beautiful image, and as a child I absorbed it in exactly that spirit: kindly pastoral artwork, a gentle man with a hooked staff, sunny hills, a flock of woolly friends, perhaps one little sheep who has wandered off and needs to be carried back to safety.

But it is worth pausing to remember what shepherding actually meant in that time and place. Sheep were not kept as pets. They were livestock: valued for wool and milk, yes, but also raised for meat—and sometimes for sacrifice. As we saw in the previous chapter, sacrifice was no tidy abstraction. In the rite set out in Leviticus and elaborated in the later rabbinic descriptions of the Temple, the animal was secured with rings before the altar, its throat was cut and its blood caught in a special container and dashed against the altar; the carcass was then hung from an iron hook, skinned, and its organs removed and burned.

A shepherd's role was not only protection and guidance; it also involved ownership, control, and—eventually—decisions about which animals would be killed, sacrificed, or eaten. In that light, "being shepherded" contains an unsettling double meaning: you are kept from straying, guarded from wolves, and held within the safety of the flock—but you are also being managed toward ends that are not your own.

And if we push the image just one step closer to lived reality, it gets darker in a way the children's illustrations never hinted at. Imagine being a sheep in the flock. Every so often the younger males—your brothers, or your cousins—are taken away. Perhaps they are led toward a little shed at the edge of the field, or down a path behind a stand of trees, and they are simply never seen again. The flock goes on grazing. The shepherd is still "protecting" the flock. But the protection is inseparable from a system in which some members are quietly designated for disappearance.

To be fair, the Christian image in particular tries to invert the usual arrangement: the "Good Shepherd" is portrayed as laying down his life for the sheep. That is morally striking. And the traditions are not naïve about bad shepherds, either: the prophets turn the very same metaphor into an indictment of corrupt leadership. Ezekiel devotes an entire chapter to denouncing the "shepherds of Israel" who feed themselves while the flock starves, who rule harshly and leave the weak uncared for, and Jeremiah does much the same—a reminder that the image was always double-edged, and could be turned against the powerful as readily as it sanctified them.

Still, the metaphor does something psychologically and socially important: it trains us to admire a certain kind of relationship—one in which docility is a virtue, "straying" is a moral failure, and the authority to define what counts as straying belongs to the shepherd. This intuition has a long intellectual lineage, and the vocabulary itself gives the game away: a "pastor" is literally a shepherd, "pastoral care" is the tending of a flock, and a "congregation" is, etymologically, a gathered herd. Friedrich Nietzsche gave the underlying suspicion its most savage form. He attacked what he called "slave morality"—a value system he attributed above all to Christianity, which (in his reading) inverts aristocratic virtues, rebrands meekness, obedience, and self-suppression as goodness, and treats the docile herd animal as the moral ideal. Michel Foucault, more coolly, gave the structure a name: "pastoral power," a distinctively Western form of authority, inherited from the Church, that governs not a territory but souls—caring, all-seeing, devoted to your welfare, and expecting obedience in return—and that was later secularized into the modern state's management of whole populations. Tellingly, even Foucault's admirers note that his pastoral idyll leaves one thing out: the slaughter.

Now I think the metaphor is more revealing than its gentle surface suggests. The phrase "sheep gone astray" recurs throughout scripture, almost always as a rebuke for human misbehaviour—but notice that the actual sheep who never stray do not thereby graduate into freedom. They remain in the flock, under management, until the flock's purposes require otherwise. Taken to its literal conclusion, the image idealizes a passive, domesticated existence in which obedience is the highest virtue and "straying" the cardinal sin; in which the protection on offer is inseparable from ownership; and in which the reward for a lifetime of perfect docility is to remain in a flock whose ends are set entirely by someone else. As a child, I never thought about any of this. The pastoral image comforts us, in part, precisely because it asks us not to look too closely at where the path behind the trees actually leads.

References

Foucault, M. (2007). Security, territory, population: Lectures at the Collège de France, 1977–1978 (M. Senellart, Ed.; G. Burchell, Trans.). Palgrave Macmillan. (Original lectures delivered 1978)

Develops the concept of "pastoral power": a distinctively Western mode of authority, inherited from the Christian Church, that governs individuals as a shepherd governs a flock—caring, individualizing, and totalizing at once, and oriented toward obedience and salvation. Foucault traces how this logic was later secularized into the modern state's management of whole populations. Commentators have noted that his account conspicuously omits the slaughter that real shepherding entails.

Nietzsche, F. (1967). On the genealogy of morals (W. Kaufmann & R. J. Hollingdale, Trans.). Vintage Books. (Original work published 1887)

Nietzsche argues that a "slave morality," which he attributes above all to Christianity, inverts aristocratic values so that meekness, obedience, and self-suppression are recast as virtues and the docile herd animal becomes the moral ideal. The companion argument appears in Beyond Good and Evil (1886). The vocabulary of the "herd" (Herde) recurs throughout both works.

The "Good Shepherd" passage is John 10:11–18; the Lord as shepherd, Psalm 23. The motif of straying sheep appears at Isaiah 53:6, Luke 15:3–7, and Matthew 18:12–14. The prophetic use of the shepherd metaphor to condemn exploitative leadership is found in Ezekiel 34 and Jeremiah 23:1–4. The burnt-offering rite is described in Leviticus 1; the Second Temple details—rings to secure the animal, iron hooks for flaying, and the blood caught in vessels and dashed against the altar—are preserved in the Mishnah (Tamid 3; Middot 3:5; Pesaḥim 5:9).

← Chapter 16 | Contents | Chapter 18 →

The Psychology of Religion, Chapter 16: Sacrifice

Most religions have some form of sacrifice woven into their theology. Sometimes this involves literal offerings—killing and burning animals, or destroying valuable objects. Other times it is "bloodless": giving money, time, obedience, or the renunciation of pleasures through fasting, abstinence, or celibacy. In all these cases the underlying idea is the same: something costly is offered up, in the hope of securing meaning, favour, purity, forgiveness, protection, or belonging.

Some sacrificial motifs move disturbingly close to human sacrifice. In the Abrahamic traditions, the willingness of Abraham—Ibrahim—to sacrifice his son is presented as the peak test of obedience. In Islam it is commemorated each year at Eid al-Adha, the "Festival of Sacrifice," where the animal offering serves as a memorial of that story. And in Christianity the theme is carried into the central story of Jesus—a dramatic reframing of sacrifice into self-sacrifice, offered "for others."

These stories share a hidden architecture: substitution. The ram caught in the thicket dies in Isaac's place; the Levitical scapegoat is loaded with the community's sins and driven into the wilderness to carry them away; and Christianity universalizes the logic completely, casting Jesus as the "Lamb of God" whose single sacrifice is meant to end the need for any further sacrifice at all. Much of the moral force of sacrifice lives here—in the idea that suffering can be transferred, that one life or one death can be made to stand in for another.

Across much of the ancient world, sacrificial traditions were common, and often brutal. Ancient Greek religion had animal sacrifice. Vedic religion in India revolved around yajña, the sacrificial fire ritual. Ancient China had elaborate sacrificial practices directed toward ancestors and higher powers, sometimes involving animals and—as the royal tombs at Anyang attest, on a large scale—human beings. The Aztecs are especially notorious for human sacrifice; the numbers reported by their Spanish conquerors were very likely inflated, but archaeology at the Templo Mayor has confirmed that the practice itself was real and substantial.

Why would an all-powerful deity, especially one associated with the highest moral standards, want a dead animal or a burnt work of art as a gift? One might think a god worth revering would count it a gift if you helped other people or cared for the natural world, rather than destroyed objects or killed living things. But sacrificial systems do not usually work that way.

This objection is not a modern or anti-religious invention; the traditions raised it against themselves long ago. The Hebrew prophets are scathing on exactly this point: "I desire mercy, not sacrifice," says Hosea; Micah asks whether the Lord wants "thousands of rams," or rather that one do justice and love mercy; Amos has God declare that he despises the festival offerings and demands instead that "justice roll down like waters." Socrates presses the same nerve in Plato's Euthyphro, mocking the notion of piety as a kind of commercial exchange with the gods—a trading relationship in which the right offering buys favour. The most humane corrections to religion have often come from inside it, and the critique of sacrifice is a striking early example.

Reciprocity, Magical Thinking, and Social Technology

Sacrifice is, in my view, an extension of ordinary human ideas about reciprocity and gratitude—infused with magical thinking. In a community we do favours, give gifts, and care for one another; these behaviours can be altruistic, but they are also held in place by norms of reciprocity. If one believes that a mystical power controls destiny, fertility, weather, health, wealth, or victory in war, it becomes psychologically "reasonable," within that worldview, to give that power a gift—and to hope for a return.

And once a person enters this mindset, the logic can become self-sealing. If you make a sacrifice and misfortune still comes, you can conclude the offering was not large enough, or not sincere enough, or not given with the right purity of heart—so next time you must give more. If something good happens afterward, it feels like proof that the sacrifice worked, and should be repeated. There is no result that can disconfirm the practice; every outcome feeds it. In this way sacrifice can escalate into something brutal and destructive—and the sacrificed animals, often the most vulnerable and least able to "consent" to the human story being told about them, get no say in the matter.

Another motivation for sacrifice likely came from the grim necessities of ancient life: hunting, or killing domestic animals for food. Most humans bond to animals easily, and it is troubling to watch an animal struggle and suffer. Ritual can function as a kind of moral anaesthetic—a way to consecrate violence, assuage guilt, and turn a hard necessity into a story of gratitude, order, and meaning.

Two scholars loom over any modern account of sacrifice, and it is worth setting these intuitions against theirs. Walter Burkert, in Homo Necans, argued that Greek sacrificial ritual grew out of the guilt and anxiety of the Palaeolithic hunt—that the solemn choreography around the killing is, at bottom, a way of managing the unease of taking life. That is close to the "moral anaesthetic" idea. René Girard, in Violence and the Sacred, proposed something darker: that sacrifice channels the contagious, escalating violence of a community onto a substitute victim—a scapegoat—whose death discharges the tension and restores peace, after which the victim is often made sacred. One need not accept either theory wholesale to notice that both converge with the argument here: sacrifice is a technology for metabolizing the violence and anxiety that group life generates.

Sacrifice can also be political performance. Public ritual consolidates hierarchy—especially priestly hierarchy—displays power, intensifies fear, and signals unity. It makes shared belief visible and costly. It puts loyalty on display: it shows who is serious, who is obedient, who can be trusted, and who has the authority to declare what counts as holy.

This is also where sacrifice connects to group psychology. The anthropologist Richard Sosis has argued that costly rituals—the things you would not do unless you were genuinely committed—work as hard-to-fake signals that strengthen trust and cooperation within a group, partly by screening out free riders. A community bound together by shared sacrifice can feel safer, warmer, and more morally serious to its members. But the same mechanism can harden boundaries and deepen suspicion of outsiders.

And costly sacrifice does not only send a signal to other people; it also works on the person making it. As we saw earlier with initiation, people are deeply reluctant to admit they have suffered for nothing. So the greater the sacrifice, the stronger the pressure to reinterpret the suffering as meaningful, noble, or necessary. That is what makes sacrificial systems self-protective: the cost itself becomes part of the "evidence" that the belief must matter.

Which raises a deeper question. Sacrifice is only the most vivid of a whole set of shared, costly markers—shared dress, shared food, shared song, shared taboo—and all of them seem to produce the same effect: they make strangers feel like family. But why should a shared ritual generate the feeling of kinship at all? To answer that, we have to look at something far older than any religion: the machinery by which living things came to favour their own kind in the first place.

Kin Altruism

Reciprocity has a deeper and older root than social custom. There is a strongly selected tendency to favour and help genetic relatives, sometimes even at real cost to oneself. The logic was formalized in 1964 by W. D. Hamilton: a gene that inclines an individual to help close relatives can spread down the generations, because those relatives are disproportionately likely to carry the very same gene. Hamilton's rule states the condition precisely—an allele prompting costly help toward kin can spread when the benefit to those relatives, weighted by the probability that they share the allele, outweighs the cost to the helper, even when that cost is steep.

But humans do not walk around calculating degrees of genetic relatedness. Instead we rely on crude, fast cues—ones that, over most of human history, were reasonably well correlated with kinship. People who live near one another, marry one another, and raise children together will, over generations, tend to share not only genes but language, accent, customs, dress, habits, and norms. On average, they will also tend to resemble one another physically more than they resemble people from a distant village or lineage. Conversely, people who look different, speak differently, or keep very different customs were, historically, more often from a different family network—and so, on average, somewhat less closely related.

Similarity of appearance, familiarity of accent, shared habits, shared rituals, shared dress, shared taboos—all of these can become proxies, very imperfect proxies, for "one of us." And religion supplies them in abundance: common dress, common restrictions, common foods, common sacrifices, common songs, common stories, common enemies. In effect, it manufactures the feeling of kinship among people who are not literally kin.

A caution is in order, because the step from "looks different" to "treated as an enemy" is looser than it first appears. In a well-known set of experiments, Robert Kurzban, John Tooby, and Leda Cosmides found that the mind does not seem to track race, or physical similarity, as a fundamental category at all; what it tracks is coalition—who is allied with whom. Subjects' tendency to sort others by race dropped sharply, within minutes, once they were given a more salient cue to who was on which team. If that is right, then physical appearance is encoded only insofar as it has historically predicted alliance, and it can be overwritten when a better predictor comes along. This does not weaken the argument so much as relocate it: religion's real power is that it floods the mind with vivid, durable coalition markers—shared dress, shared rites, shared sacrifice—and so recruits a coalitional psychology far older and deeper than any doctrine.

The mind appears to have evolved to be slightly more generous, trusting, and self-sacrificing toward those who are more likely to be "one of us," and correspondingly less generous, more wary, or more distant toward those who feel like "not us." These tendencies are not destiny, and they are not moral justification—but they do appear to be among the psychological and evolutionary foundations of prejudice. They are precisely the sort of inherited inclination we have to learn to recognize, question, and actively override.

Belonging and Group Boundaries

Religion can widen the circle of felt family. But it can also sharpen the line between those inside the group and those outside it. Once sacrifice, loyalty, and identity are fused, shared customs take on unusual emotional and moral weight, and group boundaries begin to feel enormously important. The stronger those boundaries, the easier it becomes to view outsiders with suspicion, distance, or moral distrust. This does not mean religion always produces hostility, or that it does so uniquely—these are broad features of human social psychology, and the willingness to sacrifice for one's own group has been linked, in some evolutionary models, to hostility toward rival groups, the two evolving together. But religion can give the whole pattern a sacred language, a ritual structure, and a heightened sense of seriousness. In that way, stronger religious boundaries can feed exclusion and, in some cases, open hostility between groups. Religion does not invent this psychology—but it can consecrate it.

References

Aronson, E., & Mills, J. (1959). The effect of severity of initiation on liking for a group. The Journal of Abnormal and Social Psychology, 59(2), 177–181. https://doi.org/10.1037/h0047195

The classic effort-justification experiment: people who underwent a more embarrassing initiation in order to join a group afterward rated the group more highly, not less. The result has been replicated many times and is counterintuitive—the steeper the price of entry, the greater the reported attachment.  The cost of a sacrifice tends to deepen commitment to the belief it serves rather than weaken it.

Burkert, W. (1983). Homo necans: The anthropology of ancient Greek sacrificial ritual and myth (P. Bing, Trans.). University of California Press. (Original work published 1972)

A foundational anthropological account tracing Greek sacrificial ritual to the guilt and anxiety of Palaeolithic hunting, and reading the solemn choreography of the kill as a way of managing the unease of taking life. Burkert treats sacrifice as something humans developed to live with violence rather than to celebrate it. 

Choi, J.-K., & Bowles, S. (2007). The coevolution of parochial altruism and war. Science, 318(5850), 636–640. https://doi.org/10.1126/science.1144237

A formal evolutionary model showing how in-group self-sacrifice ("altruism") and hostility toward out-groups ("parochialism") can evolve together, each making the other adaptive: groups with altruistic warriors outcompete groups without them. The authors argue the two traits are not opposites but partners. The warmth of belonging and the suspicion of outsiders are two faces of a single mechanism.

Girard, R. (1977). Violence and the sacred (P. Gregory, Trans.). Johns Hopkins University Press. (Original work published 1972)

Advances the "scapegoat" theory of sacrifice: that a community discharges its own escalating, imitative violence onto a substitute victim, whose killing restores order and who is then made sacred. For Girard, the mechanism is largely hidden from the participants, which is part of how it works. 

Hamilton, W. D. (1964). The genetical evolution of social behaviour. I. Journal of Theoretical Biology, 7(1), 1–16. https://doi.org/10.1016/0022-5193(64)90038-4

The founding paper of kin-selection theory, introducing the rule now named for Hamilton: a gene for helping relatives can spread when the benefit to kin, discounted by the degree of relatedness, exceeds the cost to the helper. It gave a precise mathematical basis for the evolution of self-sacrificial behaviour toward family. 

Kurzban, R., Tooby, J., & Cosmides, L. (2001). Can race be erased? Coalitional computation and social categorization. Proceedings of the National Academy of Sciences, 98(26), 15387–15392. https://doi.org/10.1073/pnas.251541498

Experimental evidence that the mind encodes race not as a primitive category but as a cue to coalition or alliance, such that race-based sorting drops sharply once a competing cue to team membership is made salient. The finding suggests that what looks like an automatic focus on physical difference is really a search for who is allied with whom.  Appearance matters only as a proxy for "us versus them," and can be overwritten.

Sosis, R., & Alcorta, C. (2003). Signaling, solidarity, and the sacred: The evolution of religious behavior. Evolutionary Anthropology, 12(6), 264–274. https://doi.org/10.1002/evan.10120

A widely cited review of costly-signalling theories of religion, arguing that hard-to-fake ritual commitments stabilize cooperation by screening out free riders—those who would take the benefits of group membership without paying its costs. The more expensive the ritual, on this account, the more reliably it identifies the genuinely committed. 

The Binding of Isaac appears in Genesis 22:1–19; the parallel Qur'anic narrative is in Surah as-Saffat (37:99–113), where the son is not named. The prophetic critique of sacrifice is found at Hosea 6:6, Micah 6:6–8, Amos 5:21–24, 1 Samuel 15:22, Psalm 51:16–17, and Isaiah 1:11–17.

Socrates' treatment of piety as a form of exchange with the gods appears in Plato's Euthyphro*, especially at 14e–15a.*

← Chapter 15 | Contents | Chapter 17 →