The Folly of Fools: The Logic of Deceit and Self-Deception in Human Life - Robert Trivers (2011)

Chapter 12. Religion and Self-Deception


A book could be written on this subject—no, a twelve-volume treatise. Religion is a deep and complex subject, and so are its interactions with deceit and self-deception. Religions range from animists to monotheists to nontheists to atheists and then from Christian to Hindu to Buddhist to Muslim to Jew, with many subspecies. Here I can only hope to sketch out some of the major biological forces favoring religion and some of the important ways in which religion may encourage deceit and self-deception. This is a very tentative chapter, one that is heavily biased toward my own limited knowledge, namely that of the monotheistic religions of the West, chiefly Christianity, and not toward polytheism or the great Eastern religions. Nevertheless, I hope to show how religion and self-deception can interact in important ways and invite others to make the more important advances.

Some people think of religion itself as complete self-deception, all of it nonsense on its face, counterfactual, and in the extreme having nothing but negative side effects. In this view, the entire enterprise is self-deluded at the outset, so religion should be studied as a well-developed system of self-deception, as it certainly is—but is it only that? This may in fact be true, but these people have no theory for how this malady could have spread so far—to every culture and almost every human being in every culture—by self-deception alone.

What some have is a metaphor. Religion is a viral meme; that is, it is not an actual virus, which can easily bring a population to its knees, but rather it is merely a thought system that happens to propagate as if it were a virus, to the detriment of those with the belief system. Despite its negative effects, it apparently generates insufficient selection pressure to suppress the spread of this non-coevolving nonorganism. This is not a very impressive foundation for an evolutionary theory of religion, and it easily invites undue optimism regarding the life span of current religions. For example, from one such proponent of the meme-centered view: “I expect to live to see the evaporation of the powerful mystique of religion. I think that in about twenty-five years almost all religions will have evolved into very different phenomena, so much so that in most quarters religion will no longer command the awe it does today.” I think it more likely (though not the most likely) that twenty-five years from now, evolutionary biologists and philosophers will be in hiding from the then-dominant religious groups. Fifty years from now, no one stubbing his or her toe will say, “Charles Darwin, Charles H. Darwin, this hurts!” but they will still be saying, “Jesus H. Christ, this fucking hurts.”

At the same time, many, many people believe religion is the received truth from the Almighty—or, more to the point, that their religion is. Some have a book—a Torah, a Bible, a Koran—all of whose words are true, often literally so. Their view has as little backing as the viral-meme story and appears at first to be nothing more than a deep form of self-justification. If their own religion is God’s own truth, then competing religions are often seen as anti-truth, or the work of the devil, the ultimate target. So we begin with two very extreme views of religion, with the truth probably somewhere in between, but where exactly and why?

First we need to separate the truth value of religious statements from the possible benefits of believing in them, and likewise separate partaking in religious ceremonies from the truth value attributed to them. Then we need to analyze beliefs and behavior in a more fine-grained way so that we can evaluate the meaning and function of particular beliefs. In my own view, there is often an internal struggle within religions between general truth and personal or group falsehood. That is, the essence of religion is neither self-deception nor deep truth, but a mixture of the two, with self-deception often overwhelming truth.

Religions tend to increase within-religion cooperation at the cost of lowered cooperation with outsiders. Often this involves a false historical narrative and shared group self-deception: “We are the chosen people or the original people from creation or those whose beliefs [e.g., in the divinity of Jesus] cause God to favor us [or whatever].” In short, religions often act as templates for in-group/out-group biases. Insofar as they encourage in-group cooperation, many benefits may accrue, but insofar as they encourage in-group cooperation in aggressive attacks on out-groups, they both inflict harm on others as a price of their cooperation and inflict harm on self when they fail (which, in warfare, is roughly half the time).

At the same time, certain features of religion provide a recipe for self-deception, removing nearly all restraints from rational thought. The universal system of truth espoused by a religion usually gives special status to the believer. Various phantasmagorical things are easily imagined, and “faith” is permitted to supersede reason.

Religion has a complex relationship with health and disease. On the one hand, health may be a major selective factor favoring religious behavior and beliefs. Not only do religions often preach healthy behavior, but there also is evidence that religious belief and association improve individual survival, immune function, and health. Even music, so common in religion and courtship, has positive immune effects. Medicine was originally embedded within religion, and both provide strong placebo benefits to at least part of the population.

A completely unexpected association between disease and religion emerges when we study the entire globe for degree of religious diversity (number of religions per unit area) as a function of parasite load (roughly, degree of human loss due to parasites). Here we find many more religions (and languages) per square inch when parasites are high. Since splitting of religions is also naturally associated with ethnocentrism and ethnic differentiation, parasites are a factor expected to degrade general religious truth value over time and thus to be positively associated with in-group deceit and self-deception. This may be especially true of the polytheistic religions, but with monotheism came additional forces of self-deception associated with global conquest and a single, dominant spirit.

Finally, we consider the role of prayer and meditation, specific teachings against self-deception, and the contrast between the social and internal sides of religious devotion.


By logic, religion ought to increase altruistic and cooperative behavior among group members—of obvious potential benefit—but it may do so along with reduced such behavior toward nonmembers and, worse still, outright aggression and murder. That is, an increased degree of hostility toward neighboring groups can heighten the within-group bias (and vice versa). This is the double-edged sword of religion, inside and outside: a religion urges its own members to treat each neighbor as they would treat themselves, yet also to slaughter every nonbeliever and outsider, as is ordered in the good book, for group after group, down to every last man, woman, and child. At the extremes, some religions advocate in-group love and out-group genocidal hatred.

In some religions, people imagine that God is watching and evaluating their every action. Reputational concerns are expected to have obvious effects on human cooperative tendencies. One study shows that even a pair of eyelike objects on a small part of a computer screen can unconsciously increase cooperative behavior in an anonymous economic game. An awareness of observing, judging god(s) may have similar effects. Indeed, providing a “God prime” hidden in a game of sentence creation increases cooperative tendencies to about the same degree that primes of secular retribution do (police, courts, etc.). Insofar as fear of God’s judgment entrains more moral behavior on our part toward others, it can be seen either as a device that costs us some occasional selfish behavior but protects us from the greater cost of such behavior being detected by others and of being aggressed against, or as a form of imposed self-deception by others, in effect, scaring us into greater group orientation.

A tendency to detect agency in nature likely supplies the cognitive template supporting belief in supernatural agents transcending the usual limitations of nature. Since only in some religions do these gods watch, monitor, and respond to human behavior, it would be most interesting to know which religions do so and why. Is this, in part, a means of increasing in-group cooperation?

Although those Christians who frequently pray and attend religious services reliably report more altruistic behavior—such as charity donations and volunteer work—it is uncertain how much this applies only within the religious group or even whether it applies at all. This is because various measures of religiosity repeatedly have been shown to correlate with higher false opinions of self, suggesting an obvious self-deceptive effect of religion: you think better of yourself than you otherwise would. In Islam, it is mandatory to give to the poor, but there must be variability in doing so, and it would be most interesting to know what such variability correlates with.

One interesting fact on the effect of religion on cooperation emerges from comparing small religious organizations—“sects”—with small nonreligious communes. There is a striking tendency for the religious to outlast the secular (at least in the United States). In each year, the religious sect is four times as likely to survive into the next year as the secular. So religion provides some kind of social glue that makes organizations based on them more likely to endure than those based on nonreligious themes. Living in a cohesive and mutually supporting organization would be expected to have immune benefits as well, since one is less isolated and more likely, in a crisis, to be able to draw on the resources of others. As we have noted, the placebo effect is based partly on its expected association with caring acts by others.

Another interesting difference between the two kinds of communes is that the more costly the requirements imposed on group members in a commune (regarding food, tobacco, clothing, hairstyle, sex, communication with outsiders, fasts, and mutual criticism), the longer the survival of a religious commune, though there is no association between cost and survival in the nonreligious. This raises two questions: Why should cost be positively associated with commune survival, and why should this hold only for religious ones? According to cognitive dissonance theory, greater cost needs to be rationalized, leading to greater self-deception, in this case in the direction of group identity and solidarity. Why do religions provide more fertile ground for this process than secular communes? Perhaps because religions provide a much more comprehensive logic for justifying beliefs and actions. In religious communes, men’s participation in group prayer predicts their degree of sociality in an experimental economic game.


Whether religion is entirely devoted to self-deception from its very foundation to its every last branch seems unlikely, but the fact that this is even a theoretical possibility suggests the degree to which religion has been infected by forces of self-deception. Even a casual glance at most religions suggests that there is far more nonsense than revealed truth. Some of the key features of Western religions (and some Eastern ones) are the following.

A Unified, Privileged View of the Universe for Your Own Group

Most religions propose this view. Either you are the founding people and all others degenerate dogs, or else yours are the “chosen people” either by ethnicity (Jewish) or by attachment to this or that prophet (Jesus, Muhammad). Of course, any general system of thought that places you at the center is useful to you in interactions with others. In defense of religion’s inadequacies, it should be remembered that for many thousands of years, there was nothing else other than religion. Certainly no organized science, no Newton or Darwin, but still this alone can’t justify the strong egocentric biases of religion.

There May Be a Series of Interconnected Phantasmagorical Things

For example, there may be an afterlife; a giant spirit who controls all but is amenable to human persuasion on the most trivial matters; a prophet capable of performing miracles, whether parting the seas, raising the dead, or feeding the masses; a prophet who is born without a human father, only God himself, and who stays dead for only three days; and so on. Once you have signed on to a few of these notions, there are hardly any boundaries left, and very small details can turn out to be critical features of dogma.

The supreme spirit (or God) is typically given a masculine name that on biological grounds seems most dubious. Besides imparting an image of God as a fearsome tyrant, there is no such thing as an all-male species in nature. Not a single one. Only females can reproduce by themselves, females preceded males in evolution, and to this day they are still the critical sex as far as biological work is concerned. God should be interpreted as mostly female, and I will do so throughout. A male God has many unfortunate features, including the heartlessness and aggression associated with men and their divorce from reproduction, producing a series of horrors—pedophilia in all-male “celibate” castes; hostility toward women’s interests, especially efforts to control their reproduction and sexuality (banning sexual activity, abortion, in vitro fertilization, etc.); honor killings; indeed every kind of anti-female horror including mass rape during warfare in the name of God.

The Deification of a Prophet

The deification of Jesus is unlike the treatment of prophets in either Islam or Judaism. His birth by unheard-of means, miracles ascribed, and of course, his very brief death, so that now he is one-third of the show: Father, Son, and Holy Ghost. The basic story was put together after his death in the years that Christianity was a small, persecuted sect. To believe in his divinity became the key test, one that automatically shrank and exalted the group. The bigger you make Jesus, the smaller you make God. Not only are other gods no longer real but also God herself has lost a good part of her powers to a (dead) human being.

It is also ironic that the more you deify the prophet, the less attention you pay to his actual teachings, since the key distinction then becomes whether you believe in his divinity, not whether you believe in any of his teachings. “I believe, Jesus, I believe in you as the Lord, my personal savior.” Yes, but do you believe that the meek shall inherit the earth, that blessed are the peacemakers, that you should treat all others as you wish to be treated yourself, and so on? I doubt it. Deification of Jesus also makes more likely patently absurd beliefs, such as intercessory prayer, since Jesus now joins God as someone you can beg favors from (and the Catholics add yet another layer, Jesus’s mother, the Virgin Mary), no matter how many laws of nature need to be violated in the process. Among prophets, Jesus was an extreme case—hung on a cross until he gave up his life—but do not imagine that the earlier prophets in the same tradition were welcomed (whether Isaiah, Jeremiah, Ezekiel, or whoever). During their time, they were often persecuted and rebuked, only later restored in memory to prophetic status.

Sometimes a Book Is Treated as Received Wisdom Direct from God

This allows plenty of room for interpretation. Sometimes every word is literally true, even if this results in numerous contradictions within the book itself, never mind the larger world. Other times, metaphor is permitted, and indeed encouraged, giving plenty of latitude for how this divinely generated document is interpreted. The key is that you—or your group—control the document and the interpretation. If God literally created the world in seven days about six thousand years ago, then all of astronomy, geology, and evolutionary biology must be nonsense. Did God really give “the land of Israel” in perpetuity to a people who wrote a book a few thousand years ago saying he did?

Faith Supersedes Reason

Sometimes anti-logic is directly pushed, as in the notion that “by faith ye shall know them”—indeed, an attachment to reason may be evidence of sacrilege. The degree to which we believe something now becomes a determinant of its truth value. Once again, this joins a long line of features that tends to remove all rational boundaries from religious thought, permitting any and every deceptive ploy and self-deceptive concept.

We Are Right

And here comes the critical, all-encompassing self-deception: we are the measure of what is good, we represent the best, we have the true religion, and as believers we are superior to those around us. (We have been “saved”; they have not.) Our religion is one of love and concern for the world, our God a just God, so our actions can’t be evil when they are done in God’s name.

Given the ease with which religion slides toward self-deception, what are the larger forces that might propel a religion toward more or less self-deception? One important factor is the degree to which the religion is associated with the powerful in a society. Another important force has to do with religious fragmentation. Because religions almost always preach within-religion mating, fragmentation is expected to lead to intergroup conflict over minor religious distinctions. I will argue that parasite load—average pressure on a society every generation from coevolving parasites—may be an important force fragmenting religions and thus encouraging parochial self-deception. The evidence for an association with parasite load is strong, but the evidence for a connection to self-deception not nearly so strong. First, let us turn to the positive association between religion and health.


Religious behavior and practice appear to be positively correlated with health, a well-established fact with dozens of careful studies in support, on both sick people and well. Longitudinal studies suggest that variables such as degree of attendance at religious service are positively associated with survival years into the future.

Part of this effect may result from the tendency of religions to establish rules related to health: avoid tobacco and alcohol, pork, top predators such as sharks and lions (which tend to concentrate toxins as they move up the food chain), and generally risky or unwise behavior, such as gambling. One long-term study of US Christians showed that degree of religious attendance in 1965 predicted a change to more positive health behaviors thirty years later.

Under Islam, some behavior is prohibited, some encouraged, and some required. The forbidden (haram) tend to relate directly to health:

• Gambling

• Alcohol

• Eating pigs or dogs

• Eating dead meat

• Eating meat of animals not slaughtered the Islamic way (cutting throat at aorta and bleeding animal)

• Eating predatory fish

• Eating shellfish

• Usury (charging interest on money)

• Saying oiff to parents (an expression of impatience or annoyance), or yelling at them

• Suicide

All of the prohibitions regarding eating probably reduce parasite acquisition. Predatory fish are like sharks and lions in other religions—top predators that may be forbidden because they strongly concentrate toxins. Bleeding presumably reduces exposure to blood parasites. Only avoiding usury and saying oiff may not be directly related to personal health.

It is perhaps interesting to note that of the requirements in Islam (wajeb), three have positive connections to health (among other effects):

• Daily prayer (five times per day)

• Cleanliness (must be clean to pray: use only running water or sand)

• Fasting

• Alms to poor

• Pilgrimage to Mecca (if possible)

• Testifying (“there is only one God and Muhammad is his prophet”)

The latter three are clearly social: two showing off, and one helping a group member, all with unknown possible immune effects.

But the relation between religion and health goes deeper than health-related behavior. Some effects may come from the benefits of positive belief itself—for example, on immune function—as well as benefits that flow from being a member of a mutually supporting group, including musically supported activities that raise group consciousness, a very common feature of religion. As we have seen (Chapter 6), music has positive immune effects, while noise has negative ones. The exalting, positive music of so many religions is probably on the high end for positive immune effects (in contrast to, say, jazz or rap). Even confessing sins to God and disclosing trauma may have beneficial immune effects. The private confessional in the Catholic Church facilitates this, as do numerous public rituals of confession common to Amerindian religions. It seems likely that private, verbal confession in prayer has similar immune benefits, an example of a personal benefit to private religious behavior because it mimics a social interaction.

Whatever the precise causes, the links between religion and health seem strong enough on their own to select directly for religious behavior and belief. As biologists, we need not view religion phobically, as some negative, nonliving force of unknown nature that has us in its viruslike grip. We might remember that before the advent of modern science, almost all medicine was practiced within religion, often by special castes, medicine men and women, faith healers, and so on. Some medicinal benefits were certainly real, for example, consuming plants for their real chemical effects, a behavior that reaches deep into our monkey past (although the causal connection was usually unknown to the actors), and some may merely be the blessed placebo effect, itself probably the dominant benefit throughout two thousand years of Western medical “science.” Belief kills and belief cures.

One benefit of religion is that it does provide a framework for understanding and acting within our world, a framework we might expect to provide some psychological and mental benefits. Recent work in neurophysiology suggests one such benefit. Scientists concentrated on the anterior cingulated cortex (ACC), a region involved in many processes, including self-regulation and the experience of anxiety. EEG neural activity in the ACC was recorded while people were taking the Stroop test (name the color in which words are written, though the words denote a different color). The stronger people’s religious zeal (as measured by a scale) or the more they professed a belief in God, the less their ACC fired in response to errors and the fewer errors they made. It was as if religion was providing them a buffer against error. There must be many such possible effects.


Religions have repeatedly split into subreligions that are sometimes at one another’s throats. Religions occasionally join together, but this occurs much more rarely than splitting. There is thus a bias in the propagation of religions over time, with a tendency for major faiths to split into subgroups, which may split further, typically emphasizing relatively minor doctrinal differences on which to disagree: from universal truth widely shared to smaller within-breeding units at war with each other on the basis of intellectually false distinctions. If this is an important feature of splitting—corruption of the religion’s generality and logic—then we need to understand its origins.

Recent work suggests that parasites and, in particular, parasite load may drive religions to split. These splits, in turn, entrain changes in doctrine to justify them, and thus tend to degrade the universal truth value of religion with parochial arguments whose true meaning is usually hidden. Parasite load is meant as an aggregate measure of the number of parasites and their degree of damage on a local population. Ideally, parasite load would be measured as something like the degree of overall mortality (or loss in reproduction) due to disease, but it is usually measured as a simple count of the major diseases present and the relative strength of their negative effects.

The argument goes as follows: Where parasite load is low, an in-group and out-group member may be almost equivalent where risk of transmitting a new infection is concerned, namely, low. But where parasite load is high, an asymmetry emerges. An in-group member will in general have been exposed to the same set of parasites as the other members and will carry some of the same genes that give at least partial resistance to many of these parasites. But an out-group member will be subject to selection from a slightly different set of parasites and will carry a subset to which it may be partly resistant but in-group members are not. From the standpoint of each group, the other is a threat—you may transmit your parasites to one another far faster than the genes that would protect against them. Hence, individuals in both groups may be selected to avoid one another. In short, other things equal, high parasite load is expected to increase ethnocentrism, within-group love, and hostility toward strangers. By this argument, degree of self-deception across religions and cultures is expected to correlate positively with parasite load.

What is the evidence? Two broad factors are of interest: religious and linguistic diversity. That is, how many languages and religions coexist per unit area? With high parasite load, we expect many of each, since splitting into smaller groups facilitates language formation. Regarding the evidence, there can be little doubt. Across the entire globe, religious and linguistic diversity map directly on parasite load, as does ethnic diversity—the higher the parasite pressure, the more religions, languages, and ethnic groups per unit area. The exact overlap between religion and language has not been described, but these results have been corrected for numerous possible confounding variables, and the associations remain strong and unambiguous. For language, the correlations are significant for all five of the great continents.

Canada and Brazil are roughly the same size, yet Canada has 15 religions and Brazil, 159. Canada is located in the far north, where parasite load is low; Brazil is in the American tropics, high in parasite load. Likewise, Norway, in the far north, has thirteen religions, while Cote d’Ivoire is the same size but is located in the parasite-rich African tropics and has seventy-six religions. Of course, if there is a bias toward interactions based on shared language and religion, this ought usually to intensify within-group mating, with resulting ethnic differentiation (and hostility). It is certainly striking how often out-groups are characterized as if they were flea-ridden and scabrous, if not syphilitic.

Whether this argument applies to major splits in religion is unknown. Did Shia and Sunni really split in response to parasites? And Roman and Greek Catholics? The peeling off of various Protestant sects from Roman Catholicism was associated with the publication of the Bible in modern languages, as well as with a great European outward surge of warfare, plunder, and colonialism. Where is the parasite connection, if any? Did the newly fragmented groups interact less frequently? In short, the general trend seems clear, but particular major cases may have little or nothing to do with this rule, at least given our current understanding.

One subject that requires analysis is the degree to which the formation of cities and more widespread trade conspired with monotheism to create a world less fractured along parasitic lines. We know that the appearance of agriculture and the subsequent explosion in both population numbers and rate of adaptive evolution preceded the invention and spread of monotheism, but we know little about the interaction with parasite pressure. In general, higher density increases parasite pressure, resulting in such horrors as the Black Plague, which wiped out one-third of Europe in the Middle Ages, or the influenza incubated in the trenches of World War I that consumed twenty million lives before it was done. On the other hand, we are completely ignorant of the subtler dimensions of this subject. A series of other variables have been shown to covary with parasite load, so these will very likely covary with religious features as well. High-parasite-load societies appear to be more xenophobic, more in-group oriented and homogeneous, more suppressive of women, less permissive of casual sex—in short, a suite of characteristics that can at least by logic be linked to parasite defense. So far as I know, no one has studied the interaction of these variables with religion, yet surely we would expect many connections: the more numerous religions there are in parasite-rich areas, the more the religions are expected to be xenophobic, harsh on women, conformist, and so on.

In this situation, underlying correlations are expected to bubble up from the unconscious, requiring post-hoc justification. Presumably, no one is saying, “Look, worm density has increased alarmingly in ourselves in this area for the past ten years. Perhaps it would be wise for us to be more focused on in-group interactions, including mating. Let’s up our racism level.” Instead, as I imagine it, religion provides substitute logics with similar effects—let’s emphasize minor doctrinal differences: “We scratch our asses with our right hands, they with their left [note the parasite implications], so let’s avoid the nasty left-scratchers entirely.”


There is one important problem hidden in the above account: the assumption that in-group mating will be as strongly selected for as in-group favoritism. This is counter to expectation. We know that sexual reproduction—and the recombination it promotes—is strongly associated with evolutionary protection from coevolving parasites. Thus, parasite load may generate impulses toward in-group favoritism while at the same time heightening interest in sex with an out-group member.

Consider greater sexual promiscuity, or diversity of mating partners, well known to be higher in both birds and humans in the tropics, and presumed to represent an adaptive response to parasite load by increasing genetic quality of offspring. So why should this kind of sex be more prohibited in parasite-rich regions? Is it precisely because in these situations women would benefit more from such activity (improved genetic quality of their offspring) and thus provoke greater male countermoves, the kind of behavior we described so vividly in Chapter 5: mutilation, beating, terror, and murder? Certainly religions are overwhelmingly patriarchal in logic and structure, with numerous resulting effects.

One such effect is the bizarre recent claim from the Holy Roman Catholic Church that male celibacy does not contribute to priestly pedophilia, but homosexuality does. Certainly the latter should bias molestation toward male children, but what could be more conducive to sex with children than a complete prohibition on sex between adults? And what is more conducive to abusing boys than an all-male priesthood that presumably attracts men who like men? What continually haunts me when I think about such matters is the function of all this nonsense. Who benefits from an all-male priesthood? As for the priests’ being nonreproductive, at least this guards in principle against narrow kin interests. There are very few genetic dynasties in the Catholic Church (contrast North Korea, Syria, Egypt, Jordan, India, Haiti, and the United States), so the Church is likely to be corrupt but not nepotistically so.

But why all-male? This makes celibacy easier, but all-male priesthoods coexist with priestly reproduction in Islam, Judaism, and many Protestant sects, among others. And why the association with distortions against women’s interests? Is female reproduction to be subordinated to male interests for group benefit or male benefit—and at what cost to females? The Catholic Church outlaws all control by a woman over her own reproduction short of abstinence from sex at the very moment that she is most eager for it. She is not allowed to prevent conception if copulation occurs, and she is not allowed to terminate a pregnancy, however induced (rape and incest included). This appears to be a simple strategy for maximizing group reproduction, or at least male group interest. Female interests appear to count for little.


As we have seen, power corrupts: the powerful are less attentive to others, see the world less from their standpoint, and feel less empathy for them. The converse is that the powerless are more apt to see things from the other person’s standpoint, to be committed to the principle of fairness, and to identify with people like themselves. The religious effects are that humility, fairness, forgiveness, and neighborly love are more apt to be virtues preached among the powerless. It is no accident that in both Christianity and Islam, this dynamic has been played out. The Christian gospels were all written while the church was a small, underground, persecuted sect. Islam’s more peaceful injunctions came when it was an oppressed minority, its more assertive when it reemerged with military power.

It has been said that when after three centuries Constantinople elevated Christianity to the state religion, it went in one century from being the persecuted church to the persecuting church. This is a recurring theme in monotheistic religions: with state power comes a new source of bias. They change from emphasizing the universal principles of brotherhood that would especially benefit the oppressed and those needing alliances with other groups to emphasizing principles of dominance and imperialism—the lesser orders should remain so and unbelievers and outsiders may be attacked more or less at will. Racism is a valuable handmaiden. If the others are biologically inferior, is it then not God’s will that they should be supplanted by their superiors? How else is evolution supposed to work?

Islam provides a nice example of these forces, because we know the order in which the Sura of the Koran were written, its verses—that is, the actual words of the prophet—recorded while he lived. (In contrast, all of Jesus’s teachings were written long after he died.) Just like Jesus, Muhammad began as a marginal prophet of the marginalized, but unlike Jesus, he ended up as the head of a reinvading army of true believers. Muhammad began his ministry in Mecca, where he formed a small sect, often persecuted and vulnerable, so he preached an ideology of peace, respect for other groups, humility, and universal brotherhood. He then moved to Medina, where he initially faced the same situation and talked the same talk, but he then came to power in Medina and was able to head an invading army back to Mecca, which he promptly took over. During all this time, his Sura became more self-assertive and less tolerant the more powerful he became, sometimes urging attack on the infidels by the faithful. Similarly, in the Jewish tradition, it is King Josiah who is said to have both consolidated monotheism and been the first bloodthirsty advocate of it.

Consider a much more recent example from the Catholic Church. Pope Paul XXIII and Vatican II inspired in the Latin American Church a new “liberation theology” in the 1980s closer to the humble, persecuted church (prior to Constantin), the time when Jesus’s teachings were actually written down. This liberation theology explicitly favored “the preferential option for the poor” and urged their organization into self-supporting communes. This entire movement was crushed by the US military, explicitly so, and of course by the Catholic Church itself, always eager to bend theology to local power. Assassinations were the preferred means of enforcing orthodoxy, especially in El Salvador, whether of nuns traveling innocently on the road or of the Archbishop (Romero) while saying Mass, or of a courageous Jesuit priest who cried out in prophesy, “Very soon the Bible and the gospel will not be allowed in our country. We’ll get the covers and nothing more.” If Jesus were to reappear, he would be arrested as subversive, said the priest a few weeks before his own assassination. Thus is religion degraded by very regressive forces.


Religions tend to impose their own mating systems, and these in turn affect degrees of relatedness within and between religions. Religions typically ask (or require) of their adherents that they marry within the religion or subreligion: Catholic with Catholic, Protestant with Protestant, Shia with Shia, Jewish with Jewish, and so on.

The pressure to breed within the group leads to a degree of inbreeding, that is, nonrandom mating with those to whom one is (at least marginally) more closely related. We are not here talking about close inbreeding—parent /offspring, brother/sister—but typically more distant, first cousin to second cousin and beyond. But repeated generation after generation, inbreeding inflates degrees of relatedness between group members (that is, genetic similarity because of common ancestry). At the same time, it creates a chasm in relatedness to other groups—one is less related than one otherwise would be.

Two important kinds of migration are important here. People may outbreed—that is, marry outside their group—and people may convert, or join another group.

When a man (for example) outbreeds and his children are raised outside his original group, his out-migration is experienced as a “selective death” to his original group. Whatever genetic traits he has are lost to that group, including his outbreeding tendencies. To put a fine point on it, if he is on average less ethnocentric, less self-loving, and less narrow in outlook than members of his original group, his out-migration lowers the frequency of these traits in that group as surely as if he had died young.

On the other hand, his arrival in the new group has the opposite effect. It is experienced as a selective birth, as if someone had been born (at full reproductive age) with the same traits we just described. Returning to the composition of the original group, the key question is, how much in-migration occurs and under what conditions? If a man marrying a woman has his children accepted as being of her faith, then the same kind of traits lost through out-migration will tend to return through in-migration. But are the two processes equally strong? If there are more men leaving than arriving, then this group will become more inbred. I have not had the chance to pursue this subject in greater detail, but if I were interested in the genetics of religious diversity, I would pay attention to biases in between-group transfers, by sex and by magnitude.

As for genetics, inbreeding has well-known effects. Products of inbreeding show less internal variability than do products of outbreeding. This genetic similarity can have two detrimental effects. On the one hand, relatively rare negative traits that require two copies of the same gene for expression (for example, sickle-cell anemia, Tay-Sachs disease) become more common. On the other, greater genetic variability has well-known benefits in defending against rapidly coevolving diseases, so that outbreeding becomes a genetic defense.

The second form of in-migration is simple conversion (initially unconnected to marriage), and religions differ in their rules regarding this. Thus, Christianity has usually been a proselytizing religion, continually seeking converts, wherever and however, as has Islam. Sunni and Shia Muslims may stretch from Senegal to Sudan to Lebanon to Pakistan to India to Indonesia, with similar opportunities for interbreeding all along the continuum within each group, but limited exchange between the two. With some notable exceptions, Judaism has not been a proselytizing religion, although Jews have been subject to forced conversion (for example, in Spain in the sixteenth century).


Many religions have teachings that are either explicitly or implicitly against self-deception. It is often argued that self-deception interferes with one’s ability to know not only oneself and others but also God herself. For one thing, there is a presumptive case for the utility and validity of general principles. What is true here should be true there. What applies to you should apply to me. The very universality argues against the usual biases of deceit and self-deception. If you are told to treat others as you wish to be treated, then you have a rule, which, if actually followed, would counter much of your unconscious self-deceptive tendencies in favor of self over others. Similar general rules could reduce self-deception further. Of course, as we have seen, the generality of these “general” principles is easily undercut by forces of fragmentation, in-group and out-group formation, and the rule of the powerful.

Religions also preach explicitly against self-deception. Consider Jesus’s famous teachings about not judging others (Matthew 7:1–5):

Judge not that ye be not judged. For with what judgment you judge, you shall be judged: and with what measure you mete, it shall be measured to you again. And why do you behold the mote that is in your brother’s eye, but consider not the beam that is in your own eye? Or how will thou say to your brother, let me pull out the mote out of your eye; and behold, a beam is in your own eye? You hypocrite, first cast out the beam out of your own eye; and then you shall see clearly to cast out the mote out of your brother’s eye.

I translate this directly into the language of self-deception. Beware of self-righteousness, because it easily invites self-deception. You may be projecting onto others your own faults. And beware lest you come to be judged by the same criteria you are enforcing on them. Why do you see the minor fault in your neighbor but fail to see the major one in yourself? Instead of denying your own fault and projecting it onto others, admit your fault, the better to see whether any fault lies elsewhere. Otherwise, you are a hypocrite, criticizing the wrong person, in the wrong order.

Another argument against the speed—and injustice—with which we judge others comes from the case where Jesus is presented with a woman about to be stoned to death for committing adultery. His reaction? “Let he who is without sin cast the first stone.” And it is said that everyone left the room in reverse order of age, the oldest—who had the most sins—the first. In both of these cases, it is internal contradictions that drive the argument, precisely the reason that universally valid principles tend naturally to argue against self-deception.

Other teachings are less explicitly opposed to self-deception but have similar implications just the same. Here is one that is opposed to the in-group/out-group bias. In the parable of the Good Samaritan (really the good Arab or Palestinian), Jew after Jew passes by the badly injured fellow Jew. It is an outsider, an Arab, a Samaritan, who responds to the sufferer’s needs by binding his wounds, giving him water and food and finding him safe lodging. Who is the admirable person here, the heartless in-group member or the otherwise hated out-group one? Or what about Nicodemus, the man who came by night? It is precisely his willingness to meet Jesus at night, out of sight of others, that made him a hypocrite, one who eventually voted to condemn Jesus but then made sure to help bind the body for burial.

Another example is the structure of the Lord’s Prayer, which has interesting features where self-deception is concerned. First, it is short. Then it is divided into only three parts, the first an assertion of humility: “hallowed be thy name” and “thy will be done.” When landing at an airport, I often pray that “thy will be done” and add the hope that this does not include flipping the plane upside down on arrival but, if so, thy will be done. In other words, let us accept a larger plan than our own and not seek to change the plan through personal begging. Let us humble our own self-interest to the larger plan. In any case, if the plane is going to flip, the plane is going to flip; the only thing we can pray for is to be calm upon arrival.

The second part of the prayer has an interesting feature—you are allowed to beg for only two things on your behalf, and one of these is contingent. You can ask for your daily handout, what every creature needs: its daily bread. And then you may ask that your own sins be forgiven but only insofar as you forgive those of others. This is critical: no blanket amnesty. You must give to get; you must forgive to be forgiven. This binds you to a psychological commitment—one that ought to reduce self-deception on the spot.

Then comes the final part, where you ask not to be led into temptation—really an injunction against allowing yourself to be tempted—and to be protected from all evil (self-induced included). No intercessory prayer here. No “and may the president continue to make wise decisions and may God bless America,” so commonly heard in US churches (or, more absurdly, in President George W. Bush’s words, “may God continue to bless America,” as if an obligation had developed). Indeed, the ability on Sunday of so many Christian preachers to forget the only teachings by Jesus on prayer is astonishing, were it not for the power of deceit and self-deception.

Sometimes the teaching against self-deception is only a metaphor. In the twenty-seventh psalm (vs 8), David says, “When Thou said, seek ye my face, my heart said unto Thee, Thy face, Lord, will I seek.” It is hard to imagine looking God straight in the face and lying—to God or to oneself.

For a parallel in Islam, there is an important distinction in Sufi thinking between the jihad (struggle) against the outside world, called the small jihad, and the jihad against oneself, called the greater jihad. The small jihad is relatively simple: one struggles in group activities against an out-group in order to convert them. In the extreme case, either they are killed or you are, at which point you ascend to heaven. No great problem. But the struggle against oneself is far more difficult, and to reach God’s light, one must succeed in controlling one’s own body. This is a personal struggle that requires controlling your bodily desires (for money, pleasure, satisfaction) in order to purify your soul. These desires occlude self-knowledge, in our system of logic, by encouraging self-deception. In the Sufi system, you must enslave your desires or they will enslave you. And finally, controlling the self is also a useful tool for controlling the outside world. The Greek sage Thales once put the general matter succinctly. “Oh master,” he was asked, “what is the most difficult thing to do?” “To know thyself,” he replied. “And the easiest?” “To give advice to others.” Various Eastern religions also sometimes urge rather extreme systems of physical self-denial to free the individual from its egocentric center.


A bizarre belief widespread in many Christian circles is that of the power of intercessory prayer. That is, many people seem to believe that a group of people in a room, scrunching up their foreheads in intense concentration on behalf of someone miles away about to undergo surgery, can have a positive effect on the outcome. Were this to be true, the laws of physics would have to be violated on a daily, even minutely basis, by a deity who chooses to alter reality in response to the pleas of petitioners according to some unknown criterion—a most unlikely structure to the real world. The matter has been put to a test a number of times but often with poorly controlled studies and small sample sizes, precisely the conditions expected to produce a conflicting array of positive and negative findings, feeding the illusion that something may actually be going on.

Then came a multimillion-dollar study, carefully organized with six hospitals in which groups prayed for given patients from the day before they entered surgery until two weeks later, while another group of patients received no such prayer. Meanwhile, some of those being prayed for were told that they were being prayed for and others were not. Patients were followed for a month after surgery. The results were unambiguous: no effect whatsoever of intercessory prayer on the outcome, no hint of a benefit. So our first question is answered: it has no direct effect.

But does it have a placebo effect? Does belief in the efficacy of intercessory prayer by the victim give any kind of efficacious benefit? Quite the contrary. Those told they were prayed for had more postoperative complications of every sort than did those who did not know they were being prayed for. One hypothesis is that when told people are praying for you, you interpret your situation as being more dire than it really is, with associated stress. The patients are not being offered anything more than a useless prayer: no talk of cleaning the apartment or keeping their dog alive, no investment in their future, nothing—just the claim of people in intense wishful thinking on their behalf.

Note that the truly devout have no problem with these new scientific results—God responds to these experiments by simply withholding the usual benefits of intercessory prayer the better to keep scientists (and unbelievers more generally) in the dark. Did not Jesus say, “I will reveal unto babes what I will keep hidden from the wise”?


There has been an exponential increase in suicide attacks worldwide, at least as measured over the past twenty years. This is a device by which a member of one group sacrifices his or her life to inflict damage (death and otherwise) to numerous or highly important members of another group. There is no question that this behavior could in principle be an effective political (and reproductive) strategy with return benefits to the martyr’s much larger kin group, but there is also no doubt that such behavior easily induces massive return spite. In any case, suicide bombing can serve as a sensitive measure of the degree of willingness to commit violence against an out-group at great personal cost.

It is of some interest to know the role of religion in all this—pro, con, or otherwise. Recent work has provided a most interesting answer. When measured one way, religious activity makes the participation in (and support of) suicide bombings more likely. When measured another way, religion has no effect. What is the difference? Religion has an external, social aspect and an internal, contemplative one. Across a variety of suicidal conditions (Palestinian surveys, a hostile prime for Israeli settlers), religious attendance (the social aspect) is positively correlated with support for suicide bombings, but prayer (the contemplative) is not. This holds for study after study. In a summary of six religions in as many countries, regular attendance at religious services predicted both out-group hostility and in some cases willingness to be a suicide martyr— but prayer did not. The Sufi outer jihad is run by social interactions, the greater inner jihad by achieving independence through prayer. This is, I think, the double face of religion—outward, hostile, and egocentric; internal, contemplative, and anti-egoistic.


Religions tend to contribute to war in several ways. They encourage an in-group mentality, backed up by a breeding system that increases within-group relatedness (while decreasing between-group relatedness) and they readily provide the shared self-deceptions on which to base group action. But there is one final gift of many religions: self-righteousness. Murder is not only not prohibited (as it is within the group), but it is also sometimes required. It is your moral duty to kill the infidel, the unbeliever, the other. You are doing the Lord’s work—not just your own or that of your group. You are fulfilling more than your manifest destiny—you are the Lord’s executioner. You are helping natural selection along its ordained path. The Bible, as it turns out, warns against this path: “Vengeance is mine,” saith the Lord.