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

Chapter 4. Self-Deception in the Family—and the Split Self


We usually begin our lives—the first twenty years, at least—embedded in a family, typically one or both parents and one or more siblings. This is often part of a larger extended family including grandparents, uncles, cousins, and so on. The key to the biology of all this is genetic relatedness (r). That is, family members are all related to one another in the sense that there is a chance that any given gene in any one individual has an identical copy in another by direct descent from a common ancestor. A typical gene in a parent is found in its offspring half the time (hence r to offspring = ½), while a typical gene in the offspring is also found in either parent half of the time. Siblings are related by ½ but half-siblings by only ¼ and so on. This leads to “Hamilton’s rule,” which states that the benefit of an altruistic act toward a relative times the relevant degree of relatedness must be greater than the cost suffered by the altruist in order for selection to favor the altruism. For example, if you are helping your half-sister, then (other things being equal) the benefit to her had better be greater than four times the cost to you. Likewise, selection will oppose a selfish act that harms her four times more than it benefits you. In sum, degrees of relatedness in families are high—which tends to induce investment and restrict conflict—but degrees of relatedness are far from unitary (r = 1) so that conflict is also expected between the actors. For our purposes, the key is that relatedness adds an extra dimension and logic to the kinds of deception and self-deception that will evolve.

Parents can pretend to base their actions on shared relatedness to the child (parental investment) when, in fact, it is based on the unrelated part (parental exploitation). They may be unconscious of this bias. In turn, offspring may pretend greater need in order to induce more parental investment than is optimal for the parent, and they may be more effective when they believe it themselves. And so on. Relatedness in fact leads to a series of ramifying complexities where deceit and self-deception are concerned. These have to do with misrepresentation, manipulation, and internal bifurcation. Let us look at each in turn.

Since an individual is selected to act both altruistically and selfishly toward family members, there are chances for misrepresentations regarding motivation and orientation that are deeper than those occurring toward more distantly related people. For example, there is no presumption that a person with a low r to you is programmed to act in your self-interest, but that is precisely the presumption with related individuals—and the more so the more closely they are related to you. So your relatives can pretend an interest in you that is plausible on its face, even if their real motivation is completely manipulative. A relative can also lay a claim to you. Aren’t I related to you by one-fourth, so if you are messing up in life, aren’t you messing with my quarter interest in it? Get yourself together for both our sakes.

Or consider the following. Although one is selected to invest parental care in one’s offspring, one is not selected to give as much as requested, or always to give anything at all. Hence, deeper—and, probably often, more painful—misrepresentations are possible between close relatives. Are you investing in a child or exploiting it? Do you love the child or not? Do you have in mind a separate self-interest in the child that you are willing to support or is the child entirely conceived as instrumental to your larger projects? It makes a whale of a difference to the offspring which of these is true, and there is plenty of scope for deceit and self-deception on the part of the parent, as well as of the offspring.

Second, with the added factor of a long period of parental investment soaked with language, there are many opportunities for conscious and unconscious manipulation, including induced self-deception, in which the parent can induce a pattern of self-deception in the offspring that serves the parent’s interests but not those of the offspring expressing the self-deception. The child may grow to believe that its parents are acting in its true interests when in fact they are not. The offspring may not be in a position to free itself of such an imposed self-deception until it no longer requires parental investment, giving an added reason for emotional turbulence in late adolescence, along with open hostility toward parents. Adults, in turn, may differ in the degree to which they suffer costly effects from earlier parental manipulation. In addition, parents are not a unit; they are a father and a mother, with different interests in offspring manipulation because the manipulation affects them and their differing sets of relatives.

Third and unexpectedly, relatedness considerations automatically split the organism into multiple selves, with differing interests, the most important for our selves being our maternal self and our paternal one. Formerly, we used to believe that an organism had a single self-interest. It had a unitary aim—to maximize its genetic reproduction. Kinship theory says this can’t be true. Different genes within us have differing rules of inheritance, and this will give them contradictory interests. For example, the Y chromosome is always passed father to son. It is not selected to have any interest in daughters. Does that mean we expect fathers to be at least slightly biased toward their sons? Not at all. The male’s X chromosome is passed only to his daughters and it is more than ten times as gene-rich as his Y, so if anything, men should show a slight genetic bias toward their daughters. No one knows whether this is true, but there is some evidence that paternal grandmothers favor their granddaughters over their grandsons according to the differing chances that their X chromosome will be found in them (½ versus 0).

The Y and the X are only small parts of the whole genome. The main genetic split within us is between our maternal and paternal halves, which are equally strong. There are a few hundred genes in us that are active only if inherited from our mother, so-called maternally active genes, and about an equal number from the father, so-called paternally active ones. Maternally active genes are selected to promote maternal interests and paternally active, paternal. This generates internal genetic conflict in which two separate genetic selves compete for control of our behavior and larger phenotype. This conflict has two important effects. We expect deception between these two halves—not directed toward outsiders but toward each other. For example, maternal genes in you may overemphasize the benefits to the organism as a whole of acting on its special relatedness to others (when these are maternally biased), while paternal genes may be selected to discount such maternal effects. Second, we also expect differences between our two halves over whom to deceive in the outside world (with or without self-deception). As we shall see below, this split in us runs deep, both from early-acting genes affecting growth and consumption of parental resources to later-acting ones affecting adult behavior.


Because parents typically are related to each offspring by ½ but not by 1—and vice versa—there is ample scope for conflict between the two parties. This conflict usually concerns how much parental investment the offspring receives and what its behavioral tendencies are, as these affect its relatives. The parent is selected to maximize the number of surviving offspring it produces, but the child is twice as related to itself as to its full siblings, so it is selected to try to gain more than its fair share of resources—though not so much more that it inflicts twice the cost on its siblings as the gain it enjoys itself. Deception is an important part of the child’s repertoire, pretending greater need than is actually being experienced and manipulating the parent psychologically, sometimes against the parent’s better instincts. The parent may be selected to minimize the appearance of available resources, the better to save some for other offspring. One critical choice the parent has is whether to impose its will, insofar as it can, or opt for a fair split with the offspring. The latter, in principle, should reduce future conflict with the offspring, especially if in response it adopts a similar posture. One danger of complete domination is the turmoil that may erupt when the offspring is as physically large as the parent and is cognizant of the parental style to which it has long been exposed.

Regarding the offspring’s general behavior, it is selected to act altruistically toward a relative only when the benefit times degree of relatedness is greater than the cost to itself (B>2C for full siblings), but the parent would prefer to see altruism whenever there is a net benefit to the parent’s offspring—in this example, B>C. Thus, parents are selected to mold their offspring into being better people (more altruistic, less selfish) than they are inclined to act on their own. This may take the form of punishing behavior as being generally immoral (instead of merely counter to the parent’s self-interest).


The long period of parental investment in humans means that there are many opportunities for each party to respond to the other’s actions. One important consequence is that a child who is receiving insufficient investment or actual abuse maybe put in an awkward position where resistance is concerned. In the extreme case, resistance is likely to only make matters worse; it will provoke additional abuse and withdrawal of investment. Thus, until children reach the teen years, they may, in general, have to submit—and the more so the harsher the regime under which they live. There also are more things they need to hide from the outside world, so lack of resistance includes lack of disclosure to others, and here the evidence is clear. For abuse in general (physical, emotional, and sexual), the more the abuse is perpetrated by a close relative (or stepparent), compared to that from a more distant figure, the longer the children take to disclose the abuse, if at all. We are talking about delays of a year or more. Intervention is less likely and caregivers less supportive. And there are negative immune effects that endure into adulthood (see Chapter 6).

Here the child may be favored by natural selection to keep up a good front, which may involve self-deception, such as disassociation and selective recall. In disassociation, the mind is split into two (or more) relatively separate parts, one of which fails to recall the abuse or see it as such—perhaps the self that is usually shown to the parent. Disassociation is more common than selective recall in those who have been abused, and this disassociation compromises intellectual performance, for example, on the Stroop test (recognizing words denoting color that are written in different colors).

The notion that children completely repress memory of extreme trauma, only to recall it years later in full detail, has been shown to be unlikely in most cases, but this does not mean that amnesic factors are not at work in trauma, of which disassociation is only one example. Again, it is the closeness of the abuser that is associated with the greatest memory defects. For all forms of impairment, abuse by a caregiver induces more memory impairment than similar abuse by non-caregivers. Is this because it is inherently more offensive and in need of memory eradication or because pressure from caregivers to keep one’s silence is especially strong? It may be both. We know that the tendency to share with others is less frequent when the abuse comes from a caregiver.


As mentioned already, one of the most striking discoveries in the past thirty years of genetics is that we are expected not to be unitary creatures with a single self-interest, but to have a paternal genetic interest and a maternal one, which may differ, with each acting to promote a view of the world from its standpoint. Biologists used to think that genes had no memory of where they came from, thus they computed the average degrees of relatedness cited earlier—half chance through Mom, half chance through Dad. In the 1980s, biologists began to discover a minority of genes whose expression level depended on which parent contributed it. Often one copy was active and one inactive. So there are paternally active genes and maternally active ones. With activity limited by parental origin, these genes can act not on average relatedness, but on exact relatedness to each parent (0 or 1) and their relatives.

The first two imprinted genes described in mice tell the whole story. Igf2 (insulin-like growth factor 2) is a paternally active gene that activates growth in fetal life by increasing rates of cell division. A single active copy increases size at birth by 40 percent compared to no active copies. Why does this make sense? In competition over access to maternal investment, paternal genes in offspring are inevitably less related to siblings than are maternal genes. Multiple mating by a female for each litter or changing fathers between litters lowers paternal relatedness among the resulting siblings while leaving maternal relatedness unchanged. Thus, paternal genes will weigh effects on self relatively more heavily than effects on siblings (compared to unimprinted genes or maternally active ones), preferring faster fetal growth rates and relatively larger size at birth.

The proof is in the pudding. An oppositely imprinted gene has exactly opposite effects. Igf2r (insulin-like growth factor 2 receptor) is maternally active and, in mammals, its protein has evolved a secondary binding site to Igf2, which is carried to lysosomes and degraded. Indeed, Igf2r gets rid of 70 percent of all of the Igf2 that is produced. As a result, it lowers the fetal growth rate by about 30 percent. This is no way to build a railroad. Here are two large, costly opposing effects that virtually wipe each other out. This is not good for the individual but is exactly what you would expect if there were two opposing forces within the offspring. Evidence confirms that imprinted genes that affect early development almost always obey “Haig’s rule”—paternally active genes have positive effects on growth during maternal investment, while maternally active ones have negative effects.

One final line of evidence is worth mentioning. Although mice that are artificially manipulated to have a doubly paternal or doubly maternal genome fail to develop, individuals will develop successfully if only a fraction of their cells are doubly paternal (their nuclei from two sperm cells) or doubly maternal (nuclei from two eggs) and the rest are normal cells. Such chimeras reveal a striking fact. The more doubly maternal cells, the smaller the newborn; the more doubly paternal cells, the larger the newborn, exactly as expected. But there is a surprise: the relative size of organs inside the mouse is also changed. For example, the greater the number of doubly paternal cells, the smaller the neocortex and, hence, the brain. The hypothalamus is affected in the opposite way: doubly paternal cells do well in the hypothalamus, while doubly maternal ones disappear. Let us see why.


Just as conflict between individuals sets the context for deception between them (including self-deception), so conflict within the individual sets the stage for deception between its competing parts—something we might call “selves-deception,” which may involve different parts of the brain. The neocortex is largely the social brain, differentially involved in interactions with close relatives and other social relationships; the hypothalamus is involved in hunger and growth, much more egocentric motives. One can well imagine an argument between the two, with the (maternal) neocortex saying, “Family is important; I believe in family; I will invest in family,” while the (paternal) hypothalamus replies, “I’m hungry.” That is, each argues for its favored position as if arguing for the good of the entire organism (“I”).

And there can be no doubt that the requisite genetic variability is available. It is a striking discovery regarding imprinted genes (in mice, at least) that more than half of them affect neural development and later adult behavior. Work is still in its infancy, but here is one striking example. In mice, paternally active genes in females are especially important in directing maternal behavior. A few paternally active genes in adult females mediate such important maternal activities as retrieving the pups, licking them, and huddling over them to transfer heat. Sound like a paradox? Not really. Absent inbreeding, the two kinds of genes in a female—maternal and paternal—have the same chance of showing up in her progeny, so no bias is expected on this basis. But females also invest in their sisters’ progeny and other relatives, and they are more closely related to these on their maternal side, so such genes are more likely to compromise on personal reproduction, saving some investment for others, while paternal genes will emphasize investment in their offspring.

Or consider a young woman contemplating whether to enjoy a sexual adventure with her cousin (let us say, the son of her father’s sister). Her paternal genes will at once see an increase in relatedness to any resulting progeny, from one-half to five-eighths (the upside to inbreeding), while maternal genes will see no increase in relatedness at all—but both sets of genes will suffer the resulting decrease in quality of the offspring due to increased genetic homogeneity (the downside to inbreeding). In short, paternal genes in her are more likely to seek out the sexual relationship and maternal ones to resist it. The first declares that “kissing cousins are cute”; the other speaks moralistically about the dangers of defective young via inbreeding. To the individual, this may be experienced as internal argumentation, without any necessary resolution and with each side tempted to overstate its case.

Imagine also a possible society-wide effect. Imagine a patri-local society, in which a woman moves into her husband’s village at marriage, rarely if ever to return to her village of origin. This is common in rural India and many other parts of the world. All of her children will grow up in a world in which they are more related to most surrounding individuals on their paternal side and not their maternal one (mother and full siblings excepted). Thus, growing up in such societies, youngsters are expected to experience internal conflict between their two genetic selves over behavior affecting others. Altruistic behavior that will increase inclusive fitness of paternal genes will not necessarily do so for maternal ones, and so forth. Sons are destined to remain in this patri-local world while daughters will, like their mothers, migrate to other villages, so sons should be especially conflicted. The mother, in turn, will support the maternal genes in her sons, urging sons especially to be less kin-group oriented than the rest of his genes (and his father) might wish.


Parents are selected to manipulate their offspring to serve parental interests, and offspring are selected to resist such manipulation. A key variable is the offspring’s degree of altruistic and selfish tendencies, insofar as these affect other relatives. Parents will tend to encourage an equality ethic among their offspring, because the parents are equally related to all, but each offspring is more related to self than to siblings, so that a more personally biased ethic would seem more appropriate.

Of course, each parent is expected to represent its own interests and not those of both parents, so there will be maternal manipulation and paternal manipulation, and in turn possible conflict between the two representations in the offspring. What is more to the point is that maternally active genes in the offspring are expected to be receptive to maternal manipulation and vice versa for paternal genes. Thus, parental manipulation should coevolve with imprinted genes in progeny, each reinforcing the other. This strengthens the case for a “maternal voice” and a “paternal voice,” each based on effects from the same-sex parent as reinforced by imprinted genes.

I must say this interaction first occurred to me when I was trying to poison the minds of my three daughters against their mother’s people. Not against their mother, God forbid—I was not crazy—only against her relatives. As their faces lit up with, so far as I could tell, full agreement, I felt good, another case of successful parental manipulation in the guise of teaching. Then as they walked away, it hit me: I had been looking only at the paternal genes in them, vibrating in unison with my paternally biased argument. As soon as they were on their own, they would take a more balanced view of the matter, and what was worse, as soon as they were with their mother, the whole matter would be reversed.

Incidentally, as people age, their important categories of relatives change from those with important genetic asymmetries (parents, half-siblings, and cousins) to those without asymmetries (children and grandchildren)—in short, from relatives over whom genomic conflict is expected to occur to those in whom it is not. So perhaps we become less internally conflicted as we age because our relatedness structure to the outside world becomes more symmetrical.


The above line of thinking leads to a very important question: What is the genetic effect of marital strife on the psyche of the child who is both witness and actor in the drama? By logic, one would expect the child’s paternal genome to accept or acquiesce in the paternal viewpoint, while the maternal genome would be biased to embrace the maternal position. With increasing strife, one can easily imagine that the two genetic sides in the child—maternal and paternal—are hyped by the escalating conflict toward excessive production of their products (proteins, small-interfering RNAs, or anti-sense RNAs, all capable of regulating other genes). Thus with greater marital strife, the intensity of the child’s internal conflict may increase at the genetic level and the biochemical, as well as at the psychological. If so, this must be an important factor in intensifying the child’s internal suffering.

A striking feature of children, noted in anecdotes, is how often they respond to the news of an impending half-sibling—let us say, Dad’s child by his new wife—with intense hostility. Rather than gladdening their hearts at the arrival of a half-sibling, children seem instead to see the less related sibling as a threat to investment in themselves (and in their full siblings). Again, one would expect maternal genes to take the lead in such reactions, because they have no interest in Dad’s new progeny. Thus, the genetic conflict induced by this new situation may be more intense than the direct psychological one.


The relevance of genomic imprinting to deceit and self-deception is several-fold, of which the most important is the internal fragmentation and conflict it generates. In important parts of our family lives, we are two separable people (not one) with partly divergent aims, theories of reality, and degrees of deceit and self-deception—two people who are also tempted to deceive each other. We call these two people our maternal and our paternal selves.

What difference is expected between the two sides in degree of consciousness? This depends, of course, on which personality we are most inclined to hide from others. Let us say the maternal side is more selfish in its orientation (fewer relatives to interact with). It will wish the more to hide itself from the outside world, as well as its other genetic half (the paternal). Thus, the conscious mind will show paternally oriented behavior and be unaware of its maternal biases, while the maternal side will have full opportunity to study (and exploit) the paternal, much as happens in dual personalities, where the unconscious one knows the conscious one but not vice versa. These are merely the first speculations. Inevitably, the subject of the two halves of our minds—their interactions and their differing effects on deceit and self-deception—will grow to become a major subarea regarding family.

Here is an interesting possibility. Can one half of you feel guiltier than the other? Yes. Can half of you feel ashamed and the other not? I believe so. If guilt concerns harm to other, then by logic, harm to a relative is worse than harm to a stranger, so your paternal side could feel guilty for a hurt to a paternal relative, while your maternal side scarcely notices. If shame deals with damage to the self, especially in public, then when the public includes relatives related, say, through Mom, you may feel strong shame through your maternal genes and much less or none at all through your paternal. Guilt and shame are feelings that are both produced by us and induced in us. Someone may try to make us feel guilty when there is no good reason to do so, and someone may also attempt to shame us. Their own relatedness asymmetries may affect their tendency to induce these feelings in us. The induction may split each of us in two, which can produce both internal conflict and confusion.


At what ages do humans become capable of deception? We talk of the innocence of children, but dissembling and lying show up at very early ages—both in everyday observations and in scientific studies. Children show a wide array of deception by ages two and three, and the earliest clear signs appear at about six months. Fake crying and pretend laughing are among the earliest. Fake crying can be discerned because infants often stop to see whether anyone is listening before resuming. This shows that they are capable of moderating the deception according to the victim’s behavior. By eight months, infants are capable of concealing forbidden activities and distracting parental attention. By age two, a child can bluff a threat of punishment, for example, by saying, “I don’t care,” about a proposed punishment when he or she clearly cares. In one study, two-thirds of children age two and a half practiced deception at least once in a two-hour period. Motives for children’s lies seem broadly similar to those of adults. Lies to protect the feelings of others—so-called white lies—appear only by age five.

Temper tantrums, violent instances of rage with the child threatening even self-harm, are well known in humans, but also in chimpanzees and even pelicans. Pelican chicks will work themselves into a frenzy, swirling around violently and, in the process, chasing away their siblings, before falling prostrate at their parent’s feet, in effect demanding immediate investment, which indeed they often receive. Instead of banging their head on the ground, as a child or a young chimp might do, the pelican attacks its most critical part and bites its own wing.

Offspring deception can be extremely subtle, as the following two anecdotes suggest. A woman with a close, loving relationship to her happy five-month-old daughter picks her daughter up at day care. The girl is playing happily with a staff member, but when she spots her mother there is a flash of joy, followed at once by collapse and tears. The mother’s interpretation? The daughter is genuinely happy to see her but then immediately hides the happiness to express her suffering at not being cared for continuously by her mother, in other words, to induce guilt in Mom. In another anecdote, the same girl, now more than two years of age, uses “need” when she wants something (“I need . . . ”), as if to stress how critical the matter is, but when she does not want something, she no longer speaks of need but says more gently that she does not “want” it, both asserting that she too has wants and now speaking more slowly, almost plaintively, “But, Mom, I don’t want that.” She is manipulating her mother toward greater investment in the first case, and in the second, trying to get her mother to sympathize with her as someone with her own wants.

In fact, deception in children starts even before birth. In the last trimester of pregnancy, there is a striking change in the control of the mother’s major blood variables—pulse rate, blood-sugar level, and distribution of her blood. Normally these are under the control of maternal hormones, produced at very low levels. In the third trimester, control shifts to the offspring, who either produces the same chemicals or their very close mimics, but does so at one hundred to one thousand times higher concentrations. Why this shift in control to the offspring and to a grossly inefficient signaling system, at that?

Control has shifted to the fetus, to its own advantage. It acts to increase maternal blood-sugar levels and pulse rate above what the mother favors, because this will increase nutrient transfer to itself via the placenta. For the same reason, it also acts to deprive the mother’s legs and arms of blood and to concentrate the blood near itself. If one assumes a coevolutionary struggle in which increases of fetal hormones are matched by increasing maternal insensitivity to them, one can easily see how hormone levels could grow over evolutionary time to many times greater than when mother alone controls her own blood. As an expert in this field put it, when there is no disagreement, a whisper will do; shouting suggests conflict.

As children mature, they become increasingly intelligent and increasingly deceptive. This is not an accident. The very maturing capacity that gives them greater general intelligence also gives them greater ability to suppress behavior and create novel behavior. There is also clear evidence that natural variation in intelligence, corrected for age, is positively correlated with deception. A child is left in a room and told not to look in a box. By the time the experimenter returns, most children have peeked. Now they are asked whether they peeked. Most say no, and the brighter the children are on simple cognitive tests, the more likely they are to lie. Even health of the child at birth (as measured by a weighted sum of multiple factors) is positively correlated with lying. Because we experience deception aimed toward ourselves as negative does not imply that as deceivers we experience it as negative, at least when undetected.

Although the critical evidence is lacking for adult humans, smarter ones, as we saw in monkeys and apes (Chapter 2), are expected to practice more deception, not less, and more skillfully. By theory, they are also expected to be more self-deceived than the less gifted. This creates special dangers—high intellectual ability combined with high self-deception—for example, a malevolent person who is good at being malevolent. It is easy for the intellectually gifted to argue otherwise, that their special talents will save them from the failings lesser mortals are prone to, but by evidence and logic, we expect the opposite. Until shown otherwise, we should assume that the intellectually gifted are often especially prone to deceit and self-deception, including in many of the academic disciplines they produce (see Chapters 10 and 13). Those who take pride in their alleged intellectual gifts or of their particular group might well contemplate whether they are also more regular liars and self-deceivers. They are expected to be better at it.

When children are told to tell white lies (for example, that they like a gift when they do not), they direct all their smiling toward the intended victim (the gift giver). When receiving a gift they actually like, they share their smiles more broadly. Like adults, children tend to suppress true facial expressions more often than they invent novel ones, and they are better at it—when inventing faces, people of all ages tend to exaggerate, while suppression is achieved more exactly.

It is interesting that more dominant five-year-old children of both sexes are better at fooling observers in laboratory experiments, but in the same experiments, dominance confers no advantage in detecting deception by others. Among adults, the same is true for men, but women’s deceptive behavior is unaffected by dominance (as is their ability to spot it). As we saw in the first chapter, when people are given a “power prime,” they see the emotional expressions of others less accurately, so if anything, we expect them to be more vulnerable to deception.

It is noteworthy that parents play “pretend” with their children at very early ages, that children play pretend with one another and with themselves, and that most of children’s literature is fantasy. Consider how common (and popular) games are that involve deception—hide-and-seek, card tricks, magic, liar’s dice, and so on. Hence, there seems to exist some drive to incorporate pretense into life at very early stages. It certainly stimulates imagination and learning and also prepares the child for living in a world where practicing and spotting deception are important. I have certainly never seen any signs of natural guilt in children when practicing deception. Quite the contrary, regarding their parents at least, children seem to regard deception as their first line of defense, as well they might. Their parents are bigger, stronger, more experienced, and in control of most of the resources at issue.


Even though parents may encourage white lies in their children, they often seek to penalize, suppress, and (sometimes) harshly punish deceptive behavior (especially directed toward them). Parents have the power; they need only the facts. A common parental device is to stare into the child’s face at close range and force the child to look into the parent’s eyes. I have seen parents use this successfully with their twenty-year-old children. College students consistently tell me that they think their parents read their deception better than anyone else and sometimes with near-perfect accuracy. The threat of punishment in general tends to induce deception in children to avoid it, and this is true also for punishment in response to deception itself. Punishment (especially harsh varieties) may drive the deception deeper, perhaps inducing greater self-deception to hide rising fear and pain (with unknown downstream immune effects).

Parents may also have a huge effect when they themselves indulge in deceptive behavior that the offspring are then tempted to mimic. Children may learn that it is fine to deceive; it may even be a legitimate lifestyle. This may range from lying to friends to hide misdeeds (“Oh, I’m sorry I didn’t pick you up; I had a medical emergency with one of my children”) to more serious misrepresentations. If a parent is a drug addict and tells a lot of “stories” trying to cover up the addiction, the children may tell lies to cover up the parent’s addiction and may then grow to lie to people in general. On the other hand, children are notoriously sensitive to parental contradictions and hypocrisy, especially when directed toward them. If you get caught by your child doing something you have prohibited the child from doing (throwing trash into a flower bush near your front porch), you may be in for a long afternoon of recriminations.

Psychologists have argued that a key initial stage in a child’s development is whether the child has learned to trust the world around it. This is usually navigated successfully with considerable parental care, but not always: diminished care may mean that the child can’t trust the world to provide the necessary care. In the extreme case, parents can so abuse the child’s trust that it develops no trust and lies for fear of telling the truth. It is as if the child has learned to fear reality itself, certainly its own representation of reality. If a child can’t trust its parents to act appropriately with the truth, then it may lie out of defense and distrust. This syndrome can be deep enough to endure in relations more generally. After all, parents are closely related to their children and are expected mostly to have their children’s interests at heart, so that distrust engendered by them may easily extend more broadly, to individuals with less interest at stake in them.

Parents will often act to deceive their children regarding the degree of their commitment and care. “I am doing this for your own good,” a child may hear while being beaten, or later, “I only have your best interests at heart,” while the child’s behavior is being further restricted. Really? People are expected to have their own best interests at heart, and these may conflict with their children’s. More extreme opportunities for parental deception of children are nicely illustrated in some single-parent households. “Where is my father?” asks the child. “He left us,” says the mother (in fact, it is the other way around). “He doesn’t want to have anything to do with you, so get over it.” Here the mother’s initial behavior inflicts a cost on the child, as does its continuation—no relationship whatsoever with one’s father, nor an image of the paternal half of oneself. Or says the mother, “He is dead” (in fact, he is in prison). Later the child learns the truth and is angry at the deception and its associated costs—again, no chance to develop a relationship with the father, through visits to prison, correspondence, and phone calls. Here is a particularly unfortunate example: One child reported that Mom said the man living in the house was her brother. They did sleep in separate rooms, yet the child is sure they have sex together. So which is it: They are not brother and sister and Mom is lying, or they are and she is committing incest? Family and sex could hardly be a more volatile psychological combination. To gain a deeper understanding of the family, we need, indeed, to include sex. Parent must be replaced by mother and father, offspring by son and daughter, and sibling by brother and sister. At the same time, the two sexes have meaning beyond families and attract deceit and self-deception specific to their roles. We turn to this topic next.