Abstract: Most parents lie to their children. They do it for fun, as a method of behavior control, and to protect children from what they consider to be dangerous truths. At the same time, most parents bring their children up with the message that honesty is a virtue and that lying is usually wrong. How should our practice and our preaching be reconciled? In this paper, I examine the ethics of parental lies. Most philosophers who have written on the ethics of deception have focused on deception of and by autonomous adults. I therefore begin by surveying this literature. Contemporary philosophers have given three types of reason to explain what makes lying wrong (when it is wrong): negative consequences, breaches of trust, and interference with autonomy. I briefly analyze what constitutes a breach of trust and identify four factors that affect how bad a breach is. I then explicate how lying can constitute a wrongful interference with autonomy. A long-running debate concerns whether lying is ethically different than other forms of deception. I argue—briefly—that we do not need to resolve this debate in order to evaluate parental deception. Armed with a framework for what makes lying to autonomous adults wrongful, I turn to the special case of parental lying. Since the parent-child relationship is typically very close, lying to one’s child is a relatively serious breach of trust. This is exacerbated in the case of serious lies that implicate the parent-child relationship or the child’s identity. On the other hand, at least for young children, concerns about autonomy are less significant than for autonomous adults. I close by applying my analysis, along with data on the consequences of parental deception, to different types of parental lie. I argue that lying to one’s child is more rarely justified than is commonly thought, and delineate the circumstances in which it can be justified.

Keywords: parenting, moral obligations, parental lies, education.

"Children are often deceived with the fewest qualms. They, more than all others, need care, support, protection. To shield them, not only from brutal speech and frightening news, but from apprehension and pain—to soften and embellish, and disguise—is as natural as to shelter them from harsh weather."[1]

1. Introduction 

Each night, Amy Kind reports, “as I am putting my 5-year-old son to bed, I tell him a lie.”[2] She sprays his bedroom with air freshener, which she has dubbed “Bad Dream Spray.” It works! His regular nightmares have stopped. Not all parental deception is so playful. For example, parents who experience serious illness often struggle with what to tell their children. A study of 39 Chinese parents with HIV revealed that many did not tell their children and some outright lied—saying that they had another disease, like tuberculosis, hepatitis, or influenza. Some were afraid that their children would tell other people about their diagnosis and they would be stigmatized. Others wanted to spare their children from worrying.[3] These examples just scratch the surface. Parents lie all the time. Kid won’t leave the playground? Tell her you’re going without her. Toy finally stopped making that damn electronic noise? Say that the batteries can’t be replaced. Dog died? We had to send it to the farm. Child proudly shows you a painting? That is so beautiful! Let’s put it on the fridge. 

Despite the litany of lies that parents tell their children, most adults purport to be very keen on honesty. One study asked 127 American undergraduates about how they were parented and 127 American parents about how they raised their children.[4] Four-fifths of the undergraduates reported that they had been taught that lying is never acceptable and similar numbers of the parents reported teaching their kids that message. Nevertheless, both groups also reported high rates of what the lead researcher, Gail Heyman, labels “parenting by lying.” Heyman’s team invented a series of scenarios in which a mother lied to her six-year-old child, either to get the child to behave or to promote positive feelings. Of the undergraduates, 88% identified at least one of these scenarios in which their parents had said something similar to them; 78% of current parents did the same. Heyman concludes: “There was no evidence that the parents who strongly promoted the importance of honesty were less likely to lie to their children than were other parents.”[5] In fact, even when instructing their children on the value of honesty, some parents lied. One parent told their children: “Lying is for bad people and witches; good people and fairies never tell lies.”[6]  

In summary, most parents lie to their children. At the same time, most parents bring their children up with the message that honesty is a virtue and that lying is usually wrong. How should our practice and our preaching be reconciled? In this paper, I examine the ethics of parents lying to their children. I argue that lying to one’s child is more rarely justified than is commonly thought, and delineate the circumstances in which it can be justified.

I begin with a brief taxonomy of different types of parental lie. I then turn to the normative question of what makes lying wrong. Most philosophers who have written on the ethics of deception have focused on deception of and by autonomous adults. I begin by surveying this literature. Contemporary philosophers have articulated three types of reason to explain what makes lying wrong (when it is wrong): lying has negative consequences; lying breaches trust; and lying to someone constitutes interference with their autonomy. I briefly analyze what constitutes a betrayal of trust and identify four factors that affect how bad a betrayal is. I then explicate how lying can constitute a wrongful interference with autonomy. Armed with a framework for what makes lying to autonomous adults wrongful, I turn to the special case of parental lying. Since the parent-child relationship is typically very close, lying to one’s child is a relatively serious breach of trust. This is exacerbated in the case of serious lies that implicate the parent-child relationship and the child’s identity. On the other hand, at least for young children, concerns about autonomy are less significant than for autonomous adults. I close by applying my analysis, along with data on the consequences of parental deception, to different types of parental lie.

A word about the scope of my discussion. I mostly consider the question of parental deception in terms of whether and when lies are ethically permissible. However, there is a long-running philosophical debate over whether lying is ethically different than other forms of deception. For example, rather than directly lie about whether the batteries to the noisy toy can be replaced, a parent might say, “I’ve never seen anyone repair one of these.” This is literally true, but misleads by implying that the toy is irreparable. Or, rather than repeating the myth of Santa Claus, a child’s parents might leave clues, such as “reindeer footprints” or direct their kid to NORAD’s Santa tracker. In section 4, I consider whether the type of deception is relevant to the ethics of deceiving one’s child. I conclude that for the most part it is not—most of the time what we really want to know is whether and when deception of any type is permissible. What I say about parental lies in this paper therefore should be read as applying equally to other forms of parental deception. 

2. A taxonomy of parental lies 

In thinking about the ethics of parental lying it helps to have an idea of what sorts of lies parents tell. A review of parenting manuals, online forums, and relevant social science and bioethics literatures suggests a threefold classification on the basis of the function of the lie.

First, lying for fun. This includes lies told for the entertainment of the parents—“If you unscrew your belly button, your bum will fall off!”—and lies, like the Santa Claus myth, that are intended to be fun for the child.

Second, parenting by lying, where lies are told in order to facilitate the day-to-day job of parenting.[7] In this category are lies told to encourage correct behavior. For example, a child may be told that if he does not eat his vegetables a goblin will come and eat his toes. Parents may also lie in order to affect their child’s emotional state or maintain their relationship with the child. Feigning interest in the child’s own interests is a common example of this. You’re into shiny pebbles? Sure, I’ll pretend to be into the shiny pebbles so that we can have a nice time together looking at them. And many parental lies are told for the convenience of the parents—that is, to make the job easier. For example, we want to get home quicker to make sure that dinner is ready in time, so we say that the park is shutting.

Third, lies told to protect children from the truth. Common lies in this category include lies about illness or death and lies about genetic relationships. For example, a parent may lie about whether they have a fatal illness because they think that their children cannot handle the truth about an incipient death in the family. Or, they may tell their adopted child that she is her parents’ biological offspring because they worry that otherwise she won’t feel like their “real” child.

3. What makes lying wrong? 

Contemporary philosophers have articulated three types of reasons to explain what makes lying wrong (when it is wrong): lying has negative consequences; lying breaches trust; and lying to someone constitutes interference with their autonomy.

3.1. The consequences 

The first of these—the consequences of lying—is used as both a justification for lying and to condemn it. If I lie to my friend and tell her that I like her poetry, I will justify my lie on the grounds that it will have better consequences on balance than telling her the truth. It will protect her feelings, maintain our friendship, and, besides, there is no real chance that she will change how she writes on the basis of my feedback. Lies can be criticized in terms of their consequences in the same way. If only you hadn’t told her how much you liked her poetry, my friend’s spouse might say, then she wouldn’t have quit her engineering job and decided to focus on poetry full-time.

There is not much to be said in terms of philosophical analysis here that is special to lying. Different ethical theories take consequences into account in different ways—some regard the consequences as all that matter, some say there are constraints that apply independently of the consequences, and so on—but how they take consequences into account does not seem to change when they are evaluating the consequences of a lie versus the consequences of some other act. What matters for my purposes, then, are what the actual consequences of lying to children are. In section 5, I return to this question when I attempt to bring together all the ethical considerations that are relevant to parental lies.

3.2. Trust 

One common way to describe what is wrong with lying (and other forms of deception) is to say that it constitutes a breach of trust.[8] Philosophers differ in their analysis of exactly what the wrong consists in, but the basic idea is clear: when we lie to someone, we invite them to trust our sincerity about some matter and then we fail to do what we invited them to trust us to do. This is, depending on the writer, a “violation,” an “abuse,” or a “betrayal” of trust.[9]

Most philosophers who discuss the relationship of deception to trust are primarily interested in the question of whether trust can explain why lies are worse than other forms of deception—for example, do lies break an implicit promise that what the speaker asserts is true, while merely misleading someone does not involve a promise? Are we using the other person’s trust in some different way in lying than in other forms of deception? For reasons that I come to in section 4, I am not particularly interested in that debate. I too want to know what is it that makes deception wrongful, when it is, in order to make comparative judgments. But my ultimate interest is in determining whether and when lying—or any other deception—is justified. That requires a comparison of deceptive to honest actions, not just a comparison among deceptive actions. We must look then to how philosophers have explained how considerations of trust can explain what makes lying, as a form of deception, wrongful.

Collin O’Neil provides a perspicuous account in his discussion of deception and “betrayals” of trust.[10]   According to O’Neil, the sentiment of betrayal—as opposed to, say, mere resentment—arises when we are wronged by someone and we are surprised that they would wrong us. When we trust someone, we assume that they bear good will towards us and when they betray our trust they reveal that we were mistaken in our assumption. This prompts us to re-evaluate our relationship with the person whom we trusted.

Of course, this is not yet to explain what is particularly wrongful about betrayals of trust. To do that, we need to examine why others have a moral reason not to betray our trust—that is, a moral reason to act as though they genuinely bear the good will to us that we assume, based on our relationship, that they do. O’Neil argues that to trust someone—to believe that they will discharge their commitment to us because of their good will to us—is to honor them. And, he argues, the correct response to being honored is gratitude. We should be grateful for the trust bestowed on us and the correct response to this honor is to endeavor to live up to it. In other words, the very fact that we are trusted to tell the truth gives us a further reason to tell the truth so that we live up to the honor of being trusted.[11]

This account can help us to distinguish the wrongfulness of different sorts of lie, depending on whether they betray trust and, if so, how great the betrayal is.

First, some lies will not betray trust at all. Where there is no expectation of good will or no presumption of honesty, there can be no betrayal of trust because there is no trust to begin with.[12] This is true, for example, in games which are understood to involve bluffing or outright lying. Other lies may not betray trust overall, even though the person deceived believes that the liar has a good will towards them and has sufficient expectation of honesty that they are successfully deceived. Suppose that I am about to go into an important job interview and I ask my friend if I look nervous. Though he thinks I do, he judges that what I need at this moment is reassurance, and tells me that no one can see that I’m nervous from the outside. Again, such a lie may not betray my trust if I trust my friend to tell me what I need to hear in such circumstances. In these cases, we might say that insofar as the action involves deception it betrays trust, but insofar as the action expresses good will towards the deceived it does not undermine the relationship on which the trust is based. On balance, trust may be enhanced, not reduced.

Most lies, however, will involve some sort of betrayal of trust. For these, the analysis in terms of failure to live up to the honor implied by being trusted can be helpful in working out how bad the lie is. Lies may constitute worse betrayals of trust along four axes. First, they will be worse insofar as they imply more ill will. Suppose I lie to you about the time a work meeting starts. It would be worse if I do so because I want to sabotage your relationship with our boss than if I do it simply so I can have cover when I also turn up late to the meeting. Second, they will be worse the larger the re-evaluation of the relationship discovering the lie would require. If someone discovered that the man she has been dating for three months lied to her about whether he is married, this might prompt a wholesale re-evaluation of their relationship (making some assumptions about how people commonly view relationships). On the other hand, if she discovered that he lied about his height in his dating profile—he is 5’ 10”, not six feet as he claimed—this might require very little re-evaluation. So he’s a little insecure. That doesn’t undermine her confidence in how much he likes her, how he really views their relationship, and so on. Third, and relatedly, the closer the relationship, the worse a betrayal of trust, since the more the betrayal calls into question the relationship. If my (alleged) best friend cheats me out of £10, that is worse than being cheated by a stranger. Finally, a betrayal of trust will be worse when it concerns something that matters more to the person who is deceived. If you lie to me about whether my work is respected by my peers, that is a greater betrayal of trust than if you lie to me about whether they like my taste in shirts.

3.3. Autonomy 

The third explanation of the special wrong of lying analyses it as an illegitimate interference with autonomy. Suppose Ana tells Bernardo that she is collecting money for a lung cancer research charity when in fact she is planning to use the money for a holiday. He gives her some cash. By deceiving him, she prevents him from deciding for himself how to use his money. This is a violation of his right to dispose of his property as he sees fit.

The idea that deception is—in at least some cases—inimical to autonomy is a Kantian notion. Lying to someone treats them as a mere means, not an end in themselves, since it deprives them of the opportunity to decide for themselves and substitutes the liar’s own will for theirs. But one need not be a Kantian for this explanation to be plausible. Provided that we attach some value to autonomy, when lying interferes with a person making their own decisions regarding matters about which they have a right to decide, it will to that extent be wrongful.

Describing the wrong of deception in this latter way opens up the possibility that more lies are permissible than the strict Kantian might allow. First, there may be cases in which lying is permissible because the subject matter falls outside those about which the victim of the lie has a right to decide. For example, suppose, at a group dinner, Artur asks Ajay if he is gay. Ajay is gay but he is not out. In some social contexts, it may be impossible for Ajay to preserve his privacy regarding his sexual orientation without lying. Staying silent, or responding, “That’s none of your business!” may imply to his listeners that there is something he is concealing about himself.[13] In such cases, we might think it permissible for Ajay to lie. Here, lying would not be a violation of his audience’s autonomy because the subject matter of the lie falls within his—not their—sphere of privacy.[14] Second, if we think that autonomy is only one among the values that matter, then we can make sense of the idea that the wrong of lying can, in principle, be outweighed by other reasons that favor the deception (such as important benefits to the deceived that could not otherwise be realized).[15] Such a view seems much more plausible to me than one that regards autonomy rights as absolute.

4. Is lying special? 

Thus far I have sketched arguments for why lying can be wrongful over and above the consequences of any particular lie. I have also used the terms “lying” and “deception” interchangeably. However, as mentioned earlier, philosophers have spilt a considerable amount of ink on the question of whether and why lying might be worse than other forms of deception. For example, many people seem to think that it is worse to lie than it is to deceive through misleading implicature.[16] At this point in my analysis, it is worth pausing to explain why I am not that interested in the question of whether parental lying is different than other forms of parental deception.

First, all forms of deception betray trust and impede autonomy in the ways that I have been discussing. When we take someone’s word for something, we trust them, and in doing so we assume that they will tell us the truth because they bear us good will. But we also trust people not to mislead us and that means that we assume they will be—for example—honest in implicature because they bear us good will.[17] Likewise, suppose that Ana does not literally lie to Bernardo about what she is collecting money for. She just tells him that she works for a lung cancer charity and then immediately asks if he will donate some money. Aha!, she thinks, I’ve not actually said that the donation is for the charity, so at least I’m not lying! But still she has interfered with his autonomy by substituting her own judgment about what he should do with his money for his.

Second, depending on the details of the case, it may sometimes be better to lie or better to deceive in some other way. Consider the murderer at the door asking after my friend. I assume I ought to deceive him. But if I am a terrible liar, then I probably ought to go for some other form of deception, so as not to give it away with my blushes. On the other hand, if I cannot think of a clever way to mislead him, I ought to just lie and say, “No, she’s not here,” rather than give the impression of weaseling my way out of responding. All this says nothing about whether one method of deception is worse in itself—deception is justified in this situation and I should just pick the method of deception that is most likely to be effective.

Third, there are, of course, special contexts in which lying is worse than misleading because the speaker’s responsibility is known by all participants to be limited to just saying the literal truth. Courtrooms are such a case.[18] But parents are generally not confronting their children in court (and I can’t give you advice if you are).

Fourth, despite the various arguments that have been given for why lying is typically worse than mere misleading (or the other way round), I struggle to see how the difference between forms of deception would be very important, even if it were to exist. Outside of special contexts like courtrooms, there do not seem to be many cases, for example, where it seems plausible to say that a lie would be wrongful but false implicature regarding the same proposition would be permissible.[19] At best, if you are convinced by the argument that lying is to some extent worse than misleading, you should prefer to mislead rather than lie (and vice versa if you have the opposite view about which form of deception is preferable). But in the vast majority of cases where lying is wrongful it will also be wrongful to mislead. What we really want to know is whether and when deception of any type is permissible.

5. Lying to one’s children 

Let us return to parents lying to their children. I have sketched the considerations that explain why lying (and other forms of deception) can be wrongful, over and above their expected consequences. Thus far, however, I have dealt only with the case of adults lying to other, presumed competent, adults. Children are different than competent adults in morally important ways. Their social and legal status as children leaves them more dependent on others than most adults are. This is true even for older children who have the cognitive capacities to make their own autonomous decisions. Younger children are even more vulnerable and dependent—they are often very trusting, they are less capable of autonomous living, and they rely on their caregivers to bring them up to become flourishing, competent adults themselves.[20] These facts are relevant to the evaluation of parents lying to their children.

5.1. Trust and lying to children 

Consider, first, how lying can betray a child’s trust. Almost any deception of a child is liable to betray their trust to some degree. But, as I discussed above, some betrayals of trust are worse than others. I argued that deception betrays trust more when it reveals more ill will, when it prompts greater re-evaluation of the presumed relationship, when the relationship is closer, and when the content of the lie is more significant. It is clear from this why children still often feel that serious lies—lies about whether they were adopted or how sick they or a family member is—are such huge betrayals. They are about matters that are highly significant, such as the child’s identity, they may call into question core aspects of the child-parent relationship, and the parent-child relationship is typically as close as any relationship can be.[21] Serious lies, then, betray trust to a high degree.

More trivial lies can also substantially betray trust. First, the fact that the parent-child relationship is close means that any lie will be a worse betrayal of trust than if the parent lied to an acquaintance or friend. Telling your kid that her bum will fall off if you unscrew the belly button still betrays her trust! In fact, it only works because she trusts you—her authoritative, wise, loving parent—so much. Second, many such lies do not evince good will towards the child. The example just given is told for the parent’s amusement, not out of good will.

One further point about trust may help with justifying some deception of children. This is that our children trust us with much more than just telling the truth. They trust us, for example, to keep them safe, emotionally and physically. In certain circumstances this obligation to care for one’s child can ground trust-based reasons that actually favor deception. For example, maybe your five-year-old is not ready to hear about the grisly crime on your street. Saving him from anxiety and nightmares might be worth a little deception. Or maybe he’s really worried about his parents splitting up. Assume that you are not, in fact, anticipating a break-up. Still, the truth is that no one can predict the future and that couples sometimes grow apart. That is not reassuring. You might honor his trust more with a straightforward, “No, we’re not going to get divorced.”[22]

In cases like these, where a parent judges that deception is necessary to honor their child’s trust, considerations of trust seem to pull in different directions. It may be true both that your child trusts you to tell the truth and that they trust you to keep them safe. If the truth will undermine their safety, then completely honoring their trust may be impossible. This is a tragic situation, in that some breach of trust is inevitable. It falls then to the parent to judge whether the child’s trust overall is better served through honesty or deception.

5.2. Autonomy and lying to children 

While lying to one’s children often involves a worse betrayal of trust than lying to most adults, the opposite is true with respect to autonomy. Young children are not fully autonomous. In many areas of life, they lack stable values, and are not able to reflect on their goals and decide what to do on the basis of reasons. Where they lack the capacity to make their own decisions, paternalism is justified. The objection to lying to children that it illegitimately substitutes another’s judgment for their own therefore does not apply.[23]

However, as children develop, they develop values, they get better at reasoning, and they get better at making their own decisions. As that happens, parents ought to take their child’s developing autonomy into account. Paternalism can be problematic even with children, when they are capable of making their own choices. This can matter when we’re weighing up the pros and cons of lying to an older child or adolescent. Within reason, we should let them make their own mistakes. We should respect them by telling the truth even in some cases when we think they will make bad decisions. This has its limits, of course. Because children are not yet fully autonomous, some paternalism is still justifiable. We should not manipulate a ten-year-old into going to play with her friends, even though we know she’d be happier once out of the house. But, I would argue, we can step in if she is doing something that risks life or limb—with deception if need be.

5.3. The consequences of lying to children 

So much for trust and autonomy. In order to render ethical verdicts about lying to one’s children it is necessary to consider not just those considerations that make deception special, but also what consequences—good and bad—deceiving one’s child is liable to have. Here, of course, it is impossible to cover every possible case. Plus, I do not have a general theory of which consequences matter and to what extent. Nevertheless, there is some valuable data from the social sciences that we can draw upon to evaluate the effects of different types of lie. In what follows, I summarize some of that data. Along with the ethical analysis of how lying implicates trust and autonomy, and some non-contentious judgments about which consequences matter, this can give us rough guidance about parental lies.  

People usually justify the lies they tell in terms of their consequences. A white lie is told in order not to hurt someone’s feelings with the truth. A practical joke is justified by the entertainment it supposedly provides. This reasoning applies to more serious lies too. Someone who lies to their partner about an affair may think the lie was justified, even if the affair was not. The affair is in the past, but if their partner found out about it the relationship would be destroyed. So, they lie to save the relationship. Parents typically justify their lies in terms of the consequences too. Lying for fun, parenting by lying, and lying to protect children from the truth are all done because of their anticipated benefits. If a parental lie is justified, it will be so in virtue of the good consequences the parent expects as a result. However, just as with adults, lies told to children can have bad consequences as well as good.

Before looking at the data on the effects of parental lies, I should note a prior question concerning which consequences count. Lies can have effects on more than just the individual who is deceived. For example, if adopting parents tell their child that she is their biological offspring, they must decide what to tell other family members too. Involving them in the lie makes them complicit, risks them revealing the truth, and may affect their relationship with the parents and the child. On the other hand, some lies might be thought to have additional beneficial effects for other parties. A parent who lies in order to avert an afternoon of whining might do so primarily for their own peace of mind, not because they think it benefits their kid. Which of these consequences should be weighed in the moral calculus?

Limited space precludes full analysis of this complex topic. In brief, it seems plausible that any predictable negative effects on third parties should count. Harms to others require justification no matter whether they are caused directly or indirectly. I doubt, though, whether harms and benefits to the potential deceiver should count in the majority of cases. This is because most lies involve at least a minor wrong—they normally breach trust, at least. Outside of exceptional circumstances, we do not generally think it permissible to wrong someone for one’s own benefit. For example, it would be impermissible for me to lie to someone about the bus timetable in order to ensure that I would get a seat to myself or to cheat at cards in order to win the pot. This verdict holds, even if the balance of overall benefits is positive—perhaps I am somewhat poorer than the other gamblers and so overall welfare would be enhanced if I won. Exceptions to this generalization might arise for cases when the other party has themselves acted wrongly or for self-protection against substantial harms or infringements of rights. For example, I think it would be permissible for me to lie in order to feed myself and, as described above, I think that it can be permissible to lie to protect one’s private information. I see no reason to think that the justificatory bar would be lower when thinking about parents and children; if anything, the bar for wronging one’s child for one’s own benefit should be higher given that you owe them a duty of care. So, in most cases, potential benefits to the parents from lying to their children should be set aside.[24]

Turn now to what we know of the effects on children of parental deception. Researchers in various countries have asked young adults—generally college students—about how they were parented.[25] Their studies reveal that parenting by lying is common across cultures. Young adults who recall more lying by parents generally report worse relationships with their parents. They are also more likely to lie themselves, and score higher on measures of antisocial behavior, mood disorders, and psychopathy.

I should note right away that these studies do not show that lying to one’s children causes criminal behavior or mental illness.  First, there is no evidence that these studies enrolled a lot of psychopaths or people with severe mental illness. The studies asked participants to complete questionnaires that would put them on a scale for each attribute the researchers were interested in.[26] The studies show only that there is a correlation between reporting more parenting by lying and higher scores on the scales, not a correlation between, say, reporting some level of parenting by lying and actual psychopathy.

Second, correlation is not causation. People who report a lot of lying in their childhood and people who report very little are probably different in many other ways too. For example, if you think that your parent is dishonest, you probably have a poor relationship with them. Having a poor relationship with a parent is also a very plausible cause of psychological difficulties, such as mood disorders. In sum, there is positive evidence, but weak evidence, of negative effects from telling children lots of lies.

Asking adults to remember how much they were lied to as children has some obvious drawbacks. Some researchers have instead conducted experiments with children to see directly how they respond to lies. Chelsea Hays and Leslie J. Carver randomized young children to a lie or a no-lie condition.[27] Half were told, “There is a huge bowl of candy in the next room, want to go get some?” There was no candy. When they got into the room, the experimenter explained that he just wanted to get the child to come with him to play a game. The children in the no-lie condition were simply asked if they wanted to play a game in the next room. Both sets of children were then given the opportunity to cheat on a game and they were asked whether or not they had cheated. The researchers divided the children by age. For pre-school children, it made no difference whether they’d been lied to. But for the school-age children (aged 5-7) it mattered. If they had been lied to, they were more likely to cheat at the game and more likely to lie about whether they cheated.

One interpretation of the lying experiment is that it shows the dangers of lying to children—if you lie then you will bring up liars! But these children may be making more nuanced judgments. Perhaps, Hays and Carver suggest, “the children did not feel the need to uphold their commitment to tell the truth to someone whom they perceived as a liar.”[28] And perhaps this is a good result: children shouldn’t feel like they have to tell the truth to dishonest strangers.

Other research on lying supports the idea that children care a lot about character. Several studies have involved asking primary school-aged children about stories. In the stories, a character either lies or tells the truth. The reason they lie or tell the truth is varied so that sometimes it’s for selfish reasons and sometimes to benefit others.[29] Children, particularly as they get older, tend to regard those who lie to benefit others as trustworthy, despite their lying. In terms of the consequences of lying to children, I think this might be one of the more important findings. It suggests that lying to children for benevolent reasons is much less likely to damage the parent-child relationship than if parents lie selfishly.

The data I’ve discussed so far has looked at the effects of trivial lies. What about the more serious lies that parents tell their children—about adoption, illnesses, and so on? The literature on the consequences of telling (or not telling) children about serious matters is pretty consistent. Almost without exception, the results of parents telling the truth are better for the child and for the relationship, than not telling or having the child find out from other sources.

Amanda Turner and Adrian Coyle asked the offspring of sperm donors about their experiences finding out how they were conceived. Feelings of betrayal and mistrust were common. “Sarah” reported: “I felt my entire life was based on a lie and I was furious with my mother for dying with this secret.”[30] No one said that they were glad to have been deceived. There are similar findings regarding the responses of adopted children. An Australian Institute of Family Studies study of children of “closed adoptions” that included 823 adoptees concluded: “the earlier they knew and the more openly and freely discussed adoption was as a topic within the adopted family, the higher the level of wellbeing of the adoptee.”[31] Here is one adoptee who discovered as an adult that she had been adopted:

“When I found out I was adopted, it came completely out of the blue as a note on a birthday card sent from an aunt [by marriage], telling me that although I was adopted, I would always be her nephew. From this point, my life was shattered; the life I had been living up till now was a complete lie.”[32]

In addition to how children feel when they discover lies about their origins, they may also need to know certain facts to make good decisions later. Suppose your genetic father had a heart condition and died of bowel cancer. That is really important information for you and your doctor to know. If you don’t know that your father is not your genetic father, you may not realize that you could be at risk. So, there can be objective harms from telling serious lies, too.

Turn now to lies about family illness and death. Parents of a very sick child may pretend that she’ll get better, even though she won’t. Though it may feel loving and protective, this is usually unhelpful.[33] Dying children are often aware that they are not being given the complete picture. If they are not communicated with honestly, they don’t have the chance to address their fears or prepare for death. Ulrika Kreicbergs and colleagues sent questionnaires to every parent in Sweden who had lost a child to cancer over a six-year period.[34] They found a striking result. About a quarter of the parents who did not talk to their child about impending death regretted that they had not done so. None of the parents who talked to their child about death said that they regretted it.

Studies of parental illness and disclosure reveal a complex picture, but one that also generally supports disclosure. Children of mothers with HIV sometimes react negatively to being told and a small number may have long-term negative reactions, such as regressive behaviors.[35] However, children whose mothers do not disclose also frequently experience guilt and anxiety due to not knowing the source of problems within their family environment, but being aware that something is amiss. Most parents who have regrets about disclosure, regret the manner of disclosure not the fact that they told their child. Matters are similar with respect to parental cancer: families benefit from support in talking with their children, but the results of doing so are largely positive on balance.[36]

6. Fantasy and make-believe 

Children, especially young children, engage in a lot of make-believe when they play. This extends from pretending to be a doctor or serving pretend tea to having imaginary companions and giving thoughts and feelings to soft toys. Many parents enjoy playing along (or are pressured into doing so by their insistent child). But children sometimes seem to blur the line between fantasy and reality. They are frightened of the monsters and protective of their imaginary friends. If I condemn deceiving children, does that also mean I should condemn any encouragement of make-believe as tantamount to lying?

No. Despite their strong emotional reactions to fantasy events and creatures, and contrary to earlier assertions by child psychologists, it appears that children appreciate the difference between reality and pretend from a very early age.[37] A child can be genuinely scared of the monster under the bed she’s just invented but also be able to invent the ritual that her parent must use to appease it.[38] To initiate, encourage, or go along with make-believe is not, in fact, to encourage false beliefs in your child, and so it is not deception.

7. Some verdicts on lying to our children 

Insofar as lying breaches trust, interferes with autonomy, or predictably puts the victim at risk of harm it is pro tanto wrongful. To say that an action is pro tanto wrongful is to say that there are moral reasons that count against it. However, those moral reasons do not entail final verdicts on what should be done, since the pro tanto wrongs could nevertheless be justified.[39] A relatively simple framework can help us think through whether lying is likely to be overall justified in a particular case. First, we should assess how the different ethical considerations apply to the proposed deception. Would it breach trust (and, if so, how great would the breach of trust be)? Would it interfere with the child’s autonomy (and, if so, how substantial would that interference be)? What are the expected consequences for the child and, where relevant, for other parties? Where only the child’s interests are at stake, lying will only be justified when there are positive expected consequences for them. This is because breaching trust and interfering with autonomous action never speak in favor of an action, and the cases where trust is maintained overall by lying will be ones where the deception is expected to protect the child. So, second, if the overall expected consequences for the child are positive, we must consider whether they justify the deception. They will justify the deception just in case the good consequences are sufficiently good that they outweigh the considerations that tell against the lie and there are no morally better alternative actions. Third, in the rare case that the interests of other parties are deemed ethically relevant, the expected harms and benefits to them should be included in the moral calculus.

This framework does not give a formula for deciding what to do. In applied ethics such formulas are not available. It is up to the reader to assess how important the different ethical considerations are, to estimate how likely and substantial are the effects, and to judge when it seems that the good consequences of a lie are good enough. That said, I think that the framework plus the foregoing normative analysis does suggest some rules of thumb about what types of parental lies are likely to be permissible and impermissible. The lies that are most likely to be justified will be those where the negative consequences are low, the positive consequences sufficiently great, the betrayal of trust from lying outweighed by the maintenance of trust overall, and the wrongs the minimum necessary.

This suggests that in most cases we should not lie to protect children from the truth concerning serious matters such as a child’s genetic relationships, illness in the family, and the like. Serious lies are likely to be discovered and can have insidious effects on family relationships even when they are not. The betrayal of trust is much greater when it comes to lies about subjects like terminal illness or adoption. These are matters that are very significant and affect a child’s identity.[40]

At the other extreme, and to my disappointment, I think the verdict on lying to children for one’s own entertainment is also pretty clear. Unless the child is going to be in on the joke, messing with him in this way is just taking advantage of his trust for your own trivial pleasure. I sound like a killjoy, but sometimes that’s where we end up in philosophy. It’s not okay to tell your child that if you unscrew his belly button then his legs will be detached. Likewise, lies that are told for the alleged joy of the child—that fairies exist or that Santa Claus rewards good children with gifts—are hard to justify.[41] If genuinely done for the sake of the child, they betray trust less, but there is still some betrayal. Plus, they might have negative consequences and there are surely alternative ways to make one’s child happy or excited. As just noted, this does not rule out stories and fantasy play—those are exactly the cases in which children are (or should be) in on the pretense.

Most parenting by lying—that is, lying to facilitate the day-to-day job of parenting—also appears unjustified. As a habit, there is some evidence that it is destructive to the parent-child relationship. It is a breach of trust. And it doesn’t set up children well to make good decisions. If someone is threatened with a consequence that doesn’t arrive, then they learn to disregard their caregivers. If a child is told lies just to make them feel good, then praise becomes hollow. I’m not criticizing any parent who has gotten to the end of their tether and lied, but parenting by lying should be avoided if possible.

Two types of parental lie are most likely to be justified. One is parenting by lying in a case where it is hard to find a way to achieve one’s morally legitimate aim by a morally better method. With a young child it might just be the nicest way to get them to leave the park if you tell them that it’s closing. You know, as your child does not, that if you don’t leave now you won’t be back in time for dinner, and so your child will be grumpy, and so forth. Alternatives—getting angry, physical removal—are potentially more ethically fraught.

The other type of case that could be justified would be a lie that really does protect the child. This is in fact not true of a lot of serious lies—the data on adoption, for example, are pretty clear that telling one’s child the truth is better for them. But it might well be true in response to questions about matters that do not implicate the child’s identity, but where the truth has the potential to provoke substantial anxiety. If there is good evidence that the truth really is too hard to hear or too hard at the child’s particular age, lying might then be permissible in virtue of the consequences for the child. In both these cases of justified lies, parents are ultimately doing what their children trust them to do—look after them.[42]

[1] Bok, Sissela. Lying: Moral Choice in Public and Private Life. Vintage (1979): 217. [^]

[2] Kind, Amy. “Creative Mothering: Lies and the Lying Mothers Who Tell Them.” In Motherhood: The Birth of Wisdom, Sheila Lintott ed. (2011). [^]

[3] Zhou, Y., Zhang, L., Li, X., & Kaljee, L. (2013). Do Chinese parents with HIV tell their children the truth? A qualitative preliminary study of parental HIV disclosure in China. Child: Care, Health and Development, 39(6), 816-824. [^]

[4] Heyman, Gail D., Diem H. Luu, and Kang Lee. “Parenting by lying.” Journal of Moral Education 38.3 (2009): 353-369. [^]

[5] Heyman et al. 2009: 359. [^]

[6] Heyman et al. 2009: 360. [^]

[7] This includes the category of lies that Kira Tomsons calls “white lies for children.” (“Lying to Children and the Cultivation of Epistemic Virtue.” Canadian Journal of Practical Philosophy Vol. 4, 2020: Practical Ethics: Issues and Perspectives: 44-57, at 45). [^]

[8] See, e.g., O'Neil, Collin. “Lying, trust, and gratitude.” Philosophy & Public Affairs 40.4 (2012): 301-333; Simpson, D. I. (1992). “Lying, liars and language.” Philosophy and Phenomenological Research, 52 (3), 623-639; Williams, Bernard. Truth and Truthfulness. Princeton University Press (2002). [^]

[9] O’Neil 2012; Williams 2002. [^]

[10]O’Neil also distinguishes “betrayals” of trust from “abuses” of trust and argues that forms of deception can differ in how bad they are because though all betray trust only some abuse trust (O’Neil 2012: 325-329). [^]

[11] See, also, Philip Pettit, “The Cunning of Trust,” Philosophy & Public Affairs 24 (1995): 202–225. Some might find the language of “honor” odd, especially in the context of young children—do children really go around honoring their parents? To my ear, it is much more natural (and plausible) to say that it is an honor to be trusted by a child, and that we should be grateful that they trust us so deeply. [^]

[12] Depending on the reason that the recipient of the lie does not expect good will or presume honesty, deception could still be wrongful under these circumstances. For example, I might know that someone is attempting to scam me and so assume that they will not be honest. Their deception will therefore fail. But they do not know that. The fact that they attempt to deceive me can then still be wrongful (just as it can be wrongful to stab someone with a stage knife that one believes to be real). [^]

[13] This is an example of what Sam Berstler labels a “loud silence problem.” (Berstler, S., “Bad Question!” Draft manuscript (2023).) Berstler also judges that where one’s interlocutor has put one in a position where refusing to answer a question would give them private information it is permissible to lie. [^]

[14] To spell out the ethical analysis in full. Lying does not illegitimately interfere with the audience’s autonomy for the reasons given. It might involve a minor breach of trust, depending on the social context, including whether the question is asked innocently or malevolently. We can assume that Ajay predicts no serious negative consequences will result from the lie. Given the stringency of an individual’s privacy right to control such personal information, and that there are not morally better ways for Ajay to protect his privacy, the moral reasons against lying are easily outweighed. [^]

[15] By “wrong” here I therefore mean pro tanto wrong: interference with autonomy gives a moral reason against lying to someone, but that reason could be outweighed by countervailing moral reasons. [^]

[16] See, for example, Strudler, Alan. “The distinctive wrong in lying.” Ethical Theory and Moral Practice 13.2 (2010): 171-179 and Webber, Jonathan. “Liar!” Analysis 73.4 (2013): 651-659. But Jennifer Saul argues that lying is not worse (“Just go ahead and lie.” Analysis 72.1 (2012): 3-9). And Clea Rees argues that lying is actually morally better (“Better lie!” Analysis 74.1 (2014): 59-64.). [^]

[17] 17 Cf. O’Neil 2012: 318-319. Note that communication is impossible without the assumption that one’s conversation partner is being cooperative, including that they intend to communicate the truth in most conversational contexts—this is one of the basic insights of H.P. Grice and the philosophers of language who followed him (Grice, H. Paul. Studies in the Ways of Words. Harvard University Press (1989). [^]

[18] Saul 2012: 7-8. [^]

[19] Strudler gives one example involving a representative of the Walt Disney company who is trying to buy a plot of land from a “tough-minded used car salesperson” (Strudler 2010: 174). According to Strudler it would be permissible for the rep to mislead the landowner by implying that he is buying the land for his personal use, but it would not be permissible to lie about it. I must confess that I think honesty is required in this situation, so the example does not have the intuitive pull for me that it does for Strudler. [^]

[20] Cf. Tomsons 2020: 51. Note that some children, due to terminal illness or substantial cognitive impairment, may not be expected to become competent adults. Nevertheless, they rely on their caregivers to bring them up in a way appropriate to their specific situation. [^]

[21] On “foundational lies” and how they can undermine the relationship of trust between parent and child, see Charles, Sonya. “On the Immorality of Lying to Children About Their Origins.” Philosophy in the Contemporary World 18;2 (2011): 22-33. [^]

[22] Note that having a good will towards one’s child is not on its own sufficient to justify a lie. According to the above analysis, the person whose trust is betrayed by a lie believes that they will be told the truth because their conversation partner bears them good will. This generates an obligation to tell the truth—to do what one is trusted to do—not simply an obligation to act out of good will. [^]

[23] Here the relevant form of autonomy is internal: when someone has the capacity to make decisions for themselves on the basis of their values, they are autonomous, and that grounds a right to make their own decisions. Children also generally have diminished autonomy in an external sense, in that they are usually materially and legally dependent on adults. [^]

[24] For more discussion of how parents should balance the competing interests of family members, see Millum, Joseph, The Moral Foundations of Parenthood. Oxford University Press (2018): 128-153. I confess to feeling a great deal of sympathy for parents who lie in order to reduce their own stress. Parenting is hard and every parent at some point falls short of what is ideal. [^]

[25] Setoh, Peipei, et al. “Parenting by lying in childhood is associated with negative developmental outcomes in adulthood.” Journal of Experimental Child Psychology 189 (2020): 104680; Santos RM, et al. “Exposure to Parenting by Lying in Childhood: Associations with Negative Outcomes in Adulthood.” Front. Psychol. 8 (2017):1240; Joscelyn R. Cargill and Drew A. Curtis. “Parental Deception: Perceived Effects on Parent-Child Relationships.” Journal of Relationships Research, Volume 8, e1 (2017): 1–8; Dodd, B., Malm, E.K. “Effects of Parenting by Lying in Childhood on Adult Lying, Internalizing Behaviors, and Relationship Quality.” Child Psychiatry Hum Dev (2021). [^]

[26] Take the Levenson Self-Report Psychopathy Scale that some of these studies used. A score of 58 or more is the threshold for psychopathy. Non-psychopathic individuals typically score between 0 and 48. [^]

[27] Hays, Chelsea, and Leslie J. Carver. “Follow the liar: the effects of adult lies on children's honesty.” Developmental Science 17.6 (2014): 977-983. [^]

[28] Hays and Carver 2014: 981. [^]

[29] Xu, F., et al. “The role of honesty and benevolence in children’s judgments of trustworthiness.” International Journal of Behavioral Development, 37(3) (2013): 257-265; Fu, Genyue, et al. “Children trust people who lie to benefit others.” Journal of Experimental Child Psychology 129 (2015): 127-139. [^]

[30] Turner, A. J., & Coyle, A. “What does it mean to be a donor offspring? The identity experiences of adults conceived by donor insemination and the implications for counselling and therapy.” Human Reproduction, 15(9) (2000): 2041-2051, at 2045. [^]

[31] Kenny, P., et al. Past adoption experiences: National Research Study on the Service Response to Past Adoption Practices (Research Report No. 21). Melbourne: Australian Institute of Family Studies (2012), at 92. [^]

[32] Kenny et al. 2012: 90. [^]

[33] Dunlop, Susan. “The dying child: should we tell the truth?” Nursing Children and Young People 20.6 (2008). [^]

[34] Kreicbergs, Ulrika, et al. “Talking about death with children who have severe malignant disease.” New England Journal of Medicine 351.12 (2004): 1175-1186. [^]

[35] Murphy, Debra A. “HIV-positive mothers' disclosure of their serostatus to their young children: a review.” Clinical Child Psychology and Psychiatry 13.1 (2008): 105-122. See, also, Zhou et al. 2012: 816-817. [^]

[36] See, e.g., Barnes, Jacqueline, et al. “Qualitative interview study of communication between parents and children about maternal breast cancer.” Bmj 321.7259 (2000): 479-482; Landry-Dattée, Nicole, et al. “Telling the truth… with kindness: retrospective evaluation of 12 years of activity of a support group for children and their parents with cancer.” Cancer Nursing 39.2 (2016): E10-E18.[^]

[37] Gopnik, Alison. The Gardener and the Carpenter: What the New Science of Child Development Tells Us About the Relationship Between Parents and Children. Macmillan (2016). [^]

[38] Adults, too, have strong emotional reactions to fiction, including their own fictional creations. We may engage in less whole-hearted pretending than children do, but it’s a difference of degree not of kind. [^]

[39] To say that an action is pro tanto wrongful is therefore quite different than saying that it is prima facie wrongful. Pro tanto is an ontological notion—a pro tanto wrong is a reason that counts morally against the action. Prima facie is an epistemological notion—a prima facie wrong is an apparent reason that may turn out not to apply at all. [^]

[40] What about lying to protect the parent’s own private information? For matters—like serious parental illness—that substantially implicate the interests of other close family members, I am skeptical that our standard model of privacy rights held against strangers applies. Do I really have the right to keep secret from my child the fact that I am likely to die soon? However, we might imagine cases where the child’s interests are not so intertwined with those of the parent, such as when a parent wishes to keep their sexual orientation private (section 3.3). In such cases, the stringency of the privacy right might justify a necessary lie, but I would question how often such lies are truly necessary to avoid revealing information: in general, one can more easily refuse to answer a child’s question without implying anything about the true answer, than one can an adult’s.[^]

[41] For more discussion of lying about Santa Claus see Johnson, David Kyle. “Against the Santa Claus lie: The truth we should tell our children.” In Scott C. Lowe (ed.), Christmas‐Philosophy for Everyone: Better Than a Lump of Coal (2010): 137-150; Gavrielides, Era. “Lying to Children about Santa: Why It's Just Not Wrong.” In Lowe (ed.) (2010: 151-160). [^]

[42] For valuable feedback on this paper, I thank Kirstin Borgerson, Danielle Bromwich, Benjamin Davies, Tom Douglas, Michael Garnett, Doug MacKay, Liam Shields, Marika Warren, two anonymous reviewers for this journal, and audiences at the Institute of Philosophy, the Manchester Centre for Political Theory, the New Scholarship in Bioethics Annual Symposium, the Society for Applied Philosophy Annual Conference, and the St Andrews Philosophy Society. [^]

Forthcoming