Skip to main content

Verified by Psychology Today


Should Parents Make Their Children Apologize?

This is what decades of research say about children and apologies.

 Royalty Free
Source: Boy asks girl for forgiveness | File ID: 45556246 License: Royalty Free

In a recent Huffington Post blog post, parenting author Sarah Ockwell-Smith claimed parents should not insist that their children apologize because doing so will make them "less kind and thoughtful." She justifies this claim by making additional claims about the development of theory of mind reasoning (TOM) in children. She states emphatically that

Young children have a very underdeveloped 'theory of mind'. In essence this means that they have a hard time understanding the viewpoint of others. They struggle to understand how others are feeling, particularly if it differs from what they themselves are feeling… Toddlers and preschoolers are notoriously lacking in empathy skills…

She goes on to claim that, since young children can't yet feel empathy, their apologies can be nothing more than lies:

Why does empathy matter when it comes to saying "sorry"? Because it implies that the child feels bad for what they have done, and in order to feel bad they have to understand how they have made another feel...If they have poor empathy skills (as is normal for this age) they will not have such a train of thought…Forcing the child to apologize in this instance doesn't make the child sorry, in fact all it does is force them to lie.

This blog post was so popular that it was shared on Facebook over 3500 times. There is only one small problem:

None of the claims made are true.

Let's start with her claims regarding Theory of Mind (TOM). TOM refers to the ability to attribute mental states—beliefs, intentions, desires, pretending, knowledge, etc.—to oneself and others, and to understand that others have beliefs, desires, intentions, and perspectives that are different from one's own. Even toddlers empathize with others, know that others can hold preferences different from their own preferences, and engage freely in pretense play. What causes them trouble is reasoning about other people's false beliefs.


Suppose a child watches as an adult say, "Yuck! I don't like that!" when she tastes a cracker, and "Yum! I like that!" when she tastes broccoli. Then she asks the child "Would you give me some?" At 14 months of age, children hand over more of what they like (the crackers), but by 18-months of age, they give the adult what she likes (the broccoli), even if it tastes yucky to them. Here is a brief video showing this kind of interaction.

By 14 months of age, infants show surprise when an adult reaches for an object that is hidden from the adult's perspective but is visible to the infant. By 24 months, they are coordinated enough to reliably reach for and hand the hidden object to an adult who asks for help finding the object.

Eighteen-month-olds can figure out the goal of an adult's efforts to pull apart two attached objects, even though they have not seen the adult accomplish the intended action.

In short, a number of studies using a variety of different methods have shown that by the the second year of life, we already appreciate that other people can have preferences, perspectives, and intentions that differ from our own, and we can use that knowledge to engage in prosocial behavior.


Whether very young children can understand that others' have false beliefs depends on how their knowledge is tested. If we deliberately and explicitly ask young children about another person's beliefs, they respond in ways that suggest they don't. The child in this video answers the way most children younger than age five respond to the classic "Smarties" version of the false belief test.

But if we test their knowledge visually, we find that infants as young as 15 months show false belief understanding. You can see how this is done by watching this brief video. (Fast forward to 4:00)


A wealth of developmental research also shows that infants make implicit value judgments as a matter of course. For example, infants as young as six-months of age prefer individuals who help others over those who hinder or harm others. Here is a video from 60 Minutes demonstrating one of the techniques used to investigate children's responses to helpers and hinderers.


Now what about empathy—when does a true empathetic capacity emerge? It is well established that babies will cry when they hear another baby cry, and will become distressed when they see a crying face. But these don't really count as evidence of true empathy. Crying in response to other another baby's crying could simply be a matter of emotion "contagion" and imitation. Becoming distressed at the sight of a sad or crying face could simply mean that such displays upset babies.

But we usually mean much more by the term "empathy". We usually mean an emotional sharing of experience that often motivates people to try to alleviate the other person’s distress. When does this kind of emotional response begin to emerge?

In one study, researchers investigated the emotional responses of 6-month-olds in a naturalistic situation. Pairs of infants were observed in the same playroom, with their mothers present, and the responses of each infant to the spontaneous crying of the other infant in the pair were examined. Infants rarely responded by getting upset themselves. Instead, the majority of infants directed their attention to the crying peer, and about half of them also reached out or touched the crying infant. These responses appear to be other-oriented rather than self-focused, and so they count as the beginning of empathy.

In another study, 18- and 25-month-olds watched as an adult either harmed another adult by destroying or taking away her possessions (harm condition) or else did something similar that did not harm her (neutral condition). The “victim” expressed no emotions in either condition. Nevertheless, in the harm as compared with the neutral condition, children showed more concern and subsequent prosocial behavior toward the victim.

Suppose we raise the bar even higher by insisting on evidence of cognitive, emotional, and behavioral responses consistent with empathy. Emotional responses would include facial expressions, vocalizations, and gestures directed toward the sufferer. Cognitive responses would include inquiry, or hypothesis testing behavior, as evidenced by active attempts to explore and cognitively comprehend what is happening to the victim. Behavioral responses would include prosocial behavior, such as the child's attempts to help or comfort the distressed other.

A large body of developmental studies have shown quite clearly that these very stringent markers of empathy reliably emerge in the second year of life. So toddlers and young children are indeed capable of feeling empathy, contrary to Ms. Ockwell-Smith's conjectures.

The important point is that in each of these cases, there is insufficient time for infants to have acquired such complex knowledge through experience. These are the basic building blocks of prosocial behavior that are inborn or emerge very early without much deliberate intervention from parents, other than making children feel welcome, wanted, loved, and cared for. Once these foundations are in place, parents can build on them to help their children develop fully into socially competent and decent people.


Now let's return to the question of apologies—do they make children less moral? Adults believe that the purpose of an apology is to publicly accept responsibility for wrong- doing and acknowledge that their behavior that caused harm. Apologies reaffirm a shared value system, and communicates the intention to repair a relationship threatened by one's actions. Do children understand this?

In one study, children ages 4 to 9 years heard stories about moral transgressions in which the wrongdoers either did or did not apologize. Children in the no-apology condition condoned the bad behavior, expecting the ‘happy victimizer’ to feel good about gains they won by being bad. In contrast, in the apology condition, children attributed negative feelings to the transgressor and improved feelings to the victim.

Another study found similar results among children ages 4 to 7: Compared to children who did not receive the apology, those who received an apology reported feeling better, viewed the other child as more remorseful, and rated the other child as nicer.

Another study investigated the impact of apologies among slightly older children. Six‐ and seven‐year‐olds participated in a building activity, but then had their towers knocked over. Those who knocked over the towers apologized, offered to make restitution, neither, or both. Children shared more with a transgressor who offered restitution, a spontaneous apology, or a prompted apology than with one who failed to offer any apology. This was true even though the only response that actually made children feel better was when the transgressor offered restitution. The researchers concluded that apologies serve mainly to repair relationships, while restitution can both mitigate hurt feelings and repair relationships in children.

Researchers documented the natural social interactions of children within 40 families when the children were 2 ½, 4 ½, and 6 ½ years old. Each session lasted nine hours. Over these time periods, they noted that siblings offered 270 apologies. Spontaneous apologies were more frequent (n = 167) than apologies mandated by parents (n = 103 ). Apologies were more frequent after physical harm than after rights violations or verbal harms, and spontaneous apologies increased with age, as did reconciliation following apologies. By 6 years of age, children responded better to spontaneous than to parent-mandated apologies, which indicates that they didn't think mandated apologies were sincere.

Which brings us back to the original question: Should parents insist that children apologize when they cause intentional or unintentional harm to another? The evidence seems to indicate the following practical course of action:

1. Encourage young children to apologize to each other because it will help mend hurt feelings, and very young children don't seem to distinguish between prompted apologies and spontaneous apologies.

2. As children reach elementary school age, spontaneous apologies have more impact, so parents should think twice about whether and how to prompt apologies. If your child's apology doesn't seem sincere, it will have less impact in mending broken relationships.

3. Offers to make restitution not only heal hurt feelings, they also make the repair broken relationships. Parents can be helpful in making suggestions of how children can make up for the damage or hurt they caused either intentionally or unintentionally.

What do you think about this? Feel free to share your experiences with apologies, either as a parent or a child.

Copyright Dr. Denise Cummins April 12, 2016

Dr. Cummins is a research psychologist, an elected Fellow of the Association for Psychological Science, and the author of Good Thinking: Seven Powerful Ideas That Influence the Way We Think.

More information about me can be found on my homepage.

My books can be found here.

Follow me on Twitter.

And on Google+.

And on LinkedIn.