Unlearning parenting, part II
The Gardener and the Carpenter by Alison Gopnik, continued
In part I of my review of The Gardener and the Carpenter, I covered Alison Gopnik’s summary of how childhood and caregiving evolved. In part II, I’ll cover her chapters on how we learn and grow.
Kids learn through social osmosis; it’s impossible to predict what will make it through those semipermeable membranes.
Most people who have spent time with young children have probably experienced their frighteningly good observation and imitation skills. At restaurants, my toddler would arrange her hands under her chin at the precise angle of a random adult diner’s hands. And she can tell when I’m frustrated or amused by watching my face and calculating the velocity of air passing through my nose.
“Learning through looking” is more than just copying funny facial expressions. Gopnik argues that it is a deep and potent form of causal learning, which is foundational to humans’ ability to develop physical and psychological tools. Experiments with children show that imitation is not just mimicry — they have sophisticated levels of discernment about what and who to imitate. A few of the findings from her and others’ research:
Kids imitate people, not robots: In experiments, kids will imitate adults who take actions with interesting results (like bumping a car into a box that then lights up) but not uninteresting ones (bumping a car into a regular box). Kids won’t imitate robots or cars that move by themselves; in this version of the experiments, they just sit there when asked to make the box flash.
Kids imitate intent: Even 18-month-olds don’t blindly copy what they see. Experimenters will show toddlers a person who bangs her head on a box to make it light up. In scenario 1, the person has her hands constricted because she’s wrapped in a blanket. Instead of copying precisely, kids will do it more efficiently by pushing the box with their hands. In scenario 2, the person has her hands free but still uses her head to make the box light up. The kids will imitate her exactly by banging the box with their head, assuming there’s a good reason for this method.
Kids imitate experts: When the experimenter said she had no idea how a toy worked, the kids copied only the necessary actions. When she said she was an expert, they precisely imitated every action, including unnecessary ones.
Kids imitate rituals that otherwise seem pointless: If kids observe something without a causal outcome, they use it as a clue that they’re seeing a ritual. If the experimenter twirls a pen in the air and puts it into a box, kids just put the pen in the box. If the experimenter twirls the pen and puts it back in the same spot, kids will copy the twirling. This may be an important mechanism for transferring cultural rituals to the next generation.
“Learning from testimony,” or listening to what people say, is another important version of social osmosis. While this type of learning seems more straightforward to control as an adult, there are many hidden factors that affect how kids learn.
Attachment: In an experiment, kids were shown a hybrid animal. Their caregiver called it a fish, but a stranger said it was a bird. It looked more like a bird. Children who were securely attached (sad when parents leave but happy when they come back) agreed with their caregiver only if they were more likely to be right. Avoidant babies (who look away when mother leaves and avoid eye contact when she returns) were just as likely to learn from a stranger as the caregiver. Anxious babies (who are inconsolable when caregiver leaves and returns) agreed with their mother, even when she was likely to be wrong.
Track record: Kids trust someone who has been right before more than someone who made a mistake. (I sense this is why parental consistency is crucial in setting expectations about sleep, food, etc.)
Confidence: Three-year-olds will listen to the person who sounds more confident. In my professional career, it has been annoying to continuously hear that women need to sound more confident in the workplace. And yet our desire for confidence is built into our earliest understanding of the world, even to the point of leading us astray: When a person with a poor track record still projects confidence, kids will go with the confident-sounding person.
Stories are another way we learn. Gopnik spends most of this section explaining how kids can surprisingly differentiate fact from fiction from “the third realm, the realm of the supernatural, magical, and religious.” Experiments suggest that young kids don’t seem to confuse real and pretend — e.g., they know that bumping into a tree is real while talking to a tree is pretend. They also segment different fictional worlds — e.g., Batman can talk to Robin but not SpongeBob. They do, however, have strong and real emotional reactions to fictional stories.
Unlike fiction, the “third realm” is a place where invisible entities can have real, causal consequences — spirits that make you sick, presents from Santa Claus, money from the Tooth Fairy. So how do kids differentiate between reality and the third realm?
Gopnik suggests that they pay attention to how adults “explicitly mark the supernatural by using phrases such as ‘I believe.’ No one ever says ‘I believe in oranges,’ or ‘some people believe in oranges and some don’t.’ But we use these locutions to describe magical and religious beliefs. Ironically, saying ‘I believe in magic’ may actually make children less likely to think that your belief is true.” This quote got me thinking: As we become more polarized, do slogans like “believe science” or “in this household, we believe that love is love…” work against the goal to ground us in a common reality?
Gopnik presents a different example of how subtle differences in wording can shape our beliefs. Generic language (”birds fly”) is more likely to lead to essentialist thinking, which means that we believe categories are deep, permanent, and rooted in the reality of the world. “Some birds fly” doesn’t tell you much about what makes the category special, but “birds fly” does. Generic language from adults leads young children to make new predictions, accurate or not (e.g., they might guess that a crow or a penguin can fly if categorized as a bird).
Gender essentialism is, of course, a hot topic. Even mothers who were committed to sexual equality used generics, such as “boys play with trucks” and “girls can play with trucks, too.” Gopnik notes that the latter statement subtly reinforces gender stereotypes but doesn't fully explain why. Based on a brief tour of (free) literature on generic language, I gather that this is because it confirms that the category is a useful, essential way to predict different preferences, even if you’re trying to dispel a specific stereotype. This preprint paper suggests that one effective way to respond is to broaden the scope of the statement (”lots of kids play with trucks”).
Still, this entire section on learning demonstrates how difficult it is to consciously teach specific things to kids, given that they are incorporating subtle cues to decide what’s interesting and who to trust. Gopnik’s conclusion is that parents don’t need to teach — just give kids the opportunity to observe and talk with as many different people as possible while creating a trusted and stable environment. Preschoolers average 75 questions per hour. If they’re satisfied with your answer, they’ll express agreement and use the next question to elaborate; if they’re not, they’ll ask again and again. They are persistent about learning — we just have to give them the opportunities and resources.
Kids learn through unstructured play.
Animals that play — like us — typically have long childhoods, high parental investment, and large brains. By definition, play doesn’t accomplish anything specifically, which makes it an evolutionary puzzle (most confusing for intellectuals who don’t know how to have fun). Why waste energy and risk injury through play?
Rough and tumble play may help animals and children learn to interact with others. Play helps us learn to react in a swift, flexible way in new and uncertain circumstances. Rat pups who play end up having more plastic (flexible) brains because it helps produce certain chemicals in social parts of the frontal cortex.
Exploratory play helps us learn how things work. In experiments, kids were driven to play when something contradicted their theories. Instead of simplifying the world through confirmation bias (like adults), “babies have a positive hunger for the unexpected.”
Pretend play may help us develop counterfactual thinking and empathy for other people’s minds. In experiments, kids who were able to play pretend were also the kids who could think about counterfactuals and alternative possibilities. About one-third of 3-4 year olds in the experiment were “sternly literal,” refusing to entertain counterfactuals — and this wasn’t due to cognitive ability or executive function; it was specifically linked to pretending. Kids who have imaginary friends are better at predicting how someone else might see the world. In one experiment, they’ll fill a bandaid box with paperclips and ask what someone else might think is in the box. Some kids will say someone else would know it’s filled with paperclips; the kids with imaginary friends will say someone else would think it’s filled with bandages.
This section continues to build the case for stepping back from explicit instruction to let kids learn best. Kids played with a new toy longer and tried more actions if the grownup squeaked a toy accidentally versus teaching them to squeak it. As Gopnik puts it, teaching seemed to discourage the children from discovering all the possibilities the toy had to offer.”
“Guided play” works better than lecturing. Putting on detective hats and asking kids to help discover the secrets of shapes helped them figure out the common principle behind triangles or pentagons, including irregular shapes. The grownups’ role was to elaborate on what kids said and ask questions. The guided play approach worked better than didactic instruction, which was still better than “free play” (giving kids flashcards about shapes).
Growing up
A friend who ran a tutoring center once told me that by the time girls turned seven or eight, they were poised and serious, without the free-wheeling zaniness of their younger selves. We talked about how sad it was that the world had taken a scalpel to their silliness and whether there was anything we could do to protect their girlhood personalities.
Gopnik notes that this change may be developmental: “There is a real and striking change from the wild, crazy, poetic preschooler to the sane and sober 7- and 8-year-olds. Indeed, there may be no saner and more sober creature on earth than your average 8-year-old.” Around this age, children may continue some aspects of young kids’ play, but they also begin to engage in “mastery learning” in which they practice skills to make them second nature. Historically, this was the age when children became informal apprentices or began working. She quips: “With adult teeth, apparently, come new adult-like responsibilities.”
She writes that children “learn especially well when they interact with particularly skilled adults in a distinctive cycle of trial and error” and that we should approach more kids’ activities like coaching or apprenticeships:
Imagine if we taught baseball the way we teach science. Until they were twelve, children would read about baseball technique and history, and occasionally hear inspirational stories of the great baseball players. They would fill out quizzes about baseball rules. College undergraduates might be allowed, under strict supervision, to reproduce famous historic baseball plays. But only in the second or third year of graduate school, would they, at last, actually get to play a game.
This relatively recent approach of quarantining kids from real-life skills is one way to explain modern adolescence. She writes about a two system theory of adolescence. In the first system, biological and chemical changes of puberty trigger changes in how teenagers respond to rewards: “Adolescents are reckless not because they underestimate risks but because they overestimate rewards — or rather, find rewards more rewarding than adults do,” especially around peers. The second system, which changes the prefrontal cortex to encourage long-term planning and delayed gratification, depends on learning and experience.
The theory is that these systems were historically in sync. By the time puberty arrived, children would have had lots of time to practice skills they would need as adults. Today, puberty arrives earlier for reasons that are still not fully known, while social changes mean that children have less practice of adult skills. Gopnik’s analogy: “Today’s adolescents acquire an accelerator long before they can steer and brake.” She concludes that instead of only debating when young brains are developmentally mature, we should also discuss how to give them the experience needed to shape an adult brain (e.g., driving apprenticeships as opposed to increasing the driving age).
A way station for the future
For me, this book felt like diving into a warm bath of confirmation bias, with that unique Berkeley tone that’s both laid back and semi-moralizing. I had already read enough bad parenting books to be judicious about which advice I take and to prioritize advice that’s more about changing the parent than the kid. But while I think this book’s message works well to tone down the anxieties of people who, say, have a well-connected grandparent scientist to fall back on, I’m not convinced that a chill gardener vibe is the only good strategy for all parents.
Still, The Gardener and the Carpenter is a nice antidote to constant millennial optimization. Gopnik misses this generational nuance when she questions why we judge “parenting” by its improvement of children, since we don’t assess romantic relationships or friendships by whether we’re improving our spouses or friends. She missed the memo on millennial hustle culture quotes like “you’re the average of the five people you hang out with” or Fabolous and Ne-Yo’s 2007 power couple anthem “Make Me Better.”
So when the frenzy of childcare sets in, settle down with the grunt work and long-term thinking of a gardener: Childhood mess and variability makes us resilient and adaptable to uncertain futures. Get all the caregiving help you can, because we’re supposed to grow up that way. Think less about teaching and more about creating opportunities for kids to observe and talk to many different people. Don’t mourn the loss of different life stages; prepare for the exciting possibilities of the new ones. And if we’re absolutely lucky, we get to “watch our beloved children glide irretrievably into the future we can never reach ourselves.”
Read with:
📺 Mister Rogers’ Neighborhood: A soothing way to spend a half-hour of screen time with your toddler; you get to watch scenes like cellos being assembled at a 1980s cello factory followed by Yo-Yo Ma playing an intimate concert. The pace of the show lets kids observe and learn calmly. (I myself get somewhat distracted by wondering how many of these small, sub-scale factories are left in 2023).
📺 Daniel Tiger’s Neighborhood is excellent, high-quality toddler propaganda, reducing life lessons and emotional intelligence into catchy little tunes that you can sing when they, say, refuse to go to the bathroom before leaving the house. (”Do you need to go potty? Maybe yes, maybe no, why don’t you sit and try to go?”)
📕The Good Neighbor: The Life and Work of Fred Rogers by Maxwell King: I can only recommend this to the most avid Fred Rogers fans because the writing is somewhat repetitive and borders on discipleship (adjectives like “genius” are used liberally) — I realized at the end that this is because the author is associated with the Fred Rogers Center. However, it covers Rogers’ philosophy on children and learning, which are very gardener-esque, starting with his respect for kids as thinking, feeling people. It also covers how a lonely, sickly boy’s fascination with puppets and music led to a national treasure of a TV show, illustrating how powerful it can be to cultivate children’s unique interests and encourage creativity.
📚What’s Going on in There? How the Brain and Mind Develop in the First Five Years of Life by Lise Eliot
📄Parental Ethnotheories on Children’s Learning by Harkness et al - A very accessible anthropological perspective on parenting, comparing parents’ beliefs across cultures. A reminder that this whole focus on cognitive development and stimulation is quite American.