The Gardener and the Carpenter: What the New Science of Child Development Tells Us About the Relationship Between Parents and Children

by Alison Gopnik

We can aspire to love better without thinking of love as a kind of work. We might say that we try hard to be a good wife or husband, or that it’s important to us to be a good friend or a better child. But I would not evaluate the success of my marriage by measuring whether my husband’s character had improved in the years since we wed. I would not evaluate the quality of an old friendship by whether my friend was happier or more successful than when we first met—indeed, we all know that friendships show their quality most in the darkest days. Nevertheless, this is the implicit picture of parenting—that your qualities as a parent can be, and even should be, judged by the child you create.

Love doesn’t have goals or benchmarks or blueprints, but it does have a purpose. The purpose is not to change the people we love, but to give them what they need to thrive. Love’s purpose is not to shape our beloved’s destiny, but to help them shape their own. It isn’t to show them the way, but to help them find a path for themselves, even if the path they take isn’t one we would choose ourselves, or even one we would choose for them. The purpose of loving children, in particular, is to give those helpless young human beings a rich, stable, safe environment—an environment in which variation, innovation, and novelty can blossom. This is true both from a biological and evolutionary point of view and from a personal and political one. Loving children doesn’t give them a destination; it gives them sustenance for the journey.

“Mother” and “father” are as old as English itself, and “parent” has been around since at least the fourteenth century. But the word “parenting,” now so ubiquitous, first emerged in America in 1958, and became common only in the 1970s.

Smaller families, greater mobility, and older first-time parents radically altered the learning curve. For most of human history, people grew up in large extended families with many children. Most parents had extensive experience of taking care of children before they had children themselves. And they had extensive opportunities to watch other people, not just their own parents, but grandmothers and grandfathers, aunts and uncles and older cousins, take care of children. Those traditional sources of wisdom and competence—not quite the same as expertise—have largely disappeared. Parenting how-to books, websites, and speakers are appealing because they seem to fill that gap.

But, in fact, parenting is a terrible invention. It hasn’t improved the lives of children and parents, and in some ways it’s arguably made them worse. For middle-class parents, trying to shape their children into worthy adults becomes the source of endless anxiety and guilt coupled with frustration. And for their children, parenting leads to an oppressive cloud of hovering expectations.

We ate pie, pasta, or pot stickers because our mothers cooked them, and they cooked that way because their mothers did before them. Those many and varied traditions all led to reasonably healthy outcomes. In the twentieth century, especially the American middle-class twentieth century, the erosion of those traditions led to a culture of “nutrition” and “dieting” that has a lot in common with the culture of parenting.

Cooking and caring for children are both essentially and distinctively human—we couldn’t survive as a species without them. But the more we intentionally and deliberately cook and eat in order to become healthy, or raise children in order to make them happy and successful adults, the less healthy and happy we and our children seem to become. The preponderance of parenting books, like the preponderance of diet books, should, just by itself, be a sign of their futility; if any of them actually worked, that success ought to put the rest of them out of business.

The central scientific idea of this book is that the answer lies in disorder. Children are incontrovertibly and undeniably messy.

A long tradition, going back to the Greek rationalist philosophers, sees these forces of disorder as the enemies of knowledge, progress, and civilization. But another tradition, going back to the nineteenth-century Romantics, sees disorder as the wellspring of freedom, innovation, and creativity. The Romantics also celebrated childhood; for them, children were the quintessential example of the virtues of chaos.

Research in epigenetics shows how features of the environment, even quite subtle ones such as the quality of caregiving, can make genes active or inactive. For example, mice who are stressed early in life express certain genes differently. The same thing happens in human children. Genes vary, but children’s experiences also vary, and this influences how those genes are expressed.

If you have a child with a small genetic tendency toward risk-taking, you will probably treat him very differently, even just unconsciously, than you treat his more timid brother, and that difference in nurture will greatly amplify the difference in nature.

Young brains are designed to explore; old brains are designed to exploit.

Giving children the chance to intimately observe what many different people do is the best way to help them learn by looking. Giving children the chance to talk with many different people is the best way to help them learn by listening.

Pretending is closely related to another distinctively human ability, hypothetical or counterfactual thinking—that is, the ability to consider alternative ways that the world might be. And that, in turn, is central to our powerful human learning abilities.

Bayesianism is one of the most influential recent accounts of human learning. It’s named after the Reverend Thomas Bayes, an eighteenth-century theologian and pioneer of probability theory. Bayesians think that learning is much like scientific progress. We consider a range of different hypotheses, different pictures of how the world might work. Some of the hypotheses may be more likely than others, but none of them are absolutely sure to be true. When we say that we believe a hypothesis is true, what we really mean is that, right now, it’s the best guess we’ve got.

But what happens when children get older? For our children, of course, what happens is school. Schools become the caregivers of older children. In fact, psychology textbooks divide children up into preschool and school age as if school defined an essential biological difference. But “schooling” is actually only a little older than “parenting.” Schools have really only been around for a couple of hundred years, an eye-blink in human history. There is a parallel between the contemporary dilemmas of parenting and the equally ferocious dilemmas of schooling. Like parents, educators often have a scientifically inaccurate picture of learning and development. In fact, they share the same inaccurate picture. The misleading idea is that education is supposed to shape a child into a particular kind of adult. The remarkable spread of standardized testing is a particularly good example of this. A school’s job becomes creating children who will score well on standardized tests. At least the parenting project has sensible goals, such as turning your child into a happy, successful adult, even if there’s actually no clear prescription about how to accomplish those goals. But in practice, if not in principle, schooling ends up being aimed at arbitrary goals like the high test score, the A grade, and admission to the next school.

School requires even more extreme forms of focused attention than ordinary adult life. Many children are able to develop that kind of attention as they get older. But many more continue to have difficulty focusing even well into the school-age years. In particular, there is a close connection between the rise of schools and the development of attention deficit disorder. In the past two decades, the number of children who are diagnosed with attention deficit hyperactivity disorder has nearly doubled. One in five American boys receives a diagnosis by age seventeen. More than 70 percent of those who are diagnosed—millions of children—get drugs. Many people have suspected that there is a relationship between the ADHD explosion and the increasing emphasis on school performance. In the same two decades that saw a precipitous rise in diagnoses, more states began to evaluate schools and teachers based on test scores.

Steve Hinshaw and Richard Scheffler used a kind of “natural experiment” to test this. Different parts of the country introduced new education policies at different times. The researchers looked at the relationship between when a state introduced new policies that included high-stakes testing and the rate of ADHD diagnoses. Immediately after the policies were introduced, the diagnoses increased dramatically.

Socrates feared that reading and writing would undermine the kind of interactive, critical dialogue that was so important for reflective thought. You couldn’t talk back to a written text or ask it questions, and you were far too likely to think that something was true just because it was written down. Socrates also thought that writing would undermine the capacity for memory. In the ancient world, poets had developed an amazing ability to memorize thousands of lines of verse. Homer’s epics were orally transmitted from bard to bard through memory alone. But if you had a written copy of Homer’s Iliad, why would you go through all the trouble of memorizing it? Those impressive and hard-won memory skills would disappear. And, of course, Socrates was completely right. Reading is different from talking; we do tend to accept things just because they’re written down; and no one now knows the Iliad by heart. Indeed, reading reshaped many broader aspects of our culture and thought. The rise of literacy was connected to the emergence of the modern concepts of individualism and privacy, as well as the birth of Protestantism.

I’ve argued for a different picture of the relations between parents and children than the typical parenting picture. Caring for children is an absolutely fundamental, profoundly valuable part of the human project. But it isn’t carpentry, it isn’t a goal-directed enterprise aimed at shaping a child into a particular kind of adult. Instead, being a parent is like making a garden. It’s about providing a rich, stable, safe environment that allows many different kinds of flowers to bloom. It’s about producing a robust, flexible ecosystem that lets children themselves create many varied, unpredictable kinds of adult futures.

Classical philosophical approaches to morality don’t apply very well to parents and children. One important approach is John Stuart Mill’s utilitarianism. Utilitarians think that we should make decisions by calculating what will lead to “the greatest good for the greatest number.” Another philosophical theory, Kant’s “deontology,” claims that there are absolute and universal moral principles all of us should follow.

Think only of the difficult decision many of us make about whether to send our children to public schools or private ones. Sending them to public schools is arguably better for everyone, but sending them to private ones is arguably better for them. The utilitarian principle of the greatest good for the greatest number should make you choose public schools. But a genuinely moral impulse to do the best for your children might lead you to private ones instead.

The philosopher Isaiah Berlin argued for “value pluralism,” contra both Mill and Kant. We have a multiplicity of diverse ethical values and those values are often simply incompatible. There is no way to measure or weigh them against one another, no single value that trumps the others. Justice or mercy, altruism or autonomy, what Yeats called “perfection of the life or of the work”—these values can’t simply be weighed in some single objective scale. They can’t be measured against each other in a way that reveals the single best thing to do. And yet often, in real life, we have to choose between them. Berlin thought this made human life inescapably tragic, and he was right. But it is also what makes it rich and deep. Caring for children fits Berlin’s picture better than Mill’s or Kant’s.

I personally place a very high value, the highest, on caring for children. That is also true for many people in religious traditions, and is often articulated as the justification for opposing contraception or abortion. But I would argue that it’s just because caring for children is so valuable, so transformative, and so morally central that contraception and abortion have to be freely available. When I was forty I unintentionally got pregnant, and decided to have an abortion. It was a difficult decision but it was, thankfully, my decision to make.

Berlin argues that in these cases of conflicting values the best we can do is muddle through and make the best decision we can given the particular context. There is no decision that is best in some absolute way, and we need to accept both the guilt and regret, and the consolations, that follow from this.

Kind of a shallow exploration of a big question

Poverty, of course, is relative. Most of these still have a roof over their heads, electricity, and enough to eat. But I agree with her broader point: as a society we should rework the social safety net to be more pronatalist.

For some time, of course, the solution to this problem was to link resources for children specifically to marriage. This is the classic “nuclear family” picture. Fathers exclusively gather resources at jobs outside the home, and then share them with mothers who don’t work and exclusively look after children.

Ignores millennia of experience from various cultures? Nomadic tribes, Roman society, Confucian China, etc?

Until the nineteenth and even the twentieth century, most people lived and worked on farms, or in small local workshops or businesses. In 1830, 70 percent of American children lived in families with two parents who both worked on a farm; only 15 percent lived in a “nuclear family” with a working father and a stay-at-home mom. In 1930, only 30 percent of children had two farming parents, and 55 percent were in nuclear families. In the 1970s, the composition of families began to change again, and by 1989 less than a third of children were being raised in “nuclear families”—most children were raised by two parents who both worked or by a single parent who did. The percentage of children in one-parent families has continued to rise. By 2014 more than 30 percent of children were being raised by single parents.

On a farm, fathers and mothers, and families more generally, both work and care for children at the same time. It was only when home and work became separated that caregiving and work became separated as well. The flaws of the stay-at-home-mother solution are, of course, obvious by now. Women don’t get to experience the satisfactions of a career. It makes both women and children entirely dependent on fathers, and so profoundly vulnerable. And at the same time it isolates fathers from children and child-rearing. These flaws become especially apparent when divorce is widely available, as it should be for independent reasons. Children are especially likely to suffer if the only way to get them the resources they need is legal pressure on an embittered ex-husband. Since at least the 1970s the stay-at-home-mother model has crumbled. Partly, this is the result of the women’s movement, as women have rightly acted to gain more autonomy. It is also partly the result of economic forces. The average family income has been maintained only because women have gone to work. But other institutions haven’t emerged to take its place. This has had disastrous effects for children, particularly low-income children. In a foraging or farming society children had extended families to take care of them. In a traditional “nuclear family” they at least had a couple who were both dedicated to providing their resources and care. But in America right now children often end up depending on a single woma...

One set of solutions is well known and straightforward and has already been adopted in most civilized countries. This is to recognize that providing resources for children is the responsibility not only of biological mothers or even biological mothers and fathers but also of the community at large. There is no mystery about the moral advantages or the good practical consequences of policies such as universal prenatal care, home visitors and nurses, paid parental leave for men and women, universal and free preschool, and direct subsidies to parents. These policies really do lead to better outcomes for children as a whole—in fact, this is one ...

Preschool teachers and child-care workers tend to have something more like a “gardener” model: kindergarten, after all, was traditionally supposed to be just what it says—a garden for children that was different from school. But they are currently caught in a kind of pincer movement between two groups who assume the “carpenter” model. One group includes the parents who want to shape their three-year-olds into Harvard freshmen. The other group includes the policy makers who want to be sure of “outcomes” in the form of high test scores. Both these groups tend to take an instrumental carpenter’s approach to preschool. One New York mother sued her child’s preschool because the children were just playing instead of preparing for tests. You can even see this in the way the word “preschool” has replaced “child care” in these discussions, as if caring for children is less valuable than schooling them.

The way we treat the old is as much of a slow-motion invisible disaster as the way we treat the young. It’s hard to visit almost any “assisted living facility” or “retirement home,” no matter how dedicated and loving a child you are, and not feel your heart sink. We have spectacularly failed to provide the people we love with a good and dignified end, and we fear that this will be our fate as well.

The myth of Medicare and Social Security is that they are savings plans. In fact, of course, they are, as they should be, a way that the current generation collectively takes care of the past generation, just as parental leave or universal preschool should be ways that the current generation takes care of the future one.

We should start thinking about caring for the people we love, young or old, as an intrinsic value, a fundamental good that deserves both recognition and support. Just as we should provide paid parental leave, we should provide paid leave for caring for the elderly. And just as we should recognize formally that the demands of work should sometimes give way to the needs of our children, we should also recognize that they should sometimes give way to the needs of our parents.

One of the most striking examples of the value of play, for adults and for children, that I have ever seen comes in the great 1920s silent film Nanook of the North. The film traces the life of an Inuit hunter, Nanook, and his family as they struggle to survive in one of the harshest climates on earth, with only their own skill at hunting and gathering to sustain them. At one point in the movie Nanook makes a little toy sled for his adorable toddler son. Father and child have a grand romp in the snow, familiar to any Canadian parent like me. This may seem unremarkable, but think about it for a minute. Consider just how much of the gross domestic product of that barely subsisting family, in time and materials, went into building and playing with that sled. What Nanook knew, though, was that there was no better investment in a future life than the playful exploration of snow and ice. In the richest country on earth we still haven’t learned that lesson.

If I look back on the questions I started out with—Did I do the right thing when I raised my own children? How did I influence the way they turned out?—I am more convinced than ever that those questions were simply ill-conceived.