Last year, Henrik Karlsson wrote an essay about the childhoods of exceptional people. Like Erik Hoel, who had previously written about aristocratic tutoring, Karlsson proposed that the intensive educations of earlier times were responsible—at least in part—for producing the remarkable minds of people like Bertrand Russell, Virginia Woolf, Mozart, and John von Neumann. Erik Hoel believes that the decline of aristocratic tutoring—the one-on-one education many geniuses in the past received—is why we lack geniuses today.
These essays prompted responses from Scott Alexander, Ian Leslie, and others, debating how effective these educations really are—and whether they make the children involved unhappy. Ian Leslie warned that there is a trade off between brilliance and happiness: exceptional childhoods make miserable adults. This is not an uncommon view.
Scott Alexander says that if there is a lack of genius now, the reason is more likely to be a combination of: ideas are harder to find, more researchers work on each project, and cultural norms emphasise the team around the genius. Scott identifies Geoffrey Hinton—who left Google to warn about AI risks—as a true modern genius, just not a genius many people will be familiar with.
Interestingly, Hinton makes a good case study for the question of whether exceptional childhoods are necessary to create geniuses—and whether that intensity leaves the children with lifelong unhappiness.
Autonomy creates success
Geoffrey Hinton had a highly autonomous career: he switched his undergraduate degree several times, didn’t get into AI until his PhD, and struggled to get funding early on. He had been interested in neural networks (now the basis of AI technology) and the way brains work since school. To turn that into a workable career, he had to start with an early, open-ended “explore” phase: his academic career gave him the freedom to investigate a range of ideas. He later moved to an “exploit” phase, applying those ideas in practical ways, when he spent a decade working at Google.
Hinton’s ability to follow his own interests, and explore ideas about AI, was enabled by the fact that he worked in a relatively small and new area of academia. Importantly, Hinton chose to work at the Canadian Institute for Advanced Research (known as CIFAR). CIFAR investigates ideas other places won’t, and offered Hinton autonomy to work on ideas, which, until recently, were dismissed by many academics.
When he later went to Google, Hinton was able to start implementing his ideas in voice recognition, image tagging, and other applications. He started by working in an area which was very little understood—how exactly does the brain work?—and ended up reaching the pinnacle of technological development.
First he needed an environment where he could explore different ideas: then he needed an environment where he could exploit the knowledge he had found and put it to use. Both the explore and the exploit phase are important to Hinton’s flourishing.
What links both sides of Hinton’s career is autonomy—and that goes all the way back to his exceptional childhood.
Geoffrey Hinton’s exceptional childhood
Hinton grew up in the sort of expectational milieu that Karlsson describes. His father was an academic who expected rigorous intellectual standards; the family tree is full of heavyweight scientists. The house was full of animals, which Hinton was free to explore as a child, including vipers which once nearly killed him.
It was also the sort of environment that seems destined to make someone unhappy: his mother told him to become an academic or become a failure.
After he graduated, Hinton was at a loss about how to proceed. He had done an undergraduate degree in experimental psychology, then worked odd jobs before starting a PhD in artificial intelligence. He became depressed and felt indifferent to his research. At a self-actualisation session one weekend, a revelation struck. Toronto Life reports:
On the last day, each attendee had to announce what they really, really wanted in life. People were saying they really wanted to be loved. “Primal and uninhibited things,” Hinton recalls. He was freezing up and didn’t know what to say. As they went around the group, shouting their secret desires, Hinton surprised himself: “What I really want is a PhD!” he bellowed. The declaration reignited his passion for neural networks research.
And so he went off on the long road to becoming Geoff Hinton, the man Scott Alexander expects to be one day known as the Darwin of artificial intelligence.
Hinton seems to prove both Scott Alexander’s and Henrik Karlsson’s ideas about genius: he had an exceptional childhood and he worked in a field where ideas aren’t so hard to find, because the subject is new. What ties these two theories together is autonomy: Hinton was free to explore as a child, an undergraduate, and an academic at CIFAR, so that when the opportunity at Google came, he was ready to exploit.
What cost excellence?
Hinton, though, also serves as a warning. As Ian Leslie would expect, he has been depressed for much of his life.
“For a long time,” he says, “I felt I wasn’t—well, I finally made it, and it’s a huge relief.”
This is the strongest criticism made of the Karlsson/Hoel system. Expecting children to be exceptional makes them unhappy. Among people interested in education reform, Bloom’s 2-sigma problem is often cited as an example of what could be achieved if more children were educated intensively. But there is a worry that this pressure would lead to emotional problems. Examples like Hinton seem to bear out the idea that pushing children to achieve makes them unhappy. “Be an academic or be a failure” isn’t the sort of parenting that leads to happy, well-adjusted children.
But according to a 2016 literature review, intellectually gifted people are more likely to have “decreased levels of psychopathology” and “high self-esteem and well-developed social skills.” Being a prodigy is associated with being happier and better adjusted. There is often a combination of freedom and intensity in exceptional educations, as noted by Karlsson:
A common theme in the biographies is that the area of study which would eventually give them fame came to them almost like a wild hallucination induced by overdosing on boredom. They would be overcome by an obsession arising from within.
In these educations, there are clear expectations about what you ought to achieve, but plenty of freedom about how you go about achieving it. Not everyone who is brought up to autonomy has a domineering father and demanding mother like Hinton—it is possible to get a happier balance of control and motivation.
We should also remember that depression is about 50% heritable. Sometimes people like Hinton are depressed for reasons unrelated, or less related than we think, to their childhoods and education. According to the psychologist Jay Belsky, the same childhood conditions will affect different people in very different ways: what causes trauma to one child goes almost unnoticed by another.
Exceptional educations do not mean we have to choose between brilliance and happiness. The key is autonomy.
Personality and will are crucial factors in becoming an innovator or revolutionizer of a domain. Creators have a desire to shake things up. They are restless, rebellious, and dissatisfied with the status quo. They are courageous and independent. They are able to manage multiple related projects at the same time, engaging in what Gruber calls a “network of enterprise”. For these two reasons, we should never expect a prodigy to go on to become a creator. The ones who do make this transition are the exception, not the rule.
As Adam Grant said, “practice makes perfect, but it doesn’t make new.” That’s a good thing, of course: we all benefit from having superb minds in unoriginal jobs. The very best doctors, engineers, lawyers are essential to the way society functions. But there is room to invoke more originality in the best minds.
And originality is the result of autonomy. Exceptional educations should emphasise freedom. If we can enable more autonomy, we can have high accomplishment and happiness.
How to raise a (happy) genius
This was originally posted on my Subtack: https://www.commonreader.co.uk/p/how-to-raise-a-happy-genius
Where does genius come from?
Last year, Henrik Karlsson wrote an essay about the childhoods of exceptional people. Like Erik Hoel, who had previously written about aristocratic tutoring, Karlsson proposed that the intensive educations of earlier times were responsible—at least in part—for producing the remarkable minds of people like Bertrand Russell, Virginia Woolf, Mozart, and John von Neumann. Erik Hoel believes that the decline of aristocratic tutoring—the one-on-one education many geniuses in the past received—is why we lack geniuses today.
These essays prompted responses from Scott Alexander, Ian Leslie, and others, debating how effective these educations really are—and whether they make the children involved unhappy. Ian Leslie warned that there is a trade off between brilliance and happiness: exceptional childhoods make miserable adults. This is not an uncommon view.
Scott Alexander says that if there is a lack of genius now, the reason is more likely to be a combination of: ideas are harder to find, more researchers work on each project, and cultural norms emphasise the team around the genius. Scott identifies Geoffrey Hinton—who left Google to warn about AI risks—as a true modern genius, just not a genius many people will be familiar with.
Interestingly, Hinton makes a good case study for the question of whether exceptional childhoods are necessary to create geniuses—and whether that intensity leaves the children with lifelong unhappiness.
Autonomy creates success
Geoffrey Hinton had a highly autonomous career: he switched his undergraduate degree several times, didn’t get into AI until his PhD, and struggled to get funding early on. He had been interested in neural networks (now the basis of AI technology) and the way brains work since school. To turn that into a workable career, he had to start with an early, open-ended “explore” phase: his academic career gave him the freedom to investigate a range of ideas. He later moved to an “exploit” phase, applying those ideas in practical ways, when he spent a decade working at Google.
Hinton’s ability to follow his own interests, and explore ideas about AI, was enabled by the fact that he worked in a relatively small and new area of academia. Importantly, Hinton chose to work at the Canadian Institute for Advanced Research (known as CIFAR). CIFAR investigates ideas other places won’t, and offered Hinton autonomy to work on ideas, which, until recently, were dismissed by many academics.
When he later went to Google, Hinton was able to start implementing his ideas in voice recognition, image tagging, and other applications. He started by working in an area which was very little understood—how exactly does the brain work?—and ended up reaching the pinnacle of technological development.
First he needed an environment where he could explore different ideas: then he needed an environment where he could exploit the knowledge he had found and put it to use. Both the explore and the exploit phase are important to Hinton’s flourishing.
What links both sides of Hinton’s career is autonomy—and that goes all the way back to his exceptional childhood.
Geoffrey Hinton’s exceptional childhood
Hinton grew up in the sort of expectational milieu that Karlsson describes. His father was an academic who expected rigorous intellectual standards; the family tree is full of heavyweight scientists. The house was full of animals, which Hinton was free to explore as a child, including vipers which once nearly killed him.
It was also the sort of environment that seems destined to make someone unhappy: his mother told him to become an academic or become a failure.
After he graduated, Hinton was at a loss about how to proceed. He had done an undergraduate degree in experimental psychology, then worked odd jobs before starting a PhD in artificial intelligence. He became depressed and felt indifferent to his research. At a self-actualisation session one weekend, a revelation struck. Toronto Life reports:
And so he went off on the long road to becoming Geoff Hinton, the man Scott Alexander expects to be one day known as the Darwin of artificial intelligence.
Hinton seems to prove both Scott Alexander’s and Henrik Karlsson’s ideas about genius: he had an exceptional childhood and he worked in a field where ideas aren’t so hard to find, because the subject is new. What ties these two theories together is autonomy: Hinton was free to explore as a child, an undergraduate, and an academic at CIFAR, so that when the opportunity at Google came, he was ready to exploit.
What cost excellence?
Hinton, though, also serves as a warning. As Ian Leslie would expect, he has been depressed for much of his life.
This is the strongest criticism made of the Karlsson/Hoel system. Expecting children to be exceptional makes them unhappy. Among people interested in education reform, Bloom’s 2-sigma problem is often cited as an example of what could be achieved if more children were educated intensively. But there is a worry that this pressure would lead to emotional problems. Examples like Hinton seem to bear out the idea that pushing children to achieve makes them unhappy. “Be an academic or be a failure” isn’t the sort of parenting that leads to happy, well-adjusted children.
But according to a 2016 literature review, intellectually gifted people are more likely to have “decreased levels of psychopathology” and “high self-esteem and well-developed social skills.” Being a prodigy is associated with being happier and better adjusted. There is often a combination of freedom and intensity in exceptional educations, as noted by Karlsson:
In these educations, there are clear expectations about what you ought to achieve, but plenty of freedom about how you go about achieving it. Not everyone who is brought up to autonomy has a domineering father and demanding mother like Hinton—it is possible to get a happier balance of control and motivation.
A study of Chinese adolescents found that those with high academic achievement and parents who gave them supportive autonomy (parents who were not controlling) were at lower risk of developing depression. Other studies have found that a combination of care and autonomy lowers depression risks. And the pressure to be a perfectionist might bad for your mental health, a neurosis which gets worse as you age: excessive parental expectations correlate with higher levels of anxiety. This is consistent with the other findings that what damages children is not high standards, but being too controlling.
We should also remember that depression is about 50% heritable. Sometimes people like Hinton are depressed for reasons unrelated, or less related than we think, to their childhoods and education. According to the psychologist Jay Belsky, the same childhood conditions will affect different people in very different ways: what causes trauma to one child goes almost unnoticed by another.
Exceptional educations do not mean we have to choose between brilliance and happiness. The key is autonomy.
Unoriginal prodigies
This matters today because few child prodigies end up being highly creative or original. Mastering a domain is one thing: changing it, quite another. Psychologist Ellen Winner summarised the research findings like this:
As Adam Grant said, “practice makes perfect, but it doesn’t make new.” That’s a good thing, of course: we all benefit from having superb minds in unoriginal jobs. The very best doctors, engineers, lawyers are essential to the way society functions. But there is room to invoke more originality in the best minds.
And originality is the result of autonomy. Exceptional educations should emphasise freedom. If we can enable more autonomy, we can have high accomplishment and happiness.