Steven Pinker · Human nature and the blank slate
Monterey, California, the end of February 2003. Steven Pinker is on the TED stage reading out what colleagues told him after they saw his manuscript:
"Better get a security camera for your house." … "Tell your publisher not to list your hometown in your author bio." "Do you have tenure?"— Steven Pinker, TED2003
The room laughs. The book was The Blank Slate, published the previous October, and its target was one of the twentieth century's most influential ideas: that the human mind starts out empty and gets all its structure from upbringing, culture and experience. It waded into race, gender and violence. Yet of the 80-odd reviews, Pinker said, about 90 percent of the heat went to just two topics. He asked the audience to guess, then answered: "the arts and parenting."
The first half of the talk is evidence. Children arrive with temperaments of their own. A child and a house pet hear the same speech, and only the child learns a language. Some 6,000 cultures share a long list of behaviors. Identical twins separated at birth turn out astonishingly alike when they finally meet as adults. He told the story of one pair: one raised Catholic in Germany, the other in a Jewish family in Trinidad. They walked into a Minnesota lab wearing the same navy blue shirts with epaulettes.
So why was the blank slate so appealing? Politics, he said. If everyone starts blank, everyone is equal, because zero equals zero. His reply: fairness is not sameness. When Jefferson wrote that all men are created equal, he did not mean all men are clones.
The second half is where the trouble was. Elite art and criticism in the twentieth century, he argued, threw out beauty and clarity, so audiences walked away. Then parenting. Most parenting studies never ruled out heredity: talkative parents raise articulate children, but maybe they just passed on their genes. The studies that do control for genes found something hard to sit still for. Twins raised apart are no less alike than twins raised together, and adopted siblings who grow up in one home with no shared genes end up not similar at all. In his words:
Everything that happens to you in a given home over all of those years appears to leave no permanent stamp on your personality or intellect.— Steven Pinker, TED2003
What shapes children, he said, is genes, the culture of their peers, and a great deal of plain chance.
What happened next? Partly held. The "not a blank slate" half has been confirmed. A 2015 meta-analysis pooling fifty years of research on more than 14 million twin pairs put the average heritability of human traits at 49 percent, and for most traits found little sign of shared family environment. But "no permanent stamp" went too far. On stage Pinker said we could not know whether parenting findings hold until the studies were redone with adoptive families. In 2015 Sweden did exactly that: of two full brothers, the one who was adopted away scored on average 4.4 IQ points higher at age 18 to 20. Among poor American families, the home turned out to matter far more for seven-year-olds' IQ than Pinker allowed. On average, family matters less than parents assume. But "no stamp" holds only where homes don't differ much, and many twin samples are drawn from exactly that kind of population.
Halfway through, Pinker puts up a list: the arts, cloning, crime, free will, gender differences, God, infanticide, Nazism, parenting, race, rape, religion… every minefield his new book had walked into. His own words for them: "hot buttons, hot zones, Chernobyls, third rails."
Then he reads out what colleagues said after seeing the first draft. "Better get a security camera for your house." "Don't expect to get any more awards, job offers or positions in scholarly societies." "Tell your publisher not to list your hometown in your author bio." The last one is four words: "Do you have tenure?" The room cracks up.
Then he turns the joke over. Half a year out, 80-odd reviews, and by his estimate 90 percent of the reaction went to two topics. He asks the audience to guess:
It's not violence and war, it's not race, it's not gender, it's not Marxism, it's not Nazism. They are: the arts and parenting.— Steven Pinker, TED2003
More laughter. That reveal is also the shape of the whole talk: the first half is about why we are not blank slates, the second about how the two most harmless-looking topics in the book turned out to be the sorest.
Pinker was then a cognitive psychologist at MIT who had made his name in language acquisition by studying something that sounds trivial: how children learn past-tense verbs. His popular books kept getting bolder. The Language Instinct (1994) argued that language is innate; How the Mind Works (1997) used evolution to explain the whole mind. His first sentence sets the timeline: a year earlier he had come to TED to talk about a book he was still finishing, and now he was back to talk about the fights it had started.
The book was The Blank Slate: The Modern Denial of Human Nature, published in October 2002. The science in it was not new. Behavioral genetics had been measuring heredity with twin and adoption data for decades, and the Minnesota study of twins reared apart had begun in 1979. Four years earlier Judith Rich Harris, a textbook writer without a PhD, had published The Nurture Assumption, arguing that in the long run peers shape children more than parents do; Pinker wrote its foreword. What was new was where Pinker took these findings: to a political verdict that twentieth-century intellectuals had denied human nature with good intentions, and that the denial had costs. On stage he joked that "I haven't been exiled from the city of Cambridge." MIT is in Cambridge.
Pinker opens with three quotations standing in for the slate he wants to break, among them the line from the late Stephen Jay Gould: "The human brain is capable of a full range of behaviors and predisposed to none." The rebuttal starts with common sense: a child and a house pet hear the same speech, and only the child learns a language. Then science. The anthropologist Donald Brown listed what the world's 6,000-odd cultures have in common, from aesthetics and affection down to weaning, weapons and weather. And in brain scans by the neuroscientist Paul Thompson, two people paired at random show no correlation in how their gray matter is distributed; fraternal twins show large correlated regions; identical twins show "huge areas of cortex" with massive correlations.
To show that the resemblance goes beyond anatomy, he puts up a Charles Addams cartoon: twin inventors separated at birth meet by chance in a patent attorney's waiting room, identical contraptions in their laps. Then his favorite real case: one twin raised Catholic in a Nazi family in Germany, the other in a Jewish family in Trinidad. Both wore rubber bands on their wrists, both flushed the toilet before using it as well as after, and both liked to sneeze in crowded elevators to watch people jump. He grants that the story sounds too good to be true, then says batteries of psychological tests give the same result.
The turn comes with a question: if common sense and data both cut against it, why was the blank slate so attractive? His answer is politics. A blank slate means equality, "because zero equals zero equals zero." It also promises that humanity can be perfected through social engineering. Both, he says, are non sequiturs:
first of all, the concept of fairness is not the same as the concept of sameness.— Steven Pinker, TED2003
Jefferson's "all men are created equal" did not mean all men are clones. It meant equal rights, and that each person should be treated as an individual rather than prejudged by the statistics of a group.
Next come the minefield list and the colleagues' warnings. In passing he tells of two psychologists who published data suggesting left-handers die younger, then drew so many furious letters and death threats that they were afraid to open their mail. He adds, carefully, that it is not actually clear whether the finding was accurate.
Then the two hot buttons. The arts: art is a human universal, and lines of poetry all over the world run close to three seconds. Yet critics from T. S. Eliot onward have mourned the arts' decline. Judged by books, albums and opera tickets, Pinker says, the arts have never flourished more. What has declined is elite art since the 1930s, literary criticism and university humanities, and that decline coincided with the denial of human nature. Modernism gave us visual art without beauty, literature without plot, music without melody, and "criticism without clarity." The room laughs. He reads a single sentence by the Berkeley professor Judith Butler as evidence, then notes that people are staying away from elite art and criticism: "What a puzzle -- I wonder why."
Parenting comes last. Most parenting studies are useless, he says: they see what parents do and how children turn out, assume the first caused the second, and never rule out that parents passed on genes too.
Until the studies are redone with adoptive children, who provide an environment but not genes to their kids, we have no way of knowing whether these conclusions are valid.— Steven Pinker, TED2003
The genetically controlled studies say this: identical twins separated at birth are no less similar than those raised together, and adopted siblings reared together without shared genes end up "not similar at all." So in the long run children are shaped not by their parents but partly by genes, partly by culture, especially their own peer group, and "to a very large extent, larger than most people are prepared to acknowledge, by chance": chance in how the brain gets wired in the womb, chance in what happens over a life.
He closes on the conference theme of choices. The sciences of human nature will keep upsetting dogmas and political beliefs, and we must choose whether to treat some topics as taboo or to explore them honestly. He gives the last word to Chekhov:
Man will become better when you show him what he is like.— Steven Pinker quoting Chekhov, TED2003
The brain-scan figure came from 20 pairs of twins. Pinker described Thompson's study as "a large sample of pairs of people." The original 2001 paper in Nature Neuroscience used 10 identical and 10 fraternal pairs. Its finding is strong: in two language areas and frontal cortex, identical twins' gray matter matched with r² above 0.8. But the paper itself calls its link between frontal gray matter and general intelligence "preliminary," and Thompson told Scientific American, "But this is quite a mild correlation," and "You can't predict an individual's IQ from a brain scan."
The twins' story is more tangled than the stage version. They were Oskar Stöhr and Jack Yufe, born in Trinidad in 1933. Their parents divorced when they were six months old, and Oskar went to Germany with their Catholic mother. According to the Jewish Telegraphic Agency's 2015 obituary, he hid his Jewish ancestry and joined the Hitler Youth "out of self preservation." "A Nazi family" flattens that history. More important, Pinker said the uncanny resemblance shows up in "every pair" ever studied, while the Minnesota team's 1990 summary in Science reports averages. More than 100 sets of reared-apart twins or triplets each went through a week of testing; about 70 percent of the variance in IQ was associated with genetic variation; and on personality, interests and social attitudes, identical twins reared apart were "about as similar as are monozygotic twins reared together." That matches Pinker. What he left out are the abstract's last two sentences: genes probably act largely indirectly, by shaping the environment a child actually experiences, and the evidence "does not detract from the value or importance of parenting, education, and other propaedeutic interventions." Same data; a gentler conclusion from the people who collected it.
"No permanent stamp" is a statistical claim. Behavioral genetics splits population variation into genes, shared environment and nonshared environment, and Pinker's claim is that the second slice is small. In 2000 the psychologist Eric Turkheimer distilled the field's consensus into "three laws": all human behavioral traits are heritable; the effect of being raised in the same family is smaller than the effect of genes; and a substantial part of the variation in complex traits is explained by neither. Pinker's line is close to a plain-language Second Law. But Turkheimer's abstract says flatly that some behavior-genetic partisans had concluded family environment is not an important cause of development, and that this is incorrect. The better reading, he argues, is that the part of the family environment siblings share doesn't matter, while each child's own environment, peers, and the parenting they don't share still do.
The chance claim, oddly, has the hardest data. Turkheimer cites a review of 43 studies: the nonshared component accounts for upwards of 50 percent of the variance in outcomes, yet the environmental differences between siblings that researchers actually measured explained a median of less than 2 percent. The rest, he concludes, is unsystematic chance: "The gloomy prospect is true."
An offhand example later fell apart. The left-hander study was Diane Halpern and Stanley Coren's 1991 paper in the New England Journal of Medicine, which reported that left-handers die nine years younger. A 2023 paper modeled every US death in 1989 and found that cultural pressure had kept older generations from reporting left-handedness; that shift in reported rates over time fully explains the nine-year gap. In hindsight, Pinker's "it's not clear" was right.
PARTLY HELD · as of 2026-10
Half a century of twin data has confirmed "not a blank slate," and the finding that shared family environment is small on average mostly holds. But "no permanent stamp" is true only where homes are broadly similar. Add poverty, adoption or schooling, and the claim is too broad.
2003: a straw man? The day before the talk, the February 27 issue of The New York Review of Books ran a review by the biologist H. Allen Orr. Pinker builds a straw man and torches it, Orr wrote: "the Blank Slate he assails is something few thinking people believe in." On evolutionary psychology: "A Darwinian story is a story." And: "Pinker seems never to have met an adaptive tale he doesn't like." These were aimed at the book, but "straw man" applies just as well to the three quotations that opened the talk.
2003: in poor homes, family matters far more. Turkheimer's team analyzed IQ scores of seven-year-old American twins in Psychological Science, a sample in which many families lived near or below the poverty line. In impoverished families, shared environment accounted for 60 percent of the variance in IQ and genes for close to zero; in affluent families the result was almost exactly reversed. A 2016 meta-analysis by Elliot Tucker-Drob and Timothy Bates added another layer: the effect is real in US studies but zero or reversed in Western Europe and Australia, where access to education and health care is more even. Whether the home leaves a stamp depends on which society you are born into.
2015: two big studies, one on each side. On one side, Tinca Polderman and colleagues published a meta-analysis in Nature Genetics covering virtually every published twin study: 2,748 papers, 17,804 traits, 14,558,903 partly overlapping twin pairs. Average heritability was 49 percent; for 69 percent of traits, twin resemblance fit a model with additive genetic effects alone; and the data were "inconsistent with substantial influences from shared environment." It is the strongest support yet for the first half of the talk. On the other side, Kenneth Kendler, Turkheimer and colleagues in PNAS ran exactly the adoption design Pinker had asked for, using Swedish records: 436 pairs of full brothers, one raised by a biological parent and one adopted away, IQ measured at the military conscription exam at 18 to 20. The adopted brother scored 4.41 points higher on average; in 2,341 half-brother pairs the gain replicated at 3.18 points; and each step up in the rearing parents' education added about 1.7 points. The home left a measurable mark on intellect, lasting at least into early adulthood.
2018: parents' genes acting through the environment. Augustine Kong and colleagues in Iceland reported "genetic nurture" in Science: the half of the parents' genetic variants that is not passed to a child still predicts that child's educational attainment, with an effect about 29.9 percent as large as the transmitted half. Genes that were never inherited can only work through the environment the parents create. Nurture is real; some of it just gets booked to genes.
Roughly, the consensus today: nobody seriously defends the blank slate any more, and among middle-class populations in rich countries, growing up in the same home does not make siblings' personalities more alike in adulthood. That much largely stands. But "no permanent stamp" overreaches: for IQ and education, once homes differ enough, the stamp shows up clearly. The arts half has stayed an assertion for twenty years; nobody has turned it into a testable study.