What this covers

Mathematician David Bessis and economist Russ Roberts examine the scientific foundations of the famous "twins reared apart" studies that have long been cited as evidence that intelligence is largely inherited. The conversation centers on why these studies—particularly Thomas Bouchard's landmark 1990 research—do not establish genetic determinism. Bessis walks through multiple methodological failures: Bouchard collected a control group of fraternal twins but never published it, the twins were not actually separated at birth but shared nine months in the womb plus roughly five months together, and adoptive placements were non-random, meaning the natural experiment was compromised from the start. The sample size—about 56 to 58 pairs collected globally—was too small to support sweeping claims about billions of people.

The speakers then examine what the data actually tells us. They discuss Turkheimer's Three Laws of Behavioral Genetics, which establish that genes matter but that a vast portion of human variation comes from non-shared environment—the unique experiences, friendships, chance encounters, and personal narratives that never appear in datasets. They show how common patterns in the twin data can be reinterpreted: the high correlation between identical twins raised together, for instance, does not rule out that family and individual circumstances shape outcomes in ways heritability statistics fail to capture. They also address why the dramatic coincidences in the "Jim twins" case (both marrying women named Linda then Betty) reveal shared social context rather than genetic influence, and explain why educational interventions that fail to scale are often misread as evidence that teaching doesn't matter—a determinist fallacy. Throughout, they distinguish between genes being one influence among many and genes being the whole story.

Sharpest takeaway

Bessis argues that the famous 'twins reared apart' studies do not establish genetic determinism of intelligence: the natural experiment is fatally compromised (shared womb, late separation, non-random adoptive placements, tiny samples, withheld control data), and Turkheimer's Third Law shows that idiosyncratic individual experience, not just genes or family, is decisive in who we become.

  • The Bouchard study collected a fraternal-twin control group but never published it, making the heritability estimate undecidable
  • Twins share nine months in the womb and ~five months together before separation, and adoptive families are non-random, so the 'separated at birth' premise is false
  • Turkheimer's Third Law attributes a huge share of variance to unique, non-shared individual experience that never appears in datasets

The argument · threads4 threads · 34 claims
0.85

Heritability figures claiming high genetic determination of IQ are scientifically imprecise and misleadingly presented as settled fact.

2 pointscentrality 5/5
  • In economics, researchers can publish papers where a scatterplot of data points looks like a random cloud, yet fitting a line through it sometimes yields a statistically significant result that gets trumpeted as an important finding.

    “there's a scatterplot of the data points and it's a cloud, and it's a circle... but, if you put a line through it, sometimes you can get a significant result; and they trumpet that as if it's an important finding.”

  • Statements like 'IQ is 80% genetic' are treated as precise and scientific but are not: IQ is not a precise measure of intelligence, and a heritability figure of 30% or 50% still leaves substantial room for individual agency and self-determination, whereas 80% moves closer to determinism.

    “these numbers get thrown around like they're scientific--you know, that IQ is 80% genetic. As if that means something really exact. It doesn't in so many ways.”

0.71

Individual inner narrative and personal experience are decisive but unmeasurable, leaving heritability claims incomplete and omitting crucial variables.

2 pointscentrality 4/5
  • People care intensely about nature versus nurture even though, behaviorally, the precise heritability figure barely changes how anyone lives their life; the passion arises because the question strikes at the heart of how we see our humanity and whether our sense of our own journey is an illusion, making the debate emotionally unbearable rather than practically decisive.

    “nobody lives their life saying, 'Well, if it's only 50%, I'm going to do X, but if it's 80%, oh my gosh, then I won't.'”

  • The inner journey—one's inner narrative and personal experiences, which Turkheimer's Third Law identifies as decisive—will never appear in any dataset in our lifetime; statistically these are 'omitted variables' that are sometimes incredibly important, a reality we dislike but must accept.

    “The thing that's happening in your head, your inner journey, your inner narrative, your inner experiences, is never going to be in the dataset... The fancy name for this in statistics is there are 'omitted variables'.”

0.70

Great mathematicians achieve excellence through learnable methods, hard work, and focus, not predetermined genetic genius.

1 pointcentrality 5/5
  • Great mathematicians like Einstein, Descartes, and Grothendieck are not uniquely born geniuses; almost anyone can do math if they approach it correctly, and greatness comes from a powerful learnable method combined with dogged focus and hard work rather than predetermined genetics.

    “The great mathematicians are not uniquely genius-like. They just have an approach that's very powerful that can be learned by normal human beings.”

0.56

Twin studies claiming to isolate genetic contribution often fail because separated twins are placed in too-similar environments.

1 pointcentrality 4/5
  • Some 'separated' twins were assigned not just to similar families but to related households (such as an uncle and brothers in nearby neighborhoods), so their nurture was extremely similar; this means we would expect similar IQs from environment alone, which is exactly why the fraternal-twin control is needed to isolate the genetic contribution.

    “some of them were assigned to families--the same family, just not the same house. It's, like, an uncle and two brothers: two different houses in nearby neighborhoods.”