EconTalk · Russ Roberts

Nature, Nurture, and Identical Twins (with David Bessis)

January 19, 2026·1 hr 5 min·5 clips

David Bessis admits he was deeply disturbed by viral twin study claims, fearing his entire mathematical journey might be a delusion.

The title does real work. Bessis's essay, "Twins Reared Apart Do Not Exist, The Shaky Science of Genetic Determinism," questions the confidence built around separated twin cases as proof of genetic determinism. Roberts introduces Bessis as a mathematician and author, then sets up the conversation through the Substack essay. The lure is obvious. Identical twins raised apart seem to give economists the clean thing they love: a natural experiment where nature appears to have handled the setup. Bessis does not mock that hope. He understands why Thomas Bouchard fell for the idea, because it seems to offer a clean answer to nature versus nurture. Then he makes the inference do some work. These are not ordinary observations from ordinary family life. They are a tiny group created by separations so unusual that most people would never design them on purpose. Fifty something pairs cannot carry the whole argument. From that odd setup, Bessis says, researchers and audiences try to make claims about cognition that are meant to apply to billions of people. Patterns make the trap more tempting. Roberts notes that listeners already know how much human beings love patterns, including the ones that may be fooling us. Three Identical Strangers makes the danger easier to feel. The separated triplets look alike, share coincidences, and make a messy question feel cleaner than it is. The conversation stays careful, though. Bessis is willing to call the idea beautiful before asking whether beauty, a small sample, and strange circumstances have been treated like settled evidence. By the end, the discussion has moved from twins to brains, machines, and inquiry itself, with Bessis warning that ideology here can turn into censorship.

Listen to the show on

Apple Podcasts Spotify

Play clips from this episode More from EconTalk