What this covers

Paige Harden joins Sean Carroll to defend a measured position: DNA measurably affects life outcomes, yet this fact has been widely weaponized to justify discrimination, creating a reluctance to study genetics at all. Harden argues this avoidance is counterproductive. She contends that polygenic scores—summary statistics predicting traits like college completion—carry roughly the same predictive power as family income, and that understanding genetic variation is essential to designing interventions that actually work. The episode works through why genetics is a uniquely useful tool for causal inference: DNA does not change over a person's lifetime and is randomly assigned conditional on parents' genetics, properties that make it almost impossible to find in observational social science. Yet Harden is careful to distinguish what the science shows from what it does not. Genes do not work in isolation, traits are not determined by single variants but by thousands of small probabilistic effects, and a genetic correlation with an outcome tells us only which embodied characteristics a given society happens to reward.

The conversation spans several contentious turns. Harden explains why antagonistic pleiotropy—the fact that variants can raise risk for both valued and disvalued traits—makes eugenics conceptually confused. She unpacks treatment effect heterogeneity, showing how genetic information can reveal which children benefit most from an intervention and which benefit not at all, a pattern masked by simple averages. She distinguishes merit as earned from merit as instrumental fitness for a role, arguing the confusion between these drives much of the heated rhetoric. Drawing on Elizabeth Anderson, she articulates relational equality—equalizing people's ability to participate in common life—as a framework for thinking about genetic diversity without flattening human difference. Throughout, Harden addresses the fraught history of misuse while making the case that the actual scientific challenge is methodological rigor, not silence.

Sharpest takeaway

Harden argues that DNA measurably affects life outcomes through small, probabilistic, polygenic effects, and that refusing to study these effects perpetuates the myth of meritocracy; taking genetic luck seriously is a necessary tool for building a more equitable society organized around relational equality rather than identical outcomes.

  • Polygenic scores predict outcomes like college completion at roughly the same effect size (~11-15%) as family income, providing non-redundant information.
  • Genetic variation, conditional on parents, is randomly assigned and temporally inert, making it a uniquely useful causal tool for identifying which environments help which people.
  • Genes do not map cleanly onto 'good' or 'bad' traits, so a just society should accommodate genetic diversity like a meadow rather than enforce a monoculture.

The argument · threads3 threads · 39 claims
0.71

Emotional intensity on divisive topics impairs cognition; noticing and processing that feeling enables clearer thought.

1 pointcentrality 4/5
  • When emotional valence rises sharply on a contentious topic, it lowers people's cognitive abilities, so we should notice and savor that feeling in order to move beyond it and think clearly.

    “it lowers your cognitive abilities when your emotional valence goes way up like that.”

0.71

Genetic inequality from birth raises fundamental questions about justice since people cannot choose their DNA yet it shapes who they become.

1 pointcentrality 4/5
  • We do not choose our DNA and should not get credit or blame for it, even though it affects who we grow up to be; this raises the question of how to live justly in a world where people receive unequal talents from the start.

    “we don't choose our DNA. Our DNA affects who we grow up to be, but we don't get credit or blame for it, we shouldn't anyway.”

0.57

People will pursue genetic enhancement of desired traits despite biological complexity, as shown by consumer genetic testing trends.

1 pointcentrality 3/5
  • Even though biologists rightly stress that engineering desired babies is far more complicated than the public imagines, people will try anyway, because while no one wants a monoculture, when asked whether they want their babies to be smart and beautiful they will say yes—as already evidenced by direct-to-consumer genetic testing marketing matching DNA to wine or Spotify playlists.

    “if you ask them, do they want their babies to be smart and beautiful, they're gonna say yes.”