Racial Hereditarianism in Science: Trees or Forest?

• In Event: Mathematical Sociology Section Flash Talks
  In Refereed Roundtable (60 minute): Table 9. Racial Hereditarianism in Science: Trees or Forest?

Abstract

How do scientific fields police ethically problematic work, and do their efforts matter beyond the academy? We investigate this question in intelligence psychology, where controversial hereditarian research on group differences in intelligence, crime, and behavior has appeared in mass shooting manifestos, viral social media posts, and mainstream political discourse. Recent attention has focused on sanctioning authors affiliated with Mankind Quarterly, a journal associated with scientific racism. But individual sanctions are only one tool available to scientific communities. Fields also police through retraction, non-citation and, more recently, through bioethics standards that articulate norms for what types of research is inappropriate.

We build an LLM classifier trained on expert bioethical assessments to identify potentially problematic abstracts across approximately 70,000 papers by intelligence psychologists published between 1980 and 2020. We compare the sanctioned Mankind Quarterly network against a broader pool of classifier-flagged papers, tracing both through authorship networks, citation trajectories across 2.4 million citing papers, and public reception across 15,000 Reddit posts.

The Mankind Quarterly network accounts for only about half of the ethically concerning work in the field. The remainder is produced by mainstream researchers indistinguishable from their peers on measures of productivity and scientific status. The field's governance tools have been deployed unevenly: retraction targets Mankind Quarterly authors specifically, while the broader pool of concerning papers has never been retracted. Citation suppression extends more broadly but is focused heavily on race, while structurally similar work on gender and personality can still be highly cited. On Reddit, both paper pools are dramatically overrepresented in toxic rhetoric, with little relationship to academic citation counts. Papers spread through copy-pasted bibliographies that treat scientific publications as interchangeable with non-academic sources. These findings suggest that governance in intelligence psychology isn't absent but misdirected in nuanced ways. AI tools provide new opportunities to help scientific communities identify where self-governance succeeds, where it fails, and where intervention is most needed.

Authors

• Bernard Joseph Koch, University of Chicago

• Kushan Dasgupta, University of Louisville

• BJ Moses-Rosenthal, University of Chicago

• Aaron Panofsky, University of California-Los Angeles