https://razib.substack.com

This is where you will find all the podcasts from Razib Khan's Substack and original video content.

On this episode of Unsupervised Learning Razib talks to Dr. Cory J. Clark, a behavioral scientist and executive director of the Adversarial Collaboration Project at the University of Pennsylvania. Clark got her Ph.D. in social psychology at UC Irvine, but her interests have broadened over her career as is clear in a diverse oeuvre.

First, Razib and Clark talk about the culture of self-censorship within science due to politicization and intra-scientific politics. They discuss whether fraud is more damaging to the career of a senior or junior scientist, and the crisis coming for behavioral economics in the wake of the Francesco Gino and Dan Ariely ethics scandals. While Razib offers the prescription of viewpoint diversity, Clark argues that a recommitment to objectivity and truth as the fundamental values of science is needed. They then move on to her major current project on “adversarial collaboration.” Whereas in “normal science” two rival research groups may hold to conflicting hypotheses for decades, with outsiders unable to adjudicate, Clark argues that researchers with differing views should come together to converge upon the truth.

Her interest in the culture of science leads naturally to a broader concern about human cultural equilibria. In The Evolution of Relentless Badassery, Clark argues that a particular personality type is socially and evolutionarily favored. Razib and Clark discuss whether we live in a time of peace so that disagreeable violent characters are at a low ebb in their stature, and perhaps in the face of cultural chaos the “badass,” figures like Michael Corleone in The Godfather films may reemerge to establish order and ruthless justice.

The discussion loops back to a consideration of the values that unite scientists, and the cultural and political winds moving through the profession that might threaten to blow it off course as an enterprise, might leave it more a social club than a venerable institution to generate information. Clark is candid that she is not sure she would recommend heterodox students even attempt to join the academy.

Cory Clark: adversarial collaborations in science

Razib Khan's Unsupervised Learning