Colloq (JF 507): Patricia Rich (University of Bayreuth, Germany): Epistemic Diversity: A Re-Evaluation

Friday, October 30, 2020 - 17:00
John Freely Hall, Bebek, BÜ Güney Kampüsü No:6, 34342 Beşiktaş/İstanbul, Turkey

ABSTRACT: It is a familiar problem that individuals often follow the crowd instead of relying on their own good judgment. Such behavior is especially troubling because it can cause the whole group to go badly astray, for example becoming convinced of falsehoods. Paradoxically, however, the standard view in philosophy is that it is typically the individuals who do *not* follow the crowd who are in fact irrational. If individuals were all perfectly rational, we would have much more conformity than we in fact have. I illustrate this state of affairs using the example of 'information cascades'; the term is used to describe events as varied as stock market crashes and book, restaurant and fashion trends. In information cascades, individuals follow the crowd because their private evidence is outweighed by public evidence in the form of others' actions. I explain why the standard analysis, which concludes that the non-conformists are irrational, is problematic, and how it reveals a basic limitation of our modeling approach. I argue that we can only really explain the non-conformist behavior we observe by adopting a game-theoretic perspective. From this perspective, non-conformist behavior is naturally construed as cooperative. I provide new results, using computer simulations, which show that sticking to one's own evidence can constitute a rational cooperative strategy for individual group members. This shows that the diversity of opinion that we see need not be the result of individual irrationality, contrary to the standard account.