120. Cristina Bicchieri | Social Norms

What really makes a behavior a social norm, and how can measuring people’s expectations about what others do and approve of help us change it?My links: https://linktr.ee/frictionphilosophy.1. GuestCristina Bicchieri is the S.J.P. Harvie Professor of Social Thought and Comparative Ethics in the Philosophy and Psychology Departments at the University of Pennsylvania, professor of Legal Studies in the Wharton School, and director of the Master in Behavioral Decision Sciences program and the Philosophy, Politics and Economics program.2. Interview SummaryBicchieri starts by pushing back against how loose and “catch-all” talk of social norms can be—especially definitions that treat a norm as simply “what people commonly do.” Instead, she frames social norms as genuinely social rules of behavior: shared within a group (sometimes local, sometimes society-wide) and “alive” in the sense that they’re sustained by what people expect of one another. On her view, two kinds of expectations matter—empirical expectations about what others do, and normative expectations about what others think one ought to do—and a behavior counts as governed by a social norm only when people’s preference to comply is conditional on those expectations. A key selling point is that this makes norms testable: you can measure the relevant expectations and the conditionality, which in turn makes norm talk more predictive and scientifically usable.From there she emphasizes both the psychology and the methodology. Psychologically, she notes that many everyday norms function like default rules: we often follow them automatically because they coordinate social life and spare us constant deliberation (her example is familiar “greeting” behavior like handshaking). Methodologically, she’s interested in detecting conditionality rather than just asking people what they personally endorse. So she stresses the importance of specifying the relevant reference group, then measuring empirical and normative expectations (via surveys), and then using vignette designs to reduce “experimenter demand” and to see whether predicted behavior shifts across different combinations of “most people do X” and “most people approve of X.” Those off-diagonal cases let her test what happens when descriptive and normative signals conflict—and she reports that, in her data, descriptive/empirical expectations typically “win the day”: when people believe most others do something, that tends to guide what they’ll do even if they also think most others don’t approve.A big theme in the latter part of the interview is application: if policymakers want to shift behavior, they need a clearer picture of the “cognitive plumbing” behind how people interpret social information. She treats “norm nudging” as a kind of black box—sometimes it works, sometimes it doesn’t—and points to systematic asymmetries in the inferences people draw from descriptive vs. approval messages, depending on whether the behavior is seen as positive or negative. She also highlights how crucial it is to target the right reference network: in a successful intervention aimed at curbing antibiotic overprescribing, the message was framed around what the majority of doctors in London do, so recipients couldn’t dismiss it as irrelevant. At the same time, she warns that you can’t design good interventions if you misclassify behaviors: in work for the Bill & Melinda Gates Foundation on open defecation in India, she found it wasn’t a social norm at all but a widely shared custom, which changes what it would take to replace it with a toilet-usage norm. She also connects this to why norm change is hard: trendsetters and tipping points exist, but thresholds are difficult to know in advance, so we can often explain change better than we can predict it. Finally, she notes that messaging can backfire when it clashes with what people observe locally—prompting a “pragmatic” (suspicious) rather than “semantic” (straightforward) interpretation—before closing with a defense of philosophy’s analytic mindset as a real asset for doing this kind of careful conceptual and empirical work.3. Interview Chapters00:00 - Introduction01:00 - What are social norms?05:23 - Formal definition12:42 - Context16:14 - Beliefs19:52 - Empirical research28:34 - Inferences and norm nudging32:20 - Historical development36:04 - Game theory40:56 - Norms and normativity44:28 - Changing social norms49:49 - Too broad?52:42 - Upshots of account56:50 - Open questions1:00:12 - Value of philosophy1:01:52 - Conclusion This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit fric.substack.com/subscribe