A number of legal scholars have recently posited a "group herding" account of legal institutions and precedent. Their arguments draw heavily from the emergent literature on information cascades -- a set of theories that attempts to explain certain pathologies of group behavior. Explicitly, cascade theory demonstrates how rational individuals, when making ostensibly individual decisions, may ignore their own private inclinations, choosing to emulate others instead. When aggregated, such behavior can lead to significant errors of judgment on behalf of the entire group, particularly if exacerbated by other cognitive effects. If legal precedent is a species of cascade, then the consequences for legal theory are profound. Indeed, not only would such a theory call into question the wisdom of celebrated "watershed" cases from the last century, but it would also severely compromise virtually any positive account of law that conceives of precedent as a central mechanism for judicial learning. This article critically analyzes the viability of a cascade theory of precedent, asking whether such phenomena are (a) possible, (b) plausible, and (c) empirically verifiable. Although I find that a precedential cascade is certainly possible as a matter of theory, the necessary conditions for its occurrence are either extremely rare or easily avoidable. Moreover, even the more modest task of diagnosing when precedential cascades occur may be unavailing: for the outward symptoms of a cascade are essentially indistinguishable from those of more plausible (and less troubling) behavioral hypotheses.
Eric L. Talley,
Precedential Cascades: A Critical Appraisal,
USC Law School, Olin Working Paper No. 99-6
Available at: https://scholarship.law.columbia.edu/faculty_scholarship/1179