Recasting Reasonable Doubt: Decision Theory and the Virtues of Variability

Abstract:     
According to the traditional understanding, proof beyond a reasonable doubt requires a high level of certainty before a jury convicts someone of a crime. This view is often framed in explicitly utilitarian terms: that a high standard of proof is justified because the costs of erroneously convicting an innocent person are so much higher than the costs of erroneously acquitting a guilty person. Empirical evidence, though, suggests that the standard reasonable doubt jury instruction does not actually require as much certainty as we generally assume. The common response to this observation is to propose improvements to the instruction to ensure that juries will require a high level of certainty. This Article contends that this solution is misguided because it wrongly presumes that the instruction needs to be fixed. Instead, what needs to change is our understanding of reasonable doubt. Using the expected utility model of decision theory, as well as insights from behavioral economics and the social norms literature, this Article suggests that the reasonable doubt standard of proof is inevitably flexible in nature: in some cases juries will require more proof than in other cases. The Article goes on to suggest that this result is in fact preferable to a standard of proof that requires a high level of certainty in all criminal cases.

Number of Pages in PDF File: 113

JEL Classification: K14, K41, K42

Accepted Paper Series