Document Type


Publication Date



This Article critically examines a cluster of rules that use the concept of prejudice to restrict the scope of criminal defendants’ procedural rights, forming what I call prejudice-based rights. I focus, in particular, on outcome-centric prejudice- based rights—rights that apply only when failing to apply them might cause prejudice by affecting the outcome of the case. Two of criminal defendants’ most important rights fit this description: the right, originating in Brady v. Maryland, to obtain favorable, “material” evidence within the government’s knowledge, and the right to effective assistance of counsel. Since prejudice (or equivalently, materiality) is an element of these rights, no constitutional violation occurs when the government suppresses favorable evidence, or defense counsel furnishes ineffective assistance, unless there is a reasonable probability that the outcome was affected thereby.

Outcome-centric, right-restricting prejudice rules serve understandable functions: they enable courts to preserve finality by rationing appellate and postconviction relief. To the extent this is their aim, however, such rules sweep far too broadly. For they narrow the scope of defendants’ rights not only in appellate and postconviction proceedings, as intended, but also at the trial court level, where finality is not at stake. The overbreadth of these rules is deeply problematic. For one thing, efforts to predict outcome-determinative prejudice ex ante usually amount to little more than guesswork, since relevant information is in short supply during the early stages of a prosecution. And in any event, procedural fairness is essential in every criminal case—not just when fair processes are needed to prevent unfair outcomes. Outcome- centric prejudice-based rights are at odds with that premise, because the entitlements they bestow vanish once a trial judge, prosecutor, or police officer determines that the outcome is a foregone conclusion.

After fleshing out these concerns, I propose a two-pronged strategy for reforming Brady, effective assistance of counsel, and potentially other outcome-centric prejudice-based rights. First, courts should remove prejudice from the definition of these rights, treating prejudice instead as a remedial question—one that would come into play when a convicted defendant seeks relief from an appellate or postconviction court, but generally not in other settings. Second, courts should dismantle the outcome-centric conception of prejudice embedded in these doctrines and replace it with a non-outcome-centric framework that I call contextual harmless error review. These reforms would greatly improve the fairness of the criminal process, and would do so without unduly disturbing the finality of trial court judgments.


University of Pennsylvania Law Review, Vol. 168, Issue 2 (January 2020), pp. 277-330