Too often appellate decisions decided on harmless error grounds are not serious read by lawyers (OK and trial judges as well). They appellate court spends most of the opinion focused on why the error is harmless and rarely state emphatically: DO NOT DO THIS AGAIN!!! So perhaps we should rethink harmless error.

Justin Murray (New York Law School) has posted Policing Procedural Error in the Lower Criminal Courts (Fordham Law Review, Forthcoming) on SSRN. Here is the abstract: The criminal justice system depends on reviewing courts to formulate norms of procedural law and to make sure those norms are actually followed in the lower courts. Yet reviewing courts are not performing either of these functions very well. No single factor can fully explain why this is the case, for there is plenty of blame to go around. But the harmless error rule is a major culprit. The conventional approach to harmless error review prohibits reversal of a defendant’s conviction or sentence, even when the law was violated during proceedings in the lower court, unless that violation influenced the outcome below. This limitation impedes effective oversight of the lower courts in two significant ways. First, it enables trial judges, prosecutors, and other relevant entities (such as a district attorney’s office, to name one example) to persistently evade accountability for procedural errors, diminishing their incentives to comply with legal norms. And second, it provides reviewing courts with a handy tool to avoid resolving legal claims on their merits. Instead of holding that an error did or did not occur, thereby helping trial judges, prosecutors, and others learn what the law requires going forward, reviewing courts can—and often do—affirm on factbound harmless error grounds without ever adjudicating the legality of the challenged conduct.

These failings call for a major shift in how courts review procedural error. I propose that, in addition to examining whether an error affected the outcome, as current law directs, a reviewing court should also consider whether (1) reversal would substantially help to prevent future errors, (2) the error caused substantial harm to a legally protected interest unrelated to the outcome, and (3) the benefits of reversal, as tabulated in the previous steps, outweigh its costs. After making the case for this framework and discussing how to operationalize each of its components, I then explore, a bit more tentatively, whether the same set of ideas could help stimulate much-needed rethinking of other controversial rules that further obstruct the policing of procedural error in the lower criminal courts.