Harmless error review is profoundly important, but arguably broken, in the form that courts currently employ it in criminal cases. One significant reason for this brokenness lies in the dissonance between the reductionism of modern harmless error methodology and the diverse normative ambitions of criminal procedure. Nearly all harmless error rules used by courts today focus exclusively on whether the procedural error under review affected the result of a judicial proceeding. I refer to these rules as “result-based harmlesserror review.” The singular preoccupation of result-based harmless error review with the outputs of criminal processes stands in marked contrast with criminal procedure’s broader ethical vision, which also encompasses non-result-related interests such as providing defendants with space for autonomous decisionmaking, enforcing compliance with nondiscrimination norms, and making transparent the inner workings of criminal justice.
The vast scholarship relating to result-based harmless error review, though deeply critical of its current role in the administration of justice, has not put forward an alternative method of harmless error review that courts might realistically consider using. Commentators in this area have devoted much of their energy toward persuading courts to exempt large swaths of criminal procedure from harmless error review entirely and thus to require automatic reversal for errors involving exempted rules. Instead, courts have done just the opposite by subjecting an ever-expanding list of errors to harmless error review, and there is no reason to think this trend will abate in the foreseeable future.
I attempt in this Article to chart a different course. My proposal, called “contextual harmless error review,” has two essential features. First, it would assess harm in relation to the constellation of interests served by the particular procedural rule that was infringed and would not, as under existing law, automatically confine the harmless error inquiry to estimating the error’s effect on the outcome. Second, contextual harmless error review would examine whether the error harmed the interests identified in the first step of the analysis to a degree substantial enough to justify reversal.
Murray, Justin, "A Contextual Approach to Harmless Error Review" (2017). Articles & Chapters. 1273.