* Associate Professor of Law, Washington University in St. Louis. I am grateful to the editors of the Columbia Law Review Forum for helpful comments.
My article Harmless Errors and Substantial Rights 1 Daniel Epps, Harmless Errors and Substantial Rights, 131 Harv. L. Rev. 2117 (2018). challenged conventional wisdom about the harmless constitutional error doctrine in criminal procedure. Specifically, I contended that the traditional way of understanding harmless error as a remedial doctrine rooted in so-called “constitutional common law” 2 For the canonical articulation of this view, see Daniel J. Meltzer, Harmless Error and Constitutional Remedies, 61 U. Chi. L. Rev. 1, 23–34 (1994); see also Richard H. Fallon, Jr. & Daniel J. Meltzer, New Law, Non-Retroactivity, and Constitutional Remedies, 104 Harv. L. Rev. 1731, 1771–72 (1991); Craig Goldblatt, Comment, Harmless Error as Constitutional Common Law: Congress’s Power to Reverse Arizona v Fulminante, 60 U. Chi. L. Rev. 985, 986–87, 1005–12 (1993). created significant anomalies. The remedial perspective does not explain which errors can properly be treated as harmless, provides no guidance for how harmless error analysis should work in practice, leaves mysterious the relationship between the doctrine and the relevant statutory framework, and creates deep uncertainty about the Supreme Court’s power to impose the doctrine of harmless constitutional error on state courts. 3 See Epps, supra note 1, at 2142–58. Instead, harmless constitutional error doctrine can only be understood as part of the definition and judicial enforcement of constitutional rights. A rights-based theory of harmless error helps explain which errors can be treated as harmless, and how harmless error analysis should work in practice; makes sense of the governing statutory law; shows why the Supreme Court has the power to require state courts to follow federal harmless error rules; and prevents courts from surreptitiously undermining the value of constitutional rights. 4 See id. at 2164–70.
Few legal scholars have thought as deeply about the mysteries of harmless error as Professor John M. Greabe, 5 See generally John M. Greabe, The Riddle of Harmless Error Revisited, 54 Hous. L. Rev. 59 (2016) [hereinafter Greabe, Revisited] (assessing the Supreme Court’s “unnecessarily complicated” application of harmless error analysis and arguing for a simplified harmless error test). and he is well equipped to give the remedial perspective the best possible defense. I am grateful to him for his generous response 6 John M. Greabe, Criminal Procedure Rights and Harmless Error: A Response to Professor Epps, 118 Colum. L. Rev. Online 118 (2018), https://columbialawreview.org/ content/criminal-procedure-rights-and-harmless-error-a-response-to-professor-epps/ [https:// perma.cc/TPP3-KKFH] [hereinafter Greabe, Criminal Procedure Rights]. to my article, and believe he makes several good criticisms of my approach that deserve a reply. Nonetheless, despite Professor Greabe’s able efforts, I remain persuaded of the correctness of the rights-based approach to harmless error laid out in my article.
In this short Reply, I will explain why. I divide my remarks into two Parts. First, I respond to Professor Greabe’s most significant criticisms of my rights-based approach, and show why I think they miss the mark. Second, I explain why Professor Greabe’s own remedy-based theory is itself problematic and unable to solve the enduring riddles of harmless error.
I. Greabe’s Criticisms
In Harmless Errors and Substantial Rights, I identified a number of significant problems with the traditional perspective on harmless constitutional error. When conceived of as a remedial inquiry into the appropriate solution for an already-completed constitutional violation, the harmless error doctrine runs into a number of anomalies and difficulties. To list the most significant problems with that approach: It provides a dubious basis for the Supreme Court to require state courts to follow the Chapman v. California 7 386 U.S. 18 (1967). harmless error standard; 8 See Epps, supra note 1, at 2142–51. it provides no good way to identify which constitutional errors should properly be subject to harmless error review; 9 See id. at 2151–55. it provides no useful guidance on how that review should be conducted; 10 See id. at 2155–58. and it is hard to reconcile with the statutory framework ostensibly governing appellate courts’ use of harmless error doctrine. 11 See id. at 2144–46. A rights-based approach, by contrast, provides a simple answer to metaphysical questions about the constitutional basis for Chapman’s beyond a reasonable doubt test; 12 Under Chapman, “[B]efore a federal constitutional error can be held harmless, the court must be able to declare a belief that it was harmless beyond a reasonable doubt.” Chapman, 386 U.S. at 24; see also, e.g., Delaware v. Van Arsdall, 475 U.S. 673, 681 (1986) (“[A]n otherwise valid conviction should not be set aside if the reviewing court may confidently say, on the whole record, that the constitutional error was harmless beyond a reasonable doubt.”). it necessarily implies answers to questions about which errors should be subject to harmless error analysis, and how that review should be conducted; and it provides a simple way to reconcile the practice of harmless error with relevant statutes and rules. 13 See Epps, supra note 1, at 2164–70.
For the most part, Professor Greabe does not quarrel with these arguments. He neither defends the traditional remedial approach from my criticisms, 14 The one possible exception is Professor Greabe’s claim that his approach, like mine, makes sense of 28 U.S.C. § 2111, the federal harmless error statute that provides that “[o]n the hearing of any appeal or writ of certiorari in any case, the court shall give judgment after an examination of the record without regard to errors or defects which do not affect the substantial rights of the parties.” 28 U.S.C. § 2111 (2012). I address that claim in Part II. nor does he suggest that my claim that the rights-based approach offers advantages on all these fronts is misguided. That seems like a significant concession, since the problems with the traditional remedial approach—and the corresponding benefits of my revisionist approach—are quite significant.
Instead, Professor Greabe argues that my approach cannot be right because it would create significant problems when applied to other procedural contexts beyond appellate review of convictions —specifically, trial administration, postconviction review, and congressional enforcement of criminal procedure rights under Section 5 of the Fourteenth Amendment. I address each in turn.
First, Professor Greabe contends that the rights-based approach creates serious problems for the administration of criminal trials. 15 See Greabe, Criminal Procedure Rights, supra note 6, at 124–27. Here is his objection: Under my theory, the Confrontation Clause is best understood as creating a right against conviction based on unconfronted testimonial hearsay (thus explaining why supposed “violations” of the right can be considered harmless) and not as creating a right against the admission of unconfronted testimonial hearsay, full stop. But if that is right, Professor Greabe asks, what authority do trial courts have to exclude all unconfronted testimonial hearsay without any additional finding that the evidence will actually be harmful—and, thus, without finding that the evidence will violate the defendant’s constitutional rights? 16 See id. at 125–26. Twisting the knife further, Professor Greabe contends that the only way to justify this practice would be to recognize some kind of “constitutional common law evidentiary principle.” 17 See id. This argument cuts deep, as one of the very problems with the remedial approach that my theory is supposed to solve is the remedial approach’s need to rely on a questionable account of “constitutional common law” to justify the imposition of a harmless error doctrine on state courts. 18 See Epps, supra note 1, at 2150; see also Henry P. Monaghan, The Supreme Court, 1974 Term—Foreword: Constitutional Common Law, 89 Harv. L. Rev. 1, 2–3, 9–10 (1975) (“If the Supreme Court is not mistaken in its insistence on the application of the exclusionary rule in state cases . . . we are driven to conclude that the Court has a common law power.”).
I commend Professor Greabe for identifying what may be the toughest challenge for my theory of harmless error. Yet I don’t think he has landed a death blow for the rights-based approach by any means. Let us continue with the helpful example of a Confrontation Clause violation. Even if we recognize that a constitutional violation is not actually complete until, say, the erroneous admission of evidence results in a conviction, it does not follow that trial courts should (or will) start admitting evidence that could result in an unconstitutional conviction willy-nilly.
Consider how things would work in practice. Even under my theory, what justification would a trial court have for admitting evidence on the basis that it might end up being harmless? If the unconfronted testimonial hearsay is offered as substantive proof of the defendant’s guilt, how on earth could a trial court be confident at that moment—before the rest of the trial record was established—that it would not eventually contribute to the defendant’s conviction under the demanding Chapman beyond a reasonable doubt standard? Perhaps a trial court could reach that level of certainty if the proffered evidence did nothing to actually establish the defendant’s guilt. But if so, one might wonder whether the evidence should even be admissible under ordinary principles of evidentiary relevance. 19 See Fed. R. Evid. 401–403; see also Dale A. Nance, The Best Evidence Principle, 73 Iowa L. Rev. 227, 273 n.221 (1988) (“[G]enerally the court ought to admit evidence if it finds that a reasonable juror could find the evidence relevant. This kind of standard is available for the determination of preliminary facts that supposedly condition relevance.”). Perhaps the trial court might admit the unconfronted testimonial hearsay on the theory that the substance of the statement was not being offered to prove the defendant’s guilt. But here, Confrontation Clause doctrine already permits trial courts to admit such statements. 20 See Crawford v. Washington, 541 U.S. 36, 59 n.9 (2004) (noting that the Confrontation Clause “does not bar the use of testimonial statements for purposes other than establishing the truth of the matter asserted”). It is hard, then, to see what exactly the problem is.
Moreover, even Professor Greabe’s seemingly strongest argument—about the supposed need to recognize a “constitutional common law evidentiary principle” governing the admissibility of evidence under my theory—crumbles under scrutiny. Recall that the remedial approach needs to rely on constitutional common law in order to explain why the Supreme Court has the power to reverse state court convictions for failure to adhere to the Chapman beyond a reasonable doubt standard. 21 See Epps, supra note 1, at 2149–51 (summarizing the constitutional common law argument); see also Meltzer, supra note 2, at 26–29 (arguing that Chapman is best understood as constitutional common law); Monaghan, supra note 18, at 2–3 (originating the concept of constitutional common law); Goldblatt, supra note 2, at 1005–07 (making a similar claim). My theory explains away that problem by recognizing that Chapman is just part of the constitutional decision rule for enforcing a constitutional right. 22 See Epps, supra note 1, at 2164–65. But when it comes to the admission of evidence, what practice, exactly, stands in need of a constitutional justification? Does the Supreme Court routinely reverse state-court convictions when trial courts admitted unconfronted testimonial hearsay that was harmless? Not at all; instead, the Supreme Court would only reverse if the Chapman standard were satisfied. There thus seems to be nothing that a theory of constitutional common law needs to explain under my approach.
Second, Professor Greabe argues that my approach will create conceptual difficulties for habeas corpus. Under my approach, at least if Brecht v. Abrahamson 23 507 U.S. 619 (1993). Brecht held that the harmless error test applicable in the habeas context is the far less defendant-friendly “substantial and injurious effect or influence” test rather than the Chapman beyond a reasonable doubt standard. Id. at 638 (internal quotation marks omitted) (quoting Kotteakos v. United States, 328 U.S. 750, 776 (1946)). is not overturned, “habeas courts would be required to apply a different definition of the underlying constitutional right than the one that direct-review courts apply.” 24 Greabe, Criminal Procedure Rights, supra note 6, at 128. In Professor Greabe’s view, this would “transform habeas into a fundamentally different regime.” 25 Id. Going further, Professor Greabe imagines a confusing procedural morass, in which state criminal defendants wishing to preserve their federal habeas rights during direct appeals and collateral proceedings in state courts would have to assert both that an alleged constitutional trial error was not harmless beyond a reasonable doubt and that it had a substantial and injurious effect or influence on the jury’s verdict. 26 Id. at 129.
I agree that if my theory commanded such a patently absurd result, that alone would be reason to reject it. But here, I fear that Professor Greabe has gotten a bit wound around the axle in his efforts to rebut my theory. The problem, as I see it, is that he does not acknowledge the distinction between a constitutional right and judicially created doctrinal tests designed to enforce that right. A constitutional right is stated at a fairly high level of generality—we speak of the right to free speech, even if in practice that right is mediated through doctrinal tests that apply various “tiers of scrutiny” depending on the particular type of speech and the particular context at issue. 27 See Richard H. Fallon, Jr., The Supreme Court, 1996 Term—Foreword: Implementing the Constitution, 111 Harv. L. Rev. 54, 67–73 (1997) (describing various kinds of doctrinal tests).
So too with rights like the one created by the Confrontation Clause. The right, properly understood, bars conviction on the basis of particular kinds of evidence. But when a defendant seeks to vindicate that right on direct appeal, it is mediated through the Chapman doctrinal framework. The Chapman test cannot be derived from the text of the Constitution, but it is nonetheless a legitimate judicial creation by the Supreme Court in its efforts to implement the Constitution. 28 See id. at 56–61 (“Even when general agreement exists that the Constitution reflects a particular value or protective purpose, questions of implementation often remain. . . . The Supreme Court has responded accordingly.”). Adopting the rights-based theory requires acknowledging that the particular doctrinal test used to enforce some constitutional rights can change based on procedural context. The right remains the same, but—at least if Brecht remains the law—the doctrinal test that courts will use to evaluate claims that that right has been violated will change based on context, with courts using a more government-friendly doctrinal test in the postconviction context in light of the differing calculus of interests in that procedural context. There is no serious conceptual problem with this possibility, even if it is not the way we are used to thinking about things. I also note that Professor Greabe’s revisionist theory, like mine, sees Brecht as wrongly decided. 29 Greabe, Criminal Procedure Rights, supra note 6, at 133. Thus, on that score, his theory seems to offer no advantages over my own.
Third, Professor Greabe argues that my approach is problematic because it would severely limit Congress’s ability to enforce criminal procedure rights using legislation under Section 5 of the Fourteenth Amendment. 30 See id. at 130–31. My approach would make it harder for Congress to enact legislation abrogating state sovereign immunity in the face of severe violations of constitutional rights by state prosecutors and judges, he argues. 31 See id. Conceding that such legislation does not presently exist, and that it does not seem likely to be enacted anytime soon, 32 See id. at 131 (“This hypothetical may be fanciful given our current political climate.”). Professor Greabe nonetheless worries that my approach would water down the value of constitutional protections if such a scenario ever came to pass.
I confess to being unconcerned by this objection. As I have argued, one leading problem with our present approach to harmless error is how it enables courts to undermine constitutional rights surreptitiously, by “declar[ing] a broad scope of a constitutional right while in the same breath undercutting that right’s effective value through the use of harmless error analysis.” 33 Epps, supra note 1, at 2186. The rights-based approach would address this failing by forcing courts to acknowledge the ways in which they are substantively redefining the scope of constitutional rights by applying harmless error doctrine too generously. If, in the course of addressing this very real and present problem of courts watering down constitutional rights, my approach at the same time might make it harder to enforce those same rights through legislation in some hypothetical future that I (and, admittedly, Professor Greabe 34 See supra note 32 and accompanying text. )find hard to imagine, that is a trade I am more than willing to make.
Yet even considering it on its own terms, I fail to grasp the force of Professor Greabe’s imagined worst-case scenario. Consider the precise details of the hypothetical scenario he is concerned with:
Imagine a wave election in which Congress comes under the control of politicians who are deeply concerned about whether federal rights are being sufficiently observed during state criminal trials. This new Congress holds hearings and develops an extensive body of evidence showing that in states with elected judiciaries in which judicial candidates are frequently former prosecutors who campaign on promises to be tough on crime, prosecutors regularly seek to admit—and trial judges regularly do in fact admit—evidence obtained in violation of constitutional criminal procedure rights spelled out in the Fourth, Fifth, and Sixth Amendments, at least as those rights are currently understood. Moreover, appellate courts in such states (also staffed with elected judges) regularly withhold remedies for such violations under Chapman. 35 Greabe, Criminal Procedure Rights, supra note 6, at 130.
Professor Greabe hopes that in this future, Congress will be able to pass legislation enabling “convicts who are later exonerated to file constitutional tort actions under 42 U.S.C. § 1983 against prosecutors and trial judges who participated in particularly egregious rights violations at their trials.” 36 See id. at 130–31.
In this fact pattern, I see two possibilities. One is that the rights violations at issue are actually contributing to exonerees’ false convictions, and that the state appellate court judges are misapplying Chapman. If so, there would be no problem with Congress providing a money damages remedy, since those exonerees would have had their constitutional rights violated even under my theory (since they would have been convicted on the basis of impermissible evidence). The other possibility is that the state courts applied Chapman correctly, and the supposedly improper evidence did not contribute to the exonerees’ false convictions. In that scenario, then, I agree that Congress might face more obstacles to providing a damages remedy, for all the reasons Professor Greabe identifies. But I’m puzzled about why we should care about this scenario; if the improper evidence really was harmless, I fail to see why it should be a priority for Congress to provide a damages remedy in those cases—even if I am willing to entertain the possibility that Congress might care to do so at some point in the future. Put another way, what interest does it serve to require state judges and prosecutors to pay damages awards (potentially out of their own pockets) for constitutional “errors” that made no difference to the outcome? Professor Greabe does not provide an answer.
Thus, the transcontextual concerns that Professor Greabe raises pose no serious challenge to the rights-based approach to harmless error. Whatever seeming difficulties that theory creates are decidedly second order when compared to the serious conceptual and practical problems that the remedial approach creates. I laid out those problems in detail in my article and will not do so again here. But suffice it to say that while no theory is perfect, the balance appears to weigh strongly in favor of my theory.
Some may harbor a deeper reservation about the rights-based approach. Professor Greabe notes at several points that my approach is very much at odds with some things the Supreme Court has said. 37 See id. at 123, 128 (“Epps’s argument runs up against the fact that it has been considered and rejected by the Court. But such is the nature of doctrinal-reform scholarship. . . . But Van Arsdall got it right.”). This is not a fatal problem, he says, and he himself is willing to urge serious doctrinal course corrections in support of his own view. 38 See id. at 133 (urging the Supreme Court to abandon Chapman, the structural defect–trial error dichotomy, and Brecht). Nonetheless, because this objection may be particularly troubling to some readers, it deserves some response. 39 I also wonder whether this objection might inform some of Professor Greabe’s arguments—despite his statements to the contrary—given his emphasis of the point.
Let me begin by discussing the purpose and methods of the kind of doctrinal-reform effort that my article sought to undertake, as Harmless Errors and Substantial Rights did not contain any lengthy explanation of what exactly it sought to do. The premise of this kind of work is that careful analysis of the legal doctrine can reveal inconsistencies between different legal sources—conflicts between two different cases, or between a case and what a statute or constitutional provision says, or between what the cases say and some deeper intuition about the right answer to particular disputed questions. 40 For a thoughtful explanation of how careful examination of shared legal premises can lead to a conclusion that a widely shared view about the law might be wrong, see Stephen E. Sachs, The “Constitution in Exile” as a Problem for Legal Theory, 89 Notre Dame L. Rev. 2253, 2253–60 (2014). Harmless constitutional error is one place where the law, such as it is, is rife with difficult-to-reconcile inconsistencies. Existing doctrine is nearly impossible to square with statutory law, and it provides no good explanation of why certain results that seem obviously wrong 41 For example: Could an appellate court simply affirm a conviction where the defendant was deprived of a jury trial on the ground that the defendant is guilty anyway? That result seems intuitively wrong—yet the remedial approach cannot explain why. See Epps, supra note 1, at 2154–55; see also Sullivan v. Louisiana, 508 U.S. 275, 280 (1993) (“The Sixth Amendment requires more than appellate speculation about a hypothetical jury’s action, or else directed verdicts for the State would be sustainable on appeal; it requires an actual jury finding of guilty.”). are off the table.
My article sought to drill down to the deeper principles and build the legal doctrine back up in a way that offered the most coherent synthesis of the cases, the constitutional text and statutory law, and our intuitions. My theory requires relatively little in the way of reconsidering actual holdings of past Supreme Court decisions, 42 The only case that I say with certainty was wrongly decided under my theory is Neder v. United States, 527 U.S. 1 (1999). Epps, supra note 1, at 2178 (“Neder is clearly wrong . . . ; there simply is no way to reconcile a right to have the jury make all the relevant factual findings . . . with a harmless error rule that permits appellate courts to ignore flagrant violations of that right . . . .”). though it does require recognizing that some of the language the Supreme Court has used to talk about the scope of rights and the harmless error doctrine has been wrong or at least incomplete. While the best theory might make perfect sense of all possible sources of law, the rights-based theory offers the most coherence and does the least damage to the law as a whole given the remedial approach’s failure to provide satisfactory answers to so many puzzling questions about harmless error.
II. The Remedial Approach Reconsidered
After offering his criticisms of my rights-based theory, Professor Greabe proceeds to defend his own version of a remedial approach to the harmless constitutional error doctrine. Professor Greabe has thought carefully about his theory, and he offers a significant improvement on the current doctrinal approach. In particular, I admire his willingness to address the federal harmless error statute and to offer an account that tries to explain how that statute fits into the practice of harmless error—something that the Supreme Court has steadfastly refused to do.
Nonetheless, I think Professor Greabe’s approach is still deficient to my own, and that his theory fails to resolve some of the most glaring problems with the remedial approach. In this Part, I briefly explain why. Rather than recapping all of the flaws of the remedial approach, I will focus on those that Professor Greabe emphasizes in his response.
The main problem is Professor Greabe’s conception of harmless error as a doctrine that is primarily concerned with “substitutionary relief for a wholly concluded wrong.” 43 Greabe, Criminal Procedure Rights, supra note 6, at 133 (quoting Greabe, Revisited, supra note 5, at 64). In his view, an appellate court’s responsibility is only to ask whether a rights violation in the past is sufficiently severe as to justify the remedy of reversal. In performing that analysis, under his view, courts should exercise discretion to reverse depending on whether reversal would vindicate various constitutional values. 44 See id. at 134.
This approach, in my view, leaves judges completely at sea. And it provides no good answers to hard questions. Imagine a case where a dangerous serial killer who is obviously guilty was deprived of his right to a jury trial and convicted by a judge after a cursory bench trial. If he appeals, is he entitled to a new trial, or can an appellate court say that because the defendant is clearly guilty, the cost of providing the remedy of reversal simply isn’t worth it? Professor Greabe’s theory provides no good explanation of why that course is out of bounds. To be sure, Professor Greabe might believe that the balance of constitutional values counsels in favor of providing a remedy of reversal here. But it is not clear what his response is to a judge who simply weighs those values differently.
To ask the question another way: Would it be permissible for Congress to declare that henceforth defendants will not receive jury trials, but that they are entitled to money damages awards afterwards as compensation for the completed violation of their rights? Again, I suspect many people’s strong intuition would be that such a course is impermissible. But if harmless error merely involves an inquiry into the permissible remedy for a completed constitutional violation, it is unclear why substituting a remedy in this way is off the table. As Professor Daniel Meltzer put it, “[T]he Constitution often requires some adequate remedy but not necessarily any particular one.” 45 Meltzer, supra note 2, at 25.
Yet the deprivation of a right to a jury trial is perhaps the core example of one that our intuitions tells us can never be harmless. 46 See Sullivan, 508 U.S. at 280 (holding that without an “actual jury finding of guilty,” there is no basis “upon which harmless-error scrutiny can operate”). The answer, then, is to recognize that some kinds of constitutional errors are not over and done the moment they first arise; they render any resulting conviction invalid, making the remedy of reversal mandatory. 47 Richard Re’s due process approach to harmless error treats harmless error as an inquiry into whether a conviction rests on an ongoing constitutional violation. See Richard M. Re, The Due Process Exclusionary Rule, 127 Harv. L. Rev. 1885, 1915–17 (2014). While my article expressed some questions about that premise, see Epps, supra note 1, at 2147–48, Professor Greabe’s emphasis on the distinction between completed and ongoing violations has given me more sympathy for Professor Re’s view. See, e.g., Greabe, Criminal Procedure Rights, supra note 6, at 132–33. There is nothing to balance, no competing considerations to weigh. A defendant was denied the right to a jury trial, and the obvious and necessary conclusion is that the conviction cannot stand. The remedial approach simply cannot provide courts with this needed clarity.
Instead, it seems to give appellate courts free rein to recognize the remedy of reversal, or not, depending on how they feel about the complex constellation of values implicated by a particular constitutional violation. In Professor Greabe’s words, courts should determine whether “an exercise of remedial discretion is needed to vindicate [fundamental constitutional] values.” 48 Greabe, Criminal Procedure Rights, supra note 6, at 134. He gives a few examples of circumstances that would satisfy this standard in his view. 49 See id. (suggesting that such circumstances include “judicial proceedings marred by unconstitutional discrimination on the basis of race, religion, ethnicity, national origin, or gender and intentional misconduct by government officials such as judges, prosecutors, and policy or probation officers”). But can we be confident that the judges applying this malleable standard—which would include state and federal appellate judges across the country—would agree with Professor Greabe’s instincts on what exactly our “fundamental constitutional values” are, let alone when exactly reversal is necessary to vindicate them? Such a flexible standard may give us more of what we already have—a harmless error doctrine that can be used to make many constitutional protections meaningless in practice.
For this reason, Professor Greabe’s approach seems to provide no clearer answers than other scholarly attempts to make sense of the remedial approach to harmless error. One such example comes from Professor Justin Murray, who has argued for a “contextual” approach to harmless error. 50 See Justin Murray, A Contextual Approach to Harmless Error Review, 130 Harv. L. Rev. 1791, 1795–96 (2017). Under his approach, courts “would begin by identifying the interest (or range of interests) protected by whichever procedural rule was infringed.” 51 Id. at 1795. Having done so, then “[t]he court would balance the redressable harm caused by the error against the social cost of reversal and reverse if the former outweighs the latter.” 52 Id. Like Professor Greabe, Professor Murray has offered an able defense and reconstruction of a remedial approach to harmless constitutional error. But as with Professor Greabe’s approach, Professor Murray’s approach gives courts far too little guidance in individual cases. If this is where the remedial approach leads, that is all the more reason to prefer the rights-based course.
A more minor problem with Professor Greabe’s approach is precisely how he deals with 28 U.S.C. § 2111, the federal harmless error statute. 53 See supra note 14 and accompanying text. Despite my admiration for Professor Greabe’s willingness to grapple with the statutory text, I don’t think his solution has things quite right. In his view, § 2111, rather than “constitutional common law,” provides the basis for the Chapman ruling. 54 See Greabe, Criminal Procedure Rights, supra note 6, at 133. This view has an illustrious lineage, given that it was first suggested by Justice Roger Traynor in his classic treatment of the subject of harmless error. 55 See Roger J. Traynor, The Riddle of Harmless Error 57 (1970) (“Since Section 2111 does not distinguish constitutional violations from other errors, it apparently governs them also . . . .”). Nonetheless, this view of § 2111 strikes me as deeply flawed—largely for reasons offered by Professor Meltzer. 56 Meltzer, supra note 2, at 20.
The first problem Professor Meltzer identified is the provision’s text. It provides as follows:
On the hearing of any appeal or writ of certiorari in any case, the court shall give judgment after an examination of the record without regard to errors or defects which do not affect the substantial rights of the parties. 57 28 U.S.C. § 2111 (2012).
The difficulty is that this text reads like a limit on a reviewing court’s ability to provide the remedy of reversal, given that it merely instructs courts to ignore certain kinds of errors. In Professor Meltzer’s words, “[T]he statute enjoins the Court to ignore errors that do not affect substantial rights; it does not mandate reversal where substantial rights are affected.” 58 Meltzer, supra note 2, at 20. Another problem is that Chapman requires state courts to follow the federal harmless error standard, and it is far from obvious that § 2111 was meant to apply to state appellate courts. 59 See id. at 21 (“[T]he statutory text hardly makes clear that it means to regulate the standards of harmlessness applied by the states in their own courts, rather than by the Supreme Court on review of state court decisions.”). Nor is it obvious that Congress even has the power to require state appellate courts to follow a particular remedial regime for constitutional violations (if we accept the premises of the remedial view). 60 As Professor Meltzer has observed, the Supreme Court has long insisted that there is no constitutional right to an appeal in the first place. See id. at 5–6. That being so, it is not clear why Congress would have the power to insist that state courts follow particular remedial regimes, given that an appeal is merely a matter of grace. Perhaps such a law could be justified as an exercise of congressional power to enforce the Fourteenth Amendment, but if so, Congress would likely need to show that such legislation reflected “congruence and proportionality between the injury to be prevented or remedied and the means adopted to that end.” City of Boerne v. Flores, 521 U.S. 507, 520 (1997). I am aware of no evidence that Congress thought it was exercising this power when it passed § 2111 or its predecessors. If Congress amended § 2111 to require automatic reversal for even the most trivial violations of the Confrontation Clause, would it have the authority to do so? Perhaps, but proving the case would require arguments that Professor Greabe has not provided.
In contrast to Professor Greabe’s view, the rights-based approach provides clear answers—or at the very least, a clear question for courts to answer. And it provides a straightforward account of how § 2111 interacts with the doctrine. The statutory provision is, as the test suggests, a command for courts not to overenforce constitutional rights by ordering reversal for prima facie violations that do not actually violate any meaningful right of the defendant. 61 See Epps, supra note 1, at 2165.
Professor Greabe has offered difficult counterarguments for the rights-based approach to harmless error. And he has made a thoughtful refinement to the prevailing remedial approach, one that offers perhaps the best possible reconstruction and defense of that view. Nonetheless, I remain unpersuaded, and if anything, I believe Professor Greabe’s efforts ultimately illustrate why only a rights-based approach can solve the enduring riddles of harmless constitutional error.