What standard of review should federal habeas courts use when assessing state courts' prior determinations that constitutional trial errors were harmless?
A 6-to-3 divided Supreme Court answered that question yesterday in Brown v. Davenport, No. 20-826 (Apr. 21, 2022), with Justicer Gorsuch wrting the majority opinion, availabe here. The Court held that when a state court has ruled on the merits of a state prisoner's claim, a federal court cannot grant habeas relief without applying both the test the Court outlined in Brecht v. Abramson, 507 U.S. 619 (1993), and the one Congress prescribed in the Antiterrorism and Effective Death Penalty Act (AEDPA), 28 U.S.C. SS 2254(d). Further, the Court held the state court's determination, that the due process violation from petitioner's shackling during trial was harmless, was neither contrary to nor an unreasonable application of clearly established federal law under SS 2254(d)(1).
By way of background, on direct appeal both state and federal courts apply the harmless error test from Chapman v. California, 386 U.S. 18, 24 (1967), which requires the prosecution to show that a constitutional error is harmless beyond a reasonable doubt. See also Deck v. Missouri, 544 U.S. 622, 635 (2005) ("[W]here a court, without adequate justification, orders the defendant to wear shackles that will be seen by the jury ... [t]he State must prove 'beyond a reasonable doubt that the [shackling] error complained of did not contribute to the verdict obtained[.]'"). In Brecht, the Supreme Court adopted a different, more state friendly harmless-error review standard in federal habeas cases, based on interests of finality and comity. Brecht requires a showing of "actual prejudice," meaning that a state prisoner seeking to challenge his conviction on the basis of a state court's constitutional violation must show that the error had a "substantial and injurious effect or influence" on the trial's outcome. Under AEDPA, federal courts are not permitted to grant state prisoners relief for claims adjudicated on the merits in state court unless the petitioner shows that a state court's decision was contrary to or an unreasonable application of clearly established federal law, or was based on an unreasonable determination of the facts. 28 U.S.C. SS 2254(d)(1)-(2).
Ervine Davenport was convicted of first-degree murder and sentenced to life in prison without the possibility of parole after a trial where he was visibly shackled in front of the jury. On direct appeal, he argued that his conviction should be set aside in light of Deck v. Missouri. Finding no special need, the Michigan Supreme Court agreed that a Deck violation occurred and remanded the case to the trial court to determine whether the state could prove the error was harmless under Chapman. The trial court conducted an evidentiary hearing at which jurors testified that the shackles had not affected their verdict, and the trial court concluded the error was harmless. The Michigan Court of Criminal Appeals affirmed, and the Michigan Supreme Court declined review.
Davenport filed a petition for federal habeas relief which was denied by the district court. A divided Sixth Circuit panel reversed, declining to analyze the case under AEDPA. Instead, the Sixth Circuit held that its review was governed only by Brecht.
The Supreme Court said this was a mistake: "When a state court has ruled on the merits of a state prisoner's claim, a federal court cannot grant relief without first applying both the test this Court outlined in Brecht and the one Congress prescribed in AEDPA." A state court's harmless error determination under Chapman qualifies as an adjudication on the merits under AEDPA. Thus, "satisfying Brecht is only a necessary, not a sufficient, condition to relief. AEDPA too much be satisfied."
Justice Kagan, with whom Justices Breyer and Sotomayor joined, dissented:
Twice in recent years, this Court has addressed how a federal habeas court is to evaluate whether a state trial error was harmless. See Fry v. Pliler, 551 U.S. 112, 119-120 (2007); Davis v. Ayala, 576 U.S. 257, 267-270 (2015). And twice, we have made clear that the habeas court need apply only the standard prescribed in Brecht v. Abrahamson []; it need not also run through the test set out in the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA). That is because, we have both times explained, the Brecht standard "obviously subsumes" the "more liberal" AEDPA one: If a defendant meets the former, he will "necessarily" meet the latter too. Fry, 551 U.S. at 120; Ayala, 576 U.S. at 270.
Today, the Court discards those crystal-clear statements, subscribed to on each occasion by every Justice. The majority reverses the Court of Appeals for following our prior guidance, allowing the use of the Brecht test alone. And in declaring Brecht insufficient, the majority consigns future habeas courts to a regimen of make-work. Now those courts will have to jump through AEDPA's hoops as well, even though that extra analysis will never lead to a different result. I respectfully dissent from that pointless demand.