on Apr 22, 2022
at 10:35 am
In Brown v. Davenport, the Supreme Court held on Thursday that a federal court cannot grant habeas relief to a convicted state prisoner whose constitutional rights were violated at trial unless that prisoner satisfies both the judicially-created Brecht v. Abrahamson harmless-error standard and the statutorily-created requirements of the Antiterrorism and Effective Death Penalty Act. The decision will make it marginally harder for state prisoners to obtain relief, because they will have to jump over two hurdles rather than one, but it is unlikely to affect the outcome in most cases. Perhaps the most important part of the case is how Justice Neil Gorsuch, writing for a six-member majority, sets the stage for future narrowing of the scope of the Great Writ – a choice that prompted a fiery dissent from Justice Elena Kagan (joined by Justices Stephen Breyer and Sonia Sotomayor).
Everyone agreed that Ervine Lee Davenport’s due process rights were violated when he was visibly shackled in front of the jury while facing first-degree murder charges. (The Supreme Court held in 2005 in Deck v. Missouri that such shackling practices violate a defendant’s due process rights.) The only issue was whether Davenport should get federal habeas relief because of that violation. The answer to that question turned on the relationship between a Supreme Court decision and a congressional statute. In Brecht v. Abrahamson, the Supreme Court adopted a harmless-error standard for federal habeas review, holding that a habeas petitioner is not ordinarily entitled to relief unless a constitutional violation had a “substantial and injurious effect or influence on the verdict.” Three years later, Congress passed AEDPA, which provides that no state prisoner may obtain habeas relief on the basis of a constitutional claim that has been “adjudicated on the merits in State court proceedings” unless the state court adjudication “resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States” or involved an unreasonable determination of the facts.
In Davenport’s case, the U.S. Court of Appeals for the 6th Circuit had found Brecht satisfied and, as a result, held that Davenport need not also satisfy AEDPA to obtain relief because, according to language in two Supreme Court cases (Fry v. Pliler and Davis v. Ayala), the Brecht inquiry is a more demanding one that “subsumes” the AEDPA inquiry. Gorsuch, writing for himself, the chief justice, and Justices Clarence Thomas, Samuel Alito, Brett Kavanaugh, and Amy Coney Barrett, disagreed.
Gorsuch began by outlining his view of the historical scope of federal habeas review. He first laid out this historical interpretation in a concurring opinion a year ago in Edwards v. Vannoy, but now he has a majority willing to sign on to it. Under this view, habeas was historically a writ used to test the legality of an executive detention but not a means of challenging a final judgment of conviction, unless the judgment was issued by a court that lacked jurisdiction over the defendant or his offense. Those who ascribe to this more narrow view of habeas history believe that habeas was and is not meant to be a tool for error correction and argue that, originally, it was not so used. That changed, according to the majority, when the Supreme Court decided Brown v. Allen in 1953 and permitted federal courts to re-adjudicate state prisoners’ post-conviction constitutional claims. This shift, Gorsuch wrote, unmoored habeas from its traditional function, and “[f]ull-blown constitutional error correction became the order of the day.” Gorsuch then explained how the Supreme Court and Congress’ subsequent attempts to limit the scope of federal habeas review of state court criminal convictions – including the adoption of Brecht’s harmless-error standard and the enactment of AEDPA – were a necessary response to the flood of petitions courts received after Brown v. Allen.
None of this “amateur,” “law-chambers history,” as Kagan called it in her dissent, was necessary to the resolution of Davenport’s case. Kagan recognized it as an attempt by the majority to plant seeds now to “yield more succulent fruit in cases to come.” In other words, Gorsuch laid out his vision (joined by five other justices) of a much narrower federal habeas writ in the hopes of limiting the scope of the writ in future cases. According to the dissenters, however, the historical picture that the majority presents is wrong. Kagan presented a number of cases from the mid-19th century to show that federal courts were extending habeas relief to convicted prisoners long before Brown v. Allen was decided. And she quoted language from articles, treatises, and modern Supreme Court cases acknowledging that history. Whose view of the history will prevail in future cases remains to be seen, but it is significant that Gorsuch has now managed to get his more constricted view of the scope of the Great Writ into a majority decision.
Once the majority turned to the merits of Davenport’s case, it emphasized that AEDPA represented “a sea change in federal habeas law” that was designed to restrict the availability of relief and noted that AEDPA is “entirely different in kind” from the Brecht analysis. According to the majority, “where AEDPA asks whether every fairminded jurist would agree that an error was prejudicial, Brecht asks only whether a federal habeas court itself harbors grave doubt about the petitioner’s verdict.” Additionally, the majority noted, the legal materials that a court may consult when addressing the two inquiries are distinct: AEDPA requires the court to focus on U.S. Supreme Court holdings that existed at the time of the relevant state court decision whereas, under Brecht, the courts can draw on circuit precedent, dicta in Supreme Court cases, and cases that were decided after the relevant state court decision.
The majority also disagreed with Davenport and the dissenters’ claims that precedent required them to adopt the 6th Circuit’s position. Neither Fry v. Pliler, nor Davis v. Ayala, addressed the circumstances presented in Davenport’s case. As for the language in both decisions that Brecht “subsumes” AEDPA, the majority admitted that there are some cases where Brecht subsumes AEDPA – for example, if a state court has not adjudicated the petitioner’s claims on the merits and AEDPA falls away (as in Fry) or if a petitioner fails to satisfy Brecht and there is no need to apply AEDPA (as in Ayala). But, the majority noted, none of this means that Brecht always subsumes AEDPA.
The majority then assumed that Davenport could satisfy Brecht (as the 6th Circuit had held) and addressed whether he could also satisfy AEDPA. This first required the court to determine what was the last reasoned state court decision. The majority noted that the Michigan Supreme Court had denied leave to appeal and that such discretionary denials are not typically adjudications of the underlying merits of the claims. As a result, it decided that the Michigan Court of Appeals’ decision was the last reasoned state court decision. It then held that nothing in that decision was contrary to or involved an unreasonable application of its holdings. The state court had properly noted that Davenport would be entitled to relief under Chapman v. California unless the state could prove that the shackling error was harmless beyond a reasonable doubt. And, though reasonable jurists may disagree on whether the evidence was sufficient to meet that standard, the majority did not feel that every fairminded jurist would come to the same conclusion. Thus, the state court’s determination that the error was harmless was a reasonable one.
Nor was the state court’s decision to consider the post-trial testimony from jurors about whether the shackling influenced their decisions unreasonable under Holbrook v. Flynn. In Holbrook, the Supreme Court rejected reliance solely on jurors’ assessments of whether they were prejudiced when addressing whether courtroom security protocols were so inherently prejudicial as to violate due process. But, the majority pointed out, Holbrook was not a harmless-error case, and AEDPA does not permit litigants to try to extend precedent into new contexts. Additionally, the majority noted that Holbrook was distinguishable because the Michigan Court of Appeals did not rely solely on the jurors’ testimony in reaching its decision and because Holbrook involved pretrial, speculative testimony about potential prejudice whereas this case involved post-trial testimony from jurors about whether they were prejudiced.
Because Davenport could not satisfy both AEDPA and Brecht, the majority concluded that he was not entitled to habeas relief and reversed the 6th Circuit.
Kagan’s dissent cited language in both Fry and Ayala that repeatedly said that Brecht subsumes an AEDPA/Chapman inquiry and argued that the majority’s decision to the contrary “makes no sense,” because it will result in needless work for federal courts. As the dissenters explained, the Brecht standard is actually harder for a prisoner to meet than AEDPA as applied to Chapman. Even with the AEDPA overlay, it is easier for a defendant to satisfy the defendant-friendly Chapman harmless-error standard than the more-exacting Brecht requirement of “actual prejudice.” So, if a habeas court confines itself to using AEDPA-approved materials, the dissenters wrote, it should only need to find the Brecht standard met to grant relief.
Kagan noted that there are no cases in which a court has found that a petitioner satisfied Brecht but not AEDPA/Chapman. (And Davenport’s case is not an appropriate example in her view since the majority never decided that Davenport satisfied Brecht but only assumed so.) Thus, she wrote, “today’s holding will make no difference to habeas outcomes.” It will only create more work for the federal courts.
Even if Gorsuch’s opinion in Davenport will not change the outcome in many habeas cases, it remains to be seen if the seeds he has planted about narrowing the scope of habeas review will “yield more succulent fruit in cases to come.”