
Maya James
Partner Keker, Van Nest & Peters LLP

Max Alderman
Associate Keker, Van Nest & Peters LLP
On Feb. 25, 2025, the
U.S. Supreme Court issued its long-awaited opinion in Glossip v. Oklahoma
(Barrett, J. concurring and dissenting in part; Thomas, J. & Alito, J.,
dissenting; Gorsuch, J., taking no part), overturning Richard Glossip's murder
conviction and ordering a new trial because the State violated its
constitutional obligation to correct the false testimony of its star witness,
Justin Sneed. This was not Glossip's first appearance before the Court. Since
his 1998 conviction, Glossip has consistently asserted his innocence, receiving
a new trial in 2004, as well as filing several unsuccessful habeas petitions in
state and federal court. In 2022, his decades-long efforts drew the attention
of the Oklahoma Legislature, which commissioned a report casting "grave doubt
as to the integrity of Glossip's murder conviction." That, in turn, prompted the state to disclose
improperly withheld evidence from Glossip's trial, including a state
prosecutor's handwritten notes indicating she knew Sneed had lied under oath
yet failed to correct it. When Glossip filed his latest habeas petition--this
time seeking relief under both Brady v. Maryland, 373 U.S. 83 (1963) and
Napue v. Illinois, 360 U.S. 264 (1959)--he had the public support of
Oklahoma's attorney and solicitor generals. And his story generated national
headlines, as numerous commentators queried how the state could put a man to
death when his trial had been plagued by admitted constitutional deficiencies.
Public opinion aside,
Glossip did not necessarily face a receptive panel. The prior year, the Court
summarily denied a similar request for review from Toforest
Johnson, who sought post-conviction relief under Brady because the
prosecutor failed to disclose that its star witness had received a $5,000
reward for her testimony. That denial followed on the heels of the Court's
refusal to review David Brown's and Davel Chinn's convictions, both of whom had
also raised Brady claims. In Brown's case, the prosecution withheld the
co-defendant's confession that he, not Brown, was the one who made the decision
to kill the victim; and in Chinn's case, "there [was] no dispute" that the
State had suppressed evidence that its key witness had an intellectual
disability that could affect his ability to testify accurately. All told, the
Court's denial of these applications (among many others) did little more than
cement its current reputation for "contentment with capital punishment," and
its "assumption that last-minute claims" for death penalty relief are "abusive
even if they might succeed on the merits."
See Stephen Vladeck, The Shadow Docket 149 (2023).
So why did Glossip
succeed where so many other petitioners have failed? In our view, that answer may turn, in part,
on the strength of Glossip's affirmative prosecutorial misconduct claim under Napue
and the Court's burgeoning desire to distinguish Napue from Brady.
For decades, the
Court has lumped Brady and Napue together, given that Brady
"has its roots in a series of cases dealing with convictions based on the
prosecution's knowing use of perjured testimony." United States v. Bagley, 473 U.S. 667,
679 n.8 (1985) (citing Mooney v. Holohan, 294 U.S. 103 (1935) and Napue,
360 U.S. at 269). Indeed, during oral argument, the Glossip panel's
questioning largely treated the two claims as coextensive. But Brady and
Napue are not identical, and over the years, the Court has quietly
declined to treat them as such.
Since deciding Brady
in 1963, the Court has repeatedly cabined its broad rule prohibiting
the suppression of favorable and material evidence during criminal proceedings.
For example, it has held that Brady does not extend to postconviction
proceedings, District Attorney's Office v. Osborne, 557 U.S. 52, 68-69
(2009); that defendants bear some affirmative burden in requesting favorable
pretrial evidence to sustain a Brady claim, United States v. Agurs,
427 U.S. 97, 109-10 (1976); and that Brady's materiality standard
requires a showing that withheld evidence must, when evaluated in the context
of the entire record, have a "reasonable probability" of generating a different
result, see Turner v. United States, 582 U.S. 313, 324-25 (2017). In
many respects, this last alteration is the most important: By heightening Brady's
materiality standard, the Court has, effectively, weakened its role as a
tool to challenge constitutionally suspect criminal proceedings.
The same cannot be
said for Napue. Although Napue has only infrequently appeared in
the Court's jurisprudence, the Court has left in place its narrow directive
that the state "may not knowingly use false evidence, including false
testimony, to obtain a tainted conviction" and cannot allow such evidence or
testimony "to go uncorrected when it appears." 360 U.S. at 269. And critically,
the Court has conclusively affirmed that a petitioner can show materiality for
Napue purposes "if there is any reasonable likelihood that the false
testimony could have affected the judgment of the jury," United States v.
Agurs, 427 U.S. 97, 103 (1985) (emphasis added)--a standard that has, in
practice, created less openings for the state to brush off misconduct as
immaterial. Put simply, while the Court facially "adhere[s] to the principles
of Brady and Napue," Moore v. Illinois, 408 U.S. 786, 798
(1972), these two households, both alike in dignity, have long remained
divided.
Therefore, it should
not come as too much of a surprise that, in a case where even the trial
prosecutor attested that he was "horrifie[d]" by the
extent of the constitutional violations, Op. at 21, the Court took the
opportunity to rectify prosecutorial misconduct. In so doing, the Court
explicitly distinguished Glossip's Napue claim from his Brady claim.
Op. at 12 n.5. To that end, the Court focused not on the prosecutor's withheld
(and arguably ambiguous) interview notes that formed the basis of the Brady
claim, but on the prosecutor's knowing failure to correct Sneed's false
testimony. Op. at 17. Specifically, when Sneed testified against Glossip at
trial, stating that he had been mistakenly prescribed lithium for a cold and
had not seen a psychiatrist, that testimony was false. As the prosecution
"almost certainly" knew, Sneed had been treated by Dr. Larry Trombka and been prescribed lithium to treat his bipolar
disorder. Op. at 17. Given that "Sneed's testimony was the only direct evidence
of Glossip's guilt of capital murder," the majority concluded that such
evidence undermining Sneed's credibility "was necessarily determinative." Op.
at 19.
This result may only have been possible under the less onerous
materiality standard of Napue. As Justice Thomas's dissent highlights,
there were plausible reasons to think that Sneed's challenged testimony would
not have changed the jury's verdict (not least because the defense already
knew that Sneed suffered from an "atypical mood swing disorder," and chose
not to raise it). Dissent at 26. Indeed, this is precisely the line of
reasoning that petitioners have challenged--and which this Court has repeatedly declined
to disturb--under Brady's "reasonable probability" standard. It is, of
course, too early to say whether Glossip marks a true shift in the
Court's death penalty jurisprudence. But the majority's decision may
well reflect a new direction for the Court--one that focuses its attention on
the state's clear, affirmative, and knowing misconduct, while narrowing the
path for claims that, like Brady, increasingly place the burden on the
defendant to prove that the particular unfairness was
dispositive.