Glossip Swirl: Supreme Court will hear criminal discovery obligation case

In its October 2024 Term, the United States Supreme Court will hear Glossip v. Oklahoma, No. 22-7466. Glossip will give the High Court an opportunity to address three issues regarding the criminal discovery obligations of prosecutors as well as an issue of judicial administration.

In January 1997, the owner of the Best Budget Inn in Oklahoma City was found bludgeoned to death with a baseball bat in one of the motel’s rooms. The motel’s handyman admitted to being the bludgeoner. But the handyman told police he was acting at the direction of the motel’s manager (Richard Glossip) as part of a Machiavellian scheme to rise from motel manager to motel owner; the handyman said Glossip had paid him to carry out the killing. The handyman was the sole witness implicating Glossip as the architect of the killing, and the handyman’s account of the facts differed with each telling.

The handyman pled guilty to the hotel owner’s murder and was sentenced to life in prison without the possibility of parole.

The case against Glossip went to trial and, eventually, a retrial. At the retrial, the prosecutor called the handyman as a witness. The handyman testified that he “never seen no psychiatrist or anything”; he also testified that he had taken lithium while in jail, but did not know why. The jury ultimately convicted Glossip of murder and recommended the death penalty.

In September 2022, the prosecutor’s office released the prosecuting attorney’s file. That file revealed that the handyman had been seeing a psychiatrist, and that the psychiatrist had diagnosed him with bipolar disorder and had prescribed him lithium.

Glossip sought post-conviction relief. He argued that the prosecutor’s failure to disclose the handyman’s psychological diagnosis and medication violated Brady v. Maryland, 373 U.S. 83 (1963), and that the prosecutor’s failure to correct the handyman’s denial of obtaining mental health treatment or medication violated Napue v. Illinois, 360 U.S. 264 (1959). The State of Oklahoma conceded error.

The Oklahoma Court of Criminal Appeals rejected the parties’ mutual concession to vacate Glossip’s murder conviction and death sentence, ruling that Glossip’s Brady claim could have been presented in an earlier post-conviction challenge and that his Napue claim lacked merit because the handyman’s testimony (1) was “not clearly false” because the handyman “was more than likely in denial of his mental health disorders,” and (2) was not “material” because it would not reasonably have led to a different outcome. Glossip v. State, 529 P.3d 218, 226-27 (Okla. Crim. App. 2023).

Glossip presents two issues regarding the scope of a prosecutor’s Brady obligation.

Brady provides that a prosecutor violates a criminal defendant’s right to due process if they do not disclose to the defense evidence that is both (1) “favorable” and (2) “material to … guilt or punishment.” Skinner v. Switzer, 562 U.S. 521, 536-37 (2011); Brady, 373 U.S. at 87; Giglio v. United States, 405 U.S. 150, 153-54 (1972) (evidence is “suppress[ed]” if the prosecution does not disclose it). “Favorable” evidence means evidence (1) that is exculpatory, because it tends to exonerate the defendant, or (2) that impeaches the prosecution’s case or its witnesses. Turner v. United States, 582 U.S. 313, 323-24 (2017).

The first Brady issue that Glossip presents is: When does a witness’s mental health diagnosis and medication regimen constitute impeachment evidence that must be disclosed under Brady?

The federal and California courts generally agree that mental health information can constitute impeachment evidence if it casts doubt on the witness’s ability to perceive, recall, or narrate the events about which the witness is testifying. Browning v. Trammell, 717 F.3d 1092, 1105 (10th Cir. 2013); Wilson v. Beard, 589 F.3d 651, 666 (3d Cir. 2009); People v. Gurule, 28 Cal. 4th 557, 591-92 (2002). Thus, evidence that a witness’s psychological condition causes “memory deficits, magical thinking, blurring of reality and fantasy, and projection of blame onto others” (as in Browning) or causes blackouts (as in Wilson) falls within the ambit of this type of impeachment evidence, which somewhat tracks the common law requirements for basic witness competency. E.g., Mitchell v. Kemna, 109 F.3d 494, 496 (8th Cir. 1997).

Do mental health diagnoses and medication regimens constitute impeachment evidence in situations beyond these competence-focused concerns?

California courts have viewed such broader use skeptically, noting that “[t]he use of the results of a psychological examination to impeach a witness’s credibility is, in the main, disfavored.” People v. Ayala, 23 Cal.4th 225, 263 (2000); People v. Marshall, 13 Cal.4th 799, 835 (1996).

But many federal courts have taken a different view. Some have ruled that diagnoses can impeach because they can constitute evidence of bias. For instance, in Fuentes v. Griffin, 829 F.3d 233, 241, 247-48 (2d Cir. 2016), the court held that a rape victim’s psychological diagnosis of depression and suicidal ideation, including “‘frequent crying spells,’” was subject to disclosure as impeachment evidence under Brady because it could show bias and an ulterior motive of malice or vindictiveness. Others have ruled that diagnoses can impeach because they can constitute evidence of propensity, whether it be a propensity to lie (due to a history showing deceitful and manipulative behavior), Gonzalez v. Wong, 667 F.3d 965, 982-84 (9th Cir. 2011), or a propensity to engage in violence (as argued by Glossip here).

These cases have yet to grapple with the broadest use of such information—namely, whether being diagnosed with a psychological condition, by itself and without more, is a basis to believe a witness less credible (and thus a basis to require disclosure of all psychological diagnoses of witnesses).

The second Brady issue that Glossip presents is a latent one that has been lingering in plain sight for a while now. Most of the Supreme Court’s cases hold that evidence is “material” if it would have affected the outcome of the prosecution. Cone v. Bell, 556 U.S. 449, 469-70 (2009); Turner, 582 U.S. at 324; United States v. Bagley, 473 U.S. 667, 682 (1985). But a few have held that evidence is “material” merely if it could. Wearry v. Cain, 577 U.S. 385, 392 (2016); Brown v. Louisiana, 143 S. Ct. 886, 887 (2023) (citing Wearry). This “could”-based standard is more permissive than a “would”-based standard. E.g., Mastracchio v. Vose, 274 F.3d 590, 601 (1st Cir. 2001). Here, Glossip urges the Court to apply the more defendant-friendly “could” formulation.

If the Supreme Court elects to reach the merits of Glossip’s Brady claim rather than affirm on procedural grounds, the Court could provide guidance on these questions. (Whether it will is still anybody’s guess.)

Glossip also presents an issue regarding the scope of a prosecutor’s duty under Napue.

Under Napue, a prosecutor violates a criminal defendant’s due process rights if, as pertinent here, (1) the prosecutor allows false testimony to go uncorrected, (2) the prosecutor knows or should know that the testimony is false, and (3) the false testimony is “material” (under the more defendant-friendly “could” standard). Fong Soto v. Ryan, 760 F.3d 947, 957-58 (9th Cir. 2014); Mastracchio, 274 F.3d at 601.

The Oklahoma Court of Criminal Appeals seemed to add a fourth requirement to Napue’s test—namely, that the witness must intend that his testimony be false. Other cases have disclaimed any such requirement. Hein v. Sullivan, 601 F.3d 897, 911 n.11 (9th Cir. 2010).

Glossip gives the Supreme Court an opportunity to address this split.

Glossip lastly presents an issue about the power of courts to independently assess the merits of issues. In his petition for certiorari, Glossip argues that the Oklahoma Court of Criminal Appeals violated his right to due process by disregarding the parties’ joint concession that his conviction be vacated. Typically, appellate courts give considerable weight to a party’s concession that its opponent’s position has merit, but are not bound by such a concession and independently evaluate the merits of the appeal. E.g., People v. Therman, 236 Cal. App.4th 1276, 1278 (2015) (rejecting concession of error).

Of all these issues, the two Brady issues have the broadest potential impact—the first defining the boundary of accommodation between a witness’s ability to seek mental health treatment privately and a criminal defendant’s right to impeach that witness, and the second defining “materiality” and hence the scope of a prosecutor’s obligation to disclose evidence to the criminal defense.

All four issues are addressed in the merits briefing. Which of this swirl of issues the Court will ultimately address remains to be seen.