Breaking News

The evidentiary challenges of confessions in co-defendant trials

A side view of the Supreme Court building in pink sunlight.

The Sixth Amendment’s confrontation clause guarantees defendants in criminal cases the right to be “confronted with the witnesses against” them. In trials with multiple defendants, this complicates the introduction of a confession by a non-testifying defendant that incriminates a co-defendant. On Wednesday, the Supreme Court will hear argument in Samia v. United States, a case that will determine whether it is sufficient to simply modify confessions in these circumstances by replacing the co-defendant’s name with a neutral reference, like “other guy.”

“Crimes worthy of a James Bond villain”

The case arises from the federal prosecution of “a transnational criminal organization,” responsible for “an array of crimes worthy of a James Bond villain.” Among those crimes, the prosecution alleged, the organization hired Carl Stillwell and Adam Samia to murder Catherine Lee, a real estate broker in the Philippines.

After Lee’s murder, Stillwell admitted to a Drug Enforcement Administration agent that he was driving a van with Samia in it when Samia shot Lee, and that Stillwell was later paid $20,000 to $30,000, presumably for the murder.

The prosecution tried Stillwell and Samia (and the person alleged to have hired them) in a single trial. The judge required that Stillwell’s out-of-court statements – admissible only against Stillwell – be modified to exclude any explicit references to Samia. Consequently, the DEA agent testified that Stillwell confessed to being present for the murder with an “other person,” and that “the other person he was with” pulled the trigger. The judge then instructed the jurors that Stillwell’s confession was “only admissible as to Mr. Stillwell and not as to Mr. Samia.”

Jury instructions and the Bruton rule

A confession in a trial with multiple defendants is just one of many scenarios in which evidence is admissible for one purpose but not another. The usual solution is to admit the evidence but instruct the jury on the limited purpose for which it can be used. While the effectiveness of jury instructions is a matter of dispute outside the courtroom, judges indulge the “almost invariable assumption of the law that jurors follow their instructions.”

Even the courts do not indulge this assumption, however, when one defendant’s confession not only reveals that defendant’s guilt but “powerfully incriminat[es]” another co-defendant. The Supreme Court held in Bruton v. United States that, in such a circumstance, “the risk that the jury will not, or cannot, follow instructions is so great, and the consequences of failure so vital to the defendant, that the practical and human limitations of the jury system cannot be ignored.”

The specifics of Bruton illustrate the point. In that case, a postal inspector testified at a joint trial that William Evans confessed to “committing the robbery with Bruton.” The Supreme Court held that a limiting instruction was insufficient to protect Bruton’s rights because the jury would likely disregard the instruction and consider Evans’ confession (properly) as evidence of Evans’ guilt, but also (improperly) as evidence of Bruton’s.

Does context matter?

Given this backdrop, the question in Samia appears simple: Did Stillwell’s redacted confession “powerfully incriminate” Samia? If so, the jury instruction was insufficient, and Samia’s confrontation clause rights were violated since Stillwell did not testify.

The case is not simple, however, due to confusion in the lower courts about what a trial court can consider when deciding whether one co-defendant’s confession powerfully incriminates another. The U.S. Court of Appeals for the 2nd Circuit, in rejecting Samia’s appeal, suggested that the trial court should consider only the confession itself: “[A] court views the redacted statement ‘separate and apart from any other evidence admitted at trial.’” Samia, by contrast, argues that the courts must view the confession alongside the other trial evidence – evidence that, in this case, strongly suggested that the “other guy” Stillwell referenced was Samia.

The primary justification for excluding other evidence from consideration of a confession’s admissibility is administrability. Rulings on this kind of evidence typically occur before trial, when it may not even be possible to consider the full context, since some of the trial evidence may be unforeseeable. Bowing to these concerns, Samia’s brief does not push courts to consider all context, but only “those aspects of the case that are either knowable in advance of trial or within the prosecution’s control.”

Of course, administrability concerns are not supposed to trump constitutional rights, a point the Supreme Court emphasizes in other confrontation clause cases. Thus, if Samia is right that context should be considered, his proposed limitation seems strained. If a non-testifying defendant’s statement becomes powerfully incriminating against a co-defendant, through context or otherwise, the defendant becomes a “witness against” the co-defendant and the confrontation clause is violated.

Still, administrability holds a special place in the Bruton line of cases. The prosecution can always avoid confrontation clause violations by declining to use confessions in joint trials or by trying defendants separately. It is only because the Supreme Court has emphasized the value of joint trials (and confessions) to the efficient administration of justice that Bruton dilemmas exist at all.

All of that said, the purported conflict in the lower courts on the context question may be overblown. It seems clear – as a “friend of the court” brief from 12 law professors stresses – that some context must be considered. For example, if the confession is, “I committed the murder with my roommate,” that confession only incriminates the co-defendant once the evidence at trial establishes that they were roommates. Yet it would be odd if the confession was deemed admissible in a joint trial because it did not incriminate the codefendant on its face. The Supreme Court said as much in Gray v. Maryland, its most recent case addressing Bruton redactions. There, the court explained that confessions that refer to codefendants by a “nickname [or] specific descriptions” still violate the Bruton rule even though they only incriminate when linked to other evidence at the trial.

But incorporating some degree of context into Bruton analysis does not mean Samia prevails.

Even if the Supreme Court rebukes the 2nd Circuit for failing to consider context, that only returns us to the difficult line-drawing exercise where we began: When does a defendant’s confession so “powerfully incriminate” a co-defendant that a jury instruction becomes ineffective? The Supreme Court has already given us the answer in two of the three most frequent redaction scenarios: (1) When the confession explicitly names a codefendant, Bruton; and (2) when it implicitly names a codefendant via obvious redaction (“Me and [Redacted] committed the murder”) or a nickname, Gray.

Samia is the court’s chance to round out the trilogy, addressing the third scenario: When a redacted confession includes a neutral reference (“other person”), but the jury can nevertheless discern from other evidence that the “other person” is the co-defendant.

Samia proposes that third-scenario questions always come down to degree: courts must consider the other evidence in assessing these confessions and exclude those where the neutral reference, in context, powerfully incriminates the defendant. The Supreme Court may well adopt this approach. Or the justices may accept that context matters, but conclude that neutral references, while incriminatory, will rarely be so “powerfully incriminatory” that jury instructions cannot suffice.

Yes, but what would the founders say?

The 1968 Bruton opinion contains no historical analysis, and the post-Bruton case law and the parties’ briefing in this case largely follow Bruton’s lead. But since 2004, the Supreme Court’s confrontation clause cases purport to turn on historical analysis. Consequently, the justices could reframe the question as a historical rather than logical inquiry. In fact, the court recently pointed to history as the primary place to look for exceptions to the confrontation right: in his 2015 opinion in Ohio v. Clark, Justice Samuel Alito wrote that the court had “recognized that the Confrontation Clause does not prohibit the introduction of out-of-court statements that would have been admissible in a criminal case at the time of the founding.”

The United States nods toward this possibility in a short section towards the end of its brief. There, it cites a handful of treatises that suggest that, at the time of the founding, courts tolerated the introduction of a co-defendant’s confession along with a jury instruction. Importantly, this suggests not just that the redactions in this case were sufficient, but that they were unnecessary.

Sensing danger, Samia beefs up his historical analysis in his reply brief. He points out weaknesses in the sparse historical record and critical differences in criminal trials of that era, and he concludes with the indisputable point that the “historical evidence has nothing to say about the specific question presented here.”

The United States does not push its historical argument to its conclusion: that Bruton itself is ahistorical and should be overturned – an inconsistency Samia flags as suggesting that the United States’ historical argument is mere “posturing” since “the government does not have the courage to urge that Bruton be overruled.” Whether the justices will defer to the United States’ temerity and, if not, how they will view the (sparse) historical record is a wildcard in this case, and something to look for at oral argument.

Recommended Citation: Jeffrey Bellin, The evidentiary challenges of confessions in co-defendant trials, SCOTUSblog (Mar. 27, 2023, 1:25 PM),