Amicus Curiae Brief — Adam Samia, aka Sal, aka Adam Samic, Petitioner v. United States

No. 22-196

In the Supreme Court of the United States

_______________

ADAM SAMIA,

Petitioner,

v.

UNITED STATES OF AMERICA,

Respondent

_________________________

On Writ of Certiorari to the

United States Court of Appeals for the Second Circuit

___________________

BRIEF FOR NATIONAL ASSOCIATION OF

FEDERAL DEFENDERS AS AMICUS CURIAE

IN SUPPORT OF PETITIONER

___________________

David Watnick

PERKINS COIE LLP

1120 N.W. Couch Street

Tenth Floor

Portland, OR 97209-4128

Keith M. Donoghue

Jessica Stengel

Stacie Fahsel

Members, Amicus

Committee

NATIONAL ASSOCIATION OF

FEDERAL DEFENDERS

Michael R. Huston

Counsel of Record

PERKINS COIE LLP

700 Thirteenth Street N.W.

Suite 800

Washington, DC 20005-3960

(202) 434-1630

mhuston@perkinscoie.com

Luci D. Davis

PERKINS COIE LLP

2901 North Central Avenue

Suite 2000

Phoenix, AZ 85012-2788

TABLE OF CONTENTS

Interest of Amicus Curiae ................................................... 1

Summary of Argument ........................................................ 2

Argument ............................................................................... 3

A. This Court’s precedent establishes that whether

the introduction of a non-testifying codefendant’s

out-of-court confession violates the Sixth

Amendment requires a contextual inquiry ................. 4

B. The lower courts’ experience shows that

analyzing Bruton issues in light of a trial’s

context is an administrable rule ................................... 7

1. Courts consider the number of defendants,

how many participants the confession or

evidence identifies, and the complexity of the

facts ............................................................................ 8

2. Courts consider the frequency and type of

redactions ................................................................ 12

3. Courts consider the strength of the inferences

from a confession, and how those inferences

relate to other evidence ......................................... 13

4. Courts consider the prosecution’s questioning

and arguments ........................................................ 15

C. Considering trial context in Bruton challenges

will not disrupt the jury system ................................. 18

1. Procedures for resolving Bruton issues are

well-established ...................................................... 18

2. Considering context gives effect to the Sixth

Amendment without unduly constraining

prosecutors’ ability to pursue joint trials............ 21

Conclusion ........................................................................... 23

ii

TABLE OF AUTHORITIES

CASES

Page(s)

Brown v. Superintendent Greene SCI,

834 F.3d 506 (3d Cir. 2016) ............................... 15, 16, 17

Bruton v. United States,

391 U.S. 123 (1968) ............... 2, 3, 4, 5, 6, 7, 8, 13, 14, 15,

16, 17, 18, 19, 20, 21, 22

Davis v. State,

528 S.E.2d 800 (Ga. 2000) ............................................. 16

Eley v. Erickson,

712 F.3d 837 (3d. Cir. 2013) .................................... 10, 11

Foxworth v. St. Amand,

570 F.3d 414 (1st Cir. 2009) ........................................ 6, 8

Gray v. Maryland,

523 U.S. 185 (1998) .............. 3, 4, 5, 6, 7, 8, 12, 15, 18, 21

Greer v. Miller,

483 U.S. 756 (1987) ........................................................ 17

Harrington v. California,

395 U.S. 250 (1969) ........................................................ 21

Holland v. Attorney Gen. of N.J.,

777 F.2d 150 (3d Cir. 1985) ........................................... 19

Jefferson v. State,

198 S.W.3d 527 (Ark. 2004) .......................................... 11

Neal v. State,

806 So.2d 1151 (Miss. Ct. App. 2002) .......................... 11

Ohio v. Clark,

576 U.S. 237 (2015) ........................................................ 17

Priester v. Vaughn,

382 F.3d 394 (3d. Cir. 2004) ............................................ 9

iii

Richardson v. Marsh,

481 U.S. 200 (1987) ............................ 3, 4, 5, 6, 15, 18, 22

State v. Johnson,

111 P.3d 784 (Or. Ct. App. 2005).................................. 13

State v. McDonald,

771 S.E.2d 840 (S.C. 2015) ............................................ 16

State v. Medina,

48 P.3d 1005 (Wash. Ct. App. 2002) ............................ 11

Turner v. United States,

137 S. Ct. 1885 (2017) .................................................... 17

United States v. Campbell,

986 F.3d 782 (8th Cir. 2021),

cert . denied, 142 S. Ct. 751 (2022) .............................. 20

United States v. Ceballos,

789 F.3d 607 (5th Cir. 2015) ......................................... 19

United States v. Coleman,

349 F.3d 1077 (8th Cir. 2003) ....................................... 20

United States v. Damra,

621 F.3d 474 (6th Cir. 2010) ......................................... 20

United States v. Davis,

534 F.3d 903 (8th Cir. 2008) ......................................... 16

United States v. de Leon-De La Rosa,

17 F.4th 175 (1st Cir. 2021) ............................................ 7

United States v. Field,

756 F.3d 911 (6th Cir. 2014) ......................................... 20

United States v. Gibson,

875 F.3d 179 (5th Cir. 2017) ......................................... 15

United States v. Green,

648 F.3d 569 (7th Cir. 2011) ......................................... 14

United States v. Hernandez,

330 F.3d 964 (7th Cir. 2003) ........................................... 8

iv

United States v. Hoover,

246 F.3d 1054 (7th Cir. 2001) ......................................... 7

United States v. Javell,

695 F.3d 707 (7th Cir. 2012) ......................................... 19

United States v. Lung Fong Chen,

393 F.3d 139 (2d Cir. 2004) ........................................... 15

United States v. Macias,

387 F.3d 509 (6th Cir. 2004) ............................. 13, 14, 21

United States v. Moore,

651 F.3d 30 (D.C. Cir. 2011) ......................................... 19

United States v. Nash,

482 F.3d 1209 (10th Cir. 2007) ..................................... 13

United States v. Padilla-Galarza,

990 F.3d 60 (1st Cir. 2021) ...................................... 19, 20

United States v. Park,

421 U.S. 658 (1974) ........................................................ 17

United States v. Peterson,

140 F.3d 819 (9th Cir. 1998) ......................................... 15

United States v. Ramos-Cardenas,

524 F.3d 600 (5th Cir. 2008) ................................... 13, 21

United States v. Schwartz,

541 F.3d 1331 (11th Cir. 2008) ................................. 7, 15

United States v. Straker,

800 F.3d 570 (D.C. Cir. 2015) ..................................... 7, 8

United States v. Williams,

429 F.3d 767 (8th Cir. 2005) ......................................... 12

Vazquez v. Wilson,

550 F.3d 270 (3d Cir. 2008) ........................... 9, 10, 11, 18

Wynn v. United States,

241 A.3d 277 (D.C. Ct. App. 2020) ............................... 16

In the Supreme Court of the United States

_______________

NO. 22-196

ADAM SAMIA,

v.

Petitioner,

UNITED STATES OF AMERICA,

Respondent

_________________________

On Writ of Certiorari to the

United States Court of Appeals for the Second Circuit

___________________

BRIEF FOR NATIONAL ASSOCIATION OF

FEDERAL DEFENDERS AS AMICUS CURIAE

IN SUPPORT OF PETITIONER

____________

INTEREST OF AMICUS CURIAE 1

The National Association of Federal Defenders

(NAFD) is a nationwide volunteer organization of attorneys who work for federal public defender offices and

community defender organizations authorized under the

Criminal Justice Act, 18 U.S.C. § 3006A. Each year, federal defenders represent tens of thousands of indigent

criminal defendants in federal court. Those representations include defendants prosecuted in joint trials where

the introduction of a codefendant’s out-of-court confession

implicates the defendant’s rights under the Sixth AmendPursuant to this Court’s Rule 37.6, counsel for amicus curiae

states that no counsel for any party authored this brief in whole or in

part, and no party or counsel for a party, or any other person other

than amicus curiae and its counsel, made a monetary contribution to

fund the preparation or submission of this brief.

1

2

ment. NAFD members therefore have experience with,

and a substantial interest in, this Court’s resolution of the

question presented.

SUMMARY OF ARGUMENT

When the prosecution seeks to introduce a nontestifying codefendant’s testimonial confession in a joint

trial, the Sixth Amendment requires the trial court to apply—and requires the prosecution to heed—the fundamental principle that context matters. For at least three

reasons, this Court should reaffirm that command here

and vacate the judgment of the court of appeals.

A. First, this Court’s precedents establish that careful attention to context is necessary to determine whether

the introduction of a non-testifying codefendant’s confession violates the Confrontation Clause. This Court has

already recognized that aspects of a trial’s context are relevant to that analysis, including the number of defendants

and other alleged participants in a case, as well as the

prosecutor’s arguments and line of examination through

which the confession is admitted. How a jury is likely to

understand a confession, and whether a jury instruction

to consider evidence for one purpose but not another is

unreasonable under the circumstances, are necessarily

contextual inquiries. By so holding here, the Court will

break no new ground but rather merely confirm the prevailing interpretation of Bruton v. United States, 391 U.S.

123 (1968), and its progeny in the courts of appeals.

B. Second, the lower courts’ experience shows that a

requirement to analyze Bruton issues in light of a trial’s

structure and context is administrable. In the decades

since Bruton, various courts have coalesced around several common factors relevant to the analysis, which indicates that those considerations are both workable and

important.

3

C. Third, considering context for Bruton questions

does not disrupt trial management or unduly limit the

availability of joint trials. The procedures for raising and

resolving Sixth Amendment issues are well-established,

but also flexible to the needs of each case. And courts are

already considering context in determining whether and

how to redact proffered confessions consistent with the

Confrontation Clause. This sort of analysis is nothing new.

Trial courts are well-suited to direct the redaction of confessions and to enforce pretrial Bruton rulings, just as

they have been doing for years.

ARGUMENT

In Bruton v. United States, 391 U.S. 123 (1968), this

Court recognized that “there are some contexts in which

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.” Id. at 135. As this

Court held, that intolerable risk arises when, in a joint

criminal trial, the prosecution seeks to introduce the outof-court confession of a non-testifying codefendant that

expressly incriminates the defendant. Id. at 135-136. The

Sixth Amendment thus prohibits admitting such evidence,

even with a limiting instruction. See id. at 136-137.

The Court has elaborated on the Bruton rule in two

subsequent cases. Under Richardson v. Marsh, 481 U.S.

200 (1987), the prosecution may introduce a confession

that has been redacted so as “to omit all indication that

anyone other than [the codefendant had] participated in

the crime.” Id. at 203. But under Gray v. Maryland, 523

U.S. 185 (1998), the prosecution may not introduce a confession that “refers directly to the ‘existence’ of the nonconfessing defendant,” even if it has been redacted to

substitute a blank space or the word “deleted” for the

4

defendant’s name. Id. at 192. In that situation, this Court

observed, the jury will naturally still “realize that the confession refers specifically to the defendant.” Id. at 193.

This Court should now use this case to reaffirm that

Bruton challenges require careful analysis of a trial’s context to determine how to redact a codefendant’s out-ofcourt confession consistent with the Sixth Amendment.

Even when the defendant’s name is redacted, if the trial’s

context—including the number of defendants, the factual

circumstances, or the prosecutor’s questioning and arguments—effectively unmasks the defendant as the subject

of a redacted confession, then it as if there was no redaction at all. The Sixth Amendment prohibits that result.

That rule flows from, and faithfully harmonizes, the

Court’s precedents, and the experience of the lower courts

proves that a context-driven inquiry is administrable.

Further, considering context will not upset current trial

procedure. To the contrary, this rule is consistent with

how many courts already address Bruton questions. And

importantly, a context-matters rule best protects the core

confrontation right enshrined in the Sixth Amendment by

recognizing that courts must take special care to safeguard a defendant’s constitutional rights when the prosecution introduces such potent evidence.

A. This Court’s precedent establishes that whether

the introduction of a non-testifying codefendant’s

out-of-court confession violates the Sixth

Amendment requires a contextual inquiry

This Court’s trilogy of cases regarding admission of

a non-testifying codefendant’s out-of-court confession—

Bruton, Richardson, and Gray—illustrates how the Confrontation Clause applies along a spectrum of cases involving various types and degrees of redaction. Bruton

and Richardson are the bookends to that spectrum:

5

Bruton prohibits an unredacted confession that refers to

a defendant by name. 391 U.S. at 135-136. And Richardson approves the most protective type of redaction that

eliminates all reference to the defendant’s existence. 481

U.S. at 211. Gray then refined the analysis and began

providing guidance for other kinds of cases in the middle

space. See 523 U.S. at 192, 195-196.

This case picks up where Gray left off, falling between

Gray and Richardson on the spectrum. But the sum of

this Court’s precedents already establishes four guiding

principles for analyzing Bruton questions.

1. First, this Court clarified in Gray that the introduction of a codefendant’s out-of-court confession can violate the Sixth Amendment even if that confession does not

name the defendant. 523 U.S. at 195-196. The Sixth

Amendment is still implicated when a confession, “despite

redaction, obviously refer[s] directly to someone, often

obviously the defendant, and which involve[s] inferences

that a jury ordinarily could make immediately.” Id. at 196.

Gray thus necessarily rejected a four-corners approach

that focuses exclusively on the confession itself. See id. at

195-196. 2

Second, Gray recognized that whether a codefendant’s confession violates the Sixth Amendment requires a

case-specific determination, noting that “in some

Richardson, too, cannot be read to endorse a four-corners

approach, both on its own terms and in light of Gray. Although Richardson suggested that any “evidence introduced later at trial” is

“linkage” that is irrelevant under the Sixth Amendment, the only

such “linkage” at issue was “the defendant’s own testimony.” 481 U.S.

at 208 (emphasis added). Further, Richardson’s reasoning necessarily must be limited to what a defendant does at trial, given the

Court’s recognition that a prosecutor’s conduct can negate a Bruton

instruction. Richardson, 481 U.S. at 211.

2

6

instances the person to whom the blank [in a confession]

refers may not be clear,” 523 U.S. at 194, but in others, it

might. Gray explained that, in addition to how a confession is redacted, other aspects of a trial can affect how

“transparent” is the out-of-court confession, including the

number of defendants and whether “the trial indicates

that there are more participants than the confession has

named.” Id. at 195-196.

Third, this Court has twice confirmed that the relevant

context includes the prosecutor’s arguments and the line

of examination through which the confession is introduced. In Richardson, the prosecutor impermissibly

“sought to undo the effect of the limiting instruction by

urging the jury to use [the codefendant’s] confession in

evaluating [the defendant’s] case.” 481 U.S. at 211. Likewise, in Gray, the prosecution “blatantly link[ed] the

defendant to the deleted name” through its questioning,

which “eliminated all doubt” about the person to whom

the confession referred. 523 U.S. at 193, 194.

Fourth, Richardson establishes a limiting principle

for a context-based analysis: when a prosecutor properly

redacts a confession, the defendant cannot generate a

Sixth Amendment problem by introducing evidence that,

viewed in connection with the redacted confession, inculpates himself. 481 U.S. at 208-209. 3

2. These principles reflect how multiple lower courts

have understood this precedent. See, e.g., Foxworth v. St.

Amand, 570 F.3d 414, 430, 433 (1st Cir. 2009) (Gray “refined and extended the Bruton rule … Under this regime,

an inquiring court must judge the efficacy of redaction on

In rare cases, it is conceivable that a defendant could need to

introduce certain evidence for a valid purpose that nonetheless would

be inculpatory with respect to the codefendant’s redacted confession.

In that circumstance, a severance would be especially appropriate.

3

7

a case-by-case basis, paying careful attention to both a

statement’s text and the context in which it is offered.”)

(citation omitted), certified question on other grounds answered, 929 N.E.2d 286 (Mass. 2010); United States v. de

Leon-De La Rosa, 17 F.4th 175, 194 (1st Cir. 2021)

(“[L]anguage is always used in context, as Gray instructs

us to remember in assessing whether a Bruton violation

occurred.”); United States v. Straker, 800 F.3d 570, 596

(D.C. Cir. 2015) (“Evaluations of [the] effectiveness [of redactions under the Confrontation Clause] are necessarily

contextual.”); United States v. Hoover, 246 F.3d 1054,

1059 (7th Cir. 2001) (“Very little evidence is incriminating

when viewed in isolation; even most confessions depend

for their punch on other evidence. To adopt a four-corners

rule would be to undo Bruton in practical effect.”); United

States v. Schwartz, 541 F.3d 1331, 1351 (11th Cir. 2008)

(“[A] defendant’s confrontation right is violated when the

court admits a codefendant statement that, in light of the

Government’s whole case, compels a reasonable person to

infer the defendant’s guilt.”).

In short, the Bruton trilogy already provides the

framework for a context-matters rule, including foundational principles that the trial court must apply and that

the prosecution must abide.

B. The lower courts’ experience shows that

analyzing Bruton issues in light of a trial’s

context is an administrable rule

To reaffirm that trial context is critical in a Bruton

analysis, this Court does not need to develop a new test

out of whole cloth. In the 55 years since Bruton, and the

25 years since Gray, lower courts have been analyzing

proffered confessions in light of the structure and context

of trial, and those decisions are instructive. Indeed, the

fact that courts have consistently identified similar factors

8

illustrates both the importance and workability of those

factors.

1. Courts consider the number of defendants, how

many participants the confession or evidence

identifies, and the complexity of the facts

Consistent with this Court’s suggestion in Gray, 523

U.S. at 195, several courts have resolved Bruton issues by

considering the number of defendants on trial and how

many other alleged criminal participants are identified by

the confession or other evidence.

Depending on the factual complexity in a given case,

the number of defendants and other actors can make an

important difference to how the jury understands a confession. Some cases “may involve numerous actors and

events, such that no compelling inference can be drawn

that a symbol or neutral pronoun refers to a specific

defendant.” Foxworth, 570 F.3d at 433. But others “may

involve few actors and events, such that a symbol or neutral pronoun becomes transparent.” Ibid.; see also, e.g.,

Straker, 800 F.3d at 599 (where “evidence identified more

than a dozen different men involved in the crimes” it was

“unlikely that the jury would readily link a statement’s

mention of a ‘person’ or ‘guy’ to a specific defendant”).

In United States v. Hernandez, 330 F.3d 964 (7th Cir.

2003), for example, the indictment accused 21 individuals

of drug and conspiracy offenses. Id. at 967-968. Of the

eight defendants who went to trial, one had confessed to

being a member of the “Project Kings” as a gang enforcer

and stated that the gang had taken “disciplinary action …

against ‘Latin Kings.’ ” Id. at 972. The Seventh Circuit

rejected the argument that “Project Kings” and “Latin

Kings” were “overly obvious” references to the other

seven defendants. Id. at 974. Citing the number of individuals in the indictment and the size and complexity of the

9

gang overall, the court found that those terms were not

“ ‘obvious stand-ins’ for the co-defendants.” Ibid.

For similar reasons, the Third Circuit found no Sixth

Amendment violation in Priester v. Vaughn, 382 F.3d 394

(2004), which involved a multi-car drive-by shooting and a

confession that replaced the names of several individuals,

including the defendant, “with phrases such as ‘the other

guy.’ ” Id. at 396. The court observed that “[t]here were at

least fifteen perpetrators in various cars involved in the

shooting”; the redacted statement was “unclear as to the

people in the first car, in the second car, who was shooting

when and from which car”; and the substitute phrases

lacked “any innuendo” that identified the defendant. Id.

at 399-401.

By contrast, the Third Circuit found the Sixth Amendment was violated in a pair of cases analogous to petitioner’s here.

In the first, the defendant (Vazquez) and one codefendant (Santiago) were tried together for murder.

Vazquez v. Wilson, 550 F.3d 270, 271 (3d Cir. 2008). Santiago confessed to police that he had been driving with the

defendant and another person (Rivera) when the defendant shot the victim. Id. at 272-273, 281. The defendant

later testified that Rivera had fired the fatal shot. Id. at

273-274. Santiago did not testify but sought to introduce

his unredacted confession to show that he had immediately cooperated with police. Id. at 274. But the defendant

argued that even a redacted confession would identify him

given the full context of the case. Ibid.

The district court denied both defendants’ motions to

sever and admitted Santiago’s redacted confession, which

replaced the defendant’s and Rivera’s names with “my

boy” or “the other guy” more than twenty times. 550 F.3d

at 274, 281. The court allowed Santiago’s counsel to

10

establish that Santiago had identified the two individuals

in the car. Id. at 274. And during closing argument, the

prosecutor identified Rivera as “the man who’s not the

shooter.” Id. at 275. But those arguments by counsel and

the prosecutor merely underscored what the jury was already “almost certain to conclude”—Santiago had identified the defendant as the shooter. Id. at 281. Although the

confession was redacted, the circumstances offered no

other conclusion; “there were only two possible shooters

under Santiago’s statement,” “Rivera was not on trial,”

and the state argued that the defendant, not Rivera, was

the shooter. Ibid. 4

The number of defendants was similarly critical to the

Third Circuit’s analysis in Eley v. Erickson, 712 F.3d 837

(2013). After the trial court denied defendant Eley’s motion to sever, he was tried with codefendants Eiland and

Mitchell for murder and robbery. Id. at 841, 854. Among

other statements, Eley challenged Eiland’s confession to

a cellmate that he was the one who had shot the victim,

but that “[i]t was the other two’s idea.” Id. at 854, 858. Although that statement did not name Eley, it “expressly referred to the existence of exactly three people,” which, the

court found, “could not have been lost on the jury” because “the Commonwealth emphasized shortly before introducing the confession” that “there were exactly ‘three

defendants’ sitting at the defense table.” Id. at 859.

Vazquez is an example of a case in which a limiting instruction

was very likely insufficient to prevent the jury from drawing the

obvious inference that the confession referred to the defendant. Although the jury received the proper limiting instruction, it submitted

this question during deliberation: “Are we supposed to not consider

Santiago’s statement that [defendant] was the shooter?” 550 F.3d at

275. The court answered by repeating the limiting instruction; the

jury then acquitted Santiago and convicted the defendant. Ibid.

4

11

The Third Circuit thus found in Eley “an even more

compelling case for habeas relief than in Vazquez,” where

the jury had to decide whether Santiago’s statement implicated the charged defendant or the absent Rivera. 712

F.3d at 860. In Eley’s case, the inference was even clearer:

the “confession expressly implicated exactly three people

in the crimes and exactly three defendants appeared at

the joint trial.” Id. at 860-861.

State courts have conducted similar analyses. In Jefferson v. State, 198 S.W.3d 527 (Ark. 2004), for example,

the Arkansas Supreme Court found a Sixth Amendment

violation in the joint robbery trial of the defendant (Jefferson) and his codefendant (Starr). Id. at 529. The third

alleged participant (Foster) pleaded guilty before trial.

Id. at 530. The trial court admitted Starr’s pre-trial statement to police that implicated Jefferson and Foster, but it

was redacted to replace Jefferson’s name with “he” or

“some other guy.” Id. at 530-531, 533. Nonetheless, the

court held, “the jury easily could have drawn” the inference that Starr’s statement “obviously directly referred

to Jefferson,” because the prosecution made clear that

three individuals were involved and conceded that Foster

had been the shooter. Id. at 536; see also, e.g., State v. Medina, 48 P.3d 1005, 1012 (Wash. Ct. App. 2002) (finding

“references to ‘the guys’ and a ‘guy’ ” did not impermissibly implicate two of the three defendants, in part, because

“there were approximately six individuals involved”);

Neal v. State, 806 So.2d 1151, 1156 (Miss. Ct. App. 2002)

(admission of statement violated the Sixth Amendment

“because there were only two people on trial” and therefore “the subject” “obviously” referred to the defendant).

In short, courts understand that the number of

defendants and other participants involved in a case is an

especially relevant factor that must inform the Sixth

12

Amendment inquiry. Jurors can count; they know how to

put two and two together. In some instances, this aspect

of a trial may reveal such clear, immediate, and powerfully

damaging inferences that a juror “need only lift his eyes

to … counsel table” to understand to whom the redacted

confession refers. Gray, 523 U.S. at 193.

2. Courts consider the frequency and type of

redactions

Lower courts have also correctly observed that the

way a confession is redacted, and the number of substituted phrases, can affect how the jury understands the

confession.

In United States v. Williams, 429 F.3d 767 (8th Cir.

2005), for instance, the confession at issue had “more than

forty instances where [the defendant’s] name was replaced with the word ‘someone.’ ” Id. at 773. Although the

Eighth Circuit ultimately did not need to resolve the Sixth

Amendment issue, the court observed that the “kind and

degree” of the redactions made it obvious that the defendant’s name had been removed. Id. at 774. The “replacements were not seamlessly woven into the narrative …

and the neutral pronoun ‘someone’ may have lost its anonymity by sheer repetition.” Ibid. Under those circumstances, “[i]t may well have been clear to the jury that the

statement had obviously been redacted and that the

‘someone’ of the statement was defendant.” Ibid. (citation

omitted).

Similarly, the contrast between anonymized and

named individuals may be an obvious tell. In a case before

the Oregon Court of Appeals, the court found that the

terms “the person” and “the individual” clearly referred

to the defendant because everyone else “in the confession

[was] named with one conspicuous exception,” and the unnamed person’s “anonymity [was] reemphasized with

13

every use of some antecedentless pronoun or generic

term.” State v. Johnson, 111 P.3d 784, 788 (2005).

By contrast, the Fifth Circuit held that the Confrontation Clause was not violated by an officer’s accidental use

of the plural pronoun “they” during his trial testimony recounting a declarant’s statement, even though the government had agreed to omit all plural references so as not to

implicate the dozen other defendants. United States v.

Ramos-Cardenas, 524 F.3d 600, 603-604, 607-608 (2008).

Among other reasons, the court noted that the officer

made this error only twice and corrected himself. Id. at

608. And given the nature of the charges against the other

defendants, and the substance of the officer’s testimony,

the officer’s use of “they” did not implicate specific individuals. Ibid. The court observed, however, that “[r]epeated use of an indefinite pronoun, may, in some circumstances, give rise to a Confrontation Clause violation.” Id. at 609 n.6.

3. Courts consider the strength of the inferences

from a confession, and how those inferences

relate to other evidence

In determining whether it is possible to redact a confession consistent with the Confrontation Clause, and to

what degree it must be redacted, lower courts account for

the inferences that may arise from a redacted confession

and how those inferences relate to the rest of the prosecution’s case. See, e.g., United States v. Nash, 482 F.3d

1209, 1219 (10th Cir. 2007) (“[O]ur judgment is informed

by the context in which the Bruton statement was admitted, how it was used at trial, and how it compares to the

properly admitted evidence.”) (citation omitted).

The Sixth Circuit’s decision in United States v.

Macias, 387 F.3d 509 (2004) is a helpful example. The redacted confession at issue referred to the defendant as

14

“subject two,” but it described subject two’s residence as

being located in the same area where the evidence showed

that the defendant lived. Id. at 514. In the context of the

whole trial, “the description of subject two was sufficiently specific that it could have referred only to [the defendant],” and therefore “was incriminating on its face.”

Id. at 519. The statement was thus “comparable to Bruton” in how it affected the rest of the evidence. Id. at 518.

As the Sixth Circuit observed, the statement “transformed the government’s case into a direct evidence

case … rendering it largely unnecessary for the jury to

infer [the defendant’s] involvement in the conspiracy

based on the circumstantial evidence.” Ibid.

Although it is “not always easy” to analyze whether a

redacted confession still obviously implicates the defendant, trial courts have experience making this “delicate determination” and recognize that it “requires case-by-case

consideration rather than a brightline rule.” United

States v. Green, 648 F.3d 569, 575 (7th Cir. 2011). Green

was an edge case that fell “close to that subtle line,” but

the Seventh Circuit carefully compared precedent and ultimately found that the substitute term “straw buyer” in

an out-of-court confession did not violate Bruton. Ibid.

The court reasoned, in part, that the evidence necessary

to identify the defendant as the “straw buyer” was “farther removed” than in other cases, although the court considered the case “very close to the Bruton line.” Id. at 576.

Considering how certain evidence relates to other evidence, how the jury is likely to receive information at trial,

and how relevant are any inferences that the jury might

draw from the evidence presented at trial is nothing new.

Criminal trials constantly require that kind of analysis,

and courts are as capable of making those determinations

in the Bruton context as in any other. E.g., United States

15

v. Gibson, 875 F.3d 179, 195 (5th Cir. 2017) (finding no

Sixth Amendment violation where “several inferential

leaps” were necessary for the jury to connect a statement

about a hospital entity to specific defendants in a charged

conspiracy); United States v. Lung Fong Chen, 393 F.3d

139, 150 (2d Cir. 2004) (finding no Sixth Amendment violation where “substantial evidence was necessary to link

defendants … with [challenged] statements”).

4. Courts consider the prosecution’s questioning

and arguments

Since Richardson and Gray, lower courts have continued to recognize that a prosecutor’s questioning and

arguments regarding an out-of-court confession can violate the Sixth Amendment—either in connection with

other evidence or standing alone.

This Court’s precedents establish that the Constitution does not permit the prosecutor to unmask the defendant as the individual implicated in a confession. Gray, 523

U.S. at 193; Richardson, 481 U.S. at 211; see, e.g., Brown

v. Superintendent Greene SCI, 834 F.3d 506, 517 (3d Cir.

2016) (“[A] prosecutor’s inadmissible use of a confession

during closing arguments runs afoul of Bruton.”); United

States v. Peterson, 140 F.3d 819, 822 (9th Cir. 1998)

(“[T]he prosecutor’s closing argument that Peterson was

‘person X,’ an argument which would be clearly prohibited

under Gray, further compounded the constitutional violation.”); Schwartz, 541 F.3d at 1353 (“[T]he inference was

made inevitable … when the prosecutor expressly made

that connection for the jury in his closing argument.”).

But directly identifying the defendant as the person

referred to in a redacted confession is not the only way for

a prosecutor to violate the Confrontation Clause. Courts

have recognized that a prosecutor’s statements violate the

16

Constitution just the same when they clearly point to the

defendant in context. E.g., Brown, 834 F.3d at 517-518.

In Brown, the non-testifying codefendant (Garcia)

confessed before trial that defendant Lambert had shot

the victim, got into Garcia’s car afterwards, and then Garcia drove them to Garcia’s house; at trial, Lambert’s name

was redacted and replaced with the “guy.” 834 F.3d at 510.

During closing argument, however, the prosecutor “revealed that Garcia took Lambert to his house” after the

incident. Id. at 518. She stated, in part: “If Garcia had not

been part of what happened, how easy would it have been

for him to drop Lambert off, go home … [But] he takes

Lambert to his house … and he says the guy I’m with

brings the gun into my house ….” Id. at 510. By placing

Lambert in Garcia’s car and at his house, the prosecutor

“conveyed a message—that Lambert was the person

whose name was withheld in the redacted confession—as

clearly as” if she had expressly identified him as the

“guy.” Id. at 518; see also United States v. Davis, 534 F.3d

903, 915 (8th Cir. 2008) (acknowledging that “a Bruton violation occurs ‘when the unnamed defendant is tied directly to the confession in the manner and context in

which the confession is presented’ ”) (citation omitted).

A prosecutor can also convey the same message by

introducing evidence before the redacted confession that

“impermissibly prim[es] the jury to implicate” the defendant later. Wynn v. United States, 241 A.3d 277, 284-285

(D.C. Ct. App. 2020). Likewise, the prosecutor can “eradicate[ ]” any doubt the jury might have after hearing the

redacted confession by following it up with a question

about whom the confession led law enforcement to investigate or arrest. E.g., Davis v. State, 528 S.E.2d 800, 805806 (Ga. 2000); see also State v. McDonald, 771 S.E.2d

840, 844 n.3 (S.C. 2015) (similar).

17

In sum, as these and other courts have recognized,

there would be “no point in redacting and sanitizing otherwise inculpatory statements” if those protections “could

be deliberately and directly undone by lawyer commentary.” Brown, 834 F.3d at 517 (citation omitted). The Sixth

Amendment cannot be so easily diminished.

* * *

Particularly in criminal matters, trial and appellate

courts are accustomed to resolving legal issues in light of

the full context and unique circumstances of each trial,

and they are well-suited to identify the relevant considerations in each case. See, e.g., Turner v. United States, 137

S. Ct. 1885, 1893 (2017) (considering “the withheld

[Brady] evidence ‘in the context of the entire record’ ”) (citation omitted); Ohio v. Clark, 576 U.S. 237, 249 (2015)

(observing that the primary-purpose test requires “evaluat[ing] challenged statements in context”); Greer v. Miller, 483 U.S. 756, 765-766 (1987) (“When a defendant

contends that a prosecutor’s question rendered his trial

fundamentally unfair, it is important ‘as an initial matter

to place the remark in context.’ ”) (citation and original alterations omitted); United States v. Park, 421 U.S. 658,

674 (1974) (“[A] single instruction to a jury may not be

judged in artificial isolation, but must be viewed in the

context of the overall charge.”) (citation omitted).

That lower courts have coalesced around several manageable factors to resolve Bruton challenges underscores

that those factors are not only relevant but plainly significant. Those decisions, spanning myriad factual circumstances, illustrate that courts can apply a context-matters

rule reliably, consistently, and effectively—as they have

been doing for many years.

18

C. Considering trial context in Bruton challenges

will not disrupt the jury system

1. Procedures for resolving Bruton issues are

well-established

The substance of a context-matters rule is not new, nor

is the procedure. A holding by this Court in this case that

a trial court must consider the full context to evaluate the

impact of a redacted confession will clarify, not disrupt,

what is already fairly commonplace in the lower courts.

In a case involving a non-testifying codefendant’s confession that inculpates another defendant, the caselaw

establishes four general options for addressing Confrontation concerns. If the prosecutor (1) tries the defendant

separately; or (2) declines to introduce the confession at

the joint trial, then Bruton is not implicated. But if the

prosecutor both pursues a joint trial and seeks to introduce the confession, then the trial court must determine

whether the confession can be redacted consistent with

the Confrontation Clause at all, and if so, to what degree

it must be redacted. Specifically, the court must determine whether, under the circumstances of the case, the

Sixth Amendment (3) requires the prosecutor to redact

the confession to omit all references to the inculpated

defendant’s existence, as in Richardson; or (4) permits

the prosecutor to do something less and substitute the defendant’s name with neutral pronouns or phrases, as suggested in Gray. The structure and context of a trial—

including the factors above—inform whether and what

degree of redaction is constitutionally sufficient.

In practice, these questions are typically raised in a

motion to sever—whether by the prosecution or the

defense—or a motion in limine. E.g., Vazquez, 550 F.3d at

274, 281. And the prosecution, being familiar with its own

case, is able to foresee these issues and has ample

19

opportunity to raise the possibility of a Confrontation

Clause problem to resolve the issue before trial or before

introducing particular evidence. Cf. United States v. Ceballos, 789 F.3d 607, 615 (5th Cir. 2015) (discussing prior

case where the “government [had] announced during a

pretrial proceeding that it intended to introduce an incriminating written statement” and “offered to introduce

a redacted version”) (citation omitted); United States v.

Moore, 651 F.3d 30, 85 (D.C. Cir. 2011) (observing that,

before the relevant witnesses were called, the “prosecutor

advised the district court that [part of its case] might

cause a potential” Bruton issue).

Once raised, courts have adopted different procedures

for resolving these issues, including holding a pre-trial

“Bruton hearing” to consider the evidence and possible

solutions. E.g., Holland v. Attorney Gen. of N.J., 777 F.2d

150, 152 (3d Cir. 1985) (“[T]he court held a Bruton hearing, and ruled that references to [the defendant] could be

effectively excised.”); United States v. Padilla-Galarza,

990 F.3d 60, 76 (1st Cir. 2021) (“[A]t a pretrial conference,

prompted by the appellant’s severance motion … the

prosecutor spelled out the government’s planned procedure for handling [the] confession.”). By way of example,

the district court in United States v. Javell, 695 F.3d 707

(7th Cir. 2012), “ordered the government to submit a Bruton statement detailing exactly what they intended to introduce at trial,” held a hearing, and then further

redacted the proposed statement. Id. at 710, 713 (affirming admission).

After a court rules on the degree of a redaction, the

prosecution must take measures necessary to comply,

such as carefully managing its witness examinations to

avoid reanimating a Bruton problem—which a diligent

prosecutor should be able to do. Cf. Padilla-Galarza, 990

20

F.3d at 75 (discussing how the prosecutor instructed witness appropriately and “conducted the remainder of the

examination in accordance with her assurance” to the

court to “take care to ‘lead [the witness] through [the]

questions [to] avoid the Bruton issue”). This is not a new

or unreasonable expectation.

For instance, in “an attempt to avoid a Bruton problem” in United States v. Coleman, 349 F.3d 1077 (8th Cir.

2003), the prosecutor substituted the word “someone” for

the defendant’s name, and then led the testifying agent

through questions by “only [asking] for a yes or no

answer.” Id. at 1085. Assuming the redaction is otherwise

constitutionally sufficient, such structured questioning is

just one of several ways that a prosecutor might control

the admission of the confession. Because the prosecution

alone has control over the preparation of its witnesses and

the presentation of its case, it is reasonable—indeed,

necessary—for the prosecution to bear the responsibility

to ensure that it introduces evidence in a way that avoids

a Confrontation Clause problem.

In some cases, the trial court might conclude that “the

statements [cannot] be sufficiently redacted … without

substantially compromising their evidentiary value.”

United States v. Campbell, 986 F.3d 782, 803-804 (8th Cir.

2021) (affirming grant of the government’s severance

motion), cert . denied, 142 S. Ct. 751 (2022), and cert. denied, 142 S. Ct. 784 (2022). Even so, that finding does not

necessarily foreclose trying the defendants jointly; the

prosecutor might simply decide to forego offering the

statement in order “to preserve a joint trial.” United

States v. Damra, 621 F.3d 474, 482 (6th Cir. 2010); see also

United States v. Field, 756 F.3d 911, 913 (6th Cir. 2014)

(“Explaining it faced a Bruton problem, the government

moved to dismiss the charges against [one of the

21

defendants] without prejudice.”). And importantly, even if

the prosecutor successfully sought severance (or made

that choice from the outset), she need not try every defendant separately. In a large joint trial, it may be possible to peel off only the affected defendant’s case and leave

the rest of the joint trial intact.

Because courts routinely oversee this process to protect defendants’ rights and efficiently manage trials, they

are well-equipped to address unexpected testimony that

might compromise pretrial Bruton rulings and redactions. In rare cases, a mistrial may be the only constitutionally permissible response when the violation is “so

prejudicial [that] an admonition would have been useless.”

Macias, 387 F.3d at 522. But that is far from the typical

case. Clear pre-trial rulings, diligent witness preparation,

and careful redactions and case management—that is,

what courts and parties already strive for in proceedings

across the country—will minimize the risk of significant

errors at trial. Other “slip-ups” in testimony may be minor

and not raise impermissible inferences about the redacted

confession at all. E.g., Ramos-Cardenas, 524 F.3d at 608609. And in all events, Bruton errors are still subject to

harmless error review. See Harrington v. California, 395

U.S. 250, 254 (1969).

Simply put, the caselaw does not reveal “significant

practical difficulties arising out of [courts’] administration” of a context-based rule. Gray, 523 U.S. at 197. To the

contrary, the procedural paths for resolving Bruton issues are well-established and flexible.

2. Considering context gives effect to the Sixth

Amendment without unduly constraining

prosecutors’ ability to pursue joint trials

This Court has recognized that, when the prosecution

pursues a joint trial and seeks to introduce a non-

22

testifying codefendant’s confession, the court must diligently protect the defendant’s Sixth Amendment confrontation right. Bruton, 391 U.S. at 135-136. At the same

time, the Court has identified important interests that

joint trials serve. Richardson, 481 U.S. at 209-210. Not

only is a context-matters rule the more faithful application

of the Confrontation Clause than the four-corners approach, it also harmonizes the Court’s precedent to impose symmetrical responsibilities on the parties and

balance adversarial interests.

A requirement to consider trial context does not meaningfully limit the prosecution’s procedural paths to trial—

if it does at all. And if in some cases the circumstances

require the prosecution to use the most protective redaction or pursue a separate trial, then that is a feature, not

a bug, of the Sixth Amendment’s confrontation guarantee.

By contrast, a bright-line rule that ignores how a jury

will likely understand certain evidence would permit the

sort of practical deprivations of confrontation that the

Sixth Amendment forbids. The prosecution is nearly always the party that reaps the benefits of a joint trial. As

such, the trial court may impose on the prosecution,

through appropriate orders, the corresponding responsibility to manage constitutionally intolerable risks to the

defendant. The system’s interests in joint trials—even

when substantial—end where the Constitution’s protections begin.

At the same time, a defendant has nothing to gain

from a context-grounded approach other than what the

Constitution promises. If the prosecution complies with

the trial court’s Bruton rulings, then the defendant cannot

obtain a mistrial or other strategic advantage by implicating himself in the redacted confession. Accordingly, both

sides have every incentive to fully flesh out Bruton issues

23

before trial and manage their cases pursuant to that plan.

This Court can trust lower courts to continue overseeing

that process; they have proven well-equipped to do so.

CONCLUSION

The judgment of the court of appeals should be

vacated, and the case remanded for further proceedings.

Respectfully submitted,

David Watnick

PERKINS COIE LLP

1120 N.W. Couch Street

Tenth Floor

Portland, OR 97209-4128

Keith M. Donoghue

Jessica Stengel

Stacie Fahsel

Members, Amicus

Committee

NATIONAL ASSOCIATION OF

FEDERAL DEFENDERS

February 1, 2023

Michael R. Huston

Counsel of Record

PERKINS COIE LLP

700 Thirteenth Street N.W.

Suite 800

Washington, DC 20005-3960

(202) 434-1630

mhuston@perkinscoie.com

Luci D. Davis

PERKINS COIE LLP

2901 North Central Avenue

Suite 2000

Phoenix, AZ 85012-2788