Opposition Brief — Oliver v. Quarterman (No. 08-833)

FILED

MAR 2 0 2009

fn

mA

Va OFFICE OF THE CLERK

No. 08-833 “SUPREME COURT, U.S.

nu the Supreme Court of the United States

KHRISTIAN OLIVER, PETITIONER

V.

NATHANIEL QUARTERMAN, DIRECTOR, TEXAS

DEPARTMENT OF CRIMINAL JUSTICE, CORRECTIONAL

INSTITUTIONS DIVISION

(CAPITAL CASE)

ON PETITION FOR A WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE FIFTH CIRCUIT

BRIEF FOR THE RESPONDENT IN OPPOSITION

GREG ABBOTT JAMES C. HO

Attorney General of Texas Solicitor General

Counsel of Record

C. ANDREW WEBER

First Assistant Attorney DANIEL L. GEYSER

General Assistant Solicitor General

ERIC J. R. NICHOLS OFFICE OF THE ATTORNEY

Deputy Attorney General GENERAL

for Criminal Justice P.O. Box 12548 (MC 059)

Austin, TX 78711-2548

EDWARD L. MARSHALL (512) 936-1700

Chief, Postconviction

Litigation Division

QUESTIONS PRESENTED

In 1996, Congress revised federal law to restrict the

authority of federal courts to grant habeas relief to

prisoners challenging their state-court criminal

convictions. Habeas relief, as a threshold matter, is now

unavailable unless a _ petitioner can identify a

constitutional defect in light of “clearly established

Federal law, as determined by the Supreme Court.” 28

U.S.C. 2254(d)(1). Federal courts are now also ordered to

defer to state-court factual findings absent any “clear and

convincing” contrary evidence. 28 U.S.C. 2254(e)(1).

The questions presented are:

1. Whether a juror’s consultation of the Bible during

the sentencing phase of a capital case violates this Court’s

“clearly established” law, in light of the clear split of

authority over whether this conduct even implicates a

defendant’s Sixth Amendment rights in the first place.

2. Whether a habeas petitioner alleging that a jury

wrongly reviewed outside evidence must prove the

evidence had a “substantial and injurious effect” on the

verdict (the standard for collateral review), or must satisfy

only a “presumptive prejudice” analysis (the standard for

direct review), or must satisfy no standard at all, under a

theory of an “irrebuttable presumption of prejudice” (never

before recognized for this claim at any stage of review).

3. Whether the lower court’s fact-bound application of

Section 2254(e)(1)—deferring to a finding of historical fact

that the jury was not influenced by a Bible that a handful

of jurors read at some point during sentencing—was

correct under the unique circumstances of this case.

(I)

{I

TABLE OF CONTENTS

TD nk oe ewe eye wwe es eu ee gam I

ee oe dae eee eee eee il

I Bd ee ny oa pene 2

Reasons for Denying the Petition ................. 7

I. This Case Is Not a Proper Vehicle for

Resolving the First Question

UI oaks wy eg ao eee es 8

IT. The Second Question Presented Fails

to Implicate Any Circuit Split and Is

Otherwise Unworthy of Review ...... 18

Ill. The Third Question Presented Only

Seeks to Revisit a Fact-bound Dispute

Unworthy of Review ........... oe a

i cg Saal ag keg on gi ee 25

Ii

TABLE OF AUTHORITIES

Cases:

Beardv. Banks,

Oe 0 Oe no on ss 14

Bellv. Cone,

ee I ns va Weve ac ew baw ees 9

Bibbins v. Dalsheim,

Zi Fa@ 18 Cd Cir. 1904) .. 0... ccs ce ws 19

Billings v. Polk,

441 F.3d 238 (4th Cir. 2006) ........... 12,13

Brecht v. Abrahamson,

507 U.S. 619 (1993) ............. 2, 16, 3 2h

Brown v. Payton,

TES Me Ee | a arr 9

Burch v. Corcoran,

972 F.3d 577 Goth Cir. 2001) .........5...; 10

Carey v. Musladin,

eee er . s c oe e e ee ween 8,9

Crease v. McKune,

189 F.3d 1188 (10th Cir. 1999) ............ 19

Duncan v. Walker,

ee ee AS. balan seus we ees 9

Early v. Packer,

og SS LS a ae ae aa 10

Fields v. Brown,

431 F.3d 1166 (Oth Cir. 2005) ............. 11

Fields v. Brown,

503 F.3d 755 (9th Cir. 2007) 11, 13, 15,17-19, 24

Fry v. Pliiler,

bye > we 2). | 18, 21

IV

Haley v. Blue Ridge Transfer,

802 F.2d 1532 (4th Cir. 1986) .............. 7

Hornv. Banks,

eae 14, 15

Kane v. Espitia,

i | a ae aire a 12

Kansas v. Marsh,

ee Ee ar 24

McNair v. Campbell,

416 ¥.36 1201 (1 ith Cir. 2005) ............ 19

Miller-Elv. Cockrell,

rr eee EE). vec se ve ced evewueweus 9

Mitchellv. Esparza,

I MMs cn ds ew ewcesuuwas 20

Neill v. Gibson,

278 F.3d 1044 (10th Cir. 2001) ............ 11

Parker v. Gladden,

eR Ee | a ee 11

People v. Danks,

Oa F.3d 3240 Wal. 2006) ... 2... eee 13

Remmerv. United States,

- vhie 2. gc) 0 | 7, 11, 2, 22

Robinson v. Polk,

438 F.3d 350 (4th Cir. 2006) ..... 11, 12, 15, 17

Robinson v. Polk,

444 F.3d 225 (4th Cir. 2006) ........... 11, 16

Rushen v. Spain,

gc Ee, 21, 22, 24

Springtield v. Kibbe,

gk oS. ys a ee rnd 16

TVannerv. United States,

Po Bee Bek. 5) 11, 23

Teague v. Lane,

rr 14

Turner v. Louisiana,

EE 11, 16

Tylerv. Cain,

EE ee 14

United States v. Lara-Ramirez,

519 F.3d 76 (1st Cir. 2008) ............. 12, 20

United States v. Wells,

es 16

Wiggins v. Smith,

a 9

Willams v. Taylor,

eee ee cee 9

Wilsonv. Vermont Castings,

Soe eee Ge cae. 1905) ................ 7

Woodford v. Visciotti,

ce ce ewe eee ess 10

Wright v. Van Patten,

ee 14

Yarborough v. Alvarado,

ee 13

Constitution, Statutes, and Rules:

SS L1.238&8 12

De ED kw ewe wee 3, 20, 22, 24

pee ED 5... twee I, 3, 20, 22, 24

Tee ewe ee eweueees 23

EE Se ee 23

Ss 7

Miscellaneous:

H.R. Conf. Rep. No. 104-518, at 111 (1996) ......... 3

BRIEF FOR THE RESPONDENT IN OPPOSITION

This case presents the unusual circumstance in which

the existence of a clear split is precisely the reason to deny

review. Respondent readily agrees that petitioner’s main

constitutional claim (under the Sixth Amendment) is both

highly important and frequently recurring. But under the

controlling AEDPA standards, the threshold question is

not whether petition’s constitutional claim is valid, but

whether the claimed right was clearly established hy this

Court at the time of the state-court decision. See 28

U.S.C. 2254(d)(1). By refusing to engage the critical

AEDPA framework, petitioner has presented the wrong

question at the wrong time: if his constitutional claim was

not clearly established, a federal court would be required

to reject it on that basis alone—without any occasion or

need to resolve the underlying constitutional issue.

It accordingly follows that petitioner himself has

uncovered the most debilitating problem for his own

petition. Since federal courts are statutorily forbidden

from granting habeas relief in the absence of clearly

established law, the very split petitioner has identified

confirms that the law is anything but clearly established

in this area. Indeed, the circuits are not simply divided

over the availability of habeas relief, but at least one

circuit has held that the conduct in question is not even

unconstitutiona/in the first place. This claim is therefore

not subject to redress in this collateral proceeding.

Because the Court can resolve this petition without

resolving the constitutional question it purports to

present, the split should instead be addressed and

resolved in a proper vehicle arising on direct review.

(1)

2

Nor are petitioner's other questions worthy of

certiorari. Petitioner’s request for the Court to reconsider

the standard of review for “extrinsic evidence” claims is

insubstantial and warrants no further review. Contrary

to petitioner’s contention, there is no circuit split on this

question—presumably because this Court has already

definitively resolved the issue in Brecht v. Abrahamson,

507 U.S. 619 (1993) (imposing the “substantial and

injurious effect” standard). Petitioner has misread the

single Eleventh Circuit decision that he believes conflicts

with Brecht. That decision found the claim at issue

procedurally defaulted, and then, without meaningful

analysis, noted that the defendant could not satisfy any

standard of harmless-error review. The circuit accordingly

had no occasion to decide which standard should

apply—and it certainly did not use that case as a vehicle

to create an (unacknowledged) split while flouting this

Court’s controlling precedent.

Petitioner’s final question presents only a fact-bound

issue unworthy of review. The Fifth Circuit’s deference to

the state-court factual findings in this case operated

exactly how Congress intended—and petitioner raises no

meaningful legal question to undercut the circuit’s

(correct) analysis.

Because the only question warranting review is not

adequately presented in this (collateral-proceeding)

posture, the petition for a writ of certiorari should be

denied.

STATEMENT

1. Under the Antiterrorism and Effective Death

Penalty Act of 1996, Pub. L. No. 104-132, Tit. I, 110 Stat.

3

1218, Congress strictly limited a federal court’s power on

habeas review to void state criminal convictions. Congress

provided that a state-court decision can be set aside on

collateral attack only if it is “contrary to, or involved an

unreasonable application of, clearly established Federal

law, as determined by the Supreme Court.” 28 U.S.C.

2254(d)(1). Congress also commanded federal courts to

defer to state-court factual findings absent “clear and

convincing” conflicting evidence. 28 U.S.C. 2254(e)(1); see

also 28 U.S.C. 2254(d)(2) (similarly insulating state-court

factual determinations from challenge unless the decision

“was based on an unreasonable determination of the facts

in light of the evidence presented in the State court

proceedings”). These reforms collectively were designed to

“curb the abuse of the statutory writ of habeas corpus” and

to preserve the proper degree of deference for state-court

determinations. H.R. Conf. Rep. No. 104-518, at 111

(1996).

2. Petitioner murdered Joe Collins on March 17, 1998,

during an armed burglary of Collins’s house. Pet. App.

263a. Collins had arrived home during the burglary and

confronted petitioner and an accomplice. After Collins

shot the accomplice in the leg, petitioner pursued Collins,

shooting him five times—including at least twice in the

face—and continuing to “beat{[]” Collins with the butt of

his rifle as Collins lay on his back. /d. at 267a, 277a. The

beating (described as “brutal”) had severely “fracturled]”

Collins skull and “disfigurled] his face nearly beyond

recognition.” Jd. at 28la. Photographs of the attack

depicted injuries so “gruesome” that petitioner moved at

trial to have the pictures excluded from evidence. Pet.

App. 272a-273a.

4

This murder and petitioner’s subsequent arrest were

the culmination of an 18-month “crime spree” that

included multiple armed robberies, a car-jacking (where

petitioner was found to have pointed a cocked gun at the

victim’s head), and a violent encounter with a security

guard during an attempted burglary at a high school

(where petitioner struck the guard, aimed a gun at him,

and fired on an approaching janitor). Pet. App. 279a-281a.

As the state court described it, the prior offenses

“reflectled] an escalating pattern of criminal activity,” “an

increasing willingness to take risks,” and “a propensity for

violence.” Jd. at 281a.

3. Ajury found petitioner guilty of capital murder and

sentenced him to death. Pet. App. 2a-3a. Petitioner

thereafter filed a motion for new trial based on the claim

that some jurors had purportedly read passages of the

Bible at some point during the sentencing deliberations.

Id. at 2a-3a, 285a-286a.

The trial court held a hearing on the motion, and

petitioner called four jurors to testify. Pet. App. 285a.

The trial court instructed the jurors not to testify about

their mental processes, but only to describe the objective

circumstances of what occurred in the jury room. Jd. at

129a-130a. The testimony revealed that a handful of

jurors, individually or in small groups, had reviewed

certain passages in the Bible before the sentencing verdict

was read. /d. at 3a-5a, 285a-286a. One of those passages

included a section that included the following statement:

if a person “smite” someone “with an instrument of iron,

so that he die, he is a murderer,” and “the murderer shall

surely be put to death.” Jd. at 3a.

5

The jurors also testified that no juror at any time

suggested the Bible should be considered as law or

evidence, and the Bible was never discussed “as part of or

during” any group deliberations. Pet. App. 286a. ‘I'wo

jurors also recalled not reading the Bible until after the

final vote was cast on sentencing and while the jurors

were waiting to return to the courtroom. /bid. Each juror

also confirmed that they received and followed the

instructions in the court’s charge. J/hid.

At the conclusion of testimony, the trial court ruled

that, in light of “all the evidence pertaining to the

occurrence in the jury room in question in reference to the

Biblical quotation,” “the conduct of the jury was not

improper.” Pet. App. 234a. The court further ruled that

“a conscientious, dedicated and carrying [sic] jury

considered this case in accord with the Court’s Charge and

the instructions of the Court and rendered their verdict in

accord with the evidence they heard in this case

uninfluenced by any outside influence of any kind.” Jbid.

4. The Court of Criminal Appeals of Texas affirmed

the trial court’s ruling on direct review. Pet. App. 261a,

286a. It recited the evidence on this claim, and

determined that petitioner “failed to prove the presence of

a Bible in the jury room was outside influence that

affected the jury’s verdict in violation of Article 36.13.” Jd.

at 286a.

5. This Court thereafter denied petitioner’s first

certiorari petition. Pet. App. 59a.

6. a. Petitioner subsequently file? an application for

a writ of habeas corpus in federal! district court. Pet. App.

57a. That court denied petitioner’s request for an

6

evidentiary hearing (a decision petitioner does not

challenge here), and also rejected his claims on the merits.

Id. at 62a-66a.

b. The district court also granted petitioner’s request

for a certificate of appealability on his Sixth Amendment

claims. Pet. App. 67a-68a.

7. A panel of the Fifth Circuit affirmed. Pet. App.

30a.

a. The panel first held that the “jury’s use of the Bible

during the sentencing phase * * * amounted to an

improper external influence on the jury’s deliberations.”

Pet. App. 30a. The panel examined the existing Supreme

Court case law on the issue, and concluded that the Court

has “clearly established a constitutional rule forbidding a

jury from being exposed to an external influence.” Jd. at

12a. The panel then reviewed the existing circuit law,

openly acknowledged the split of authority on this

question, and determined that the jurors erred in

consulting material that was “not part of the law and

evidence that the jury was to consider in its deliberations.”

Id. at 20a-21a.

b. The panel next turned to a harmless-error analysis.

It held, first, that the Brecht standard of review applied to

constitutional errors of this nature, Pet. App. 24a; and,

second, that the state court’s factual findings were

entitled to deference: petitioner “failed to demonstrate

that the state court’s finding that the Bible did not

influence the jury lacks ‘even fair support in the record,”

1d. at 28a-29a.

7

REASONS FOR DENYING THE PETITION

Petitioner has indeed identified an important issue

that has sharply divided the circuits—an issue that

implicates the constitutional right to an impartial jury,

and one that presumably applies to civil as well as

criminal matters. See, eg, U.S. Const. amend VI;

Remmer v. United States, 347 U.S. 227, 229-30 (1954);

Wilson v. Vermont Castings, 170 F.3d 391, 395 (3d Cir.

1999) (applying comparable standards in the civil context);

Haley v. Blue Ridge Transfer, 802 F.2d 1532, 1535 (4th

Cir. 1986) (same). This issue, therefore, has the very real

potential to deeply affect jury deliberations in virtually

every single trial, civil and criminal, across the nation.

Nevertheless, though the question truly is important and

recurring, this is not the right vehicle for resolving it.

Petitioner's first question presented—focusing on his

main constitutional claim—suffers from a variety of

critical vehicle problems, and in any event lacks merit.

His second question presented—focusing on the standard

of review—is not the subject of any genuine circuit split,

and is otherwise unworthy of the Court’s attention. And

his third question presented—a fact-bound attack on the

lower court’s application of settled law to the unique

circumstances of this case—is both incorrect in substance

and insubstantial in importance.

The critical constitutional question raised in this

petition should be resolved in a case where it will have at

least a realistic chance of affecting the outcome. That is

not the situation here. Given its recurring nature, another

opportunity should soon arise to evaluate the claim where

it properly belongs: on direct review, outside the stringent

AEDPA framework. Given the substantial barriers to

8

reaching the merits in the procedural posture of this case,

the petition should be denied.

I. THIS CASE IS NOT A PROPER VEHICLE FOR

RESOLVING THE FIRST QUESTION PRESENTED

Petitioner’s first claim—that the jury’s limited use of

the Bible during sentencing proceedings violated his right

to an “impartial jury” (Pet. 21-26)—suffers from at least

four substantial vehicle problems and (in any event) lacks

merit. Because the Court can resolve this case without

resolving the question presented, further review is not

warranted.

A. This petition is an unsuitable vehicle for reviewing

the underlying constitutional issue for a fundamental

reason: the case arises not on direct review but collateral

attack. As discussed below, petitioner cannot possibly win

under the controlling AEDPA standard. The question is

not whether the state courts erred, but instead whether

the error was “clearly established” under this Court’s

precedent. 28 U.S.C. 2254(d)(1). The very split among the

circuits—including carefully reasoned decisions by

unanimous panels on each side—eliminates any

conceivable argument that the law is so clear that the

state-court decision is subject to correction on collateral

review. This accordingly is the rare case in which the

undeniable existence of a circuit split warrants a quick

and obvious deny. See, e.g., Carey v. Musladia, 549 U.S.

70, 76 (2006) (finding that evidence of wide lower-court

“diverge[nce]” over an issue “[rJeflect[s] the lack of

guidance from this Court”).

1. In a long series of cases, this Court has repeatedly

reinforced that Congress meant what it said when it

9

ordered federal courts not to upset state convictions not

obtained in violation of “clearly established” law, 28 U.S.C.

2254(d)(1). See, e.g., Brown v. Payton, 544 U.S. 133, 143

(2005) (reviewing state-court decision applying Supreme

Court precedent “to similar but not identical facts” and

concluding that “lelven on the assumption that its

conclusion was incorrect, it was not unreasonable, and is

therefore just the type of decision that AEDPA shields on

habeas review”); Wiggins v. Smith, 539 U.S. 510, 520

(2003) (AEDPA “circumscribels]” the Court’s consideration

of petitioner’s habeas claims and “limit[s]” its analysis “t

the law as it was ‘clearly established’ by our precedents at

the time of the state court’s decision”); see also Carey, 549

U.S. at 81 (Kennedy, J., concurring) (a rule should be

“established in this Court before it can be grounds for

relief’ on habeas review).

Nor were these congressional constraints on habeas

relief mere technical requirements. These restrictions

reflect the important and substantial policy interests in

“comity, finality, ane federalism.” Willamsv. Taylor, 529

U.S. 420, 436 (2000); see also Duncanv. Walker, 533 U.S.

167, 178 (2001) (same). “A federal court’s collateral review

of a state-court decision must [accordingly] be consistent

with the respect due state courts in our federal system.”

Miller-El v. Cockrell, 537 U.S. 322, 340 (2003). Part of

that respect entails leaving in place reasonable state-court

decisions, even if a reviewing court disagrees with the

ultimate conclusion on the merits. Bel/v. Cone, 535 U.S.

685, 697 (2002).

These principles collectively doom petitioner’s claim.

In inviting the Court to discard AEDPA’s limitations—and

reach the merits of a constitutional question over which

10

the law is unavoidably unclear—petitioner plainly errs.

Indeed, petitioner’s proposed course of action is

indistinguishable from the analysis that resulted in this

Court summarily reversing the Ninth Circuit twice on a

single day at the start of the 2002 Term. See Woodfordv.

Visciotti, 537 U.S. 19 (20C2) (per curiam); Harlyv. Packer,

537 U.S. 3 (2002) (per curiam). Because it makes no

difference “[w]hether or not [the Court] would reach the

same conclusion” as the state court here, Woodford, 537

U.S. at 27, petitioner’s first question presented is

irrelevant to the outcome of this proceeding. Review

should accordingly be denied. See also Karly, 537 U.S. at

11 (noting that the Court would reach the same

result—reversing the Ninth Circuit for improperly

discarding a state-court decision—“lelven if we agreed

with the Ninth Circuit majority” on the merits).

2. Nor is there any real dispute that the law is

unsettled in this area. See Pet. 5 (recognizing the circuit

conflict over the “impartiality” question). The very fact

that circuits have divided over whether there is any crror

(much less clearly established error) eliminates any

plausible argument that existing Supreme Court law

resolves this question.

Indeed, the Fourth Circuit has alternately recognized

these kinds of juror actions as legitimate (on the one hand)

and recognized the uncertainty itself as a basis for

denying habeas review (on the other). Compare, e.g.,

Burch v. Corcoran, 273 F.3d 577, 590-91 (4th Cir. 2001)

(holding that juror’s quoting of the Bible during

deliberations—both from memory and by reading—“did

not constitute an improper jury communication”), with

Robinson v. Polk, 438 F.3d 350, 361-66 (4th Cir. 2006)

11

(declaring the pre-existing case law non-controlling and

distinguishing Vanner v. United States, 483 U.S. 107

(1987), Parkerv. Gladden, 385 U.S. 363 (1966), Turnerv.

Louisiana, 379 U.S. 466 (1965), and Remmer v. United

States, 347 U.S. 227 (1954)); and Robinson v. Polk, 444

F.3d 225, 225-30 (4th Cir. 2006) (Wilkinson, J., concurring

in the denial of reh’g en banc). Moreover, the Ninth

Circuit—after first outright declaring the practice

constitutional, Fields v. Brown, 431 F.3d 1186, 1208-09

(9th Cir. 2005)—granted en-banc rehearing and ultimately

left the question unanswered by concluding that any error

was harmless in the context of that case, Fie/ds v. Brown,

503 F.3d 755, 780-81 (9th Cir. 2007) (en banc).

Notwithstanding the technical backtracking, the en-banc

majority's opinion still suggested with strong language

that the practice was _ permissible. id. at 780

(characterizing the Biblical verses as “notions of general

currency that inform the moral judgment that capital-case

jurors are called upon to make”). And, finally, the Tenth

Circuit has apparently considered the practice so well

settled—and so plainly constitutional—that it refused to

grant a certificate of appealability to review an improper-

juror-conduct claim. Nez//v. Gibson, 278 F.3d 1044, 1064

n.8 (10th Cir. 2001).

The Fifth Circuit, by contrast, reached exactly the

opposite conclusion on the same issue in its decision

below, relying on cases from the First, Sixth, and Eleventh

Circuits for support. See Pet. App. 14a-23a. The split is

therefore real, defined, and openly acknowledged—and is

precisely the reason the petition is manifestly

inappropriate for review.

12

In short, when lower courts sharply split over a

constitutional question—with some suggesting the conduct

is perfectly permissible—it is difficult to characterize the

law as “clearly established.” 28 U.S.C. 2254(d)(1). For

that very reason, circuit “splits” on the underlying merits

are rarely properly resolved in the habeas context, as the

“necessary condition for federal habeas relhef’—clearly

established law—is routinely left unsatisfied. Kane v.

Espitia, 546 U.S. 9, 10 (2005) (per curiam).

3. This split is perhaps unsurprising in light of the

fact that existing Supreme Court case law does notresolve

this question. See United States v. Lara-Ramurez, 519

F.3d 76, 88 (ist Cir. 2008) (“We have never decided a case

involving a Bible in a jury room, nor has the Supreme

Court.”).

As a preliminary matter, the Fifth Circuit has

overstated the reach of existing Supreme Court precedent

on this subject. Every single decision cited in the panel’s

opinion addressed factua/ information—the kind of

extrinsic evidence that aids a jury’s understanding of the

defendant’s factua/ innocence or guilt—or cases in which

third parties sought to coerce jurors into voting a certain

way. See, eg., Robinson, 438 F.3d at 363 (finding it

reasonable to conclude that “the Bible had no bearing on

any fact relevant to sentencing, and was therefore not

tantamount to ‘evidence’ that was used against him at

sentencing,” distinguishing Parkerand Turner (emphasis

in original); Billings v. Polk, 441 F.3d 238 (4th Cir. 2006).

Petitioner has wholly failed to identify any of this Court’s

precedent applying the same rule to information

implicating only mora/ beliefs—not the kind of evidence

typically introduced at trial, but the personal thoughts

13

and moral outlook that a juror is supposed to bring to

deliberations. See, e.g. Fields, 503 F.3d at 780

(contrasting “the [factual] question of innocence or guilty

of the offense” with capital sentencing, which is “a matter

of reasoned moral judgment”).

These distinctions undoubtedly matter: “the task of

jurors at the penalty phase is qualitatively different form

that at the guilt phase”: “[a]t the penalty phase, jurors are

asked to make a normative determination—one which

necessarily includes moral and_ ethical

considerations—designed to reflect community values.”

Peoplev. Danks, 82 P.3d 1249, 1277 (Cal. 2004). It stands

to reason that a rule governing the use of information at

one phase therefore might not control the use of a

categorically different kind of information at a

qualitatively different stage of trial. Because “Section

2254(d)(1) would be undermined if habeas courts

introduced rules not clearly established under the guise of

extensions to existing law,” Yarborough v. Alvarado, 541

U.S. 652, 666 (2004), the court below erred in assuming

that a general rule, structured to meet the needs of

controlling a factual record, would apply across the

-board—even where the kind of inputs affecting moral

outlook are not required to be introduced through

witnesses at trial. See Fields, 503 F.3d at 780 (“It is

difficult to see how sharing notes can be constitutionally

infirm if sharing memory isn’t.”); Billings, 441 F.3d at 248

(explaining that Remmer's rule does not obviously apply

“whenever a juror reads a book that influences his

thinking about the case”).

Because the Court can dispose of petitioner’s claim

solely on the lack of clearly established law, there is no

14

reason to reach the underlying merits: “Our own

consideration of the merits * * * is for another day, and

this case turns on the recognition that no clearly

established law contrary to the state court’s conclusion

justifies collateral relief.” Wrightv. Van Patten, 1285. Ct.

743, 747 (2008) (per curiam). This alone is a compelling

reason to deny review.

B. For predominantly the same reasons, the anti-

retroactivity rule in Teague v. Lane, 489 U.S. 288 (1989)

(per curiam), stands as another substantial barrier to

reaching the underlying question on the merits. See Horn

v. Banks, 536 U.S. 266, 272 (2002) (per curiam) (holding

that Teague stands as an independent bar to certain relief

sought in habeas proceedings). Under Teague’s command,

new constitutional rules of criminal procedure are

inapplicable on collateral review unless they qualify as

“watershed” rules. 489 U.S. at 311. And if petitioner’s

new rule cannot apply immediately, this case does not

present an adequate vehicle for addressing it.

As an initial matter, there is no serious contention

that this rule, unlike virtually all others preceding it, fits

within the exceedingly narrow category of “watershed”

developments. See, e.g., 7ylerv. Cain, 533 U.S. 656, 665,

667 n.7 (2001). And, critically, petitioner’s theory, if

adopted, would plainly stand as a “new” rule: the holding

is not so “dictated” by existing precedent that no

“reasonable juristl]” would decline to adopt it. Beard v.

Banks, 542 U.S. 406, 413 (2004); see also sd. at 411 (asking

whether existing precedent “compels the rule”). The same

division of authority cited above—and the same

distinctions between the subject-matter of existing

precedent (factual record) andthe impartiality objection at

15

issue here (moral outlook)—demonstrate that reasonable

jurists not only can but Aave found existing precedent not

to “compel” the rule sought in this case. See, e.g., Fields

v. Brown, 503 F.3d 755, 780 (9th Cir. 2007) (en banc);

Robinson v. Polk, 438 F.3d 350, 362-66 (4th Cir. 2006).

Because Jeague presents its own distinct obstacle to

relief irrespective of the underlying merits of petitioner’s

claim, it demonstrates yet another reason that this case is

an inappropriate vehicle for deciding the Sixth

Amendment question. Horn, 536 U.S. at 272 (“in addition

to performing any analysis required by AEDPA, a federal

court considering a habeas petition must conduct a

threshold Teague analysis when the issue is properly

raised by the state”).

C. There is an additional reason why this petition is

a particularly poor vehicle to address this constitutional

issue: petitioner prevailed on this issue below, and simply

failed to obtain relief on other grounds. See Pet. App. 23a-

30a. Because these constitutional errors are subject to the

Brecht standard of review, see Part II, znfra, there is no

“need to decide whether there was juror misconduct

because even assuming there was,” it is apparent that the

Bible consultation “had no substantial and injurious effect

or influence in determining the jury’s verdict.” Fie/ds v.

Brown, 503 F.3d 755, 781 (9th Cir. 2007) (en banc); see

also Pet. App. 23a-30a (so holding). Because the Court has

no reason to resolve the underlying constitutional question

when any recognized error would lack the requisite impact

on the ultimate disposition, this is not a suitable vehicle

for resolving this question.

D. The petition also suffers from a final potential

vehicle problem: invited error. As respondent noted in his

16

supplemental brief before the Fifth Circuit (at 20-21),

petitioner’s counsel, in his closing argument, explicitly

invited the jury to consider passages from the Bible,

including one focused on “forgivel[ness].” 43 Rep. Rec. 195.

Having asked the jury to consider the Bible when

exercising moral judgment during sentencing, petitioner

cannot obviously now fault the jury for following his

counsel’s own suggestion. This further undermines this

petition as the right vehicle for resolving this important

and difficult question. See, e.g., United States v. Wells,

519 U.S. 482, 488 (1997) (noting invited error as a

“consideration|] bearing on whether to decide a question”

even after a petition has been granted for review) (citing

Springtield v. Kibbe, 480 U.S. 257, 259-60 (1987) (per

curiam)).

E. In any event, petitioner’s theory is wrong on the

merits. As established above, there is a clear and material

distinction between information affecting the factual

record and information shaping moral beliefs.

The Court’s case law on external influence is focused

on the need to ensure that factua/ evidence is admitted

through the witness stand, where it is challenged by

cross-examination, subject to judicial scrutiny, and tested

through the adversarial process. See, e.g., Turner, 379

U.S. at 472-73. It logically follows that the category of

information barred by the “external evidence” rule is the

kind of information naturally subject to these constraints:

evidence that ordinarily enters the record through official

court proceedings, not through outside sources. Robinson,

444 F.3d at 225 (Wilkinson, J., concurring in the denial of

en-banc reh’g).

bY

By contrast, the kind of information that typically

shapes one’s moral beliefs typically arises outside the

judicial process (whether gained from religious sources,

non-religious sources, philosophy texts, or any other

resource). None of this information is subject to cross-

examination, introduced through a witness, or tested by

the adversary process. And yet no one “suggests that

[jurors] are not free to recite these points, including those

from the Bible, or to resort to their reasoning.” Fie/ds, 503

F.3d at 780; Robinson, 438 F.3d at 364 (“reading the Bible

is analogous to the situation where a juror quotes the

Bible from memory, which assuredly would not be

considered an improper influence”).

If “morality based” information is permitted prior to

the empaneling of a jury, it is difficult to see why a

different rule should attach if the same information arises

during jury deliberations. The timing has no obvious

constitutional significance under any controlling case law.

The standard is appropriately focused on substance, not

timing. If the information is permissible at all, it must be

permissible no matter when it is obtained. Along the same

lines, extrinsic factual evidence concerning questions of

guilt or innocence is problematic no matter when a juror

receives the information—for example, a juror learning of

a suppressed confession before the jury is empaneled

would implicate the same impartiality concerns as a juror

learning of the same evidence during deliberations.

Jurors are charged with exercising moral judgment,

and it is assumed that they do so when addressing

difficult moral issues. It is widely accepted that these

judgments are dependent upon a person’s life experiences

and ordinary interactions. Yet these very influences are

18

not introduced in open court; they do not become part of

the evidentiary record; they are not expressed through the

witness stand, cross-examined, or tested against any legal

standard by a judge. They are, in fact, part of who the

jurors are—and therefore are properly recognized as the

collective wisdom brought together when a cross-section of

the community is selected to form a 12-person jury. The

information shaping a jurors moral outlook is not

automatically invalid simply because it arises during the

course of trial.

Il. THE SECOND QUESTION PRESENTED FAILS TO

IMPLICATE ANY CIRCUIT SPLIT AND IS

OTHERWISE UNWORTHY OF REVIEW

A. Petitioner is incorrect that this case presents a

second purported split. There is no genuine conflict at all

over the correct standard for determining harmlessness.

Every circuit to have squarely addressed the question has

uniformly held that the standard in Brecht v.

Abrahamson, 507 U.S. 619 (1993)—the same standard

that this Court has repeatedly held applies to virtually all

constitutional claims on collateral review, see, e.g., Fry v.

Pliler, 127 S. Ct. 2321, 2325 (2007)—is also the standard

that applies here: unless petitioner can show that the

error had a “substantial and injurious effect” on the

outcome, he cannot obtain relief. See, eg., Pet. App.

24a-25a (“on habeas review, we do not use the normal

harmless error analysis”; “habeas petitioners are not

entitled to relief based on a constitutional error unless the

error ‘had [al substantial and injurious effect” on the

outcome); Fields v. Brown, 503 F.3d 755, 781 (9th Cir.

2007) (en banc) (applying Brecht to an analogous claim);

Crease v. McKune, 189 F.3d 1188, 1192 (10th Cir. 1999)

19

(same); Brbbins v. Dalsheim, 21 F.3d 13, 16 (2d Cir. 1994)

(same).

Petitioner claims (Pet. 26-28) that the Eleventh

Circuit has in fact adopted a lower standard in A*e Nair v.

Campbell, 416 F.3d 1291 (11th Cir. 2005), but that is

wrong. In that case, the panei found the prisoner’s claim

procedurally defaulted, so any comment on _ the

appropriate standard of review was merely dicta. See 416

F.3d at 1307 (“the district court should have dismissed

this claim as procedurally barred”). It also was

particularly meaningless dicta: the panel said only that

the error was harmless beyond a_ reasonable

doubt—meaning that the state in that case could have

satisfied any standard (including the stricter Brecht

standard), so any holding that a lesser standard applied

was dicta once again. See zd. at 1309 (“Because the state

could successfully rebut the presumption of prejudice

arising from the jury’s consideration of extraneous

evidence, McNair would not have been entitled to relief on

this claim even if he had properly raised it in the state

court.”).

What’s more, the panel nowhere confronted the

“standard of review” question as framed here. It did not

acknowledge Brecht (in order to adopt or reject it). And it

did not ask whether different standards should apply in

direct and collateral review. McNair accordingly does not

create a reasoned split warranting any further review.

See also Fields, 503 F.3d at 782 n.21 (“the court in McNair

held that the issue was procedurally defaulted” and

“stated that even if it weren’t there was no prejudice”).

B. In any event, petitioner’s theory is irreconcilable

with Brecht and its controlling rationale. Every one of

20

Brechts concerns—fostering comity and federalism,

preserving finality, lmiting the habeas writ to its

traditional function of protecting only those “grievously

wronged,” and respecting state sovereignty over criminal

matters—applies with full force to this case. See 507 U.S.

at 633-34.’

C. In the alternative, petitioner argues that this kind

of constitutional error should be subject to an irrebuttable

presumption of prejudice. Pet. 28°34. The novel

contention that this type of ordinary constitutional error

should be subject in this collateral challenge to automatic

reversal—a standard not even applicable on direct review,

see Remmer v. United States, 347 U.S. 227, 229

(1954)—would stand every single one of Congress’s core

restrictions on habeas review (as all notions of comity,

federalism, and finality) straight on its head. Cf. United

States v. Lara-Ramirez, 519 F.3d 76, 88 (1st Cir. 2008)

{rejecting the argument that the “taint” arising from a

“Bible in the jury room * * * cannot be cured”). It would

also impose great social costs—requiring countless retrials

for “errors” that had no conceivable effect on the

il. In any event, this question is not adequately presented here,

because petitioner clearly fails to satisfy any standard of

harmlessness. The state court determined that no juror would have

changed his or her vote as a result of reading the Bible. Because that

finding is supported by the record and is not “unreasonable” (28 U.S.C.

2254(d)(2)) or contradicted by “clear and convincing evidence” (28

U.S.C. 2254(e)(1)), it is controlling on collateral review. See Mitchell

v. Esparza, 540 U.S. 12, 17-18 (2004). Petitioner therefore cannot

satisfy any standard of harmless error: if, as a historical matter, the

error was in fact harmless, it necessarily follows that it was legally

harmless, too. This is yet another reason that the petition should be

denied.

21

ae

outcome—without any countervailing benefit. See Rushen

v. Spain, 464 U.S. 114, 118-19 (1983).

Finally, Petitioner’s theory has no obvious limiting

principle and hence would prove unworkable in practice.

The petition reflects a clear inability to distinguish this

putative constitutional error from any other constitutional

error that the Court has repeatedly found is appropriately

evaluated under harmless-error review. After all, the

claim at its core is that the jury considered “evidence” that

should have been excluded from the record. But this

complaint describes the vast majority of trial errors: the

admission of evidence from wrongful searches and

seizures; the consideration of coerced or otherwise flawed

confessions; the introduction of tainted evidence; the use

of hearsay evidence from witnesses not subject to

confrontation; prosecutorial comments about a defendant’s

refusal to testify; and so on—all leave the jury considering

another evidentiary piece that should have been excluded

from the puzzle. Yet the Court has routinely found such

errors subject to harmless-error analysis, because the

impermissible evidence can always be evaluated against

the permissible evidence, and a conclusion reached about

how an objective juror might have been swayed (or not) by

what should not have been there. See Fry, 127 S. Ct. at

2325-26; Brecht, 507 U.S. at 629-30.

There is absolutely no reason to impose a different

standard here—and, perhaps for that very reason, the

Court’s precedent directly addressing the effects of

external evidence has always held that the hint of

prejudice could be proved harmless on appeal (even on

direct review). See, eg, Remmer, 347 U.S. at 229.

22

Petitioner has failed to identify aiy reason to back away

from this sound standard.

Iii, THE THIRD QUESTION PRESENTED ONLY

SEEKS TO REVISIT A FACT-BOUND DISPUTE

UNWORTHY OF REVIEW

A. Peiitioner’s final argument is that the lower courts

‘incorrectly applied” AEDPA’s provisions to the facts of

this case. Pet. 38. This is a fact-bound challenge.

Petitioner does not question the applicable legal

framework, and for good reason: Congress was hardly

equivocal in its command that federal courts defer to

state-court fact-finding. See 28 U.S.C. 2254(d)(2), (e)(1);

see also Rushen v. Spain, 464 U.S. 114, 120 (1983). He

simply disagrees with how all the courts to date have

understood the factual record.

This is a capital case, and so alleged factual errors do

deserve careful scrutiny. But this claim has been carefully

addressed by multiple state and federal courts on direct

and collateral proceedings. These courts have all

scrutinized the factual record and all reached the same

conclusion: petitioner has failed to show that any reference

to the Bible had any effect on the jury’s deliberations. See,

e.g., Pet. App. 28a-30a. There is no reason for the Court

to grant review for the single purpose of reassessing (and,

inevitably, reaffirming) the particular factual-findings in

this single case. Thisclaim consequently does not warrant

further view.

B. Nor is Petitioner correct that the state courts

improperly truncated his factual examination.

Specifically, Petitioner argues that he was prevented from

questioning the jury about the Bible’s impact on the actual

23

deliberations, yet punished for not demonstrating that

very kind of irmpact. This is incorrect. The state court

properly restricte! the inquiry into the jury’s mental

processes—such an inquiry is forbidden under both state

and federal law, see Fed. R. Evid. 606; Tex. R. Evid. 606,

and for significant policy reasons. See, e.g., Tanner v.

United States, 483 U.S. 107, 120-21 (1987).

Furthermore, the factual findings that the state courts

made—and the Fifth Circuit found worthy of AEDPA

deference—consisted of two categories of Aistorica/ facts,

and each were equally open for petitioner to explore: first,

the objective facts surrounding the jury’s deliberations

(including what materials were reviewed at what time by

which jurors); and second, a reasonable inference based on

those objective facts. Pet. App. 28a-30a. The state court’s

ultimate findings accordingly were not based on direct

testimony about how the Bible may have affected an

individual juror’s internal thoughts. They were instead

based on what happened (from the view of a neutral

outside observer) factua//y during deliberations.

And the state court’s findings, far from clearly wrong,

in fact had ample record support. A number of jurors, for

example, testified that the Bible was not even consulted

until after the sentencing vote was submitted: they were

simply looking for consolation while awaiting their return

to the courtroom. Since a post-hoc reading cannot possibly

affect a vote already cast, the court’s findings are easily

supported by this testimony alone.

Moreover, all jurors testified that the Bible was read

and discussed only in small groups, never during full

deliberations—an objective fact that supports an inference

that the Bible was not used as a material factor in the

24

overall discussions or decision (since the entire jury was

never once focused on it).

Finally, every single juror at the hearing testified that

the jury was properly instructed to consider only the

Court’s legal charge and apply it to the evidence in the

record, and that no one, at any point, suggested that the

legal standard be supplanted by a religious one. Given the

traditional presumption that jurors are presumed to follow

the court’s instructions, see, e.g., Fieldsv. Brown, 503 F.3d

755, 782 (9th Cir. 2007) (en banc) (citing Kansasv. Marsh,

548 U.S. 163 (2006)), this too supports a finding that the

jury was not unduly swayed by any identified “outside”

infiuence.

There is no basis in the record for upsetting these

factual findings. Petitioner has not cited any admissible

evidence that undercuts these facts, and does not suggest

how his view is even consistent with the objective record

evidence. This dooms his claim: once the historical] facts

are established, it is plainly reasonable to infer that the

jury was not unduly influenced—if influenced at all—by

what a handful of jurors read in the Bible during a time

when the group was not deliberating as a whole. Because

that conclusion reflects the most natural reading of the

factual record, it certainly was not “cbjectively

unreasonable” or undercut by “clear and convincing”

evidence. See 28 U.S.C. 2254(d)(2), (e)(1); Rushen, 464

U.S. at 121 n.6. The court below hence was correct to

defer to the factual findings.’

2. Petitioner has not challenged (and therefore has waived) any

objection to the Fifth Circuit’s decision on the ground that it asked

what these individual jurors (as opposed to an objective juror) would

25

CONCLUSION

The petition for a writ of certiorari should be denied.

Respectfully submitted.

GREG ABBOTT JAMES C. HO

Attorney General of Texas Solicitor General

Counsel of Record

C. ANDREW WEBER

First Assistant Attorney DANIEL L. GEYSER

General Assistant Solicitor General

ERIC J. R. NICHOLS

Deputy Attorney General

for Criminal Justice

EDWARD L. MARSHALL

Chief, Postconviction

Litigation Division

MARCH 2009

have thought. To be sure, it is far from obvious that the Fifth Circuit's

holding actually turned on anything but an objective standard. But to

the extent an objective standard applies (and, furthermore, to the

extent the lower court failed to follow it), any challenge is plainly

waived. Petitioner’s argument is premised not on the notion that the

beliefs of the actual jurors were irrelevant, but rather that they were

so highly relevant that the state court erred in refusing to permit a

searching inquiry into the specific mental processes of each juror in

reaching a fina] decision. Pet. 34-36. That contention is surely wrong

on the law—this Court alone has rejected it on a number of occasions.

E.g., Rushen, 464 U.S. at 121 n.5. But any contrary challenge, based

on the (flawed) notion that the factual findings were wrongly focused

on individual juror’s mental perceptions, is just as surely waived.