Hassine v. Zimmerman

Opinions of the United

1998 Decisions States Court of Appeals

for the Third Circuit

11-9-1998

Hassine v. Zimmerman

Precedential or Non-Precedential:

Docket 97-1969

Follow this and additional works at: http://digitalcommons.law.villanova.edu/thirdcircuit_1998

Recommended Citation

"Hassine v. Zimmerman" (1998). 1998 Decisions. Paper 260.

http://digitalcommons.law.villanova.edu/thirdcircuit_1998/260

This decision is brought to you for free and open access by the Opinions of the United States Court of Appeals for the Third Circuit at Villanova

University School of Law Digital Repository. It has been accepted for inclusion in 1998 Decisions by an authorized administrator of Villanova

University School of Law Digital Repository. For more information, please contact Benjamin.Carlson@law.villanova.edu.

Filed November 9, 1998

UNITED STATES COURT OF APPEALS

FOR THE THIRD CIRCUIT

No. 97-1969

VICTOR HASSINE,

Appellant

v.

CHARLES ZIMMERMAN, Superintendent, and THE

ATTORNEY GENERAL OF THE STATE OF PENNSYLVANIA

ON APPEAL FROM THE

UNITED STATES DISTRICT COURT

FOR THE EASTERN DISTRICT OF PENNSYLVANIA

(D.C. Civil No. 86-6315)

Argued: August 4, 1998

Before: NYGAARD, ALITO, and RENDELL, Circuit Judges

(Opinion Filed: November 9, 1998)

Donald J. Goldberg (Argued)

Leslie H. Smith

Ballard Spahr Andrews &

Ingersoll, LLP

1735 Market Street, 51st Floor

Philadelphia, PA 19103-7599

Attorneys for Appellant

C. Theodore Fritsch, Jr. (Argued)

Stephen B. Harris

Alan M. Rubenstein

District Attorney's Office

Bucks County Courthouse, 4th Floor

Doylestown, PA 18901

Attorneys for Appellees

OPINION OF THE COURT

RENDELL, Circuit Judge:

On January 4, 1983, a Pennsylvania state court

sentenced Victor Hassine to life in prison, following his

1981 conviction on charges of first degree murder,

attempted murder, criminal conspiracy, and criminal

solicitation. Now, more than fifteen years later, we are faced

with the question of whether to vacate Hassine's conviction

by granting his petition for habeas relief brought pursuant

to 28 U.S.C. S 2254. Hassine contends that relief is

warranted because the state prosecutor sought to use his

post-arrest silence for impeachment purposes at trial in

violation of the due process principles established in Doyle

v. Ohio, 426 U.S. 610 (1976). The district court found that

a Doyle violation had occurred, but it concluded that any

constitutional error was harmless under the standard

announced in Brecht v. Abrahamson, 507 U.S. 619 (1993).

We agree that the prosecutor violated Doyle by seeking to

elicit testimony concerning Hassine's post-arrest silence.

We also agree that Brecht is the proper standard to apply

on collateral review. Because we agree further that the

Doyle violation was harmless under Brecht, we will affirm.

I. FACTUAL AND PROCEDURAL HISTORY

A. The Murder Conspiracy

This case arises out a conspiracy involving Hassine and

his co-conspirators George Gregory Orlowski and William

Eric Decker, which culminated in the August 22, 1980,

2

murder of James Puerale and the shootings of Albert "Skip"

Kellet and Lois Kellet.

The evidence at trial showed that Hassine first met

Orlowski in 1979, and that shortly thereafter they decided

to open a store in Morrisville, Pennsylvania, called Greg's

Quality Meat Market ("the Market"). The store was financed

by Hassine's family, overseen and supervised by Hassine

himself, and operated on a daily basis by Orlowski.

However, the Market soon experienced financial problems,

and, with Hassine's knowledge, Orlowski began selling

marijuana and methamphetamine from a back room in the

store.

In June 1980, Albert Kellet, a close friend of Orlowski,

purchased $150 of methamphetamine at the Market. Upon

returning home, Kellet discovered that the drugs were of

inferior quality and he became enraged. He called Orlowski

and invited him to his apartment under the false pretense

of wishing to buy more methamphetamine. When Orlowski

arrived, Kellet threatened him with a club, stole all of his

money and drugs, and threw him out of the apartment.

Several days later, Orlowski, Hassine, and Decker, among

others, met at the Market to discuss Kellet's actions. The

State's witnesses testified that at this meeting, Hassine

announced that he wanted Kellet "wasted" and that if Lois

Kellet, Albert's wife, was present, she "was to go also,

because any witnesses had to go." As a result, Hassine,

Decker, and Orlowski made several attempts over the next

month to obtain a gun with which to kill Kellet, and they

investigated the possibility of paying two other individuals

to have Kellet murdered. A number of confrontations

between Hassine and Kellet also erupted during this time,

and, on at least one occasion, Hassine instructed Decker to

shoot Kellet and to kill him. Because it was daylight and a

witness was present, Decker declined. Nevertheless, on

August 22, 1980, Decker had his own encounter with

Kellet, and he returned to the Market to tell Hassine that

"[t]onight's the night -- the cat's got to go. We'll use your

gun."

According to the State's witnesses, Hassine picked

Decker up later that evening and drove him to Hassine's

3

parents' house in Trenton, New Jersey. While Decker waited

in the car, Hassine entered the house and obtained his

father's .380 caliber Llama handgun. Hassine then drove

Decker to Kellet's apartment building and gave him the gun

and a New York Yankees batting helmet to cover his hair.

As Decker approached the building, he saw Kellet in a first-

floor apartment watching television with his wife Lois, and

with James Puerale and George Sofield. Decker entered the

apartment, surveyed the room, and opened fire, killing

Puerale instantly and injuring Skip and Lois Kellet with

shots to the head.

The police arrested Decker the next day, and arrested

Hassine three months later, charging Hassine withfirst

degree murder, attempted murder, conspiracy, and

solicitation.

B. The Murder Trial

The State presented thirty-four witnesses at Hassine's

trial, including Decker, who had negotiated a plea bargain

with the District Attorney to avoid the death penalty.

Decker and the other witnesses testified as to the details of

the murder conspiracy and described Hassine's extensive

involvement in the plot to murder Kellet.

Hassine then took the stand in his own defense and, for

the first time since his arrest, offered an innocent

explanation for his role in the conspiracy. He testified that

he had obtained his father's gun and stored it in the

Market after Orlowski had asked him for protection from

Kellet. He then claimed that Decker entered the Market on

the day of the murder, and, without Hassine's knowledge,

took the gun and left the store "ranting and raving, saying

he was going to get Skip Kellet." Hassine thus maintained

that he never told Decker to kill Kellet, that he never gave

Decker a gun with which to carry out the crime, and that

he was never part of the murder conspiracy.

Attempting to discredit these claims, the prosecutor

asked Hassine a series of questions on cross-examination

regarding Hassine's post-arrest silence. In particular, he

inquired three times as to why Hassine had not offered the

same exculpatory story to the authorities following his

4

arrest. Hassine's attorney objected each time, believing that

the questions violated Hassine's rights under Doyle v. Ohio,

which provides that the government cannot use a

defendant's post-arrest, post-Miranda silence for

impeachment purposes at trial. 426 U.S. at 610. The trial

judge agreed, sustaining the objections and preventing

Hassine from answering the prosecutor's questions. The

prosecutor also made two general references to Hassine's

silence in his closing argument. Nevertheless, the court did

not provide curative instructions during Hassine's

testimony or in the jury charge.

The case was sent to the jury at the close of all evidence,

and on June 11, 1981, the jury found Hassine guilty and

recommended that he be sentenced to life in prison on the

charge of first degree murder. After denying Hassine's

motions for a new trial and arrest of judgment, the court

adopted the jury's recommendation and sentenced Hassine

to life imprisonment, in addition to several consecutive

prison terms extending from two to twenty years.

C. Review of the State Court Conviction

Hassine appealed his conviction to the Superior Court of

Pennsylvania, assigning fifteen reversible errors to the trial

court. See Commonwealth v. Hassine, 490 A.2d 438, 443-

44 (Pa. Super. Ct. 1985) ("Hassine I"). Among other things,

he argued that the state prosecutor had violated the due

process principles established in Doyle by interrogating him

on the witness stand about his post-arrest silence. On

February 8, 1985, the Superior Court denied Hassine's

appeal in its entirety and affirmed his sentence,finding

that his list of alleged errors was "short on merit," id. at

444, and rejecting his claim of prosecutorial misconduct

based on Doyle. Relying on state precedent, the court noted

that testimony regarding post-arrest silence can be elicited

"to refute contrary statements volunteered by the defendant

to demonstrate his cooperation at the time of questioning,"

id. at 451 (quoting Commonwealth v. Bey, 439 A.2d 1175,

1177 (1982)), and concluded that "[w]e believe this occurred

here. The district attorney elicited testimony designed to

rebut [Hassine's] claim that he cooperated with the police,

5

not to suggest [Hassine's] guilt to the jury. Hence we find

no error." Hassine I, 490 A.2d at 451.

On May 13, 1985, Hassine applied for a Petition for

Allowance of Appeal in the Supreme Court of Pennsylvania,

claiming again, inter alia, that the prosecution asked

improper questions at trial about his post-arrest silence.

On April 15, 1986, however, the state Supreme Court

denied the Petition and Hassine's conviction became final.

Hassine chose not to seek a writ of certiorari from the

Supreme Court of the United States, but instead filed a

petition for a writ of habeas corpus with the district court

on October 28, 1986. Among other things, Hassine argued

once again that the prosecutor had exploited his post-arrest

silence as an impeachment tactic in violation of due

process and Doyle. The district court referred the petition to

a Magistrate Judge for a Report and Recommendation in

accordance with 28 U.S.C. S 636(b)(1)(B), and on June 15,

1988, the Magistrate Judge recommended that relief be

denied. After reviewing the record in detail, the Magistrate

Judge found strong evidence "that the jury may have

impermissibly been able to draw an inference between

petitioner's silence and his guilt, thus clearly infringing

upon petitioner's due process rights as established in

Doyle." Nevertheless, he concluded that any constitutional

error was harmless, because "[t]he evidence against

Hassine was overwhelming."

Hassine filed several objections to the Report and

Recommendation, and the district court granted de novo

review of the Magistrate Judge's findings consistent with 28

U.S.C. S 636(b)(1)(C). One year later, on November 15,

1989, the district court issued a Memorandum and Order

agreeing with the Magistrate Judge's ruling that the jury

had been permitted "to draw an impermissible inference of

guilt from Hassine's silence" in violation of Doyle and its

progeny. Hassine v. Zimmerman, No. Civ. A. 86-6315, 1989

WL 140491, at *5 (E.D. Pa. Nov. 15, 1989) ("Hassine II").

However, the district court withheld a final decision on

Hassine's petition, because it believed that oral argument

was needed on the question of whether the trial error had

been harmless.

6

Oral argument was held on December 1, 1989, but the

court did not immediately issue a ruling, and no further

events transpired in the case for the next three and a half

years. On April 21, 1993, the Supreme Court announced its

decision in Brecht v. Abrahamson, 507 U.S. 619 (1993),

altering the standard for harmless error in habeas cases

involving constitutional trial flaws. Prior to Brecht, the

Court had held in Chapman v. California that habeas relief

should be denied in the event of constitutional trial error

only if the prosecution could prove that the error was

"harmless beyond a reasonable doubt." 386 U.S. 18, 24

(1967). In Brecht, however, the Supreme Court announced

a new test for harmless error in habeas cases, namely,

whether the error "had substantial and injurious effect or

influence in determining the jury's verdict." 507 U.S. at 623

(quoting Kotteakos v. United States, 328 U.S. 750, 776

(1946)). If not, habeas relief should be denied.

Realizing the potential impact of the new standard on

Hassine's petition, the district court scheduled a July 29,

1993, conference call to discuss several matters, including

"the significance, if any" of the Supreme Court's decision in

Brecht. Following that call, the parties submitted letter

briefs to the district court regarding whether Hassine's

claims should be evaluated under Brecht or Chapman and

whether harmless error could be found under either

standard. However, two more years passed, and on May 4,

1995, the district court conducted yet another oral

argument concerning whether Brecht should be applied to

Hassine's petition. Still, no decision issued, and the parties

filed at least ten additional letter briefs in the ensuing two

years.

Finally, on October 30, 1997 -- more than eleven years

from the date on which Hassine first filed his habeas

petition -- the district court reached a decision on the

merits of the harmless error claim. The court examined the

caselaw and the rationales supporting the rule in Brecht

and concluded that "[t]his court must apply the Brecht

standard" to Hassine's petition for relief. Hassine v.

Zimmerman, No. Civ. A. 86-6315, 1997 WL 677152, at *5

(E.D. Pa. Oct. 30, 1997) ("Hassine III"). After conducting an

extensive review of the record and the evidence supporting

7

the jury's verdict, the court then held that the Doyle error

at Hassine's trial was harmless, because "the record,

considered as a whole, demonstrates that the direct and

circumstantial evidence of [Hassine's] guilt was

overwhelming." Id. at *7. The court determined that "the

prosecutor's improper reference to Hassine's post-arrest

silence, although a constitutional violation under Doyle, did

not have a substantial and injurious effect or influence on

the jury's guilty verdict" within the meaning of Brecht. Id. at

*10.

The district court issued a certificate of probable cause

allowing Hassine to file a timely appeal, and we now have

jurisdiction to hear the appeal pursuant to 28 U.S.C.

SS 1291 and 2253.1 Because the district court relied entirely

on the state court record and did not hold an evidentiary

hearing, our review is plenary. See Johnson v. Rosemeyer,

117 F.3d 104, 109 (3d Cir. 1997). Acting in the same

capacity as the district court, we will presume that the

factual findings of the state courts are correct if fairly

supported by the record, but we will exercise plenary review

over state court conclusions on mixed questions of law and

fact and pure issues of law. See Ahmad v. Redman , 782

F.2d 409, 412 (3d Cir. 1986).

II. THE DOYLE VIOLATION

The Supreme Court established in Doyle v. Ohio that it is

improper for a prosecutor to cause the jury to draw an

impermissible inference of guilt from a defendant's post-

arrest silence. 426 U.S. at 610. Doyle involved two

defendants who, upon taking the witness stand in their

own defense, claimed that a narcotics informant had

framed them for the crime. The prosecutor asked each

defendant on cross-examination whether he had told the

"framing" story to the police after his arrest. Defense

counsel objected strenuously, arguing that the prosecution

could not impeach the defendants on this basis because

_________________________________________________________________

1. This case is not subject to the terms of the Antiterrorism and

Effective

Death Penalty Act of 1996 ("AEDPA"), Pub. L. No. 104-132, 110 Stat.

1214 (1996), because Hassine filed his habeas petition well before the

April 24, 1996, date on which the AEDPA took effect.

8

the defendants were merely exercising their right to remain

silent after arrest. However, the objections were overruled

and the defendants were compelled to admit that they had

not provided an exculpatory story to the authorities

following their arrests and the reading of their Miranda

rights.

On appeal, the defendants argued that the prosecutor's

line of questioning violated the due process guarantees

inherent in the right to remain silent. The Supreme Court

agreed, reversing their convictions in light of Miranda's

implicit assurance that silence will have no adverse

consequences. The Court stated that "it would be

fundamentally unfair and a deprivation of due process to

allow the arrested person's silence to be used to impeach

an explanation subsequently offered at trial." Id. at 618.

Therefore, the Court held, "the use for impeachment

purposes of petitioners' silence, at the time of arrest and

after receiving Miranda warnings, violate[s] the Due Process

Clause of the Fourteenth Amendment." Id. at 619.

In this case, we believe that the state prosecutor violated

the due process principles established in Doyle by seeking

to elicit testimony about Hassine's post-arrest, post-

Miranda silence. The record shows that the prosecutor's

cross-examination of Hassine proceeded as follows:

Prosecutor: How long have you been sitting in jail, sir?

Hassine: Close to seven months.

Prosecutor: And you have been sitting in Bucks Country

Prison?

Hassine: No, sir.

Prosecutor: You were in Bucks County Prison for a time?

Hassine: About a month and a half.

Prosecutor: You were sitting in Holmesburg Prison?

Hassine: For about five months.

Prosecutor: And another prison?

Hassine: Delaware County.

Prosecutor: And conditions are not very good?

Hassine: No, sir.

Prosecutor: You sat for seven months in prison with the

knowledge of what was really involved in

regard to this gun, and you just kept it to

9

yourself because your attorney said to keep it

to yourself?

Hassine's Attorney: Objection.

The Court: Sustained.

Prosecutor: But you kept it to yourself until you came in

to a court of law today and said it for the first

time, in any event, outside of perhaps your

family or your lawyer?

Hassine's Attorney: Objection.

Prosecutor: For the first time?

Hassine's Attorney: That is objected to.

The Court: Sustained.

On three occasions, and over three objections, the

prosecutor asked Hassine why he had remained silent

about the crime after his arrest and after Miranda warnings

had been given. It is clear from these questions that the

prosecutor was attempting to elicit the precise inferences

that the State is prohibited from exploiting under Doyle.

Further, the trial court did not provide, and defense

counsel did not request, any curative instructions for the

jury. In this regard, the instant case is distinguishable from

Greer v. Miller, 483 U.S. 756 (1987), where the Supreme

Court found no Doyle violation. In Greer, the sequence of

events "-- a single question, an immediate objection, and

two curative instructions -- clearly indicate[d] that the

prosecutor's improper question did not violate [the

petitioner's] due process rights." Id. at 766. Here, the

prosecutor asked questions concerning the amount of time

Hassine had spent in prison, questioned Hassine about the

poor conditions he had experienced in confinement, and

then, despite objections by the defense, asked three

questions that are clearly prohibited by Doyle, and then

made statements in his closing that could be understood to

refer to that same silence. Moreover, the trial court gave no

curative instructions at all. We thus find that the

government violated Hassine's right to due process under

Doyle.

In so ruling we reject the State's argument, and the view

of the Pennsylvania Superior Court in Hassine I, 490 A.2d

10

at 451, that the prosecutor's questions were nonetheless

permissible because Hassine had "opened the door" in his

direct testimony to inquiries about his post-arrest silence.

Hassine had told the jury on direct examination that his

attorney had offered to make him "available to the police,"

by, for example, offering to present his father to the police

"in connection with identifying the gun." The State

contends that by asking Hassine about his post-arrest

silence, the prosecutor was merely challenging Hassine's

testimony as to his behavior following arrest by impeaching

Hassine's claim on direct examination that he was

"available" and willing to cooperate with the authorities.

The State's argument is based on an exception contained

in footnote eleven in Doyle, which provides that "post-arrest

silence could be used by the prosecution to contradict a

defendant who testifies to an exculpatory version of events

and claims to have told the police the same version upon

arrest." 426 U.S. at 619 n.11. As the Supreme Court

explained, "[i]n that situation the fact of earlier silence

would not be used to impeach the exculpatory story, but

rather to challenge the defendant's testimony as to his

behavior following arrest." Id.

We agree that the State's argument has some allure

under Doyle as a basis for the prosecutor to use Hassine's

post-arrest silence for impeachment. However, we find, as

the district court did, that Hassine's direct testimony was

far too innocuous and ambiguous to constitute an

exculpatory version that would justify the pointed cross-

examination in this case. The Doyle footnote applies when

a witness testifies on the stand to a version of events and

indicates that he previously told that version to law

enforcement. The government can then pursue its position

that the story was not previously told, and may bring out

on cross-examination the fact that the defendant was silent

following arrest. As the one case cited in the Doyle footnote

makes clear, "to be admissible, keeping silence must be

much more than ambiguous. It must appear to be an act

blatantly inconsistent with the defendant's trial testimony."

United States v. Fairchild, 505 F.2d 1378, 1382 (5th Cir.

1975). In this case, however, there is no contention that

Hassine described a version of events on the stand and

11

indicated that he had previously told that same story to the

police. Rather, the only testimony which harkens back to

Hassine's previous behavior deals with his declaration that

he was "available" for further questioning. Hassine did not

claim to have told the police an exculpatory story after

arrest, did not imply that he had participated actively in the

investigation, and never suggested that he had surrendered

his right to silence by speaking directly with the

authorities. The only relevance of the prosecutor's

questioning to Hassine's previous silence, therefore,

implicates his right to remain silent more than it

constitutes questioning probative of his truth or credibility.2

The Doyle footnote exception only permits the

prosecution to use post-arrest silence to impeach the

credibility of the defendant's version of what he did

following arrest; the government cannot use the silence to

impeach the exculpatory story itself or to draw inferences

suggesting the defendant's guilt. See Doyle, 426 U.S. at 619

n.11; United States v. Gant, 17 F.3d 935, 941 (7th Cir.

1994) ("[T]he government may use defendant's silence for

the limited purpose of impeaching his testimony; it may not

argue that the defendant's silence is inconsistent with his

claim of innocence."); Alo v. Olim, 639 F.2d 466, 468 (9th

Cir. 1980) (questions implying that the defendant's silence

is substantive evidence of guilt are not permitted under the

Doyle exception). In the present case, rather than simply

asking Hassine if he had told his story to the police after

arrest, the prosecutor asked incredulously, "[y]ou sat for

seven months in prison with the knowledge of what was

really involved in regard to the his gun, and you just kept

it to yourself because your attorney said to keep it to

yourself?" We believe that questions like this clearly invite

_________________________________________________________________

2. Hassine's appeal thus differs greatly from Fairchild, where the court

determined that post-arrest silence was admissible because the defense

attorney had specifically asked the defendant if he had "cooperated fully

with the FBI and U.S. Attorney's office in responding with anything that

you all wanted." 505 F.2d at 1383; see also Leecan v. Lopes, 893 F.2d

1434, 1442 (2d Cir. 1990) (emphasis added) (finding that an inquiry into

post-arrest silence was warranted where the defendant's testimony "left

the clear implication that he had proffered his alibi to the police upon

surrender").

12

the jury -- in violation of Doyle -- to reject Hassine's story

and to infer that Hassine's post-arrest silence was a sign of

his guilt.

Consequently, we conclude that the footnote eleven

exception in Doyle is of no aid to the State. The prosecutor

violated Hassine's right to due process at trial by seeking to

draw impermissible inferences about Hassine's post-

Miranda silence following arrest.

III. HARMLESS ERROR

Having found a violation of Hassine's federal

constitutional rights, we turn next to the issue of whether

the error at Hassine's trial can be characterized as

harmless. The Supreme Court has identified two types of

constitutional errors: structural and trial. See Arizona v.

Fulminante, 499 U.S. 279, 306-10 (1991). "A structural

error is a defect in the trial mechanism itself, affecting the

entire trial process, and is per se prejudicial." Yohn v. Love,

76 F.3d 508, 522 (3d Cir. 1996) (citations omitted). By

contrast, "trial error occurs during the presentation of the

case to the jury, and may be quantitatively assessed in the

context of all other evidence." Id. The Supreme Court has

written that "Doyle error fits squarely into the category of

constitutional violations which we have characterized as

trial error." Brecht, 507 U.S. at 629 (quotation omitted).

Therefore, we must assess whether the Doyle violation at

Hassine's trial constitutes harmless error in the context of

all the evidence presented by the State.

The nature of the harmless error inquiry, however, has

been modified during the pendency of Hassine's habeas

corpus proceedings, as discussed above. For many years,

the standard in both direct and collateral appeals for

determining whether a constitutional trial error was

harmless was the test established in Chapman v. California,

which held that a conviction would be vacated unless the

government could prove that the error was "harmless

beyond a reasonable doubt." 386 U.S. at 24. In 1993,

however, the Supreme Court held in Brecht v. Abrahamson,

507 U.S. at 619, that harmless error in habeas cases

should be judged instead by the standard for

13

nonconstitutional errors set forth in Kotteakos v. United

States. Thus, under Brecht, habeas relief can now be

granted on collateral review only if the constitutional trial

error "had substantial and injurious effect or influence in

determining the jury's verdict." 507 U.S. at 623 (quoting

Kotteakos, 328 U.S. at 776).3

Ordinarily, this would present no problem on appeal,

because we would apply the Brecht standard to a petitioner

seeking habeas relief. In this case, though, Hassine argues

that he is entitled to the benefit of the older, more stringent

Chapman standard because no state court has ever

evaluated his claim under Chapman on direct review.

Before proceeding to an application of the harmless error

test, we must therefore resolve the following threshold

issue: should a federal court apply the Brecht standard for

harmless error in a habeas case where the state courts

have never conducted a review of the error on direct appeal

under Chapman? The district court found that Brecht

should apply regardless of the actions of the state courts,

and we agree.

A. Whether Brecht or Chapman Should Apply

The federal courts remain divided as to whether Brecht or

Chapman applies in a situation in which a habeas

petitioner has never had his claim evaluated under

Chapman on direct appeal. The Eighth Circuit, and at least

two district courts, have concluded that the "rule

announced in Brecht does not apply and that the Chapman

harmless error standard is the appropriate test in this

case." Orndorff v. Lockhart, 998 F.2d 1426, 1430 (8th Cir.

_________________________________________________________________

3. Brecht adopts a more deferential standard on collateral review for

cases involving constitutional trial errors. Nevertheless, the degree of

the

difference in practice remains to be seen as the new test is applied. As

Justice Stevens, whose concurrence provided the decisive fifth vote in

Brecht, noted, "[g]iven the critical importance of the faculty of judgment

in administering either standard, however, th[e] difference [in emphasis]

is less significant than it might seem . . . . In the end, the way we

phrase

the governing standard is far less important than the quality of the

judgment with which it is applied." 507 U.S. at 643 (Stevens, J.,

concurring).

14

1993); see also Starr v. Lockhart, 23 F.3d 1280, 1292 (8th

Cir. 1994); Lyons v. Johnson, 912 F. Supp. 679, 687-89

(S.D.N.Y. 1996), aff'd, 99 F.3d 499 (2d Cir. 1996); Rickman

v. Dutton, 864 F. Supp. 686, 712 (M.D. Tenn. 1994), aff'd

sub nom. Rickman v. Bell, 131 F.3d 1150 (6th Cir. 1997).

By contrast, the Fourth, Fifth, Seventh, Tenth, and

Eleventh circuits have all held that "Brecht, rather than

Chapman, enunciates the appropriate standard for

determining whether a constitutional error was harmless in

a federal habeas challenge to a state conviction or

sentence," regardless of the standard employed by the state

courts. Hogue v. Johnson, 131 F.3d 466, 499 (5th Cir.

1997), cert. denied, ___ U.S. ___, 118 S. Ct. 1297, (1998);4

see also Sherman v. Smith, 89 F.3d 1134, 1140-41 (4th Cir.

1996), cert. denied, ___ U.S. #6D6D 6D#, 117 S. Ct. 765 (1997);

Brewer v. Reynolds, 51 F.3d 1519, 1529 (10th Cir. 1995);

Tyson v. Trigg, 50 F.3d 436, 446-47 (7th Cir. 1995), cert.

denied, 516 U.S. 1041 (1996); Horsley v. Alabama, 45 F.3d

1486, 1492 n.11 (11th Cir. 1995).5 Hassine contends that

we should not implement Brecht as we normally would, but

that we should cling to the old Chapman standard,

notwithstanding the new formulation handed down by the

Supreme Court. However, we are unwilling to find an

exception to Brecht's "standard formula," Tyson, 50 F.3d at

446, because we disagree with the four arguments that

Hassine offers in contending that Chapman should apply.

First, the plain language of Brecht makes no distinction

_________________________________________________________________

4. A panel of the Fifth Circuit recently noted its disagreement with

Hogue, but concluded that "we may not ignore the [Hogue] decision, for

in this circuit one panel may not overrule the decision of a prior panel."

Barber v. Johnson, 145 F.3d 234, 237 (5th Cir. 1998), petition for cert.

filed, (U.S. Sep. 10, 1998) (No. 98-6001). Hogue thus remains good law

in the Fifth Circuit.

5. Two courts passed on the issue after finding that the trial error would

not have been harmless under either standard. See Lyons v. Johnson, 99

F.3d 499, 503 (2d Cir. 1996) ("[D]ue to the magnitude of the trial court's

error . . . we need not decide whether . . . Chapman or Brecht is the

appropriate standard to apply in this case."); Hanna v. Riveland, 87 F.3d

1034, 1038 n.2 (9th Cir. 1996) ("[W]e do not need to decide this issue

because, even under Brecht's less stringent standard of review, the error

cannot be deemed harmless.").

15

that requires us to employ Chapman on collateral review.

The issue in Brecht was "whether the Chapman harmless-

error standard applies in determining whether the

prosecution's use for impeachment purposes of petitioner's

post-Miranda silence, in violation of due process under

Doyle v. Ohio, entitles petitioner to habeas corpus relief."

507 U.S. at 622-23 (footnote and citation omitted). The

Court held that "the Kotteakos harmless-error standard

applies in determining whether habeas relief must be

granted because of constitutional error of the trial type." Id.

at 638; see also id. at 623 ("[T]he standard for determining

whether habeas relief must be granted is whether the Doyle

error `had substantial and injurious effect or influence in

determining the jury's verdict.' ") (quoting Kotteakos, 328

U.S. at 776). The Court never restricted the issues or the

holding in Brecht to situations where a petitioner has

already had his or her claim evaluated by the state courts

under Chapman. Therefore, we are not persuaded that we

should read an exception into Brecht's harmless error rule

when no such exception is even implicit in the Court's

opinion.

Second, in contrast to the views of some courts, we do

not believe that we should regard the facts in Brecht as

limiting the Court's holding to cases in which Chapman has

previously been applied. Because the Court decided Brecht

after the Wisconsin Supreme Court had already ruled on

the harmless error issue under Chapman, several courts

have viewed prior Chapman review as integral to Brecht's

result.6 We disagree.

_________________________________________________________________

6. For example, the Eighth Circuit has written that Brecht was "based

largely on the notion that because the state courts can properly apply

the Chapman harmless error standard on direct review, the federal

habeas courts need only review those decisions under the Kotteakos

harmless error standard." Orndorff, 998 F.2d at 1430. Likewise, the

district court in Lyons stated that Brecht "was premised largely on

respect for the actual, conscientious review of harmlessness performed

by four prior courts -- both federal and state-- that applied Chapman."

912 F. Supp. at 689. These courts thus believe that Brecht intended to

alter the harmless error standard only for those habeas cases which, like

Brecht, involved a previous application of Chapman on direct appeal.

16

While the Court noted that "it scarcely seems logical to

require federal habeas courts to engage in the identical

approach to harmless-error review that Chapman requires

state courts to engage in on direct review," 507 U.S. at 636,

its holding was based primarily on a finding -- apart from

the particular facts or history of the case -- that "the costs

of applying the Chapman standard on federal habeas

outweigh the additional deterrent effect, if any, that would

be derived from its application on collateral review." Id.

In Brecht, the Court identified four considerations in

justifying the application of a more relaxed harmless error

standard for habeas petitions, including the States'

interests in the finality of convictions, comity, federalism,

and the recognition that "liberal allowance of the writ

degrades the prominence of the trial itself." Id. at 635

(quoting Engle v. Isaac, 456 U.S. 107, 127 (1982)). The

Court then reasoned comprehensively that:

[o]verturning final and presumptively correct

convictions on collateral review because the State

cannot prove that an error is harmless under Chapman

undermines the States' interest in finality and infringes

upon their sovereignty over criminal matters. Moreover,

granting habeas relief merely because there is a

reasonable possibility that trial error contributed to the

verdict, is at odds with the historic meaning of habeas

corpus -- to afford relief to those whom society has

grievously wronged. Retrying defendants whose

convictions are set aside also imposes significant social

costs, including the expenditure of additional time and

resources for all the parties involved, the erosion of

memory and dispersion of witnesses that accompany

the passage of time and make obtaining convictions on

retrial more difficult, and the frustration of society's

interest in the prompt administration of justice.

507 U.S. at 637 (citations and quotations omitted). This

analysis led the Court to conclude that "[t]he imbalance of

the costs and benefits of applying the Chapman harmless-

error standard on collateral review counsels in favor of

applying a less onerous standard on habeas review of

constitutional error." Id. at 637; see also id. at 637-38 ("The

Kotteakos standard is . . . better tailored to the nature and

17

purpose of collateral review and more likely to promote the

considerations underlying our recent habeas cases."). Thus,

Brecht was clearly not premised on the notion that the state

courts had already conducted a Chapman review of

harmless error on direct appeal. The fact that the state

courts may or may not have performed such a review was

of minor importance to the Court's view that the Kotteakos

standard should be applied consistently in collateral

proceedings.

Third, contrary to Hassine's argument, we believe that

the policy interests identified in Brecht are nonetheless

implicated in cases where the state courts have not

evaluated a petitioner's claim under Chapman. Hassine

urges that we follow Lyons, in which a district court stated

that Brecht should be used only when federal courts would

be repeating Chapman analyses performed by the state

courts, because it is only in these situations that the

federal courts demonstrate a "lack of respect for state

courts' ability as federal constitutional interpreters, and

hence . . . violate obligations of comity and principles of

federalism." 912 F. Supp. at 689; see also Barber, 145 F.3d

at 238 (Dennis, J., concurring) (stating that the federal

courts cannot defer "to the systemic values offinality,

federalism, and comity . . . unless there has in fact been a

good-faith State effort to protect constitutional rights by

applying the Chapman standard").

Once again, however, we disagree. Finality, comity, and

federalism do not turn on whether the state courts have

performed a Chapman analysis, but depend, rather, on the

fact that they have rejected a defendant's direct appeal.

Indeed, a federal habeas court which overturns a state

conviction on collateral review still upsets a State's interest

in finality, and still raises federalism, comity, and trial

process concerns, regardless of whether the state courts

have previously identified the constitutional error or have

performed a Chapman analysis on direct appeal.

In light of the Court's reasoning and holding in Brecht, it

is anomalous to say that a Chapman standard should apply

in any habeas case based on concerns such as federalism,

comity, or finality. Brecht stands for the proposition that

these concerns are to be advanced by employment of the

18

Brecht standard itself. Thus, not to apply Brecht, and to

apply instead a less deferential standard, is to nullify not

merely the policies that the Court has advanced, but the

very import of the Brecht decision. In fact, the Supreme

Court has repeatedly emphasized the importance of

deference to State "sovereignty over criminal matters,"

Brecht, 507 U.S. at 637, and to the finality of criminal

judgments, outside the context of Chapman. In Teague v.

Lane, for example, the Court restricted the ability of habeas

petitioners to establish new constitutional rules on

collateral review, in part because the "[a]pplication of

constitutional rules not in existence at the time a

conviction became final seriously undermines the principle

of finality which is essential to the operation of our criminal

justice system." 489 U.S. 288, 309 (1989). Likewise, in

Wright v. West, the Court stressed the "significant costs" of

habeas review. 505 U.S. 277, 293 (1992) (quoting Engle,

456 U.S. at 126). "Among other things," the Court wrote, "it

disturbs the State's significant interest in repose for

concluded litigation, denies society the right to punish

some admitted offenders, and intrudes on state sovereignty

to a degree matched by few exercises of federal judicial

authority." Wright, 505 U.S. at 293 (quotations omitted).

Therefore, while the absence of a state court Chapman

review may make the considerations identified in Brecht

less compelling, it certainly does not eliminate them. As our

sister circuits have concluded, Brecht is still the general

rule to apply in habeas cases, even in situations where the

state courts never employ Chapman, because "principles of

federalism, comity, and finality apply regardless of the

harmless error standard used by the state court." Sherman,

89 F.3d at 1141; see also Tyson, 50 F.3d at 446 ("The

reasons the Court gave in Brecht for adopting a less

stringent rule are independent of the rule applied in the

state appellate process.").7

_________________________________________________________________

7. We do not mean to imply that habeas relief should be limited in the

name of federalism and comity alone. We note merely that Supreme

Court jurisprudence requires us to consider the systemic costs of

applying certain standards on collateral review, and that, in this case,

those costs are substantial even if the state courts never perform a

Chapman analysis on direct appeal.

19

Finally, we note that we are wary of the practical impact

of applying Chapman to every case where a prior Chapman

review has not occurred. As the Seventh Circuit recognized

in Tyson, "ordinarily . . . the state court will not have found

any error and therefore will have had no occasion to apply

any standard of harmless error. Brecht itself was a fluke in

this regard; the state courts had considered the issue of

harmless error and applied the Chapman standard." 50

F.3d at 446.8 Consequently, any rule requiring Chapman to

be used when the state courts fail to address harmless

error would render Brecht inapplicable to the majority of

habeas petitions and would "rob the decision of any general

significance." Id. Given the substantial interests identified

in Brecht, and given the reasoning of the opinion, we find

it highly unlikely that the Supreme Court could have

sanctioned such a result.

Cases like Brecht and Teague express the Court's belief

that collateral review entails heightened deference and

respect to the state courts, and we must follow governing

precedent when it applies, as it does here. Accordingly, we

hold that a federal habeas court performing a harmless

error inquiry on collateral review must employ the standard

for harmless error articulated in Brecht, even if the state

courts have never reviewed the error on direct appeal under

Chapman.

B. Whether Brecht Should be Applied to Hassine

Even if Brecht applies generally to collateral proceedings,

Hassine contends that we should still review his claims

under Chapman due to the unique and unusual facts of his

case. In particular, he argues that Chapman review is

appropriate here because (1) he forfeited the right to pursue

an appeal in the Supreme Court when Chapman was still

the law for harmless error on collateral review, and because

_________________________________________________________________

8. The state courts may not use Chapman to evaluate harmless error on

direct appeal if, as here, they conclude that there was no constitutional

error and thus do not perform the harmless error analysis at all, if they

find that the petitioner has defaulted procedurally on his claim, or if

there is no opportunity to address the error because it is discovered only

after the defendant has already exhausted his state appeals.

20

(2) the district court proceedings were marked by such

extensive delay as to violate his right to due process.9 We

are not persuaded by either contention.

First, the fact that Hassine made the deliberate choice to

forego seeking relief in the Supreme Court is irrelevant.

Hassine could not, of course, have anticipated an

intervening change in the law, but that alone is insufficient

to prohibit a retroactive application of Brecht on collateral

review. Hassine made a tactical decision to bypass the

Supreme Court and to proceed to the district court for

habeas relief, and we will not reward that decision by

employing a less stringent standard when reviewing his

petition.

Second, while we agree that Hassine experienced

excessive delay in the processing of his petition, we do not

believe that he is entitled to habeas relief, including any

altered standard for harmless error, even if the delay in the

district court gave rise to a violation of his due process

rights.10 In Heiser v. Ryan, 951 F.2d 559, 563 (3d Cir.

1991) ("Heiser I"), we acknowledged that a delay in certain

collateral proceedings can violate a prisoner's fundamental

right to due process. We reviewed the claim of a petitioner

whose state action for post-conviction relief had been

pending for nearly four years, and we stated that"delays in

post-verdict process may violate the Due Process Clause.

. . . The due process clause protects against more than

delay in trial and sentencing. It guarantees as well the right

to attack a conviction." Id.

After Heiser I was remanded to the district court,

however, we clarified our earlier remarks by revisiting the

_________________________________________________________________

9. Hassine also argues that his appeal is different because no state court

has ever conducted a Chapman review of his claims. As we discussed

above, however, we believe that this does not alter our analysis in

determining whether Brecht applies.

10. While we certainly do not condone the extraordinary delay in

processing Hassine's petition, we wish to make clear that we are not

holding that the delay constituted a violation of due process. We hold

merely that Hassine would not be entitled to relief through habeas

corpus proceedings even if the delay did amount to a constitutional

violation.

21

due process issue in a subsequent appeal. See Heiser v.

Ryan, 15 F.3d 299 (3d Cir. 1994) ("Heiser II"). In Heiser II,

we still recognized that excessive delay in state collateral

proceedings could violate due process, but we made clear

that while the defendant could seek damages against

counsel responsible for the delay, a writ of habeas corpus

was not the proper remedy in such a case. Id. at 307. In

other cases, we have indicated that petitioners experiencing

delay can seek a writ of mandamus to compel the district

court to reach a decision on the habeas claim. For example,

in Madden v. Myers, we addressed a petitioner's argument

that mandamus was warranted where seven months had

elapsed since the filing of the Magistrate Judge's Report

and Recommendation without any action by the district

court. 102 F.3d 74, 76 (3d Cir. 1996). We ultimately denied

the request, but, in our only published opinion on this

issue, we noted that the petitioner's claims of delay had

"force" and were "of concern." Id. at 79. Moreover, we stated

that "[a]lthough this delay is of concern, it does not yet rise

to the level of a denial of due process," implying that at

some point, delay by the district court could become so

excessive as to warrant the issuance of a writ of

mandamus. Id. (emphasis added).11 We are not alone in

entertaining writs of mandamus due to extraordinary

federal delay. See, e.g., Johnson v. Rogers, 917 F.2d 1283,

1285 (10th Cir. 1990) (granting a writ of mandamus after a

fourteen-month delay by the district court); McClellan v.

Young, 421 F.2d 690, 691 (6th Cir. 1970) (granting a writ

of mandamus where the district court had delayed its

habeas ruling pending a decision by the Supreme Court).12

However, while we have shown a willingness to issue

_________________________________________________________________

11. In fact, a review of our unpublished opinions reveals that we

routinely entertain mandamus petitions complaining of extraordinary

delay in habeas corpus proceedings at the district court level.

12. We hasten to add that "[t]he remedy of mandamus is a drastic one,

to be invoked only in extraordinary circumstances." Kerr v. United States

Dist. Ct., 426 U.S. 394, 402 (1976). Thus, a habeas petitioner seeking

mandamus in these cases must experience extraordinary delay, "must

have no other adequate means to obtain the desired relief, and must

show that the right to issuance is clear and indisputable." Madden, 102

F.3d at 79.

22

writs of mandamus and to entertain lawsuits for damages,

we have never granted habeas relief because of a delay in

the processing of a petition, and we decline to do so here.

The federal courts are authorized to provide collateral relief

where a petitioner is in state custody or under a federal

sentence imposed in violation of the Constitution or the

laws or treaties of the United States. 28 U.S.C.SS 2254,

2255. Thus, the federal role in reviewing an application for

habeas corpus is limited to evaluating what occurred in the

state or federal proceedings that actually led to the

petitioner's conviction; what occurred in the petitioner's

collateral proceeding does not enter into the habeas

calculation. We have often noted the general proposition

that habeas proceedings are "hybrid actions"; they are

"independent civil dispositions of completed criminal

proceedings." See, e.g., Santana v. United States, 98 F.3d

752, 754 (3d Cir. 1996). Federal habeas power is"limited

. . . to a determination of whether there has been an

improper detention by virtue of the state court judgment."

Henderson v. Frank, No. 97-3041, 1998 WL 456254, at *9

(3d Cir. Aug. 6, 1998). As the Seventh Circuit has noted,

"[d]elay in processing [a] collateral claim does not make the

continued imprisonment of the defendant unlawful, and

hence, does not warrant federal habeas corpus relief."

Montgomery v. Meloy, 90 F.3d 1200, 1206 (7th Cir. 1996),

cert. denied, ___ U.S. ___, 117 S. Ct. 266 (1996); see also

Jackson v. Duckworth, 112 F.3d 878, 880 (7th Cir.) (stating

that "federal habeas corpus cannot remedy a delay in state

collateral proceedings because such an error has absolutely

nothing to do with the reason for a defendant's

confinement"), cert. denied, ___ U.S. ___, 118 S. Ct. 380

(1997).13

Therefore, to the extent that a petitioner wishes to claim

that delay in the processing of a collateral petition violates

due process, we hold that the petitioner's remedy, if any, is

through such avenues as a lawsuit for damages or a writ of

mandamus rather than through the habeas corpus

_________________________________________________________________

13. Providing relief to a petitioner in state custody because of a delay

in

the federal courts would also greatly undermine principles of federalism

and comity by disturbing a state court judgment due to a federal court

error.

23

proceeding itself. In the case at bar, we thus conclude that

Brecht still applies in evaluating whether the Doyle violation

at Hassine's trial constituted harmless error.

C. Application of the Brecht Standard for Harm less Error

Having found that Brecht applies to Hassine's petition, we

proceed next to an examination of whether the district

court properly concluded that the prosecution's Doyle

violation was harmless.14 Under Brecht and its progeny, a

constitutional trial error is not harmless if the court is in

"grave doubt" as to whether the error had a substantial and

injurious effect or influence in determining the jury's

verdict. O'Neal v. McAninch, 513 U.S. 432, 436 (1995).

"Grave doubt" exists when, "in the judge's mind, the matter

is so evenly balanced that he feels himself in virtual

equipoise as to the harmlessness of the error." Id. at 435.

Moreover, it is "inappropriate to ask whether there was

sufficient evidence to support the result, apart from the

phase of the trial affected by the error. The correct inquiry

is whether the error had a substantial influence on the

verdict despite sufficient evidence to support the result

apart from the error." Yohn, 76 F.3d at 523 (citations

omitted).

Hassine contends that the Doyle violation could not have

been harmless because it went to the heart of his defense.

He maintains that because he was the only witness

testifying as to his version of the events, and because the

Doyle error undermined his credibility, it necessarily had a

"substantial and injurious effect or influence in determining

the jury's verdict." Brecht, 507 U.S. at 623 (quoting

Kotteakos, 328 U.S. at 776). In evaluating Hassine's claim,

we must thus determine whether, and to what extent, the

jury's decision to accept the State's version of the facts

rather than Hassine's was influenced by the prosecutor's

_________________________________________________________________

14. The district court also evaluated Hassine's claim under Chapman,

stating in a footnote that "the record establishes sufficient evidence of

Hassine's guilt to suggest that the Doyle error would have been harmless

under the Chapman reasonable-doubt standard as well." Hassine III,

1997 WL 677152, at *10 n.8. However, we need not reach this issue

because we have determined that Chapman does not apply.

24

Doyle violation. If the jury disbelieved Hassine and

convicted him because of the Doyle violation, the error was

not harmless and we will grant the petition. The crucial

inquiry is the "impact of the error on the minds of the

jurors in the total setting." Yohn, 76 F.3d at 523 (citing

Kotteakos, 328 U.S. at 764). While the nature of the

evidence against Hassine is important, we must also

examine the phases of the trial affected by the error, and

determine whether the error had a substantial influence on

the verdict despite sufficient evidence to support the result

apart from the error. See id. (citing Kotteakos, 328 U.S. at

765). In so doing, we must of necessity weigh the impact of

evidence on the jury and cannot help but make a judgment

as to how the jury would reasonably perceive Hasssine's

version of events with and without the Doyle violation.

At the outset, we note our agreement with the district

court's finding that the evidence, as mounted through a

parade of Commonwealth witnesses, was substantial

indeed. The witnesses testifying as to Hassine's planning of,

and intent to commit, the crime were numerous, and their

stories consistent.15 A number of witnesses testified to an

initial meeting in which Hassine called for Kellet's

execution. (N.T. 34, 111-12, 743-44, 756-57, 810, 822).16

Several witnesses also recalled that Hassine had actively

searched for a weapon to carry out his plan: at the initial

meeting, he demanded that the other employees try to

procure a weapon (N.T. 36, 744, 774); he had discussed

buying a .357 with Breed members Tuite and Schwab (N.T.

56-59, 555-57, 626-27); he asked his employee Ron

Wharton to buy a gun and to participate in the plan to kill

_________________________________________________________________

15. The Commonwealth put 34 witnesses on the stand. Among those

witnesses central to the Commonwealth's case were: 1) Eric Decker, a

sometime employee of Hassine's who shot the Kellets, Sofield, and

Puerale; 2) Greg Orlowski, the manager and day-to-day owner of the

Market; 3) Ron Wharton and Billy Hayes, employees at the Market; 4)

Valerie Lynch, Decker's girlfriend; 5) Ron Tuite and Joseph "Critter"

Schwab, members of the Breed motorcycle gang; and 6) Theodore

Camera, a tenant of the Hassine family and a former classmate of

Hassine's.

16. Ron Wharton recalled Hassine stating that, "He wanted him wasted.

He had to set an example." (N.T. 744).

25

Kellet (N.T. 747, 760); and he visited his tenant Theodore

Camera and asked Camera to find him a gun. (N.T. 485-86,

487-88, 490).

Hassine settled on a weapon for the shootings: a stolen

.25 caliber automatic handgun. (N.T. 933). Both Decker

and Wharton testified that Hassine gave Decker the.25

while the three were in the Market van. (N.T. 53, 749).17

Hassine and Decker also went to a gun store where Hassine

bought a box of ammunition and signed for the purchase.

(N.T. 59-60, 708-10).

After buying the bullets, Decker and Hassine drove by

Kellet's house, where Hassine told Decker to shoot him

then and there with the .25. (N.T. 61). Decker declined

because it was daytime, and a witness was present. (N.T.

61-62). Hassine and Decker repeatedly test fired the gun

into the back wall of the store, but the .25 was

malfunctioning and had to be put aside, despite their

attempts to fix it. (N.T. 58-60, 62-63, 750-52, 811-12).18

Hassine and his co-defendant Orlowski also met with

Breed members Tuite and Schwab to discuss killing Kellet.

After bailing Tuite out of jail on a gun charge involving the

.357, Hassine contacted him about a "deal" involving Kellet,

and the four met to discuss a contract killing. (N.T. 557,

560, 560-61, 562, 627, 630, 753-55).19 Tuite and Schwab's

asking price of $1500 was too high, however, so the four

agreed instead that they would beat Kellet up for $500

($250 credit for the bail and $250 for the beating). (N.T.

560-62, 563, 628-629, 632).20 Rather than attack Kellet, as

_________________________________________________________________

17. Decker testified that Hassine handed him the gun and stated "here,

hit him in the head and leave it there." (N.T. 53).

18. The .25, a spent shell, and photos of the holes in the Market wall

were entered into evidence. (N.T. 51-52, 792-97, 800-01).

19. Both Tuite and Schwab testified that Hassine appeared upset and

angered by the challenge Kellet posed to his authority. Tuite: "Hassine

was pissed off--pissed off that somebody would step on his toes." (N.T.

563). Schwab: Hassine was "ruffled because somebody messed with one

of his boys." (N.T. 628).

20. Schwab mentioned that Orlowski and Hassine wanted to continue to

consider the possibility of killing Kellet for money: "They were haggling

back and forth about the fifteen hundred bucks, so I took the two fifty.

They wanted a night to think about if it was worth fifteen hundred bucks

to waste Kellet." (N.T. 629-30).

26

agreed, however, Tuite and Schwab decided to warn Kellet

that Hassine and Orlowski wanted to kill him and kept the

$250. (N.T. 566, 630).21

The Commonwealth also presented evidence of Hassine's

conduct both at and after the incident that pointed to his

guilt. On the night of the shootings, Decker testified that he

met Hassine at the market and said, "Tonight's the night--

this cat has got to go. We'll use your gun. I want two

hundred and fifty dollars." (N.T. at 77, 218-19). Hassine

agreed and said he would come back to pick Decker up

with his car. Id. Decker further testified that Hassine then

drove him to Trenton and retrieved the .380 Llama

handgun used in the shootings from his parents' house;

gave him directions to Kellet's house; told him that he

would wait for him and drive him away after the shootings;

and gave him a Yankees batting helmet to disguise himself

during the shootings. (N.T. 83-85, 87-88).

Hassine then went to Orlowski's house, where Hassine

and Orlowski met a third person, Michael Thompson

("Thompson"), who testified that all three got into his car.

(N.T. 861-63). Hassine was carrying a metal pipe. (N.T.

870). Hassine told Thompson to drive to the parking lot

across the street from Kellet's residence and then to other

locations; Hassine whistled out the window from time to

time, and at one point Hassine told Thompson to get out of

the car and call for "Eric". (N.T. 863-870, 874-884, 905-07).22

In the course of their drive, they heard two shots, and

Hassine stated, "Oh shit, that's my father's gun, I hope that

asshole doesn't get caught." (N.T. 868-69, 881, 1224).

After the murder, when Hassine and Orlowski returned

from their attempts to find Decker, Hassine told Orlowski

that they still had to find Decker, because he was afraid

Decker would be arrested. (N.T. 1227-28). The following

morning, Hassine told his employee Wharton to drive the

pair to Decker's apartment, so they could try once again to

retrieve the .380 from Decker. (N.T. 761). When they

_________________________________________________________________

21. Hassine later complained that Tuite and Schwab had "ripped him

off." (N.T. 755).

22. In the course of their drive, the three went to the Dunkin' Donuts,

where Hassine mentioned that they had been seen. (N.T. 867).

27

realized Decker had been arrested, Hassine told Decker's

girlfriend, Valerie Lynch, to get rid of the gun. (N.T. 761-63,

1043-44, 1046-47). As word spread that the case against

Hassine was building, Tuite and Schwab decided to try to

intimidate Hassine into giving them $10,000 to kill

Commonwealth witnesses, and Hassine considered their

demands but did not ask them to follow through on the plan.23

(N.T. 568, 572, 577-78, 631-32).

Not only did the Commonwealth's case present evidence

of Hassine's guilt but also evidence of both a coverup and

instances of lying that detracted from Hassine's credibility.

On the night of the shootings, Hassine told his co-

defendant Orlowski to keep his "mouth shut" if the police

questioned him. (N.T. 1228). Hassine's attempts to

influence others continued as the investigation intensified.

After Decker was arrested for the shootings, Hassine tried

to make Orlowski sign a statement holding him responsible

for the crimes, and the two held practice sessions in which

Hassine and Orlowski practiced Orlowski's story. (N.T.

1388-90). Valerie Lynch testified that Hassine told her to lie

about his and Decker's whereabouts the night of the

shootings, saying: "Remember, Valerie, I was in New York,

and Eric Decker was at home." (N.T. 764-65, 1051-52). In

the course of an intercepted phone conversation, Hassine

queried Thompson about what Thompson had told the

police during questioning, and in the course of that

conversation the two agreed, ostensibly, that they would

keep each other out of the case.24 (N.T. 892-94, 901).

_________________________________________________________________

23. Schwab testified as follows:

He was all frazzled; he was confused -- "I can't think--I don't

know

what to do" -- imbalanced -- all that shit, and he walked around.

He said he would talk to his old man to see if he could come up

with some bucks.

Q: Did you talk to him any more after that, Mr. Schwab?

A: Yeah -- one more time . . . . I told him the pressure was coming

down on him. And he said he would try again to get the money. I

told him I would take a van load of steaks--pack the van with

steaks and give me five grand. (N.T. 631-32).

24. V: So, as far as they're concerned, you know, you were with

somebody else that night, right?

28

Hassine had made promises to Decker as a part of his

agreement to kill Kellet -- an apartment, money, and a

future on "easy street" -- and he continued the practice of

offering money or services for cooperation from others after

the shootings had taken place. (N.T. 63, 111, 752-53, 773).

In the course of the conversation in which he urged her to

lie, Hassine also promised Lynch that he would help her

and her child out with money, cuts of meat, or anything

else she needed. (N.T. 1052). Hassine also promised

Thompson, in the course of their intercepted phone

conversation, that Thompson should look into buying

Kellet's house, and that the two could enter into a deal

whereby Hassine would put up the money for the house

and the surrounding property, and Thompson could keep

the house while Hassine developed the lot. (N.T. 896-97,

897-900). In his own testimony, Hassine admitted that he

had no intention of buying the house or developing the lot,

but he was making the promise because he wanted to know

what Thompson told the police. (N.T. 1563-64).25

Notwithstanding the weight of the evidence presented by

the prosecution, we would have little difficultyfinding that

the Doyle violation had substantially influenced the jury if,

apart from the violation, the phase of the trial most directly

impacted, namely, Hassine's testimony, presented a strong

counter to the state's evidence. But that is not the case.

Hassine's story was undermined not only by evidence of his

_________________________________________________________________

M: Yeah, as far as they're concerned I wasn't even with you.

V: You weren't.

M: Yeah. (N.T. 901).

25. The weaknesses of the Commonwealth's witnesses were exposed in

the course of both direct and cross-examination. Decker was cross

examined on his frequent use of methamphetamine, prior record,

shifting stories, and his cooperation with the Commonwealth. (N.T. 116-

17, 130-32, 132-35, 136, 138, 215). Other witnesses were questioned

about their prior denials to the police, changes in accounts, drug usage,

conflicts with Hassine, bias in favor of Orlowski, or reluctance to

testify.

(N.T. 569-70, 632-33, 769-71, 784-86, 787-88, 789, 789-90, 872, 903,

963-64). As such, the jury had ample opportunity to consider the

credibility and relative sophistication of the Commonwealth's witnesses.

29

disregard for the truth, but by his own testimony in which

he set forth strained and unrealistic explanations for his

conduct and portrayed himself as a mild mannered store

manager and law student -- a portrayal strikingly at odds

with other evidence depicting his behavior as controlling,

crude, and vulgar. In the course of his testimony, Hassine

portrayed himself as a diligent law student who spent the

bulk of his summer studying for the New Jersey and New

York bar exams, which he did not pass. (N.T. 1436-38,

1450-52). Hassine also presented himself as a sophisticated

individual versed in business practices who was concerned

about the management of the market. Hassine testified that

he made Orlowski run the store efficiently, and that he had

threatened to close the store if Orlowski did not run it in a

cost-effective manner. (N.T. 1414-1422, 1492-93). He also

testified that he was aware of the drug dealing at the store

after Kellet had beaten and stolen money from Orlowski,

but that he never participated in or profited from the drug

dealing. (N.T. 1424, 1510).26 He testified that when Wharton

took $40 from the register to front a drug deal, he punched

him, but did so because he wanted nothing to do with

Wharton's activities. (N.T. 1426-27, 1492-93).

Hassine denied having any feud with Kellet, having any

animosity toward Kellet, or saying that he wanted Kellet

killed or "wasted," claiming instead that he had only

wanted Kellet to be put in jail. (N.T. 1428-29, 1432-34,

1496, 1497-98, 1500, 1553).27 Hassine also testified that by

saying he wanted to make an "example of Kellet," he meant

that any time Kellet made any motion toward him or any

contact with him the police should be called. (N.T. 1429).

Hassine also testified that he had, in fact, rebuffed Tuite

and Schwab's offer to kill Kellet for him by indicating that

_________________________________________________________________

26. Wharton had testified that Hassine insisted that the drugs be "cut"

to increase profits. (N.T. 780-81). Hassine also referred to Thompson as

"Mr. Reds" in his testimony and in the course of the recorded

conversation played for the jury. (N.T. 1470, 1507-08).

27. Kellet himself testified to the pattern of escalating threats and

harassment between himself, Orlowski, and Hassine prior to the

shootings, as well as prior tensions between himself and Hassine. (N.T.

382-83, 394-99).

30

the situation with Kellet was being handled by the police.

(N.T. 1455-59). Hassine claimed to have no idea why Tuite

and Schwab would think he would want Kellet killed. (N.T.

1455, 1492).

Hassine admitted that he had engaged in a complex

course of behavior involving guns, but testified that his

search for a weapon was undertaken in order to find

adequate protection for the store. He conceded that he had

contacted Camera about obtaining a gun, even though he

knew he could purchase a gun legally; that he had bought

bullets and signed the register at the gun store because

Decker had asked him to do so, even though he knew that

Decker was on probation for a gun charge; and that he had

shot up the wall of the store with Decker because Decker

wanted to demonstrate problems with the gun and he just

got "carried away" in the process. (N.T. 1430-31, 1436,

1445-46, 1449-50, 1549-52, 1553, 1556-58). Hassine also

explained his conduct on the night of the shootings as

being motivated by his desire to get his father's gun back

and to prevent a crime from occurring. (N.T. 1466-68,

1526-29, 1521-32).

Hassine's version of events was further undercut by

evidence that he was not the mild-mannered person he

portrayed himself to be. There was testimony as to his

threatening and controlling Orlowski, his need to be a boss

of others and his desire to protect his turf and his

authority, even if doing so entailed resort to violence. (N.T.

488-89, 490, 555, 563, 628, 755, 783, 861-62, 816, 1051,

1441, 1454-55). Other evidence, some of which was

conceded in Hassine's own testimony, demonstrated that

Hassine had violent tendencies, through incidents such as

his pulling Orlowski's finger out of its socket during an

argument and punching Wharton, and through his

practices of engaging in threats and feuding. (N.T. 382-83,

1498, 1508, 1556).

The Doyle violation occurred as a series of three

questions in the midst of Hassine's testimony, and, later, in

two ambiguous references to Hassine's silence in the

prosecutor's closing argument. (N.T. 1543-1544, E.H. 223,

259-60). Based on our assessment of the trial evidence, we

conclude that the questions posed in the midst of Hassine's

31

testimony did not have a substantial or injurious effect on

the jury's verdict. Nor did the two ambiguous references

made to a general pattern of silence result in an

overemphasis on this silence that destroyed the jury's

ability to look at the evidence in the case, as Hassine

contends. Rather, these statements were couched within a

general framework of asking the jury to consider the

credibility of the witnesses as a whole, and as such, the

closing provided the jury with ample latitude to evaluate

Hassine's credibility and the evidence of his guilt apart from

those particular statements.

While not minimizing the importance of the right at issue

or condoning the prosecutor's conduct, we find that the

actual effect of the Doyle violation on Hassine's credibility

before the jury and the inferences drawn by that jury, and

on the jury's determination of guilt or innocence, was

minimal, as the improper questions and statements

occurred within a contextual presentation of a story that

was, on its own, not likely to be viewed as credible. Given

the overwhelming evidence of Hassine's guilt, and his lack

of credibility apart from the Doyle violation, we conclude

that the constitutional error at Hassine's trial was

harmless; there is little doubt that the prosecutor's Doyle

violation did not have a "substantial and injurious effect or

influence in determining the jury's verdict." Brecht, 507

U.S. at 623 (quoting Kotteakos, 328 U.S. at 776).

D. The Footnote Nine Exception to Brecht

Hassine's final argument is that even if the Doyle error

did not substantially influence the jury, he is still entitled

to relief under an exception to the harmless error rule

noted in Brecht. In footnote nine of the Brecht opinion, the

Supreme Court stated:

Our holding does not foreclose the possibility that in

an unusual case, a deliberate and especially egregious

error of the trial type, or one that is combined with a

pattern of prosecutorial misconduct, might so infect

the integrity of the proceeding as to warrant the grant

of habeas relief, even if it did not substantially

influence the jury's verdict. Cf. Greer v. Miller, 483 U.S.

32

756, 769, 107 S. Ct. 3102, 3110, 97 L. Ed. 2d 618

(1987) (Stevens, J., concurring in judgment).

507 U.S. at 638 n.9. Hassine contends that this casefits

squarely within the Footnote Nine exception, because the

state prosecutor committed a "deliberate and especially

egregious error" at trial. He argues that "it was nothing

other than deliberate for the very experienced prosecutor to

persist in pointedly cross-examining Hassine about his

post-arrest silence," and maintains that it was "anything

short of deliberate" and egregious when the prosector

commented on Hassine's post-arrest silence during closing

arguments. (Pet'r Br. at 34 n.14.)28 However, after reviewing

the origins of the Footnote Nine exception and the few cases

in which it has been applied, we conclude that the

prosecutor's Doyle violation did not rise to the level needed

to warrant habeas relief.29

As the citation in Footnote Nine acknowledges, the

exception to Brecht's harmless error standard has its roots

in the views of Justice Stevens, who has written on several

occasions that habeas petitions should be granted where

trial errors are truly extraordinary, even if they are found to

constitute harmless error. Brecht itself cites to Justice

Stevens' concurrence in Greer v. Miller, 483 U.S. at 767,

which, in turn, refers to the Justice Stevens dissent in Rose

v. Lundy, 455 U.S. 509, 538 (1982). In Rose, Justice

Stevens argued that there are at least four types of

constitutional error:

The one most frequently encountered is a claim that

attaches a constitutional label to a set of facts that

_________________________________________________________________

28. Hassine does not contend that the error falls within Footnote Nine's

prohibition on errors "combined with a pattern of prosecutorial

misconduct." Brecht, 507 U.S. at 638 n.9. He argues only that the Doyle

violation was sufficiently "deliberate and . . . egregious" to warrant

habeas relief. We will thus limit our analysis to this aspect.

29. It is worth noting that Brecht does not truly establish an exception

for "deliberate and especially egregious errors;" it merely "does not

foreclose the possibility" that such an exception could exist. 507 U.S.

638 n.9. Nevertheless, for purposes of this discussion, we will assume

arguendo that Footnote Nine does support awarding habeas relief for

egregious errors in an appropriate case.

33

does not disclose a violation of any constitutional right.

. . . The second class includes constitutional violations

that are not of sufficient import in a particular case to

justify reversal even on direct appeal, when the

evidence is still fresh and a fair retrial could be

promptly conducted. . . . A third category includes

errors that are important enough to require reversal on

direct appeal but do not reveal the kind of fundamental

unfairness to the accused that will support a collateral

attack on a final judgment. . . . The fourth category

includes those errors that are so fundamental that they

infect the validity of the underlying judgment itself, or

the integrity of the process by which that judgment

was obtained.

Id. at 543-44 (dissenting opinion) (citations and footnote

omitted).

It is, of course, this fourth category that the Court

referenced eleven years later in Brecht as a potential

exception to the harmless error standard. Justice Stevens

has admitted that "[t]his category cannot be defined

precisely," id. at 544, but his opinions in Rose and Greer do

provide some guidance as to which infractions are to be

considered `fundamentally' unfair. In Rose, Justice Stevens

explained that the "error that falls in this category is best

illustrated by recalling the classic grounds for the issuance

of a writ of habeas corpus -- that the proceeding was

dominated by mob violence; that the prosecutor knowingly

made use of perjured testimony; or that the conviction was

based on a confession extorted from the defendant by

brutal methods." Id. (footnotes omitted). In his view,

"[e]rrors of this kind justify collateral relief no matter how

long a judgment may have been final and even though they

may not have been preserved properly in the original trial."

Id. (footnotes omitted). Further, in Greer, Justice Stevens

applied these principles to Doyle violations, writing that

"there may be extraordinary cases in which the Doyle error

is so egregious, or is combined with other errors or

incidents of prosecutorial misconduct, that the integrity of

the process is called into question. In such an event,

habeas corpus relief should be afforded." 483 U.S. at 769

(concurring).

34

Brecht itself sheds light on the Footnote Nine exception

as well. Brecht involved a petitioner convicted of first-degree

murder and sentenced to life imprisonment. The petitioner

took the stand at trial and admitted that he had fatally

wounded his brother-in-law, Roger Hartman, but he

claimed that the killing was accidental. Over the objections

of defense counsel, and in violation of Doyle , the State

asked the petitioner on cross-examination and again on re-

cross-examination "whether he had told anyone at any time

before trial that the shooting was an accident, to which

petitioner replied `no.' " 507 U.S. at 625. The prosecutor

then made three glaring references to the petitioner's pre-

trial silence during closing arguments, urging the jury to

" `remember that Mr. Brecht never volunteered until in this

courtroom what happened in the Hartman residence,' "

stating that " `[h]e sits back here and sees all of our

evidence go in and then he comes out with this crazy

story,' " and reminding the jury that the petitioner " `didn't

say' " this was an accident, " `[n]o, he waited until he hears

our story.' " Id. at 625 n.2 (quoting the record in the case).

Nevertheless, the Supreme Court still applied a harmless

error standard, so that the facts in Brecht did not involve "a

deliberate and especially egregious error of the trial type, or

one that is combined with a pattern of prosecutorial

misconduct" as reflected in Footnote Nine. Id. at 638 n.9.

As the Court wrote, "[w]e, of course, are not presented with

such a situation here." Id.30

With this in mind, we too conclude that we are not here

presented with the type of egregious situation alluded to in

Footnote Nine. First, the error in this case was much less

_________________________________________________________________

30. We are aware of only two federal circuit court opinions discussing the

Footnote Nine exception in detail, but neither of these cases is helpful

to

our discussion. We addressed the Brecht exception in Robinson v.

Arvonio, 27 F.3d 877, 886 (3d Cir. 1994), but Robinson was later vacated

on appeal to the Supreme Court and is no longer good law. See Robinson

v. Arvonio, 513 U.S. 1186 (1995). In addition, the Ninth Circuit

discussed Footnote Nine in Hardnett v. Marshall, 25 F.3d 875, 879-81

(9th Cir. 1994), but, in contrast to the instant case, Hardnett involved

an

error that was ultimately found to be irrelevant to the jury's verdict. We

will thus limit our analysis to Supreme Court precedents discussing the

Brecht exception.

35

egregious than in Brecht where the Court did employ a

harmless error analysis. At Hassine's trial, objections were

immediately raised and sustained, Hassine never answered

the improper questions, and the prosecutor's references to

Hassine's silence during closing arguments were

ambiguous. See supra. Therefore, if the Supreme Court

believed that Brecht was not "of course" a Footnote Nine

case, we fail to see how Hassine's case could possibly

qualify under the Brecht exception.

Second, we do not believe that the prosecutor's actions

here were "deliberate and especially egregious" within the

meaning of Footnote Nine. As we discussed supra, there

was some evidence that Hassine had opened the door to

questions about his post-arrest silence by testifying that he

was "available" for questioning by the police. In fact, this is

precisely what the Pennsylvania Superior Court found on

direct appeal of Hassine's claim. See Hassine I, 490 A.2d at

451. Thus, while we now disagree with the Superior Court

and find that the questions were inappropriate, we cannot

say that the prosecutor's actions were so deliberate and

egregious as to warrant habeas relief that bypasses

harmless error scrutiny.

Finally, we conclude that the Doyle violation did not "so

infect the integrity of the proceeding," Brecht, 507 U.S. at

638 n.9, or "call[ ] into question" the integrity of the trial

process, Greer, 483 U.S. at 769 (Stevens, J., concurring), so

as to warrant the grant of habeas relief to Hassine. As we

have discussed, the Doyle violation in this case consisted of

three unanswered questions, occupying just one page in a

transcript of more than 1500 pages of testimony, in

addition to two ambiguous references to Hassine's silence

during closing arguments. Moreover, the State called thirty-

four witnesses on its behalf and impeached Hassine's

testimony on several different grounds. Under these

circumstances, we do not believe that Hassine's case can be

classified as the "unusual," Brecht, 507 U.S. at 638 n.9, or

"extraordinary," Greer, 483 U.S. at 769 (Stevens, J.,

concurring), proceeding in which relief can be granted

pursuant to Footnote Nine. The Doyle violation at Hassine's

trial was significant, but we remain confident that its

impact on the entire case was not so profound as to infect

the very integrity of Hassine's conviction.

36

IV. CONCLUSION

For the foregoing reasons, the district court's Order dated

October 30, 1997, will be affirmed.

37

NYGAARD, Circuit Judge, Concurring.

I believe the majority reaches the correct result, agree

with its reasoning, and, so I concur in its judgment. I write

separately only to add my thoughts to Section III, D of its

opinion, and express my indignation with the prosecuting

attorney's misbehavior.

The concept of an accused's right to remain silent, to be

free from prosecutorial comment about that silence, and to

be free from any adverse consequences from accepting this

constitutional protection, is so fundamental in our

jurisprudence, and is so well known among practitioners,

that I cannot conceive of any alternative but that the

prosecutor's use of Hassine's silence was both deliberate

and calculated to invite the jury to draw an adverse

inference from it. Moreover, I believe this deliberate

violation of Mr. Hassine's constitutional rights is egregious

-- indeed, it is outrageous. The prosecutor not only

questioned Hassine about his silence, but continued to do

so even after objections to the questions were sustained by

the trial court. This arrogance towards the court's ruling

and the accused's rights, is reprehensible.

Nonetheless, on this habeas corpus review, I cannot say

that his behavior, although unprofessional, "so infected" the

trial's integrity, that it requires us to grant Hassine relief.

Moreover, even though the prosecutor again made reference

to Hassine's silence in his closing argument, considering

the entire record, I agree with the majority -- we cannot say

that his action constituted a "pattern" of misconduct. The

prosecutor's act was unfair, for which he deserves a stern

rebuke. Hassine, nonetheless, received a fair trial.

Hence, although the prosecutor's behavior came very

close to crossing the line of harm, and came even closer to

that "unusual" case in which relief could be granted

regardless of harm, I agree with the majority that it did not.

A True Copy:

Teste:

Clerk of the United States Court of Appeals

for the Third Circuit

38