November 22, 1994 UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT

No. 93-1795

NAZZARO SCARPA, Petitioner, Appellee,

v.

LARRY E. DUBOIS, ETC., Respondent, Appellant.

ERRATA SHEET ERRATA SHEET

The opinion of the Court issued on October 18, 1994, is corrected as follows:

On page 26, line 17, "449" should be "499"

UNITED STATES COURT OF APPEALS FOR THE FIRST CIRCUIT

No. 93-1795

NAZZARO SCARPA,

Petitioner, Appellee,

v.

LARRY E. DUBOIS, ETC.,

Respondent, Appellant.

APPEAL FROM THE UNITED STATES DISTRICT COURT

FOR THE DISTRICT OF MASSACHUSETTS

[Hon. William G. Young, U.S. District Judge]

Before

Selya, Cyr and Boudin,

Circuit Judges.

William J. Duensing, Assistant Attorney General, with whom

Scott Harshbarger, Attorney General, was on brief, for appellant.

Nazzaro Scarpa, pro se, orally and on original briefs, and

Seth M. Kalberg, by appointment of the court, orally and on

supplemental brief, for appellee.

October 18, 1994

SELYA, Circuit Judge. This appeal requires that we SELYA, Circuit Judge.

address an important question, not authoritatively resolved by

controlling precedent: When (if ever) does defense counsel's

substandard performance in a criminal case never a pretty sight

become so unattractive that a habeas court must forgo the

customary inquiry into the harmful effects of attorney error and,

instead, conclusively presume that counsel's blunders prejudiced

the defendant?

The question arises in the following context.

Petitioner-appellee Nazzaro Scarpa brought a pro se application

for habeas corpus in the federal district court.1 See 28 U.S.C.

2241-2254 (1988). He denominated a state correctional

official, in his representative capacity, as the respondent. The

district court discerned a Sixth Amendment violation: it

concluded that Scarpa's trial counsel in the state court rendered

grossly ineffective legal assistance to him, see Strickland v.

Washington,

(1984) (elucidating applicable

test); see also Hill v. Lockhart,

(1985)

(applying Strickland in the habeas context), and that counsel's

woeful performance gave rise to a per se presumption of

prejudice. The district court relied principally on dictum

contained in United States v. Cronic,

(1984), for

the proposition that it did not need to inquire into the

1On appeal, petitioner has also appeared pro se, preparing a

brief and arguing orally on his own behalf. To assist him, we appointed counsel who filed a supplemental brief and presented additional oral argument.

3

existence of actual prejudice.

Respondent appeals. Although the district court's

reading of Cronic finds some support in the case law, including

isolated cases decided by the Ninth and Tenth Circuits, see

United States v. Swanson,

(9th Cir. 1991);

Osborn v. Shillinger,

(10th Cir. 1988), we

believe that Cronic is not nearly so wide-ranging as the district

court assumed. Hence, we reverse.

I. BACKGROUND I. BACKGROUND

We glean the essential facts from the transcript of

petitioner's trial in Suffolk Superior Court. On June 10, 1987,

Joseph Desmond, an agent of the federal Drug Enforcement

Administration (DEA), posing as a would-be cocaine purchaser, met

with his initial target, Robert Ricupero, at a pub in East

Boston. At Ricupero's request, petitioner joined them. The trio

discussed a possible cocaine purchase and then crossed the street

to a parked limousine that bore the insignia of the "Snow White

Limousine Service." Ricupero and Scarpa entered the vehicle. As

Desmond later testified, Scarpa passed roughly 28 grams of

cocaine to Ricupero, who handed it to Desmond in exchange for

$1500 in cash. Ricupero kept $100 and gave the remainder to

Scarpa. These events occurred under police surveillance.

The next encounter between Desmond and his prey

occurred on July 18, 1987. In preparation for it, the

authorities again assigned a cadre of law enforcement officers to

surveillance duties. Desmond and Ricupero met at the same pub.

4

At Ricupero's invitation, Scarpa again joined them. On this

occasion, the actual exchange occurred in the deserted stairwell

of a nearby apartment building, and a fourth man, James Marcella,

entered the equation. Desmond testified that Marcella handed a

package containing roughly 55 grams of cocaine to Scarpa, who

passed the package to Ricupero. When Ricupero placed the drugs

within Desmond's reach, Desmond handed him $3000. Ricupero

slipped the money to Scarpa, who turned it over to Marcella.

In due season, the Commonwealth indicted petitioner for

drug trafficking and unlawful distribution. A jury convicted him

on all charges after a four-day trial. The trial judge sentenced

him to serve a lengthy prison term. Petitioner's motion for a

new trial failed; the Massachusetts Appeals Court affirmed the

conviction, see Commonwealth v. Scarpa,

,

(1991) (table); and the Supreme Judicial Court

(SJC) summarily denied petitioner's application for leave to

obtain further appellate review (alofar), see Commonwealth v.

Scarpa,

,

(1991).

Undaunted, Scarpa filed an application for a writ of

habeas corpus in federal district court. After hearing arguments

presented by Scarpa and by the Commonwealth, the district court

granted the petition. It found that defense counsel's

performance not only fell below an objectively reasonable

standard of proficiency but also caused a breakdown in the

adversarial system. This, the district judge thought,

constituted prejudice per se. Accordingly, he vacated the

5

conviction, ordered petitioner released from state custody, and

directed the Commonwealth to retry him if it sought to exact

further punishment. The court refused respondent's application

for a stay, and petitioner is at liberty.

II. EXHAUSTION OF REMEDIES II. EXHAUSTION OF REMEDIES

The Commonwealth is the real party in interest in these

proceedings, and we treat the case as if it were the named

respondent. At the outset, the Commonwealth seeks to sidestep

habeas relief by convincing us that petitioner failed to present

his constitutional claim to the state courts before bolting to a

federal forum. We are not persuaded.

A. Governing Principles. A. Governing Principles.

Under our federal system, both the federal and state

courts are entrusted with the protection of constitutional

rights. See Ex parte Royall,

(1886). In order

to ease potential friction between these two sovereigns, a

federal court will ordinarily defer action on a cause properly

within its jurisdiction until the courts of another sovereign

with concurrent powers, already cognizant of the litigation, have

had an opportunity to pass upon the matter. See Rose v. Lundy,

(1982). This practice, reflecting concerns of

comity, has been codified in 28 U.S.C. 2254,2 and memorialized

2The statute provides in pertinent part:

* * *

(b) An application for a writ of habeas corpus in behalf of a person in custody pursuant to the judgment of a State court

6

in our case law, see, e.g., Mele v. Fitchburg Dist. Court,

(1st Cir. 1988).

In order to present a federal claim to the state courts

in a manner sufficient to satisfy exhaustion concerns, a

petitioner must inform the state court of both the factual and

legal underpinnings of the claim. See Picard v. Conner,

(1971). The test is substantive: was the claim

presented in such a way as to make it probable that a reasonable

jurist would have been alerted to the existence of the federal

question? See Nadworny v. Fair,

(1st Cir.

1989). While the answer to the question must not be made to

depend on "ritualistic formality,"

, neither is the

answer wholly in the eye of the beholder.

In Gagne v. Fair,

(1st Cir. 1987), we

catalogued four ways in which the requirement of fair presentment

may be fulfilled: "1) citing a specific provision of the

Constitution; 2) presenting the substance of a federal

shall not be granted unless it appears that the applicant has exhausted the remedies available in the courts of the State, or that there is either an absence of available State corrective process or the existence of circumstances rendering such process ineffective to protect the rights of the prisoner. (c) An applicant shall not be deemed to have exhausted the remedies available in the courts of the State, within the meaning of this section, if he has the right under the law of the State to raise, by any available procedure, the question presented.

28 U.S.C. 2254(b), (c) (1988).

7

constitutional claim in such manner that it likely alerted the

state court to the claim's federal nature; 3) reliance on federal

constitutional precedents; and 4) claiming a particular right

specifically guaranteed by the Constitution." We did not,

however, attribute exclusivity to this compendium. In Nadworny,

, we mentioned a fifth possibility, namely,

the assertion of a state law claim that is functionally identical

to a federal claim. These possibilities recognize that certain

constitutional violations have the capacity to rest on a variety

of factual bases. While the facts and legal theories need not be

propounded in precisely the same terms, fair presentation

requires that the constitutional analysis necessary to resolve

the ultimate question posed in the habeas petition and in the

state court proceedings, respectively, be substantially the same.

See Lanigan v. Maloney,

(1st Cir. 1988), cert.

denied,

(1989).

B. Analysis. B. Analysis.

Here, petitioner's odyssey through the Massachusetts

court system involved a trial, a motion for a new trial, a full-

dress appeal in the state appeals court, and an alofar. At all

three post-trial stages, petitioner raised claims anent counsel's

proficiency (or, more precisely put, counsel's lack of

proficiency) and couched his claim in terms that remained largely

unchanged. In his pleadings and memoranda at all three stages,

petitioner alleged three principal shortcomings on counsel's

part: a failure to attack the prosecution's star witness; a

8

mindless solicitation to the jury to believe that star witness;

and the ill-advised pursuit of a defense, not legally cognizable,

that virtually conceded the elements of the charged offenses.

Throughout the appellate process, petitioner described his claim

as "ineffective assistance of counsel."

Moreover, at the first two stages he cited three state

cases, Commonwealth v. Pope,

(Mass. 1984);

Commonwealth v. Satterfield,

(Mass. 1977);

Commonwealth v. Saferian,

(Mass. 1974), that dealt

squarely with this issue.3 In his motion for new trial,

petitioner cited the Sixth Amendment by name, accompanying the

motion with the affidavit of his trial counsel, Arthur Tacelli,

attesting to Tacelli's self-professed ineffectiveness. Scarpa's

federal habeas petition again asserted "ineffective assistance of

counsel," and cited the same three factual bases in support of

the assertion.

On these facts, we agree with the district judge that

the arguments presented by petitioner sufficiently alerted the

state courts to the substance of the constitutional claim. In

the first place, an argument phrased as "ineffective assistance

of counsel" certainly "claim[s] a particular right specifically

3The Commonwealth makes much of the fact that these cases were not cited in the alofar, and insists that Mele,

, requires a federal court to restrict the exhaustion inquiry to that document. This crabbed reading of Mele wrenches the case

out of its context. There, the defendant raised his constitutional issue before the intermediate appellate court, abandoned it in his alofar, and then attempted to raise it anew in his habeas petition. See

. In contrast, Scarpa

has consistently asserted his ineffective assistance claim.

9

guaranteed by the Constitution." Gagne,

. In the

second place, by identifying the Sixth Amendment in his motion

for a new trial, petitioner "cite[s] a specific provision of the

Constitution,"

and, at the same time, provided a backdrop

against which his later filings had to be viewed.

If any doubt remains, the sockdolager is that, as a

general rule, presenting a state-law claim that is functionally

identical to a federal-law claim suffices to effectuate fair

presentment of the latter claim. See Nadworny,

-

1100. So it is here: petitioner brought himself within the

encincture of this rule by his repeated citation to the trio of

Massachusetts cases that we have mentioned cases that evaluate

the effectiveness of an attorney's performance in terms

reminiscent of the federal constitutional standard. As in

Strickland,

, the Massachusetts cases call for a

deferential evaluation of counsel's performance, and, if the

performance is found to be substandard, an inquiry into whether

counsel's incompetence injured the defendant's substantial

rights. See Pope,

; Satterfield,

; Saferian,

.4

4The SJC has made clear that it ordinarily considers questions involving "assistance of counsel" as coming "within the

meaning of the Sixth Amendment." Saferian,

(emphasis supplied). A defendant must show that

there has been serious incompetency, inefficiency, or inattention of counsel behavior of counsel falling measurably below that which might be expected from an ordinary fallible lawyer and, if that is found, then, typically, whether it has likely

10

Despite minor differences in phraseology, the two

standards state and federal strike us as equivalent. Indeed,

the Commonwealth does not contend that a claim of ineffective

assistance of counsel arising under Massachusetts law differs

from such a claim arising out of the Sixth Amendment. We readily

appreciate why this contention is not voiced. The essence of

each inquiry looks to the likelihood that effective assistance of

counsel would have produced a different trial outcome. The SJC

itself, while leaving open the theoretical possibility that there

might be some difference between the state and federal standards,

has concluded that if their state's test is satisfied, "the

Federal test is necessarily met as well." Commonwealth v.

Fuller,

, 385 n.3 (Mass. 1985). Finally, we deem

it highly relevant that the SJC has continued to apply the

Saferian analysis to ineffective assistance of counsel claims in

the post-Strickland era. See, e.g., Commonwealth v. Charles, 489

deprived the defendant of an otherwise available, substantial ground of defence.

Id. at 883. This is functionally identical to the federal

standard, which calls for a defendant to show

that counsel's performance was deficient. This requires showing that counsel made errors so serious that counsel was not functioning as the "counsel" guaranteed the defendant by the Sixth Amendment. Second, the defendant must show that the deficient performance prejudiced the defense. This requires showing that counsel's errors were so serious as to deprive the defendant of a fair trial, a trial whose result is reliable.

Strickland,

.

(Mass. 1986); Commonwealth v. Licata,

(Mass. 1992).

To be sure, petitioner failed to cite directly to

federal precedent in his journey through the state appellate

process. In our view, however, this omission is not fatal.

Although such citation is strongly recommended if only to avoid

controversies of this nature, we have specifically declined to

adopt a bright-line rule. See Nadworny,

& n.4.

The guidelines we have promulgated in respect to exhaustion are

intended to be instructive, rather than to comprise the sole

corridors through which the "actual embodiment of fair

presentation" may pass.

.

To say more would be supererogatory. For the reasons

stated above, we conclude that petitioner's Sixth Amendment claim

was put to the state courts with the requisite clarity. See

Twitty v. Smith,

(2d Cir. 1979) (finding a

similar claim exhausted, under analogous circumstances, because

"the mention of `effective assistance of counsel' instantly calls

to mind the Sixth Amendment's guaranty of the accused's right `to

have the Assistance of Counsel for his defence'") (citations

omitted); see also Daye v. Attorney General,

(2d Cir. 1982) (en banc) (reaffirming Twitty holding); Brady v.

Ponte,

(D. Mass. 1988) (stating that explicit

reference to "ineffective assistance of counsel" suffices to

exhaust a Sixth Amendment claim) (dictum).

III. THE MERITS III. THE MERITS

12

We segment our consideration of the merits, first

outlining certain legal principles of general applicability, then

essaying an overview of petitioner's trial, and thereafter

synthesizing the fruits of these endeavors by applying the

relevant principles to the relevant circumstances.

A. Governing Principles. A. Governing Principles.

The Sixth Amendment guarantees criminal defendants the

right to effective assistance of counsel. See Strickland,

. The touchstone for determining whether an

attorney's performance falls below the constitutional norm is

whether counsel has brought "to bear such skill and knowledge as

will render the trial a reliable adversarial testing process."

. The inquiry has two foci. First, a reviewing court

must assess the proficiency of counsel's performance under

prevailing professional norms. See United States v. Natanel,

(1st Cir. 1991), cert. denied,

(1992). This evaluation demands a fairly tolerant approach;

after all, the Constitution pledges to an accused an effective

defense, not necessarily a perfect defense or a successful

defense. See, e.g., Lema v. United States,

(1st

Cir. 1993); Natanel,

. And, moreover, since even

the most celebrated lawyers can differ over trial tactics in a

particular case, a reviewing court must "indulge a strong

presumption that counsel's conduct falls within the wide range of

reasonable professional assistance." Strickland,

.

13

The second line of inquiry is needed because, in

itself, dreary lawyering does not offend the Constitution.

Rather, a finding that counsel failed to meet the performance

standard merely serves to advance the focus of the Strickland

inquiry to the question of whether the accused suffered prejudice

in consequence of counsel's blunders. See

. This

entails a showing of a "reasonable probability that, but for

counsel's unprofessional errors, the result of the proceeding

would have been different."

. A defendant who alleges

ineffective assistance of counsel must carry the devoir of

persuasion on both tiers of the Strickland test. See, e.g.,

Lema,

. The same holds true of a habeas petitioner

who claims that counsel mishandled his case in the state courts.

See Perron v. Perrin,

(1st Cir. 1984).

An inquiry into the effectiveness of counsel is almost

always a mixed question of law and fact. See Strickland,

. In federal courts, mixed questions of law and fact

arising in section 2254 cases are ordinarily subject to de novo

review. See Chakouian v. Moran,

(1st Cir.

1992). This includes claims premised on ineffective assistance

of counsel. See, e.g., McAleese v. Mazulkiewcz,

, 165

(3d Cir. 1993); Fields v. Attorney General,

, 1297

n.18 (4th Cir.), cert. denied,

(1992).

Comfortable with this precedent, and mindful that the district

court's "prejudice per se" ruling derives from a conception of

law rather than from a finding of fact, we apply a de novo

14

standard of review here.5

B. The State Court Trial. B. The State Court Trial.

In the superior court, the prosecution conveyed its

case principally through two witnesses. Desmond supplied

detailed, firsthand testimony anent the cocaine sales and a

Boston police detective, Joseph Mugnano, testified that Scarpa

admitted owning the Snow White Limousine Service.

Scarpa's defense counsel did not attempt to impeach

Desmond, but, rather, rehashed the direct examination, extracting

from Desmond the following facts: that Ricupero, not Scarpa, was

the primary target of the DEA's investigation; that Ricupero

initially indicated to Desmond that his repository for drugs was

a pickup truck, not a limousine; that, with respect to the first

transaction, (1) Desmond did not know who put the cocaine in the

limousine, and (2) someone other than Scarpa actually handed the

5Some courts have suggested that a standard of independent review which we have described in a different context as "an intermediate level of scrutiny, more rigorous than the abuse of discretion or clear-error standards, but stopping short of plenary or de novo review," United States v. Tortora,

(1st Cir. 1990) applies in habeas cases. See, e.g.,

Battle v. Dell,

(8th Cir. 1994); Hamilton v.

Ford,

(11th Cir. 1992), cert. denied,

(1993); see also S. Childress & M. Davis, Federal

Standards of Review 13.05, at 13-37 (1992). We are satisfied

that de novo review is appropriate in the case at bar, and we

need not decide today whether a standard of independent review should ever be employed in habeas cases. Withal, it strikes us that where, as here, the district judge does not himself take any evidence, the gap between independent review and de novo review,

if one exists at all, is necessarily very small. Cf. Tortora,

(explaining that lesser deference is warranted when district court essays no "new or different factfinding," but, instead, acts on the basis of a magistrate's findings and report).

15

drugs to Desmond; that, with respect to the second transaction,

(1) Scarpa was a middle link in the chain of drugs and cash, and

(2) Desmond did not know whether Scarpa received any money

referable to that transaction. Attorney Tacelli declined to

question Mugnano and produced no witnesses in Scarpa's defense.

His closing argument consisted of a terse explanation of the

concept of reasonable doubt and a solicitation to the jury to

accept the government's testimony:

So, I'm asking you, as finders of fact, to believe Detective Mugnano, because his testimony, I suggest, is innocuous. The second witness that the Government and the prime witness that the Government produced in support of their argument that Mr. Scarpa was guilty of cocaine trafficking and distribution, was Agent Desmond . . . And you listen to DEA Agent Drug Enforcement Agent Desmond. And I ask you: What motive would that man have to come into a superior court courtroom, with a varied jury, a superior court judge, what motive would he have for lying? What motive would he have to tell an untruth? What motive would he have to color the fact situation as he remembered it? And I suggest to you and I hope you

find resoundingly that he has no motive but

that of following the truth. . . . [I]'m

asking you to find that man a credible human being; a man who came in, took the oath and told the truth [emphasis supplied].

Speaking of Scarpa, Attorney Tacelli continued:

Was he a user of cocaine? Was he a dupe? What happened to that money? What was its final destination? Is Scarpa a user of drugs? Is Scarpa someone that Ricupero, the target of the investigation is Scarpa was he used by Ricupero to shield himself? . . . And I'm suggesting to you again, at the expense of being repetitious, Scarpa is not found and it is undetermined that is the word that Agent Desmond used on July 8th it's undetermined if Scarpa had any of that

16

money. . . . And clearly, the source of the cocaine on the 8th was not Scarpa. At best

he was a conduit; someone through whom it

passed, and through whom the money passed

[emphasis supplied].

During summation, the prosecutor agreed that Desmond

had no reason to lie. He told the jury that the Commonwealth had

no obligation to prove either the source of the cocaine or the

ultimate destination of the money. And he labelled defense

counsel's closing argument "a smokescreen."

In due course, the judge instructed the jury on the

elements of the trafficking offense. He told the jurors, in

substance, that to convict, they must find that the defendant (1)

knowingly (2) possessed cocaine; (3) with the intent to

distribute it; and (4) that the quantity of cocaine must be in

excess of 28 grams. See Mass. Gen. Laws ch. 94C, 32E(b)

(1992). The judge instructed the jurors to much the same effect

in regard to the distribution charge, but substituted

distribution for possession and eliminated any reference to a

minimum quantity of cocaine. See

32A(a). The judge also

informed the jury that the identity of "the kingpin" did not bear

upon the charges at hand. The jury convicted Scarpa on both

counts.

C. The Attorney's Performance. C. The Attorney's Performance.

The district court deemed defense counsel's argument as

tantamount to arguing that petitioner was a "mere conduit" for

the contraband. Believing that this approach effectively

conceded the only disputed elements of the charged crimes and

17

relieved the prosecution of its burden of proof, the court found

Attorney Tacelli's use of it to be objectively unreasonable, and

therefore, substandard.

We uphold this finding. At the least, defense counsel

in a criminal case should understand the elements of the offenses

with which his client is charged and should display some

appreciation of the recognized defenses thereto. See Young v.

Zant,

(11th Cir. 1982) (explaining that defense

counsel falls below performance standard by failing to understand

his client's factual claims or the legal significance of those

claims); Baty v. Balkcom,

(5th Cir. 1981)

(holding that defense counsel's unfamiliarity with his client's

case transgressed performance standard), cert. denied,

(1982). Unless counsel brings these rudiments to the table,

a defendant likely will be deprived of a fair "opportunity to

meet the case of the prosecution," Strickland,

(quoting Adams v. United States ex rel. McCann,

(1942)), and, thus, will be placed at undue risk of

having no effective advocate for his cause. Phrased another way,

if an attorney does not grasp the basics of the charges and the

potential defenses to them, an accused may well be stripped of

the very means that are essential to subject the prosecution's

case to adversarial testing. See id. at 688.

We agree with the district court that this is such a

case. Defense counsel's pursuit of his half-baked theory

evidenced a blatant misunderstanding of the charged crimes.

18

Being a "conduit" denotes acting as an agent or intermediary.

Persons who knowingly serve as agents or intermediaries in

narcotics transactions are punishable as principals under

Massachusetts law. See Commonwealth v. Murillo,

(Mass.), rev. denied,

(1992); Commonwealth v.

Poole,

(Mass. 1990). Thus, the line of

defense that counsel selected was altogether irrelevant to

petitioner's guilt or innocence; and, to compound the problem,

the steps taken in pursuit of it such as urging the jury to

accept Desmond's testimony played into the prosecution's hands.

Serious errors in an attorney's performance, unrelated to

tactical choices or to some plausible strategic aim, constitute

substandard performance. See United States v. Weston,

(7th Cir.) (examining only those errors not reasonably

classifiable as tactical choices to determine the existence of

grossly unprofessional conduct), cert. denied,

(1983); see also Francis v. Spraggins,

(11th

Cir. 1983) (stating that "complete concession of the defendant's

guilt" may constitute ineffective assistance), cert. denied,

(1988); cf. United States v. Tabares,

(1st Cir. 1991) (finding no ineffective assistance when counsel's

concession is strategic); Underwood v. Clark,

(7th Cir. 1991) (similar). This verity has particular force

where, as here, counsel's blunders not only failed to assist in

fashioning a defense but also cemented the prosecution's theory

of the case. There are times when even the most adroit advocate

19

cannot extricate a criminal defendant from a pit; but when

counsel, to no apparent end, digs the hole deeper, the Sixth

Amendment performance standard is likely implicated.

The Commonwealth's rejoinder is lame. First, it

contends that Attorney Tacelli rendered constitutionally

effective assistance because the conduit defense is a "common

defense which raises issues considered good strategy." This is

no more than an ipse dixit, unsupported by authority. To be

sure, the Commonwealth cites a quadrat of cases in a

conspicuously unsuccessful effort to bolster this claim but

none of them is persuasive on the point. Two of these cases

stand for the unremarkable proposition that "mere presence" is

not enough to convict in a narcotics case, in the absence of

other evidence. See Commonwealth v. Cruz,

(Mass. 1993); Commonwealth v. Brown,

(Mass.

1993); see also United States v. Ortiz,

(1st

Cir. 1992) (explaining difference between "mere presence" and

"culpable presence" in drug-trafficking cases), cert. denied,

(1993). The other two cases are easily

distinguishable on the facts. See Commonwealth v. Johnson,

, 559 & n.8 (Mass. 1992); Commonwealth v. Claudio,

(Mass. 1988).

Second, respondent attempts to cast Attorney Tacelli's

pratfalls as an argument for jury nullification. This is pure

conjecture. The record contains no indication that counsel

strove to implant the notion of nullification in the jurors'

20

minds. In any event, "although jurors possess the raw power to

set the accused free for any reason or for no reason, their duty

is to apply the law as given to them by the court." United

States v. Sepulveda,

(1st Cir. 1993), cert.

denied,

(1994); see also Commonwealth v. Leno,

(1993) ("We do not accept the premise that

jurors have the right to nullify the law on which they are

instructed . . . ."). Consequently, defense counsel may not

press arguments for jury nullification in criminal cases, see

Sepulveda,

; United States v. Desmarais,

(1st Cir. 1991); Leno,

, and we will

not permit the Commonwealth to pretend that it sat idly by and

allowed Attorney Tacelli to violate this rule.

D. Prejudice. D. Prejudice.

Having found substandard performance, we come next to

the second prong of the Strickland inquiry. The district court,

while acknowledging that Scarpa's plight was "well nigh

hopeless," bypassed a case-specific inquiry into prejudice,

instead finding prejudice per se on the theory that counsel was

so derelict in his duty that petitioner, in effect, had no

counsel at all. We reject the application of a per se standard

to this case. Moreover, after conducting the full Strickland

analysis in the appropriate way, we find that petitioner suffered

no actual prejudice.

1. 1.

As mentioned above, the district court relied primarily

21

on dictum contained in United States v. Cronic,

-

60, for the proposition that, in the circumstances at bar, it

could forgo an inquiry into actual prejudice. The Cronic Court

stated that in rare instances prejudice might be presumed

"without inquiry into counsel's actual performance at trial."

(dictum). But, the approach suggested in this

statement is in all events the exception, not the rule and it

can be employed only if the record reveals presumptively

prejudicial circumstances such as an outright denial of counsel,

a denial of the right to effective cross-examination, or a

complete failure to subject the prosecution's case to adversarial

testing.6 See

. The Cronic Court itself warned that,

in most cases, a showing of actual prejudice remained a necessary

element. See

The Court stated: "there is generally no

basis for finding a Sixth Amendment violation unless the accused

can show how specific errors of counsel undermined the

reliability of the finding of guilt."

at 659 n.26.

For the most part, courts have been cautious in

invoking the exception limned in the Cronic dictum. Cronic like

principles have been applied, for example, in situations in which

6The facts of Cronic illustrate the narrowness of the

exception. In that case, the defendant was charged in a complicated check-kiting scheme. The government had spent over four years investigating the case, but when the defendant's counsel withdrew, the trial court appointed a young real estate lawyer only 25 days before trial. The Supreme Court held that this brief period for preparation was "not so short that it even arguably justifies a presumption that no lawyer could provide the [defendant] with the effective assistance of counsel required by the Constitution." 466 U.S. at 665.

22

defense counsel labored under an actual conflict of interest, see

Cuyler v. Sullivan,

(1980), or in which no attorney

appeared despite a defendant's unwaived right to appointed

counsel, see United States v. Mateo,

(1st Cir.

1991), or in which defendant's lawyer sat in total silence

throughout the relevant proceeding, see Tucker v. Day,

(5th Cir. 1992) (involving resentencing); Harding v.

Davis,

(11th Cir. 1989) (holding that defense

counsel's muteness throughout trial, including his utter silence

as the judge directed a verdict against his client, is per se

prejudicial), or in which the defense attorney was absent from

the courtroom during a critical part of the trial, see Green v.

Arn,

(6th Cir.), cert. granted, vacated

and remanded to consider mootness,

(1987); Siverson

v. O'Leary,

(7th Cir. 1985), or, pre-Cronic,

in which counsel snoozed through much of the proceedings, see

Javor v. United States,

(9th Cir. 1984).

A few courts have extended the exception's boundaries

beyond the circumstances surrounding representation and found

that a lawyer's particular errors at trial may cause a breakdown

in the adversarial system and thus justify invocation of the

Cronic dictum. See Swanson,

(holding that

knowingly and explicitly conceding reasonable doubt in closing

argument is per se prejudicial); Osborn,

(finding per se prejudice when defense counsel intentionally

stressed the brutality of his client's crime, admitted that the

23

evidence against his client was overwhelming, and made statements

to the press that his client had no evidence to support his

claims). We believe that these cases misperceive the rationale

underlying the Cronic exception. In our view, the Court's

language in Cronic was driven by the recognition that certain

types of conduct are in general so antithetic to effective

assistance for example, lawyers who leave the courtroom for

long stretches of time during trial are unlikely to be stellar

advocates in any matter that a case-by-case analysis simply is

not worth the cost of protracted litigation. No matter what the

facts of a given case may be, this sort of conduct will almost

always result in prejudice. See Cronic,

. But

attorney errors particular to the facts of an individual case are

qualitatively different. Virtually by definition, such errors

"cannot be classified according to likelihood of causing

prejudice" or "defined with sufficient precision to inform

defense attorneys correctly just what conduct to avoid."

Strickland,

. Consequently, the Court has

declined to accord presumptively prejudicial status to them. See

We are not alone in our attempt to harmonize Cronic

with Strickland by drawing an easily visible line separating

those few cases in which prejudice may be presumed from the mine-

run (in which actual prejudice must be shown). When confronted

by particular errors on the part of defense counsel, best

evaluated in the context of the defendant's trial, other federal

24

courts have refused to march under the Cronic banner, and,

instead, notwithstanding the seriousness of the errors, have

performed both parts of the requisite Strickland analysis. Thus,

in McInerny v. Puckett,

(5th Cir. 1990), the

defendant claimed that his lawyer's lack of preparedness and

failure to raise an insanity defense justified the invocation of

the Cronic dictum. See

. In requiring a showing of

prejudice, the Fifth Circuit noted that "bad lawyering,

regardless of how bad, does not support the [per se] presumption;

more is required."

; see also United States v.

Thompson,

(D.C. Cir. 1994) (finding no prejudice

per se in defense counsel's failure to inform defendant before

guilty plea that, as a career offender, he faced possible life

imprisonment); United States v. Baldwin,

(9th Cir.) (finding no prejudice per se where attorney conceded

his client's guilt at pretrial conference and neglected to

request jury instruction on overt act requirement for conspiracy

charge), cert. denied,

(1993); Woodard v.

Collins,

(5th Cir. 1990) (requiring showing

of prejudice where defense counsel advised the accused to plead

guilty to a charge that counsel had not investigated); United

States v. Reiter,

(2d Cir.) (applying full

Strickland standard in spite of defendant's claim that counsel's

errors were so pervasive as to amount to "no counsel at all"),

cert. denied,

(1990); Green v.Lynaugh,

(5th Cir.) (applying full Strickland analysis to

25

attorney's decision to conduct "almost no investigation"), cert.

denied,

(1989); Henderson v. Thieret,

(7th Cir. 1988) (applying second prong of Strickland to

attorney's lack of preparation in connection with sentencing),

cert. denied,

(1989); Gardner v. Ponte,

(1st Cir.) (refusing to extend Cronic to attorney's

failure to object to jury instructions), cert. denied,

(1987); State v. Savage,

(N.J. 1990)

(finding no prejudice per se in a capital case where counsel only

met once with defendant). Similarly, in reviewing claims of

ineffective assistance of counsel at the appellate level, courts

have declined to apply Cronic to attorney errors that do not

amount to the constructive absence of counsel. See, e.g.,

Hollenback v. United States,

, 1276 & n.1 (7th Cir.

1993) (finding no per se prejudice in appellate counsel's

citation to wrong provision of money-laundering statute); United

States v. Birtle,

(9th Cir. 1986) (finding

no per se prejudice when defendant's appellate counsel failed to

appear at oral argument or file a reply brief).7

7Of course, courts have not required a showing of prejudice when the attorney's inadequate performance completely denies the defendant his right to an appeal. See, e.g., Bonneau v. United

States,

(1st Cir. 1992) (requiring no showing of

prejudice when the defendant's appeal was dismissed due to his lawyer's failure to file a brief); United States ex rel. Thomas

v. O'Leary,

(7th Cir. 1988) (finding

prejudice per se when defense counsel filed no brief during

state's appeal of a suppression order and the ensuing decision was thus based only on the record and the government's brief); Williams v. Lockhart,

, 1137 n.3 (8th Cir. 1988)

(finding prejudice per se in attorney's failure to bring appeal

after promising to do so).

26

These authorities suggest that attorney error, even

when egregious, will almost always require analysis under

Strickland's prejudice prong. We agree. Thus, we decline to

adopt the expanded version of Cronic embraced by the district

court. Our reasons are manifold, but four of them are paramount.

First, we do not think that the Court intended such an

expansion to occur. Second, once it is necessary to examine the

trial record in order to evaluate counsel's particular errors,

resort to a per se presumption is no longer justified by the wish

to avoid the cost of case-by-case litigation. An overly generous

reading of Cronic would do little more than replace case-by-case

litigation over prejudice with case-by-case litigation over

prejudice per se.

Third, in our judgment the proper approach to the

intended reach of the Cronic dictum is informed by the

refinements of the harmless-error doctrine contained in a battery

of recent Supreme Court cases. Some constitutional errors,

denominated "trial errors," will not result in reversal of a

conviction if they are shown to be harmless. See Brecht v.

The counterpoint, however, is that in deciding whether to require a showing of prejudice for inadequate legal assistance on appeal, courts have traced a line, analogous to the one we draw today, distinguishing between inept performance and no performance. See, e.g., Penson v. Ohio,

(1988)

(requiring no showing of prejudice when defendant's lawyer withdrew without filing a brief on appeal, and distinguishing this situation from "a case in which counsel fails to press a particular argument on appeal or fails to argue an issue as effectively as he or she might") (citation omitted); Bonneau,

(requiring no showing of prejudice but distinguishing its facts from "a case of sloppy briefing that missed some vital issues" or a case of "inadequate oral argument").

27

Abrahamson,

(1993); Arizona v. Fulminante,

(1991). Examples of such trial errors

include overbroad jury instructions used during the sentencing

stage of a capital case, see Clemons v. Mississippi,

(1990); jury instructions containing an erroneous (but

rebuttable) presumption, see Carella v. California,

(1989); and improper prosecutorial comment on the

defendant's silence, see United States v. Hasting,

(1983). However, other more fundamental errors, denominated

"structural errors," jar the framework in which the trial

proceeds and, accordingly, are said to "defy analysis by

`harmless-error' standards," Brecht,

(quoting

Fulminante,

), and, thus necessitate "automatic

reversal of [a] conviction because they infect the entire trial

process,"

In effect, then, the harmfulness of structural

errors can be conclusively presumed. Examples of structural

errors, in addition to total deprivation of the right to counsel,

see Gideon v. Wainwright,

(1963), include failing to

give a constitutionally sufficient "reasonable doubt"

instruction, see Sullivan v. Louisiana,

(1993); permitting a trial to proceed before a biased

adjudicator, see Tumey v. Ohio,

(1927); and

discriminatorily excluding members of a defendant's race from a

grand jury, see Vasquez v. Hillery,

(1986),

or a petit jury, see Batson v. Kentucky,

(1986).

The "common thread" connecting the numerous examples of trial

28

error listed by Chief Justice Rehnquist in Fulminante is that all

such errors occur "during the presentation of the case to the

jury," and therefore may "be quantitatively assessed in the

context of [the] evidence presented" in order to gauge

harmlessness. Fulminante,

.

We are confident that what transpired in this case

bears a much stronger resemblance to trial error than to

structural error. Like the line separating trial errors from

structural errors, the line past which prejudice will be presumed

in cases involving claims of ineffective assistance ought to be

plotted to exclude cases in which a detailed contextual analysis

is required. Drawing the line in this way is especially fitting,

we suggest, because like the harmless-error doctrine, the

prejudice prong of Strickland helps to promote the salutary tenet

that "the central purpose of a criminal trial is to decide the

factual question of the defendant's guilt or innocence, and

promote[] public respect for the criminal process by focusing on

the underlying fairness of the trial rather than on the virtually

inevitable presence of immaterial error." Delaware v. Van

Arsdall,

(1986) (citation omitted).

Our fourth and final reason for taking a somewhat

narrow view of the Cronic dictum is closely related to the

concerns that the Court has expressed in the harmless-error

cases. In addition to comity and federalism concerns, the state

has an important interest in the finality of its jury verdicts

and in keeping behind bars criminals who have been fairly tried

29

and justly convicted. Forcing a state to retry its criminals

imposes social costs, including the expenditure of time and

resources for all concerned; the dispersal of witnesses and the

erosion of witnesses' memories; and the occurrence of sundry

other events that make obtaining a conviction more difficult on

retrial. See Brecht,

; cf. Barker v. Wingo,

, 522 n.16 (1972) (admonishing that the public has an

"interest in trying people accused of crime, rather than granting

them immunization because of legal error") (citation omitted).

For this reason, federal courts should not rush to overturn the

state-court conviction of a defendant who, although represented

by mistake-prone counsel, is unable to show how (if at all) the

lawyer's bevues undermined the fairness or reliability of the

trial's result.8

8At any rate, this is not the case in which to push the envelope. Even if one were to accept the expansive view of Cronic exemplified by Swanson,

, the record here

simply does not justify a finding of a complete failure to subject the prosecution's case to meaningful adversarial testing. Indeed, in the unpublished rescript accompanying its summary affirmance of Scarpa's conviction, the Massachusetts Appeals Court did not even find Attorney Tacelli's conduct to be "manifestly unreasonable." See Commonwealth v. Scarpa, No. 90-P-

694, at 2 (Mass. App. Ct. Mar. 7, 1991). While we do not necessarily agree with this evaluation, see supra Part III(C), we

recognize that whatever his failings, Attorney Tacelli strove to impress the jury with the gravity of the prosecution's burden. For example, he focused in his summation on "the obligation of the Government to prove their [sic] case beyond a reasonable doubt"; reminded the jurors that, in deciding the case, they must "have an abiding conviction"; and told them that they could "choose to believe everything a witness says, disbelieve it, [or] believe half of it." Although Attorney Tacelli weakened his presentation by his later remarks, quoted ante, he still left it

up to the jury to decide the ultimate question of Scarpa's guilt. Hence, we do not find in this record such a deliberate rolling over as might warrant a finding of an absolute breakdown of the

30

To summarize, we hold that Strickland controls

inquiries concerning counsel's actual performance at trial, and

that substandard performance, in the nature of particular

attorney errors, cannot conclusively be presumed to have been

prejudicial. Silhouetted against this backdrop, we consider it

supremely important that Attorney Tacelli's blunders cannot be

judged solely by the "surrounding circumstances" of the

representation, but, rather, must be judged in light of the whole

record, including the facts of the case, the trial transcript,

the exhibits, and the applicable substantive law. We conclude

that this characterization places the case beyond Cronic's reach.

Put bluntly, because Attorney Tacelli's errors are more an

example of maladroit performance than of non-performance,

Strickland necessitates an inquiry into the existence of actual

prejudice.

2. 2.

Since the district court presumed prejudice, it made no

explicit findings on the second prong of the Strickland test. We

have considered the advisability of remanding for this purpose,

but we conclude that it is unnecessary to do so. The origins of

the case date back to 1987; the parties have briefed and argued

the issue of actual prejudice; and the record is sufficient to

permit us to perform the decisionmaking task. Moreover, all the

evidence was taken in the state courts; thus, we are in as good a

position as the federal district judge to probe the matter. And,

adversarial process.

31

finally, even if a finding were made below, we would be obliged

to exercise de novo review, see supra p. 13. This combination of

factors persuades us to undertake the inquiry into actual

prejudice.

A convicted defendant can establish the requisite

prejudice in an ineffective assistance case by demonstrating a

reasonable probability that, but for counsel's bevues, the trial

outcome would have been different. For this purpose, a

reasonable probability is defined as that which undermines

confidence in the result of the proceeding. See Strickland,

; see also Kotteakos v. United States,

(1946). We caution however, that the analysis does not focus

solely on outcome determination, but also takes into prominent

consideration "whether the result of the proceeding was

fundamentally unfair or unreliable." Lockhart v. Fretwell,

(1993). This question must be answered without

reference to certain extraneous factors, such as "the possibility

of arbitrariness, whimsy, caprice, `nullification,' and the

like," which do not legitimately enter the jury's deliberations.

Strickland,

. With these omissions, our analysis

proceeds "on the assumption that the decisionmaker is reasonably,

conscientiously, and impartially applying the standards that

govern the decision."

Despite Attorney Tacelli's ineptitude, we discern no

actual prejudice here. The government presented clear,

uncontroverted eyewitness testimony from an agent who

32

participated in both drug-trafficking transactions and who had

conducted more than 30 undercover operations during his career.

Eight other law officers assisted agent Desmond and stood ready

to testify in a substantially similar fashion if the need arose.

The risk of prejudice from Attorney Tacelli's ill-advised request

that the jury credit the government's witness was minimized by

the one-sidedness of the evidence; here, there was no

contradictory version of the critical events that a skeptical

jury otherwise might have chosen to believe. Similarly, any

facts tacitly conceded during Attorney Tacelli's misconceived

"conduit" argument were overwhelmingly supported by the proof; as

we have mentioned, the record contains not one scintilla of

exculpatory evidence. To this day, petitioner has failed to

identify any promising line of defense or to construct a

plausible scenario that, if exploited, might have given the jury

pause.

We agree with the district court's observation that, on

this record, it is difficult to imagine any rational jury failing

to convict. Because there is neither a reasonable probability

that the outcome of the trial would have differed if counsel had

been more adept nor any solid basis for believing that the trial

was fundamentally unfair or unreliable, no Sixth Amendment

violation inheres.

IV. CONCLUSION IV. CONCLUSION

We need go no further. Petitioner's habeas claim is

ripe for review, but, upon due consideration, the claim fails.

33

Hence, the judgment below must be reversed and the case remanded

to the district court for the entry of an appropriate order

clearing the way for the Commonwealth to resume custody of

petitioner.

Reversed and remanded. Reversed and remanded.

34