In the Supreme Court of the United States
JOSEPH MASSARO, PETITIONER
UNITED STATES OF AMERICA
ON WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE SECOND CIRCUIT
BRIEF FOR THE UNITED STATES
THEODORE B. OLSON
Counsel of Record
Assistant Attorney General
MICHAEL R. DREEBEN
Deputy Solicitor General
Assistant to the Solicitor
STEVEN L. LANE
Department of Justice
Washington, D.C. 20530-0001
Whether the court of appeals correctly affirmed the denial of petitioner's motion to vacate his conviction under 28 U.S.C. 2255 on the basis of ineffective assistance of counsel because petitioner did not demonstrate cause and prejudice for his failure to raise that claim on direct appeal of his conviction.
In the Supreme Court of the United States
JOSEPH MASSARO, PETITIONER
UNITED STATES OF AMERICA
ON WRIT OF CERTIORARI
TO THE UNITED STATES COURT OF APPEALS
FOR THE SECOND CIRCUIT
BRIEF FOR THE UNITED STATES
The opinion of the court of appeals (Pet. App. A3-A10) is not published in the Federal Reporter, but is available at 27 Fed. Appx. 26. The opinion of the district court (Pet. App. A16-A26) also is not published in the Federal Reporter.
The judgment of the court of appeals was entered on November 2, 2001. A petition for rehearing was denied on January 14, 2002. Pet. App. A1-A2. The petition for a writ of certiorari was filed on April 15, 2002, and was granted on October 1, 2002. The jurisdiction of this Court rests on 28 U.S.C. 1254(1).
CONSTITUTIONAL AND STATUTORY
The relevant constitutional and statutory provisions are reprinted in an appendix to this brief. App., infra, 1a-2a.
Following a jury trial in the United States District Court for the Southern District of New York, petitioner was convicted of conspiracy to conduct and participate in the conduct of the affairs of a racketeering enterprise, in violation of 18 U.S.C. 1962(d); participating in the conduct of the affairs of a racketeering enterprise, in violation of 18 U.S.C. 1962(c); murder in aid of racketeering, in violation of 18 U.S.C. 1959(a)(1); three counts of conspiracy to commit extortion, in violation of 18 U.S.C. 1951; conspiracy to make extortionate extensions of credit and to use extortionate means to collect extensions of credit, in violation of 18 U.S.C. 892, 894; loansharking, in violation of 18 U.S.C. 893; and traveling interstate in aid of the extortion counts, in violation of 18 U.S.C. 1952. Pet. App. A31. Petitioner was sentenced to life imprisonment and a $240,000 fine, and was ordered to pay $104,100 in restitution. Id. at A18; Pet. 4. The court of appeals affirmed. United States v. Massaro, 57 F.3d 1063 (2d Cir.) (Table) (reprinted in Pet. App. A30-A37), cert. denied, 516 U.S. 933 (1995).
Petitioner then filed a motion under 28 U.S.C. 2255 challenging his convictions. The district court denied the motion. Pet. App. A16-A26. The court of appeals affirmed. Id. at A3-A10.
1. From at least the late 1970s until his arrest on June 24, 1992, petitioner Joseph Massaro was a "soldier" in the Luchese Organized Crime Family and actively participated in its various illicit affairs. As a Luchese soldier, petitioner's racketeering activity included using threats, violence, extortion and arson to secure and expand his control over the topless bar industry in Long Island, New York, extorting several other business establishments, and operating a loansharking business throughout Long Island, Brooklyn, and Queens, as well as a horse betting parlor in Queens. Pet. App. A4-A5, A31-A33.
On September 20, 1990, petitioner murdered his racketeering associate, Joseph Fiorito, because of Fiorito's failure to remit gambling proceeds and other monies. Petitioner enlisted the help of Patrick Esposito, another racketeering associate, to lure Fiorito to a vacant house in Hauppauge, Long Island, by telling Fiorito that they were going to commit an arson at the home. Petitioner was waiting in the basement of the house. Esposito asked Fiorito to go to the basement for some gasoline and rags, and when Fiorito went down the stairs, petitioner shot him in the head. Petitioner and Esposito then carried Fiorito's body up the stairs and through the first floor into the garage. They placed the body in a sitting position in the rear passenger seat of Fiorito's car, with the head resting on the middle of the seat over the transmission hump. Pet. App. A5, A18-A19, A33; Trial Tr. 371-444.
As petitioner and Esposito drove away from the murder site in Fiorito's car, petitioner expressed concern that Fiorito was not dead and shot him a second time. Petitioner instructed Esposito to park Fiorito's car in a residential area in Queens, and to pull Fiorito's body across the back seat so that Fiorito would appear to be asleep. Petitioner and Esposito then got into a car driven by Joseph Kern, another of petitioner's associates. At petitioner's direction, Kern and another man returned to the vacant house later that day and recovered the shell casing from the first shot. Esposito confessed to his girlfriend on returning home that "they" had just killed Fiorito. Pet. App. A19, A33; Trial Tr. 431-444.
In the early morning of September 21, 1990, a New York City Police Officer discovered Fiorito's body in his parked car. When the police removed the body, they discovered a spent shell casing on the floor behind the passenger seat, but did not find a bullet. Later that same day, the police, alerted by Esposito's girlfriend, arrested Esposito outside his home and recovered his bloodstained clothing from a dumpster. Esposito agreed to cooperate with police and implicated petitioner in the crime. Subsequently, the medical examiner determined that the bullet from the initial shot fired in the house remained lodged in Fiorito's brain in fragments and that the bullet from the second shot fired in the car had passed through and exited his head. Pet. App. A19, A33; Trial Tr. 438-448, 1529-1532, 1586.
2. On September 6, 1993, the day before opening statements in petitioner's trial, the new owner of Fiorito's car notified police investigators that, while removing the carpet and insulation in the rear of the car, he had discovered a bullet, as well as a hole in the carpet near the transmission hump. Government prosecutors were notified of the development the following day, September 7, 1993. They informed petitioner's counsel of the bullet's discovery by a faxed letter on Saturday, September 11, 1993, and explained that a ballistics laboratory was conducting testing on the bullet. J.A. 176-177. In the meantime, defense counsel had delivered an opening statement arguing that Esposito had murdered Fiorito alone and that no scientific evidence linked petitioner to the crime. The government made no reference to the bullet in its opening statement. Pet. App. A5, A29, A34; J.A. 42- 43, 50-55; Trial Tr. 39-42.1
When trial resumed on Monday, September 13, 1993, defense counsel did not request a continuance or otherwise raise any issue concerning the bullet. Defense counsel first broached the subject of the bullet with the court ten days later, on September 23, 1993, when counsel sought to expedite the government's ballistics tests. The government advised the court that, because forensic reports were not yet available, it had not decided whether to introduce the bullet into evidence. Before trial resumed on Monday, September 27, 1993, the government informed defense counsel that the ballistics expert had concluded that the bullet matched the shell fragments found in Fiorito's head and had been fired from the same gun. At defense counsel's request, the government made the bullet and shell fragments available for examination by a defense expert. Pet. App. A34; J.A. 24-25, 178-179.
On the following day, September 28, 1993, defense counsel moved for a mistrial, or, in the alternative, to preclude the admission of the bullet into evidence, on the basis of the government's belated disclosure of the bullet. J.A. 19, 25; Pet. App. A35. In connection with that motion, counsel claimed that the new evidence severely undermined the defense theory chosen in advance of trial. That strategy, which entailed arguing that petitioner was not in the car and that Esposito's account of the murder was not credible, was based in part on the absence of any blood spatter or scientific evidence indicating that a shot had been fired in Fiorito's car. Defense counsel acknowledged that a shell casing had been found in the car when Fioroto's body was first discovered, but argued that, notwithstanding the shell casing, the absence of a bullet in the car was critical to the defense case. J.A. 16-24.
The district court found that the circumstances did not justify a mistrial. J.A. 24-26, 30. The court also concluded that the bullet could be admitted into evidence on the basis that the government's delay had not irreparably prejudiced petitioner's defense. Pet. App. A20, A35; J.A. 73-108. The court explained that the bullet did not physically link petitioner to Fiorito's murder, and therefore did not compromise the defense's argument that no scientific evidence placed petitioner in the car or connected him to the crime. Instead, the bullet only tended to corroborate Esposito's account of the events. J.A. 81-83.
At the same time, the district court repeatedly offered defense counsel a continuance to examine the bullet and to adjust his defense strategy if necessary, and ruled that petitioner could recall any of the witnesses who had previously testified. Pet. App. A14, A35; J.A. 20, 24, 55-56, 89-101, 107-108. Defense counsel declined the offer of a continuance, explaining that a continuance was not necessary for the defense ballistics expert to examine the bullet.2 Counsel also argued that there would be no use in obtaining a continuance because the bullet had irreparably damaged his plans to argue that there had been no shooting in the car. J.A. 20-21, 24-25. Defense counsel did elect, however, to recall one of the crime scene investigators. J.A. 107-109.
3. On appeal, petitioner retained new counsel, who argued, inter alia, that the district court had erred in admitting the bullet into evidence. The court of appeals affirmed. Pet. App. A30-A37.
The court of appeals, like the district court, concluded that the government's belated disclosure of the bullet's discovery, while "regrettable," did "not irreparably damage [petitioner's] strategy." Pet. App. A36. In particular, the court observed, "[n]othing in [petitioner's] opening statement * * * was rendered false by the bullet evidence" and "the defense had ample other detrimental information with which to attack Esposito's credibility." Ibid. The court added that its conclusion was supported by trial counsel's decision to decline a continuance, explaining that petitioner's "decision to proceed with his strategy despite his knowledge that the bullet might be matched to existing shell fragments weakens his claim" of "substantial prejudice." Id. at A37. Petitioner's appellate counsel did not raise a claim on appeal that trial counsel had rendered ineffective assistance by forgoing a continuance after learning about the bullet.
4. On April 24, 1997, petitioner filed a motion under 28 U.S.C. 2255 seeking to vacate his conviction. Pet. App. A11. The motion asserted, inter alia, that trial counsel's failure to accept the district court's offer of a continuance amounted to ineffective assistance of counsel. Petitioner argued that, if trial counsel had accepted a continuance, he could have mounted a more persuasive challenge to the bullet evidence and to Esposito's testimony about the murder. Pet. App. A6, A20. In connection with the motion, petitioner submitted the opinions of three newly-retained experts who attempted to cast doubt on the ballistics evidence and on Esposito's account of Fiorito's murder. J.A. 143-175. The expert opinions argued generally that Esposito's account of the murder was suspect, but none of the opinions challenged the conclusion of both the government's and petitioner's ballistics experts that the bullet matched the shell fragments recovered from Fiorito's head.
The district court denied petitioner's Section 2255 motion. Pet. App. A16-A26. Relying on Billy-Eko v. United States, 8 F.3d 111 (2d Cir. 1993), the court found petitioner's claim that trial counsel was ineffective for failing to seek a continuance to be procedurally barred because the claim could have been raised on direct appeal and petitioner could not establish cause or prejudice for his default. Pet. App. A24-A25. The district court granted petitioner a certificate of appealability limited to that issue. Id. at A27.3
5. The court of appeals affirmed. Pet. App. A3-A10. The court explained that, under its decision in Billy Eko, supra, the failure of a defendant to raise a claim of ineffective assistance of counsel on direct appeal constitutes a procedural default if new counsel represents the defendant on appeal and if the claim is based solely on the record already developed at trial. Pet. App. A7. Applying those standards to this case, the court concluded that petitioner could have raised on direct appeal his claim that trial counsel rendered ineffective assistance in failing to accept a continuance.
The court first observed that new counsel represented petitioner on direct appeal. Next, it determined that petitioner's claim of ineffective assistance was "based solely on the record adduced at trial and available to [his] appellate counsel on direct appeal." Pet. App. A7. The court explained in that regard that petitioner's trial "counsel engaged in many discussions about the decision not to accept a continuance" in the trial record. Ibid. The court added that, although "a new set of experts allegedly reached different conclusions after trial" about the ballistics evidence, that did "not render the basis for [petitioner's] ineffective assistance claim outside the scope of the record adduced at trial." Ibid. "Moreover," the court explained, "each of the issues raised by [petitioner's] most recent forensic analysis-the absence of blood in the vehicle, the absence of blood 'spatter' or other remnants, the belated discovery of the bullet after several searches- all were extensively addressed at trial." Id. at A7-A8. As a result, the court concluded, "the record available to [petitioner's] appellate counsel fully revealed the implications of the failure to accept a continuance." Id. at A8. Because petitioner was unable to demonstrate cause for having failed to raise his claim on direct appeal, the court found his claim of ineffective assistance of trial counsel to be procedurally barred. Ibid.
SUMMARY OF ARGUMENT
I. A basic principle of procedural default is that the failure to raise a claim on direct appeal bars assertion of the claim on collateral review absent a showing of cause
and prejudice. Petitioner's principal argument is that procedural default rules should have no application to claims asserting ineffective assistance of trial counsel. That argument rests on the recognition that most ineffectiveness claims cannot fairly be expected to be raised and addressed on direct appeal, either because the factual basis for a claim is not fully developed in the trial record, or because, when trial counsel continues to represent the defendant on appeal, it is unreasonable to expect counsel to argue his own ineffectiveness. Petitioner's bright-line approach would be easy to administer, and would avoid the need to apply the cause-and-prejudice standard to identify those ineffectiveness claims that can fairly be expected to be raised and resolved on direct appeal.
While those considerations have force, the better approach on balance is to apply the procedural default rule to ineffective assistance claims. The fact that many ineffective assistance claims cannot fairly be raised and resolved on direct appeal does not mean that no ineffectiveness claim should be required to be asserted on direct appeal. When the factual basis for a claim is fully developed in the record on appeal and new counsel represents the defendant on appeal, it is reasonable to require the claim to be raised at that time. And when those conditions are not present, the defendant would establish "cause" for not asserting the claim on appeal. Channeling ineffective assistance claims to direct appeal rather than collateral review in appropriate situations serves the general societal interests in respecting the finality of criminal judgments and encouraging resolution of legal challenges to convictions at the earliest feasible opportunity.
The Second Circuit and Seventh Circuit are thus correct to apply the procedural default rule to ineffective assistance claims, and to refuse to entertain an ineffectiveness claim on collateral review when new counsel represented the defendant on appeal and the claim is based solely on the record developed at trial. Administering that rule does not impose undue burdens on courts and litigants, and the rule does not work unfairness against defendants. When a defendant fears that his ineffectiveness claim may be procedurally defaulted if he waits to assert it in a motion under 28 U.S.C. 2255, he can raise the claim on direct appeal while explaining to the court that further factual development would advance the claim. In that situation, the court of appeals would resolve the claim if it is able to rule definitively in one direction or the other, or, if the claim cannot be resolved, would decline to decide the claim and allow the defendant to raise it on collateral review. In either case, the defendant will not have been unfairly prejudiced.
II. The court of appeals correctly ruled that petitioner procedurally defaulted his ineffective assistance claim by failing to raise it on direct review. Petitioner's claim alleges that trial counsel was ineffective in failing to accept a continuance upon learning of the discovery of the bullet in the car in which Fiorito's body was found. Petitioner was represented by new counsel on appeal, and trial counsel's reasons for declining to accept a continuance are fully developed in the trial record. Contrary to petitioner's argument, moreover, the expert affidavits filed in connection with his Section 2255 motion do not constitute extrinsic evidence that bears on the prejudice prong of his ineffectiveness claim. Each of the arguments raised in the affidavits was made at trial. The affidavits therefore do not advance petitioner's argument on the prejudice allegedly caused by trial counsel's failure to accept a continuance. Accordingly, petitioner cannot establish "cause" for failing to raise his ineffectiveness claim on direct appeal.
Even if petitioner could establish "cause," this Court should affirm the judgment on the alternate ground that petitioner cannot establish "prejudice." The prejudice inquiry for cause-and-prejudice purposes is coextensive with the prejudice prong of the underlying ineffectiveness claim. Petitioner cannot establish that, if trial counsel had accepted a continuance, there is a reasonable probability that the result at trial would have been different. None of the theories raised in petitioner's expert affidavits challenges the ballistics tests establishing that the bullet matches shell fragments found in Fiorito's head. The affidavits thus offer no explanation for how the bullet ended up in the car other than the one proved by the government at trial.
I. CLAIMS ASSERTING INEFFECTIVE ASSISTANCE OF TRIAL COUNSEL SHOULD BE SUBJECT TO THE GENERAL RULE OF PROCEDURAL DEFAULT THAT, WHEN A CLAIM IS NOT RAISED ON DIRECT APPEAL, IT CAN BE CONSIDERED ON COLLATERAL REVIEW ONLY UPON A SHOWING OF CAUSE AND PREJUDICE
A basic principle of procedural default is that the failure to raise a claim on direct appeal bars its assertion on collateral review absent a demonstration of cause and prejudice. See, e.g., Bousley v. United States, 523 U.S. 614, 621-622 (1998). The issue in this case is whether claims alleging ineffective assistance of trial counsel-as a class-should be excepted from the procedural default rule. While many ineffective assistance claims cannot be properly litigated on direct appeal, others can be. The normal rules of procedural default thus should be applied in this context, in order to promote respect for the finality of criminal judgments and to encourage resolution of legal challenges at the earliest feasible opportunity.
A. A Claim That Is Not Raised On Direct Appeal Generally Cannot Be Asserted On Collateral Review Absent A Showing Of Cause And Prejudice
Section 2255 "is an extraordinary remedy and 'will not be allowed to do service for an appeal.'" Bousley, 523 U.S. at 621 (quoting Reed v. Farley, 512 U.S. 339, 354 (1994)); see United States v. Frady, 456 U.S. 152, 165 (1982). Accordingly, the failure of a defendant to raise a claim on direct appeal generally constitutes a procedural default that bars consideration of the claim on collateral review. Bousley, 523 U.S. at 621; Reed, 512 U.S. at 354. In particular, "[w]here a defendant has procedurally defaulted a claim by failing to raise it on direct review, the claim may be raised [under Section 2255] only if the defendant can first demonstrate * * * 'cause' and actual 'prejudice.'" Bousley, 523 U.S. at 622; see Reed, 512 U.S. at 354 ("Where the petitioner- whether a state or federal prisoner-failed properly to raise his claim on direct review, the writ is available only if the petitioner establishes 'cause' for the waiver and shows 'actual prejudice resulting from the alleged . . . violation.'") (quoting Wainwright v. Sykes, 433 U.S. 72, 84 (1977)); Frady, 456 U.S. at 167-168.4
The strict limitations on collateral review of claims not presented on direct appeal reflect the basic societal interest in respecting final criminal judgments. See Frady, 456 U.S. at 164 (recognizing the "legitimate interest in the finality of [a] judgment * * * perfected by the expiration of the time allowed for direct review or by the affirmance of the conviction on appeal"); United States v. Timmreck, 441 U.S. 780, 784 (1979) (discussing "the concern with finality served by the limitation on collateral attack"). "Once the defendant's chance to appeal has been waived or exhausted," this Court has explained, "we are entitled to presume he stands fairly and finally convicted, especially when * * * he already has had a fair opportunity to present his federal claims to a federal forum." Frady, 456 U.S. at 164. A defendant's "fair opportunity to present his federal claim to a federal forum" includes his direct appeal, ibid., and his "[f]ailure to raise a claim on appeal reduces the finality of appellate proceedings" and "deprives the appellate court of an opportunity to review trial error," Murray v. Carrier, 477 U.S. 478, 491 (1986). Raising the claim on collateral review in that situation therefore requires a showing of cause and prejudice. Bousley, 523 U.S. at 621; Reed, 512 U.S. at 354.5
B. Claims Asserting Ineffective Assistance Of Counsel Should Not Be Exempt From The General Rule Of Procedural Default
Petitioner argues (Br. 21-29) that claims asserting ineffective assistance of trial counsel should be categorically excepted from the procedural default rule, and thus should be heard on collateral review even if they could have been raised and resolved on direct appeal, without any need to demonstrate cause and prejudice. A bright-line rule to that effect would have certain advantages, and the government advocated that approach some time ago. See United States v. Cronic, 466 U.S. 648, 667 n.42 (1984); see also Billy-Eko v. United States, 509 U.S. 901 (1993); Diaz-Albertini v. United States, 498 U.S. 1061 (1991); Chappell v. United States, 494 U.S. 1075 (1990). That approach is straightforward to administer, and avoids the expenditure of judicial resources on applying the cause-and-prejudice standard to ineffectiveness claims, with the goal of identifying that set of claims that could fairly be raised and resolved on direct appeal.
On balance, however, adhering to the procedural default rule in the context of ineffective assistance claims is more in keeping with this Court's general application of procedural default principles. It is also more consistent with the attendant societal interests in according respect to final criminal judgments and encouraging resolution of claims at the earliest feasible opportunity. Advancement of those interests would not come at the cost of unfairness, because a defendant would be required to raise an ineffectiveness claim on direct appeal only if he were represented by new counsel and the factual basis for the claim were fully developed in the record. The experience of the Second and Seventh Circuits and of the many States that apply the procedural default rule to ineffective assistance claims shows that the rule's application in that context is not unworkable and does not impose unwarranted burdens on judicial administration. For those reasons, claims alleging ineffective assistance of counsel should not be exempt from the general rule that, if a claim was not raised on direct appeal, its consideration on collateral review is conditioned upon a showing of cause and prejudice.6
1. Ineffective assistance claims often cannot be fairly raised or fully addressed on direct appeal
a. Petitioner's argument for a blanket exception to the procedural default rule for ineffective assistance claims begins with the recognition that many ineffective assistance claims cannot fairly be expected to be raised and resolved on direct appeal. Pet. Br. 22-23. That is the case for two separate reasons.
First, if the same lawyer represents the defendant at trial and on appeal, it is unrealistic to expect that lawyer to argue on appeal that his own performance at trial was ineffective. See Kimmelman v. Morrison, 477 U.S. 365, 378 (1986) ("Indeed, an accused will often not realize that he has a meritorious ineffectiveness claim until he begins collateral review proceedings, particularly if he retained trial counsel on direct appeal.") (emphasis added); Billy-Eko v. United States, 8 F.3d 111, 114 (2d Cir. 1993) ("[E]ven the scrupulous attorney searching the record in good faith would likely be blind to his derelictions at the trial level."); cf. Wood v. Georgia, 450 U.S. 261, 265 n.5 (1981) ("The party who argued the appeal and prepared the petition for certiorari was the lawyer on whom the conflict-of-interest charge focused. It is unlikely that he would concede that he had continued improperly to act as counsel."); Evitts v. Lucey, 469 U.S. 387, 391 n.3 (1985) (noting that counsel was referred to bar "disciplinary proceedings for 'attacking his own work'").
Second, even if new counsel represents a defendant on direct appeal, the grounds for an ineffective assistance claim often either are not apparent from the trial record or are not sufficiently developed to permit meaningful consideration of the claim on appeal. As the Second Circuit has explained:
Ineffective assistance claims are often based on assertions that trial counsel made errors of omission, errors that are difficult to perceive from the record: for example, neglecting to call certain witnesses or introduce certain evidence. The claims might also be based on a conflict of interest not apparent at trial. Proof is sometimes provided in attorney-client correspondence, or in other documents not introduced at trial. Even if a new attorney represents the accused on direct appeal, she might not come across reasons to suspect ineffective assistance in preparing a direct appeal.
Billy-Eko, 8 F.3d at 114 (citations omitted); see McCleese v. United States, 75 F.3d 1174, 1178 (7th Cir. 1996). As a result, resolution of most ineffective assistance claims requires additional proceedings in the district court on collateral review, at which the defendant can present evidence about counsel's allegedly deficient performance and its effect on the verdict, and trial counsel can shed added light on his or her representation. See Kimmelman, 477 U.S. at 378-379 n.3.
b. For those reasons, the courts of appeals agree that ineffectiveness claims frequently should be addressed in the first instance by the district courts on collateral review.7 At the same time, because an ineffectiveness claim can be fully and fairly resolved on direct appeal in certain situations, and because defendants desire to have the claim addressed on direct appeal in some circumstances, all courts of appeals allow the assertion of an ineffectiveness claim on direct appeal. See Guinan v. United States, 6 F.3d 468, 472 (7th Cir. 1993). The courts of appeals disagree on whether, with respect to those ineffectiveness claims that could be raised and addressed on direct appeal, failure to present the claim on appeal is subject to the general rule of procedural default barring assertion of the claim under Section 2255 absent a showing of cause and prejudice.
Some courts of appeals, in accord with petitioner's position in this case, hold that the failure to raise an ineffective assistance claim on direct appeal can never constitute a procedural default subject to the "cause and prejudice" limitation. Those courts permit defendants to raise a claim of ineffective assistance of counsel for the first time on collateral attack without showing cause and prejudice, even if the claim could have been fully considered and resolved on direct appeal. See United States v. Galloway, 56 F.3d 1239, 1242-1243 (10th Cir. 1995) (en banc); United States v. DeRewal, 10 F.3d 100, 103-105 (3d Cir. 1993), cert. denied, 511 U.S. 1033 (1994); see also Knight v. United States, 37 F.3d 769, 774 (1st Cir. 1994) ("[T]he failure to bring a claim of ineffective assistance of counsel on direct appeal is not subject to the cause and prejudice standard.").
By contrast, the Second and Seventh Circuits, in agreement with the government's position in this case, do not categorically exempt ineffectiveness claims from the procedural default rule. Those courts recognize that, "in most cases there is good reason to allow a defendant to make ineffective assistance claims on collateral attack even if those claims were not brought on direct appeal." Billy-Eko, 8 F.3d at 114 (2d Cir.); McCleese 75 F.3d at 1178 (7th Cir.); see Guinan, 6 F.3d at 471 (7th Cir.). Those courts hold, however, that where a defendant is represented by new counsel on appeal and his ineffectiveness claim is grounded solely in the trial record, failure to raise the claim on direct appeal is a procedural default requiring a showing of cause and prejudice to bring the claim on collateral review. McCleese, 75 F.3d at 1178; Billy-Eko, 8 F.3d at 115-116. The approach of the Second and Seventh Circuits, in our view, is sound.8
2. Applying the procedural default rule appropriately targets those ineffective assistance claims that could fairly be raised and resolved on direct appeal
Application of the procedural default rule to ineffective assistance claims bars collateral review only of those ineffectiveness claims that could reasonably be expected to be raised on direct appeal. First, if the defendant was represented by the same attorney at trial and on direct appeal, he would establish "cause" for not having raised an ineffectiveness claim on appeal. See Prewitt v. United States, 83 F.3d 812, 816 (7th Cir. 1996); Billy-Eko, 8 F.3d at 115. In addition, if the factual basis for an ineffectiveness claim were not fully developed in the record on direct appeal, there also would be "cause" for failing to assert the claim on appeal. See, e.g., Amiel v. United States, 209 F.3d 195, 198 (2d Cir. 2000) (holding that an ineffective assistance claim could be asserted on collateral review because it was based on "events outside the trial record, namely, counsel's alleged off- the-record statements and alleged disloyal[ty]" to the defendant); see also Bond v. United States, 1 F.3d 631, 636-637 (7th Cir. 1993).9
If "cause" were established on either of those grounds, there would be no bar against collateral review on the merits. That is because the remaining requirement to show "prejudice" is coextensive with the "prejudice" prong of the underlying claim of ineffective assistance. This Court has described the two inquiries in precisely equivalent terms. Compare Strickler v. Greene, 527 U.S. 263, 289 (1999) (prejudice prong of cause-and-prejudice test entails showing that, absent the alleged error, "'there is a reasonable probability' that the result of the trial would have been different") (quoting Kyles v. Whitley, 514 U.S. 419, 434 (1995)), with Strickland v. Washington, 466 U.S. 668, 694 (1984) (prejudice prong of ineffective assistance test requires demonstrating "that there is a reasonable probability that, but for the counsel's unprofessional errors, the result of the proceeding would have been different"). In Strickler v. Greene, supra, in fact, the Court collapsed its analysis of prejudice for "cause and prejudice" purposes into its analysis of materiality on the underlying claim under Brady v. Maryland, 373 U.S. 83 (1963), see 527 U.S. at 282, 289, 296, and the Brady test for materiality mirrors the Strickland test for prejudice, see Strickland, 466 U.S. at 694 ("appropriate test for prejudice finds its roots in the test for materiality of exculpatory information not disclosed to the defense by the prosecution"). As a result, once a defendant establishes "cause" for failing to raise an ineffectiveness claim on direct appeal, the inquiry effectively shifts to the merits of the substantive claim.10
3. Applying the procedural default rule to ineffective assistance claims promotes important societal interests
a. By channeling the resolution of ineffective assistance claims to direct appeal instead of collateral review in appropriate situations-where new counsel represents the defendant on appeal and the claim is fully developed in the trial record-the procedural default rule advances the basic societal interest in respecting the finality of criminal judgments. See Carrier, 477 U.S. at 491; Frady, 456 U.S. at 164-165 ("Our trial and appellate procedures are not so unreliable that we may not afford their completed operation any binding effect beyond the next in a series of endless postconviction collateral attacks. To the contrary, a final judgment commands respect.").
In certain cases, the factual basis for a claim of ineffective assistance is sufficiently developed in the trial record to permit adjudication of the claim on direct appeal. See Strickland, 466 U.S. at 697 (assuming resolution of "ineffectiveness claims * * * on direct appeal"); see also Kimmelman, 477 U.S. at 380 (discussing "ineffective-assistance claim asserted on direct review"). In such a case, the record on direct appeal might conclusively demonstrate a defendant's entitlement to relief. A number of decisions have granted relief on direct appeal on the ground that trial counsel rendered ineffective assistance.11
Conversely, the record might establish that the ineffectiveness claim could never prevail. That might be the case because, as a matter of law, the attorney's performance could not have prejudiced the defendant. E.g., United States v. Finley, 245 F.3d 199, 204 (2d Cir. 2001) (failure to move for judgment of acquittal could not have affected trial outcome because district court, as a matter of law, could not have granted the motion), cert. denied, 534 U.S. 1144 (2002); McCleese, 75 F.3d at 1179 (failure to advise defendant of mandatory minimum sentence could not have affected decision to plead guilty because record establishes defendant's belief that minimum sentence was greater than actual statutory minimum). A defendant also might be unable to prevail because, as a matter of law, counsel's performance could not be found deficient. E.g., United States v. Brooks, 125 F.3d 484, 496 (7th Cir. 1997) (record on appeal establishes that performance was not unreasonable because counsel explained that decision not to recall witness was based on strategic assessment that potential damage from testimony outweighed possible benefits).
In either situation-whether relief is granted or denied-resolving the claim on direct appeal serves the interest of finality by averting the need for post-conviction proceedings to adjudicate the claim. See United States v. Timmreck, 441 U.S. 780, 784 (1979) ("Every inroad on the concept of finality * * * by increasing the volume of judicial work, inevitably delays and impairs the orderly administration of justice.") (quoting United States v. Smith, 440 F.2d 521, 528-529 (7th Cir. 1971)). Rules requiring the timely presentation of claims that are ripe for review serve those finality interests.
Procedural default principles advance the interest in finality not only by encouraging resolution of certain claims on direct appeal, but also by accelerating disposition of claims on collateral review. The rule requires dismissal of some ineffectiveness claims brought under Section 2255 in advance of a merits inquiry. If a claim raised on collateral review is "based solely on the record developed at trial," Billy-Eko, 8 F.3d at 115; see McCleese, 75 F.3d at 1178, denial of the claim at the outset of the proceedings because of a failure to establish "cause" pretermits adjudication of a legal challenge that otherwise could unsettle the final judgment. Moreover, the procedural default rule, by the fact of its existence, tends to dissuade the filing of ineffectiveness claims that are susceptible to dismissal because they are grounded solely in the trial record. Those effects on ineffectiveness claims both on direct appeal and on collateral review advance the finality interest in "minimizing the number of collateral proceedings so that the direct appeal will be the main bout and not just the warm-up." Guinan, 6 F.3d at 472.12
For substantially similar reasons, procedural default principles promote efficiency in the administration of criminal proceedings by encouraging resolution of legal challenges at the earliest feasible opportunity. See Guinan, 6 F.3d at 472. In the case of a meritorious claim of ineffective assistance, for instance, the award of relief "on appeal, rather than on postconviction review," would afford "the opportunity to resolve the issue shortly after trial, while evidence is still available both to assess the * * * claim and to retry the defendant effectively if he prevails in his appeal." Reed v. Ross 468 U.S. 1, 10 (1984). In the case of a non-meritorious ineffective assistance claim, it may be more efficient to resolve the claim on direct appeal when the case may be before the court of appeals on other issues, rather than through the institution of a round of collateral review perhaps followed by another appeal.
Requiring assertion on direct appeal of certain ineffectiveness claims also "forc[es] the defendant to litigate all of his claims together" on appeal, "as quickly after trial as the docket will allow, and while the attention of the appellate court is focused on the case." Ross, 468 U.S. at 10. If an ineffective assistance claim is closely connected with other claims raised on direct appeal, efficiency considerations weigh in favor of addressing all the related claims in one proceeding. The procedural default rule also streamlines the administration of claims at the collateral review stage by affording the district courts a basis for denying relief at the outset of Section 2255 proceedings based on the failure to establish "cause"-without the need to entertain the claim on the merits, and often, without the need to conduct an evidentiary hearing. For all of those reasons, the important objectives advanced by the procedural default rule in other contexts are served by enforcing the rule against ineffective assistance claims.
b. Petitioner's amicus, the National Association of Criminal Defense Lawyers (NACDL), contends that the values served by procedural default principles are not implicated when a defendant fails to raise an ineffectiveness claim on appeal, because, according to NACDL, procedural default principles exist only to ensure the making of contemporaneous objections at trial. NACDL Br. 11-13. That argument lacks merit. See Carrier, 477 U.S. at 490 (finding "unpersuasive" the contention "that the concerns that underlie the cause and prejudice test are not present in the case of defaults on appeal"). Even if it were the case, as NACDL erroneously presumes (Br. 12-13), that procedural default rules exist solely to discourage defense counsel from withholding an objection for tactical reasons, see, e.g., Wainwright v. Sykes, 433 U.S. 72, 89 (1977), that concern arises on appeal as well as at trial: "the possibility of 'sandbagging'" does not "vanish once a trial has ended in conviction, since appellate counsel might well conclude that the best strategy is to select a few promising claims for airing on appeal, while reserving others for federal habeas review should the appeal be unsuccessful." Carrier, 477 U.S. at 492; see Ross, 468 U.S. at 14 ("In general, therefore, defense counsel may not make a tactical decision to forgo a procedural opportunity-for instance, an opportunity to object at trial or to raise an issue on appeal.") (emphasis added).
In any event, the procedural default rule and the attendant cause-and-prejudice standard are grounded in broader societal interests in according respect to the finality of criminal proceedings and encouraging the prompt resolution of legal challenges to criminal judgments. Those important values, for the reasons explained, are advanced by applying the procedural default rule to claims asserting ineffective assistance of counsel.
4. The procedural default rule is not unworkable, unfair, or unduly burdensome when applied to ineffective assistance claims
The lynchpin of petitioner's argument for a categorical exception to the procedural default rule is that, in the particular context of ineffective assistance claims, enforcement of the procedural default rule and the "cause and prejudice" standard entails an excessive burden on judicial administration and works unfairness against defendants. Pet. Br. 25-29. The procedural default rule, however, is not unworkable or unfair when applied to ineffective assistance claims.
a. Petitioner, joined by NACDL, argues that applying procedural default principles to ineffective assistance claims imposes an unwarranted burden on defendants and the courts. Petitioner and NACDL argue (Pet. Br. 17; NACDL Br. 23), for instance, that the imperative to avoid a procedural default imposes excessive burdens on newly appointed appellate counsel to scour the trial record in search of ineffective assistance claims. Any such obligation, however, does not seem appreciably out of step with counsel's basic responsibility in all cases to examine the trial record in order to identify potential legal claims to raise on appeal. Moreover, because the prospect of procedural default arises only if an ineffectiveness claim brought on collateral review is based solely on the record that was available on appeal, Billy-Eko, 8 F.3d at 115; McCleese, 75 F.3d at 1178, counsel need only identify those ineffectiveness claims that are fully apparent and developed in the record.
Petitioner and NACDL also assert (Pet. Br. 22; NACDL Br. 22) that the procedural default rule burdens the courts of appeals by saddling them with ineffective assistance claims raised prematurely on direct appeal. But no court of appeals prohibits assertion of an ineffectiveness claim on direct appeal. Accordingly, all circuits-not just those that apply the procedural default rule-must deal with ineffectiveness claims raised on direct appeal and determine whether to resolve the claims or remit them to collateral review.13 The government's experience in the Second and Seventh Circuits does not support the suggestion (Pet. Br. 22; NACDL Br. 22) that, in those courts that apply the procedural default rule to ineffective assistance claims, defendants will inundate the courts with ineffectiveness claims to protect against the possibility of a default. Only those ineffectiveness claims that are fully developed in the trial record are subject to the procedural bar; and to the extent that the effect is to cause defendants to raise fully developed claims on direct appeal, that is the object of the rule.
Even when defendants raise claims on appeal that are not ripe for resolution and require further factual development, the assertion of the claim on appeal does not compel its resolution by the court of appeals. Instead, courts of appeals have flexibility to direct consideration of the claim in the most appropriate and efficient manner-by addressing the claim on the merits or by remitting the defendant to collateral review under Section 2255 for initial consideration of the claim with other post-conviction claims.14 That framework is not at odds with the normal appellate function (Pet. Br. 16; NACDL Br. 16-17), because it contemplates resolution on appeal only of claims for which the trial record is adequately developed.
Finally, petitioner suggests (Br. 28-29) that district courts will be burdened by having to apply the cause-and-prejudice standard when an ineffectiveness claim is raised on collateral review. It is not clear, however, why applying the cause-and-prejudice standard exacts more of a toll on district courts in the context of ineffective assistance claims than in the context of all other claims. In fact, because, with ineffective assistance claims, the analysis of prejudice merges with the prejudice prong of the underlying claim on the merits, see pp. 21-22, supra, any added burden on district courts in that context flows solely from the "cause" requirement. And even on the question of "cause," there would be no need for extensive examination of the issue when it is apparent that cause exists for having failed to raise an ineffectiveness claim on direct appeal-either because trial counsel represented the defendant on appeal or because the claim is based on extrinsic evidence outside the trial record.15
b. Petitioner and NACDL also contend that the procedural default rule is unpredictable when applied to ineffectiveness claims and thus results in unfairness to defendants. According to petitioner and NACDL (Pet. Br. 16, 25; NACDL Br. 20-21), the circumstances in which an ineffectiveness claim is required to be raised on direct appeal are not clear. As a result, they assert (Pet. Br. 27; see NACDL Br. 20-21), a defendant represented by new counsel, when deciding whether to raise an ineffective assistance claim on appeal, is put to the unfair choice between, on one hand, presenting the claim and risking its resolution against him based on an inadequate factual record, and, on the other hand, forgoing assertion of the claim and risking procedural default when the claim is raised on collateral review. The dilemma outlined by petitioner and NACDL, however, should not often arise in practice.
First, contrary to the premise of their argument, the standard for determining when a claim is sufficiently developed on the trial record to require its assertion on direct appeal is not unduly ambiguous or unworkable. The Second Circuit and the Seventh Circuit apply essentially the same test: there is no "cause" for having failed to raise the claim on direct appeal if the claim, when presented on collateral review, "is based solely on the record developed at trial." Billy-Eko, 8 F.3d at 115; McCleese, 75 F.3d at 1178 (absence of cause if "no extrinsic evidence" offered in support of claim). That standard is straightforward for district courts to apply in Section 2255 proceedings: the question is whether the claim relies on new evidence outside the record that was available on appeal. If it does not-if the claim is grounded solely on the record that was available on appeal-then the same claim based on the same facts could have been raised on direct appeal, and there is no "cause" for having failed to assert it at that time (at least if the defendant was represented by new counsel on appeal).16
At the time of the direct appeal, moreover, the standard applied by the Second and Seventh Circuits confronts a defendant deciding whether to raise an ineffectiveness claim with a fairly straightforward question: whether the claim is likely to be based on extrinsic evidence if it is later brought on collateral review. If the claim requires development of extra-record evidence for its resolution, it can safely be held over for proceedings under Section 2255. That will often be the case. For instance, a claim based on an alleged conflict of interest frequently turns on extrinsic facts concerning counsel's relationships with third parties or extra-record communications with the client. See, e.g., Amiel, 209 F.3d at 198; United States v. Stoia, 22 F.3d 766, 768 (7th Cir. 1994); but see United States v. Stantini, 85 F.3d 9, 20 (2d Cir.) (rejecting claim based on asserted conflict of interest because clear from trial record that conflict could not have affected defense), cert. denied, 519 U.S. 1000 (1996). A claim based on counsel's lack of preparation also often requires development of extrinsic evidence, including counsel's testimony concerning preparation of the defense. See, e.g., Abbamonte v. United States, 160 F.3d 922, 925 (2d Cir. 1998); Billy-Eko, 8 F.3d at 116. Likewise, allegations based on counsel's failure to take certain actions-e.g., to call a particular witness, or to investigate a particular issue -sometimes require development of evidence on counsel's reasons for the decision or on the nature of the testimony or evidence at issue. See, e.g., Ellerby v. United States, 187 F.3d 257, 259-260 (2d Cir. 1998) (per curiam); United States v. Downs, 123 F.3d 637, 641-642 (7th Cir 1997); United States v. Workman, 80 F.3d 688, 700-701 (2d Cir.), cert. denied, 519 U.S. 938 and 955 (1996); but see Finley, 245 F.3d at 204 (prejudice could not be proved as a matter of law); United States v. Ben Zvi, 242 F.3d 89, 96-98 (2d Cir. 2001) (same); McCleese, 75 F.3d at 1178-1179 (same).
In contrast, if the factual basis for the claim of ineffective assistance is developed in the record on appeal and the defendant has concerns about whether the claim will be based on additional extra-record evidence if it were raised on collateral review, he can protect against the possibility of procedural default by asserting the claim on appeal. The Second Circuit, to that effect, has urged "new appellate counsel * * * to err on the side of inclusion on direct appeal." Billy-Eko, 8 F.3d at 116; see ibid. (suggesting that counsel should "[i]nclud[e] the claim, complete with the attorney's assertion that further factual development is necessary"). In that situation, the court could decide the claim if the record conclusively established either an entitlement to relief or that the claim cannot prevail. See pp. 23-24, supra. If the court is unable to resolve the claim definitively in either direction, it should decline to address it and permit development of the record on collateral review. See, e.g., United States v. Williams, 205 F.3d 23, 35-36 (2d Cir.), cert. denied, 513 U.S. 885 (2000); Workman, 80 F.3d at 700-701. Neither outcome would prejudice the defendant: resolution of the claim by the court of appeals, by assumption, would give the defendant the same relief he would have obtained had the claim been pressed on collateral review; and the court's declining to address the claim would leave the defendant free to pursue the claim under Section 2255. For those reasons, insofar as there is some ambiguity concerning when a claim is sufficiently developed in the record to require its assertion on appeal, the rule nonetheless should not work unfairness against a defendant in practice.17
c. Finally, the experience of state courts suggests that application of the procedural default rule to ineffective assistance claims is not unworkable and does not entail an unwarranted burden on judicial administration. The procedural rules applied in the States serve the same objectives as their federal counterparts. See Ross, 468 U.S. at 10. ("The criminal justice system in each of the 50 States is structured both to determine the guilt or innocence of defendants and to resolve all questions incident to that determination, including the constitutionality of the procedures leading up to the verdict. Each State's complement of procedural rules facilitates this complex process, channeling, to the extent possible, the resolution of various types of questions to the stage of the judicial process at which they can be resolved most fairly and efficiently.").
It therefore is significant that a substantial number of States apply the procedural default rule to ineffective assistance of counsel claims. The governing rule in 20 States is that, when a claim of ineffective assistance is sufficiently developed in the trial record, the failure to raise the claim on direct appeal bars its consideration in a post-conviction proceeding, at least if the defendant was represented by new counsel on appeal.18
* * * * *
The bright-line exception to the procedural default rule for ineffectiveness claims urged by petitioner would be easy to administer and would avoid any burdens and uncertainties associated with applying the "cause and prejudice" standard in particular cases. But the interest in ease of administration alone does not justify avoiding the procedural default rule, because the important societal interests in respecting the finality of criminal judgments compels adherence to the rule's restrictions on collateral review. Application of the procedural default rule to ineffective assistance claims advances those interests, and enforcement of the rule in that context is not unworkable or unfair.
II. THE COURT OF APPEALS CORRECTLY DENIED PETITIONER'S INEFFECTIVE ASSISTANCE OF COUNSEL CLAIM UNDER THE PROCEDURAL DEFAULT RULE
A. The Court Of Appeals Correctly Concluded That Petitioner's Ineffective Assistance Claim Is Grounded Solely In The Record Developed At Trial
Petitioner's claim of ineffective assistance alleges that his trial counsel was ineffective in failing to seek or accept a continuance after learning that a bullet had been found in the car in which Fiorito's body was discovered. The court of appeals correctly concluded that petitioner's ineffectiveness claim is "based solely on the record adduced at trial and available to [petitioner's] appellate counsel on direct appeal," Pet. App. A7, that "the record available to [petitioner's] appellate counsel on direct appeal fully revealed the implications of the failure to accept a continuance," id. at A8, and that petitioner therefore could not establish "cause" for failing to raise his ineffective assistance claim on direct appeal, ibid.19
1. By the time petitioner's trial counsel received notice of the bullet's discovery, the parties had given their opening statements and the government had completed its direct examination of Esposito. The government had made no reference to the bullet, and petitioner's trial counsel had not cross-examined Esposito about the murder. Pet. App. A34, A36. Petitioner's counsel was aware before trial that Esposito would testify that petitioner had shot Fiorito a second time in the car; moreover, Esposito had already given that testimony when petitioner's counsel learned about the bullet. Petitioner's counsel also knew that a shell casing had been found in the car when Fiorito's body was discovered there, substantiating Esposito's testimony that Fiorito had been shot a second time in the car.
The discovery of the bullet, for those reasons, did not alter the government's theory of how the murder occurred and did not affect the defense's understanding of the government's theory. Moreover, the bullet did not physically link petitioner to the car or the murder. Instead, the discovery of the bullet in the car-like the earlier discovery of the shell casing in the car- corroborated Esposito's testimony that Fiorito had been shot a second time in the car. Esposito's testimony was further confirmed by the determination that the bullet was of the same caliber as the shell casing, Trial Tr. 908, 3757-3758, and by ballistics tests finding that the bullet matched shell fragments discovered in Fiorito's head, J.A. 178-179.
2. By the time of petitioner's direct appeal, the Second Circuit had made clear that a defendant must raise an ineffective assistance claim on appeal if he is represented by new appellate counsel and the claim is fully developed in the trial record. Billy-Eko, 8 F.3d at 115. Because petitioner was represented by new counsel on appeal, and because he did not raise an ineffective assistance claim at that time, petitioner's ability to establish "cause" for not asserting the claim on appeal turns on whether his claim is "based solely on the record developed at trial." Ibid.; McCleese, 75 F.3d at 1178. Petitioner's ineffectiveness claim therefore must be based, at least in part, on extrinsic evidence not developed in the trial record. Petitioner, as the court of appeals correctly found, cannot make that showing.
a. Petitioner does not dispute that the trial record was fully developed on the performance prong of his ineffective assistance claim. Pet. Br. 32. The record contains numerous discussions about the district court's offer of a continuance and trial counsel's reasons for not accepting one. On September 28, 1993, for example, when petitioner first moved for a mistrial and the district court instead offered a continuance, counsel explained that a continuance was unnecessary because the defense had already retained a ballistics expert who could complete testing on the bullet in a short time. J.A. 20, 24-25. Counsel also argued that a continuance could not undo the effect of the bullet's discovery on the defense theory that no shot had been fired in the car. J.A. 24-25. Later in the trial, counsel reiterated on several occasions his belief that a continuance would not have helped the defense, which was committed to attacking Esposito's testimony that petitioner had shot Fiorito once in the car. J.A. 55-57, 80-85, 100-101. Counsel also acknowledged that he perhaps should have accepted a continuance, observing that he would "take the weight for that," and that he "may have made a mistake. I made a mistake." J.A. 56, 57.
b. Although essentially acknowledging that the performance prong of his ineffectiveness claim rests solely on the trial record, petitioner argues that, as to the prejudice prong, his claim is based on new evidence that was not developed in the trial record. Pet. Br. 32. To establish that he was prejudiced by counsel's failure to accept a continuance, petitioner would need to prove that, if counsel had accepted a continuance, there is a "reasonable probability" that "the result of the [trial] would have been different." Strickland, 466 U.S. at 694. According to petitioner, the prejudice that allegedly resulted from trial counsel's failure to accept a continuance was not apparent until he retained three experts on collateral review, who, after reviewing the trial transcripts and some of the evidence, submitted affidavits attempting to cast doubt on whether the shooting in the car had occurred as Esposito described and whether the bullet recovered from the car in fact was related to Fiorito's murder. The three affidavits, however, add nothing of significance to the trial record on whether counsel's failure to accept a continuance caused prejudice to petitioner.
First, nothing in the affidavits suggests that a continuance would have enabled petitioner to challenge the evidence at trial that the bullet is the same caliber as the shell casing found in the car. More significantly, none of the experts contests the conclusion of the ballistics tests introduced at trial that the bullet matches shell fragments recovered from Fiorito's head. In fact, none of the three experts conducted tests on the bullet or compared the bullet with the shell fragments in Fiorito's head, and only one purports to have physically examined the bullet. J.A. 166.20 At trial, by contrast, petitioner-without the need for a continuance- retained an expert who conducted ballistics testing on the bullet and who quickly concluded, in agreement with the government's experts, that the bullet matched the shell fragments in Fiorito's head. Pet. App. A7. There is no suggestion in the expert affidavits of a reason to doubt the scientific match, and no indication that a continuance somehow could have enabled a direct challenge to that undisputed evidence. In fact, the affidavits offer no new evidence (or theory) on how a bullet that matches shell fragments found in Fiorito's head could have wound up in the car other than in the manner Esposito described.
The affidavits raise questions about whether Esposito's account of the shot fired in the car can be squared with the physical evidence. The first affidavit was submitted by a forensic scientist, Robert Kopec, who reviewed various reports and photographs and portions of the trial transcript. J.A. 143. He observed that there was no blood from petitioner's or Esposito's clothing in the front seat of the car when Fiorito's body was found, J.A. 145, that no blood or tissue "spatter" from the bullet wounds was discovered in the car, J.A. 146-147, that the failure of previous searches to uncover the bullet in the car raised suspicions about the bullet's late discovery, J.A. 147-148, that the bullet should have been extremely deformed rather than only moderately deformed if it had passed through Fiorito's head and the carpeting and insulation of the car, J.A. 148-149, and that Esposito's description of the position of Fiorito's body when he was shot could not be squared with the angle of the wound and the location of the bullet, J.A. 149-150. He opined for those reasons that Esposito's testimony was "not consistent with the physical evidence." J.A. 151.
The remaining two affidavits make some of the same observations. The second expert is Rene Trasorras, a criminal investigator, whose opinion was based on a review of the "facts" of the case. J.A. 161. He concluded that the shooting could not have occurred as Esposito described because the delayed discovery of the bullet cast doubt on its evidentiary value and because the shooting should have produced blood or tissue "splatter" in the car. J.A. 162. Finally, Charles V. Morton, a criminologist, conducted what he described as a "cursory examination" of several items of evidence. J.A. 166. After cataloging the evidence and documents he reviewed, J.A. 165-174, he reached the "tentative conclusion" that there are "serious questions" about the evidence, J.A. 174. He viewed the failure to find the bullet in initial searches of the car to cast doubt on the bullet's evidentiary value, and he believed that proper examinations were not conducted on the hole in the carpet of the car. J.A. 174-175.
Those affidavits, according to petitioner (Br. 18, 32), constitute extrinsic evidence not in the trial record that advance the claim that counsel's failure to seek a continuance was prejudicial. The thrust of petitioner's argument (Br. 18) is that, if trial counsel had secured a continuance, he could have devised the same arguments as are now in the affidavits to cast doubt on Esposito's testimony, and to argue that the bullet-notwithstanding the ballistics match-was unrelated to the murder. If the affidavits in fact presented new theories for calling the bullet evidence into question, petitioner might be able to establish that his claim of prejudice is based on evidence outside the trial record. That argument, however, does not succeed in this case. That is because, as the court of appeals explained, "each of the issues raised" in the three affidavits-"the absence of blood in the vehicle, the absence of blood 'spatter' or other remnants, the belated discovery of the bullet after several searches-all were extensively addressed at trial." Pet. App. A7-A8.
For instance, petitioner's trial counsel cross-examined the crime scene detective and medical examiner about the absence of blood and "spatter" in the car. Trial Tr. 922, 939, 1580, 1621, 1627-1628. He also presented a videotaped demonstration in an effort to cast doubt on Esposito's account of the position of Fiorito's body, and argued in his summation that the shot could not have been fired in the manner Esposito testified given the positioning of the body in the car. Id. at 1589-1595, 4905-4948, 5309-5319. In addition, counsel questioned several witnesses about the circumstances surrounding the bullet's discovery and the failure to discover it earlier. Id. at 3711-3721, 4161-4163, 4171-4179, 4186-4194, 4217-4222, 4467-4471. Finally, counsel asserted in his summation that the condition of the bullet was not consistent with its allegedly having traveled through Fiorito's head and the carpet of the car into the floor. Id. at 5317.
The upshot is that the affidavits do not offer new evidence on how, if counsel had obtained a continuance, the result of the trial would have been different. The affidavits do not reveal a new theory that was unknown to counsel and that would have contributed to the argument that failure to seek a continuance was prejudicial. Nor do the affidavits explain how a continuance would have enabled trial counsel to develop more effectively the theories that he knew about and argued to the jury. Instead, the affidavits reiterate theories that were explored and argued at trial, and that presumably were rejected by the jury. It follows that petitioner's claim of ineffective assistance does not rest on extrinsic evidence concerning the prejudice caused by counsel's failure to secure a continuance. Instead, petitioner's claim is based "solely on the record developed at trial," Billy-Eko, 8 F.3d at 115, and he therefore cannot establish "cause" for failing to assert the claim on direct appeal.21
B. This Court Should Affirm The Judgment On The Alternate Ground That, Even If Petitioner Can Establish "Cause," He Cannot Establish "Prejudice"
Even assuming that petitioner can establish "cause" on the basis that his claim is now based on new evidence that was not developed in the trial record, this Court should affirm the judgment on the ground that petitioner cannot demonstrate "prejudice." The question presented (Pet. i) concerns whether a defendant is procedurally barred from raising his ineffective assistance claim under Section 2255 in the circumstances of this case, and it is settled that, to avoid a procedural bar, a defendant must establish both "cause" and "prejudice." E.g., Bousley, 523 U.S. at 622; Carrier, 477 U.S. at 485. Accordingly, petitioner argues (Br. 33) that he has demonstrated both cause and prejudice for failing to raise his claim on appeal. As a matter of law, however, petitioner cannot establish prejudice in this case.
Demonstrating "prejudice" for purposes of the cause-and-prejudice standard, as explained, entails establishing "prejudice" on the underlying claim of ineffective assistance. See pp. 21-22, supra. Petitioner thus must show that, if trial counsel had accepted a continuance, there is a reasonable probability that the result at trial would have been different. Even assuming that a continuance would have assisted petitioner in developing further the theories reviewed in the expert affidavits, there is no reasonable likelihood that the jury would have acquitted petitioner on the charge that he murdered Fiorito.
First, none of the theories advanced in the affidavits calls into question Esposito's testimony about the first shot petitioner fired into Fiorito's head in the vacant house. As to the second shot fired in the car, a shell casing was found in the car at the time Fiorito's body was discovered, a bullet was later found in the car, the shell casing and the bullet are of the same caliber, and, most importantly, ballistics tests conducted both by the government and by petitioner's expert confirmed that the bullet matches shell fragments found in Fiorito's head. The theories outlined in the expert affidavits contain no explanation for the discovery in the car of a bullet scientifically matched to shell fragments in Fiorito's head. And the attacks on whether Esposito's account conforms to the physical evidence were already placed before the jury. The jury nonetheless was persuaded by Esposito's testimony, and it rejected petitioner's attempt to argue that Esposito alone was responsible for the murder. In those circumstances, there can be no reasonable likelihood that a continuance to further develop arguments along the lines contemplated in the affidavits could have affected the jury's verdict against petitioner.
The judgment of the court of appeals should be affirmed.
THEODORE B. OLSON
Assistant Attorney General
MICHAEL R. DREEBEN
Deputy Solicitor General
Assistant to the Solicitor
STEVEN L. LANE
1 When asked by the district court about the delay in disclosing the discovery of the bullet, the prosecutor explained that he had waited to inform petitioner's trial counsel because, initially, he was informed that the bullet could not be linked to Fiorito's murder. J.A. 50-51. Petitioner's trial counsel advised the district court that he did not believe the government had acted in bad faith. J.A. 30-31.
2 Petitioner had retained a ballistics expert by the time he moved for a mistrial, J.A. 20, and petitioner's expert soon reached agreement with the government's conclusion that the bullet matched the shell fragments recovered from Fiorito's head, J.A. 55. Pet. App. A7.
3 The court of appeals subsequently expanded the certificate of appealability to include a second claim concerning whether trial counsel had been ineffective in failing to call petitioner to testify at trial. Pet. App. A28-A29. That claim is not before this Court.
4 In the absence of a showing of "cause" and "prejudice," a demonstration of "actual innocence" would allow assertion of a defaulted claim on collateral review. See Bousley, 523 U.S. at 622; Murray v. Carrier, 477 U.S. 478, 495-496 (1986). The exception for actual innocence is not at issue in this case.
5 When a federal court conducts habeas review of a state court conviction, the requirement to show cause and prejudice to overcome a state law procedural bar rests on both considerations of comity and the interest in finality. See, e.g., Coleman v. Thompson, 501 U.S. 722, 730 (1991). Although collateral review of federal convictions under Section 2255 does not implicate concerns of comity, the limitation on collateral review of claims that were not raised on direct appeal applies with full force to review under Section 2255. See Bousley, 523 U.S. at 621; Reed, 512 U.S. at 354. That is because "the Federal Government, no less than the States, has an interest in the finality of its criminal judgments." Frady, 456 U.S. at 166.
6 In its brief in opposition to the petition for certiorari in this case, the government argued that this Court need not grant review to establish a uniform national rule on the circumstances in which the failure to raise an ineffectiveness claim on direct appeal bars its assertion on collateral review. See Br. in Opp. 11-13. The grant of certiorari, however, squarely presents the Court with a choice among competing approaches, and the government submits, in light of its experience with the different approaches over the years, that the approach of the Second and Seventh Circuits is workable and fair.
7 See, e.g., United States v. Mala, 7 F.3d 1058, 1063 (1st Cir. 1993), cert. denied, 511 U.S. 1086 (1994); Billy-Eko, 8 F.3d at 114; United States v. DeRewal, 10 F.3d 100, 103-104 (3d Cir. 1993), cert. denied, 511 U.S. 1033 (1994); United States v. Smith, 62 F.3d 641, 651 (4th Cir. 1995); United States v. Sevick, 234 F.3d 248, 252 (5th Cir. 2000); United States v. Pruitt, 156 F.3d 638, 646 (6th Cir. 1998), cert. denied, 525 U.S. 1091 and 526 U.S. 1012 (1999); United States v. Kellum, 42 F.3d 1087, 1094-1095 (7th Cir. 1994); United States v. Thompson, 972 F.2d 201, 203-204 (8th Cir. 1992); United States v. Mal, 942 F.2d 682, 689 (9th Cir. 1991); United States v. Galloway, 56 F.3d 1239, 1240 (10th Cir. 1995) (en banc); United States v. Casamayor, 837 F.2d 1509, 1516 (11th Cir. 1988), cert. denied, 488 U.S. 1017 (1989); United States v. Richardson, 167 F.3d 621, 626 (D.C. Cir.), cert. denied, 528 U.S. 895 (1999).
8 Other courts of appeals, while allowing ineffectiveness claims to be raised on direct appeal, have expressed a general preference that those claims be brought in a motion under Section 2255. See, e.g., Smith, 62 F.3d at 651 (4th Cir.); Sevick, 234 F.3d at 251 (5th Cir.); Pruitt, 156 F.3d at 646 (6th Cir.); Thompson, 972 F.2d at 203-204 (8th Cir.); Mal, 942 F.2d at 689 (9th Cir.). Those courts have not squarely addressed whether, for those ineffectiveness claims that could be addressed on direct appeal, failure to present the claim on appeal is subject to the procedural default rule.
9 The procedural default rule requires raising on direct appeal legal claims that arise from the trial, and any asserted explanation for the failure to raise a claim on appeal-such as an insufficient development of the factual basis for the claim or the unfairness of expecting an attorney to argue his own ineffectiveness on appeal -pertains to whether there is "cause" for not asserting the claim. See Carrier, 477 U.S. at 488 ("a showing that the factual * * * basis for the claim was not reasonably available to counsel * * * would constitute cause"). Even assuming, however, that those sorts of considerations bear on the antecedent question whether there is a procedural default at all-rather than the question of "cause" for a default, cf. Bousley, 523 U.S. at 621 (suggesting "an exception to the procedural default rule for claims that could not be presented without further factual development")-the Court should hold that at least those ineffectiveness claims where the defendant has new counsel on appeal and the factual basis for the claim is fully developed are procedurally defaulted if not raised on direct appeal. Under either approach, the ability to raise a claim on collateral review turns on whether the defendant was represented by new counsel on appeal and the factual basis for the claim was developed in the trial record.
10 Conceivably, a defendant could establish "cause" not just on the grounds that trial counsel continued to represent him on appeal or that the factual basis for the claim was not fully developed, but also on the ground that appellate counsel was constitutionally ineffective in failing to preserve the claim of trial counsel's ineffectiveness. See Carrier, 477 U.S. at 488 ("Ineffective assistance of counsel * * * is cause for a procedural default."); McCleese, 75 F.3d at 1180 (rejecting claim that appellate counsel was ineffective in failing to preserve ineffectiveness claim). As this Court has made clear when addressing the failure of appellate counsel to preserve a claim, however, "the mere fact that counsel failed to recognize the factual or legal basis for a claim, or failed to raise the claim despite recognizing it, does not constitute cause," Carrier, 477 U.S. at 486, nor does a deliberate decision not to press the claim in order to emphasize other claims perceived as more likely to prevail, Smith v. Murray, 477 U.S. 527, 535-536 (1986).
11 See United States v. Jimenez Recio, 258 F.3d 1069, 1074 (9th Cir. 2001), cert. granted on other grounds, 122 S. Ct. 2288 (2002); United States v. Soto, 132 F.3d 56, 59 (D.C. Cir. 1997); United States v. Swartz, 975 F.2d 1042, 1050 (4th Cir. 1992); United States v. Tatum, 943 F.2d 370, 379-380 (4th Cir. 1991); United States v. Swanson, 943 F.2d 1070 (9th Cir. 1991); United States v. Headley, 923 F.2d 1079 (3d Cir. 1991); United States v. Ford, 918 F.2d 1343, 1350 (8th Cir. 1990); United States v. Loughery, 908 F.2d 1014 (D.C. Cir. 1990); United States v. Iorizzo, 786 F.2d 52, 57-59 (2d Cir. 1986); Government of the Virgin Islands v. Zepp, 748 F.2d 125 (3d Cir. 1984); cf. United States v. Russell, 221 F.3d 615, 619-623 (4th Cir. 2000) (granting relief on direct appeal on ineffective assistance claim based on record developed in post-trial proceedings on motion for new trial under Federal Rule of Criminal Procedure 33); United States v. Patterson, 215 F.3d 776 (7th Cir.) (same), judgment vacated in part on other grounds, 531 U.S. 1033 (2000); United States v. Fulton, 5 F.3d 605, 613-614 (2d Cir. 1993) (granting relief on direct appeal on grounds of ineffective assistance after district court rejected post-trial motion under Section 2255).
12 Section 105 of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), Pub. L. No. 104-132, 110 Stat. 1220, establishes a default rule that Section 2255 motions must be brought within one year of the date on which a conviction becomes final. See 28 U.S.C. 2255 para. 6(1). That provision does not affect the ordinary operation of the procedural bar against collateral review of claims not raised on direct appeal. Rather, the procedural default rule and AEDPA's default one-year limitations period work in complementary fashion to reduce delay in the assertion of legal challenges to a conviction and to advance the interest in respecting finality.
13 See, e.g., United States v. Haywood, 155 F.3d 674, 678 (3d Cir. 1998); United States v. King, 119 F.3d 290, 295 (4th Cir. 1997); United States v. Rivas, 157 F.3d 364, 369 (5th Cir. 1998); United States v. Neuhausser, 241 F.3d 460, 474-475 (6th Cir.), cert. denied, 122 S. Ct. 181 (2001); United States v. Richardson, 167 F.3d 621, 626 (D.C. Cir.), cert. denied, 528 U.S. 895 (1999).
14 Some decisions appear to suggest a blanket preference for remanding an ineffectiveness claim that requires further factual development rather than remitting the defendant to pursuing the claim on collateral review. See United States v. Geraldo, 271 F.3d 1112, 1116 (D.C. Cir. 2001); United States v. Leone, 215 F.3d 253, 256 (2d Cir. 2000). Generally, however, the proper course is to decline to resolve the claim and grant the defendant leave to develop it on collateral review. See United States v. Reyes-Platero, 224 F.3d 1112, 1116-1117 (9th Cir. 2000) (inappropriate to remand claim), cert. denied, 531 U.S. 1117 (2001); Haywood, 155 F.3d at 678 & n.3 (same). A routine resort to remand would delay imposition of a final judgment and would have the effect of undermining AEDPA's strict limitations on the filing of successive Section 2255 motions. See 28 U.S.C. 2255 para. 8.
15 NACDL argues (Br. 18-19) that ineffectiveness claims should be addressed on collateral review in order to ensure that trial counsel has a full opportunity to explain his or her challenged actions. The government shares an interest in ensuring that relief is not granted on grounds of ineffective assistance without taking into account trial counsel's strategic reasons for a challenged decision. The Seventh Circuit has addressed those concerns by adopting a presumption on direct appeal that trial counsel acted for tactical reasons. See, e.g., United States v. Trevino, 60 F.3d 333, 338 (7th Cir. 1995), cert. denied, 516 U.S. 1061 (1996). Relief therefore is available on direct appeal only if there is no possible strategic explanation for counsel's actions.
16 The application of that standard requires that the purportedly new non-record evidence make a substantive addition to the trial record. It is not enough for a defendant to proffer evidence that purports to add a new basis for a claim of ineffective assistance, but that, in fact, substantively is equivalent to matter already in the trial record. Otherwise, defendants could easily circumvent the procedural default rule by tendering affidavits that purport to offer new evidence even when they add nothing to the claim. Such an approach-as occurred in this case-should not be permitted. See pp. 37-44, infra.
17 NACDL identifies a handful of district court decisions (NACDL Br. 21) that, in NACDL's view, unfairly hold that the defendant procedurally defaulted an ineffectiveness claim. In each case, however, the court went on to address the claim on the merits. See Moskowitz v. United States, No. 01-Civ-10644, 2002 WL 31119629, at *3-*4 (S.D.N.Y. Sept. 24, 2002); Saldarriaga v. United States, No. 99-CIV-4487, 2002 WL 449651, at *3-*4 (S.D.N.Y. Mar. 21, 2002); Peets v. United States, 55 F. Supp. 2d 275, 277-279 (S.D.N.Y. 1999); Grice v. United States, No. 98-CV-622, 1998 WL 743718, at *2-*3 (N.D.N.Y. Oct. 15, 1998); cf. James v. United States, No. 00-CIV-8818LAKGWG, 2002 WL 1023146, at *9-*10, *14-*16 (S.D.N.Y. May 20, 2002) (claims asserting ineffectiveness of trial counsel not sufficiently meritorious to render appellate counsel ineffective for failing to raise claims on appeal).
18 That result is reflected in the decisions of the highest courts in 16 states, see Sullivan v. United States, 721 A.2d 936, 937 (D.C. 1998); White v. Kelso, 401 S.E.2d 733, 734 (Ga. 1991); State v. Silva, 864 P.2d 583, 591-592 (Haw. 1993) (assuming procedural bar); People v. Coady, 622 N.E.2d 798, 802 (Ill. 1993); Jones v. State, 479 N.W.2d 265, 271 (Iowa 1991); Commonwealth v. Chase, 741 N.E.2d 59, 64 & n.3 (Mass. 2001); Robledo-Kinney v. State, 637 N.W.2d 581, 585 (Minn. 2002); Lockett v. State, 656 So. 2d 76, 80 (Miss.), cert. denied, 515 U.S. 1150 (1995); State v. Courchene, 847 P.2d 271, 276 (Mont. 1992); State v. Williams, 609 N.W.2d 313, 317-318 (Neb. 2000); State v. Fair, 557 S.E.2d 500, 524-525 (N.C. 2001), cert. denied, 122 S. Ct. 2332 (2002); State v. Lentz, 639 N.E.2d 784, 785 (Ohio 1994); McCracken v. State, 946 P.2d 672, 676 (Okla. Crim. App. 1997); Commonwealth v. Moore, 805 A.2d 1212, 1214-1216 (Pa. 2002); Pascual v. Carver, 876 P.2d 364, 366 (Utah 1994); State v. Escalona-Naranjo, 517 N.W.2d 157, 158-159, 163-164 (Wis. 1994); Amin v. State, 774 P.2d 597, 598-599 (Wyo. 1989), and in the decisions of intermediate courts in four states, see Cain v. State, 712 So. 2d 1110, 1112 (Ala. Crim. App. 1997); Maggard v. State, 11 P.3d 89, 93 (Kan. Ct. App. 2000); People v. Wong, 682 N.Y.S.2d 689, 691 (N.Y. App. Div. 1998); McCarver v. Lee, 221 F.3d 583, 589 (4th Cir. 2000) (reviewing North Carolina decisions). See generally Anne M. Voigts, Note, Narrowing the Eye of the Needle: Procedural Default, Habeas Reform, and Claims of Ineffective Assistance of Counsel, 99 Colum. L. Rev. 1103, 1128 (1999) ("in a growing number of cases, state courts are finding procedural defaults based on failure to raise ineffective assistance of counsel claims on direct appeal").
19 Petitioner's claim that trial counsel was ineffective in failing to seek a continuance upon learning about the bullet, by its own terms, can only relate to counsel's conduct after he was apprised of the bullet's discovery. As a result, any events that preceded counsel's awareness of the bullet could have no bearing on petitioner's ineffectiveness claim. In any event, the court of appeals, in petitioner's direct appeal, held that the government's delay of four days in informing petitioner's trial counsel about the bullet, although "regrettable," did not prejudice petitioner. Pet. App. A36.
20 Petitioner asserts without elaboration (Br. 12) that "the Government permitted only a restricted examination" of the physical evidence, repeating a suggestion made in the proceedings below that petitioner's experts were not allowed to conduct testing on the evidence. See J.A. 188-189. In fact, however, when petitioner's new counsel on collateral review initially sought to have trial evidence examined, the government expressed its concern that the integrity of the evidence be maintained. To that end, counsel was asked to specify the types of testing the experts desired to conduct and the effects of those tests on the evidence, or to seek a court order permitting the tests to be conducted. That recommendation was repeated in a letter to the district court and to counsel. Petitioner thereafter did not pursue further examination of the evidence or seek a court order. See Gov't Mem. of Law in Opp. to Joseph Massaro's Mot. for a New Trial and for Relief Under 28 U.S.C. 2255 at 45 n.29.
21 Petitioner suggests (Br. 33) that the government's delay in disclosing
discovery of the bullet suffices to establish "cause." That is
incorrect. The government's belated disclosure of the bullet at trial cannot
constitute "cause" for petitioner's failure to raise a claim of
ineffective assistance on appeal. Moreover, the court of appeals specifically
found that petitioner had not been prejudiced by the delay. Pet. App. A36.
1. The Sixth Amendment to the United States Constitution provides in relevant part: "In all criminal prosecutions, the accused shall enjoy the right * * * to have the Assistance of Counsel for his defence."
2. Section 2255 of Title 28, United States Code (2000) provides in relevant part:
A prisoner in custody under sentence of a court established by Act of Congress claiming the right to be released upon the ground that the sentence was imposed in violation of the Constitution or laws of the United States, or that the court was without jurisdiction to impose such sentence, or that the sentence was in excess of the maximum authorized by law, or is otherwise subject to collateral attack, may move the court which imposed the sentence to vacate, set aside or correct the sentence.
Unless the motion and the files and records of the case conclusively show that the prisoner is entitled to no relief, the court shall cause notice thereof to be served upon the United States attorney, grant a prompt hearing thereon, determine the issues and make findings of fact and conclusions of law with respect thereto. If the court finds that the judgment was rendered without jurisdiction, or that the sentence imposed was not authorized by law or otherwise open to collateral attack, or that there has been such a denial or infringement of the constitutional rights of the prisoner as to render the judgment vulnerable to collateral attack, the court shall vacate and set the judgment aside and shall discharge the prisoner or resentence him or grant a new trial or correct the sentence as may appear appropriate.
* * * * *