Victor D. Marzullo's appeal from the denial of his application for a writ of habeas corpus raises this question: What is the appropriate standard for determining when criminal defendants have been denied their constitutional right to the effective assistance of counsel? We conclude that the standard applied by the district court is no longer acceptable, and that properly tested, the representation Marzullo received was ineffective.

In considering whether Marzullo had been denied effective assistance of counsel, the district court applied the standard set forth in Root v. Cunningham, 344 F.2d 1, 3 (4th Cir. 1965):

Ordinarily, one is deprived of effective assistance of counsel only in those extreme instances where the representation is so transparently inadequate as to make a farce of the trial.

The farce and mockery of justice test to which Root referred gained wide currency in the era when an accused tried in a state court had no constitutional entitlement to counsel unless he could satisfy the requirements of Betts v. Brady, 316 U.S. 455, 62 S. Ct. 1252, 86 L. Ed. 1595 (1942).1 That case compelled him to show circumstances rendering the lack of counsel so "offensive to the common and fundamental ideas of fairness and right" as to deny him due process of law. 316 U.S. at 473, 62 S.Ct. at 1262. The overruling of Betts by Gideon v. Wainwright, 372 U.S. 335, 83 S. Ct. 792, 9 L. Ed. 2d 799 (1963), foretold a different standard for determining counsel's adequacy. Following Gideon, courts and commentators recognized that the right to counsel assured by the sixth and fourteenth amendments lacks substance unless counsel is reasonably competent.2

We implicitly departed from the farce and mockery test when, in Coles v. Peyton, 389 F.2d 224 (4th Cir. 1968), we imposed specific requirements for counsel's preparation of his client's defense. Coles has been cited frequently as offering an improved measure for counsel's performance.3 Nevertheless, some of our subsequent opinions quoted the older test, and district courts, justifiably relying on them, have continued to apply it.4 In other instances, we have not referred to that test.5 Our decisions, however, have been more consistent than reference to the test or lack of it would indicate. Since Coles, we have usually judged effective representation by determining whether counsel furnished reasonably adequate services instead of inquiring whether the representation was so poor as to make a farce of the trial.6 Be that as it may, our ambivalence has persisted long enough. We now expressly disavow the farce and mockery of justice test which we approved in Root v. Cunningham, 344 F.2d 1 (4th Cir. 1965).

Two years after Coles, the Supreme Court decided McMann v. Richardson, 397 U.S. 759, 90 S. Ct. 1441, 25 L. Ed. 2d 763 (1970), which dealt with the validity of a guilty plea following a coerced confession. There the Court reiterated that defendants "are entitled to the effective assistance of competent counsel." It explained that a court should not measure the competency of counsel's advice by retrospectively considering whether it was right or wrong. The proper test, the Court stated, is whether counsel's advice "was within the range of competence demanded of attorneys in criminal cases." 397 U.S. at 770-71, 90 S.Ct. at 1449.

Though McMann involved an attorney's advice about pleading guilty, many courts have recognized that it provides a suitable general standard for determining whether the representation afforded the accused satisfied his constitutional right to effective counsel. One of the first courts to adopt the McMann standard was the Court of Appeals for the Third Circuit in Moore v. United States, 432 F.2d 730, 736 (3d Cir. 1970) (en banc). Numerous other courts have followed Moore's lead in adopting some version of the normal competency test: United States v. DeCoster, 159 U.S.App.D.C. 326, 487 F.2d 1197, 1202 (1973); Herring v. Estelle, 491 F.2d 125, 128 (5th Cir. 1974); Williams v. Twomey, 510 F.2d 634, 641 (7th Cir. 1975); Beasley v. United States, 491 F.2d 687, 696 (6th Cir. 1974); United States v. Easter, 539 F.2d 663, 665-67 (8th Cir. 1976); but see Rickenbacker v. Warden, Auburn Correctional Facility, 550 F.2d 62 (2d Cir. 1976).

We, too, are persuaded that McMann furnishes the proper standard for determining the effectiveness of counsel. Therefore, paraphrasing the Court's opinion, we adopt as an appropriate measure: Was the defense counsel's representation within the range of competence demanded of attorneys in criminal cases? See 397 U.S. at 771, 90 S. Ct. 1441.

By this standard, effective representation is not the same as errorless representation. An attorney may make a decision or give advice which in hindsight proves wrong. Such errors, as McMann pointed out, are not necessarily grounds for post-conviction relief. 397 U.S. at 770-71, 90 S. Ct. 1441.7 A convict generally must establish that his counsel's error was so flagrant that a court can conclude that it resulted from neglect or ignorance rather than from informed, professional deliberation.8

The McMann normal competency standard is similar in many respects to those which judges must apply in other contexts.9 Moreover, because it measures an attorney's conduct by comparison with the competence generally found in the profession, it requires an objective assessment of counsel's adequacy.10 While the normal competency standard does not purport to list the things counsel should or should not do, it does not preclude resorting to specifics for ascertaining the "range of competence demanded of attorneys in criminal cases." For example, in Coles v. Peyton, 389 F.2d 224, 226 (4th Cir. 1968), we said:

Counsel for an indigent defendant should be appointed promptly. Counsel should be afforded a reasonable opportunity to prepare to defend an accused. Counsel must confer with his client without undue delay and as often as necessary, to advise him of his rights and to elicit matters of defense or to ascertain that potential defenses are unavailable. Counsel must conduct appropriate investigations, both factual and legal, to determine if matters of defense can be developed, and to allow himself enough time for reflection and preparation for trial.

We adhere to this statement, for it is a definitive, objective description of the competency normally demanded of counsel in certain aspects of their service.

The normal competency standard is necessarily broad and flexible because it is designed to encompass many different factual situations and circumstances. Consequently, its fair and effective administration rests primarily on the district judges. Speaking for the Court in McMann on this subject, Mr. Justice White said:

(W)e think the matter, for the most part, should be left to the good sense and discretion of the trial courts with the admonition that if the right to counsel guaranteed by the Constitution is to serve its purpose, defendants cannot be left to the mercies of incompetent counsel, and that judges should strive to maintain proper standards of performance by attorneys who are representing defendants in criminal cases in their courts. 397 U.S. at 771, 90 S.Ct. at 1449.

In exercising its discretion, a trial court may refer to other sources to determine the normal competency of the bar. Among these are precedent from state and federal courts, state bar canons, the American Bar Association Standards Relating to the Defense Function (App. Draft 1971), and in some instances, expert testimony on the particular conduct at issue. These, of course, do not supplant the test that we have prescribed, but they can aid in objectively ascertaining the range of competency normally expected of attorneys practicing criminal law.

Although this appeal was not heard en banc, all of the judges of the court approve Parts I and II.

We now apply the normal competency standard to Marzullo's claim of ineffective representation.

The Public Defender named an attorney to represent Marzullo in the Circuit Court of Baltimore, Maryland, where he had been charged in separate indictments with raping two women. Shortly after the appointment, Marzullo complained in a letter to the trial judge about the lawyer: "The impression he gave me by talking to him, I really don't think for one minute he is going to try and defend me adequately except to go through the formalities of my trial." The court denied Marzullo's request for a substitute, and subsequently, Marzullo alleged many deficiencies in the representation he received. We believe that only his complaint about the selection of the jury merits discussion.

When the first rape case was called for trial, the prosecuting witness told the judge, in the presence of the jurors, that she could not identify Marzullo as her attacker because she did not see his face. The prosecutor responded that she had previously said she could identify him, but the case was dismissed with the State's consent.

The prosecutor then announced that the State was ready to proceed on the second indictment charging Marzullo with rape, assault with intent to rape, common law assault, statutory mayhem, and perverted sexual practice. Marzullo was arraigned and pled not guilty. The court, observing that only 19 jurors were present, noted that this was insufficient to allow all peremptory challenges.11 The prosecutor said the state would waive its challenges, and Marzullo's attorney stated, "If Your Honor please, my client advises me he believes we would end up with 12 jurors out of that many even with the challenges."

The jury panel was sworn and the court conducted the following voir dire:

THE COURT: The first question is: Each of you heard, I am sure, what I said about being advised by the State's attorney that neither of these girls could identify the alleged assailant. You have also observed the fact that Ms. Yingling openly stated in Court she could not identify him, or he was not the man. So that the case was confessed not guilty. There was a conversation between the other girl and the State's Attorney obviously indicating that young lady believes she can identify the defendant. Would that fact in any way affect your ability to render a fair and impartial verdict in this case?

(No response.)

THE COURT: Would you base your verdict solely on what you hear from the witness stand?

(No response.)

THE COURT: Is there any juror who feels she or he cannot render a fair and impartial verdict, in view of what has taken place in Court this morning?

(No response.)

THE COURT: The answers to all the foregoing questions are in the negative.

The jury was then impanelled, ten minutes after court had convened.

The evidence discloses that Marzullo and the prosecuting witness, an 18-year-old married woman, met at the home of a mutual friend. Marzullo offered to drive the woman to the home of another friend. They left with Marzullo's girlfriend whom they soon dropped off. For the next several hours Marzullo and the woman drove around, making a few stops and eventually parking at a secluded spot in the country.

The woman testified that Marzullo threatened to kill her, wounded her with a knife, forced her to disrobe, raped her, and compelled her to commit fellatio. She said that after she dressed, she fled in fear and asked some people in a nearby car to take her to a friend's house in Baltimore.

Marzullo testified that after they parked the woman kissed him, they both voluntarily disrobed and got in the back seat where she willingly committed fellatio. He denied threatening her or wounding her, or having intercourse with her. He claimed she left his car because as they were about to drive away they got into an argument and he slapped her.

Approximately five hours elapsed from the time the woman started out with Marzullo until she complained to the police. A medical examination revealed minor cuts and abrasions, but no sign of vaginal penetration or sperm. The police later found a knife in Marzullo's car.

The jury convicted Marzullo of assault with intent to rape and of perverted practices. The court sentenced him to 10 years on the first charge and to a concurrent sentence of 5 years on the second. After exhausting his state remedies, Marzullo petitioned for a writ of habeas corpus in the district court.

Relying on the state trial and post-conviction records, the district judge denied relief. He held that since Marzullo and his counsel were aware that the jurors were present when the first indictment was dismissed, the failure of counsel to object resulted in a waiver of the right to collateral relief. He noted that Marzullo himself made no objection when his lawyer assured the court that a jury could be selected, and he characterized the lawyer's conduct as a permissible trial tactic.

The district judge's decision rested in large part on his justifiable reliance on our adherence to the farce and mockery test to which he expressly referred. If the material facts were in dispute or incompletely developed, we would remand the case for reconsideration in the light of the standard that we adopted in Part II of this opinion. But here the attorney's conduct at the voir dire and the jury's exposure to the proceedings involving the first indictment are undisputed matters of record. We believe therefore that a remand for an evidentiary hearing on the habeas corpus petition is unwarranted.

Before the trial, Marzullo's attorney learned at a bench conference that the State could not proceed on the first indictment. There is no doubt that he also was aware of the conflicting accounts as to what took place in the parked car, and he should have realized that the defense rested almost entirely on the jury's acceptance of Marzullo's testimony. The information about the first rape charge had no probative value and could only serve to discredit Marzullo; therefore, it was important to withhold from the jury all prejudicial reference to it. Nevertheless, the attorney did not move to exclude the jury during the interrogation of the prosecuting witness about the crime charged in the first indictment.

Furthermore, the attorney's conduct at the voir dire again failed to protect Marzullo from the prejudicial effects of the jury's exposure to the first rape charge. Before the jurors were even questioned, he assured the court that a panel could be selected. At that time, he could not possibly have known that the colloquy about the first indictment did not bias any of the jurors. Furthermore, he never asked the court to instruct the jury that remarks about the dismissed rape charge could not be considered in their deliberation about the second indictment.

The state's defenses in the habeas proceeding of waiver and trial tactics are not supported by the records of either the state trial or the post-conviction hearing. Before the trial, Marzullo's attorney explained the procedure for striking the jury, but there is no evidence that he advised Marzullo of his right to have the proceedings about the first indictment conducted in the absence of the jury. Without proof that Marzullo knew his rights, waiver cannot be presumed. See Johnson v. Zerbst, 304 U.S. 458, 464, 58 S. Ct. 1019, 82 L. Ed. 1461 (1938).

Furthermore, at the post-conviction hearing, Marzullo's attorney made no claim that his decision not to challenge the jury was a trial tactic. On the contrary, he testified that according to his recollection, the jury had not been present when the prosecuting witness stated she could not identify Marzullo. His explanation is clearly refuted by the record, particularly by the trial judge's observation during the voir dire about the jury's presence.

As we noted in Part II, the ABA Standards Relating to the Defense Function (App. Draft 1971) do not provide a test for determining whether an accused has received effective representation. They do however, furnish a reliable guide for determining the responsibilities of defense counsel, including those related to the selection of the jury.12 We conclude that Marzullo's attorney failed to discharge his duty effectively by not moving to exclude the jury while the first indictment was dismissed, and then by foregoing Marzullo's peremptory challenges before any voir dire examination. His actions exemplified his perfunctory approach to the important task of selecting an unbiased jury. We are persuaded that his representation of Marzullo with respect to this aspect of the case was outside the range of competence expected of attorneys in criminal cases.

The judgment of the district court is reversed, and this case is remanded with instructions to issue a writ of habeas corpus discharging Marzullo from custody unless the State retries him within a reasonable time.