Criminal Justice Section  


Criminal Justice Magazine
Fall 2003
Volume 18 Number 3

Exonerations Change Judicial Views on Ineffective Assistance of Counsel

by Adele Bernhard

Adele Bernhard is an associate professor of law at Pace Law School in White Plains, New York, where she teaches the Criminal Defense Clinic and the Post-Conviction Project.

The American public is fascinated with criminal justice. Crime stories titillate. They’re the daily fare on television newscasts, the subject of "true crime" bestsellers, and the front-page stories in newspapers and magazines nationwide. Today, it is the story of exonerations of the wrongly convicted that has captured the public’s imagination. In recent months, the New York Times painstakingly explored how a group of young men convicted of raping and assaulting the "Central Park jogger" had been wrongly convicted—despite their taped confessions. Newsday published a four-part series on 13 wrongly convicted individuals in New York alone. Parade magazine featured "It Could Happen to Any of Us," by Jack Newfield, describing how Ray Krone, a former Boy Scout and Little League ballplayer, was convicted of a murder he did not commit in Arizona. PBS aired an hour-long special that illustrated the post-exoneration lives of wrongly convicted men. And The Exonerated—a play combining fiction with real-life events, was an off-Broadway hit.

It is not hyperbole to suggest that the interest and excitement generated by the stream of exoneration stories has encouraged hundreds of young people to attend law school, invigorated J.D. curriculum, revamped crime laboratories, and influenced jury verdicts.

Law evolves more slowly than pop culture or public attitude. Because most exonerations have not resulted in written legal opinions, their impact is only slowly seeping into case law. However, courts are influenced by the same news that sways the rest of us. Even without explicitly referring to innocence or wrongful convictions, modern trial courts are undoubtedly more likely to admit expert testimony on the question of eyewitness identification because they are painfully aware of just how easily such witnesses—no matter how honest or passionate—can be wrong. They are certainly more inclined to view confessions suspiciously, especially when it involves the very young, and to consider whether and to what extent police slant evidence. Finally, the fact that innocent people are routinely convicted—despite a full-blown jury trial at which they were represented by defense counsel—suggests that courts should play a more active role in supervising the quality of criminal defense services.

Without overstating the case, there is some evidence that courts are doing more to protect the rights of the accused to effective assistance of counsel. Some courts are relaxing the overly restrictive standard by which individual post-conviction claims are judged. Others have become more receptive to affirmative litigation challenging the provision of criminal defense services on Sixth Amendment grounds. Finally, at least one circuit has abrogated the virtual immunity that currently protects assigned and public defenders from malpractice liability. This article discusses each of these developments, focusing on the federal courts and on New York State, where I live and practice.

Ambivalence to enforcing right to counsel

Even as its criminal justice jurisprudence has evolved more restrictively, the Supreme Court has steadfastly insisted that the Constitution requires provision of counsel to anyone facing a loss of freedom as the result of a criminal charge. ( Gideon v. Wainwright, 372 U.S. 335 (1963); Alabama v. Shelton, 535 U.S. 654 (2002).) Nonetheless, a right is only as potent as its enforcement, and the vigor of the Sixth Amendment right to counsel has been undercut by judicial reluctance to supervise the provision of criminal defense services. In part, this reticence is due to appropriate concern for finality and the difficulty of devising a standard of review that would spare appellate courts the task of combing through lengthy transcripts looking for trial errors. But the reserve can also be attributed to the unstated belief that excellence in defense services is unnecessary. If everyone is guilty, it doesn’t matter who does the defending. As Richard Posner puts it:

I can confirm from my own experience as a judge that criminal defendants are generally poorly represented. But if we are to be hardheaded we must recognize that this may not be an entirely bad thing. The lawyers who represent indigent criminal defendants seem to be good enough to reduce the probability of convicting an innocent person to a very low level. If they were much better, either many guilty people would be acquitted or society would have to devote much greater resources to the prosecution of criminal cases. A bare-bones system for defense of indigent criminal defendants may be optimal.

(Richard A. Posner, The Problematics of Morals and Legal Theory 163–64 (1999), cited in Abbe Smith, Difference in Criminal Defense and the Difference It Makes, 11 Wash. U. J.L. & Pol’y 83 (2003).)

Twenty years ago, when the Supreme Court decided Strickland v. Washington, 466 U.S. 668 (1984), the probability of convicting an innocent person seemed very low indeed. Our federal and state multilayered criminal justice systems, replete with complex pretrial procedures and constitutional protections, were thought to accurately differentiate between guilty persons and innocent ones. Judicial comfort with the operation of the criminal justice system is reflected in the Strickland decision, which diminishes the significance of counsel’s role in the trial process. However, as the courts lose faith in police and prosecutorial ability (or inclination) to distinguish between the guilty and the innocent, they are sure to refocus attention on the significance of zealous defense advocacy and search for ways to compel defense counsel to do a better job of advocating for their clients—so as to reduce the chances of convicting the innocent.

Revisiting Strickland

One way to influence the quality of defense services is to reverse more cases for ineffective assistance of counsel. Judicial tolerance of malpractice has insulated ineffective counsel, and the lack of adverse consequences for bad lawyering has retarded efforts at reform. Certainly, reversal of an individual case gives that particular defendant a better shot at an accurate verdict. Moreover, every reversal teaches a general lesson about counsel’s obligations, and potentially requires a greater investment in defense resources.

Currently, for a conviction to be reversed on the ground of ineffective assistance of counsel, Strickland requires: 1) that the defense attorney’s performance fell below an objective standard of reasonableness, and 2) a reasonable probability that the malfeasance prejudiced the outcome of the trial. ( See Peter A. Joy and Kevin C. McMunigal, Has Gideon ’s Promise Been Fulfilled?, 18:4 Crim. Just. 46 (Summer 2003).) In other words, where there was overwhelming proof of guilt at trial, malpractice will be excused even if that malpractice involved such egregious behavior as sleeping, taking drugs, or drinking during trial, suffering through a psychotic break, or any number of other disasters that have been so extensively reported by journalists and scholars alike.

Although it is unlikely that the Supreme Court will overrule Strickland, recognition that ineffective assistance of counsel is contributing to convictions of innocent people may spur courts to: (1) either find prejudice more frequently, or (2) to characterize counsel’s error or omission as egregious enough to avoid a careful prejudice analysis.

Focusing on prejudice

Courts could increase the number of circumstances where malfeasance is presumed to have prejudiced the outcome of the trial—without resorting to a careful weighing of attorney malfeasance against prosecutorial proof. There are already three situations where courts will presume prejudice and reverse a conviction without measuring how the attorney’s performance affected the outcome. Prejudice is presumed when counsel and client are divided by a completely antagonistic relationship, rising to the level of an "irreconcilable conflict," United States v. Moore, 159 F.3d 1154 (9th Cir. 1998), or when one lawyer actively represents multiple individuals with inconsistent defenses, Cuyler v. Sullivan, 446 U.S. 335 (1980). Third, and more relevant to this discussion, prejudice will be presumed when an accused can claim legitimately that representation was so inadequate as to constitute a complete deprivation of counsel, United States v. Cronic, 466 U.S. 648 (1984), decided with Strickland. The Cronic exception to the Strickland standard applies to that small number of cases where there has either been: 1) a complete deprivation of counsel at a critical stage in the life of a criminal case; or 2) where counsel has been asked to provide representation in an unusually difficult situation (such as that which occurred in Powell v. Alabama, where a number of young black men accused of capital rape in a hostile southern town were assigned counsel moments before the start of the trial); or 3) where counsel fails "to subject the prosecution’s case to meaningful adversarial testing," such as by conceding guilt in closing arguments, United States v. Swanson, 943 F.2d 1070 (9th Cir. 1991).

Although Cronic clearly provides an opening for courts willing to more strictly scrutinize counsel’s performance, federal courts have been reluctant to step through it. That may be changing. Recent cases suggest that the federal bench is more willing to presume prejudice in cases where, just a few years ago, it would have dismissed claims of ineffectiveness. For starters, in 2002, the Supreme Court let stand the Fifth Circuit decision in Burdine that condemned as ineffective a defense attorney’s sleeping through parts of a capital trial. ( Cockrell v. Burdine, 535 U.S. 1120 (2002) (mem.).) There is no doubt that Burdine marks a significant change, since less than 10 years earlier the Court had denied relief to another defendant whose attorney had also slept through long portions of his client’s capital trial. ( McFarland v. Texas, 519 U.S. 1119 (1997) (mem.) (denying review of McFarland’s capital conviction).) Also last year, the First Circuit in Ouber v. Guarion, 293 F.3d 19 (1st Cir. 2002), held that an attorney who failed to call her client, after promising the jury that she would testify, was ineffective and that the verdict ought to be overturned despite the court’s inability to pinpoint exactly how the omission impacted the jury’s decision.

Courts may be hesitant to use the Cronic exception out of fear that such analysis will be impracticable, requiring appellate courts to spend more time scrutinizing trial records, or necessitating an unacceptable number of reversals. These worries are overstated. New York State has built a workable ineffective assistance of counsel jurisprudence on an approach very similar to what Cronic suggests. New York’s highest court has rejected the federal Strickland standard in favor of its slightly different concept of "meaningful representation," which focuses on the "‘fairness of the process as a whole rather than [any] particular impact on the outcome of the case.’" ( People v. Henry, 95 N.Y.2d 563, 565 (2000).)

So, for example, New York State’s intermediate appellate court has found that trial errors—counsel’s lack of familiarity with the rules of evidence, failure to review impeachment material, inability to effectively cross-examine witnesses, solicitation of inadmissible identification testimony during cross-examination, failure to object when inadmissible testimony was elicited on redirect, and misstatement in summation—can cumulatively deprive a person of meaningful representation without regard for whether that person would have been convicted in the absence of those errors. ( People v. Cortez, 296 A.D.2d 465, 745 N.Y.S.2d 467 (2002); see also People v. Gil, 285 A.D.2d 7, 729 N.Y.S.2d 121 (2001) (defense counsel’s decision to start trial on the day of arraignment, despite his failure to conduct discovery or motion practice, is ineffective); People v. Erick Brown, 2002 N.Y. AD Lexis 11617 (2d Dep’t 2002) (defendant was deprived of a fair trial by counsel’s failure to prepare for trial, his inability to effectively cross-examine the complaining witness, his unfamiliarity with the law regarding the admissibility of prompt outcry; and his indication in summation that he found the complaining witness’s testimony believable).) Thus, New York cases may assist the federal courts to implement the Cronic exception to the Strickland prejudice requirement.

Requiring counsel to investigate

In a parallel development, some federal courts are construing the Sixth Amendment to require defense counsel to conduct an investigation—at least in certain circumstances. This is most dramatically apparent in the recent Supreme Court case, Wiggins v. Smith, 123 S. Ct. 2527 (2003), which reversed a capital defendant’s death sentence because his counsel did not conduct a thorough investigation of his childhood trauma in preparation for the mitigation hearing.

Auspiciously, it doesn’t look as though the emerging consensus that the Constitution requires effective counsel to conduct fact investigation is going to be confined to death penalty jurisprudence. The Court of Appeals for the Second Circuit has suggested that counsel’s significant trial decisions must be supported by a sound strategy, and that a sound strategy can’t be formulated in the absence of an investigation. ( See Eze v. Senkowski, 321 F.3d 110 (2d Cir. 2003) (remanding to district court for factual hearing because the circuit was "unable to assess with confidence whether strategic considerations accounted for . . . counsel’s decisions.").) In that determination, the circuit relied upon Strickland’s admonition that "as a general matter, strategic choices made by counsel after a thorough investigation of the facts and law are ‘virtually unchallengeable,’ though strategic choices ‘made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgment support the limitation on investigation.’ " ( Strickland, 466 U.S. at 690–91.) Thus, counsel "has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary." ( Id. at 691.)

"Where counsel fails to make a reasonable investigation that is reasonably necessary to the defense, a court must conclude that the decision not to call an expert [for example] cannot have been based in strategic considerations and will thus be subject to review under Strickland’s prejudice prong." ( Pavel v. Hollins, 261 F.3d 210, 223 (2d Cir. 2001) (counsel ineffective in a child sexual abuse case where his failure to call a medical expert was based on an insufficient investigation.).) ( Also see Lindstadt v. Keane, 239 F.3d 191, 201 (2d Cir. 2001) (to the same effect); and Thomas v. Kuhlman, 255 F. Supp. 2d 99 ( E.D.N.Y. 2003) (counsel failed to investigate the scene of the crime and as a result did not realize that prosecution witnesses were mistaken in their testimony, and there could be no strategic rationale for such an omission.).)

Several circuits in addition to the Second have held similarly that counsel may be deemed ineffective for failure to conduct an investigation. ( See Williams v. Washington, 59 F.3d 673, 680–81 (7th Cir. 1995) (ineffective assistance, in part, for failure to investigate crime scene where doing so would have revealed evidence that, "given the layout of the home and the relatively crowded conditions, the alleged assault could not have taken place as claimed."); Washington v. Smith, 219 F.3d 620 (7th Cir. 2000) (counsel was ineffective when he failed to interview or subpoena alibi witnesses, neglected to read police reports, and did not present any semblance of a tactical reason for that minimal diligence); Matthew v. Abramajtys, 319 F.3d 780, 789 (6th Cir. 2003) (affirming the district court’s grant of petitioner’s habeas petition, because "[f]undamentally, the lawyer in this case, at best, occupied a space next to his client, but did not assist him. He did nothing to present potential alibi witnesses, whose testimony would have been quite useful, even if not conclusive.); Jennings v. Woodford, 290 F.3d 1000 (9th Cir. 2002) (where there was no basis to conclude that trial counsel’s failure to investigate and present evidence that might have defeated jury’s finding of intent and provided mitigation of the crime at sentence was tactical, court would find ineffective assistance of counsel.).)

I don’t mean to overstate the importance of what are still only a limited number of decisions, and I’m sure that some readers will think I am being unreasonably optimistic in characterizing the holdings as a development. Only time will tell.

Prospective systemic litigation gets new life

Another way for courts to enforce the Sixth Amendment and improve the quality of defense services is through the use of the injunction. Courts can order institutional change. Classic and dramatic examples of judge-ordered reform include transformation of the prison and educational systems. Thus, for years scholars have urged advocacy groups to undertake structural reform litigation to enforce the right to effective assistance of counsel. ( See Richard J. Wilson, Litigative Approaches to Enforcing the Right to Effective Assistance of Counsel in Criminal Cases, 14 N.Y.U. Rev. L. & Soc. Change 203, 216–17 (1986); Margaret H. Lemos, Note, Civil Challenges to the Use of Low-Bid Contracts for Indigent Defense, 75 N.Y.U. L. Rev. 1808 (2000); Note, Gideon ’s Promise Unfulfilled: The Need for Litigated Reform of Indigent Defense, 113 Harv. L. Rev. 2062 (2000).)

Of course, bringing systemic litigation is extremely difficult. The Abstention Doctrine prevents federal courts from intervening in state indigent defense systems, Younger v. Harris, 401 U.S. 37 (1971), and, until recently, state courts have been hesitant to insert themselves in what can be described as the legislative domain. ( See, e.g., Kennedy v. Carlson, 544 N.W.2d 1 (Minn. 1996) (Minnesota Court of Appeals rejected the Minnesota public defender’s request for additional funds so that his staff could provide services that comported with constitutional requirements.).) Nonetheless, judicial reluctance to intervene in the operation of local criminal court operations may be waning.

Confronted with a lawsuit brought on behalf of all criminal defendants and family court respondents represented by private assigned counsel, a trial court judge in New York has done more to improve the quality of criminal defense services than the legislature had accomplished in almost two decades. ( New York City Lawyers’ Association v. Pataki, 727 N.Y.S.2d 851 (Sup. Ct. N.Y. Cty. 2001) (denying the state’s motion to dismiss), aff’d, 294 A.D.2d 69, 742 N.Y.S.2d 16 (1st Dep’t 2002) and 2003 N.Y. Misc. LEXIS 453 (Feb. 2003) (granting plaintiff’s motion for a mandatory permanent injunction pending further action by the legislature).)

Some background is necessary to appreciate the significance of the NYCLA decision. Each of New York State’s 62 counties pays for and manages its own public defense. New York City (which includes five counties) provides defense services to poor people through what is known as a mixed delivery system. The Legal Aid Society is the primary defender, representing approximately 85 percent of all the indigent criminal matters. Small, boutique alternate defenders exist in each of the boroughs. Private attorneys are assigned to handle conflict cases through an assigned counsel plan.

New York City’s assigned counsel plans (ACP) have some excellent characteristics. Full-time administrators manage the plans. Attorneys seeking to join must meet certain qualifications. A screening committee reviews applications, adjudicates complaints, and has undertaken extensive recertification drives. Finally, staff provides continuing legal education tailored to the needs and schedules of the ACP attorneys, and circulates legal updates and information about investigators, experts, and alternative to incarceration options for clients. However, none of these structural advantages offsets the impact of resource deprivation. Fees paid to ACP attorneys were set by the state in 1986 at $40 an hour for in-court work and $25 an hour for out-of-court work and were not raised until 2003.

Combined with the high cost of doing business in New York City, ACP rates were driving attorneys off the criminal court and family panels. Those attorneys willing to take assigned work were handling too many cases. Further, the distinction between in- and out-of-court pay scales discouraged preparation, investigation, and legal research, as well as client contact.

The New York County Lawyers’ Association brought suit in state court claiming that the low rates violated the Sixth Amendment rights of those defendants and family court respondents who were represented by ACP attorneys. State Supreme Court Justice Lorenzo Suarez rejected the State’s argument that a decision to raise the rates would improperly interfere with the State’s sovereign ability to allocate funds. He relied on Luckey v. Harris, 860 F.2d 1012 (11th Cir. 1988), and the New York State standard for evaluating ineffective assistance of counsel, to find that the low rates were adversely impacting the ability of ACP attorneys to provide meaningful assistance, and ordered the state to raise its rates to $90 an hour for both in- and out-of-court work.

Justice Suarez reasoned that, while "ordinarily, federal claims of ineffective assistance are judged case by case, after conviction, and measured against the Strickland standard," (192 Misc. 2d at 429), in New York a different standard prevails. In New York

[w]hile the inquiry focuses on the quality of the representation provided to the accused, the claim of ineffectiveness is ultimately concerned with the fairness of the process as a whole rather than its particular impact on the outcome of the case. The purpose is to ensure that a defendant has the assistance necessary to justify society’s reliance on the outcome of the proceedings. Notably, New York is concerned as much with the integrity of the judicial process as with the issue of guilt or innocence.

( Supra at 431–32.) (citations omitted.)

The State of New York immediately appealed the decision, which stayed the court order raising the rates. However, the New York legislature didn’t wait to see what would happen, and voted to raise the assigned counsel rates to $75 for time spent in and out of court.

The New York Supreme Court’s willingness to intervene in the operation of the local criminal justice system and force a reallocation of funds may result from the growing consensus that public defense systems are inadequately protecting the rights of the accused.

Litigation designed to improve the quality of criminal defense services has been successful in other parts of the country, as well. In 1996, Connecticut’s public defender settled a lawsuit brought by the American Civil Liberties Union challenging the state indigent defense system. The settlement raised the fees for private attorneys accepting those indigent cases that the public defender is unable to handle and permitted a number of new hires in the office of the public defense. Five years ago, the ACLU settled a class-action lawsuit against Allegheny County, Pennsylvania, with a consent decree designed to overhaul the public defender’s office there.

This year in Georgia, the legislature has voted to create public defender offices in each of the state’s 49 counties. It is fair to say that the state legislature responded to relentless pressure from the Southern Center for Human Rights, which had filed a total of six lawsuits seeking systemic reforms. Litigation designed to improve the quality of criminal defense services is currently pending in Quitman County, Mississippi, and in Montana where the ACLU of Montana is alleges that indigent defense services in seven counties are constitutionally deficient. "The state has failed to provide the counties with the funding and guidance needed to administer an indigent defense program adequately," said Scott Crichton, executive director of the ACLU of Montana, in a news release. "Under these circumstances, even the most diligent attorneys cannot provide competent representation."

Court holds that public defender can be sued

Finally, there is another tool for courts desiring to enhance the quality of criminal defense services, albeit one that is bound to be unpopular in the defender community. Courts could make it easier to sue public defenders for malpractice—especially when administrative decisions contributed to it.

Although defenders aren’t guaranteed the same kind of immunity that shields judges, police, and prosecutors from civil liability for errors made in the course of fulfilling their responsibilities, courts have made it difficult for criminal defendants to sue their counsel. On the one hand, the U.S. Supreme Court has held that when a public defender is performing the traditional role of an individual attorney for an individual client, that lawyer is not a state actor and is thus not amenable to suit under the federal civil rights laws. ( Polk County v. Dodson, 454 U.S. 312 (1981).) On the other hand, it is equally difficult to sue a public defender under state malpractice tort theory. Not only are claims of legal malpractice cumbersome to bring and difficult to prove, but also many states treat individual public defenders as civil servants with individual immunity ( Scott v. City of Niagara Falls, 407 N.Y.S.2d 103, 105 (1978) (holding that there is "no valid reason to extend . . . immunity to state and federal prosecutors and judges and to withhold it from state-appointed and state-subsidized defenders.).) Further, damages won against public defender offices are often capped.

Commentators deplore these protections, which some see as an abridgment of the rights of the accused. ( See Harold Chen, Malpractice Immunity: An Illegitimate and Ineffective Response to the Indigent-Defense Crisis, 45 Duke L.J. 783 (1996); David Sadoff, The Public Defender as Private Offender: A Retreat from Evolving Malpractice Liability Standards for Public Defenders, 32 Am. Crim. L. Rev. 883 (Spring 1995); David J. Richards, The Public Defender Defendant: A Model Statutory Approach to Public Defender Malpractice Liability, 29 Val. U. L. Rev. 511 (Fall 1994).) Their voices are being heard.

In Nevada, an accused individual whose conviction had been reversed for ineffective assistance of counsel and whose indictment was subsequently dismissed—although not on grounds of innocence—brought suit in the federal district court against his individual public defender, the office that trained and supervised him, and the county that funded the office. ( Miranda v. Clark County, 319 F.3d 465 (9th Cir. 2003).) Miranda claimed that the chief defender required lie detector tests to be administered to all of his clients, and provided fewer investigative and defense resources to those who failed. Miranda alleged that this policy was not an isolated instance, but a deliberate pattern and policy—part of a general refusal to properly train and supervise lawyers. Miranda claimed that no investigation was conducted on his case as a result of the test results. ( Id. at 468.)

Affirming the district court determination, the Ninth Circuit found that Miranda’s allegations, if proven, would be sufficient to establish against the public defender office and the county a claim of deliberate indifference to constitutional rights, reachable under the leading Supreme Court decisions on state and municipal liability, such as Monell v. Department of Social Services, 436 U.S. 658 (1978).

I doubt there is another public defender office that uses polygraph tests as Nevada is alleged to have done. But all individual public defenders prioritize cases and allocate resources in some way. They must. No one can carry the caseloads that defenders shoulder without deciding which clients are going to get the most attention. Most public defender organizations provide little guidance to their staff about making those decisions and fail to review the decisions that are made. It seems entirely plausible that other innocent clients, upon release from jail, will sue for failing to investigate, to devote resources, or to train and evaluate staff. The Miranda v. Clark County decision condemned an affirmative policy as systemically ineffective, but there is no reason why another organization’s omissions or failures might not likewise be considered bureaucratic malfeasance establishing liability.

Exonerations of innocent individuals have taught the public and the courts to be more demanding of the police and the prosecutors. The police are already paying for careless work through larger jury verdicts and settlements. (In February 2003, the Seventh Circuit affirmed a $15 million jury verdict in favor of James Newsome, a wrongly convicted man.) The defense bar may find itself in the same position soon. The public will demand better performance from the defense bar, and the courts will enforce those demands.

Return to Table of Contents - Fall 2003

Return to Criminal Justice magazine home page