The Yale Law Journal


The Myth of Prosecutorial Accountability After Connick v. Thompson: Why Existing Professional Responsibility Measures Cannot Protect Against Prosecutorial Misconduct

25 Oct 2011


On March 29, 2011, the Supreme Court—by a vote of five to four—overturned a $14 million jury verdict in favor of John Thompson, a Louisiana man who spent fourteen years on death row because prosecutors withheld exculpatory blood evidence from his defense attorneys. Thompson had sued the Orleans Parish District Attorney’s Office based on a failure-to-train theory, arguing that the office had denied him due process of law through its deliberate indifference toward the need to train its attorneys in proper disclosure procedures. Thompson’s failure-to-train theory relied on Brady v. Maryland, a 1963 Supreme Court decision that requires prosecutors to share evidence with defendants in criminal cases when that evidence is “material either to guilt or to punishment.” The Connick Court, in an opinion authored by Justice Thomas, disagreed with Thompson’s argument. According to Justice Thomas’s majority opinion, a single Brady violation—i.e., a one-time failure to disclose “material” evidence—is insufficient to establish liability on a failure-to-train theory.

While seemingly narrow in its holding, Connick is significant because it forecloses one of the few remaining avenues for holding prosecutors civilly liable for official misconduct. The likelihood that a plaintiff will be able to prove the pattern of recurrent misconduct necessary to sustain a § 1983 action is remote. In the wake of Connick, then, advocates of enhanced prosecutorial accountability must look beyond civil liability in search of alternative mechanisms for combating misconduct.

One alternative is readily apparent from the Court’s Connick decision itself: state professional disciplinary procedures. In holding that district attorneys are reasonably entitled to rely on the “professional training and ethical obligations” of their subordinates, the Court noted that “[a]n attorney who violates his or her ethical obligations is subject to professional discipline, including sanctions, suspension, and disbarment.” Implicit in the Court’s reasoning is a belief that disciplinary procedures effectively deter prosecutorial misconduct. This position echoes the Court’s earlier holding in Imbler v. Pachtman, in which Justice Powell noted that “a prosecutor stands perhaps unique, among officials whose acts could deprive persons of constitutional rights, in his amenability to professional discipline by an association of his peers.”

In reality, prosecutors have rarely been subjected to disciplinary action by state bar authorities. This Essay asks why that is so and what may be done to make bar associations more responsive to allegations against prosecutors. Our findings, based on an investigation into the professional conduct rules and attorney discipline procedures of all fifty states, suggest that disciplinary systems as they are currently constituted do a poor job of policing prosecutors. Nonetheless, we argue that with a few modest reforms, grievance procedures can function as an effective deterrent to prosecutorial misconduct. In Part I, we briefly review the Connick decision, and in Part II we discuss the widespread problem of prosecutorial misconduct in the United States and the limited capacity of the civil and criminal justice systems to hold prosecutors accountable for their misdeeds. In Part III, we examine the current state of ethics rules and disciplinary procedures. We then conclude in Part IV with recommendations for enhancing disciplinary procedures with the aim of constructing an efficacious check on prosecutorial misconduct.

I. Connick v. Thompson

John Thompson’s execution date was only weeks away when an investigator working for his defense team found an exculpatory blood-evidence report in an obscure file buried in the New Orleans Police Crime Laboratory. By then, Thompson had already spent fourteen years on death row. When Thompson was first tried in 1985, prosecutors in the Orleans Parish District Attorney’s Office had strategically pursued the attempted robbery charge prior to the murder charge, to dissuade Thompson from testifying at his murder trial for fear of his criminal record being introduced to the jury. Prosecutors also neglected to inform Thompson’s public defender that the perpetrator of the robbery had left his blood on the pants leg of one of the victims. A test performed on a swatch of fabric taken from the pants conclusively established that the perpetrator’s blood was type B; Thompson’s blood, which prosecutors never tested, is type O. Years later, when Thompson was finally cleared of the robbery charge and free to testify on his own behalf, the jury at his retrial for murder acquitted him after only thirty-five minutes of deliberation.

Five prosecutors were implicated in the failure to turn over the exculpatory blood evidence. In 1994, nearly ten years after the misconduct occurred, Gerry Deegan, an assistant district attorney on the armed robbery case, confessed to a friend and former prosecutor, Michael Riehlmann, that he had “intentionally suppressed blood evidence.” Three other prosecutors—Bruce Whittaker, James Williams, and Eric Dubelier, all of whom worked with Deegan—knew of the blood evidence and failed to turn it over to Thompson’s attorneys.

Prior to trial, Thompson’s attorneys made a motion to inspect all material evidence and scientific reports and all materials favorable to the defendant. In her dissent from the majority opinion in Connick, Justice Ginsburg outlined three ways in which the prosecution’s response to Thompson’s motion “fell far short of Brady compliance.” First, Dubelier’s response stated that “[i]nspection [was] to be permitted,” but the swatch was signed out of the property room the next day and was not returned until a week later, one day before Thompson’s robbery trial. Second, after the swatch was initially returned, Deegan checked it out again almost immediately, on the morning of the first day of trial. The swatch itself, however, was never produced at trial or returned to the evidence room. To this day, it has never been recovered; Thompson’s investigator could only locate a microfiche of the lab report. Finally, either Dubelier or Whittaker ordered a pretrial test of the swatch to be rushed; Whittaker received the results, addressed to him, and immediately put them on Williams’s desk. Though the test conclusively established the perpetrator’s blood type, the prosecution never turned it over to the defense.

After the discovery of the exculpatory evidence and his subsequent acquittal, Thompson sued Harry Connick, Sr., the District Attorney of Orleans Parish, alleging that Connick’s deliberate indifference to an obvious need to train the prosecutors in his office caused theprosecutors’ failure to turn over exculpatory evidence in Thompson’s case. It became apparent from evidence presented at trial that Connick’s office offered no formal training to its prosecutors regarding Brady evidence. Connick himself misstated Brady’s requirements in his testimony, as did the other prosecutors questioned. Connick also conceded that he stopped reading legal opinions after he came to office in 1974 and was therefore unaware of important Supreme Court rulings concerning the scope of Brady obligations. Shortly after Connick’s retirement, “a survey of assistant district attorneys in the Office revealed that more than half felt they had not received the training they needed to do their jobs.” Based on this evidence, a jury in the Eastern District of Louisiana awarded Thompson $14 million in damages. The verdict was affirmed by the Fifth Circuit, then reheard and reaffirmed by an equally divided en banc court.

In overturning the Fifth Circuit, the Supreme Court put its full faith in the efficacy of professional standards and disciplinary procedures. Notably, the Supreme Court recognized that Connick knew both that prosecutors in his office encountered Brady issues frequently and that “erroneous decisions regarding Brady evidence would result in constitutional violations.” But a “licensed attorney making legal judgments, in his capacity as a prosecutor,” the Court asserted, “simply does not present . . . [a] ‘highly predictable’ constitutional danger.” In reaching this conclusion, the Court determined that professional training provided an adequate safeguard against constitutional violations. Justice Thomas, writing for the majority, specifically referenced lawyers’ education in law school, their completion of the bar exam, continuing education requirements, character and fitness standards, on-the-job training from more experienced attorneys, and the potential imposition of professional discipline as reasons for rejecting single instance failure-to-train liability.

Although Connick purported to answer a narrow question—“whether a district attorney’s office may be held liable under § 1983 for failure to train based on a single Brady violation”—the holding has great significance for those who hoped to see a move toward accountability for prosecutors and their offices, as well as for those who favor prosecutors’ relative freedom from traditional disciplinary measures. In rejecting Thompson’s jury award, the Supreme Court reaffirmed its commitment to prosecutorial immunity, sharply limiting one of the few remaining avenues of redress for prosecutorial misconduct. In the next Part, we provide a broader overview of the problem of prosecutorial misconduct in the United States and the troubling lack of accountability for such misconduct. The history of prosecutorial immunity in particular demonstrates that, in the wake of Connick, state bar disciplinary procedures stand as one of the few—and perhaps the only—means of holding prosecutors accountable for gross misconduct.

II. Prosecutorial Misconduct and Immunity in the United States

A. Prosecutorial Misconduct: The Scope of the Problem

Several empirical problems hamper efforts to provide an accurate assessment of prosecutorial misconduct in the United States. First, prosecutors who engage in willful misconduct presumably do not want to be discovered and therefore take steps to conceal their misdeeds. Even a scrupulous prosecutor who witnesses a colleague engage in misconduct may nevertheless fail to report it for fear of professional repercussions.

Second, prosecutors’ offices enjoy considerable autonomy in shaping their internal policies. Although judicial oversight should theoretically check this autonomy, courts are generally loath to interfere with the inner workings of a coordinate branch of government. Likewise, individual prosecutors exercise almost unlimited discretion over whom to prosecute and which offenses to charge. Pretrial hearings ostensibly exist to cabin these powers but in practice rarely operate as an effective safeguard. The lack of any external oversight of prosecutors’ offices creates an environment in which misconduct can go undetected and undeterred.

Third, the vast majority of known instances of prosecutorial misconduct come to light only during the course of a drawn-out trial or appellate proceeding. John Thompson’s ordeal is illustrative: the blood evidence that ultimately exculpated Thompson was obtained at the eleventh hour through the “chance discovery” of a lone investigator hired by his defense team. But most criminal cases in the United States result in plea bargains, which are rarely the subject of extensive investigation or judicial review, creating a heightened risk of undetected prosecutorial misconduct in the plea bargaining context.

Finally, those in the best position to report misconduct—namely judges, other prosecutors, and defense attorneys and their clients—are often disincentivized from doing so for both strategic and political reasons. From the defendant’s perspective, there is little to gain from filing a bar complaint and much to lose. As one state judge has written:

  It flies in the face of reason to expect a defendant to risk a prosecutor’s actual or imagined displeasure by instituting proceedings that cannot directly benefit him. The defendant may not unreasonably believe such action will adversely affect his case in subsequent proceedings . . . or his later chances for parole.  

In other words, a bar complaint could itself negatively impact the outcome of ongoing litigation, if the prosecutor’s need to defend against disciplinary proceedings, or simple resentment at being reported to the authorities, results in less favorable treatment of the defendant. From the defense attorney’s perspective, there is little time for bar complaints when trying a case or handling an appeal. These attorneys are also understandably reluctant to turn in their colleagues, especially given their ongoing professional relationships.

What little evidence we do have indicates that prosecutorial misconduct is a serious problem. A 2003 study by the Center for Public Integrity, for instance, found over two thousand appellate cases since 1970 in which prosecutorial misconduct led to dismissals, sentence reductions, or reversals. Another study of all American capital convictions between 1973 and 1995 revealed that state post-conviction courts found “prosecutorial suppression of evidence that the defendant is innocent or does not deserve the death penalty” in one in six cases where the conviction was reversed. Other scholars and journalists have also documented widespread prosecutorial misconduct throughout the United States.

Available statistics significantly underreport the extent of prosecutorial misconduct, not only because of the empirical challenges discussed above, but also because courts have embraced a “harmless error” standard when reviewing criminal convictions. In order to win a reversal, a defendant must not only prove misconduct, but must also show that the misconduct substantially prejudiced the outcome of his or her trial. Courts can therefore avoid making a finding of misconduct altogether by finding that the alleged error, even if proven, was harmless. By reducing the likelihood of reversal, the harmless error standard substantially weakens one of the primary deterrents to prosecutorial misconduct. Knowing that “minor” misconduct is unlikely to jeopardize a conviction on appeal, prosecutors may be more likely to bend the rules in the pursuit of victory.

There is an obvious need for an effective check on prosecutorial misconduct. Yet, as this Essay will show, no such check currently exists. The next Section reviews five potential means of providing accountability for prosecutorial misconduct and explains how each has been rejected by the courts, left unutilized, or diluted to the point of total ineffectiveness.

B. Prosecutorial Immunity and the Decline of Accountability

In the United States, five main avenues have been explored as potential mechanisms to punish the official misbehavior of prosecutors. As this Section explains, two of these—common-law personal tort liability and personal tort liability under 42 U.S.C. § 1983—have been explicitly rejected by the Supreme Court. The third form of civil liability—municipal liability under § 1983—was, prior to Connick, generally recognized as a viable mechanism for keeping prosecutors’ offices in check. The potential for such liability was thought necessary in part because criminal punishment for prosecutorial misconduct, the fourth avenue, is almost never been utilized in practice. And, as we shall demonstrate, the final avenue—professional responsibility measures—is almost always ineffective in the prosecutorial misconduct context. This is precisely why Connick’s narrowing of municipal liability is so troubling. Indeed, it calls for reform of professional discipline systems to enable them to hold prosecutors accountable in a way that weakened civil remedies cannot.

Since the nineteenth century, American courts have recognized that prosecutors are immune from tort liability for actions performed in the line of duty. After decades of general adherence to this principle by state courts, the Supreme Court recognized prosecutors’ common-law tort immunity from suits for malicious prosecution in 1927, affirming per curiama decision of the Court of Appeals for the Second Circuit which held that “[t]he immunity is absolute, and is grounded on principles of public policy.” The purposes underlying prosecutorial immunity, as stated by the Supreme Court, are “concern that harassment by unfounded litigation would cause a deflection of the prosecutor’s energies from his public duties, and the possibility that he would shade his decisions instead of exercising the independence of judgment required by his public trust.”

While general tort liability for official misconduct by prosecutors has been regarded as unwise as a matter of policy, the specific issue of prosecutorial liability under 42 U.S.C. § 1983 has a more dynamic, contentious, and recent history. Congress enacted § 1983 during Reconstruction as part of an effort to permit federal courts to supervise compliance with the Fourteenth Amendment, particularly in former Confederate states. The statute creates a cause of action for damages or equitable relief against “[e]very person who, under color of” state law, “subjects, or causes to be subjected, any citizen of the United States or other person within the jurisdiction thereof to the deprivation of any rights, privileges, or immunities secured by the Constitution and laws.” As Justice Douglas wryly noted in his dissent from the Court’s opinion in Pierson v. Ray, in which the majority held that state judges are absolutely immune from damage suits under § 1983, “[t]o most, ‘every person’ would mean every person, not every person except judges.” Yet the Supreme Court extended Pierson’s absolute judicial immunity to state prosecutors in Imbler v. Pachtman.Despite the objections of commentators who note that § 1983’s very purpose was to provide an otherwise unavailable tort remedy for federal constitutional violations committed through the ultra vires abuse of state law power, the Court held that § 1983 “is to be read in harmony with general principles of tort immunities and defenses rather than in derogation of them.” Thus, the Court applied the longstanding policy of prosecutorial immunity to § 1983 interpretation. The Imbler Court expressly held that prosecutors are absolutely immune from § 1983 damage suits alleging Brady violations.

Perhaps the lack of a personal civil remedy against misbehaving prosecutors would be less consequential if other effective remedies were available. Municipal liability—the avenue for relief advanced by Thompson in Connick—has been considered one such alternative. The Supreme Court’s decision in Monell v. Department of Social Services overturned part of a previous decision, Monroe v. Pape, which had held that Congress did not intend for § 1983 to create liability for constitutional violations on the part of municipalities. In Monell, the Court delved deeply into the history of the statute and concluded:

  Congress did intend municipalities and other local government units to be included among those persons to whom § 1983 applies. Local governing bodies, therefore, can be sued directly under § 1983 for monetary, declaratory, or injunctive relief where, as here, the action that is alleged to be unconstitutional implements or executes a policy statement, ordinance, regulation, or decision officially adopted and promulgated by that body’s officers. Moreover, although the touchstone of the § 1983 action against a government body is an allegation that official policy is responsible for a deprivation of rights protected by the Constitution, local governments, like every other § 1983 “person,” by the very terms of the statute, may be sued for constitutional deprivations visited pursuant to governmental “custom” even though such a custom has not received formal approval through the body’s official decisionmaking channels.  

Municipal defendants enjoy neither absolute nor qualified immunity, and—while they cannot be sued under a respondeat superiortheory—they are liable when a municipal policy or custom causes a constitutional injury. Before Connick, it appeared that municipal liability could exist where a supervising prosecutor had failed to train line prosecutors regarding their constitutional obligations. However, Connick suggests that plaintiffs will have great difficulty proving that a supervising prosecutor acted as a policymaker in failing to train subordinates—the showing necessary to obtain a remedy under § 1983. Because civil rights plaintiffs must establish that their rights were violated as a result of an official policy or custom, Connick’s holding that a failure-to-train showing can only be made by demonstrating a pattern of violations—information that might be difficult for individual plaintiffs to access—will make such suits exceedingly difficult to win. Moreover, the Court appeared to signal in Connick that a pattern of extremely similar specific violations, rather than overall misconduct, would be necessary to establish municipal liability. Thus, the class of facts potentially giving rise to municipal liability in prosecutorial misconduct cases is significantly narrowed after Connick.

Alternatives to civil liability have proven no more successful. In the course of upholding official immunity, the Supreme Court in Imbler wrote that prosecutorial misconduct “is reprehensible, warranting criminal prosecution as well as disbarment,” rather than civil damages. Unfortunately, history has not borne out the notion that criminal sanctions or bar discipline are effective tools for deterring and punishing prosecutorial misconduct. In the popular imagination, the idea of criminal liability sometimes appears as a kind of poetically just punishment for unethical prosecutors. As John Thompson wrote in an op-ed published shortly after the Supreme Court’s decision: “I don’t care about the money. I just want to know why the prosecutors who hid evidence, sent me to prison for something I didn’t do and nearly had me killed are not in jail themselves.” However, criminal sanctions for prosecutors who violate Brady are exceedingly rare. The 1999 Illinois trial of the so-called “DuPage Seven,” police officers and prosecutors accused of perjury and obstruction of justice for allegedly framing an innocent defendant in a capital murder case, appears to be the first time in American history that a felony prosecution of former prosecutors for misconduct reached the verdict stage. All of the defendants, however, were acquitted. Although it is difficult to comprehensively determine exactly how many prosecutors have been subject to criminal sanctions for official misconduct throughout U.S. history, the number is surely extremely low. Criminal sanctions are most likely rare because they are seen as an overly harsh punishment for “technical” errors made by people with demanding and stressful jobs. Moreover, the federal criminal statute that allows for punishment of prosecutorial misconduct that violates a defendant’s civil rights—18 U.S.C. § 242—requires that the misconduct be willful, rendering the government’s burden in pursuing criminal punishment for unethical prosecutors under the law daunting and making criminal sanctions available only for a small fraction of instances of misconduct.

Similarly, bar discipline procedures have not proved a fruitful sanction for deterring prosecutorial misconduct. Many state bar disciplinary systems barely seem to contemplate prosecutorial misconduct as a cognizable complaint, focusing instead on fee disputes and failure to diligently pursue a client’s claim. Indeed, only one of the five prosecutors responsible for violating John Thompson’s constitutional rights has ever been disciplined by the attorney grievance system in place in Louisiana. Ironically, that prosecutor is Michael Riehlmann, the only one of the five who was not directly involved in prosecuting Thompson’s case or implicated in any of the Brady violations that occurred and the only attorney to ever report the violations to Louisiana’s Office of Disciplinary Counsel (ODC). Five years after Gerry Deegan had confessed to him about suppressing the blood evidence, Riehlmann reported his conversation with Deegan to ODC after Thompson’s attorneys inquired about his knowledge of the newly discovered crime lab report. The Louisiana Attorney Discipline Board subsequently recommended that Riehlmann’s law license be suspended for six months because he failed to report Deegan’s confession within a “reasonable time” and this failure was “prejudicial to the administration of justice.” The Supreme Court of Louisiana, however, determined that Riehlmann’s behavior was “merely negligent” and that a public reprimand was the appropriate sanction.

The lack of action taken in Thompson’s case is emblematic of the broader failure of state bar disciplinary procedures to punish those directly engaged in prosecutorial misconduct. The following Part first reviews the ethics rules and disciplinary procedures of all fifty states, highlighting the pervasive flaws in those rules and procedures; and, second, explains how the existing disciplinary regime is ineffective at addressing prosecutorial misconduct.

III. Disciplining Prosecutors?

Given the Supreme Court’s repeated endorsement of professional discipline as the appropriate vehicle for addressing allegations of prosecutorial misconduct, one might suppose that state bar agencies frequently sanction prosecutors. In fact, prosecutors are rarely held accountable for violating ethics rules. In 1999, Chicago Tribune reporters Maurice Possley and Ken Armstrong identified 381 homicide cases nationally in which Brady violations produced conviction reversals. Not a single prosecutor in those cases was publicly sanctioned. Four years later, a study by the Center for Public Integrity found 2012 appellate cases between 1970 and 2003 in which prosecutorial misconduct led to dismissals, sentence reductions, or reversals.Yet prosecutors faced disciplinary action in only forty-four of those cases, and seven of these actions were eventually dismissed. The most recent study indicates that depressingly little has changed since 2003, at least in California. The Northern California Innocence Project identified 707 cases between 1997 and 2009 in which courts made explicit findings of prosecutorial misconduct, 159 of which were deemed harmful. The Project’s review of the public disciplinary actions reported in the California State Bar Journal, however, revealed a mere six—out of a total of 4741—that involved prosecutorial misconduct.

As these studies indicate, infrequent punishment of prosecutors cannot be blamed on a paucity of discoverable violations. Even when judicial findings of misconduct result in conviction reversals, disciplinary sanctions are almost never imposed against the offending prosecutor. This Part endeavors to explain why prosecutors are rarely sanctioned by state bar authorities.

Our conclusions derivee from a comprehensive survey of the ethical rules and disciplinary practices of all fifty states. As part of the survey, we compiled comparative data on each state’s rules of professional conduct and rules of disciplinary procedure. In addition, we consulted all fifty discipline agency websites and conducted telephone interviews with bar personnel to glean additional information about the complaint process. We further supplemented our research with statistical data compiled by the American Bar Association as part of its 2009 Survey on Lawyer Discipline Systems.

The data from our survey suggest four broad causes for the breakdown in attorney discipline systems with respect to prosecutors. First, the ethical rules that govern prosecutorial behavior fail to proscribe most forms of prosecutorial misconduct. Second, the procedures governing attorney discipline systems afford complainants too few rights and administrators too much discretion. Third, those who are in the best position to discover prosecutorial misconduct—judges, prosecutors, and defense attorneys—routinely fail to report it. Fourth, overlapping policing mechanisms create confusion about the appropriate locus of disciplinary authority.

A. Model Rule 3.8: A Weak Check on Prosecutorial Misconduct

Ethics rules create legally enforceable obligations that can shape norms of behavior. Accordingly, this Part begins by discussing the ethical obligations of prosecutors as defined by Rule 3.8 of the American Bar Association’s Model Rules of Professional Conduct. In the next Section, we consider the degree to which individual states have deviated from this Model Rule in enacting their own rules of professional conduct. In the final Section, we examine the existing disciplinary mechanisms used by states to enforce their rules of professional conduct.

For over one hundred years, states have looked to the ABA for guidance when constructing their local rules for attorney discipline. The Model Rules of Professional Conduct, first promulgated in 1983 and substantially revised in 2002, have proven especially influential. Every state save California has adopted attorney ethics codes that substantially mirror the Model Rules.

The Model Rules generally do not distinguish between private attorneys and prosecutors. All lawyers are expected to conduct themselves in accordance with its general provisions. Model Rule 3.8 is exceptional, however, in that it defines certain “special” ethical duties unique to prosecutors, including the obligation not to pursue charges against an individual in the absence of probable cause and the affirmative responsibility to disclose exculpatory evidence in a timely fashion. While other Model Rule provisions apply equally to prosecutors and private attorneys, Rule 3.8 is the only rule that directly addresses the prosecutorial function. Consequently, its provisions serve as a baseline for measuring prosecutorial misconduct.

Rule 3.8 embodies Justice Sutherland’s general admonition in Berger v. United States that “while [a prosecutor] may strike hard blows, he is not at liberty to strike foul ones.” Commentary on the Rule underscores this admonition by noting that a prosecutor’s role is to be a “minister of justice and not simply . . . an advocate.” Accordingly, the Rule places both negative and affirmative responsibilities on prosecutors. A prosecutor shall:

  (a) refrain from prosecuting a charge that the prosecutor knows is not supported by probable cause;  
  (b) make reasonable efforts to assure that the accused has been advised of the right to, and the procedure for obtaining, counsel and has been given reasonable opportunity to obtain counsel;  
  (c) not seek to obtain from an unrepresented accused a waiver of important pretrial rights, such as the right to a preliminary hearing;  
  (d) make timely disclosure to the defense of all evidence or information known to the prosecutor that tends to negate the guilt of the accused or mitigates the offense, and, in connection with sentencing, disclose to the defense and to the tribunal all unprivileged mitigating information known to the prosecutor, except when the prosecutor is relieved of this responsibility by a protective order of the tribunal;  
  (e) not subpoena a lawyer in a grand jury or other criminal proceeding to present evidence about a past or present client unless the prosecutor reasonably believes:  
  (1) the information sought is not protected from disclosure by any applicable privilege;  
  (2) the evidence sought is essential to the successful completion of an ongoing investigation or prosecution; and  
  (3) there is no other feasible alternative to obtain the information;  
  (f) except for statements that are necessary to inform the public of the nature and extent of the prosecutor’s action and that serve a legitimate law enforcement purpose, refrain from making extrajudicial comments that have a substantial likelihood of heightening public condemnation of the accused and exercise reasonable care to prevent investigators, law enforcement personnel, employees or other persons assisting or associated with the prosecutor in a criminal case from making an extrajudicial statement that the prosecutor would be prohibited from making under Rule 3.6 or this Rule.  
  (g) When a prosecutor knows of new, credible and material evidence creating a reasonable likelihood that a convicted defendant did not commit an offense of which the defendant was convicted, the prosecutor shall:  
  (1) promptly disclose that evidence to an appropriate court or authority, and  
  (2) if the conviction was obtained in the prosecutor’s jurisdiction,  
  (i) promptly disclose that evidence to the defendant unless a court authorizes delay, and  
  (ii) undertake further investigation, or make reasonable efforts to cause an investigation, to determine whether the defendant was convicted of an offense that the defendant did not commit.  
  (h) When a prosecutor knows of clear and convincing evidence establishing that a defendant in the prosecutor’s jurisdiction was convicted of an offense that the defendant did not commit, the prosecutor shall seek to remedy the conviction.  

On its face, Rule 3.8 appears to justify the Supreme Court’s confident assertion that a “well-developed and pervasive mechanism” exists for policing prosecutorial misconduct. The Rule addresses many of the prosecutor’s most important ethical duties, including those related to his charging discretion in (a), discovery obligations in (d), subpoena power in (e), duty to inform the public in (f), and review of wrongful conviction claims in (g) and (h). Moreover, in some cases the Rule imposes obligations on prosecutors broader than those required by constitutional case law or rules of criminal procedure. For instance, under Brady a prosecutor is only required to produce evidence “upon request” that he determines is “material either to guilt or to punishment.” Rule 16 of the Federal Rules of Criminal Procedure similarly confines a federal prosecutor’s discovery obligations to the production of Brady or Giglio evidence “upon a defendant’s request.” By contrast, Rule 3.8(d) obligates prosecutors to voluntarily turn over all favorable evidence and to do so in a timely manner. In this way, Rule 3.8(d) is more rigorous than Brady’s material standard by requiring disclosure of exculpatory or mitigating evidence regardless of whether the favorable evidence is dispositive of the ultimate issue of guilt. As the ABA noted in a formal advisory opinion interpreting Rule 3.8, section (d) “requires prosecutors to steer clear of the constitutional line, erring on the side of caution.”

While Rule 3.8 might expose prosecutors to a heightened standard of conduct in theory, the Rule’s vague terminology undermines its efficacy and enforceability in practice. Rule 3.8(d) exemplifies this problem. Neither the text of that provision nor the accompanying commentary explains the proper standard for determining whether evidence is “favorable” to an accused. Likewise, the rule provides little guidance regarding the knowledge and timeliness requirements. In its formal advisory opinion, the ABA interpreted evidence that is “known to the prosecutor” to mean evidence of which the prosecutor has actual, rather than constructive, knowledge. Consequently, according to the ABA’s interpretation, “Rule 3.8(d) does not establish a duty to undertake an investigation in search of exculpatory evidence.” This interpretation greatly limits the Rule’s prophylactic potential and actually imposes an ethical standard below the constitutional minimum. As the Supreme Court explained in Kyles v. Whitley, under Brady, “the individual prosecutor has a duty to learn of any favorable evidence known to the others acting on the government’s behalf in the case, including the police.” But the ABA’s interpretation would permit a prosecutor to pursue a conviction without having familiarized himself with the most basic aspects of the case, such as the arresting officer’s police report and witness statements. Even if a prosecutor did read these materials and in doing so discovered certain inconsistencies, it is not clear under the ABA’s interpretation of the Rule that he would be ethically bound to undertake further investigation.

Rule 3.8’s prescriptive force is also greatly diminished by its failure to address many important aspects of the prosecutorial function. Over ninety percent of federal criminal prosecutions result in guilty pleas, yet the Model Rule nowhere explains how prosecutors should conduct themselves in plea negotiations. Rule 3.8(a) obliquely addresses the issue of charging discretion by urging that a prosecutor should “refrain from prosecuting a charge that the prosecutor knows is not supported by probable cause.” But the probable cause determination is a minimal standard that is typically decided by the grand jury; a body that, as the popular saying goes, could be convinced to indict a ham sandwich. Furthermore, the ethics rules do not prohibit a prosecutor who wishes to gain leverage in plea negotiations from filing a charge that he has no intention of bringing to trial.

Returning to the hypothetical posited above, suppose a prosecutor does undertake an investigation into inconsistent witness statements and learns of information unquestionably favorable to the defense, such as a disagreement between two primary witnesses over the defendant’s race. Under the Model Rules, a prosecutor who is nonetheless convinced of the defendant’s guilt is arguably under no obligation to present this information to the grand jury. Nor is he necessarily obligated to disclose the information to defense counsel during plea negotiations. Although in its formal opinion the ABA interprets “timely” to mean “as soon as reasonably practical,” that is not enforceable. A prosecutor looking to obtain a tactical advantage during plea negotiations may make a calculated decision to interpret “timely” as meaning any time prior to trial.

In sum, Model Rule 3.8 promises on its face more than it delivers in practice. While there are many instances of prosecutorial misconduct that clearly fall within its ambit, the Rule fails to address some of the more significant aspects of the prosecutor’s justice-seeking role. As one commentator has aptly noted, “there is no principled reason for a disciplinary code to include only the particular provisions now included in Model Rule 3.8.” By failing to cover the full scope of prosecutorial misconduct, Model Rule 3.8 offers states a flawed template upon which to base their own ethics rules.

B. Diluting Rule 3.8 at the State Level

State disciplinary authorities have the potential to rein in unethical behavior by prosecutors. They can only perform this function, however, if states adopt ethics rules with bite. This Section describes the inconsistent and incomplete implementation of Rule 3.8 or other similar provisions by local disciplinary authorities. The failure of many states to adequately define the special role of a prosecutor in their rules casts doubt on the Supreme Court’s optimism about professional discipline’s potential to check prosecutorial misconduct.

While every state save California has adopted a version of Model Rule 3.8, our research shows that few have gone beyond its minimal standards. Many states, in fact, have compounded Rule 3.8’s weaknesses by adopting watered-down versions of the Rule that omit or materially alter its most substantive provisions. States have also been slow historically in adopting strengthening amendments to the Rule. This trend has continued with the two most recent amendments promulgated in 2008, provisions (g) and (h), both of which focus much-needed attention on the steps a prosecutor must take when confronted with credible evidence of a convicted person’s innocence. The failure of many states to ratify these and other amendments in a timely fashion, together with the substantive deviations mentioned above, has resulted in a patchwork of ethics rules that lacks rhyme or reason.

The table below offers a visual depiction of the degree to which states have adopted Model Rule 3.8’s various provisions. The data paint a decidedly mixed picture of state compliance. Only one state, Idaho, has adopted Model Rule 3.8 in its entirety. While specific provisions garner nearly unanimous approval, others have proven less popular. Rule 3.8(g) and (h), which both deal with wrongful convictions, were approved by the ABA’s House of Delegates in February of 2008. However, as explained below, the slow pace of states’ adoption of those provisions is itself indicative of a ratification process that is dysfunctional.

Figure 1.

State Adoption of Model Rule 3.8 Provisions


As Figure 1 demonstrates, every state that follows the Model Rules has adopted some version of sections (a) and (d). This reflects the fact that (a) and (d) comprised the entirety of Rule 3.8 as first promulgated in 1969 in the predecessor to the Model Rules. States’ modifications to these provisions, however, have not followed a consistent pattern. Divergence between North Dakota and South Dakota in their respective adoptions of Rule 3.8(d) illustrates the point. North Dakota moderately strengthened the provision by clarifying that the requirement of timely disclosure means disclosure that occurs “at the earliest practical time.” South Dakota, on the other hand, kept the Model Rule’s vague timeliness requirement and replaced its emphasis on broader disclosure with Brady’s considerably less demanding standard.

Provisions (b) and (c), two of the first amendments to the Rule, have also been widely adopted. Yet, such widespread adoption may be of little consequence, as the importance of these provisions has been questioned by commentators. Rule 3.8(b) requires a prosecutor to “make reasonable efforts” to ensure a defendant is advised of his right to counsel, and 3.8(c) prohibits a prosecutor from seeking a waiver of that defendant’s pretrial rights, including the right to a preliminary hearing. Defendants, however, are normally advised of these rights during their first appearance before a judge. Neither provision is therefore likely to arise during the ordinary course of a prosecutor’s work. One state—Wisconsin—has, in fact, adopted versions of provisions (b) and (c) that impose a far more substantive standard on prosecutors by requiring that they identify their “role and interest in the matter” when questioning a defendant in addition to apprising him of his right to counsel. However, unless the ABA opts to update its rule to reflect these modifications, other states are unlikely to follow Wisconsin’s lead.

Rule 3.8(e), which concerns intrusions into the lawyer-client relationship through the use of lawyer subpoenas, has only been adopted in full or modified form by thirty-two states. Since its introduction in 1990, Rule 3.8(e) has stirred substantial controversy. The Justice Department, whose attorneys’ aggressive use of the subpoena power prompted the provision in the first place, immediately expressed its disagreement with 3.8(e) by seeking court rulings exempting federal prosecutors from its reach. After the Third Circuit issued a decision endorsing the Justice Department’s position, the ABA’s House of Delegates voted to remove a clause from 3.8(e) that required prosecutors to obtain a judicial order before issuing a subpoena.Even with this amendment, many states have opted to forego the new rule.

Finally, provisions (g) and (h) have only been adopted by five states since their introduction in 2008. These important provisions extend a prosecutor’s Brady obligations to evidence of non-guilt that comes to light after trial. The story of their enactment by the ABA and subsequent implementation by individual states is simultaneously encouraging and troubling. On the positive side, Rule 3.8(g) and (h) originated from a proposal made by the Bar of the City of New York to the New York State Bar. This grassroots approach stands in stark contrast to the normal mechanism by which Rule 3.8 is amended. Ordinarily, revisions are made by an ethics committee whose members boast no particular expertise in criminal justice matters. For instance, in preparation for the new millennium, the ABA instituted a major reform initiative called Ethics 2000, the object of which was to encourage state uniformity in rule adoption as well as to update those rules to better reflect the pace of technological change. Despite a report identifying Rule 3.8’s shortcomings, however, the Ethics Commission made only one minor alteration—one that arguably weakened the Rule. The Ethics 2000 reforms did have the salutary effect, however, of causing state bar associations to revisit their ethics rules. When the New York State Bar turned its attention to that task in 2005, its members found themselves troubled by a spate of wrongful convictions uncovered by the Innocence Project. Rule 3.8(g) and (h) subsequently grew out of a reasoned debate among New York prosecutors, defense attorneys, and judges about what obligations prosecutors should have when confronted with new evidence that raises credible doubts about the validity of a conviction. In further contrast to the way amendments to Rule 3.8 are normally adopted, provisions (g) and (h) were proposed to the ABA’s House of Delegates by the body’s Criminal Justice Section rather than its ethics committee.

The process of Rule 3.8(g) and (h)’s adoption, however, should also serve as a cautionary tale. To begin with, New York never adopted the rule that its own bar agency proposed. This is because state bar associations only have the power to propose rules; each state’s highest court possesses the ultimate authority to issue rules governing attorney behavior. In the abstract, there may be both substantive and symbolic reasons that justify this separation; however, in actuality, New York’s highest court chose to reject the proposed changes without offering a single reason for its decision.

The slow pace of Rule 3.8(g) and (h)’s adoption offers a second cause for concern. To date, only five states have adopted the provisions in full or modified form. Eleven other states are currently considering amending their versions of Rule 3.8. The remainder, thirty-four states in total, have taken no action. The lesson appears to be that piecemeal ethics reforms are unlikely to garner significant attention from state bar associations. Indeed, the only reason the newest amendments have received as much attention as they have is because some states are still in the process of implementing the Ethics 2000 reforms. The poor track record of amendment adoptions prior to Ethics 2000 is further evidence that rulemaking inertia may be the biggest stumbling block to meaningful reform efforts.

C. Disciplinary Systems

The corollary to the ethics rules are the disciplinary systems established to enforce those rules. Without consistent enforcement by the bodies charged with overseeing attorney discipline, ethics rules are little more than empty promises. This Section therefore catalogs the various features of all fifty state disciplinary systems in an effort to explain the lax enforcement of prosecutorial ethics rules. In examining these systems to better understand why they fail to discipline prosecutors, we highlight the wide divergence between state disciplinary systems in terms of their transparency and responsiveness. Many states actively discourage potential grievance filers by erecting procedural barriers like statutes of limitations, notarized document requirements, or mandatory referral programs. Moreover, disciplinary agencies rarely initiate investigations sua sponte, preferring instead to rely on those personally affected by lawyer misconduct to bring claims to the agency’s attention. While these deficiencies in state disciplinary systems are not peculiar to matters involving prosecutorial misconduct, their significance is heightened in that context given the potential liberty interests involved.

Like state ethics rules, most state disciplinary systems follow a model code developed by the ABA. The current version of the code, the Model Rules for Lawyer Disciplinary Enforcement, was adopted in 1989 and last amended in 2002. Under this model, complaints are received by a central intake office, which determines whether the complaint states a colorable claim that merits further investigation. Statistics show that, in most jurisdictions, the majority of complaints are dismissed at this stage. Those that remain open are forwarded to an administrator for further review. The attorney named in the complaint is then afforded an opportunity to respond before the disciplinary agency decides whether to file a formal complaint. At this stage, many states offer attorneys accused of minor offenses the opportunity to participate in diversion programs or accept a private reprimand in lieu of further action. Should an attorney accept either option, the investigation will remain confidential. Once the administrator or hearing board files a formal complaint, however, the proceedings are made public. Because most disciplinary agencies do not publish statistics concerning the number of prosecutorial misconduct claims they receive, there is no method to determine how many claims of that nature result in private sanctions. If a formal complaint is filed, adversarial hearings are scheduled to review the allegations and solicit testimony from the parties involved. The hearing committee will subsequently issue findings of fact and recommend one of several possible dispositions: dismissal, reprimand, censure, probation, suspension, or disbarment. Each state’s court of last instance, under whose authority bar organizations operate, acts as an appellate body and retains final review over the imposition of any sanctions.

While the ABA aspires to offer a “simple and direct procedure for making a complaint,” even this modest aim has proven elusive. Only four states, for example, offer complainants the opportunity to submit their complaints online. Most other states offer a complaint form that can be downloaded and mailed, but twelve states do not. Complainants in the latter must either file their complaints over the telephone, request that a form be mailed to them, or enter into mandatory consumer assistance programs. Although no state charges a filing fee, both Kentucky and New Hampshire require complaints to be notarized.

Some states actively discourage complainants from filing allegations of misconduct. Mississippi’s bar association, for instance, goes to great lengths to warn complainants of the serious consequences that can result from filing a complaint. The bar association’s website begins its appeal by reminding potential filers that “lawyers are human.” The website continues, “The lawyer [complained against] inevitably suffers from the accusation, regardless of whether any misconduct is ultimately found. But, if you believe the complaint is well-founded, by all means make it! A complaint cannot be withdrawn once it has been received in this office.” Georgia discourages complaints in a different way by requiring prospective filers to go through a mediation program before deciding whether to pursue a formal complaint. The mediation program reflects a disciplinary system whose primary focus is private disputes between attorneys and their clients. In designing its disciplinary system, Georgia’s bar officials apparently did not envision complaints concerning prosecutorial misconduct, which ordinarily would not be amenable to mediation.

Filing a complaint is only a minor hurdle compared to the subsequent steps that must be taken before a complaint is finally resolved. Primarily, the problem is that complaints must work their way through a byzantine structure of state disciplinary systems. Compounding the problem is confusion over where the authority of the court ends and the disciplinary system begins. In the context of prosecutorial misconduct in particular, disciplinary system administrators may be wary of inserting themselves into ongoing court proceedings. The complexity of the procedure also results in substantial delay in resolving complaints. Data compiled by the ABA reveal that the amount of time between the filing of a complaint and the imposition of a public sanction in some states can take more than one thousand days. This lag time is likely to disincentivize those who might have a legitimate grievance from pursuing a disciplinary remedy.

Statutes of limitation pose a further barrier to potential claimants that can be especially problematic in the context of prosecutorial misconduct because such violations often come to light only years after their occurrence. At least twenty-one states impose some kind of statute of limitations on grievance filers. These range in length from as little as two years from the occurrence of the incident giving rise to the misconduct, to as many as ten years after its discovery. These statutes of limitations pose barriers to grievance filers and are fundamentally at odds with the ABA’s Model Rules, which caution that such statutes are “wholly inappropriate in lawyer disciplinary proceedings.”

Although states that have statutes of limitations in place will generally toll them if the misconduct was not discovered due to fraud or concealment, time limitations can be a major impediment to holding prosecutors responsible for misconduct, as a recent case in North Carolina demonstrates. In 2005, the State Bar of North Carolina brought a series of charges against two district attorneys, Scott Brewer and Kenneth Honeycutt. The complaint charged the attorneys with violating a host of ethics rules for failing to report an immunity deal given to a witness in exchange for his testimony in a capital murder trial. The North Carolina State Bar, however, requires that all grievances be filed within six years of the offense, exempting only actions involving felonious criminal conduct. Because they were concerned about the potential negative impact of a bar complaint on their client’s trial, the defendant’s attorneys chose to wait before filing. Consequently, the State Bar Disciplinary Hearing Commission held that the complaint was time-barred. The Commission also invalidated the felonious criminal conduct exception because the North Carolina Supreme Court had failed to publish the rule as required by statute, effectively precluding any possible ethical sanctions against the prosecutors. In upholding the Commission’s decision, a panel for the North Carolina Court of Appeals wrote that it was “cognizant” that its decision would “leave the State Bar unable to act if an aggrieved party learns of concealed misconduct by an attorney but does not report it to the State Bar.” Nonetheless, the court felt bound by traditional canons of statutory interpretation to affirm the Commission’s ruling.

State disciplinary authorities, which are comprised almost entirely of lawyers, also exercise nearly unbridled discretion in deciding whether to pursue individual complaints. While every state will dismiss a complaint for failing to state a colorable claim, it does not follow that every colorable claim is fully investigated. Instead, a disciplinary authority may decide not to pursue a complaint as a matter of resource allocation or because a reviewing attorney merely suspects that it lacks merit. In some states, like Florida, an investigation may be closed even where ethics violations are shown to have occurred, under the theory that “[t]he investigation of a complaint frequently has deterrent value in and of itself.” Furthermore, disciplinary authorities often conduct their proceedings in secret and require strict confidentiality from complainants. They may also decide to dispose of a case by issuing a private reprimand to the attorney involved. The lack of laypersons on hearing boards and review panels compounds the problem by creating the appearance of bias toward lawyers.

Measuring state disciplinary systems’ responsiveness to prosecutorial misconduct in particular is hampered by a paucity of available statistics. Only one state, Illinois, publishes data on the number of complaints of prosecutorial misconduct received and investigated on an annual basis. But if that data are indicative of the way most states handle such claims, they paint a bleak picture. The statistics show that, in 2010, charges against 4016 attorneys were docketed by the Illinois Attorney Registration and Disciplinary Commission, of which ninety-nine involved charges of prosecutorial misconduct. Only one of these ninety-nine cases, however, actually reached a formal hearing. In other words, the Illinois disciplinary commission held as many formal hearings involving charges of prosecutorial misconduct as it did charges of “bad faith avoidance of a student loan.”

To make matters worse, the grievance process in many states does not provide complainants with the opportunity to appeal the dismissal of their complaint unless it has reached the hearing stage. As Florida explains to prospective filers in its bar consumer pamphlet: “Your role in a disciplinary complaint is that of the complaining witness, similar to the role of a victim in a criminal proceeding. As such, you are not a party to the adjustment proceeding in that the Bar counsel does not represent you as your lawyer.” Twenty-three states provide no recourse for complainants wishing to appeal a disciplinary staff attorney’s decision to dismiss their complaint, while several other states provide for redress in only limited circumstances.

In light of the foregoing shortcomings in state disciplinary procedures, the Supreme Court’s faith in the ability of those procedures to adequately check prosecutorial misconduct seems misplaced. Yet, rather than simply lamenting Connick as another barrier to holding prosecutors accountable for their misdeeds, scholars and advocates alike should take seriously the Court’s insistence that bar disciplinary procedures are the appropriate mechanism for policing misconduct. Accordingly, in the next Part, we offer suggestions for strengthening ethics rules and disciplinary procedures to achieve a regime of greater accountability for prosecutors.

IV. Recommendations

There are many important steps that prosecutors’ offices, state judiciaries, and bar associations should take to ensure an environment in which proper incentives and adequate training enable prosecutors to seek justice. Our recommendations here reflect our findings and focus on the responsibilities of the state courts of last instance, as well as on the role of state attorney grievance procedures, in building mechanisms that inform the responsibilities of prosecutors and that live up to the U.S. Supreme Court’s expectations regarding the efficacy of lawyer discipline.

Prosecutors have different professional and ethical obligations than private attorneys. Grievance mechanisms should be strong enough to hold prosecutors to these heightened obligations. To better protect the rights of the accused and to ensure a just system that adequately checks prosecutorial misconduct, state supreme courts and state bar associations should take the following actions to ensure that the rules governing prosecutorial conduct are adequate and that lawyer discipline procedures ensure the efficacy of the rules.

A. Rules of Professional Conduct

The starting point for improving state attorney grievance mechanisms is the promulgation of an effective rule defining the ethical obligations of prosecutors. The ABA should begin a dialogue with states and the Department of Justice about expanding Rule 3.8 to more completely address the unique ethical challenges that face prosecutors. Important areas of prosecutorial function include investigating crimes, negotiating pleas, and exercising discretion in charging crimes. These responsibilities are not adequately addressed in the Model Rules, an oversight that leaves much of the prosecutorial function outside the scope of ethical regulation or guidance.

First, states should expedite the review and adoption of sections (g) and (h), which create new ethical obligations for a prosecutor who becomes aware of evidence suggesting or establishing the non-guilt of a convicted defendant. Since sections (g) and (h) were added to Model Rule 3.8 in February 2008, only one state—Idaho—has adopted the modified rule in its entirety. Three more—Colorado, Tennessee, and Wisconsin—have adopted section (h) and modified versions of section (g), and one further state—Delaware—has adopted a hybrid version of (g) and (h). The Criminal Justice Section of the ABA emphasized the importance of these additions, noting that “[t]he obligation to avoid and rectify convictions of innocent people, to which the proposed provisions give expression, is the most fundamental professional obligation of criminal prosecutors.”

Second, states can influence the ABA and the Department of Justice in the design of Model Rule 3.8. By promulgating tougher rules in their state codes, states can pressure the ABA to address the deficiencies in the Model Rule. Proactively defining the scope of the ethical obligations that should govern prosecutorial conduct can inform the ABA’s own deliberation and amendment of its rules.

B. State Grievance Procedures

Attorney grievance procedures must inspire confidence in the regulatory system governing attorney behavior. As such, bar associations and state supreme courts should take pains to avoid the pitfalls of a system that relies on self-regulation. State supreme courts should assume full and independent control over disciplinary processes. Elected bar officials governing the disciplinary process create the impression of self-regulation that can lead to suspicion of bias in the proceedings. Laypersons should have an active and substantial role in the grievance process. Non-lawyers comprise a third of the grievance boards (i.e., appellate review) of nine states (Arizona, Connecticut, Delaware, Idaho, Louisiana, Massachusetts, Oregon, New Jersey, and New Mexico). Two states (Kansas and California) do not have any non-lawyers participating in grievance process, at the committee (i.e., trial) or board level. In other states, disciplinary organizations are overwhelmingly controlled by the bar; South Dakota’s disciplinary board, for instance, consists of six bar members appointed by the President of the State Bar Association and only one layperson appointed by the Chief Justice. A more balanced distribution of influence between the judiciary and the bar would signal that disciplinary bodies take seriously the dangers of self-policing. Legitimacy and the perception of fairness decrease when self-regulation is the chosen method for governing attorney conduct.

Grievance procedures should be simple and accessible so that potential claimants are incentivized to file colorable claims of misconduct. Structural disincentives may dissuade potential claimants from using bar grievance mechanisms. As our findings show, procedures for bringing a grievance complaint vary greatly in their accessibility and form from state to state. Any interested party—including third parties, such as advocacy organizations, law school clinics, and the general public—should be able to bring a grievance alleging prosecutorial misconduct. States should also lengthen or abolish statutes of limitations and explicitly provide for tolling where misconduct has been concealed or where equitable factors, such as the pursuit of a criminal appeal, have impeded parties from pursuing an ethics complaint. Furthermore, providing access to easy-to-use complaint forms in courthouses and online would facilitate filings. Many state supreme court and bar association websites are discouragingly difficult to navigate. State grievance procedures and infrastructure, as administered by state judiciaries and bar associations, should invite claims of prosecutorial misconduct by ensuring that both the public and interested parties can easily file a claim.

Procedures to investigate and sanction prosecutorial misconduct should also encourage adjudication of colorable claims. Because of infrequent adjudication and the opacity of most bar systems, the standards Rule 3.8 imposes on prosecutorial conduct are neither clearly defined nor given substance by precedential case law. State grievance agencies need to be properly resourced, both financially and with experienced investigators knowledgeable in the intricacies of criminal justice and the role of prosecutors. Further, disciplinary committees should institute automatic filing of ethics complaints, triggered whenever a court finds (whether on direct appeal, collateral review, or otherwise) that a prosecutor has behaved unethically. Judges in particular should be compelled to flag an instance of misconduct for review by the grievance committee. Automatic filing will trigger investigation and take the process of submitting a formal complaint out of the hands of busy or disincentivized attorneys and court officials. All states should enforce rules requiring attorneys who are aware of prosecutorial misconduct to report it promptly; the entire profession should be held responsible for the administration of justice. Increased adjudication of ethics complaints would better inform both bar investigators and prosecutors of the obligations and standards of prosecutorial behavior.

State grievance committees should undertake regular and randomized auditing of cases in their jurisdictions to increase the likelihood that prosecutorial misconduct will be discovered and remedied. A grievance investigator could be tasked with reviewing a randomly selected sample of cases and undertaking an investigation to ascertain whether professional and ethical rules are being followed. If the investigation uncovers errors, a state bar committee could—in addition to the regular grievance process in place—work directly with the prosecutor’s office to explain the errors, thus demonstrating the professional rules specifically applicable to prosecutors. Such a system of audits would contribute to prosecutors’ incentives to understand and comply with their legal and ethical obligations and serve a pedagogical role in educating prosecutors about the scope of those duties.

Finally, transparency of process is important both to legitimize the grievance process and to inform laypersons and prosecutors alike of the appropriate standards of conduct. To varying degrees in the states surveyed, bar disciplinary investigations are largely confidential. When a grievance committee dismisses charges before a public hearing, there is no record or published opinion. An increased practice of issuing written and public findings in disciplinary cases that do not result in sanctions by the state supreme court would increase transparency and legitimacy. Written opinions are, of course, routine in the context of dismissed civil lawsuits and criminal cases. Making all grievance decisions available to the public and easily searchable on an online database would serve to inform interested parties of the grievance’s disposition and would educate prosecutors about their ethical and professional responsibilities.

Further, complainants and interested parties should be able to discern the path their complaint will take once filed. The grievance procedures of many states make adjudicating a claim unnecessarily complex. A quick and efficient system with as few steps as possible between complaint and investigation is important to prevent colorable claims from falling through the cracks.


The Connick decision reflects the Supreme Court’s historical reliance on ethics rules and state disciplinary procedures to regulate prosecutorial behavior. Irrespective of the wisdom of the Court’s reasoning, the ethics rules governing prosecutorial behavior need to be expanded and strengthened, and the disciplinary procedures tasked with enforcing them reformed, if our legal system is to justifiably rely on professional sanctions to deter prosecutorial misconduct. The job of a prosecutor is to do justice; the structure in which the prosecutor works should, at a minimum, enable and encourage ethical behavior in this pursuit.

The authors are current students at Yale Law School. The research underlying this Essay began as part of the Prosecutorial Ethics and Accountability Project of the Arthur Liman Public Interest Program at Yale Law School. The authors would like to extend their gratitude to the project’s other participants, Isabel Bussarakum, Ester Murdukhayeva, and Emily Washington; the project’s supervisor, Fiona Doherty; the director of the Liman Program, Hope Metcalf; and Professors Judith Resnik and Dennis Curtis. The authors would also like to thank David Menschel and Jeff Meyer for their comments on earlier drafts of the piece. Finally, the authors would like to express their gratitude to Nick Hoy and his colleagues at The Yale Law Journal Online for their thoughtful editorial suggestions.

Preferred citation: David Keenan, Deborah Jane Cooper, David Lebowitz & Tamar Lerer, The Myth of Prosecutorial Accountability After Connick v. Thompson: Why Existing Professional Responsibility Measures Cannot Protect Against Prosecutorial Misconduct, 121 Yale L.J. Online 203 (2011),

[Editor's note - the Appendix to this Essay can be found in the PDF version.]