Skip navigation

Prosecutorial Oversight, Innocence Project, 2016

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Prosecutorial Oversight:
A National Dialogue in the Wake of
Connick v. Thompson

march 2016

The Innocence Project would like to acknowledge and thank the following individuals who contributed to this report and its
success: Lawanna Kimbro, Liz Jansky, Emily West, Stephen Saloom, Cookie Ridolfi and Ellen Yaroshefsky.
Resurrection After Exoneration
Resurrection After Exoneration (RAE) was founded in 2007 by exonerees to promote and sustain a network of support among
formerly wrongfully incarcerated individuals in the South. RAE works to reconnect exonerees to their communities and provide
access to those opportunities of which they were robbed.
Innocence Project New Orleans
Innocence Project New Orleans (IPNO) is a nonprofit law office that represents innocent prisoners serving life sentences in
Louisiana and Mississippi at no cost to them or their loved ones, and assists them with their transition into the free world upon
their release. IPNO uses its cases to explain how wrongful convictions happen and what we can all do to prevent them. Since its
inception in 2001, IPNO has freed or exonerated 27 innocent men. We devote the majority of our time and resources to freeing
poor people who will otherwise die in prison for crimes they did not commit.
Veritas Initiative at Santa Clara University School of Law
The Veritas Initiative is dedicated to advancing the integrity of our justice system by researching and providing critical data that
shines a light on such crucial issues as the misconduct of public prosecutors.
The report Preventable Error: A Report on Prosecutorial Misconduct in California 1997-2009 marks the launch of the Veritas
Initiative. By shining a light on issues like prosecutorial misconduct, the Veritas Initiative and the studies it publishes will serve
as a catalyst for reform.
Innocence Project
The Innocence Project was founded in 1992 by Barry C. Scheck and Peter J. Neufeld at the Benjamin N. Cardozo School of Law
at Yeshiva University to assist prisoners who could be proven innocent through DNA testing. To date, more than 300 people in
the United States have been exonerated by DNA testing, including 20 who served time on death row. The Innocence Project was
involved in 177 of the DNA exonerations. Others were helped by Innocence Network organizations, private attorneys and by pro
se defendants in a few instances.
The Innocence Project’s full-time staff attorneys and Cardozo clinic students provide direct representation or critical assistance
in most of these cases. The Innocence Project’s groundbreaking use of DNA technology to free innocent people has provided
irrefutable proof that wrongful convictions are not isolated or rare events but instead arise from systemic defects. Now an
independent nonprofit organization closely affiliated with Cardozo School of Law at Yeshiva University, the Innocence Project’s
mission is nothing less than to free the staggering numbers of innocent people who remain incarcerated and to bring substantive
reform to the system responsible for their unjust imprisonment.

EMBARGOED Until March 29

table of contents
A Call for Action................................................................................................................................................... 4
Section I: Introduction........................................................................................................................................ 6
The Case of John Thompson.............................................................................................................................................................................. 6
Formation of Prosecutorial Oversight Coalition............................................................................................................................................ 7
The Need for Greater Prosecutorial Accountability....................................................................................................................................... 8
What is Prosecutorial Misconduct?.................................................................................................................................................................. 9
Difficulties Identifying the Scope of the Problem........................................................................................................................................... 9
Section Ii: Scope of This Report....................................................................................................................... 10
Coalition Partners.............................................................................................................................................................................................10
Questions Explored...........................................................................................................................................................................................10
Methods Used....................................................................................................................................................................................................10
Section III: Review of existing research on Prosecutorial Misconduct................................................ 11
SECTION IV: JUDICIAL CASE LAW REVIEW OF PROSECUTORIAL MISCONDUCT....................................................... 12
Absence of Disciplinary Action in the Wake of Prosecutorial Misconduct..............................................................................................12
SECTION V: CONCLUSIONS DRAWN FROM FINDINGS AND FORUMS........................................................................ 14
Culture of Underreporting...............................................................................................................................................................................14
Impractical Reliance upon Bar Disciplinary Committees...........................................................................................................................15
Appellate Review: Harmless Error Doctrine Greatly Minimizes Problem of Misconduct.....................................................................15
Civil Remedies Impossible to Obtain.............................................................................................................................................................16
Dearth of Criminal Sanctions..........................................................................................................................................................................17
Section Vi: Recommendations to improve systems of prosecutorial oversight.................................. 18
Bar Oversight Entities.......................................................................................................................................................................................19
Legislative Action..............................................................................................................................................................................................19
Law Schools........................................................................................................................................................................................................19
Section VIi: ConCLUSION...................................................................................................................................... 20
APPENDIXES............................................................................................................................................................ 21
Endnotes............................................................................................................................................................... 25

EMBARGOED Until March 29

March 29, 2011
The Honorable Eric H. Holder, Jr.
Attorney General of the United States
United States Department of Justice
950 Pennsylvania Avenue, NW
Washington, D.C.  20530-0001
James M. Reams, Esq.
President of the National District Attorneys Association
Rockingham County Attorney’s Office
Rockingham County Superior Courthouse,
10 Rte. 125
Brentwood, N.H. 03833
Roy Cooper
President of the National Association of Attorneys General
2030 M Street NW, 8th Floor
Washington, D.C. 20036
Dear Sirs,
Today the U.S. Supreme Court in Connick v. Thompson took away most of the only remaining means those of us who
have been wrongfully convicted of a crime had for holding prosecutors liable for their misconduct. Although all other
professionals, from doctors to airline pilots to clergy, can be held liable for their misconduct, the Supreme Court has effectively
given prosecutors complete immunity for their actions. We, the undersigned and our families, have suffered profound harm at
the hands of careless, overzealous and unethical prosecutors.  Unfortunately, today’s ruling only threatens to further embolden
those prosecutors who are willing to abandon their responsibility to seek justice in their zeal to win convictions. 
Now that the wrongfully convicted have virtually no meaningful access to the courts to hold prosecutors liable for their
misdeeds, we demand to know what you intend to do to put a check on the otherwise unchecked and enormous power that
prosecutors wield over the justice system.
Former United States Attorney General and Supreme Court Justice Robert H. Jackson once said, “The prosecutor has
more power over life, liberty, and reputation than any other person in America.” Unfortunately recent reports have shown
that prosecutors are abusing this power at alarming rates and are facing no consequences for their actions. According to
Preventable Error: A Report on Prosecutorial Misconduct in California 1997-2009, prosecutors were guilty of misconduct 707
times from 1997 to 2009, yet were disciplined only seven times. A USA Today investigation by Brad Heath and Kevin McCoy
that was published on Sept. 23, 2010, documented 201 instances where federal prosecutors violated laws or ethics rules since
1997 and noted that only one of those prosecutors was suspended from practicing law—and that was only for one year.   
In many of our wrongful conviction cases prosecutorial misconduct was found but later declared “harmless” by the
courts. Nothing could be further from the truth. In our cases, each act had profoundly harmful effects on our lives. 
Together we represent hundreds of years in prison, separated from our wives, husbands, children, parents, brothers,
sisters, grandparents and other loved ones, who suffered their own shame and wasted hundreds of thousands of dollars

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


on lawyers and spent countless sleepless nights worrying about our well being. The misconduct contributed to nearly unbearable
depression and unhappiness, loss of jobs and career opportunities, the derailing of educations and forever destroyed hopes and
dreams. Each of us has worked long and hard to repair what has happened to us, but we will never regain the lives we had before
we were wrongfully convicted at the hands of careless or deceitful prosecutors. 
According to the friend-of-the-court briefs submitted in recent Supreme Court cases dealing with prosecutorial misconduct the
National District Attorneys Association, the National Association of Assistant United States Attorney’s Attorney Generals and the
Solicitor General claim that there are already plenty of systems in place to cure the problems of misconduct, including: internal
disciplinary systems, state bar disciplinary systems, monitoring by the courts and, in extreme cases, criminal prosecution. These
systems didn’t do a thing to prevent prosecutorial misconduct in our cases. As far as we can tell, none of these systems were
brought to bear on the prosecutors in our cases. Our sense is that they do nothing at all. 
We demand to know what you are doing to stop these abuses. What has happened to the prosecutors whose knowing acts
contributed to our suffering? What have you done to make them understand that they cannot do it again? How have the systems
been fixed to prevent future misconduct and errors? What makes you think these solutions have worked when so many people
continue to be wrongfully convicted?   
The power to charge and prosecute someone with a crime comes with grave responsibility. Now that the Supreme Court has
said that prosecutors cannot be held civilly liable for their actions, it’s up to you to make sure that prosecutors take that
responsibility as seriously as the job demands.  As you consider the significance of today’s decision, please know that those
of us who have been the victims of prosecutorial misconduct are eager to hear what you intend to do to ensure that others
don’t suffer injustice as we have.
Kennedy Brewer
Noxubee County, MS

Dennis Fritz
Pontotoc County, OK

Ray Krone
Maricopa County, AZ

Roy Brown
Cayuga County, NY

James Giles
Dallas County, TX

Curtis McCarty
Oklahoma County, OK

Darryl Burton
Saint Louis, MO

Bruce Godschalk
Montgomery County, PA

Kirk Bloodsworth
Baltimore County, MD

Thomas Goldstein
Los Angeles County, CA

Gene and Elizabeth Sodersten,
parents on behalf of Mark Sodersten,
who died in prison
Tulare County, CA

Algie Crivens
Cook County, IL

Michael Green
Cuyahoga County, OH

Clarence Elkins
Summit County, OH

Lesly Jean
Onslow County, NC

Michael Evans
Cook County, IL

Joshua Kezer
Cole County, MO

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson

John Thompson
Orleans Parrish, LA
Keith Turner
Dallas County, TX


Section I: Introduction
“The prosecutor has more control over life, liberty and reputation than any other person in America.”1
- Former United States Attorney General and Supreme Court Justice Robert H. Jackson
The 19 people who signed the letter beginning this report
were each imprisoned for a crime they did not commit
that was secured in whole or in part by prosecutorial
misconduct. Notably, not one of the prosecutors involved in
these cases was sanctioned for their role in contributing to
the wrongful conviction. The United States Supreme Court
in Connick v. Thompson2 faced with a particularly egregious
case of deliberate misconduct by a prosecutor found that
existing oversight systems are adequate to respond to and
prevent future prosecutorial error and misconduct in
overturning an award of monetary damages that the wrongly
convicted man received for the misconduct that nearly led
to his execution. What explains this contradiction? Did the
court get it wrong? Why did these systems fail to hold any
of these prosecutors accountable despite clear evidence of
misconduct and such significant harm?
This report is the culmination of an inquiry into these
questions—whether existing oversight systems are
adequate to respond to and prevent prosecutorial error
and misconduct—and concludes that the court was wrong.
There are almost no adequate systems in place to keep
prosecutorial error and misconduct in check and, in fact,
prosecutors are rarely held accountable even for intentional
misconduct. There is a critical need for a national dialogue
on the importance of prosecutorial accountability and the
implementation of functioning systems that will prevent
and address prosecutorial error and misconduct when
it occurs. This report is intended to spark meaningful
conversations and action among key criminal justice leaders
and policymakers about the need for greater oversight and
accountability for prosecutors. To help those audiences
develop more efficient systems, the report includes a list of
recommendations that, if implemented, could help achieve
those goals.

The Case of John Thompson
In 1984, John Thompson, a 22-year-old father of two, was
wrongfully convicted of two separate crimes—an armed
robbery and a murder—at two separate trials. He was
prosecuted and convicted first for the robbery. By obtaining
a felony conviction for the armed robbery, prosecutors were
then able to charge Thompson with capital murder and seek
the death penalty. Thompson was found guilty and sent to
death row to await his execution.

With less than a month before Thompson’s seventh
scheduled execution date, a private investigator hired
by his appellate attorneys discovered scientific evidence
of Thompson’s innocence from the robbery case that
had been concealed for 15 years by the Orleans Parish
District Attorney’s Office. The new evidence showed that
prior to Thompson’s armed robbery trial, the prosecution
ordered blood type testing of bloodstains on the victim’s
pant leg and shoe. The blood test results, which excluded
Thompson, were never disclosed to the defense and the
samples themselves were withdrawn from the evidence
locker and destroyed by one of the assistant district
attorneys prosecuting the armed robbery case. Further
investigation revealed additional undisclosed Brady
information, including a payment to the primary informant
in Thompson’s murder case that had not been disclosed to
Thompson or his attorneys before the murder trial.
Thompson was eventually exonerated of both crimes and
filed a civil law suit against the district attorney’s office for
the violations of his civil rights that resulted in his wrongful
conviction and near-execution. In order to hold the Orleans
Parish District Attorney’s Office—a municipal entity—liable
under 42 U.S.C. § 1983, Thompson was required to show
that the district attorney’s failure to train employees on
their legal obligations to avoid constitutional violations of
defendants’ rights was so persistent and widespread in the
office as to amount to a “deliberate indifference to the rights
of persons with whom the [untrained employees] come
into contact.”3 “[D]eliberate indifference” must amount to
a disregard of the “known or obvious consequence[s]” that
such a failure to institute a training program on prosecutors’
legal duties to disclose exculpatory evidence would result in
district attorneys’ violation of defendants’ due process rights.
A pattern of similar constitutional violations committed
by the untrained employees is sufficient evidence of such
deliberate indifference.4
In a jury trial, Thompson prevailed and was awarded $14
million—$1 million for every year he spent on death row.
The award was affirmed by the U.S. Court of Appeals for
the Fifth Circuit and reaffirmed in an en banc rehearing of
the case. However, in a controversial 5-4 decision, the U.S.
Supreme Court overturned the jury’s award,5 finding that the
Orleans Parish District Attorney’s Office’s could not be held
civilly liable based on a “failure to train” theory of liability
for what the court concluded was a single Brady violation
perpetuated by one prosecutor’s actions.6 In the majority

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


opinion that reversed the jury verdict, the Supreme Court
appeared to find persuasive an argument outlined in an
amicus brief by the National District Attorneys Association
The brief argued that the combination of legal education,
professional ethics, on-the-job supervision and training and
existing lawyer oversight systems are sufficient to ensure that
prosecutors will act within the bounds of the law, and that
the courts should not “punish” the district attorney’s office
“for not having a formal Brady training program” in light of
the “very scarce resources” available to prosecutors.7
The court’s majority agreed, ruling that because of what
it described as well-established systems and training
procedures in the legal profession as a whole, prosecutors’
offices could not be held constitutionally liable for a single
act of misconduct by an employee, no matter how egregious
the misconduct in question.8 The court went on to find that
the record in Thompson’s case was insufficient to show a
pattern and practice of Brady violations sufficient to hold the
prosecutor’s office liable under Section 1983.
By contrast, Justice Ginsburg and her fellow dissenters found
compelling evidence in the record developed at Thompson’s
civil trial that “inattention to Brady was standard operating
procedure at the district attorney’s office” over many years.9
Because even the egregious violations in Thompson were
held insufficient to support a verdict against the district
attorney’s office, the decision effectively eliminated virtually
any avenue for holding prosecutors civilly liable, even for
intentional misconduct. 10

Formation of Prosecutorial
Oversight Coalition
After the controversial decision in Thompson, several
members of the Innocence Network formed the
Prosecutorial Oversight Coalition. The coalition was
formed to investigate the absence of accountability for
prosecutorial misconduct and error, and to propose policy
reforms dedicated to prosecutorial accountability, with an
aim of reducing both the deliberate misconduct and the
preventable errors that have contributed to innocent people
spending decades in prison for crimes they did not commit.
Given the number of wrongful convictions caused, at
least in part, by prosecutorial error and misconduct, the
Prosecutorial Oversight Coalition felt that it was important
to investigate the Supreme Court’s conclusion that existing
oversight systems are adequate to respond to and prevent
prosecutorial error and misconduct. The coalition reviewed
the existing literature and research on prosecutorial
misconduct. It conducted independent research to try
to quantify the prevalence of prosecutorial error and
misconduct in the United States and held forums in six
states across the country to determine what systems are
currently in place to prevent prosecutorial misconduct and
error, what we know about the effectiveness of these systems
and what can be done to improve them.
Contrary to the Supreme Court’s assertions, the coalition’s
investigation has found that there are almost no adequate
systems in place to keep prosecutorial error and
misconduct in check and, in fact, prosecutors are rarely

Lesly Jean
Lesly Jean was wrongfully convicted of rape in Jacksonville, North Carolina, in 1982.
Shortly after the victim was sexually assaulted, a Jacksonville police officer stopped a
man who seemed to fit the victim’s description of her assailant. The man fled on foot. The
chief detective of the Jacksonville Police Department decided that the officer should be
hypnotized in order to more accurately recall the description of the man he stopped; after the
hypnotization, the officer’s description of the man changed dramatically.
Soon after, Jean was arrested when an officer saw him in a local Dunkin’ Donuts and thought
that he matched the suspect’s description. The officer radioed the original officer, who arrived
on the scene and confirmed that Jean was the same man he approached on the street. Jean
was placed under arrest. After the victim twice failed to identify Jean as the perpetrator during
a photo lineup, she was also hypnotized to “improve her memory,” which ultimately resulted
in her positive identification of Jean’s photograph. The critical fact that hypnosis was an integral part of the identification
process for both witnesses was never disclosed to the defense. This Brady violation led to the reversal of Jean’s conviction
in 1991 and his release from prison. Jean v. Rice, 945 F.2d 82, 87 (4th Cir. 1991). He was officially exonerated in 2001 after
DNA testing proved his innocence and he was pardoned. No action was taken to examine the prosecutor’s conduct.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


held accountable even for intentional misconduct. This
mirrors the experiences and opinions of many legal and
advocacy organizations, scholars and researchers who
have previously explored the efficacy of prosecutorial
accountability systems.11 It also conforms with the
experiences of the 19 signatories to the opening letter to this
report, all former prisoners whose wrongful convictions
were secured in whole or in part through prosecutorial
misconduct. Significantly, based on public records, none of
the prosecutors involved in these cases—each of which later
resulted in an exoneration—had their conduct subject to any
kind of meaningful review, much less any type of sanction.
The Supreme Court decision in Thompson and the findings
of this coalition, make clear that there is a critical need for
a national dialogue on the importance of prosecutorial
accountability and the implementation of functioning
systems that will prevent and address prosecutorial error
and misconduct when it occurs. This report hopes to spark
meaningful conversations and action among key criminal
justice leaders and policymakers about the need for greater
oversight and accountability for prosecutors. In an effort
to help those audiences develop more efficient systems for
stemming prosecutorial error and misconduct, this report
provides a list of recommendations for how that may be

The Need for Greater
Prosecutorial Accountability
Most prosecutors do not act with intent to conceal
exculpatory evidence. Indeed, the vast majority of
prosecutors perform their duties in good faith with the aim
to fulfill their constitutional and legal obligations. This good
faith intention does not, however, eliminate the need for
systems that address both intentional misconduct and error
when it occurs.
Like the rest of us, prosecutors are susceptible to the stress
of their very demanding jobs, cognitive biases and a host
of other human realities.12 Mistakes are bound to occur, no
matter how experienced or thorough a prosecutor may be.
And in some rare cases, prosecutors’ eagerness to secure
convictions has led them to commit deliberate violations
of the law, leading to wrongful convictions and even death
sentences of persons we now know were factually innocent.
Moreover, in a substantial number of wrongful conviction
cases, the exculpatory material that prosecutors did not
disclose at trial would have provided important leads to
the true perpetrator of the crime.13 As a result, justice was
denied to the crime victims in these cases, not just the
wrongfully convicted defendants, and public safety was
The need for a vigorous accountability system is, of course,
not specific to prosecutors, and this report is offered in the

context of a larger, national conversation about heightened
accountability in all areas of the criminal justice system.
We have previously reported on the problem of ineffective
assistance of the defense as a contributing factor in wrongful
convictions and believe these lawyers should also be held
accountable and subject to appropriate discipline.14 But
as we discovered through the research for this report,
inadequate accountability for prosecutors is unique among
the other actors in the criminal justice system.
Legal ethics scholars have been calling for increased
accountability for prosecutors for some time.15 Prosecutors
are the most powerful figures in the American criminal
justice system.16 They exercise significant discretion,
and their decisions are not subject to external review.17
Prosecutors decide how to investigate a case, what charges to
bring, what plea bargains to offer, what penalties to seek and
what evidence to turn over to the defense. These decisions
have an enormous impact on defendants, victims, their
families and the public at large.18
Yet, very few prosecutor offices have any internal review
policies, and following the Thompson decision, prosecutors
enjoy almost complete immunity from civil liability.
Given their broad powers, it is critical that effective systems
of accountability are implemented to incentivize prosecutors
to act within their ethical and legal bounds. Efforts to bring
greater accountability to government actors as well as a host
of other industries—from healthcare, aviation, construction,
to the food industry—offer effective approaches for
prosecutorial accountability.
Instead of looking at errors through a single-cause lens,
where the focus of the search is to identify the one mistake
or “bad apple” that caused the problem, experts advocate for
a more systemic approach for reviewing errors. This process
avoids simply blaming individuals and instead concentrates
on understanding the organizational factors that contribute
to errors.19
Using this systems-based approach, every mistake
is evaluated from the premise that examining and
understanding errors can lead to improvements in practice.
In fact, this approach is being advanced for improving
policing,20 the indigent defense system and the judicial
system and it is in that spirit that we offer this report.21
Given the number of innocent people who have suffered
injustice as a result of prosecutorial error and misconduct,
the members of the coalition have a unique perspective
to offer on this issue. We offer this perspective and
suggestions for reform without pointing blame, but with
the hope of developing greater systems of accountability
for prosecutors. The coalition’s national tour, multi-state
surveys of court decisions and this report are part of an
effort to generate and contribute to a growing conversation
that is focused on improvement, looking specifically at the

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


current prosecutorial oversight systems and assessing how
we can make them better. We also view this report as a call
to policymakers to more closely review their systems of
oversight and enact the changes necessary to ensure that
these systems maintain the highest standards of professional
integrity and promote the interest of justice and public

What is Prosecutorial Misconduct?
As defined by courts, prosecutorial misconduct includes
any conduct by a prosecutor that violates a defendant’s
rights, regardless of whether that conduct was known or
should have been known to be improper by the prosecutor,
or whether the prosecutor intended to violate legal
While intentional misconduct and unintentional error
are different, and suggest the need for different kinds of
oversight mechanisms, both can result in injustice and must
be addressed. This report examines systems of oversight for
both extreme acts of intentional behavior and less extreme,
but also important, instances where prosecutors commit
errors or mistakes. A prosecutor’s failure to turn over a
piece of exculpatory evidence can have a devastating effect
regardless of whether the nondisclosure was intentional or a
genuine oversight by the prosecutor.
Further, as we heard from many of the panelists at the
forums, solely focusing on the minority of cases that contain
outrageous, unethical, and illegal acts by a prosecutor will
not address the full scope of the problem and is unlikely to
lead to any productive, systemic reform. It is precisely the
opposite of the “learning from error” approach that experts
recommend and that has become the standard that we
demand of other professions.23 By looking at prosecutorial
error and misconduct in its entirety, we will be able to create
a range of mechanisms sensitive enough to recognize the
difference between this wide range of actions and flexible
enough to move towards more significant responses when
Since the legal definition could lead policymakers into
focusing solely on instances of intentional misconduct, we
will use the term “prosecutorial error and misconduct” to
include any type of prosecutorial action that falls outside
of the profession’s legal and ethical guidelines, regardless of
intent or knowledge of wrongdoing. We view each of these
actions as opportunities to learn where there is a need for
more training, supervision, and in certain cases, sanction.
Prosecutorial error and misconduct can occur at any
stage of a criminal proceeding, although the behaviors
generally recognized are those that occur during trial. This
encompasses a wide range of behaviors but the following are
the most common: 24

•	 improper arguments or examination at trial
•	 inflammatory comments in the presence of the jury
•	 mischaracterizing evidence
•	 allowing false testimony to stand uncorrected
•	 failure to disclose evidence favorable to the accused,
which includes evidence that tends to negate a defendant’s
guilt, that would provide grounds to mitigate or reduce
a defendant’s potential sentence, or evidence going
to the credibility of a witness (known, respectively, as
“exculpatory evidence” or “impeachment evidence”),
which generally occurs prior to or during trial.25

Difficulties Identifying the
Scope of the Problem
Defining the universe of prosecutorial error and misconduct
is difficult, if not impossible. Because of the challenges in
discovering and recording prosecutorial actions, many
instances of prosecutorial misconduct and error never reach
public view, suggesting that the problem is much more
widespread than the number of reported judicial findings
of prosecutorial misconduct would indicate. Although
the occurrence of prosecutorial error and misconduct
is universally acknowledged, there is considerable
disagreement about how widespread a problem it is. One
federal chief judge called it “epidemic” in a blistering and
often-cited dissent.26 But whether epidemic or episodic, a
problem that results in the conviction of innocent people
must be addressed.
Being unable to document the extent of the problem
illustrates one of the biggest hurdles to greater accountability
for prosecutors. In order to prevent error and misconduct
from happening, we need to know when and how often it
There are currently only three data sources for investigating
the scope of prosecutorial error and misconduct: the media,
written public court decisions and disciplinary decisions
of the individual state bars. As discussed in greater detail
in the sections below, these are inadequate mechanisms for
identifying the prevalence of error and misconduct.
While attempts to document the extent of the problem have
proven difficult, one common theme remains: even when
prosecutorial error and misconduct are discovered, the
actions are rarely addressed in a meaningful way, regardless
of whether the conduct was intentional or merely error. This
tells us that generally the prosecutorial oversight systems
that currently exist are not adequate and that both the
federal government and individual states must explore other
steps to ensure meaningful and constructive oversight.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Section Ii: Scope of This Report
Coalition Partners
This report was a collaborative effort by several
organizations that oversaw the design and implementations
of the research process. The organizations primarily
responsible were the Innocence Project, the Veritas Initiative
at Santa Clara University School of Law, Innocence Project
New Orleans and Resurrection After Exoneration.
A number of additional organizations also provided support
in hosting and presenting the prosecutorial oversight
forums. These included the Arizona Justice Project, the
Pennsylvania Innocence Project and the Actual Innocence
Clinic at the University of Texas School of Law.

Questions Explored
The three following fundamental questions informed and
guided the tour and this report:
1.	What systems are in place, inside and outside of
prosecutors’ offices, to identify and address prosecutorial
error and misconduct and ensure accountability?
2.	What do we know from research and experience about
how well these internal and external systems are working?
3.	What improvements must be made to these systems,
and what are the associated hurdles to overcome, to
ensure that prosecutors act within their legal and ethical
guidelines, deter error and misconduct and maintain
public confidence in the integrity of the criminal justice

Methods Used
In exploring these questions, coalition investigators drew
from both previously conducted research on prosecutorial
misconduct from other sources and new research conducted
by the coalition and expert opinions generated through
panel discussions from the prosecutorial oversight forums.
1. Literature Review: The coalition examined pre-existing
studies and investigative journalism pieces on prosecutorial
misconduct and error, and the oversight systems currently
in place. These studies were conducted by a variety of news
organizations and research centers, including the Center for
Public Integrity, ProPublica, Chicago Tribune, USA Today,
the Veritas Initiative and Yale University. The findings from
the literature review are presented in Section III of this

2. Judicial Case Review: In preparation for the prosecutorial
oversight forums, the Veritas Initiative, with support from
the Innocence Project, conducted independent research
into judicial findings of prosecutorial misconduct and
error. Cases from five states of the tour were reviewed.27
Researchers examined public, federal and state judicial
rulings that address prosecutorial error and misconduct
from 2004 to 2008. Public disciplinary decisions associated
with the reviewed cases were also examined.28 Westlaw
was used to search for public judicial rulings.29 Public
disciplinary decisions were obtained through state grievance
Researchers looked at all cases that captured any
acknowledgment by the courts of prosecutorial error or
misconduct, not only cases where the most egregious
misconduct occurred.31 As a result, the range of error and
misconduct represents a full spectrum—from simple error,
such as a prosecutor who made an isolated inappropriate
comment, to serious misconduct, such as a prosecutor who
knowingly allowed a witness to lie on the stand. Results
from these reviews were analyzed and presented at each
forum during the prosecutorial oversight tour.32 The results
of this analysis are presented in Section IV of this report.
3. Prosecutorial Oversight Forums: In 2012, coalition
partners conducted a national prosecutorial oversight
tour to begin a conversation in the wake of the Thompson
decision. From January to October of that year, coalition
partners conducted forums in five geographically
diverse states: Arizona, California, Pennsylvania, New
York and Texas. The aim was to bring together panelists
with backgrounds from all aspects of the criminal
justice system to spark a meaningful dialogue about the
problem and recommendations for greater prosecutorial
accountability. Each stop of the tour explored a variety of
expert perspectives on prosecutorial accountability issues,
including those of current and former prosecutors, ethics
professors, members of state bar disciplinary committees,
defense counsel and judges.33 Following the tour, the
discussions from each forum were transcribed, coded and
analyzed for overarching themes. The results from this
analysis are presented in Section V of this report.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Section III: Review of existing research
on Prosecutorial Misconduct
“We’re not only dealing with willful and intentional misconduct or unavoidable instances where a
prosecutor’s error is inadvertent, but with a level of recklessness by prosecutors who practice dangerously
close to the ethical line.”
– Prof. Kathleen (Cookie) Ridolfi, Director of the Veritas Initiative and California panelist
Multiple organizations and journalists have investigated the
prevalence of prosecutorial error and misconduct in the
United States and how it is handled. They vary in geographic
focus, sample size and research methodology, but every
study arrived at the same conclusion: though allegations
of prosecutorial error or misconduct are widespread, few
prosecutors are formally disciplined for their actions, even
in cases of egregious intentional misconduct. Further, in the
rare case in which a prosecutor is sanctioned, consequences
are insignificant.34 Some of the more noteworthy
investigations include the following:
•	 The Veritas Initiative’s report Preventable Error: A Report
on Prosecutorial Misconduct in California 1997-2009,
looked at more than 4,000 California state and federal
appellate decisions and identified 707 cases in which
courts found prosecutorial misconduct. Of those cases, 67
prosecutors were found to have committed prosecutorial
misconduct in multiple cases. A total of seven prosecutors
were disciplined formally.35
•	 USA Today documented 201 instances between 1997 and
2010 where federal prosecutors were found by a judge to
have violated a law or an ethical rule.36 In many of these
cases, the misconduct was so severe that defendants’
convictions were overturned and some prosecutors were
reprimanded for their misconduct. However, only one
prosecutor was sanctioned and he received a one-year
suspension from practicing law.
•	 In a multi-part series for the Chicago Tribune, “Trial and
Error,” Ken Armstrong and Maurice Possley reviewed
more than 11,000 homicide cases across the country
involving prosecutorial misconduct between 1963 and
1999. Qualifying their results as “only a fraction of how
often prosecutors commit such deception—which is
by design hidden and can take extraordinary efforts
to uncover,” they found that courts reversed homicide
convictions against at least 381 defendants because
prosecutors either concealed exculpatory information
or presented false evidence.37 Of the 381 defendants, 67
had been sentenced to death.38 Despite the hundreds of
cases involving substantiated prosecutorial misconduct
and error, not a single state disciplinary agency publicly
sanctioned any of the prosecutors.

	One district attorney’s office fired a prosecutor who was
reinstated with back pay after successfully appealing the
termination. In a second case, the district attorney’s office
imposed a 30-day in-house suspension. A third prosecutor
received a 59-day suspension from practicing law, but only
because of other misconduct that occurred in the case.
Two others were indicted but the charges were dismissed
before trial.39
•	 The Liman Prosecutorial Misconduct Research Project
at Yale University surveyed the ethical rules and
disciplinary practices of all 50 states and the District of
Columbia. The investigators concluded that, in general,
the ethical rules that govern prosecutorial behavior fail to
adequately mitigate most forms of prosecutorial error and
misconduct. The study found that disciplinary systems are
largely inadequate, most state bar disciplinary authorities
are not designed to address prosecutorial misconduct, and
many discourage complaints through procedural barriers,
and these authorities have tremendous discretion and
often choose not to investigate prosecutors.40
	The study also noted that judges, prosecutors and defense
attorneys—those most likely to discover prosecutorial
misconduct—often fail to report it for myriad reasons,
including a culture that does not support reporting, poor
administrative processes and professional disincentives.41
•	 In the 2011 Fordham Law Review article, “The Supreme
Court Assumes Errant Prosecutors Will be Disciplined
by Their Offices or the Bar: Three Case Studies that
Prove that Assumption Wrong,” New York attorney Joel
Rudin provided case studies from the Bronx, Queens
and Brooklyn showing that misconduct by prosecutors
in those boroughs is almost never disciplined. Citing
deposition testimony and other discovery from civil
litigation relating to the misconduct, he revealed that the
three district attorney offices “have no codes of conduct,
no formal disciplinary rules or procedures, and no history
of imposing sanctions or any other negative consequences
on prosecutors who violate Brady and related due process
rules intended to guarantee defendants the right to a fair

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


“They [prosecutors] need to understand that it isn’t all about wins and losses, how many convictions
did I get; it’s how many meritorious convictions did I get? And until we get over that, we’ll never really
see the kind of justice system we want.”43
– Texas panelist Robert Schuwerk, professor, University of Houston Law Center,
author of leading treatise on Texas rules of professional conduct
In addition to reviewing existing research, the coalition
conducted independent research in each of the states
where forums were held.44 This review, conducted in 2011
in advance of the tour, included publically available court
decisions addressing allegations of prosecutorial misconduct
in Arizona, California, Pennsylvania, New York and Texas
and any publicly reported sanctions of prosecutors that
occurred during the same time frame.45
The research relied exclusively on reported court decisions
addressing findings of prosecutorial misconduct employing
a methodology with significant inherent limitations. There
are undoubtedly other instances of misconduct, including
findings by courts that are not published. For example, in
many cases handled by members of the Innocence Network,
prosecutors’ offices have agreed to vacate and dismiss old
convictions in the interests of justice based on credible
evidence of prior misconduct, but because the parties
agreed to the dismissal, there is typically no published
appeal or even a published order by the court that details the
Additionally, the overwhelming majority of criminal
prosecutions are resolved by guilty pleas and even innocent
defendants pled guilty in nearly 10% of the cases later
overturned by DNA evidence. These cases are rarely
subject to meaningful judicial review so it is impossible to
know how frequently prosecutorial error or misconduct
occurred even recognizing the different standard for review
established by the Supreme Court.46

Nearly 10% of the wrongful convictions later overturned
by DNA were people who entered guilty pleas, many of
whom took the pleas on the strong recommendation of
their lawyers. Had their lawyers been privy to undisclosed
evidence pointing to innocence, it is entirely possible that
those lawyers would not have recommended pleading guilty.

Absence of Disciplinary Action in the
Wake of Prosecutorial Misconduct
Results from our review revealed patterns similar to those
identified in earlier research. The coalition’s researchers
identified 660 criminal cases where courts confirmed
prosecutorial misconduct across the five forum states
(133 leading to reversals), and only one prosecutor from
these cases was disciplined (see infographic: Confirmed
Cases of Prosecutorial Misconduct, 2004-2008.47
While many of these cases may not have merited
suspensions, public censure, disbarment or criminal charges,
any instance that gives rise to a judicial finding of error
or misconduct deserves internal review at a minimum, so
that prosecutors’ offices can learn from past mistakes and
better train staff to prevent future misconduct. In each of the
states researched there were at least a handful of findings of
prosecutorial misconduct that were serious enough to merit
some form of disciplinary action.
It is troubling that the systems that the public should have
been able to rely upon to properly identify, review and

Confirmed Cases of Prosecutorial Misconduct, 2004-2008

660 527
total court findings
of prosecutorial
error or misconduct

of those errors were

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson

133 1
of those errors were
deemed HARMFUL

of those
was disciplined


address those misbehaviors were not equipped to handle
that responsibility. But perhaps the most troubling aspect
of all is that when our systems of prosecutorial oversight
of error and misconduct are not working, we miss valuable
opportunities to constructively review these actions, and
where necessary, prevent their reoccurrence or address them
to ensure a better response the next time.
Types of Crimes: Prosecutorial misconduct was
acknowledged most often in murder and sexual assault cases
(35% and 14%, respectively). Violent crime cases made up
the bulk (>75%) of the sample, however, violent crime cases
were less likely to lead to reversals than non-violent crime
cases (18% and 28%, respectively).
Types of Errors: The specific problems that were
acknowledged by courts varied widely, including improper
arguments/comments; improper witness examination;
Batson violations (the improper exclusion of a potential
juror on the basis of race or gender);48 Brady violations
(suppression of exculpatory evidence) and violations of
a defendant’s Fifth Amendment right not to testify, such
as improper commentary by the prosecutor as to the
defendant’s failure to take the stand.
Improper argument was the most common error found
by the courts, but it was the least likely to lead to reversal
(39 of 371). While the failure to disclose exculpatory
evidence (Brady violations) was found less frequently, when
confirmed, courts were more likely to reverse (find the
misconduct harmful) compared to any other category (38 of
66), with the exception of Batson violations, which, if found,
require automatic reversal.49
The following are a few examples of the wide-ranging error
or misconduct findings (as mentioned above, not one of the
prosecutors in these cases was publicly sanctioned):
1.	In Willis v. Cockrell,50 the prosecutor failed to disclose
to defense attorneys a pretrial psychological report
that concluded there was no evidence to support a
conclusion that the defendant posed a “future danger”
for the purposes of the Texas capital sentencing statute.
Despite these findings, prosecutors charged the defendant
with capital crimes and obtained a death sentence. On
appeal, the court found the nondisclosure both unlawful
and material and ordered a stay of Willis’ execution.
His conviction was vacated later that year because of
ineffective assistance of counsel. The district attorney
reinvestigated the arson murder, and new fire investigators
concluded that the testimony of the fire investigator who
claimed that Willis intentionally set the fire that killed two
people was based on outdated arson science.
2.	In People v. Spruill,51 the defendant was charged with
attempted sexual abuse. During summation, the
prosecutor asked the jury five times “if they would want
their own children in the place of the complainant.”52

The court found that the references to the jurors’ children,
which had a natural tendency to stir up “emotional
turmoil . . . cloud[ing] the mind and interfer[ing] with
the jury’s function to weigh and evaluate the evidence
objectively,” were decidedly “inappropriate.” However,
the court concluded that the other evidence against the
defendant was so strong that there was no reasonable
likelihood that the prosecutor’s improper comments
substantially influenced the outcome of the trial.
3.	In U.S. v. Rivas,53 a narcotics smuggling case, federal
prosecutors failed to disclose that their chief witness, the
defendant’s fellow seaman, had told the government that
he, not the defendant, had brought the package of drugs
on board the vessel in question. The U.S. Court of Appeals
for the Second Circuit reversed the conviction based on
the Brady violation, and Rivas was exonerated in 2004
after the government dismissed his indictment.54
While it is impossible to know the full extent of the problem,
it is clear that even among cases where serious misconduct
has been documented, the overwhelming majority of
prosecutors involved in those cases did not face substantial
discipline, if any at all.
The prosecutorial oversight system is clearly not as strong
and efficient as the Supreme Court in Thompson assumed
it to be. Like the research conducted by other organizations
and journalists, our investigation revealed a severely
inadequate, essentially non-functioning external disciplinary
process and a problematic lack of transparency. There is
a need to shift the criminal justice culture from a place of
secrecy and disregard for errors to a place where errors are
viewed as a deep reservoir of useful information that can
greatly improve not just individual prosecutorial work but
the system as a whole.
Panelist Jim Leitner, a former first assistant district
attorney in Texas, relayed a story from his time as
a defense attorney that exemplifies how even an
unintentional error can have a major impact on a
case: “I had a case, and it was fixing to go to trial,
and got a great offer, took it to my client, the client
took the offer, and afterwards, I sat down with the
prosecutor and asked him why he did that. He told
me some fault in the case that was obviously Brady.
And I said, ‘Why didn’t you tell me that?’ And his
response was, ‘Well, I didn’t believe it. And if I didn’t
believe it, it would be perjury, so therefore it is not
Brady.’” The fact that the prosecutor believed that,
since he did not find the evidence favorable to the
defendant reliable, he was under no obligation to
disclose it to the defense, exemplifies much of the
confusion regarding Brady and the lack of training
that contributes to the prevalence of error and

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


At the forum events in the five states, the panelists’
comments mirrored the conclusions that can be drawn
from the research: their states’ systems of prosecutorial
oversight were weak and/or had significant gaps, and these
shortcomings made it highly unlikely that prosecutorial
error and misconduct would be properly identified,
addressed and prevented.
There is not one particular issue that causes this failure
to properly address prosecutorial error and misconduct.
Arizona panelist Paul Charlton, a former U.S. attorney,
stressed: “By and large, prosecutors go to work within the
system of justice because they want to do what’s right. They
are men and women of good faith. But they can go wrong.”56

Culture of Underreporting
Despite ethical requirements to report attorney misconduct
in most jurisdictions,57 the criminal justice system’s culture
is not conducive to reporting misconduct and error. Judges,
fellow prosecutors and defense attorneys are reluctant to
report their colleagues out of fear of retribution, being
stigmatized as whistleblowers, and hurting relationships
with individuals who they work with on a daily basis.
Defense attorneys, in particular, have strong reason to
be fearful of alienating prosecutors who exercise such
enormous, unilateral discretion when it comes to plea offers
and sentencing recommendations for their clients. Most
actors in the justice system do not see reporting as a means
of improving the system and learning from mistakes; on
the contrary, panelists identified many reasons, even patent
disincentives, for lawyers and judges to choose not to report
instances of prosecutorial misconduct and error.
New York panelist Ellen Yaroshefsky said: “We had judges
talk about why they don’t do it. And part of it is that these
are people they deal with every single day. They don’t want
them to lose their licenses. They don’t want them to lose
their livelihoods.”58
She also detailed the issues defense lawyers face when
contemplating reporting prosecutorial misconduct, stating
that, “defense lawyers are fearful of reporting a prosecutor
because they feel that in the next case, their client will

Arizona panelist Judge Colin Campbell added, “If you as
a judge try to impose the rules of criminal procedure, the
attorneys will kill you. And they kill you in a thousand
different ways.”59
State disciplinary systems that require a formal complaint
to be made before reviewing a prosecutor’s actions—as
opposed to counting on prosecutors and bar oversight
entities to note problems as they occur—are therefore, as
a practical matter, a barrier to oversight. For example, in
Texas, the state disciplinary system is very much complaintdriven. Therefore, in order to investigate prosecutorial error
or misconduct, the disciplinary commission must receive an
actual complaint.
Panelist Betty Blackwell, former Chair of the Texas
Commission for Lawyer Discipline, explained: “One of the
biggest drawbacks is people don’t file complaints. Judges
don’t particularly want to file complaints on prosecutors
who they’re going to see every day in their court. [Defense]
lawyers don’t particularly want to file complaints on
prosecutors if [they] have to go back to that same prosecutor
on another day, on another case.”60
In most jurisdictions, trial and appellate judges are not
required to report prosecutorial misconduct and error. Yet,
even in jurisdictions where there is a requirement to report,
judges are failing to do so.61
California panelist Cookie Ridolfi explained that in the 10year period reviewed by the Veritas Initiative, there was not
one case where the court had actually reported misconduct,
even though state law requires them to do so.62 California
panelist Judge James Emerson added: “I know very few
judges who report to the state bar. And the issues that are
mandatory reportable, such as contempt—contempt is rarely
found by a court. As a matter of fact, in training we’re almost
advised against finding attorneys in contempt.”63
The failure of stakeholders to report prosecutorial
misconduct and error allows even the most egregious acts
of misconduct to remain unaddressed. It creates a false
sense that there is not a problem, prevents prosecutors who
repeatedly abuse their power from being identified and
sanctioned and does not allow for prosecutors’ offices to
use these mistakes as opportunities to educate and prevent
future harm.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


At the Texas forum, panelist Bob Perkins, a former
district court judge from Travis County, noted the
reluctance judges have in reporting prosecutors for
misconduct, saying: “I’ve only taken one case to the
grievance committee—a lawyer that lied to the court
and lied on the record. I took that to the grievance
committee here in Austin, and they found that it was
not an act of professional misconduct. So, I never
took anything more to the grievance committee
after that.”64

Impractical Reliance upon Bar
Disciplinary Committees
The panelists’ views on their respective states concur with
the research that states’ bar disciplinary committees (or
their counterparts) rarely investigate prosecutorial error
or misconduct. In the handful of situations where an
investigation is launched, the committees generally failed
to properly discipline the prosecutor who committed the
misconduct. The panelists described disciplinary systems
that lack adequate resources, are far more focused on
allegations directed at private attorneys who represent civil
clients, and are largely hidden from public scrutiny.
State bar disciplinary bodies typically possess the power to
sanction prosecutors professionally through censure (both
private and public), license suspension and disbarment.
Although the process varies from state to state, complaints
are typically received by a central intake office, which
determines whether there is a colorable claim worthy of
further investigation. In most jurisdictions, the majority of
complaints are dismissed at this stage.
The complaints that survive are forwarded to an
administrator for further investigation. The attorney against
whom the complaint was made is notified and given an
opportunity to respond before a formal complaint is filed. If
no formal complaint is filed, the proceedings and complaint
remain confidential. Additionally, there is no way to know
how many prosecutors received non-public sanctions
because most state bar disciplinary committees (or their
counterparts) do not provide this information.65
State bar disciplinary committees (or their counterparts)
have historically paid little attention to allegations of
misconduct by prosecutors.66 Many states have unnecessary
barriers, such as time limitations on filing and requirements
that documents be notarized, that discourage people from
filing grievances.67 Typically, state bar disciplinary or
grievance counsel websites provide instructions on how
individual citizens can file grievances against their lawyer
but fail to outline procedures for those who wish to report
prosecutorial error and misconduct.68

Bar disciplinary committee officials also described how, due
to limited resources, it is difficult to investigate instances of
extensive prosecutorial misconduct and error. According to
panelist Tom Wilkinson, president of the Pennsylvania Bar
Association and former chair and co-chair of its Legal Ethics
and Legal Responsibility Committees: “The disciplinary
council has very limited resources to plow through what
may be a very extensive record, going back a number of
years, involving a conviction . . . . They really don’t have the
resources to delve into what may be thousands of pages of
transcripts and other materials.”69
Panelist Joanne Hamilton, former secretary to the
Committee of Character and Fitness and former employee
at the Department Disciplinary Committee for the First
Department in New York, reported a lack of emphasis on
investigating egregious prosecutorial misconduct. “It’s my
experience that no one in the grievance and disciplinary
committee is assigned specific authority to root out
prosecutorial misconduct, cases or complaints. No one that
I’m aware of in any of the departments is the prosecutor of
prosecutors,” said Hamilton.70
The lack of transparency in disciplinary proceedings is
another element of the oversight system that undermines the
ability to document the extent to which allegations of serious
misconduct by prosecutors are being properly addressed.71
For example, in New York, a statute requires confidentiality
with respect to the disciplinary committee information,
including documents, complaints and procedures.72
“We don’t have any data. We have spent years trying to
gather data about New York’s disciplinary system, just
general data—how many people, without names, are ever
brought before a disciplinary committee for allegations of
prosecutorial misconduct? How many of those people are
ever disciplined? That data is unavailable,” explained New
York panelist Ellen Yaroshefsky.73
The lack of transparency or meaningful access to the state
bar disciplinary committees’ process and decisions makes
it virtually impossible for the public to assess the level of
protection it actually receives from such committees.

Appellate Review: Harmless Error
Doctrine Greatly Minimizes Problem
of Misconduct
Appellate courts use harmless error analysis to determine
whether an error that occurred in a lower court proceeding
is serious enough to require reversal of a criminal
conviction.74 If prosecutorial error or misconduct violates a
defendant’s constitutional right to a fair trial, a defendant’s
conviction must be overturned on appeal.75 However,
reversals are limited because the harmless error doctrine

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


generally precludes relief when the court finds that the
prosecutor’s actions did not fundamentally prejudice the
The data shows that the majority of prosecutorial
misconduct findings by courts are ruled harmless, meaning
that a court has concluded that it would not have changed
the outcome of the case had the error or misconduct not
been committed. Such rulings inherently minimize the
problem of prosecutorial misconduct and error, effectively
signaling to prosecutors that error or misconduct is
acceptable—completely disregarding that an ethical
violation has likely occurred—as long as it would not have
altered the case’s result.
This is troubling, particularly because research has shown
that misconduct in harmful error cases and misconduct
in harmless error cases is comparable—the harmless/
harmful distinction is not based on the prosecutor’s conduct
but on the perceived strength of the evidence against the
defendant.76 Without any real consequences, these findings
represent wasted, precious opportunities to address
the behavior that prompted the underlying finding of
prosecutorial misconduct.
Having a case reversed on appeal may be perceived as a
sanction for prosecutors because the state must re-try the
case or lose the conviction, but many argue that it does little
to effectively deter misconduct.77 Some argue that reversals
for misconduct punish the criminal justice system but not
the individual who is causing the problem. Because of that,
judges are often reluctant to find misconduct even though
they might see it and would want it correctly addressed.78

As New York Supreme Court Judge and former Queens
Assistant District Attorney Richard Butcher—a New York
panelist—explained: “The problem here is that what we’re
really doing is punishing the people . . . and the prosecutor
is only punished collaterally because it is the case that’s
damaged and not him or her personally.”79
Further, as mentioned earlier, courts rarely publish the
names of prosecutors (or defense lawyers) at the center of
misconduct decisions, so any deterrence that might develop
because prosecutors do not want their names publicly
attached to cases where error and misconduct has been
established is lost.80

Civil Remedies Impossible to Obtain
Civil lawsuits have proven equally ineffective as remedies for
prosecutorial error and misconduct. Prosecutors are granted
immunity from civil suits, even if their conduct at trial is
unlawful and malicious, or causes serious harm to defendants.
Prior to Thompson, the Supreme Court established a broad
rule of absolute immunity from civil liability for prosecutors
in Imbler v. Pachtman.81 The Imbler rule provides a state
prosecuting attorney who acts within the scope of his duties
in initiating and pursuing a criminal prosecution and in
presenting the state’s case absolute immunity from a civil
suit for damages under the Civil Rights Act of 1871.82
In the majority opinion, the court expressed concern that
prosecutors might be deterred from zealously pursuing their
law enforcement responsibilities if they faced the possibility
of civil liability and suggested that prosecutorial error and
misconduct should be referred to state attorney disciplinary

Joshua Kezer
Joshua Kezer was wrongfully convicted of murder in Benton, Missouri, in 1994.
In November 1992, the body of the victim, a female college student, was discovered by Mark Abbott, who claimed that
while attempting to report the crime from a pay phone at a gas station, he saw a white hatchback drive into the station.
Four months later, three inmates at the local jail told authorities that Joshua Kezer confessed to committing the murder.
In a fifth interview with police, Abbott altered his story and for the first time identified Kezer from a photo lineup as the
driver of the hatchback. Before trial, two of the informants recanted to Kezer’s attorney, but then later retracted their
recantations. No physical evidence connected Kezer to the crime. The state’s case was centered on the testimony of the
jailhouse informants and Abbott’s identification. Kezer was convicted of second degree murder in 1994.
A 2006 reinvestigation of the case revealed that the state failed to disclose key documents which demonstrated that Abbott
had made a statement to police 10 days after the murder that described and identified another man as the individual
he saw near the crime scene. Additionally, police notes that prosecutors said were destroyed later emerged, revealing
that Abbott, the eyewitness, had originally been a suspect in the murder. These Brady violations, and subsequent DNA
testing that excluded Kezer as the source of blood found under the victim’s fingernails, prompted a court to overturn his
conviction. Kezer was ultimately released and the murder charge dismissed in 2009. Kezer v. Dormire, Findings of Fact,
Conclusions of Law, and Judgment, Cause No. 08AC-CCOO293, (Circuit Ct. Cole Cty, Feb. 17, 2009). No action was taken to
examine the prosecutor’s conduct.
Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


The court went even farther in Thompson, holding that
individual prosecutors and the offices that employ them will
be largely beyond civil legal scrutiny when they err—or even
when they commit serious misconduct.
The court has distinguished prosecutors from many other
actors doing critical and difficult work in our society,
including doctors, other lawyers and police. Consequently,
victims of wrongful convictions in which prosecutorial
misconduct played a key role can almost never sue the
individual prosecutors responsible, or even the office that
failed to train or supervise the prosecutor in question.
In fact, in a recent letter to the editor of the New York
Times, retired Supreme Court Justice John Paul Stevens
advocated that the most effective means of avoiding future
injustices caused by prosecutors’ concealment of evidence is
through “[t]he rule of respondent superior—which requires
employers to pay damages for torts committed by their
employees in the ordinary course of business . . .” The rule,
he said, “should apply to state law enforcement agencies.”84

Dearth of Criminal Sanctions
If state bar authorities are hesitant to bring disciplinary
actions against prosecutors, it is not surprising that judges
are equally reluctant.85 Laws providing criminal sanctions for
intentional prosecutorial misconduct do exist in a number
of states, but in reality, prosecutions are rarely sanctioned
under such statutes, even in the most egregious cases.86
The Michael Morton case was an extremely rare instance
where a prosecutor was criminally charged and convicted
for his misconduct. The prosecution was possible only
because the record showed that the prosecutor made a
false statement on the record when, in response to a direct
question from the trial court, he claimed that he had nothing
favorable to disclose to the defense. Twenty-five years later,
Morton’s attorneys discovered that the prosecutor possessed,
but did not disclose, a highly detailed and exculpatory
statement from a child eyewitness to the murder.
The trial court also ordered the prosecutor to submit the
lead investigator’s reports for in camera review under Brady,
yet the prosecutor failed to include the key eyewitness

statement in his submission. Such direct questioning by
the court and order for in camera review are not common
and, in Morton’s case, only came about because the original
defense attorneys became suspicious when the prosecution
announced that it would not be calling its lead investigator
to testify.87
The handling of grave prosecutorial misconduct in the
prosecution of former Alaska Senator Ted Stevens is a much
more typical example. In that case, the special prosecutor
charged with investigating possible misconduct found that
the Justice Department lawyers committed ethics violations
by the deliberate and “systematic” withholding of critical
evidence pointing to Stevens’ innocence. However, the court
concluded it was unable to punish them because there was
no prior, express court order directing the government to
turn over the evidence.88
As these two cases illustrate, criminal prosecution of
prosecutors is complicated. Some panelists expressed the
view that prosecutors enjoy an unfair advantage when the
allegations of misconduct turn on the parties’ respective
credibility—for example, the lawyers’ conflicting claims as to
whether Brady material was disclosed or not.
As Arizona panelist Karen Clark, a principal with Adams &
Clark who represents lawyers in disciplinary proceedings,
said: “The judges believe the prosecutors and they don’t
believe the defense attorneys. And why is that? Because
three-quarters of the judges were prosecutors themselves.”89
Further, in most instances there is an inherent conflict of
interest because the prosecutor’s office is responsible for
initiating criminal proceedings against one of its own.90 The
individuals charged with investigating and indicting the
prosecutor are usually current or former coworkers of the
offending prosecutor.91 Such a proceeding also generates
negative publicity and scrutiny for the office, which is a
huge disincentive, especially when one considers that most
district attorneys are elected to office. This conflict of interest
was on display in the Thompson case, as the only prosecutor
who was ever actually disciplined in the case—a grand
jury was convened to prosecute others but subsequently
disbanded without indictments—was a former prosecutor
who took proactive steps to prevent the execution of an
innocent man.92

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Section Vi: Recommendations to improve
systems of prosecutorial oversight
The following is a list of recommended mechanisms that
could improve systems of oversight and accountability
by addressing the problems identified by the research
and panelists who spoke at the forums. This list does not
exhaustively discuss or comment on each of the ideas but
instead aims to identify a range of solutions policymakers
and practitioners can consider when designing or enhancing
current oversight systems.
Different stakeholders across the system, including
prosecutors, the local bar and the judiciary would be
responsible for implementing these system improvements.
Some are hardly controversial, calling for better internal
supervision and management within local prosecutors’
offices. Others suggest strengthening existing structures like
bar and judicial disciplinary powers. Still others go even
further, recommending newly created independent external
oversight entities or legislative clarification of liability for the
most egregious intentional acts.

1.	Formal Written Policies: Prosecutors’ offices should be
required to provide written guidelines and tools that are
updated and provided to all prosecutors annually to help
prosecutors make decisions about discovery and other
matters that implicate defendants’ fundamental rights
fairly, ethically, equitably and effectively. These policies
would need to be enforced through internal supervision.93

the prosecutor’s corrective actions and resulting history
into performance reviews, compensation reviews and
	On the flip side, prosecutors’ offices should ensure that
compliance with Brady and other fundamental ethical
obligations is recognized and rewarded. Supervisors
should ensure, for example, that a prosecutor who
discloses exculpatory evidence that leads to a favorable
outcome for a defendant or elects to dismiss a case
because of concerns about the reliability of the evidence,
is appropriately commended for his or her actions; hence,
securing a conviction at trial or a plea to serious charges
are not the only “wins” on the prosecutor’s docket that are
singled out for internal recognition.

1.	Pretrial Order to Comply with Model Rule 3.8 (Brady
Order): Judges should be encouraged to convene a
conference with the prosecutor and the defense attorney—
at a reasonable time prior to trial—and issue a specific
order directing prosecutors to produce all evidence that
“tends to negate the guilt of the accused or mitigate the
offense,” as required by the American Bar Association's
ethics rules. This should include the requirement that
prosecutors certify that they have contacted the relevant
law enforcement personnel to ensure that they are made
aware of any favorable evidence in the possession of other
agencies (as is required by Brady and its progeny) and
make certain all such evidence is disclosed as soon as

2.	Enhanced Training: Prosecutors should be required to
conduct training both at the outset of employment and
periodically throughout their tenure in ethical obligations
and specific discovery practices. Prosecutors could be
encouraged to collaborate with defender organizations on
training where possible to ensure mutual understanding
of rules and obligations. These training programs could
be specifically designed to address repeated mistakes
identified by internal supervision and courts—whether
considered harmless or harmful.

2.	Mandatory Reporting: In jurisdictions that do not
already have reporting requirements, judges should
be required to routinely report findings of error or
misconduct, regardless of the impact on the case, to state
entities with oversight responsibility. Procedures for
reporting should not be burdensome thereby discouraging

3.	Increased Transparency: Prosecutors’ offices should
be required to increase transparency through annual
reporting on office performance including, among other
indicators, those that show identified error, misconduct
and response.94

3.	State Supreme Court Monitoring: State supreme courts
should be charged with actively monitoring compliance
with requirements of judicial reporting and notification of
attorneys. Annual reports could be made available to the

4.	Internal Review: Prosecutors’ offices should be required
to develop internal review systems to review findings of
prosecutorial misconduct and other transgressions of
prosecutorial ethics rules. Offices should also incorporate
reports of error and misconduct (internal or judicial),

4.	Documentation of Agreements and Potential Benefits:
Courts should be required to make prosecutors produce
documentation (or otherwise notify the defense) as to all
agreements they have made with witnesses and jailhouse

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


informants, particularly ones concerning conferment of
benefits of any kind. This disclosure requirement should
also include notice of any discussions they may have had
with such witnesses regarding potential benefits that may
result from the witnesses’ cooperation, even if no formal
“agreement” has been reached.

Bar Oversight Entities
1.	Streamline Grievance Procedures: Grievance procedures
should be improved to ensure that they are more
accessible to potential claimants who have colorable
claims of misconduct and ineffective assistance of counsel.
Procedures should not have unnecessary barriers such as
time limitations for filing and notary requirements.
2.	Automatic Filing: Disciplinary committees should be
required to institute automatic filing of ethics complaints
that are triggered whenever a court finds (whether on
direct appeal, collateral review or otherwise) that a
prosecutor has behaved unethically. At a minimum,
disciplinary committees should be required to create a
system listing the names of prosecutors where there has
been a court finding of misconduct so that the disciplinary
committee will know if a prosecutor has a history of
committing misconduct if and when a complaint is filed
against that individual.
3.	Modify Statute of Limitations: State bar disciplinary
committees should lengthen their statutes of limitations
and provide for tolling in cases where the misconduct has
been found to be intentional.
4.	Transparency/Public Records: All grievance decisions
should be made available to the public and easily
searchable on an online database, which would serve to
inform interested parties of grievance dispositions. (At a
minimum, disciplinary committees could be required to
provide summary data on the number of complaints that
are filed and reviewed, the nature of the complaints, and
the outcomes of each investigation.)

Legislative Action
1.	Open File Discovery: To address concerns regarding
the disclosure of exculpatory material, states should
require open file discovery. One model is the North
Carolina statute which requires prosecutors to provide to
the defense before trial the complete investigative files,
including any material obtained by law enforcement,
investigators’ notes, the required recordation of all oral
statements and any other information obtained during
the investigation. In 2013, Texas passed a similar law that
includes protections sought by prosecutors to ensure that
witness privacy and safety is not jeopardized by disclosure
of such information. The rules should provide for work
product privileges to protect the prosecuting attorney’s
mental process while allowing the defendant access to
factual information collected by the state.96
2.	Independent Oversight: Prosecutorial oversight should
be vested in an independent state agency or within an
existing state agency (inspector general or attorney
general—except where the attorney general supervises the
prosecutor who is the subject of the investigation) with
the authority and resources to investigate allegations of
prosecutorial misconduct and impose remedies and/or
3.	Enact a Law Limiting Immunity for Prosecutors: States
should define the immunity for prosecutors by statute. As
retired Justice John Paul Stevens has powerfully argued
in the wake of the Thompson decision, prosecutorial
immunity is a federal judge-made rule of law that can and
should be overturned or modified by Congress through an
amendment to Section 1983 of the Civil Rights Act.98

Law Schools
1.	Ethics Training: Law schools should be required to
ensure that all students receive training in Brady and other
leading prosecutorial ethics rules as part of their standard
professional responsibility curriculum.

5.	Capacity to investigate: State bar disciplinary committees
should have adequate resources to investigate complaints
of prosecutorial error or misconduct. Similarly, staff
should be trained to identify and understand the issues
raised in these complaints, different from the complaints
made against private lawyers in civil proceedings, and
have access to transcripts and records as needed.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Section VIi: ConCLUSION
Despite the Supreme Court’s assertion in Thompson that
the public can rely on numerous systems of prosecutorial
oversight to ensure that prosecutors will act within ethical
and legal bounds, it is abundantly clear that, across the
country, our systems of prosecutorial oversight are either
failing or nonexistent.
No prosecutor begins his or her career with the intention
of imprisoning innocent people, and the vast majority of
prosecutors work hard every day to ensure that justice
is served. However, like other people, prosecutors are
susceptible to cognitive biases that distort perception and
like other human beings, tend to look at information in a
way that fits with their already constructed mindsets. There
is also a small, but important, minority of prosecutors who
commit intentional and egregious misconduct.
Regardless of the extent of error or misconduct committed
by a prosecutor—from the simplest of mistakes to the
intentional withholding of exculpatory evidence—these
actions undermine the accuracy of criminal trials and
threaten to create wrongful convictions at unacceptably high
rates. The courts, prosecutors’ offices, defense attorneys,
legislators and state bar disciplinary authorities must work
together to develop a comprehensive system of prosecutorial
oversight to ensure the quality of prosecutorial behavior.
This system must include training for all prosecutors
throughout their careers; effective internal systems at
prosecutors offices to identify, review and address error and
misconduct; improved bar oversight to enable and handle

legitimate claims regarding prosecutorial misbehavior; and
civil and criminal remedies in response to the most serious
and unconscionable abuses of prosecutorial power.
Given the magnitude of power granted to prosecutors
and the legal system’s reliance on their discretion to act
in the interests of justice, states across the nation must
ensure that there are effective systems in place to ensure
the appropriateness of prosecutorial action. The federal
government and every state in the nation should convene
experts from across their criminal justice and legal
communities to examine their state’s oversight systems and
develop action plans to initiate effective reforms.
The leadership for such a convening may come from the
judicial, executive or legislative branch leadership in any
state. However, any effort would be wise to include members
from all three branches, because the development and
maintenance of effective systems of prosecutorial oversight
will require the participation and support of everyone.
If such systems are established and properly supported
and administered, the quality of prosecutorial action—and
therefore criminal justice—will be much more reliably
dispensed. If such actions are given lip service only, then
we can expect the same for the quality of justice. Given
the importance of the problem and the impotence of
prosecutorial oversight systems in virtually every state in
the nation, the creation of effective systems of prosecutorial
oversight is a fundamental and necessary first step in
restoring public confidence in the justice system.

Keith Turner
Keith Turner was wrongfully convicted of aggravated sexual assault in Texas in 1983.
Turner, who worked at the same company as the victim, became the suspect after the victim
misidentified him both visually and by voice as the perpetrator. At his trial, the prosecutor commented
on Turner’s silence post-arrest, and repeatedly questioned Turner on the stand as to whether or not he
immediately protested his innocence or offered an alibi at the time of his arrest.
An intermediate appellate court found that the prosecutor committed error by questioning Turner about
his post-arrest silence and remanded the case for a new trial. The Texas Court of Criminal Appeals
reversed the intermediate court’s decision, finding that any error committed by the prosecutor’s
questioning was harmless, as the jury had been instructed to disregard the questions that infringed upon Turner’s
constitutional rights. Turner v. State, 719 S.W.2d 190 (Tex. Crim. App. 1986) rev’g 690 S.W. 2d 66, 68 (Tex. Ct. App. 1985).
Turner was released on parole in 1989, but still fought to prove his innocence. In May 2004, he obtained DNA testing which
exonerated him. Two days before Christmas 2005, Turner received a full pardon. No action was taken to examine the
prosecutor’s conduct.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Appendix A
Sources for State by State Search
of Public Disciplinary Decisions

Veritas Initiative reviewed public disciplinary records
to see if prosecutors were sanctioned for their error and

New York

Searched Pennsylvania online disciplinary decisions
tracking attorney discipline for the period 2004-2010 to see
whether any prosecutors were publicly sanctioned.

Searched Texas online disciplinary decisions between 2004
and November 2011 to determine how many prosecutors
have been publicly disciplined.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Appendix B
Prosecutorial Accountability Tour
List of Panelists
New York


Honorable Richard Buchter: New York Supreme Court
Judge and former Queens Assistant District Attorney.

Betty Blackwell: Former Chair of the Texas Commission
for Lawyer Discipline.

Maddy deLone (Moderator): Executive Director of the
Innocence Project.

Jennifer Laurin (Moderator): Assistant professor at the
University of Texas School of Law.

Ross E. Firsenbaum and Shauna Friedman: Senior
Associates at Wilmer Hale, who represented Arthur Ashe
Courage Award Winner Dewey Bozella, who was wrongly
convicted of murder due to police and prosecutorial
misconduct and was exonerated after serving 26 years in
New York prisons.

Jim Leitner: Harris County First Assistant District Attorney
in Austin Texas.

Sarah Jo Hamilton: Principal at Scalise & Hamilton, LLP,
and a former trial counsel and first deputy chief counsel to
the Departmental Disciplinary Committee for New York’s
First Judicial Department.

Honorable Robert Perkins: Former district court judge
in the 331st District Court in Travis County.

Michael Morton: After spending nearly 25 years in prison
for the murder of his wife, Michael Morton was exonerated
in 2011.

Robert Schuwerk: Professor at the University of Houston
Law Center.

Honorable Elisa Koenderman: New York Supreme Court
Judge and former Bronx Assistant District Attorney.
*John Thompson: Exoneree whose $14 million civil
award for the prosecutorial misconduct that caused him
to spend 14 years on death row was overturned by the U.S.
Supreme Court, Founder and Director of Resurrection
After Exoneration and Voices of Innocence.
**Emily West: Research Director of Innocence Project.
Ellen Yaroshefsky: Clinical Professor of Law and Director,
Jacob Burns Center for Ethics in the Practice of Law at
Cardozo School of Law.

Jim Belanger: Defense Attorney currently working at
Coppersmith Schermer & Brockelman PLC.
Hon. Colin Campbell: Former Judge of the Maricopa
County Superior Court.
Paul Charlton: Former prosecutor and former United
States Attorney. He is currently a partner at the Phoenix,
Ariz., law firm Gallagher & Kennedy.
Karen Clark: Former ethics counsel for the state bar
of Arizona.

	 *	 Panelist presented at all panels.
	**	 Panelist presented at the Texas and Pennsylvania panels.
Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Lindsay Herf (Moderator): Co-director of the Arizona
Justice Project.
Ray Krone: Exoneree who spent more than a decade
in prison, some of it on death row, before DNA testing
cleared his name.

David Angel: Special Assistant District Attorney,
and Director of the Conviction Integrity Unit in Santa
Clara County.

Kathy Mayer: Former Deputy Pima County Attorney.

Obie Anthony III: Exonerated after spending 17 years
behind bars for a crime he did not commit.

Keith Swisher: Associate Dean of Faculty Development
and Associate Professor at the Phoenix School of Law.

Robin Brune: Senior Trial Counsel for the California
State Bar.

Kathleen “Cookie” Ridolfi: Law professor at Santa
Clara University School of Law and the Director of
the Veritas Initiative.

Honorable James Emerson: Former Santa Clara County
Superior Court judge.

John Todd: Special Assistant Attorney General at
Arizona Attorney General’s Office.

Anne Bowen Poulin (Moderator): Professor of Law,
Villanova University School of Law.
Greg Rowe: Legislative Liaison for the Pennsylvania
District Attorneys Association.

Rachel Myrow (Moderator): Host of The California
Report for KQED Public Radio in San Francisco.
Tom Nolan: Defense attorney of Nolan, Armstrong
and Barton.
Kathleen “Cookie” Ridolfi: Law professor at Santa
Clara University School of Law and the Director of
the Veritas Initiative.
Jeff Rosen: District Attorney for Santa Clara County.

Honorable William R. Carpenter: Court of Common
Pleas, Montgomery County
Thomas G. Wilkinson, Jr.: President of the Pennsylvania
Bar Association and Partner at Cozen O'Connor.

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


1	 On April 1, 1940, then-Attorney General Robert Jackson gave a speech
to the United States attorneys who then were serving in each Federal Judicial
District across the country. In the speech, Jackson, who had been attorney
general for only three months, offered his views on what constituted proper,
ethical conduct by federal prosecutors. Attorney General Robert H. Jackson,
the Federal Prosecutor, Remarks at the Second Annual Conference of United
States Attorneys, United States Department of Justice, Washington, D.C. (Apr.
1, 1940) published in 31 J. of Crim. L. & Criminology 3-6 (1940).
2	 Connick v. Thompson, 131 S. Ct. 1350 (2011)
3	 Connick, 131 S. Ct. 1350, 1353 (2011) (holding that a single Brady violation
committed by one attorney in the New Orleans County District Attorney’s
office was insufficient to place the district attorney on notice of the need for
further training and therefore the office was not civilly liable based on a failureto-train theory of municipal liability).
4	Id. at 1358-66.
5	Id.
6	Id. at 1375 (Ginsburg, J., dissenting). All of the facts regarding the failure of
prosecutors to disclose are stated in the Supreme Court’s majority opinion or in
Justice Ginsburg’s dissenting opinion.
7	 Brief for Harry Connick et al. as Amici Curiae Brief of the National District
Attorneys Association and the California District Attorneys Association in
Support of Petitioners, Connick v. Thompson, 131 S. Ct. 1350 (2011) (No. 09571) 2010 WL 2394353.
8	 Connick, 131 S. Ct. at 1365-66.
9	Id. at 1370 (Ginsburg, J., dissenting).
10	 See 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), available at http:// [hereinafter Yale Study].
11	 See Section V.
12	 See Mary Nicole Bowman, Mitigating Foul Blows, Ga. L. Rev. (forthcoming
13	 See James R. Acker, The Flipside Injustice of Wrongful Convictions: When the
Guilty Go Free, 76 Alb. L. Rev. 1629, 1632 (2012).
14	 Emily M. West, Court Findings of Ineffective Assistance of Counsel Claims
in Post‐Conviction Appeals Among the First 255 DNA Exoneration Cases,
Innocence Project (September 2010), available at http://www.innocenceproject.
15	 In 2009, the Jacob Burns Ethics Center at Cardozo School of Law hosted
a conference with leading experts and scholars on Brady and other disclosure
obligations. In advance of the conference, the 75 participants were split into
six working groups, and during the conference, the groups met to develop best
practices on various issues related to disclosure obligations, including systems
for information management, training on disclosure and oversight systems
for accountability. The recommendations of the six groups were published in a
law review article for the Cardozo Law Review. New Perspectives on Brady and
Other Disclosure Obligations, 31 Cardozo L. Rev. 1943 (2010).
16	 See Robert L. Misner, Recasting Prosecutorial Discretion, 86 J. Crim. L.
& Criminology 717, 718 (1996) (“In the past thirty years . . . power has
increasingly come to rest in the office of the prosecutor. Developments in the
areas of charging, plea bargaining, and sentencing have made the prosecutor
the preeminent actor in the system.”); James Vorenberg, Decent Restraint of
Prosecutorial Power, 94 Harv. L. Rev. 1521, 1522 (1981) (“There is a broad
and rather casual acceptance of the fact that prosecutors often exercise greater
control over the administration of criminal justice than do other officials.”).
17	 See Angela J. Davis, Arbitrary Justice: The Power of the American
Prosecutor 125 (2007) (“Prosecutors make the most important of these
discretionary decisions behind closed doors and answer only to other
prosecutors. Even elected prosecutors, who presumably answer to the
electorate, escape accountability, in part because their most important
responsibilities—particularly the charging and plea bargaining decisions—are

shielded from public view.”). See also Tracey L. Meares, Rewards for Good
Behavior: Influencing Prosecutorial Discretion and Conduct with Financial
Incentives, 64 Fordham L. Rev. 851, 862 (1995) (“The prosecutor’s charging
discretion is, for the most part, unreviewable.”).
18	 See Brandon K. Crase, When Doing Justice Isn’t Enough: Reinventing the
Guidelines for Prosecutorial Discretion, 20 Geo. J. Legal Ethics 475, 477
(2007) (“The discretion afforded to prosecutors extends from the finest detail
of the case to the questions of whether to investigate, grant immunity, or even
whether to bring the charges at all.”). See also, Bordenkircher v. Hayes, 434 U.S.
357, 364 (1978) (“In our system, so long as the prosecutor has probable cause to
believe that the accused committed an offense defined by statute, the decision
whether or not to prosecute, and what charge to file or bring before a grand
jury, generally rests entirely in his discretion.”).
19	 James M. Doyle, Learning about Learning from Error, 14 Ideas
in American Policing 7-13 (May 2012), available at http://www.
20	 In October 2013, the U.S. Attorney for the Southern District of California
filed a motion to reverse and vacate the conviction of a defendant upon
reviewing the videotape of an oral argument which revealed that the prosecutor
had made an improper argument. Notably, the U.S. Attorney’s office stated that
it planned to use the video as a training tool for its Assistant U.S. attorneys. U.S.
v. Maloney, Motion to Summarily Reverse the Conviction, Vacate the Sentence,
and Remand to the D. Ct., C.A. No. 11-50311 (9th Cir. Oct. 7, 2013).
21	 John Hollway, “A Systems Approach to Error Reduction in Criminal Justice”
(2014). Faculty Scholarship. Paper 976.
22	Criminal Justice Section, Report to the House of Delegates, American Bar
Association at 1 (Aug. 2009) (accepted by the House of Delegates Aug. 9-10,
2009), available at
23	 See Anthony Batts, Maddy DeLone & Darrel Stephens, “Policing and
Wrongful Convictions,” New Perspectives in Policing Bulletin, U.S. Department
of Justice, National Institute of Justice 15-17 (2014), available at
pdffiles1/nij/246328.pdf (advocating for an organization accident model which
encourages police department to review errors as system wide weaknesses
instead of single-cause mistakes).
24	 Angela J. Davis, The Legal Profession’s Failure to Discipline Unethical
Prosecutors, 36 Hofstra L. Rev. 275, 277 (2007)(“Prosecutorial misconduct
encompasses a wide range of behaviors, including courtroom misconduct
(such as making inflammatory comments in the presence of the jury,
mischaracterizing evidence, or making improper closing arguments),
mishandling physical evidence (destroying evidence or case files), threatening
witnesses, bringing a vindictive or selective prosecution, and withholding
exculpatory evidence.”).
25	 See Smith v. Cain, 132 S. Ct. 627(2011). For a discussion of the case law
governing prosecutorial misconduct, see infra Section IV.
26	 United States v. Olsen, 737 F.3d 625, 632 (9th Cir. Dec. 10, 2013) (ord.
denying reh’g en banc), (C.J. Kozinski, dissenting).
27	 In the forum states, to the knowledge, prosecutors did not routinely keep
records of sanctions against prosecutors. Presumably, such records would be a
reliable source of data, but to our knowledge this is not systematically done.
28	 Schulte, Roth and Zabel provided invaluable assistance in the data analysis
and research for prosecutorial misconduct in New York.
29	 The search criteria included:
“prosecut! misconduct” & da(aft 12/31/2003 & bef 1/1/2009) ; “Brady
v. Maryland” & da(aft 12/31/2003 & bef 1/1/2009) % “ prosecut!
misconduct”;”Batson v. Kentucky” & da(aft 12/31/2003 & bef 1/1/2009) % “
prosecut! misconduct”; “Doyle v. Ohio” & da(aft 12/31/2003 & bef 1/1/2009)
% “ prosecut! misconduct”; “Griffin v. California” & da(aft 12/31/2003 &
bef 1/1/2009) % “ prosecut! misconduct”; “improper argument” & da(aft
12/31/2003 & bef 1/1/2009) % “ prosecut! misconduct”; “improp! cross!”
& da(aft 12/31/2003 & bef 1/1/2009) % “prosecut! misconduct”; “improp!
question!” & da(aft 12/31/2003 & bef 1/1/2009) % “prosecut! misconduct” %
“improp! cross!”; “failure to correct” & da(aft 12/31/2003 & bef 1/1/2009) %
“prosecut! misconduct”

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


30	 See Appendix A for state-by-state source information.
31	 It is important to provide some context about how courts rule on issues
relating to prosecutorial misconduct. When a defendant brings a claim of
prosecutorial misconduct, the courts examine the claim and generally rule in
one of the following ways:
	1.The claim is rejected—the claim is without merit and the prosecutor acted
within his or her professional bounds.
	2. The claim is waived due to the defense’s failure to make a proper
objection; the trial court correcting any possible error by sustaining
objections and/or jury instructions; or the claim is untimely.
	3. The claim is deemed harmless—the court offers language which
acknowledges the error or misconduct, but concludes that the conduct did
not fundamentally affect the fairness of the trial.
	4. The claim is deemed harmful—the court acknowledges the error or
misconduct and believes that the conduct unduly affected the outcome of
the trial, leading the court to set aside a conviction or sentence; declare a
mistrial; bar evidence from court.
32	 The Issue, Prosecutorial Oversight,
about-the-issue (last visited Apr. 13, 2015).
33	 See Appendix B for a list of panelists who participated in each state forum.
34	 Kathleen Ridolfi & Maurice Possley, N. Cal. Innocence Project, Preventable
Error: A Report on Prosecutorial Misconduct in California 1997-2009,(2010);
Ken Armstrong & Maurice Possley, Verdict: Dishonor, CHI.TRIB., Jan. 10, 1999,
available at
35	Ridolfi & Possley, supra note 31 at 3.
36	 Brad Heath & Kevin McCoy, “Prosecuting Offices’ Immunity Tested,”
USA Today, Oct. 6, 2010, available at
37	 Armstrong & Possley, supra note 31.
38	Id.
39	Id.
40	 Yale Study, supra note 9 at 234.
41	 Yale Study, supra note 9 at 234-40. The Yale Study’s findings were
developed from the following pieces of comparative data for each state: rules
of professional conduct, rules of lawyer disciplinary procedure, information
from each state’s disciplinary authority website, telephone interviews with bar
personnel, and supplemental statistical data compiled by the American Bar
Association (ABA) as part of its 2009 Survey on Lawyer Disciplinary Systems.
42	 Joel B. Rudin, The Supreme Court Assumes Errant Prosecutors Will Be
Disciplined By Their Offices or The Bar: Three Case Studies That Prove That
Assumption Wrong, 80 Fordham L. Rev. 537, 539-41 (2011).
43	Robert Schuwerk, Professor of Law, University of Houston Law Center,
Prosecutorial Oversight Forum at the University of Texas Law School (Mar. 29,
44	 The coalition looked at the published trial and appellate court decisions
addressing allegations of prosecutorial misconduct between 2004-2008.
In order to give the states’ bar disciplinary committees sufficient time to
investigate and decide whether or not to take action against a prosecutor, the
groups looked at disciplinary records from 2004 through 2010 for Arizona and
Pennsylvania and 2011 for California, New York and Texas.
45	 In the forum states, prosecutors did not routinely keep records of sanctions
against prosecutors. Presumably, such records would be a reliable source of
data, but to our knowledge this is not systematically done.
46	 United States v. Ruiz, 122 S.Ct. 2450 (2002).
47	 From 2004 to November 2011, 15 prosecutors across these five states were
publicly disciplined for prosecutorial misconduct, but none for conduct relating
to the cases in our sample. ​However, after the research for this report was
concluded, at least one prosecutor was disciplined for acts of​misconduct that
had occurred during the time frame studied here. In that case,​Burleson and
Washington County Texas District Attorney Charles Sebe​sta was disbarred for
his misconduct in the wrongful capital murder conviction of Anthony Graves,

whose conviction was reversed in 2006 by the United States Court of Appeals
for the Fifth Circuit because the prosecution failed to turn over evidence
pointing to Graves’​innocence. After wrongly serving 18 years (including 12 on
death row), Graves was exonerated in 2011. After the Texas legislature passed
a law amending the statute of limitation to allow exonerees to file a grievance
four years after their release from prison, Graves did so in 2015. In June 2015,
after a rare four-day disciplinary hearing, ​a panel of the State Bar determined
that Sebe​sta withheld evidence and used false testimony in securing Graves’​
capital murder conviction and stripped him of his law license. The disbarment
was rendered final in February 2016, when the Board of Disciplinary Appeals
upheld the State Bar of Texas’s findings and recommendation.
48	In Batson v. Kentucky, 476 U.S. 79 (1986), the Supreme Court held that the
Equal Protection Clause prohibits the use of a peremptory strike during voir
dire solely on the basis of the potential juror’s race. In J.E.B. v. Alabama, 511
U.S. 127 (1994) extended the Batson rule to prohibit gender-based peremptory
challenges. In order to prove a Batson violation, the party making the objection
must establish by evidence a prima facie case of purposeful discrimination, at
which point the other party has the burden of providing a neutral explanation
for the use of the peremptory strike.
49	 See Kathleen Ridolfi, Tiffany M. Joslyn & Todd Fries, National Association
of Criminal Defense Lawyers, Material Indifference: How Courts Are Impeding
Fair Disclosure In Criminal Cases (2014) (concluding based on its review of
federal Brady claims that late disclosure or complete nondisclosure of favorable
information is rarely found to violate Brady).
50	 No. P–01–CA–20, 2004 WL 1812698 (W.D. Tex. Aug. 9, 2004).
51	 5 A.D.3d 318 (N.Y. App. Div. 2004).
52	Id. at 320.
53	 U.S. v. Rivas, 2004 WL 1658368.
55	 Jim Leitner, First Assistant District Attorney, Harris County, Texas,
Prosecutorial Oversight Forum at the University of Texas Law School (Mar. 29,
56	 Paul Charlton, former United States Attorney for the District of Arizona,
Prosecutorial Oversight Forum at the Phoenix School of Law (Apr. 26, 2012).
57	 See Hon. William M. Hoeveler, Ethics and the Prosecutor, 29 Stetson L.
Rev. 195, 195 (1999).
58	 Ellen Yaroshefsky, clinical professor of law, Benjamin N. Cardozo School of
Law, Prosecutorial Oversight Forum at the Benjamin N. Cardozo School of Law
(Feb. 6, 2012).
59	 The Honorable Colin Campbell, Judge (Ret.), Maricopa County Superior
Court, Prosecutorial Oversight Forum at the Phoenix School of Law (Apr. 26,
60	 Betty Blackwell, former chair, Texas Commission for Lawyer Discipline,
Prosecutorial Oversight Forum at the University of Texas Law School (Mar. 29,
61	 Final report, California Commission on the Fair Administration of Justice
at 71, available at
62	 Kathleen “Cookie” Ridolfi, former director, Northern California Innocence
Project, Prosecutorial Oversight Forum at Santa Clara University School of Law
(Oct. 1, 2012).
63	 The Honorable James Emerson, Judge (ret.), Santa Clara County Superior
Court, Prosecutorial Oversight Forum at Santa Clara University School of Law
(Oct. 1, 2012).
64	 The Honorable Robert Perkins, District Court Judge (ret.), 331st District
Court, Travis County, Tex., Prosecutorial Oversight Forum at the University of
Texas Law School (Mar. 29, 2012).
65	 See Yale Study, supra note 9 at 235.
66	 Walter W. Steele, Jr., Unethical Prosecutors and Inadequate Discipline, 38 Sw.
L.J. 965, 966 (1984) (noting “both scholars and bar grievance committees have
paid scant attention to prosecutorial ethicality, and consequently, prosecutors
may have developed a sense of insulation from the ethical standards of other

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


67	 See Yale Study, supra note 9 at 234.

86	 Id. at 85.

68	 Id. at 243.

87	 In re Honorable Ken Anderson (A Court of Inquiry), Probable Cause Order
and Findings of Fact and Conclusions of Law, Cause No. 12-0420-K26, (26th
Dist. Ct., Williamson Cty., Tex.) (Apr. 19, 2013) (Sturns, J.), available at http://

69	 Thomas G. Wilkinson, Jr., president, Pennsylvania Bar Association,
Prosecutorial Oversight Forum, Philadelphia, PA. (Oct. 1, 2012).
70	 Sarah Jo Hamilton, former trial counsel and deputy chief counsel,
Departmental Disciplinary Committee for New York’s First Judicial
Department, Prosecutorial Oversight Forum at the Benjamin N. Cardozo
School of Law (Feb. 6, 2012).
71	Id. at 245 (“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.”).
72	 See N.Y. Jud. Law § 90(10) (2013).
73	 Ellen Yaroshefsky, clinical professor of law, Benjamin N. Cardozo School of
Law, Prosecutorial Oversight Forum at the Benjamin N. Cardozo School of Law
(Feb. 6, 2012).
74	 See generally, Harry T. Edwards,To Err Is Human, But Not Always Harmless:
When Should Legal Error Be Tolerated? 70 NYU L. Rev. 1167 (1995) (discussing
the flaws of Harmless Error Doctrine).
75	Id.
76	Ridolfi & Possley, supra note 31 at 19-23.
77	 Id. at 71.
78	Id. at 50.
79	 The Honorable Richard Butcher, judge, New York Supreme Court, Queens
County, N.Y., and former assistant district attorney, Queens County, New York,
Prosecutorial Oversight Forum at the Benjamin N. Cardozo School of Law
(Feb. 6, 2012).
80	Ridolfi & Possley, supra note 31 at 50.
81	 424 U.S. 409 (1976).
82	Id. at 423-26.
83	Id. at 422-26.
84	 John Paul Stevens, “Letter to the Editor,” New York Times, Feb. 18, 2015,
available at
85	 See Alexandra White Dunahoe, Revisiting the Cost-Benefit Calculus of the
Misbehaving Prosecutor: Deterrence Economics and Transitory Prosecutors, 61
NYU Ann. Surv. AM. L 45, 84 (2005) (“[E]ven where a knowing deprivation
is proven, many judges and juries are hesitant to impose criminal sanctions
for ‘technical’ constitutional violations …. Even in the context of extreme
prosecutorial abuse, however, judges may prefer to use a less severe, quasicriminal remedy available to sanction the misconduct, such as the contempt

88	 United States v. Theodore F. Stevens, Order, Crim. No. 1:08-cr-00231-EGS
(D.D.C. 2009)(Nov. 21, 2011)(Sullivan, J.).
89	 Karen Clark, former ethics counsel, State Bar of Arizona, Prosecutorial
Oversight Forum at the Phoenix School of Law (Apr. 26, 2012).
90	 See generally, Heather Schoenfeld, Violated Trust: Conceptualizing
Prosecutorial Misconduct, 21 J. Contemp. Crim. Just. 250 (2005); Barbara
O’Brien, A Recipe for Bias: An Empirical Look at the Interplay Between
Institutional Incentives and Bounded Rationality in Prosecutorial Decision
Making, 74 Mo. L. Rev. 999 (2009); Bennett L. Gershman, Why Prosecutors
Misbehave, 22 Crim. L. Bull. 131 (1986).
91	 See Angela Davis, Prosecutors Who Intentionally Break the Law, 1 Am.
U. Crim. L. Brief 16 (2006) (describing the reasons for the criminal justice
system’s seemingly deferential posture toward prosecutors).
92	In re Riehlmann, 891 So. 2d 1239 (La. 2005) (regarding formal charges filed
against respondent by the Office of Disciplinary Counsel).
93	 The ABA Standing Committee on Ethics and Professional Responsibility
in a recently issued formal opinion stated that Model Rules 5.1 and 5.3 require
prosecutors with managerial authority and supervisory lawyers to make
“reasonable efforts to ensure” that all lawyers and non-lawyers in their offices
conform to the Model Rules, including 3.8. See ABA Comm. on Ethics & Prof ’l
Responsibility, Formal Op. 467 (2014).
94	Rachel Barkow, Organizational Guidelines for the Prosecutor’s Office, 31
Cardozo L. Rev. 2089, 2116 (2010) (recommending an office culture that
emphasizes compliance will improve prosecutors’ behavior).
95	 Barry Scheck & Nancy Gertner, Combating Brady Violations With An
“Ethical Rule” Order for the Disclosure of Favorable Evidence, NACDL, 37
Champion 40 (May 2013).
96	 In response to the prosecutorial misconduct that led to Michael Morton’s
wrongful conviction, Texas passed the Michael Morton Act, which established a
more extensive open file discovery process. Michael Morton Act, S.B.1611, 83rd
Leg., 1st Sess. (Tex. 2013).
97	Yaroshefsky, Wrongful Convictions: It Is Time to Take Prosecution Discipline
Seriously, 8 D.C. L. Rev. 275 (2004)
98	 John Paul Stevens, retired associate justice, United States Supreme Court,
Remarks at the Equal Justice Initiative Dinner Honoring Justice Stevens 1012 (May 2, 2011) (transcript available at

Innocence Project
Prosecutorial Oversight: A National Dialogue in the Wake of Connick v. Thompson


Contact Paul Cates
at (212) 364-5346 for
more information
about this report.

Innocence Project, Inc.
40 Worth Street, Suite 701
New York, NY 10013

Benjamin N. Cardozo
School of Law,
Yeshiva University

Donate online at