THE JUSTICE PROJECT

Prosecutors relied heavily

Jailhouse Snitch
Testimony
A Policy Review

on the testimony of a
jailhouse snitch to convict
Wilton Dedge of rape.
Dedge spent twenty-two
years in prison before
he was released and
exonerated of the crime.
Proper safeguards on
jailhouse snitch testimony
could have prevented
this injustice.

Introduction .............................................. 1
Recommendations & Solutions ........ 2
Grounds for Reform .............................. 3
The Legal Landscape
Benefits & Costs

.......................... 6

.................................... 7

Profiles of Injustice ................................ 8
Snapshots of Success ........................ 14
Voices of Support ................................ 15
Questions & Answers ........................ 16
A Model Policy...................................... 18
Statistics

.................................................. 20

Literature

................................................ 21

THE JUSTICE PROJECT

INTRODUCTION

T

perate to attain sentence reductions, snitch testimony is
estimony from in-custody informants, often
widely regarded as the least reliable testimony encounreferred to as “jailhouse snitches,” has been
tered in the criminal justice system.
widely used in the American criminal justice system.
In the face of serious concerns about the inherent
Witnesses with special knowledge of criminal activity
unreliability of jailhouse snitches and the miscarriages
enable police and prosecutors to apprehend and prosof justice they cause, there are measures that states can
ecute criminal suspects. Thus, utilizing cooperating
implement to help ensure that the use of cooperating
witnesses in order to obtain evidence of criminal
witness testimony does not undermine fairness and
activity is an important tool.
accuracy in criminal trials. Pragmatic changes requirNonetheless, the motive to fabricate testimony is
ing corroboration of the facts to which an informant
inherent in a system in which snitches are often
testifies, pretrial disclosures, reliability hearings, and
rewarded for their testimony. Jailhouse snitches, who
special jury instructions raise the
often testify at pivotal
evidentiary threshold and
moments in criminal
All wrongful convictions detract
improve the quality of evidence
prosecutions, have been
from the public’s faith in the
presented at criminal trial.
shown to go to great
fair administration of justice,
Courts raise standards for the
lengths to deceive and
admissibility of snitch testimony
misinform in the hopes
but the cost is especially high
and ensure that judges and juries
of improving their curwhen wrongful convictions
are able to make more informed
rent situations. With litdecisions about the relative
tle or nothing to lose,
result from the testimony of
credibility of jailhouse snitch
and everything to gain,
questionable
witnesses.
testimony by requiring greater
cunning and unscrupuscrutiny. By implementing these
lous jailhouse snitches
pragmatic changes within the context of courtroom
invent narratives and crime details that mislead law
procedures already in place, states can improve the
enforcement officers and contribute to appalling misquality of evidence presented at criminal trials.
carriages of justice.
This policy review has been designed to facilitate
A 2005 report by the Center on Wrongful
communication among local law enforcement officers,
Convictions at Northwestern University School of
prosecutors, defense attorneys, judges, and others
Law found that snitch-dependent prosecutions are a
1
regarding the best practices and methods for enhancleading cause of wrongful convictions in capital cases.
ing the evidentiary value of a highly unreliable brand of
In fact, a survey of all cases involving individuals later
cooperating witness testimony. By presenting the sucexonerated by DNA testing showed that in over fifteen
cessful methods employed in individual jurisdictions, as
percent of cases, a jailhouse snitch testified against the
2
well as the reasoning behind them, we hope to create a
defendant.
dialogue around recommendations that will enhance
The problems that arise when prosecutions rely on
the quality of evidence relied upon in criminal trials, as
cooperating witnesses vary with the type of benefit
well as confidence in our system of justice.
conferred upon a witness in exchange for his or her tesAll wrongful convictions detract from the public’s
timony. Compensation of “jailhouse snitches” who
faith in the fair administration of justice, but the cost
provide incriminating testimony against a suspect, freis especially high when wrongful convictions result
quently one with whom they share a jail or prison cell,
from the testimony of questionable witnesses. While
often takes the form of a favorable plea to a lesser
this review is limited to a discussion of the problems
charge or a reduction in sentence. Other types of criminherent in the use of jailhouse snitch testimony,
inal witnesses, such as accomplice witnesses and outmany of the policy improvements recommended here
of-custody informants, can be compensated by the
could be considered in the context of other types of
state either through immunity from prosecution or
cooperating witnesses compensated by the state.
reduced charges. Because jailhouse snitches are so des1
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

RECOMMENDATIONS & SOLUTIONS

J

conclude that the jailhouse snitch’s testimony is sufficiently reliable to submit to the jury by considering all
factors that bear on the credibility of the jailhouse
snitch, based on all information made available
through written pretrial disclosures.
The testimony of a jailhouse snitch can often be
powerful evidence at trial, overshadowing the obvious
incentives for fabrication with compelling accounts of
criminal conduct. Through improved standards,
states can ensure that evidence presented in a courtroom and before a jury is of a sufficient quality to
enable more reliable outcomes.

ailhouse snitch testimony poses special challenges
to fairness and accuracy in criminal trials. When
the state offers a benefit in exchange for testimony,
whether that benefit is explicit or implied, the incentive for incarcerated individuals to fabricate evidence
dramatically increases. Some informants may fabricate testimony in an effort to curry favor with prosecutors apart from any promise or implied benefit.
Though the legal system is designed to weed out
perjured testimony through adversarial procedures such
as cross-examination, the protections currently in place
have proven starkly inadequate to safeguarding against
unreliable testimony by witnesses with powerful incentives to lie. Remarkably, the use of jailhouse snitch testimony continues to be largely unregulated by state legislatures or courts despite frequent, documented miscarriages of justice and instances of wanton abuse.
The costs to the individual and to the state are high
when snitch testimony leads to the wrongful conviction
of an innocent person. Because perjured testimony has
played a prominent role in documented cases of wrongful conviction in this country, jurisdictions must examine
and implement safeguards designed to subject jailhouse
snitch testimony, and the process by which such testimony is acquired, to higher levels of scrutiny and care.

CORROBORATION
States should adopt corroboration requirements
for jailhouse snitches to mitigate the inherent risks
incentivised witness testimony carries.
Many law enforcement officers and prosecutors
seek to corroborate at least a portion of the information provided by informants for the purpose of determining witness credibility, which has bearing on charging decisions as well as trial strategy. Nonetheless, the
manner in which the prosecution may seek internal
corroboration of jailhouse snitch testimony is largely a
closed-door process. To inject a greater degree of
transparency, oversight, and neutrality into the process,
prosecutors should be required to disclose and present
any information corroborating the witness’ testimony.
If the state is unable to corroborate the facts of snitch
testimony, courts should limit the purposes for which
such unsubstantiated testimony is used at trial.

WRITTEN PRETRIAL DISCLOSURES
States should adopt rules requiring mandatory,
automatic pretrial disclosures of information related to
jailhouse snitch testimony. Specifically, states should
require the prosecution to make written disclosures
regarding the circumstances of cooperation agreements and any other information about the credibility
of a jailhouse snitch. Such disclosures should occur
prior to any criminal trial or proceeding in which the
prosecution intends to call the informant to testify.
Disclosure of this information ensures that defendants
can conduct meaningful cross-examination.

CAUTIONARY JURY INSTRUCTIONS
States should adopt cautionary jury instructions in
all cases where the testimony of a jailhouse snitch is
used. The jury should be instructed to take into
account several factors indicating the extent to which
the testimony is credible, including: 1) explicit or
implied inducements that the jailhouse snitch received,
may receive, or will receive; 2) the prior criminal history of the informant; 3) evidence that he or she is a
“career informant” who has testified in other criminal
cases; and 4) any other factors that might tend to render the witness’ testimony unreliable. Special jury
instructions ensure that jailhouse snitch testimony is
examined and weighed with proper caution.

PRETRIAL RELIABILITY HEARINGS
States should adopt rules mandating pretrial
determinations of reliability in cases where the prosecution intends to employ jailhouse snitch testimony.
In a pretrial reliability hearing, the court is able to
perform a “gatekeeper” function when admitting the
testimony of the jailhouse snitch. The court must
2

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

GROUNDS FOR REFORM

A

n incarcerated individual has particular incentive
to provide information in exchange for leniency,
a reduced sentence, or other remuneration. Incarcerated
individuals, in a system that relies on jailhouse snitches, risk little and can potentially gain much from lying
to authorities.
Though fabricated snitch testimony continues to
contribute to the mounting record of wrongful convictions in this country, state legislatures and courts
have been slow to curb excesses or abuse. In large
part, well-meaning police and prosecutors demonstrate due diligence in utilizing testimony by jailhouse
snitches; however, few safeguards are currently in
place to guide prosecutorial discretion or to ensure
that juries weigh the testimony of these informants
with proper care.
The recommendations in this policy review,
explained in greater detail below, improve the informant process by ensuring greater access to critical
information and giving the court a greater hand in
determining reliability. Through more neutral and
transparent use of snitch testimony, states ensure that
proper safeguards are in place to protect against perjured testimony and increase the reliability of outcomes in criminal cases. By improving the quality of
snitch testimony at trial through these reforms, states
improve the use of snitch testimony at all phases of
the criminal justice process.

required to disclose any “material” information that
might exculpate the defendant in pretrial discovery;4
however, this rule does not mean that prosecutors are
required to disclose all of the circumstances under
which informant witnesses come to cooperate with the
state — information that is critical to proper determinations of reliability.5 The additional burden of implementing greater pretrial disclosures would be minimal
considering the existing systems in place for the
exchange of information as a requirement of Brady.
States should adopt or extend rules to mandate
written pretrial disclosure of the following: statements made by the accused to the jailhouse snitch;
incentives that the witness received, will receive, or
may receive in exchange for testimony (e.g., promises
for sentence reductions, offers to lesser pleas,
improved incarceration conditions for in-custody witnesses, or anything else of value); whether the witness
has agreed to testify at prior criminal trials and, if so,
how many times he or she has done so (or agreed to
do so) and whether the witness has received any previous benefits for testimony; the complete criminal
history of the jailhouse snitch; whether at any time
prior to trial the witness has recanted his or her testimony or made statements inconsistent with the testimony to be presented at trial; and anything else bearing on the witness’ credibility.6
By specifying that the disclosures be in written
form, this recommendation helps ensure the accumulation of detailed records of all interactions between the
government and the informant witness prior to trial.

WRITTEN PRETRIAL DISCLOSURES
The adoption of mandatory, automatic pretrial
disclosures related to jailhouse snitch testimony
would allow for a complete airing of all relevant information bearing on a jailhouse snitch’s credibility.
Mandatory disclosures create a more transparent
process, allowing for meaningful oversight and adversarial challenge. In fact, the effectiveness of the legal
system’s built-in safeguard of cross-examination is
almost entirely dependent upon the level of pretrial
disclosures. Because the processes by which jailhouse
snitches are compensated and their testimony is
developed are largely hidden from view (and from triers of fact), current procedural safeguards are unable
to guard against untruthful testimony.3
Under the rule articulated by the U.S. Supreme
Court in Brady v. Maryland, prosecutors are already

PRETRIAL RELIABILITY HEARINGS
American jurisprudence has long wrestled with
the problems inherent in compensating witnesses —
monetarily or otherwise — in exchange for truthful
testimony. Payment of any sort in exchange for testimony creates a motive for a witness to lie. Though
paying witnesses is largely considered unethical and
even illegal as a general rule,7 there are several commonly held exceptions.8
In the context of expert witnesses, for example,
payment for testimony (or expertise) is an accepted
practice. American courts allow witnesses who are
leading professionals in their fields to receive compensation for their testimony because the subject
3

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

or her testimony, will generally incriminate himself or
matter of expert testimony is beyond the common
herself to some degree; jailhouse snitches, on the
knowledge of a layman or of the court.9 In civil trials,
other hand, expect a potential gain while risking little
courts require that certain indicia of reliability be met
or nothing in testifying against a defendant.
before an expert is allowed to testify in exchange for
Illinois has recognized the fallibility of jailhouse
money. For example, a pretrial “Daubert hearing” is a
snitch testimony and its potential harm. In April 2002,
requirement established by the U.S. Supreme Court
the Illinois Governor’s Commission on Capital
specifying that courts must determine the reliability
Punishment identified “a number of cases where it
of expert witnesses before their testimony is presentappeared that the prosecution relied unduly on the
ed to a jury.10 In terms of scientific expert testimony,
uncorroborated testimony of a witness with somefor example, the court must not only determine
thing to gain.” 15 As a result, Illinois has passed a prowhether the scientific expert is knowledgeable of the
issues presented, but must also establish that the convision allowing a court to decertify a death penalty
tent of the expert’s testimony is reliable under acceptcase when it finds that the evidence against the defened standards within the field.11
dant, which led to the conviction, was limited to the
uncorroborated testimony of an accomplice or a jailOur criminal justice system does not afford the
house snitch.16 Similarly, the California Commission
same pretrial procedural safeguards in criminal cases
involving compensated jailhouse snitches — even in
on the Fair Administration of Justice, established to
capital cases. Similar in
examine the adtheory to the function it
ministration of
By improving the quality of snitch
serves with respect to
criminal justice
testimony at trial through these reforms,
expert witnesses, courts
in California and
should perform this
recommend
states improve the use of snitch testimony
“gatekeeper” function
safeguards, has
at
all
phases
of
the
criminal
justice
system.
in any criminal proceedproposed three
ing or trial in which the
bills designed to
state presents a jailhouse snitch witness.12 Because the
address the leading causes of wrongful convictions,
including a bill to curb false testimony by jailhouse
stakes are so high in felony cases, and the propensity for
informants by requiring corroborating evidence for all
inadvertent bias is so great in the criminal adversarial
such testimony.17 The American Bar Association, in a
system, a reliability determination with respect to jailhouse snitches should be made by a neutral, objective
2005 resolution, urged “federal, state, local, and terriparty and not by the prosecutor alone. The best policy
torial governments to reduce the risk of convicting the
for ensuring the integrity of the criminal justice system
innocent, while increasing the likelihood of convicting
is a requirement that the prosecution bear the burden of
the guilty, by ensuring that no prosecution should
proof in showing that jailhouse snitch testimony is sufoccur based solely upon uncorroborated jailhouse
ficiently reliable to be put before a jury in all criminal
informant testimony.” 18
prosecutions. At the very least, this determination
While at least seventeen states have taken steps
should be made in capital cases, as in Illinois.13
toward expanding corroboration requirements to
include testimony offered by jailhouse snitches, most
states do not have legal safeguards against this risk.19
CORROBORATION
Several states, including California, Illinois, and
The California Commission on the Fair Administration
New York, have recognized the inherent unreliability
of Justice recommends that state legislatures enact
of testimony offered by an accomplice, which has
statutory requirements for the corroboration of jailresulted in legislation requiring that accomplice testihouse snitch testimony. Courts, according to the
14
mony be corroborated. While testimony provided
California Commission recommendations, must only
admit testimony corroborated by evidence that conby an accomplice is inherently suspect, and corrobonects the defendant with the commission of the offense
ration requirements should be implemented across
charged or the special circumstance(s) or aggravating
the board, the testimony of a jailhouse snitch presents
factor(s) to which the jailhouse snitch testifies. Such evipotentially greater risks. An accomplice, through his
4
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

to take into account several factors that shed light on
the extent to which the testimony is reliable.
Specifically, the presiding jury should consider all factors required through pretrial disclosures and/or considered in pretrial determinations of reliability. The
factors should include incentives that the witness
received, will receive, or may receive in exchange for
testimony (e.g., promises for sentence reductions,
offers to lesser pleas, improved incarceration conditions for in-custody witnesses, monetary rewards, or
anything else of value). Judges should also consider
whether the witness has agreed to testify at prior criminal trials and, if so, how many times he or she has
done so (or agreed to do so) and whether the witness
has received any previous benefits for testimony, as
well as the complete criminal history of the informant
witness. Finally, judges should also consider whether
at any time prior to trial the witness has recanted testimony or made statements inconsistent with the testimony to be presented at trial; and anything else bearing on the witness’ credibility.23
While cautionary jury instructions should not be
considered a sufficient safeguard against informant
perjury in and of themselves, they should be given by
courts as follow-through measures to reinforce the
dependability of the determinations made by judges
at pretrial reliability hearings.

dence must go beyond demonstrating merely that the
offense took place or that special circumstances or
aggravating factors occurred. Corroborative evidence
must demonstrate not only that the events described by
the snitch are correct, but must also demonstrate that
the snitch’s story factually links the offense to the
accused. Further, the testimony of another snitch must
not be considered adequate corroboration.20
It is important to note that corroboration
requirements alone are not sufficient to prevent the
risks inherent in jailhouse snitch testimony. While
corroboration requirements for jailhouse snitch testimony are critical because “the existence of corroboration is usually a threshold question for the
judge,” 21 in many cases it may prove to be an insufficient measure to counteract the inherent unreliability of this type of testimony. Consequently, without other measures such as written disclosures, reliability hearings, and jury instructions, a corroboration requirement for jailhouse snitch testimony is
likely to fall short of its intended purpose.

CAUTIONARY JURY INSTRUCTIONS
If the court allows the state to present snitch testimony, it appropriately falls to a jury to decide whether
the testimony is credible. Nonetheless, the record of
wrongful convictions based on perjured testimony has
reinforced the need for greater guidance in making this determination.
ACCOMPLICE AND OUT-OF-CUSTODY INFORMANTS
Thus, states should adopt rules
requiring the court to provide a more
his review deals specifically with jailhouse snitches, but
specific framework to juries who
there are other types of informants that can compromise
wrestle with the numerous reliability
the criminal justice system. Accomplice testimony, and even
issues presented by snitch testimony.
out-of-custody informant testimony, can be problematic.
This is especially true because such
Although accomplice informants or out-of-custody informants
testimony is presented by the state;
generally have much to lose from a perjury conviction, they
so, absent a limiting instruction,
often have something to gain from testifying as well.24 While it
jurors are often inclined to assume
is illegal in the United States to give bribes or compensation in
the existence of some threshold of
exchange for testimony, out-of-custody informants can wreak
22
witness credibility.
havoc on an otherwise fair trial by testifying because of a
With little
grudge, or other personal motive, and desiring to see the
expense or burden on the courts, caudefendant behind bars. Additionally, even if an informant is not
tionary jury instructions tailored to
in state custody, there are circumstances in which witnesses can
the reliability issues specifically preget immunity from prosecution for suspected crimes or possible
sented by jailhouse snitches provide a
charges.25 Despite these potential problems with other types of
necessary added safeguard.
When the state presents the testestimony, jailhouse snitch testimony is still regarded as the
timony of a jailhouse snitch, the preleast reliable type of testimony in the criminal justice system.
siding judge should advise the jury

T

5
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

THE LEGAL LANDSCAPE

T

Less than two weeks later, however, on July 10,
1998, the court granted a rehearing en banc. On
January 8, 1999, the Tenth Circuit en banc decided
Singleton II, reversing its previous ruling by reading
the statute much more narrowly. The reversing
majority rationalized its reading by touting notions of
sovereignty — that “whoever” cannot be deemed to
include the sovereign government of the United
States, and that a “thing of value” cannot be construed to include benefits received from the state.28
Though prosecutors are persons, when they make
plea bargains with defendants, they act in their official
capacity as agents of the United States government.
The United States government is not a person, and
therefore not encompassed by the word “whoever.” 29
Following Singleton I, defense attorneys in all of
the federal circuits filed motions to suppress the testimony of jailhouse snitches who had received leniency
in exchange for testimony. When the Tenth Circuit
reversed itself, the other circuits quickly followed suit,
dismissing the motions.30
The holding in Singleton I, though reversed, shook
the bedrock of the informant witness system and, in so
doing, brought to light the complicity with which the
criminal justice system accepts, without screening, the
use of testimony that is inherently unreliable. The
Singleton I holding is a reminder that the justice system’s reliance on snitch testimony enjoys, at best, an
uneasy relationship to foundational principles of
American jurisprudence, and that reforms are necessary to avoid the pitfalls of bestowing benefits on witnesses in exchange for their testimony.

he issues presented by the use of informant witness testimony do not exist in a vacuum. Courts
in many jurisdictions have recognized that special
requirements are necessary to address the specific
reliability concerns inherent in this type of testimony. The following is a brief overview of a number of
ways in which states, and their courts, have enhanced
procedural safeguards for defendants on the receiving end of informant-dependent prosecutions.
SINGLETON I AND SINGLETON II
Perhaps the most noteworthy decisions to come
from any court regarding snitch testimony are the
Tenth Circuit cases known as Singleton I and Singleton
II, each of which dramatically changed the playing field
for prosecutors and the defense bar. The implications
of Singleton I were so far reaching as to cause some
amount of internal crisis in District Attorneys’ offices
across the country. In turn, the defense bar lamented
Singleton II, which was handed down shortly thereafter.
On July 1, 1998, a panel of the United States
Court of Appeals for the Tenth Circuit decided
United States v. Singleton, or Singleton I, ruling that the
common practice of federal prosecutors conferring a
benefit (be it money or a sentence reduction) on a
witness in exchange for his or her testimony constitutes bribery of the witness. In coming to this decision, the panel relied on Section 201 of the Title
XVIII of the U.S. Code, which reads in part:
Whoever … directly or indirectly, corruptly
gives, offers, or promises anything of value to
any person, with intent to influence the testimony under oath or affirmation, such firstmentioned person as a witness upon a trial,
hearing, or other proceeding, before any court
…shall be fined under this title or imprisoned
for not more than two years, or both.26

FEDERAL CIRCUITS
Singleton I is one of a long list of cases that have
raised concerns about the reliability of snitch testimony.
In addition to the Tenth Circuit ruling in the Singleton
decisions, a number of other federal courts of appeal
have addressed the issue of cooperating informants.
For example, in 1987, the Fifth Circuit Court of
Appeals ruled that the trial court should give a special
instruction cautioning the jury to question the credibility of witnesses who have been compensated for
their testimony.31
In 1993, the Ninth Circuit discussed the unreliability of informants in United States v. Bernal-

The panel read this statute broadly, finding that it
applied to prosecutors and government officials giving
sentence reductions to cooperating witnesses.27
According to the rules of statutory interpretation used
by the court in Singleton I, the word “whoever” referred
to federal prosecutors, and “anything of value” included intangibles, such as sentence reductions.
6

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

conviction, the conviction must be overturned as a
matter of law.37
In 2000, the Oklahoma Criminal Court of
Appeals ruled that courts must give a special instruction when jailhouse snitches testify, cautioning the
jury that it must examine the testimony with special
care. Courts ask jurors to take into account specific
factors. The facts relevant to these factors must be
disclosed by the prosecution prior to the trial.38
In 2004, the Colorado Court of Appeals reaffirmed its 1996 ruling that juries should be given a
cautionary instruction when there is no corroborating
evidence to support the testimony of an accomplice:
“An instruction that directs the jury to use caution
when considering accomplice testimony ‘is to be
given only when the prosecution’s case is based on
uncorroborated testimony of an accomplice.’” 39
Ohio courts have similarly held that evidence
corroborating an informant’s testimony obviates the
need for cautionary instructions.40
In 2001, the Wisconsin Appellate Court also
ruled that “[i]t is an error to deny a request for an
accomplice instruction only in a case where the
accomplice’s testimony is totally uncorroborated.” 41
In a 2005 decision, the Connecticut Supreme
Court overruled a case in which the court had not
allowed a credibility instruction, extending their special jury instruction law from including only accomplices to include jailhouse snitches. In the opinion the
court stated that “an informant who has been promised a benefit by the State in return for his or her testimony has a powerful incentive, fueled by self-interest, to falsely implicate the accused. Consequently,
the testimony of such an informant, like that of an
accomplice, is inevitably suspect.” 42

Obeso: “The use of informants to investigate and
prosecute persons engaged in clandestine criminal
activity is fraught with peril. This hazard is a matter
‘capable of accurate and ready determination by
resort to sources whose accuracy cannot reasonably
be questioned’ and thus of which we can take judicial notice.” 32
In 2002, the Fourth Circuit expressed its concern about snitch testimony, writing that compensated testimony “create(s) fertile fields from which
truth-bending or even perjury could grow, threatening the core of a trial’s legitimacy.” 33
In 2005, the Ninth Circuit Court of Appeals again
called for heightened judicial scrutiny of deals between
informants and prosecutors when it held: “where the
prosecution fails to disclose evidence such as the existence of a leniency deal or promise that would be valuable in impeaching a witness whose testimony is central to the prosecution’s case, it violates the due process
rights of the accused and undermines confidence in the
outcome of the trial.” 34 Later in 2005, the same court
called a lack of disclosure of deals between prosecutors
and informants “unscrupulous.” 35
STATE COURTS
Likewise, some state courts have independently
adopted general rules for different classes of informants, indicating a widely-held distrust of incentivebased testimony.36
For example, in 1999, the Montana Supreme
Court ruled that when an informant testifies for personal gain rather than an independent law enforcement purpose, the court must give a special cautionary instruction to the jury. If the trial court fails to
give the instruction, and the testimony is crucial to

BENEFITS & COSTS
would not have access — information that can be
extremely helpful for incarcerating the guilty.
Nonetheless, in addition to its inherent interest in
the economical administration of justice, the state must
maintain credibility with its citizenry as it prosecutes
crime. All wrongful convictions detract from the public’s faith in the fair administration of justice, but the

BENEFITS OF REFORM
The practice of inmates exchanging testimony
for more lenient sentences has its roots in British
common law.43 The main reason for its institution
then, and its continued use today, is simple: it results
in noticeably higher conviction rates.44 Inmates may
have information about suspects to which others
7

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

costs associated with a slightly higher workload for
cost is especially high when wrongful convictions result
judges, more extensive pretrial investigation necessitatfrom the testimony of witnesses who have received a
ed by corroboration requirements, and education or
benefit in exchange for their testimony. False snitching
retraining programs associated with implementation.
— and the misguided prosecution that it enables —
The expenses related to pretrial disclosures and jury
erodes the relationships between citizens and the state.45
instructions, on the other hand,
For each person wrongly convicted, a
are negligible. It is axiomatic that
guilty party remains free to commit
When perjured
such procedural costs, incurred
more crimes.
testimony leads to
in the interests of justice, are a
Higher scrutiny and transparenbare minimum expenditure for a
cy of jailhouse snitch testimony will
wrongful convictions,
criminal justice system in pursuit
allow law enforcement, courts, and
taxpayers shoulder
of more reliable outcomes in
the criminal justice system as a whole
the financial burden.
criminal cases. The values of fairto focus their limited resources on
ness and accuracy are of far
convicting the guilty. By ensuring
greater worth than the marginal procedural costs
available resources will be used to capture the actual
expended by the state.
perpetrator, the criminal justice system simultaneously
When perjured testimony leads to wrongful conhelps prevent wrongful convictions and further victimvictions, taxpayers shoulder the financial burden. From
ization of the community. With codified requirements
the state’s initial investigation and prosecution through
for determining the reliability of jailhouse snitch testiadditional investigation, multiple, subsequent appeals,
mony, the benefits to law enforcement, prosecutors,
and exoneration (where sizeable compensation is posand the community will accrue through stronger prossible), the public pays for perjured testimony.
ecutions and more reliable outcomes in criminal cases.
Compared to the costs of wrongful convictions, the
expense to the state associated with implementing
COSTS OF REFORM
these reforms is extremely low. Given the pay-offs, and
The policy improvements outlined in this review
given that the proposed improvements fit easily within
are generally accepted as high-yielding safeguards that
existing procedures, the reforms recommended here
do not overburden taxpayers or the courts. The main
constitute pragmatic proposals for improved policies.
expenses in terms of implementation are procedural

PROFILES OF INJUSTICE
The Marietta Seven

had a witness to the Matthews crime in custody on a
shoplifting charge. Deborah Ann Kidd, a habitual
drug abuser, prostitute, and shoplifter, claimed to
have pertinent information and asked for immunity in
exchange for her testimony.46 Desperate for clues,
then-Cobb County D.A. Ben Smith immediately sent
a letter to Kidd promising blanket immunity in
exchange for information about the crimes.
In discussions with authorities, Kidd implicated
herself and nine other individuals in the murders: James
Creamer, George Emmett, Hoyt Powell, Larry Hacker,
Bill Jenkins, Wayne Ruff, Charles Roberts, Mary Ann
Morphus, and Carolyn Sue Bowling Johnson. The
handprint and fingerprints found at the scene did not
match Kidd or any of the nine she implicated; however,

James Creamer and six co-defendants were wrongfully
convicted of murdering two pathologists in Marietta,
Georgia, almost entirely on the word of an informant
named Deborah Ann Kidd. Transcripts of inconsistencies
in Kidd’s statements were withheld from the defense.
In 1975, the convictions of the Marietta Seven were
reversed, and the state dropped all charges. Despite
the dropped charges, the District Attorney declined to
prosecute Kidd for perjury.

O

n July 26, 1972, more than a year after the well
publicized killings of two pathologists, Drs.
Warren and Rozina Matthews, South Carolina State
Police notified Cobb County authorities that they
8

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

manufactured than “recovered.” Astonishingly, while
the prosecution continued to work with Kidd in an
attempt to shape her testimony into credible evidence, Kidd stayed for several weeks at the home of a
detective with whom she developed a sexual relationship while continuing to abuse amphetamines supplied by the police.
During the appeals process, defense lawyers discovered numerous documents in police and prosecutors’ files, hidden from the defense at trial, that shattered Kidd’s credibility. The files revealed that during
the summer of 1972, Kidd gave three substantially
different accounts of the crime that contradicted the
physical evidence. For example, she said Rozina
Matthews had been severely beaten before being
shot, although an autopsy showed no cuts or bruises,
and no torn clothing. Additionally, Kidd initially stated that the crime occurred during cold weather,
sometime around Christmas or New Year’s Day,
when it had, in fact, occurred in May.
She originally described Creamer’s bleeding at
the scene as profuse, but later testified that it was light
— a more plausible claim, given that numerous samples of blood from the scene all matched the victims’
blood type and could not have come from Creamer.
No weapons were recovered from the scene, but ballistics tests indicated that three different .38 caliber
guns were fired. Police knew that Creamer had a gunshot wound and that a bullet was lodged in his body.
When it was surgically removed, the .38 slug was
found to have been fired by a gun other than the
Smith and Wesson owned and allegedly used by
Rozina Matthews, and it matched none of the slugs
found at the scene. During appeals, Creamer testified
that he was shot during an attempted robbery near
the Atlanta airport on the 19th or 20th of May, 1971.
In Kidd’s first three versions of events given to
police, she unequivocally claimed that Carolyn Sue
Bowling Johnson participated in the murders.
Investigators determined, however, that Johnson was
in Hamilton, Ohio on the day of the crime, a fact confirmed by medical records and the testimony of a doctor that had treated her on that day. Further investigation did indicate that Johnson had been involved in
a different crime — one in which Creamer was shot
— but that this crime had occurred weeks later,
around May 21, 1971. This was consistent with
Creamer’s explanation of his wound.

all nine individuals were indicted for murder based on
Kidd’s story, and seven were prosecuted.
THE SNITCH
Testifying under immunity, Kidd said she met
Creamer on May 2, 1971 and became his girlfriend
right away. They went to Georgia on May 4th with
Powell and Ruff and checked into an Atlanta motel
where she met the other men. After a party at the
motel with drugs and alcohol, she said the group
embarked on an armed robbery ending in the murder
of the Matthews couple.
The neighbor who first reported hearing gunshots had a clear view of the house and rear yard, but
when he looked out the window, he saw no people or
automobiles. According to Kidd, however, the murderous party involved no fewer than ten people who
traveled to the home in three cars.
Kidd said Ruff and Creamer killed Mr. Matthews.
Before it was over, she claimed Mrs. Matthews shot
Creamer, apparently with her own .38 pistol. Kidd tearfully claimed that she tried to flee the bloody scene, but
Roberts caught her and made her shoot Mrs. Matthews
in the head with her own gun. Kidd testified that she was
able to recall the crime with greater clarity as a result of
sessions with a psychologist who used hypnosis.
THE TRIALS
The Marietta Seven were convicted in five separate trials: Creamer and Emmett were tried separately in early 1973; Jenkins, Hacker, Powell, and Ruff
were tried jointly in July 1973; and Roberts was tried
in January 1975 after an earlier mistrial. All seven
were convicted of murder. Despite Kidd’s testimony,
they all consistently maintained their innocence.
After the original trials, it became clear that Kidd
had told several significantly different stories about the
crime — stories that were at odds with known facts.
Authorities had worked extensively with her, including
retaining a psychologist, Dr. Edwin P. Hall, who guided Kidd’s story over twelve visits totaling some thirtyfive hours (some with police and prosecutors present).
Dr. Hall conducted several “age regression” hypnosis
sessions that were supposed to help Kidd “recover”
memories and remove inconsistencies.
Defense attorneys were aware of the sessions, but
were denied access to tapes and transcripts until much
later. The records showed that Kidd’s story was more
9

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

a specialized law enforcement investigator, the
judge found, who was provided with detailed information about the crime by the police to help build
a case out of Kidd’s testimony.
The tapes and transcripts of their sessions
revealed that Hall told Kidd to read media accounts
of the case, including one taped comment in which
Hall tells Kidd that she “ought to read that newspaper and get those names straight.” 49 The judge found
the sessions to be “a thinly veiled effort to prop up the
prosecution’s case.” 50 Although the hypnotic sessions
were taped, the prosecution claimed that some tapes
and transcripts of the sessions were inadvertently
destroyed. The court concluded that the evidence had
been deliberately destroyed, constituting an unlawful
obstruction of justice.
Cobb County District Attorney Darden acknowledged during the hearings that the Matthews investigation had been “bungled,” and Judge Moye noted
the tunnel vision of investigators in his ruling, writing, “The number and significance of the investigative gaps in this case is truly astounding.” The court’s
conclusion was stinging:

Defense lawyers also discovered suppressed documents showing that police had a witness who
described seeing two teenagers driving a Mercedes
sports car like the Matthews’ near where their car was
found. The description matched none of the defendants. A neighbor also told investigators he saw a car
in front of the Matthews home near the time of the
crime, and gave a description of its two occupants that
matched none of the defendants. Other documents
showed that on Aug 1, 1971, two witnesses told police
that a different man, Willie Lloyd Gauldin, had confessed to them that he was the killer. Gauldin was
arrested and taken to the police psychologist who performed a “hypnotic interrogation” and concluded
that he was not involved.
Emmett’s and Creamer’s cases advanced first
through the state appeals process. During the unsuccessful state court appeals, defense lawyer Bobby Cook
dispatched an investigator to South Carolina to look
into Kidd’s past. The investigator found dated documents, including checks and divorce papers signed and
dated by Kidd, showing that she was actually in
Greenville, South Carolina on the very day she claimed
to be in Marietta with the defendants committing the
murders. After exhausting state appeals, their cases
went to the U.S. District Court. When presented with
the documentary evidence in a federal court hearing,
Kidd denied the signatures were hers, but three document experts testified that they were Kidd’s.

The prosecutorial suppression of nearly all
evidence concerning Deborah Kidd resulted
in a criminal proceeding that bordered on the
Kafkaesque ... the extreme measures to which
the state resorted in extracting information (or
more accurately, in supplying information to)
this witness and the use of her testimony at
trial … the suppression of documents, the firing of police officers skeptical of Kidd’s story,
all raise grave questions regarding the singleminded zeal with which these convictions
appear to have been sought and obtained. The
predictable result is that this Court has before
it a pair of criminal convictions obtained in a
manner so manifestly and fundamentally
unfair that they must be vacated.51

TOWARD JUSTICE
After seventeen days of hearings, United States
District Judge Charles Moye overturned Emmett
and Creamer’s convictions on June 17, 1975. The
court, finding numerous and pervasive instances of
suppression and destruction of exculpatory evidence, described the undisclosed report of Kidd’s
three varying accounts of the crime as “utterly devastating to Kidd’s credibility.” Judge Moye wrote,
“The prosecution, though it knew full well the
exculpatory and devastating nature of the documents it possessed, did not divulge their existence or
contents to either petitioner.” 47
In addition, the court found that “by the end of
August, Kidd’s scenario, riddled as it was with
inconsistencies, implausibilities and gaps, was in
dire need of shoring up if the prosecution were to
obtain convictions.” 48 Dr. Hall acted essentially as

TWO CONFESSIONS
During appeals, Billy Sunday Birt came forward to
confess to the Matthews killings and implicated two
others he said participated in the crime: Billy Wayne
Davis and Willie Hester. Birt had been convicted and
sentenced to death for the murder of an elderly couple
in Wrens, Georgia. Davis was in federal prison for
10

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

bank robbery, while Hester was never apprehended.
Birt’s wife initially contacted lawyers representing
the seven defendants and told of her husband’s
involvement in the Marietta murders. Birt himself
contacted Cobb County authorities to tell his story.
He was doing time in Illinois for bank robbery when
he was extradited to Georgia to face charges in the
Wrens killings. In a signed confession, Birt said he
killed Warren Matthews and Davis killed Rozina
Matthews. He was indicted for the Matthews murders in 1979. Birt was already on death row, however,
and the case never went to trial.
Kidd, too, finally confessed to her lies. On
Monday, August 25, 1975, she admitted on tape to
police and prosecutors that she lied in testimony that
convicted the seven men. Two days later, after intense
negotiations among prosecutors, defense lawyers, and
federal and state judges, three of the men (Roberts,
Powell, and Emmett) were released on personal
recognizance bonds and eventually saw all charges
dropped. The other four remained incarcerated for
charges unrelated to the Marietta murders.

On September 2, 1975, Cobb District Attorney
Buddy Darden announced he was dropping all
charges against the seven. He conceded that Kidd,
his star witness in the five trials, had admitted to
lying, but he refused to prosecute her for perjury.
Darden cited several reasons, including possible
involvement of others in manufacturing her testimony, legal complications associated with the initial
promise of blanket immunity, and “a waste of taxpayer money.” Critics charged that authorities wanted to
avoid the embarrassment that would follow shining a
spotlight on their gross mishandling of this unreliable witness.
By the time of his release, Emmett had served
thirty-five months, Powell two years, and Roberts
twenty-three months. Roberts pled guilty to drug
and gun charges upon release, with credit for time
spent on the charges for which he was exonerated.
While these innocent men served time in jail, Kidd
suffered no repercussions for committing perjury.
Proper safeguards monitoring snitch testimony may
have prevented this tragic injustice altogether.

Wilton Dedge’s Story

was denied bond. While his sentence from the first
trial was thirty years, at his retrial he was sentenced to
two consecutive life sentences plus two consecutive
fifteen year sentences. The conviction and increased
sentence were due in large part to Zacke’s testimony.

Arrested at age twenty, Wilton Dedge spent twentytwo years in prison for the rape of a seventeen-year-old
Florida woman before DNA testing finally proved his
innocence. The prosecution relied heavily on identification testimony from the victim and testimony from a
jailhouse snitch who testified that Dedge had confessed
to committing the crimes. After years of fighting for a
DNA test, Dedge won his freedom in August 2004.
The state of Florida awarded Dedge $2 million in
compensation for his wrongful imprisonment.

THE TRIALS
On December 8, 1981, around 4:30 p.m., a seventeen-year-old woman was repeatedly raped and
assaulted in her home in Canaveral Groves, Florida. A
month later, on January 8, 1982, Wilton Dedge was
arrested based on the victim’s identification, which had
wavered substantially in the month since the crime.
Dedge was first tried for burglary, sexual battery,
and aggravated battery in 1982. The prosecution
relied heavily on the victim’s identification, scent
identification from a police dog, and analysis of a hair
found at the crime scene. The Florida jury took four
hours and twenty-five minutes to convict Dedge of
burglary with assault, sexual battery with a weapon,
and aggravated battery. On December 22, 1983, however, the Fifth District Court of Appeals reversed
Dedge’s conviction, finding that while the scent identification was persuasive, the trial judge had erred in

O

n January 23, 1984, Clarence Zacke and Wilton
Dedge were placed in a prison transport van
together. They were the only two inmates in the van.
Dedge was awaiting a bond proceeding and a retrial
for the 1981 rape of a seventeen-year-old Florida
woman. Zacke, in prison for murder and conspiracy
to commit murder, was a jailhouse snitch. A little over
a week after their time together in the transport van,
Zacke testified at Dedge’s bond proceedings. He
claimed that Dedge had confessed the crime to him,
calling the victim “an old hog”, and saying that he
would kill her if he ever got out of prison.52 Dedge
11

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

unsuccessful in parlaying his snitch testimony into an
early release. On December 21, 2005, jurors convicted
Zacke of raping his adopted daughter over a multi-year
period in the 1970s after deliberating for only two-anda-half hours. He was sentenced to five consecutive life
sentences. The victim came forward to publicly accuse
him of rape upon learning of his impending release.57
In 2006, attorneys for Wilton Dedge called for an
investigation after discovering that Florida authorities
had prior knowledge of these allegations against Zacke,
and that they may have hidden the allegations in order
to secure Dedge’s conviction.58

disallowing the defense to present the testimony of an
expert on human scent discrimination and in allowing
hearsay during the examination of the prosecution’s
expert witness. Because the eyewitness testimony was
equivocal and the forensic evidence inconclusive, the
Court of Appeals found these errors to be harmful.
Dedge was convicted a second time in August of
1984 based on questionable eyewitness identification,
snitch testimony, limited forensic hair comparison, and
dog sniffing evidence from a since-discredited handler.
THE JAILHOUSE SNITCH
Prosecutors relied heavily on testimony of prison
inmate Zacke at Dedge’s second trial in 1984. Based
on Zacke’s testimony, the open-ended forensic hair
analysis, and the victim’s identification, Dedge was
convicted a second time of burglary with assault, sexual battery with a weapon, and aggravated battery.
This conviction was affirmed on appeal. Assistant
State Attorney Chris White, who prosecuted the
case, noted that Zacke wasn’t promised anything
specifically in exchange for his testimony. Still, Zacke
received a reduction in his sentence after testifying
against Dedge.53
Notably, the testimony against Dedge was not
the first time Zacke had come forward with information to help an investigation. He had previously testified against convicted serial killer Gerald Stano,
claiming that Stano had confessed to murdering
Cathy Lee Schraf. Following conviction for the
Schraf murder, Stano was sentenced to death. Zacke
later recanted this testimony during a phone interview with a freelance writer.54 Zacke had over a century shaved off of his original 180 year sentence. He
later admitted that he had been hoping to receive
parole by testifying against Dedge.55 On November
11, 1989, a hearing examiner requested a twenty-six
year reduction of Zacke’s sentence in return for
Zacke’s alleged cooperation in providing authorities
with information about a potential prison escape.
Assistant State Attorney Chris White and Assistant
State Attorney Michael Hunt both spoke at the hearing, calling Zacke a liar and a con artist (incidentally,
their statements at the hearing were never shared
with Dedge’s defense attorney).
The culmination of Zacke’s snitch testimonies
allowed him to negotiate his sentence to sixty years or
less with good behavior.56 But Zacke was ultimately

THE LONG ROAD TO EXONERATION
Throughout the course of his trials and appeals,
Dedge continually proclaimed his innocence. At the
time of his original and second trials, however, DNA
testing was not available. In fact, DNA testing was
not used in commercial laboratories until 1987.59
Florida courts first used DNA analysis in October of
1988,60 and it wasn’t until 1990 that federal courts
authorized its use.61
On March 30, 1988, Dedge’s attorney first wrote
the State Attorney seeking DNA testing. Though the
State Attorney had the authority to grant the request
for DNA testing, he advised Dedge’s attorney to file a
motion with the court. Dedge’s attorney subsequently
verified that the state attorney’s office was maintaining
the forensic evidence from the crime scene so that testing could be performed. During this same time, Dedge
himself was inquiring into different possibilities of
exoneration. He tried to show that Zacke had lied, the
eyewitness identification had been contradictory, and
that the prosecution had misused the hair analysis. He
also contacted DNA testing services, including advocacy groups, to seek help in getting tested.
On October 17, 1994, Dedge contacted attorneys
at the Innocence Project after seeing a television
report about their work in post-conviction DNA testing. Less than two months later, the Innocence
Project decided to take Dedge’s case. When the
Innocence Project contacted the State Attorney’s
office seeking release of certain evidence, the
Assistant State Attorney requested that they obtain a
court order. Though they could hardly know this at
the time, following this initial opposition by the state,
Dedge and his attorneys would face ten more years of
appeals before finally winning release.
12

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

On April 24, 1997, the Innocence Project filed the
first motion for DNA testing. The State Attorney’s
office opposed this motion, claiming that the statute of
limitations had passed despite the fact that the state
had received the first request for DNA from Dedge’s
attorney in 1988. The trial court agreed with the state,
and denied the motion for DNA testing. After multiple appeals, the court ordered the release of certain
evidence for DNA testing and, in March of 2001,
Dedge motioned to vacate his conviction based on
determinative proof of his innocence.
Later that spring, the legislature passed a new
statute that allowed for post-conviction DNA testing. In November of that year, Dedge returned to
court, filing yet another motion to vacate the judgment against him. The state argued that his conviction rested upon more than forensic evidence, relying on the snitch testimony and the dog scent lineup.
According to the prosecutors, any of this evidence
would have been sufficient to convict Dedge; thus,
the exculpatory DNA evidence should not be determinative in the case. After an initial hearing in which
Dedge’s motion to vacate his sentence was denied,
the Florida Fifth District Court of Appeal affirmed
without prejudice, allowing Dedge to file under the
newly passed post-conviction DNA statute. Dedge’s
attorneys filed a new motion under this statute, and
on April 27, 2004, a new trial was ordered. After
twenty-two years behind bars for a crime he did not
commit, and after years of arduous appeals and disappointments, Dedge at last won his freedom. He
was released on August 12, 2004.62
On June 23, 2003, Governor Jeb Bush signed legislation, inspired in part by Dedge’s case, which
extended prisoners’ rights to DNA testing that could
exonerate them by removing any deadline for seeking
evidence to prove innocence.63 The law also mandates
that evidence collected at the time of the crime must
be preserved until an inmate’s sentence is completed.64
On December 14, 2005, the state of Florida awarded Dedge a $2 million settlement for his twenty-two
year ordeal.65 Dedge was the first Florida inmate exonerated by DNA testing to receive compensation from
the state. In 2006, attorneys for Wilton Dedge called
for an investigation after discovering that the Florida
authorities had prior knowledge of the allegations
against Zacke, and may have hidden the information in
order to secure Dedge’s conviction.66

A SNITCH’S STORY

L

eslie Vernon White, a self-confessed career
criminal, has provided prosecutors with testimony in as many as forty cases. In an appearance on 60 Minutes, White described the
process by which inmate informers fabricate
evidence and claimed that he often lied when
giving testimony as a jailhouse snitch.67 In a
1988 interview with Time Magazine, White had
this to say about his prison stints: “Every time I
come in here, I inform and get back out.”68
After perjuring himself in a 1981 trial, and
falsely claiming that the Hillside Strangler had
confessed to him in 1982, White lost any
remaining shards of credibility. Nevertheless,
prosecutors continued to use his testimony, and
in November of 1988, the Los Angeles Times
reported that White had been called as a witness in three murder cases.69
In a 1990 interview with 60 Minutes, White
gave a first hand account of how he was able
to render perjured testimony believable. First,
White would determine the last name of a person recently charged with a murder in Los
Angeles County (available in the public
record). Using the prison chaplain’s phone,
White then called the Document Control
Center of the Los Angeles County Sheriff’s
Office to obtain a case number and arrest date.
White would then call the District Attorney’s
Record’s Bureau and pose as a Deputy District
Attorney to get the names of prosecutors
assigned to the case and names of key witnesses. White would then identify himself as a Los
Angeles police officer to the County Coroner’s
Office, where he learned how the victim was
killed. Finally, White would call families of the
victim and accused to learn characteristics
personal to each. Armed with this information,
White would fabricate a seemingly credible
“confession” on the part of the accused.70
Although White was crafty in his pursuit of
details, he claimed to Los Angeles Times
reporters that his methods were both known to
and employed by many looking for early
release from California’s prisons.71

13
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

SNAPSHOTS OF SUCCESS

A

The principal testimony against Fritz came from jailhouse snitches, with little corroboration. Five days
before Fritz’s codefendant, Ron Williamson, was scheduled to be executed, DNA testing was finally performed. The DNA results matched one of the informants who had testified against Fritz and Williamson, and
both men were exonerated. The Commission has proposed three bills designed to address the leading causes
of wrongful convictions, including a bill to curb false
testimony by jailhouse informants by requiring corroborating evidence for all such testimony.75 In its April 17,
2007 press release, the Commission argued: “Jailhouse
informants have strong reasons to lie because they are
offered lenience in return for information. The leading
cause of wrongful convictions in death penalty cases in
the United States is false testimony by informants.” 76
The Commission made a number of recommendations, most of them similar to the best practices outlined
in this review. These include the disclosure of any benefit a government informant receives or may receive,
required independent corroboration of snitch testimony,
and recording of all contact with in-custody informants.
The Commission’s recommendation that prosecutors seek independent corroboration of snitch information largely reflects the internal policies of District
Attorneys in a number of California jurisdictions. For
example, in response to the exploits of Leslie Vernon
White, a Los Angeles jail inmate who made national
news after detailing methods for fabricating testimony
to gain lenience, the Los Angeles County Grand Jury
convened a comprehensive investigation regarding the
use of jailhouse snitches.77 In response to the report
and recommendations that resulted from the investigation, the Los Angeles County District Attorney’s office
adopted policy guidelines to strictly control the use of
jailhouse snitches as witnesses.
The California legislature has addressed the need
for jury instructions. California currently requires an
instruction to juries to make an independent reliability determination when the state presents jailhouse
snitch testimony. In every California criminal proceeding in which the jury hears snitch testimony,
upon request of either party, the judge instructs the
jury, “The testimony of an in-custody informant
should be viewed with caution and close scrutiny. In
evaluating such testimony, you should consider the

number of states and jurisdictions have taken
measures to ensure that perjured snitch testimony
does not result in egregious miscarriages of justice such
as wrongful convictions. States like Illinois, California,
and Oklahoma represent case studies in snitch reform
— and in successful methods for enhancing the evidentiary value of jailhouse snitch testimony.
ILLINOIS
Illinois has recognized the need for proper disclosures of information relevant to incentive agreements
with jailhouse snitches, and that courts should perform
a “gatekeeper” function when criminal prosecutors
present jailhouse snitch testimony. The Illinois House
Special Committee on Prosecutorial Misconduct, after
holding extensive hearings, proposed that the Illinois
Supreme Court adopt jury instructions cautioning
about the reliability of such testimony.72 In April 2002,
the Illinois Governor’s Commission on Capital
Punishment, expanding on the prior work of the House
Special Committee, concluded that “[t]estimony from
in-custody witnesses has often been shown to have been
false, and several of the thirteen cases of men released
from death row involved, at least in part, testimony
from an in-custody informant.” The Commission recommended that the state require pretrial reliability
screenings of jailhouse snitch testimony.73
Illinois courts are now required by statute to hold
pretrial reliability hearings in capital cases that employ
jailhouse snitches. In reaching a decision, Illinois courts
consider information provided by prosecutors, including the criminal history of the informant, any benefit
conferred or to be conferred to the informant in
exchange for his or her testimony, other cases in which
the informant has testified, and other information relevant to the informant’s credibility.74 These practices
match this report’s best practices for disclosure.
CALIFORNIA
California established the California Commission
on the Fair Administration of Justice to examine
California’s administration of criminal justice and to
recommend safeguards to ensure its fairness. On
September 20, 2006, the Commission conducted a public hearing, which included the testimony of Dennis
Fritz, a man wrongly convicted of rape and murder.
14

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

Supreme Court’s 1963 holding in Brady v. Maryland,
which requires prosecutors to turn over to defense
counsel any “material” evidence that might impeach
government witness testimony. The Dodd rule
expands what the court considers “material” to
include any information that might lead a fact finder
to deem snitch testimony unreliable. The second rule
adopted in Dodd requires trial courts to issue a special
cautionary instruction to juries who hear snitch testimony. The instruction requires juries to take into
account several factors similar to those set forth in the
Illinois statute. The Court wrote, “Courts should be
exceedingly leery of jailhouse informants, especially if
there is a hint that the informant received some sort
of a benefit for his or her testimony.” 79

extent to which it may have been influenced by the
receipt of, or expectation of, any benefits from the
party calling that witness. This does not mean that
you may arbitrarily disregard such testimony, but you
should give it the weight to which you find it to be
entitled in the light of all the evidence in the case.” 78
OKLAHOMA
In Dodd v. State (2000), the Oklahoma Court of
Criminal Appeals adopted two rules that limit
improper reliance on snitch testimony. First, the
Oklahoma Court now requires that prosecutors share
any information with defense counsel that might discredit the reliability of snitch testimony. In practice,
this rule operates as a supplement to the U.S.

VOICES OF SUPPORT
“The use of informers, accessories, accomplices, false
friends, or any of the other betrayals which are ‘dirty
business’ may raise serious questions of credibility.” 82

“The need for disclosure is particularly acute where
the government presents witnesses who have been
granted immunity from prosecution in exchange for
their testimony … We said that informants granted
immunity are by definition ... cut from untrustworthy
cloth, and must be managed and carefully watched
by the government and the courts to prevent
them from falsely accusing the innocent, from manufacturing evidence against those under suspicion of
crime, and from lying under oath in the courtroom. ...
Accordingly, we expect prosecutors and investigators
to take all reasonable measures to safeguard the system against treachery.” 80

United States Supreme Court
On Lee v. U.S., June 2, 1952
“A legally unsophisticated jury has little knowledge as
to the types of pressures and inducements that jail
inmates are under to ‘cooperate’ with the state and to
say anything that is ‘helpful’ to the state’s case. It is up
to the trial judge to see that there are sufficient assurances of reliability prior to admitting this kind of
amorphous testimony to keep this kind of unreliable
evidence out of the hands of the jury ...” 83

United States Court of Appeals
for the Ninth Circuit
Carriger v. Stewart, December 17, 1997

Supreme Court of Nevada
D’Agostino v. State, December 30, 1991

“If I worked with a cooperator and came to trust him
and I corroborated six of the eight major facts he told
me, I would tend to believe the other two uncorroborated ones and use those at trial. I would not always
try to corroborate those additional two facts. I’ve gotten burned by such an approach.” 81

“Jailhouse informant testimony has come under increasing criticism and has contributed to a great number of
wrongful convictions in [Illinois] and the country … In
Illinois, of the 13 wrongful convictions from death row,
five were convicted based on jailhouse informant testimony. Clearly, there is need for a legislative response.” 84

Anonymous
Assistant United States District Attorney
Fordham Law Review, December, 1999

James B. Durkin
Illinois State Representative
Chicago Daily Law Bulletin, April 26, 2003
15

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

“When used properly, informants can be a powerful
and appropriate investigative tool. But they can also
be destructive, crime-producing, and corrupting.
The widespread use of informants means that much
of the real adjudicative process takes place underground, without rules, records, or lawyers, and without public or judicial scrutiny of the fairness and
accuracy of the process.” 87

“The jailhouse informant is often a seasoned witness
who can appear convincing even during tough crossexamination. And it’s been shown that juries tend to
give weight to the evidence of a defendant’s confession, even after warnings as to the credibility of jailhouse informants in general. I believe the only effective way to deal with this problem is to provide a pretrial exclusion process to ensure the reliability of an
informant’s testimony.” 85

Alexandra Natapoff
Associate Professor
Loyola School of Law, Los Angeles
San Francisco Chronicle, November 19, 2006

Robert M. Bloom
Professor, Boston College Law School
ABA Criminal Justice Magazine, Spring 2003
“I’m not the first guy who went to prison because
someone lied, and I won’t be the last. But it’s wrong,
and something should be done to try to prevent this
because no one can give me back all the years I lost.” 86

“How can we prevent informants from testifying
falsely? We can’t. But we can reduce the number of
wrongful convictions based on false testimony with
steps designed to level the playing field and open the
process to daylight.” 88

Timothy Atkins
Exoneree
Sacramento Bee, May 18, 2007

George C. Harris
Director of the Center for Advocacy and
Dispute Resolution, University of the Pacific
McGeorge School of Law
San Jose Mercury News, November 14, 2006

QUESTIONS & ANSWERS
determine, but scholars generally agree that the
number is very high.

Are unreliable jailhouse snitches so pervasive
in criminal cases, and in the record of
wrongful convictions, as to warrant
substantive policy change?
Yes. The “informant institution” is an everexpanding one.89 The incentives to cooperate are
often irresistible given the benefits offered in
exchange for testimony. Unfortunately, the number
of wrongful convictions incurred as a result of perjured informant witness testimony is correspondingly high. Of 111 wrongful convictions in capital
cases recently examined by the Center on Wrongful
Convictions at Northwestern University School of
Law, fifty-one involved “incentivised” witness
testimony. Perjured snitch testimony was determined to be the most common cause of wrongful
convictions in capital cases.90 The total number of
cases in which perjured informant witness testimony has led to wrongful convictions is impossible to

Our system of criminal justice already has a
safeguard in using cross-examination to
discredit unreliable witness testimony.
Why isn’t this safeguard a sufficient tool to
weed out false jailhouse snitch testimony?
Cross-examination is an insufficient safeguard
against the perils of unreliable snitch testimony
because of the special problems that arise from limited disclosure requirements related to informant witnesses. Oftentimes, defense counsel will not have
access to all of the information to discredit the testimony of an unreliable state informant, because discovery requirements do not, as a general rule, extend to
evidence that is not “material” to guilt or innocence.
Materiality as defined by courts is a very high threshold, describing evidence that, if disclosed, would have
16

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

cial to a defendant as to warrant both corroboration
and a pretrial determination by the court that the testimony can be presented to the jury at all. Once such
a determination has been made, it is indeed the role
of the jury to make a determination as to whether to
believe the informant witness’ testimony based on the
guidelines of a limiting instruction.

resulted in a “reasonable probability” that the defendant would be found not guilty.91 The materiality
standard certainly does not apply to all evidence that
could be used to show the implausibility of a jailhouse
snitch’s testimony. Without pretrial disclosures of all
information relevant to credibility determinations,
meaningful cross-examination is impossible.
Furthermore, it is difficult to “un-toll the bell.”
Though effective cross-examination might convince a
jury to give less weight to informant testimony, jurors
are somewhat predisposed to infer some degree of reliability because the witness is presented by the state.
Therefore, a pretrial, independent determination by
the court that the witness is credible is necessary to
prevent improper reliance by juries on informant testimony. Similarly, use of the common tool of implied
inducements allows for prosecution witnesses to state
to a jury, unequivocally, that they have not received
any benefit in exchange for their testimony. The fact
that such informers will or may receive such benefits,
even if not explicitly promised, is often overwhelmed
by the informer’s second-hand account of criminal
activity. Proper instruction to the jury is necessary to
balance this precarious practice.
Our lengthy and unfortunate history of wrongful
convictions has shown that the procedural safeguards
currently in place do not effectively remedy the problems presented by the unreliable testimony of jailhouse snitches. In order to offer adequate protection
to innocent individuals, and to ensure reliable outcomes in criminal cases, states must implement meaningful procedural safeguards that supplement the
tools currently available to defendants.

Aren’t the policy recommendations implicitly
displaying a general mistrust of prosecutors
and law enforcement officers who are simply
trying to keep criminals of the streets?
Most police and prosecutors subscribe to high
standards of corroboration and witness scrutiny before
utilizing snitch testimony. When a state informant witness is not credible, the credibility of those who employ
that testimony is also undermined, as is their ability to
successfully prosecute and enhance public safety.
Often, the problems discussed here arise as a
result of unscrupulous informers deceiving law
enforcement, whose resources are often over-extended. Informants are often so desperate to escape incarceration that they will go to great lengths to weave
elaborate narratives in exchange for sentence reductions. Though prosecutors have an ethical duty to
ascertain the truthfulness of information from cooperating witnesses, it shouldn’t fall entirely to prosecutors,
or to police, to weed out the bad apples.
Most of the policy improvements discussed here are
not designed to place any additional burden on state
attorneys or law enforcement officers who already
employ basic corroboration techniques. Furthermore,
expanding the role of the courts in determining informant witness reliability will ensure that public confidence
in our law enforcement officers remains intact.

Shouldn’t it be left to a jury to decide
whatever testimony is credible or not?
As is the case when a party to a civil action wishes to
present expert testimony, there are times when it is necessary for a court to make an independent legal determination as to the admissibility of witness testimony. Such
legal determinations fall squarely within the jurisdiction
of the judge in criminal trials. Once the judge has ruled
on the legal implications of allowing a jury to hear
snitch testimony, the jury, as fact-finder, should give the
testimony whatever weight they feel is appropriate.
Evidence that is unduly prejudicial is always
excluded from the total body of evidence presented to
a jury. Perjured snitch testimony is so highly prejudi-

Are reforms related to snitch testimony
difficult to implement?
The policy recommendations are designed to fit
readily within the context of processes already in place,
including discovery, jury instructions, and consideration
of adversarial motions (with argument and presentation
of evidence). Courts are already employing these procedures in their daily practice and in the context of criminal trials. What’s more, courts are already conducting
hearings to determine the reliability of expert witnesses.
The reforms would be an extension of this rule to a class
of witnesses that demands equal, if not higher, scrutiny.
17

W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

A MODEL POLICY
MODEL BILL FOR INCREASING THE EVIDENTIARY VALUE
OF JAILHOUSE INFORMANT TESTIMONY
An Act:
Section I. Purpose.
The purpose of this Act is to ensure that only reliable jailhouse informants are permitted to testify at trial, and to ensure that when such an informant testifies, the jury
is fully informed. Because in-custody informants have very strong incentives to fabricate or elaborate testimony in order to receive lenient treatment, courts should view
such testimony with skepticism. This act should be interpreted consistent with the
goal of keeping unreliable informant testimony out of court.
Section II. Definitions.
A. As used in this section, “in-custody informant” means a person, other
than a co-defendant, percipient witness, accomplice, or co-conspirator
whose testimony is based upon statements made by the defendant while
both the defendant and the informant are held within a correctional
institution.
B. As used in this section, “consideration” means any plea bargain, bail consideration, reduction or modification of sentence, or any other leniency,
benefit, immunity, financial assistance, reward, or amelioration of current or future conditions of incarceration in return for, or in connection
with, the informant’s testimony in the criminal proceeding in which the
prosecutor intends to call him or her as a witness.
Section III. Disclosure Obligations of the Prosecution.
In any criminal trial or proceeding in which the prosecution intends to call an in-custody informant to testify, pursuant to relevant state rules governing discovery, the
prosecution must obtain and disclose the following information to the defense:
A. A written statement setting out any and all consideration promised to,
received by, or to be received by the in-custody informant. This requirement applies even if the prosecution is not the source of the consideration.
B. The complete criminal history of the in-custody informant.
C. The names and addresses of any and all persons with information concerning the defendant’s alleged statements, including but not limited to:
law enforcement and/or prison officers to whom the informant related
the alleged statements; other persons named or included in the alleged
statement; and other persons who were witness and who can be reasonably expected to have been witness to the alleged statements.
D. Any prior cases in which the in-custody informant testified and any consideration promised to or received by the in-custody informant, provided such information may be obtained by reasonable inquiry.
18
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

E. Any and all statements by the in-custody informant concerning the
offense charged.
F. Any other information that tends to undermine the in-custody informant’s credibility.
G. This section does not alter other disclosure or discovery obligations
imposed by state or federal law.
H. Any materials that the prosecution must disclose under this section are
admissible to impeach the credibility of the in-custody informant if such
informant testifies at trial.
Section IV. Requirement for a Pre-Trial Admissibility Hearing.
A. Prior to trial, the prosecution must apply to the trial court and request
that the trial court admit the testimony of the in-custody informant. In
such hearing, the court must only admit the testimony of the in-custody informant if it concludes that the informant is reliable, considering
such factors as the consideration offered to the in-custody informant,
the complete criminal record of the in-custody informant, the alleged
statements made by the accused, the time, place, and circumstances of
the alleged statements, the time, place, and circumstances of the alleged
disclosure to law enforcement officials, any inconsistent statements by
the in-custody informant, other cases in which the in-custody informant testified, and any consideration promised or received in those cases,
the quality of corroborating evidence, and any other evidence relevant
to the in-custody informant’s credibility. The prosecution shall bear the
burden of proof.
B. The judge should only admit the in-custody informant’s testimony if
corroborated by other such evidence as independently tends to connect the defendant with the commission of the offense charged or the
special circumstance(s) or aggravating factor(s) to which the in-custody
informant testifies. Such corroboration is not sufficient if it merely
shows the commission of the offense or the special circumstance(s) or
aggravating factor(s).
Section V. Jury Instructions.
Prior to sending the charges to the jury, the court should instruct the jury that in-custody informant testimony can be especially unreliable and must be given special
scrutiny. The court should also instruct the jury that they may consider all of the factors listed in Section IV in evaluating the credibility of the in-custody informant. The
jury shall not be instructed that the court has already found that the in-custody
informant is reliable.

19
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

STATISTICS

A

PERCENTAGE OF EXONERATIONS

NUMBER OF EXONERATIONS

s of May 11, 2007, over 120 people
Leading States with Wrongful Conviction
have been exonerated from death
by Snitch Testimony
row since capital punishment was rein10
stated in 1973. A 2005 study by the
Center for Wrongful Convictions at
8
9
the Northwestern School of Law exam6
ined 111 of those exonerations and
4
found that fifty-one of those 111 peo5
4
4
ple were wrongfully sentenced to death
2
based at least in part on the testimony
0
of “witnesses with incentives to lie.” 92
Illinois
Oklahoma
Florida
New Mexico
In fact, testimony from snitches and
Source: Center on Wrongful Convictions: The Snitch System (2002)
other informants is the leading cause of
wrongful convictions in capital cases.93
In a related study published in 2002, the Center
Main Causes of Wrongful Convictions
for Wrongful Convictions examined ninety-seven
in U.S. Capital Cases from 1973–2004
cases in which evidence presented subsequent to sentencing conclusively exonerated the defendants. In
OTHER
SNITCH
15%
TESTIMONY
thirty-eight of those ninety-seven cases, informant
15%
46%
witness testimony was shown to be a primary factor in
FALSE
CONFESSIONS
the jury’s decision to convict.94 And in sixteen of
14%
14%
those ninety-seven cases, jailhouse snitches simply
46%
fabricated confessions that were never actually made
by the defendant. In each instance, the testifying government witness received some benefit in exchange
25%
for the testimony.95
ERRONEOUS
IDENTIFICATION
As of November 1999, two months prior to the
25%
Illinois moratorium on the death penalty, four of
twelve Illinois cases that resulted in wrongful death
sentences for individuals who were later exoneratSource: Center on Wrongful Convictions: The Snitch System (2005)
ed, relied on jailhouse snitch testimony. In another
two of those twelve cases,
Illinois prosecutors had jailFactors Contributing to Exoneration
house snitch testimony at the
(Of 38 Wrongful Convictions Involving Snitch Testimony)
ready, but opted not to present
100
it to the jury.96
80
Finally, according to the
California Commission on the
60
45%
Fair Administration of Justice, in
37%
40
the state of California, twenty
26%
percent of all wrongful convic13%
13%
20
tions are the result of perjured
5%
97
snitch testimony.
0
Recantation
of Snitch

Media
Investigation

DNA
Testing

New
Witnesses

Source: Center on Wrongful Convictions: The Snitch System (2002)

20
W W W . T H E J U S T I C E P R O J E C T. O R G

New
Evidence

Investigation
by Activists

THE JUSTICE PROJECT

LITERATURE
S. Clark, Procedural Reforms in Capital Cases Applied to
Perjury, 34 J. MARSHALL L. REV. 453, 453-67
(2001).

SUGGESTED READINGS
The following materials are essential reading for
individuals interested in enhancing the evidentiary
value of jailhouse snitch testimony.

George C. Harris, Testimony for Sale: The Law and
Ethics of Snitches and Experts, 28 PEPP. L. REV. 1,
1-74 (2000).

Robert M. Bloom, Jailhouse Informants, Criminal
Justice Magazine 18, no. 1 (2003), available at
http://www.abanet.org/crimjust/spring2003/
jailhouse.html.

Alexandra Natapoff, Beyond Unreliable: How Snitches
Contribute to Wrongful Convictions, 37 GOLDEN
GATE U. L. REV. 107, 107-29 (2006).

Mark Curriden, No Honor Among Thieves, ABA
Journal 52, no. 75 (1989).

Alexandra Natapoff, Snitching: The Institutional and
Communal Consequences, 73 U. CIN. L. REV.645,
645-702 (2004).

The Snitch System: How Snitch Testimony Sent Randy
Steidl and Other Innocent Americans to Death Row,
A Center on Wrongful Convictions Survey,
Northwestern University School of Law (Winter
2004-2005), available at http://www.law.north
western.edu/depts/clinic/wrongful/documents/
SnitchSystemBooklet.pdf.

Sam Roberts, Note, Should Prosecutors be Required to
Record their Pretrial Interviews with Accomplices and
Snitches?, 74 FORDHAM L. REV. 257, 257-302
(2005).
Symposium, The Cooperating Witness Conundrum:
Is Justice Obtainable?, 23 CARDOZO L. REV. 747
(2002).

SELECTED BIBLIOGRAPHY
The following listing includes some of the key
source material used in developing the content of this
policy review. While by no means an exhaustive list of
the sources consulted, it is intended as a convenience
for those wishing to engage in further study of the
topic of jailhouse snitch testimony. Many of the
entries contain hyperlinks for ease in locating an article, report, or document on the web.

Stephen S. Trott, S.S. Words of Warning for
Prosecutors Using Criminals as Witnesses, 47
HASTINGS L. J. 1381, 1381-1432 (1996).
Ellen Yaroshefsky, Cooperation with Federal
Prosecutors: Experiences of Truth Telling and
Embellishment, 68 FORDHAM L. REV. 917, 917-64
(1999).

1. Journals and Law Reviews
C. Elliott Blaine, Life’s Uncertainties: How to Deal
with Cooperating Witnesses and Jailhouse Snitches,
16 CAP. DEF. J. 1, 1-17 (2003).

2. Reports, Policies and Motions

Jack Call, Judicial Control of Jailhouse Snitches,
22 JUST. SYS. J. 73, 73-83 (2001).

American Bar Association. Resolution 108B. Adopted
by the House of Delegates February 14, 2005,
available at http://www.abanet.org/leadership/
2005/midyear/daily/108B.doc.

R. Michael Cassidy, “Soft Words of Hope”: Giglio,
Accomplice Witnesses, and the Problem of Implied
Inducements, 98 NW. U. L. REV. 1129, 1129-77
(2004).

California Commission on the Fair Administration
of Justice. Official Recommendations on the Use of
Jailhouse Informants (November 20, 2006), available at http://www.ccfaj.org/rr-use-official.html.

21
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

Center on Wrongful Convictions, Northwestern
University School of Law. The Snitch System:
How Snitch Testimony Sent Randy Steidl and Other
Innocent Americans to Death Row. Chicago:
Northwestern University Press, 2005, available at
http://www.law.northwestern.edu/wrongful
convictions/documents/SnitchSystemBooklet.pdf.

FPT Heads of Prosecutions Committee. Report of the
Working Group on the Prevention of Miscarriages
of Justice (September 2004), available at
http://www.justice.gc.ca/en/dept/pub/hop/toc.html.
Governor’s Commission on Capital Punishment.
Report (April 15, 2002), available at http://www.idoc.
state.il.us/ccp/ccp/reports/commission_report/
index.html.

3. Selected Media

Motion to Exclude Accomplice/Snitch Testimony,
Sample Motion 59. Louisiana Indigent Defense
Board (last visited September 13, 2007), available
at http://lidab.com/motions/59accomp.htm.

Bloom, Robert M. “Jailhouse Informants.” Criminal
Justice Magazine 18, no. 1 (2003), available at
http://www.abanet.org/crimjust/spring2003/
jailhouse.html.

Motion to Preclude Creation of Snitch Testimony,
Sample Motion 58. Louisiana Indigent Defense
Board (last visited September 13, 2007), available
at http://lidab.com/motions/58creat.htm.

Curriden, Mark. “No Honor Among Thieves.”
ABA Journal 52, no. 75 (1989).
Mills, Steve and Ken Armstrong. “The Failure of
the Death Penalty in Illinois – Part 3: The jailhouse Informant.” Chicago Tribune November 16,
1999, available at http://www.chicagotribune.com/
news/nationworld/chi-dpdpillinois-special,
1,2049367.special.

Warden, Rob. “The Snitch System: How
Incentivised Witnesses Put 38 Americans on
Death Row.” Research report, Arizona State
University College of Law, Tempe, AZ (April 25
2002), available at http://www.law.northwestern.
edu/depts/clinic/wrongful/documents/Snitch.htm.

ENDNOTES
1
Center on Wrongful Convictions, Northwestern University School of Law,
The Snitch System: How Snitch Testimony Sent Randy Steidl and other Innocent
Americans to Death Row, Center on Wrongful Convictions, 2005, available at
http://www.law.northwestern.edu/wrongfulconvictions/documents/Snitch
SystemBooklet.pdf.
2
The Innocence Project, “Informants/Snitches,” available at http://innocence
project.org/understand/Snitches-Informants.php (accessed Jun. 28, 2007).
3
George C. Harris, Testimony for Sale: The Law and Ethics of Snitches and
Experts, 28 PEPP. L. REV. 1 (2000).
4
Brady v. Maryland, 373 U.S. 83 (1963).
5
Jack Call, Judicial Control of Jailhouse Snitches, 22 JUST. SYS. J. 73 (2001).
6
These factors were largely derived from a statute enacted in Illinois that
requires pretrial disclosures and reliability hearings for jailhouse informants
in capital cases; see 725 ILL. COMP. STAT. 5/115-21(c) (2003).
7
Harris, supra note 3, at 1.
8
See United States v. Singleton, 144 F.3d 1343 (10th Cir. 1998), overruled by
United States v. Singleton, 165 F.3d 1297 (10th Cir. 1999).
9
Harris, supra note 3, at 3.
10
Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993); Kumho Tire v.
Carmichael, 526 U.S. 137 (1999).
11
Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993).
12
Many law review articles endorse the courts’ gatekeeper role in determining reliability of jailhouse testimony. See, e.g., Alexandra Natapoff, Beyond
Unreliable: How Snitches Contribute to Wrongful Convictions, 37 GOLDEN GATE
U. L. REV. 107, 113 (2006); Harris, supra note 3, at 63-64.

725 ILL. COMP. STAT. 5/115-21(c) (2003).
See TEX. CODE CRIM. PROC. ANN. art. 38.14 (Vernon 2005); N.Y. CRIM.
PROC. LAW § 60.22 (McKinney 2003); CAL. PENAL CODE § 1111 (West 2004).
15
Report of the Illinois Governor’s Commission on Capital Punishment, 8
(April 15, 2002), available at http://www.idoc.state.il.us/ccp/ccp/reports/
commission_report/index.html.
16
See 725 ILL. COMP. STAT. 5/115-21(2003); John Cullerton, et al, “Death
Penalty Reforms and ‘Fundamental Justice,’” Chicago Daily Law Bulletin
(June 30, 2004).
17
California Commission on the Fair Administration of Justice, Report and
Recommendations Regarding Informant Testimony (Nov. 20, 2006).
18
A.B.A. RESOL.108B, House of Delegates (Feb.14, 2005), available at
http://www.abanet.org/leadership/2005/midyear/daily/108B.doc.
19
Id.
20
California Commission on the Fair Administration of Justice, Report and
Recommendations Regarding Informant Testimony (Nov. 20, 2006).
21
Christine J. Saverda, Accomplices in Federal Court: A Case for Increased
Evidentiary Standards, 100 YALE L.J 785, 791 (1990).
22
Harris, supra note 3, at 56.
23
These factors were largely derived from a statute enacted in Illinois that
requires pretrial disclosures and reliability hearings for jailhouse informants
in capital cases; see 725 ILL. COMP. STAT. 5/115-21(c) (2003).
24
See Saverda, supra note 20.
25
See Amanda Schreiber, Dealing with the Devil: An Examination of the FBI’s
Troubled Relationship With its Confidential Informants, 34 COLUM. J.L. & SOC.
13
14

22
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT

PROBS. 301 (Summer 2001).
26
18 U.S.C. § 201 (c) (2) (2007).
27
United States v. Singleton, 144 F.3d 1343.
28
United States v. Singleton, 165 F.3d 1297.
29
Id. at 1299-1301.
30
C. Elliott Blaine, Life’s Uncertainties: How to Deal with Cooperating Witnesses
and Jailhouse Snitches, 16 CAP. DEF. J. 1, 5-6 (2003).
31
United States v. Cervantes-Pacheo, 826 F.2d 310, 316 (5th Cir. 1987) (citing
United States v. Beard, 761 F.2d 1477, 1481 (11th Cir. 1985)).
32
United States. v. Bernal-Obeso, 989 F.2d 331, 333 (9th Cir. 1993) (citing FED
R. EVID. 201(b) (2)); cf. Hudson v. Palmer, 468 U.S. 517, 526-27 (1984).
33
Natapoff, supra note 12, at 120 (citing United States v. Levenite, 277 F.3d
454, 462 (4th Cir. 2002)).
34
Natapoff, supra note 12, at 121 (citing Horton v. Mayle, 408 F.3d 570, 581
(9th Cir. 2005)).
35
Natapoff, supra note 12, at 121 (citing Silva v. Brown, 416 F.3d 980, 991
(9th Cir. 2005)).
36
Call, supra note 5, at 73.
37
Call, supra note 5, at 78-79 (citing State v. Grimes, 982 P.2d 1037 (Mont. 1999)).
38
Dodd v. State, 993 P.2d 778 (Okla. Crim. App. 2000). See also Call, supra
note 5, at 79.
39
People v. Petschow, 119 P.3d 495, 504 (Colo. Ct. App. 2004) (citing People v.
Montoya, 942 P.2d 1287, 1293 (Colo. Ct. App. 1996)).
40
See State v. James, No. 96-CA-17, 1998 WL 518135 (Ohio Ct. App.
Mar. 25, 1998).
41
State v. Spiller, No. 00-2897-CR, 2001 WL 1035213 (Wis. App. Sept. 11,
2001) (citing Linse v. State, 286 N.W. 2d 554, 558 (Wis. 1980)).
42
State v. Patterson, 886 A.2d 777 (Conn. 2005) (rev’g State v.Santiago, 708
A.2d 969 (Conn. 2005), and modifying, State v. Colon, 864 A.2d 666 (Conn.
2004), State v. Colton, 384 A.2d 343 (Conn. 1977)).
43
R. Michael Cassidy, “Soft Words of Hope”: Giglio, Accomplice Witnesses, and the
Problem of Implied Inducements, 98 NW. U. L. REV. 1129 (2004).
44
Daniel C. Richman, Cooperating Defendants: The Costs and Benefits of
Purchasing Information from Scoundrels, 8 FED. SENT. REV. 292 (1996).
45
Alexandra Natapoff, Snitching: The Institutional and Communal Consequences,
73 U. CIN. L. REV. 645 (2004).
46
Emmett v. State, 205 S.E. 2d 231, 238 (Ga. 1974). See also Emmett v. Ricketts,
397 F. Supp. 1025, 1030-34 (N.D. Ga. 1975).
47
Ricketts, 397 F.Supp. at 1042.
48
Id. at 1042.
49
Id. at 1025.
50
Id. at 1042.
51
Id. at 1047.
52
Jan Pudlow, “Dedge prosecutor details decision-making process: ‘I am truly
sorry that he got convicted wrongly and spent 22 years in prison,’” 3 Florida
Bar News 14, Vol. 33, (2006).
53
Id.
54
Lelis Ludmilla, “Jailhouse snitch calls new charges payback,” Orlando
Sentinel, December 19, 2005.
55
“Questions about a Snitch,” St. Petersburg Times, December 30, 2005.
56
Mariano, Willoughby, et al., “Snitch receives 5 life terms in rape of girl,”
Orlando Sentinel, December 22, 2005.
57
See Ludmilla, “Jailhouse snitch calls new charges payback,” and Linda
Jump, “Zacke guilty of rape,” Florida Today, December. 22, 2005.
58
John Torres, “Prior Zacke knowledge may have been hidden,” Florida Today,
January 23, 2006.
59
Dedge v. State, 723 So.2d 322, 323 (Fla. Dist. Ct. App. 1998).
60
Id. at 323 (citing Andrews v. State, 533 So.2d 841 (Fla. Dist. Ct. App. 1988)).
61
Id. at 323 (citing United States v. Jakobetz, 747 F.Supp. 250 (D.Vt. 1990),
aff’d, 955 F.2d 786 (2d Cir. 1992)).
62
Bill Berlow, “Dedge paid dearly for authorities lapses,” Tallahassee Democrat,
Dec. 9, 2005.
63
FLA. STAT. ANN. § 925.11 (West 2006).

64
Paul Flemming, “Governor signs DNA bill prompted by Dedge case,”
Florida Today (Brevard County, FL), Jun. 24, 2006.
65
“Snitch Receives 5 life terms in rape of girl,” Orlando Sentinel, Dec. 22, 2005.
66
Torres, “Prior Zacke knowledge may have been hidden.”
67
See, e.g., People v. Griffin, 235 Cal. App.3d 1740, 1742-43 (Cal. Ct. App.
1991) (discussing Leslie Vernon White and his reputation as an informer).
68
“A Snitch’s Story,” Time, December 12, 1988.
69
“A Look at Jailhouse Informants,” Los Angeles Times, November 11, 1988.
70
Interview by Harry Reasoner of Leslie Vernon White (June 17, 1990).
71
Ted Rohrlich, “D.A. Officials to Study Means of Plugging Data Leaks,” Los
Angeles Times, metro sec., October 30, 1998.
72
Illinois Coalition to Abolish the Death Penalty, “From a Broken System at
Work: Report on the State of the Death Penalty in Illinois in the Year of the
Moratorium” (Feb. 5, 2001). See also ILL. SUP. CT. R. 451 (There is no provision for cautionary instructions when dealing with informant testimony).
73
Illinois Governor’s Commission on Capital Punishment, Chapter 8 (Apr. 2002).
74
725 ILL. COMP. STAT. 5/115-21(c) (2003). See also Natapoff, supra note 12,
at 112-15.
75
California Commission on the Fair Administration of Justice, Report and
Recommendations Regarding Informant Testimony (Nov. 20, 2006).
76
California Commission on the Fair Administration of Justice, Press Release,
April 17, 2007, available at http://www.ccfaj.org/documents/press/Press11.pdf
77
See Report of the 1989-90 Los Angeles County Grand Jury: Investigation of
the Involvement of Jail House Informants in the Criminal Justice System in
Los Angeles County (1990).
78
CAL. PENAL CODE § 1127(a) (West 2004).
79
Dodd v. State, 993 P.2d 778, 783 (Okla. Crim. App. 2000).
80
Carriger v. Stewart, 132 F.3d 463, 479 (9th Cir. 1997).
81
Ellen Yaroshefsky, “Cooperation with Federal Prosecutors: Experiences of
Truth Telling and Embellishment,” 68 FORDHAM L. REV. 917 (December
1999): 943.
82
On Lee v. United States, 343 U.S. 747, 757 (1952); see also United States v.
Swiderski, 539 F.2d 854 (2nd Cir.1976); United States v. Sarvis, 523 F.2d
1177, 1180 (D.C. Cir. 1975); United States v. Wasko, 473 F.2d 1282 (7th Cir.
1973); United States v. Leonard, 494 F.2d 955, 961 (D.C. Cir. 1974); United
States v. Garcia, 528 F.2d 580 (5th Cir.1976); Sandoval v. United States, 426
U.S. 952 (1976).
83
D’Agostino v. State, 823 P.2d 283, 285 (Nev. 1991).
84
James Durkin, “Examining prosecutorial misconduct in Illinois,” Chicago
Daily Law Bulletin, April 26, 2003.
85
Robert M. Bloom, “Jailhouse Informants,” ABA Criminal Justice Magazine,
Spring 2003.
86
Aurelio Rojas, “A hard look at jail snitches; Bill would require that their testimony be corroborated,” Sacramento Bee, May 18, 2007.
87
Alexandra Natapoff, “California Reconsiders Snitching,” San Francisco
Chronicle, November 19, 2006.
88
George C. Harris, “Open up process of using informant witnesses,” San
Jose Mercury News, November 14, 2006.
89
Natapoff, supra note 43.
90
Center on Wrongful Convictions, The Snitch System: How Snitch Testimony
Sent Randy Steidl and other Innocent Americans to Death Row.
91
United States v. Bagley, 473 U.S. 667 (1998).
92
Center on Wrongful Convictions, The Snitch System: How Snitch Testimony
Sent Randy Steidl and other Innocent Americans to Death Row.
93
Id.
94
Rob Warden, The Snitch System: How Incentivised Witnesses Put 38 Innocent
Americans on Death Row. Research Report, Arizona State University College of
Law, Tempe, AZ, April 25, 2002, available at http://www.law.northwestern.edu/
depts/clinic/wrongful/documents/Snitch.htm.
95
Id.
96
Steve Mills and Ken Armstrong, “The Inside Informant,” Chicago Tribune,
November 16, 1999.
97
Nina Martin, “Innocence Lost,” San Francisco Magazine, November 2004, 78.

23
W W W . T H E J U S T I C E P R O J E C T. O R G

THE JUSTICE PROJECT
Dedicated to fighting injustice and to creating a more humane and just world.
1025 Vermont Avenue, NW • Third Floor • Washington, DC 20005
202 638-5855 • Fax 202 638-6056 • www.thejusticeproject.org

ACKNOWLEDGEMENTS
For their contributions toward the development of this policy review,
The Justice Project would like to thank (in no particular order)
Rob Warden, Center on Wrongful Convictions at Northwestern University School of Law;
Alexandra Natapoff, Loyola Law School; Whitney Price; McKenzie Millar,
Delia Herrin; Kate Ory; Rebecca Estes; Dardi Harrison; Leah Lavin;
Margaret Tucker; Megha Desai; Dave Seitzer; and Rosa Malley.

STAFF
John F. Terzano — President
Joyce A. McGee — Executive Director
Robert L. Schiffer — Senior Vice President
Kirk Noble Bloodsworth — Program Officer
Laura Burstein — Director of Communications
Daniel Aaron Weir — Director of National Campaigns
Edwin Colfax — Director of State Campaigns
Bill Redick — Director of Tennessee Campaign
Brad MacLean — Assistant Director of Tennessee Campaign
Shane Truett — Campaign Coordinator, Tennessee
Melissa Hamilton — Campaign Coordinator, Texas
Michelle Strikowsky — Communications Coordinator
Ginger Eldridge — Office Manager, Tennessee
Adeela Gaither — Office Manager

For information on ordering additional copies of this policy review, contact:
Daniel Weir at (202) 557-7562 or dweir@thejusticeproject.org.
©2007 The Justice Project — All Rights Reserved.

The Justice Project is comprised of two
nonpartisan organizations dedicated to
fighting injustice and to creating a more
humane and just world. The Justice Project,
Inc., which lobbies for reform, and The
Justice Project Education Fund, which
increases public awareness of needed
reforms, work together on the Campaign for
Criminal Justice Reform to reaffirm America’s
core commitment to fairness and accuracy
by designing and implementing national and
state-based campaigns to advance reforms
that address significant flaws in the American
criminal justice system, with particular focus
on the capital punishment system.

This report is made possible primarily
through a grant from The Pew Charitable
Trusts to The Justice Project Education Fund.
The opinions expressed are those of the
author(s) and do not necessarily reflect
the views of the Trusts. For additional
information, questions or comments,
please contact our offices at (202) 638-5855,
or email info@thejusticeproject.org.

THE JUSTICE PROJECT
1025 Vermont Avenue, NW • Third Floor • Washington, DC 20005
202 638-5855 • Fax 202 638-6056 • www.thejusticeproject.org