Benjamin N.

Cardozo School of Law, Yeshiva University

TESTIMONY OF
REBECCA BROWN, DIRECTOR OF STATE POLICY REFORM
& DANIELLA HENRY, POLICY ANALYST, INNOCENCE PROJECT
BEFORE MEMBERS OF THE VERMONT SENATE & HOUSE JUDICIARY COMMITTEES
RE: EYEWITNESS IDENTIFICATION & RECORDING OF INTERROGATIONS
JANUARY 15, 2013

Members of the Vermont Senate & House Judiciary Committees:

On behalf of the Innocence Project, we greatly appreciate the opportunity to testify before you today.
Indeed, Vermont has shown great leadership in the area of wrongful conviction. During the course of one
legislative session in 2007, Vermont passed a post-conviction DNA testing law and a statute providing
compensation to its wrongfully convicted, aspects of which we offer to your counterparts in other states
seeking a fair and equitable framework. Today, we are before you to seek simple changes to Vermont
police practice that will prevent wrongful convictions from ever taking place.

To date, there have been 312 DNA exonerations across the nation. In 153 of these DNA exoneration
cases, the process of exonerating the innocent also led to the identification of the true perpetrators of these
heinous crimes, who committed more than 70 rapes and 30 murders - of which we are aware - while the
innocent spent time behind bars for their crimes.

We regard each exoneration as an opportunity to identify how the system fell short and promote factually-
supported policies and procedures that make the system more reliable and accurate. The two proposals
for your consideration address two of the leading causes of wrongful conviction – eyewitness
misidentification and false confessions.

Eyewitness Misidentification
At least one mistaken eyewitness identification was a contributing factor in a full 73% of cases of
wrongful conviction proven through DNA testing. The problem of misidentifications is not unique to
certain geographic regions, but afflicts all law enforcement agencies nationwide, regardless of size or
location. While horrible harm is done to innocent people who are wrongfully convicted due to an
eyewitness misidentification, they are not the only ones who suffer. Public safety is greatly diminished,
as misidentifications cause the police to focus their investigation on an innocent person, leading them
away from the real perpetrator, who is then free to commit further crimes. Furthermore, in the rare
instances when the police return their focus on the actual perpetrator, the eyewitness who had previously
identified an innocent person is ―burned,‖ and thus not of use in the criminal prosecution. Simply put,
nobody – not the police, prosecutors, judge, jury, or indeed, the public at large – benefits from a
misidentification. The only beneficiary is the actual perpetrator.

Barry C. Scheck, Esq. and Peter J. Neufeld, Esq., Directors Maddy deLone, Esq., Executive Director
100 Fifth Avenue, 3rd Floor • New York, NY 10011 • Tel: 212/364-5340 • Fax: 212/264-5341
Innocence Project, Inc.
Page 2

The good news is that over the past 30 years, a large body of peer-reviewed research and practice has
been developed, demonstrating how simple, inexpensive reforms to eyewitness identification procedures
can greatly reduce the rate of identification error, particularly by minimizing the inadvertent misleading
influences present in traditional procedures.

In the wake of leadership from the National Institute of Justice at the U.S. Department of Justice1, the
American Bar Association2, the Police Executive Research Forum3, the International Association of
Chiefs of Police4, states across the nation have taken significant steps toward eyewitness identification
reform. In the past few years alone, the Georgia5, North Carolina6, California7, Connecticut8, and Ohio9
legislatures passed legislation to advance reform, and many other states are currently considering similar
legislation.

Misidentification is the Largest Contributor to Wrongful Convictions

Of all the causes of wrongful conviction, the most prevalent is mistaken eyewitness identification. In
fact, in many wrongful convictions, it was not just one, but multiple eyewitnesses who mistakenly
identified an innocent person:
 Luis Diaz, a Florida cook who was married with three children at the time of his arrest, was
convicted of a string of sexual assaults and served 25 years in Florida prisons. He had been
misidentified by eight witnesses.
 Kirk Bloodsworth, a former United States Marine, was convicted of having raped and murdered a
little girl in Baltimore County, Maryland based on the mistaken identification of five
eyewitnesses. Prior to his exoneration, Mr. Bloodsworth had been sentenced to death.
 Brandon Moon, an Army veteran and college student who was released in 2005 from the Texas
prison system after serving 17 years for a rape that DNA proved he did not commit, was
misidentified by five witnesses.
 Dennis Maher, a Massachusetts man, and who is here today, served 19 years for a series of rapes,
having been misidentified by three different victims.
 Stephen Phillips, a Texas man, was exonerated of a string of sexual assaults after serving 25 years
in prison. In the 11 crimes for which Phillips was wrongfully convicted, there were at least 60
victims. At least ten of those victims erroneously identified Phillips as the perpetrator. Mr.
Phillips was exonerated in 2008.

1
Eyewitness Evidence, A Guide For Law Enforcement, United States Department of Justice (Oct. 1999).
2
See ACHIEVING JUSTICE: FREEING THE INNOCENT AND CONVICTING THE GUILTY at 23-45 (Paul Giannelli et. al. eds.,
2006).
3
See JAMES M. CRONIN ET. AL., PROMOTING EFFECTIVE HOMICIDE INVESTIGATIONS at 35-60 (2007).
4
See Int‘l Ass‘n of Chiefs of Police, Training Key #600.
5
H.R. 352, 2007 Leg. (Ga. 2007).
6
H.B. 1625, 2007 Leg. (N.C. 2007).
7
S.B. 756, 2007 Leg. (Cal. 2007).
8
H.B. 5501, 2012 Leg. (Ct 2012).
9
S.B. 77, 2010 Leg (Oh 2010).

2
Innocence Project, Inc.
Page 3

Mistaken Eyewitness Identifications Also Harm Victims

Jennifer Thompson, whom you will hear from today, and Penny Beernstein were each crime victims who
identified the wrong person as their assailants, and even after DNA proved the innocence of those men,
continued to believe in their guilt – until DNA also identified the real perpetrator. It was difficult for
them to accept, not to mention horrifying for them to learn, that their memories of the actual perpetrator
were wrong and that their mistakes sent innocent people to prison. Yet as a result of their experiences,
Ms. Thompson and Ms. Beernstein are now strong advocates for the eyewitness identification reform
procedures being rapidly adopted in jurisdictions around the country.

Every time a witness makes a misidentification, the entire system suffers. Erroneous eyewitness
identifications harm crime victims, unintentionally distract police and prosecutors' attention from the true
culprit, mislead witnesses, undercut their credibility, and force innocent people to defend their innocence
and possibly go to prison for crimes they did not commit. It is, therefore, imperative that scientifically-
supported eyewitness identification procedures be uniformly applied across this state and, indeed, the
nation.

Lineup Protocols Should be Grounded in Best Practices & Social Science Research
From DNA exonerations we have learned that the standard non-blind lineup procedures provide many
opportunities for the lineup administrator to inadvertently cause a witness to select the suspect even when
the witness is unsure that this is the person from the crime scene. In other words, traditional procedures
increase identifications made as a result of witnesses guessing as opposed to actual recognition.
Traditional eyewitness identification protocol (if there is any protocol at all) also often reinforces a
witness‘s wrong choice through confirming feedback that ultimately increases their confidence in that
pick, regardless of initial hesitance, in addition to contaminating the witness‘s memory of the actual
event. Indeed, social science research has consistently confirmed not only the fallibility of eyewitness
identifications but also the unwitting tainting of witness memory through many standard eyewitness
identification procedures.

A decade ago, the Department of Justice (DOJ) addressed the problem of misidentification in a technical
working group, which sought to identify best practices supported by rigorous social science research. The
National Institute of Justice, the research arm of the DOJ, formed the ―Technical Working Group for
Eyewitness Evidence,‖ composed of membership from the scientific, legal and criminal justice
communities, which recommended a series of protocols in a report and an attendant training manual.10

Since its publication, a number of bar associations, police groups, and state commissions have conducted
more comprehensive consideration of these reforms. The American Bar Association‘s House of
Delegates adopted Resolution 111C in 2004, a statement of Best Practices for Promoting Accuracy of

10
Technical Working Group for Eyewitness Evidence. (1999) Eyewitness evidence: A Guide for Law Enforcement.
Washington, DC. United States Department of Justice, Office of Justice Programs; and Technical Working Group
for Eyewitness Evidence. (2003) Eyewitness evidence: A Trainer’s Manual for Law Enforcement. Washington, DC.
United States Department of Justice, Office of Justice Programs.

3
Innocence Project, Inc.
Page 4

Eyewitness Identification Procedures, which delineated general guidelines for administering lineups and
photo arrays. In a report of the American Bar Association‘s Criminal Justice Section‘s Innocence
Committee to Ensure the Integrity of the Criminal Process, the ABA resolved that federal, state and local
governments should be urged to adopt a series of principles consistent with those contained in its
resolution, incorporating scientific advances in research that has been developed over time.

In 2006, the International Association of Chiefs of Police (IACP) published a ―Training Key on
Eyewitness Identification,‖ which concludes that ―of all investigative procedures employed by police in
criminal cases, probably none is less reliable than the eyewitness identification. Erroneous identifications
create more injustice and cause more suffering to innocent persons than perhaps any other aspect of police
work. Proper precautions must be followed by officers if they are to use eyewitness identifications
effectively and accurately.‖ The IACP Training Key endorses a number of key reforms, including blind
administration, sequential presentation recording the procedure, instructing the witness and obtaining a
confidence statement. Then in 2010, the IACP issued a Model Policy on the same, embracing the core
reforms that will ensure the reliability of eyewitness evidence. And just last month, the IACP announced
the publication of its groundbreaking report, ―Building a Systemic Approach to Prevent Wrongful
Convictions,‖ which grew out of a national summit co-sponsored by the IACP and the DOJ. The report
calls for the adoption of the very reforms you will find in SB 184.

Efforts to address misidentification have also taken place on the state level. In April 2001, New Jersey
became the first state in the nation to officially adopt the NIJ recommendations when the Attorney
General issued Guidelines for Preparing and Conducting Photo and Live Lineup Identification
Procedures, mandating implementation of the recommendations – in addition to requiring that lineups be
administered blind and presented sequentially – by all law enforcement agencies statewide. In May 2005,
the Criminal Justice Standards Division of the North Carolina Department of Justice endorsed
recommendations set forth in the North Carolina Actual Innocence Commission‘s report,
Recommendations for Eyewitness Identification, which included ―blind‖ and ―sequential‖ lineups.11 In
September 2005, the Wisconsin Attorney General‘s Office followed New Jersey‘s lead and issued a
similar set of policies for statewide use, Model Policy and Procedure for Eyewitness Identification, which
also mandated the ―blind-sequential‖ reform package.12 In 2006, the California Commission on the Fair
Administration of Justice, comprised of key criminal justice stakeholders from across the state of
California, embraced a set of reforms in its Report and Recommendations Regarding Eye Witness
Identification Procedures.13 In 2007, the North Carolina legislature mandated the ―blind-sequential‖
reform package when it passed HB 1625, perhaps the most comprehensive piece of eyewitness
identification reform legislation to date. In 2012, the Connecticut legislature adopted the same.

11
North Carolina Department of Justice, Criminal Justice Standards Division. Recommendations for Eyewitness
Identification, May 19, 2005.
12
State of Wisconsin, Office of the Attorney General. Model Policy and Procedure for EyewitnessIdentification,
2005.
13
Please see http://ccfaj.org/documents/reports/eyewitness/official/eyewitnessidrep.pdf.

4
Innocence Project, Inc.
Page 5

Scientific Support for Eyewitness Reform

The large body of scientific research that supported these groundbreaking guidelines devised by NIJ‘s
working group nearly a decade ago has only been bolstered by a significant amount of additional peer-
reviewed studies on every aspect of these reforms. Simply put, today there is solid research and
experiential support for all of these reforms.

Blind Administration
We strongly support a statewide requirement that identification procedures be conducted double-blind,
ensuring that the lineup administrator does not know which photograph or live lineup member being
viewed by the eyewitness is the suspect. Over forty years of general social science research has
demonstrated that test administrators‘ expectations are communicated either openly or indirectly to test
subjects, who then modify their behavior in response.14 A prominent meta-analysis conducted at Harvard
University, which combined the findings of 345 previous studies, concluded that in the absence of a blind
administrator, individuals typically tailor their responses to meet the expectations of the administrator.15

Eyewitnesses themselves may seek clues from an identification procedure administrator. A recent
experiment examining the decision-making processes of eyewitness test subjects concluded that,
―witnesses were more likely to make decisions consistent with lineup administrator expectations when the
level of contact between the administrator and the witness was high than when it was low.‖16 The only
way to avoid the influence of the administrator‘s expectations on the eyewitness is through the use of a
blind administrator.

Advocating for the use of a blind administrator does not call into question the integrity of law
enforcement; rather it acknowledges a fundamental principle of properly conducted experiments – that a
person administering an experiment (or an eyewitness identification) should not have any predisposition
about what the subject‘s response should be – and applies it to the eyewitness procedure. This eliminates
the possibility – proven to exist in the eyewitness identification process – that a witness could seek, and
an administrator might inadvertently provide, cues as to the expected response.

Some worry that double-blind administration is not feasible, potentially too expensive or resource-heavy,
but this has not proven true in the field and, moreover, need not be the case. First, both large and small
police departments that have progressed to using double-blind lineups, including those in New Jersey,
North Carolina, Northampton, Denver, Dallas, Minneapolis- St. Paul, most of Wisconsin, etc., are doing
so routinely without complaint, problems, or prohibitive expenses. The experience of these departments
should quell concerns about the practicality of conducting blind lineups.

14
e.g. Adair, J. G., & Epstein, J. S. (1968). Verbal cues in the mediation of experimenter bias. Psychological
Reports, 22, 1045–1053; Aronson, E., Ellsworth, P. C., Carlsmith, J. M., & Gonzales, M. H. (1990). On the
avoidance of bias. Methods of Research in Social Psychology (2nd ed., pp. 292–314). New York: McGraw-Hill.
15
Rosenthal, R., & Rubin, D. B. (1978). Interpersonal expectancy effects: The first 345 studies. Behavioral
and Brain Sciences, 3, 377-386.
16
Haw, R. M. & Fisher, R. P. (2004). Effects of administrator-witness contact on eyewitness Identification
accuracy. Journal of Applied Psychology, 89, 1106-1112.

5
Innocence Project, Inc.
Page 6

Second, jurisdictions that have been concerned about expending any additional manpower have
implemented an alternative form of blind administration in which they ―blind‖ the non-blind
administrator. This can be done using a ―folder shuffle method,‖ as used in Wisconsin and Minnesota, as
well as through the use of laptop computers, as employed in Charlotte, NC.17 This option is contained in
the proposal before you. Implementing blind administration carries the pricetag of ten manila folders, so
those jurisdictions with limited manpower, unable to use a second administrator to perform an
identification procedure, will not experience fiscal strains.

Instructing the Eyewitness

In addition to blind lineups, ―cautionary instructions,‖ or what we also call ―witness warnings,‖ are a key
component of reform aimed at reducing the rate of mistaken identifications. Indeed, studies have
demonstrated the dramatic decrease in mistaken identifications when witnesses understand that they are
not required to identify someone at a lineup. See Nancy Steblay, Social Influence in Eyewitness Recall: A
Meta-Analytic Review of Lineup Instruction Effects, 21 L. and Hum. Behav. 283 (1997) (finding a
reduction in misidentifications when the culprit was not present from 78% to 33%, while still resulting in
87% identification of the culprit when the culprit was present).

These instructions deter the eyewitness from feeling compelled to make a selection or seek clues or
feedback from the administrator during the identification procedure about whom to pick or whether or not
a selection was correct, and otherwise help minimize the likelihood of a misidentification. This ―best
practice‖ is generally accepted by law enforcement and easily administered.

Proper Composition of the Lineup

Clearly, the optimal composition of a lineup assures more accurate selections. Therefore, the Innocence
Project advocates that the fillers be selected for a live and/or photo lineups based on their similarity to the
witness‘s description rather than on their resemblance to the suspect. As found by Gary Wells, ―the
match-description strategy is as effective as the resemble-suspect strategy at holding down false-
identification rates. In addition, our results show that the match-description strategy is much better than
the resemble-suspect strategy at promoting high rates of accurate identification. These results bolster the
argument that selecting distractors who resemble a suspect can be detrimental to maintaining high
accurate-identification rates.‖ Wells, G.L., Rydell, S.M. and Seelau, E.P., On the selection of distractors
for eyewitness lineups, 78 J. of Applied Psychol. 835 (1993).

In light of this research, the match-to-description basis for selecting lineup fillers has been recommended
by the National Institute of Justice in both its Eyewitness Evidence: A Guide for Law Enforcement and
Eyewitness Evidence: A Trainer’s Manual for Law Enforcement, the New Jersey Attorney General‘s
Guidelines for Preparing and Conducting Photo and Live Lineup Identification Procedures, the
Wisconsin Department of Justice‘s Model Policy and Procedure for Eyewitness Identification, the
California Commission On The Fair Administration Of Justice‘s Report And Recommendations

17
Based on the best practices we have advocated for some time, the Innocence Project has included attached with
this submission its recommended practices for ―blinding‖ the administrator.

6
Innocence Project, Inc.
Page 7

Regarding Eyewitness Identification Procedures, and the American Bar Association‘s Statement Of Best
Practices For Promoting The Accuracy Of Eyewitness Identification Procedures.

Consequently, non-suspect photographs and/or live lineup fillers should be selected based on their
resemblance to the description provided by the witness – as opposed to their resemblance to the police
suspect – yet in such a way that the suspect does not unduly stand out from the fillers.

Obtaining a Confidence Statement

A significant body of peer-reviewed research clearly indicates that post-identification feedback to the
eyewitness at the time the identification is made both artificially inflates the confidence of a witness in his
or her identification and also contaminates the witness‘s memory of the event.18 In other words, In
addition to the danger of confidence inflation and false certainty, when post-identification confirming
feedback is provided to an eyewitness who has incorrectly identified an innocent person, it can produce
―strong effects‖ on witnesses‘ memory, including recollection of their opportunity to view the perpetrator
and their degree of attention on the perpetrator. 19 This contaminating effect of confirming feedback,
therefore, confounds the efforts of courts to assess the reliability of identification evidence, since it
distorts and renders untrustworthy three of the five ―reliability‖ factors enunciated in Neil v. Biggers, 409
U.S. 188 (1972) (a witness‘s degree of certainty, opportunity to view the perpetrator at the time of the
incident, and degree of attention on the perpetrator). It also makes it difficult, if not impossible, for the
jury to properly assess the witness‘s confidence at the time of the out-of-court confrontation, leaving it
only with the witness‘s testimonial certainty months later. No one benefits in this situation – save for the
real perpetrator, who becomes that much more sheltered from ever being identified, prosecuted, and
convicted.

Given the corrupting effect of confirming feedback, documenting the witness‘s certainty, in his or her
own words, immediately at the time of the identification, is critical, particularly in light of research that
has consistently shown that the eyewitness‘s degree of confidence in his identification at trial is the single
largest factor affecting whether jurors believe that the identification is accurate.20 The more confidence
the eyewitness exudes – irrespective of accuracy –, the more likely jurors will believe that the
identification is accurate.

18
See, e.g., Bradfield, A. L., Wells, G. L., & Olson, E. A. (2002). The damaging effect of confirming feedback on
the relation between eyewitness certainty and identification accuracy. Journal of Applied Psychology, 87, 112-120.
and Wright, D. B., & Skagerberg, E. M. Post-identification feedback affects real eyewitnesses. Psychological
Science, 18, 172-178 (2007).
19
Wells, G.L., & Bradfield, A.L. (1998). ―‗Good, You Identified the Suspect‘: Feedback to Eyewitnesses Distorts
Their Reports of the Witnessing Experience,‖ Journal of Applied Psychology, 83, 360-376.
20
Bradfield, A. L. & Wells, G. L. (2000). The perceived validity of eyewitness identification testimony: A test of
the five Biggers criteria, Law and Human Behavior, 24, 581-594 and Wells, G.L., Small, M., Penrod, S., Malpass,
R.S., Fulero, S.M., & Brimacombe, C.A.E. (1998). Eyewitness identification procedures: Recommendations for
lineups and photospreads, Law and Human Behavior, 22, 603-647. (Surveys and studies show that people believe
strong relation exists between eyewitness confidence and accuracy).

7
Innocence Project, Inc.
Page 8

Showups
Research has demonstrated that innocent suspects are at a greater risk in showups than in lineups,
particularly (and not surprisingly) those who bear a resemblance to the actual perpetrator and/or are
wearing similar clothing. Showups can be problematic because, as social scientists have argued, the
format of an identification procedure should not directly communicate law enforcement‘s hypothesis of
the perpetrator‘s identity to the eyewitness.21 Further, an alternative format, such as a photo or live
lineups, can rule out at least some incorrect identifications, while a show-up does not present the
opportunity to identify any errors. Consequently, some criminal justice practitioners have concluded that
the show-up procedure is inherently suggestive.22

Despite the intrinsic suggestiveness of the show-up procedure, there are occasions when it might be
necessary for law enforcement. The show-up procedure can be useful for police officers who may lack
the probable cause necessary for an arrest but believe the suspect, detained close in time and proximity to
the incident, matches a general description of the perpetrator and should therefore participate in an
identification procedure. While increasing the risk to innocent suspects of being mistakenly identified,
the show-up can also afford protection to innocent suspects who are not identified and thus may be
immediately shielded from further suspicion, excluded as potential suspects, and protected from an
otherwise humiliating arrest and investigation process.

It is critical, however, given the inherent suggestiveness of the show-up identification procedure format,
that any perceived benefits be balanced against the inherent risks. Therefore, several safeguards should
be built into all show-up procedures to minimize the deleterious effects of its format.23 The police should
also take measures to minimize potentially damaging or prejudicial inferences that could be drawn about
the suspect‘s guilt, including removing the suspect from the squad car, removing handcuffs before the
arrival of the witness, and avoiding any words or conduct that may imply that the suspect is the
perpetrator of the crime. The police should not conduct showups inclusive of more than one suspect or to
more than one witness at a time. If one eyewitness makes a positive identification of the suspect, this
should provide the police with sufficient probable cause to arrest the suspect, and thus each additional
eyewitness should instead participate in either a photo or live line-up. Lastly, the police should document
the show-up procedure (using video or audio recording if practicable), including the eyewitness‘s verbal
reaction to the suspect presented and degree of certainty, in the eyewitness‘s own words, in his or her
identification. We commend the authors of this legislation in crafting provisions relating to the show-up
procedure.

21
G. Wells, G.L., Small, M. & Penrod, S. et al. (1998), Eyewitness Identification Procedures: Recommendations for
Lineups and Photospreads, 22 Law & Human Behavior, 603, 619-20.
22
See State v. Dubose, 699 N.W.2d 582, 592 (Wisc. 2005)(the Wisconsin Supreme Court ruled that show-up
identification evidence is inadmissible unless, on the basis of the totality of the circumstances, it was shown to be
necessary).
23
These safeguards are derived from Wisconsin‘s Avery Task Force‘s ―Eyewitness Identification Procedure
Recommendations,‖ which was based upon a comprehensive review and analysis of best practices, as well as from
anecdotal recommendations and other existing research.

8
Innocence Project, Inc.
Page 9

The Experiences of Those Jurisdictions that have Adopted Reforms

These changes have proven to be successful across the country. In the states of North Carolina, Texas,
and Connecticut, for instance, all jurisdictions were directed to promulgate their own policies and
procedures for implementing these reforms, and, after an exhaustive review of research and practitioner
experience, opted to implement the ―blind-sequential‖ reform package. These states reported that while
there was initial resistance from many about the need for and value of such reforms, after police were
provided the opportunity to learn more about them, receive training about how to properly implement
them, and to participate in the formation of the specific adaptations of the reforms in their jurisdictions,
those initial concerns have been replaced with acceptance of and appreciation for eyewitness
identification procedures that increase the accuracy of criminal investigations and the effectiveness of
criminal prosecutions and, by virtue of employing the most accurate eyewitness procedures available,
strengthen the persuasive and probative value of eyewitness identifications before, during and after trial.24

In addition to these states, large cities such as Minneapolis and St. Paul, MN and Milwaukee WI,
medium-sized jurisdictions such as Santa Clara, CA, and Madison, WI, and small towns such as
Norwood, MA have implemented best practices, including blind administration, and have found that they
have improved their quality of their eyewitness identifications, strengthened prosecutions, and reduced
the likelihood of convicting the innocent.

False “confessions” are far more prevalent than one might think.
A false confession, admission, or dream statement was found to have contributed to nearly 30% of
America‘s 312 DNA exonerations. Electronically recording custodial interrogations from Miranda
onward removes serious questions about the ―confession‖ in question, by enabling the finder of fact to
consider the most accurate presentation of the confession evidence at trial, thus narrowing the possibility
of a wrongful conviction.

Ancillary Benefits of Recording Interrogations

There are a number of ancillary benefits that can be achieved through the implementation of mandatory
recording. A record of the interrogation can resolve disputes about the conduct of law enforcement
officers—allegations of police misconduct can be disproven. Investigators will not have to focus upon
writing up a meticulous account of the statements provided by the suspect, and may instead focus his
attention on small details, such as subtle changes in the narrative, which he might have otherwise missed.
Having a record of good interrogation techniques can be a useful training device for police departments,
particularly as cases with distinctive characteristics come to light. Overburdened courts will welcome a
huge reduction in defense motions to suppress unrecorded statements and confessions as well as pretrial
and trial hearings focused upon establishing what transpired during the course of an interrogation.

The single best reform available to hinder the occurrence of false confessions, the recording of
interrogations, is voluntarily being embraced by police departments around the country, now estimated at

24
The North Carolina initiative described above flowed from a working group led by their Chief Justice. It is worth
noting, however, that the North Carolina Legislature chose to require the implementation of such reforms when –
after the Duke Lacrosse case and other incidents – it became clear that guidelines were not enough.

9
Innocence Project, Inc.
Page 10

850 law enforcement agencies. A total of twenty states and the District of Columbia already require
recording of interrogations statewide. Vermont should follow suit, taking the lead in instigating a reform
whose innumerable benefits will undeniably bolster the investigations of criminal cases.

In the summer of 2004, Thomas P. Sullivan, the former U.S. Attorney for the Northern District of Illinois,
whom you will also hear from today, published a report detailing police experiences with the recording of
custodial interrogations. Researchers interviewed officers in 238 law enforcement agencies which
have implemented the reform in 38 states and concluded, “virtually every officer with whom we
spoke, having given custodial recordings a try, was enthusiastically in favor of the practice.”
(Sullivan, Thomas, ―Police Experiences with Recording Custodial Interrogations.‖ Report presented by
Northwestern University School of Law‘s Center on Wrongful Convictions, p. 6.)

Assuring that all custodial interrogations be videotaped, or audiotaped, at minimum, in serious violent
crimes will assure protections to the innocent, which in turn will allow law enforcement to focus its
attention on the apprehension of the true culprit. Less than ideal interrogation procedures have
contributed to or been the main factor in nearly one in five wrongful convictions of individuals later
exonerated through DNA evidence. In each of these cases, the true perpetrator remained at large, able to
commit additional crimes. The mandatory recording of interrogations is a reform whose time has come.

Vermont’s Progress to Date

At this stage, we would like to build on the significant gains made possible by earlier efforts in Vermont
and urge the Vermont legislature to consider two proposals, which would address two of the leading
causes of wrongful conviction: one proposal that would simply require all local law enforcement agencies
to adopt scientifically-supported best practices related to eyewitness evidence; and another law that would
require the recording of custodial interrogations in serious violent crime categories, including, at
minimum, murder and sexual assault.

With respect to eyewitness identification best practices, reform has begun to take hold in Vermont, thanks
in large part to the great work of Richard Gauthier and the Vermont Criminal Justice Training Council.
Our office, along with the New England Innocence Project, worked closely with Mr. Gauthier, and his
wonderful predecessor, Jim Baker, and the training council on the development of a model policy for use
in Vermont. We also helped to sponsor a ―train the trainers‖ event, inviting a certified trainer on best
practices from Massachusetts to present before law enforcement in Vermont.

We understand from Mr. Gauthier, that to date, Vermont has trained 375 officers on this excellent model
policy, representing 25% of all law enforcement officers in Vermont. By the end of 2013, the Council
plans to have trained 50-60% of agencies and by 2014, Mr. Gauthier intends to make eyewitness
identification a mandatory component of training to uphold certification. That said, some agencies have
been trained in best practices but still have not adopted those practices at the agency level. In other
words, the Council has done all that is possible to enable training in best practices but, without a law,
is unable to assure uniform adoption across the state of Vermont.

10
Innocence Project, Inc.
Page 11

Put simply, the requirement that a mandatory training in eyewitness identification ―best practices‖
become a prerequisite of certification represents a huge step forward but will not assure uniform adoption
of these scientifically-supported best practices at the agency level. This legislative proposal would enable
the uniform application of best practices across the state of Vermont. We have drafted such a proposal,
which is modeled on Vermont‘s Fair and Impartial Policing law, which simply takes the essential,
uncontroversial, core elements of the Vermont Model Policy and ensures their adoption at the local
agency level, and have submitted this proposal for this Committee‘s consideration.

With respect to recording of interrogations, we have proposed a law that would only require electronic
recording of all homicide and sexual assault cases. For those agencies that do not possess video recording
equipment, audio recording is permitted, thereby alleviating any lingering cost concerns. With those
concerns in mind, however, we built a task force into the proposal to further consider implementation
issues in advance of the phased-in requirement. It is our hope that, in coordination with law enforcement,
we can forge a proposal that addresses concerns previously raised that such a requirement would amount
to an unfunded mandate.

We remain available to all members of the legislature who seek additional information from us and look
forward to working with stakeholders from all corners of the criminal justice community as these
proposals move forward.

######

Thank you for the opportunity to testify before you about these critically important innocence reforms.

11