Skip navigation

Report on Wrongful Conviction of Jeffrey Deskovic Ny 2007

Download original document:
Brief thumbnail
This text is machine-read, and may contain errors. Check the original document to verify accuracy.
Report on the Conviction
of Jeffrey Deskovic

Prepared at the Request of
Janet DiFiore
Westchester County District Attorney

Judge (ret.) Leslie Crocker Snyder,
Member, Kasowitz, Benson, Torres & Friedman
Judge (ret.) Peter J. McQuillan
Hon. William L. Murphy,
Former Richmond County District Attorney
Richard Joselson, Esq.,
Supervising Attorney, Criminal Appeals Bureau
The Legal Aid Society of New York City

June 2007

TABLE OF CONTENTS
SUMMARY OF THE FACTS…………………………………………………........................... 1
WHAT WENT WRONG…........................................................................................................... 3
Introduction………………………………………………………………………………. 3
What Went Wrong - - An Overview……………………………………………………... 5
Police and Prosecutorial Failures………………………………………………… 5
Defense Failures Before and During Trial………………………………………. 6
Other Errors……………………………………………………………………… 7
Errors by Police and Prosecution……………………………………………………….. 7
Police & Prosecutorial Tunnel Vision…………………………………………… 7
Police Over-Reliance on NYPD Profile………………………………………... 10
Selective Reading of Deskovic’s Statements………………………………….. 11
Troubling Police Tactics in Dealing with Deskovic…………………………… 15
Carelessness or Misconduct in Police Investigation…………………………. 17
Prosecution Decision to Proceed with Grand Jury
Prosecution Before Receiving DNA Results………………………………… 20
Prosecution’s Questionable Presentation of Scientific Evidence……………….21
Defense Failures……………………………………………………………………….. 24
Defense Failure to Use Evidence of Deskovic’s
Psychological Vulnerabilities………………………………………………. 24
Defense Failure to Maximize the Exculpatory Value of the
Scientific Evidence ………………………………………………. . . . . . . . 26
Defense Conflict of Interest in Representing Freddy Claxton……………….. 27
Other Errors…………………………………………………………………………… 29
Court’s Midtrial Loss of Evidence…………………………………………… 29

i

CORRECTIVE ACTION…………………………………………………………………. .30
Introduction………………………………………………… ………………………30
The Unidentified DNA Profile…………………………………………………….. 31
Videotaping the Interrogation………………………………………………………32
Commission of Inquiry……………………………………………………………. 34
Evidence Collection and Storage…………………………………………………. 35

ii

SUMMARY OF THE FACTS
Jeffrey Deskovic was a 16 year-old sophomore at Peekskill High School in Peekskill,
New York in November 1989.
On the morning of November 17, 1989, Peekskill police found the raped, strangled and
beaten body of 15 year-old Angela Correa, a classmate of Deskovic’s, in a wooded area in
Hillcrest Park known as “the Pit”; it was covered with leaves. Her body was mostly naked, with
items of clothing nearby.
Correa had last been seen on November 15, 1989.
She had left school at about 2:45pm, gone home and changed into casual clothes. When
she left home, she carried a camera and telephoto lens in a plastic bag and wore a walkman-type
cassette player with a “New Kids on the Block” cassette inside and headphones.
She was reported missing by her family on November 16th and considered a missing
person by the Peekskill Police Department by November 17th.
A witness, William Harrison, had been taking his daily walk through Hillcrest Park the
afternoon of November 15th at approximately 3:30pm and saw Angela. He heard voices which
sounded like people arguing, but could not hear what was said.
After Angela’s body was discovered on November 17th, numerous police officers and
Detectives responded to the scene and sealed off the entire area of the park. For 6 hours, no one
was permitted to enter the area, after which the police ended their crime scene investigation and
the area was open to the public.
In addition to learning that Angela had been raped, the police found a piece of torn, wet
white paper under her body which appeared to be part of a note written by Angela to “Freddy”
and dated 11/15/89. Also found at the crime scene were Angela’s torn bra (a one piece pullover), among other items, including 3 different type of head hairs, none ever matched to
Deskovic Ultimately, 3 separate crime scenes were denoted by the police.
Detectives Thomas McIntyre and David Levine headed the subsequent investigation into
Correa’s murder. Dr. Louis Roh performed the autopsy on Correa’s body. The cause of her
death was a fractured skull, internal hemorrhage and asphyxiation due to ligature strangulation.
There were various internal and external injuries. Among them were indicia that the body had
been dragged, face down, over a dirt surface. Abrasions and contusions in the vaginal area led to
an examination of the underlying tissue which showed fresh hemorrhage indicating recent force
was applied to the vaginal surface consistent with forcible sexual intercourse. In addition,
multiple tearing on her hymen indicated to Roh prior sexual activity. Roh opined that the date
and time of Correa’s death were consistent with occurring between 3:30 p.m. and 4:30 p.m. on
November 15, 1989.

1

During the course of the two month-long police investigation, the police interviewed
numerous classmates of Deskovic’s and Angela’s at Peekskill High School and learned that
Deskovic was allegedly absent from school at the Medical Examiner’s estimated time of her
death (3:30-4:30pm on Nov. 15), had attended all 3 wakes for Angela and had been observed
distraught and crying over her death. Freddy Claxton, a classmate presumed to be “Freddy”, had
an alibi.
From December 12 until Deskovic’s arrest on January 25, 1990, numerous encounters
occurred between various Detectives and Deskovic, especially Detective Thomas McIntyre and
David Levine, the lead detectives in the case. Some were initiated by the police, some by
Deskovic. Deskovic was Mirandized on most occasions. Deskovic asked the police not to tell
his mother and, despite his age, 16, they never contacted her at any point in the investigation.
They continued their contact with Deskovic, even after Deskovic’s mother made clear she did
not want Deskovic talking to the police and, had attempted unsuccessfully to hire an attorney to
represent him.
Ultimately Detective McIntyre asked Deskovic to take a polygraph examination.
Deskovic said he would think about it and continued to conduct his own “investigation,” eager to
share his notes with the police, including various “revelations” about the crime and crime scenes
which included drawing an accurate diagram of the crime scene. He also gave a blood sample to
the police (January 10, 1990), after being told by Det. Levine that this would resolve whether he
had been involved in the crime or not.
During the various times the police questioned Deskovic, most importantly on January 10
and January 25, a tape recorder was available. On January 10, 35 minutes of a four hour session
with Detective Levine and Lieutenant Tumulo were recorded, Levine having turned the tape on
and off 3 times. The January 25 session, summarized below, was not recorded at all.
On January 25, Deskovic voluntarily submitted to a polygraph, as requested by the
police. He arrived at police headquarters alone at about 9:30am, without either a lawyer, his
mother, a friend or family member. He was driven to Brewster, N.Y. by Detectives McIntyre
and Levine. The polygraph was administered by police investigator Stephens. Deskovic was
questioned sporadically until 5pm, at which time Deskovic asked for Det. McIntyre, after being
informed he had failed the exam. Deskovic “confessed” to McIntyre that he killed Angela, fell
on the floor in a fetal position, where McIntyre physically comforted him by holding Deskovic’s
head on his lap and rubbing his back. During the course of eliciting “background” information,
Stephens learned that Deskovic “sometimes hears voices and they make me to things I
shouldn’t.” The entire day’s session was not tape recorded.
During the course of the investigation, swabbings of Correa’s vaginal cavity had revealed
seminal fluid and intact spermatozoa. In January, 1990, the fluid was sent to the FBI laboratory
for DNA analysis and compared with the sample of Deskovic’s blood. On March 2, the police
were informed that Deskovic had been conclusively excluded as the source of the seminal fluid
found inside Angela. No action was taken by the police as a result. No physical evidence or
eyewitness testimony was found to connect Deskovic to Angela’s rape and murder. The
scientific evidence exonerated him.

2

It should be noted that Deskovic was described as “suicidal” or having “suicidal ideation”
as of January 26, 1990, the day after his arrest, and was hospitalized from early March 1990 until
August 13, 1990 when he was released from Rockland Children’s Psychiatric Hospital. A
competency examination had been ordered and completed in June, 1990.

WHAT WENT WRONG
Introduction
On January 18, 1991, Deskovic, then seventeen, faced the court for sentencing after being
convicted of murdering and raping Correa. Granted an opportunity to speak, Deskovic thanked
his attorney and family for standing by him, and implored the court to overturn the guilty verdict:
I didn’t do anything. I’ve already had a year of my life taken from
me for something I didn’t do, and I’m about to lose more time and
I didn’t do anything. 1
Deskovic’s proclamation of innocence carried more weight than many similar pleas. The
whole case against him had been built upon a series of his own statements, most of them
unrecorded, culminating in an allegedly incriminating statement, which was entirely unrecorded
and which was elicited only after a lengthy, confrontational polygraph examination. No
eyewitness had identified him as the perpetrator or placed him near the scene of the homicide.
No physical evidence connected him to the crime. Indeed, seminal fluid and live sperm found in
the victim’s body following the rape/murder definitively excluded Deskovic as it source. So did
several hairs removed from the victim during the autopsy. Deskovic was sixteen years old with
no prior record at the time of the crime. Nevertheless, his plea left the Assistant District
Attorney not only persuaded of his guilt, but “convinced that he might very well do this again.” 2
The court called the case a “classic tragedy” in which “two young people’s lives [were]
destroyed.” 3 It said that it had observed Deskovic more closely than it had observed any other
defendant, recognized his family and community support and acknowledged that “maybe [he
was] innocent.” 4 Still, the “jury ha[d] spoken and the court could neither “quarrel” nor

1

2

3

4

See Sentencing Tr. at 10.
See Sentencing Tr. at 6.
See Sentencing Tr. at 11.
See Sentencing Tr. at 10-11.

3

“disagree” with the verdict. 5 It imposed the minimum lawful sentence of imprisonment of 15
years to life. 6 As the proceeding ended, Deskovic spoke once again:
I will be back on appeal. Justice will yet be served. I will be set
free. 7
Deskovic was right. Tragically, however, his vindication would not come for sixteen
years.
Following his conviction and sentence, Deskovic received all of the traditional elements
of post-judgment due process, but he obtained no relief. The Appellate Division, Second
Department unanimously affirmed his conviction, opining that there was no “indication in the
record” that police elicited the inculpatory statement in a manner that “could [have] induce[d] a
false confession” and that the evidence against Deskovic was “overwhelming.” People v.
Deskovic, 210 A.D.2d 579 (2d Dept. 1994). That same court then summarily denied a motion to
reargue. Thereafter, a judge of the Court of Appeals denied Deskovic’s application to bring his
case before New York’s highest court. People v. Deskovic, 83 N.Y.2d 1003 (1994).
His state appeals exhausted, Deskovic tried to obtain federal habeas corpus review, but,
owing to an error by his attorney, he missed the statute of limitations. Deskovic v. Mann, 1997
WL 811524 (S.D.N.Y. 1997). That determination was affirmed by the United States Court of
Appeals for the Second Circuit. Deskovic v. Mann, 210 F.3d 354 (2d Cir. 2001). Finally, the
United States Supreme Court ended Deskovic’s collateral attack on his conviction by denying his
petition for certiorari. Deskovic v. Mann, 531 U.S. 1088 (2001).
In the ensuing years, Deskovic repeatedly asked the then-Westchester District Attorney to
run the DNA samples from the case against the State and federal DNA databases. His requests
were unavailing. Only after Deskovic had obtained counsel from the Innocence Project and a
new Westchester District Attorney, Janet DiFiore, had taken office did his luck change. DiFiore,
to her great credit, promptly agreed to the necessary testing and consented to Deskovic’s release
when the sample matched that of a man serving a life sentence for an unrelated murder. That
man subsequently confessed to raping and killing Angela Correa. He recently pled guilty to the
charge.
On November 2, 2006, the First Deputy District Attorney sought dismissal of the
indictment on the ground that “Deskovic [was] actually innocent.” The prosecutor offered
apologies on behalf of her office and the Peekskill Police Department. The court expressed
regret as well. When the case against him was finally dismissed, Deskovic was 33 years old. He
had been incarcerated half his life for a crime he did not commit.
5

6

7

See Sentencing Tr. at 11.
See Sentencing Tr. at 13.
See Sentencing Tr. at 14.

4

It is, of course, theoretically possible that, even as this terrible tragedy unfolded, the
criminal justice system functioned exactly as it should. One can imagine a situation in which
police, prosecutors, defense counsel and the courts each discharged their functions in a perfectly
appropriate way, yet the result achieved was calamitously wrong. One can imagine such a case,
but Jeffrey Deskovic’s case is not such a case. We have reviewed the voluminous public record
of the proceedings below, and we are persuaded that, from the outset and continuing at every
stage thereafter, errors were made that propelled the case towards its unjust outcome. We do not
catalogue these mistakes to assign fault or to apportion blame. There may be a place for that, but
that place is not here. Rather, we attempt to analyze what went wrong for Jeffrey Deskovic in
the hope that a broader understanding of his tragedy will help those who work in the criminal
justice system take the steps necessary to protect others from his fate.
What Went Wrong – An Overview
Before embarking on a step-by-step review of what we believe went wrong in Deskovic’s
case, we provide a summary of the problems that we have detected:
POLICE AND PROSECUTORIAL FAILURES
·

Police & Prosecutorial Tunnel Vision – The police focused too early on Deskovic as their
prime suspect due, in part, to an inaccurate NYPD profile of the offender. Because they
believed he was guilty, detectives interrogated Deskovic in a manner that improperly
exploited his youth, naiveté and psychological vulnerability, thereby eliciting a false
inculpatory statement. The prosecution, which, like the police, believed it had its man,
failed to undertake a necessary reassessment of its case when scientific facts emerged
(e.g., DNA and hair evidence) that appeared to exculpate Deskovic. Specifically, the
record indicates that all investigation ceased after police obtained Deskovic’s purported
confession. The prosecution apparently did little or nothing to corroborate the theories it
employed to square the scientific evidence with Deskovic’s guilt. There is no evidence,
for example, that much was done to locate the “boyfriend” who was the supposed source
of semen or even to document Correa’s movements in the 24 hours before her death
when, according to the prosecution theory, she must have had consensual sex. The
prosecution also affirmatively decided not to seek hair samples from the ME and ME’s
assistant, who, in the prosecution’s view, were the sources of hair found on Correa’s
body.

·

Police Over-Reliance on NYPD Profile – Shortly after they discovered Correa’s body,
Peekskill police sought an offender “profile” from the NYPD. The profile ultimately
proved inaccurate in almost every respect, but it appeared to fit Deskovic, prompting a
premature focus on him as a prime suspect.

·

Selective Recording of Deskovic’s Statements – The two most inculpatory statements in
the case were the 1/10 statement during which Deskovic supposedly drew an accurate
diagram of the crime scene and the 1/25 polygraph exam, which culminated in
Deskovic’s confession. Although on 1/10, Deskovic spent approximately four hours with
Detective Levine and Lieutenant Tumulo, only about 35 minutes of that session was

5

recorded because Levine turned the tape on and off three times. The 1/25 session was not
recorded at all. As to both, the opportunity for complete taping undeniably existed. Had
those statements been recorded, jurors would have been in a far better position to
evaluate them at trial. In the absence of recordings, however, jurors were forced to rely
exclusively on accounts by the police. That other, less consequential statements were
recorded, in whole or in part, makes the failure to record the key statements appear
possibly deliberate and certainly tactical.
·

Troubling Police Tactics In Dealing with Deskovic – Though legal, the police tactics in
dealing with Deskovic did not take adequate account of his youth, naiveté, inexperience
with the justice system and psychological vulnerabilities, particularly in light of their
knowledge that Deskovic’s mother did not want him involved in the police
“investigation” and that, at least at one point, the family had sought to retain counsel on
Deskovic’s behalf.

·

Carelessness or Misconduct in the Police Investigation – Much of the prosecution’s effort
to persuade the jury that Deskovic’s statements established his guilt hinged on the
argument that Deskovic knew things about the crime that only the killer could know (e.g.,
crime scene details, the existence of a note, found with Correa’s body, to Freddy Claxton,
another high school classmate). Given Deskovic’s innocence, two scenarios are possible:
either the police (deliberately or inadvertently) communicated this information directly to
Deskovic or their questioning at the high school and elsewhere caused this supposedly
secret information to be widely known throughout the community.

·

Prosecution’s Decision to Proceed with the Grand Jury Presentation Before Receiving the
DNA Results – Though the prosecution acted lawfully in proceeding as it did, it would
have been wiser to seek delay of the grand jury presentation until after the DNA results
arrived. The case certainly looked different at that point and a thorough re-examination
might have been easier to pursue had a murder indictment not yet been obtained.

·

Prosecution’s Questionable Presentation of the Scientific Evidence – Confronted with
scientific evidence that did not support its case, the prosecutor developed strained and
shifting theories to explain that evidence away. The prosecutor refused, until the end of
the trial, to commit to a theory of the DNA evidence. Specifically, the prosecution
waffled about whether the source of the semen found in Correa was an unidentified
boyfriend who had had consensual sex with her at some unspecified time before her death
or an unapprehended accomplice, who had deposited the sperm during the rape. It is also
difficult to justify the prosecutor’s decision not to test hair samples from the Medical
Examiner or his assistant when the prosecution argued to the jury that they were the
likely source of hair found on Correa’s body.

DEFENSE FAILURES BEFORE AND DURING THE TRIAL
·

Defense Failure to Use Evidence of Deskovic’s Psychological Vulnerabilities – The
defense faced the difficult task of explaining to jurors why someone might confess to a
murder that he did not commit. Ample evidence existed in the record demonstrating

6

Deskovic’s psychological vulnerabilities. Nevertheless, the defense did not attempt to
introduce psychiatric evidence that might have persuaded jurors that Deskovic was
particularly vulnerable to the police tactics employed against him and that those tactics
induced a false confession. In the absence of such evidence, the defense attack on the
statements seemed scattershot and unfocused.
·

Defense Failure to Maximize the Exculpatory Value of the Scientific Evidence – There is
no question that, from the defense perspective, the most favorable evidence in the case
was the scientific proof that excluded Deskovic as the source of the semen present on the
vaginal swabs taken from the victim. Despite this, the defense made no effort to
meaningfully question the forensic biologist or the FBI DNA expert, who were the
sources of this evidence. As a consequence, the defense failed to make the most of the
evidence that should have served as the centerpiece of its case.

·

Defense Conflict of Interest in the Representation of Freddy Claxton – Before Deskovic’s
arrest, the Legal Aid Society of Westchester began representing Freddy Claxton, who had
been questioned by police following Correa’s death. Eventually, Claxton became an
important, yet unseen, player in the case against Deskovic: the prosecutor suggested that
Claxton may have had consensual sex with Correa before the crime and Deskovic’s
knowledge of Correa’s note to Claxton was cited as an incriminating fact. The defense
should have had every incentive to establish that Claxton did not have consensual sex
with Correa and that Claxton may have spoken to others at school about the note (police
had questioned him about it). The defense representation of Claxton necessarily
hampered these efforts. Further, the court should have considered appointing
independent counsel to explain the potential conflict to Deskovic, but it did not do so.

OTHER ERRORS
·

Court’s Mid-trial Loss of Evidence – As the end of the trial neared, a courthouse cleaning
crew inadvertently discarded many of the exhibits that had been introduced in evidence,
including the clothing worn by Correa when she died. The evidence had inexcusably
been left unsecured in a black plastic garbage bag in the courtroom. There was
controversy between the parties about the condition of some of these items and, because
the evidence had been lost, the jury’s deliberation request to examine some of it could not
be accommodated. Even in the absence of bad faith, given the stakes, this cavalier
treatment of the evidence was unacceptable.
ERRORS BY THE POLICE AND PROSECUTION CONTRIBUTED TO DESKOVIC’S
CONVICTION

Police & Prosecutorial Tunnel Vision
Time and again, academics, advocates and independent investigators alike have identified

7

tunnel vision as a primary cause of wrongful convictions 8 . Jeffrey Deskovic’s wrongful
conviction is no exception.
Tunnel vision by law enforcement authorities does not imply a malicious intent to frame
the innocent. Nor does it suggest that authorities fixate on particular suspects arbitrarily. Rather,
tunnel vision is a “natural human tendency” that causes “lead actors in the criminal justice
system to focus on a suspect, [and then] select and filter evidence that will build a case for
conviction while ignoring or suppressing evidence that points away from guilt.” 9 When tunnel
vision operates in criminal investigations, a conclusion by police or prosecutors, however
tentative, becomes the lens through which all subsequently uncovered information is assessed.
And there is the inevitable institutional pressures on police to “clear cases,” especially serious or
high profile ones, and on prosecutors to obtain convictions following arrests.
Jeffrey Deskovic’s case provides a textbook illustration of tunnel vision in action.
Deskovic’s behavior following Correa’s death was certainly odd and it was perfectly appropriate
for police to take a closer look at him. Early in the investigation, however, Peekskill authorities
obtained a so-called offender profile from the NYPD. 10 This profile ultimately proved wildly
inaccurate, but it appeared to match Deskovic in many respects. Through the filter of this
inaccurate profile, detectives regarded Deskovic’s unusual behavior with rapidly increasing
suspicion.
The tunnel vision that led police to hone in exclusively on Deskovic caused them to
overlook or undervalue exculpatory aspects of Deskovic’s unusual behavior. By one
interpretation – the one favored by police – Deskovic was a disturbed killer who was inserting
himself farther and farther into the police investigation until he could finally take responsibility
for what he had done. By another light, however, Deskovic was a lonely and immature young
man, who was genuinely conducting his own “investigation” in a juvenile attempt to gain
acceptance from the police. Supporting the latter theory was the fact that much of what
Deskovic provided was silly or dead wrong. For example, he urged the authorities to investigate
one of his classmates, with whom he had previously quarreled. 11 He offered them the surmise,
8

See Keith A. Findley & Michael S. Scott, “The Multiple Dimensions of Tunnel Vision in
Criminal Cases,” 2006 Wisc. L.Rev. 291, 293-94 (2006) (hereinafter “Findley & Scott”)(citing,
as examples, the findings of former Illinois Governor George Ryan’s Commission on Capital
Punishment, numerous Canadian government inquiries on the causes of wrongful convictions
and the report of the Innocence Commission for Virginia). According to the website of the
Innocence Project of the Benjamin N. Cardozo School of Law, “One commonality in almost all
of the [160-plus post-conviction DNA exonerations] is that they feature some form of tunnel
vision.” Innocence Project, Benjamin N. Cardozo School of Law, “The Problem of Tunnel
Vision in Criminal Justice,” posted at www.innocenceproject.org (last visited February, 2007)
9 Findley & Scott at 292, 305.
10
See Complaint Follow-up Report of Det. McIntyre re: profile (undated).
11

See Testimony of Det. Levine, Trial Tr. at 829; Testimony of Inv. Stephens, Trial Tr. at

1014.

8

wholly unsupported by anything police had unearthed, that the victim’s sister was covering for
someone. 12 Finally, he searched the crime scene and located a key, which he thought might
belong to Correa, but which, in truth had absolutely nothing to do with the crime. 13 These
actions and ones like them might well have prompted open-minded investigators to conclude that
Deskovic merely was an inept (but innocent) amateur detective, but the Peekskill detectives had
already become convinced of his guilt.
Tunnel vision continued to operate. As the police became increasingly convinced of
Deskovic’s culpability, the tactics they employed against him made him appear more guilty in
their eyes. At times, they were highly confrontational, accusing him of the crime and sharply
rejecting his professions of innocence. 14 On other occasions, they were collegial, seemingly
inviting this teenager to play a critical role in their investigation. 15 These tactics, themselves the
product of the narrowing police focus on Deskovic, laid the groundwork for a false confession. 16
Ultimately, detectives drove Deskovic to another town and interrogated him in a 10x 10 room for
hours against the pressure-filled backdrop of a polygraph test. They proceeded as they did
because, by then, they were fully convinced of Deskovic’s guilt. That their target was a
psychologically vulnerable sixteen-year-old boy, who had no previous experience with the
justice system, was of no moment under these circumstances. On the day of the polygraph, they
acted for the avowed purpose of getting a confession. 17 The tactics they employed were
designed for this purpose and, although those tactics were left largely unexamined by the finders
of fact because the interrogation was unrecorded, they succeeded.
Once Deskovic was arrested, tunnel vision also distorted the prosecution’s behavior.
Convinced that its man was in custody, the District Attorney’s office successfully pressed for an
12

13

See Testimony of Det. Levine, Trial Tr. at 834.
See Testimony of Det. McIntyre, Trial Tr. at 1175, 11247.

14

See Complaint Follow-up Report of Det. McIntyre (acknowledging that he questioned
Deskovic’s account during their first conversation and requested he take polygraph); Testimony
of Det. McIntyre, Trial Tr. at 1159-60 (same); Testimony of Det. David Levine, Trial Tr. at 710,
815, 849, 852 (told Deskovic that he was a suspect during questioning on January 9th and
accused him of committing the crime during interrogation on January 10th).
15

See Testimony of Inv. Stephens, Trial Tr. at 163 (Deskovic told Stephens that he was
taking the polygraph on January 25th because police would permit him to participate in their
investigation if he passed).
Findley & Scott at 333-40 Saul M. Kassin, “On the Psychology of Confessions: Does
Innocence Put Innocents at Risk?,” 2005 American Psychologist 215, 219-22 (April 2005)
(hereinafter cited as “Kassin”)
16

17

See Testimony of Inv. Stephens, Trial Tr. at 1034 (acknowledging that goal of
polygraph procedure was to elicit confession).

9

indictment, rather than await the results of potentially exculpatory scientific testing. When DNA
and hair analysis seemed to exclude Deskovic as the perpetrator, the prosecution did nothing.
Still firmly committed to its belief in Deskovic’s guilt, it eschewed the opportunity to re-open the
investigation. Rather, it seemed determined to square the exculpatory scientific proof with the
case against Deskovic. As for the DNA, the prosecution veered between contending that the
sperm had been deposited by an unapprehended accomplice of Deskovic (even though their
profile described a crime committed by a loner) and urging that sperm had nothing at all to do
with the rape, but rather was the product of consensual sex at an unspecified time and an
unknown place with a boyfriend who never could quite be identified. Not until summation was
it clear that the prosecution would press the latter DNA theory and jettison the former.
As for the hair evidence, the prosecution insisted that the hairs found on the body – hairs
that could not have belonged to Deskovic – must have come from the Westchester County
Medical Examiner or his assistant. Even as they made this argument, the prosecution steered
scrupulously clear of asking these men for hair samples, although their own expert suggested that
this would be a good idea. 18 The tactical desire to win thus prevented the prosecution from
taking an easy and obvious step that could have contributed to discovering the truth.
From start to finish, tunnel vision propelled the police investigation and then the District
Attorney’s prosecution. No one would suggest that the intended result was to convict the
innocent, but that is exactly what occurred.
Police Over-Reliance on NYPD Profile
On November 24, 1989, only a week after the discovery of Correa’s body, Detectives
McIntyre and Levine met Detective Pierce of the NYPD Criminal Assessment and Profiling
Unit. After reviewing the facts of the investigation and examining crime scene photos and a map
of the location, Pierce offered a detailed profile of the offender. He predicted that Correa’s killer
would be a white or Hispanic man, less than 25 years old and probably less than 19 years old,
who was shorter than 5'10". According to Pierce’s profile, the offender knew his victim. The
murderer was a loner who was unsure around women. He had little involvement in school
activities. He likely had a physical handicap or was mentally slow. He could well have been a
“trouble maker” with an assaultive history and involvement with drugs or alcohol. 19
The first and most obvious effect of the profile was that it focused police attention on the
students of Peekskill High School, which Correa had attended. While it surely was not
unreasonable to investigate Correa’s classmates as possible suspects, the profile may well have
served prematurely to foreclose other potentially fruitful areas.
The profile also directed police to a particular kind of Peekskill student. Early police
investigative reports strongly suggest that the detectives’ initial suspicions of Deskovic were
18

19

See Testimony of Peter DeForest, Trial Tr. at 385.
See Complaint Follow-up Report of Det. McIntyre re: profile (undated).

10

validated and then elevated precisely because Deskovic appeared to fit Pierce’s description of the
offender. Deskovic was, after all, a 5'10" white man who was under 19 at the time. As Detective
McIntyre noted in a complaint follow-up report that was apparently authored before Deskovic’s
first encounter with police, Deskovic knew the victim and was described as a “loner.” Unnamed
sources told McIntyre that Deskovic was “emotionally handicapped,” “emotionally distraught”
and delusional, and suggested that he had previously “assaulted” his mother. 20 This seemed to
gibe perfectly with the profile, which described an offender who was a “trouble maker” with an
“assaultive history.”
Subsequent police reports regarding Deskovic continued to hew closely to the terms of
the profile: Deskovic was deemed “hostile and agitated,” 21 “a very troubled youngster,” who
was “schizo,” “very secretive,” “very nasty,” “prone to violence,” and “extremely violent
towards his mother, hitting and pushing her and screaming obscenities at her.” 22 We cannot
now know, but it would take no great leap to infer that police actually used the profile in framing
their questions to others about Deskovic. Because questions often suggest answers, as the
investigation continued along this route, Deskovic appeared to match the profile even more. If
the police reports are any indication, before long, Deskovic became the exclusive focus of the
investigation. The profile marked the first step – actually, the first misstep – down that path.
Throughout the early stages of the investigation, the profile reinforced the police perception that
Deskovic was guilty.
As it turned out, of course, the initial profile of the killer was wrong in almost all critical
respects. The man presently charged with Correa’s murder – the man whose DNA was found in
her body and who has confessed to the crime – is African-American, not white or Hispanic. He
was nearly 30 years old, not less than 19. He was a total stranger, not an acquaintance of the
victim. As the profile predicted, he had no involvement in school activities, but that was because
he had absolutely no association with the high school. 23
Selective Recording of Deskovic’s Statements

20

See Complaint Follow-up Report of Det. McIntyre re: Deskovic’s conduct at Correa’s
wake and funeral (undated).
21

See Complaint Follow-up Report of Det McIntyre re: 12/12 interview with Deskovic
(undated). Notably, Deskovic became hostile and agitated only after McIntyre had suggested
that his reactions to Correa’s death made no sense, expressed disbelief for his story and asked
him to take a polygraph.
22

See Complaint Follow-up Report of Det. Brovarski re: interview with mother of John
Larino (undated).
23

See Jonathan Bandler, “Killer, I Strangled Her,” The Journal News (Westchester
County, NY, October 6, 2006, p. 1A.

11

The police had no problem getting Jeffrey Deskovic to talk to them about Angela Correa.
On seven separate occasions between December 12, 1989, and January 25, 1990, Deskovic
spoke with the authorities about the case. 24 The Peekskill detectives also had the wherewithal to
record their interrogations of Deskovic. 25 By recording those sessions only selectively, and by
failing to record those portions of the interrogations in which Deskovic allegedly made his most
incriminating remarks, the police denied the jury an essential tool for evaluating the setting in
which the statements were made. The police failure to create a complete taped record of their
interrogations of Deskovic was likely a major cause of this erroneous conviction.
Legal academics as well as many who have scrutinized the causes of wrongful
convictions have long advocated the videotaping of police interrogations, at least those involving
serious crimes 26 . Significantly, even when it is not mandatory, law enforcement in many
jurisdictions has embraced electronic recording, believing that it both provides a complete record
of an interrogation and that it protects police from unwarranted claims of misconduct 27 .
In 1989 and 1990, when Deskovic’s case arose, the Peekskill police had the capacity to
record interrogations and, indeed, they did some of the time or selectively. Both Detective
Levine and Detective McIntyre had ready access to micro cassette recorders and standard
recorders for this purpose. 28 Levine and McIntyre taped portions of their police station
interrogations of Deskovic on January 10th. 29 In addition, Levine recorded his entire telephone
24

See Notice to Defendant of Intention to Offer Evidence Pursuant to Section 710.30 CPL
(describing Deskovic’s statements to police on December 12, 1989, and January 9,10, 22, 23, 24,
25, 1990).
25

See id. (indicating that at least portions of the January 10, 1990 statement and the
January 22, 1990 statement were recorded).
26 Findley & Scott at 391-93; Kassin at 225; Canadian Commissions of Inquiry, “Report
of the Working Group on the Prevention of Miscarriages of Justice,” October 20, 2005.
Currently, Great Britain and four U.S. states – Minnesota, Alaska, Illinois and Maine – all
require that interrogations be recorded in some form. Kassin at 225. In Massachusetts, the
Supreme Judicial Court stopped just short of mandating videotaping, holding instead that, if an
interrogation resulting in a confession is unrecorded, then the defendant will be entitled to a jury
instruction urging caution in the use of that confession. Commonwealth v. DiGiambattista, 442
Mass. 423, 813 N.E.2d 516 (Mass. 2004)
27

See Thomas P. Sullivan, “Electronic Recording of Custodial Interrogations: Everybody
Wins,” Crim. L. & Criminology, 1127 (2005) (cataloguing jurisdictions where recording occurs).
28

See Testimony of Det. Levine, Trial Tr. at 637-43; Testimony of Det. McIntyre, Trial
Tr. at 1094-99, 1144, 1199.
29

See Testimony of Det. Levine, Trial Tr. at 637-43; Testimony of Det. McIntyre, Trial
Tr. at 1094-99.

12

conversation with Deskovic on January 22nd as well as a brief conversation he had with Deskovic
outside a friend’s home later that evening. 30 The Westchester District Attorney’s office also
recognized the potential evidentiary value of taped statements. The prosecution introduced the
tapes of those portions of the interrogations that had been recorded during its direct case at
Deskovic’s trial. 31
That the detectives chose to record some of their interrogations with Deskovic renders
their failure to record others particularly questionable. For example, Levine acknowledged that
he and Lieutenant Tumulo spoke with Deskovic for approximately four hours at the police
station on January 10th, but had the recorder running for only about 35 minutes. 32 Specifically,
Levine turned on the tape at 3:55 p.m. that afternoon, but turned it off around 4:15 p.m. He did
not reactivate the recorder until 5:20 p.m. and he shut it off again, for good this time, at 5:40
p.m. 33 Levine offered a series of explanations for his conduct, all of them unpersuasive. On one
occasion, he contended that he turned the tape off when he went to get coffee and forgot to turn it
back on. 34 Later in the interrogation, he claimed that he turned off the machine because he
believed Deskovic was uncomfortable with it on. He did not explain, however, why, only a
moment before, he had decided to remove the recorder from his pocket and place it on the table,
where Deskovic could see it, thereby triggering his supposed discomfort. 35
These explanations notwithstanding, the record strongly suggests that the decision about
when to press play and when to press stop was governed, at least in part, by a tactical desire to
choreograph which parts of the interrogation a fact-finder would ultimately hear. The recorder
was fully operational, for example, when Levine professionally read Deskovic Miranda warnings
and Deskovic calmly waived his rights. 36 When Deskovic drew his supposedly incriminating
crime scene diagram, in contrast, Levine had switched off the recorder. 37 Similarly, by the time
the interrogation turned angry and confrontational, the tape had stopped running. 38 Jurors thus
never got to hear the exchange that occurred when Levine branded Deskovic a liar and accused

30

31

32

33

34

35

36

37

38

See Testimony of Det. Levine, Trial Tr. at 723-39.
See Trial Tr. at 637, 725, 739, 1169.
See Testimony of Det. Levine, Trial Tr. at 780.
See Testimony of Det. Levine, Trial Tr. at 821-23.
See Testimony of Det. Levine, Trial Tr. at 668, 671.
See Testimony of Det. Levine, Trial Tr. at 708-09.
See Testimony of Det. Levine, Trial Tr. at 663.
See Prosecutor’s Summation, Trial Tr. at 1508.
See Testimony of Det. Levine, Trial Tr. at 710.
13

him of murdering Correa, and Deskovic responded by professing his innocence. 39 As to these
crucial aspects of the interrogation, jurors could rely only on the detective’s after-the-fact incourt recapitulation.
That only snippets of the January 10th interrogation were recorded is particularly
disturbing because of that statement’s importance to the police investigation and, eventually, to
the prosecution’s case as whole. During an unrecorded segment of this interrogation, Deskovic,
supposedly without prompting, drew a map of the area behind the Hillcrest elementary school
that portrayed three discrete crime scenes – 1) the place where he believed the assailant had
accosted Correa; 2) the spot on the path where the rape had occurred; and 3) the location where
Correa’s body had been found. 40 The prosecution placed great weight on the diagram at
Deskovic’s trial, contending both that it was forensically accurate 41 and that the information it
contained had not been released to the public and, thus, could only have been known by the real
killer. 42 Because there was no recorded account of the creation of this pivotal evidence,
Deskovic’s jurors were denied an essential tool for evaluating the prosecution’s case.
Deskovic’s January 25th statement was far and away the most important evidence at the
trial. Without it, the State had no case against him. He would never have been prosecuted for
killing Correa. He would never have been convicted. He would never have spent a day – let
alone 16 years – in prison. Reasonable minds can differ about the credibility of the police
explanations for sporadically recording Deskovic’s January 10th statement. No similar
controversy exists about the police rationale for failure to record any of the January 25th
polygraph examination. The police simply offered none.
Deskovic arrived at police headquarters in Peekskill at 9:30 a.m. on January 25, 1990.
About half an hour later, he left the station with detective McIntyre to make the one hour drive to
Brewster, where the polygraph examination would be administered. They arrived at the testing
site around 11:00 a.m. There, Deskovic met Investigator Stephens, the polygraph examiner, and
was placed in a 10 x10 room, where he would remain for the next eight hours. 43 Stephens then
put Deskovic through an arduous series of interviews, the avowed purpose of which was to elicit
a confession. 44 During part of this period, Deskovic was connected to the polygraph machine.
39

40

41

See Complaint Follow-up Report of Det. Levine, dated 1/27/90.
See Complaint Follow-up Report of Det. Levine, dated 1/27/90.
See Testimony of Det. Paul Astrologo, Trial Tr. 192, 264.

42

See Testimony of Det. McIntyre, Trial Tr. at 1267; Prosecution Summation, Trial Tr. at
1504, 1526.
43

See Complaint Follow-up Report of Det. McIntyre re: 1/25 interrogation (undated);
Testimony of Inv. Stephens, Trial Tr. at 1042.
44

See Testimony of Inv. Stephens, Trial Tr. at 955, 1031,1034.

14

After the examination was complete, Stephens told Deskovic that he had failed it. 45 Thereafter,
Stephens left and was replaced by McIntyre. 46 The interrogation continued. By its end,
Deskovic was lying under a desk in a fetal position, sobbing uncontrollably. 47 He had confessed
to murdering Correa. 48
Nothing whatsoever prevented the Peekskill detectives from recording the January 25th
session in its entirety. The polygraph room was equipped with a monitoring device, which
allowed others to listen in an adjoining room. 49 Detectives Levine and McIntyre and Lieutenant
Tumulo were present in that room and heard everything through the intercom. 50 There is no
indication that the acoustics were anything less than acceptable. The setting thus seems uniquely
suited to tape recording. Such a recording would have allowed jurors to hear for themselves
precisely what Deskovic said and to comprehend fully the circumstances surrounding his saying
it. No recording was made, however, because, as Levine and McIntyre testified at trial, they
simply left their recorders behind when they left for Brewster. 51 No additional explanation was
offered, either by the police or the prosecution.
It is, of course, impossible to say whether jurors would have acquitted Deskovic had they
heard, first hand, what occurred during those eight hours in Brewster. What is certain, however,
is that in the absence of a taped record, they were treated only to an incomplete, self-serving
account of events. The unexcused and inexcusable failure to provide jurors with this essential
tool looms large in explaining Deskovic’s wrongful conviction.
Troubling Police Tactics In Dealing with Deskovic
It is not difficult to understand why the police were interested in talking to Deskovic in
the wake of Correa’s murder. He had behaved oddly at Correa’s wake and his conduct in
embarking on an independent investigation, complete with proposed lines of inquiries and
45

46

See Testimony of Inv. Stephens, Huntley Hearing Tr. at 271; Trial Tr. at 1028.
See Testimony of Inv. Stephens, Trial Tr. at 1092.

47

See Testimony of Det. Levine, Trial Tr. at 766; Testimony of Det. McIntyre, Trial Tr. at
1188, 1262; Testimony of Inv. Stephens, Huntley Hearing Tr. at 291.
48

See Testimony of Det. McIntyre, Trial Tr. at 1184-88; Complaint Follow-up Report of
Det. McIntyre re: 1/25 interrogation (undated).
49

50

See Testimony of Inv. Stephens, Trial Tr. at 950.
See Testimony of Det. Levine, Trial Tr. at 762; Testimony of Det. McIntyre, Trial Tr. at

1178.
51

See Testimony of Det. Levine, Trial Tr. at 760; Testimony of Det. McIntyre, Trial Tr. at

1255.

15

unnamed sources, was suspicious. As the police theorized, Deskovic’s actions could have been
those of a killer inserting himself into the investigation and working up the gumption to confess.
By the same token, Deskovic’s actions could have been – and, indeed, apparently were – those of
a troubled sixteen year old, who having experienced the death of a peer for the first time, became
obsessed with it and naively believed that the detectives, who fed him pizza and spent time with
him discussing the case, would invite him into their investigation if he could prove himself
worthy. Had the detective’s initial theory been correct, then their tactics would likely have been
credited with snaring a killer. Because their gut reaction proved tragically wrong, however,
those same tactics produced a false confession and procured a wrongful conviction.
It is important to acknowledge that, as the court found following the Huntley hearing, the
police tactics in dealing with Deskovic were lawful. 52 Investigator Stephens described the
passive/active techniques that police employ when interrogating a suspect they believe to be
guilty. Passive interrogation involves low-key non-stressful questioning. If such inquiry bears
no fruit – if it does not elicit the desired admission – then the questioning shifts to active
interrogation, which is more aggressive, confrontational and accusatory. The shift from passive
to active interrogation sometimes involves a change in the questioner. 53
The police employed these methods in their interrogations of Deskovic from December
12, 1989, until January 25, 1990. McIntyre acknowledged that he considered Deskovic a suspect
on December 12th and it is evident from the record that police thereafter did little to search for
other would-be killers. 54 When Deskovic refused to implicate himself on December 12th,
McIntyre shifted to active interrogation, becoming more forceful, expressing his disbelief in
Deskovic’s account and opining that Deskovic had something to do with Correa’s death. 55 On
that date, however, Deskovic held firm. The pattern repeated itself during Levine’s January 10th
interrogation. Again the conversation began in a low-key way with Levine even going to fetch
coffee. 56 Deskovic maintained his innocence, however, and, at some point, Levine became
confrontational, accusing Deskovic of being responsible for Correa’s death. 57 Again, Deskovic
remained steadfast. Finally, on January 25th, after Stephens’s low-key questioning failed to elicit
the desired confession, McIntyre took over the interrogation and, following one final
confrontation, Deskovic confessed. 58
52

53

54

55

56

57

See Decision on Motion to Suppress Statements (Colabella, J.).
See Testimony of Inv. Stephens, Huntley Hearing Tr. at 288-91, Trial Tr. at 1046-50.
See Testimony of Det. McIntyre, Trial Tr. at 1231.
See Testimony of Det. McIntyre, Trial Tr. at 1159-60, 1209.
See Testimony of Det. Levine, Trial Tr. at 668.
See Testimony of Det. Levine, Trial Tr. at 710, 849.

58

See Testimony of Inv. Stephens, Huntley Hearing Tr. at 291-92; Trial Tr. at 1044, 1092;
Testimony of Det. McIntyre, Huntley Hearing Tr. at 446-47; Trial Tr. at 1178-84.

16

There were warning signs. Deskovic was a sixteen-year-old boy, who had no prior
experience with the criminal justice system. Through their early conversations with him and
their interviews of others, the police learned further that Deskovic might well have serious
psychological difficulties, making him even less capable of fending for himself. 59 Moreover,
the detectives knew that Deskovic was speaking to them without the permission and against the
wishes of his mother, but they proceeded anyway. 60 Indeed, they scheduled the polygraph for a
school day, making it unlikely that Deskovic’s family would realize that he was missing. In
addition, police were aware that, following their first conversation with Deskovic, the family had
attempted unsuccessfully to retain counsel to represent him. 61
Further, in their dealings with him, the police encouraged Deskovic’s naive belief that
they would accept him as a partner in their investigation. They reviewed his proposed
investigative questions with him. They urged him to sketch his impressions of the crime scene
and of Correa’s route on the day she disappeared. They allowed him to accompany them to the
area behind Hillcrest elementary school. They ate pizza with him. 62 Even as he sat strapped to
the polygraph, Deskovic explained that he was taking the test because, if he passed, he believed
he would be permitted to participate fully in the police investigation of Correa’s death. 63
Carelessness or Misconduct in the Police Investigation
Lacking any physical evidence or eyewitness testimony connecting Deskovic to the crime
and forced to contend with scientific evidence that appeared to exonerate him, the prosecution
needed something to corroborate the January 25th confession. It found what it was seeking in
information gleaned from Deskovic’s statements that the prosecution insisted could only have
been known to the real perpetrator. Deskovic must be the killer, the prosecution urged, because
he had information about the crime available to no one else.

59

See Complaint Follow-up Report of Det. McIntyre re: Deskovic’s conduct at Correa’s
wake and funeral (undated); Complaint Follow-up Report of Det. Brovarski re: interview with
mother of John Larino (undated).
60

See Testimony of Det. McIntyre, Huntley Hearing Tr. at 481, 482; Trial Tr. at 1176-77.

61

See Complaint Follow-up Report of Det. McIntyre re: Lt. Tumulo’s conversation with
Atty. Louis Echer (undated). Plainly, had Deskovic’s family been able to retain counsel or had
they been referred to an agency that could provide them legal services at no cost, then this entire
tragedy could have been averted.
62

Complaint Follow-up Report of Det. Levine, dated ½7/90; Complaint Follow-up Report
of Det. McIntyre re: 1/10 crime scene visit (undated); Testimony of Det. Levine, Trial Tr. at 721;
Testimony of Det. McIntyre, Trial Tr. at 1161-69.
63

See Testimony of Inv. Stephens, Huntley Hearing Tr. at 279, Trial Tr. at 1063.

17

We now know that Deskovic had nothing to do with Correa’s death. Therefore, two
scenarios are possible. First, the police, perhaps inadvertently, may have fed information
directly to Deskovic. This is a possibility, but it is one for which no supporting evidence
presently exists. Second, the information that the prosecution claimed only the killer could know
may, despite the investigator’s efforts, have found its way into the community at large. Either
way, Deskovic’s jurors were misled.
At trial, the prosecution took great pains to suggest that the confidentiality of the police
investigation had been strictly maintained. Sergeant O’Buck and Detective Astrologo, who were
involved in the search of crime scene, both testified that they had disclosed nothing about their
observations to members of the media. 64 Detective Levine made the same point. 65 Detective
McIntyre insisted that he provided no information to any of the people that he interviewed during
the course of the investigation. 66 He added that the information about the three discrete crime
scenes had appeared in no media accounts of the case. 67 The prosecution even introduced the
crime scene diagram that had appeared in the local paper following Correa’s death to show that it
lacked the detail of Deskovic’s sketch. 68
The prosecutor employed this testimony in making powerful arguments against Deskovic
at trial. When Deskovic met with Levine on January 10th to discuss his proposed areas of
investigation, he made reference to a note that Angela had written to Freddy Claxton shortly
before her death. 69 Indeed, remnants of this note were recovered from the area beneath Correa’s
body. 70 In both his opening and closing arguments, the prosecutor suggested that Deskovic’s
knowledge of Correa’s note to Claxton evidenced Deskovic’s guilt. 71 Similarly, regarding

64

See Testimony of Sgt. O’Buck, Trial Tr. at 132; Testimony of Det. Astrologo, Trial Tr.

at 301.
65

66

67

See Testimony of Det. Levine, Trial Tr. at 598.
See Testimony of Det. McIntyre, Trial Tr. at 1265.
See Testimony of Det. McIntyre, Trial Tr. at 1267.

68

See Testimony of Det. Levine, Trial Tr. at 777; Testimony of Det. McIntyre, Trial Tr. at
1267-70.
69

See Testimony of Det. Levine, Trial Tr. at 653, 902.

70

See Testimony of Officer Ubben, Trial Tr. at 174-76; Testimony of Det. McIntyre, Trial
Tr. at 1116, 1137.
71

See Prosecutor’s Opening Statement, Trial Tr. at 27; Prosecutor’s Summation, Trial Tr.

at 1517.

18

Deskovic’s crime scene sketch, the prosecutor urged jurors to conclude that Deskovic’s
knowledge of the three discrete sites proved he was the killer. 72
We cannot now answer the question of how Deskovic acquired this knowledge. We
know only that his knowledge was not the knowledge of a killer. Perhaps, the police conveyed
the information to him directly, although the defense did not appear to allege this at trial and, as
far as we know, this charge has not been made elsewhere, nor is there any proof of it. More
likely, Deskovic learned what he learned because information about the crime had been
disseminated widely throughout the Peekskill community. This was, after all, a terrible, highprofile crime and the public must have been keenly interested in seeing it solved. Officials
acknowledged that the crime scene had not been entirely cordoned off from public view at the
time it was processed. 73 It was reopened entirely after only six hours. 74 Deskovic had read
media accounts of the case and acknowledged visiting the scene the day after police found
Correa as he commenced his own “investigation.” He returned to it several times after that. 75 In
all likelihood, Deskovic learned of the “three crime scenes” from people who watched the crime
scene technicians at work on November 17th or from his own observations the next day. Despite
their protestations to the contrary, 76 investigators may simply have left evidence of their work
behind.
Similar carelessness also could well explain how Deskovic learned of the Claxton note.
From the start, Peekskill High School was a focus of police attention in this case. In the course
of their investigation, officers spoke with many students, including, of course, Claxton. 77
Anyone who has ever attended high school knows full well how quickly information, both
accurate and not, spreads there. It takes no great speculative leap to conclude that rumors
abounded about Angela and Freddy and a possible note before the police had left the building.
The bottom line is this. If, as the prosecution contended, Deskovic had information about
Correa’s death that was not widely known, flaws in the investigation, in some measure, must be
to blame.

72

73

See Prosecutor’s Summation, Trial Tr. at 1526.
See Testimony of Det. Astrologo, Trial Tr. at 289; Testimony of Det. Levine, Trial Tr. at

798.
74

75

76

See Testimony of Det. Astrologo, Trial Tr. at 286.
See Testimony of Det. McIntyre, Trial Tr. at 1204.
See Testimony of Det. Levine, Trial Tr. at 598.

77

See Complaint Follow-up Report of Det. McIntyre re: 11/20/89 interview of Freddy
Clacton (undated). According to the report, police specifically asked Claxton about Correa’s
note.

19

Prosecution’s Decision to Proceed with the Grand Jury Presentation Before Receiving the DNA
Results
Chief Judge Judith Kaye has written that, “[i]n our State justice system, the critical
functions of investigating criminal activity and protecting citizens from unfounded accusations
are performed by the Grand Jury, whose proceedings are conducted by the prosecutor, beyond
public scrutiny.” People v. Huston, 88 N.Y.2d 400, 401 (1996). It is the historical function of
the Grand Jury “to determine whether sufficient evidence exists to accuse a citizen of a crime.”
Id. at 406.
Following Deskovic’s arrest on January 25, 1990, prosecutors moved quickly to present
the case to the grand jury. Samples from the vaginal swabs taken from Correa’s body and from
Deskovic’s blood had already been forwarded to the FBI laboratory for DNA comparison, but
the test results had not yet been received. 78 The Grand Jury voted an indictment without hearing
the DNA results and that indictment was filed on February 27, 1990. 79 Only three days later, on
March 2, 1990, the District Attorney’s office received word that the semen found in Correa’s
body had not come from Deskovic. 80 Deskovic was arraigned on the indictment two weeks
later, on March 15, 1990. 81 Thereafter, the defense moved unsuccessfully to dismiss the
indictment, urging that the exculpatory DNA evidence should have been presented to the Grand
Jury. 82
As the trial court found, the District Attorney’s Office acted in accordance with the law in
obtaining the indictment as it did. The DNA test results had yet to be received at the time of the
Grand Jury presentation or the filing of the indictment. Even if it had been, it is far from clear
that the failure to present this evidence would, as a legal matter, have required dismissal of the
indictment.
Minimum legal requirements aside, the District Attorney’s Office exercised questionable
judgment in proceeding as it did. By the time of the Grand Jury presentation, the DNA samples
had already been forwarded to the FBI. There is no indication in the record that the prosecution
took any steps to ascertain whether the testing could be expedited under the circumstances. Nor
does it appear that the District Attorney explored the possibility of briefly delaying the Grand
Jury presentation until after the test results had been received. Given the stakes, it is highly
unlikely that the defense would have opposed such a request had one been made.
78

79

See Testimony of FBI Special Agent Deadman, Trial Tr. at 415.
See People’s Affidavit in Opposition to Defense Omnibus Motion at 3.

80

See People’s Affidavit in Opposition to Defense Omnibus Motion at 3; Testimony of
Special Agent Deadman, Trial Tr. at 418-19.
81

82

See People’s Affidavit in Opposition to Defense Omnibus Motion at 3.
See Defense Omnibus Motion at 3-4.

20

As it was, the Grand Jury was denied the opportunity to consider critical evidence
in the case. The District Attorney was not legally required to defer the presentation until
the receipt of this evidence. A brief delay, however, would have provided the grand
jurors with a far more complete picture. Had the Deskovic’s DNA matched the DNA on
the vaginal swab, the prosecution would have been strengthened immeasurably. If, as
turned out to be the case, there was no match, then grand jurors could have determined
“whether sufficient evidence exist[ed] to accuse a citizen of a crime.” Huston, 88 N.Y.2d
at 406. By declining to wait, the prosecution deprived the Grand Jury of the
opportunity to fulfill this historic function.
Prosecution’s Questionable Presentation of the Scientific Evidence
Towards the end of his opening statement, the prosecutor confessed to
jurors that his case had some “twists” and “wrinkles.” 83 Specifically, the prosecutor
acknowledged that seminal fluid found in Angela Correa’s body after her death did not
come from Deskovic.84 Neither was Deskovic the source of hairs found on Correa
during the autopsy. 85 These “wrinkles” posed considerable difficulties. To
accommodate them, the prosecution advanced shifting and inconsistent hypotheses,
which either contradicted its core theory of the case, were unsupported by any record
evidence or both. The resulting confusion undoubtedly contributed to Deskovic’s
conviction.
In late August of 1990 as the trial date neared, the prosecution, in an effort to
explain the DNA evidence, first advanced the theory that Deskovic might not have acted
alone. 86 Indeed, in his opening statement, the prosecutor urged jurors to consider
whether the proof would establish “more than one person’s involvement in committing
one or more of these various crimes.” 87 The accomplice hypothesis flew in the face of
the prosecution’s core theory of the case: that Deskovic was a maladjusted loner who
harbored a burgeoning private obsession with Correa. That obsession, the theory went,
had prompted Deskovic to follow Correa to the area behind Hillcrest Elementary School,
where he confronted her, fought with her, raped her and killed her. This tale of lonely
and ultimately violent fixation did not easily accommodate a partner in crime. The
purported evidence supporting the accomplice theory 88 – Deskovic’s use of the third
83

84

85

86

87

88

See Prosecutor’s Opening Statement, Trial Tr. at 46.
See Prosecutor’s Opening Statement, Trial Tr. at 47.
See Prosecutor’s Opening Statement, Trial Tr. at 47-48.
See Huntley Hearing Tr. at 11.
See Prosecutor’s Opening Statement, Trial Tr. at 49.
See Prosecutor’s Opening Statement at 48-49.

21

person at some points in his January 25th statement – was woefully deficient, as
evidenced by the fact that the prosecution simultaneously contended that these third
person pronouns represented Deskovic’s inability or unwillingness to take responsibility
for what he had done. By the close of the evidence, the court opined that there was no
record support for the existence of an accomplice 89 and the prosecutor withdrew his
request to argue the accomplice theory in summation. 90 Nevertheless, this argument had
already been placed before the jury. Although it was bereft of evidentiary support, it may
have contributed unfairly to Deskovic’s conviction.
The prosecution’s alternative explanation for the presence of another man’s
semen was equally strained. In his opening, the prosecutor pointed out that seminal fluid
could persist in the vagina for “upwards of seventy-two hours and longer.” 91 Indeed, the
prosecution serologist testified at trial that this was a possible, if not a particularly likely
scenario. 92 Because sperm was detected on the vaginal swab taken from Correa
approximately 48 hours after her death, 93 the prosecution hypothesized that its source
might not be the rapist at all, but a consensual partner who had deposited it at some
earlier time. 94 In his summation, the prosecutor speculated that Correa’s partner was “in
all probability” Freddy Claxton, to whom Correa had addressed her final note. 95
The problem with this prosecution theory was that it, too, was utterly unproven.
The prosecution had traced Correa’s movements from the time she left school on
November 15th to the time she arrived at the scene of the crime. Based on that time line,
there appeared to be no opportunity for Correa to have engaged in sexual relations that
afternoon. In the preceding hours, Correa had attended school, again making sexual
activity unlikely. While, under the prosecution’s scenario, it was still possible for Correa
to have had consensual sex that morning or the night before, the scientific plausibility of
the theory diminished as the clock was moved backwards. And, the viability of the
prosecution’s consensual partner theory hinged on the dubious proposition that Correa
had not bathed or washed between the time she had sexual relations and her death.
89

90

91

See Trial Tr. at 1134.
See Trial Tr. at 1391.
See Prosecutor’s Opening Statement, Trial Tr. at 48.

92

See Testimony of Linda Duffy, Trial Tr. at 326, 350 (noting that sperm usually
survived approximately 24 hours in the vagina, but could survive longer after the
woman’s death).
93

94

95

See Testimony of Linda Duffy, Trial Tr. at 332.
See Prosecutor’s Summation, Trial Tr. at 1492-93.
See Prosecutor’s Summation, Trial Tr. at 1492-93.

22

More fundamentally, however, there was simply no evidence that, at the time of
her death, Correa was involved in a consensual sexual relationship with anyone. While to
the jury’s less-than-fully-informed ears, the summation speculation that Freddy Claxton
could have been her sexual partner may have sounded reasonable, the prosecutor should
have known better. For he was aware, as the jury was not, that Claxton had a verified
alibi for the entire afternoon of November 15th, from the time of school dismissal until
well after Correa’s death. During that time, Claxton was playing basketball with four
friends, not having sex with Correa. 96 More, the police had interviewed Claxton, scores
of other Peekskill students as well as members of Correa’s family and had received no
indication that Correa was sexually involved with Claxton or anyone else.
In the final analysis, the prosecution’s “consensual partner” theory was both
scientifically dubious and unsupported by the evidence. Like the “unapprehended
accomplice” scenario, it could only have served to confuse the jury.
The prosecution’s tactics in dealing with the apparently exculpatory hair evidence
were also unacceptable. As the prosecution’s own expert testified, none of the hairs
recovered from Correa’s body could be matched to Deskovic. 97 In addition, hairs were
found that matched neither Deskovic nor Correa. One of these hairs was a “negroid hair”
– the source of which was an African-American – found on Correa’s right foot. 98 To
accommodate this piece of evidence, the prosecutor elicited testimony that the Medical
Examiner’s assistant was an African-American man 99 and urged jurors to presume that
the hair found on Correa’s body must have belonged to him. 100 This may or may not
have been true, but the prosecution should not have left the matter to speculation. Rather,
it should have sought, and doubtlessly could have obtained, a sample from the assistant
for comparison. Indeed, the prosecution’s own expert testified that he had made this
precise suggestion, but that his request had gone unheeded. 101
There is no excuse for the prosecution’s failure to follow the advice of its expert.
Had a sample been tested and been strongly associated with the hair of the Medical
Examiner’s assistant, then the exculpatory value of the hair evidence would have been
96

See Complaint Follow-up Report of Det. McIntyre re: 11/20/89 interview of
Freddy Claxton (undated).
97

98

99

See Testimony of Peter DeForest, Trial Tr. at 359-83.
See Testimony of Peter DeForest, Trial Tr. at 375.
See Testimony of Det. McIntyre, Trial Tr. at 1114.

100

101

See Prosecutor’s Summation, Trial Tr. at 1487-88.
See Testimony of Peter DeForest, Trial Tr. at 385.

23

largely neutralized. Had the sample been inconsistent the assistant hair, however, then
the outcome of the trial could well have been different. Since, as we now know, Correa’s
assailant was indeed African-American, this possibility cannot be gainsaid.
DEFENSE FAILURES
CONTRIBUTED TO DESKOVIC’S CONVICTION
Defense Failure to Use Evidence of Deskovic’s Psychological Vulnerabilities
Why would a man confess to a crime he did not commit? That was the question
that the defense at Deskovic’s trial had to answer. The scientific evidence strongly
supported Deskovic’s innocence, but his January 25th statement created a significant
barrier to vindication. Lay jurors and, indeed, criminal justice professionals, are simply
loath to accept the possibility that someone would falsely confess to a serious crime,
particularly to a brutal rape and homicide such as this one. If the defense could not
credibly address this issue, then Deskovic would remain at grave risk.
On the record before us, there is no indication that the defense explored the
possibility of presenting expert testimony to assist the jury in understanding how
someone like Deskovic could have been induced to confess falsely. 102 We believe this
was a mistake. Under New York law, expert testimony is admissible when its subject
matter is “beyond the ken of the typical juror.” People v. Taylor, 75 N.Y.2d 277, 288
(1990). Such testimony is most frequently offered to explain matters that otherwise seem
counterintuitive or to dispel common misconceptions. Id. The phenomenon of false
confessions seems to fall comfortably within this definition.
In concluding that the defense should have explored the use of such an expert in
this case, we are mindful that existing decisional law may have complicated such efforts.
In People v. Lea, 144 A.D.2d 863 (1988), the Appellate Division, Third Department had
upheld the exclusion of a proposed defense expert’s testimony that the “defendant’s
personality was such that he was deferential to the wishes and attitudes of others, making
it more likely that the defendant was intimidated by the atmosphere in the interrogation
room.” Id. at 864. No issue had been raised in Lea as to the defendant’s mental state or
psychological condition. The issue of competency also had not arisen. Id.
The defense here could have distinguished Lea from Deskovic’s case. Unlike in
Lea, the record is replete with evidence that Deskovic suffered from psychological
problems. A friend’s mother described him as “emotionally handicapped” and, by his
own account, he was “distraught” about what had happened to Correa, so distraught that
he had attended three out of four sessions of her wake. 103 During the polygraph itself,
102

See Colloquy, Trial Tr. at 30, 42.

103

See Complaint Follow-up Report of Det. McIntyre re: Deskovic’s conduct at
Correa’s wake and funeral (undated); Testimony of Det. McIntyre, Trial Tr. at 1155-56.

24

Deskovic told Stephens that he periodically heard voices. 104 Finally, and again in
contrast to Lea, the court in Deskovic’s case had ordered a 730 examination and
Deskovic spent some time during the pretrial period committed to the Rockland County
Children’s Psychiatric Hospital. 105
On this record, the trial judge may well have exercised his discretion to admit
such testimony had the defense made the appropriate application. Such testimony could
have provided jurors with the explanation they desperately needed to follow the DNA
evidence to its logical exculpatory conclusion. At the very least, by moving to introduce
expert testimony, the defense would have preserved a viable claim in the event an appeal
became necessary.
The Defense was Scattered, Unfocused and Confusing:
Without the guidance an expert could have provided, the defense efforts to attack
the confession were rendered largely ineffectual. At times, the defense seemed to
acknowledge that Deskovic had made the January 25th inculpatory statements, but urged
jurors to find that those statements were false. 106 At other points, the defense flatly
questioned whether Deskovic had made the statements at all, suggesting instead that the
police may have fabricated them. 107 On still other occasions, the defense urged the jury
to exclude the statements as involuntary, whether or not the statements were true. 108 In
one breath, counsel suggested that Deskovic’s statements were product of his access to
rampant high school “rumors” or his own “fertile imagination,” but, in the next, he
speculated that Deskovic might actually have witnessed the crime and gained his
knowledge that way. 109 This scattershot approach was reflected in the court’s lengthy
and confusing charge on voluntariness. 110 While the defense in a criminal case is, of
104

105

See Testimony of Inv. Stephens, Trial Tr. at 1016.
See Case Endorsements, People v. Deskovic.

106

See Defense Summation, Trial Tr. at 1429 (if Deskovic had been “questioned
for another hour” he would have confessed to assassinating JFK)
107

See Defense Summation, Trial Tr. at 1429 (“the day after this statement was
allegedly made, and I say allegedly was made.”); Trial Tr. at 1433-34 (referring to
“statements made in the case by Jeffrey. Again, I use the word statements. I don’t use
the word confessions.”)
108

See Defense Summation, Trial Tr. at 1428 (“isn’t it interesting that [the January
25 ] statement was obtained under those circumstances.”); Trial Tr. at 1434-35 (if jurors
determine that the “statements were not voluntarily given, then you must completely
disregard them” even if that “may be offensive to some of you”)
th

109

See Defense Summation, Trial Tr. at 1430-31.

25

course, free to advance alternative or inconsistent theories, we believe that Deskovic
would have been served better by a more focused attack on the only damning evidence
against him.
Defense Failure to Maximize the Exculpatory Value of the Scientific Evidence
From the defense perspective, the most important evidence in this case was the
DNA profile that scientifically excluded Deskovic as a source of the semen present on the
vaginal swabs taken from the victim. Nevertheless, when Dr. Harold Deadman, a DNA
expert from the Federal Bureau of Investigation and Linda Duffy, a forensic biologist
from the Westchester County Department of Laboratories and Research, took the stand to
testify on this pivotal matter, defense counsel did virtually nothing to ensure that the jury
understood the centrality of this evidence. His questioning of Deadman was brief and
perfunctory and he did not examine Duffy at all. As a consequence, much of the
potential value of this evidence to the defense may well have been lost.
On the prosecutor's direct examination, Dr. Deadman testified that the semen
recovered from Correa did not originate from Deskovic. While defense counsel had
Deadman reiterate this conclusion during his brief cross, he otherwise focused on
seemingly irrelevant issues such as the “actual chemical substance” of DNA and the
ability to distinguish DNA samples drawn from identical twins. Counsel failed
meaningfully to examine Deadman about procedures existing in his laboratory to ensure
the accuracy of his testing procedures. Such testimony would have served to enhance the
reliability of the favorable results. Nor did counsel have Deadman stress the uniqueness
of human DNA or the absolute nature of Deskovic’s exclusion. Finally, counsel never
elicited from Deadman that the sample he tested contained genetic material from only
one male donor. Particularly back in 1990, when this case was tried, jurors undoubtedly
needed assistance in assessing the probative value of DNA evidence. Through his
superficial examination of Deadman, defense counsel squandered this opportunity.
Counsel also failed to derive full benefit from the testimony of forensic biologist
Duffy. Duffy testified on direct that the presence of intact sperm on a vaginal swab
generally means that intercourse has occurred within 24 hours of the swabbing. While
this time may be extended somewhat in a dead body, the 24 hour limitation still held true
in the majority of cases. Duffy also noted that no semen at all was found on the victim’s
underwear.
Duffy’s testimony in this regard could have been used to undermine significantly
the prosecution claim that Correa had engaged in consensual sexual intercourse at some
point prior to the murder. Yet counsel did nothing to emphasize this evidence when his
turn came to examine Duffy. Indeed, he asked her no questions at all. In this way as
well, counsel forfeited a potentially valuable opportunity to undermine the prosecution
case, thereby contributing to Deskovic’s wrongful conviction.
110

See Court’s Jury Instructions, Trial Tr. at 1580-1601.

26

Defense Conflict of Interest in the Representation of Freddy Claxton
Early in their investigation, the police questioned Freddy Claxton, who like
Deskovic was a student at Peekskill High School. Detectives asked Claxton about his
relationship with Correa and about his whereabouts at the time of the crime. 111 Two
days later, the Legal Aid Society of Westchester advised police that it represented
Claxton and that he should not be questioned again without his attorney. 112 There is no
record indication of any further communication between the police and Claxton.
After Deskovic’s arrest, Legal Aid was assigned to represent him as well. Around
six weeks later, on March 8, 1990, and on numerous occasions thereafter, the prosecution
raised concerns about Legal Aid’s representation of both Deskovic and Claxton. At the
prosecution’s urging, the court raised the issue of a potential conflict with Deskovic and
defense counsel several times during this early part of the case. Deskovic consistently
maintained that he wished to continue with Legal Aid counsel. Deskovic’s mother also
acknowledged that she understood the court’s explanation of the issue and agreed to
Legal Aid’s retention on the case.
At the start of the Huntley hearing, the court, referencing these earlier
conversations, told Deskovic that, based on Legal Aid’s representation of Claxton,
“maybe there was a conflict of interest and maybe there wasn’t.” 113 The court wanted to
ensure that Deskovic was aware of this and that he remained “comfortable with [his]
lawyer.” 114 Both Deskovic and his mother reaffirmed their desire to continue with Legal
Aid as counsel. 115
In fact, the dual representation of Claxton and Deskovic presented a troubling
issue for Legal Aid, which, we believe, required further exploration. An attorney’s (or
law firm’s) previous representation of someone who had been suspect or might be a
witness in the defendant’s case always raises a potential conflict of interest. People v.
Brown, 235 A.D.2d 563, 564-65 (3d Dept. 1997). Here, the risk was particularly acute
because Legal Aid not only had represented Claxton, but they represented him in
connection with this very case. Further, it was by no means clear that this representation
111

See Complaint Follow-up Report of Det Mcintyre re: 11/20/89 interview with
Freddy Claxton (undated).
112

See Complaint Follow-up Report of Det. McIntyre re:11/22/89 communication
with the Legal Aid Society.
113

114

115

See Huntley Hearing Tr. at 26-27.
See Huntley Hearing Tr. at 27.
See Huntley Hearing Tr. at 27-28.

27

had terminated by the time of Deskovic’s trial. On this basis alone, all of the principals –
the defense, the prosecution and the court – should have been particularly cautious.
To make matters worse, although Claxton did not testify, he was still a significant
player in the prosecution’s case. First, according to the prosecution’s theory, Claxton
was the intended recipient of the note found under Correa’s body at the time of her death.
In both his opening and closing statements, the prosecutor urged jurors to infer that
Deskovic’s apparent knowledge of the note’s existence necessarily constituted guilty
knowledge. 116 Claxton, whom police had questioned about the note, thus possessed
information material to the case: if he had told Deskovic or others at Peekskill High about
the note’s existence, then that fact would have tended to exculpate Deskovic by providing
an innocent explanation for his knowledge of the note. Counsel’s relationship to Claxton
may well have hampered the defense investigation of this issue and at the very least
supports the conclusion that the defense also failed Deskovic.
In addition, the prosecutor expressly argued in his summation that Claxton had
been sexually involved with Correa at the time of her death and was the probable source
of the semen found in her body. 117 Unbeknownst to the jury, however, Claxton’s
statements to the police during their investigation strongly suggested that, had he been
called to testify, Claxton would have denied this. 118 This denial would have lent
significant support to Deskovic’s defense. If the prosecution failed to come up with a
consensual source for the semen, then the defense argument that Correa’s rapist and killer
was the source would have been strengthened considerably. Once again, counsel’s
attorney-client relationship to Claxton may well have prevented full exploration of this
issue in Deskovic’s behalf.
In fairness, we should note that Claxton’s centrality to the case may not have been
clear to either the defense or the court at the outset. At the start of the trial, the defense
did not know how the prosecution would deal with the DNA evidence and the
prosecution steadfastly refused to illuminate the issue. Only in summation did the
prosecution finally settle on Claxton as the likely source of the DNA.
Under New York law, it is incumbent upon each of the parties to bring to light
what it knows about potential conflicts of interest as early as possible. People v. Smart,
96 N.Y.2d 793, 796 (2001). On the record presented, it appears that both parties may
have fallen short on this issue. Although the prosecution laudably and repeatedly brought
the conflict issue to the fore, it could have been more forthcoming about Claxton’s
116

See Prosecutor’s Opening Statement, Trial Tr. at 27; Prosecutor’s Summation,
Trial Tr. at 1517.
117

See Prosecutor’s Summation, Trial Tr. at 1492-93.

118

See Complaint Follow-up Report of Det. McIntyre re: 11/20/89 interview of
Freddy Claxton.

28

anticipated role in the State’s case. In addition, it is a court’s obligation to undertake an
inquiry sufficient to dispose of the conflict problem. Id. at 795-96. Here, the trial court,
in its effort to tread lightly, may also have fallen short. Indeed, it would have been more
prudent for the court to assign Deskovic independent counsel for the purpose of resolving
the potential conflict.
As it was, the problem of the defense representation of Freddy Claxton never
received the careful treatment it warranted. Consequently, we believe that it impeded the
fairness of Deskovic’s trial and was a significant failure by the defense.
OTHER ERRORS
Court’s Mid-trial Loss of Evidence
Shortly before summations, the parties realized that four prosecution exhibits, Ex.
37, Correa’s shoes; Ex. 35, her jeans, Ex. 36, a paper bag containing her underwear and
Ex. 31, her white bra, had been lost. 119 The prosecutor, who made the relevant record,
explained that, during the trial, the exhibits were stored in the courtroom in a black
plastic garbage bag either near the bench, the stenographer’s chair or the witness box. 120
At some point during the Thanksgiving recess, the garbage bag, as well as the exhibits it
contained, was inadvertently discarded by a substitute cleaning crew. 121 Counsel moved
for a mistrial, for preclusion of all evidence relating to the lost exhibits or for an adverse
inference charge. 122 The court denied those applications. 123
At the outset, it is important to emphasize that there is no evidence of bad faith on
anyone’s part in connection with the loss of these exhibits. For that reason, the court
appeared on solid legal ground when it denied counsel’s motion for a mistrial and other
relief. That said, there was absolutely no excuse for the negligent failure to safeguard the
evidence in this case. Surely, court personnel should have anticipated the risk that the
cleaning crew would throw out a black garbage bag left unsecured in the courtroom.
That the courtroom doors were themselves locked to the public and other intruders is thus
simply beside the point. If trial evidence is to be stored in the courtroom, it must be
stored in a secure place – either in a locked drawer or cabinet. That plainly was not done
here. Lax procedures such as these would have been unacceptable in a misdemeanor
trial. In a case of this magnitude, they were beyond the pale.
119

120

121

122

123

See Colloquy, Trial Tr. at 1341-61.
See Colloquy, Trial Tr. at 1342, 1344.
See Colloquy, Trial Tr. at 1407-08.
See Colloquy, Trial Tr. at 1359-61, 1393-94.
See Colloquy, Trial Tr. at 1393-94.

29

Thankfully, the lost evidence did not include the DNA samples taken from
Correa’s body. Had that evidence disappeared, the tragedy of this case would have been
made incalculably worse. Even as it was, however, Deskovic was potentially prejudiced.
In his January 25th statement, Deskovic had purportedly asserted that he ripped off
Correa’s bra during the attack. 124 Before the loss of the evidence, defense counsel had
planned to argue that the nature of the bra – a pullover, with no snaps or clips – made this
claim unlikely, thus casting doubt on the veracity of the entire confession. 125 Without
the physical exhibit available to illustrate his claim, counsel evidently elected to forgo the
argument. In his summation, however, the prosecutor went on the offensive, arguing
from a photograph that the physical condition of the bra corroborated the claim regarding
it in Deskovic’s statement. 126 Undoubtedly in an effort to assess the validity of this
assertion, the jury, during deliberations, asked not for the photograph, but for the physical
exhibit itself. 127 Only then did the court explain what had happened. 128 It told jurors
that it could not comply with their request.
As counsel pointed out in renewing his mistrial bid, 129 the bra had become
important evidence in the case because the jury asked to see it during their deliberations.
Jurors asked for it, presumably, because it was relevant to their determination of the
matter before them. Because the inexcusable loss of the evidence prevented the court
from offering jurors a meaningful response to their inquiry, it adversely affected the
fairness of Deskovic’s trial.
CORRECTIVE ACTION
INTRODUCTION
It is obvious that an enormous and horrific injustice was imposed upon Jeffrey
Mark Deskovic by the State of New York.
Detectives testified at trial that Deskovic confessed to committing this crime. But
the crime scene DNA evidence introduced at trial proved that the semen recovered from
the victim's body was not his. You would expect that DNA would trump a "confession.”
124

125

126

127

128

129

See Testimony of Det. McIntyre, Trial Tr. at 1185.
See Colloquy, Trial Tr. at 1398-1403.
See Prosecutor’s Summation, Trial Tr. at 1511-12.
See Colloquy, Trial Tr. at 1677.
See Trial Tr. at 1690-92.
See Colloquy, Trial Tr. at 1680-87.

30

But it didn't. The jury gave more weight to an immature and distraught youngster's
unrecorded "confession" than to the DNA and other forensic evidence (e.g., microscopic
hair comparison). The juror's gave credence to the prosecutor’s theory – unsupported by
any evidence – that the semen found in the victim's body was from a consensual sexual
relationship with someone else.
As previously discussed, for some time after the conviction, the then-DA refused
to run the unknown DNA profile through the ever expanding State and Federal DNA
Databases. No New York statute offered a remedy to a wrongfully convicted defendant
who was seeking a Database comparison. Courts continually refused to order the DA to
act. In June 2006, Janet DiFiore, the newly elected Westchester County District
Attorney, met with Barry Scheck of the Innocence Project and promptly agreed to have
the Crime Lab conduct the more sophisticated STR analysis on the semen found on the
victim and then to seek a match in the CODIS databases for the unknown crime scene
DNA profile.
There was a cold hit in November 2006 – almost 17 years after Deskovic was
wrongfully arrested. District Attorney DiFiore consented to his immediate release when
the Crime Lab reported that the unknown DNA profile matched the DNA profile of a
man who was serving a life sentence for an unrelated murder committed in Westchester
County only a few years after the 15-year-old high school student was raped and killed.
This man has confessed to killing the young girl Deskovic was convicted of killing. It
cannot be said enough times that Janet DiFiore made the command decision that ended
Deskovic's nightmare.
On November 2, 2006, the Westchester First Deputy DA told the Court that the
People were moving to dismiss the indictment because "Deskovic is actually innocent."
Unwritten in the prosecution motion, he added, were "the most sincere apologies we can
muster on behalf of the Westchester County District Attorney's Office and the Peekskill
police." Deskovic said he appreciated that apology, the first public one he has received.
The judge commended the work of Nina Morrison and Barry Scheck and the
other lawyers and students at The Innocence Project who pushed for Deskovic's release.
He praised the open-mindedness of Westchester prosecutors, who agreed to have the
evidence retested so that a match could be obtained. "Mr. Deskovic, despite my
recognition of these laudable efforts undertaken on your behalf, I must also admit the
undeniable fact that nothing can be done in this courtroom here today to erase the pain
and suffering endured by you and your loved ones over the past 16 years," the judge said.
1.

THE UNIDENTIFIED DNA PROFILE: According appropriate defendants basic
rights

During the very long post-conviction period, Deskovic made a number of requests
to the then DA that the extracted DNA profile be run through CODIS. These requests
were continually denied until late in 2006 when the new Westchester County DA – Janet
DiFiore -- ordered that the DNA Database be searched for a match. This not only

31

exonerated Deskovic but equally important the real rapist-killer was identified and has
been brought to justice.
Had this been done a few years earlier, Deskovic’s wrongful imprisonment would
not have lasted 16 years. And the real rapist-killer would have been identified and
already prosecuted. Why did the former DA consistently reject defendant's application?
It may be unlikely today that a crime scene DNA profile would not be run through
CODIS. But to avoid another Deskovic case in the future, there is a need for legislation
according a court the authority to direct a CODIS search when the defendant's petition is
non-frivolous.
A defendant -- either pretrial or post-conviction -- should have a right to have an
unidentified DNA profile, whether extracted from crime scene evidence or otherwise, run
through the DNA Databases to see if the real perpetrator or an accomplice can be
identified. The defendant's application should be granted unless the prosecutor can show
(perhaps by a preponderance of the evidence) that it is frivolous and devoid of all merit.
Whenever a DNA profile is extracted from crime scene evidence and the donor of
the DNA is unknown, there is never a valid reason for not seeking DNA samples from
others or running that profile through CODIS particularly if a defendant has been charged
with or convicted of the crime and makes a request that the profile be compared with
other profiles in CODIS or with another person's DNA profile. No DA should be able to
oppose a reasonable DNA Database search request, or a reasonable DNA profile
comparison request.
VIDEOTAPING THE INTERROGATION
The prosecution's proof against Deskovic rested almost exclusively on the
"confession" that he allegedly made during his final police interrogation on January 25,
1990, and on some 16 statements made to detectives on and after December 12, 1989
regarding his "opinions" as to how the crime had occurred. Although there was a wealth
of forensic evidence collected from the crime scene and the victim's body at autopsy, all
of which was subjected to extensive analysis, not a single item was attributed to
Deskovic. For its part, the defense contended that the confession was given by a
vulnerable young man under extreme police pressure, and that the forensic evidence
(particularly from the semen) established that the confession was false and that Deskovic
was innocent of the rape and murder.
The jury was told about the exclusionary results from the FBI Crime Lab's RFLPDNA testing on the semen from the victim's vaginal swabs. At trial, the DA conceded
that Deskovic was not the source of the semen, but argued that the confession nonetheless
established his guilt. Although no further forensic testing was performed to identify the
actual source of the semen, the DA contended that it was from a prior consensual sexual
partner of the victim.

32

Significantly, the State was not able to identify the actual source of the semen
found in the victim. "Freddy," whom the DA argued was "probably” the source, was
never ordered to give a DNA sample for comparison purposes. (See discussion, Supra, of
conflict problems and defense inadequacies.)
In addition, the prosecution experts excluded Deskovic as the source of the
numerous hairs and fibers found on the victim and on evidence in the surrounding area.
On April 16, 1990, a hair analyst received numerous slides containing hairs and fibers for
trace and transfer evidence analysis. On September 7, 1990, he was also provided with
hair exemplars from Deskovic, including head and pubic hairs. He identified the hair
removed from the victim's right foot as a fragment of "Negroid origin," and the hair
found on the victim's right breast as a "Caucasoid" head hair, for which he could not
exclude the victim as the source.
The expert also examined the hair retrieved from the pubic combing, which he
identified as a non-pubic, growing "Mongoloid" hair. A shed, medium-brown, nongrowing Caucasian pubic hair was also compared with both the victim and Deskovic and
the latter was excluded as its source. Deskovic fingerprints were not found when the
victim's cassette tape, cassette player, “Dear Freddy" note, and the bottles, twigs, and
Gatorade bottle top found at the scene were forensically examined.
The Deskovic case unmistakably demonstrates the desirability of videotaping the
entire interrogation of all persons suspected of involvement in a violent felony. Deskovic
was questioned by detectives at length, on multiple occasions and over many days. On a
very few occasions a voice tape recorder was used, as discussed at length, previously.
The two most inculpatory statements in the case -- the portion, of the January 10
statement during which Deskovic drew his diagram of the three crime scenes and the
January 25 polygraph exam -- were not recorded, although the opportunity to record
those statements plainly existed. That other statements were recorded, in whole or in
part, makes the failure to record the key statements appear deliberate or tactical. By
turning the recorder on-and-off, the interrogation process suggests selectivity and
opportunity to record only incriminating statements.
It is a generally unrecognized phenomenon that innocent people have confessed to
crimes that they did not commit. Videotaping all interrogations of violent felony suspects
not only may improve interrogation techniques but, more importantly, will aid the jury or
other fact finder to identify a false confession.
Videotaping custodial interrogations is not only feasible, it is being used in an
increasing number of jurisdictions as a way to protect the innocent and ensure the
conviction of the guilty.
Thanks to the many DNA exonerations over the last decade, the problem of false
confessions by the innocent is now well documented. In fact, in approximately 24% of
the 180 post-conviction DNA exoneration cases to date, the wrongful convictions were

33

based, in large part, on confessions, admissions, or inculpatory statements that DNA
evidence later proved to be false.
Special caution must be exercised when a juvenile is interrogated by the police.
Deskovic was only sixteen years-old at the time of his confession. Social science
research indicates that juveniles, possessing marked developmental differences from their
adult counterparts, are far more susceptible to interrogative suggestibility and thereby to
confess falsely to crimes they did not commit. This conclusion is supported by a range of
research, which reveals that adolescents have difficulty understanding lexical language,
including legal terminology, have a higher susceptibility to negative feedback, present
differences in decision-making, present behaviors more often than adults that are
considered "deceptive" by interrogators, and have more negative responses to situational
risk factors, such as stress, the presence of authority figures, physical custody and
isolation, and confrontation, than adults.
These factors in isolation or taken together place adolescents at a higher risk for
confessing to a crime they did not commit. Videotaping the interrogation is a prudent
way to minimize wrongful convictions founded on false confessions.
COMMISSION OF INQUIRY
The Legislature should create a "Commission of Inquiry" to deal with persons like
Deskovic who are exonerated in New York by post-conviction DNA testing. There is a
growing movement throughout the US, UK, Canada, Australia and elsewhere to create
such commissions. Canada conducts the most exhaustive -- and expensive -- type of
investigation. [The "inquiry" movement does not include the many thousands of
defendants -- the exact number is unknown -- exonerated in the past decade by "postarrest preconviction" DNA testing.]
The purpose of the inquiry, of course, is to learn what went wrong. Was it a
systemic error or an individual's mistake or misconduct? The inquiry group should then
recommend changed procedures or practices -- or legislation -- to prevent a repetition of
the injustice.
Recent recommended changes include, e.g., reform of eyewitness identification
procedures; mandatory recording by video of all interrogations of violent felony suspects;
training investigators and judges on the special vulnerability of the young and the
emotionally or mentally impaired during the interrogation process and on the
phenomenon of a "false confession"; a special charge instructing the jury to view certain
evidence with care and extreme caution, e.g., testimony of a "jailhouse" informant; closer
gate keeping by the trial judge of the admissibility of “questionable science” such as bite
mark testimony; reform by law enforcement of its procedures for collecting, preserving
and retrieving forensic evidence; and enactment of legislation that explicitly declares that
defendants have a right to have crime scene evidence run through the DNA or fingerprint
database to see if the real perpetrator can be identified.

34

An October 8, 2006 editorial in a Westchester County newspaper stated: "Pending
in the Assembly is a bill ... that would establish a commission to investigate confirmed
instances of wrongful conviction and issue reports and recommendations, so mistakes can
be avoided in future cases .... The bill has been pending in the Assembly since February
2005; it is worthy of consideration by the full Legislature."

EVIDENCE COLLECTION AND STORAGE
There is a need for legislation that standardizes the procedures for evidence
collection and storage in New York. In too many cases where a defendant is seeking
post-conviction DNA testing to prove his innocence, the crime scene evidence is simply
lost, misplaced or discarded. In the Deskovic case, fortunately, the biological evidence
was in fact properly preserved by the Westchester County Crime Lab and readily
available for new STR-DNA testing. In almost half the states - but not New York - there
are statutes requiring the preservation of biological evidence. In New York each law
enforcement agency has its own practices. A statute formalizing and making uniform the
way DNA evidence is collected, stored and retrieved would further the cause of justice.

35