Amicus Curiae Brief — Toforest Onesha Johnson, Petitioner v. Alabama

No. 22-7337

In the

Supreme Court of the United States

──────────────────────────

TOFOREST ONESHA JOHNSON,

Petitioner,

v.

STATE OF ALABAMA,

Respondent.

──────────────────────────

On Petition for Writ of Certiorari to

the Alabama Court of Criminal Appeals

──────────────────────────

BRIEF FOR THE INNOCENCE PROJECT

AS AMICUS CURIAE

IN SUPPORT OF PETITIONER

──────────────────────────

Adnan Sultan

THE INNOCENCE

PROJECT, INC.

40 Worth Street, Suite 701

New York, NY 10013

Lindsey C Boney IV

Counsel of Record

Virginia N. Adamson

BRADLEY ARANT BOULT

CUMMINGS LLP

1819 Fifth Avenue North

Birmingham, AL 35203

Phone: (205) 521-8914

lboney@bradley.com

Counsel for Amicus Curiae

May 22, 2023

i

TABLE OF CONTENTS

TABLE OF AUTHORITIES....................................... ii

INTEREST OF AMICUS CURIAE.............................1

INTRODUCTION AND SUMMARY OF

ARGUMENT ...............................................................3

ARGUMENT ...............................................................4

I.

Disclosure of Ellison’s ulterior motive for

testifying would have demolished an

already weak case. ............................................5

II. This case bears numerous hallmarks of a

wrongful conviction. .........................................7

A. The State pursued inconsistent

prosecutorial theories. ...................................7

B. The State relied on an unreliable

“earwitness identification.” ......................... 10

C. The State’s star witness was

incentivized to provide testimony

leading to a conviction. ................................ 16

D. The prosecutors now doubt the

validity of Johnson’s conviction. .................. 18

CONCLUSION .......................................................... 22

ii

TABLE OF AUTHORITIES

Page(s)

Cases

Bankhead v. State,

182 S.W.3d 253 (Mo. Ct. App. 2006) ..................... 9

Baze v. Rees,

553 U.S. 35 (2008) .................................................. 1

Berger v. United States,

295 U.S. 78 (1935) .................................................. 7

Bradshaw v. Stumpf,

545 U.S. 175 (2005) ................................................ 8

Drake v. Kemp,

762 F.2d 1449 (11th Cir. 1985) .............................. 8

Elonis v. United States,

575 U.S. 723 (2015) ................................................ 1

Ford v. Wainwright,

477 U.S. 399 (1986) .............................................. 16

Furman v. Georgia,

408 U.S. 238 (1972) .............................................. 12

Glossip v. Gross,

576 U.S. 863 (2015) ................................................ 1

Imbler v. Pachtman,

424 U.S. 409 (1976) .............................................. 21

Kyles v. Whitley,

514 U.S. 419 (1995) ................................................ 5

iii

Perry v. New Hampshire,

565 U.S. 228 (2012) .............................................. 11

Ramos v. Louisiana,

590 U.S. ----, 140 S. Ct. 1390 (2020) ...................... 7

Smith v. Groose,

205 F.3d 1045 (8th Cir. 2000) ................................ 9

Strickler v. Greene,

527 U.S. 263 (1999) ............................................ 5, 7

Thompson v. Calderon,

120 F.3d 1045 (9th Cir. 1997) ................................ 8

United States v. Bagley,

472 U.S. 667 (1985) ................................................ 5

United States v. Wade,

388 U.S. 218 (1967) .............................................. 11

Young v. United States,

315 U.S. 257 (1942) ........................................ 18, 21

Statutes

18 U.S.C. § 201(c)(2) .................................................. 16

Other Authorities

ALA. R. OF PROF’L CONDUCT 3.4(b) ............................ 16

iv

Beth Shelburne, District Attorney Urges

New Trial for Man on Alabama's

Death Row, WBRC (June 12, 2020,

12:22 PM),

https://www.wbrc.com/2020/06/12/dis

trict-attorney-urges-new-trial-manalabamas-death-row/ ........................................... 20

Christopher Sherrin, Earwitness

Evidence: The Reliability of Voice

Identifications, 52 OSGOODE HALL

L.J. 819 (2015)...................................................... 14

Christopher T. Robertson & D. Alex

Winkelman, Incentives, Lies, and

Disclosure, 20 U. PA. J. CONST. L. 33

(2017) .................................................................... 16

Cindy E. Laub, Lindsey E. Wylie, &

Brian H. Bornstein, Can The Courts

Tell An Ear From An Eye? Legal

Approaches to Voice Identification

Evidence, 37 LAW & PSYCHOL. REV.

119 (2013) ............................................................. 14

Conviction Integrity Units, NAT’L

REGISTRY OF EXONERATIONS, https://

www.law.umich.edu/special/exonerat

ion/Pages/Conviction-IntegrityUnits.aspx (last visited May 3, 2023) .................. 21

Innocence, DEATH PENALTY INFO. CTR.,

https://deathpenaltyinfo.org/policyissues/innocence (last visited May

19, 2023) ................................................................. 1

v

Ex-Louisiana Prosecutor Apologizes for

‘the misery I have caused’ Freed

Inmate, GUARDIAN (Mar. 25, 2015,

4:20 PM),

https://www.theguardian.com/usnews/2015/mar/25/former-louisianaprosecutor-exonerated-death-rowinmate-apologizes ................................................ 18

How Eyewitness Misidentification Can

Send Innocent People to Prison,

INNOCENCE PROJECT,

https://innocenceproject.org/howeyewitness-misidentification-cansend-innocent-people-to-prison/ (last

visited May 18, 2023) ........................................... 11

INNOCENCE PROJECT,

https://innocenceproject.org/exonerat

ions-data/ (last visited May 19, 2023) ................... 2

The Innocence Project, Larry Fuller,

NAT’L REGISTRY OF EXONERATIONS

(before June 2012),

https://www.law.umich.edu/special/e

xoneration/Pages/casedetail.aspx?cas

eid=3224 ............................................................... 14

Jessica K. Swanner & Denise R. Beike,

Incentives Increase the Rate of False

but Not True Secondary Confessions

from Informants with an Allegiance

to a Suspect, 34 LAW & HUM. BEHAV.

418 (2010) ....................................................... 10, 16

vi

Jessica Roth, Informant Witnesses and

the Risk of Wrongful Convictions, 53

AM. CRIM. L. REV. 737 (2016) ................................. 1

Kaitlin Jackson & Samuel Gross,

Tainted Identifications, NAT’L

REGISTRY OF EXONERATIONS (Sept.

22, 2016),

https://www.law.umich.edu/special/e

xoneration/Pages/taintedids.aspx ....................... 12

Maurice Possley, Frank Sealie, NAT’L

REGISTRY OF EXONERATIONS (last

updated Oct. 31, 2016),

https://www.law.umich.edu/special/e

xoneration/Pages/casedetail.aspx?cas

eid=4659 ............................................................... 18

Maurice Possley, James C. Anderson,

NAT’L REGISTRY OF EXONERATIONS

(Feb. 16, 2014),

https://www.law.umich.edu/special/e

xoneration/Pages/casedetail.aspx?cas

eid=4378 ............................................................... 13

Michael Q. English, A Prosecutor’s Use

of Inconsistent Factual Theories of a

Crime in Successive Trials: Zealous

Advocacy or a Due Process

Violation?, 68 FORDHAM L. REV. 525

(1999) ...................................................................... 9

NAT’L RESEARCH COUNCIL, IDENTIFYING

THE CULPRIT: ASSESSING EYEWITNESS

IDENTIFICATION 2 (Nat’l Academies

Press 2014) ........................................................... 13

vii

Radley Balko, An Illusion of Justice:

The Baffling Conviction and Death

Sentence of Toforest Johnson Reveal

a Broken System, WASH. POST (Sept.

5, 2019) ................................................................. 19

Rob Warden, How Mistaken and

Perjured Eyewitness Identification

Testimony Put 46 Innocent

Americans on Death Row, CTR. ON

WRONGFUL CONVICTIONS, NW. L.

SCH. (May 2, 2001),

https://files.deathpenaltyinfo.org/leg

acy/files/pdf/StudyCWC2001.pdf ......................... 12

Sara Conway, A New Era of Eyewitness

Identification Law: Putting

Eyewitness Testimony on Trial, 50

NEW ENG. L. REV. 81 (2015) ................................. 12

Stephanie Denzel, Freddie Lee Gaines,

NAT’L REGISTRY OF EXONERATIONS

(before June 2012)

https://www.law.umich.edu/special/e

xoneration/Pages/casedetail.aspx?cas

eid=3226 ............................................................... 13

Thomas Albright & Jed Rakoff,

Eyewitnesses Aren’t as Reliable as

You Might Think, WASH. POST (Jan.

13, 2015) ............................................................... 12

Tom Jackman, Va. Attorney General

Launches Conviction Integrity Unit

to Identify Wrongful Convictions,

WASH. POST (Jan. 16, 2021) ................................. 21

1

INTEREST OF AMICUS CURIAE1

Wrongful convictions are not an “‘unreal dream’—

[but] instead an undeniable reality.” Jessica Roth,

Informant Witnesses and the Risk of Wrongful

Convictions, 53 AM. CRIM. L. REV. 737, 738 (2016)

(citation omitted); see, e.g., Elonis v. United States,

575 U.S. 723, 742 (2015) (Alito, J., concurring in part

and dissenting in part) (observing that if certain

criminal-procedure protections are not followed “a

defendant may be wrongly convicted”); Baze v. Rees,

553 U.S. 35, 86 (2008) (Stevens, J., concurring in the

judgment) (“[A]bundant evidence accumulated in

recent years has resulted in the exoneration of an

unacceptable number of defendants found guilty of

capital offenses.”); cf. Glossip v. Gross, 576 U.S. 863,

895 (2015) (Scalia, J., concurring) (convictions can be

“unreliable”).2

The Innocence Project is a non-profit organization

and law-school clinic dedicated primarily to providing

pro bono legal services to indigent prisoners whose

actual innocence may be established through postconviction evidence.

Its mission is to free the

staggering number of innocent people who remain

Pursuant to Supreme Court Rule 37, amicus states that no

counsel for any party authored this brief in whole or in part, and

that no entity or person, aside from amicus or its counsel, made

any monetary contribution intended to fund the preparation or

submission of this brief. Counsel for the parties received notice

of the intent to file this brief at least ten days before its filing.

1

2 To date, 192 death-row inmates have been exonerated.

See

Innocence, DEATH PENALTY INFO. CTR., https://death

penaltyinfo.org/policy-issues/innocence (last visited May 19,

2023).

2

incarcerated and to bring reform to the system

responsible for their unjust imprisonment.

For more than thirty years, the Innocence Project

has fulfilled this mission by helping courts identify

the signs of wrongful convictions and prevent future

injustice. It has provided representation or assistance

in most of the 375 DNA exonerations in the United

States, as well as numerous exonerations based on

constitutional violations.3 The Innocence Project also

works to prevent future miscarriages of justice by

researching the root causes of wrongful convictions,

participating as amicus curiae in cases of broader

significance, and pursuing reform initiatives designed

to enhance the truth-seeking function of the criminaljustice system. As a leading national advocate for the

imprisoned, the Innocence Project is dedicated to

improving the criminal-justice system and has a

compelling interest in ensuring the fundamental

dignity of those held in our nation’s prisons.

The Innocence Project files this brief as amicus

curiae to urge the Court to grant certiorari because

this capital case presents numerous warning signs—

far beyond the alleged Brady violation—that strongly

point to Petitioner Toforest Johnson’s innocence.

Since its founding in 1992, the Innocence Project has been

involved in 243 exonerations.

See INNOCENCE PROJECT,

https://innocenceproject.org/exonerations-data/ (last visited May

19, 2023).

3

3

INTRODUCTION AND

SUMMARY OF ARGUMENT

If ever a case bore the hallmarks of a wrongful

conviction, Toforest Johnson’s is it. Inconsistent

prosecutorial theories, a prosecution that depends on

a single “earwitness” identification (who was paid an

undisclosed reward for her testimony, no less), and

later-expressed prosecutorial doubt about the

strength of the case . . . this case checks all the boxes.

Indeed, the signs of wrongful conviction fairly cry

out from the record. Individually, each warrants a

new trial—but together, they cannot be easily swept

away, particularly now that both the lead trial

prosecutor and the current district attorney support

Johnson’s request for a new trial. The Innocence

Project respectfully urges the Court to grant the

petition for certiorari.

* * *

Over the course of grand-jury proceedings and four

trials, the State of Alabama presented at least five

different, contradictory theories of who shot Deputy

Sheriff William Hardy. This prosecutorial tactic is

suspect in general. But here, it virtually ensures that

the State knowingly prosecuted at least one innocent

person—the forensic evidence and witness testimony

made clear that this was a single-shooter crime.

The State’s case against Johnson, who has always

maintained his innocence, depended on the testimony

of Violet Ellison, who claims to have overheard

Johnson confess to the crime on a three-way phone

call. Eye- or ear-witness identifications are generally

unreliable, but here, doubly so: Ellison was not

present when the crime occurred; she claimed to

4

overhear facts that were inconsistent with the

physical evidence, the State’s prosecutorial theories,

and even her own notes; she was paid by the State for

her testimony after she testified, in an undisclosed

payment; and although she identified Johnson based

on overhearing a phone call, she did not know him and

did not know his voice.

Considering these facts, it is unsurprising that the

lead trial prosecutor has expressed doubt about the

strength of the State’s case and that the current

district attorney, following an independent review,

now supports a new trial “in the interest of justice.”4

Each of these aspects of Johnson’s case is a

hallmark of wrongful conviction, a red flag that should

stop reviewing courts in their tracks. All of them

combined in a single case should create unavoidable

doubt in the outcome of the trial.

ARGUMENT

The Court should grant review because the

troubling case against Toforest Johnson is not limited

to a Brady violation—the case was incredibly weak

from the start and is only made weaker in the light of

the State’s suppressed evidence. Johnson correctly

argues that the court below should have found,

consistent with the evidence, that the State

suppressed evidence that Violet Ellison testified for a

reward that she anticipated receiving at the time of

4 See C3. Supp-3 11.

This brief employs the same record citation

conventions as the petition for certiorari. “T.C.” and “T.R.” refer

to the trial record. “C.” and “R.” refer to the record from the

initial post-conviction proceedings. “C1.” and “R1.” refer to the

first remand record. “C2.” and “R2.” refer to the second remand

record. “C3.” and “R3.” refer to the third remand record.

5

her testimony. This Court should correct that error

because the disclosure of the ulterior motive of the

State’s star witness would have shattered the already

weak case against Johnson. Indeed, the conclusion of

the district attorney’s recent review of the case

highlights the materiality of that suppression.

I.

Disclosure of Ellison’s ulterior motive for

testifying would have demolished an

already weak case.

A correct resolution of the Brady claim is especially

important because Johnson’s case bears so many

hallmarks of a wrongful conviction. This will be

significant at the merits stage because if this Court

agrees with Johnson that the State suppressed the

evidence of Ellison’s motive for testifying, the next

step in the Brady analysis—either on remand or in

this Court, see Pet. 23–25—is whether the disclosure

would have made a different result “‘reasonab[ly]

probab[le].’” Strickler v. Greene, 527 U.S. 263, 280

(1999) (quoting United States v. Bagley, 472 U.S. 667,

682 (1985)). This materiality standard is met when

the suppressed evidence would produce a

“significantly weaker case” for the prosecution, based

on the “cumulative effect of the evidence.” Kyles v.

Whitley, 514 U.S. 419, 440–41, 454 (1995).

The Court should grant certiorari here because the

Brady violation described in the petition is just one of

many troubling aspects of the prosecution’s case

against Toforest Johnson.

The discussion below reveals just how weak the

case for conviction was in the first place. Johnson’s

case displays four hallmarks of a wrongful conviction.

All of them should independently undermine

6

confidence in the trial’s outcome, but together—in

combination with the suppressed Brady evidence—

they are devastating to the State’s case:

•

The State’s inconsistent prosecutorial theories

show that no forensic evidence pointed to

Johnson and emphasize that Ellison’s

testimony was essential to his eventual

conviction. See infra Part II.A.

•

Eyewitness or earwitness identification is

problematic

in

general—and

Ellison’s

identification was particularly suspect—yet the

State relied on Ellison’s testimony as the

lynchpin of its case. This highlights both the

influence her identification likely exerted on

the jury and the significance of impeaching it.

See infra Part II.B.

•

The real danger of conviction based on

incentivized witness testimony stresses the

importance of safeguards, such as disclosure,

cross-examination, and a cautionary jury

instruction—but none of that happened here.

See infra Part II.C.

•

The lead prosecutor’s own belief that the case

against Johnson was not strong because it

rested on Ellison’s testimony—an opinion

affirmed more recently by an independent reinvestigation—confirms that the disclosure of

the promised reward would have dramatically

weakened

Ellison’s

already

unreliable

testimony.

Indeed, the district attorney’s

recent review expressly identifies the

suppression of the reward as a reason why a

new trial is necessary. See infra Part II.D.

7

The non-disclosure of Ellison’s motive to testify for

reward money is just one disquieting piece of a

concerningly weak case for conviction.

II.

This case bears numerous hallmarks of a

wrongful conviction.

A.

The State pursued inconsistent

prosecutorial theories.

Prosecutors must “search for truth in criminal

trials,” Strickler v. Greene, 527 U.S. 263, 281 (1999),

and have a corresponding “duty to refrain from

improper methods calculated to produce a wrongful

conviction,” Berger v. United States, 295 U.S. 78, 88

(1935). See also Ramos v. Louisiana, 590 U.S. ----, 140

S. Ct. 1390, 1401 (2020) (recognizing the

constitutional importance of “safeguarding against

overzealous prosecutions”).

That responsibility

precludes advancing inconsistent theories on the

same facts about the same crime, when only one

person can be responsible. Doing otherwise promotes

unreliability in convictions and is incompatible with

the constitutional mandate that a prosecutor’s

obligation “is not that it shall win a case, but that

justice shall be done.” Berger, 295 U.S. at 88.

The State presented five different theories, over

the course of three years, regarding who killed Deputy

Hardy. See C3. Supp-3 10 ¶ 1; see also generally C3.

Supp-1 63.

The inconsistent theories were

particularly problematic because the forensic

evidence, supported by the testimony of State

witnesses, made clear that a single shooter fired two

shots in rapid succession from the same gun. See T.R.

389, 545–51, 886–87.

8

First, in January 1996, the lead detective testified

before the grand jury that his investigation revealed

that Ardragus Ford and Omar Berry shot Deputy

Hardy. C3. 495. Second, in November 1997, at

Ardragus Ford’s first trial, the State advanced the

theory that Ford was the shooter. C3. Supp-1 63.

Third, at Johnson’s first trial in December 1997, the

State argued that Johnson pulled the trigger and

killed Deputy Hardy. C3. Supp-1 63. Fourth, at

Johnson’s second trial in August 1998—the trial that

led to his wrongful conviction, which is the basis of

this petition—the State proceeded on the theory that

both Johnson and Quintez Wilson shot Deputy Hardy.

C3. Supp-1 63; T.R. 903–04. Fifth, and finally, in

Ardragus Ford’s second trial in June 1999 (i.e., after

Johnson was convicted), the State returned to the

theory that Ford shot Deputy Hardy. C3. Supp-1 63.

This history—which begins and ends with the

State advancing the theory, embraced by the State’s

lead detective, that Ardragus Ford killed Deputy

Hardy—proves that the State has prosecuted at least

one innocent person in connection with the crime.

That cannot be squared with a “search for truth.”5

5 Cf. Bradshaw v. Stumpf, 545 U.S. 175, 186–88 (2005) (holding

prosecutor’s use of inconsistent theories could have affected

defendant’s sentencing and remanding for consideration of

whether it violated due process); see also Drake v. Kemp, 762 F.2d

1449, 1479 (11th Cir. 1985) (Clark, J., concurring) (inconsistent

theories of the same crime “reduce criminal trials to mere

gamesmanship and rob them of their supposed purpose of a

search for truth”); Thompson v. Calderon, 120 F.3d 1045, 1058–

59 (9th Cir. 1997) (en banc) (holding that it violates due process

for the State to argue in one defendant’s trial that he alone

committed a murder, then argue in a subsequent trial that

another defendant actually committed the same crime), vacated

9

The concept that inconsistent prosecutorial

theories can lead to a wrongful conviction is not

theoretical—it unfortunately happens. In Bankhead

v. State, 182 S.W.3d 253, 259 (Mo. Ct. App. 2006), for

example, the court affirmed vacatur of a murder

conviction because the State “selectively presented

contradictory evidence and arguments in three

different cases depending upon which defendant was

before the trial court.” And in Smith v. Groose, 205

F.3d 1045, 1051–54 (8th Cir. 2000), the Eighth Circuit

vacated a murder conviction because “[t]he State’s use

of factually contradictory theories . . . fatally infected

[the] conviction.”

That inconsistent prosecutorial theories similarly

infected this capital case is the first wrongfulconviction hallmark that reveals the weakness of the

case against Johnson—and that should give pause as

to his trial’s outcome and the significance of Ellison’s

non-disclosed motive to testify. In fact, five months

after Violet Ellison approached the police with her

story implicating Johnson, the State’s lead detective

provided sworn testimony to the grand jury that

someone else killed Deputy Hardy. C3. 488; T.R. 693.

In other words, the lead detective’s grand-jury

testimony shows that the State itself did not believe

on other grounds, 523 U.S. 538 (1998); Michael Q. English, A

Prosecutor’s Use of Inconsistent Factual Theories of a Crime in

Successive Trials: Zealous Advocacy or a Due Process Violation?,

68 FORDHAM L. REV. 525, 528 (1999) (“[A] prosecutor violates

both the Due Process Clause and her ethical obligations when

she argues inconsistent factual theories of a crime in successive

trials without taking affirmative steps to repudiate the factual

theory used in the first trial.”).

10

Ellison’s testimony, even though it eventually formed

the basis of the case against Johnson.

B.

The State relied on an unreliable

“earwitness identification.”

This case involves “earwitness identification”—i.e.,

Ellison testified that she overheard Johnson admit, on

a three-way phone call, to involvement in the crime.6

In fact, Ellison was the State’s star witness: The

prosecution described her expected testimony in detail

at the climax of its opening statement, exhorting the

jury that if they found Ellison to be someone “they

c[ould] trust,” then they would “be convinced” that

Johnson shot Deputy Hardy.7

Yet, earwitness identification is an identification

method even less reliable than its troubled cousin,

eyewitness identification. Nearly half of death-row

exonerations have involved this sort of false

“secondary confession”—i.e., a statement made by a

third-party about someone else’s supposed (but later

proved to be false) admission of guilt.8 That number

See T.R. 312–16 (State’s opening), 663–719 (V. Ellison

testimony), 905–11 (State’s closing).

6

7 T.R. 316 (“I want you to listen to Mrs. Ellison, I want you to pay

attention to her testimony. Look at her, ask yourself if she is a

person that you would trust, that you can trust, that you believe

if there is any way that she could have gotten the information

that she tells you from anywhere else other than [Johnson]. And

I submit to you that if you do that, at the close of this case you’ll

be convinced . . . .”).

8 See Jessica K. Swanner & Denise R. Beike, Incentives Increase

the Rate of False but Not True Secondary Confessions from

Informants with an Allegiance to a Suspect, 34 LAW & HUM.

BEHAV. 418 (2010) (noting that false secondary confessions were

“present in 46% of the wrongful convictions in death row cases”).

11

alone is staggering, but in light of the fact that

Johnson was convicted on the basis of a single

earwitness, who overheard a phone call in which

Johnson purportedly confessed to the crime—even

though the witness had never met him and did not

know his voice (T.R. 713–14; C3. Supp-1 64 (citing

C3.202–03))—the

presence

of

earwitness

identification should be a huge red flag.

* * *

Witness identification is problematic—not just in

earwitness identifications, but even in the far more

common (but no more reliable) realm of eyewitness

identification. See, e.g., Perry v. New Hampshire, 565

U.S. 228, 244–45 (2012) (“We do not doubt either the

importance or the fallibility of eyewitness

identifications.” (citing “studies showing that

eyewitness misidentifications are the leading cause of

wrongful convictions” and “research indicating that as

many as one in three eyewitness identifications is

inaccurate”)); United States v. Wade, 388 U.S. 218,

228 (1967) (“The vagaries of eyewitness identification

are well-known; the annals of criminal law are rife

with instances of mistaken identification.”). The

Innocence Project’s work has revealed that mistaken

eyewitness

identifications

contributed

to

approximately 69% of the 375 wrongful convictions in

the United States that have been overturned by postconviction DNA evidence.9

9 How Eyewitness Misidentification Can Send Innocent People to

Prison, INNOCENCE PROJECT, https://innocenceproject.org/howeyewitness-misidentification-can-send-innocent-people-toprison/ (last visited May 19, 2023). The pairing between

mistaken eyewitness identification and wrongful conviction is

not limited to cases where innocence is proven by DNA evidence,

12

This

propensity

of

mistaken

eyewitness

identification to lead to wrongful convictions is just as

prevalent in capital cases. In fact, the Center on

Wrongful Convictions conducted a study of eighty-six

defendants who were sentenced to death after

Furman v. Georgia, 408 U.S. 238 (1972), but were

later exonerated, and they concluded that eyewitness

testimony played a role in forty-six of those eighty-six

wrongful convictions.10 More than half. And, as in

this case, in thirty-three of those forty-six cases,

eyewitness (or, here, earwitness) testimony was the

only evidence connecting the defendant to the crime.11

Simply put, convictions (like this one) that are

based on the testimony of a single witness are

unreliable, even when (unlike here) those witnesses

were present at the scene, and even when (again,

unlike here) the witnesses were themselves the

victims of the crime.12 There are numerous examples

of course. As of April 2014, 75% of the 1,365 exonerations

examined by the National Registry of Exonerations involved

some type of false identification. See Kaitlin Jackson & Samuel

Gross,

Tainted

Identifications,

NAT’L

REGISTRY

OF

EXONERATIONS

(Sept.

22,

2016),

https://www.law.umich.edu/special/exoneration/Pages/taintedid

s.aspx.

See Rob Warden, How Mistaken and Perjured Eyewitness

Identification Testimony Put 46 Innocent Americans on Death

Row, CTR. ON WRONGFUL CONVICTIONS, NW. L. SCH. at 2–3 (May

2, 2001), https://files.deathpenaltyinfo.org/legacy/files/pdf/Study

CWC2001.pdf.

10

11 Ibid.

See generally Sara Conway, A New Era of Eyewitness

Identification Law: Putting Eyewitness Testimony on Trial, 50

NEW ENG. L. REV. 81 (2015); see also, e.g., Thomas Albright & Jed

Rakoff, Eyewitnesses Aren’t as Reliable as You Might Think,

WASH. POST (Jan. 13, 2015), https://www.washingtonpost.com/

12

13

of the dangers of relying so heavily on witness

identifications—again, even when the witness is the

actual victim—but to highlight just a few:

•

In Jefferson County, Alabama, Freddie Lee

Gaines was convicted of second-degree murder.

A survivor of the shooting identified Gaines,

but years later, another man confessed, and

Gaines’s conviction was vacated.13

•

In Bronx County, New York, James Anderson

was identified by the victim as the intruder who

threatened her with a knife and burgled her

apartment. Anderson insisted that he had been

hospitalized at the time, but a jury convicted

him.

Before sentencing, hospital records

arrived corroborating Anderson’s story; his

convictions were vacated.14

•

In Dallas County, Texas, a rape victim

identified Larry Fuller as her attacker and a

opinions/eyewitnesses-arent-as-reliable-as-you-might-think/

2015/01/30/fe1bc26c-7a74-11e4-9a27-6fdbc612bff8_story.html

(noting that “accounts of events promulgated by attorneys and

news media” can reinforce a witness’s beliefs regardless of the

accuracy of the identification); NAT’L RESEARCH COUNCIL,

IDENTIFYING

THE

CULPRIT:

ASSESSING

EYEWITNESS

IDENTIFICATION 2 (Nat’l Academies Press 2014) (“The fidelity of

our memories to actual events may be compromised by many

factors at all stages of processing, from encoding to storage to

retrieval.”).

Stephanie Denzel, Freddie Lee Gaines, NAT’L REGISTRY OF

EXONERATIONS (before June 2012) https://www.law.umich.edu/

special/exoneration/Pages/casedetail.aspx?caseid=3226.

13

14 Maurice Possley, James C. Anderson, NAT’L REGISTRY OF

EXONERATIONS (Feb. 16, 2014), https://www.law.umich.edu/

special/exoneration/Pages/casedetail.aspx?caseid=4378.

14

forensic analyst testified that Fuller could not

be excluded. Eighteen years later, DNA testing

excluded Fuller, and his conviction was

vacated.15

Earwitnesses who (as in this case) identify a

suspect based solely on hearing the suspect’s voice—

especially when they do not know the suspect or his

voice—are even less reliable than eyewitnesses. See,

e.g., Cindy E. Laub, Lindsey E. Wylie, & Brian H.

Bornstein, Can The Courts Tell An Ear From An Eye?

Legal Approaches to Voice Identification Evidence, 37

LAW & PSYCHOL. REV. 119, 124–25 (2013) (“[V]oice

recognition by itself is poor, such that earwitness

identifications are less accurate than eyewitness

identifications.”); Christopher Sherrin, Earwitness

Evidence: The Reliability of Voice Identifications, 52

OSGOODE HALL L.J. 819, 822 (2015) (collecting studies

identifying earwitness testimony as “extremely

inaccurate[] and likely to produce high false

identifications” (internal quotations omitted)).

* * *

That is the kind of notoriously unreliable witness

testimony that put Toforest Johnson on death row.

Both times that Johnson was on trial, the State’s case

hinged on Ellison’s testimony. See C3. Supp-3 11 ¶ 4;

T.R. 316, 905–16; C3. Supp-1 63. Yet, Ellison was not

present when the murder occurred, so she was not an

eyewitness. See C3. Supp-1 63–64; C3. Supp-3 11 ¶ 4.

Her contribution to the evidence was far more

attenuated: She listened in on a three-way phone call

15 The Innocence Project, Larry Fuller, NAT’L REGISTRY OF

EXONERATIONS (before June 2012), https://www.law.umich.edu/

special/exoneration/Pages/casedetail.aspx?caseid=3224.

15

initiated by her daughter, and she testified that she

overheard a man who said his name was “Toforest”

say he shot Deputy Hardy and that another man had

shot a second time. See T.R. 668–85, 711; C3. Supp-1

63–64.

That testimony was, of course, inconsistent with

the physical evidence; as already explained in Part

II.A above (see also T.R. 545, 551, 886–87), there was

only one shooter, not two different people who shot as

Ellison testified (T.R. 683–84, 711). Moreover, like

unreliable eyewitness testimony later shown to have

contributed to a wrongful conviction, Ellison’s

testimony is rife with other indicia of unreliability.

She had never met Toforest Johnson and did not know

his voice. T.R. 713–14; C3. Supp-1 64 (citing C3.202–

03). And, critically, she had heard media coverage of

the murder and Johnson’s status as a suspect before

she overheard the conversation about which she

testified. C3. Supp-1 72–73 (citing T.R. 667, 706–07,

711; C3. 134–137). The influence of that knowledge

on the significance Ellison attached to what she

thought she heard is reflected in her own allegedly

contemporaneous notes of the eavesdropped

conversation, in which she identified the person she

overheard speaking as “Johnson,” despite the fact that

she later testified that the speaker referred to himself

only as “Toforest.” C3. Supp-1 73–74 (citing T.R. 717;

C3. 537–40).

Just as can occur even with eyewitnesses, Ellison

was influenced by reports about the case, and her

recollection of what she heard was not hers alone.

Those facts further evidence the weakness of the case

against Johnson and should undermine confidence in

his conviction.

16

C.

The State’s star witness was

incentivized to provide testimony

leading to a conviction.

Incentivized witnesses are yet another hallmark of

wrongful convictions. Any incentive offered to a

witness is highly relevant because “[i]ncentives

provide a motive to lie.” See Swanner, supra, at 427.

To protect against tainted testimony, statutes and

professional-conduct codes prevent inducement or

payment, other than reimbursement of actual

expenses, to any witness testifying under oath. See

e.g., 18 U.S.C. § 201(c)(2); ALA. R. OF PROF’L CONDUCT

3.4(b).

The dangers of incentivized witnesses are present

with all witnesses who are offered an incentive, but in

capital cases, where life and liberty are on the line,

the consequences of introducing those dangers into

trial are much more serious. Cf. Ford v. Wainwright,

477 U.S. 399, 411 (1986) (“[D]eath is different.”). So,

the State’s promise to pay a reward—and its later,

undisclosed payment to Ellison—for testimony

leading to a conviction for Deputy Hardy’s murder is

another huge red flag in this case.

Historically, it has been assumed that dangers

from incentives given to witnesses are “adequately

mitigated by disclosure, cross-examination, and

cautionary jury instructions.” Christopher T.

Robertson & D. Alex Winkelman, Incentives, Lies, and

Disclosure, 20 U. PA. J. CONST. L. 33, 42 (2017)

(internal quotations omitted). The undeniable reality

is that innocent people have been and continue to be

convicted based on the testimony of witnesses who

have an incentive to help obtain a conviction. See

Swanner, supra, at 418.

17

This danger is even more acute when the

prosecution fails to disclose a witness’s incentive.

Non-disclosure prevents the jury from learning about

evidence that renders the witness’s testimony

potentially unreliable. It also forecloses the defense

from effectively cross-examining the witness and

precludes the defense and the trial court from

recognizing the need for a cautionary jury instruction.

Thus, Johnson’s petition rightly argues that the

State’s suppression of evidence about Ellison’s

incentive to testify warrants reversal.

Indeed, here, not only was there a motive for

Ellison to fabricate what she heard—because the

State offered to pay a financial reward for helpful

information—but the State never disclosed Ellison’s

knowledge of the offer and instead emphasized, in

both opening and closing the trial, that she came

forward on her own and that she had no reason to lie.

See T.R. 316 (opening); T.R. 905–11 (closing); see also

supra n. 7; C3. Supp-1 64. Not so.16

The (hidden) incentive payment to Ellison brands

Johnson’s case with yet another hallmark of wrongful

conviction.

Johnson’s trial counsel requested the disclosure of Brady

material and the trial court ordered discovery (see C3. 482–87),

but the State never disclosed anything, before or during trial,

that connected Ellison to the reward offer. Moreover, not only

did she have a financial motive for testifying, but Ellison and the

victim’s wife had known each other for more than fifteen years.

See C1. Supp-1 242. The State’s promise of a reward for assisting

in the successful prosecution of someone for the murder of her

friend’s husband made Ellison’s testimony doubly unreliable.

16

18

D.

The prosecutors now doubt the

validity of Johnson’s conviction.

In the light of a prosecutor’s charge to pursue

justice, not convictions, prosecutorial doubt about the

reliability of a conviction, or the strength of the State’s

case, should not be taken lightly. See, e.g., Young v.

United States, 315 U.S. 257, 258 (1942) (“The public

trust reposed in the law enforcement officers of the

Government requires that they be quick to confess

error when, in their opinion, a miscarriage of justice

may result from their remaining silent. . . . The

considered judgment of the law enforcement officers

that reversible error has been committed is entitled to

great weight . . . .”).

The Innocence Project has found over the course of

its more than thirty years that when a prosecutor

becomes convinced of holes in the State’s case, that is

a crucial tipping point. Indeed, when prosecutors

arrive at doubts about the strength of their case on

their own, that is a particularly (and extremely)

powerful indicator of a wrongful conviction. When

that happens, the reasons for the doubt should be

extensively examined, and the prosecution should be

encouraged to reconsider the case. Innocent people

can walk free: The exonerations of Glenn Ford in

Louisiana and Frank Sealie in Alabama are just two

examples that evidence the point.17

17 See Ex-Louisiana Prosecutor Apologizes for ‘the misery I have

caused’ Freed Inmate, GUARDIAN (Mar. 25, 2015, 4:20 PM),

https://www.theguardian.com/us-news/2015/mar/25/formerlouisiana-prosecutor-exonerated-death-row-inmate-apologizes;

Maurice Possley, Frank Sealie, NAT’L REGISTRY OF

EXONERATIONS (last updated Oct. 31, 2016), https://www.law.

19

There is serious prosecutorial doubt here too. Both

the lead prosecutor in Johnson’s trial, Jeff Wallace,

and the current Jefferson County district attorney,

Danny Carr, have expressed their concerns about the

validity of Johnson’s conviction. That is incredibly

significant and another powerful indicator of a

wrongful conviction—particularly alongside the other

evidentiary issues that have surfaced in the years

since Johnson’s conviction.

Lead Trial Prosecutor Jeff Wallace. Wallace

was the lead prosecutor at Johnson’s trial. He has

since testified under oath that he “[did not] think the

State’s case [against Johnson] was very strong,

because it depended on the testimony of Violet

Ellison.” See C3. 455.18

That is a telling concession because without

Ellison’s testimony, there would have been no case

against Johnson. It also recognizes that the State’s

dependence on a single earwitness identification, by a

witness whom the State promised to pay for her

testimony (without informing the defense or the jury),

introduces serious reliability problems and leads to

the increased risk that an innocent man will be

punished for someone else’s crime.

umich.edu/special/exoneration/Pages/casedetail.aspx?caseid=46

59.

See also Radley Balko, An Illusion of Justice: The Baffling

Conviction and Death Sentence of Toforest Johnson Reveal a

Broken

System,

WASH.

POST

(Sept.

5,

2019),

https://www.washingtonpost.com/opinions/2019/09/05/analabama-man-has-been-death-row-years-he-is-almost-certainlyinnocent/?arc404=true.

18

20

According to a recent filing by the district

attorney’s office, Wallace continues to “express[]

concerns about this case” and “supports [Johnson’s]

request” for a new trial. C3. Supp-3 11 ¶ 5.

District Attorney Danny Carr.

The

prosecutorial doubt extends beyond the original trial

team. In June 2020, the current Jefferson County

district attorney, having “determined that [his

office’s] duty to seek justice requires intervention in

this case,” filed a brief in support of Johnson, asking

the circuit court to “in the interest of justice . . . grant[]

a new trial.” C3. Supp-3 10–11.

The district attorney’s filing was the culmination

of a nine-months-long, independent review of this case

that included meeting with Wallace, “reviewing the

case files, interviewing witnesses, and consulting with

the family of Deputy Hardy.”19 See C3. Supp-3 at 11

¶ 5. In addition to reinforcing Wallace’s continued

“concerns about this case,” the district attorney was

moved by, among other things, these various

wrongful-conviction hallmarks—e.g., the inconsistent

prosecutorial theories, the significance of Ellison’s

earwitness testimony, and the fact that Ellison “was

subsequently paid $5,000 which was never mentioned

during trial.” C3. Supp-3 10–11 ¶¶ 1, 4–5. Indeed,

the fact that the district attorney expressly

highlighted the Brady violation that is the subject of

this petition is powerful evidence of the materiality of

that suppressed evidence.

19 Beth Shelburne, District Attorney Urges New Trial for Man on

Alabama’s Death Row, WBRC (June 12, 2020, 12:22 PM),

https://www.wbrc.com/2020/06/12/district-attorney-urges-newtrial-man-alabamas-death-row/.

21

The Court should give “great weight” to Wallace’s

lingering concerns and the district attorney’s

independent review of this case. Young, 315 U.S. at

258. These sorts of conviction integrity reviews are

crucial to the legitimacy of the criminal-justice

system. Increasingly, prosecutors have exercised

their duty to ensure justice—and to remedy unjust

convictions—by establishing formal convictionintegrity-review units. See Imbler v. Pachtman, 424

U.S. 409, 427 n.25 (1976).20 And the mission could not

be more important: Wrongful convictions distort the

criminal-justice system by punishing the innocent

while the guilty go free. Conviction integrity reviews,

such as District Attorney Carr’s independent review

in this case, exist to fix that distortion.

* * *

Against that backdrop—(1) a prosecution pursued

via inconsistent theories of who committed the crime;

(2) a conviction secured on the back of testimony from

a single earwitness, who identified the defendant

(whom she did not know) by overhearing his voice

20 As of May 2023, at least seven states have established state-

wide “Conviction Integrity Units” or “Conviction Review Units”

that work to prevent, identify, and remedy false convictions. In

addition to statewide Conviction Integrity Units, at least ninety

independent units have been established by prosecutor’s offices

around the nation. See Conviction Integrity Units, NAT’L

REGISTRY OF EXONERATIONS, https:// www.law.umich.edu/

special/exoneration/Pages/Conviction-Integrity-Units.aspx (last

visited May 3, 2023). These units have helped to exonerate more

than 461 people in the last decade alone. See Tom Jackman, Va.

Attorney General Launches Conviction Integrity Unit to Identify

Wrongful Convictions, WASH. POST (Jan. 16, 2021),

https://www.washingtonpost. com/crime-law/2021/01/16/herring

-wrongful-convictions/.

22

while eavesdropping on a phone call; (3) a publicized

reward for testimony leading to a conviction that was

later paid out in an undisclosed payment; (4) a case so

tenuous that even the prosecutor decries it as weak

and a conviction that the district attorney’s office no

longer stands behind—the Court’s review takes on

heightened significance.

Dozens of innocent people have been convicted in

cases that present just one hallmark of wrongful

conviction.

Here, the presence of all of these

hallmarks in a single case cries out for relief for

Toforest Johnson—particularly in the light of the

district attorney’s support for a new trial, predicated

in part on the very subject of this petition, following

his recently completed independent review of the case.

CONCLUSION

The Innocence Project is proud to have assisted in

freeing hundreds of innocent people since 1992, and it

is working to free those still-wrongfully-incarcerated

people who have been convicted despite their

innocence. In that spirit, the Innocence Project

respectfully urges this Court to consider Johnson’s

petition for a writ of certiorari against the backdrop of

the numerous troubling indicia of a wrongful

conviction that his capital case presents, and to grant

the petition.

23

Respectfully submitted,

Adnan Sultan

THE INNOCENCE

PROJECT, INC.

40 Worth Street, Suite 701

New York, NY 10013

Lindsey C Boney IV

Counsel of Record

Virginia N. Adamson

BRADLEY ARANT BOULT

CUMMINGS LLP

1819 Fifth Avenue North

Birmingham, AL 35203

Phone: (205) 521-8914

lboney@bradley.com

Counsel for Amicus Curiae

May 22, 2023