Amicus Curiae Brief — Davis v. Georgia, 129 S. Ct. 397 (2008) (No. 08-66)

AUG 15 2008 (3)

No. 08-66

IN THE

Supreme Court of the Wnited States

TROY ANTHONY DAVIS,

Petitioner,

v.

STATE OF GEORGIA

Respondent.

ON PETITION FOR A WRIT OF CERTIORARI TO THE

SUPREME Court OF GEORGIA

MOTION FOR LEAVE TO FILE BRIEF

AMICUS CURIAE AND BRIEF AMICUS CURIAE

OF THE INNOCENCE PROJECT

IN SUPPORT OF PETITIONER

Barry C. SCHECK DEIRDRE O’CONNOR

PETER J. NUEFELD Counsel of Record

Davin Lortis PO. Box 210787

INNOCENCE PROJECT Chula Vista, CA 91921

100 Fifth Avenue (404) 392-0249

3rd Floor

New York, NY 10011

(212) 364-5340

Attorneys for Amicus Curiae

MOTION FOR LEAVE TO FILE AMICUS

Pursuant to Rule 37.3(b) of the Rules of the

Supreme Court of the United States, Deirdre

O'Connor! (Of Counsel) and Barry C. Scheck,

Peter J. Neufeld, and David Loftis of the Innocence

Project, Inc., (“the Project”) hereby request leave to

file the accompanying amicus curae brief. This

brief is submitted in support of the petition for

writ of certiorari to the Supreme Court of Georgia.

Petitioner Troy Davis has consented to the filing of

this brief. Respondent State of Georgia has

withheld consent.

As set forth in the accompanying brief, the

Project provides pro bono legal services to indigent

prisoners for whom post-conviction DNA testing of

evidence can yield conclusive proof of innocence.

The Project is greatly concerned that the

Georgia Supreme Court has set an impossibly

high standard preventing Mr. Davis and other

innocent Georgians from having evidence of

innocence heard. Accordingly, the Project

respectfully requests leave to file’ the

accompanying amicus cunae brief.

Respectfully submitted,

Deirdre O'Connor

Of Counsel of Record for

Innocence Project, Inc.

' Deirdre O'Connor's application for admission to the United States

Supreme Court bar has been processed. She is scheduled for admission

on August 18, 2008.

ii

TABLE OF CONTENTS

BRIEF AMICUS CURIAE OF THE INNOCENCE

PROJECT IN SUPPORT OF PETITIONER i

TABLE OF CONTENTS il

TABLE OF AUTHORITIES iv

INTEREST OF AMICUS CURIAE 1

SUMMARY OF ARGUMENT 2

ARGUMENT 4

I. A Deeply Divided Court Created A Rule Of

Law Authorizing Categorical Denial Of Due

Process For Innocent Georgians Convicted

On The Word Of Perjurers. 4

II. The Impossible High Purest Fabrication

Standard Would Have Failed to Protect At

Least Three Known Innocent Death Row

Inmates Who Were Convicted Based On

Perjured Testimony. 10

Ill. The Majority’s Blind Adherence To This

New Standard And Its Failure To Apply

Today’s Science To The Undisputed Facts,

Which Demonstrate The Unreliability Of The

Now-Disavowed Identification Evidence, Is

Further Proof Of This Standard’s Failure. 12

A. The Court Ignored Recent Scientific

Studies Establishing That The Undisputed

Viewing Conditions And Circumstances Of

This Crime Precluded A Genuine Basis For

Subsequent Recognition. 12

ili

1. The Extraordinarily Short Duration Of

The Assault Of Young And Shooting Of

Officer MacPhail Did Not Provide Sufficient

Time To Encode Crucial Detail. 14

2. The Presence And Discharge Of A Gun

Took Focus Away From The Face Of The

Shooter. 15

3. It Is Scientifically Impossible For Ferrell

To Subsequently Recognize A Stranger Seen

In A Dimly Lit Parking Lot From 160 Feet

Away. 16

B. The Court Did Not Consider The

Scientifically Supported Probability That

Witnesses Selected The Police Suspect

During A Suggestive Identification Process,

Which Created An Ideal Situation For

Memory Source Error And False

Identifications. 19

1. The Saturation Of Davis’ Image Exposed

Witnesses To A Contaminant That

Increased The Probability Of Memory

Source Error And Rendered The Witnesses’

Later Selection Of Davis Unreliable. 20

2. Non-Blind Administration After Repeated

Exposure To Davis’ Image Destroyed The

Procedural Protections Of A Properly

Conducted Photo Array And Resulted In

Unreliable Evidence Of Guilt. 21

CONCLUSION 23

iV

TABLE OF AUTHORITIES

Federal Cases

Board of Pardons v. Allen, 482 U.S. 369 (1987)._ 5

Boddie v. Connecticut, 401 U.S. 371, 379 (1971) 5

Drake v. Kemp (Drake II), 762 F.2d 1449 (11 Cir.

1985) 11

Guibert v. California, 388 U.S. 263 (1967) ___—s 20

Manson v. Brathwaite, 432 U.S. 98 (1976) ___—_— 21

Stovall v. Denno 388 U.S. 293 (1967). 20

Stovall, 388 U.S. at 301-302. 20

Tennessee v. Lane, 541 U.S. 509, 523 (2004) ss 55

United States v. Wade, 388 U.S. 218 (1967)_ 20

State Cases

Drake v. State, 248 Ga. 891, 894 (1982)__‘:10, 11

Ex parte Adams, 768 S.W.2d 281 (1989)___—s«10

State ex rel. Amrine v. Roper, 102 S.W.3d 541 (Mo.

2003) 9

Timberlake v. State, 271 S.E.2d 792 (1980).__2, 4

Statutes

0.C.G.A § 5-5-23 2,4

Other Authorities

Alexandra Natapoff, The Faces Of Wrongful

Conviction Symposium: Beyond Unreliable: How

Snitches Contribute To Wrongful Convictions, 37

Golden Gate U.L. Rev. 107 (2006) 7

Blind to Change, Even as It Stares Us in the Face,

New York Times, April 1, 2008 14

Bradfield, A.L & Wells, G.L., ‘Good, You Identified

the Suspect’: Feedback to Eyewitnesses Distorts

Their Reports Of the Witnessing Experience, 83

J. Appl. Psychol. 360 (1998) 21

Brian Murray & Joseph C., He Lies, You Die:

Criminal Trials, Truth, Perjury, and Fairness, 27

N.E. J. on Crim. & Civ. Con. 1 (2001) 7

Daniel Medwed, Anatomy Of A Wrongful

Conviction: Theoretical Implications And Practical

Solutions, 51 Vill. L. Rev. 337 (2006) 7

Daniel Wolf, I Cannot Tell a Lie: The Standard for

New Trial in False Testimony Cases, 83 Mich. L.

Rev. 1925 (1985) 7

Deffenbacher, K.A., et al., A Meta-Analytic Review

of the Effects of High Stress on Eyewitness

Memory, 28 Law and Human Behavior 687

(2004) 13

vi

Douglas & Steblay, Memory Distortion in

Eyewitness: A Meta-Analysis of the Post-

Identification Feedback Effect, 20 App. Cognitive .

Psychol. 991 (2006) 21

Dysart, J., Lindsay, R.C.L., Hammond, R.,

Dupuis, P, Mugshot Exposure Prior to Lineup

Identification: Interference, Transference, and

Commitment Effects, 86 Journal of Applied

Psychology 1280 (2001) 19

Hodes, William W., Executing The Wrong Person,

29 Loy. L.A. L. Rev. 1547 11

Janice J. Repka, Rethinking The Standard For

New Trial Motions Based Upon Recantations As

Newly Discovered Evidence, 134 U. Pa. L. Rev.

1433 (1986) 7

Jonathan Liebman and Joel Cohen, Perjury and

Civil Litigation, 20 Litig. 43 (1994); Lisa Harris,

Perjury Defeats Justice, 42 Wayne L. Rev. 1755

(1996) 6

Leippe, M.L. et al, Crime Seriousness as a

Determinant of Accuracy in Eyewitness

Identifications, 63 J. of Applied Pyschol. 345

(1978) 13

Loftus, E. F. & Greene, E., Warning: Even memory

for faces may be contagious, 4 Law And Hum.

Behav. 323 (1980). 19

Loftus, E., Eyewitness Testimony, Harvard

University Press, 25-27 (1996) 14

Vil

Loftus, G.R. & Harley, E.M., Why Is It Easier to

Identify Someone Close Than Far Away, 12

Psychonomic Bull. & Rev. 43 (2005) 17

Mark Curriden, The Lies Have It: Judges Maintain

That Perjury Is on the Rise, but the Court System

May Not Have Enough Resources to Stem the

Tide, A.B.A.J., May 1995, at 68 5

Memon, A., Hope, L., & Bull, R., Exposure

duration: Effects on eyewitness accuracy and

confidence, 94 British J. of Psychol. 339 (2003).

13

Richard H. Underwood, Perjury! The Charges and

the Defenses, 36 Dua. L. Rev. 715 (1998) 6

_Ross, D. F., Ceci, S. J.. Dunning. D., & Toglia, M.

P., Unconscious transference and mistaken

identity: Toward a memory blending approach,

(1994) in Ross, D. F., Read, J.D., & Toglia, M. P.

(Eds.) Adult eyewitness memory: Current trends

and developments (Cambridge University Press

1994). 19

Sharon Cobb, Gary Dotson As Victim: The Legal

Response To Recanting Testimony, 35 Emory

L.J. 969 (1986) 7

Shawn Armbrust, Reevaluating Recanting

Witnesses: Why The Red-Headed Stepchild Of

New Evidence Deserves Another Look, 28 B.C.

Third World L.J. 75 (2008) ae

Vili

Steblay, N. M. A Meta-Analytic Review of the

Weapon Focus Effect, 16 Law And Hum. Behav.

413, 420-21 (1992). is

The Supreme Court on Trial: How the American

Justice system Sacrifices Innocent Defendants,

George C. Thomas III 6

Wells, G. L. & Olson, E. (2003). Eyewitness

identification. Annual Review of Psychology, 54,

277-295. 13

Wells, G., et al, Eyewitness Identification

Procedures: Recommendations for Lineups and

Photospreads, 22 Law and Hum.Behav. 603

(1998). 21

INTEREST OF AMICUS CURIAE!

The Innocence Project, Inc. (“the Project”) is

a nonprofit legal clinic and resource center

created by Barry C. Scheck and Peter J. Neufeld.

Founded at the Benjamin N. Cardozo School of

Law in 1992, the Project provides pro bono legal

services. to indigent prisoners for whom post-

conviction DNA testing of evidence can yield

conclusive proof of innocence. The Project

pioneered the post-conviction DNA litigation

model that has to date exonerated 218 innocent

persons, and served as counsel or provided

critical assistance in a majority of these cases.

The advent of forensic DNA testing and the

use of such testing to review criminal convictions

have provided scientific proof that our system

convicts innocent people, and that wrongful

convictions are not isolated or rare events.

Although there are untold numbers of cases in

which people have been wrongfully convicted but

there is no DNA evidence that can scientifically

prove their innocence, DNA testing has opened a

window into wrongful convictions so that we may

study the causes of this injustice and recommend

1 Counsel of record for all parties received notice of the

Amicus curiae’s intention to file this brief at least 10 days

prior to the due date. Petitioner has consented to the

Network’s filing of this bnef and Respondent has withheld

consent. No counsel for a party authored this brief in whole

or in part, and no counsel or party made a monetary

contribution intended to fund the preparation or submission

of this brief. No person other than amicus cunae, its

members, or its counse! made a monetary contnbution to

its preparation or submission.

2

practices to minimize the chance of its

occurrence. 2

SUMMARY OF ARGUMENT

Innocence does not fade with the passage of

time. Perjured testimony and other forms of

unreliable evidence, though, often cannot

withstand the test of time or the evolution of

science. At the heart of the Troy Davis case are

questions of innocence and fabricated evidence.

Seven recanting witnesses - many of whom

provided highly implausible testimony at trial —

and the nine additional witnesses implicating

Redd Coles, undermine our confidence in the

verdict and compel a meaningful reexamination of

all the evidence.

The Georgia Legislature created a state right

to a new trial for the wrongfully convicted who can

establish that new evidence of innocence “would

probably produce a different verdict.” O.C.G.A §

5-5-23; Timberlake v. State, 246 Ga. 488; 271

S.E.2d 792 (1980). By its terms it provided for a

meaningful review of Mr. Davis case, and indeed a

new trial. He received neither.

In Davis, the Georgia Supreme Court

changed the statute’s materiality standard to

require proof of the purest fabncation in witness

recantation cases. As interpreted, this is an

unattainable measure that necessarily thwarts

the statutorily conferred right to obtain relief from

a wrongful conviction, thus denying due process

—_

2 Appendix contains summary biographies of other

signatones.

3

to anyone whose conviction was secured via

perjured testimony.

As nearly half of the Georgia Supreme Court

justices recognized, the purest fabrication

standard is so “rigid” that the “fundamental

question [of] whether or not an innocent person

might have been convicted or even, as in this

case, might be put to death” went unanswered.

App. A. at 17a.

This decision creates a rule of law

authorizing categorical denial of due process for

innocence cases in Georgia. It gives the trial court

complete freedom to dismiss’ evidence of

innocence without evaluating its_ reliability,

subverting the well-considered, truth-seeking

function of post-conviction proceedings proscribed

by the legislature. This new judicially-minted,

purest fabrication standard will result in the

imprudent disregard of reliable evidence of

innocence.

Critically, the Davis majority, constricted by

its own newly-crafted standard, ignored

scientifically supported concerns’ that’ the

evidence used to convict Mr. Davis was unreliable.

The now-disavowed trial testimony included

reports of perception and memory beyond human

capacity. Even the two non-recanting witnesses,

Redd Coles (who has now been. identified by nine

people as the shooter) and Stephen Sanders,

presented inherently incredible and scientifically

implausible testimony.

Evidence of innocence should not _ go

unheard especially in a case, like Mr. Davis’,

where the conviction was based on scientifically

unreliable evidence and the legislature has

4

provided a remedy to hear such claims. Deeply

rooted in our notion of justice is the recognition of

the court’s duty to protect the innocent. A panoply

of constitutional safeguards guarantees the

accused a fair trial so that society can trust

findings of guilt. The hallmark of a fair trial is its

ability to successfully fulfill its truth-seeking

function. In any quest for truth, the reliability of

the evidence should take center stage. When new

information surfaces after trial that causes

reasonable people to lose confidence in the jury’s

guilty verdict — whether it is due to the ineffective

performance of defense counsel, the failure of the

prosecutor to disclose material evidence, or the

realization that unreliable and false testimony

resulted in the conviction of an innocent person —

our commitment to justice and truth compels

corrective action and a process through which

that can occur. The Georgia legislature has

provided an avenue for redress of such claims,

but its Supreme Court arbitrarily and capriciously

foreclosed claims stemming from perjury, and

subsequent recantation.

The time has come for this Court to

articulate the minimum requirements for a

constitutionally adequate procedure so_ thai

claims of wrongful convictions based on fabricated

testimony can be properly evaluated and innocent

people protected.

ARGUMENT

I. A Deeply Divided Court Created A Rule Of

Law Authorizing Categorical Denial Of Due

5

Process For Innocent Georgians Convicted On

The Word Of Perjurers.

The Georgia legislature created a right to a

new trial for any wrongfully convicted person who

can establish that new evidence of innocence

“would probably produce a different verdict.”

O.C.G.A § 5-5-23; Timberlake v. State, 246 Ga.

488; 271 S.E.2d 792 (1980). The Legislature did

not limit or classify the new evidence, but rather

broadly asserted that a petitioner could present

“any material evidence” to meet this standard.

This statute unambiguously creates a liberty

interest by conferring the right to a new trial once

the “substantive predicates’ have been met.”

Board of Pardons v. Allen, 482 U.S. 369 (1987).

This right exists independent of the evidentiary

cause of the wrongful conviction and recantation

evidence that is “material” fits comfortably within

the statute’s ambit.

It is a well-established due process principle

that once a state has created a statutory scheme

affecting a litigant's rights and interests, it must

provide “‘a meaningful opportunity to be heard by

removing obstacles to their full participation in

judicial proceedings” Tennessee v. Lane, 541 U.S.

509, 523 (2004) (citing Boddie v. Connecticut, 401

U.S. 371, 379 (1971); MLB. v. S.L.J., 519 U.S.

102 (1996)). Here, the court’s’. arbitrary

construction of the statute, under its rigid purest

fabrication standard, imposes unconstitutional

obstacles to full participation for those seeking

relief from wrongful convictions based on

3 Amict incorporate by reference the petitioner’s arguments

on this point.

6

demonstrably false testimony. This arbitrary

construction will undoubtedly have the innocent

languish in prison, and unduly risks wrongful

execution — concerns, which the post-conviction

statutory scheme undoubtedly sought to acdress.

It is axiomatic that some witnesses commit

perjury. And, despite the system’s truth-seeking

function, some perjurers get away with it. Mark

Curriden, The Lies Have It: Judges Maintain That

Perjury Is on the Rise, but the Court System May

Not Have Enough Resources to Stem the Tide,

A.B.A.J., May 1995, at 68; see also Jonathan

Liebman and Joel Cohen, Perjury and Civil

Litigation, 20 Litig. 43 (1994); Lisa Harris, Perjury

Defeats Justice, 42 Wayne L. Rev. 1755 (1996)

("Perjury has thus far evaded all solutions, is

pervasive in the courtroom, and is on the

increase."). Id. at 1755. Indeed, “[mJany lawyers

answer that [perjury] is the natural result, and

the tolerable cost, of an adversary system of

justice.” Richard H. Underwood, Perjury! The

Charges and the Defenses, 36 Duq. L. Rev. 715

(1998). In The Supreme Court on Trial: How the

American Justice system Sacrifices Innocent

Defendants, George C. Thomas III concludes that

“we have a problem with [ ] perjury” and other

categories of false evidence. Id. at 39. “It is the

too frequent failure of the adversary system itself

that produces wrongful convictions.” Id..

For perjurers who are later moved to speak

the truth, their recantations are received with a

disproportionate skepticism that can perpetuate

- 7

eternal injustice.’ Shawn Armbrust, Reevaluating

Recanting Witnesses: Why The Red-Headed

Stepchild Of New Evidence Deserves Another Look,

28 B.C. Third World L.J. 75 (2008); Daniel

Medwed, Anatomy Of A _. Wrongful Conviction:

Theoretical Implications And Practical Solutions, 51

Vill. L. Rev. ~37 (2006); Alexandra Natapoff, The

Faces Of Wrongful Conviction Symposium: Beyond

Unreliable: How Snitches Contribute To Wrongful

Convictions, 37 Golden Gate U.L. Rev. 107 (2006);

Brian Murray & Joseph C., He Lies, You Die:

Criminal Trials, Truth, Perjury, and Fairness, 27

N.E. J. on Crim. & Civ. Con. 1 (2001); Sharon

Cobb, Gary Dotson As Victim: The Legal Response

To Recanting Testimony, 35 Emory LJ. 969

(1986); Janice J. Repka, Rethinking The Standard

For New Trial Motions Based Upon Recantations As

Newly Discovered Evidence, 134 U. Pa. L. Rev.

1433 (1986); Daniel Wolf, I Cannot Tell a Lie: The

Standard for New Trial in False Testimony Cases,

83 Mich. L. Rev. 1925 (1985).

The Georgia Legislature developed a sound

analytical framework for undertaking this problem

4 Amici acknowledge that not all recanting witnesses have

committed perjury. Every recanter necessarily asserts that

their trial testimony was inaccurate. Some were simply

mistaken or improperly influenced to believe the falsity;

their testimony was inaccurate, but honest. Others

knowingly told an untruth. It is the latter ilk that is most

prevalent in Mr. Davis case and the most troubling for

courts. For simplicity sake, this brief uses perjurer

interchangeably with recanting witness. Amici recognize

that witnesses who lie out of fear prompted by police

intimidation or out of concern that the true perpetrator will

cause them harm are deserving of greater empathy than

connoted by the label of perjurer.

8

within its statutory scheme. If the recantations

were material, i.e., probably produce a different

outcome, a new trial is required. The Georgia

Supreme Court, however, arbitrarily placed its

own effectively insurmountable hurdled onto such

claims, requiring proof of the purest fabrication.

Even worse, here, it decided that the failure to

meet such a standard could be determined

without a hearing though there were seven

recantations.

Indeed, the Davis court held that each of

the seven recantation “affidavits lack[ed] the type

of materiality required to support’ an

extraordinary motion for new trial, as they do not

show the witnesses’ trial testimony to have been

the ‘purest fabrication.” App. A. at 7a; see also

8a-10a. It is hard to imagine any recantation — or

collection of recantations — that could satisfy this

standard. As noted by the dissent, the urdeniable

effect of the holding is to categorically exclude

recantations. App. A. at 18a.

By definition, a recantation always involves

a prior inconsistent version, e.g., “I lied when I

said X, Y is really the truth.” It can never, on its

own, conclusively prove the falsity of the first

version. The most to which a recantation can

aspire is to ultimately, at trial, be the more

persuasive of two competing versions of fact.

Requiring proof of the purest fabrication as a

threshold to justify a new trial necessarily

translates into a categorical exclusion.

-Indeed, the court’s analysis in this very case

demonstrates the impossibility of meeting its

standard. The court parsed each _ statement

individually trying to determine whether it alone

9

met the pure fabrication standard. It, however,

refused to engage in reviewing the collective

impact of the seven recantations taken together.

Moreover, by deeming the evidence mere

recantation, it did not properly weigh the scientific

testimony. The statute, which requires an

assessment of any material evidence, would

require the court to weigh the impact of the

statements collectively and in conjunction with

the supporting scientific evidence.

The Georgia court denied even a hearing on

the issue -- though, nonetheless, claiming not to

foreclose the possibility of hearings to subsequent

litigants. - App. A. at 3a. But it is unclear what

purpose a hearing would serve if the post-

conviction court were not then authorized to make

a finding that a recantation was_ sufficiently

reliable to warrant a new trial. No matter how

great the indicia of reliability, the recantation will

always compete with the earlier version. A judge

may find the recantation more credible, but could

not find, as a matter of law, that the original

version was the purest fabrication.

No longer tethered to any _ statutory

moorings, post-conviction courts now have

unfettered discretion to deny claims where there

is objectively material evidence mandating a new

trial. Indeed, in contradiction to legislative intent,

it would now be an abuse of discretion in Georgia

to grant a new trial if there is “material evidence”

but no proof of the “purest fabncation.” A trial

court in DeKalb County, then, could grant a

petitioner a new trial based on an ineffectiveness

claim, but a trial court in Chatham County would

have to deny such a nght to a petitioner who

10

provides equally compelling evidence of innocence

through recantation. Hence, Davis — a death row

inmate with seven recanting witnesses and nine

others implicating another - under the pure

fabrication standard was arbitrarily denied his

statutorily conferred right to a new trial. Indeed,

he never received a hearing on the merits.

Il. The Impossible High Purest Fabrication

Standard Would Have Failed to Protect At

Least Three Known Innocent Death Row

Inmates Who Were Convicted Based On

Perjured Testimony.

This standard’s failure is already proven by

cases of known innocents. At least three death

penalty cases depended entirely on recantations

to prove their innocence. State ex rel. Amrine v.

Roper, 102 S.W.3d 541 (Mo. 2003); Ex parte

Adams, 768 S.W.2d 281 (1989); Drake v. State,

248 Ga. 891, 894 (1982). Each man ultimately

found relief; two found relief in court, the

Georgian from the Board of Pardons and Paroles.

Id. All three men would have been executed

under the purest fabrication standard.

Joseph Amrine was_ convicted and

sentenced to death on the word of three prison

inmates; all of whom ultimately recanted. If Davis

were applied to Amrine, the court would have been

forced to deny relief because the three

recantations “do not show the witnesses’ tnal

testimony to have been the ‘purest fabrication.”

App. A. at 7a. Fortunately Amrine was wrongfully

convicted in Missouri, not Georgia.

If proof of the purest fabrication were

required in Randall Adams’ case, he too would

11

have been executed. His conviction was based on

several purported eyewitnesses. One, David

Harris, later confessed and recanted his trial

testimony against Adams. The court granted

relief even though two other’ purported

eyewitnesses did not recant. Adams would not

have fared so well in Georgia. The belief that

Harris “could subvert the ends of justice by falsely

admitting the crime to others and then absenting

himself”? would have rendered Harris’ confession

meaningless. App. A. at lla. Fortunately for

Adams his ordeal took place in Dallas, Texas and

not Savannah, Georgia.

Henry Drake was one of six known innocent

Georgia death row inmates to ultimately secure

his freedom. Drake was convicted based entirely

on the uncharged perjury of the actual killer,

William Campbell. The only way Drake could

prove his innocence, beyond the alibi witnesses he

presented at trial, was through Campbell’s

_ recantation. The Georgia Supreme Court denied

him relief and held that the “effect of Campbell’s

new testimony was clearly to impeach the

credibility of his earlier sworn statements.” Drake

v. State, 248 Ga. 891, 894 (1982).

Ultimately, Drake was pardoned.> Had he

been required to meet the purest fabrication

5 The Eleventh Circuit ordered a new trial due to Sandstrom

error and improper sentencing arguments. Drake v. Kemp

(Drake HI), 762 F.2d 1449 (11 Cir. 1985). At retrial, he was

convicted and sentenced to life. The Board freed Drake in

November 1987, finding that Campbell lied. Hodes, William

W., Executing The Wrong Person, 29 Loy. L.A. L. Rev. 1547;

Georgia’s Death Row; Waiting to die. Drake’s reaction: “If |

didn't have all those appeals, I’d be dead. I always thought

12

standard, it is unlikely that he would have been

granted a hearing, which produced evidence that

captured the attention of others and ultimately

led to his freedom.

Ill. The Majority’s Blind Adherence To This

New Standard And Its Failure To Apply Today’s

Science To The Undisputed Facts, Which

Demonstrate The Unreliability Of The Now-

Disavowed Identification Evidence, Is Further

Proof Of This Standard’s Failure.

When today’s accepted scientific truths are

applied to the undisputed’ circumstances

surrounding witnesses’ initial observations, the

reliability of their subsequent selection of Davis is

greatly undermined. As a matter of science, key

“estimator variables” and contaminating pre-

identification exposure produced highly unreliable

claims of recognition. Current science then

constitutes intrinsic evidence supporting the

recantations. The application of the purest

fabrication standard, however, caused the court to

ignore these studies, and fail to conduct hearings

where these scientific studies could help establish

the validity of the recantations. The court’s blind

adherence to the purest fabrication standard

arbitrarily prevented it from considering objective

proof supporting the reliability of the recantations.

A. The Court Ignored Recent Scientific

Studies Establishing That The Undisputed

Viewing Conditions And Circumstances Of This

I'd be electrocuted. I worried they’d mix up the papers, and

poof you’re gone.” /d.

13

Crime Precluded A Genuine Basis For

Subsequent Recognition.

The uncontroverted evidence at trial

demonstrated that the sudden unexpected assault

and murder® in a poorly lit parking lot late at

night’? was witnessed by people under the extreme

stress of being in the line of fire.? Scientific

research demonstrates that heightened stress at

the moment of perception reduces identification

accuracy. Deffenbacher, K.A., et al., A Meta-

Analytic Review of the Effects of High Stress on

Eyewitness Memory, 28 Law and Human Behavior

687 (2004). Science also confirms that which

should appeal to common sense: event-related

factors (e.g., duration, awareness of crime,

lighting, distance/vantage point, the presence of a

weapon, high levels of stress, intoxication, and

other physical or mental limitations) affect the

ability of a witness to accurately encode

identifying information. Wells, G. L. & Olson, E.

(2003). Eyewitness identification. Annual Review

of Psychology, 54, 277-295.

° Officer MacPhail was shot within seconds of the assault on

Young. The shooter fled immediately.

7 At the time of shooting the Burger King was closed and its

exterior lights were turned off. The parking lot was dimly lit.

According to Leu Bishop, Burger King’s manager, there was

enough light so that “you’re not going to walk into anything

or trip over anything.” But it was too dark for Bishop to

recognize the familiar officer as he laid in the parking lot

until Bishop moved close enough (10-15 feet away) to see

Officer MacPhail’s uniform. The shooting occurred in “the

worst area in the parking lot,” as the street light on the

corner was bdilocked by a large tree located directly

underneath. According to Young, “I couldn’t see, you, it was

just dark, right the particular spot we were standing.”

8 Moreover, many were also intoxicated and/or injured.

14

1. The Extraordinarily Short Duration Of

The Assault Of Young And Shooting Of

Officer MacPhail Did Not Provide

Sufficient Time To Encode Crucial Detail.

Common. sense. dictates, and_ science

confirms, that the shorter the exposure the more

difficult to encode enough information for later

recognition. Memon, A., Hope, L., & Bull, R.,

Exposure duration: Effects on eyewitness accuracy

and confidence, 94 British J. of Psychol. 339

(2003). The degree of a witness’ attention as

events unfold is as critical to the witness’ ability

to take in meaningful information as the duration

of observation and proximity to the criminal. See

Leippe, M.L. et al, Crime Seriousness as a

Determinant of Accuracy in Eyewitness

Identifications, 63 J. of Applied Pyschol. 345

(1978). If a witness is unaware that a crime is

occurring, interactions or mere proximity will

often not leave a lasting impression. A recent

Cornell study illustrates the concept of inattention

blindness. Researchers conducted a series of

experiments in which pedestrians who were giving

directions to someone posing as a lost tourist did

not notice when, midway through the exchange,

the sham tourist was replaced by another person

altogether. Blind to Change, Even as It Stares Us

in the Face, New York Times, April 1, 2008; see

also Loftus, E., Eyewitness Testimony, Harvard

University Press, 25-27 (1996). Duration, then,

should be assessed in conjunction with the time

period that the witness is paying attention to the

perpetrator.

In this case, the pistol-whipping of Young

first captured the attention of the stranger

15

eyewitnesses. Officer MacPhail was shot within

seconds of the assault on Young. The shooter fled

immediately. According to Daniel Kinsman, one

of Sanders’ military friends in the van, the

incident was “over as soon as it began.” App. A.

at 13a.

2. The Presence And Discharge Of A Gun

Took Focus Away From The Face Of The

Shooter.

Forensic psychologists have documented

that, during a crime, the witness’ attention is

drawn to any visible weapon and away from the

culprit’s facial and physical characteristics.

Experiments involving videotaped robberies with

some culprits brandishing a gun and others

concealing it repeatedly demonstrated that

eyewitness identifications were less accurate when

the gun was brandished. This is known as

weapon focus effect. In a recent meta-analysis,

Dr. Nancy Steblay examined 19 studies of weapon

focus effect and found that “|t]he presence of a

weapon does make a significant difference in

eyewitness performance... particularly in crimes

of short duration in which a threatening weapon

is visible.” Steblay, N. M. A Meta-Analytic Review

of the Weapon Focus Effect, 16 Law And Hum.

Behav. 413, 420-21 (1992).

In this case, even if witnesses had not

sought cover, they would have instinctively

focused on the gun, not the shooter’s face. Given

the extremely short duration of this volatile crime

in which witnesses were immediately made aware

of the gun, the opportunity for meaningful

observations were significantly diminished. The

16

well-documented weapon focus effect supports the

probability that some influence other than true

recognition accounted for the witnesses’

subsequent selection.

In all probability, most witnesses did not

maintain a vantage point once the gun was

displayed. Common sense dictates that people

will remove themselves from the line of fire.? The

brief duration, the weapon, and the instinctive

inclination toward self-preservation, renders

Murray and Williams’ purported ability § to

“recognize” the shooter incredibl!°

3. It Is Scientifically Impossible For

Ferrell To Subsequently Recognize A

* Indeed, Redd’s threat of producing a gun caused Murray

and two unidentified companions to run for cover. The

sound of gunfire caused Antoine Williams to seek shelter

under the dash of his car. Securing their safety ended their

period of observation.

© Though Stephen Sanders claimed he maintained visual

contact after, shots were fired, he candidly admitted that

night that his observations were insufficient to enable

identification. Sanders was seated in a van with seven U.S.

Air Force buddies ordering food at the drive thru window

after a night of partying. As Young was trying to get

assistance from Sanders and his friends in the van, shots

suddenly rang out. Bishop told his employees and the van’s

occupants to duck for cover.

Even assuming Sanders did not heed Bishop’s advice, his

inability to describe the shooter beyond “a black man in

twenties” and his unequivocal assertion that he “wouldn't

recognize them again” rendered his in-court identification

two years later extremely implausible. Indeed, his selection

is the most troubling of all in a field where the competition

for that distinction is at an all-time high. Something other

than true recognition was at work

17

Stranger Seen In A Dimly Lit Parking Lot

From 160 Feet Away.

A famous woman is depicted

here as she would appear from

a distance of 172 feet. The lack

of sufficient visual details

makes identifying her a

challenge beyond our human

capacity. As Dr. Geoffrey

Loftus states:

When you see anything at a distance the

human visual system starts to lose small

details. The greater the distance the coarser

the detail you lose... . . At 10 feet you might

not be able to see individual eyelashes on a

person's face. At 200 feet you would not

even be able to see a person's eyes. At 500

fect you could see the person's head but

just as one big blur. There is equivalence

between size and blurriness. By making

something smaller you lose the fine detail.

Loftus, G.R. & Harley, E.M., Why Is It Easier

to Identify Someone Close Than Far Away,

12 Psychonomic Bull. & Rev. 43 (2005).

As the photos below illustrate, a known face

may be clearly recognizable at 5 feet or even 43

feet; however, at distance of 100 feet or more, the

fine detail necessary for true recognition is lost so

that even an easily recognizable face like Julia

Roberts’ is but a blur. If it were possible to

recognize the blurred woman, then her name

would have occurred to the viewer immediately.

18

. ie ) oe 2;

172 feet 100 feet 43 feet

5 feet

Yet, Dorothy Ferrell testified at Davis’ trial

that she recognized a complete stranger whom she

viewed briefly from a distance of at least 160 feet!!

as he quickly moved about in a poorly lit parkig

lot with a group of other black men. The only

plausible explanation is that she did not recognize

Davis from the lot that night, but, rather,

accepted Detective Wilson’s suggestion when he

displayed a single photo of Davis to her a few days

after the shooting.

11 Ferrell claimed she witnessed the event from the edge of

the hotel parking lot, near a palm tree, where she remained

for the duration. The Thunderbird Inn is located on the

southeast corner of the intersection of Oglethorpe Avenue

and Fahm Street. The Burger King parking lot is located on

the northeast corner of the intersection of Oglethorpe

Avenue and Fahm Street. There are two small palm trees at

the hotel parking edge; one is 160 feet away from the

location of shooting, the other 200.

19

B. The Court Did Not Consider The

Scientifically Supported Probability That

Witnesses Selected The Police Suspect During

A Suggestive Identification Process, Which

Created An Ideal Situation For Memory Source

Error And False Identifications.

A true identification occurs when the

witness recognizes the person as the actual

perpetrator. When the police expose witnesses to

the suspect’s image prior to the “official”

identification procedure and make it abundantly

clear that he is the police suspect then the

witnesses’ subsequent selection of that man is

likely the product of the unnecessary suggestivity

and/or memory source error. In other words, the

witnesses did not recognize the suspect; instead,

they simply tollowed the lead of the police who

were determined to build a case against Troy

Davis. Depending on the witnesses’ motives, they

may have genuinely believed that the police

suspect was the man that they observed (memory

source error) or they may have acquiesced out of

self-interest and implicated the suspect to

appease the police (false identification). The

court’s categorical disqualification of the

recantations caused it to fail to consider this

undisputed and objective corroboration of those

recantations.

20

1. The Saturation Of Davis’ Image

Exposed Witnesses To A Contaminant

That Increased The Probability Of

Memory Source Error And Rendered The

Witnesses’ Later Selection Of Davis

Unreliable. :

Memory source error occurs when a witness

incorrectly attributes a familiar face to the wrong

source because he unconsciously transfers

memory of the familiar face from one context to

another. A witness who has seen the suspect

before may, through transference, mistake him for

the perpetrator in a later photo array. Decades of

scientific studies reveal that prior exposure to a

suspect’s image will make a subsequent selection

of the suspect more likely. Dysart, J., Lindsay,

R.C.L., Hammond, R., Dupuis, P, Mugshot

Exposure Prior to Lineup Identification:

Interference, Transference, and Commitment

Effects, 86 Journal of Applied Psychology 1280 —

(2001); Loftus, E. F. & Greene, E., Warning: Even

memory for faces may be contagious, 4 Law And

Hum. Behav. 323 (1980). Ross, D. F., Ceci, S. J.,

Dunning. D., & Toglia, M. P., Unconscious

transference and mistaken identity: Toward a

memory blending approach, (1994) in Ross, D. F.,

Read, J.D., & Toglia, M. P. (Eds.) Adult eyewitness

memory: Current trends and developments

(Cambridge University Press 1994).

Prompted by Redd Coles’ self-serving

accusation against Davis, the police acquired

Davis’ photo on the evening of August 19, 1989.

Yet, none of the witnesses were shown a photo

array until August 24, 1989; some were not asked

until August 28 and Sanders was not shown the

21

array until just before he testified at trial two

years later. In the 5-10 day delay preceding the

“official” identification process, the witnesses were

repeatedly exposed to Davis’ image in a variety of

mediums, including Wanted Posters prominently

displayed around Savannah and at their place of

employment; a single photo display by Detective

Wilson as he canvassed the neighborhood during

the days immediately following the shooting; and

through daily displays of Davis’ image in print and

televised news media.

Under those circumstances, even welil-

intentioned witnesses free from police intimidation

could unconsciously transfer the memory of

Davis’ image.

2. Non-Blind Administration After

Repeated Exposure To Davis’ Image

Destroyed The Procedural Protections Of

A Properly Conducted Photo Array And

Resulted In Unreliable Evidence Of Guilt.

The Court has long recognized the danger of

misidentification brought about by suggestive

identification procedures. United States v. Wade,

388 U.S. 218 (1967); Gilbert v. California, 388 U.S.

263 (1967); Stovall v. Denno 388 U.S. 293 (1967).

“The vagaries of eyewitness identification are well-

known; the annals of criminal law are rife with

instances of mistaken identification.” Wade, 388

U.S. at 228. In Stovall, the Court recognized that

the “the conduct of a confrontation” may be “so

unnecessarily suggestive and conducive to

irreparable mistaken identification” as to deny

due process of law. Stovall, 388 U.S. at 301-302.

22

Over the past four decades, scientific

research has provided indisputable proof that

eyewitnesses are vulnerable to police suggestions.

See Douglas & Steblay, Memory Distortion in

Eyewitness: A Meta-Analysis of the _ Post-

Identification Feedback Effect, 20 App. Cognitive

Psychol. 991 (2006); Bradfield, A.L & Wells, G.L.,

‘Good, You Identified the Suspect’: Feedback to

Eyewitnesses Distorts Their Reports Of the

Witnessing Experience, 83 J. Appl. Psychol. 360

(1998); Wells, G., et al, Eyewitness Identification

Procedures: Recommendations for Lineups and

Photospreads, 22 Law and Hum.Behav. 603

(1998). This research has led scientist to

recommend “blind” administration of the

identification process to avoid cues that may lead

a witness to falsely identify an innocent person.

Id.

“The reason for keeping the tester blind is

to prevent the tester from unintentionally

influencing the outcome of the results.” Wells,

2006 Wis. L. Rev. at 629. Double-blind

administration works not only to prevent the

investigator from influencing which person the

eyewitness picks, but also to prevent him from

“influencing the certainty of the eyewitness” by

giving the witness confirming feedback. Id.

it is also a well-settled that a single photo

display of the primary suspect, in the absence of

some emergency, is unnecessarily suggestive.

Manson vu Brathwaite, 432 U.S. 98 (1976).

In this -case, non-blind administrators

conducted the photo arrays on_ non-blind

witnesses — everyone knew who the suspect was.

The prior extensive exposure to and disclosure of

23

the police suspect eliminated the intended

procedural benefits of a using a photo array, i.e.,

to accurately measure true recognition. The

resulting selections are unreliable because they

do not measure the witnesses’ recognition of the

shooter. Even setting aside the consistent reports

of heavy-handed police intimidation, these

identifications were extraordinarily suggestive.

From aé_ scientific perspective, the suggestive

nature of this process coupled with the poor

viewing conditions rendered the _ selections

extremely unreliable evidence of guilt.

CONCLUSION

The innocent take their wrongful

convictions as they come. The means by which

they can prove their innocence is beyond their

control. In a case of perjured testimony, a

recantation may be as good as it gets. Assessing

the reliability of recanting witnesses can be

achieved without denying an innocent person his

right to relief. Georgia’s categorical exclusion of

seven recantations - which, taken together,

present a consistent explanation for the false trial

testimony and are objectively corroborated by

science — and its refusal to hear other compelling

evidence of innocence constitute a denial of due

process and may result in the execution of an

innocent man.

24

Respectfully submitted,

DEIRDRE O’CONNOR

P.O. Box 210787

Chula Vista, CA 91921

404-392-0249

Of Counsel for Amicus Curiae

Counsel of Record for Amicus Curiae

BARRY C, SCHECK

PETER J. NUEFELD

DAVID LOFTIS

Innocence Project, Inc.

100 Fifth Avenue, 3rd Floor

New York, NY 10011

212-364-5340

August 13, 2008

APPENDIX

APPENDIX A

I

SUMMARY OF BIOGRAPHIES OF MEMBERS

FROM THE INTERNATIONAL LEGAL

COMMUNITY JOINING IN THIS AMICUS BRIEF

Today, Europe is the only region in the world

where death penalty is no longer applied. All the

Council of Europe's 47 member states have

either abolished capital punishment or

instituted a moratorium on executions. This

determination to eradicate the death penalty was

reflected in Protocol No.6 to the European

Convention on Human Rights. In 2002, another

important step was taken with the adoption of

Protocol No. 13 on the abolition of capital

punishment in all circumstances, even for acts

committed in time of war. The Council as well as

he Euroepan Union have made abolition of the

death penalty a prerequisite for membership. As

a result, no execution has taken place on the

The European Bars, law societies and lawyer’s

professional organisation aim at playing a

leading role in the battle for abolition, believing

that the death penalty has no place in

democratic societies. They strongly supported

the adoption on the 18th of December 2007, by

the General Assembly, of a resolution calling for

a global moratorium on executions.

APPENDIX A

I]

Moreover, the entire european legal community

calls on nations that do impose the death

penalty to ensure they meet internationally

agreed minimum standards: on the safeguards

for those facing execution. Further, European

Bars and lav’ Societies permanently ask

countries to progressively restrict the use of the

death penalty, such as by reducing the number

of offences for which it may be imposed, and

calls on those States that have abolished the

practice to not reintroduce it.

England and Wales: Paul Marsh, President of

the Law Society of England and Wales, joins in

this brief on behalf of the Law Society, a

professional body representing more _ than

138,000 solicitors in England and Wales. It is

concerned to see the independence of the legal

profession, the rule or law and human rights up

held throughout the world.

Italy: Aldo Bulgarelli, a private Italian attorney

and member of the ABA, Consiglio Nazionale

Forense joins in his individual capacity. Mr.

Bulgarelli launched a special three year long

celebration by the lawyers all around the world

of the anniversaries of the Universal Declaration

on Human Rights (New York, 2008) and of the

European Declaration on Human Rights and

Fundamental Freedoms (Rome, 2010), passing

APPENDIX A

Il

through Paris. This project was accepted by the

, approved by ABA, Consiglio Nazionale Forense

and the Paris Bar.

Madrid, Spain: Juan Antonio Cremades,

President of the International Commission of the

Council of the Bar of Madrid joins on behalf of

that organization.

Paris, France: Charriére Bournazel, President

of the Paris Bar joins on behalf the organization

with almost 18 000 members. (Over 40% of the

lawyers practicing in France are members of the

Paris Bar.)

The Paris Bar is very often consulted when

human rights are threatened. Whenever,

throughout the world, basic freedoms are

challenged on the principles of a fair hearing are

hampered or-even non existent. It was at the

initiative of the President of the Paris bar that

the International Convention to safeguard Due

Process was signed by more than 58 bar

associations in over 40 countries in 1987. The

Paris Bar is a founding member of the World

Coalition against death penalty.