Amicus Curiae Brief — McCleskey v. Kemp

(\))

No. 84-6811 Supreme Court, US.

IN THE 8B

SUPREME COURT OF THE UNITED |.

JOSEPH r. SPANIOL, JR,

CLERK

October Term, 1986 —— =

WARREN MCCLESKEY,

Petitioner,

vs.

RALPH M. KEMP, Superintendent,

Georgia Diagnostic and

Classification Center,

Respondent.

On Writ of Certiorari to the

United States Court of Appeals

for the Eleventh Circuit

Brief of Amici Curiae State of California,

by John K. Van de Kamp, Attorney General,

and County of Los Angeles, by Ira Reiner,

District Attorney, In Support of Respondent

JOHN K. VAN DE KAMP IRA REINER

Attorney General District Attorney

of the State of of Los Angeles

California County

MICHAEL C. WELLINGTON GEORGE M. PALMER

Supervising Deputy Deputy District

Attorney General Attorney

SUSAN LEE FRIERSON HARRY B. SONDHEIM

Deputy A.torney [Counsel of Record]

General Head Deputy

District Attorney

3580 Wilshire 849 South Broadway

Boulevard, Suite 800 llth Floor

Los Angeles, Los Angeles,

California 90010 California 90014 r/

(213) 736-2236 (213) 974-5917 miu

TABLE OF CONTENTS

Page

INTEREST OF AMICI CURIAE 1

SUMMARY OF ARGUMENT 10

ARGUMENT 13

I

THE NATURE OF THE DECISION-MAKING

PROCESS IN A CONSTITUTIONAL CAPITAL

SENTENCING SYSTEM JUSTIFICS REQUIRING

MORE THAN THE LEVEL OF DISPARATE

IMPACT PROFFERED BY PLTITIONER TO

ESTABLISH A PRIMA FACIE CASE OF

PURPOSEFUL INVIDIOUS DISCRIMINATION

IN THE IMPOSITION OF THE DEAT!

PENALTY 13

A. The Strict Procedural

Safeguards Built Into

the Capital Sentencing

Process Justify Applying

the General Rule That

Disparate Impact Alone

Is Insufficient to

Support a Claim of

Discrimination 12

B. The Number, Complexity

and Subjectivity of

Factors Considered in

Capital Sentencing Make

Evidence of Disparate

Impact Alone Insufficient 23

C. Petitioner's Showing 28

TABLE OF CONTENTS

(Continued)

II

EACH CAPITAL CASE IS UNIQUE AND

THE COMPARISON OF ONE CASE WITH

ANOTHER, THROUGH THE USE OF

STATISTICAL ANALYSIS, CANNOT

REASONABLY BE EXPECTED TO YIELD

VALID RESULTS

A. Use of Generalized

Statistical Studies of

Capital Sentencing

Decisions Has Been

Uniformly Rejected by

Lower Courts

B. Capital Sentencing

Decisions Are Different

From Decisions In Other

Contexts

C. Critical Factors in

Capital Sentencing

Decisions Cannot Be

Accurately and Reliably

Measured

D. A Generalized Statistical

Analysis of Capital

Sentencing Decisions in

Georgia Cannot Explain

the Reasons Why Petitioner

Was Sentenced to Death

E. Conclusion

ii.

Page

34

35

43

44

TABLE OF CONTENTS

(Continued)

Page

II!

PETITIONER'S STATISTICAL ARGUMENT

UNDERMINES THE RIGHT TO TRIAL BY

JURY AND SUBSTITUTCS IN ITS PLACE

TRIAL BY STATISTICAL ANALYSIS 46

CONCLUSION 50

iii.

TABLE OF AUTHORITIES

Cases

Page

Abney v. United States,

431 U.S. 651 (1977) 47

Adams v. Wainwright,

709 F.2d 1443

(llth Cir. 1983) 34

Alexander v. Louisiana,

405 U.S. 625 (1972) 38,39

Arlington Heights v.

Metropolitan Housing Corp.,

429 U.S. 252 (1977) 15,16,17, 28,33

Batson v. Kentucky,

U.S. f

90 L.Ed. 2d 69 (1986) 16,17,20

Castaneda v. Partida,

430 U.S. 482 (1977) 24,33,38

Dothard v. Rawlinson,

433 U.S. 321 (1977) 40

Duncan v. Louisiana,

391 U.S. 14& (1968) 46,47

e

Eddings v. Oklahoma,

455 U.S. 104 (1982) 25-26

Gomillion v. Lightfoot,

364 U.S. 339 (1969) 15,31,33

Gregg v. Georgia,

428 U.S. 153 (1976) 19,49-50

iv.

TABLE OF AUTHORITIES

(Continued)

Cases

Hernandez v. Texas,

347 U.S. 475 (1954)

Keely v. Westinghouse

Electric Corp.,

404 F.Supp. 573

{E.D.Mo. 1975)

Lockett v. Ohio,

438 U.S. 586 (1978)

Lockhart v. McCree,

i

90 L.Ed.2d 137 (1986)

McCleskey v. Kenp,

753 F.2d 877

(llth Cir. 1985)

McCleskey v. Zant,

580 F.Supp. 338

(N.D. Ga. i984)

McCorquodale v. State,

211 S.E.2d 577 (Ga. 1974)

People v. Frierson,

25 Cal.3d 142 (1979)

People v. Harr.s,

28 Cal.3d 935 (1981)

People v. Jackson,

28 Cal.3d 264 (1980)

Page

27

50-51

25

39-40

21,27

27

37

4,42

TABLE OF AUTHORITIES

(Continued)

Cases

Page

Pulley v. Harris,

465 U.S. 37 (1984) 4,22,41

Shotwell Mfg. Co. v.

United States,

371 U.S. 341 (1963) 47

Smith v. Balkcom,

660 F.2d 573, as mod.

671 F.2d 858 (5th Cir. 1982) 34,35

Spaziano v. Florida,

468 U.S. 447 (1984) 19

Spinkellink v. Wainwright,

578 F.2d 582

(Sth Cir. 1978) 34

Stephens v. Kemp,

464 U.S. 1027 (1983) 34-35

Taylor v. Louisiana,

419 U.S. 522 (1975) 47

Teamsters v. United States,

431 U.S. 324 (1977) 33-34,44-45

Turner v. Murray,

U.S. td

90 L.Ed.2d 27 (1986) 20

Washington v. Davis,

426 U.S. 229 (1976) 16,17, 24-25, 33

vi.

TABLE OF AUTHORITIES

(Continued)

Cases

Page

Wayte v. United States,

0 L.ba. 20-547 (1985) 16

Yick Wo v. Hopkins,

118 U.S. 356 (1886) 15,31,32

Constitution

United States Constitution:

Eighth Amendment 5,13,17

Fourteenth Amendment 5,13,15,17

Sixth Amendment 46

Statutes

Cal. Pen. Code:

§ 190 3

Cal. Stats.:

1977, ch. 316 3

Rules

U.S. Supreme Court Rule 36.4 1

vii.

TABLE OF AUTHORITIES

(Continued)

Miscellaneous

Kleck, Life Su rt for Ailing

Hypotheses: odes of

Summarizing the Evidence for

Racial Discrimination in

Sentencing, 9 Law and Human

Behavior, at 271 (1985)

Baldus and Cole, Statistical

Proof of Discrimination,

at 5 (1980)

Walker & Walker, The English

Legal System, at 229 (1580)

viii.

Page

12

44

46

Amici curiae, the State of California

by John K. Van de Kamp, Attorney General,

and the County of Los Angeles (a political

subdivision of the State of California),

by Ira Reiner, District Attorney, submit

this brief in support of respondent pur-

suant to Rule 36.4 of the Rules of the

Supreme Court of the United States.

INTEREST OF AMICI CURIAE

John K. Van De Kamp, Attorney General

for the State of California and Ira |

Reiner, District Attorney for the aw

of Los Angeles, State of California,

jointly represent the People of the State

of California in the case of In re Earl

Lloyd Jackson, Crim. 22165, pending before

the California Supreme Court on petition

for writ of habeas corpus. Said case is

pending before a referee appointed by the

California Supreme Court to take evidence

on three issues, one of which is highly

pertinent to the instant case: Whether

"death sentences in California have been

discriminatorily imposed on the basis of

(1) the race of the victim; (2) the race

of the defendant; and/or (3) the gender of

the defendant."1/ Amici curiae have been

litigating just the discovery aspect of

this case for over two years. This order

for a reference hear‘ng was granted on

the basis of a statistical analysis of

limited data on death and 1life-without-

possibility-of-parole cases. It is the

theory of the defense in Jackson that a

Statistical analysis of death and li fe-

without-possibility-of parole cases will

show that persons who kill white victims,

1. All of the factual representations

made in this brief are based upon matters

set forth in the record as well as the

personal experiences of the government

attorneys who have litigated, before the

California Supreme Court and its appointed

referee, the petition for writ of habeas

corpus in the Jackson case.

2.

and male, black defendants are more likely

to be charged with and to receive the

death penalty because o these unconstitu-

tional racial/gender factors than are

persons in other racial/gender categories.

Defendant Jackson, who is black, was

charged with murdering two elderly white

women in two separate burglaries of their

residences in August and September 1977.2/

These charges made him eligible for the

death penalty pursuant to California Penal

3/

Code section 190 et seq.— After a jury

2. The race of defendant Jackson as well

as the race of his two victims are not

alleged or referred to in the Information.

3. The law under which Jackson was con-

victed and sentenced (Stats. 1977, Ch.

316), enacted August 11, 1977, requires

that one or more “special circumstances”

be alleged and found true by the trier of

fact before capital punishment may be

imposed. This law was repealed, and

essentially reenacted as modified, by the

"Briggs Initiative", passed by the voters

and effective November 7, 1978, princi-

pally to expand the number of special cir-

cumstances making a person eligible for

Capital punishment.

3.

4

verdict finding him guilty as charged and

imposing the death penalty, a judgment

was rendered in March 1979, sentencing him

to death. On his automatic appeal to the

California Supreme Court, the judgment was

affirmed and a concurrent petition for

writ of habeas corpus was denied. People

v. Jackson, 28 Cal.3d 264 (1980). The law

under which defendant Jackson was sen-

tenced has been held constitutional on its

face by this Court and the California

Supreme Court. Pulley v. Harris, 465 U.S.

37 (1984); People v. Frierson, 25 Cal.3d

142, 1-195 (1979).

Defendant Jackson filed a subsequent

petition for writ of habeas corpus, which

is the basis for the reference hearing

ordered by the California Supreme Court.

That court first ordered the reference

hearing to address two unrelated issues.

Defendant Jackson then moved to

expand the reference hearing on the theory

that a statistical analysis of capital

case data showed evidence of race and

gender discrimination in violation of the

Eighth and Fourteenth Amendments to the

Federal Constitution.

In support of his application, he

offered inter alia the declaration of

Dr. James Cole, Ph.D., a statistician, who

analyzed race and gender homicide data

published annually by the Bureau of

Criminal Statistics, a division of the

State Attorney General's Office, and data

Supplied by the State Public Defender's

Office. Using a total of three variables

(victim race, defendant race, defendant

sex) for all state-wide homicides, all

state-wide robbery murders, and all

robbery-murders in Los Angeles County, in

various combinations of what is princi-

: 5.

pally a cross tabulation analysis,

Dr. Cole concluded, without reference to

other circumstances of any cases, that

killers of white victims are five times

more likely to receive the death penalty

than killers of non-white victims.

Similar proportions were found for black

defendants when compared to other groups.

On this basis, the reference hearing

was ordered expanded to address the issue

of whether death sentences in California

have been discriminatorily imposed on the

basis of race of victim, race of defen-

dant, or gender of defendant.

Subsequently, defendant Jackson moved

for discovery of a virtual mountain of

statewide homicide data. Jackson

requested and was granted an order com-

pelling the District Attorney of Los

Angeles County to provide this data, even

though most of the data is a matter of

6.

Pt

public record, located outside the juris-

diction of Los Angeles County.4/

To comply with this order, amici sub-

poenaed homicide data from all of che

Superior Court Clerks in the 58 counties

throughout the State as well as other

entities such as the Administrative Office

of the Court. Because of the complex

nature of the task of obtaining even

limited data from the Clerks, and because

not one single Clerk's Office maintains

such data on computers, the process of

Obtaining the data was time-consuming and

expensive. Clerks" records in literally

thousands of cases had to be individually

identified, categorized and reviewed to

4. For a more detailed exposition of the

Order and what followed, the Court is

respectfully referred to Argument I of the

Brief of Amici Curiae, State of California

and County of Los Angeles, filed in the

case of Hitchcock v. Wainwright,

No. 85-6756, now pending before this Court

on Writ of Certiorari.

7.

obtain the required data. When, after six

months, this effort by severa. lawyers and

numerous Court Clerks and their staffs was

completed, the product of this effort was

found to be highly questionable in terms

of its quality. For example, some cate-

gories of data by the Los Angeles County

Clerk's Office are subject to a 50% plus

error rate and there is reason to believe

that data submitted by other Clerks from

throughout the State may aiso be subject

to error.

The discovery process itself heigh-

tens the interest of amici in the instant

case. Data gachering must take place

before a statistica’? challenge to the

death penalty can be mounted. The fact

that the data gathering process may differ

from one jurisdiction to another and the

fact that it may occur in the i sence of a

court order, as in the instant case, are

ee _

i

not significant. Regardless of who

gathers the data, it will be a time-

consuming, expensive process. This, in

turn, causes inordinate delay in the

judicial process. The quality of the

product of discovery (the data) may be

highly questionable. It may, as in

Jackson, be subject to significant error.

More importantly, as we set forth in

Argument II, infra, a capital case cannot

be reduced to statistical data which

accurately reflects how and why the jury

reached its decision.

Since the issues presented in the

instant case are so closely related to

those of the Jackson case, amici curiae

have concluded that the outcome of the

instant case will have a substantial

impact upon the administration of criminal

justice, and the death penalty law in

particular, throughout California.

9.

Amici's experience in the Jackson case has

made us familiar with the nature of the

discrimination issues and the arguments

offered by petitioner in this case.

SUMMARY OF ARGUMENT

When a S.iate imposes its death

penalty under a constitutional system

which by its very design minimizes any

risk of arbitrariness, generalized claims

Of arbitrariness in the imposition of that

State's death penalty should he fore-

Closed. Only a particularized and

factually supported claim of purposeful

invidious discrimination in the imposition

of petitioner's own death sentence should

have entitled petitioner to a hearing.

The nacure of the decision-making

process in a constitutionally valid

Capital-sentencing system justifies

requiring more than the evidence of dis-

parate impact proffered by petitioner to

10.

establish a prima facie case of purposeful

invidious race discrimination. This

decision-making process is distinctly

different from other decision-making

contexts in that it is more complex and it

contains many more safeguards against

purposeful discrimination. Thus, only

evidence of a stark pattern could ever

suffice to demonstrate a prima facie case

of discrimination in the imposition of the

death penalty.

Moreover, such a stark pattern of

race discrimination can never be demou-

Strated through the use of a statistical

analysis, no matter how sophisticated the

methodology. Each case is unique,

involving its own quantum of variables,

which are not comparable tc any other set

of variables. The factors found in the

evidence which move a jury to impose

capital punishment, even when identified,

ll.

are impossible to measure accurately.

Thus, no statistical analysis of capital

eligible cases will yield a valid result.

Finally, petitioner's argument, when

reduced to its essence, is an assault upon

the judicial system itself, for it postu-

lates that no jury's decision can ever be

trusted unless it passes the litmus test

of a statistical analysis. This proposi-

tion is unacceptable as a matter of

5/

constitutional law.

5. Petitioner cites many articles from

law reviews and other treatises to demon-

Strate that study after study has found

evidence of race discrimination in the

imposition of the death penalty speci fi-

cally, and in sentencing generally, in

Georgia and other states in the South.

Neither time nor space permits us the

luxury of answering the contentions made

in these many articles. However, a

recent, objective review of some of these

studies and their conclusion may be found

in Kleck, Life Support for Ailing

Hypotheses: Modes of Summarizing the

Evidence for Racial Discrimination in

Sentencing, 9 Law and Human Behavior, at

271 (1985).

i2.

ARGUMENT

I

THE NATURE OF THE DECISION-

MAKING PROCESS IN A CONSTI-

TUTIONAL CAPITAL SENTENCING

SYSTEM JUSTIFIES REQUIRING

MORE THAN THE LEVEL OF

DISPARATE IMPACT PROFFERED

BY PETITIONER TO ESTABLISH

A PRIMA FACIE CASE OF PUR-

POSEFUL INVIDIOUS DISCRIMI-

NATION IN THE IMPOSITION OF

THE DEATH PENALTY

Petitioner contends that he presented

a prima facie case of discrimination in

the imposition of the death penalty in

Georgia, that his proof was unrebutted and

that it was sufficient to support a

finding that Georgia's entire capital

sentencing system has been unconstitu-

tionally applied in violation of the

Eighth and Fourteenth Amendments on the

basis of the race of the victim. Amici

Curiae urge that petitioner's proof

consisted, at most, of little more than a

relatively small pattern of disparate

13.

impact which was legally insufficient to

constitute a prima facie case of discrimi-

nation, much less to support a finding

that Georgia's entire facially constitu-

tional capital sentencing system has been

applied unconstitutionally.

The essence of petitioner's submis-

sion is that the minimal standards

required to prove racial discrimination in

the context of job promotion or selection

of a jury should apply in the context of

capital sentencing. Brief for Petitioner

at 31-32. Amici curiae urge that such

minimal standards should not apply to

proof of racial discrimination in the

Capital sentencing context. As we shall

demonstrate, given the nature of the

decision-making process in a constitu-

tional capital sentencing system, the

general rule should be followed that,

when proof of disparate impact alone is

14.

offered, only "a pattern as stark as that

6/

in Gomillion — or Yick wo"/ will be

determinative on the issue of purposeful

invidious .iscrimination. Arlington

Heights v. Metropolitan Housing Corp., 429

U.S. 252, 266 and fn. 13 (1977).

Whenever governmental action is

claimed to be racially discriminatory in

violation of the Equal Protection Clause

of the Fourteenth Améndment, the

"invidious quality” of that action “must

ultimately be traced to a racially

6. In Gomillion v. Lightfoot, 364 U.S.

339 (1960), a state redefined a city's

boundaries in such a manner that the

formerly square-shaped city became a

28-sided city with the result that all but

four or five of 400 black voters were

disenfranchised while no white voters

were,

7. In Yick Wo v. Hopkins, 118 U.S. 356

(1886), a city administered an ordinance

in such a manner that permission to

operate a laundry was denied to all 200

Chinese who sought permission during the

same time period that such permission was

granted to 80 non-Chinese.

15.

discriminatory purpose." Washington v.

Davis, 426 U.S. 229, 240 (1976). The

burden of proof is on the claimant and the

showing required of the claimant to

establish a prima facie case of purposeful

invidious discrimination depends on the

context in which the claim arose. See

Batson v. Kentucky, U.S. P , 90

L.Ed.2d 69, 85-87 (1986); Wayte v. United

States, U.S. ; , 84 L.Ed. 2d 547,

556-557 and fn. 10 (1985); Washington v.

Davis, supra, 426 U.S. at 253 (Stevens, J.

concurring).

The general rule is that unless there

is a "pattern as stark as that in

Gomillion or Yick Wo, impact alone is not

determinative." Ariington Heights v.

Metropolitan Housing Corp., supra, 429

U.S. at 266. In some exceptional con-

texts, proof of a less than stark pattern

of disparate impact may demonstrate

16.

purposeful racial discrimination because

the very nature of the disputed decision-

making task itself makes a racially

disparate impact unexplainable except on

racial grounds. For example, “[p]Jroof of

systematic exclusion from the venire

raises an inference of purposeful discri-

mination because the ‘result bespeaks

discrimination.' [Citations.]" Batson v.

Kentucky, supra, 90 L.Ed.2d at 86; see

also Washington v. Davis, supra, 426 U.S.

at 238-245. “But such cases are i2-:e"

(Arlington Heights v. Metropolitan Housing

Corp., supra, 429 U.S. at 266), and

important distinction may be drawn to

separate them from those in which the

8/

general rule applies.—

8. Whether petitioner's claim is

presented in terms of an Eighth Amendment

cruel and unusual punishment concern or in

terms of a Fourteenth Amendment equal

protection concern, the basic thrust of

his claim is the same: governmental action

aVe

A. The Strict Procedural Safeguards

Built Into the Capital Sentencin

Process Justify Applying the

General Rule That Disparate Impact

Alone Is Insufficient to Support

a Claim of Discrimination

The decision-making process in the

imposition of the death penalty is unique.

Unlike any other decision-making process

(such as in selecting the venire, or

hiring or promoting employees or selling

Or renting a home, or drawing city voting

boundaries, or issuing permits for

laundries), the decision-making process

involved in the imposition of the death

penalty is replete with built-in proce-

dural safeguards against purposeful

invidious discrimination on the part of

the decision makers. First, a

constitutional capital sentencing system

has impacted in an invidiously discrimina-

tory manner on a group of which he is a

member. Thus, no matter how his claim is

clothed, petitioner should be required to

prove purposeful invidious discrimination.

18.

itself is “suitably directed and limited

so as to minimize the risk of wholly

arbitrary and capricious action.” Gregg

v. Georgia, 428 U.S. 153, 179 (1976). A

constitutional capital sentencing system

"can rationally distinguish between those

individuals for whom death is an appropri-

ate sanction and those for whom it is

not." Spaziano v. Florida, 468 U.S. 447,

460 (1984). Additionally, as a criminal

defendant, the capital defendant is

entitled to insist that both the venire,

from which the decision-making petit jury

will be drawn, and the decision-making

petit jury itself are selected pursuant to

non-discriminatory criteria. Even the

historically unfettered exercise of the

peremptory challenge is restricted (for

the prosecution at least), and the defen-

dant may question at trial the peremptory

exclusion of veniremen from the petit jury

19.

on account of their race. Batson v.

Kentucky, supra, U.S. at ; 90

L.Ed.2d at 87. As an added precaution

against purposeful invidious racial

discrimination on the part of the decision

makers, a Capital defendant is entitled to

have prospective jurors questioned on the

issue of racial bias if there is a risk of

racial prejudice infecting the sentencing

proceeding. Turner v. Murray, U.S.

’ ; 90 L.Ed.2d 27, 37 (1986). These

are but a sampling of the panoply of

safeguards protecting the capital

sentencing decision-making process.

The procedural safeguards against

purposeful invidious discrimination which

are an integral part of the capital

sentencing decision-making process readily

distinguish that process from the job

promotion and jury selection decision-

making processes. In those processes

20.

— ee

f

4

:

z

~

- there are no comparable built-in

safeguards against purposeful invidious

discrimination on the part of the decision

makers. Thus, an examination of their

decisions cannot begin with the same

confidence. The sa fequards present in

capital sentencing justify applying the

general rule that disparate impact alone

will not establish a prima facie case of

purposeful invidious discrimination

unless, as the Court of Appeals held in

the case below, the "disparate impact is

so great that it compels a conclusion that

the system is unprincipled, irrational,

arbitrary and capricious such that

purposeful [racial] discrimination...

can be presumed to permeate the system."

McCleskey v. Kemp, 753 F.2d 877, 892 (11th

Cir. 1985).2/

9. Contrary to petitioner's contenti

that the Court of Appeals "fashioned

unprecedented standards of proof” and

21.

The Court has previously recognized

and applied the principles underlying this

conclusion in Pulley v. Harris, supra, 465

U.S. at 51-54. Therein, the Court

addressed the issue whether mandatory

comparative proportionality review was an

essential element of a constitutional

capital sentencing system. The Court

found it was not, if the capital

sentencing system already had in place

other extensive procedural safeguards

against arbitrariness. Clearly, if a

System's in-place procedural safeguards

against arbitrariness are factors to be

considered in determining whether other

such safeguards will be required,

a fortiori, a system's in-place procedural

"announced the abolition of the prima

facie standard," the Court of Appeals in

the case below merely restated this

Court's general rule concerning proffers

of disparate impact evidence. See Brief

of Petitioner at 45, 62.

22.

eee .

safeguards against purposeful invidious

discrimination are also factors to be

considered in determining what standard of

proof should be applied to claims of

discrimination within that system.

B. The Number, Complexity and

Subjectivity of Factors

Considere n Capital Sentencing

Make Evidence of Disparate

Impact Alone Insufficient

In addition to the built-in proce-

dural safeguards which distinguish the

capital sentencing decision-making process

from other decision-making processes, the

greater number, complexity, subjectivity,

and interactivity of factors legitimately

affecting the capital sentencing decigions

further distinguish the capital sentencing

decision-making process from others.

Likewise, this difference also justifies

applying the general rule, in claims of

Capital sentencing discrimination, that

proof of disparate impact which reflects

23.

anything less than a stark pattern will

not establish a prima facie case of

purposeful invidious discrimination.

There are comparatively few factors

which can legitimately affect the deci-

sions whether to select a person to be a

part of the venire or a grand jury or

whether to hire an applicant for a posi-

tion as a police officer. Many of these

factors, such as the prospective grand

juror's county of citizenship or the

prospective police officer's score on a

Civil service vocabulary examination, are

also relatively simple, objective factors

for the decision maker to weigh. Further,

the same set of these factors are appli-

Cable in each decision whether to hire an

individual for a job or to select an

individual to sit on a grand jury. See

Castaneda v. Partida, 430 U.S. 482,

484-485 (1977); Washington v. Davis,

24.

supra, 426 U.S. at 232-236. In these

contexts, a racially disparate impact

evidenced by the decisions may itself hint

of purposeful invidious diccrimination

merely because of the sparsity of alterna-

tive explanations.

The situation is starkly different as

to decisions whether to sentence a person

to death. These decisions are affected by

countless legitimate factors, most of

which are complex and subjective. Each

individual case has its own set of uni yue

legitimate factors. Indeed in each

individual case, the capital-sentence

decision maker is required to take into

account “any aspect of a defendant's

character or record and any of the circum-

stances of the offense that the defendant

proffers as a basis for a sentence less

than death." Lockett v. Ohio, 438 U.S.

586, 604 (1978), emphasis added; Eddings

25.

v. Oklahoma, 455 U.S. 104, 111, 113-114

(1982).

It is patent that the specific set of

factors legitimately applicable to the

capital sentencing decision in one case

will not be the same set of factors legi-

timately applicable to the vast majority

of other capital sentencing decisions.

Petitioner does not bring to the Court's

attention a single Georgia case other than

his own in which the decision maker was

faced with evidence sufficient to find the

defendant guilty beyond a reasonable doubt

of killing a police officer to prevent his

Own arrest for the public-endangering

daytime armed robbery the officer caught

him committing in a retail store, in which

the robbery had been planned, in which the

defendant had accomplices, in which the

defendant boasted of the kiliing after his

arrest, in which no mitigating evidence

26.

was presented to the penalty decision

maker, and in which the defendant had

three prior convictions for armed robbery.

See McCleskey v. Kemp, supra, 753 F.2d at

882; McCleskey v. Zant, 580 F.Supp. 338,

345-346 (N.D. Ga. 1984). Consequently, in

the context of capital sentencing deci-

sions, a racially disparate impact of

those decisions does not itself suggest

purposeful invidious discrimination

because of the veritable ocean of alter-

native explanations.

Since a bare showing of a raciaily

disparate impact of capital sentencing

decisions does not begin to reflect the

thousands of unique factors considered by

the decision makers in all the cases, it

Cannot be said that such a disparate

impact "bespeaks discrimination." See

Hernandez v. Texas, 347 U.S. 475, 482

(1954). Thus, proof of disparate impact

27.

alone cannot suffice to demonstrate

purposeful racial discrimination in the

imposition of the death penalty.

Arlington Heights v. Metropolitan Housing

Corp., supra, 429 U.S. at 266.

om Petitioner's Showing

Petitioner's evidence, at most, was

nothing more than a showing of disparate

impact. The “bottom line" of his argument

is that, even when 39 legitimate

sentencing factors are taken into account,

killers of white victims in Georgia are on

an average over 4.3 times more likely to

receive a death sentence than similarly

10/

Situated killers of black victims.—

Brief for Petitioner at 55.

10. According to petitioner, Professors

Baldus and Woodworth collected data on

over 500 factors. Brief for Petitioner at

53. However, they considered onty 39

factors in what they called "their most

explanatory model", reflecting a logistic

regression analysis. Id. at 55, 80,

emphasis added. Although 230 variables

28.

The fact that Professor Baldus consi-

dered 39 legitimate sentencing factors

does not alter the disparate impact nature

of his showing. It is no more suggestive

of the conclusion that the race of the

victim influenced the entire capital

sentencing process in Georgia than it is

suggestive of the conclusion that other

legitimate factors, somehow associated

with the race of the victim, but distinct

from the race of the victim, influenced

the process. In fact, if any conclusion

can be drawn from Professor Baldus'

figures it is the latter one. When

Professor Baldus first examined Georgia's

Capital eligible cases and took into

were considered in another model,

reflecting a multiple regression analysis,

Professor Baldus apparently was of the

opinion that the "most meaningful summary

indicators of the magnitude of the racial

factors found" were those that he

calculated under the logistic regression

analysis. Id. at 80.

29.

consideration only the race of the victim,

he found that the death sentencing rate in

Georgia was nearly 11 times higher in

white victim cases than in black victim

cases. Id. at 52-53. This disparity

plummeted from 11 to 4.3 when only 39

legitimate race-neutral factors were

considered. Id. at 55. Thus, it would

appear that when only a fraction of the

innumerable possible legitimate capital

sentencing factors were taken into

account, the initial disparity was reduced

by more than half. This would suggest

that the race of victim disparity in

Georgia merely reflects that white victin:

in Georgia are more likely to be targets

of the aggravated type of killings which

gualify the killer for the death penalty.

In the enormously complex and subjec-

tive context of capital sentencing, this

"4.3" disparity based on a mechanical

30.

consideration of only 39 factors is

relatively small and does not present a

pattern resembling that found in Gomillion

or Yick Wo. Accordingly, petitioner did

not meet his burden of proof.

II

EACH CAPITAL CASE IS UNIQUE

AND THE COMPARISON OF ONE

CASE WITH ANOTHER, THROUGH

THE USE OF STATISTICAL

ANALYSIS, CANNOT REASONABLY

BE EXPECTED TO YIELD VALID

RESULTS

The defect in petitioner's showing

goes beyond his failure to demonstrate a

level of disparate impact sufficient to

make a prima facie case of purposeful

invidious discrimination in the imposition

of Georgia's death penalty. Amici curiae

urge that, in the unique context of capi-

tal sentencing decisions, a generalized

Statistical showing of disparate impact

does not even reliably show disparate

impact. While it may be theoretically

31.

possible to reduce capital sentencing

decisions to a statistical analysis, in

reality no statistical analysis of those

decisions will yield a valid result.

As petitioner characterizes it, his

argument is at heart simple and direct:

"Evidence of racial discrimination that

would amply suffice if the stakes were a

job promotion, or the selection of a jury,

should not be disregarded when the stakes

are life and death. Methods of proof and

fact finding accepted as necessary in

every other area of law should not be

jettisoned in this one." Brief for

Petitioner, at 31-32.

This contention demonstrates on its

face why it is unsound. The methods of

proof and factfinding accepted as neces-

sary in other areas of the law are not

jettisoned here. No one suggests that the

Principles established in Yick Wo,

32.

Gomillion, Arlington Heights, ard

Washington v. Davis, supra (to name just a

few pertinent cases) be ignored. Indeed,

they are relied upon more strongly than

ever. However, this is not a problem of

discrimination in employment, housing or

jury selection. Statistical analysis of

capital cases is almost infinitely more

Complex than the statistical analysis of

a job promotion or jury selection case.

Petitioner has failed to meet the

challenge of this argument. He masks over

the near insuperable difficulties he faces

with legal rretoric which fails to address

the problems of a statistical analysis of

Capital cases. If this were a simple case

and the data analyzed by petitioner's

experts were limited as it is in other

types of discrimination cases (e.g.,

Castaneda v. Partida, supra, 430 U.S. 482

[jury panel composition]; Teamsters v.

33.

United States, 431 U.S. 324 (1977)

[employment discrimination]), the problems

we outline below would be considerably

less important. But this is not a simple

case. As we shall show, there is

virtually no hope of success of showing

race discrimination through a statistical

analysis.

A. Use of Generalized Statistical

Studies of Capital Sentencin

Decisions oe Ee

Rejected by Lower Courts

Other courts which have addressed the

issue of whether such generalized

Statistical studies as were presented in

the instant case can succeed have

concluded such studies have virtually no

hope of success. Smith v. Balkcom, 660

F.2d 573, as modified 671 F.2d 858, 859-

860 (Sth Cir. 1982); Spinkellink v.

Wainwright, 578 F.2d 582, 614-615 (Sth

Cir. 1978); Adams v. Wainwright, 70S F.2d

1443, 1449 (llth Cir. 1983); Stephens v.

34.

Kemp, 464 U.S. 1027, 1030, n. 2 (1983)

(Powell, J., dissenting). As the Court

stated in Smith v. Balkcom, supra, 671

F.2d at 859: "The raw data selected for

the statistical study bear no more than a

highly attenuated relationship to capital

cases actually presented for trial in the

state. The leap from that data to the

conclusion of discriminatory intent or

purpose leaves untouched countless

racially neutral variables.”

> samt) Berens Eertsioe

In Other Contexts

Petitioner's argument that his

Statistical analysis is only different in

degree from statistical analyses in other

contexts such as jury panel composition

and employment discrimination fails to

address and appreciate the difficulties

inherent in a statistical analysis of

Capital cases. Upon reflection, it will

35.

be evident that there are qualitative

differences which distinguish statistical

analysis of capital cases from all other

types of cases considered thus far by the

courts.

Focusing first on employment discri-

mination cases reveals striking differ-

ences. In this context, the factors about

an employee's background that are relevant

to job performance are in general directly

comparable across employees. They include

education (does the employee have a high

school diploma or a college degree),

previous relevant job experience (has the

employee or applicant any previous secre-

tarial experience; can he/she drive a

large tractor-trailer truck), supervisor

evaluations (the employee's typing ability

is nonexistent, poor, excellent), and the

like. A comparison of these factors to

the factors pertinent to death penalty

36.

decisiois reveals there is no analogue in

employment discrimination cases to such

factors as the presence of torture in a

killing. See McCorgquodale v. State, 211

S.E.2d 577, 579-580 (Ga. 1974).

In addition, the decision makers and

the decisions in capital sentencing have

an entirely bine shesestes than in

employment cases. In the employment

Situation, one company hires or promotes

employees from a group of potential appli-

cants. In capital cases, there is a

separate decision maker (the trier of

fact) for each case rather than one

decision maker for all cases. In employ-

ment decisions, a subset of employees is

selected from a pool for a given number of

jobs. In capital cases, each case is

decided on its own merits. There is no

quota. In many hiring and licensing situ-

ations, all applicants have to pass

37.

exactly the same objectively scored test.

A charge of discrimination in this context

can be supported if the test does not meet

the standards for job relatedness. There

is no analogy to these situations in

capital cases.

Other contexts such as whether a

constitutionally racial balance has been

achieved in the formation of a grand jury

panel are even simpler than employment

discrimination cases. See, e.g.,

Alexander v. Louisiana, 405 U.S. 625

(1972); Castaneda v. Partida, supra.

Thus, relatively little statistical data

may result in a compelling case. For ~-

example, in Alexander, a black defendant

was able to show that although 21% of the

adult local population was black, only one

of 20 persons (5%) on the grand jury panel

was black and none of the twelve persons

on the grand jury which indicted him was

38.

black. This, together with evidence that

the jury commissioners knew the race of

all prospective jurors, was sufficient to

prove a prima facie case. Clearly, the

data in Alexander was reliable and the

Statistical analysis simple and

compelling.

Sentencing a person to death has

elements not shared by these other types

Of decisions. Thus, one cannot expect

Statistical analyses aimed at detecting

racial influences in death sentencing

decisions to be the same as those that

perform well in analyzing racial

influences in other more simple social

science contexts.

Cc. Critical Factors in Capital

Sentencing Decisions Cannot

Be Accurately and Reliably

Measured

This Court has indicated its concern

in evaluating the reliability of quantita-

tive evidence. Lockhart v. McCree,

39.

U.S. , ; 90 L.Ed. 2d 137, 144-147

(1986) [reliability of social science data

purporting to show conviction-proneness

of juries]; Dothard v. Rawlinson, 433 U.S.

321, 338 (1977) (concurring opinion of

Rehnquist, J.) [reliability of statistical

data purporting to show job disqualifica-

tion of males versus females by reason of

height and weigh requirements]. The

reliability of the quantitative evidence

submitted by petitioner in the instant

case is open to great doubt.

Petitioner has failed to adequately

respond to the issue of how a statistical

analysis can accurately and reliably

measure such factors as torture, prior

Criminal record, the circumstances of the

crime, the helplessness of the victim(s),

the life experience of the defendant, and

unusual aggravating factors. For example,

it is clearly inadequate to simply

40.

determine that torture was either present

Or not present because there are varying

degrees of torture. How does one compare

cases when the criminal records of the

defendants are not identical? Is the

helplessness of a young brutalized female

victim the same as the helplessness of a

bound and gagged police officer? How does

One compare the age and experience of a

22-year-old hostile, angry young male with

the age and experience of a 35-year old,

cold, calculating, sadistic middle-aged

male? How do unusual aggravating factors

enter into the equation? For example, in

the facts behind Pulley v. Harris, supra,

the defendant coolly finished eating the

hamburgers which two teenage boys had been

in the process of eating when the defen-

dant kidnapped and murdered them for use

of their car in a bank robbery. People v.

Harris, 28 Cal.3d 935, 943-945 (1981).

41.

How is such a factor measured? What

measurable impact did it have on the jury?

More importantly, how is it compared with

other unusual but vastly different aggra-

vating factors in other cases?ll/ What of

the attitude displayed by a defendant

during trial? Evidence of this factor in

the record may be sparse if it exists at

all. If it does exist, how can it be

measured in such a way that it can be

compared with evidence of another defen-

dant's attitude in a different case?

The courts have accepted as valid

Statistical analyses done in jury panel

composition and employment discrimination

11. A crucial case in point for amici is

the California case of People v. Jackson,

supra, 28 Cal.3d at 282-284, 303. During

the course of one of his burglary-murders,

Jackson raped his victim -- a 90-year old

female -- with a wine bottle. Later, he

described his victims to an acquaintance

as "'two old bags [who] were a nuisance

and... got what they deserved.'"

42.

cases but they have not accepted as valid

a statistical analysis of death penalty

cases which claimed to prove race discri-

mination in the imposition of the death

penalty because of these important

distinctions.

D. A Generalized Statistical

Analysis of Capital Sentencing

Decisions in Georgia Cannot

Explain the Reasons Why

Petitioner Was Sentenced to

Death

Finally, the premise upon which

petitioner's analysis is based deliber-

ately ignores what happened in his case.

A statistical analysis can never prove

Girectly that race was a factor considered

by the jury in petitioner's case. As

petitioner's foremost expert, David C.

Baldus, has stated in his book on the use

of statistics to prove discrimination:

"The primary limitation of quantita-

tive proof in the discrimination context

is its inability to support an inference

about the reasons for a particular deci-

sion, such as why a certain individual

43.

was hired or fired, or why a particular

law was adopted. Statistics can provide

powerful insight into general or long-

run behavior, but as for a particular

decision -- and many cases are concerned

with just one decision -- at best it can

provide a presumption by inferring from

the general to the particular." Baldus

and Cole, Statistical Proof of

Discrimination, at 5 (1980).

E. Conclusion

Amici is not impugning the role of

Statistical analyses in the law as a

general proposition. After all, this

Court has made it “unmistakably clear that

‘[s]tatistical analyses have served and

will continue to serve an important role'

in cases in which the existence of

discrimination is a disputed issue.

[Citations.]" Teamsters v. United States,

supra, 431 U.S. at 339. However, even in

the context of employment discrimination,

where the number of significant variables

operating is limited, this Court recog-

nizes that “statistics are not irrefut-

able; they come in infinite variety and,

44.

like any other kind of evidence, they

may be rebutted. In short, their useful-

ness depends on all of the surrounding

facts and circumstances. See, e.g., Hester

v. Southern R. Co., 497 F.2d 1374,

1379-1381 (CA5)." Id. at 340. Our point

is that no court has ever validated the

use of statistical énalyses for the

purpose of determining whether jury

verdicts of capital cases, which involve

hundreds if not thousands of significant

variables, are constitutionally defective

because the jury allegedly considered race

of victim or defendant in arriving at

their verdict. Capital cases are qualita-

tively different from other types of

discrimination cases: the number of

Significant variables operating in this

context is exponentially greater than in

any context heretofore considered by this

Court. For this reason, petitioner's

45.

analysis snould be rejected as without

merit.

II!I

PETITIONER'S STATISTICAL

ARGUMENT UNDERMINES THE

RIGHT TO TRIAL BY JURY

RND SUBSTITUTES IN ITS

PLACE TRIAL BY STATISTI-

CAL ANALYSIS

Petitioner's position is an attack on

the jury system itself.

The right to a jury trial is one of

the foremost protections of our legal

System. "It is fundamental to the

American scheme of justice." Duncan v.

Louisiana, 391 U.S. 145, 150 (1968). Its

lineage can be traced to the time of the

Norman Conquest. Walker & Walker, The

English Legal System, at 229 (1980). It

is a fundamental tenet that a criminal

defendant is entitled to a trial by an

impartial jury drawn from a representative

cross-section of the community. This

right is guaranteed by the Sixth Amendment

46.

to the Constitution. Taylor v. Louisiana,

419 U.S. 522, 530 (1975). This right,

thus, guarantees a defendant a trial by

his peers and, together with other funda-

mental rights, ensures a fair and just

determination of the cause. Duncan v.

Louisiana, supra, 391 U.S. at 151-156.

Although juries are generally

presumed to follow the law given to them

by the court (Abney v. United States, 431

U.S. 651, 665 (1977); Shotwell Mfg. Co. v.

United States, 371 U.S. 341, 367 (1963)),

petitioner's statistical analysis impli-

citly assumes this presumption to be

incorrect or inoperative. Notwithstanding

the absence of any jury instruction

permitting race to be considered by the

jury, petitioner's statistical analysis

rests on the conclusion that juries in

fact do consider race in determining

whether to impose the death penalty.

47.

Petitioner's statistical argument

postulates that the death penalty verdicts

reached by presumptively lawfully consti-

tuted juries, acting pursuant to constitu-

tionally valid laws, are constitutionally

invalid because statistically it can be

shown that persons who kill white victims

are more likely to receive the death

penalty than es who kill non-whites.

This argument strikes at the heart of

the judicial system. A jury's verdict,

based on literally hundreds (perhaps

thousands or millions) of individual bits

of information, arrived at through the

collective reasoning process of twelve

separate persons, is reduced to mere

Statistical data. Petitioner would, in

essence, substitute a statistical analysis

for the jury's verdict. The end result

would be the emasculation of the right to

a jury trial.

48.

Petitioner's argument postulates that

regardless of the observance of his

constitutional rights in the course of a

jury or court trial, conducted pursuant to

constitutionally valid laws, the verdict

is always subject to further statistical

analysis. Petitioner would, thus, create

a super appellate process whereby after a

verdict has been found legally valid on

appeal to the highest court of a state,

the verdict may nevertneless be tested

again by being subjected to a statistical

analysis. There is no constitutional

basis for such procedure and a hearing

aimed at subjecting jury verdict data in

capital cases to such analysis is contrary

co our system of criminal jurisprudence.

In his concurring opinion in Gregg v.

Georgia, supra, 428 U.S. at 226, Justice

White disposed of a similar argument:

"Petitioner has argued, in effect, that

no matter how effective the death

49.

penalty may be as a punishment, govern-

ment, created and run as it must be by

humans, is inevitably incompetent to

administer it. This cannot be accepted

as a proposition of constitutional law.

Imposition of the death penalty is

surely an awesome responsibility for any

system of justice and those who partici-

pate in it. Mistakes will be made and

discriminations will occur which will be

difficult to explain. However, one of

society's most basic tasks is that of

protecting the lives of its citizens and

one of the most basic ways in which it

achieves the task is through criminal

laws against murder. I decline to

interfere with the manner in which

Georgia has chosen to enforce such laws

on what is simply an assertion of lack

of faith in the ability of the system of

justice to operate in a fundamentally

fair manner."

CONCLUSION

Petitioner's statistical analysis of

capital cases and the conclusions he

reaches should be rejected. He has failed

to prove even a prima facie case of race

discrimination in the system. A fortiori

he has failed to prove race discrimination

by the jury in his case. In the instant

case, petitioner has used "statistics as a

50.

drunk man uses a lamp post -- for support

and not illumination." Keely v.

Westinghouse Electric Corp., 404 F.Supp.

573, 579 (E.D.Mo. 1975).

Petitioner asks this Court to apply a

standard for weighing evidence completely

out of context. Then he asks this Court

not just to accept but to validate a

Statistical analysis which inherently

fails to identify and accurately measure

all significant variables operating in

Capital cases. Finally, he asks this

Court to reject his individual sentence of

death on the anes theory that it must be

infected with race bias because a general

Statistical analysis suggests race bias in

other cases. All of this he asks be done

after decisions by the Georgia Supreme

Court, various federal courts, and this

Court, upholding the jury's sentence.

None of these requests have merit. To

51.

validate any of them would be contrary to

law previously laid down by this Court.

To grant them all wiil be tantamount to

rejecting one of the principal elements of

our judicial system: trial by jury.

Surely, such request must be denied as

without any foundation in the law. The

judgment of the Court of Appeals should be

affirmed.

Respectfully submitted,

John K. Van de Kamp, Ira Reiner,

Attorney General District Attorney of

of the State of Los Angeles County

California

Michael D. Wellington George M. Palmer

Supervising Deputy Deputy District

Attorney General Attorney

Susan Lee Frierson

Deputy Attorney

General

Harry B. Sondheim

[Counsel of Record]

Head Deputy

District Attorney

Appellate Division

9/86

52.