McCleskey v. Kemp Brief for Petitioner

This item is featured in:

Jack Boger Oral History

    No. 84-6811
IN THE

SUPREME COURT OF THE UNITED STATES 
October Term, 1935

WARREN McCLESKSY,
Petitioner,

- v . -

RALPH M. KEMP, Superintendent, 
Georgia Diagnostic & Classification 
Center.

On Writ of Certiorari to the United States 
Court of Appeals for the Eleventh Circuit

3RIEF FOR PETITIONER

JULIUS L . CHAMBERS 
JAMES M. NA3RIT, III 

* JOHN CHARLES SOGER 
DEVAL L. PATRICK 
VIVIAN 3ERGER

99 Hudson Street
Mew York, New York 10013
(212) 219-1900

ROBERT H. STROUP
141 Walton Street 
Atlanta, Georgia 30303

TIMOTHY K. FORD
600 Pioneer Building 
Seattle, Washington 38104

ANTHONY 3. AMSTERDAM
New York University 
School of Law
40 Washington Square South 
New York, New York 10012

^Attorney of Record
ATTORNEYS FOR PP T T T n 3NER



QUESTIONS PRESENTED
1. To make out a prima facie case 

under the Equal Protection Clause of the 
Fourteenth Amendment, must a condemned 
inmate alleging racial discrimination in 
a State's application of its capital 
sentencing statutes present statistical 
evidence "so strong as to permit no 
inference other than that the results 
are a product of racially discriminatory 
intent or purpose?"

2. Is proof of intent to 
discriminate a necessary element of an 
Eighth Amendment claim that a State has 
applied its capital statutes in an 
arbitrary, capricious and unequal 
manner?

3. Must a condemned inmate present 
specific evidence that he was personally 
discriminated against in order to obtain 
either Eighth or Fourteenth Amendment 
relief on the grounds that he was

i



sentenced to die under a statute
administered in an arbitrary or racially 
discriminatory manner?

4. Does a proven racial disparity 
in the imposition of capital sentences, 
reflecting a systematic bias against 
black defendants and those whose victims 
are white, offend the Eighth or 
Fourteenth Amendments irrespective of 
its magnitude?

5. Does an average 20-point racial 
disparity in death-sentencing rates 
among that class of cases in which a 
death sentence is a serious possibility 
so undermine the evenhandedness of a 
capital sentencing system as to violate 
the Eighth or Fourteenth Amendment 
rights of a death-sentenced black 
inmate?

ii



TABLE OF CONTENTS

QUESTIONS PRESENTED ................
CITATIONS TO OPINIONS BELOW ........
JURISDICTION ......................
CONSTITUTIONAL PROVISIONS INVOLVED 
STATEMENT OF THE CASE ..............

A. Course of Proceedings . . .
B. Petitioner's Evidence of
Racial Discrimination: The 
Baldus Studies ................
C. The Decisions Below . . . .

SUMMARY OF ARGUMENT ................
I. RACE IS AN INVIDIOUS AND 

UNCONSTITUTIONAL CONSIDERATION IN 
CAPITAL SENTENCING PROCEEDINGS . . .

A. The Equal Protection
Clause Of The 
Fourteenth Amendment 
Forbids Racial 
Discrimination In The 
Administration Of 
Criminal Statutes . .

B. The Eighth Amendment
Prohibits Racial Bias 
In Capital Sentencing

II. THE COURT OF APPEALS 
FASHIONED UNPRECEDENTED STANDARDS 
OF PROOF WHICH FORECLOSE ALL 
MEANINGFUL REVIEW OF RACIAL

i
1

2
2
2
2

7

18
23

32

32

41

iii



DISCRIMINATION IN CAPITAL
SENTENCING PROCEEDINGS ..........  45

A. The Court of Appeals
Ignored This Court's 
Decisions Delineating 
A Party's Prima Facie 
Burden Of Proof Under 
The Equal Protection 
Clause 47

B. The Court of Appeals
Disregarded This 
Court's Teachings On 
The Proper Role Of 
Statistical Evidence 
In Proving Intentional 
Discrimination . . . .  64

C. The Court Of Appeals 
Erroneously Held That 
Even Proven 
Patterns Of Racial 
Discrimination Will 
Not Violate The 
Constitution Unless 
Racial Disparities Are
Of Large Magnitude ... 77

D. The Court Of Appeals 
Erred in Demanding 
Proof of "Specific 
Intent To
Discriminate" As A 
Necessary Element Of 
An Eighth Amendment 
C l a i m .............. 97

III. THE COURT SHOULD EITHER GRANT 
PETITIONER RELIEF OR REMAND THE CASE 
TO THE COURT OF APPEALS FOR FURTHER 
CONSIDERATION UNDER APPROPRIATE LEGAL 
STANDARDS........................ 104

iv



TABLE OF AUTHORITIES
Cases Pages
Alabama v. Evans,

461 U.S. 230 ( 1983)............  95
Alexander v. Louisiana, 405 U.S.

625 ( 1972)...................  47,48
Avery v. Georgia, 345 U.S.

559 (1953)....................  76
Ballew v. Georgia,

435 U.S. 223 ( 1978)............  84
Batson v. Kentucky, ___U.S.__,

90 L.Ed. 2d 69
( 1986)...........  24,26,27,33,47,74

Bazemore v. Friday, ___U.S___,
___ L.Ed. 2d___ "
(1986)   27,29,64,73,75,78,106

Briscoe v. Lahue,
460 U.S. 325 (1983)............  38

Brown v. Board of Education,
346 U.S. 483 ( 1954)..........  32

Castaneda v. Partida,
430 U.S. 482
( 1977)...........  27,49,56,65,73,79

Chapman v. California,
386 U.S. 18 ( 1967).............  108

Cleveland Board of Education v.
LaFleur, 414 U.S. 632 (1974)...39,44

vi



Coble v. Hot Springs School District 
No.6, 682 F. 2d 721 (8th 
Cir . 1982 ).....................  66

Eastland v. TVA,
704 F. 2d 613 (11th Cir. 1983).. 66

Eddings v. Oklahoma,
455 U.S. 104 ( 1982)............  98

EEOC v. Ball Corp.,
661 F. 2d 531 (6th Cir. 1981)... 66

Furman v. Georgia,
408 U.S. 238
(1972)............  24,31,41,97,107

Gardner v. Florida,
430 U.S. 349 (1977).......  44,98,99

General Building Contractors Ass'n,
Inc. v. Pennsylvania, 458 U.S.
375 ( 1983).....................  34

Giglio v. United States,
405 U.S. 150 ( 1972)............  4

Godfrey v. Georgia,
446 U.S. 420
(1980)............  25,31,42,57,98

Graves v. Barnes,
405 U.S. 1201 ( 1972)...........  95

Gregg v. Georgia,
428 U.S. 153
( 1976)......... 25,40,4 2,57,59,89,98

Hazelwood School District v. United
States, 433 U.S. 299 (1977)....  65

Ho Ah Kow v. Nunan, 12 Fed. Cas. 252
(No. 6546) (C.C. D. Cal. 1879).. 34

vii



Hunter v. Underwood, ___U.S.___,
85 L. Ed, 2d 222
(1985)...................  33,60,91

Jones v. Georgia,
389 U.S. 24 ( 1967)............. 48

Loving v. Virginia,
388 U.S. 1 ( 1967).............. 35

Lyons v. Oklahoma,
322 U.S. 596 ( 1944)............ 108

McClesky v. State, 245 Ga. 108, 263
S.E. 2d 14, cert, denied, 449 
U.S. 891 ( 1980)................  5

McCleskey v. Zant,
454 U.S. 1093 ( 1981)........... 6

McLaughlin v. Florida,
379 U.S. 184
(1964)..................  34,35,39

Mt. Healthy City Board of Educ. v.
Doyle, 429 U.S. 274 (1977)...107,108

Neal v. Delaware, 100 U.S. 370
( 1881).........................  49

Nixon v. Herndon, 273 U.S. 536
(1927).........................  33

Papasan v. Allain, ___U.S.___,
___L. Ed. 2d__(1986).......  29,78

Parker v. North Carolina,
397 U.S. 790 ( 1970)............  108

Patton v. Mississippi,
332 U.S. 463 (1947)............  76

Personnel Administrator of 
Massachusetts v. Feeney,

viii



442 U.S. 256 
(1976)..... 35,74

Rhodes v. Chapman, 452 U.S. 337
( 1981 ).........................  99

Roe v. Wade, 410 U.S. 113
(1973).... 39,43

Rogers v. Lodge, 458 U.S. 613
(1982)......................  50,60

Rose v. Mitchell, 443 U.S. 545
( 1979).........................  33

Rozecki v. Gaughan,
459 F. 2d 6 (1st Cir. 1972).....  99

Segar v. Smith, 738 F. 2d 1249
(D.C. Cir. 1984)............  66,76

Skinner v. Oklahoma,
316 U.S. 535 ( 1942)............  39

Skipper v. South Carolina, ___U.S.
, 90 L. Ed. 2d 1 ( 1986)......  104

Smith v. Texas, 311 U.S. 128
( 1940).....................  32,45

Spain v. Procunier, 600 F. 2d 189
(9th Cir. 1979)................  99

Stanley v. Illinois,
405 U.S. 645 (1972).........  39,44

Strauder v. West Virginia,
100 U.S. 303 (1880).........  34,41

Sullivan v. Wainwright, 464 U.S. 109,
(1983).........................  93

Teamsters v. United States,
431 U.S. 324 ( 1977)...........  65

ix



Texas Dep't of Community Affairs v.
Burdine, 450 U.S . 248
(1981)......... ......  29,48,75,

Turner v. Murray, U.S .
90 L. Ed. 2d 27
(1986)......... 24,33,56,76,1

Vasquez v. Hillery, U.S.
88 L. Ed. 2d 598
(1986)........................

Village of Arlington Heights v.
Metropolitan Housing Development
Corp., 429 U.S. 252
(1977)................  28,50,52,

Vuyanich v. Republic National Bank,
505 F. Supp. 224 (N.D. Tex.
1980) vacated on other grounds,
732 F. 2d 1195 (5th Cir.1984)...

Wainwright v. Adams, 466 U.S. 964
(1984).........................

Wainwright v. Ford, 467 U.S. 1220
(1984).........................

Washington v. Davis,
426 U.S. 229
(1976).............  27,32,47,49,

Wayte v. United States, ___U.S.___,
84 L. Ed. 2d 547 (1985)........

Whitus v. Georgia, 385 U.S. 545
( 1967 ).......................  47 ,

Wilkins v. University of Houston,
654 F. 2d 388 (5th Cir. 1981), 
vacated and remanded on other 
grounds, 459 U.S. 809 (1982)....

76

03

24

59

66

93

93

74

49

56

66

X



Wolfe v. Georgia Ry. & Elec. Co.,
2 Ga. App. 499, ___, 58 S.E. 899
(1907)....................... 61

Wong Sun v. United States,
371 U.S. 471 ( 1963)......... 108

Yick Wo v. Hopkins,
118 U.S. 356 (1886).........33,56

Zant v. Stephens, 462 U.S. 862
(1983).......................43,57

Zant v. Stephens, 456 U.S. 410
(1982) (per curiam).......... 43

Statutes

28 U.S.C. § 1254 (1)............  2
28 U.S.C. § 2241 (c) (3)......... 106
Rule 406, F. Rule Evid.............  72
Former Ga. Code Ann. § 27-2534.1

(b)(2).......................  5
Former Ga. Code Ann. § 27-2534.1

(b)(8).......................  5

Other Authorities

D. Baldus & J. Cole, Statistical
Proof of Discrimination (1980).. 8

Baldus, Pulaski & Woodworth,
Arbitrariness and Discrimination 
in the Administration of the

xi



Death Penalty: A Challenge to State 
Supreme Courts, 15 Stetson L.
Rev. 133 ( 1986)...............  8

Baldus, Pulaski & Woodworth,
Comparative Review of Death 
Sentences: An Empirical Study of 
the Georgia Experience, 74 J.
Crim. Law & Criminology 661 
(1983)...................... ; . . 8

Baldus, Pulaski, Woodworth & Kyle, 
Identifying Comparatively 
Excessive Sentences of Death: A 
Quantitative Approach, 33 Stan.
L. Rev. 1 ( 1977).............  8

Baldus, Woodworth & Pulaski,
Monitoring and Evaluating 
Contemporary Death Sentencing 
Systems: Lessons from Georgia,
18 U.C. Davis L. Rev. 1375
(1985).........................  8

Barnett, Some Distribution Patterns 
for the Georgia Death Sentence,
18 U.C. Davis L. Rev. 1327 
( 1985).........................  51

Bentele, The Death Penalty in Georgia: 
Still Arbitrary, 61 Wash. U.L.Q.
573 (1985).....................  59

Bowers & Pierce, Arbitrariness and
Discrimination Under Post-Furman 
Capital Statutes, 26 Crime &
Delinq. 563 ( 1980)..............  51

Finkelstein, The Judicial Reception of 
Multiple Regression Studies in Race 
and Sex Discrimination Cases, 80 
Colum. L. Rev. 737 (1980).......  66

Fisher, Multiple Regression in Legal
xii



Proceedings, 80 Colum. L. Rev. 737 
(1980)..........................  66

Gross, Race and Death: The Judicial 
Evaluation of Evidence of 
Discrimination in Capital 
Sentencing, 18 U.C. Davis L. Rev.
1275 ( 1985)................... 81,90

Gross & Mauro, Patterns of Death:
Disparities in Capital Sentencing 
and Homicide Victimization, 37 
Stan. L. Rev. 27 (1985).... 51

H. Kalven & H. Zeisel, The American
Jury (1966 )....................  84

B. Nakell & K. Hardy,
The Arbitrariness of the Death 
Penalty, (1986)
(forthcoming).................  100

Report of the Joint Committee on 
Reconstruction at the First 
Session, Thirty-Ninth Congress,
( 1866).........................  37

Statement of Rep. Thaddeus Stevens, 
Cong. Globe, 39th Cong., 1st 
Sess. 2459 (1966); Accord, 
statement of Sen. Pollard, Cong. 
Globe, 39th Cong., 1st Sess.
2961 ( 1866)....................  37

Wolfgang & Riedel, Race, Judicial
Discretion and the Death Penalty,
407 Annals 119 (May 1973)......  51

Wolfgang & Riedel, Race, Rape, and the 
Death Penalty in Georgia, 45 Am. J. 
Orthopsychiat. 658 (1975)......  51

xiii



No. 84-6811
IN THE

SUPREME COURT OF THE UNITED STATES 
October Term, 1985

WARREN McCLESKEY,
Petitioner,

- v. -
RALPH M. KEMP, Superintendent, 
Georgia Diagnostic & 
Classification Center.

On Writ of Certiorari to the 
United States Court of Appeals for 

the Eleventh Circuit

BRIEF FOR PETITIONER

CITATIONS TO OPINIONS BELOW
The opinion of the United States 

Court of Appeals for the Eleventh 
Circuit is reported at 753 F .2d 877 
(11th Cir. 1985)(en banc). The opinion 
of the United States District Court for 
the Northern District of Georgia is 
reported at 580 F. Supp. 338 (N.D. Ga.



1984) .
JURISDICTION

The judgment of the Court of Appeals 
was entered on January 29, 1985. A
timely motion for rehearing was denied 
on March 26, 1985. The Court granted
certiorari on July 7, 1986. The
jurisdiction of this Court is invoked 
pursuant to 28 U.S.C. § 1254(1).

CONSTITUTIONAL 
PROVISIONS INVOLVED

This case involves the Eighth and 
the Fourteenth Amendments to the 
Constitution of the United States.

STATEMENT OF THE CASE
A. Course of Proceedings

Petitioner Warren McCleskey is a 
young black man who was tried in the 
Superior Court of Fulton County, 
Georgia, for the murder of a white 
police officer, Frank Schlatt. The 
homicide occurred on May 13, 1978 during 
an armed robbery of the Dixie Furniture

2



Store in Atlanta. In a statement to 
police, petitioner admitted that he had 
been present during the robbery, but he 
denied that he had fired the shot that 
killed Officer Schlatt. (Tr.T. 453).1

Petitioner was tried by a jury 
comprised of eleven whites and one 
black. (Fed.Tr.1316). The State's case 
rested principally upon certain disputed 
forensic and other circumstantial 
evidence suggesting that petitioner may 
have fired the murder weapon, and upon

1 Each reference to the trial 
transcript will be indicated by the 
abbreviation "Tr.T," and to the 
federal habeas corpus transcript, by the 
abbreviation "Fed.Tr."

References to the Joint Appendix 
will be indicated by the abbreviation 
"J.A." and to the Supplemental Exhibits, 
by "S.E." Petitioner's exhibits
submitted to the District Court during 
the federal hearing were identified 
throughout the proceedings by the 
initials of the witness during whose 
testimony they were introduced, followed 
by an exhibit number. For example, the 
first exhibit introduced during the 
testimony of Professor David Baldus was 
designated "DB 1."

3



purported 
defendant 
Evans. 2

confessions made to a co- 
and to a cellmate. Offie

2 The co-defendant, Ben Wright, had 
a possible personal motive to shift 
responsibility from himself to 
petitioner. Inmate Evans testified 
without any apparent self-interest that 
petitioner had boasted to him in the 
cell about shooting Officer Schlatt. 
However, the District Court later found 
that Evans had concealed from 
petitioner's jury a detective's promise 
of favorable treatment concerning 
pending federal charges. Holding that 
this promise was "within the scope of 
Giqlio [v. United States, 405 U.S. 150 
(1972)]," (J .A . ), the District Court 
granted petitioner habeas corpus relief: 
"[G]iven the circumstantial nature of 
the evidence that McCleskey was the 
triggerman who killed Officer Schlatt 
and the damaging nature of Evans' 
testimony as to this issue and the issue 
of malice . . . the jury may reasonably 
have reached a different verdict on the 
charge of malice murder had the promise 
of favorable treatment been disclosed." 
(J.A . ).

The Court of Appeals reversed, 
holding that the detective's promises to 
witness Evans were insufficiently 
substantial to require full disclosure 
under Giglio, and that any errors in 
concealing the promises were harmless. 
(J.A. ). Five judges dissented, 
contending that Giqlio had plainly been 
violated; four of the five also believed 
that the concealed promise was not

4



The jury convicted petitioner on all 
charges. Following the penalty phase,
it returned
aggravating
recommending

a verdict finding two 
circumstances * 3 and 

a sentence of death. On 
October 12, 1978, the Superior Court
imposed a death sentence for murder and 
life sentences for armed robbery. (J.A.)

After his convictions and sentences 
had been affirmed on direct appeal, 
McClesky v. State, 245 Ga. 108, 263 
S.E.2d 146, cert, denied, 449 U.S. 891 
(1980), petitioner filed a petition for 
habeas corpus in the Superior Court of 
Butts County, alleging, inter alia, that

harmless. (J.A. ) (Godbold, Ch.J., 
dissenting in part); id. at (Kravitch, 
J., concurring).

3 The jury found that the murder 
had been committed during an armed 
robbery, former Ga. Code Ann. § 27- 
2534.1(b)(2)(current version O.C.G.A. § 
17-10-30(b)(2)), and that it had been 
committed against a police officer. 
Former Ga. Code Ann. § 27- 
2534.1(b)(8)(current version O.C.G.A. § 
17-10-30(b)(8)).

5



he had been condemned pursuant to 
capital statutes which were being 
"applied arbitrarily, capriciously and 
whimsically" in violation of the Eighth 
Amendment (State Habeas Petition, SI 10), 
and in a "pattern . . .  to discriminate 
intentionally and purposefully on 
grounds of race," in violation of the 
Equal Protection Clause. (Id. SI 11). 
The Superior Court denied relief on 
April 8, 1981.

After unsuccessfully seeking review 
from the Supreme Court of Georgia and 
this Court, see McCleskey v. Zant, 454 
U.S. 1093 (1981)(denying certiorari), 
petitioner filed a federal habeas corpus 
petition reasserting his claims of 
systemic racial discrimination and 
arbitrariness. (Fed. Habeas Pet. SHI 45- 
50; 51-53). The District Court held an 
evidentiary hearing on these claims in 
August of 1983.

6



The evidence presented by petitioner
at the federal hearing is integrally
related to the issues now on certiorari.
In the next section, we will summarize
that evidence briefly; fuller discussion
will be included with the legal
arguments as it becomes relevant.̂
B. Petitioner's Evidence of Racial 

Discrimination: The Baldus Studies
Petitioner's principal witness at

the federal habeas hearing was
Professor David C. Baldus, one of the
nation's leading experts on the legal 4

4 Discussion of the research 
design of the Baldus studies appears at 
pp. 50-55 infra. Statistical methods 
used by Professor Baldus and his 
colleagues are described at pp. 66-71. 
The principal findings are reviewed at 
pp. 80-89.

A more detailed description of 
the research methodology of the Baldus 
studies —  including study design, 
questionnaire construction, data 
sources, data collection methods, and 
methods of statistical analysis —  can 
be found in Appendix E to the Petition 
for Certiorari, McCleskey v. Kemp, No. 
84-6811.

7



use of statistical evidence. 5

Professor Baldus testified concerning 
two meticulous and comprehensive studies 
he had undertaken with Dr. George 
Woodworth.5 6 and Professor Charles

5 Professor Baldus is the co­
author of an authoritative text in the 
field, D.Baldus & J. Cole, Statistical 
Proof of Discrimination (1980), as well 
as a number of law review articles 
relevant to his testimony in this case. 
Baldus, Pulaski, Woodworth & Kyle,
Identifying____Comparatively Excessive
Sentences of Death, 33 Stan. L. Rev. 601 
(1980); Baldus, Pulaski & Woodworth, 
Comparative Review of Death Sentences: 
An Empirical Study of the Georgia 
Experience, 74 J. Crim. Law &
Criminology 661 (1983); Baldus,
Woodworth & Pulaski, Monitoring and 
Evaluating Contemporary Death Sentencing 
Systems: Lessons From Georgia, 18 U.C.
Davis L. Rev. 1374 (1985); Baldus,
Pulaski & Woodworth, Arbitrariness and 
Discrimination in the Administration of 
the Death Penalty: A Challenge to State 
Supreme Courts, 15 Stetson L. Rev. 133
(1986).

6 Dr. Woodworth is Associate
Professor of Statistics at the 
University of Iowa and the founder of 
Iowa's Statistical Consulting Center. 
(Fed.Tr.1203-04). He has consulted on 
statistical techniques for over eighty 
empirical studies (id. 1203-04) and has 
taught and written widely on statistical 
issues. (GW 1).

3



Pulaski.7 Professor Baldus explained
that he had undertaken the studies to 
examine Georgia's capital sentencing 
experience under its post-Furman 
statutes. The studies drew from a 
remarkable variety of official records 
on Georgia defendants convicted of 
murder and voluntary manslaughter, to 
which Professor Baldus obtained access 
through the cooperation of the Georgia 
Supreme Court, the Georgia Board of 7

7 Professor Charles A. Pulaski, 
Jr., is Professor of Law at Arizona 
State University College of Law, 
specializing in criminal procedure. 
Professor Pulaski did not testify during 
the federal hearing.

Petitioner also presented expert 
testimony from Dr. Richard A. Berk, 
Professor of Sociology and Director of 
the Social Process Research Institute at 
the University of California at Santa 
Barbara, and a nationally prominent 
expert on research methodology, 
especially in the area of criminal 
justice research. He was a member of 
the National Academy of Sciences' 
Committee on Sentencing Research. Dr. 
Berk gave testimony evaluating the 
appropriateness of Baldus' method and 
the significance of his findings.

9



Pardons and Paroles, and other state
agencies. These records included not 
only trial transcripts and appellate 
briefs but also detailed parole board 
records, prison files, police reports 
and other official documents. (S.E. 43).

Using a carefully tailored 
questionnaire, Professor Baldus gathered 
over five hundred items of information 
on each case concerning the defendant, 
the victim, the crime, the aggravating 
and mitigating circumstances, and the 
strength of the evidence. In addition, 
the Baldus questionnaire required 
researchers to prepare a narrative 
summary to capture individual features 
of each case. The full questionnaire 
appears as DB 38 in the Supplemental 
Exhibits. (S.E. 1-42). Employing 
generally accepted data collection 
methods at each step, Professor Baldus 
cross-checked the accuracy of the data

10



both manually and by computer-aided
systems. (Fed.Tr.585-616).

Professor Baldus found that during 
the 1973-1979 period, 2484 murders and 
non-negligent manslaughters occurred in 
the State of Georgia. Approximately
1665 of those involved black defendants; 
819 involved white defendants. Blacks 
were the victims of homicides in
approximately 61 percent of the cases, 
whites in 39 percent. When Professor 
Baldus began to examine the State's 
subsequent charging and sentencing 
patterns, however, he found that the 
racial proportions were heavily
inverted. Among the 128 cases in which 
a death sentence was imposed, 108 or Ql% 
involved white victims. As exhibit DB
62 demonstrates, white victim cases were 
nearly eleven times more likely to
receive a sentence of death than were 
black victim cases. (S.E. 46). When the

11



cases were further subdivided by race of 
defendant. Professor Baldus discovered 
that 22 percent of black defendants in 
Georgia who murdered whites were 
sentenced to death, while scarcely 3 
percent of white defendants who murdered 
blacks faced a capital sentence. (S.E. 
47) .

These unexplained racial disparities 
prompted Professors Baldus and Woodworth 
to undertake an exhaustive statistical 
inquiry. They first defined hundreds of 
variables, each capturing a single 
feature of the cases.8 Using various 
statistical models, each comprised of 
selected groups of different variables 
(see Fed. Tr. 689-705), Baldus and 
Woodworth tested whether other

8 For example, one variable might 
be defined to reflect whether a case was 
characterized by the presence or absence 
of a statutory aggravating circumstance, 
such as the murder of a police victim. 
(See Fed.Tr.617-22).

12



characteristics of Georgia homicide 
cases might suffice to explain the 
racial disparities they had observed. 
Through the use of multiple regression 
analysis, Baldus and Woodworth were able 
to measure the independent impact of the 
racial factors while simultaneously 
taking into account or controlling for 
more than two hundred aggravating and 
mitigating factors, strength of evidence 
factors, and other legitimate sentencing 
considerations. (See. e.q., S.E. 51).

Professors Baldus and Woodworth 
subjected the data to a wide variety of 
statistical procedures, including cross- 
tabular comparisons, weighted and 
unweighted least-squares regressions, 
logistic regressions, index methods, 
cohort studies and other appropriate 
scientific techniques. Yet regardless 
of which of these analytical tools 
Baldus and Woodworth brought to bear,

13



race held firm as a prominent determiner 
of life or death. Race proved no less 
significant in determining the
likelihood of a death sentence than 
aggravating circumstances such as 
whether the defendant had a prior murder 
conviction or whether he was the prime 
mover in the homicide. (S.E. 50).
Indeed, Professor Baldus testified that 
his best statistical model, which 
"captured the essence of [the Georgia] .
. . system" (Fed.Tr.808), revealed that
after taking into account most 
legitimate reasons for sentencing 
distinctions, the odds of receiving a 
death sentence were still more than 4.3 
times greater for those whose victims 
were white than for those whose victims 
were black. (Fed.Tr. 818; DB 82). 
Focusing directly on petitioner's case, 
Baldus and his colleagues estimated that 
for homicide cases "at Mr. McCleskey's

14



level of aggravation the average white 
victim case has approximately a twenty 
[20] percentage point higher risk of 
receiving a death sentence than a 
similarly situated black victim case." 
(Id. 1740).9 Professor Baldus also 
testified that black defendants whose 
victims were white were significantly 
more likely to receive death sentences 
than were white defendants, especially 
among cases of the general nature of

9 These figures represent a
twenty percentage point, not a twenty 
percent, increase in the likelihood of 
death. Among those cases where the
average death-sentencing rate is .24 or 
24-in-100, the white-victim rate would 
be approximately .34 or 34-in-100, the 
black-victim rate, only .14, or 14-in- 
100. This means that the sentencing rate 
in white victim cases would be over 
twice as high (.34 vs. .14) as in black 
victim cases. Thus, on the average,
among every 34 Georgia defendants
sentenced to death at this level of 
aggravation for the murders of whites,
20 would likely not have received a 
death sentence had their victims been black.

15



petitioner's. (Fed. Hab. Tr. 863-64).
Professor Baldus demonstrated that 

this "dual system" of capital sentencing 
was fully at work in Fulton County where 
petitioner had been tried and sentenced 
to death. Not only did county 
statistical patterns replicate the 
statewide trends, but several non- 
statistical comparisons of Fulton County 
cases further emphasized the importance 
of race. For example, among those 17 
defendants who had been charged with 
homicides of Fulton County police 
officers between 1973 and 1980, only 
one defendant other than petitioner had 
even received a penalty trial. In that 
case, where the victim was black, a life 
sentence was imposed. (Fed.Tr.1050-62).

The State of Georgia produced little 
affirmative evidence to rebut 
petitioner's case. It offered no 
alternative model that might have

16



reduced or eliminated the racial 
variables. (Fed. Tr. 1609). It did not 
even propose, much less test the effect 
of, additional factors concerning 
Georgia crimes, defendants or victims, 
admitting that it did not know whether 
such factors "would have any effect or 
not." (Id. 1569). The State expressly 
declined Professor Baldus's offer, 
during the hearing, to employ 
statistical procedures of the State's 
choice in order to calculate the effect 
of any factors the State might choose to 
designate and to see whether the racial 
effects might be eliminated. 10

Instead, the State simply attacked

10 The District Court did accept 
Professor Baldus's invitation and 
designated a statistical model it 
believed would most accurately capture 
the forces at work in Georgia's capital 
sentencing system. (Fed. Tr. 810; 1426; 
1475-76; 1800-03; Court's Exhibit 1). 
After analyzing this model, Professor 
Baldus reported that it did nothing to 
diminish the racial disparities. (See R. 
731-52) .

17



the integrity of Professor Baldus1s data 
sources (see Fed. Tr. 1380-1447), its 
own official records. It also presented 
one hypothesis, that the apparent racial 
disparities could be explained by the 
generally more aggravated nature of 
white victim cases. The State's 
principal expert never tested that 
hypothesis by any accepted statistical 
techniques (id. 1760-61), although he 
admitted that such a test "would . . 
.[have been] desirable." (Id. 1613). 
Professors Baldus and Woodworth did test 
the hypothesis and testified 
conclusively on rebuttal that it could 
not explain the racial disparities. 
(Fed.Tr.1290-97; 1729-32; GW 5-8).
C. The Decisions Below

The District Court rejected 
petitioner's claims. It faulted 
petitioner's extraordinary data sources 
because they had "not capture[d] every

18



nuance of every issue." (J.A. ). The 
extensive Parole Board records, the 
court complained, "present a 
retrospective view of the facts and 
circumstances . . . after all 
investigation is completed, after all 
pretrial preparation is made." (J.A. ). 
Since such files, the court reasoned, 
did not measure the precise quanta of 
information available to each decision 
maker —  police, prosecutor, judge, jury 
—  at the exact moment when different 
decisions about the case were made, "the 
data base . . i s  substantially 
flawed." (Id.) As a related matter, 
the District Court insisted that all of 
Professor Baldus's statistical models of
the Georgia system --  even those
employing more than 230 separate 
variables —  were "insufficiently 
predictive" since they did not include 
every conceivable variable and could not

19



predict every case outcome. (J.A. ).
The District Court ended its opinion

by rejecting the legal utility of such
statistical methods altogether:

[M]ultivariate analysis is ill 
suited to provide the court with 
circumstantial evidence of the 
presence of discrimination, and 
it is incapable of providing the 
court with measures of
qualitative difference in
treatment which are necessary to 
a finding that a prima facie 
case has been established . . .
To the extent that McCleskey 
contends that he was denied . .
. equal protection of the law, 
his methods fail to contribute 
anything of value to his cause.

(J.A. )(italics omitted).
The majority of the Court of Appeals

chose not to rest its decision on these
findings by the District Court; instead
it expressly "assum[ed] the validity of
the research" and "that it proves what
it claims to prove." (J.A. ). Yet the
Court proceeded to announce novel
standards of proof that foreclose any
meaningful review of racial claims like

20



petitioner's. As its baseline, the 
Court held that statistical proof of 
racial disparities must be "sufficient 
to compel a conclusion that it results 
from discriminatory intent and purpose." 
{J .A . ) (emphasis added). 
"[S]tatistical evidence of racially 
disproportionate impact [must be] . . . 
so strong as to permit no inference 
other than that the results are the 
product of a racially discriminatory 
intent or purpose." (J.A. ). The Court 
also announced that even unquestioned 
proof of racially discriminatory 
sentencing results would not suffice to 
make out an Equal Protection Clause 
violation unless the racial disparities 
were of sufficient magnitude: "The key 
to the problems lies in the principle 
that the proof, no matter how strong, of 
some disparity is alone insufficient." 
(J.A. ). "In any discretionary system,

21



some imprecision must be tolerated," the 
Court stated, and petitioner's proven 
racial disparities were "simply 
insufficient to support a ruling . . .
that racial factors are playing a role 
in the outcome sufficient to render the 
system as a whole arbitrary and 
capricious." (J.A. ). Finally, the
majority held that no Eighth Amendment 
challenge based upon race could succeed 
absent similar proof of purposeful State 
conduct. Although "cruel and unusual 
punishment cases do not normally focus 
on the intent of the government actor .
. . where racial discrimination is
claimed . . . their purpose, intent and
motive are a natural component of the 
proof" (J.A. ) and "proof of a disparate
impact alone is insufficient 
unless . . .  it compels a conclusion 
that . . . race is intentionally being
used as a factor in sentencing." Id.

22



SUMMARY OF ARGUMENT
The principal questions before the 

Court on certiorari involve intermediate 
issues of evidence and proof. 
Fundamental constitutional values are 
nonetheless at the heart of this appeal. 
Our primary submission is that the lower 
courts, by their treatment of 
petitioner's evidence, have effectively 
placed claims of racial discrimination 
in the death penalty —  no matter how 
thoroughly proven —  beyond effective 
judicial review. To appreciate the 
impact of the lower court's holding, it 
is necessary at the outset to recall the 
constitutional values at stake.

This country has, for several 
decades, been engaged in a profound 
national struggle to rid its public life 
of the lingering influence of official, 
state-sanctioned racial discrimination. 
The Court has been especially vigilant

23



to prevent racial bias from weighing in
the scales of criminal justice. See,
e.q., Batson v. Kentucky, __U.S.__, 90
L.Ed.2d 69 (1986); Turner v. Murray,
__U.S.__, 90 L .Ed.2d 27, 35 (1986);
Vasquez v. Hillery, __U.S.__, 88
L.Ed.2d 598 (1986). A commitment
against racial discrimination was among 
the concerns that led the Court to 
scrutinize long-entrenched capital 
sentencing practices and to strike down 
statutes that permitted arbitrary or 
discriminatory enforcement of the death 
penalty. See, e.q., Furman v. Georgia, 
408 U.S. 238 (1972).

In 1976, reviewing Georgia's then 
new post-Furman capital statutes, the 
Court declined to assume that the 
revised sentencing procedures would 
inevitably fail in their purpose to 
eliminate "the arbitrariness and 
capriciousness condemned by Furman."

24



153, 198Gregg v. Georgia, 428 U.S.
(1976)(opinion of Stewart, Powell & 
Stevens, J.J.). Accord, id. at 220-26 
(opinion of White, J.); see also 
Godfrey v. Georgia, 446 U.S. 420, 428 
(1980). It was appropriate at that time 
for the Court to clothe Georgia's new 
statutes with a strong presumption of 
constitutionality —  to assume, 
"[a]bsent facts to the contrary," Gregg 
v. Georgia, 428 U.S. at 225 (opinion of 
White, J.), that its statutes would be 
administered constitutionally: to reject 
"the naked assertion that the effort is 
bound to fail." Id. at 222. Yet the 
presumption extended to Georgia in 1976 
was not —  and under the Constitution 
could never have been —  an irrevocable 
license to carry out capital punishment 
arbitrarily and discriminatorily in 
practice.

Petitioner McCleskey has now

25



presented comprehensive evidence to the
lower courts that Georgia's post-Furman 
experiment has failed, and that its 
capital sentencing system continues to 
be haunted by widespread and substantial 
racial bias.

Faced with this overwhelming 
evidence, the Court of Appeals took a 
wrong turn. It accorded Georgia's 
death-sentencing statutes what amounts 
to an irrebuttable presumption of 
validity, one no capital defendant could 
ever overcome. It did so through a 
series of rulings that "placed on 
defendants a crippling burden of proof." 
Batson v. Kentucky, 90 L.Ed.2d at 85. 
Henceforth, a capital defendant, rather 
than proving a prima facie case of 
discrimination by demonstrating the 
presence of substantial racial 
disparities within a system "susceptible 
of abuse" —  thereby shifting the

26



burden of explanation to the State, see,
e.q., Castaneda v. Partida, 430 U.S. 
482, 494-495 (1977); Washington v. 
Davis, 426 U.S. 229, 241 (1976); Batson 
v. Kentucky, supra —  must present proof 
so strong that it "permits no inference 
other than . . . racially discriminatory 
intent." No room is left in this 
formulation for proof by ordinary fact­
finding processes. Instead, a capital 
defendant must anticipate and exclude at 
the outset "every possible factor that 
might make a difference between crimes 
and defendants, exclusive of race." 
(J.A. ).

This new standard for proof of 
racial discrimination has no precedent 
in the Court's teachings under the 
Equal Protection Clause; it is contrary 
to everything stated or implied in 
Batson v. Kentucky, supra; Bazemore v. 
Friday, __U.S.__, __L.Ed.2d__ (1986);

27



Arlington____Heights v. Metropolitan
Housing Development Corp., 429 U.S. 252
(1977), and a host of the Court's 
decisions expounding the principle of a 
prima facie case.

Compounding the Court of Appeals'
new standard is the burden it imposed
upon statistical modes of proof, which 
virtually forecloses any demonstration 
of discriminatory capital sentencing by 
means of scientific evidence. To be 
sufficient, a statistical case must 
address not only the recognized major 
sentencing determinants, but also a host 
of hypothetical factors, conjectured by 
the Court, whose systematic relation to 
demonstrated racial disparities is 
dubious to say the least. (See J.A. ) 
This cannot be the law, unless there is 
to be a "death penalty exception" to the 
Equal Protection Clause. Just last Term, 
the Court unanimously held that such a

28



approach torestrictive judicial 
statistical evidence was unacceptable
error. Bazemore v. Friday, __L.Ed.2d at

(1986). See also Texas Department of 
Community Affairs v. Burdine, 450 U.S. 
248, 252 (1981) .

The Court of Appeals also concluded 
that even proven, persistent racial 
disparities in capital sentencing are 
constitutionally irrelevant unless their 
magnitude is great. This holding strays 
far from the Constitution and the 
record. The Equal Protection Clause 
protects individuals against a little 
state-sanctioned racial discrimination 
as well as a lot; the law does not 
permit a State to use the death penalty 
infrequently, or discriminate when it 
does, and defend by saying that this 
discrimination is rare. Only last Term,
in Papasan____v. Allain, __U.S.__,
_L.Ed.2d __ (1986), the Court expressly

29



declined to apply "some sort of 
threshold level of effect . . . before 
the Equal Protection Clause's strictures 
become binding."

In any event, the Court of Appeals 
plainly misconceived the facts as much 
as the law on this issue. As we will 
show, one central flaw pervading its 
decision was a serious misapprehension 
of the degree to which race played a 
part in Georgia's capital sentencing 
system from 1973 through 1979.

Finally, the court announced that, 
henceforth, in a capital case, proof of 
"purposeful discrimination will be a 
necessary component of any Eighth 
Amendment claim alleging racial 
discrimination. Such a rule contradicts 
both precedent and principle. Under the 
Eighth Amendment, this Court has held 
that it is the State's obligation "to 
tailor and apply its laws in a manner

30



that avoids the arbitrary and capricious
infliction of the death penalty." 
Godfrey v. Georgia, 446 U.S. 420, 428 
(1980). The federal task in reviewing 
the administration of those laws "is not 
restricted to an effort to divine what 
motives impelled the[] death penalties," 
Furman v. Georgia, 408 U.S. at 253 
(Douglas, J., concurring), but, having 
"put to one side" the issue of 
intentional discrimination, id. at 310 
(Stewart, J., concurring), to discern 
whether death sentences are "be[ing] . .
. wantonly and . . . freakishly 
imposed." Id. at 312.

Reduced to its essence, petitioner's 
submission to the Court is a simple one. 
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

31



proof and fact-finding accepted as 
necessary in every other area of law 
should not be jettisoned in this one.

I.
RACE IS AN INVIDIOUS AND UNCONSTITUTIONAL 

CONSIDERATION IN CAPITAL SENTENCING 
PROCEEDINGS

A. The Equal Protection Clause Of
The Fourteenth Amendment Forbids 
Racial Discrimination In The 
Administration Of Criminal Statutes
In the past century, few judicial

responsibilities have laid greater claim
on the moral and intellectual energies
of the Court than "the prevention of
official conduct discriminating on the
basis of race." Washington v. Davis,
426 U.S. at 239. The Court has striven
to eliminate all forms of state-
sanctioned discrimination, "whether
accomplished ingeniously or
ingenuously." Smith v. Texas, 311 U.S.
128, 132 (1940). It has forbidden
discrimination required by statute, see,
e.q ., Brown v. Board of Education, 346

32



U.S. 483 {1954); Nixon v. Herndon, 273
U.S. 536 (1927), and has not hesitated
to "look beyond the face of . . . [a] 
statute . . . where the procedures 
implementing a neutral statute operate . 
. . on racial grounds." Batson v. 
Kentucky, 90 L.Ed.2d at 82; Turner v. 
Fouche, 396 U.S. 346 (1970); Yick Mo v. 
Hopkins, 118 U.S. 356, 373-74 (1886).

The Court has repeatedly emphasized 
that "the core of the Fourteenth 
Amendment is the prevention of 
meaningful and unjustified official 
distinctions based on race." Hunter v.
Erickson, 393 U.S. 385, 391 (1969). In
the area of criminal justice, where
racial discrimination "strikes at the 
fundamental values of our judicial 
system and our society as a whole," Rose 
v. Mitchell, 443 U.S. 545, 556 (1979), 
the Court has "consistently" articulated 
a "strong policy . . .  of combating

33



racial discrimination." ^d* at 558.
One of the most obvious forms that 

such discrimination can take in the 
criminal law is a systematically unequal 
treatment of defendants based upon their 
race. See McLaughlin v. Florida, 379 
U.S. 184, 190 n.8 (1964), citing
Strauder v. West Virginia, 100 U.S. 303, 
306-08 (1880); Ho Ah Kow v. Nunan, 12
Fed. Cas. 252 (No. 6546)(C .C.D .Cal. 
1879). Certainly, among the evils that 
ultimately prompted the enactment of the 
Fourteenth Amendment and cognate post- 
Civil War federal legislation were state 
criminal statutes, including the 
infamous Black Codes, which prescribed 
harsher penalties for black persons than 
for whites. See General Building
Contractors_____Ass 1 n . ,______ Inc ._____ v ■
Pennsylvania, 458 U.S. 375, 386-87

34



( 1982).11 jn this case, Professor Baldus 
has reported that the race of the 
defendant —  especially when the 
defendant is black and the victim is 
white —  influences Georgia's capital 
sentencing process. The State of 
Georgia has disputed the truth of this 
claim, but has offered no constitutional 
defense if the claim is true. Georgia 
has never articulated, or even 
suggested, any "permissible state 11

11 The Court has accordingly 
insisted "that racial classifications, 
especially suspect in criminal statutes, 
be subjected to the 'most rigid 
scrutiny' and, if they are ever to be 
upheld . . .  be shown to be necessary 
to the accomplishment of some 
permissible state objective, independent 
of the racial discrimination which it 
was the object of the Fourteenth 
Amendment to eliminate." Loving v. 
Virginia, 388 U.S. 1, 11 (1967). See
also____Personnel____Administrator____of
Massachusetts v. Feeney, 442 U.S. 256, 
272 (1979); cf. McLaughlin v. Florida, 
379 U.S. at 198 ("I cannot conceive of a 
valid legislative purpose under our 
Constitution for a state law which makes 
the color of a person's skin the test of 
whether his conduct is a criminal 
offense")(Stewart, J., concurring).

35



interest" that would justify the 
disproportionate infliction of capital 
punishment in a discriminatory fashion 
against black defendants.

Nor has Georgia claimed any 
constitutional warrant to execute 
murderers of white citizens at a greater 
rate than murderers of black citizens. 
The history of the Equal Protection 
Clause establishes that race-of-victim 
discrimination was a major concern of 
its Framers, just as Professor Baldus 
has now found that it is a major feature 
of Georgia's administration of the death 
penalty. Following the Civil War and 
immediately preceding the enactment of 
the Fourteenth Amendment, Southern 
authorities not only enacted statutes 
that treated crimes committed against 
black victims more leniently, but 
frequently declined even to prosecute 
persons who committed criminal acts

36



against blacks. When prosecutions did 
occur, authorities often acquitted or 
imposed disproportionately light
sentences on those guilty of crimes 
against black persons. 12

The congressional hearings and

12 See, e.q.. Report of the Joint 
Committee on Reconstruction, at the 
First Session, Thirty-Ninth Congress, 
Part II, at 25 (1866)(testimony of 
George Tucker, commonwealth 
attorney)(The southern people "have not 
any idea of prosecuting white men for 
offenses against colored people; they do 
not appreciate the idea."); id. at 209 
(testimony of Lt. Col Dexter Clapp)("Of 
the thousand cases of murder, robbery, 
and maltreatment of freedmen that have 
come before me, . . . .  I have never yet 
known a single case in which the local 
authorities or police or citizens made 
any attempt or exhibited any inclination 
to redress any of these wrongs or to 
protect such persons."); id. at 213 
(testimony of Lt. Col. J. Campbell); 
id., Part III, at 141 (testimony of 
Brevet M.J. Gen. Wagner Swayne)("I have 
not known, after six months' residence 
at the capital of the State, a single 
instance of a white man being convicted 
and hung r sic 1 or sent to the 
penitentiary for crime against a negro, 
while many cases of crime warranting 
such punishment have been reported to 
me."); id., Part IV, at 76-76 (testimony 
of Maj. Gen. George Custer).

37



debates that led to enactment of the
Fourteenth Amendment are replete with 
references to this pervasive race-of- 
victim discrimination; the Amendment and 
the enforcing legislation were intended, 
in substantial part, to stop it. As the 
Court recently concluded in Briscoe v. 
Lahue, 460 U.S. 325, 338 (1983), "[i]t 
is clear from the legislative debates 
that, in the view of the . . . sponsors, 
the victims of Klan outrages were 
deprived of 'equal protection of the 
laws' if the perpetrators systematically 
went unpunished." See discussion in 
Petition for Certiorari, McCleskey v. 
Zant, No. 84-6811, at 5-7.

Even without reference to the 
Amendment's history, race-of-victim 
sentencing disparities violate long- 
recognized equal protection principles 
applicable to all forms of state action. 
The Court has often held that whenever

38



either "fundamental rights" or "suspect 
classifications" are involved, state 
action "may be justified only by a 
'compelling state interest1 . . . and .
. . legislative enactments must be 
narrowly drawn to express only the 
legitimate state interests at stake." 
Roe v . Wade, 410 U.S. 113, 135 (1973); 
see also Cleveland Board of Education v. 
LaFleur, 414 U.S. 632 (1974); Stanley v. 
Illinois, 405 U.S. 645 (1972).

Discrimination by the race of victim 
not only implicates a capital 
defendant's fundamental right to life, 
cf. Skinner v. Oklahoma, 316 U.S. 535, 
541 (1942), but employs the paradigmatic 
suspect classification, that of race. In 
McLaughlin v. Florida, supra, the Court 
examined a criminal statute which 
singled out for separate prosecution 
any black man who habitually occupied a 
room at night with a white woman (or

39



vice versa) without being married. The 
statute, in essence, prosecuted only 
those of one race whose cohabiting 
"victims" were of the other race. 
Finding no rational justification for 
this race-based incidence of the law, 
the Court struck down the statute.

The discrimination proven in the 
present case cannot be defended under 
any level of Fourteenth Amendment 
scrutiny. Systematically treating 
killers of white victims more harshly 
than killers of black victims can have 
no constitutional justification. *3 This 13

13 The Court identified in Gregg 
v. Georgia, 428 U.S. at 183-84 (1976), 
at least two "legitimate governmental 
objectives" for the death penalty—  
retribution and deterrence. The Court 
noted that the death penalty serves a 
retributive purpose as an "expression of 
society's moral outrage at particularly 
offensive conduct." 428 U.S. at 183. 
The race of the victim obviously has no 
place as a factor in society's 
expression of moral outrage. Similarly, 
if the death penalty is meant to deter 
capital crime, it ought to deter such 
crime equally whether inflicted against

40



would set the seal of the state upon the 
proposition that the lives of white 
people are more highly valued than those 
of black people —  either an "assertion 
of [the] . . . inferiority" of blacks,
Strauder v. West Virginia, 100 U.S. at 
308, or an irrational exercise of 
governmental power in its most extreme 
form.
B. The Eighth Amendment Prohibits

Racial Bias In Capital Sentencing
Petitioner McCleskey has invoked the 

protection of a second constitutional 
principle, drawn from the Eighth 
Amendment. One clear concern of both the 
concurring and dissenting Justices in 
Furman v. Georgia, 408 U.S. 238 (1972), 
was the possible discriminatory 
application of the death penalty at that 
time. Justice Douglas concluded that 
the capital statutes before him were

black or against white citizens.

41



"pregnant with discrimination," 408 U.S. 
at 257, and thus ran directly counter to 
"the desire for equality . . . reflected 
in the ban against 'cruel and unusual 
punishments' contained in the Eighth 
Amendment." Id. at 255. Justice 
Stewart lamented that "if any basis can 
be discerned for the selection of these 
few sentenced to die, it is the 
constitutionally impermissible basis of 
r a c e . T h e s e  observations illuminate 
the holding of Furman, reaffirmed by the 
Court in Gregg and subsequent cases, 
that the death penalty may "not be 
imposed under sentencing procedures that 
created a substantial risk that it 
[will] . be inflicted in an 
arbitrary and capricious manner." Gregg 
v. Georgia, 428 U.S. at 188; Godfrey v. 14

14 See id. at 364-66 (Marshall, 
J., concurring); cf. id. at 389 n.12 
(Burger, C.J., dissenting); id. at 449- 
50 (Powell, Jr., dissenting).

42



at 428; Zant v .Georgia, 446 U.S.
Stephens, 456 U.S. 410, 413 (1982)(per
curiam).

The Court itself suggested in Zant 
v. Stephens, 462 U.S. 862, 885 (1983), 
that if "Georgia attached the 
'aggravating'' label to factors that are 
constitutionally impermissible or
totally irrelevant to the sentencing 
process, such as . . . the race . . .  of
the defendant . . . due process of law
would require that the jury's decision 
to impose death be set aside." This 
Eighth Amendment principle tracks the 
general constitutional rule that, where 
fundamental rights are at stake, 
"legislative enactments must be narrowly 
drawn to express only the legitimate 
state interests at stake." Roe v . Wade, 
410 U.S. at 155. Legislative
classifications that are unrelated to 
any valid purpose of a statute are

43



arbitrary and violative of the Due

Process Clause Cleveland Board of

Education v. LaFleur, 414 U.S. 632 
(1974); Stanley v. Illinois, 405 U.S. 
645 (1972). A legislative decision to 
inflict the uniquely harsh penalty of 
death along the lines of such an 
irrational classification would be still 
more arbitrary under the heightened 
Eighth Amendment standards of Furman. 
Cf. Gardner v. Florida, 430 U.S. 349, 
357-58, 361 (1977)(plurality opinion); 
id. at 362-64 (opinion of White, J.). 
And nothing could be more arbitrary 
within the meaning of the Eighth 
Amendment than a reliance upon race in 
determining who should live and who 
should die.

44



II.
THE COURT OF APPEALS FASHIONED 

UNPRECEDENTED STANDARDS OF PROOF WHICH 
FORECLOSE ALL MEANINGFUL REVIEW OF 
RACIAL DISCRIMINATION IN CAPITAL 

SENTENCING PROCEEDINGS

The crucial errors of the Court of 
Appeals involve the "crippling burden of 
proof" it placed upon petitioner and any 
future inmate who would seek the 
protections of the Federal Constitution 
against racial discrimination in capital 
sentencing. "[E]qual protection to
all," the Court long ago observed, "must 
be given —  not merely promised." Smith 
v . Texas, 311 U.S. at 130. The opinion
below was all promise, no give. It 
held, in effect: You can escape being
judged by the color of your skin, and by 
that of your victim, if (but only if) 
you can survey and capture every 
ineffable quality of every potentially 
capital case, and if you then meet 
standards for statistical analysis that

45



are elsewhere not demanded and nowhere 
susceptible of attainment.

Judged by these standards, the 
research of Professor Baldus—  
described by Dr. Richard Berk as "far 
and away the most complete and thorough 
analysis of sentencing that's ever been 
done" (Fed.Tr.1766) —  is simply not 
good enough. Nor would any future 
studies be, absent evidence that 
apparently must "exclud[e] every 
possible factor that might make a 
difference between crimes and 
defendants, exclusive of race." (J.A.
). As we shall demonstrate in the 
following subsections, these manifestly 
are not appropriate legal standards of 
proof. They depart radically from the 
settled teachings of the Court. They 
have no justification in policy or legal 
principle, and they trivialize the 
importance of Professor Baldus's real

46



and powerful racial findings.
A. The Court of Appeals Ignored This 
Court's Decisions Delineating A Party's 
Prima Facie Burden Of Proof Under The 
Equal Protection Clause

(i) The Controlling Precedents
In Batson v. Kentucky, the Court

recently outlined the appropriate order
of proof under the Equal Protection
Clause. " [ I ] n any equal protection
case, 'the burden is, of course,' on the
defendant. . . 'to prove the existence
of purposeful discrimination.' Whitus v.
Georgia, 385 U.S. [545], at 550 [1967]
..." 90 L.Ed. 2d at 85. "[The
defendant] may make out a prima facie
case of purposeful discrimination by
showing that the totality of relevant
facts gives rise to an inference of
discriminatory purpose." Washington v.
Davis, [426 U.S.] at 239-242:"

Once the defendant makes the 
requisite showing, the burden 
shifts to the State to explain 
adequately the racial exclusion. 
Alexander v. Louisiana, 405 U.S.

47



[625], at 632 [(1972)]. The 
State cannot meet this burden on 
mere general assertions that its 
officials did not discriminate 
or that they properly performed 
their duties. See Alexander v. 
Louisiana, supra, at 632; Jones 
v. Georgia, 389 U.S. 24, 25 
(1967). Rather the State must 
demonstrate that "permissible 
racially neutral selection 
criteria and procedures have 
produced the . . . result."

90 L.Ed.2d 85-86.
The approach is "a traditional 

feature of the common law," Texas Pep't 
of Community Affairs v. Burdine, 450 
U.S. at 255 n.8, which, in the context 
of discrimination litigation, requires a 
complainant to "eliminate[] the most 
common nondiscriminatory reasons for the 
[observed facts]," id. at 254, and then
places a burden on the alleged wrongdoer
to show "a legitimate reason for" those 
facts, _id. at 255, thereby 
"progressively . . . sharpening] the 
inquiry into the elusive factual
question of intentional discrimination."

48



Id. at 255 n.3.15

Although the initial showing of 
race-based state action required depends 
upon the nature of the claim and the 
responsibilities of the state actors 
involved, Washington v. Davis, 426 U.S. 
at 253 (Stevens, J., concurring), 
Castaneda v. Partida, 430 U.S. 482, 494- 
95 (1977); cf. Wayte v. United States,
__U.S.__, 84 L .Ed.2d 547, 556 n.10
(1985), the guiding principle is that 
courts must make "a sensitive inquiry 
into such circumstantial and direct 
evidence of intent as may be available." 15

15The roots of this approach run 
back at least as far as Neal v. 
Delaware, 103 U.S. 370 (1881), where the 
Court refused to indulge a "violent 
presumption," offered by the State of 
Delaware to excuse the absence of black 
jurors, that "the black race in Delaware 
were utterly disqualified, by want of 
intelligence, experience or moral 
integrity to sit on juries." 103 U.S. 
at 397. Absent proof to support its 
contention, the State's unsupported 
assertion was held insufficient to rebut 
the prisoner's prima facie case. Id.

49



Village of____Arlington Heights v.
Metropolitan Housing Development Corp., 
429 U.S. 252, 266 (1977). Accord,
Rogers v. Lodge, 458 U.S. 613, 618
(1982). Among the most important 
factors identified by the Court as 
probative have been (i) the racial 
impact of the challenged action, (ii) 
the existence of a system affording 
substantial state discretion, and (iii) 
a history of prior discrimination.

(ii) Petitioner's Evidence 
The prima facie case presented by 

petitioner exceeds every standard ever 
announced by this Court for proof of 
discrimination under the Equal
Protection Clause. The centerpiece of 
the case, although not its only feature, 
is the work of Professor Baldus and his 
colleagues, who have examined in 
remarkable detail the workings of 
Georgia's capital statutes during the

50



first seven years of their 
administration, from 1973 through 1979. 
The Baldus studies are part of a body of 
scientific research conducted both 
before and after Furman that has 
consistently reported racial 
discrimination at work in Georgia's 
capital sentencing system.16 Baldus1s 
research reached the same conclusions as 
the earlier studies, but there the 
resemblance ends: his work is vastly 
more detailed and comprehensive than any 
prior sentencing study in Georgia or

16 see, Wolfgang & Riedel, Race, 
Judicial Discretion and the Death 
Penalty, 407 Annals 119 (1973); Wolfgang 
& Riedel, Race, Rape and the Death
Penalty in____Georgia, 45 Am. J.
Orthopsychiat. 658 (1975); Bowers & 
Pierce, Arbitrariness and Discrimination 
Under Post-Furman Capital Statutes, 26 
Crime & Deling. 563 (1980); Gross & 
Mauro, Patterns of Death: An Analysis 
of Racial Disparities in Capital 
Sentencing and Homicide Victimization, 
37 Stan. L. Rev. 27 (1984); Barnett, 
Some Distribution Patterns for the 
Georgia Death Sentence, 18 U.C. Davis L. 
Rev. 1327 (1985).

51



elsewhere.
The Baldus research actually 

comprised two overlapping studies: the 
first, a more limited examination of 
cases from 1973-1978 in which a murder 
conviction had been obtained at trial 
(Fed.Tr.170); the second, - a wide- 
ranging study involving a sample of all 
cases from 1973 through 1979 in which 
defendants indicted for murder or 
voluntary manslaughter had been 
convicted and sentenced to prison. (Id. 
263-65). Most of Baldus' findings in 
this case are reported from the second 
study.

a. The Racial Disparities
"The impact of the official action 

—  whether it 'bears more heavily on one 
race than another' . . . —  provide[s]
an important starting point." Arlington 
Heights. 429 U.S. at 266. Here, the 
Baldus studies reveal substantial,

52



unadjusted racial disparities: a death- 
sentencing rate nearly eleven times 
higher in white-victim cases than in 
black-victim cases. (Fed.Tr.730-33; S.E. 
46). Professor Baldus testified that 
these figures standing alone did not 
form the basis for his analysis, 
because they offered no control for 
potential legitimate explanations of the 
observed racial differences. (Fed. Tr. 
734).66
Professor Baldus thus began collecting 
data on every non-racial factor 
suggested as relevant by the literature, 
the case law, or actors in the criminal 
justice system. His final 
questionnaires sought information on 
over 500 items related to each case 
studied. (Fed.Tr.278-92; S.E. 1-42).

After collecting this vast 
storehouse of data, Professor Baldus and 
his colleagues conducted an exhaustive

53



series of analyses, involving the 
application of increasingly 
sophisticated statistical tools to 
scores of sentencing models. The great 
virtue of the Baldus work was the 
richness of his data sources and the 
extraordinary thoroughness of his 
analysis. Throughout this research, 
Baldus and his colleagues forthrightly 
tested many alternative hypotheses and 
combinations of factors, in order to 
determine whether the initial observed 
racial disparities would diminish or 
disappear. (Fed.Tr.1082-83). Far from 
concealing their results from scrutiny, 
they exposed them to open and repeated 
inquiry by others, soliciting from the 
State and obtaining from the federal 
judge in this case an additional 
"sentencing model" which they then 
tested and reported. (Fed.Tr.810; 1426; 
1475-76)(R. 731-52).

54



The results of these analyses were 
uniform. Race-of-victlm disparities not 
only persisted in analysis after 
analysis —  at high levels of 
statistical significance —  but the race 
of the victim proved to be among the 
more influential determiners of capital 
sentencing in Georgia. Professors
Baldus and Woodworth indicated that 
their most explanatory model of the 
Georgia system, which controlled 39 
legitimate factors, revealed that, on 
average, the murderers of white victims 
faced odds of a death sentence over 4.3 
times greater than those similarly 
situated whose victims were black. (See 
DB 82). Moreover, black defendants like 
petitioner McCleskey whose victims were 
white were especially likely to receive 
death sentences.

b. The Opportunity for Discretion
The strong racial disparities shown

55



by Professor Baldus arise in a system
affording state actors extremely broad 
discretion, one unusually "susceptible 
of abuse." Castaneda v. Partida, 430 
U.S. at 494. The existence of discretion 
is relevant because of "the opportunity 
for discrimination [it] . . . present[s] 
the State, if so minded, to discriminate 
without ready detection." Whitus v. 
Georgia, 385 U.S. at 552. The 
combination of strong racial disparities 
and a system characterized by ample 
State discretion has historically 
prompted the closest judicial scrutiny. 
See, e.q., Yick Wo v. Hopkins, 118 U.S. 
at 373-74.

Post-Furman capital sentencing 
systems in general are characterized by 
a broad "range of discretion entrusted 
to a jury," which affords "a unique 
opportunity for racial prejudice to 
operate but remain undetected." Turner

56



v. Murray, 90 L.Ed. 2d at 35. The
Georgia system is particularly 
susceptible to such influences, since 
Georgia: (i) has only one degree of 
murder, Gregg v. Georgia, 428 U.S. 153, 
196 (1976); (ii) permits a prosecutor to 
accept a plea to a lesser offense, or to 
decline to submit a convicted murder 
case to a sentencing jury, even if 
statutory aggravating circumstances 
exist, _id. at 199; (iii) includes 
several statutory aggravating 
circumstances that are potentially 
vague and overbroad, id. at 200-02 (at 
least one of which has in fact been 
applied overbroadly, Godfrey v. Georgia, 
446 U.S. 420 (1980)); and (iv) allows a 
Georgia jury "an absolute discretion" in 
imposing sentence, unchecked by any 
facts or legal principles, once a single 
aggravating circumstance has been found. 
Zant v. Stephens, 462 U.S. 862, 871

57



( 1983) .
Petitioner presented specific

evidence which strongly corroborated 
this general picture. The District 
Attorney for Fulton County, where 
petitioner was tried, acknowledged that 
capital cases in his jurisdiction were 
handled by a dozen or more assistants. 
(Dep. 15, 45-48). The office had no 
written or oral policies or guidelines 
to determine whether a capital case 
would be plea-bargained or brought to 
trial, or whether a case would move to a 
sentencing proceeding upon conviction. 
(Dep. 12-14, 20-22, 28, 34-38). The
District Attorney admitted that his 
office did not always seek a sentencing 
trial even when substantial evidence of 
aggravating circumstances existed. (Dep. 
38-39). Indeed, he acknowledged that 
the process in his office for deciding 
whether to seek a death sentence was

58



"probably . . . the same" as it had been
in the pre-Furman period. (Dep. 59-61). 
These highly informal procedures are 
typical in other Georgia jurisdictions 
as well. See Bentele, The Death Penalty 
in Georgia: Still Arbitrary, 61 Wash.
U. L.Q. 573, 609-21 (1985)(examining
charging and sentencing practices among 
Georgia prosecutors in the post-Furman 
period).
b. The History of Discrimination

Finally, "the historical background" 
of the State action under challenge "is 17

17 This evidence is sufficient to 
overcome the constitutional presumption 
"that prosecutors will be motivated in 
their charging decisions [only by] . . . 
the strength of their case and the 
likelihood that a jury would impose the 
death penalty if it convicts." Gregg v. 
Georgia, 428 U.S. at 225. Professor 
Baldus performed a number of analyses on 
prosecutorial charging decisions, both 
statewide (Fed.Tr.897-910; S.E. 56-57), 
and in Fulton County (Fed.Tr.978-81; 
S.E. 59-60), which demonstrate racial 
disparities in prosecutorial plea­
bargaining practices.

59



one evidentiary source." Arlington 
Heights, 429 U.S. at 267. See generally
Hunter v. Underwood,__U.S.__, 85 L.Ed.2d
222 (1985); Rogers v. Lodge, 458 U.S. 
613 (1982). Petitioner supplemented
his strong statistical case with
references to the abundant history of 
racial discrimination that has plagued 
Georgia's past. Some of that history 
has been set forth in the petition for 
certiorari, and it will not be reviewed 
in detail in this brief.

It suffices to note here that, for 
over a century, Georgia possessed a 
formal, dual system of crimes and 
penalties, which explicitly varied by 
the race of the defendant and that of 
the victim. (See Pet. for Certiorari, 
3-4). When de jure discrimination in 
Georgia's criminal law ended after the 
Civil War, it was quickly replaced by a 
social system involving strict de jure

60



segregation of most areas of public 
life, with consequent rampant de facto 
discrimination against blacks in the 
criminal justice system.18 (Id., 8-11). 
This Court and the lower federal courts 
have been compelled repeatedly to 
intervene in that system well into this 
century to enforce the basic 
constitutional rights of black citizens. 
(See cases cited in Pet. for Certiorari, 
10n.l8. Unfortunately, the State's 
persistent racial bias has extended to 
the administration of its capital 
statutes as well.

* * * *

In sum, petitioner presented the
District Court with evidence of

18 As a Georgia court held in 
1907: "[E]quality [between black and 
white citizens] does not, in fact, 
exist, and never can. The God of nature 
made it otherwise and no human law can 
produce it and no tribunal enforce it." 
Wolfe v. Georgia Ry. & Elec. Co.. 2 Ga. 
App. 499, 58 S.E. 899, 903 (1907).

61



substantial racial discrimination in
Georgia's capital sentencing system, 
after controlling for hundreds of non- 
racial variables. He noted that this 
highly discretionary system was open to 
possible abuse, and he recited a long 
and tragic history of prior 
discrimination tainting the criminal 
justice system in general and the 
administration of capital punishment in 
particular. Nothing more should have 
been necessary to establish a prima 
facie case under this Court's settled 
precedents.

(iii) The Opinion Below
A majority of the Court of Appeals 

found petitioner's evidentiary showing 
to be "insufficient either to require or 
to support a decision for petitioner." 
(J.A. ). The court in effect announced 
the abolition of the prima facie 
standard, and required instead that

62



petitioner produce evidence "so great
that it compels a conclusion that the
system is arbitrary and
capricious," (J.A. ) and "so strong as
to permit no inference other than that
the results are the product of a
racially discriminatory intent or
purpose." (J.A. ). Petitioner failed
this test, the court concluded, in part
because his studies failed to take
account of "'countless racially neutral
variables,'" including

looks, age, personality,
education, profession, job, 
clothes, demeanor and remorse, 
just to name a few . . There
are, in fact, no exact 
duplicates in capital crimes and 
capital defendants.

(J.A. ).
To meet the lower court's standard 

of proof, in other words, would have 
required petitioner to anticipate and 
control for factors the court frankly 
acknowledged to be "countless." Such a

63



standard seems squarely, Irretrievably 
at odds with the whole notion of a prima 
facie case. If a petitioner's evidence 
must "compel a conclusion" of 
discriminatory intent —  if it must 
anticipate and dispel every conceivable 
non-racial explanation — then the so- 
called "prima facie" case is logically 
irrebuttable and required to be so. 
This insatiable demand for unspecified 
information is precisely what the Court 
condemned as error last Term in Bazemore
v .____ Friday, __L.Ed.2d at __.
(petitioner's1 evidence need "not
include 'all measurable variables
thought to have an effect on [the matter 
at issue]"). It is no less error in this 
case.
B. The Court of Appeals Disregarded 
This Court's Teachings On The Proper 
Role Of Statistical Evidence In 
Proving Intentional Discrimination

(i) The Controlling Precedents
Closely related to its repudiation

64



of the prima facie principle was the
Court of Appeals' disparagement of 
statistical proof. Once again, the 
court's opinion clashed sharply with the 
pronouncements of this Court. "[0]ur 
cases make it unmistakably clear," 
Justice Stewart wrote in Teamsters v. 
United States, 431 U.S. 324, 339 (1977), 
"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." "Where gross 
statistical disparities can be shown, 
they alone may in a proper case 
constitute prima facie proof of a 
pattern or practice of discrimination." 
Hazelwood School District v. United 
States. 433 U.S. 299, 307-08 (1977). 
See, e.q. Castaneda v. Partida, 430 U.S. 
482, 493-96 (1977). The statistical 
method chiefly relied upon by petitioner

65



McCleskey multiple regression
analysis —  was specifically discussed 
with approval by the Court in Bazemore
v. Friday, __L.Ed.2d at ___, and has
received wide acceptance in the lower 
courts. 19

(ii) Petitioner's Evidence
In the District Court, Professors 

Baldus and Woodworth explained in 
painstaking detail every major
methodological issue they faced, how 
they addressed the issue, and how it

19 See, e.g,____  Wilkins v.
University of Houston, 654 F.2d 388, 
402-03 (5th Cir. 1981), vacated and 
remanded on other grounds, 459 U.S. 809 
(1982); EEOC v. Ball Corp., 661 F.2d 531 
(6th Cir. 1981); Coble v. Hot Springs 
School District No, 6, 682 F.2d 721,731- 
32 (8th Cir. 1982); Eastland v. TVA, 704 
F.2d 613 (11th Cir. 1983); Segar v.
Smith. 738 F.2d at 1261, 1278-79;
Vuyanich v. Republic Nat'1 Bank, supra. 
See generally Finkelstein, The Judicial 
Reception of Multiple Regression Studies 
in Race and Sex Discrimination Cases, 80 
Col-um. L. Rev. 737 (1980); Fisher,
Multiple_____Regression_____ in_____Legal
Proceedings, 80 Colum. L. Rev. 702 (1980).

66



affected their findings. See, e . q . , 
Fed. Tr. 683; 704-05; 713; 783; 820;
917-18; 1222-24; 1279-82). In virtually
every instance of significance, they 
conducted their analysis by alternative 
methods, and demonstrated that the 
choice of methods made no difference in 
the racial disparities.

The Baldus studies drew accolades
from Dr. Richard Berk, who evaluated
their quality and soundness in light of
his prior comprehensive review of
sentencing research as a member of a
National Academy of Sciences panel:

[Baldus' studies] ha[ve] very 
high credibility, especially 
compared to the studies that 
[the National Academy of
Sciences] . . . reviewed. We
reviewed hundreds of studies on 
sentencing . . . and there's no
doubt that at this moment, this 
is far and away the most 
complete and thorough analysis 
of sentencing that's ever been 
done. I mean there's nothing 
even close.

(Fed.Tr.1766).

67



Baldus and Woodworth conducted
analyses with simple cross-tabular 
methods and with complex multivariate 
methods. (Tr. 122-28; S.E. 47-49).
They used "weighted" and "unweighted" 
data. (Fed.Tr.621-26; S.E. 68-69).
They used multiple regression models 
employing enormously large numbers of 
variables (230 or more) (Fed.Tr.802-04; 
S.E 51), and they used medium-sized and 
small models as well. (Fed.Tr.773-92; 
S.E. 58). Professor Baldus selected 
variables by employing his legal and 
professional expertise concerning the 
factors most likely to influence capital 
sentencing decisions (Tr. 808-09). Then 
he permitted a computer to refine his 
selection by the use of "stepwise" 
regressions and other objective 
statistical means. (Fed.Tr.821-23).

Professors Baldus and Woodworth 
conducted analyses on the variables as

68



coded; then, when the State challenged 
those particular coding values, they 
recoded the variables and ran the 
analyses again. (Fed.Tr.1677-1700). 
They employed acceptable statistical 
conventions to "impute" values in the 
small number of cases where some data 
were actually missing (Fed.Tr.1101-
02), but they also performed "worst- 
case" analyses in which they adopted 
assumptions most contrary to their 
theories and re-ran their analyses under 
such assumptions. (Fed.Tr.1101; 1701-
07; S.E. 64-67).

Dr. George Woodworth, petitioner's 
statistical expert, testified to the 
appropriateness of the major statistical 
conventions used in the studies.
(Fed.Tr.1265). He also testified about 
a series of "diagnostic" analyses he 
conducted to verify the statistical 
appropriateness of each procedure

69



selected.20 (Fed.Tr.1251-65).
Finally, indulging professional 

skepticism even as to the use of 
statistical methods. Professor Baldus 
conducted additional non-statistical, 
"qualitative" analyses in which he 
evaluated (a) all post-Furman Georgia 
cases with the "(b)(2)" or 
"contemporaneous felony" aggravating 
circumstance (see DB 86); (b) all 
capital cases arising in Fulton County 
(Fed.Tr.842-45; see DB 109); and (c) all 
Fulton County cases involving police 
officer victims. (Fed.Tr.1051-55; S.E. 
61-63). He evaluated those cases 
through recognized scientific means, 
comparing the qualitative features and 
facts of each case to ascertain whether 
racial factors continued to play a 20

20 Dr. Richard Berk confirmed 
during his testimony that the methods 
employed by Baldus and Woodworth were 
statistically appropriate. (Fed. Tr. 
1766; 1784-86).

70



1055-56) .
It is difficult to imagine a more 

wide-ranging and searching series of 
statistical and non-statistical
analyses. The results were not only 
internally consistent; they were 
essentially consistent with all other 
research that has been conducted on 
Georgia's post-Furman capital system.

(iii) The Opinion Below
The Court of Appeals treated 

statistical evidence as going to two 
distinct points, and ended by dismissing 
its utility for either purpose. The 
majority first held that statistical 
studies can never prove discrimination 
against an individual defendant.21 This

21 The Court cf Appeals states this 
proposition in varying forms:
" [Generalized statistical studies are 
of little use in deciding whether a 
particular defendant has been
unconstitutionally sentenced to death." 
(J .A . ). "No single petitioner

role. They did. (Fed.Tr.864-65; 993;

71



thesis appears to rest in part upon the 
unobjectionable premise that statistics, 
dealing as they do with probabilities 
and averages, cannot purport to speak 
directly to the events in any particular 
case. Where it goes wrong is in denying 
that specific events can and often must 
be proved indirectly, by inferences 
drawn from probabilities.22 It is 
unclear why the majority was unwilling 
to permit recourse to ordinary fact­
finding procedures for proof of of 
racially discrimination in capital 
sentencing. It may be unwarranted 
skepticism regarding the probative power

could, on the basis of these statistics 
alone, establish that he received the 
death sentence because, and only 
because, his victim was white." (J.A.
). "The statistics alone are 
insufficient to show that McCleskey's 
sentence was determined by the race of 
his victim, or even that the race of his 
victim contributed to the imposition of 
the penalty in his case." (J.A.)

22 Cf. Fed. Rule Evid. 406.
72



of statistics "[w]here intent and
motivation must be proved" (J.A. ). Cf. 
Castaneda v. Partida, 430 U.S. at 495-97 
& n.17 (finding statistical evidence 
sufficient to make out a prima facie 
case of intentional racial 
discrimination). Or it may reflect the 
improvident burden of proof announced by 
the Court of Appeals in capital cases, 
under which a condemned inmate must 
present evidence "so strong as to permit 
no inference other than that . . . of a 
racially discriminatory intent or 
purpose" (J.A. ). Either way, the 
result is incorrect and reversible. For 
the proper rule, of course is that "as 
long as the court may fairly conclude, 
in the light of all the evidence, that 
it is more likely than not that 
impermissible discrimination exists, the 
[claimant] . . .  is entitled to 
prevail." Bazemore v. Friday, __L.Ed.2d

73



at __.
The Court of Appeals took a somewhat 

different tack regarding the bearing of 
statistical evidence on the second issue 
it perceived —  whether there was 
discrimination in "the system" as 
distinguished from discrimination aimed 
at "a particular defendant." (J.A. ).
The majority tacitly conceded, as 
precedent requires, that statistical 
evidence might suffice in principle to 
compel an inference of system-wide 
discrimination.23 (J.A. ). Yet the
Court immediately faulted any systemwide

23 "[D]iscriminatory impact . . . 
may for all practical purposes 
demonstrate unconstitutionality [where]
. . . the discrimination is very 
difficult to explain on nonracial 
grounds." Washington v. Davis, 426 U.S. 
at 242. Accord: Batson v. Kentucky, 90 
L.Ed.2d at 85. See also Personnel
Administrator____of Massachusetts v.
Feeney. 442 U.S. 256, 275 (1979) ("[i ] f 
the impact of this statute could not 
plausibly be explained on a neutral 
ground, impact itself would signal that 
the real classification made by the law 
was in fact not neutral.")

74



statistical study that did not take into
account "every possible factor," e.q., 
each of the "'countless racially neutral 
variables'" that it hypothesized must 
exist. Id. at 894. It faulted even 
Professor Baldus's largest statistical 
models for this failure, (J.A. ), and 
concluded that "[t]he type of research 
submitted here . . .  is of restricted 
use in showing what undirected factors 
control" Georgia's capital sentencing 
system. Id- at 899.

A prima facie statistical case has 
never been supposed to require the 
anticipatory negation of "every possible 
factor" that might explain away an 
apparent pattern of discrimination. 
Accounting for "the most common 
nondiscriminatory" factors is 
sufficient. Texas Dept't of Community 
Affairs v. Burdine, 450 U.S. at 254; 
see, e.q., Bazemore v. Friday, __L.Ed.2d

75



at __. Here, petitioner not only
demonstrated substantial racial
disparities; he then voluntarily 
assumed, and amply met, the burden of 
discounting every plausible non-racial 
explanation ever suggested. At that 
point, if not earlier, he met his prima 
facie burden.24

24 Having done so, "'[i]f there 
[was] . . .  a "vacuum" it [was] . . . 
one which the State [had to] . . . fill, 
by moving in with sufficient evidence to 
dispel the prima facie case of 
discrimination.1" Turner v. Fouche, 396 
U.S. at 361, quoting Avery v. Georgia, 
345 U.S. 559, 562 (1953). See also 
Patton v. Mississippi, 332 U.S. 463, 
468-69 (1947). To do so, the State was 
obligated to "make a 'clear and 
reasonably specific showing,' based on 
admissible evidence, that [an] alleged 
nondiscriminatory explanation in fact 
explains the disparity." Seqar v. 
Smith, 738 F.2d at 1268, quoting Texas 
Pep't of Community Affairs v. Burdine, 
450 U.S. at 253-55. The State of 
Georgia never identified such a factor, 
much less made a "clear and reasonably 
specific showing" of its impact on 
Georgia's racial disparities.

76



C. The Court Of Appeals Erroneously 
Held That Even Proven Patterns Of Racial 
Discrimination Will Not Violate The 
Constitution Unless Racial Disparities 
Are Of Large Magnitude

The Court of Appeals committed two 
egregious errors —  one legal and the 
other factual —  in its treatment of 
petitioner's racial results. First, it 
held that the Equal Protection Clause 
prohibits discriminatory state conduct 
only if such conduct is of "substantial" 
magnitude. Secondly, it found 
petitioner's racial disparities to be 
"marginal."

Yet the Fourteenth Amendment 
prohibits every instance of state- 
sanctioned discrimination, irrespective 
of its magnitude. And petitioner's 
racial findings are in fact quite 
substantial in magnitude: race ranks 
among the factors, whether legitimate or 
illegitimate, that exert the largest 
influence on Georgia's capital

77



sentencing system.
(i) The Controlling Precedent
The Equal Protection Clause does not 

admit of partial performance. A State 
engaged in discrimination on the basis 
of race must cease its unconstitutional 
conduct altogether. This principle was 
confirmed last Term in Papasan v .
Allain, supra. Responding to an argument 
that the Equal Protection Clause was not 
implicated in that case because school 
funds at issue there were "'an
insignificant part of the total payments 
from all sources made to Mississippi's
school d i s t r i c t s __L.Ed.2d at__, the
Court expressly "decline[d] to append to 
the general requirements of an equal 
protection cause of action an additional 
threshold effects requirement." Id. at 
__ n.17.

The same principle emerges
inferentially from Bazemore v. Friday,

78



which involved a dispute over a 
disparity of $331 in the average yearly
wages of black and white employees—
less than 3% of the wage for white
workers. The lesson of Bazemore is
plain: if blacks prove that they
regularly receive only 95 cents on the 
dollar from a State agency, the State 
cannot defend on the ground that a 
nickel is de minimus.25

25 The Court's jury discrimination 
cases are no exception to this rule. 
The Court's tolerance of minor 
differentials in racial representation 
between the jury-eligible populations 
and the representation on grand or petit 
jury lists reflects not constitutional 
indifference toward small acts of 
discrimination, but a recognition of the 
statistical properties of random 
selection: small differences can 
sometimes be attributed to chance. See 
Castaneda v. Partida, 430 U.S. at 496 
n.17. "The idea behind the rule of 
exclusion is not at all complex. If a 
disparity is sufficiently large, then it 
is unlikely that it is due solely to 
chance or accident . . . ." Id. at 494 
n.13. In this case that problem is 
absent. Petitioner has amply proven 
that the racial disparities found here 
are statistically significant and were 
not chance findings.

79



(ii) Petitioner's Evidence 
The extraordinary array of

alternative analyses conducted by 
Professor Baldus yielded, naturally 
enough, an extraordinary array of 
statistical and nonstatistical results- 
- virtually all showing racial
disparities. Professor Baldus testified 
that the most meaningful summary 
indicators of the magnitude of the 
racial factors found were the "death 
odds-multipliers" that he calculated 
using logistic regression analysis, a 
particularly appropriate statistical 
method for the data at issue in this 
case since the overall rate of death 
sentencing is quite low. (See Fed. Tr. 
1230-34). The odds-multiplier for the 
race-of-victim factor under the best 
statistical model was 4.3, meaning that, 
on average, a Georgia defendant's odds 
of receiving a death sentence were 4.3

80



times greater if his victim was white
than if the victim was black. As
Professor Gross has observed:

It might be useful . . .to put
these numbers in perspective. 
Coronary heart disease, it is 
well known, is associated with 
cigarette smoking. But what is 
the magnitude of the effect? . .
. [Controlling for age, smokers 
were 1.7 times more likely to 
die of coronary artery disease 
than nonsmokers. . . .[s]moking
cigarettes increases the risk of 
death from heart disease 
greatly, but by a considerably 
smaller amount than the race-of- 
victim effect that the Eleventh 
Circuit dismisses as marginal.26

The Tables and Figures in the Joint 
Appendix are exemplary of additional 
evidence presented in the District Court 
on the magnitude of the racial 
disparity. One of Professor Baldus' most 
important findings was that the impact 
of the racial factors varies with the

26Gross, Race and Death:____ The
Judicial Evaluation of Evidence of 
Discrimination in Capital Sentencing, 18 
U.C. Davis L. Rev. 1275 (1985).

81



seriousness of the cases:
Race is a factor in the system 
only where there is room for 
discretion, that is, where the 
decision maker has a viable 
choice. In a large number of 
cases, race has no effect. 
These are the cases where the 
facts are so mitigated the death 
penalty is not even considered 
as a possible punishment. At 
the other end of the spectrum 
are the tremendously aggravated 
murder cases where the defendant 
will very probably receive the 
death penalty, regardless of his 
race or the race of the victim.
In between is the mid-range of 
cases where there is an 
approximately 20% racial 
disparity.

(J.A . ) (Clark, J., dissenting in 
part.) Professor Baldus prepared two 
tables, employing an "index method," 
that demonstrate this impact among more 
than 450 of the most aggravated Georgia 
cases. (Fed.Tr.880-83). In the tables, 
one of which appears in the Supplemental 
Exhibits at 54, the cases were arrayed 
into eight groups according to their 
level of seriousness, with the least 
aggravated cases in group 1 and the most

82



The death-aggravated in group 8. 
sentencing rates were then calculated 
and reported for each group. In the
first two groups, no one was sentenced 
to death and consequently no racial 
disparities appear. Once death sentences 
begin to be imposed, however, in groups 
3 through 8, a gap quickly opens between 
the death-sentencing rates in white- 
victim cases and in black-victim cases, 
with the white-victim cases showing a 
consistently higher incidence of capital 
sentences. 27 A similar pattern of

27Dr. Woodworth constructed a 
number of figures to capture this 
pattern visually. One of them, GW 8, 
appears in the Supplemental Exhibits at 
page 72. In GW 8, the horizontal axis 
moving toward the right reflects 
increasingly more aggravated groups of 
cases. The vertical line represents the 
percentage increase in the likelihood of 
a death sentence. As GW 8 makes clear, 
once cases become sufficiently 
aggravated so that juries begin imposing 
death sentences, the death-sentencing 
rate rises more sharply among white- 
victim cases than among black-victim 
cases. Thus, at any particular level of 
aggravation (until the two bands finally

83



disparities measured by race of the
defendant among all white-victim cases,
is reflected in DB 91 (Fed.Tr.885-86).
Professor Baldus observed:

[W]hen you look at the cases in 
. . . the mid-range, where the
facts do not call clearly for 
one choice or another, that's 
where you see there's room for 
the exercise of discretion . . .
the facts liberate the decision 
maker to have a broader freedom 
for the exercise of discretion, 
and it is in the context of 
those decisions that you see the 
effects of . . arbitrary or
possibly impermissible factors.

(Fed.Tr.844). 28
Dr. Woodworth testified without 

contradiction that petitioner
McCleskey's own crime fell into the

converge at the upper levels of 
aggravation), a significantly higher 
percentage of white-victim cases receive 
death sentences.

28 These findings support the 
"liberation hypothesis" advanced by 
Professors Harry Kalven and Hans Zeisel 
in their influential work, The American 
Jury 164-67 (1966). See generally 
Ballew v. Georgia, 435 U.S. 223, 237-38
(1978) .

84



aggravated cases. After reviewing the
results of three separate statistical
techniques, Dr. Woodworth concluded:

[A]t Mr. McCleskey's level of 
aggravation the average white 
victim case has approximately a 
twenty [20] percentage point 
higher risk of receiving the 
death sentence than a similarly 
situated black victim case.

(Fed.Tr.1740) .
However, Professor Baldus also 

testified concerning the average impact 
of the racial factors across all of the 
cases. The Court of Appeals focused 
upon one regression coefficient29 29

middle of the midrange of moderately

29 The regression coefficient, as 
petitioner's experts explained, measures 
the average affect of a particular 
factor on the outcome of a multiple 
regression analysis, after controlling 
for the cumulative impact of all of the 
other factors considered. For example, 
a coefficient of .06 for the race-of- 
victim factor in a multiple regression 
analysis measuring the death-sentence 
outcome means that, independently of 
every other factor considered, the race 
of the victim would increase the average 
likelihood of a death sentence by six 
percentage points. (Fed. Tr. 691-94).

85



reported in DB 83, which was derived 
from an analysis employing a 230- 
variable model. That coefficient, .06, 
indicates that when the race of the 
victim was white, the probability of a 
death sentence increased by 6-in-100.

Petitioner offered additional 
evidence, some of it statistical and 
some non-statistical, to identify more 
precisely the likely impact of Georgia's 
pervasive racial disparities on 
petitioner McCleskey's case. First, 
Baldus reported upon his analysis of 
data from Fulton County, where 
petitioner was tried. He testified that 
his performance of progressively more 
sophisticated analyses for Fulton

The number in parentheses in DB 83 under 
the .06 coefficient "(.02)" reflects the 
statistical significance of the 
coefficient. It indicates that the 
likelihood that this result would have 
occurred by chance if no racial 
disparities in fact existed is less than 
2 per cent.

86



County, similar to those he had employed 
statewide, "show a clear pattern of race 
of victim disparities in death 
sentencing rates among the cases which 
our analyses suggested were death 
eligible." (Fed.Tr.983; 1043-44).

To supplement this statistical 
picture, Baldus examined a "cohort" of 
17 Fulton county defendants arrested and 
charged, as was petitioner, with 
homicide of a police officer during the 
1973-1979 period. Only two among the 
seventeen, Baldus found, even faced a 
penalty trial. One, whose police victim 
was black, received a life sentence. 
(Fed.Tr.1050-62; S.E. 61-63). 
Petitioner, whose police victim was 
white, received a death sentence. 
Although the small numbers require 
caution, "the principal conclusion that 
one is left with," Baldus testified, "is 
that . . . this death sentence that was

87



imposed in McCleskey's case is not 
consistent with the disposition of cases 
involving police officer victims in this 
county." (Fed.Tr.1056).

Professor Baldus devised one 
additional measure of the magnitude of 
the influence of the racial factors. He 
first computed the regression 
coefficients for those factors and for 
other important aggravating and 
mitigating factors. Then he rank- 
ordered them. As DB 81 demonstrates 
(S.E. 50), the race of the victim in 
Georgia exerts as much influence on the 
sentence outcome as whether the 
defendant had a prior murder conviction. 
It is more important in determining life 
or death than the fact that the 
defendant was the prime mover in the 
homicide, or that he admitted guilt and 
asserted no defense. This measurement 
reveals the power of race at work in the

88



Georgia death penalty system. Quite 
simply: its effects are of the same
magnitude as those of statutory 
aggravating factors identified by the 
Georgia legislature as "prerequisite^] 
to the imposition of the death penalty." 
Gregg v. Georgia, 428 U.S. at 198.

(iii) The Opinion Below 
The Court of Appeals centered its 

attention on two statistics drawn from 
the Baldus studies: (i) the 6
percentage point average disparity in 
death-sentencing rates between all
white-victim and all black-victim
homicide cases; and (ii) the 
corresponding 20 percentage point
disparity within the subgroup of
moderately aggravated cases that
included petitioner McCleskey's.

Toward the six percentage point 
figure, the court displayed equal 
measures of incomprehension, skepticism

89



and toleration. The court's
incomprehension is reflected in its
repeated characterization of the
significance of the figure as "marginal"
(J.A. ) or "insufficient." (J.A. ).
This is a serious error. As one
commentator has noted, although

[i]t sounds right when the court 
describes the '6% disparity' 
found by Baldus as a 'marginal 
difference [i]n fact it is 
nothing of the sort. Although
the court seems to have missed 
the point entirely, this 
disparity actually means that 
defendants in white-victim cases 
are several times more likely to 
receive death sentences than 
defendants in black-victim 
cases.

Gross, supra, 18 U.C. Davis L. Rev. at 
1298.
What the court apparently did not 
appreciate is (a) that this figure
represents an average race-of-victim 
disparity of 6 percentage points, not 6 
percent, and (b) that the 6 percentage 
point average disparity occurs across an 
entire system in which overall death-

90



sentencing rates are only five per cent.
(See Fed. Tr. 634; S.E. 45).
Consequently, if the death-sentencing 
rate among a given group of black-victim 
cases were 6 percent, the rate for 
comparable white-victim cases would be 
12 percent, a 100% increase. However, 
since the 6 percentage point disparity 
is an average effect, it is more 
relevant to compare it to the average 
.01 death sentence rate among all black 
victim cases (S.E. 47), which it exceeds 
by a factor of 6 (.06/.01), a 600%
increase over the black-victim rate. It 
is obviously a gross mistake to view 
this difference as a "marginal" one.
Cf. Hunter v. Underwood, __U.S.__, 85
L.Ed.2d 222, 228-30 (1985)(striking down 
a statute which disqualified blacks from 
voting at 1.7 times the rate of whites).

The court's admixture of skepticism 
is reflected in its remarks that "[n]one

91



of the figures mentioned above is a 
definitive quantification of the 
victim's race in the overall likelihood 
of the death penalty in a given case" 
(J.A. ), and that this evidence proves 
only that "the reasons for a [racial] 
difference . . . are not so clear in a 
small percentage of the cases." (J.A. 
). In other words, the court regarded 
the .06 figure as little more than a 
statistical aberration. However, this 
interpretation cannot be squared with 
the unrebutted evidence that the figure 
in question —  which, it bears 
repeating, means that those who kill 
white victims in Georgia are several 
times more likely to be sentenced to 
death than are similarly situated 
murderers of black victims on the 
average —  is a highly reliable figure, 
statistically significant at the p<.02 
level after controlling for literally

92



hundreds of rival hypotheses. It will
not be blinked away.

The court's toleration of whatever
disparity does exist comprises the
greatest portion of its opinion:

Taking the 6% bottom line 
revealed in the Baldus figures 
as true, this figure is not 
sufficient to overcome the 
presumption that the statute is 
operating in a constitutional 
manner. In any discretionary 
system, some imprecision must be 
tolerated, and the Baldus study 
is simply insufficient to 
support a ruling, in the
context of a statute that is 
operating much as intended, that 
racial factors are playing a 
role in the outcome sufficient 
to render the system as a whole 
arbitrary and capricious.

(J . A .  ).

The Court bolstered its judgment by 
citing three decisions of this Court on 
applications for stays in capital 
cases.30 It reasoned that since the

30Wainwriaht v. Ford. 467 U.S. 1220 
(1984); Wainwriqht v. Adams, 466 
U.S. 964 (1984); Sullivan v. Wainwriqht, 
464 U.S. 109 (1983).

93



petitioners in those cases had all 
proffered other studies in which "[t]he 
bottom line figure [included] . . .  a 
'death-odds multiplier' of about 4.8 to 
1" (J.A. ), and since "Baldus obtained 
a death-odds multiplier of 4.3 to 1 in 
Georgia," a rejection of the Baldus 
studies "is supported, and possibly even 
compelled, by" the disposition of these 
stay applications. "[I]t is reasonable 
to suppose that the Supreme Court looked 
at the bottom line indication of racial 
effect and held that it simply was 
insufficient to state a claim." Id*

Yet as this Court well knows, the 
Florida study involved in those three 
applications was significantly less 
comprehensive and sophisticated than the 
Baldus studies. The Court of Appeals 
overlooks (i) that none of this Court's 
summary orders ever addressed the 
magnitude of the disparities shown in

94



the Florida studies; (i i) that this 
Court's orders respecting applications 
for stays of execution "may not be taken 
. . . as a statement . . . on the 
merits," Graves v. Barnes, 405 U.S. 
1201, 1204 (1972)(Powell, J., in 
chambers); accord, Alabama v. Evans, 461 
U.S. 230, 236 n.* (1983)(Marshall, J., 
dissenting), and (iii) that under the 
constitutional principles outlined 
earlier, racial discrimination of any 
magnitude is unconstitutional.

When the Court of Appeals turned to 
the 20 percentage point statistic—  
representing the average racial 
disparity among cases similar in 
aggravation level to petitioner's —  the 
majority apparently became uncomfortable 
with any approach that treated such a 
figure as marginal. Instead, it felt 
compelled to dispense with its earlier 
assumption (J.A. ) that the Baldus

95



studies were valid. In a factual attack,
the court complained that the figures 
were not adequately explained and that 
they were not shown to be statistically 
significant. On both points the court 
ignored the record. Petitioner's experts 
carefully explained the basis of their 
calculations (Fed.Tr.1738-40), the 
importance of the numbers, the rationale 
of the "midrange" categories (id. 881- 
86; 1291-1300), and the statistical 
significance of each contributing 
figure. (Id. 1734-40; S.E. 50,54,68).

In sum, there is no constitutional 
warrant for the federal courts to 
overlook proven racial discrimination—  
especially in capital sentencing—  
merely because its impact is dubbed 
"marginal." Yet even if such a notion 
were permissible, petitioner has 
adequately demonstrated that powerful, 
biasing forces are at work shaping

96



Georgia's death-sentencing system in a 
racially discriminatory pattern, and 
that he is among those defendants most 
severely affected by the invidious 
forces.
D. The Court Of Appeals Erred in 
Demanding Proof of "Specific Intent To 
Discriminate" As A Necessary Element Of 
An Eighth Amendment Claim

(i) The Controlling Precedents
The primary concern of the 

Court's Eighth Amendment cases has 
always been with the results of the 
sentencing process: capital punishment 
is cruel and unusual if "there is no 
meaningful basis for distinguishing the 
few cases in which it is imposed from 
the many cases in which it is not." 
Furman v. Georgia, 408 U.S. at 313 
(1972)(White, J., concurring). Justice 
Stewart resolved Furman after 
"put[ting]. . . to one side" the issue 
of intentional discrimination. Id. at 
310. Justice Douglas similarly

97



disavowed that the "task . . .  to divine
what motives impelled these death 
penalties." JLd. at 253. No member of 
the Furman majority stated or hinted 
that proof of invidious intent had been 
necessary to his decision.

In its subsequent opinions, the 
Court has stressed that the ultimate aim 
of the Eighth Amendment is to "minimize 
the risk of wholly arbitrary and 
capricious action." Gregg v. Georgia, 
428 U.S. at 189. Such arbitrariness can 
afflict a system irrespective of 
conscious choice by specific actors, and 
it is the State which bears the 
"constitutional responsibility to tailor 
and apply its law in a manner that 
avoids" this outcome. Godfrey v. 
Georgia, 446 U.S. at 428; Eddings v. 
Oklahoma. 455 U.S. 104, 118 
(1982)(0'Connor, J., concurring); 
Gardner v. Florida, 430 U.S. 349, 357-58

98



(1977). These rulings in capital cases 
are consistent with the law of the 
Eighth Amendment in other contexts, 
where the constitutional touchstone has 
long been effects, not intentions. See 
Rhodes v. Chapman, 452 U.S. 337, 364 
(1981)(Brennan, J., concurring). See 
also id. at 345-46 (plurality opinion); 
Spain v. Procunier, 600 F.2d 189, 197 
(9th Cir. 1979); Rozecki v. Gaughan, 459 
F .2d 6, 8 (1st Cir. 1972).

The evil identified in Furman, the 
evil which the Eighth Amendment seeks to 
prevent, is the unequal treatment of 
equals in the most extreme sentencing 
decision our society can make. Gardner
v .____ Florida, 430 U.S. at 361.
Considerations of race are legally 
irrelevant to that decision; their 
systematic influence produces, by 
definition, a pattern of sentencing that 
is legally "arbitrary and capricious."

99



See generally, B. Nakell & K. Hardy, The
Arbitrariness of the Death Penalty
(1986)(forthcoming). The task of 
identifying precisely where and how, 
consciously or unconsciously, race is 
influencing the literally thousands of 
actors involved in capital sentencing—  
prosecutors, judges, jurors who assemble 
to make a single decision in a single
case, only to be replaced by other
jurors in the next case, and still
others after them —  is virtually 
impossible. Yet "the inability to
identify the actor or the agency has 
little to do with the constitutionality 
of the system." (J.A. ) (Hatchett, J.,
dissenting in part and concurring in 
part).

(ii) Petitioner's Evidence 
Whatever disagreements may surround 

the issue of intent, there is no room 
for dispute on the question of impact.

100



Georgia's gross racial disparities are 
stark: white victim cases are nearly
eleven times more likely to result in a 
death sentence- than black victim cases. 
As we have shown, even under the most 
searching statistical analyses, this 
disproportionate racial impact remains 
substantial and highly statistically 
significant. The State has never
refuted these results.

(iii) The Opinion Below 
The Court of Appeals held that 

"purposeful discrimination" is an 
element of an Eighth Amendment challenge 
to the arbitrary administration of a 
capital statute, at least where the 
challenge is based in part upon proof of 
racial disparities. (J.A. ). The court
acknowledged that "cruel and unusual 
punishment cases do not normally focus 
on the intent of the government actor." 
Id. Yet it announced that

101



where racial discrimination is 
claimed, not on the basis of 
procedural faults or flaws in 
the structure of the law, but on 
the basis of the decisions made 
within that process, then 
purpose, intent and motive are a 
natural component of the proof 
that discrimination actually 
occurred.

Id.
This opinion is plainly an exercise 

in ipse dixit reasoning. If 
"discrimination" in this passage means 
"intentional discrimination of the sort 
that violates the Equal Protection 
Clause," then the court fails to 
account for what the Eighth Amendment 
adds to the Fourteenth. If 
"discrimination" is synonymous with 
"racial disparity" —  the actual basis 
of petitioner's Eighth Amendment claim- 
- then even the court's linguistic logic 
evaporates completely. In any event, 
the majority below fails to address 
either the contrary holdings of this 
Court or the policies that lie behind

102



the Eighth Amendment cases. It supplies 
no justification for singling out race 
bias —  alone among all arbitrary 
factors that might affect a capital 
sentencing system —  and requiring that 
petitioner trace it back to an 
individual, consciously discriminating 
actor. "Identified or unidentified, the 
result of the unconstitutional 
ingredient of race . . .  is the same." 
(J.A. ) (Hatchett, J., dissenting in 
part and concurring in part). And it 
remains the same whether the racial 
ingredient comes into play through 
wilful bigotry or through more subtle 
processes of race-based empathies, 
apprehensions and value judgments 
operating within the framework of a 
highly discretionary capital sentencing 
procedure. See Turner v. Murray, 90 
L.Ed.2d at 35-36. However brought 
about, the result is nonetheless "a

103



pattern of arbitrary and capricious 
sentencing like that found
unconstitutional in Furman." Gregg v. 
Georgia, 428 U.S. at 195 n.46.

III.
THE COURT SHOULD EITHER GRANT PETITIONER 
RELIEF OR REMAND THE CASE TO THE COURT 
OF APPEALS FOR FURTHER CONSIDERATION 
UNDER APPROPRIATE LEGAL STANDARDS

In Skipper v. South Carolina,
__U.S.__, 99 L.Ed. 2d 1, 13 n.2 (1986),
Justice Powell observed in concurrence 
that "when some defendants are able to 
avoid execution based on irrelevant 
criteria, there is a far graver risk of 
injustice in executing others." The 
criterion of race —  that of a defendant 
or his victim —  is worse than 
"irrelevant": it is expressly forbidden
by the Constitution. Yet petitioner's 
evidence indicates (a) that race has 
played a substantial role in determining 
who will be executed and who will avoid 
execution in the State of Georgia, and

104



(b) that petitioner stands among the 
group of defendants upon whom Georgia's 
burden of racial bias falls most 
heavily.

The Court of Appeals, accepting the 
validity of petitioner's evidentiary 
submission, held that it failed to meet 
his burden of proof under the Eighth and 
Fourteenth Amendments. We have shown 
that this holding was error, requiring 
reversal. Since the proof of racial 
discrimination on this record is 
overwhelming and stands unrebutted 
despite its plain sufficiency to shift 
the burden of rebuttal to the State, we 
believe that nothing more is needed to 
support a decision by this Court 
upholding the merits of petitioner's 
Eighth and Fourteenth Amendment claims. 
However, inasmuch as the Court of 
Appeals pretermitted a review of the 
factual findings of the District Court

105



(J.A. ), this Court may prefer instead
to remand for further proceedings under 
appropriate constitutional standards.
See, e.q. , Bazemore v. Friday, __L.Ed.2d
at __.

While not strictly necessary to any 
holding that directs a remand, the Court 
might wish to announce standards to 
guide the Court of Appeals in addressing 
those remedial questions presented by 
petitioner's constitutional claims. In 
our judgment, the available remedial 
options would be affected considerably 
by the Court's choice of constitutional 
theory. Although this choice is a 
matter of little immediate moment to the 
present petitioner, 31 the consequences

31 The sole remedial issue in this 
habeas corpus proceeding is whether a 
petitioner "is in custody in violation 
of the Constitution or laws . . .  of the 
United States," 28 U.S.C. § 2241(c)(3); 
thus the only relief sought or possible 
under any theory is individual relief.

106



for other death-sentenced inmates in the
State of Georgia might vary
significantly depending upon it.

Under the Eighth Amendment, for 
example, proof that a particular capital 
sentencing system is being administered 
in an arbitrary or capricious pattern 
would presumably require the
invalidation of that system as a whole, 
or at least of all sentences imposed in 
the jurisdiction during the period 
covered by the proof. See Furman v. 
Georgia, supra. However, under the 
Fourteenth Amendment, the finding of an 
Equal Protection violation need not 
inevitably require a vacatur of all 
death sentences within the jurisdiction. 
In Mt. Healthy City Board of Educ. v. 
Doyle, 429 U.S. 274 (1977), the Court
reasoned that although an employee could 
not be discharged for the exercise of 
his protected First Amendment rights, an

107



employer was entitled to "show[] by a 
preponderance of the evidence that it 
would have reached the same decision . .
. even in the absence of" consideration 
of the impermissible factor. Id* at 287. 
In the capital sentencing context, an 
analogous approach, requiring proof by 
the State beyond a reasonable doubt, see 
Chapman v. California, 386 U.S. 18 
(1967),32 would allow a State, even if

32 Mt. Healthy expressly drew upon 
principles, developed in the context of 
the criminal law, "distinguish[ing] 
between a result caused by a 
constitutional violation and one not so 
caused." 429 U.S. at 286, citing Lyons 
v. Oklahoma, 322 U.S. 596 (1944); Wong 
Sun v. United States, 371 U.S. 471 
(1963); Parker v. North Carolina, 397 
U.S. 790 (1970). The Lyons line of 
cases is related to, though analytically 
distinct from, the Chapman "harmless 
error" line. The former holds that a 
constitutional violation may be 
disregarded if it did not in fact work 
any injury to a petitioner's substantive 
rights. Chapman permits a state to 
avoid a reversal by demonstrating beyond 
a reasonable doubt that, even if an 
injury to defendant's rights occurred, 
it was so insubstantial that it did not 
contribute to the defendant's conviction 
or sentence.

108



statuteits statute had been applied in 
violation of the Equal Protection 
Clause, to prove that, because of the 
extreme aggravation of a particular 
homicide, a death sentence would have 
been imposed, irrespective of racial 
considerations. Although Georgia could 
not make such a showing against inmates 
like petitioner, whose case was in the 
"midrange" of aggravation, it might have 
a stronger argument against those 
inmates whose crimes were highly 
aggravated, since race is less likely to 
have influenced the sentencing outcomes 
in their cases.

Whatever constitutional or remedial 
analysis is adopted by the Court, 
petitioner Warren McCleskey has 
presented evidence that fully 
establishes the constitutional merit of 
his claims. The sentence of death 
imposed upon him on October 12, 1978 by

109



the Superior Court of Fulton County is 
invalid.

CONCLUSION
The judgment of the Court of Appeals 

should be reversed.
Dated: August 21, 1986.

Respectfully submitted,
JULIUS L. CHAMBERS
JAMES M. NABRIT, III 
*JOHN CHARLES BOGER
DEVAL L. PATRICK
VIVIAN BERGER

99 Hudson Street
New York, New York 10013
(212) 219-1900

ROBERT H . STROUP 
141 Walton Street 
Atlanta, Georgia 30303

TIMOTHY K. FORD
600 Pioneer Building 
Seattle, Washington 98104

ANTHONY G. AMSTERDAM 
New York University 
School of Law 
40 Washington Sq. South 
New York, New York 10012
*Attorney of Record
ATTORNEYS FOR PETITIONER

110