Briefs for Petitioner and Respondent

    / J / 7 AJ | / ] \ "| { / ! 

Sa 2 ; 1 Y aotot No [= { = td Vg. “£1 |78e 
( /\ 7 

   



No. 84-6811 
  

  

IN THE 

Supreme Court of the United States 

OcToBER TERM, 1985 

WARREN MCCLESKEY, 
Petitioner, 

V. 

Rarru 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 

TimoTHY K. FORD JuLius L. CHAMBERS 
600 Pioneer Building JAMES M. NaBrIT, III 
Seattle, Washington 98104 *JoHN CHARLES BOGER 

DEeEvAL L. PATRICK 
ANTHONY G. AMSTERDAM 

ViviIAN BERGER 

Yon ere 99 Hudson Street 

40 Washi New York, New York 10013 
ashington Sq. South (212) 219-1900 

New York, New York 10012 

Attorneys for Petitioner RoBerT H. STROUP 
141 Walton Street 
Atlanta, Georgia 30303 

* Attorney of Record 

  

  

PRESS OF RAM PRINTING, HYATTSVILLE, MD 20781 (301) 864-6662  





  
    

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 vs a a ul, wa 

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 

iii 

18 

32 

32 

41 

    

 



    

DISCRIMINATION IN CAPITAL 

SENTENCING PROCEEDINGS ‘“ 5.0 vv» 45 

A. The Court of Appeals 
Ignored This Court's 
Decisions Delineating 
A Party's Prima Facie 
Burden Of Proof Under 
The Equal Protection 
Clauses . uv. + » + + 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 
CIAiM russ o>» + » » » B17 

III. THE COURT SHOULD EITHER GRANT 

PETITIONER RELIEF OR REMAND THE CASE 

TO THE COURT OF APPEALS FOR FURTHER 

CONSIDERATION UNDER APPROPRIATE LEGAL 

STANDARDS . « . + 4 os so + « « » o » 304 

° 

iv 

   



CONCLUSION: ... . ov Freres vv +» +» + 330 

 



  

TABLE OF AUTHORITIES 
  

  

Cases Pages 

Alabama v. Evans, 
481 U.S. 230. {1983 a snsussviwns ss 95 

Alexander v. Louisiana, 405 U.S. 
B25 (1972). cinivirwn es viunmmissnivioinys oe 47,48 

Avery v. Georgia, 345 U.S. 

550 £1953). citnirwis vino hovan mir bigs os 76 

Ballew v. Georgia, 
435 U.S. 223 41978) vuumris end v vers 84 

Batson v. Kentucky, U.s. ' 

90 L.Ed. 24 69 
(1986) vv rrvvvvrvins 24,26,27,33,47,74 

Bazemore v. Friday, U.S ’ 
106 8.0%. 3000 
(1986) inves 27,29,64,73,15,78,106 

Briscoe wv. Lahue, 

460 U.5. 328 (1983) cncevisbtnnmrevs 38 

Brown v. Board of Education, 

346 U.S. 483 (19548) vec eve coves 32 

Castaneda v. Partida, 

430 U.S. 482 

01972) ec nonamuns wn 27,49,56,65,73,79 

Chapman v. California, 
386 U.S. 18 (1967) «sere. Sone + vo 108 

Cleveland Board of Education v. 

LaFleur, 414 U.S. 632 (1974)...39,44 

vi 

   



RT, 

Coble v. Hot Springs School District 
No.6, 682 F. 24 721 (8th 
Cir: 1982 svn ves vases sone siisfoni 66 

Eastland v. TVA, 

704 F. 24 613 (31th Cir. 1983)... 66 

Eddings v. Oklahoma, 
458 U.8. 104 (1982 )iuiss viv s sin ois sivis 98 

EEOC v. Ball Corp., 
661 F. 24 531 {6th Cir. 1981)... B66 

Furman v. Georgia, 
408 U.S. 238 
$1972) sciniivinimin vive ois 24.,31.,41,97,107 

Gardner v. Florida, 

430 U.S. B49 {19T77) cvcvivininan 44 ,98,99 

General Building Contractors Ass'n, 
Inc. v. Pennsylvania, 458 U.S. 
B25. HL 1083) cles cunivnon nme os vivisiniv 34 

Giglio v. United States, 
408 U.S. 180 (1972). vo vrisiv.« vivian 4 

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

Graves v. Barnes, 

405 U.S, 1201 (1972) cen vnvonmns 95 

Gregg v. Georgia, 
428 U.S. 153 
{19TB8). vvins'c wininis 25,40,42,57,59,89,98 

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. y 

35 L. Bd. 24 222 

£1988) ine ee bei, 33,60,91 

Jones v. Georgia, 
B89 U.S. 24 (1967) uo vis vas von vie 48 

Loving v. Virginia, 
388 U.S. 1 (1967) viv ove’ sous sous 35 

Lyons v. Oklahoma, 
322: U.S. 896 (1944) . is vi ous oh 108 

McClesky v. State, 245 Ga. 108, 263 
S.E. 2d 14, cert. denied, 449 
U.S: 80% (R980), iv vivinivininriny vurviolind, 5 
  

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

McLaughlin v. Florida, 
379 U.S. 184 
LIDBA) os oi Te se ie 34,35,39 

Mt. Healthy City Board of Educ. v. 
Doyle, 429 U.S. 274 (1977)...3107,108 

Neal v. Delaware, 100 U.S. 370 

E1881). cctv isitdoce eas inne dnte 49 

Nixon v. Herndon, 273 U.S. 536 

GYD 27 Hee thoinic oo v 00 on oa dninian, ooeie oi 33 

Papasan v. Allain, U.S. ’ 
1068 S.Ct. 2932 (1986) . cc vn vn. 29,78 

Parker v. North Carolina, 

397 U:.S.:790 (1970) eivseiinssvecen 108 

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

Personnel Administrator of 

Massachusetts v. Feeney, 

viii



  

442 U.S. 256 
E1976) vivinvnolh oivinivin sivioiats sale ves 35,74 

Rhodes v. Chapman, 452 U.S. 337 

(1981) ce cvvcvvorvrvnrevonomstvises 99 

Roe v. Wade, 410 U.S. 113 

{1973). «+. 39,43 

Rogers v. Lodge, 458 U.S. 613 

{IOB 2) evra vis vie virions sie v vanes 50,60 

Rose v. Mitchell, 443 U.S. 545 

L 2979) 4 con vinininni nine utuiesonass 07 4:05 9 Sinn we 33 

Rozecki wv. Gaughan, 

459- F. 24 6 (1st Cir, 1972). «ves 99 

Segar v. Smith, 738 F. 2d 1249 

(DiC. Cir. 1984) ccvver rvs 66,76 

Skinner v. Oklahoma, 

316 U.S. 535 (1942) cevrvevvevnsnns 39 

Skipper v. South Carolina, U.S. 

, 90 L. Ed. 24 1 (1986). vse. 104 

Smith v. Texas, 311 U.S. 128 

(1040) sc ssnrsrmsessssovsrers 32,45 

Spain v. Procunier, 600 F. 2d 189 

{9th Cir, 1979) .ccevevsrvvevsrns 99 

Stanley v. Illinois, 
405 U.S. 648 (1972) ci ses onus 39,44 

Strauder v. West Virginia, 

100 U.S. 303 (1880). 0 ccs vv vms 34,41 

Sullivan v. Wainwright, 464 U.S. 109, 

(1983) ence eters vsvsasnssnrsreres 93 

Teamsters v. United States, 

431 U.S. 324 (3977 Yu vsissarnins cece 65 

ix 

 



    

Texas Dep't of Community Affairs v. 
Burdine, 450 U.S. 248 
{1981 seni tere e 29,48,75,76 

Turner v. Murray, g.S. ' 

90 L. BA. 24 27 
(1086)... cstv 24,33,56,76,103 

Vasquez v. Hillery, U.s. : 
88 L. Bd. 24 598 
PIOBBY , vw rvir vs vs nnn neue 24 

Village of Arlington Heights v. 
Metropolitan Housing Development 
Corp., 429. U.S. 252 
G0 Eh ee aE le SE TR 28,50,52,59 

Vuyanich v. Republic National Bank, 
505 F. Supp. 224 (N.D. Tex. 
1980) vacated on other grounds, 
232 B. 24 1195 (5th Cir.1984)... 68 
  

Wainwright v. Adams, 466 U.S. 964 
(1984) cvs rns neato nreeiiidiis 93 

Wainwright v. Ford, 467 U.S. 1220 
Bel Foss ip Eg TER Lf SE 93 

Washington v. Davis, 
426 U.S. 229 

(1976)... 00002.02:0. 27,32,47,49.,74 

Wayte v. United States, B.S. ; 
84 L. EQ. 24 547 (1988) ......... 49 

Whitus v. Georgia, 385 U.S. 545 
CY9B TY ics rsns tris vei iY, 47,56 

   



  

Wilkins v. University of Houston, 
654 F. 24 388 (5th Cir. 1981), 

vacated and remanded on other 
  

  

grounds, 459 U.S. 809 (1982).... 66 

Wolfe v. Georgia Ry. & Elec. Co., 
2 Ga. App. 499, 58 8S,.B. 899 
(1907) .c.. crvivirineir inv soBoBole de lo + sB3tode: v 003 61 

Wong Sun v. United States, 
371. U.S. 471 (19683 )uius scone iviosnincorens 108 

Yick Wo v. Hopkins, 
118 U.S. 856: {1886) wus «is sages + vive 33,56 

Zant v. Stephens, 462 U.S. 862 
{LOB IY viiis vie winisivivih viniivie wv ivieanin vais 43,57 

Zant v. Stephens, 456 U.S. 410 
(1982) A peY CUTIAM) viv vivir vivieiv ni ve 43 

Statutes 

28 U.S.C. § 1254 (1). ..0 ccnivniviee doers 2 

28 U.S:C+ 8: 2241 LC). {3 ev cit nsosis v.00 106 

Rule 406, F. Rule: BVI. civ ive vinvin wis 72 

Former Ga. Code Ann. § 27-2534.1 

IDS) i580 cates + sbi tale viVinioasie vs Tinie o vo 5 

Former Ga. Code Ann. § 27-2534.1 

{BI LB . cinie isle. citiininnn v viv vine cltioboiolivie o's 5 

Other Authorities 
  

D. Baldus & J. Cole, Statistical 
  

®i 

TT TN NE SE



    
  

Proof of Discrimination (1980).. 8 
  

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 ) . .covinviiwvvives iin 8 

  

  

  

  

Baldus, Pulaski & Woodworth, 

Comparative Review of Death 

Sentences: An Empirical Study of 

the Georgia Experience, 74 J. 
Crim. Law & Criminology 661 
11983) in ncancncimsnievavvisndnens 8 

  

  

  

Baldus, Pulaski, Woodworth & Kyle, 
Identifying Comparatively 
Excessive Sentences of Death: A 

Quantitative Approach, 33 Stan. 

L. Rev. 1 (31077) ..c caverns 8 

  

  

  

Baldus, Woodworth & Pulaski, 

Monitoring and Evaluating 
Contemporary Death Sentencing 

Systems: Lessons from Georgia, 

18 U.C. Davis L. Rev. 1375 

ution SRT i Ne EE 8 

  

  

  

Barnett, Some Distribution Patterns 

for the Georgia Death Sentence, 

18 U.C. Davis L. Rev. 1327 

{LOB Pe satis std side desir cB 51 

  

  

Bentele, The Death Penalty in Georgia: 
Still Arbitrary, 61 Wash. U.L.Q. 
573 (1988)... uvvrivvnivivedvenrdis 59 

  

  

Bowers & Pierce, Arbitrariness and 
Discrimination Under Post-Furman 
Capital Statutes, 26 Crime & 
Deling. 363 (1980), ...oo0veveves 51 

  

  

  

Finkelstein, The Judicial Reception of 
  

xii 

   



  

Multiple Regression Studies in Race 
and Sex Discrimination Cases, 80 

Colum. L. Rev. 737 (1980) ....... 66 

  

  

Fisher, Multiple Regression in Legal 
Proceedings, 80 Colum. L. Rev. 737 
pon i er SEE CP MOE EE Sen 66 

  

  

Gross, Race and Death: The Judicial 

Evaluation of Evidence of 

Discrimination in Capital 

Sentencing, 18 U.C. Davis L. Rev. 
I278 (1985) cctv eseeifogminieteils 81,90 

  

  

  

  

Gross & Mauro, Patterns of Death: 

Disparities in Capital Sentencing 
and Homicide Victimization, 37 

Stan. L. Rev, 27 (1985) cv.» 51 

  

  

  

H. Kalven & H. Zeisel, The American 

JULY (19868) wv viv ins vr vine avecon 84 
  

B. Nakell & K. Hardy, 
The Arbitrariness of the Death 

Penalty, (1986) 

(Fortheomingy ie cee ee eee ceas ves 100 

  

Report of the Joint Committee on 

Reconstruction at the First 

Session, Thirty-Ninth Congress, 
{18686 ) esas ct re nresns reassess 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). ccc. oR Ab bv eles 37 

Wolfgang & Riedel, Race, Judicial 
Discretion and the Death Penalty, 

407 Annals 119 (May 1973)..... Ee 

  

  

Wolfgang & Riedel, Race, Rape, and the 
  

xiii 

EE EEE aa



    
Death Penalty in Georgia, 45 Am. J. 

Orthopsychiat. 658 (1975)....... 51 
  

xiv { 

   



  

No. 84-6811 

IN THE 

SUPREME COURT OF THE UNITED STATES 

October Term, 1985 

  

WARREN McCLESKEY, 

Petitioner, 

- Afi 

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. (TT 0. 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 ik or Se + MR 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.B." 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." 

 



  

  

  

purported confessions made to a co- 

defendant and to a cellmate, Offie 

Evans. 2 

  

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 
Giglio [v. United States, 405 U.S. 150 

{1972)).," (J.A.188), the District Court 

granted petitioner habeas corpus relief: 
"[G]l]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.190). 

  

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.242-44). Five judges dissented, 
contending that Giglio had plainly been 
violated; four of the five also believed 
that the concealed promise was not 

4 

EEE ———————————



The jury convicted petitioner on all 

charges. Following the penalty phase, 

it returned a verdict finding two 

aggravating circumstances ¢ and 

recommending 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.312). 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, 
  

  

harmless. (J.A.287-89) (Godbold, Ch.J., 

dissenting in part); sé. at - 286; 

(Kravitch, J., concurring). 

8 The jury found that the murder 
had been committed during an armed 

robbery, former Ga. Code Ann. § 27- 

2534.1(b) (2) (current version O0.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 O0.C.G.A. § 

17-10-30(b)(8)). 

 



  

that 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, § 10), 

and in a "pattern . . . to discriminate 

intentionally and purposefully on 

grounds of race," in violation of the 

Equal Protection Clause. (Id. % 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 vv. 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. qq 45- 

50; 51-53). The District Court held an 

evidentiary hearing on these claims in 

August of 1983. 

 



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. 4 

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 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. 
  

 



  

use of statistical evidence. 5 

Professor Baldus testified concerning 

two meticulous and comprehensive studies 

he had undertaken with Dr. George 

Woodworth 6 and Professor Charles 

  

5 Professor Baldus js 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 pr. 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). 

 



Pulaski.’ 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 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. Dx. 

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 87% 

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. 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 

fe



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.g., 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.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 4&id 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." .(14...1589). 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 systen. (Fed, Tr. 810; 14265; 
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 Baldus's 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.136). 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.52.146). 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 . . . is 

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 

39 

 



ACCC ——————__ 

variable and could not predict every 

case outcome. (J.A.147). 

The District Court ended its opinion 

by rejecting the legal utility of such 

statistical methods altogether: 

[M]Jultivariate 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 y 
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.168-69) (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.246). Yet the 

Court proceeded to announce novel 

standards of proof that foreclose any 

20 

 



meaningful review of racial claims like 

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.259) (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.250). 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 problem lies in the principle 

that the proof, no matter how strong, of 

some disparity is alone insufficient." 

21 

 



  

(J.A.259). nin any discretionary 

system, some imprecision must be 

tolerated," the Court stated, and 

petitioner's proven racial disparities 

were "simply insufficient to support a 

raling . . . that racial factors are 

playing a role in the outcome sufficient 

to render the system as a whole 

arbitrary and capricious." (J.A.268). 

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 . ... then 

purpose, intent and motive are a natural 

component of the proof" (J.A.257) and 

"proof of a disparate impact alone is 

insufficient . . J. unless |, “i.e 3% 

compels a conclusion that . . . race is 

22 

  

 



  
  

intentionally being used as a factor in 

sentencing.” (J.A.258). 

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, 

23



state-sanctioned racial discrimination. 

The Court has been especially vigilant 

to prevent racial bias from weighing in 

the scales of criminal justice. See, 

  

  

  

e,d,, Batson v. Kentucky, _ U.S._ , 90 

L.EA.2d 69 (1988); Turner Vv. Murray, 

LaeBig i90 Be BEd 2, 3541086) 

Vasquez Vi Hillery, 1 i06Se ., + BB 

L.E4d.24 598 (19886). 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.g., 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 

24 

   



  

eliminate "the arbitrariness and 

capriciousness condemned by Furman." 

greqg Vv. Georgia, '428-03.S., '153,:198 
  

{1976) (opinion of Stewart, Powell & 

Stevens, J.J.). Accord, id. at 220-26 
  

{cpinion ‘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, 

"[albsent facts to the contrary," Gregg 

VY. 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 Zall."™ 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 

25 

 



practice. 

Petitioner McCleskey has now 

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.E4d.24 at 85. 
  

Henceforth, a capital defendant, rather 

than proving a prima facie case of 
  

discrimination by demonstrating the 

presence of substantial racial 

26 

   



disparities within a system "susceptible 

of abuse" -- thereby shifting the 

burden of explanation to the State, see, 

e.g., Castaneda Vv. Partida, 430 U.S. 
  

482, 494-495 (1977); Washington Vv. 
  

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 -thils 

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.261). 

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 

27 

   



Batson v. Kentucky, supra; Bazemore v. 
  

Friday, U.8. , 106 S.Ct. 3000°(1988); 

Arlington Heights Y. 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.271). 

This cannot be the law, unless there is 

to be a "death penalty exception" to the 

28 

  

    i i) 

                            

E
S
 

ia
r 

me
 

L
E
 

 



  

  

e
a
 

Equal Protection Clause. Just last Term, 

the Court unanimously held that such a 

restrictive judicial approach to 

statistical evidence was unacceptable 

error. Bazemore v. Friday, 106 S.Ct. at 
  

3009. See also Texas Department of 
  

Community Affairs vv. 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 Tern, 

29  



  

  

in Papasan v. Allain, U.S. , 106 
  

S.Ct. 2932 (1986), the Court expressly 

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, 

30 

RE hb @  t



this Court has held that it is the 

State's obligation "to tailor and apply 

its laws in a manner that avoids the 

arbitrary and capricious infliction of 

the death penalty." Godfrey v. Georgia, 
  

448 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 vv. 
  

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[1ing] «nei» WARtONly 

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 

31 

   



  

  

job promotion, or the selection of a 

jury, should not be disregarded when the 

stakes are life and death. Methods of 

proof and fact-finding accepted as 

necessary in every other area of law 

should not be jettisoned in this one. 

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. 

32 

msm



128, 132 (1940). It has forbidden 

discrimination required by statute, see, 

e.g., Brown v. Board of Education, 346 
  

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.24 at 82; Turner v. 
  

  

Fouche, 396 U.S. 346 (1970); Yick Wo 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). 1n 
  

the area of criminal justice, where 

racial discrimination "strikes at the 

fundamental values of our judicial 

system and our society as a whole," Rose 

33  



  

  

Vv, Mitchell, 443 U.S. 545, 5586 (1979), 
  

the Court has "consistently" articulated 

& "strong policy . ‘ . of combating 

racial discrimination.” Id. 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.s (1982), citing 

Strauder v. West Virginia, 100 U.S. 303, 
  

306-08 (1880); Ho Ah Kow wv. 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'n., inc. Va. 

34 

Rr A eee et f AN ASSO



Pennsylvania, 458 U.S. 375, 386-87 
  

(1982) .11 In 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 

  

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 . . By 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 wv. 
Virginia, 388. .U.8.. 1, 11 (1967). See 

also Personnel Administrator of 
  

Massachusetts vv. 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  



  

suggested, any "permissible state 

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 

36 

   



  

persons who committed criminal acts 

against blacks. When prosecutions did 

occur, authorities often acquitted or 

imposed disproportionately light 

sentences on those guilty of crimes 

against black persons. 12 

  

12 see, e.q9,., 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 111, at 1241 (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 [sic] 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 

 



  

The congressional hearings and 

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 wv. 
  

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. 

38 

 



    

The Court has often held that whenever 

either "fundamental rights" or "suspect 

classifications” are involved, state 

action "may be justified only by a 

'compelling state interest' . . . and 

legislative enactments must be 

narrowly drawn to express only the 

legitimate state interests at stake.” 

Roe v.: Wade, 410..0.5.:113, 158 (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 vv. Oklahoma, 316 U.S. 5385, 
  

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 

39 

  

 



  

room at night with a white woman (or 

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. 13 This 

  

13 The Court identified in Gregg 
Vv. 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 

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, 2100 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 

  

capital crime, it ought to deter such 
crime equally whether inflicted against 
black or against white citizens. 

41 

  

 



  

the capital statutes before him were 

"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." I4. 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 

race."l4 These observations illuminate 

the holding of Furman, reaffirmed by the 

Court in Gredq and subsequent cases, 

that the death penalty may "not be 

imposed under sentencing procedures that 

create] a substantial risk that it 

[will] ‘ : : be inflicted in an 

arbitrary and capricious manner." Gregg 

  

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



  

  

Vv. Georgia, 428 U.S. at 188; Godfrey v. 
  

Georgia, 446 U.S. at 428: Zant wv. 
  

Stephens, 456 U.S. 410, 413 (1982) (per 
  

curiam). 

The Court itself suggested 1n Zant 

VY. Stephens, 462 U.S. 862, 385 (1983), 
  

that 1% "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 wv. Wade, 
  

410 U.S. at 185, Legislative 

classifications that are unrelated to 

43 

REE... 

 



  

  

any valid purpose of a statute are 

arbitrary and violative of the Due 

Process Clause. Cleveland Board of 
  

Education vv. LaFleur, 414 U.S. 632 
  

(1974); Stanley vv. 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]l]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.261). 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 

46 

lO tC I



Baldus's real 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 vv. 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) 

ot 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 

47 

  
  
 



  

  

adequately the racial exclusion. | 
Alexander v. Louisiana, 405 U.S. 
[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 
Vv. Georgia, “389 U.S." 24, 25 

(1967). Rather the State must 
demonstrate that "permissible i 
racially neutral selection | 
criteria and procedures have 
produced the . . . result." 

  

90 L.Ed.2d 85-86. 

The approach is "a vraditional { 

feature of the common law," Texas Dep'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 i 

[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 . . . sharpen[ing] the 

inquiry into the elusive factual 

48 

ai 
 



    

question of intentional discrimination.” 

Id. at 255 n.8.15 

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 wv. Davis, 426 U.S. 
  

at 253 (Stevens, J.., concurring), 

Castaneda v. Partida, 430 U.S. 482, 494- 
  

95 (1977); cf. Wayte vy. United States, 
  

0S. , B84 L.E&.24 B47, 556 n.lo 

(1985), the guiding principle is that 

courts must make "a sensitive inquiry 

into such circumstantial and direct 

  

15The roots of this approach run 
back at least as far as Neal vv. 

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 391. Absent proof to support its 

contention, the State's unsupported 

assertion was held insufficient to rebut 

the prisoner's prima facie case. Id. 
  

49 

 



  

evidence of intent as may be available." 

Village of Arlington Heights Vv. 
  

Metropolitan Housing Development Corp., 
  

429 U.S. 252, "266 (1971). Accord, 

Rogers vv. 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 

50 

   



Georgia's capital statutes during the 

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 Baldus's 

research reached the same conclusions as 

the earlier studies, but there the 

resemblance ends: his work is vastly 

more detailed and comprehensive than any 

  

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 (1988). 

  

  

  

51  



  

prior sentencing study in Georgia or 

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' . vu mpovidel(s] 

an important starting point." Arlington 
  

Heights, 429 U.S. at 266. Here, the 

52 

 



Baldus studies reveal substantial, 

unadjusted racial disparities: a death- 

sentencing rate nearly eleven times 
  

higher in white-victim cases than in 

black-victim cases. (Fed.Tr.730-33; SE. 

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). 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 wast 

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. 131-52). 

54 

 



The results of these analyses were 

uniform. Race-of-victim 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 ' for 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 Ho 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, 45-43). 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. ee Bentele, The Death Penalty 
  

in Georgia: Still Arbitrary, 61 Wash. 
  

U. L.Q. 5723, 609-21 (1985) (examining 

charging and sentencing practices among 

Georgia prosecutors in the post-Furman 

period) .17 

c. The History of Discrimination 

Finally, "the historical background" 

of the State action under challenge "is 

  

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 
  

Bunter v. Underwood, U.S. , 88 1..E4.24 
  

222 (1985); Rogers v. Lodge, 458 U.S. 
  

813 (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 (I1d., 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.18. 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 to either require or 

support a decision for petitioner.” 

(J.A.246). The court in effect 

announced the abolition of the prima 

facie standard, and required instead 

62 

 



that petitioner produce evidence "so 

great that it compels a conclusion that 

the system is . v . arbitrary and 

capricious," (J.A.258) 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.250). 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.221-272). 

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 
  

  

Vv. Friday, 106 S.Ct. at 3009. 

(petitioner's' evidence need "not 

include all measurable variables 

thought to have an effect on [the matter 

at issuel]"). 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. "{Ooluxr 

cases make it unmistakably clear," 

Justice Stewart wrote in Teamsters v. 
  

United States, 431 U.S. 324, 339 (19717), 
  

"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 2a 

pattern or practice of discrimination.” 

Hazelwood School District. v. United 
  

States, 433 U.S. .299, .307-08 (1977). 

See, e.g. Castaneda v. Partida, 430 U.S. 
  

482, 493-96 (19177). 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, 96 S.Ct. at 3009, and has 
  

received wide acceptance in the lower 

courts. 1° 

(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.24 531 (6th Cir. 

19081); Coble vv... Hot Springs School 
  

District No. 6, 682 F.2d 721,731-32 (8th 
Cir. 1982); Eastland wv. TVA, ~704 F.24 

613..4{11th Cir. 1983); Segar v. Smith, 
1738 P.24 at 1261, 1278-79; Vuyanich v. 

  

  

  

  

  

Republic Nat'l Bank, supra. See 
generally Finkelstein, The Judicial 

  
  

Reception of Multiple Regression Studies 
in Race and Sex Discrimination Cases, 80 

Colum. .L. Rev. 1737 (1980); Fisher, 

Multiple Regression in Legal 
Proceedings, 80 Colum. LIL. Rev. 702 (1980). 

  

  

  

  

66 

 



affected their findings. See, e.qg., 

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] halve] 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; iS .E. 247-49). 

They used "weighted" and "unweighted" 

data. {Fed.Tr.6821-26; S.E. 868-89). 

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.713-92; 

sS.B. 858). 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.1301- 

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.3101; 1701- 

07; S.E. 64-617), 

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 {bY (2)" or 

"contemporaneous felony" aggravating 

circumstance (see DB 886); bh) =l1l 

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

70 

 



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

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.2l This 
  

21 The Court of Appeals states this 
proposition in varying forms: 
"[G]leneralized statistical studies are 

of little use in deciding whether a 
particular defendant has been 
unconstitutionally sentenced to death." 

(J.A.260). "No single petitioner could, 

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 «ds 

unclear why the majority was unwilling 

to permit recourse to ordinary fact- 

finding procedures for proof of of 

racially discrimination in capital 

sentencing. lt may be unwarranted 

skepticism regarding the probative power 

  

on the basis of these statistics alone, 

establish that he received the death 

sentence because, and only because, his 
victin was white.” (J.4.267). "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.270). 

22 cf. Fed. Rule Evid. 406. 

72 

 



of statistics "[w]lhere intent and 

motivation must be proved.” (J.A.250). 

Cf. Castaneda vv. Partida, 430 U.S. at 
  

495-97 & n.l7 (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.250 ). 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 Vv. Friday, 1086 
  

23 

 



S.Ct.at 3009. 

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.260). 

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.260-61). Yet 

the Court immediately faulted any 

  

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 vv. Kentucky, 90 

L.BEd.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



systemwide statistical study that did 

not take into account "every possible 

factor," e.dg., each of the "'countless 

racially neutral variables'" that it 

hypothesized must exist. {(J.A.261). It 

faulted even Professor Baldus's largest 

statistical models for this failure, 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. (J.A.272). 

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 wv. Burdine, 450 U.S. at 254; 
  

see, e.g., Bazemore v. Friday, 106 S.Ct. 
  

75 

 



  

at 3009, 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] . e «+ a "vacuum! it [was] . Ce 

one which the State [had to] . . . fill, 

by moving in with sufficient evidence to 
dispel the prima facie case of 
discrimination.'" Turner vv. Fouche, 396 
U.S. at 361, quoting Avery v. Georgia, 
345 U.S, 589, 562 (1953). See also 
Patton ..v. Mississippi, 332 .U.S. 4863, 
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.” Segar v. 
Smith, 738 F.2d at 1268, quoting Texas 
Dep't of Community Affairs wv. Burdine, 
450 U.S, at 253-885, 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 

G a 
 



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. Pirst, 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 

117   
 



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 vv. 
  

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 districts," 106 S.Ct. “at 2951~ 

53, the Court expiessly "decline[d] to 

append to the general requirements of an 

equal protection cause of action an 

additional threshold effects 

requirement.” Id. at 2946 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.?295 
  

  

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.i7. "The idea behind the rule of 
exclusion is not at all complex. If a 
disparity is sufficiently large, then it 
is unlikely that it lis 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 af 

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 victin 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? 
.[Clontrolling for age, smokers 
were l.7. times .more likely to 
die of coronary artery disease 
than nonsmokers. wees LS ACK ING 
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 

Supplemental Exhibits 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 

  

26Gross, Race and Death: The 
  

Judicial Evaluation Of Evidence of 
  

Discrimination in Capital Sentencing, 18 
U.C. Davis L. Rev. . 1275, 1307 (1985). 
  

81 

 



with the 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 ofthe victim: 
In between is the mid-range of 
cases where there is an 

approximately 20% racial 
disparity. 

1J.5.315) (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 

——



aggravated in group 8. The death- 

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 

27py. 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-victin 
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: 

[Wlhen you look at the cases in 
the mid-range, where the 

facts Go 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 . Lh. 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 

 



middle of the midrange of moderately 

aggravated cases. After reviewing the 

results of three separate statistical 

techniques, Dr. Woodworth concluded: 

[A]Jt 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 coefficient?2® 

  

29 The regression coefficient, as 
petitioner's experts explained, measures 

the average effect of a particular 

factor on the outcome of ‘a multiple 
regression analysis, arter 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~-82; S. EB. 81-83). 

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 

firs: computed the regression 

coefficients for those factors and for 

other important aggravating and 

mitigating factors. Then he rank- 

ordered then. As DB 81 demonstrates 

{S.E. 530), 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(s] 

to the imposition of the death penalty." 

Sregc v. Georgla, 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-victinm 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.273)y or "insufficient." (J.4.,268). 

This 1s a serious error. As one 

commentator has noted, although 

[i]t sounds right when the court 

describes the '8% 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. i 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.BE, 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 800% 

increase over the black-victim rate. It 

is obviously a gross mistake to view 

this difference as a "marginal" one. 

CE. "Hunter -v. Underwood, . U.S.  , 85 
  

L.E4d.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 

3 

 



  

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.266), 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.3.273). "In other “words, the court 

regarded the .06 figure as little more 

than a statistical aberration. However, 

this interpretation cannot be squared 

with the wunrebutted 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 wild 

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 1n 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.268). 

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 

  

3O0wainwright v. Ford, 467 U.S. 1220 
(1984) ; Wainwright A Adams, 466 

U.S. 964 (1984); Sullivan v. Wainwright, 
464 U.5. 109 (1983). 

  

  

  

93 

 



petitioners in those cases had all 

proffered other studies in which "[t]he 

bottom line figure [included] : . . BR 

'death-odds multiplier' of about 4.8 to 

in {J.A.268), and since "Baldus 

obtained a death-odds multiplier of 4.3 

to 1 in Georgia," a rejection of the | 

Baldus studies ris 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.” (J.A.269). 

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 

94 

 



magnitude of the disparities shown in 

the Florida studies; {ii) that this 

Court's orders respecting applications 

for stays of execution "may not be taken 

  

as a statement : . iv onithe 

merits," Graves Vv. Barnes, 405 U.S. 

1201, 1204 (1972) (Powell, wa in 

chambers); accord, Alabama v. Evans, 461 
  

U.8.: 230, 2368: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 

95 

 



assumption (J.A.246) that the Baldus 

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. (J.A.269-70). 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, {I4. 1734-40; S.E. 

80,54 ,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 

96 

 



  

adequately demonstrated that powerful, 

biasing forces are at work shaping 

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 1s no 

meaningful basis for distinguishing the 

few cases in which it is imposed from 

the many cases in which it is not.” 

Furmah Vv, Georgia, 408 U.S. ‘at” 313 
  

{1972)Y{White, J., concurring). Justice 

Stewart resolved Furman after 

out{ting). . . 0 one side" the issue 

87  



of intentional discrimination. 14, at 

  310. Justice Douglas similarly 

disavowed that the "task . . . to divine | 

what motives impelled these death | 

penalties." 4. 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 Vv. 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 vv. 
  

Georgia, 446 U.S. at 428; Eddings v. 
  

Oklahoma, 455 U.S. 104, 118 
  

98 

 



  
(1982) (O'Connor, Tsp concurring) ; 

Gardner v. Florida, 430 U.S. 349, 357-58 
  

(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 

w
 6 rs
 Rhodes v. Chapman, 4582 U.S, 337, 

  

(1981) (Brennan, J. concurring). See 

also id. at 345-46 (plurality opinion); 

Spain vv. Procunier, 800 F.24 189, 197 
  

(9th Cir. 1979); Rozecki v. Gaughan, 459 
  

F.28. 6, 8 (lst Cir. 1972). 

The evil identified in Furman, the 

evil which the Eighth Amendment seeks to 

prevent, 1s the unequal treatment of 

equals in the most extreme sentencing 

decision our society can make. Gardner 

Vv. Florida, 430 U.s. at 3681. 
  

Considerations of race are legally 

irrelevant to that decision; their 

systematic influence produces, by 

99  



definition, a pattern of sentencing that 

is legally "arbitrary and capricious.” | 

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 "[tlhe  dnability to 

identify the actor or the agency has 

little to do with the constitutionality 

of the system." (J.A.314) (Hatchett, J., 

dissenting in part and concurring in 

part). 

(ii) Petitioner's Evidence 

Whatever disagreements may surround 

100 

 



the issue of intent, there is no room 

for dispute on the question of impact. 

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.258). The court 

acknowledged that "cruel and unusual 

punishment cases do not normally focus 

101 

 



on the intent of the government actor." 

Id. Yet it announced that 

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. 

{(J.5.257). 

This opinion is plainly an exercise 

in ipse digic 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. 13 

"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 

102 

 



either the contrary holdings of this 

Court or the policies that lie behind 

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.  MNldentified or unidentified, the 

result of the unconstitutional 

ingredient of race . . « is the same." 

{J.5.314) (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 
  

103 

 



L..Fd.24 at 35-36. However brought 

about, the result is nonetheless "a 

pattern of arbitrary and capricious 

sentencing like that found 

unconstitutional in Furman." Gregg v.   

Georgia, 428 U.S. at 195 n.46. 

Iz. 

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 Ve. South Carolina, 
  

Su.s, 99 LL Bd. 24d 1,713" 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 

104 

   



who will be executed and who will avoid 

execution in the State of Georgia, and 

(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 

105 

   



Appeals pretermitted a review of the 

factual findings of the District Court 

(J.A.263), this Court may prefer instead 

to remand for further proceedings under 

  

appropriate constitutional standards. 

See, e.g., Bazemore v. Friday, 106 S.Ct. 

at 3010-11. 

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 , . i. of the 

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 

inan 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  wvacatur 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 

  

United States," 28 U.S.C. § 2241(c)i{3); 
thus the only relief sought or possible 
under any theory is individual relief. 

107  



not be discharged for the exercise of 

his protected First Amendment rights, an 

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 vv. 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 Vv, United States, 371 U.S. 471 
{1963); Parker Vv. North Carolina, 397 

U.8: 790 - {31970}. The Lyons line of 

cases is related to, though analytically 
distinct from, the Chapman "harmless 
arror” 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 

  

  

  

  

  

108 

   



its 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 

  

injury to defendant's rights occurred, 
jt was so insubstantial that it d4id not 

contribute to the defendant's conviction 

or sentence. 

109 

 



establishes the merit of his claims. The 

sentence of death imposed upon him on 

October 12, 1978 by “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, 111 

* 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 X. FORD 
600 Pioneer Building 
Seattle, Washington 98104 

ANTHONY G. AMSTERDAM 
New York University 
School of Law 
40 Washington Sg. South 
New York, New York 10012 

*Attorney 
of Record ATTORNEYS FOR PETITIONER 

110  



 



 



No. 84.6811 

  

  

In The 

= Supreme Court of the United States 
October Term, 1985 

PS 
v 

WARREN MOCTESKRY, 
Ae Petitioner, : 

  

% 

RALPH KEMP, Snparintiident. 
Soren Diagnostie and 

~ Classification Center, | 
(eis Respondent 

O 

REE ha ONY WRIT OF CERTIORARI q 
9 LTO THE UNITED STATES COURT OF APPEALS 
Gs FOR THE ELEVENTH CIRCUIT 

  

i. 
v 
  

— ft FOR REN iL 
OO 
WV   

‘Mary Bera ‘WESTMORELAND 
Assistant Attorney General 

Counsel of Record 
For Respondent = 

i FREE , MrcHAEL J. Bowers 
~ Please Seryorpond Li Attorney General 

he a & © Marion O. Goroox 
~~ Mary Bern WESTMORELAND First Assistant 
132 State Judicial Bldg. Attorney General 
40 Capitol Square, S.W. 
Atlanta, Georgia 30334 e WiLriam B. Hur, In. 

Senior Assistant 
ry (104) 65 fe Ci Attorney General 
  

  

~ COCKLE LAW BRIEF PRINTING CO., (800) 225-6964 
Or call collect (402) 342 2-2831 :  





QUESTIONS PRESENTED 

1. 

Is the statistical analysis which was presented to the 
district court inadequate to prove a constitutional viola- 
tion, both as a matter of fact and as a matter of law? 

2. 

Are the arbitrariness and capriciousness concerns of 

Furman v. Georgia, 408 U.S. 238 (1972), removed when a 

state properly follows a constitutional sentencing proce- 

dure? 

3. 

In order to establish a constitutional violation based 

on allegations of discrimination, must a petitioner prove 

intentional and purposeful discrimination? 

 



  

QUESTIONS PRESENTED 

STATEMENT OF THE CASE 

SUMMARY OF THE ARGUMENT 

il 

TABLE OF CONTENTS 

  

  

  

ARGUMENT 

1, 

11. 

Ii 

Iv. 

CONCLUSION 

STATISTICAL ANALYSES ARE INADE- 

QUATE AS A MATTER OF FACT AND LAW 

TO PROVE DISCRIMINATION UNDER THE 
FACTS OF THE INSTANT CASE. 

THE STATISTICAL ANALYSES IN THE IN- 
STANT CASE ARE INSUFFICIENT TO 
PROVE RACIAL DISCRIMINATION. ......om 

THE ARBITRARINESS AND CAPRICIOUS- 
NESS CONCERNS OF FURMAN V. GEOR- 
GIA, 408 U.S. 238 (1972), ARE REMOVED 
WHEN A STATE PROPERLY FOLLOWS A 
CONSTITUTIONAL SENTENCING PROCE- 
DURE. 

PROOF OF DISCRIMINATORY INTENT IS 
REQUIRED TO ESTABLISH AN EQUAL 
PROTECTION VIOLATION. 

  

  

  

  

16 

23 

37 

   



iii 

TABLE OF AUTHORITIES 

  

  

  

  

  

  

  

  

  

  

  

  

Page(s) 

Cases CITED: 

Bazemore v. Friday, — U.S. —, 106 S.Ct. 3000 
(1986) 10, 20 

Britton v. Rogers, 631 F.2d 572 (5th Cir. 1980), 
cert. demied, 451 U.S. 939 (1981) 8 

Caldwell v. Mississippi, 472 U.S. —, 105 S.Ct. 2633 
(1985) 13 

California v. Ramos, 463 U.S. 992 (1983) 28 

Castaneda v. Partida, 430 U.S. 482 (1977) 32 

Eastland v. Tennessee Valley Authority, 704 F.2d 
613 (11th Cir. 1983) 11 

Eddings v. Oklahoma, 455 U.S. 104 (1982) 13 

Enmund v. Florida, 458 U.S. 782 (1982) 27 

Equal Employment Opportunity Commission wv. 
Datapowmt Corporation, 570 F.2d 1264 (5th Cir. 
1978) 10 

Estelle v. Gamble, 429 U.S. 97 (1976) 24 

Furman v. Georgia, 408 U.S. 238 (1972) ee. 8, 24, 25,27, 
28, 29, 30 

Godfrey v. Georgia, 446 U.S. 420 (1980) «ooo. 27 

Gomillion v. Laghtfoot, 364 U.S. 339 (1960) 33,35 

Gregg v. Georgia, 428 U.S. 153 (1976) ......... 25, 26, 27, 28, 29 

Ingraham v. Wright, 430 U.S. 651 (1977) 24 

International Brotherhood of Teamsters v. United 

States, 431 U.S. 324 (1977) 9 

Johnson v. Uncle Ben’s Inc., 628 F.2d 419 (5th Cir. 
1980) 11 
  

  
Lockett v. Ohio, 438 U.S. 586 (1978) 13, 26, 27  



  

1v 

TABLE OF AUTHORITIES—Continued 

  

  

Page(s) 

Louisiana ex rel. Francis v. Resweber, 329 U.S. 
459, rhng. dented, 330 U.S. 853 (1947) 24 

Maxwell v. Bishop, 398 F.2d 138 (8th Cir. 1968), 
remanded on other grounds, 398 U.S. 262 (1970) ...... 12 

Mayor of Philadelphia v. Educational Equality 
League, 415 U.S. 605 (1974) 8 

McCleskey v. Kemp, 753 F.2d 877 (11th Cir. 1985) 
(em banc) 4 
  

McCleskey v. Zant, 580 F.Supp. 338 (N.D.Ga. 1984) L2 
3,4,17,18, 20,23 

McCorquodale v. Balkcom, 525 F.Supp. 408 (N.D. 
Ga. 1981), affirmed, 721 F.2d 1493 (11th Cir. 1983)..... 13 

  

  

  

  

  

McGautha v. California, 402 U.S. 183 (1971) eee... 12 

Oyler v. Boles, 368 U.S. 448 (1962) 31 

Personnel Admwmistrator of Massachusetts wv. 
Feeney, 442 U.S. 256 (1979) 33 

Proffitt v. Florida, 428 U.S. 242 (1976) 26 

Pullman-Standard v. Swint, 456 U.S. 273 (1982) .... 17 

Rogers v.Lodge, 468 U.S. B13 (1987) ......commmmmmimm 33 

Shaw v. Martin, 733 F.2d 304 (4th Cir. 1984) ..... 13 

Smith v. Balkcom, 660 F.2d 584 (5th Cir. 1981), on 
rehearing, 671 F.2d 858 (5th Cir. Unit B, 1982) .... 13 

Spinkellink v. Wainwright, 578 F.2d 582 (5th Cir. 
1978) 13 

Stephens v. Kemp, — U.S. —, 104 S.Ct. 562 (1983) ..... 28 

Trop v. Dulles, 356 U.S. 86 (1958) 14, 24   

Turner v. Murray, — U.S. —, 106 S.Ct. 1683 (1986) ..... 14 

United States v. Ironworkers Local 86, 443 F.2d 
544 (9th Cir. 1971) 10   

   



TABLE OF AUTHORITIES—Continued 

  

  

  

  

  

  

  

  

  

Page(s) 

United States v. United States Gypsum Co., 333 
U.S. 364 (1948) 17 

Valentino v. United States Postal Service, 674 
F.2d 56 (D.C.Cir. 1982) 11 

Village of Arlington Heights v. Metropolitan Hous- 
ing Development Corp., 429 U.S. 252 (1977) eee. 32, 33 

Wade v. Mississippi Cooperative Extension Ser- 
vice, 528 F.2d 508 (5th Cir. 1976) 10 

Washington v. Davis, 426 U.S. 229 (1976) 31, 32 

Wayte v. United States, — U.S. —, 105 S.Ct. 1524 
(1985) 33 

Whitus v. Georgia, 385 U.S. 545 (1967) 31 

Wilkerson v. Utah, 99 U.S. 130 (1878) 23 

Wilkins v. University of Houston, 654 F.2d 388 
(5th Cir. Unit A 1981) 11 

Witherspoon v. Illinois, 391 U.S. 510 (1968) wo. 14 

Woodson v. North Carolina, 428 U.S. 280 (1976) ........ 26 

Yick Wo v. Hopkins, 118 U.S. 356 (1886) 31, 33 

OTHER AUTHORITIES : 

Baldus & Cole, 4 Comparison of the Work of Thor- 
sten Sellin and Isaac Ehrlich on the Deterrent 
Effect of Capital Punishment, 85 Yale I. J. 170 
  

  

(1975) 15 

Fisher, Multiple Regression in Legal Proceedings, 
80 Colum. L.Rev. 702 (1980) 15, 20 

A. Goldberger, Topics in Regression Analysis (1968) 15 

 



  

vi 

TABLE OF AUTHORITIES—Continued 

Page(s) 

McCabe, The Interpretation of Regression Analy- 
sis Results wm Sex amd Race Discrimination 
Problems, 34 Amer. Stat. 212 (1980) 16 

Smith and Abram, Quantitative Analysis and Proof 
of Employment Discrimination, 1981 U.Ill. L.Rev. 
33 (1981) 15 

  

  

G. Wesolowsky, Multiple Regression Analysis of 
Variance (1976) 15 
  

   



No. 84-6811 

  
a) 
VU 

In The 

Supreme Court of the United States 
October Term, 1985 

0 
Vv 

WARREN MCCLESKEY, 

  

Petitioner, 
Vv. 

RALPH KEMP, Superintendent, 
Georgia Diagnostic and 
Classification Center, 

Respondent. 
  

0 
A S4 

ON WRIT OF CERTIORARI 
TO THE UNITED STATES COURT OF APPEALS 

FOR THE ELEVENTH CIRCUIT 

0). 
J 

BRIEF FOR RESPONDENT 

0. 
v 

STATEMENT OF THE CASE 

  

  

In addition to the statement of the case set forth by 

the Petitioner, Respondent submits the following regard- 

ing the district court and circuit court proceedings: 

Two different studies were conducted on the criminal 

justice system in Georgia by Professors Baldus and Wood- 

worth, that is, the Procedural Reform Study and the 

Charging and Sentencing Study. See McCleskey v. Zant, 

580 F.Supp. 338, 353 (N.D.Ga. 1984). The Petitioner pre- 

sented his case primarily through the testimony of Pro- 

fessor David C. Baldus and Dr. George Woodworth. Peti- 

tioner also presented testimony from Edward Gates as 

1 

 



  

well as an official from the State Board of Pardons and 

Paroles. The state presented testimony from two expert 

statisticians, Dr. Joseph Katz and Dr. Roger Buford. 

The district court made the following specific factual 

findings regarding the trustworthiness of the data base: 

[T]he court is of the opinion that the data base has 
substantial flaws and that the petitioner has failed to 
establish by a preponderance of the evidence that it 
1s essentially trustworthy. As demonstrated above, 
there are errors in coding the questionnaire for the 
case sub judice. This fact alone will invalidate several 
important premises of petitioner’s experts. Further, 
there are large numbers of aggravating and mitigat- 
ing circumstances data about which is unknown. Also, 
the researchers are without knowledge concerning the 
decision made by prosecutors to advance cases to a 
penalty trial in a significant number of instances. The 
court’s purpose here is not to reiterate the deficien- 
cies but to mention several of its concerns. It is a 
major premise of a statistical case that the data base 
numerically mirrors reality. If it does not in substan- 
tial degree mirror reality, any inferences empirically 
arrived at are untrustworthy. 

McCleskey v. Zant, supra, 580 F.Supp. at 360 (emphasis 

in original). (J.A. 144-5). 

The district court found as fact that ‘none of the 

models utilized by the petitioner’s experts were sufficient- 

ly predictive to support an inference of discrimination.” 

McCleskey v. Zant, supra at 361. (J.A. 149). 

The district court also found problems in the data due 

to the presence of multicollinearity. The district court 

noted that a significant fact in the instant case is that 

white victim cases tend to be more aggravated, that is 

correlated with aggravating factors, while black vietim 

   



cases tend to be more mitigated, that is correlated with 
mitigating factors. Every expert who testified, with the 
exception of Dr. Berk, agreed that there was substantial 
multicollinearity in the data. The district court found, 
“The presence of multi-colinearity substantially dimin- 
ishes the weight to be accorded to the circumstantial statis- 
tical evidence of racial disparity.” McCleskey v. Zant, 
supra at 364. (J.A. 153). The court then found Petitioner 

had failed to establish a prima facie case based either on 

race of victim or race of defendant. I4. 

Additionally, the district court found ‘“that any racial 

variable is not determinant of who is going to receive the 
death penalty, and, further, the court agrees that there is 
no support for a proposition that race has any effect in 
any single case.”” McCleskey v. Zant, supra at 366 (empha- 
sis in original). (J.A. 157). ““The best models which 
Baldus was able to devise which account to any significant 

degree for the major mnom-racial variables, including 
strength of the evidence, produce no statistically sigwifi- 

cant evidence that race plays a part in either of those de- 

cisions [by the prosecutor and jury] in the State of 

Georgia.” McCleskey v. Zant, at 368 (emphasis in origi- 

nal). (J.A. 159). 

Finally, the district court found that the analyses did 

not ‘““compare identical cases, and the method is incapable 

of saying whether or not any factor had a role in the de- 

cision to impose the death penalty in any particular case.’’ 

McCleskey v. Zant at 372 (emphasis in original). (J.A. 

168). “To the extent that McCleskey contends that he was 

denied either due process or equal protection of the law, 

his methods fail to contribute anything of value to his 

 



  

4 

cause.”’ McCleskey v. Zant at 372 (emphasis in original). 

(J.A.. 169), 

The court also found the Respondent presented direct 

rebuttal evidence to Baldus’ theory that contradicted any 

prima facie case of system-wide diserimination, if one had 

been established. McCleskey v. Zant at 373. 

In examining the issues, the Eleventh Circuit Court of 

Appeals assumed, but did not decide, that the research 

was valid because there was no need to reach the question 

of the validity of the research due to the court’s legal 

analysis. The court specifically complimented the district 

court on its thorough anaylsis of the studies and the evi- 

dence. The Eleventh Circuit observed that the first study, 

the Procedural Reform Study, revealed no race of de- 

fendant effects whatsoever and revealed unclear race of 

victim effects. McCleskey v. Kemp, 753 F.2d 877, 8387 (11th 

Cir. 1985) (em banc). As to the Charging and Sentencing 

Study, the court concluded, ‘‘ There was no suggestion that 

a uniform institutional bias existed that adversely affected 

defendants in white vietim cases in all circumstances, or a 

black defendant in all cases.”” Id. Finally, the court con- 

cluded the following in relation to the data specifically re- 

lating to the county in which the Petitioner was convicted, 

that is, Fulton County, Georgia: 

Because there were only ten cases involving police 
officer victims in Fulton County, statistical analysis 
could not be utilized effectively. Baldus conceded that 
it was difficult to draw any inference concerning the 
overall race effect in these cases because there had 
been only one death sentence. He concluded that based 
on the data there was only a possibility that a racial 
factor existed in McCleskey’s case. 

Id. at 887 (emphasis in original). 

   



Any further factual or procedural matters will be 
discussed as necessary in the subsequent portion of the 
brief. 

  (=)
 

SUMMARY OF THE ARGUMENT 

Although the petition in the instant case lists five 

questions presented, the main focus of this case is simply 

one of whether there has been racial discrimination in the 
application of the death penalty in Georgia and, in par- 

ticular, whether there was racial discrimination in the im- 

position of the death penalty upon the Petitioner. An- 

other way of looking at this issue is whether the Petitioner 

was selectively prosecuted and sentenced to death based 

on his race and that of the vietim or whether Petitioner’s 

sentence 1s disproportionate. Regardless of the standard 

to be applied, an appropriate consideration is the intent 

of the decision-makers in question. A review of the cases 

of this Court dealing with death penalty statutes shows 

that the general arbitrariness and capriciousness which 

concerned the Court in 1972 is no longer a consideration 

if a state follows a properly drawn statute and if the 

jury’s discretion is properly channeled. Thus, the focus 

in an Kighth Amendment analysis becomes a question of 

whether the sentence in a given case is ‘‘arbitrary’’ in the 

sense of being an aberration. The evidence in the instant 

case shows that the Georgia statutory scheme is funection- 

ing as it was intended to function and that those cases 

which are more severe are receiving stronger penalties 

while the less severe cases are receiving lesser penalties. 

There is no evidence to show that the Petitioner’s sentence 

 



  

in the instant case was arbitrary or capricious and no evi- 

dence to show that either the prosecutor or the jury based 

their decision on race. 

In relation to an equal protection context, it has al- 

ways been recognized that intentional and purposeful dis- 

crimination must be established for a constitutional viola- 

tion to be proven. Although intent may be inferred from 

circumstantial evidence, the circumstantial evidence must 

be sufficient to establish a prima facie case of discrimina- 

tion before intent will be inferred. Even if a prima facie 

case is shown, the Petitioner would still have the ultimate 

burden of proof after considering any rebuttal evidence. 

In evaluating facts and circumstances of a given case, 

the court must consider the totality of the circumstances 

in determining whether the evidence is sufficient to find 

intentional and purposeful discrimination. Although sta- 

tistics are a useful tool in many contexts, in the situation 

presented involving the application of the death penalty, 

there are simply too many unique factors relevant to each 

individual case to allow statistics to be an effective tool in 

proving intentional discrimination. Furthermore, the Peti- 

tioner’s statistics in the instant case were found to be inval- 

id by the district court, which was the only court making 

any factual findings in relation to those statistics. Thus, 

the clearly erroneous standard should apply to those factu- 

al findings. Furthermore, when a plausible explanation is 

offered, as it was in the instant case, that is, that white 

vietim cases are simply more aggravated and less miti- 

gated than black victim cases and that various factors 

tainted the statistics utilized, statistics alone or a disparity 

alone is clearly insufficient to justify an inference of dis- 

crimination. Furthermore, the statistics in question fail 

   



to take into consideration significant factors. Thus, the 

statistics in the instant case do not give rise to an infer- 

ence of diserimination. 

When reviewing all of the evidence in the instant case, 

it is clear that the findings of fact made by the district 

court are not clearly erroneous and that the statistical 

study in question should not be concluded to be valid so 

as to raise any inference of discrimination. The Peti- 

tioner failed to make a prima facie showing of diserimina- 

tion and did not carry the ultimate burden of proof on the 

factual question of intent. Furthermore, Petitioner simply 

failed to show that his death sentence was arbitrary or 

capricious or was the result of racial discrimination either 

on the part of the prosecutor or on the part of the jury. 

  o 

ARGUMENT 

I. STATISTICAL ANALYSES ARE INADE.- 
QUATE AS A MATTER OF FACT AND LAW 
TO PROVE DISCRIMINATION UNDER THE 
FACTS OF THE INSTANT CASE. 

Respondent submits that the type of statistical an- 

alyses utilized in the instant case are not appropriate in a 

death penalty case when trying to evaluate the motivation 

behind a prosecutor’s use of his discretion and the jury’s 

subsequent exercise of discretion in determining whether  



  

or not a death sentence should be imposed.! Hach death 

penalty case is unique and even though statistics might be 

useful in jury composition cases or Title VII employment 

discrimination cases where there are a limited number of 

factors that are permissibly considered, in the instant case 

where the prosecutor has discretion to pursue a case 

through the criminal justice system and can consider any 

number of subjective factors and where a jury has com- 

plete discretion with regard to extending mercy, the sub- 

jective factors cannot be accounted for in a statistical 

analysis such as that utilized by the Petitioner in the in- 

stant case. Thus, Respondent would submit that this 

Court should completely reject the use of this type of sta- 

tistical analysis as inappropriate in this case. 

Even in the cases that have utilized statistical analysis 

in a context other than that present in the instant case, the 

courts have acknowledged various concerns with these 

analyses. This Court has recognized in another context, 

“Statistical analyses have served and will continue to 

serve an important role as one indirect indicator of racial 

diserimination in access to service on governmental bod- 

ies, particularly where, as in the case of jury service, the 

duty to serve falls equally on all citizens.”” Mayor of 

Philadelphia v. Educational Equality League, 415 U.S. 

IRespondent submits that a claim of discrimination based 
on race of victim is not cognizable under the circumstances of 
the instant case. At least one circuit court has specifically re- 
jected statistical evidence based on the race of the victim, find- 
ing that the defendant lacked standing. Britton v. Rogers, 631 
F.2d 572, 577 n.3 (5th Cir. 1980), cert. denied, 451 U.S. 939 
(1981). Even those justices raising a question of possible racial 
discrimination in Furman v. Georgia, 408 U.S. 238 (1972), seemed 
to focus on race of the defendant and not race of the victim. 
Thus, Respondent submits that the instant claim is not cognizable 
due to the lack of standing. 

   



605, 620 (1974) (emphasis added). In the instant case, 
however, there is no such uniform ‘‘duty’’ as in the jury 

composition cases, as all citizens are certainly not equally 

eligible for a death sentence, nor are even all perpetra- 

tors of homicides or murders equally eligible for a death 

sentence. 

A central case regarding the use of statistics by this 
Court arises in International Brotherhood of Teamsters v. 
United States, 431 U.S. 324 (1977). Again, this was in the 

context of a Title VII action and not in a case such as the 
instant one involving so many subjective factors. The 
Court noted prior approval of the use of statistical proof 
‘‘where it reached proportions comparable to those in this 

case to establish a prima facie case of racial discrimination 

in jury selection cases.’”’ Id. at 339. The Court also noted 

that statistics were equally competent to prove employ- 

ment discrimination, which once again is different from 

the type of discrimination sought to be proved in the in- 

stant case. The Court specifically concluded, ‘‘ We caution 

only that statistics are not irrefutable; they come in in- 

finite variety and like any other kind of evidence, they 

may be rebutted. In short their usefulness depends on all 

of the surrounding facts and circumstances.” Id. at 340. 

Thus, it is imperative to examine all of the facts and cir- 

cumstances to determine whether the statistics in a given 

case are even useful for conducting the particular analy- 

sis. In Teamsters, supra, the Court also had 40 specific 

instances of discriminatory action to consider in addition 

to the statistics and noted that even ‘‘fine tuning of the 

statistics could not have obscured the glaring absence of 

minority line drivers.”’ Id. at 342 n.23. Thus, the Court 

did not focus exclusively on the statisties.  



  

10 

Problems have also been noted revolving around the 

particular use of statistics in any given case, many of 

which occur in the studies presented to the distriet court 

in the case at bar. In Bazemore v. Friday, — U.S. —, 106 

S.Ct. 3000 (1986), the Court examined regression analyses 

and concluded that ‘‘the omission of variables from a re- 

gression analysis may render the analysis less probative 

than it otherwise might be’’ while noting that this would 

not generally make the analysis inadmissible. Id. at 3009. 

The Court did go on to note that there could be some cases 

in which the regression was so incomplete as to be inad- 

missible as irrelevant. 

Circuit courts have also utilized statistics but have 

continually urged caution in their utilization even in jury 

selection and Title VII cases. Also, the courts frequently 

had other data on which to rely in addition to the statisti- 

cal analyses. See United States v. Ironworkers Local 86, 

443 F.2d 544 (9th Cir. 1971) ; Wade v. Mississippi Coopera- 

tive Extension Service, 528 F.2d 508 (5th Cir. 1976). The 

circuit courts have also recognized that statistical evidence 

can be part of the rebuttal case itself. The Fifth Circuit 

Court of Appeals examined a Title VII case in which the 

statistics relied upon by the plaintiff actually formed the 

very basis of the defendant’s rebuttal case, that is that 

there was a showing that the statistics were not reliable. 

Equal Employment Opportunity Commission v. Datapoint 

Corporation, 570 F.2d 1264 (5th Cir. 1978). In that case, 

the court noted ‘‘while statistics are an appropriate 

method of proving a prima facie case of racial diserimina- 

tion, such statistics must be relevant, material and mean- 

ingful, and not segmented and particularized and fash- 

ioned to obtain a desired conclusion.’”’ Id. at 1269. See 

   



11 

also Johnson v. Uncle Ben’s Inc., 628 F.2d 419 (5th Cir. 
1980). 

Circuit courts have also noted that due to the ‘‘in- 

herently slippery nature of statistics’’ they are also sub- 

Jeet to misuse. See Wilkins v. University of Houston, 654 

F.2d 388 (5th Cir. Unit A 1981). In particular, that court 

focused on the fact that even though multiple regression 

analysis was a sophisticated means of determining the 

effects of factors on a particular variable, such an analy- 

sis was subject to misuse and should be employed with 

great care. Id. at 402-3. Other courts have emphasized 

that even though every conceivable factor did not have to 

be considered in a statistical analysis, the minimum ob- 

jective qualifications had to be included in the analysis 

(in an employment context). ‘‘[W]hen the statistical evi- 

dence does not adequately account for ‘the diverse and 

specialized qualifications necessary for [the positions 

in question],’ strong evidence of individual instances of 

diserimination becomes vital . . . .”” Valentino v. United 

States Postal Service, 674 F.2d 56, 69 (D.C.Cir. 1982). 

The Eleventh Circuit Court of Appeals has examined 

statistical analyses and noted that the probative value of 

multiple regressions depends upon the inclusion of all 

major variables likely to have a large effect on the de- 

pendant variable and also depends on the validity of the 

assumptions that the remaining effects were not corre- 

lated with independent variables included in the analysis. 

The court also specifically questioned the validity of step- 

wise regressions, such as those used in the instant pro- 

ceedings. Fastland v. Tennessee Valley Authority, 704 

F.2d 613, 621 n.11 (11th Cir. 1983). The court emphasized  



  

12 

that a study had to begin with a decent theoretical idea of 

what variables were likely to be important. 

Thus, examining a statistical analysis depends in part 

on the question of whether the analysis incorporated the 

requisite variables and whether there is an appropriate 

theoretical base for the incorporation of the variables. As 

found by the district court in the instant case, none of the 

models utilized by Professor Baldus necessarily reflected 

the way the system acted and specifically did not include 

important factors, such as credibility of the witnesses, 

the likelihood of a jury verdict, and subjective factors 

which could be appropriately considered by a prosecutor 

and by a jury. Thus, the district court properly rejected 

the statistical analyses in question. 

More difficult problems arise with the attempted use 

of statistics in death penalty cases. In 1968 problems were 

found with the utilization of statistics, specifically pre- 

sented by Marvin Wolfgang. The circuit court concluded 

that the study presented in that case was faulty for vari- 

ous reasons, including failing to take variables into account 

and failing to show that the jury acted with racial dis- 

crimination. The court also emphasized that it was con- 

cerned in that case with the defendant’s sentencing out- 

come and only his case. The court concluded that the sta- 

tistical argument did nothing to destroy the integrity of 

the trial. Maxwell v. Bishop, 398 F.2d 138 (8th Cir. 1968), 

remanded on other grounds, 398 U.S. 262 (1970). 

An additional factor in the death penalty situation 

comes from the unique nature of the death sentence it- 

self and the capital sentencing system. In McGautha v. 

California, 402 U.S. 183 (1971), the Court noted the diffi- 

   



13 

culty in identifying beforehand those characteristics which 

could be utilized by a sentencing authority in imposing 

the death penalty and the complex nature of those fac- 

tors. Other circuit courts have rejected statistical an- 

alyses due to just such a reason. See Spinkellink v. Wain- 

wright, 578 F.2d 582 (5th Cir. 1978); Smith v. Balkcom, 

660 F.2d 584 (5th Cir. 1981), on rehearing, 671 F.2d 858 

(5th Cir. Unit B, 1982); McCorquodale v. Balkcom, 525 

F.Supp. 408 (N.D.Ga. 1981), affirmed, 721 F.2d 1493 (11th 

Cir. 1983). 

In cases upholding the constitutionality of various 

death penalty schemes, the Court has recognized that it is 

appropriate to allow a sentencer to consider every aspect 

regarding the defendant and the crime in question in exer- 

cising the discretion as to whether to extend mercy or im- 

pose the death penalty. Thus, in Eddings v. Oklahoma, 

455 U.S. 104 (1982) the Court noted that the rule set down 

in Lockett v. Ohio, 438 U.S. 586 (1978) was a product of a 

‘‘history reflecting the law’s effort to develop a system 

of capital punishment at once consistent and principled 

but also humane and sensible to the uniqueness of the indi- 

vidual.”” Eddings, supra at 110. 

Other factors that have been recognized by courts as 

being appropriate in a death penalty case and in the prose- 

cutor’s discretion are the willingness of a defendant to 

plead guilty, as well as the sufficiency of the evidence 

available. Shaw v. Martin, 733 F.2d 304 (4th Cir. 1984). 

As recently as 1986, this Court has acknowledged that in 

a capital sentencing proceeding the jury must make a 

“highly subjective, ‘unique, individualized judgment re- 

garding the penalty that a particular person deserves.’ ’’ 

Caldwell v. Mississippi, 472 U.S. —, 105 S.Ct. 2633, 2645-6  



  

14 

n.7 (1985); Turner v. Murray, — U.S. —, 106 S.Ct. 1683 

(1986). In this context, ‘‘it is the jury that must make the 

difficult, individualized judgment as to whether the de- 

fendant deserves the sentence of death.” Turner v. Mur- 

ray, supra 106 S.Ct. at 1687. This focuses on what has 

long been recognized as one of the most important func- 

tions that a jury can perform, that is, ‘‘to maintain a link 

between contemporary community values and the penal 

system—a link without which the determination of punish- 

ment could hardly reflect ‘the evolving standards of de- 

cency that mark the progress of a maturing society.’ ”’ 

Witherspoon wv. Illinois, 391 U.S. 510, 519 n.15 (1968), 

quoting, Trop v. Dulles, 356 U.S. 86, 101 (1958). 

Thus, the myriad of factors that are available for 

consideration by a prosecutor in exercising his discretion 

and by a jury in determining whether to extend mercy to 

a given defendant simply makes the utilization of these 

types of statistical analyses unworkable in a death penalty 

context. It is simply impossible to quantify subjective 

factors which are properly considered both by the prosecu- 

tor and by the jury in reaching these determinations. In 

fact, the evidence in the instant case fails to take into ac- 

count these subjective factors, including the information 

known to the decision-maker, the likelihood a jury would 

return a verdict in a particular case, the possible credi- 

bility of individual witnesses, the availability of witnesses 

at the time of trial, the actual sufficiency of the evidence 

as determined by the prosecutor himself as well as num- 

erous other factors. 

In addition to all the above, commentators have also 

recognized that many of the factors present in the instant 

case cause problems with utilizing statistical analyses. 

   



15 

Professor Baldus himself has noted that ‘‘statistical so- 

phistication is no cure for flaws in model construction and 

research design.”” Baldus & Cole, 4 Comparison of the 

Work of Thorsten Sellin and Isaac Ehrlich on the Deter- 

rent Effect of Capital Punishment, 85 Yale L. J. 170, 173 

(1975). In that same article, Professor Baldus acknowl- 

edged that the deterrent effect of capital punishment was 

just such a type of study that would be best suited by 

simpler methods of study than statistical analysis. Id. 

Other authors have questioned the validity of statistical 

methods which include inappropriate variables in the analy- 

sis as well as those which fail to include necessary vari- 

ables. See Finkelstein, The Judicial Reception of Multi- 

ple Regression Studies in Race and Sex Discrimination 

Cases, 80 Colum. L.Rev. 737, 738 (1980). Other authors 

have also agreed with the testimony of the experts in this 

case regarding the problems presented by multicollinearity 

as well as the problems in utilizing stepwise regressions. 

See Fisher, Multiple Regression in Legal Proceedings, 80 

Colum. L.Rev. 702 (1980) ; See also G. Wesolowsky, Multi- 

ple Regression Analysis of Variance (1976); A. Gold- 

berger, Topics wn Regression Analysis (1968). 

Finally, certain authors have questioned the utilization 

of statistical analyses even in employment discrimination 

cases noting ‘‘it may be impossible to gather data on many 

of these differences in qualifications and preferences. 

Consequently, there will likely be alternative explanations, 

not captured by the statistical analysis, for observed dis- 

parities. . . . These alternative explanations must be taken 

into consideration in assessing the strength of the in- 

ference to be drawn from the statistical evidence.’”” Smith  



  

16 

and Abram, Quantitative Analysis and Proof of Employ- 

ment Discrimination, 1981 U.Ill. L.Rev. 33, 45 (1981). 

Respondent submits that a consideration of the sta- 

tistical analysis in the instant case reflects that it simply 

fails to comply with the appropriate conventions utilized 

for this type of analysis in that it fails to include appropri- 

ate variables, fails to utilize interaction variables, fails 

to specify a relevant model and has other fallacies, includ- 

ing multicollinearity which render the analysis nonpro- 

bative at best. As noted by a statistician in an article re- 

garding race and sex discrimination and regression analy- 

SiS: 

It should be again emphasized that a statistical analy- 
sis provides only a limited part of the total picture that 
must be presented to prove or disprove discrimina- 
tion. . . . “No statistician or other scientist should 

ever put himself/herself in a position of trying to 
prove or disprove discrimination.’ 

McCabe, The Interpretation of Regression Analysis Re- 

sults im Sex and Race Discrimination Problems, 34 Amer. 

Stat. 212, 215 (1980). 

II. THE STATISTICAL ANALYSES IN THE IN- 
STANT CASE ARE INSUFFICIENT TO 
PROVE RACIAL DISCRIMINATION. 

As noted previously, courts and commentators have 

expressed reservations about the use of statistics in at- 

tempting to prove discrimination. Respondent submits 

that even if the Court concludes statistical analysis is ap- 

propriate in a death penalty context, the ‘‘statistics’’ pre- 

sented to the district court are so flawed as to have no pro- 

—
p
 

   



e
e
 

17 

bative value and, thus, cannot satisfy the Petitioner’s bur- 

den of proof.? 

Petitioner claims that the studies in question are the 

product of carefully tailored questionaires resulting in the 

collection of over 500 items of information on each case. 

The Respondent has proven, and the district court found, 

that the data bases are substantially flawed, inaccurate 

and incomplete. 

As noted previously, statistical analyses, particularly 

multiple regressions, require accurate and complete data 

to be valid. Neither was presented to the district court. 

Design flaws were shown in the questionnaires utilized to 

gather data. There were problems with the format of 

critical items on the questionnaires, such that there was 

an insufficient way to account for all factors in a given 

case. ‘‘An important limitation placed on the data base 

was the fact that the questionnaire could not capture every 

nuance of every case.”” McCleskey v. Zant, supra at 356. 

(J.A. 136). 

Further, the sources of the information were notice- 

ably incomplete. Hven though the Petitioner insisted that 

2]t is clear that the findings by the district court in regard 

to the question of intent and the evaluation of the statistical 

analysis are subject to the clearly erroneous rule. In United 

States v. United States Gypsum Co., 333 U.S. 364 (1948), the 
Court acknowledged that the clearly erroneous rule set forth in 
rule 52(a) of the Federal Rules of Civil Procedure applied to 
factual findings. “A finding is ‘clearly erroneous’ when although 
there is evidence to support it, the reviewing court on the en- 
tire evidence is left with the definite and firm conviction that a 
mistake has been committed.” Id. at 395. This principle has 
been held to apply to factual findings regarding motivations 
of parties in Title VII actions and it has been specifically held 
that the question of intentional discrimination is a pure question 
of fact. Pullman-Standard v. Swint, 456 U.S. 273, 287-8 (1982).  



  

18 

he relied on State sources, obviously those sources were not 

designed to provide detailed information on each case. As 

found by the district court, ‘‘the information available to 

the coders from the Parole Board files was very summary 

in many respects.’’ McCleskey v. Zant, supra at 356. (J.A. 

137). These summaries were brief and the police reports 

from which the Parole Board summaries were prepared 

were usually only two or three pages long. (F.H.T. 1343; 

J.A. 137). As found by the district court: 

Because of the incompleteness of the Parole Board 
studies, the Charging and Sentencing Study contains 
no information about what a prosecutor felt about the 
credibility of any witnesses. R 1117. It was occasion- 
ally difficult to determine whether or not a co-perpe- 
trator testified in the case. One of the important 
strength of the evidence variables coded was whether 

or not the police report indicated clear guilt. As the 
police reports were missing in 75% of the cases, the 
coders treated the Parole Board summary as the po- 
lice report. R 493-94. Then, the coders were able to 
obtain information based only upon their impressions 
of the information contained in the file. R 349. 

McCleskey v. Zant, supra at 357. (J.A. 137). 

Furthermore, questionaires were shown to be mis- 

coded. It was also shown there were differences in judg- 

ment among the coders. (F.H.T. 387). 

Respondent also established that there were numerous 

inconsistencies between the coding for the Procedural Re- 

form Study and the Charging and Sentencing Study. (J.A. 

77-80; S.E. 78; Respondent’s Exhibit 20A). These oc- 

curred in some variables generally considered to be im- 

portant in a sentencing determination. 

   



S
w
a
y
 

19 

A further problem with the data base is due to the 

large number of unknowns. Although Petitioner claims to 

have collected information on over 500 variables relating 

to each case, the evidence showed that in the Charging and 

Sentencing Study alone there are an average of at least 33 

variables coded as unknown for each questionnaire. (J.A. 

139). A review of Respondent’s Exhibits Nos. 17A and 18A 

shows the extent to which unknowns pervade the so-called 

complete data base. For example, in the Charging and 

Sentencing Study there are 445 cases in which it was un- 

known if there was a plea bargain. (S.E. 73-74; J.A. 69- 

74). Further complicating the data is the fact that Baldus 

arbitrarily coded unknowns as if the information did not 

exist without any knowledge as to whether the information 

was known to the prosecutor or jury. 

Even though attempts were made in the district court 

to discount the unknowns, Petitioner did not succeed. In 

fact the district court concluded the so-called ‘worst case’’ 

analysis failed to prove that the coding decisions on the 

unknowns had no effect on the results. (J.A. 142). The 

Respondent also introduced evidence that the correct sta- 

tistical technique would be to discard the cases with un- 

knowns in the variables being utilized in the analysis and 

not utilize the cases in the analysis.? 

The district court also concluded that no models of- 

fered by the Petitioner were sufficiently predictive as to 

be probative. (J.A. 149). As noted previously, regres- 

sions must include relevant variables to be probative. See 

3This is precisely the reason no independent model or re- 
gression analysis was presented by the Respondent. The data 
base was simply too flawed and eliminating cases with un- 
knowns reduced the sample size to the extent that a valid 
analysis was futile. 

   



  

  
  

  

20 

Bazemore v. Friday, supra. No model was used which 

accounted for several significant factors because the in- 

formation was not in the data base, i.e., credibility of wit- 

nesses, likelihood of a jury verdict, strength of the evi- 

dence, etc.* Many of the small-scale regressions simply 

include a given list of variables with no explanation given 

for their inclusion. Even the large-scale 230-variable re- 

gression has deficiencies. ‘‘It assumes that all of the in- 

formation available to the data-gathers was available to 

each decision-maker in the system at the time that deci- 

sions were made.”’ McCleskey v. Zant, supra at 361. (J.A. 

146). This is simply an unrealistic view of the criminal 

justice system which fails to consider simple issues such 

as the admissibility of evidence. Further the adjusted 

r-squared, which measures what portion of the variance 

in the dependent variable is accounted for by the inde- 

pendent variables in the model, even in the 230-variahle 

model, is only approximately .5. (J.A. 147). Petitioner 

also fails to show the coefficients of all variables in the 

regressions. 

Major problems are also presented due to multi- 

collinearity in the data. See Fisher, supra. (J.A. 105-111). 

Multicollinearity will distort the regression coefficients 

in an analysis. (J.A. 106). It was virtually admitted that 

there is a high correlation between the race of the victim 

variable and many other variables in the study. According 

to the testimony of Respondent’s experts, this was not 

accounted for by any analysis of Baldus or Woodworth. 

Various experiments conducted by Dr. Katz confirmed the 

4Although the second study purports to include strength 
of the evidence variables, there are such a high number of un- 
knowns that it cannot be considered to be effectively included 
in any analysis. 

   



21 

correlation between aggravating factors and white victim 

cases and mitigating factors with black victim cases. See 

F.H.T. 1472, et seq.; Respondent’s Exhibits 49-52. The 

district court specifically found neither Woodworth or 

Baldus had sufficiently accounted for multicollinearity in 

any analysis. 

Petitioner has asserted that there is an average twenty 

point racial disparity in death sentencing rates which he 

asserts should constitute a violation of the Eighth or Four- 

teenth Amendments. As noted previously, the statistical 

analyses themselves have not been found to be valid by 

any court making such a determination; thus, this analy- 

sis is questionable at best. Furthermore, focusing on the 

so-called ‘‘twenty percentage point’’ effect misconstrues 

the nature of the study presented. The twenty percentage 

point ‘“‘disparity’’ occurred in the so called ‘‘mid-range’’ 

of cases. This analysis attempted to exclude the most ag- 

gravated cases from its consideration as well as the most 

mitigated cases. The analysis did not consider whether the 

cases were actually eligible for a death sentence under state 

law, but was a consideration of all cases in the study which 

have been indicted either for murder or voluntary man- 

slaughter. 

A primary problem shown with the utilization of this 

““mid-range’’ analysis is the fact that Petitioner failed to 

prove that he was comparing similar cases in this analysis. 

By virtue of the previously noted substantial variables 

which were not included in the analysis, it can hardly be 

determined that the cases were similar. 

Further, this range of cases referred to by the Pe- 

titioner was constructed based on the index method uti- 

lized extensively by Professors Baldus and Woodworth.  



  

22 

Dr. Katz testified for the Respondent concerning this in- 

dex method and noted that an index is utilized to attempt 

to rank different cases in an attempt to conclude that cer- 

tain cases had either more or less of a particular attribute. 

(J.A. 87). The numbers utilized in the comparisons men- 

tioned above were derived from these indices and the num- 

bers would ‘‘purport to represent the degree for a level of 

aggravation and mitigation in each case for the purpose 

of ranking these cases according to those numbers.”’ Id. 

Dr. Katz noted that Professor Baldus had utilized re- 

gression analysis to develop the indices and had used a 

predicted outcome to form the index for aggravation and 

mitigation. Through a demonstration conducted by Dr. 

Katz utilizing four sample regressions, it was shown that 

the index method could be shaped to give different rank- 

ings from the same cases depending on what variables 

might be included in a particular regression. Through the 

demonstration, Dr. Katz showed that by including dif- 

ferent variables in the model, the actual values for the 

index would change. ‘‘[T]he purpose of this was to show 

that at any stage, what is happening with the regression 

in terms of the independent variables it has available to 

it, is that it is trying to weigh the variables or assign co- 

efficients to the variables so that the predicted outcomes 

for the life sentence cases will have zero values and the 

predicted outcomes for the death sentence cases will have 

one value, regardless of the independent variables that 

it has to work with.”’ (J.A. 98-9). The examination of 

‘this testimony as well as the exhibits in connection there- 

with shows that the index method itself is capable of mis- 

use and abuse and, depending on the particular regression 

equation utilized, the index values can be different. No 

Ae 

-
 

   



Ae 

23 

adequate explanation was provided for the particular var- 

iables included in the regression analysis so as to justify 

utilizing the index values. Thus, it was simply not shown 

that the cases being compared to develop this ‘‘mid-range’’ 

were actually similar. See McCleskey v. Zant, supra at 

375-6. (J.A. 175). 

Additionally, the .06 figure referred to by the Petition- 

er does not represent a true disparity. The .06 so-called 

‘“‘disparity’’ does not reflect any particular comparison 

of subgroups of cases. Further the .06 figure is a weight 

which is subject to change when variables are added to 

or subtracted from the model. (J.A. 233). 

Regardless of the standard applied or the propriety 

of utilizing statistics in the instant case, the above shows 

that the data base is substantially flawed so as to be in- 

adequate for any statistical analysis. Any results of any 

such analysis are thus fatally flawed and prove nothing 

about the Georgia criminal justice system. 

ITI. THE ARBITRARINESS AND CAPRICIOUS- 

NESS CONCERNS OF FURMAN V. GEORGIA, 

408 U.S. 238 (1972), ARE REMOVED WHEN A 

STATE PROPERLY FOLLOWS A CONSTITI- 

TIONAL SENTENCING PROCEDURE. 

Throughout the history of Eighth Abendment juris- 

prudence this Court has recognized, ‘‘[d]ifficulty would 

attend the effort to define with exactness the extent of the 

constitutional provision which provides that cruel and un- 

usual punishments shall not be inflicted . . ..”” Wilkerson 

v. Utah, 99 U.S. 130, 135-6 (1878). Furthermore, ‘‘[t]lhe 

cruelty against which the Constitution protects a con- 

viected man is cruelty inherent in the method of punish- 

  

 



24 

ment, not the necessary suffering involved in any method 

employed to extinguish life humanely.’’ Louisiana ex rel. 

Francis v. Resweber, 329 U.S. 459, 464, rhng. denied, 330 

U.S. 853 (1947). Members of the Court have not agreed 

as to the extent of the applicability of the Eighth Amend- 

ment. In Trop v. Dulles, 356 U.S. 86 (1958), the Court de- 

termined that the question was whether the penalty under 

examination in that case subjected the individual to a fate 

‘‘forbidden by the principle of civilized treatment guaran- 

teed by the Kighth Amendment.” Id. at 99. The Court 

also went on to note that the Eighth Amendment was not 

a static concept but that the amendment ‘“must draw its 

meaning from evolving standards of decency that mark 

the progress of a maturing society.’’ Id. at 101. 

The Highth Amendment embodies ‘‘broad and idealis- 

tic concepts of dignity, civilized standards, humanity and 

decency . . . .”” Estelle v. Gamble, 429 U.S. 97 (1976). In 

Ingraham v. Wright, 430 U.S. 651 (1977), the Court 

acknowledged that the HKighth Amendment prohibition 

against cruel and unusual punishment circumscribed the 

criminal process in three ways: (1) it limits the particular 

kind of punishment that can be imposed on those con- 

victed; (2) the amendment proscribes punishment that 

would be grossly disproportionate to the severity of the 

crime; (3) the provision imposes substantive limits on 

what can be made criminal and punished as such. 

Not until Furman v. Georgia, 408 U.S. 238 (1972), was 

the Court squarely confronted with a claim that the death 

penalty itself violated the Eighth Amendment. The hold- 

ing of the Court in that case was simply that the carrying 

out of the death penalty in the cases before the Court con- 

stituted cruel and unusual punishment. Id. at 239. 

   



29 

In Gregg v. Georgia, 428 U.S. 153 (1976), this Court 

specifically examined the Georgia death penalty scheme. 

In so doing, the Court examined the history of the Kighth 

Amendment and the opinion in Furman v. Georgia. The 

Court noted that the Eighth Amendment was to be inter- 

preted in a flexible and dynamic manner and that the 

Eighth Amendment was not a static concept. The Court 

went on to note, however, that the HKighth Amendment 

‘““must be applied with an awareness of the limited role 

played by courts.” Id. at 174. In upholding the Georgia 

statute, the Court acknowledged that Furman established 

that the death sentence could not be imposed by sentencing 

proceedings ‘‘that created a substantial risk that it would 

be inflicted in an arbitrary and capricious manner.”’ Id. at 

188. The Court compared the death sentences in Furman 

as being cruel and unusual in the same way as being struck 

by lightning would be cruel and unusual. The Court fur- 

ther noted that Furman mandated that where discretion 

was afforded to a sentencing body, that diseretion had to 

be suitably directed and limited so as to minimize the risk 

of wholly arbitrary and capricious action. Finally, the 

Court acknowledged that in each stage of the death sen- 

tencing process an actor could make a decision which would 

remove the defendant from consideration for the death 

penalty. ‘‘Nothing in any of our cases suggests that the 

decision to afford an individual defendant mercy violates 

the Constitution. Furman held only that in order to mini- 

mize the risk that the death penalty would be imposed on 

a capriciously selected group of offenders, the decision 

to impose it had to be guided by standards so that the 

sentence authorized would focus on the particularized eir- 

cumstances of the crime and defendant.”’ Gregg, supra  



  

26 

at 199. The Court further emphasized that ¢‘[t]he isolated 

decision of a jury to afford mercy does not render uncon- 

stitutional a death sentence imposed upon defendants who 

were sentenced under a system that does not create a sub- 

stantial risk of arbitrariness or caprice . ... The propor- 

tionality review substantially eliminates the possibility 

that a person will be sentenced to die by the action of an 

aberrant jury.” Id. at 203. The Court finally found that 

a jury could no longer wantonly and freakishly impose a 

death sentence as it was always circumscribed by the 

legislative guidelines. 

The same time as the Court decided Gregg v. Georgia, 

supra, it also decided Proffitt v. Florida, 428 U.S. 242 

(1976). The Court again noted that the ‘‘requirements 

of Furman are satisfied when the sentencing authority’s 

discretion is guided and channelled by requiring the ex- 

amination of specific factors that argue in favor of or 

against the imposition of the death penalty, thus eliminat- 

ing total arbitrariness and capriciousness in its imposi- 

tion.”’ Id. at 258. 

Subsequently, the Court actually criticized states for 

restricting the discretion of the juries, thus, outlawing 

statutes providing for mandatory death sentences upon 

conviction of a capital offense. See Woodson v. North 

Carolina, 428 U.S. 280 (1976). The Court has also pro- 

hibited death penalty procedures which restrict the con- 

sideration of mitigating circumstances, consistently em- 

phasizing that there must be an individualized considera- 

tion of both the offense and the offender before a death 

sentence could be imposed. Thus, in Lockett v. Oho, 438 

U.S. 587 (1978), the plurality noted that the joint opinion 

in Gregg, Proffitt and other cases concluded that in order 

 



27 

to comply with Furman the ‘‘sentencing procedure should 

not create a substantial risk that the death penalty was 

inflicted in an arbitrary manner, only that the discretion 

be directed and limited so that the sentence was imposed 

in a more consistent and rational manner. . . .”’ Lockett, 

supra at 597. 

This Court has considered death penalty cases in an 

Eighth Amendment context, but from a different perspec- 

tive than the arbitrary and capricious infliction of a pun- 

ishment as challenged in Furman. In Godfrey v. Georgia, 

446 U.S. 420 (1980), the Court was concerned with a par- 

ticular provision of Georgia law and the question of 

whether the Georgia Supreme Court had followed the 

statute that was designed to avoid the arbitrariness and 

capriciousness prohibited in Furman. This Court essen- 

tially concluded that the state courts had not followed 

their own guidelines. This Court concluded that the death 

sentence should appear to be and must be based on reason 

rather than caprice and emotion. As the Georgia courts 

had not followed the appropriate statutory procedures in 

narrowing discretion in that case, the Court concluded 

that the sentence was not permissible under the Eighth 

Amendment. The Court did not deviate from its prior 

holding in Gregg, supra, that by following a properly 

tailored statute the concerns of Furman were met. 

The Court considered the death penalty in an Eighth 

Amendment context in Enmund v. Florida, 4568 U.S. 782 

(1982). The Court, however, did not consider the ‘‘arbi- 

trary and capricious’’ aspect but focused on the question 

of the disproportionality of the death penalty for En- 

mund’s own conduct in that case. Thus, the Court essen-  



  

28 

tially concluded that the death penalty was disproportion- 

ate under the facts of that case. 

In California v. Ramos, 463 U.S. 992, 999 (1983), the 

Court noted that ‘‘[i]n ensuring that the death penalty is 

not meted out arbitrarily or capriciously, the Court’s prin- 

cipal concern has been more with the procedure by which 

the State imposes the death sentence than with substantive 

factors the State lays before the jury as a basis for im- 

posing death. . . .”” Thus, the Court again focused on the 

state procedure in question and noted that excessively 

vague sentencing standards could lead to the arbitrariness 

and capriciousness that were condemned in Furman. 

Further, in particular reference to the study in the 

instant case, Justice Powell observed: 

No one has suggested that the study focused on this 
case. A ‘‘particularized’’ showing would require— 
as I understand it—that there was intentional race 
discrimination in indicting, trying and convicting [the 
defendant], and presumably in the state appellate and 
state collateral review that several times followed the 
trial. . . . Surely, no contention can be made that the 
entire Georgia judicial system, at all levels, operates 
to discriminate in all cases. Arguments to this effect 
may have been directed to the type of statutes ad- 
dressed in Furman. As our subsequent cases make 
clear, such arguments cannot be taken seriously un- 
der statutes approved in Gregg. 

Stephens v. Kemp, — U.S. —, 104 S.Ct. 562 n.2 (1983) 

(Powell, J., dissenting from the granting of a stay of exe- 
$¢ cution). Justice Powell went on to note ‘‘claims based 

merely on general statistics are likely to have little or no 

merit under statutes such as that in Georgia.’’ Id. 

 



29 

Respondent submits that reviewing all of the Court’s 

Eighth Amendment jurisprudence, particularly in the death 

penalty context reflects that in order to establish a claim 

of arbitrariness and capriciousness sufficient to violate 

the cruel and unusual punishment provision of the Eighth 

Amendment, it must be established that the state failed to 

properly follow a sentencing procedure which was suffi- 

cient to narrow the discretion of the decision-makers. As 

long as the state follows such a procedure, the arbitrari- 

ness and capriciousness which were the concern in Fur- 

man v. Georgia, supra, have been minimized sufficiently to 

preclude a constitutional violation, particularly under the 

Fighth Amendment. An Kighth Amendment violation 

would result in the ‘‘arbitrary and capricious’’ context, 

only if the statutory procedure either was insufficient it- 

-self or the appropriate procedures were not followed. Other 

death penalty cases under the Eighth Amendment deal 

with different aspects of the cruel and unusual punish- 

ment provision, such as disproportionality or excessive 

sentences in a given case. That is simply not the focus 

of the inquiry here. Under the circumstances of the in- 

stant case, the Petitioner has not even asserted that Geor- 

gia’s procedures themselves are unconstitutional, nor has 

the Petitioner asserted that those procedures which were 

approved in Gregg v. Georgia, supra, were not followed in 

the instant case. Thus, there can be no serious contention 

that there is an Kighth Amendment violation under the 

circumstances of this case. This is particularly true in 

light of the testimony of Petitioner’s own expert that the 

Georgia charging and sentencing system sorts cases on 

rational grounds. (F.H.T. 1277; J.A. 154).  



  

  

  

30 

Insofar as the Petitioner would attempt to assert some 

type of racial discrimination under the Eighth Amendment 

provisions, there should be a requirement of a focus on 

intent in order to make this sentence an ‘‘aberrant’’ sen- 

tence so as to classify it as arbitrary and capricious. A 

simple finding of disparate impact is insufficient to make 

a finding of arbitrariness and capriciousness such as was 

the concern in Furman, supra, particularly when a prop- 

erly drawn statute has been utilized and properly followed. 

Only a showing of purposeful or intentional discrimina- 

tion can be sufficient to find a constitutional violation un- 

der these circumstances. 

No Eighth Amendment violation can be shown in the 

instant case as Petitioner’s own witness testified that the 

system acted in a rational manner. As shown by the 

analyses conducted by Professor Baldus and Dr. Wood- 

worth, the more aggravated cases were moved through the 

charging and sentencing system and the most aggravated 

cases generally received a death sentence. The more miti- 

gated cases on the other hand dropped out at various 

stages in the system receiving lesser punishments. Thus, 

this system does function in a rational fashion. Further- 

more, it has not been shown that the death sentence in the 

instant case was arbitrary or capricious in any fashion. 

The jury found beyond a reasonable doubt that there were 

two statutory aggravating circumstances present. The 

evidence also shows that the victim was shot twice, includ- 

ing once in the head at fairly close range. The evidence 

tended to indicate that Petitioner hid and waited for the 

police officer and shot him as the officer walked by. This 

was an armed robbery by four individuals of a furniture 

   



31 

store in which several people were, in effect, held hostage 

while the robbers completed their enterprise. It was thor- 

oughly planned and thought out prior to the robbery occur- 

ring. Furthermore, the Petitioner had prior convictions 

for robbery before being brought to this trial. One of 

Petitioner’s co-perpetrators testified against him at trial 

and a statement of the Petitioner was introduced in which 

he detailed the crime and even boasted about it. (J.A. 113- 

115). Thus, under the factors in this case it is clear that 

Petitioner’s sentence is not arbitrary or capricious and 

there is clearly no Eighth Amendment violation. 

IV. PROOF OF DISCRIMINATORY INTENT IS 
REQUIRED TO ESTABLISH AN EQUAL 
PROTECTION VIOLATION. 

It is well recognized that “[a] statute otherwise neu- 

tral on its face, must not be applied so as to invidiously 

discriminate on the basis of race.” Washington v. Davis, 

426 U.S. 229, 241 (1976), citing Yick Wo v. Hopkins, 118 

U.S. 356, 369 (1886). This Court has consistently recog- 

nized, however, that in order to establish a claim of dis- 

crimination under the Equal Protection Clause, there must 

be proof that the challenged action was the product of dis- 

criminatory intent. See Washington v. Davis, supra. 

In 1962, the Court examined what was essentially an 

allegation of selective prosecution and recognized, “the 

conscious exercise of some selectivity in enforcement is not 

in itself a federal constitutional violation.” Oyler v. Boles, 

368 U.S. 448, 456 (1962). In cases finding an equal pro- 

tection violation, it is consistently recognized that the bur- 

den is on the petitioner to prove purposeful discrimination 

under the facts of the case. See Whitus v. Georgia, 385  



  

  

32 

U.S. 545 (1967). The Court specifically has recognized 

that the standard applicable to Title VII cases does not 

apply to equal protection challenges. “We have never held 

that the constitutional standard for adjudicating claims of 

invidious racial discrimination is identical to standards 

applicable under Title VIL . ..” Washington v. Davis, 

supra, 426 U.S. at 239. The Court went on in that case to 

note that the eritical purpose of the equal protection clause 

was the “prevention of official conduct discriminating on 

the basis of race.” Id. The Court emphasized that the 

cases had not embraced the proposition that an official 

action would be held to be unconstitutional solely because 

it had a racially disproportionate impact without regard 

to whether the facts showed a racially diseriminatory pur- 

pose. It was acknowledged that disproportionate impact 

might not be irrelevant and that an invidious purpose 

could be inferred from the totality of the relevant facts, 

including impact, but ‘‘[d]isproportionate impact . . . 

is not the sole touchtone of an invidious racial discrimina- 

tion forbidden by the Constitution. Standing alone it does 

not trigger the rule [cit.] that racial classes are to be sub- 

jected to the strictest serutiny. . ..” Id. at 242. 

Again in Castaneda v. Partida, 430 U.S. 482, 493 

(1977), the Court held that “an official act is not uncon- 

stitutional solely because it has a racially disproportionate 

impact.” (emphasis in original). Further, “[p]roof of 

racially discriminatory intent or purpose is required to 

show a violation of the Equal Protection Clause.” Village 

of Arlington Heights v. Metropolitan Housing Develop- 

ment Corp., 429 U.S. 252, 265 (1977). In Washington v. 

Davis the Court held that the petitioner was not required 

to prove that the decision rests solely on racially discrim- 

   



33 

matory purposes, but that the issue did demand a ‘‘sensi- 

tive inquiry into such circumstantial and direct evidence 

of intent as may be available.” Id; Village of Arlington 

Hewghts, supra. “Absent a pattern as stark as that in 

Gomullion® or Yick Wo, impact alone is not determinative, 

(footnote omitted) and the court must look to other evi- 

dence.” Id. at 266. “In many cases to recognize the lim- 

ited probative value of disproportionate impact is merely 

to acknowledge the ‘heterogeneity’ of the Nation’s popu- 

lation.” Id. at 266 n.15. 

The Court also acknowledged that the Fourteenth 

Amendment guarantees equal laws, not necessarily equal 

results. Whereas impact may be an important starting 

point, it is purposeful discrimination that offends the Con- 

stitution. Personnel Administrator of Massachusetts wv. 

Feeney, 442 U.S. 256, 273-4 (1979). A discriminatory pur- 

pose “implies more than intent as volition or intent as 

awareness of the consequences. . . . It implies that the 

decision makers selected or reaffirmed a particular course 

of action at least in part because of not merely in spite 

of its adverse effects on the identified group.” Id. at 279; 

see also Wayte v. United States, — U.S. —, 105 S.Ct. 1524, 

1532 (1985). The Court reemphasized its position in Rog- 

ers v. Lodge, 458 U.S. 613 (1982), in which the Court rec- 

ognized “the invidious quality of a law claimed to be ra- 

cially discriminatory must ultimately be traced to a racially 

discriminatory purpose,” and acknowledged that a showing 

of discriminatory intent was required in all types of equal 

protection cases which asserted racial diserimination. 

SGomillion v. Lightfoot, 364 U.S. 339 (1960).  



  
  

  
34 

Thus, it is clear from all of the above that a diserim- 

inatory purpose, requiring more than simply an awareness 

of the consequences, must be established in order to make 

out a prima facie showing of discrimination under the 

Equal Protection Clause, regardless of the type of equal 

protection claim that is raised. The burden is on the in- 

dividual alleging this discriminatory selection to prove the 

existence of the purposeful discrimination and this includes 

the initial burden of establishing a prima facie case as 

well as the ultimate burden of proof. 

In relation to the question of an Kqual Protection vi- 

olation, Petitioner has also failed to show intentional or 

purposeful discrimination. The Petitioner presented evi- 

dence to the district court by way of the deposition of 

the district attorney of Fulton County, Lewis Slaton. 

Throughout his deposition, Mr. Slaton testified that the 

important facts utilized by his office in determining wheth- 

er to proceed with a case either to indictment, to a jury 

trial or to a sentencing trial, would be the strength of the 

evidence and the likelihood of a jury verdict as well as 

other facts. Mr. Slaton observed that in a given case there 

could exist the possibility of suppression of evidence ob- 

tained pursuant to an alleged illegal search warrant which 

would also affect the prosecutor’s decision. (Slaton Dep. at 

18). In determining whether to plea bargain to a lesser of- 

fense, Mr. Slaton testified that his office would consider 

how strong the case was, how the witnesses would hold up 

under cross-examination, what scientific evidence was avail- 

able, the reasons for the crime. the mental condition of the 

parties, prior record of the defendant and the likelihood of 

what the jury might do. Id. at 30. As to proceeding to a 

   



35 

death penalty trial, Mr. Slaton testified that first of all the 

question was whether the case fell within the ambit of the 

statute and then he examined the atrociousness of the 

crime, the strength of the evidence and the possibility of 

what the jury might do as well as other factors. Id. at 31. 

He also specifically noted that his office did not seek the 

death penalty very often, for one reason because the juries 

in Fulton County were not disposed to impose the death 

penalty. Id. at 32. He also specifically testified he did 

not recall ever seeking a death penalty in a case simply 

because the community felt it should be done and did not 

recall any case in which race was a factor in determining 

whether to seek a death penalty. Id. at 78. 

This is a case in which the Petitioner has in effect by 

statistics alone sought to prove intentional discrimination. 

Although Petitioner has alleged anecdotal evidence was 

submitted, in fact, little, if any, was presented to the dis- 

trict court outside the deposition of Lewis Slaton and one 

witness who gave the composition of Petitioner’s trial 

jury. As noted previously, Respondent submits that sta- 

tistics are not appropriate in this type of analysis and the 

Petitioner’s statistics in this case are simply invalid; how- 

ever, regardless of that fact any disparity noted is simply 

not of the nature of such a gross disparity as to compel an 

inference of diserimination, unlike earlier cases before the 

court. See e.g., Gomillion v. Lightfoot, 364 U.S. 339 (1960). 

Absent the ‘‘inexorable zero’’ or a gross disparity similar 

to that, this type of evidence under the unique circumstanc- 

es of a death penalty situation should not be sufficient to 

find an inference of discrimination, particularly when both 

lower courts have found that no intentional diserimination 

was proven. Thus, Respondent submits that regardless of  



  

36 

the standard utilized, Petitioner has failed to meet this 

burden of proof. 

Regardless of the standard used for determining when 

a prima facie case has been established, it is clear where 

the ultimate burden of proof lies. Under the circumstances 

of the instant case, it is clear that the ultimate burden of 

proof rested with the Petitioner and he simply failed to 

meet his burden of proof either to establish a prima facie 

case of discriminatory purpose or to carry the ultimate 

burden of proof by a preponderance of the evidence. 

  
0). 
J 

   



37 

CONCLUSION 

For all of the above and foregoing reasons, the con- 

victions and sentences of the Petitioner should be affirmed 

and this Court should affirm the decision of the Kleventh 

Circuit Court of Appeals. 

Respectfully submitted, 

Mary BEE WESTMORELAND 
Assistant Attorney General 
Counsel of Record for Respondent 

MicuAEL J. Bowers 

Attorney General 

Marion O. GorponN 
First Assistant Attorney General 

Wirriam B. Hiri, Jr. 
Senior Assistant Attorney General 

Mary Berea WESTMORELAND 
132 State Judicial Building 
40 Capitol Square, S. W. 
Atlanta, Georgia 30334 
(404) 656-3349 

 





  

No. 84-6811 

  
£5 

A 

In The | 
Supreme Court of the United States 

October Term, 1985 

0H 
hd 

WARREN MCCLESKEY, 
- Petitioner, 

  

Ve 

RALPH KEMP, Superintendent, 
Georgia Diagnostic and 
Classification Center, 

Respondent. 
Vy On 

~ 

ON WRIT OF CERTIORARI 
TO THEE UNITED STATES COURT OF APPEALS 

FOR THEE ELEVENTH CIRCUIT 

  

BRIZF FOR RESPONDENT 

  

  

£2 

~~ 

Mary Berm WESTMORELAND 

Assistant Attorney General 

Counsel of Record for Respondent 

Micmazr J. Bowers 
Attorney General 

Maziox O. Gorpox 
‘First Assistant Attorney General 

Worm B. Hoo, Je. 
Senior Assistant Attorney General 

Mary BETTE WESTMORELAND 
132 State Judicial Building 
4) Capitol Square, S. W. 
Atlanta, Georgia 30334 
(404) 656-3349 

 



  

  

    

      

      

          

         

QUESTIONS PRESENTED 

i 

Is the statistical analysis which was presented to the 

district court inadequate to prove a constitutional viola- 

tion, both as a matter of fact and as a matter of law? 

2. 

Are the arbitrariness and capriciousness concerns of 

Furman v. Georgia, 408 U.S. 238 (1972), removed when a 

state properly follows a constitutional sentencing proce- 

dure? 

3. 

In order to establish a constitutional violation based 

on allegations of discrimination, must a petitioner prove 

intentional and purposeful discrimination? 

   

  
 



  

QUESTIONS PRESENTED 

STATEMENT OF THE CASE 

SUMMARY OF THE ARGUMENT 

i 

TABLE OF CONTENTS 

  

  

  

ARGUMENT 

L STATISTICAL ANALYSES ARE INADE- 
QUATE AS A MATTER OF FACT AND LAW 
TO PROVE DISCRIMINATION UNDER THE 
FACTS OF THE INSTANT CASE. 

I. THE STATISTICAL ANALYSES IN THE IN- 
STANT CASE ARE INSUFFICIENT TO 
PROVE RACIAL DISCRIMINATION.   

[II. THE ARBITRARINESS AND CAPRICIOUS- 

IV. 

CONCLUSION 

  

NESS CONCERNS OF FURMAN V. GEOR- 
GI4, 408 U.S. 238 (1972), ARE REMOVED 
WHEN A STATE PROPERLY FOLLOWS A 
pay SENTENCING PROCE- 
D ! : 
  

PROOF OF DISCRIMINATORY INTENT IS 
REQUIRED TO ESTABLISH AN EQUAL 
PROTECTION VIOLATION.   

  

oo
 

16 

al 

37 

  

 



  

  

TABLE OF AUTHORITIES 
Page(s) 

Caszs Crrep: 

Bazemore v. Friday, — U.S. —, 106 S.Ct. 3000 
  

  

  

  

  

(1986) 10, 20 

Brittom v. Rogers, 631 F.2d 572 (5th Cir. 1980), 
cert. demsed, 461 U.S. 939 (1981) 8 

Caldwell v. Mississipps, 472 U.S. —, 105 S.Ct. 2633 
(1985) 13 

California v. Ramos, £63 U.S. 992 (1983) 23 

Castaneda v. Partida, +30 U.S. 482 (1977) ov 32 

Eastland v. Tennessee Valley Authority, 704 F.2d 
613 (11th Cir. 1983) 1] 

Eddings v. Oklahoma. 453 U.S. 104 (1982) cen 13 

Enmund v. Florida, 468 U.S. 782 (1982) 27   

Equal Employment Opportumity Commission v. 
Datapoint Corporation, 370 F.2d 1264 (5th Cir. 
  

  

  

1978) 10 

Estelle v. Gamble, 429 U.S. 97 (1976) 24 

Furman v. Georgia, 408 U.S. 238 (1972) 3. 24, 25,27. 
28.29.00 

Godfrey v. Georgia, +46 U.S. 420 (1980) 27 

Gomillion v. Lightfoot, 364 U.S. 339 (1960) . ie. Hi 

Gregg v. Georgia, 428 U.S. 153 (1976) ..—.25, 26, 27, 28, 29 

  

  

  

Ingraham v. Wright, $30 T.S. 651 (1977) 24 

International Brotherhood of Teamsters v. United 

States, 431 U.S. 324 (1977) 9 

Johnson v. Uncle Ben's Inc.. 628 F.2d +19 (5th Cir. 

1980) 11 
  

  Lockett v. Ohio, 438 U.S. 586 (1978) 13, 26, 27 

  

 



  

iv 

TABLE OF AUTHORITIES—Continued 

  

  

  

Page(s) 

Louisiana: ex rel. Francis v. Resweber, 329 U.S. 
489, ring. denied, 330 U.S. 853 (1947) 24 

Mazwell v. Bishop, 398 F.2d 138 (8th Cir. 1968), 
remanded om other grounds, 398 U.S. 262 (1970) —_ 12 

Mayor of Philadelphia v. Educational Equality 
League, 415 U.S. 605 (1974) 8 

McCleskey v. Kemp, T33 F.2d 877 (11th Cir. 1985) 
(em bane) 4 

McCleskey v. Zant, 380 F.Supp. 338 (N.D.Ga. 1984) __1, 2, 
3, 5.17, 18,20. 33 

McCorguodale v. Baikcom, 525 F.Supp. 408 (N.D. 
Ga. 1981), affirmed, 721 F.2d 1493 (11th Cir. 1983)... 13 

  

  

  

  

  

  

UcGautha v. Califorma, 402 U.S. 183 (1971) 2 

Oyler v. Boles, 368 U.S. 448 (1962) 3 

Personnel ddmsmasirator of Massachusetts v. 
Feeney, +42 U.S. 256 ( 1979) 33 

Proffitt v. Florida, 428 U.S. 242 (1976) 26 

Pullman-Standard v. Swint, +56 U.S. 273 (1982) wee 17 

Rogers v. Lodge, 458 U.S. 613 (1982) 33 

Shaw v. Martim, 733 F.2d 304 (4th Cir. 1984) 13 

Smith v. Balkcom, 660 F.2d 584 (5th Cir. 1981), on 
rehearing, 671 F.2d 858 (5th Cir. Unit B, 1982) ..... 13 

Spinkellink v. Wainwright, 578 F.2d 582 (5th Cir. 
1978) 3 

Stephens v. Kemp, — U.S. —, 104 S.Ct. 562 (1983) ... 28 

Trop v. Dulles, 356 U.S. 36 (1958) 14, 24 

Turner v. Murray, — U.S. —, 106 S.Ct. 1683 (1986) ... 4 

United States v. Iromcorkers Local 86, 443 F.2d 
344 (9th Cir. 1971) 10 
  

  

 



  

  

TABLE OF AUTHORITIES—Continued 

  

Page(s) - 

Umsted States v. Unsted States Gypsum Co., 333 
U.S. 364 (1948) a 17 

Valentino v. United States Postal Service, 674 
F.2d 56 (D.C.Cir. 1982) 11   

Village of Arlington Heights v. Metropolitan Hous- 
  

  

  

  

  

  

img Development Corp., 429 T.S. 252 (1977) 32,33 

Wade v. Mississippi Cooperative Extension Ser- 
vice, 528 F.2d 508 (5th Cir. 1976) 10 

Washington v. Davis; 426 U.S. 229 (1976) 31,32 

Wayte v. United States, — U.S. —, 105 S.Ct. 1524 
(1985) : : 33 

Whitus v. Georgia, 385 U.S. 545 (1967) 31 
Wilkerson v. Utah, 99 U.S. 130 (1878) 23 

Wiking v. University of Houston, 654 F.2d 388 
(5th Cir. Unit A 1981) 11 
  

Witherspoon v. [llinots, 391 U.S. 510 (1968) 14 

Woodson v. North Carolina. 428 U.S. 280 (1976) —— 26 

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

OTHER AUTHORITIES: 

Baldus & Cole, 4 Comparison of the Work of Thor- 
sten Sellin and Isaac Ehrlich on the Deterrent 
Effect of Capital Punishment, 85 Yale L. J. 170 

  
(1975) 15 

~ Fisher. Multiple Regression in Legal Proceedings, 
80 Colum. L.Rev. 702 (1980) 15, 20 

  

A. Goldberger, Topics in Regression Analysis (1968) 15 

  

 



  

vi 

TABLE OF AUTHORITIES—Continued 

Page(s) 

McCabe, The Interpretation of Regressiom dnaly- 
sis Results in Sex amd Race Discrimimaiion ; 

Problems, 34 Amer. Stat. 212 (1980) 16 

Smith and Abram, Quantitative Analysis and Proof 
of Employment Discrimination, 1981 U.IIL L.Rev. 
33 (1981) 15 

G&G. Wesolowsky, Multiple Regression Analysis of 
Variance (1976) 15 

  

  

  

  

 



  

  

No. 84-6811 

  
5 
Nz 

In The 

Supreme Court of the United States 
October Term, 1986 

Fg 
a 

WARBEN MCCLESKEY, 

  

Petitioner, 

v. 

RALPH KEMP, Superintendent, 
Georgia Diagnostic and 
Classification Center, 

Respondent. 
= 

ON WRIT OF CERTIORARI 
TO THE UNITED STATES COURT OF APPEALS 

POR THE ELEVENTH CIRCUIT 

  

pos 

BRIEF FOR RESPONDENT 
ey 
a4 

STATEMENT OF THE CASE 

  

  

In addition to the statement of the case set forth by 

the Petitioner, Respondent submits the following regard- 

ing the district court and circuit court proceedings: 

Two different studies were conducted on the criminal 

justice system in Georgia by Professors Baldus and Wood- 

worth, that is, the Procedural Reform Study and the 

Charging and Sentencing Study. See McCleskey v. Zant, 

580 F.Supp. 338, 353 (N.D.Ga. 1984). The Petitioner pre- 

sented his case primarily through the testimony of Pro- 

fessor David C. Baldus and Dr. George Woodworth. Peti- 

tioner also presented testimony from Edward Gates as 

1 

  

 



  
~~

 

well as an official from the State Board of Pardons and 

Paroles. The state presented testimony from two expert 

statisticians, Dr. Joseph Katz and Dr. Roger Buford. 

The district court made the following specific factual 

findings regarding the trustworthiness of the data base: 

[(TThe court is of the opinion that the data base has 
substantial flaws and that the petitioner has failed to 
establish by a preponderance of the evidence that it 
ts essentially trustworthy. As demonstrated above, 
there are errors in coding the questionnaire for the 
case sub judice. This fact alone will invalidate several 
important premises of petitioner’s experts. Further, 
there are large numbers of aggravating and mitigat- 
ing circumstances data about which is anknown. Also, 
the researchers are without knowledge concerning the 
decision made by prosecutors to advance cases to a 
penalty trial in a significant number of instances. The 
court’s purpose here is not to reiterate the deficien- 
cies but to mention several of its concerns. It is a 
major premise of a statistical case that the data base 
numerically mirrors reality. If it does not in substan- 
tial degree mirror reality, any inferences empirically 
arrived at are untrustworthy. 

HcCleskey v. Zant, supra, 380 F.Supp. at 360 (emphasis 

in original). (J.A. 144-5). 

The district court found as fact that ‘‘none of the 

models utilized by the petitioner’s experts were sufficieni- 

ly predictive to support an inference of discrimination.” 

HcCleskey v. Zant, supra at 361. (J.A 149). 

The district court also found problems in the data due 

to the presence of multicollinearity. The district court 

noted that a significant fact in the instant case is that 

white vietim cases tend to be more aggravated. that is 
correlated with aggravating factors, while black vierim 

  

 



  

  

cases tend to be more mitigated, that is correlated with 

mitigating factors. Every expert who testified, with the 
exception of Dr. Berk, agreed that there was substantial 

multicollinearity in the data. The district court found. 

“The presence of multi-colinearity substantially dimin- 

ishes the weight to be accorded to the circumstantial statis- 

tical evidemce of racial disparity.” McCleskey v. Zant, 

supra at 364. (J.A. 153). The court then found Petitioner 

had failed to establish a prima facie case based either on 

race of victim or race of defendant. Id. 

Additionally, the district court found ‘‘that any racial 

variable is not determinant of who is going to receive the 

death penaily, and, further, the court agrees that there is 

no support for a proposition that race has amy effect in 

any single case.”” McCleskey ». Zant. sunra at 366 (empha- 

sis in original). (J.A. 157). “‘The best models which 

Baldus was able to devise which account to any significant 

degree for the major mom-racial variables, including 

strength of the evidence, produce no statistically sigmifi- 

cant evidence that race plays a part im either of those de- 

cisions [by the prosecutor and jury] in the State of 

Georqa.’”” McCleskey v. Zant, at 368 (emphasis in origi- 

nal). (J.A. 139). 

Finallv. the district court found that the analvses did 

not ‘‘compare identical cases, and the method is incapable 

of saying whether or not any factor had a role in the de- 

cision to impose the death penalty in any particular case.”’ 

McCleskey v. Zant at 372 (emphasis in original). (J.A. 

168). “To the extent that McCleskey contends that he was 

demed either due process or equal protection of the law. 

his methods fail to comiribute amything of value to his 

«ecom -ow an . . come . —.e me Che een cer seaDen wd mecmam—. e we es = = 

  

 



    

+ 

cause.”” McCleskey v. Zant at 372 (emphasis in original). 

(JA. 169). 

The court also found the Respondent presented direct 

rebuttal evidence to Baldus’ theory that contradicted any 

prima facie case of system-wide discrimination, if one had 

been established. McCleskey v. Zant at 373. 

In examining the issues, the Eleventh Cireunit Court of 

Appeals assumed. but did not decide, that the research 

was valid because there was no need to reach the question 

of the validity of the research due to the court’s legal 

analysis. The court specifically complimented the district 

court on its thorough anayisis of the studies and the evi- 

dence. The Eleventh Cirenit observed that the first study, 

the Procedural Reform Study, revealed no race of de- 

fendant effects whatsoever and revealed unclear race of 

victim effects. McCleskey v. Kemp, 753 F.2d 877, 387 (11th 

Cir. 1985) (em bamc). As to the Charging and Sentencing 

Study, the court concluded, ‘There was no suggestion that 

a uniform institutional bias existed that adversely affected 

defendants in white victim cases in all circumstances, or a 

black defendant in all cases.’”” [d. Finally, the court con- 

cluded the following in relation to the data specifically re- 

lating to the county in which the Petitioner was convicted, 

that is, Fulton County. Georgia: 

Because there were only ten cases involving police 
officer victims in Fulton County, statistical analysis 
could not be utilized effectively. Baldus conceded that 
it was difficult to draw any inference concerning the 
overall race effect in these cases because there had 
been only one death sentence. He concluded that based 
on the data there was only a possibility that a racial 
factor existed in MecCleskey’s case. 

Id. at S87 (emphasis in original). 

  

 



  

  

Any further factual or procedural matters will be 

discussed: as necessary in the subsequent portion of the 

brief. ; 

  

SUMMARY OF THE ARGUMENT 

Although the petition in the instant case lists five 

questions presented, the main focus of this case is simply 

one of whether there has been racial discrimination in the 

application of the death penalty in Georgia and, in par- 

ticular, whether there was racial discrimination in the im- 

position of the death penalty upon the Petitioner. An- 

other way of looking at this issue is whether the Petitioner 

was selectively prosecuted and sentenced to death based 

on his race and that of the victim or whether Petitioner’s 

seutence is disproportionate. Regardless of the standard 

to be applied, an appropriate consideration is the intent 

of the decision-makers in question. A review of the cases 

of this Court dealing with death penalty statutes shows 

that the gemeral arbitrariness and capriciousness which 

concerned the Court in 1972 is no longer a consideration 

if a state follows a properiv drawn statute and if the 

~ jury’s discretion is properly channeled. Thus. the focus 

in an Eighth Amendment analysis becomes a question of 

whether the sentence in a given case is ‘‘arbitrary’’ in the 

sense of being an aberration. The evidence in the instant 

case shows that the Georgia statutory scheme is function- 

ing as it was intended to function and that those cases 

which are more severe are receiving stronger penalties 

while the less severe cases are receiving lesser penalties. 

There is no evidence to show that the Petitioner’s sentence 

  

 



    

in the instant case was arbitrary or capricious and no evi- 

dence to show that either the prosecutor or the jury based 

their decision on race. 

In relation to an equal protection context, it has al- 

ways been recognized that intentional and purposerul dis- 

crimination must be established for a constitutional viola- 

tion to be proven. Although intent may be inferred from 

circumstantial evidence, the circumstantial evidence must 

be sufficient to establish a prima facie case of diserimina- 

tion before intent will he inferred. Even if a prima facie 

case is shown, the Petitioner would still have the ultimate 
burden of proof after considering any rebuttal evidence. 

In evaluating faets and cirenmstances of a given case, 

the court must consider the totality of the circumstances 

in determining whether the evidence is sufficient to find 

intentional and purposeful diseriminaton. Although sta- 

tistics are a useful tool in many contexts, in. the situation 

presented involving the application of the death penalty, 

there are simply too many unique factors relevant to each 

individual case to allow statistics to be an effective tool in 

proving intentional discrimination. Furthermore, the Peti- 

tioner’s statistics in the instant case were found to be inval- 

id by the district court, which was the only court making 

any factual findings in relation to those statistics. Thus. 

the clearly erroneous standard should apply to those factu- 

al findings. Furthermore, when a plausible explanation is 

offered, as it was in the instant case, that is, that white 

victim cases are simply more aggravated and less miti- 

gaterl than black vietim cases and that various factors 

tainterl the statistics utilized, statistics alone or a disparity 

alone is clearly insufficient to justify an inference of dis- 

crimination. Furthermore, the statistics in question fail 

  

   



  

  

to take into consideration significant factors. Thus, the 

statistics in the instant case do not give rise to an infer- 

ence of discrimination. 

When reviewing all of the evidence in the instant case, 

it is clear that the findings of fact made by the district 

court are not clearly erroneous and that the statistical 

study in question should not be concluded to be valid so 

as to raise any inference of discrimination. The Peti- 

tioner failed to make a prima facie showing of diserimina- 

tion and did not carry the ultimate burden of proof on the 

factual question of intent. Furthermore, Petitioner simply 

failed to show that his death sentence was arbitrary or 

capricious or was the result of racial diserimination either 

on the part of the prosecutor or on the part of the jury. 

  [w]
 

ARGUMENT 

I STATISTICAL ANALYSES ARE INADE- 
QUATE AS A MATTER OF FACT AND LAW 
TO PROVE DISCRIMINATION UNDER THE 
FACTS OF THE INSTANT CASE. 

Respondent submits that the type of statistical an- 

alyses utilized in the instant case are not appropriate in a 

death penalty case when trying to evaluate the motivation 

behind a prosecutor’s use of his discretion and the jury’s 

subsequent exercise of discretion in determining whether 

Wim treem cL ne ar te SEED Om EGOS Sn = Sma  ——— 

  

  

 



    

or not a death sentence should be imposed.! Each death 
penalty case is unique and even though statistics might he 

useful in jury composition cases or Title VII employment 

discrimination cases where there are a limited number of 

factors that are permissibly considered, in the instant case 

where the prosecutor has discretion to pursue a case 

through the criminal justice system and can consider any 

number of subjective factors and where a jury has com- 

plete discretion with regard to extending mercy, the sub- 

jective factors cannot be accounted for in a statistical 

analysis such as that utilized by the Petitioner in the in- 

stant case. Thus, Respondent would submit that this 

Court should completely reject the use of this type of sta- 

tistical analysis as inappropriate in this case. 

Even in the cases that have utilized statistical analysis 

in a context other than that present in the instant case, the 

courts have acknowledged various concerns with these 

analyses. This Court has recognized in another context, 

‘“Statistical analyses have served and will continme to 

serve an important role as one indirect indicator of racial 

discrimination in access to service on governmental bod- 
ies, particularly where, as in the case of jury service, the 

duty to serve falls equally on all citizens.’ Mayor of 

Philadelphia v. Educational Equality League. 415 U.S. 

Respondent submits that a claim of discrimination based 
on race of victim is not cognizable under the circumstances of 
the instant case. At least one circuit court has specifically re- 
jected statistical evidence based on the race of the victim, find- 
ing that the defendant lacked standing. Britton v. Rogers, 631 
F.2d 572, 577 n.3 (5th Cir. 1980), cert. denied, 451 U.S. 939 
(1981). Even those justices raising a question of possible racial 
discrimination in Furman v. Ceorgia, 408 U.S. 238 (1972), seemed 
to focus on race of the defendant and not race of the victim. 
Thus, Respondent submits that the instant claim is not cognizable 
due to the lack of standing. 

  

 



  

  

9 

605, 620 (1974) (emphasis added). In the instant case, 
however, there is no such uniform ‘‘duty’’ as in the jury 
composition cases, as all citizens are certainly not equally 

~ eligible for a death sentence, nor are even all perpetra- 

tors of homicides or murders equally eligible for a death 

sentence. 

A central case regarding the use of statistics by this 

Court arises in International Brotherhood of Teamsters v. 

Umied States, 431 U.S. 324 (1977). Again, this was in the 

context of a Title VII action and not in a case such as the 

instant one involving so many subjective factors. The 

Court noted prior approval of the use of statistical proof 
‘“where it reached proportions comparable to those in this 

case to establish a prima facie case of racial discrimination 

in jury selection cases.’”’ Id. at 339. The Court also noted - 

that statistics were equally competent to prove employ- 

ment discrimination, which once again is different from 

the type of discrimination sought to be proved in the in- 

stant case. The Court specifically concluded, ‘‘ We caution 

only that statistics are not irrefutable; they come in in- 

finite variety and like any other kind of evidence, they 

may be rebutted. In short their usefulness depends on all 

of the surrounding facts and circumstances.’’ Id. at 340. 

Thus, it is imperative to examine all of the facts and eir- 

cumstances to determine whether the statistics in a given 

case are even useful for conducting the particular analy- 

sis. In Teamsters, supra, the Court also had 40 specific 

instances of discriminatory action to consider in addition 

to the statistics and noted that even ‘‘fine tuning of the 

statistics could not have obscured the glaring absence of 

minority line drivers.”” Id. at 342 n.23. Thus, the Court 

did not focus exclusively on the statisties. 

  

 



    

10 

Problems have also been noted revolving around the 

particular use of statistics in any given case, many of 

which occur in the studies presented to the district court 

in the case at bar. In Bazemore v. Friday, — U.S. —, 106 

S.Ct. 3000 (1986), the Court examined regression analyses 

and concluded that ‘‘the omission of variables from a re- 

gression analysis may render the analysis less probative 

than it otherwise might be’’ while noting that this would 

not generally make the analysis inadmissible. Id. at 3009. 

The Court did go on to note that there could be some cases 

in which the regression was so incomplete as to be inad- 

missible as irrelevant. 

Circuit courts have also utilized statistics but have 

. continually urged caution in their utilization even in jury 

selection and Title VII cases. Also, the courts frequently 

had other data on which to rely in addition to the statisti- 

cal analyses. See Umited States v. [romworkers Local 36. 

443 F.2d 544 (9th Cir. 1971) : Wade v. Hississippi Coopera- 

tive Ertemsion Service, 528 F.2d 3508 (5th Cir. 1976). The 

circuit courts have also recognized that statistical avidence 

can be part of the rebuttal case itself. The Fifth Circuit 

Court of Appeals examined a Title VII case in which the 

statisties relied upon by the plaintiff actually formed the 

very hasis of the defendant’s rebuttal case. that is that 

there was a showing that the statistics were not reliable. 

Equal Employment Opportunity Commission v. Datapoint 

Corporation, 570 F.2d 1264 (5th Cir. 1978). In that case. 

the court noted ‘‘while statistics are an appropriate 

method of proving a prima facie case of racial diserimina- 

tion, such statistics must be relevant, material and mean- 

ingful, and not segmented and particularized and fash- 

ioned to obtain a desired coneclusion.’”’ Id. at 1269. See 

  

 



  

  

11 

also Johmsom v. Uncle Ben’s Inc., 628 F.2d 419 (5th Cir. 
1980). 

Circuit courts have also noted that due to the ‘‘in- 
herently slippery nature of statistics’’ they are also sub- 
ject to misuse. See Wilkins v. University of Houston. 654 
F.2d 388 (5th Cir. Unit A 1981). In particular, that court 
focused on the fact that even though multiple regression 
analysis was a sophisticated means of determining the 
effects of factors on a particular variable. such an analy- 
sis was subject to misuse and should be employed with 
great care. Id. at 402-3. Other courts have emphasized 
that even though every conceivable factor did not have to 
be considered in a statistical analysis, the minimum ob-. 
jective qualifications had to be included in the analysis 
(in an employment context). ‘‘ [When the statistical evi- 
dence does not adequately account for ‘the diverse and 
specialized qualifications necessary for [the positions 
in question],’ strong evidence of individnal instances of 
discrimination becomes vital . . ..”’ Valentino v. United 
States Postal Service, 674 F.2d 56, 69 (D.C.Cir. 1982). 

The Eleventh Circuit Court of Appeals has examined 
statistical analyses and noted that the probative value of 
multiple regressions depends npon the inclusion of all 
major variables likely to have a large effect on the de- 
pendant variable and also depends on the validity of the 
assumptions that the remaining effects were not corre- 
lated with independent variables included in the analysis. 
The court also specifically questioned the validity of step- 
wise regressions, such as those used in the instant pro- 
ceedings. Eastland v. Tennessee Valley Adwuthority, 704 
F.2d 613, 621 n.11 (11th Cir. 1983). The court emphasized 

  

  

 



  

2 

that a study had to begin with a decent theoretical idea of 
what variables were likely to be important. 

Thus, examining a statistical analysis depends in part 
on the question of whether the analysis incorporated the 
requisite variables and whether there is an appropriate 
theoretical base for the incorporation of the variables. As 
found by the district court in the instant case, none of the 
models: utilized by Professor Baldus necessarily reflected 
the way the system acted and specifically did not include 
important factors, such as credibility of the witnesses, 
the likelihood of a jury verdict, and subjective factors 
which could be appropriately considered by a prosecutor 
and by a jury. Thus, the district court properly rejected 
the statistical analyses in question. 

More difficult problems arise with the attempted use 
of statistics in death penalty cases. In 1963 problems were 
found with the utilization of statistics, specifically pre- 
sented by Marvin Wolfgang. The circuit court concluded 
that the study presented in that case was faulty for vari- 
ous reasons, including failing to take variables into account 
and failing to show that the jury acted with racial dis- 
crimination. The court also emphasized that it was con- 
cerned in that case with the defendant’s sentencing out- 
come and only his case. The court concluded that the sta- 
tistical argument did nothing to destrov the integrity of 
the trial. Mazwell v. Bishop, 398 F.2d 138 (Sth Cir. 1968). 
remanded on other grounds, 398 U.S. 262 (1970). 

An additional factor in the death penalty situation 
comes from the unique nature of the death sentence it- 
self and the capital sentencing system. In McGautha vu. 
California, 402 U.S. 183 (1971), the Court noted the diffi- 

  

 



  

  

13 

culty in identifying beforehand those characteristics which 
could be utilized by a sentencing authority in imposing 
the death penalty and the complex nature of those fac- 
tors. Other circuit courts have rejected statistical an- 

alyses dune to just such a reason. See Ipinkellink v. Wain- 
wright, 578 F.2d 582 (5th Cir. 1978); Smith v. Balkcom, 
660 F.2d 584 (5th Cir. 1981), om rehearing, 671 F.2d 858 
(5th Cir. Unit B, 1982); McCorquodale v. Balkcom, 325 
F.Supp. 408 (N.D.Ga. 1981), affirmed, 721 F.2d 1493 (11th 
Cir. 1983). 

In cases upholding the constitutionality of various 
death penalty schemes, the Court has recognized that it is 
appropriate to allow a sentencer to consider every aspect 
regarding the defendant and the crime in question in exer- 
cising the discretion as to whether to extend mercy or im- 
pose the death penalty. Thus, in Eddings v. Oklahoina, 
455 U.S. 104 (1982) the Court noted that the rule set down 
in Lockett v. Ohio, 438 U.S. 586 (1978) was a product of a 
‘‘history reflecting the law’s effort to develop a system 
of capital punishment at once consistent and principled 
but also humane and sensible to the uniqueness of the indi- 
vidual.’’ Eddings, supra at 110. 

Other factors that have been recognized by courts as 
being appropriate in a death penalty case and in the prose- 
cutor’s discretion are the willingness of a defendant to 
plead guilty. as well as the sufficiency of the evidence 
available. Shaw v. Martin, 733 F.2d 304 (4th Cir. 1984). 
As recently as 1986, this Court has acknowledged that in 
a capital sentencing proceeding the jury must make a 
‘highly subjective, ‘unique, individualized judgment re- 

garding the penalty that a particular person deserves.’ ’’ 
Caldwell v. Mississippi, 472 U.S. —, 105 S.Ct. 2633, 2645-6 

  

  
 



  
  

14 

n7 (1985); Tummer v. Murray, — U.S. —, 106 S.Ct. 1683 

(1986). In this context, ‘‘it is the jury that must make the 

difficult, individualized judgment as to whether the de- 

fendant deserves the sentence of death.’”” Turner v. JMur- 

ray, supra 106 S.Ct. at 1687. This focuses on what has 

long been recognized as one of the most important func- 

tions that a jury can perform, that is, ‘‘to maintain a link 

between contemporary community values and the penal 

system-—a link without which the determination of punish- 

ment could hardly reflect ‘the evolving standards of de- 

cency that mark the progress of a maturing society.” ”’ 

Witherspoon v. Illinois, 391 U.S. 510, 319 n.15 (1968), 

quoting, Trop v. Dulles, 356 U.S. 86, 101 (1958). 

Thus, the myriad of factors that are available for 

consideration by a prosecutor in exercising his discretion 

and by a jury in determining whether to extend mercy to 

a given defendant simply makes the utilization of these 

types of statistical analyses unworkable in a death penalty 

context. It is simply impossible to quantify subjective 

factors which are properly considered both by the prosecu- 

tor and by the jury in reaching these determinations. In 

fact, the evidence in the instant case fails to take into ac- 

count these subjective factors, including the information 

known to the decision-maker. the likelihood a jurv would 

return a verdict in a particular case. the possible credi- 

bility of individual witnesses. the availability of witnesses 

at the time of trial, the actual sufficiency of the evidence 

as determined by the prosecutor himself as well as num- 

erous other factors. 

In addition to all the above. commentators have also 

recognized that many of the factors present in the instant 

case cause problems with utilizing statistical analvses. 

  

 



  

  

15 

Professor Baldus himself has noted that ‘‘statistical so- 

phistication is no cure for flaws in model construction and 

research design.’’ Baldus & Cole, 4 Comparison of the 

Work of Thorsten Sellin and Isaac Ehriich on the Deter- 

rent Effect of Capital Punishment, 85 Yale L. J. 170, 173 

(1975). In that same article, Professor Baldus acknowi- 

edged that the deterrent effect of capital punishment was 

just such a type of study that would be best suited by 

simpler methods of study than statistical analvsis. Id. 

Other authors have questioned the validity of statistical 

methods which incinde inappropriate variabies in the analy- 

sis as well as those which fail to include necessary vari- 

ables. Jee Finkelstein, The Judicial Reception of Muiti- 
ple Regression Studies in Race amd Sex Discrimination 

Cases, 80 Colum. L.Rev. 737, 738 (1980). Other authors 

have also agreed with the testimony of the experts in this 

case regarding the problems presented by multicollinearity 

as well as the problems in utilizing stepwise regressions. 

See Fisher, Multiple Regression in Legal Proceedings, 30 

Colum. L.Rev. 702 (1980) : See also G. Wesolowsky, Julti- 

ple Regression Analysis of Variance (1976): A. Gold- 

berger, Topics in Regression Analysis (1968). 

Finally, certain authors have questioned the utilization 

of statistical analyses even in employment discrimination 

cases noting ‘‘it may be impossible to gather data on many 

of these differences in qualifications and preferences. 

Consequently, there will likely be alternative explanations, 

not captured by the statistical analysis, for observed dis- 

parities. . . . These alternative explanations must he taken 

into consideration in assessing the strength of the in- 

ference to be drawn from the statistical evidence.’ Smith 

  

   



     

  

     

  

      

  

    
    
    
    
    
      
    
        
      
     

    

     

        
    

  

      

  

    

  

     

16 

and Abram. Quantitative Analysis and Proof of Employ- 
ment Discrimination, 1981 U.Ill. L.Rev. 33, +5 (1981). 

Respondent submits that a consideration of the sta- 

tistical analysis in the instant case reflects that it simply 

fails to comply with the appropriate conventions utilized 

for this type of analysis in that it fails to include appropri- 

ate variables, fails to utilize interaction variables, fails . 

to specify a relevant model and has other fallacies, inciud- 

ing multicollinearity which render the analysis nonpro- 

bative at best. As noted by a statistician in an article re- 

garding race and sex discrimination and regression analy- 

sis: 

It should be again emphasized that a statistical analy- 
sis provides only a limited part of the total picture that 
must be presented to prove or disprove diserimina- 
tion. . . . **No statistician or other scientist should 
ever put himseif/herself in a position of trying to 
prove or disprove discrimination.’’ 

McCabe, The Interpretation of Regression Amalysis Re- 

sults iw Sex and Race Discrimination Problems, 3+ Amer. 

Stat. 212, 215 (1980). 

II. THE STATISTICAL ANALYSES IN THE IN- 
STANT CASE ARE INSUFFICIENT TO 
PROVE RACIAL DISCRIMINATION. 

As noted previously. courts and commentators have 

expressed reservations about the use of statistics in at- 

tempting to prove discrimination. Respondent submits 

that even if the Court concludes statistical analysis is ap- 

propriate in a death penalty context. the ‘‘statistics’’ pre- 

sented to the district court are so flawed as to have no pro- 

 



  

  

7 

bative value and, thus, cannot satisfy the Petitioner’s bur- 

den of proof.> 

Petitioner claims that the studies in question are the 

product of carefully tailored questionaires resulting in the 

collection of over 500 items of information on each case. 

The Respondent has proven, and the district court found, 

that the data bases are substantially flawed, inaccurate 

and incomplete. 

As noted previously, statistical analyses, particularly 

multiple regressions. require accurate and complete data 

to be valid. Neither was presented to the district court. 

Design flaws were shown in the questionnaires utilized to 

gather data. There were problems with the format of 

critical items on the questionnaires, such that there was 

an insufficient way to account for all factors in a given 

case. ‘‘An important limitation placed on the data base 

was the fact that the questionnaire could not capture every 

nuance of every case.”” McCleskey v. Zant, supra at 336. 

(J.A. 1386). 

Further, the sources of the information were notice- 

ably incomplete. Even though the Petitioner insisted that 

  

It is clear that the findings by the district court in regard 
to the question of intent and the evaluation of the statistical 
analysis are subject to the clearly erroneous rule. In United 
States v. United States Cypsum Co., 333 U.S. 364 (1948), the 
Court acknowledged that the clearly erroneous rule set forth in 
rule 52(a) of the Federal Rules of Civil Procedure applied to 
factual findings. “A finding is ‘clearly erroneous’ when although 
there is evidence to support it, the reviewing court on the en- 
tire evidence is left with the definite and firm conviction that a 
mistake has been committed.” Id. at 395. This principle has 
been held to apply to factual findings regarding motivations 
of parties in Title VII actions and it has been specifically held 
that the question of intentional discrimination is a pure question 
of fact. Pullman-Standard v. Swint, 456 U.S. 273, 287-8 (1982). 

  

 



  

18 

he relied on State sources, obviously those sources were not 

designed to provide detailed information on each case. As 

found by the district court, ‘‘the information available to 

the coders from the Parole Board files was very summary 

in many respects.’’ McCleskey v. Zant, supra at 356. (J.A. 

137). These summaries were brief and the police reports 

from which the Parole Board summaries were prepared 

were usually only two or three pages long. (F.H.T. 1343: 

J.A. 137). As fonnd by the district court: 

Because of the incompleteness of the Parole Board 
studies, the Charging and Sentencing Study contains 
no information about what a prosecutor felt about the 
credibility of any witnesses. R 1117. It was occasion- 
ally difficult to determine whether or not a co-perpe- 
trator testified in the case. One of the important 
strength of the evidence variables coded was whether 
or not the police report indicated clear gmilt. As the 
police reports were missing in 75% of the cases, the 
coders treated the Parole Board summary as the po- 
lice report. R 193-94. Then. the coders were able to 
obtain information based only upon their impressions 
of the information contained in the file. R 349. 

McCleskey v. Zant, supra at 357. (J.A. 137). 

Furthermore, questionaires were shown to be mis- 
coded. It was also shown there were differences in judg- 

ment among the coders. (F.H.T. 387). 

Respondent also established that there were numerous 

inconsistencies between the coding for the Procedural Re- 

form Study and the Charging and Sentencing Study. (J.A. 
77-80: S.E. 78; Respondent’s Exhibit 20A). These oc- 
curred in some variables generally considered to he im- 

portant in a sentencing determination. 

  

 



  

  

  

19 

" A further problem with the data base is due to the 
large number of unknowns. Although Petitioner claims to 

have collected information on over 500 variables relating 

to each case, the evidence showed that in the Charging and 

Sentencing Study alone there are an average of at least 33 

variables coded as unknown for each questionnaire. (.J.A. 

139). A review of Respondent’s Exhibits Nos. 17A and 18A 

shows the extent to which unknowns pervade the so-called 

complete data base. For example, in the Charging and 

Sentencing Study there are 445 cases in which it was un- 

known if there was a plea bargain. (S.E. 73-74; J.A. 69- 

74). Further complicating the data is the fact that Baldus 

arbitrarily coded unknowns as if the information did not 

exist without any knowledge as to whether the information 

was known to the prosecutor or jury. 

Even though attempts were made in the district court 

to discount the unknowns, Petitioner did not succeed. In 

fact the district court concinded the so-called ‘worst case’’ 

analysis failed to prove that the coding decisions on the 

unknowns had no effect on the results. (J.A. 142). The 

Respondent also introduced evidence that the correct sta- 

tistical technique would he to discard the cases with un-. 

knowns in the variables being utilized in the analysis and 

not utilize the cases in the analysis.’ 

The district court also concluded that no models of- 

fered by the Petitioner were sufficiently predictive as to 

be probative. (J.A. 149). As noted previously, regres- 

sions must include relevant variables to be probative. See 

3This is precisely the reason no independent model or re- 
gression analysis was presented by the Respondent. The data 
base was simply too flawed and eliminating cases with un- 
knowns reduced the sample size to the extent that a valid 
analysis was futile. 

  

 



20 : 

Bazemore v. Friday, supra. No model was used which 

accounted for several significant factors because the in- 

formation was not in the data base, i.e., credibility of wit- 

nesses, likelihood of a jury verdict, strength of the evi- 

dence, ete.* Many of the small-scale regressions simply 

include a given list of variables with no explanation given 

for their inclusion. Even the large-scale 230-variable re- 

gression has deficiencies. ‘‘It assumes that all of the in- 

formation available to the data-gathers was available to 
each decision-maker in the system at the time that deeci- 

sions were made,’’ McCleskey v. Zant, supra at 361. (J.A. 

146). This is simply an unrealistic view of the criminal 

justice system which fails to consider simple issues such 

as the admissibility of evidence. Further the adjusted 

r-squared, which measures what portion of the variance 

in the dependent variable is accounted for by the inde- 

pendent variables in the model. even in the 230-variahle - 

model, is only approximately .3. (J.A. 147). Petitioner 

also fails to show the coefficients of all variables in the 

regressions. : 

Major problems are also presented due to multi- 

collinearity in the data. See Fisher, supra. (J.A. 105-111). 

Multicollinearity wiil distort the regression coefficients 

in an analysis. (J.A. 106). It was virtoally admitted that 

there is a high correlation between the race of the victim 

variable and many other variables in the study. According 

to the testimony of Respondent’s experts. this was not 

accounted for bv any analysis of Baldus or Woodworth. 

Various experiments conducted hv Dr. Katz confirmed the 

‘Although the second study purports to include strength 
of the evidence variables, there are such a high number of un- 
knowns that it cannot be considered to be effectively included 
in any analysis. 

   



  

  

  

21 

correlation between aggravating factors and white victim 

cases and mitigating factors with black victim cases. Jee 
FHT. 1472, et seq.; Respondent’s Exhibits 49-52. The 

district court specifically found neither Woodworth or 

Baldus had sufficiently accounted for muiticoilinearity in 

any analysis. 

Petitioner has asserted that there is an average twenty 

point racial disparity in death sentencing rates which he 

asserts should constitute a violation of the Eighth or Four- 

teenth Amendments. As noted previously, the statistical 

analyses themselves have not been found to be valid by 

any court making such a determination; thus, this analy- 

sis is questionable at best. Furthermore, focusing on the 

so-called ‘‘twenty percentage point’’ effect misconstrues 

the nature of the study presented. The twenty percentage 

point ‘‘disparity’’ occurred in the so called ‘‘mid-range’’ 

of cases. This analysis attempted to exclude the most ag- 

gravated cases from its consideration as well as the most 

mitigated cases. The analysis did not consider whether the 

cases were actually eligible for a death sentence under state 

law, but was a consideration of all cases in the study which 

have been indicted either for murder or voluntary man- 

slaughter. 

A primary problem shown with the utilization of this 

‘““mid-range’’ analysis is the fact that Petitioner failed to 

prove that he was comparing similar cases in this analysis. 

By virtne of the previously noted substantial variables 

which were not included in the analysis, it can hardly be 

determined that the cases were similar. 

Further, this range of cases referred to by the Pe- 

titioner was constructed based on the index method uti- 

lized extensively by Professors Baldus and Woodworth. 

  

 



. 99 
A , A 

Dr. Katz testified for the Respondent concerning this in- 

dex method and noted that an index is utilized to attempt 

to rank different cases in an attempt to conclude that cer- 

tain cases had either more or less of a particular attribute. 

(J.A. 87). The numbers utilized in the comparisons men- 

tioned above were derived from these indices and the num- 

bers would ‘‘purport to represent the degree for a level of 

aggravation and mitigation in each case for the purpose 

of ranking these cases according to those numbers.”” Id. 

Dr. Katz noted that Professor Baldus had utilized re- 

gression analysis to develop the indices and had used a 

predicted outcome to form the index for aggravation and 

mitigation. Through a demonstration conducted by Dr. 

Katz utilizing four sample regressions, it was shown that 

the index method could he shaped to give different rank- 

ings from the same cases depending on what variabies 

might be included in a particular regression. Through the 

(demonstration. Dr. Katz showed that by including dif- 

ferent variables in the model, the actual values for the 

index would change. ‘‘[T]he purpose of this was to show 

that at any stage, what is happening with the regression 

in terms of the independent variables it has available to 

it, is that it is trving to weigh the variables or assign co- 

efficients to the variables so that the predicted outcomes 

for the life sentence cases will have zero values and the 

predicted outcomes for the death sentence cases will have 

one value, regardless of the independent variables that 

it has to work with.’”’ (J.A. 98-9). The examination of 

this testimony as well as the exhibits in connection there- 

with shows that the index method itself is capable of mis- 

use and abuse and. depending on the particular regression 

eanation ntilized, the index values can be different. No 

     



  

  

  

23 

adequate explanation was provided for the particular var- 

iables included in the regression analysis so as to justify 

utilizing the index values. Thus, it was simply not shown 

that the cases being compared to develop this ‘‘mid-range’’ 

were actually similar. See McCleskey v. Zant, supra at 

375-6. (J.A. 175). 

Additionally, the .06 figure referred to by the Petition- 

er does not represent a true disparity. The .06 so-called 

‘‘disparity’’ does not reflect any particular comparison 

of subgroups of cases. Further the .06 figure is a weight 

which is subject to change when variables are added to 

or subtracted from the model. (J.A. 233). 

Begardless of the standard applied or the propriety 

of utilizing statistics in the instant case, the above shows 

that the data base is substantially flawed so as to be in- 

adequate for any statistical analysis. Any results of any 

such analysis are thus fatally flawed and prove nothing 

about the Georgia criminal justice system. 

III. THE ARBITRARINESS AND CAPRICIOUS. 

NESS CONCERNS OF FURMAN V. GEORGIA, 

408 U.S. 238 (1972), ARE REMOVED WHEN A 
STATE PROPERLY FOLLOWS A CONSTITI- 
TIONAL SENTENCING PROCEDURE. 

Throughout the history of Eighth Abendment juris- 

prudence this Court has recognized, ‘‘[d]ifficulty would 

attend the effort to define with exactness the extent of the 

constitutional provision which provides that cruel and un- 

usual punishments shall not be inflicted . ...’”’ Wilkerson 

v. Utah, 99 U.S. 130, 135-6 (1878). Furthermore, ‘‘ [t]he 

cruelty against which the Constitution protects a con- 

victed man is cruelty inherent in the method of punish- 

  

  

 



  

24 

ment, not the necessary suffering involved in any method 

employed to extinguish life humanely.’’ Louisiana ex rel. 

Francis v. Resweber, 329 U.S. 459, 464, ring. denied, 330 

U.S. 853 (1947). Members of the Court have not agreed 

as to the extent of the applicability of the Eighth Amend- 

ment. In Trop v. Dulles, 356 U.S. 36 (1958), the Court de- 

termined that the question was whether the penalty under 

examination in that case subjected the individual to a fate 

‘“forbidden by the principle of civilized treatment guaran- 

teed by the Eighth Amendment.’”” Id. at 99. The Court 

also went on to note that the Eighth Amendment was not 

a static concept but that the amendment ‘‘must draw its 

meaning from evolving standards of decency that mark 

‘the progress of a maturing society.’’ Id. at 101. 

The Eighth Amendment embodies ‘‘broad and idealis- 

tic concepts of dignity, civilized standards, humanity and 
decency . . . .”” Estelle v. Gamble, 429 U.S. 97 (1976). In 

Ingraham v. Wright, 430 U.S. 65T (1977), the Court 

acknowledged that the Highth Amendment prohibition 

against cruel and unusual punishment circumseribed the 

criminal process in three ways: (1) it limits the particular 

kind of punishment that can be imposed on those con- 

victed; (2) the amendment proscribes punishment that 

would be grossly disproportionate to the severity of the 

crime; (3) the provision imposes substantive limits on 

what can be made criminal and punished as such. 

Not until Furman v. Georgia, 408 U.S. 238 (1972), was 

the Court squarely confronted with a claim that the death 
penalty itself violated the Eighth Amendment. The hold- 

ing of the Court in that case was simply that the carrying 

out of the death penalty in the cases before the Court con- 

stituted cruel and unusual punishment. [d. at 239. 

  

 



    

m
n
 
—
—
—
—
 
 
—
 

. 

  

25 

In Gregg v. Georgia, 428 U.S. 153 (1976), this Court 

specifically examined the Georgia death penalty scheme. 

In so doing, the Court examined the history of the Eighth 

Amendment and the opinion in Furman v. Georgia. The 

Court noted that the Eighth Amendment was to be inter- 

preted in a flexible and dynamic manner and that the 

Eighth Amendment was not a static concept. The Court 

went on to note, however, that the Highth Amendment 

‘“‘must be applied with an awareness of the limited role 

played by courts.’”” Id. at 174. In upholding the Georgia 

statute, the Court acknowledged that Furman established 

that the death sentence could not be imposed hy sentencing 

proceedings ‘‘that created a substantial risk thar it would 

be inflicted in an arbitrary and capricious manner.’’ Id. at 

188. The Court compared the death sentences in Furman 

as being cruel and unusual in the same way as being struck 

by lightning would be cruel and unusual. The Court fur- 

ther noted that Furman mandated that where diseretion 

was afforded to a sentencing body, that discretion had to 

be suitably directed and limited so as to minimize the risk 

of wholly arbitrary and capricious action. Finally, the 

Court acknowledged that in each stage of the death sen- 

teneing process an actor could make a decision which would 

remove the defendant from consideration for the death 

penalty. ‘“Nothing in any of our cases suggests that the 

decision to afford an individual defendant mercy violates 

the Constitution. Furman held only that in order to mini- 

mize the risk that the death penalty would be imposed on 

a capriciously selected group of offenders, the decision 

to impose it had to be guided by standards so that the 

sentence authorized would focus on the particularized cir- 

cumstances of the crime and defendant.”” Gregg, supra    



  

26 

at 199. The Court further emphasized that ‘‘ [t]he isolated 
decision of a jury to afford mercy does not render uncon- 
stitutional a death sentence imposed upon defendants who 
were sentenced under a system that does not create a sab- 
stantial risk of arbitrariness or caprice . ... The propor- 
tionality review substantially eliminates the possibility 
that a person will be sentenced to die by the action of an 
aberrant jury.’’ Id. at 203. The Court finally found that 
a jury could no longer wantonly and freakishly impose a 
death sentence as it was always circumscribed by the 
legislative guidelines. 

The same time as the Court decided Gregg v. Georgia, 
supra, it also decided Proffitt v. Florida, 428 U.S. 242 
(1976). The Court again noted that the ‘‘requirements 

- of Furman are satisfied when the sentencing authority’s 
discretion is gmided and channelled hy requiring the ex- 
amination of specific factors that argue in favor of or 
against the imposition of the death penalty, thus eliminat- 
ing total arbitrariness and capriciousness in its imposi- 
tion.”” Id. at 258. 

Subsequently, the Court actually criticized states for 
restricting the discretion of the juries, thus, outlawing 
statutes providing for mandatory death sentences upon 

conviction of a capital offense. See Woodson v. North 
Carolina, +28 U.S. 280 (1976). The Court has also pro- 
hibited death penalty procedures which restrict the con- 
sideration of mitigating circumstances, consistently em- 
phasizing that there must be an individualized considera- 
tion of both the offense and the offender before a death 
sentence could be imposed. Thus. in Lockett v. Ohio. 438 
U.S. 387 (1978), the plurality noted that the joint opinion 
in Gregg, Proffitt and other cases concluded that in order 

  

 



    

  

7 

to comply with Furmam the ‘‘sentencing procedure should 
not create a substantial risk that the death penalty was 

inflicted in an arbitrary manner, only that the discretion 

he directed and limited so that the sentence was imposed 

in a more consistent and rational manner. . . .”’ Locket, 

supra at 397. 

This Court has considered death penalty cases in an 

Eighth Amendment context. but from a different perspec- 

tive than the arbitrary and capricious infliction of a pun- 

ishment as challenged in Furman. In Godfrey v. Georgia, 

+46 U.S. 420 (1980), the Court was concerned with a par- 

ticular provision of Georgia law and the question of 

whether the Georgia Supreme Court had followed the 

statute that was designed to avoid the arbitrariness and 

capriciousness prohibited in Furman. This Court essen- 

tially concluded that the state courts had not followed 

their own guidelines. This Court concluded that the death 

sentence should appear to be and must be based on reason 

rather than caprice and emotion. As the Georgia courts 

had not followed the appropriate statutory procedures in 

narrowing discretion in that case. the Court concluded 

that the sentence was not permissible under the Eighth 

Amendment. The Court did not deviate from its prior 

holding in Gregg, supra, that bv following a properly 

tailored statute the concerns of Furnian were met. 

The Court considered the death penalty in an Eighth 

Amendment context in Emmnund v. Florida, 458 U.S. 7%2 

(1982). The Court, however, did not consider the ‘‘arbi- 

trary and capricious’’ aspect but focused on the question 

of the disproportionality of the death penalty for En- 

mund’s own conduct in that case. Thus. the Court essen-    



  

28 

tially concluded that the death penalty was disproportion- 

ate under the facts of that case. 

In Califormia v. Ramos, +63 U.S. 992, 999 (1983), the 

Court noted that ‘‘[i]n ensuring that the death penalty is 

not meted out arbitrarily or capriciously, the Court’s prin- 

cipal concern has been more with the procedure by which 

the State imposes the death sentence than with substantive 

factors the State lays before the jury as a basis for im- 

posing death. . . .”” Thus, the Court again focused on the 

state procedure in question and noted that excessively 

vague sentencing standards could lead to the arbitrariness 
and capriciousness that were condemned in Furman. 

Further, in particular reference to the study in the 

instant case. Justice Powell observed: 

No one has suggested that the study focused on this 
case. A ‘‘particularized’’ showing would require— 
as I understand it—that there was intentional race 
diserimination in indieting, trying and convicting [the 
defendant], and presumably in the state appellate and 
state collateral review that several times followed the 
trial. . . . Surely, no contention can be made that the 
entire Georgia judicial system. at all levels, operates 
to discriminate in all cases. Arguments to this effect 
may have been directed to the type of statutes ad- 
dressed in Furman. As our subsequent cases make 
clear, such arguments cannot be taken seriously un- 
der statutes approved in Gregg. 

Stephens v. Kemp, — U.S. — 104 S.Ct. 562 n.2 (1983) 
(Powell. .J., dissenting from the granting of a stay of exe- 
cution). Justice Powell went on to note ‘‘claims based 
merely on general statistics are likely to have little or no 
merit under statutes such as that in Georgia.” Id. 

  

 



  

  

  

29 

Respondent submits that reviewing all of the Court’s 

Eighth Amendment jurisprudence, particularly in the death 

penalty context reflects that in order to establish a claim 

of arbitrariness and capricicusness sufficient to violate 

the cruel and unusual punishment provision of the Eighth 

Amendment, it must be established that the state failed to 

properly follow a sentencing procedure which was suffi- 

cient to narrow the discretion of the decision-makers. As 

long as the state follows such a procedure, the arbitrari- 

ness and capriciousness which were the concern in Fur- 

mam v. Georgia, supra, have been minimized sufficiently to 

_ preclude a constitutional violation, particularly under the 

Eighth Amendment. An Eighth Amendment violation 

would result in the ‘‘arbitrary and capricious’’ context, 
only if the statutory procedure either was insufficient it- 

self or the appropriate procedures were not followed. Other 

death penalty cases under the Eighth Amendment deal 

with different aspects of the cruel and unusual punish- 

ment provision, such as disproportionality or excessive 

sentences in a given case. That is simply not the focus 

of the inquiry here. Under the circumstances of the in- 

stant case. the Petitioner has not even asserted that Geor- 

gia’s procedures themselves are unconstitutional, nor has 

the Petitioner asserted that those procedures which were 

approved in Gregg v. Georgia, supra, were not followed in 

the instant case. Thus, there can be no serious contention 

that there is an Eighth Amendment violation under the 

circumstances of this case. This is particularly true in 

light of the testimony of Petitioner’s own expert that the 

Georgia charging and sentencing system sorts cases on 

rational grounds. (F.H.T. 1277; J.A. 154).    



  
  

30 

Insofar as the Petitioner would attempt to assert some 

type of racial diserimination under the Eighth Amendment 
provisions, there should be a requirement of a focus on 

intent in order to make this sentence an ‘‘aberrant’’ sen- 

tence so as to classify it as arbitrary and capricious. A 

simple finding of disparate impact is insufficient to make 

- a finding of arbitrariness and capriciousness such as was 

the concern in Furman, supra, particularly when a prop- 

erly drawn statute has been utilized and properly followed. 

Only a showing of purposeful or intentional diserimina- 

tion can be sufficient to find a constitutional violation un- 

der these circumstances. 

No Eighth Amendment violation can be shown in the 

instant case as Petitioner’s own witness testified that the 

system acted in a rational manner. As shown by the 

analyses conducted by Professor Baldus and Dr. Wood- 

worth, the more aggravated cases were moved through the 

charging and sentencing system and the most aggravated 

cases generally received a death sentence. The more miti- 

gated cases on the other hand dropped out at various 

stages in the system receiving lesser punishments. Thus, 

this system does function in a rational fashion. Further- 

more, it has not been shown that the death sentence in the 

instant case was arbitrary or capricious in any fashion. 

The jury found bevond a reasonable doubt that there were 

two statutory aggravating circumstances present. The 

evidence also shows that the victim was shot twice, includ- 

ing once in the head at fairly close range. The evidence 

tended to indicate that Petitioner hid and waited for the 

police officer and shot him as the officer walked hy. This 

was an armed robbery by four individuals of a furniture 

  

 



  

  

  

31 

store in which several people were, in effect, held hostage 

while the robbers completed their enterprise. It was thor- 

oughly planned and thought out prior to the robbery occur- 

ring. Furthermore, the Petitioner had prior convictions 

for robbery before being brought to this trial. One of 

Petitioner’s co-perpetrators testified against him at trial 

and a statement of the Petitioner was introduced in which 

he detailed the crime and even hoasted about it. (J.A. 113- 

115). Thus, under the factors in this case it is clear that 

Patitioner’s sentence is not arbitrary or capricious and 

there is clearly no Eighth Amendment violation. 

IV. PROQF OF DISCRIMINATORY INTENT IS 
REQUIRED TO ESTABLISH AN EQUAL 
PROTECTION VIOLATION. 

It is well recogmized that “{a] statute otherwise neu- 

tral on its face, must not be applied so as to invidiousiy 

discriminate on the basis of race.” Washington v. Daris, 

426 U.S. 229, 241 (1976), citing Yick Wo v. Hopkins, 118 

U.S. 356, 369 (1886). This Court has consistently recog- 

nized, however, that in order to establish a claim of dis- 

crimination under the Equal Protection Clause, there must 

be proof that the challenged action was the prodnet of dis- 

criminatory intent. See Washington v. Davis, supra. 

In 1962, the Court examined what was essentially an 

allegation of selective prosecution and recognized, “the 

conscious exercise of some selectivity in enforcement is not 

in itself a federal constitutional violation.” Oyler v. Boles, 

368 U.S. 448, 456 (1962). In cases finding an equal pro- 

tection violation, it is consistently recognized that the hur- 

den is on the petitioner to prove purposeful discrimination 

under the facts of the case. See Whitus v. Georgia, 335 

  

 



  
  

32 

U.S. 545 (1967). The Court specifically has recognized 

that the standard applicable to Title VII cases does not 

apply to equal protection challenges. “We have never held 

that the constitutional standard for adjudicating claims of 

invidious racial discrimination is identical to standards 

applicable under Title VIL . ..” Washington v. Davis, 

supra, +26 U.S. at 239. The Court went on in that case to 

note that the critical purpose of the equal protection clause 

was the “prevention of official conduct discriminating on 

the basis of race.” Id. The Court emphasized that the 

cases had not embraced the proposition that an official 

action would be held to be unconstitutional solely because 

it had a racially disproportionate impact without regard 

to whether the facts showed a racially discriminatory pur- 

pose. It was acknowledged that disproportionate impact 

might not be irrelevant and that an invidious purpose 

could be inferred from the totality of the relevant facts, 

including impact, but ‘‘[d]isproportionate impact . . . 

is not the sole touchtone of an invidious racial discrimina- 

tion forbidden by the Constitution. Standing alone it does 

not trigger the rule [cit.] that racial classes are to be sub- 

jected to the strictest scrutiny. . ..” Id. at 242. 

Again in Castaneda v. Partida, 430 U.S. 482, 493 

(1977), the Court held that *an official act is not uncon- 

- stitutional solely because it has a racially disproportionate 

impact.” (emphasis in original). Further, “[plroof of 

racially discriminatory intent or purpose is required to 

show a violation of the Equal Protection Clause.” Village 

of Arlington Heights v. Metropolitan Houswmg Develop- 

ment Corp., 429 U.S. 252, 265 (1977). In Washington v. 

Davis the Court held that the petitioner was not required 

to prove that the decision rests solely on racially diserim- 

  

 



  

  

inatory purposes, but that the issue did demand a ‘‘sensi- 

tive inquiry into such circumstantial and direct evidence 

of intent as may be available.” Id; Village of driingtom 

Heights, supra. “Absent a pattern as stark as that in 

Gomillion® or Yick Wo, impact alone is not determinative, 

(footnote omitted) and the court must look to other evi- 

dence.” Id. at 266. “In many cases to recognize the lim- 

ited probative value of disproportionate impact is merely 

to acknowledge the ‘heterogeneity’ of the Nation’s popu- 

lation.” Id. at 266 n.13. 

The Court also acknowledged that the Fourteenth 
Amendment guarantees equal laws. not necessarily equal 
results. Whereas impact may be an important starting 

point, it is purposeful discrimination that offends the Con- 

stitution. Persommel ddmimisirator of Massachusetts v. 

Feeney, 442 U.S. 266, 273+ (1979). A discriminatory pur- 

pose “implies more than intent as volition or intent us 

awareness of the consequences. .. . [t implies that the 

decision makers selected or reaffirmed a particular course 

of action at least in part because of not merely in spite 

of its adverse effects on the identified group.” Id. at 279; 

see also Wayte v. United States, — U.S. —, 105 S.Ct. 1524. 

1532 (1985). The Court reemphasized its position in Rog- 

ers v. Lodge, 458 U.S. 613 (1982), in which the Court rec- 

ognized “the invidious quality of a law claimed to be ra- 
cially discriminatory must ultimately be traced to a racially 

discriminatory purpose,” and acknowledged that a showing 

of discriminatory intent was required in all types of equal 

protection cases which asserted racial discrimination. 

SComiilion v. Lightfoot, 364 U.S. 339 (1960). 

  

 



  

  

34 

Thus, it is clear from all of the above that a diserim- 

inatory purpose, requiring more than simply an awareness 

of the consequences, must be established in order to make 

out a prima facie showing of diserimination under the 

Equal Protection Clause, regardless of the type of equal 

protection claim that is raised. The burden is on the in- 

dividual alleging this discriminatory selection to prove the 

existence of the purposeful discrimination and this includes 

the initial burden of establishing a prima facie case as 
well as the ultimate burden of proof. 

In relation to the question of an Equal Protection vi- 

olation, Petitioner has also failed to show intentional or 

purposeful discrimination. The Petitioner presented evi- 

dence to the district court by way of the deposition of 

the district attorney of Fuiton County, Lewis Slaton. 

Throughout his deposition, Mr. Slaton testified that the 

important facts utilized by his office in determining wheth- 

er to proceed with a case either to indictment, to a jury 

trial or to a sentencing trial, would be the strength of the 

evidence and the likelihood of a jury verdict as well as 

other facts. Mr. Slaton observed that in a given case there 

could exist the possibility of suppression of evidence ob- 

tained pursuant to an alleged illegal search warrant which 

would also affect the proseentor’s decision. (Slaton Dep. at 

18). In determining whether to plea bargain to a lesser of- 

fense. Mr. Slaton testified that his office would consider 

how strong the case was, how the witnesses would hold ap 

under cross-examination, what scientific evidence was avail- 

able, the reasons for the crime. the mental condition of the 

parties, prior record of the defendant and the likelihood of 

what the jury might do. Id. at 30. As to proceeding to a 

  

 



  

  

35 

death penalty trial, Mr. Slaton testified that first of all the 

question was whether the case fell within the ambit of the 

statute and then he examined the atrociousness of the 

crime, the strength of the evidence and the possibility of 

what the jury might do as well as other factors. Id. at 31. 

He also specifically noted that his office did not seek the 

death penalty very often, for one reason because the juries 

in Fulton County were not disposed to impose the death 

penalty. Id. at 32. He also specifically testified he did 

not recall ever seeking a death penalty in a case simply 

because the community felt it should be done and did not 

recall any case in which race was a factor in determining 

whether to seek a death penalty. Id. at T8. 

This is a case in which the Petitioner has in effect by 

statistics alone sought to prove intentional discrimination. 

Although Petitioner has alleged anecdotal evidence was 

submitted, in fact, little. if any. was presented to the dis- 

trict court outside the deposition of Lewis Slaton and one 

witness who gave the composition of Petitioner’s trial 

jury. As noted previously, Respondent submits that sta- 

tistics are not appropriate in this type of analysis and the 

Petitioner’s statistics in this case are simply invalid: how- 

ever, regardless of that fact any disparity noted is simply 

not of the nature of such a gross disparity as to compel an 

inference of discrimination, unlike earlier cases before the 

court. See e.g.. Gomillion v. Lightfoot, 364 U.S. 339 (1960). 

Absent the ‘‘inexorable zero’’ or a gross disparity similar 

to that, this type of evidence under the unique cirenmstanc- 

es of a death penalty situation should not he sufficient to 

find an inference of discrimination, particularly when both 

lower courts have found that no intentional discrimination 

was proven. Thus, Respondent submits that regardless of 

  

  

 



  

  

  

  
  

36 

the standard utilized, Petitioner has failed to meet this 

burden of proof. 

Regardless of the standard used for determining when 
a prima facie case has been established, it is clear where 
the ultimate burden of proof lies. Under the circumstances 
of the instant case, it is clear that the ultimate burden of 
proof rested with the Petitioner and he simply failed to 
meet his burden of proof either to establish a prima facie 
case of discriminatory purpose or to carry the ultimate 
burden of proof by a preponderance of the evidence. 

  

  

 



  

  

37 

CONCLUSION 

For all of the above and foregoing reasons, the con- 

vietions and sentences of the Petitioner should be affirmed 

and this Court should affirm the decision of the Eleventh 

Circuit Court of Appeals. 

Respectfully submitted, 

Mary Bere WESTMORELAND 
Assistant Attorney General 
Counsel of Record for Respondent 

Mzicmagr. J. Bowens 
Attorney (General 

Marrow O. Gorobow 
First Assistant Attorney General 

Woraym B. Hoo, Jr. 
Senior Assistant Attorney General 

Mary Bere WESTMORELAND 
132 State Judicial Building 
40 Capitol Square, S. W. 
Atlanta, Georgia 3033+ 
(404) 656-3349