McCleskey v. Kemp Brief for Respondent

    In The
Supreme Court o f the United States

October Term, 1985
------------------------------------- o — — — —

WARREN MCCLESKEY,

v.
Petitioner,

RALPH KEMP, Superintendent,
Georgia Diagnostic and
Classification Center,

Respondent.
o

ON WRIT OF CERTIORARI 
TO THE UNITED STATES COURT OF APPEALS 

FOR THE ELEVENTH CIRCUIT
------------------------------ o — — ------------

BRIEF FOR RESPONDENT

Please serve:

Mart B eth W estmoreland 
132 State Judicial Bldg.
40 Capitol Square, S.W. 
Atlanta, Georgia 30334 
(404) 656-3349

•o-—-------------
Mary B eth W estmoreland 
Assistant Attorney General 
Counsel of Record 
For Respondent
Michael J. B owers 
Attorney General
Marion 0 . Gordon 
First Assistant 
Attorney General
W illiam B. H ill, Jr.
Senior Assistant 
Attorney General

COCKLE LAW BRIEF PRINTING CO., (800) 225-6964 
or call collect (402) 342-2831



1

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?



ii

TABLE OF CONTENTS
Page

QUESTIONS PRESENTED .........................................  i
STATEMENT OF THE CASE ...................................  1
SUMMARY OF THE ARGUMENT ............................ 5
ARGUMENT

I. STATISTICAL ANALYSES ARE INADE­
QUATE AS A MATTER OF FACT AND LAW 
TO PROVE DISCRIMINATION UNDER THE 
FACTS OF THE INSTANT CASE..................... 7

II. THE STATISTICAL ANALYSES IN THE IN­
STANT CASE ARE INSUFFICIENT TO 
PROVE RACIAL DISCRIMINATION...............  16

III. 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.........................................................................  23

TV. PROOF OF DISCRIMINATORY INTENT IS 
REQUIRED TO ESTABLISH AN EQUAL 
PROTECTION VIOLATION................................  31

CONCLUSION .................................................................. 37



Ill

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, denied, 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 v. 

Datapoint 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) ...........8, 24, 25,27,

28, 29, 30
Godfrey v. Georgia, 446 U.S. 420 (1980) ...................... 27
Gomillion v. Lightfoot, 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



IV

TABLE OF AUTHORITIES—Continued
Page(s)

Louisiana ex rel. Francis v. Reswebef, 329 II.S.
459, rlfmg. denied, 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)
(en banc) .......................................................................  4

McCleskey v. Zant, 580 F.Supp. 338 (N.D.Ga. 1984).....1, 2,
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) ..............  12
Oyler v. Boles, 368 U.S. 448 (1962) .......................... 31
Personnel Administrator of Massachusetts v.

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, 458 U.S. 613 (1982) ............................ 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. Waimvright, 578 F.2d 582 (5th Cir.

1978) ............................................................................... 13
Stephens v. Kemp, — U.S. —, 104 S.Ct. 562 (1983) ...  28
Prop 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



V

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)...........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
Wither son 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) ................  14
Woodson v. North Carolina, 428 U.S. 280 (1976) .........  26
Tick Wo v. Hopkins, 118 U.S. 356 (1886) .................... 31, 33

Other A uthorities :

Baldus & Cole, A 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 Colnm. L.Rev. 702 (1980) ....................................15, 20

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

TABLE OF AUTHORITIES—Continued
Page(s)



VI

TABLE OF AUTHORITIES— Continued
Page(s)

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

Smith and Abram, Quantitative Analysis and Proof 
of Employment Discrimination, 1981 U.I11. L.Rev.
33 (1981) .......................................................................  15

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



No. 84-6811

In The
Supreme Court of the United States

October Term, 1985

WARREN MCCLESKEY,
Petitioner,

RALPH KEMP, Superintendent,
Georgia Diagnostic and 
Classification Center,

Respondent.

ON WRIT OF CERTIORARI 
TO THE UNITED STATES COURT OF APPEALS 

FOR THE ELEVENTH CIRCUIT

BRIEF FOR RESPONDENT
— —  ------------ — o ----------------------------------

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 E.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



2

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 
is 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.

McGleskey 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 utilised 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 victim



3

eases tend to be more mitigated, that is correlated with 
mitigating factors. Every expert who testified, with the 
exception of I)r. 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 d isp a rityM cC leskey  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 tvho 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 non-racial variables, including 
strength of the evidence, produce no statistically signifi­
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 discrimination, 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, 887 (11th 
Cir. 1985) (en 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 victim 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).



5

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

•--------------- o----------------

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 
sentence 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 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 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



6

in the instant ease 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 ease 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 ease 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
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



7

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 discrimina­
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



8

or not a death sentence should be imposed.1 Each 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 
discrimination in access to service on governmental bod­
ies, particularly where, as in the ease 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. 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.



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. 
United States, 431 U.S. 324 (1977). Again, this was in the 
context of a Title VII action and not in a ease 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 statistics.



10

Problems have also been noted revolving aronnd 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 Basemore 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 discrimina­
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
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). “  [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 
discrimination becomes vital . . . .”  Valentino v. United 
States Postal Service, 674 F.2d 56, 69 (D.C.Oir. 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. Eastland v. Tennessee Valley Authority, 704 
F.2d 613, 621 n .ll (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 v. 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, A 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 he 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 Pinkelstein, 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 multieollinearity 
as well as the problems in utilizing stepwise regressions. 
See Fisher, Multiple Regression in Legal Proceedings, 80 
Colum. L.Rev. 702 (1980); See also Gr. Wesolowsky, Multi­
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 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 TJ.I11. 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 in 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-



17

Native value and, thus, cannot satisfy the Petitioner’s bur­
den of proof.2

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.”  McCleshey v. Zant, supra at 356. 
(J.A. 136).

Further, the sources of the information were notice­
ably incomplete. Even though the Petitioner insisted that

2lt 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.



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.3

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.



Basemore 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.4 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-variable 
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

20

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 multieollinearity 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 eases 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



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 McGleskey 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.

III. THE ARBITRARINESS AND CAPRICIOUS­
NESS CONCERNS OF FURMAN V. GEORGIA,
403 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, 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 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. 651 (1977), the Court 
acknowledged that the Eighth 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.



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 he 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 Eighth 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 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
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 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 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. Ohio, 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. . . . ”  LocJcett, 
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, 458 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 (3983), 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 ease, 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, — IJ.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 
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 ease. 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 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 Yiek 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 YII 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 VII. . . .” Washington v. Davis, 
supra, 426 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 “  [djisproportionate 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, “ [pjroof 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



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 Arlington 
Heights, supra. “Absent a pattern as stark as that in 
Gomillion5 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 v. 
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 eases which asserted racial discrimination.

5Gomillion v. Lightfoot, 364 U.S. 339 (1960).



34

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



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 discrimination, unlike earlier cases before the 
court. See e.g., Gomillion v. Lightfoot, 364TT.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 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.

—  --------------- — o ------------------------- -



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 Eleventh 
Circuit Court of Appeals.

Respectfully submitted,
Mary B eth W estmoreland 
Assistant Attorney General 
Counsel of Record for Respondent
M ichael J. B owers 
Attorney General
Marion 0 . Gordon
First Assistant Attorney General
W illiam B. H ill, Jr.
Senior Assistant Attorney General

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