McCleskey v. Kemp Brief Amicus Curiae in Support of Respondent

    No. 84-6811

In The

j^uprott? (Emtrt uf tip? lu tfrfi ^ ta te
October Term, 1986

Warren McCleskey,
Petitioner,

v.

Ralph M. Kemp, Superintendent, Georgia Diagnostic & 
Classification Center,

Respondent.

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

BRIEF AMICUS CURIAE 
OF THE WASHINGTON LEGAL FOUNDATION 

AND THE ALLIED EDUCATIONAL FOUNDATION 
IN SUPPORT OF RESPONDENT

Daniel  J. Popeo 
George C. Sm ith  *

W ashington Legal F oundation 
1705 N Street, N.W.
Washington, D.C. 20036 
(202) 857-0240 

Attorneys for Amici Curiae 
Washington Legal Foundation 

and
Allied Educational Foundation 

* Counsel of Record 
Dated: September 19,1986

W i l s o n  - Ep e s  P r i n t i n g  C o . ,  In c . - 7 8 9 -0 0 9 6  - W a s h i n g t o n , D .C . 20001



QUESTIONS PRESENTED

1. Whether a state’s system for imposing capital pun­
ishment which has been otherwise upheld as constitu­
tional in all respects may be held unconstitutional merely 
because the collective sentencing results it has produced 
during a given period of years do not conform to sub­
jective notions of racial proportionality in sentencing.

2. Whether, in the absence of any evidence of inten­
tional race discrimination causing the petitioner’s in­
dividual death sentence, that sentence may be set aside 
as unconstitutional merely because the collective sen­
tencing results of the past do not conform to subjective 
notions of racial proportionality in sentencing.

3. Whether a claim that the death penalty has been 
unconstitutionally imposed due to race discrimination 
can succeed without the necessity of proving purposeful 
or intentional discrimination by state actors merely by 
asserting the claim under the Eighth Amendment instead 
of under the equal protection clause of the Fourteenth 
Amendment.

4. Whether a claim that the death penalty has been 
unconstitutionally imposed due to race discrimination 
can be based upon evidence of disparities in sentencing 
associated solely with the race of the victim, as dis­
tinguished from the race of the defendant.

5. Whether the district court’s factual finding that 
the studies relied upon by petitioner were too flawed and 
untrustworthy to constitute cognizable evidence of ac­
tionable sentencing discrimination was clearly erroneous.

(i)





TABLE OF CONTENTS
Page

QUESTIONS PRESENTED .................................... . i

TABLE OF AUTHORITIES _______ _____ _________ v

INTERESTS OF AMICUS CURIAE..... ........ ............ 1

STATEMENT OF THE CASE_________________ _ 3

SUMMARY OF ARGUMENT..... ........ ....................... 3

ARGUMENT ______ _______ __ ____ ________________ 4

I. MERE FAILURE TO MAINTAIN AN “AC­
CEPTABLE” DEGREE OF RACIAL PROPOR­
TIONALITY IN CAPITAL SENTENCING 
PROVIDES NO GROUNDS FOR STRIKING 
AN OTHERWISE VALID CAPITAL PUNISH­
MENT SYSTEM.......... ...................... ......... .... . 5
A. A Death Sentence’s Constitutionality De­

pends Upon its Conformity with Governing 
Legal and Procedural Standards, Not upon 
its Conformity to Statistical Notions of 
Racial Proportionality ..................................  5

B. The Statistical Disparities Alleged Cannot 
Prove Discriminatory Intent, Which has 
been Consistently Required by the Courts as 
a Necessary Element of a Race-based Attack
on a Death Sentence......... .... ................... . 9

C. The Standard of Statistical Proportionality
Advocated Here is Unreasonable, Unwork­
able, and Unjust when Applied to the Out­
come of the Criminal Sentencing Process.... 13

II. EVEN IF A DISPARATE IMPACT STAND­
ARD WERE APPROPRIATE IN THE CRIM­
INAL SENTENCING CONTEXT, PETI­
TIONER FAILS TO MAKE A PLAUSIBLE 
CASE ON THAT BASIS AS WELL............ 17

(iii)



iv

TABLE OF CONTENTS—Continued
Page

A. Petitioner’s Basic Contention is Based on a
Myth  ................. ...... ........................ .......... 17

B. The Theory of Victim-based Discrimination
is Legally and Logically Invalid .............. ....  19

C. The Findings of the District Court on the
Study’s Invalidity Should be Affirmed.........  21

D. The Myriad Individualized Factors and Com­
binations of Factors Which Influence A Death 
Sentence are not Susceptible to Quantification
or Precise Comparative Analysis ....... ...... ....  22

CONCLUSION............................................................ 25



V

TABLE OF AUTHORITIES
Cases Page

Adams v. Wainwright, 709 F.2d 1443 (11th Cir.
1983)................................ ............................ ......  9, 19

AFSCME v. State of Washington, 578 F. Supp.
846 (D. Wash. 1984), rev’d, 770 F.2d 1401 (9th
Cir. 1985)__ _______________ ______ ________  17

Andrews v. Shulsen, 600 F. Supp. 408 (D. Utah 
1983), appeal pending, No. 84-2781 (10th Cir.
1986)...... .......... ............................................ 9,19, 24-25

Britton v. Rogers, 631 F.2d 571 (8th Cir. 1980),
cert, denied, 451 U.S. 939 (1981)........  .10,16,20

Brogdon v. Blackburn, 790 F.2d 1164 (5th Cir.
1986) __        9

Caldwell v. Mississippi, 105 S.Ct. 2633 (1985)...... 13
City of Cleburne v. Cleburne Living Center, 105

S.Ct. 3249 (1985)..............     16
Furman v. Georiga, 408 U.S. 238 (1972)................ 5-8
Godfrey v. Georgia, 446 U.S. 420 (1980)...............  5
Gregg v. Georgia, 428 U.S. 153 (1976)_____ ____  5
McCleskey v. Zant, 580 F. Supp. 338 (1984)_____ passim
Prejean v. Maggio, 765 F.2d 482 (5th Cir. 1985).. 9
Pulley v. Harris, 104 S.Ct. 871 (1984).... ..........8-9, 12-13
Ross v. Kemp, 756 F.2d 1483 (11th Cir. 1985) ....... 5, 9
Shaw v. Martin, 733 F.2d 304 (4th Cir.), cert, de­

nied, 83 I..Ed. 2d 159 (1984)... ..........................  9
Spinkellink v. Wainwright, 578 F.2d 582 (5th Cir.

1978), cert, denied, 440 U.S. 976 (1979)............ 7, 9,20
Stephens v. Kemp, 104 S.Ct. 562 (1983) ........... . 7
United States v. General Dynamics, 415 U.S. 486

(1974)...................... ...... ....... ........ ...................... 22
Washington v. Danis, 426 U.S. 229 (1976)........ 10-11
Whitley v. Albers, 106 S.Ct. 1078 (1986)........... . 12
Woodson v. North Carolina, 428 U.S. 280 (1976).. 15
Zant v. Stephens, 462 U.S. 862 (1983)..................  5,13

Other Authorities
Fed. R. Civ. P. 52(a) ................... ........ .................  22
R. Berger, Death Penalties (Harv. Press 1982).. 10



VI

TABLE OF AUTHORITIES—Continued
Page

Bureau of Justice Statistics Bulletin, Capital Pun­
ishment 1984, NCJ-98399 (August 1985) .......4,18-20

Note, Discrimination and Arbitrariness in Capital 
Punishment: An Analysis of Post-Furman Mur­
der Cases in Dade County, Florida, 1973-76, 33 
Stanford L. Rev. 75 (1980)................. ............. 18-19



In The

H htpm tt? (E n m l u f  t!p> B tixtm
October Term, 1986

No. 84-6811

Warren McCleskey,
Petitioner,

Ralph M. Kemp, Superintendent, Georgia Diagnostic & 
Classification Center,

________  Respondent.

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

BRIEF AMICUS CURIAE 
OF THE WASHINGTON LEGAL FOUNDATION 

AND THE ALLIED EDUCATIONAL FOUNDATION 
IN SUPPORT OF RESPONDENT

INTERESTS OF AMICI CURIAE

The Washington Legal Foundation (“ WLF” ) is a non­
profit public interest law and policy center based in 
Washington, D.C., with over 80,000 members nation­
wide. WLF engages in litigation, administrative proc­
eedings, and policy advocacy in support of the legal and 
constitutional values and principles on which America 
was founded.

WLF devotes substantial effort to asserting the rights 
of victims of crime and supporting effective law en­
forcement measures. WLF has also been a leading 
voice in support of the legitimacy of the death penalty



2

from both a constitutional and policy standpoint. The 
Foundation’s experience and expertise on this issue are 
reflected in the amicus curiae briefs it has filed in many 
of the leading Supreme Court decisions on capital pun­
ishment. E.g., Zant v. Stephens, 462 U.S. 862 (1983) ; 
Strickland v. Washington, 104 S. Ct. 2052 (1984) ; Ed- 
dings v. Oklahoma, 455 U.S. 104 (1982). WLF attorneys 
have also been repeatedly invited to testify before the 
U.S. Congress on capital punishment issues.

WLF believes the instant case is of critical importance 
for its potential impact on not only capital punishment 
law but on many broader areas where claims of racially 
disparate impact may be raised. If petitioner prevails 
here, the jurisprudence of racial and ethnic proportion­
ality will be carried to unprecedented extremes in the 
governance of this nation. The notion that the duly con­
victed murderer of a policeman could escape an other­
wise valid death sentence by invoking the race of his 
victim as a defense is repugnant to any decent sense 
of law and justice.

The Allied Education Foundational (“AEF” ), estab­
lished in 1964, is a non-profit charitable and educational 
foundation based in Englewood, New Jersey, and devoted 
to the pursuit of knowledge, education, and the broad 
public interest.

As part of its education and public interest efforts, 
AEF also supports the publication of books and studies 
on issues of law and law enforcement. Recently, for ex­
ample, AEF joined with WLF in publishing a scholarly 
legal study on the death penalty, Capital Punishment 
1986: Last Lines of Defense. A chapter of that study 
directly challenges the theory of discrimination in capital 
sentencing reflected in petitioner’s argument in this case. 
Because AEF believes that petitioner’s argument here is 
not only profoundly erroneous as a matter of law, but 
profoundly misleading in its portrayal of the American



3

criminal justice system, AEF’s commitment to the spread 
of knowledge and to the rule of law have motivated it to 
join WLF in the following brief.

STATEMENT OF THE CASE
In the interests of judicial economy, amicus adopts and 

incorporates by reference the statement of the case set 
forth in the Brief of the Respondent.

SUMMARY OF ARGUMENT
1. Georgia’s statutory scheme for imposing the death 

penalty has been repeatedly upheld as constitutional un­
der the exacting standards imposed by this Court. That 
indisputably constitutional system was fairly applied in 
petitioner’s case, and there was no evidence that inten­
tional race discrimination caused or influenced his death 
sentence. The mere fact that petitioner submits a study 
purporting to show that the collective sentencing out­
comes of other Georgia capital cases fail to conform to 
subjective notions of racial proportionality provides no 
valid basis for questioning petitioner’s sentence under 
these circumstances. Allowing death sentences to be re­
versed solely on the basis of disparate impact data, and 
without proof of actual discriminatory motive, would be 
unjust, unworkable, and a source of disastrous upheaval 
for the entire criminal sentencing process.

2. Even if an authentic and substantial race-based 
disparity in sentencing could be viewed as a valid basis 
for invalidating a death sentence, petitioner could not 
prevail on the facts of this case. Official government 
statistics demonstrate that, if anything, the death sen­
tence has been disproportionately imposed on white mur­
der defendants. Petitioner’s attempt to evade that fact 
by shifting his claim to victim-based racial disparities 
cannot salvage his case. This Court has not endorsed 
that oblique theory of discrimination, and there is no just



4

or principled basis for it to do so now. Finally, the Dis­
trict Court’s findings that the sentencing studies relied 
on by petitioner were fatally flawed and invalid were not 
clearly erroneous. They should be affirmed by this Court.

ARGUMENT
Preliminary Statement

This case addresses the extraordinary argument that 
a state’s otherwise valid system for imposing the death 
penalty should be declared unconstitutional solely because 
it fails to allocate death sentences in conformity with 
theoretical notions of racial proportionality. Neither the 
presence of meticulously fair sentencing standards nor 
the absence of any discriminatory intent is considered 
pertinent under this argument. All that counts is the 
racial breakdown of collective sentencing statistics.

Moreover, the petitioner rests his claim on the curious 
premise that juries would discriminate primarily on the 
basis of the slain victim's race, rather than that of the 
criminal defendant in the dock— despite the contradictory 
circumstance that the victim is perforce absent from the 
trial and the victim’s race is rarely a matter of relevant 
concern at trial. Petitioner’s reliance on this contrived 
theory of “victim-based” discrimination is at least under­
standable, however, in light of the fact that the more 
plausible theory of direct discrimination against black de­
fendants does not stand up. Official studies comparing the 
sentencing of white and black perpetrators now establish 
that it is actually white murderers who disproportion­
ately receive the death penalty. See Bureau of Justice 
Statistics Bulletin, Capital Punishment 1 9 8 pp. 7, 9, 
Tables 11, A -l, A-2 (August 1985). This inescapable 
fact refutes petitioner’s sweeping factual claim that the 
death penalty discriminates against minorities. His legal 
theory fares no better.



5

I. MERE FAILURE TO MAINTAIN AN “ACCEPT­
ABLE” DEGREE OF RACIAL PROPORTIONAL­
ITY IN CAPITAL SENTENCING PROVIDES NO 
GROUNDS FOR STRIKING AN OTHERWISE 
VALID CAPITAL PUNISHMENT SYSTEM
A. A Death Sentence’s Constitutionality Depends Upon 

Its Conformity With Governing Legal And Proce­
dural Standards, Not Upon Its Conformity To 
Statistical Notions of Racial Proportionality

Petitioner, the duly-convicted murderer of a policeman 
in Fulton County, Georgia, was sentenced to death by a 
judge following the binding recommendation of a jury. 
He now claims that his death sentence should be set aside 
because he is black, the policeman he murdered was 
white, and a study he cites purports to show that death 
penalties are disproportionately imposed on killers of 
white people.

The dispositive flaw in petitioner’s argument is that it 
utterly discounts the significance of the extensive legal 
safeguards incorporated in the Georgia death penalty 
scheme in conformity with post -Furman capital sen­
tencing requirements. Georgia’s current death penalty 
statute and practice have been reviewed, refined, and ap­
proved under this Court’s exacting constitutional scru­
tiny. Gregg v. Georgia, 428 U. S. 153 (1976); Godfrey 
v. Georgia, 446 U.S. 420 (1980) ; Zant v. Stephens, 462 
U.S. 862 (1983). Those cases, together with numerous 
lower court decisions upholding Georgia death sentences 
against other forms of attack, e.g., Ross v. Kemp, 756 
F.2d 1483 (11th Cir. 1985), establish that the Georgia 
capital sentencing system has satisfactorily eliminated 
the kind of standardless, arbitrary sentencing discretion 
originally condemned in Furman v. Georgia, 408 U.S. 
238 (1972). It does so by, inter alia, enumerating objec­
tive aggravating circumstances which genuinely narrow 
the class of persons eligible for the death penalty and 
by providing for “ individualized determination and ap­
pellate review at the selection stage.” Zant v. Stephens,



6

462 U.S. at 879-80. The Georgia system even exceeds 
constitutional requirements by providing for a form of 
‘proportionality review’ by the Georgia Supreme Court 
in each case. Id. at 880 n. 19.

Georgia having satisfied this Court’s exacting standards 
of fairness and procedure in capital sentencing, petitioner 
now urges the Court to superimpose a novel and funda­
mentally different requirement. He com,tends that the 
state must insure some acceptable (but unspecified) de­
gree of racial proportionality in the allocation of the 
death sentence. Not only must the state ensure that 
minority murderers receive no more than their “propor­
tional” share of death sentences, but it must also guar­
antee that those murderers -who choose to kill white vic­
tims are not disproportionately sentenced to death. This 
approach would require generalized, class-based consider­
ations to preempt the particulars of the individual crime 
in deciding whether the death penalty is justified. It is 
racial balancing run amuck.

How the state is expected to achieve and maintain this 
state of fine-tuned racial equilibrium in sentencing is 
not explained or addressed in petitioner’s arguments— 
and for good reason. For to do so would only bring peti­
tioner, full circle, to the very kind of standards which 
this Court has already established— and which the State 
of Georgia has already satisfied— as a remedy to the arbi­
trary and standardless sentencing practices struck down 
in the Furman case. Facial discrimination is merely one 
manifestation of the arbitrary and irrational sentencing 
inequities which the post-F^rmcm. capital sentencing stat­
utes were designed to minimize and contain. A capital 
sentencing system which has been carefully reviewed and 
approved by this Court on those terms is no less consti­
tutional merely because the collective sentencing results it 
produces do not conform to notions of demographic parity.

Thus, the sufficient answer to petitioner’s contentions 
was stated by the Fifth Circuit in the leading case of



7

Spinkellink v. Wainwright, 578 F.2d 582, 613 (5th Cir. 
1978), cert, denied, 440 U.S. 976 (1979) :J

The allegation that Florida’s death penalty is being 
discriminatorily applied to defendants who murder 
whites is nothing more than an allegation that the 
death penalty is being imposed arbitrarily and ca­
priciously, a contention we previously have consid­
ered and rejected.

*  *  *

As we previously noted, this Court reads Furman, 
Gregg, Proffitt, Jurek, Woodson, and Roberts as 
holding that if a state follows a properly drawn 
statute in imposing the death penalty, then the arbi­
trariness and capriciousness— and therefore the ra­
cial discrimination— condemned in Furman have 
been conclusively removed.

Petitioner’s contrary approach subordinates the signifi­
cance of the actual procedures and practices followed in 
his case to the cumulative sentencing results in hundreds 
of remote cases tried years before, involving different 
crimes, different victims, different judges, and different 
juries. Even if validated post-Furman procedures were 
scrupulously adhered to throughout his case, and even if a 
perfectly unbiased judge and/or jury decided his sen­
tence, the constitutionality of that sentence would be dic­
tated by the collective statistical profile of the unrelated 
cases of the past. This is not a rational basis for invali­
dating a given murderer’s sentence. It is a statistical 
lottery.

1 This very same point has been echoed in the opinions of mem­
bers of this Court. E.g., Stephens v. Kemp, 104 S.Ct. 562, 564-65 
(1983) (Powell, J., dissenting), where Justice Powell, joined by 
three other justices, dating asserted, “ It should be apparent from 
the decisions of this Court since Gregg was decided that claims 
based merely on general statistics are likely to have little or no 
merit under statutes such as that in Georgia.”  [emphasis added]. 
This statement squarely applies to the instant case.



8

Petitioner’s arguments make a mockery of the very 
core of the post -Furman approach to capital punishment 
— i.e., that the best means of achieving fairness and 
rationality in capital sentencing is by observing objective 
standards and procedures which limit and channel sen­
tencing discretion without eliminating it altogether. In 
effect, petitioner contends that full and faithful compli­
ance with such approved standards is futile if it does, not 
produce (and maintain) results which conform to conclu- 
sory notions of racially “proportionate” sentencing. This 
“ result-oriented” approach is alien to this Court’s post- 
Furman jurisprudence on capital punishment, and should 
be firmly rejected.

The most significant shortcoming of the Baldus Study 
in this context is that it tells us nothing about the fair­
ness and legal propriety of petitioner's trial and sen­
tencing. There is no evidence here showing that McCles- 
key’s conviction and sentencing were actually motivated 
by race discrimination—  intentional or otherwise— or by 
any other impermissible considerations. The authors of 
the Baldus study themselves concede as much. 753 F.2d 
at 895. In fact, petitioner’s entire case was conducted in 
faithful conformity to the rigorous procedures required 
for all capital proceedings under federal constitutional 
law and the law of Georgia.

To invalidate his sentence based upon flawed evidence 
of an unremarkable deviation from racial proportionality 
would be to subordinate settled standards of criminal
procedure to the vagaries and manipulations of question­
able social science theory. This Court should decline such 
a dubious invitation.

In rejecting the closely-related argument in Pulley v. 
Harris that “proportionality review” of all death sen­
tences is constitutionally required, this Court stressed 
that in light of the many other safeguards incorporated 
in the approved post-Furman death penalty statutes “pro­



9

portionality review would have been constitutionally 
superfluous.”  104 S.Ct. at 879 [emphasis added]. The 
race-based statistical analysis of past sentences in capital 
cases is but an improvised variant of proportionality re­
view, and it is redundant and unnecessary for the same 
reasons stated in Pulley v. Harris.

B. The Statistical Disparities Alleged Cannot Prove 
Discriminatory Intent, Which Has Been Consist­
ently Required By the Courts As A Necessary 
Element Of A Race-Based Attack On A Death 
Sentence

Petitioner’s arguments notwithstanding, the federal 
courts have consistently and properly required proof of 
discriminatory intent as a mandatory element of claims 
that the death penalty violates the Eighth and/or Four­
teenth Amendments by some form of race discrimina­
tion. The cases so holding are legion. E.g., Spinkellink 
v. Wainwright, supra, 578 F.2d. at 612-15; Adams v. 
Wainwright, 709 F.2d. 1443, 1449-50 (11th Cir. 1983) ; 
Ross v. Kemp, 756 F.2d 1483, 1491 (11th Cir. 1985) ; 
Shaw v. Martin, 733 F.2d. 304, 311-14 (4th Cir. 1984), 
cert, denied, 83 L.Ed. 2d. 159 (1984) ; Brogdon v. Black­
burn, 790 F.2d. 1164, 1170 (5th Cir. 1986) ; Prejean v. 
Maggio, 765 F.2d. 482, 486 (5th Cir. 1985) ; Andrews v. 
Shulsen, 600 F.Supp. 408, 426 (D.Utah 1983), appeal 
pending, No. 84-2781 (10th Cir. 1986).

Petitioner now asks this Court to hold that this im­
posing array of federal precedents is wrong, and that 
discriminatory intent really need not be proven at all. 
(Pet.’s Br. pp. 98-104). Petitioner would effectively eli­
minate the intent requirement by the simple expedient 
of recasting his equal protection/discrimination claim in 
the guise of an Eighth Amendment claim, and contend­
ing that discriminatory intent is wholly irrelevant to a 
claim of cruel and unusual punishment. (Pet.’s Br. pp. 
97-103).



10

There are numerous dispositive flaws in this argument.
Initially, as cogently expressed by the district court 

(McCleskey v. Zant, supra, 580 F.Supp. at 346-47), 
the Eighth Amendment does not even validly apply to 
death penalty appeals based upon “ race of the victim” 
disparate impact theory. Relatedly, the Eighth Circuit 
has held that perpetrators lack standing to assert a claim 
based on disparate sentencing impact in relation to the 
victim’s race. Britton v. Rogers, 631 F.2d 571, 577 n.3 
(8th Cir. 1980), cert, denied, 451 U.S. 939 (1981). See 
also Spinkellink, supra, 578 F.2d at 614 n.39 (“ the focus 
of any inquiry into the application of the death penalty 
must necessarily be limited to the persons who receive 
it rather than their victims” ). This Court should now 
hold that constitutional attacks on the death penalty 
based on claims of victim-related racial disparities in 
collective sentencing data may be maintained {if at all, 
see Point II.B, infra) only under the equal protection 
clause of the Fourteenth Amendment. Compare McCles­
key v. Zant, supra, 580 F.Supp. at 347. Such claims are 
not remotely within the scope of the cruel-and-unusual 
punishment clause as contemplated and recorded by the 
Framers of the Bill of Rights. See R. Berger, Death 
Penalties, pp. 44-58 (Harv.U.Press 1982). That amend­
ment bans only cruel and barbarous punishments, and 
does not purport to establish a standard of proportion­
ality or parity for the allocation of sentences among the 
various classes of criminals.

Further, acceptance of petitioner’s argument would 
effectively nullify the discriminatory intent element 
which is indisputably required to sustain a death penalty 
challenge on equal protection grounds. Washington v. 
Davis, 426 U.S. 229 (1976). This requirement of pur­
poseful discrimination normally requires direct proof of 
actual discriminatory motive; only in the very rare cir­
cumstances where the disparate impact is so monolithic 
as to defy explanation on any plausible non-racial



11

grounds can the intent requirement be satisfied by “ im­
pact” statistics alone. Washington v. Davis, supra, 426 
U.S. at 242. Here, there are so many alternative plausi­
ble explanations for the claimed racial disparities in 
death-sentencing2— e.g., the demonstrated fact that white- 
victim murders are a consistent “proxy” for high-aggra­
vation felony murders (see Point II. A., infra)— that a 
purely statistical mode of proof is plainly foreclosed.

Whatever the required mode of proof, the specific in­
tent requirement for claims of racially discriminatory 
action by the state cannot be evaded by simply present­
ing the claim in alternative legal garb. A claim of un­
constitutional race discrimination is still just that, 
whether asserted under the Eighth or Fourteenth Amend­
ment. The mandatory element of purposeful discrimina­
tion is grounded on decades of mature and considered 
jurisprudence; it reflects the considered judgment of our 
law that seemingly “ disproportionate”  outcomes in terms 
of race or other characteristics are generally explainable 
by a host of legitimate factors other than actionable dis­
crimination; and it is not to be dismissed by the kind of 
legal sleight-of-hand attempted by petitioner in this case.

Petitioner also errs in contending that the element of 
intent is simply irrelevant to Eighth Amendment claims. 
Any shortage of caselaw explicitly stating a discrimina­
tory intent requirement results from the simple fact that 
discrimination claims like petitioner’s are simply inap­
posite to Eighth Amendment jurisprudence, the precise 
and proper concern of which is barbarous forms of 
punishment rather than a guarantee of racial equilibrium 
in sentencing. To the extent that the Eighth Amendment

2 Among these plausible alternative explanations are the myriad 
non-racial variables which were not taken into account by the 
Baldus Study in trying to explain the sentencing “discrepancies” 
which the petitioner is pleased to ascribe to race. See Point II.D, 
infra.



12

might be held to encompass claims of racially discrimina­
tory sentencing, it would be utterly anomalous to hold 
that such claims may be established on facts which 
would plainly fail to violate the Fourteenth Amendment. 
It is only by virtue of the Fourteenth Amendment, after 
all, that the Eighth Amendment has any application to 
the State of Georgia’s sentencing practices at all.

Further, this Court only recently reiterated that the 
intent and culpability of state actors is indeed relevant 
to Eighth Amendment claims. In Whitley v. Albers, 106 
S.Ct. 1078, 1084 (1986), Justice O’Connor’s opinion for 
the Court stated as follows:

It is obduracy and wantonness, not inadvertence or 
error in good faith, that characterize the conduct 
prohibited by the Cruel and Unusual Punishments 
Clause, . . . .

While there the Court was addressing the Eighth 
Amendment’s application to conditions of confinement 
rather than methods of sentencing, the underlying prin­
ciple still applies in both instances: The cruel and un­
usual punishment clause has no legitimate application to 
the merely “ inadvertent” and unintentional imperfec­
tions and aberrations in our human system of criminal 
justice. Accord: Pulley v. Harris, 104 S.Ct. at 881.

Petitioner’s contention that inadvertent statistical dis­
parities in the distribution of death sentences violates 
the Eighth Amendment is a grotesque distortion of the 
Constitution. The Eighth Amendment has nothing to do 
with a requirement for precisely calibrated allocations of 
sentences among the various races and ethnic groups.

What the Eighth Amendment has been held to require 
in the allocation of the death sentence is that it not be 
dispensed in a wholly arbitrary and “freakish” manner, 
such that there is no rational justification for the deci­
sion that one man is sentenced to death while another



13

receives only a term of imprisonment. The death penalty 
procedures applied in this case by the State of Georgia 
have conclusively passed that test, Zant v. Stephens, 462 
U.S. at 879-80, and nothing in the Baldus studies can 
undermine that controlling fact.

C. The Standard of Statistical Proportionality Advo­
cated by Petitioner Is Unreasonable, Unworkable, 
And Unjust When Applied To The Outcome of the 
Criminal Sentencing Process

This Court has repeatedly stressed that in capital 
cases the jury is called upon to make a “ highly subjec­
tive, unique, individualized judgment regarding the 
punishment that a particular person deserves.” Caldwell 
v. Mississippi, 105 S.Ct. 2633 n.7 (1985), (quoting Zant 
v. Stephens, 462 U.S. 862, 900 (1983). That sensitive 
judgment is simply not susceptible to the crude cate­
gorizations and generalizations on which all the conclu­
sions and comparisons of the Baldus study must ulti­
mately rest.

In Pulley v. Harris, supra, 104 S.Ct. at 881, this Court 
further acknowledged that

Any capital sentencing scheme may occasionally pro­
duce abberational outcomes. Such inconsistencies are 
a far cry from the major systemic defects identified 
in Furman. As we have acknowledged in the past, 
“ there can be no ‘perfect procedures for deciding in 
which cases governmental authority should be used 
to impose death.’ ” [citations omitted]

Petitioner’s arguments cannot be reconciled with the 
foregoing observations. Petitioner’s theory holds that 
any deviation® from statistically-based norms of racially 3

3 Petitioner’s brief asserts that “ under the constitutional princi­
ples outlined earlier, racial discrimination of any magnitude is 
unconstitutional.” (Pet.’s Br., p. 95; emphasis added).



14

proportional outcomes in a capital sentencing system 
would “ require the invalidation of that system as a 
whole.” Pet. Br. p. 107. The disastrous practical im­
plications of this legal theory are perhaps the best proof 
of its invalidity.

Initially, the Court should carefully ponder exactly 
what a state would be required to do in order to “ re­
habilitate” a capital punishment system condemned un­
der petitioner’s theory of “ statistical unconstitutionality.” 
If the reason for the system’s invalidation is its failure 
to conform capital sentencing outcomes to “ acceptable” 
norms of racial balance, then the only fitting remedy 
would presumably be one that would eliminate or rectify 
such disparities to the fullest extent possible. See, e.g., 
Swann v. Charlotte Mecklenberg, 402 U.S. 1 (1971).

It would plainly not be enough for the state to enact 
and implement objective procedures and standards which 
prevent the arbitrary and unrestricted exercise of sen­
tencing discretion. The State of Georgia has already 
done precisely that, to the full satisfaction of this Court. 
See Gregg and Zant, supra. The only evident alternative, 
then, would be for the state to take more direct and 
positive measures— known in other contexts as affirma­
tive action— to assure the elimination of racially dis­
proportionate sentencing outcomes.

This would presumably and logically entail a mora­
torium on the execution of all black murderers and of 
all murderers of white victims until the offensive statis­
tical disparity was eliminated. Executions of white mur- 
dei ers of black victims could presumably go forward, 
since neither “defendant-based” nor “victim-based” racial 
bias could be credibly asserted in such cases. If this 
seems a bizarre and distorted remedy, it is because pre­
cisely such a remedy is required to fit the distorted and 
anomalous logic of petitioner’s legal theory.

There is really no remedy which could satisfy the un­
reasonable and unrealistic standards of class-based jus-



15

tiee advanced by petitioner in this case. Petitioner’s pur­
ported concern that racial factors infect the sentencer’s 
decisions in capital cases could only be resolved by the 
abolition of all jury discretion and the adoption of a 
mandatory death penalty approach (or, of course, com­
plete abolition). But this Court has already rejected 
such an approach, Woodson v. North Carolina, 428 U.S. 
280 (1976), in favor of a regime which consciously 
tolerates the occasional variances produced by the sen­
tencer’s discretion as long as they are rationally gov­
erned by objective limitations and standards. Pulley v. 
Harris, 104 S.Ct. at 881. Acceptance of petitioner’s argu­
ments in this case would require the abandonment of 
these fundamental principles of post-Furman capital 
punishment law.

The logic of petitioner’s theory entails further prac­
tical repercussions which are incompatible with any 
viable system of criminal sentencing.

If a state’s capital sentencing system is invalid for its 
failure to produce racially proportionate outcomes, then 
what of the other forms of criminal sentencing? For ex­
ample, if those sentenced to death in Georgia were in­
stead sentenced to life imprisonment without possibility 
of parole, would the racial proportionality argument lose 
all of its force— such as it is-—merely because the death 
penalty was no longer implicated? Nothing in the core 
logic of petitioner’s argument so indicates.

Indeed, petitioner’s primary argument in this case is 
phrased as follows (Pet.’s Br. p. 32) : “A. The Equal 
Protection Clause of the Fourteenth Amendment Forbids 
Racial Discrimination in the Administration of Criminal 
Statutes.” [emphasis added]. Although this point is un­
assailable by itself, petitioner insistently equates collec­
tively “ disproportionate” sentencing outcomes with the 
actionable racial discrimination he refers to. The argu­
ment therefore plainly extends the demand for racial 
equilibrium in sentencing to other serious criminal pen­



16

alties (e.g., life imprisonment), if not to all criminal 
penalties. Compare Britton v. Rogers, 631 F.2d 572 
(8th Cir. 1980), where the court rejected the argument 
that racially disparate sentencing outcomes in past rape 
cases justified habeas corpus relief.

The implication is clear. Acceptance of petitioner’s 
argument would open the door to Title VH-style “ dis­
parate impact” challenges to criminal sentences of all 
kinds. The entire criminal sentencing process would be­
come bogged down in the same morass of “ underutiliza­
tion” concepts, multivariate regression analysis, and 
“goals” or quotas which now complicate employment dis­
crimination law.

Nor do the radical implications end there.
If the Constitution requires collective sentencing out­

comes to satisfy some acceptable norm of racial propor­
tionality, what then of the other “ suspect” classifications 
under this Court’s Equal Protection jurisprudence? For 
example, discriminations based on alienage or on national 
origin now trigger the same degree of scrutiny as race 
discrimination. City of Cleburne v. Cleburne Living Cen­
ter, 105 S. Ct. 3249, 3255 (1985). Moreover, it is now 
recognized that gender-based classifications “also call for 
a heightened standard of review,” City of Cleburne, 105 
S. Ct. at 3255, as do those based upon illegitimacy. Id.

Accordingly, petitioner’s theory would also require pro­
portional allocation of capital sentences with respect to 
such classifications as alienage, ethnicity, sex, and legiti­
macy. If black murderers are entitled to invalidate their 
death sentences on grounds of statistical disparate im­
pact, it would follow that those falling within the other 
specially protected classifications are entitled to produce 
additional studies showing analogous forms of disparate 
impact as to their respective groups. Further, petition­
er’s argument would allow defendants of all classifica­
tions to challenge their sentences based on corresponding



17
variants of petitioner’s theory of victim-oriented discrim­
ination— e.g., a claim that those who murder American 
citizens are more likely to receive the death sentence 
than those who murder resident aliens. Such a claim 
would stand on the exact same constitutional footing as 
the claim at issue here. All of these predictable reper­
cussions would hopelessly complicate the state’s efforts 
to enforce capital punishment systems which have already 
been upheld as valid by this Court.

These are not exaggerated alarms, but merely acknowl­
edgement of the logical consequences that could follow 
the Court’s acceptance of petitioner’s radical theory. Just 
as theories of statistical-based employment discrimination 
have produced permutations once deemed inconceivable, 
e.g., AFSCME v. State of Washington, 578 F.Supp. 846 
(D.Wash. 1984), rev’d, 770 F.2d 1401 (9th Cir. 1985), 
so too would endorsement of petitioner’s theory of dis­
parate impact in sentencing lead to bizarre and unfore­
seen applications as well.

No workable system of criminal justice could accom­
modate the demands for race- and class-based parity in 
sentencing advanced by petitioner. Nor does the Consti­
tution require a regime of “ statistical justice” which 
would subject the validity of every criminal sentence to 
the vagaries and manipulations of fluctuating demographic 
data.

II. EVEN IF A DISPARATE IMPACT STANDARD 
WERE APPROPRIATE IN THE CRIMINAL SEN­
TENCING CONTEXT, PETITIONER FAILS TO 
MAKE A PLAUSIBLE CASE ON THAT BASIS AS 
WELL
A. Petitioner’s Basic Contention is Based on a Myth

The core premise of petitioner’s argument is the per­
sistently repeated charge that the death penalty as admin­
istered today pervasively discriminates against blacks. 
The problem with this key premise is that it is demon­
strably false.



18
In a comprehensive study of sentences imposed on thou­

sands of killers during the period 1980-1984, the Justice 
Department’s Bureau of Justice Statistics has discovered 
that it is white defendants who are disproportionately 
sentenced to death and disproportionately executed in 
this country. Bureau of Justice Statistics Bulletin, Capi­
tal Punishment 198Jp, NCJ-98399, pp. 7-9, Tables 11, A-l, 
A-2 (August 1985) (hereafter cited as “BJS Bulletin” ).

The BJS report shows that for every 1,000 whites ar­
rested on homicide charges, approximately 16 were sent 
to prison under sentence of death. BJS Bulletin, at p. 9, 
Table A-2. In comparison, fewer than 12 blacks for 
every 1,000 arrested on the same charges were sent to 
death row. The data indicates that white perpetrators 
as a group are 36% more likely to be sentenced to death 
than black perpetrators of comparable capital offenses.

Further, white homicide convicts on average run a sig­
nificantly greater likelihood than their black peers (i.e., 
55% more likely) of actually being executed subsequent 
to death sentence. From 1977 to 1984, 1.7% of all death 
row whites were actually executed, compared to only 
1.1% of blacks on death row. Id., p. 7, Table 11.

These nationwide figures are not to suggest that the 
death penalty as administered actually discriminates 
against white perpetrators. The complex combination of 
factors involved in each individual homicide is so unique 
and personalized that attempts to draw legitimate infer­
ences from such generalized class-based sentencing varia­
tions are futile.

But the BJS statistics do discredit petitioner’s sweep­
ing contention that anti-black discrimination permeates 
the capital sentencing process. Moreover, other reputable 
studies undercut the claims of victim-anchored race dis­
crimination in capital sentencing as well.4 In sum, the

4 See, e.g., Note, Discrimination and Arbitrariness in Capital 
Punishment: An Analysis of Post-Furman Murder Cases in Dade



19
image of a pervasively discriminatory criminal justice 
system which petitioner seeks to convey as a means of 
attacking the death penalty is flatly inaccurate.

Petitioner might protest that the BJS Bulletin reflects 
nationwide data and is therefor technically irrelevant to 
a murder conviction under Georgia state law. But by 
the same reduction logic, the state-wide data relied upon 
for petitioner’s most strongly-asserted contentions would 
also be over-inclusive.

A truly-focused study for purposes of legitimate, 
“ apples-to-apples” comparison between petitioner’s sen­
tence and those in like cases— and one which eliminates 
cross-regional and urban/rural factors which might also 
account for sentencing disparities'—would have to be con­
fined to (1) murders of law enforcement officers (2) in 
Fulton County only. Such a comparison with cases truly 
similar to his own would seem an obvious prerequisite 
to an individual claim of discriminatory sentencing. 
However, the limited number of such cases (i.e., s i x -  
see 580 F. Supp. at 378) is too small to allow for any 
valid statistical analysis or comparison. See, e.g., Adams 
v. Wainwright, supra, 709 F.2d at 1449; Andrews v. 
Shulsen, supra, 600 F.Supp. at 426. Accordingly, if the 
Court were to limit the proof to truly comparable cases 
within the specific prosecution venue, the statistical ap­
proach is plainly unsuitable due to insufficient data.

B. The Theory of Victim-Based Discrimination is 
Legally and Logically Invalid

Petitioner’s curious reliance on the oblique “ race-of- 
the-victim” approach is best explained by the fact that 
focusing strictly on race of the defendant simply would

County, Florida, 1973-76, 33 Stanford L. Rev. 75, 100-01 (1980), 
which demonstrates that the seeming predominance of death sen­
tences in the case of white-victim murders by blacks is fully ex­
plained by the fact that such killings disproportionately account 
for the highly aggravated felony-murders which allow and motivate 
death sentences.



20

not work. As clearly demonstrated by the district court, 
580 F.Supp. at 368, by the Court of Appeals, 753 F.2d. 
at 887, and by the BJS Bulletin, supra, the death penalty 
is not disproportionately applied to black defendants. 
On the contrary.

Although Eleventh and Fifth Circuit cases have 
broadly assumed that a death sentence may be challenged 
on the alternative grounds of victim-based disparate im­
pact statistics, that theory is by no means established as 
the Law of the Land.

Some courts have displayed well-founded skepticism 
towards this oblique and “once-removed” method of at­
tempting to prove discrimination. In Spinkellink v. 
Wainwright, 578 F.2d at 614 n.39, the Fifth Circuit 
approvingly quoted the district court’s ruling that chal­
lenges to the application of the death penalty “must nec­
essarily be limited to the persons who receive it rather 
than their victims” . In Britton v. Rogers, supra, 631 
F.2d at 577 n.3, the Eighth Circuit held that convicted 
criminals lack standing to challenge victim-based racial 
discrepancies in sentencing. And the district court in 
the instant case opined that such victim-based claims are 
not cognizable under either the Eighth Amendment of 
the equal protection clause of the 14th Amendment. 580 
F.Supp. at 347.

These concerns are well-taken, and should command the 
careful attention of this Court. A murderer freely se­
lects his own victim; it would therefore be grotesquely 
ironic for this Court to hold that the slain victim’s race 
can be subsequently invoked by the murderer as a shield 
against his just punishment. Yet that is exactly what 
the petitioner is doing in this case. A more distorted 
variant of the doctrine of jus tertii would be difficult to 
imagine.

There are other convincing reasons why the Baldus 
study’s race-of-the-victim statistics cannot serve as a



21

valid or reliable basis for overturning death sentences. 
For instance, the record shows that the Baldus study 
was unable to account for the race of the victim in 62 
of the cases it examined. 580 F.Supp at 358. This raises 
the question of precisely how the Baldus study was able 
to verify that the juries in all the studied cases had ac­
tually considered clear and reliable evidence of the race 
of the victim. After all, the murder victim is not present 
at the trial and the victim’s race is not normally a con­
tested point requiring proof or authentication. There­
fore, it is not at all clear that reliable evidence of the 
victim’s race is uniformly and unambiguously conveyed 
to the jury in every case.

Yet the Baldus study and petitioner’s arguments rest 
on the assumptions that Georgia juries invariably have 
an accurate and unambiguous understanding of the vic­
tim’s race— and that they ascribe significance to that in­
formation. We submit that such an assumption is in­
valid, providing further grounds for rejecting petitioner’s 
race-of-the-victim theory.

C. The Findings of the District Court on the Study’s 
Invalidity Should be Affirmed

In a thorough and painstaking analysis that warrants 
this Court’s careful attention, the trial court made con­
vincing first-hand findings that the Baldus study was 
riddled with errors in its data base and was not essen­
tially trustworthy; relied on statistical models which 
were not sufficiently predictive to support an inference 
of discrimination; and did not even compare like cases 
in purporting to find racially disparate impact. 580 
F.Supp. at 354-365.

For reasons not clearly expressed, the Court of Ap­
peals did not overtly pass judgment on these findings of 
fact. Instead, it chose to “ assume” the Baldus study’s 
validity and proceeded to hold that petitioner’s argu­



22

ments failed as a matter of law even given that assump­
tion. 753 F.2d at 894.

Contrary to petitioner’s disingenuous suggestions, how­
ever, the Court of Appeals in no way disturbed or ques­
tioned the trial court’s actual findings of the study’s in­
validity. Indeed, it expressly disclaimed any intent to do 
so. Id. at 894-95.

Under Fed. R. Civ. P. 52(a), the Court of Appeals 
could have set aside the district court’s findings of fact 
only if they were “ clearly erroneous.” United States v. 
General Dynamics, 415 U.S. 486 (1974). Obviously, the 
Court of Appeals did not do that in this case. So the 
trial court’s findings stand unimpeached.

Therefore, if this Court does not affirm the Eleventh 
Circuit’s holding on the legal issues, petitioner’s death 
sentence should still be affirmed on the ground that the 
Baldus study is too flawed and untrustworthy to raise 
a genuine issue of racially disparate sentencing. Given 
the manifest thoughtfulness and thoroughness of the dis­
trict court’s findings, there is no sound reason for this 
Court to avoid passing on whether they are clearly er­
roneous. And it would be a presumptuous appellate court 
indeed that would dismiss the trial court’s deliberate and 
painstaking demonstration of the study’s many palpable 
flaws as “ clearly erroneous.”

D. The Myriad Individualized Factors and Combina­
tions of Factors Which Influence A Death Sen­
tence Are Not Susceptible To Quantification Or 
Precise Comparative Analysis

Petitioner’s theory of discrimination is only as good as 
the precision and reliability of its base data, the predic­
tive capacity of its statistical models, and the essential 
equivalency of the cases it purports to compare. The dis­
trict court’s thorough scrutiny of the Baldus study pro­
duced unassailable findings that it is substantially de­
ficient in each of those critical aspects. 580 F.Supp. at



23

354-365.3 * 5 The study therefore fails to establish the 
factual predicate which is necessary even to reach peti­
tioner’s novel legal theory.

Putting aside the mere flaws, mistakes and inconsist­
encies of the study, amici would call the Court’s atten­
tion to what we consider to be a fatal and inherent fal­
lacy in petitioner’s methodology. Petitioner’s lawyers and 
“ experts” claim that they carefully recorded and ac­
counted for some 200 legitimate sentencing variables 
(e.g., various aggravating and mitigating factors) in at­
tempting to isolate the “ inexplicable” sentencing dis­
crepancies which they then blithely assigned to the race 
factor. The problem with this approach is that (a) they 
did not even thoroughly account for the factors which 
they claim to have accounted or “ controlled” for; and 
(2) the limited number of sentencing factors which they 
did choose to account for did not even begin to exhaust 
the vast range of legitimate sentencing variables (and 
combination thereof) which can result in a legitimate, 
non-discriminatory sentencing variation.6 * 8

One particular example of these fundamental flaws is 
illustrative but by no means exhaustive.

In demonstrating the numerous flaws infecting the 
data base of the Baldus studies, the district court found 
that the students who coded the various sentencing fac­
tors affecting each case were limited by the study’s 
structure to entering only one method of inflicting death. 
As the court found, 580 F.Supp. at 356:

3 Several professors or scholars who have a professional interest
in the acceptability of statistical studies as binding proof in litiga­
tion have filed a brief amicus curiae supporting the complete
validity of the Baldus studies. This Court should regard such
palpably self-serving arguments with maximum skepticism.

8 The district court expressly so found, 580 F.Supp. at 364: 
“ [The Baldus studies] do not account for a majority either of 
aggravating or mitigating circumstances in the cases.”



24

The effect of this would be to reduce the aggrava­
tion of a case that had multiple methods of inflict­
ing death. In coding this variable the students gen­
erally would list the method that actually caused the 
death and would not list any other contributing as­
saultive behavior. R463. [emphasis added].

The effect of such crude limitations on the accurate 
depiction of different capital cases can best be under­
stood by observing how they would apply to the coding 
of an actual case.

In Andrews v. Shulsen, 600 F.Supp. 408 (D.Utah 
1984), appeal pending, No. 84-2781 (10th Cir.), the de­
fendant and his accomplice murdered three people and 
brutally injured two others while robbing a Hi Fi shop 
in Ogden, Utah. The immediate cause of death in the 
murders was simply shooting. But before the fatal shoot­
ings, the defendants had (a) attempted to force the fa­
ther of one of the victims, at gunpoint, to pour poisonous 
liquid drain cleaner down the throats of his own son and 
two other bound teenage victims (he refused) ; (2) force- 
fed the poisonous drain cleaner to the hapless victims, 
then taped their mouths shut; (3) raped one of the teen­
age girl victims before methodically shooting her in the 
head; (4) attempted to strangle the father-victim with 
an electric cord; and (5) viciously kicked a long ball­
point pen deep into the father’s ear.

It is obvious from the district court’s findings that the 
Baldus study’s methodology would not begin to capture 
or account for all the hideous particulars and compounded 
variables which moved a Utah jury to vote for the death 
sentence in Andrews v. Shulsen. The cause of death 
would have been listed by the coders as a shooting (see 
580 F.Supp. at 356). Clearly, the collective horrors of 
such a case cannot be reduced to neatly coded vari­
ables in a statistician’s pigeon holes. This incapacity 
to capture the intangible but critical nuances of actual



25

murders undercuts the authenticity of all the study’s 
comparisons of supposedly similar cases.

As it turns out, the murderers in Andrews v. Shulsen 
were black and their victims were white. The perpetra­
tors in that case have appealed their death sentences, 
asserting the same claim of racially discriminatory sen­
tencing presented in the instant case. If petitioner pre­
vails here, the just death sentences of the likes of the 
“ Hi-Fi” murderers will be absurdly attributed to racial 
factors in the eyes of the law, rather than to the malici­
ous particulars which in fact account for them. Nothing 
in the Constitution or this Court’s capital punishment 
jurisprudence requires such an unreasonable and unjust 
result.

CONCLUSION
For all the foregoing reasons, the decision of the 

Eleventh Circuit should be affirmed.

Respectfully submitted,

Daniel J. P opeo 
George C. Sm ith  *

W ashington Legal F oundation 
1705 N Street, N.W.
Washington, D.C. 20036 
(202) 857-0240

Attorneys for Amici Curiae 
Washington Legal Foundation 

and
Allied Educational Foundation 

* Counsel of Record

Dated: September 19,1986