Folder
General - Attorney's Working Files, Jack Boger (Redacted)
Working File
November 3, 1982 - December 8, 1989
68 pages
Cite this item
-
Case Files, McCleskey Background Materials. General - Attorney's Working Files, Jack Boger (Redacted), 1982. e24712d8-6bcc-ef11-b8e8-7c1e520b5bae. LDF Archives, Thurgood Marshall Institute. https://ldfrecollection.org/archives/archives-search/archives-item/9d3dc5cb-e0fa-4e08-b900-625c8363a2d7/general-attorneys-working-files-jack-boger-redacted. Accessed December 05, 2025.
Copied!
MEMORANDUM
To: Anthony G. Amsterdam
From: Robert Nelson
Re: The discrimination claim in cases like McCleskv
Date: December 22, 1985
Questions Presented
[. Whether the jurv selection equal protection standard can be applied
outside of the jurv selection context?
11. Whether under the jurv selection standard there exists an affirmative
constitutional dutv bv public officials to prevent discriminatory results?
111. Whether the legislative history of 18 USC. 8242 suggests that the
framers of the Fourteenth Amendment did not contemplate that
discriminatorv purpose would be an essential requisite of equal protection
claims involving punishment?
IV. Whether the same standard utilized 10 make out a violation of 42 USC.
$1983 was originally intended 10 applv 10 Fourteenth Amendment
violations?
Introduction
This memorandum attempts to answer some of the questions raised
by my memorandum of October 20, 1985. It exclusively addresses the
traditional equal protection claim involving a showing of discrimination, as
opposed to the arbitrariness claim.
The first two questions presented focus on the jury selection standard.
The first question considers whether the requirement that a jury be a
representative body of a defendant's peers is the basis for treating jury
selection differently from other equal protection areas. If the representative
requirement is the basis for treating jury selection differently, then the jury
selection standard could not be applied to cases like McClesky where there
exist no fair representation requirement. To prove that this requirement is
not the basis for the jury selection standard. the memo traces the history of
the jury selection standard and concludes that the modern jury selection
standard has existed only since Castaneda v. Partida, while the requirement
of a representative jury has existed for one hundred years. This
representative requirement cannot. then. constitute the basis of the modern
jury selection standard. As noted in the first memorandum. the
justifications for the mosdern jury selection standard were set out in Davis,
and those justifications apply with ample force outside of the jury selection
context and to cases like McCleskv.
The second question considers the possibility of an affirmative
constitutional duty to prevent discriminatory results under the jury selection
standard. While there does in fact exist a constitutional duty "not to pursue
a course of conduct [that] operates to discriminate on racial grounds,” the
duty has been interpreted to mean, in effect, a duty not to discriminate
purposefully. Importantly, however, this duty has not been re-examined
2
since Castaneda and the onset of modern jury selection standard. Prior to
Castaneda, a showing of discriminatory purpose was required. [n the post-
Castaneda world which does not require a showing of discriminatory purpose
to make out an equal protection claim. it is possible that the duty today
could have more teeth to it.
The last two questions presented consider the the requirement of
discriminatory purpose in historical perspective. Neither the Civil Rights Act
of 1866, the statute upon [8 US.C. §242 is based. nor the Klan Act of [871],
upon which 42 US.C. §1983 -- the civl law counterpart of 18 US.C. 8242 -- is
based, expressly required a showing of willful discrimination to find a
statutory violation. [n fact, it is likely that something markedly less than
discriminatory purpose established an equal protection violation under the
Reconstruction statutes. As these statutes or their immediate predecessors
were enacted to enforce the Fourteenth Amendment. it may be possible to
infer that the framers of the Fourteenth Amendment also did not
contemplate that willful discrimination would be an essential requisite to a
prima facie equal protection claim.
I. Whether the jurv selection standard can be applied outside of the jurv
selection context?
If we apply the jury selection standard outside of the jury selection
context, we must grapple with the argument that the jury selection equal
protection standard is necessarily different from other equal protection
areas precisely because a jury must be composed of a representative body of
one's peers. Given this representative requirement, it makes good sense that
the standard would be less accepting of discriminatory results than other
equal protection areas where there exists no requirement of
3
representativeness. If it is true that the justification for the jury selection
standard is the need for representative juries, then the jury selection
standard most likely cannot be applicable outside of the jury selection
context: the standard's raison d'etre would be applicable only to jury
selection.
The history of the jury selection standard does not, however, support
this argument. The history suggests that while the requirement of a
representative jury has existed for 100 years. it has only been in the last 10
years, since Castaneda v. Partida, 450 US. 482 (1977). that the jury selection
standard has moved away {rom the purpose requirement characteristic of
other equal protection areas. The jury selection standard most likely
cannot, then. be due simply to the requirement of a representative jury. for
the requirement of a representative jury has existed long before a
substantively different equal protection standard attached to jury selection.
Since Strauder v. West Virginia, 100 US. 303 {1880), the Supreme
Court has recognized that a jury must be composed of "peers of equals of the
person whose rights it is selected or summoned to determine: that is. of his
neighbors. fellows, associates. persons having the same legal status in society
as that which he holds.” Id. at 308. Racial discrimination that results in the
systematic exclusion of otherwise qualified groups is at war with our basic
concepts of a democratic society.” Smith v. Texas, 511 US. 128. 150 (1940).
Such exclusion contravenes the very idea of a jury -- ‘a body truly
representative of the community.” Id!
| Wainwright v. Witt, 53 USL W. 4108 4121 (Brennan, J. dissenting on
other grounds); Rose v. Mitchell, 443 US. 545, 556 (1979); Tavior v.
Louisiana, 414 US. 522. S28 i19795); Carter v. Jury Commission, 396 US. 320,
330 (1970); Glasser v. US. 315 US. 60, 85 (1942).
Yet despite these expansive pronouncements about the need for a
representative jury which could certainly serve as the basis [or treating jury
selection differently from other equal protection cases, the Court did not.
prior to Castaneda, apply a markedly different equal protection standard to
jury selection. Equal protection violations were found where cases involved
either a showing of purposeful discrimination or a discriminatory jury
selection procedure. In Neal v. Delaware, 103 US. 370 (1881), for example,
the Court quashed the indictment of a black man by an all white jury from
which blacks had been excluded. No black had ever been summoned as a
juror although its black population exceeded 26,000 in a total population of
less than 150.000. [d at 377. This showing of disparate impact was,
however, accompanied by testimony from jury selection officers that they
had excluded blacks justifiably since blacks were “utterly disqualified. by
want of intelligence, experience, or moral integrity. to sit on juries.” Id.
In Smith v. Texas, 311 US. 128 (.1940), five out of 384 grand jurors
during the period from 1931-38 were black in a district in which 3.000 to
6.000 blacks met the legal qualifications for sitting on a jury. Asin Neal, the
Court also found evidence of purposeful discrimination to justify its finding
am equal protection violaiton. Even though the jury commissioners who
testified stated that they did not intentionally or systematically discriminate
against blacks, some commissioners also stated that they did nt know any
blacks and therefore would not select blacks. Id at 131-132. The Court held
that "discrimination can drige {rom commissioners who Know no negroes ds
well as from commissioners who know but eliminate them.” Id. at 152.
While the Court in Smith called this Kind of discrimination
‘ingenuous’ discrimination. id. the Court has characterized such
discrimination as ‘intentional discrimination. For example. in Cassel v.
Texas, 339 US. 282 (1950), a case similar to Smith involving disparate
impact and the defense by the jury commissioners of not knowing any
qualified blacks, the Court stated: "The statements of the jury commissioners
that they chose only whom they knew, and that they know no eligible
Negroes in an area where negroes made up [13.5%] of the population prove
the intentional exclusion that is discrimination in violation of petitioner's
constitutional rights.” Id. at 290.2 In these and other jury selection cases
more than disparate impact was required. “A purpose to discriminate must
be present...” Akins v. Texas, 316 US. 400, 403 (1945). In Fay v. New York,
332 US. 261 (1947). the Court, relying on Smith, held that a jury selection
equal protection claim cannot be established by a "mere showing” of
disparate pac but only by “a clear showing that [a group's absence on a
juryl was caused by discrimination.” Id. at 284.
In cases where an equal protection violation was found absent a
showing of discriminatory purpose. the jury selection procedure was found
to be unduly susceptible to abuse. See, e.g. Alexander v. Louisiana, 405 US.
625 (1972): Whitus v. Georgia, 385 US. 545 (1967); Avery v. Georgia, 345 US.
5359.
The Court could conclude. then. in Apodaca v. Oregon, 406 US. 404
(1972): "All that the Constitution forbids. however, is systematic exclusion of
identifiable segments of the community from jury panels and from the juries
ultimately drawn from those panels: a defendant may not. for example.
challenge the make up of a jury merely because no members of his race are
on the jurv. but must prove that his race has been systematically excluded.
2 The Court found an equal protection violation in Eubanks v. Louisiana, 356
US. S84 (1958), a similar case in which the jury commissioners attributed the
disparate impact on ‘local tradtion.”
Id. at 413, citing Swain v. Alabama, 380 US. 202, 208-209 (1965); Cassel v.
Texas, 339 US. at 286-287; Akins v. Texas, 316 US. at 403-404.
These cases suggest that the jury selection standard prior to
Castaneda was not very different from traditional equal protection analysis
outside of the jury selection context.3 Discriminatory purpose or a
discriminatory procedure alongside a discriminatory result was a
prerequisite to a prima facie equal protection claim. So despite the
importance of a representative jury of ones peers recognized first in
Strauder and continually since. the cases demonstrate that the desire for a
representative jury did not lead to a jury selection standard that did not
require a showing of discriminatory purpose. Thus any justification for
today treating jury selection differently from other equal protection
violations cannot rely solely. if at all. on the need for a representative jury of
One's peers.
The reasons for the change in the standard were, ironically.
established in Washington v. Davis, 426 US. 229 (1976). the case that firmly
established the purpose requirement outside of the jury selection context,
and were first implemented in Castaneda. In Castaneda, the Court expressiv
relied on Washington v. Davis and Arlington Heights to show that an equal
protection violation had been shown. 430 US. at 493-494. According to
Davis, jury selection should be treated differently because |) systematic
exclusion of blacks from a jury is “itself such an equal application of the law .
_. as to show intentional discrimination”; 2) and the discrimination is very
difficult to explain on other grounds.” 426 US 24[-242. Based solely on
3 But see Peters v. Kiff, where justice White, concurring, stated that Congress
put exclusions from jury service on grounds of race "in a class by
themselves.” 407 US. at 505. See infra at 20.
these justifications -- no mention was made in Castaneda or Davis about the
requirement of a representative jury as being a justification for a different
jury selection standard -- the Court in Castaneda fashioned the modern jury
selection rule. First, the person must be 2 member of a group that is a
recognizable class and is singled out for different treatment by application of
the law; second, the degree of underrepresentation must be substantial and
must be demonstrated by comparing the proportion of the group in the total
population to the proportion cdiled to serve as grand jurors, over a
significant period of time; third. the Court required that the selection
procedure be susceptible to abuse or that it be racially non-neutral. 430 US.
at 494. The Court concluded that "{olnce the defendant has shown
substantial underrepresentation of his group. he has made out a prima facie
case of discriminatory purpose.” Id. at 495. In Castaneda, perhaps for the
first time, the Court found an equal protection violation without finding
discriminatory purpose by public officials or a jury selection procedure that
was flagrantly subject to abuse.
Because the Davis justifications for treating jury selection differently
and the Castaneda requirements for making out a prima facie claim all apply
to the facts of McClesky,* it 1s possible to argue that the jury selection equal
protection standard can apply outside of the jury selection context and to
cases like McClesky.
II. Whether under the jurv selection standard there exists an affirmative
constitutional duty bv public officials to prevent discriminatorv results?
The jury selection standard does provide some basis upon which it can
be argued that the Equal Protection Clause requires public officials to
1 See my memorandum of October 20, 1985. at 14-29.
exercise a duty to prevent discriminatory results. The origins of this
possible affirmative duty are in Smith v. Texas 311 US. 128 (1940), where
jury commissioners testified that they did not select any blacks to sit on a
grand jury because they did not know any blacks.> The Court found that
where jury commissioners limit those from whom grand juries are selected
to their own personal acquaintances. ‘discrimination can arise from
commissioners who know no negroes as well as from commissioners who
know but eliminate them. Id. at 1532. The Court concluded: "If there has
been discrimination, whether accomplished ingeniously or ingenuously, the
conviction cannot stand.” Id That the jury commissioners denied that they
intentionally. arbitrarily or systematically discriminated against black jurors
was irrelevant in light of the “ingenuous” discrimination. [d. Smith may be
the first case where an equal protection violation was found absent a
showing of actual discriminatory purpose.
In Hill v. Texas, 316 US. 400 (1942). jury commissioners testified that
they did not know any qualified blacks who could serve as jurors. Yet they
had made no effort to ascertain whether any blacks were qualified to serve
as jurors. According to the Court, the jury commissioners failed to perform
their constitutional duty -- recognized by §4 of the Civil Rights Act of March
5 See supra at 5-6.
{18756 and fully established since our decision in 1881 in Neal v. Delaware,
103 US. 370 (1881) -- not to pursue a course of conduct in the
administration of their of fice which would operate to discriminate in the
selection of jurors on racial grounds.” Hill, 316 US. at 404. Therefore, the
Court found, "no State is at liberty to impose upon one charged with crime a
discrimination in its trial procedure which the Constitution and an Act of
Congress passed pursuant to the Constitution, alike forbid.” Id. at 406. The
essence of this duty in Hill seems to mean a duty not to discriminate
purposefully on the basis of race or to utilize procedures which necessarily
discriminate -- hardly helpful for equal protection claims like McClesky's
that involve far more subtle forms of discrimination. The Court in Hill
concluded that "it is our duty as well as the State's to see to it that
throughout the procedure for bringing [the defendant] to justice he shall
enjoy the protection which the Constitution guarantees.” Id. at 406. While
framed in the affirmative. this definition of the duty simply begs the
question, for its meaning is dependent on what the Constitution “guarantees.”
6 18 Stat. 336, now codified at 18 US.C. §243. This section provides: No
citizen possessing all other qualifications which are or may be prescribed by
law shall be disqualified for service as a grand or petit juror in any court of
the United States, or of any state on account of race, color, or previous
condition of servitude: and whoever, being an officer or other person
charged with any dutv in the selection or summoning of jurors, excludes or
fails to summon any citizen for such cause, shall be fined no more than
$5.000." It should be noted that this section, like the other Reconstruction
statutes noted below in questions III and IV of this memorandum, did not
expressly contain a requirement of purpose. The statute since its inception
has, however, only applied to the situation where discriminatory purpose
has been shown. See eg. Neal v. Delaware 103 US. 370 (1881), where a
showing of extreme disproportionate impact was found (no blacks ever on a
jury) alongside the testimony of jury selection officers who stated that they
had excluded blacks justifiably since blacks were “utterly disqualified. by
want of intelligence, experience, or moral integrity, to sit on juries.” Id. at
377.
10
More importantly, if the duty is simply a duty not to do what the
Constitution forbids, then the duty does not add any substantive limits to the
standard. :
The nature of this constitutional duty was next considered in Cassel v.
Texas, 339 US. 282 (1950). a similar case involving jury commissioners who
claimed that they knew no qualified black jurors. Here the Court required
jury commissioners to take affirmative steps to prevent discrimination:
"When the commissioners were appointed as judicial administrative officials.
it was their duty to familiarize themselves fairly with the qualifications of
the eligible jurors of the county without regard to race or color.” Id at 289.
As noted in Question | above, this kind of discrimination was characterized
as ‘intentional’ discrimination. Id. at 290. So here, too, it seems the duty to
prevent discrimination Kicked in because the selection procedure was so
obviously flawed that an inference of purposeful discrimination had to be
drawn. [t's important to note that the Court required the jury
commissioners take affirmative steps to correct the procedure.
This duty was found alongside slightly different facts in Alexander v.
Louisiana, 405 US. 625 (1972). In Alexander. the Court found that at two
separate points in the jury selection procedure racial identfications were
visibly attached to questionnaires; at these two points blacks were
disproportionately excluded. and no blacks were on the grand jury that
indicted the defendant.” While the progressive decimation of potential black
grand jurors was “striking.” the Court did not find a prima facie violation
7 In Alexander, 21% of the parish was black; 14% of the blacks responded to
a questionnaire from which jurors were selected. After the two culling out
procedures when racial identifications the commissioners had attached to the
forms were visible, the pool was reduced to 400, of whom 27, or 7% were
black. It was from this pool that the grand jury was selected.
11
based on the procedure. Id. at 630. Instead the Court engaged in a factual
inquiry that took into accout “all possible explanatory factors.” Id. at 630.
The Court noted that the ‘result bespeaks discrimination, whether or not it
was a conscious decision on the part of any individual jury commissioner.”8
Id. at 632. The Court then went on to quote verbatim fom Hill about the
duty of jury commissioners "not to pursue a course of conduct ... which
would operate to discriminate ...~ Id. The Court concluded by quoting
from Whitus v. Georgia, 385 US. 545 (1967). noting that the "opportunity for
discrimination was present and [that it cannot be said] on this record that it
was not resorted to by the commissioners.” Id. at 632. Here there was an
opportunity to discriminate and a duty was found not to discriminate. Based
on the record, the Court concluded that that duty was violated.
In Alexander, then. the “course of conduct” did not involve the
"ingenuous” and obviously discriminatory action reminiscent of Hill, Smith
and Cassell. Yet the jury selection procedure in Alexander was structurally
flawed. That fact, alongside the disparate impact, proved determinative.
Surely we would have a hard time showing that Georgia's death statute is
similarly structurally flawed. Still. however, the administration of the death
penalty is subject to abuse; indeed. the race of the victim of the capital
crime is apparent to public officials involved in the death sentencing
procedure. This fact. alongside a “result [that] bespeaks discrimination.” may
8 The Court was quoting here from Hernandez v. Texas, 347 US 475, 482
(1954).
constitute a sufficient factual basis to violate a constitutional duty that may
exist to prevent discrimination.’
So not unlike Alexander. then, the discrimination in McCleskv is not
purposeful in the Hill sense, and the discriminatory result is based upon a
statistical showing of disparate impact alongside a procedure that is at least
to some degree susceptible to abuse. Thus it is possible that the requisites
of a prima facie violation of this duty may exist in McClesky. Yet because
no court would find Georgia's death statute structurally defective to the
same degree as the jury selection utilized in Alexander, Whitus, or Avery,
the task in McClesky is to argue that the duty is to prevent discriminatory
results. and that duty is violated when any public conduct leads to
discriminatory results. Surely this reading is consistent with the plain
language of the duty. But if the duty is violated only by flagrantly
discriminatory actions or by procedures flagrantly subject to abuse -- a
9 It should be noted that in Peters v. Kiff, 407 US. 493 (1972), Justice White,
concurring, quoted extensively from Hill. Id. at 506. He stated that where
jury commissioners disqualify citizens on the grounds of race, they fail to
perform their constitutional duty...” Id. Thus he implied that the duty is
simply not to discriminate intentionally. It is not known in what fact pattern
he would invoke the duty in Peters because the Kind of discrimination was
not made known in the case. (The issue was whether a white defendant
could suffer a Fourteenth Amendment violation if blacks were systematically
excluded from the jury, not whether he did in that particular case. The
evidence of discrimination was never presented.)
13
reading which is entirely consistent with the cases -- then the duty would
not be violated in McClesky.!0
One explanation why this duty does not seem to be very helpful in
establishing an equal protection claim absent a showing of purpose is that
the modern jury selection equal protection standard that does not require a
strong showing of discriminatory purpose or an obviously discriminatory
procedure did not exist prior to Castaneda v. Partida. As noted in Question |
above, the jury selection standard prior to Castaneda was not all that unlike
the equal protection standard outside of the jury selection context. Read in
light of this history, the duty recognized in Hill, Cassel, and Alexander
becomes important in the evolution of the standard; these are the first cases
that began to compromise the purpose standard, if only marginally. It could
be that the duty only marginally changed the jury selection standard
because the standard was constrained within the old purpose framework.
Now that that framework has been rejected in jury selection, perhaps it is
possible to argue that the duty requirement should today be read to limit
further constitutional actions under the modern standard. If the duty
requirement limited constitutional actions under the jury selection standard
prior to the Castaneda standard. why then should it not do the same in the
post Castaneda world -- i.e. limit constitutional actions so that the duty is
10 [n light of the possible applicability of this duty requirement to McClesky.,
it is important to note that the ingenious/ingenuous distinction has been
utilized outside of the jury selection context. In Cooper v. Aaron, 358 US. 1.
17 (1958), the Court noted, "In short, the constitutional rights of children not
to be discriminated against in school admission on grounds of race or color . .
. can neither be nullified openly and directly by state legislators or state
executives or judicial officers, nor nullified indirectly by them through
evasive schemes for segregation, whether attempted ingeniously or
ingenuously.” Again, however, as in Smith, discrimination was inferred from
egregious facts.
14
violated by the existence of discriminatory results. The Court in Castaneda
did not say that the standard it enunciated reached the constitutional limits
of the jury selection standard. Perhaps. then. we could argue that the
constitutional duty not to pursue a course of conduct that operates to
discriminate on racial grounds should be read today to mean a duty to
prevent discriminatory results. Of course we are on very shaky legal turf
here.
[11. Whether the legislative history of 18 USC. 824211 suggests that the
framers of the Fourteenth Amendment did not contemplate that
discriminatory purpose would be an essential requisite of equal protection
claims involving punishment?
§2 of the Civil Rights Act of 1866.!2 made it a crime for any person
under color of any law to “subject, or cause to be subjected [freed blacks] to
... different punishment.” Senator Trumbull, chairman of the Senate
Judiciary Committee which reported the Civil Rights Bill, stated that the Bill's
purpose was to "to protect all persons in the United States in their civil
It 18 USC 8242 provides: Whoever, under color of any law, statute,
ordinance, regulation, or custom. willfully subjects any inhabitant of any
State, Territory, or District to the deprivation of any rights, privileges, or
immunities secured or protected by the Constitution or laws of the United
States, or to different punishments, pains, or penalties, on account of such
inhabitant being an alien or by reason of his color, or race, than are
prescribed for the punishment of citizens, shall be fined not more that
$1.000 or imprisoned not more than one year, or both; and if death results
shall be subject to imprisonment for any term of years or for life.” Pub L.
90-234
1214 Stat. 27.
rights, and furnish means of their vindication.”!3 It would, according to
Trumbull, destroy the discrimination against blacks in the laws of the
Southern states and enforce the 13th Amendment.!® In 1870. after the
enactment of the Fourteenth Amendment, the statute was amended by 8817
and 18 of the Act to Enforce the Provisions of the Fourteenth Amendment of
May 31, 1870.15 The clause "subject or cause to be subjected” remained in
the statute. The prohibition against the "deprivation of any rights,
I3 Cong. Globe, 39th Cong., 1st Sess. p. 211.
14 Flack, The Adoption of the fourteenth Amendment (1908) at 21. The Civil
Rights Act of 1866 was hotly debated and was passed over two Presidential
vetoes. Many thought that it was unconstitutional, as it usurped the powers
of the states. Among those Republicans who objected most strongly to the
Bill was Thaddeus Stevens. See id.
15 16 Stat. 14. The statule provided, in pertinent part: "That any person
who, under color of any law, statute, ordinance, regulation, or custom, shall
subject, or cause to be subjected, any inhabitant of any State or Territory to
the deprivation of any right secured or protected by the last preceding
secotion of this act.m or to differnt punishment, pains, or penalties on
account of such person being an alien, or by reason of his color or race, than
is prescribed for the punishment of citizens, shall be deemed guilty of a
misdemeanor. and. on conviction. shall be punished by fine not exceeding
one thousand dollars, or imprisonment not exceeding one year, or both, in
the discretion of the court.’ ;
The preceding section referred to provided: "That all persons within
the jurisdiction of the United States shall have the same right in every State
and Territory in the United States to make and enforce contracts, to sue, be
parties, give evidence, and to the full and equal benefit of all laws and
proceedings for the security of person and property as is enjoved by white
citizens. and shall be subject to like punishment, pains penalties, taxes,
licenses, and exactions of every Kind, and none other, any law, statute,
ordince, regulation, or custom to the contrary notwithstanding. No tax or
charge shall be imposed or enforced by any State upon any person
immigrating thereto from a foreign country which is not equally imposed
and enforced upon every person immigrating to such State from any other
foreign country; and any law of any State in conflict with this provision is
hereby declared null and void.” (emphasis supplied)
16
privileges, or immunities, secured or protected by the Constitution and laws
of the United States” was added in 1874.18 Like its predecessors. the 1874
revision was applicable to any person who under color of law “subjects. or
causes to be subjected” any inhabitant to the deprivation of any rights, etc.
[t was not until 1909 that the requirement of a “willful” violation was
introduced to the Criminal Code.!” "Willfully” was added to the Revised
Statutes §5510. according to Sen. Daniel. to make the section “less severe.'!8
Other than this statement by Sen. Daniel. [ have come across no other
legislative history that helps explain why Congress made this express change
in the Criminal Code. Neither the Senate!? nor House Reports2? of the bill
provide any insight into the basis for the change to willfulness. The Senate
Report did not even recommend adding the willfulness requirement into
this section of the Code 2!
Still, the fact that an express requirement of willfulness was added
more than 40 years after the law was first passed, and that the change in
the statute was to make the law “less severe,” suggest that the law was more
severe between [866-1909 -- i.e that it was easier for persons to be
prosecuted under the statute. During this period, it is likely that a cause of
action could have been contemplated under the statute without showing a
1& Revised Statutes $5510, at 1068.
17 Act of March 4, 1909, 35 Stat. 1092.
18 43 Cong. Rec., 60th Cong.. 2d Sess., p. 3599 (March 2, 1909). The statute
was then codified at 18 USC 1940 ed. 852.
19 Senate Report 10, Criminal Code Bill for Revision, S. 2982, 60th Cong., 1st
Sess.
20 House Report 2, Criminal Code Bill for Revision, HR. 11701, 60th Cong., Ist
Sess.
21 The only change the Senate committee recommended was to change
"Every person who under color of state law... "to "Whoever under color of
any law..."
17
willful violation of the law. As the 1870 Act was expressly enacted to
enforce the protections afforded by the Equal Protection Clause, it may be
possible to infer that the framers of the Fourteenth Amend ment also did not
contemplate that a showing of willful or purposeful discrimination was
required to make out a prima facie equal protection claim. A similar, more
“severe” standard, may have been intended.
This interpretation is not inconsistent with Screws v. United States
3295 US. 91 (1945). where the Court attempted to define the willfulness
requirement in §20 of the Criminal Code.22 There, §20 of the Criminal Code
was attacked as unconstitutionally vague.23 Justice Douglas argued that the
requirement of a "specific intent to deprive a person of a federal right”
would avoid those consequences which may otherwise render a vague or
indefinite statute invalid. [d. at 101-103. Such a requirement “saves” the
Act from any charge of unconstitutionality on the grounds of vagueness. [d.
at 103.
However, Justice Douglas found that this construction was consistent
with the history of the statute. This was true, he found, since the word
“willfully” was added to make the section’less severe.” 1d. Prior to 1909,
Congress intended that one who deprived a person of any right protected by
the Constitution should be liable "without more.” [d. This was the pattern of
criminal legislation which has been sustained without any charge or proof of
22 §20 of the Criminal Code was codified as 18 US.C. [940 ed.. §52 [Derived
from RS. 85510]. In 1948, 18 US.C. §242 was enacted.
23 The claim was made that a local law enforcement could violate §20 and
commit a federal offense if he does an act which some court later holds
deprives a person of due process of law or equal protection. Those who
enforced local law today might not know for many months whether what
they did deprived someone of due process of law. The enforcement of a
criminal statute so construed would cast law enforcement agencies "loose at
their own risk on a vast uncharted sea.” Id at 97-98%.
18
scienter. 1d.24 The Court concluded that since the section was amended to
be "less severe,” it is permissible to infer that that its severity was to be
lessened by making it applicable only where the requisite intent to deprive a
person of a federal right was present. 1d. at 103. Discriminatory purpose
was not a requisite of the statutory claim. Thus it would appear that
something broaching an effects or results based standard existed prior to
1909.
Justice Douglas did note that prior to the willfulness requirement, §20 .
"would have been susceptible” to this more severe interpretation. But he
added that the Equal Protection Clause was not susceptible to such an
interpretation prior to §20 since “purposeful discrimination’ must be
shown.” Id. at 103, quoting Snowden v. Hughes, 321 US. 1, 8-9. [tis
significant for us that his support for this claim is prior case law as opposed
to legislative history. Justice Douglas is not, | don't think, suggesting that
the framers of the Fourteenth Amendment had something different in mind
for the due process or privileges and immunities clauses. Nor is there any
indication that Congress intended this statute to enforce different provisions
of the Fourteenth Amendment differently. Most probably a unitary
constitutional standard was contemplated, and it was solely judicial
interpretation that made the equal protection standard a less severe
standard. Thus despite this statement by Justice Douglas in Screws, the
legislative history of 18 US.C. §242 may still be helpful in understanding
what standard the Reconstruction Congresses had in mind for the Fourteenth
Amendment. [tis at least an open question whether the lack of scienter
standard was intended to apply to the Equal Protection Clause.
24 The Court cited Shevlin-Carpenter Co v. Minnesota, 218 US. 57; US. v.
Balint 2358 US. 250
19
That the Civil Rights Act of [866 was successively amended in the
1870s after the enactment of the Fourteenth Amendment tends to suggest
that rigorous scrutiny should attach to Fourteenth Amendment violations.
Simply the fact of the statutes is important. Reconstruction Congresses were
apparently so concerned about the problem of racially discriminatory
punishment that they considered it necessary 10 enact a criminal statute to
do ostensibly what the Constitution set out to do. If punishment were so
important an issue, it could be that when the discrimination involved
punishment, the framers of the Fourteenth Amendment thought a more
severe standard was necessary to guard against even the result of unequal
punishment. This would have meant that something less than a
discriminatory purpose was required.
This reading of the statutes is perhaps also consistent with the way
some would argue the jury selection standard has evolved. The jury
selection statute, 18 US.C. §24325, which came out of the Reconstruction era,
has. arguably, had something to do with the more severe standard that jury
selection now enjoys in the equal protection context. In Peters v. Kiff, 407
US. 493 (1972). Justice White, concurring, noted that "{bly this unambiguous
provision, now contained in 18 U.S.C. 8243, Congress put cases involving
exclusions from jury service on grounds of race in a class by themselves.” He
then quoted from Fay v. New York, 332 US. 261, 282-283 (1947): "For us the
majestic generalities of the Fourteenth Amendment are thus reduced to a
concrete statutory command when cases involve race or color which is
wanting in every other case oi alleged discrimination.” In light of this,
perhaps we should argue that Congress also wanted discrimination in
punishment to enjoy 4 more severe equal protection standard; that
253 See supra note 6.
command, we should argue, can only be met today by a standard that
dispenses with the purpose requirement 2®
Thinking along these lines taps in well with #3 of Eric Multhaup's
memorandum of November 25, 1985. He argues there that in the post-Gregg
universe, courts should be particularly “alert,” rather than “deaf” to counter-
currents of applied discrimination in the otherwise unobjectionable capital
statutes. If the framers considered discrimination in punishment so
important an issue that an additional statute was necessary to enforce the
Constitution, then such claims of discriminatory punishment should be held
to the highest level of scrutiny. This arguably would compel the Equal
Protection Clause to ensure against the situation where public officials
simply condone the discriminatory implementation of the death penalty.
Certainly drawing analogies from 18 US.C. §242, a criminal statute. to
the Equal Protection Clause is a strained exercise. Yet this history should
not be altogether discounted. It suggests that a statute which was meant Lo
enforce the Fourteenth Amendment had a more severe standard than the
usual criminal standard of purpose. The [act that 18 USC. 8242.15 a
criminal statute does not seem problematic. {tis in the criminal context
where the least severe standard of discrimination would apply: indeed, the
fact that §20 was a criminal as opposed to a civil statute was a major factor
in Justice Douglas decision to require a showing of specific intent under the
statute. Perhaps we can infer from this that the framers intended the
Fourteenth Amendment -- which also attempts to prevent discrimination in
punishment -- to have a qualitatively more severe standard, one that goes
26 [t is important to note that this argument is inconsistent with the
argument noted in Question | that the modern jury selection standard
evolved from Davis and was {irst implemented in Castaneda, and that it had
nothing to do with Congressional statutes.
21
beyond a purpose requirement and approaches a duty on public officials to
prevent even the effects of discrimination.
1V. Whether the same standard utilized to make out a violation of 42 USC.
§1983 was originally intended to apply to Fourteenth Amendment
violations?
42 US.C. 5198327 is the civil law counterpart to 18 US.C. §242.
Although 42 US.C. §1983 does not expressly refer to punishment, it provides
a civil remedy to those who have been deprived of rights secured by the
Constitution or laws of the United States. In Parratt v. Taylor, 451 US. 527
(1981). the Court noted that "42 US.C. §1983, unlike its criminal counterpart
[18 US.C. 8242]. has never been found by this Court to contain a state-of -
mind requirement,” citing Monroe v_ Pape, 365 US. 167 (1961). In Monroe,
the Court distinguished Screws and the finding there of a specific intent to
deprive a person of a federal right under 18 US.C. $242 because the word
"willfully" does not appear in 42 US. C. §1983 and because §1983 provides a
civil and not a criminal remedy that could be challenged on vagueness
grounds. Monroe, 365 US. at 187. Instead of a state-of -mind requirement,
31983 “should be read against the background of tort liability that makes a
man responsible for the natural consequences of his actions.” Id. at 187.
Under §1983, then, public officials who could reasonably forsee that their
actions could result in discrimination would be liable for their discriminatory
27 42 USC. 81983 provides: "Every person who, under color of any statute,
ordinance, regulation, custom, or usage. of any State or Territory or the
District of Columbia, subjects, or causes to be subjected, any citizen of the
United States or other person within the jurisdiction thereof to the
deprivation of any rights. privileges, or immunities secured by the
Constitution and laws, shall be liable to the party injured in an action at iaw,
suit in equity, or other proper proceeding for redress.
actions. Perhaps this standard could be construed to mean that public
officials have a duty to prevent discrimination if a discriminatory result is
the forseeable consequence of their actions.
In Monroe, the Court traced the history of the42 US.C. §1983 before
concluding that a tort liability standard, as opposed to an intent standard.
was appropriate. The statute was first enacted as §1 of the Klan Act of April
20. 1871.28 This statute. like the Enforcement Act of 1870, was passed
pursuant to §5 power of the Fourteenth Amendment. Sen. Edmunds. chair of
the Judiciary Committee that reported the Bill. noted that the first section of
the Act “is merely carrying out the principles of the Civl Rights Bill which has
since become a part of the Constitution.” Indeed, the Act was also known as
"An Act to Enforce the Provisions of the Fourteenth Amendment to the
Constitution.” Thus this statute, like the precursors of 18 US.C. 8242, was
meant expressly to enforce the Equal Protection Clause, and statute never
required a showing of discriminatory purpose. Also, according to Justice
Douglas. its aim was to provide a federal remedy where the state remedy,
though adequate in theory, was not available in practice; id. at 174: the
statute was not intended to create new rights or remedies (other than
providing a federal forum. but only those which the Fourteenth Amendment
already contemplated.
Given these two facts -- i.e. that the statute was meant 10 enforce the
provisions of the Fourteenth Amendment, and that it was not intended to
create new rights -- it would not be unreasonable to argue that the
standards of proof contemplated under the Constitution and statute. laws
passed with three years of one another, were similar if not identical. As the
Court in Monroe found that $1983 has ‘never’ been found to have a state-of -
23 [7 Stat. 13.
mind requirement, perhaps it is possible also to infer that a state-of -mind
requirement was not intended by the framers to attach to Fourteenth
Amendment violations.
A review of the legislative history of the 1871 Act does suggest that
something less than a discriminatory purpose could satisfy a statutory claim.
Congresspersons who spoke about the statute focused on the impact of the
discrimination. and not the discriminatory purpose. For example, Burchard
of Illinois stated:
But if the statutes show no discrimination, yet in its judicial tribunals
one class is unable to secure that enforcement of their rights and
punishment for their infraction which is accorded to another, or if
secret combinations of men are allowed by the Executive to band
together to deprive one class of citizens of their legal rights without a
proper effort to discover, detect, and punish the violations of law and
order, the state has not afforded to all its citizens equal protection of
the laws.2?
Rep. Beatty of Ohio stated:
[Mlen were murdered, houses were burned, scourged, and officers of
the law shot down: and the State made no successful effort to bring
the guilty to punishment or afford protection or redress to the
outraged and innocent. The State, from lack of power or inclination,
practically denied the equal protection of the [aw to these persons.30
Again echoing the equal protection theme, Mr. Hoar of Massachusetts stated:
Now, it is an effectual denial by a State of the equal protection of the
laws when any class of officers charged under the laws with their
administration permanently and as a rule refuse to extend that
protection. If every sheriff in South Carolina refuses to serve a writ
for a colored man and those sheriffs are kept in office year after vear
by the people of South Carolina, and no verdict against them for their
29 Cong. Globe, 42d Cong ist Sess at 31).
30 Id. at 428.
failure of duty can be obtained before a South Carolina jury. the State
of South Carolina, through the class officers who are its
representatives to afford the equal protection of the laws to that class
of citizens, has denied that protection.3!
Mr. Lowe of Kansas said, "While murder is stalking abroad in disguise, while
whippings and lynchings and banishment have been visited upon
unoffending American citizens, the local administrations have been found
inadequate or unwilling to apply the proper corrective."32
When public officials did not, for whatever reason, vigorously enforce
the Constitution or the laws of a state, these Congresspersons contemplated
that those officials would be liable to the victims of discrimination. They
contemplated a duty of officials to guard against discriminatory application
of the laws, under both the 1871 Act and the Fourteenth Amendment.
Justice Douglas concluded after reviewing this legislative history: "It is
abundantly clear that one reason the legislation was passed was to afford a
federal right in federal courts because, by reason of prejudice, passion,
neglect, intolerance or otherwise, state laws might not be enforced and the
claims of citizens to the enjoyment of rights, privileges, and immunities
guaranteed by the Fourteenth Amendment might be denied by the state
agencies.” Id. at 180. He also noted the Act was meant as a remedy against
State officials who were "unable or unwilling to enforce state law.” Id. at 176
(emphasis in original). Discriminatory purpose was not an essential
requisite to establish a statutory violation. Indeed, if the state were, for
whatever reason, ‘unable’ to prevent discrimination, the legislative history
of the statute strongly suggests that the state would be liable under the
statute.
In light of this history, and the de-emphasis of a requirement of
discriminatory purpose, it is important to remember that the Act was meant
expressly to enforce the Fourteenth Amendment. [ndeed, the
Congresspersons who spoke to the Act considered it in relation to the Equal
Protection Clause, and nowhere did they distinguish between standards
necessary to establish either an equal protection violation or a violation
under the Act. Perhaps we can conclude. therefore, that a requirement of
discriminatory purpose under the Equal Protection Clause was not
contemplated by the framers of the Fourteenth Amendment.
Conclusion
Application of the jury selection standard outside of the jury
selection context and to cases like McClesky ought not be rejected simply
because the requirement of representativeness is unique to jury selection.
The requirement of a representative jury has been around for one hundred
years. and it has only been within the last ten years since Castaneda that a
gualitatively different equal protection standard attached to jury selection.
The basis for the change in the standard was not rooted in the
representative requirement.
It is possible, then, that the constitutional duty recognized under the
jury selection standard not to pursue a course of conduct that operates to
discriminate on racial grounds could also appiy outside of jury selection
context. As interpreted, this duty seems to have added little to the pre-
Castaneda standard beyond making ‘ingenuous’ forms of discrimination
unconstitutional. For the duty to be applicable to cases like McClesky, it
must be construed to prevent {ar more subtle forms of discrimination. One
26
way to do this is to argue under the post-Castaneda standard. the duty
should jnterpreted to be, like the standard, far more expansive, so that it
works to ensure that public officials act to prevent discriminatory results. |
am not persuaded by this argument.
The Reconstruction statutes strongly support the possibility that the
framers of the Fourteenth Amendment did not contemplate that a showing
of discriminatory purpose would necessarily accompany successful equal
protection claims. Importantly, the precursors of both 18 USC 8242 and 42
US.C 51983. most of which were enacted expressly to enforce the provisions
of the Fourteenth Amendment, did not require a showing of discriminatory
purpose to make out a statutory violation. The fact that Congress passed
several criminal laws specifically designed to address the problem of
discriminatory punishment also suggests that Congress intended that a
severe standard ought apply to claims of unequal punishment. Certainly the
Fourteenth Amendment was also aimed at stamping out discrimination in
punishment. and it is not unreasonable to infer that a similarly severe
standard was contemplated under the Equal Protection Clause.
27
November 3, 1982
The Honorable J. Owen Forrester
United States District Judge
2367 United States Courthouse
Atlanta, Georgia 30303
Re: Warren McCleskey v. Walter D. Zant
Civ. Action No. C81-2434A
Dear Judge Forrester:
I am writing as co-counsel for Warren
McCleskey to request a continuance of several
weeks in the evidentiary hearing presently
scheduled in McCleskey's case for November 19,
1632. .Counsel would nol suggest a delay inthis
case except with the greatest reluctance and
under the unusual circumstances set forth
below:
(1). 'Y have four other commitments 1in
capital cases between now and November 19th,
including (a) a successive state habeas corpus
hearing on November 8, 1982 in John Smith/Machetti
v. Zant in the Superior Court of Butts Counly;
(b) a further evidentiary hearing on November 16,
1982 in House v. Balkcom in The United States
Digtrict Court for the Northern District of
Georgia; (co) an opening brief due November 22,
1082 in ‘Ross v. Hopper in the United States
Court of Appeals for the Eleventh Circuit; and
(d) an opening brief due November 22, 1982 in
Spencer v, Zant, alsco'in the Eleventh Circuit.
id) @n light of my other commitments
cutlined above, it will be extremely difficult,
if not impossible, for me to be adequately prepared
to examine Professor Baldus and other expert wit-
nesses on the extremely complex and extended social
scientific and statistical evidence they have
assembled. Although possessed of a general famil-
jarty with their studies, their methodologies, and
their conclusions, I fear that without sufficient
+} 0 COLUMBUS CiRCILE fz 12) B'3S-8397 N EV Y OR W., NY 10019
The Honorable J. Owen Forrester
Page. 2.
November 3, 1982
time to educate myself on the significant details
of the analysis which are necessary to provide a
complete record on this issue, my presentation may
compromise the clarity and comprehensiveness of
Professor Baldus' findings.
(111) Finally, I note that the hearing
is scheduled to begin on the Friday before
Thanksgiving. It is certain that our direct case
will require more than one day's presentation, and
lt is possible that if will require three to five
days of direct testimony. Both for the Court and
for out-of-state witnesses, an uninterrupted hearing
during successive days might be a more convenient way
to proceed. While this consideration alone would, of
course, not be controlling, nevertheless in conjunction
with the points sel forth above it is apn additional
factor to be considered.
Thank you very much for considering
this request. Best regards.
John Charles Boger
JCB:agf
10 COL IMB US CrRICLE {212}:586-83897 NEWYORK, N.Y P90 195
porihain 1 Ruosell odin - Rdonéa. 04 © Si UL
| lly Fig - Bllitis
2 Coutts Prod Liber S Fhe.
hata pio. fia. deo ameweud_ Alana dunp tillers 4]
Dix frumitucdon in Wank, L) 4 2 otro wes gry b
Us. IY lniehia Jeweey gore y Clssged LT - LA <a
A & Chet cho Slt hit prot ign & glimed of.)
A
had sm suth & Gum in a BA: econ, [ze he Perko
fd thee ll bo occ rei Le
(3) Change mm pussisiihin 8 witot—= y y 7 lr 24 :
165, Chie Wet bin od uo (rom ind gti a.
(5) Chang, pum ump lormite f Uae 4 oth 2 (hl endo
) "w' J " ed
g
National Officers
William T. Coleman, Jr.
Chairman of the Board
Robert H. Preiskel
President
Barrington DD. Parker, Jr.
Vice President
Harriet Rabb
Secretary
Eleanor S. Applewhaite
Treasurer
Executive Officers
Julius LeVonne Chambers
Diirector-Counsel
James M. Nabrit HI
Associate Divector-Counsel
Elaine R. Jones
Deputy Director-Counsel
Charles Stephen Ralston
Deputy Ditector-Counsel
Board of Directors
Billye Suber Aaron
Atlanta, Georgia
Anthony G. Amsterdam
New York, New York
Clarence Avant 3
Los Angeles, California
Mario L. Baeza
New York, New York
John T. Baker
Bloomington, Indiana
Alice M. Beasley
San Francisco, California
Mary Frances Berry
Washington, D.C.
Anita Lyons Bond
St. Louis, Missouri
William H. Brown HI
Philadelphia, Pennsylvania
Helen L. Buttenwieser
New York, New York
Jack G. Clarke
New York, New York
LH. Clayborn
Dallas, Texas
William K. Coblentz
San Francisco, California
Theodore L. Cross
New York, New York
Talbot D’Alemberte
Tallahassee, Florida
Allison S. Davis
Chicago, Hlinots
Ossie Davis
New Rochelle, New York
Peter J. Deluca
White Plains, New York
Adrian W. DeWind
New York, New York
Anthony Downs
Washington, D.C.
Robert F. Drinan
Washington, D.C.
Charles T. Duncan
Washington, D.C.
“COMMITTEE OF 100” Chairman, BISHOP PAUL MOORE, JR.
Henry Aaron Dighann Carroll Marilyn Horne Eleanor Holmes Norton
Steve Allen James E. Cheek John H. Johnson Richard L. Ottinger
Arthur R. Ashe
Lown Baez
irch Bayh
Yad Beamon
Harry Belafonte
Saul Bellow
on C. Bennett
rone Bemett, Jr.
Viola W. Bernard
Leonard Bernstein
Hans A. Bethe
ulian Bond
{enry T. Bourne
George P. Brockway
Yvonne Brathwaite Burke
Helen L. Buttenwieser
Kenneth C. Edelin
Boston, Massachusetts
Marian Wright Edelman
Washington, D.C.
Christopher F. Edley
New York, New York
Helen G. Edmonds
Durham, North Carolina
David E. Feller
Berkeley, California
Clarence Finley
New York, New York
Norman C. Francis
New Orleans, Louisiana
Marvin E. Frankel
New York, New York
John Hope Franklin
Durham, North Carolina
J. Thomas Franklin
Boston, Massachusetts
Ronald T. Gault
New York, New York
Nannette B. Gibson
Princeton, New Jersey
Jack Greenberg
New York, New York
Gordon G. Greiner
Denver, Colorado
Lucy Durr Hackney
Philadelphia, Pennsylvania
Charles V. Hamilton
New York, New York
Louis Harris
New York, New York
Eliot Hubbard III
Lincoln, Massachusetts
Ann Hutchinson
New York, New York
Patricia L. Irvin
New York, New York
Herman Johnson
Kansas City, Missouri
Anna Faith Jones
Boston, Massachusetts
Jetta N. Jones
Chicago, Hlinois
Quincy Jones
Los Angeles, California
Anna J. Julian
Oak Park, llinois
Harry Kahn
New York, New York
Nicholas DeB. Katzenbach
Morristown, New Jersey
David E. Kendall
Washington, D.C.
Reginald F. Lewis
New York, New York
Connie S. Lindau
New York, New York
George E. Marshall, Jr.
Los Angeles, California
Robert McDougal, Jr.
Chicago, Hlinois
Shirley Chisholm
Ramsey Clark
Aaron Copland
Bill Cosby
Maxwell Dane
Ossie Davis
Ruby Dee
Victoria Delee
Ralph Ellison
John Hope Franklin
Mrs. A. G. Gaston
Kenneth A. Gibson
Roland B. Gittelsohn
Charles E. Goodell
Richard G. Hatcher
Theodore M. Hesburgh
‘Mrs. Percy Julian
Horace M. Kallen
Ethel Kennedy
James Lawrence, Jr.
Max Lemer
W. Arthur Lewis
John A. Mackay
‘Horace S. Manges
Henry L. Marsh III
William James McGill
Linda B. "McKean
Karl Menninger
Charles Mernll
Arthur Mitchell
Paul Newman
Anthony Newley
Paul Moore, Jr.
New York, New York
James M. Nabrit, Jr.
Washington, D.C.
Martin D. Payson
New York, New York
Stephen J. Pollak
Washington, D.C.
Glendora Mcllwain Putnam
Boston, Massachusetts
Daniel L. Rabinowitz
Newark, New Jersey
C. Carl Randolph
New York, New York
Gilbert T. Ray
Los Angeles, California
Henry T. Reath
Philadelphia, Pennsylvania
Norman Redlich
New York, New York
Charles B. Renfrew
San Francisco, California
Harvey C. Russell
Yonkers, New York
William H. Scheide
Princeton, New Jersey
Frederick A. O. Schwarz, Jr.
New York, New York
Bernard G. Segal
Philadelphia, Pennsylvania
Jacob Sheinkman
New York, New York
George C. Simkins, Jr.
Greensboro, North Carolina
Wayman F. Smith III
St. Louis, Missouri
Michael I. Sovern
New York, New York
Bonnie Kayatta-Steingart
New York, New York
Chuck Stone
Philadelphia, Pennsylvania
Jay Topkis
New York, New York
Cyrus R. Vance
New York, New York
James Vorenberg
Cambridge, Massachusetts
John W. Walker
Little Rock, Arkansas
Robert C. Weaver
New York, New York
M. Moran Weston
New Rochelle, New York
Roger W. Wilkins
Washington, D.C.
Karen Hastie Williams
Washington, D.C.
E. Thomas Williams, Jr.
Bronx, New York
Andrew Young
Atlanta, Georgia
Honorary Board Member
Dorothy Rosenman
New York, New York
Leon E. Panetta
Gordon A. B. Parks
Sidney Poitier
Joseph L. Rauh, Jr.
Carl T. Rowan
John L. Saltonstall, Jr.
William H. Scheide
Arthur Schlesinger, Jr.
Charles E. Silberman
John P. Spiegel
‘William Styron
Telford Taylor
Robert Penn Warren
Robert C. Weaver
Tom Wicker
Myrlie Evers Williams
The “Committee of 100,” a voluntary cooperative group of individuals headed by Bishop Paul Moore, Jr.,
has sponsored the appeal of the NAACP Legal Defense and Educational Fund, Inc. since 1943 to enable the
Fund to put into operation a program designed to make desegregation a reality throughout the United States.
As of April 1989
WARREN McCLESKEY
Petitioner,
vs. . HABEAS CORPUS
2 NO.
RALPH M. KEMP, Superintendent
Georgia Diagnostic and
Classification Center,
Respondent.
MOTION FOR LEAVE TO PROCEED IN FORMA PAUGPERIS
The petitioner, WARREN MCCLESKEY, bv his undersigned
counsel, moves the Court for leave to file his petition for
habeas corpus, without prepayment of costs, and to proceed in
forma pauperis. An affidavit by petitioner in support of this
motion is attached hereto.
Respectfully Submitted,
ROBERT H. STRODP
141 Walton Street
Atlanta, Georgia 30303
IN THE SUPERIOR COURT OF BUTTS
STATE OF GEORGIA
WARREN McCLESKEY st y
Petitioner, Civil Action No.
D-0023035 ’
Inmate Number Habeas Corpus
vs
RALPH KEMP, , Warden
Georgia Diagnostic & Clgssification Center,
Name of Institution
Respondent.
Request to Proceed in Forma Pauperis
I, WARREN McCLESKEY , being first duly sworn, depose and
a that I am the plaintiff in the above entitled case; that in
support of my motion to proceed without being required to prepay
fees, costs or give security therefor,I state that because of my
poverty I am unable to pay the costs of said proceeding or to give
security therefor; that I believe I am entitled to redress.
I further swear that the responses which I have made to ques-
tions and instructions below are true.
l. Are you presently employed? Yes ( ) No (4)
a. If the answer is yes, state the amount of your
salary or wages per month, and give the name and
address of your employer.
AOC-6
(7-1-85)
be. If the answer is no, state the date of last
employment and the amount of the salary and wages
per month which you received.
pe — ’ & a? — Jo -
~
3 Have you received within the past twelve months any
money from any of the following sources:.
a, Business, profession or form of self-employment?
Yes ( ) Ro (+)
b. Pensions, annuities or life insurance payments?
Yes ( ) No (+)
Ce Rent payments, interest or dividends?
Yes ( ) No. ()
a, Gifts or inheritances?
Yes ( ) No (=)
e. Any other sources?
Yes (4) NoO=:L.)
If the answer to any of the above is yes, describe each
source of money and state the amount received from each
during the past twelve months.
= pe 2 Fiat = ! 4 =)
on
3. Do you own any cash, or do you have money in a checking
or savings account? Yes (+) No ( ) (Include any
funds in prison accounts)
If the answer is yes, state the total value of the items
owned.
Ae A . -
AOC-6 2 Ss
(7-1-85)
4, Do you own any real estate, stocks, bonds, notes, auto-
mobiles, or other valuable property (excluding ordinary
household furnishings and clothing)?
Yes ( ) No (-)
If the answer is yes, describe the property and state
its approximate value.
5 List the persons who are dependent upon you for finan-
cial support; state your relationship to those persons,
and indicate how you contribute toward their support.
I understand that a false statement or answer to any ques-
tion in this affidavit will subject me to penalties for perjury
and that state law provides as follows:
(a) A person to whom a lawful oath or affirmation
has been administered commits the offense of
perjury when, in a judicial proceeding, he
knowingly and willfully makes a false state-
ment material to the issue on point in question
(b) A person convicted of the offense of perjury
shall be punished by a fine of not more than $1000
or by imprisonment for not less than one or more
than ten years, or both....0.C.G.A.§16-10-70
Signature of Petitioner
AOC-6
(7-1~-83) = 3m
VERIFICATION
State of Georgia, County of BUTTS
Personally appeared before me, the undersigned officer authorized
by law to administer oaths, the undersigned affiant, who having
first been duly sworn, says under oath: That he is the plaintiff
in this action and knows the content of the above Request to
Proceed in Forma Pauperis; that the answers he has given are true
of his own knowledge, except as to those matters that are stated
in it on his information and belief, and as to those matters he
believes them to be true. Affiant further says under oath that
he has read the perjury statute set out above and is aware of the
penalties for giving any false information on this form.
/ 4 2 r
4 ny
Signature of Affiant Petitioner
Subscribed and sworn to before me
this i day of Moo pid | IE
- - re t £, - I [} Tas
—
{Notary Public or other person authorized
by law to administer oaths.)
Notary Public. Fulion County Caorgid
4 ; - A “rm NOY OY
Mv Commission Expires May 30, 1990
CERTIFICATION
I hereby certify that the Plaintiff herein has an average
monthly balance for the last Sustve U2 months of $§ 77 ¢ C
on account to his credit at the A
institution where he is confined. I further certify that Plain-
tiff likewise has the following securities to his credit accord-
ing to the records of said
. ZZ 9
Institution: “77 nn LR
~~
=>
x 7 7)
If not confined for a full twelve (12) months, specify the number
of months confined. Then compute average monthly balance on that
number of months.
if >
| or 2, / in? nell ee /7 2
Authorized Officer of] Institution Date
(Please attach copy of Balance Sheet if available)
will
16
when pressed by the Court to clarify his remarks. (Fed. III 61-
64).
On this record, Worthy’s July 9th testimony is credible and
consistent; his August 10th testimony on this point is not. The
Court’s August 10th observation that it "could think of no reason
to discredit Worthy’s testimony" (Fed. III 119) is warranted on
the whole; the Court should, however, reject Worthy’s clumsy
attempt on August 10th to protect Detective Dorsey.
5. The Testimony of District Attorney Parker
Assistant District Attorney Russell Parker firmly denied
ever meeting with Evans prior to July ‘12, 197s, (Fed. I 142;
Fed. III 109). He also testified that he never requested that
Offie Evans be moved near McCleskey or that Evans act to overhear
or elicit information from McCleskey in the cell. (Fed. III 111-
12). There 1s no reason, we believe, to doubt Mr. Parker on
these points. Warren McCleskey’s constitutional claim does not
depend on whether Parker was personally aware of the informant
relationship.
Parker never claimed that he possessed full knowledge of all
steps taken by other officers investigating the case. To the
contrary, he had no recollection of Detective Dorsey’s role in
the McCleskey case at all (Fed. I 131: Fed. III 113), and more
specifically, he did not remember Dorsey’s presence at the July
12, 1978 meeting, even though his own notes indicate that Dorsey
attended that meeting. (Ped. T 131; Fed. 113 113). Moreover,
Parker was not aware of any interviews apart from the ones he
appekion Tek corned did wot ed fy lad
wef) (eok 1 on towne wtf alawelsiaret]
4 [vtvned oa sled be ! wo We telonad 1.
ules | Tot coud ew wet Led [Een tr oA Col log
Wey Ub aa Baus Shoes] &- re coguczed
oa Ve apeat
203 DA Ao. po / [1 sth => dane oven A
C NC Co tZ
24
Ste he cx (owl ; def , Couuned acl A CePA L S (2 ?, ©
dine. ha , Uadddid all ws sts. expt bau
Covaded On. Wp free [17%Y) wah. a dry. Clo wd
33 ps Sle Lad Legal lel Vis gel AAeep s¢ G (is
Made=e k ~> ORA
sulp SELLA
st. he ck. to vempec. (ecnd
LSD olin ny
32-3
MERITS
Fie ow Monae wel w) Pande. 7/128 = to ld ik al
alld bw basic Eads 8 MN adaidnice tra.
ie Pinkie did vobanlibo. br solid ay
Www Eo LS head ic :
CA Prunlites Won ude Bois before vA [2 / r
fel) Lo [SOVITS) / Bane eer he
is ow [il ?¢
MUHA allied Outs bu ld lo woe nid |
nWoLE Ho IM
pte: rr
Jousera en naibel 7%) baat cou als Je
TR yun en bin lr)
wo (wv. 1 EI Ve Moulben (Hoey pec) )
Sede (28; [of ne 21[79 ~~ wee Wf
cited bod” tec Lihue ou
hwoutedoe Busunt wpred
wo ore lr asl | bods aut Loos Fund
22 Ube vas elle of Lov Y¥
Dat -
Dor: SEY aero bows (there he (are = = le frou
angen eral [A BELauu Uwlrén GG ae Lf
(lel Go est id -- did us lly posi tue +,
10 A Ounage Betti 15 Jal Lindl Bc
(o|- 2 Wo yy ts © f) as Ct velod El. se0/- conte
iy omen ps
(2-b3 ~ seovyd bv vecald sowie thle ce Nad &
Elouu = helese
Ct ved a veg lest fo Placer seuuow ul
cold Whe to Sou uy 0010 L A
Coven) ¢ sli ~ by A A
| by 2k Ele big, (gf 0/8) -- (st | Masse f ! Ewan
lise, Be hp peed Oot wpe Hot Bau
Lad etl 4% ~ DA
jst hie dips. fallleet ou) Gavan, (st
biol ey, 2 quwrsue ~~ decul
brace tue 'a a Leo st
&
FEDERAL BUREAU OF INVESTIGATION
7/7/78 Date of tronscription
OFFIE GENE EVANS was interviewed at the Georgia
Bureau of Investigation (GBI) Headquarters in Atlanta, Georgia.
EVANS was advised of the official identities of the interviewing
Agents and that he was being interviewed regarding an alleged
violation of the Escaped Federal Prisoner Statute.
EVANS was read his rights as set forth on an
"Interrogation; Advice of Rights" form. EVANS was presented
this form which he appeared to read. EVANS stated he understood
his rights and signed his name to the form. .
EVANS admitted his true identity and thereafter
provided the following information:
EVANS indicated that on April 2, 1973, he appeared
in U. S. District Court, Northern District of Georgia, Atlanta,
and was convicted of forging and uttering a U. S. Treasury
check and was subsequently sentenced to six years in prison.
He was paroled from March 1975 until April 6, 1976, at which
time he was returned to the U. S. Penitentiary in Atlanta,
Georgia, to do the rest of his sentence.
#
While at the U. S. Penitentiary in Atlanta, Georgia,
EVANS assisted DEA Agents and prison officials, specifically
Lt. GEOUGE and Counselor WHITAIN, on a drug matter.
At the Federal Community Treatment Center in Atlanta,
Georgia (FCTC-A) EVANS tried to assist DEA Agents JIM WILLIAMS
and PERSHING in drug matters. EVANS was not cleared by FCTC-A
personnel to assist these agents.
-—
While assigned to FCTC-A, LAWSON advised EVANS that
his urine test was positive for quinine. LAWSON restricted
EVANS to FCTC-A, however, the next day EVANS signed out early
in the morning and did not come back. The sign-out sheet
had written across the top that he was restricted to FCTC-A
but the clerk on duty let him go out anyway. EVANS was supposed
to be back by 11:00 p.m. that evening, however, he did not make
it back sole decided to stay out.
cs39- 14
7/3/78 . Atlanta, Georgia 4 AT 76-6502»
File Interviewed on
SA BRUCE SI uson
SA DAVID J. KELSEY/rk 7/6/78
Date dictated
/ J
This document contains neither recommendations nor conclusions of the FBI. It is the property of the FBI and is loaned to
your agency; it and its contents are not to be distributed outside your agency. FBI/DOJ
AT 76-6502
~ EVANS stated that he did not intend to escape
as he had only 36 days to serve at FCTC-A with only 10
or 11 left to go. EVANS knew it was against the law not
to return to FCTC-A and EVANS knew he could be prosecuted
for not returning to FCTC-A when he was supposed to.
EVANS stated he figured that what he was doing
would justify his not returning.
EVANS telephoned KINNEBREW at FCTC-A last Friday
and was told to return to FCTC-A but EVANS decided not to.
EVANS stated he was arrested by GBI Agents at his
house early this morning.
Name
Race
Sex
Date of Birth
Place of Birth
Height
Weight
Scar
Wife
Phone
OFFIE GENE EVANS
Negro
Male
August 15, .1935
Monroe, Georgia
58
130 pounds
One inch scar over
right eye
MABLE MURPHY EVANS,
768-0723
Hi ana Sarit LPR DEIR fof 40 HEL - Eo
x . 1 EE ET £3 8 BE APD 5 v RE
Bo Heb a fe a A gj rao AG SSL HL
% gest - gy, 238 . Ay 4 54 WR a:b in a 1 « ’ +I
Td y 3 1% p GSR I
gorse si RSE we = HE 1335) 3h ES pi
fou" oh
%
Nya HO Ue TSM Lua SAMARAS Cp. RT Tm an RN I SS a pela iia
Se TURN TRG TRUE RY Sal Re SERIA an » Tr te 2)
" :
N
~
0
ph
3
Post Office Box 1683 Ios
Atlanta, Georgia 80301 i
76-6539 oe
December 28, 1978
Mr. ¥illiam L. Harper i
U. S. Attorney | |
Northern District of Georgia |
Atlanta, Georgia
Fe
Attention: Mr. James E. Baker !
Assistant U. S. Attorney i
Chief, Criminal Section ied
| 7%
Dear Sir: Rie
RE: OFFIE EUGENE EVANS | ie
ESCAPED FEDERAL PRISONER fs
This matter was discussed with Assistant U. S.
Attorney (AUSA) William F. Bartee, Jr., of vour office oi]
and Special Agent (SA) David J. Kelsey of our office. hi
SA Kelsey furnished Mr. Bartee the enclosed letterhead Biiil
memorandum which sets forth the facts concerning Evans’ Ba
background, escape, and capture. In addition, Mr. a"
Bartee was advised of the following:
{
1.) Bureau of Prisons' officials at the U. S. Wil
Penitentiary in Atlanta, Georgia, have indicated that i
Evans has furnished information to them regarding criminal | ot
activities at the U.S. Penitentiary, Atlanta, Georgia. fr
Wt
2+) Fulton County Assistant District Attorney po
Russell Parker advised Evans testified at a murder trial ery
(in which the victim was a police officer) that he Se
(Evans) overheard the perpetrator admit to shooting the ]
police officer. |
3.) Bureau of Prisons' officials at the U.S. & J
Penitentiary, in Atlanta, Georgia, have advised that as i
a result of administrative action taken against Evans
for escape, Evans' parole date of July 6, 1978, has
By
Enclosure
2 / =
aT NT ge BRL SE a ) OF
-Addressee HG HE RCE |
tlanta (76-6539) yo ; 5 278]
DJK/cg
Eo alal ex.
(2) 2 5
go 3 rE RR A Ss SOIR BR i gh a a EY spa di a ERE a, Rt pe ah HET ARREST ERE SHEL PRE 2 # = H alae SL ela fui 323 X E SP, iets iad rend. a" eer
LJ
{ 3
rescinded. Evans is now scheduled for release at an
undetermined date in 1979.
Mr. Bartee declined prosecution of Evans for
violating Title 18, Section 751 and Section 4082(4d),
U. S. Code, as Evans was in escape status for a very
short period of time; Evans’ escape was a ''walk away"
as opposed to an actual escape from a penitentiary,
and Evans' parole of July 6, 1978, was rescinded.
Very truly yours,
BENJAMIN H. COOKE
Special Agent in Charge
a
DECLAN J. HUGHES
Supervisory Special Agent
By:
UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF GEORGIA
GAINESVILLE DIVISION
JOSEPR WILSON, JR.
CIVIL ACTION NO. (C83-93G
Petitioner,
V.
HABEAS CORPUS
RALPH KEMP, WARDEN,
A
%
%
3%
%
¥
¥
*
*
Respondent.
NOTICE OF FILING
COMES NOW Ralph Kemp, Warden, Respondent in the
above-styled action, and submits the instant notice of filing
and attached exhibit to be submitted as a part of the record in
the instant case. Respondent submits as Respondent's Exhibit
No. 8, an affidavit from Joseph I... Katz, Ph.D. Said affldavit
is submitted pursuant to the hearing held on PeciBer 30, 1983
before this Court at which time Petitioner requested the
opportunity to present additional evidence and Respondent
requested the opportunity to present evidence to supplement
that presented in McCleskey v. Zant, Civil Action No. C81-4234A
and respond to any evidence submitted by the Petitioner in this
case. The affidavit submitted from Dr. Katz is specifically in
response to the brief submitted by the Petitioner in this case
and specifically supplements the material submitted previously
in McCleskey v. Zant.
PY)
WHEREFORE, Respondent requests that the instant document be
filed as a part of the record and be considered as substantive
evidence in this case.
MARY BETH WESTMORELAND
132 State Judicial Building
40 Capitol Square, S.
Atlanta, Georgia 30334
(404) 656-3349
Ww.
Respectfully submitted,
MICHAEL J. BOWERS
Attorney General
JAMES P. GOOGE, JR.
Executive Assistant Attorney General
RION O. GORDON X
irst Assistant Attorney General
A Si
LLIAM B rd ’
Senior resistaf Se rney General
Dis Lo ibs bvtie lend
ra WESTMORELAND
AC tant Attorney General
i
STATE OF GEORGIA
COUNTY OF FULTON
A FPFPIDAVLIT
Personally appeared before me, the undersigned, an officer duly authorized
to administer oaths, JOSEPH L. KATZ, Ph.D., who after first being duly sworn,
deposes and says:
l.
That he is presently under contract with the Department of Offender
Rehabilitation and has conducted studies and analyses of the data and reports
Of Professor David C. Baldus pursuant to said contract;
2%
That he testified as an expert witness for the Department of Offender
Rehabilitation in the case of Warren McCleskey v. Walter Zant, Civil Action
No. CB1-2434A in the United States District Court for the Northern District of
Georgia;
3 Ld 2 sf
That he has conducted further analyses of the data provided by Professor
David C. Baldqus to be utilized in the case of Joseph Wilson, Jr. v. Ralph
Kemp, Civil Action No. CV83-93G, presently pending in the United States
District Court for the Northern District of Georgia, Gainesville Divison;
RESPONDENT'S EXHIBIT NOC. 8
4.
That he examined the data in both the Procedural Reform Study and the
Georgia Charging and Sentencing Study for cases in which the defendant was
convicted in the Superior Court of Forsyth County, Georgia. The Procedural
Reform Study has three cases from Forsyth County and Joseph Wilson is not one
of those three cases. The Georgia Charging and Sentencing Study has a total
of eight cases from Forsyth County which include the three cases from the
Procedural Reform Study and the case of Joseph Wilson. The defendants in each
of the eight cases from Forsyth County are classified by Professor Baldus as
white. Seven of these cases are classified as having white victims with the
race of victim for the eighth case being indicated as unknown.
In the analysis that was performed earlier by Professor Baldus on the data
in the Georgia Charging and Sentencing Study, cases in which the race of the
victim was unknown were assumed to have the same race as the defendant;
therefore, the one case out of Forsyth County in which the race of the victim
is indicated to be unknown would be classified for purposes of Professor
Baldus' analysis as having a white victim as the case has a white defendant.
No variation exists in the defendant-victim racial combination in Forsyth
County; therefore, the effect of race of defendant or race of victim on
1
sentencing outcomes cannot be determined.
5.
Affiant further states that the following information is presented to
supplement the testimony given by affiant in the case of McCleskey v. Zant
before the Honorable Owen Forrester, and is in response to the order in that
case and the brief of Petitioner in the case of Wilson v. Kemp. The following
is based on affiant's analysis of the data provided by Professor David C.
Baldus.
a. The Relationship Between Regression and
Crosstabulation In Controlling For Variables
Crosstabulation is the usual statistical method for measuring the effect
of one variable after controlling for the effects of several related
variables. When all the variables of interest take on values of 0 or 1
exclusively, there also exists a direct relationship between regression and
crosstabulation methods, which is illustrated by the following example.
The Georgia Charging and Sentencing Study lists 1,018 defendants who were
indicted for murder. Professor Baldus defined the variable DPMURIDT to be 1
for those cases in which the defendant received a death sentence (127 cases)
‘and O for those cases in which a lesser sentence was imposed (891 cases).
Overall, death sentences were imposed on 12.0% (127/1018) of the defendants
who were indicted for murder. In this example, the effects of the three
variables, MULSH (multiple gunshots), BLVICMOD (family, lover, liquor, barroom
quarrel) and PREMEDK (killing planned more than five minutes), on death
sentencing rates have been examined using both the crosstabulation and
~~ ~~ rd rn a tparey ss rs nn 3 vos ee ne
regression techniques. Consider first the effect of the variable MULSH
(multiple gunshots) on death sentencing rates, which are calculated for cases
categorized based on the occurrence of the variable MULSH.
Table 1
1
. 127%"
1-11018.!
32,1200"
MULSH=1 MULSH=0
| 1
2 3
id 35 : : S92 :
! 26060 : : 7152 :
yz, 1316! ! 2,122
According to the crosstabulated cells 2 and 3, death sentences were imposed in
13.16% of the cases in which MULSH occurred (MULSH=1), whereas a death
sentence was imposed in 12.23% of the cases in which MULSH did not occur
(MULSH=0). Therefore, the variable MULSH shows little affect on the resulting
death sentencing rates since the death sentencing rate for those cases in
which MULSH occurred (.1316) is very close to the death sentencing rate in
which MULSH did not occur (.1223). Therefore, according to Table 1, the
occurrence of MULSH in a case tends to increase, on the average, the death
sentencing ‘rate by only .0093 (.1316 - .1223).
A regression model can also produce the same information that is deducible
from the crosstabulations in Table 1. A simple linear regression with
dependent variable DPMURIDT and independent variable MULSH produces the
following equation:
(1) DPMURIDT = .1223 + .0093(MULSH)
In equation (1), the constant .1223 is the death sentencing rate when the
variable MULSH has not occurred, and the coefficient of MULSH .0023 is the
increase in the death sentencing rate (.1316 - .1223) that is due to the
occurrence of the variable MULSH. Since the p-value of the variable MULSH is
.6955 in this regression, it follows that the occurrence of MULSH alone does
not have a substantial effect on death sentencing rates.
To control for the second variable BLVICMOD (family, lover, liquor,
barroom quarrel), the crosstabulation from Table 1 is augmented as presented
in Table 2.
Table 2
1
: 12d =
: 1018. !
=. 1200!
MULSH=1 MULSH=0
71
2 3
: SH ie . O21
: 66 : : 752 1
! =,1316! y = 1233!
BLVICMOD=1 BLVICMOD=0 BLVICMOD=1 # BLVICMOD=0
f i I . i
) 5 : 29. .! : Bea 2 87
: 12) : : 145 : : 327 : 3 425 2
« =.049%96! 3 =.,2000. !1'=_,0153) 1m 204871
The death sentencing rates after controlling for both variables MULSH and
BLVICMOD are summarized in Table 3.
Table 3
CELL NUMBER MULSH BLVICMOD DEATH SENTENCING RATE
4 1 1 4.90%
5 ) 0] 20.00%
6 0 : 3 1.53%
7 0 0] 20.47%
According to the information in Table 3, the variable BLVICMOD is shown to
have a mitigating effect on death sevitencink rates. This observation is based
on the low death sentencing rates 4.96% and 1.53% that are found in cells 4
and 6 when the variable BLVICMOD occurred in the case, and on the
substantially higher death sentencing rates 20.00% and 20.47% that are found
in cells 5 and 7 where the variable BLVICMOD did not occur. The numbers in
Table 3 may also be explained in the following manner. In cases where neither
variable MULSH or BLVICMOD occurs, the death Sentencing rite is .2047. 1f only
the variable MULSH occurs, then the death sentencing rate drops to .2000,
whereas if only the variable BLVICMOD occurs, the death sentencing rate drops
to .0153. Therefore, the lone occurrence of the variable MULSH has the effect
of changing (reducing) the death sentencing rate by -.0047 (.2000 - .2047).
whereas the lone occurrence of the variable BLVICMOD has the effect of
changing (reducing) the death sentencing rate by -.1894 (.0153 - .2047). If
there is only an additive effect on death sentencing rates between the two
variables MULSH and BLVICMOD, then the death sentencing rate for the cases
where both MULSH and BLVICMOD occurred would be .0106 as tabulated below.
.2047 death sentencing rate when neither MULSH nor BLVICMOD occur
- .0047 death sentencing rate when MULSH occurs
- .1894 death sentencing rate when BLVICMOD occurs
.0106
The actual death sentencing rate for the cell when both MULSH and BLVICMOD
occur (cell 4) is .0496. Therefore, a two-way interaction effect between the
two variables MULSH and BLVICMOD is said to exist that increases the death
sentencing rate by an amount .0390 (.0496 - .0106) when both MULSH and
BLVICMOD occur in a case.
This same information can be drawn directly from the regression model that
appropriately controls for both variables and the two-way interaction effect.
Let DPMURIDT again be the dependent variable and let MULSH, BLVICMOD, and the
two-way interaction variable MULSH*BLVICMOD be the three independent
variables. The two-way interaction variable for MULSH and BLVICMOD, written
MULSH*BLVICMOD is constructed for each of the 1,018 cases as follows. If both
MULSH and BLVICMOD occurred in a particular case, that is if both variables
MULSH and BLVICMOD are 1 for a particular case, then the two-way interaction
variable is defined to be 1 for that case. Otherwise, the two-way interaction
variable MULSH*BLVICMOD is 0 for that case. The regression model is presented
as equation (2).
(2) DPMURIDT = .2047 - .0047(MULSH) - .1894(BLVICMOD) + .0390 (MULSH*BLVICMOD)
The constant term, .2047, represents the death sentencing rate when both MULSH
and BLVICMOD do not occur in the case. The coefficients of the independent
variables in equation (2) represent the change in the death sentencing rate,
.2047, due to the occurrence of the independent variable in the case. For
example, the occurrence of only the variable MULSH reduces the death
sentencing rate. from .2047 down to .2000. Therefore, the change in death
sentencing rates due to the occurrence of MULSH is -.0047 which is exactly the
coefficient of the independent variable MULSH. Since the variable BLVICMOD did
not occur, then the value of both the variable BLVICMOD and the two-way
interaction variable (MULSH*BLVICMOD) would be zero yielding a predicted death
sentencing rate of .2000 (exactly the appropriate crosstabulated death
sentencing rate) for cases in that category. A similar interpretation applies
to the coefficient -.1894 for the variable BLVICMOD as the marginal effect on
the death sentencing rate .2047 for cases in which BLVICMOD occurred and MULSH
did not occur. Therefore, the death sentencing rate for those cases in which
BLVICMOD occurred and MULSH did not occur is .0153 (.2047 - .1894). Finally,
the coefficient of the two-way interaction variable MULSH*BLVICMOD represents
the effect on death sentencing rates that results due to the fact that the
marginal effects of the variables MULSH and BLVICMOD tale perfectly
additive, as is apparent from the following tabulation:
> f
.2047 death sentencing rate when neither MULSH nor BLVICMOD occur
—- .0047 death sentencing rate when MULSH occurs
- .1894 death sentencing rate when BLVICMOD occurs
.0390 coefficient of MULSH*BLVICMOD
.04926 death sentencing rate when both MULSH and BLVICMOD occur
Consider the regression with the same dependent variable DPMURIDT but with
only the two independent variables MULSE and BLVICMOD as shown in equation (3)
below:
(3) DPMURIDT = .2002 + .0128(MULSH) - .1792(BLVICMOD)
As a result of omitting the two-way interaction variable MULSH*BLVICMOD, the
coefficients for the constant term and the independent variables are slightly
distorted as compared to the respective coefficients in the earlier regression
in equation (2). This distortion in the coefficients is due to the fact that
there exists a slight interaction effect in the MULSH*BLVICMOD variable for
which the regression in equation (3) is attempting to numerically compensate.
In this example, the two-way interaction variable MULSH*BLVICMOD is not
significantly different than 0; therefore, the distortion that results from
the omission of this two-way interaction variable is only slight. If a strong
interaction effect existed, the noninclusion of this interaction effect could
seriously distort both the size and significance of variables that are
presented in the regression model.
Table 4 presents the crosstabulations after controlling for the third
of
[>
variable PREMEDK (killing planned more than 5 minutes).
Table 4
1
JE
2 Toit
! =,1200!
MULSH=1 MULSH=0
f ~3
2 3
7 357 TEE
ro 9 tog TES.
tie, 1336) ! =.,1223!
BLVICMOD=1 BLVICMOD=0 BLVICMOD=1 BLVICMOD=0
{ ; |
T A TTT T TET
ELT t+ Tas. a io F271 1 39
! '=,0496! ! =.2000! ! =.0153! ' =.2047!
PREMEDK=1| PREMEDK=0 PREMEDK=1|PREMEDK=0 PREMEDK=1| PREMEDK=0 PREMEDK=1| PREMEDX=0
| ] | a | 3 RB
8 9 10 11 12 12 14 15
PE TRY THY AT av IW OTT TI It 4 Iwi 7
HEE nr FE a FEE ay FEA WIGITE Ge BTR a 1 ER AY a Ese
ae. 0857! 1 =. 0349! =.26920 1 £.119400 =.0408! 12.0108! ! =. 8406) I =. 1304!
The death sentencing rates for
PREMEDK are summarized in Table 5. q
cases after controlling for the variable
Table 5
CELL NUMBER MULSH BLVICMOD PREMEDK DEATH SENTENCING RATE
8 X 1 1 8.57%
| 1 i 0 3.49%
10 1 0 l 26.92%
11 : 0 0 11.94%
32 0 l 1 4.08%
13 0) 1 0] 1.08%
14 0 0 1 34.06%
15 0 0 0 13.94%
s.r SAE mre antral et T—————— -.
«1 0~
The information on death sentencing rates in Table 5 can again be deduced from
a regression model. Again, the dependent variable is DPMURIDT, whereas seven
independent variables are required namely, MULSH, BLVICMOD, PREMEDK,
MULSH*BLVICMOD, MULSH*PREMEDK, BLVICMOD*PREMEDK, and MULSH*BLVICMOD*PREMEDK.
Of the 7 independent variables, three are original variables, three are
two-way interaction variables and one is a three-way interaction variable. The
regression model is given in equation (5).
(5) DPMURIDT = .1394 - .02(MULSH) - .1286(BLVICMOD) + .2012(PREMEDK)
+ .0441(MULSH*BLVICMOD) - .0514(MULSH*PREMEDK) - .1712(BLVICMOD*PREMEDK)
+ .0722(MULSH*BLVICMOD*PREMEDK)
Again, the coefficients in the regression in equation (5) are directly related
to the death sentencing rates from the crosstabulations. The constant term,
.1394, is the death sentencing rate when none of the three variables are
present in the case. The coefficient of MULSH is equal to ,1194 - .1394 or
-.02 and represents the marginal effect on death sentencing rates if only
MULSH occurs and both BLVICMOD and PREMEDK do not occur. As the lone
occurrence of BLVICMOD has a death sentencing rate of .0108, its coefficient
pe
must be (.0108 - .1394) or equal to -.1286. The coefficient for PREMEDK can be
deduced in a similar fashion. The coefficient of a two-way interaction term,
such as MULSH*BLVICMOD, is determined by the situation when both MULSH and
~1i=
MAAR ATA =~ TAILS ns SA Tn tye Aa Sh Bren ne tm
BLVICMOD occur and PREMEDK does not occur. Since the death sentencing rate in
this situation is .0349, then the coefficient is found as follows:
.1394 death sentencing rate when all 3 variables do not occur
- .0200 coefficient of MULSH
+1280 coefficient of BLVICMOD
+ coefficient of MULSH*BLVICMOD
0377. death sentencing rate when only MULSH and BLVICMOD occur
The solution for x (the coefficient of MULSH*BLVICMOD) is .0441. The
coefficients for the two other two-way interaction variables are found in a
similar manner. The coefficient for the three-way interaction variable
MULSH*BLVICMOD*PREMEDK is computed as follows:
.1394 death sentencing rate when all 3 variables do not occur
.0200 coefficient of MULSH
.1286 coefficient of BLVICMOD
+ .2012 coefficient of PREMEDK
+ .0441 coefficient of MULSH*BLVICMOD
- .0514 coefficient of MULSH*PREMEDK
- .1712 coefficient of BLVICMOD*PREMEDK
pe xX coefficient of MULSH*BLVICMOD*PREMEDK
+ .0857 death sentencing rate when all three variables occur
i
Thus the coefficient x for the three-way interaction variable
MULSH*BLVICMOD*PREMEDK is .0722.
[;
Equation (6) displays the regression model if all the two-way and
three-way interaction variables are omitted.
(6) DPMURIDT = .1552 - .0115(MULSH) - .1529(BLVICMOD) + .1353(PREMEDK)
A comparison of the coefficients in Table 6 between the regressions in
&
equation (5) and equation (6) shows substantial distortion for the variable
PREMEDK, and slight distortion for the other coefficients Que to the exclusion
of the interaction variables.
i SE
Table 6
COEFFICIENTS FOR
REGRESSION WITH
COEFFICIENTS FOR
REGRESSION WITHOUT
VARIABLE INTERACTION VARIABLES a INTERACTION VARIABLES
Constant «1394 «31552
MULSH - 0200 - «0115
BLVICMOD - «1286 -1.9 1529
PREMEDK «2012 1353
Regression can be used as a complete substitute to crosstabulations as a
method to control for variables only if all the relevant interaction variables
are also included in the regression model. If important and significant
interaction variables are not represented. ih the regression model, and assumed
to be negligible or zero, there may be a substantial distortion in the
interpretation of the controlling effect as represented in the regression
model relative to the actual crosstabulated results.
b. On The Use of Crosstabulations or Regressions
in the Georgia Charging and Sentencing Study
‘Due to the fact that the Georgia Charging and Sentencing Study has at most
1,082 observations, either method of controlling for variables,
crosstabulations or regressions, will be severely limited as to the number of
variables that could be completely controlled for. For example, if one
attempted to use the crosstabulation technique to control for 10 variables,
1024 cells (2 raised to the tenth power) would be generated for the tenth
level, where all cases would be classified according to the BecRELence Or
nonoccurrence of the 10 control variables. Since only 1,082 cases are
3
available from the Georgia Charging and Sentencing Study, the expected small
number of cases in each cell on the controlled tenth level eliminates the
possibility of meaningful statistical comparisons of these observations within
the same cell. Similarly, the equivalent regression model with all the
appropriate interaction variables would require exactly 1023 variables. In
general, controlling for n variables generates 2 (2 raised to the n power)
cells for the nth level and 2" - 1 variables for the equivalent regression
model. Therefore, the use of regression as an equivalent approach to the
crosstabulation method also suffers from the relative small sample size that
is available in relationship to the number of potential important controlling
variables. Professor Baldus has defined over 230 potential aggravating and
mitigating factors in the Georgia Charging and Sentencing Study that could
conceivably affect, in his view, death sentencing outcomes. A regression model
that would control for all 230 variables and that would be equivalent to the
crogstabulation results would require 2 raised to the 230th power minus 1 ox
approximately in scientific notation, 1.7 x 10%° (which is equivalent to _the
number 17 followed by 68 zeroes) variables. The number or interaction terms
required in the regression model may be reduced somewhat if it can be shown
that some of the potential interaction effects have negligible or no effect
with regard to death sentencing rate (that is the coefficient of the
interaction variable is zero in the regression model). However, affiant's
analysis of some interaction variables shows the existence of substantial
interaction effects. The omission of interaction variables can completely
&
distort the size, direction, and significance of the regression coefficients
for. all variables.
-14-
It is unlikely that the Georgia Charging and Sentencing System will ever
process enough offenders to produce a large enough sample size so that all 230
potentially aggravating and mitigating factors that were defined by Professor
Baldus could be controlled for.
c. Multiple Regression and Crosstabulation
Analysis With An Incomplete Data Set
One critical consideration as to the proper application of any
multivariate technique rests on the accuracy and completeness of the data set.
As tabulated in Respondent's Exhibit 18A in McCleskey v. Zant, the Georgia
Charging and Sentencing Study contains a larde number of unknown values for
the questionnaire item variables. For example, suppose one attempted to
perform a multiple regression or crosstabulation experiment that controlled
for the three variables TRIANGLE (lover's triangle was a special precipitating
event), EXECUT (execution style murder) and DRESIST (defendant actively
resisted arrest) in the Georgia Charging and Fontoncing Stay. The data file
for this study indicates that the occurrence of the variable TRIANGLE was
unknown in a total of 74 cases, the occurrence of the variable EXECUT was
unknown in 109 cases, and the occurrence of the vavishie DRESIST was unknown
in 67 cases. For the three variables listed above, there are a total of 194
cases which have one or more unknown values for the variables listed above.
Without the correct information about the true value of these unknowns, it is
impossible to correctly classify these 194 cases so that they are "similarly
situated" with regard to the three varizbies listed above. The usual
statistical method of dealing with unknowns in crosstabulation analysis is to
iy 1
exclude cases from the crosstabulations that do not have complete data for
each of the variables that are controlled for. Obviously, as more variables
are controlled for, the number of cells in the crosstabulations increases and
the number of cases with complete data on these variables decreases.
Similarly, waltiple vegrassion models with TRIANGLE, DRESIST and EXECUT as
independent variables cannot properly utilize the 194 cases in which the
unknowns occur since the correct calculation of the regression coefficients
require the correct assignment of these unknown values. Again, the usual and
statistically accepted practice, as is also the case for crosstabulation
analysis, is to exclude cases from the regression analysis in which one or
more unknowns is present in the set of tpYebendent variables. In order to
generate any regressions at all, Professor Baldus recoded all the unknown
variable values uniformly to O.
d. Examples of Regression Where The Coefficient of the
Race of the Victim Variable Is Not Significant At .05
As affiant testified at the evidentiary hearing in MeCleskey v. Zant, in
August 1983, even if all of this recoded data is accepted at face value, the
statistical significance of the race of victim recode variable (WHVICRC) that
is observed in the regressions presented by Professor Halong could be
completely eliminated if additional aggravating and mitigating effects
(interaction variables) had been controlled for. As noted earlier, the number
of possible interaction variables is very large. Affiant has run regression
models that control for several of these interaction variables-in addition to
those variables that were defined by Professor Baldus. These experiments have
-16-
wn A + 3. ah 5 5 TS pa Smeg roe +
{
|
|
|
|
yielded hundreds of different regressions, both unweighted and weighted, in
which the coefficient of the race of the victim (WHVICRC) and the coefficient
of the race of the defendant (BLACKD) are not significant at the .05 level of
: ted . : 2
significance. Furthermore, many of these regressions have higher R values
than those testified to by Professor Baldus in his analysis. As examples,
Table 7 reports the coefficients and significance levels for three unweighted
regression models for the indicated dependent variable. The list of variables
for each regression is recorded in Appendix 1 of this affidavit.
RACE OF VICTIM COEFFICIENT
TABLE 7
RACE OF DEFENDANT
DEPENDENT V1TH STATISTICAL 1.EVEL COEFFICIENT WITH LEVEL UNADJUST!
VARIABLE OF SIGNIFICANCE OF SIGNIFICANCE R-SQUARE
Death sentence 02 «01 «65
given murder (.43) (.68)
indictment.
(DPMURIDT)
Penalty trial of «07 .00 «63
murder conviction { .23) (.94)
(PSEEKALL)
Judge or jury «00 -.01 «58
imposes death {..99) {.9]1)
given penalty
trial (DEATHSNT)
wY7<
These regressions, whose set of independent variables are listed in Appendix
are presented as examples of regressions where both the race of victim
(WHVICRC) and race of defendant (BLACKD) coefficients are not significant, and
are not displayed as a model and are not submitted as being representative of
the Georgia Charging and Sentencing System.
Lrorh 4 tat;
gOSEPH IL. KATZ, Ph.DC/
Sworn to and subscribed
before me this / day
EN
Of J ned ini is XOB4
el 7 din 0
Notary Public,
Hotary Pubiie, Georgia, State at Large
Hy Cormmission Expires Feh 12, 1985
y 3 EG
18
APPENDIX 1
This Appendix lists the independent variables that were used in the
regressions whose results are tabulated in Table 7 of this ®€fidavit.
Dependent Variable = DPMURIDT
Independent Variables = WHVICRC BLACKD ARMROB KIDNAP DEFLED VWITNESS VDEFOLD
NONPROPC DPLEAS EXECUT DDULL DLOWIQ MUTILATE NUDE LDFB2 LDFB4 LDFB10 MAJAGFCX
LDFB7EXP LDFBSUM JEALOUS DEFSUR NOVIOLCR LDFB7F STRANGER DMIDCLAS DPOOR
VHISTAT SCIEVID DCONFESS INCMDRPT CPSTATEM DLEADER RECCRIM DRECPRIS SLODIE
NOKILL AVENGE INSMOT RACEMOT VDEFENLS BDGAG BEAT DEFESC MENTORT PREMEDK THROA
TWOVIC LDFB8 TRIANGLE DEFINTOX DRGHIS RAPE VDEFNSLS VFAMPRES DSELFDEF DCOERCE]
VPCARBR SMUNDRLG DEATHELG DTHINK FELMUR*DRUGDIS DRUGDIS*PROVPASS RAGE*DEFSUR
COPERP*SMUNDRLG FELMUR*PROVPASS ARMROB*VICCLOSE RAPE*MUTILATE ARMROB*VWITNESS
FELMUR*VWITNESS FELMUR*EXECUT FELMUR*UNNECERY TORTURE*PERVER KIDNAP*EXECUT
ARMROB*RAPE ARMROB*KIDNAP NUDE*MUTILATE NUDE*VICPLEAD EXECUT*VICPLEAD
FELMUR*VICPLEAD FELMUR*MITMOTVE FELMUR*EMOTION FELMUR*NOKILL
Dependent Variable = PSEEKALL
Independent Variables = WHVICRC BLACKD ARMROB FELMUR KIDNAP DEFLED VWITNESS
NOVPROV VDEFOLD NONPROPC VSEXUP CONCELBD DPLEAS DRESIST EXECUT DDULL DLOWIQ
MUTILATE NOCLOTH NUDE PLANCOF UNNECERY VICPLEAD LDFBl1 LDFB2 LDFB4 LDFB10
MAJAGFCX MINAGFCX LDFB7EXP LDFBSUM MITMOTVE RAGE JEALOUS REVENGE BLVICMOD
DISFIGHT EMOTION MADLOVER PROVPASS DEFSUR COPERP DEFFEAR SMPROVOK VBADREP
NOVIOLCR VICDRUG MITCIRX LDFB7F VINJURD STRANGER DMIDCLAS DPOOR VHISTAT
SCIEVID DCONFESS TWOWITD INCMDRPT CPSTATEM DLEADER RECCRIM DRECPRIS GUN
DEFWAIT SLODIE DKILLER NOKILL DSHOOT5 VASSAULD VINJDEAR MULHDSHT VIOLENCE FEL
2VMD SHOOTOUT CLEARID MURPRIOR VBED CPSHOOTS5 DNOINT PHYILL MITCIR POISON
DEFSURLT AVENGE INSMOT RACEMOT VDEFENLS VDEFENS ADDCRIME BDGAG BEAT DEFESC
HARMOTH MENTORT NOREMORE NOVICMIT PERVER PREMEDK THROAT TWOVIC LDFB7B LDFRBS8
HATE DRUGDIS SPOUSESK TRIANGLE VICVERB AIDVICT DEFINTOX DEFREMOR DRGHIS
MENTILL BDBLOOD VACCUSED VICARMED LDFB7E RAPE RESBKIN VDEFNSLS VFAMPRES
VSHOMON VICDRUNK STRANGLE DCOERCED DEFHURT VWEAK VPCARBR NOSPAGCR ACCIDENT
SMUNDRLG DEATHELG DTHINK FELMUR*DRUGDIS DRESIST*DRUGDIS ARMROB*VICCLOSE
RAPE*MUTILATE ARMROB*VWITNESS KIDNAP*VWITNESS FELMUR*VWITNESS FELMUR*EXECUT
FELMUR*UNNECERY NUDE*PERVER FELMUR*DPLEAS TORTURE*PERVER KIDNAP*EXECUT
ARMROB*RAPE ARMROB*KIDNAP ARMROB*RAPE*KIDNAP EXECUT*UNNECERY NUDE*MUTILATE
NUDE*VICPLEAD EXECUT*VICPLEAD FELMUR*VICPLEAD FELMUR*MITMOTVE FELMUR*EMOTION
FELMUR*NOKILL FELMUR*PROVPASS PROVPASS*DRUGDIS BLVICMOD*DEFFEAR
DEFFEAR*DEFINTOX RAGE*DEFREMOR RAGE*DEFSUR JEALOUS*PROVPASS DRUGDIS*NOSPAGCR
BLVICMOD*NOSPAGCR SMPROVOK*NOSPAGCR DEFFEAR*DRUGDIS BDBLOOD*DRUGDIS
SMPROVOK*DRUGDIS PROVPASS*VICVERB RAGE*DRUGDIS RAGE*DEFINTOX JEALOUS*DISFIGHT
BLVICMOD*SMPROVOK BLVICMOD*BDBLOOD EMOTION*DEFSUR NOSPAGCR*RAGE
NOSPAGCR*DISFIGHT COPERP*SMUNDRLG FELMUR*NOVPROV .
Dependent Variable = DEATHSNT
Independent Variables = WHVICRC BLACKD MAJAGFCX MINAGFCX MITMOTVE DMIDCLAS
DPOOR DCONFESS INCMDRPT RECCRIM DEFWAIT VINJDEAR AIDVICT DRGHIS
-] Ou