Berghuis v. Smith Brief for the NAACP Legal Defense & Educational Fund as Amicus Curiae in Support of Respondent

    No. 08-1402

In THE

Supreme (Court of the lilntteh States

MARY BERGHUIS, Warden,

v.
DIAPOLIS SMITH,

Petitioner,

Respondent.

On W rit of Certiorari to the 
U nited States Court of A ppeals for the S ixth C ircuit

BRIEF FOR THE NAACP LEGAL DEFENSE & 
EDUCATIONAL FUND, INC., AS AMICUS CURIAE 

IN SUPPORT OF RESPONDENT

V irginia  A. S eitz  
Gary F einerman 
J e ffr e y  T. G r een  
R ebecca  K. T roth 
A nand H . D as 
S idley  A ustin  llp  

1501 K Street, NW 
Washington, DC 20005 
(202) 736-8000

Sarah O’R ourke S chrup 
N orthwestern U niversity 

S upreme C ourt P racticum 
357 East Chicago Avenue 
Chicago, IL 60611 
(312) 503-8576

J ohn P ayton 
Director- Counsel 

D ebo P  A degbile  
C hristina  Swarns* 
J ohanna B. S teinberg  
J in  H ee  L ee

V incent  M. S outherland 
M ary H unter  
NAACP L egal D e fen se  

& E ducational F und , I nc . 
99 Hudson St., 16th Floor 
New York, NY 10013 
(212) 965-2200

Counsel for Amicus Curiae

December 28, 2009 * Counsel of Record



TABLE OF CONTENTS
Page

TABLE OF AUTHORITIES................................  iii
INTEREST OF THE AMICUS CURIAE............  1
SUMMARY OF ARGUMENT.................................  2
ARGUMENT............................................................. 3

I. REPRESENTATIVE DEMOCRACY IS 
THE CENTERPIECE OF THE SIXTH 
AMENDMENT RIGHT TO A JURY 
CHOSEN FROM A FAIR CROSS- 
SECTION OF THE COMMUNITY.................  3
A. The Value Of Representation In Jury-

Service ....................................   3
B. Dureris Prima Facie Test Secures The

Goals Of The Sixth Amendment.............. 5
1. First Prong: Distinctive Groups.........  5
2. Second Prong: Fair And Reasonable

Representation.....................................  6
3. Third Prong: Systematic Exclusion .... 9

II. THE MICHIGAN SUPREME COURT’S 
DECISION CONTRAVENES THIS 
COURT’S FAIR CROSS-SECTION PRE­
CEDENTS ....................................................  9
A. The State Court Proceedings...................  10
B. The Michigan Supreme Court Erred In

Holding That Mr. Smith Failed To 
Demonstrate Constitutionally Signifi­
cant African American Underrepresen­
tation ........................................................  14

(i)



ii
TABLE OF CONTENTS—continued

Page
C. Kent County’s Siphoning Of Prospective 

Jurors From The Circuit Court To The 
District Courts Systematically Excluded 
African Americans From Circuit Court
Venires....................................................... 17

III. AEDPA DID NOT REQUIRE THE 
COURT OF APPEALS TO IGNORE 
KENT COUNTY’S CONSTITUTIONAL 
VIOLATION.................................................  21
A. The Michigan Supreme Court Was 

Objectively Unreasonable In Deter­
mining That African Americans Were 
Not Underrepresented To A Constitu­
tionally Significant Degree In Kent
County’s Circuit Court Venire Pool.........  23
1. The Sixth Amendment’s Fair Cross-

Section Requirement Is Clearly 
Established Law..................................  23

2. The Michigan Supreme Court’s Un­
reasonable Application Of Clearly 
Established Law..................................  25

B. The Michigan Supreme Court Was 
Objectively Unreasonable In Deter­
mining That African Americans Were 
Not Systematically Excluded From Kent
County’s Circuit Court Venire Pool.........  27

CONCLUSION....................................................  31



TABLE OF AUTHORITIES 
CASES Page

Alexander v. Louisiana, 405 U.S. 625
(1972) ......................................................  1,7

Amadeov. Zant, 486 U.S. 214 (1988).........  1
Apodaca v. Oregon, 406 U.S. 404 (1972)....  4
Azania v. State, 778 N.E.2d 1253

(Ind. 2002)................................................  7
Ballard v. United States, 329 U.S. 187

(1946)........................................................  3
Batson v. Kentucky, 476 U.S. 79 (1986).....  2
Carey v. Musladin, 549 U.S. 70 (2006)....... 22, 24
Carter v. Jury Comm’n o f Greene County,

396 U.S. 320 (1970)..................................  1
Castaneda v. Partida, 430 U.S. 482 (1977). 9
Davis v. Warden, 867 F.2d 1003

(7th Cir. 1989)..........................................  30
Duren v. Missouri, 439 U.S. 357

(1979).................................  passim
Edmonson v. Leesville Concrete Co., Inc.,

500 U.S. 614 (1991)..................................  2
Ford v. Seabold, 841 F.2d 677

(6th Cir. 1988)..........................................  28
Georgia v. McCollum, 505 U.S. 42 (1992)... 1
Ham v. South Carolina, 409 U.S. 524

(1973) ......................................................  1
Hobby v. United States, 468 U.S. 339

(1988)........................................................  1
Holland v. Illinois, 493 U.S. 474 (1990).....  3
Johnson v. California, 545 U.S. 162

(2005)........................................................  1, 31
Knowles v. Mirzayance, -  U.S. ", 129 S.Ct.

1411 (2009)...............................................  24
Lockhartv. McCree, 476 U.S. 162 (1986)... 4, 5, 6 
Lockyerv. Andrade, 538 U.S. 63 (2003).....  22



IV

TABLE OF AUTHORITIES—continued
Page

Miller-Elv. Cockrell, 537 U.S. 322 (2003)... 1
M iller-Elv. Dretke, 545 U.S. 231 (2005)....  1
Norris v. Alabama, 294 U.S. 587 (1935).....  16
Panetti v. Quarterman, 551 U.S. 930

(2007)......................................................... 22
Paredes v. Quarterman, 574 F.3d 281

(5th Cir. 2009)..........................................  28
Parents Involved v. Seattle, 551 U.S. 701

(2007)......................................................... 17
People v. Burgener, 62 P.3d 1 (Cal. 2003)... 7
Peters v. Kiff, 407 U.S. 493 (1972)......  4, 5, 6, 9
Ramseur v. Beyer, 983 F.2d 1215

(3d Cir. 1992)............................................  17
Schriro v. Landrigan, 550 U.S. 465

(2007)......................................................... 24
Smith v. Texas, 311 U.S. 128 (1940).......  3, 4
State v. Dixon, 593 A.2d 266 (N.J. 1991).... 7
Statev. Gibbs, 758 A.2d 327 (Conn. 2000).. 7
State v. Lovell, 702 A.2d 261 (Md. 1997) .... 7
State v. Williams, 525 N.W.2d 538

(Minn. 1994).............................................  7
Strickland v. Washington, 466 U.S. 668

(1984)......................................................... 24
Swain v. Alabama, 380 U.S. 202 (1965)......  1
Taylor v. Louisiana, 419 U.S. 522

(1975).....................................................  passim
Turner v. Fouche, 396 U.S. 346 (1970)....... 1
United States v. Green, 435 F.3d 1265

(10th Cir. 2006)........................................  6
United States v. Butler, 615 F.2d 685

(5th Cir. 1980)..........................................  16
United States v. Jackman, 46 F.3d 1240

(2d Cir. 1995)........................................  7, 8, 15
United States v. Maskeny, 609 F.2d 183 

(5th Cir. 1980)..........................................  16



V

TABLE OF AUTHORITIES—continued
Page

United States v. Rogers, 73 F.3d 774
(8th Cir. 1996)...................................... 8, 16, 17

United States v. Weaver, 267 F.3d 231
(3d Cir. 2001)...............................   7,28

Washington v, People, 186 P.3d 594
(Col. 2008).................................................  7

Williams v. Taylor, 529 U.S. 362
(2000).................................................  21,22,24

Wright v. Van Patten, 552 U.S. 120 
(2008)........................................................  24

STATUTE
28 U.S.C. § 2254...................................  26

SCHOLARLY AUTHORITIES
David Kairys, Joseph Kadane and John 

Lehoczky, Jury Representativeness■ A 
Mandate for Multiple Source Lists, 65
Cal. L. Rev. 776 (1977).............................  8,15

Jeffrey Fagan, et. ah, Brief for Social 
Scientists, Statisticians, and Law 
Professors as Amici Curiae Supporting 
Respondent, Berghuis v. Smith (No. 08- 
1402) 26



INTEREST OF AMICUS CURIAE1
The NAACP Legal Defense & Educational Fund, 

Inc. (“LDF”), is a non-profit corporation formed to 
assist African Americans and other people of color 
who are unable, on account of poverty, to employ 
legal counsel to secure their rights through the 
prosecution of lawsuits. LDF has a long-standing 
concern with democracy, representation, fair jury 
service and selection, and the influence of racial 
discrimination on the criminal justice system. LDF 
has therefore represented defendants and/or served 
as amicus curiae in numerous jury 
underrepresentation and jury selection cases before 
this Court including, inter alia, Amadeo v. Zant, 486 
U.S. 214 (1988); Hobby v. United States, 468 U.S. 339 
(1984); Ham v. South Carolina, 409 U.S. 524 (1973); 
Alexander v. Louisiana, 405 U.S. 625 (1972) and 
Swain v. Alabama, 380 U.S. 202 (1965).

LDF also pioneered the affirmative use of civil 
actions to end jury discrimination in Carter v. Jury 
Comm’n o f Greene County, 396 U.S. 320 (1970), and 
Turner v. Fouche, 396 U.S. 346 (1970); and appeared 
as amicus curiae in cases challenging the 
discriminatory use of peremptory challenges. See 
Miller-El v. Dretke, 545 U.S. 231 (2005); Johnson v. 
California, 545 U.S. 162 (2005); Miller-El x. Cockrell, 
537 U.S. 322 (2003); Georgia v. McCollum, 505 U.S. 
42 (1992); Edmonson v. Leesville Concrete Co., Inc.,

1 No counsel for a party authored this brief in whole or in 
part, and no such counsel or party made a monetary 
contribution intended to fund the preparation or submission of 
this brief. No person other than the amicus curiae, or its 
counsel, made a monetary contribution intended to fund its 
preparation or submission. The parties have consented to the 
filing of this brief and such consents are being lodged herewith.



2
500 U.S. 614 (1991); and Batson v. Kentucky, 476 
U.S. 79 (1986).

SUMMARY OF ARGUMENT
The right to a jury drawn from a fair and 

representative cross-section of the community is a 
core value of the Sixth Amendment and is essential to 
the integrity and legitimacy of the judicial system. 
By embodying the value of representative democracy, 
the fair cross-section requirement secures the right to 
a fair trial, guarantees that all members of distinctive 
groups in the community will play a role in the 
administration of justice, and allows citizens to serve 
as a check on the government. These are the issues 
at the heart of Respondent Diapolis Smith’s case.

The United States Court of Appeals for the Sixth 
Circuit correctly found that Mr. Smith’s Sixth 
Amendment right was violated after he was convicted 
by a jury in Kent County, Michigan, that was drawn 
from a pool from which African Americans were 
systematically excluded. Mr. Smith established that 
in the month his jury was selected, African 
Americans were 34 percent less likely to be in the 
venire pool relative to their jury-eligible percentage 
in the County. He also presented the uncontradicted 
testimony of Kent County officials who admitted that 
the County’s established practice of “siphoning” 
prospective jurors away from its circuit court and to 
its district courts caused African Americans to be 
underrepresented in the circuit court venire pool, and 
that the practice was ultimately terminated because 
of those disparities.

Although this uncontested evidence proved that 
Kent County’s jury-selection system violated 
Mr. Smith’s Sixth Amendment rights, the Michigan 
Supreme Court rejected his claim by misapplying this



3
Court’s fair cross-section precedents. Its decision was 
not only incorrect, but also objectively unreasonable 
under the Anti-Terrorism and Effective Death 
Penalty Act (“AEDPA”).

ARGUMENT
I. REPRESENTATIVE DEMOCRACY IS THE 

CENTERPIECE OF THE SIXTH AMENDMENT 
RIGHT TO A JURY CHOSEN FROM A FAIR 
CROSS-SECTION OF THE COMMUNITY.
A. The Value Of Representation In Jury Service

As explained by Respondent and supporting amici, 
Duren v. Missouri, 439 U.S. 357 (1979), is the 
culmination of this Court’s long line of decisions 
emphasizing the unique constitutional importance of 
jury representativeness.2 This Court has repeatedly 
recognized that all qualified segments of the 
population must have the opportunity to participate 
in the jury process. See Smith v. Texas, 311 U.S. 128, 
130 (1940). Thus, “those eligible for jury service are 
to be found in every stratum of society,” Ballard v. 
United States, 329 U.S. 187, 193 (1946), and the 
systematic exclusion from jury service of distinctive 
groups is “at war with our basic concepts of a 
democratic society and a representative government.” 
Smith, 311 U.S. at 130; see also Lockhart v. McCree, 
476 U.S. 162, 175 (1986) (explaining that the

2 In advancing this argument, am icus curiae does not assert 
that Mr. Smith has the right to a petit jury “of any particular 
composition.” Taylor v. Louisiana, 419 U.S. 522, 538 (1975). 
Rather, the Sixth Amendment fair cross-section right 
contemplates a “fair possibility” of a representative jury, which 
is satisfied by the “inclusion of all cognizable groups in the 
venire.” H olland v. Illinois, 493 U.S. 474, 478 (1990) (citation 
and in ternal quotation marks omitted).



4
wholesale exclusion of women, African Americans 
and Mexican Americans “clearly contravened . . . the 
purposes of the fair-cross-section requirement”)! 
Apodaca v. Oregon, 406 U.S. 404, 413 (1972) (“[T]he 
Constitution forbids. . . systematic exclusion of 
identifiable segments of the community from jury 
panels and from the juries ultimately drawn from 
those panels,” and groups have “the right to 
participate in the overall legal processes by which 
criminal guilt and innocence are determined.”).

Although “[t]he principle of the representative jury 
was first articulated by this Court as a requirement 
of equal protection,” it long ago “emerged as an aspect 
of the [Sixth Amendment] . . . right to jury trial.” 
Peters v. Kiff, 407 U.S. 493, 500 n. 9 (1972). Thus, in 
Taylor v. Louisiana, 419 U.S. 522, 528 (1975), this 
Court made explicit that “the selection of a petit jury 
from a representative cross section of the community 
is an essential component of the Sixth Amendment 
right to a jury trial.”

Soon thereafter, in Duren v. Missouri, this Court 
established a three-part test for demonstrating a 
prima facie case of a Sixth Amendment fair cross- 
section violation. 439 U.S. at 357. Specifically, this 
Court stated that:

In order to establish a prima facie violation of 
the fair cross-section requirement, the defendant 
must show (l) that the group alleged to be 
excluded is a ‘distinctive’ group in the 
community! (2) that the representation of this 
group in venires from which juries are selected is 
not fair and reasonable in relation to the number 
of such persons in the community! and (3) that 
this underrepresentation is due to systematic 
exclusion of the group in the jury-selection 
process.



5
439 U.S. at 364. Once the prima facie showing has 
been made, “it is the State that bears the burden of 
justifying [its infringement on the Sixth Amendment 
fair cross-section right] by showing attainment of a 
fair cross section to be incompatible with a significant 
state interest.” Id. at 368.

B. Dureiis Prima Facie Test Secures The Goals 
Of The Sixth Amendment.

Each prong of Dureiis prima facie test furthers the 
purposes of the Sixth Amendment’s fair cross-section 
requirement: “guarding] against the exercise of 
arbitrary power [by] makting] available the 
commonsense judgment of the community as a hedge 
against. . . bias,” Taylor, 419 U.S. at 530, “preserving 
‘public confidence in the fairness of the criminal 
justice system,’ and implementing our belief that 
‘sharing in the administration of justice is a phase of 
civic responsibility.’” Lockhart v. McCree, 476 U.S. at 
174-75 (quoting Taylor, 419 U.S. at 530-31).

1. First Prong: Distinctive Groups
This Court has not “precisely define [d] the term 

‘distinctive group,”’ but it has made clear that its 
meaning is “linked to the purposes of the [Sixth 
Amendment’s] fair cross-section requirement.” Id. at 
174. Thus, women, African Americans and Mexican 
Americans have been properly designated distinctive 
groups because their exclusion offends the Sixth 
Amendment by undermining the legitimacy of the 
criminal justice system, denying the fundamental 
fairness guaranteed by an impartial jury trial, and 
removing a unique voice in the community from 
participation in the administration of justice. Id. at 
175; see also Peters, 407 U.S. at 499-500, 503-04 
(describing how exclusion of African Americans from



6
jury service thwarts the purposes served by the Sixth 
Amendment).3

2. Second Prong: Fair And Reasonable 
Representation

Because “the selection of a petit jury from a 
representative cross section of the community is an 
essential component of the Sixth Amendment right to 
a jury trial,” Taylor, 419 U.S. at 528, courts must 
ensure that the representation of distinctive groups 
in venires is “fair and reasonable” relative to their 
presence in the community. This is, by definition, a 
context-sensitive evaluation. Since distinctive group 
size and composition vary from county to county, the 
determination whether venire representation fulfills 
the Sixth Amendment’s requirements must take into 
account all individual circumstances.

Thus, there is no “one size fits all” mathematical 
standard or formula for measuring the 
reasonableness of a distinctive group’s representation

3 Petitioner’s contention that smaller groups (groups th a t 
constitute less than  10% of the venire pool) are not “distinctive 
groups” for this purpose, Pet. Br. a t 45-46, is wrong because it 
would eviscerate the fair cross-section requirem ent in most 
counties in this country and contravene the purposes of the 
requirem ent as set forth in Duren and Taylor. To the contrary, 
a group is distinctive whenever its exclusion implicates the 
Sixth Amendment. See, e.g. Lockhart, 476 U.S. a t 174-75; 
Peters, 407 U.S. a t 499-500, 503-04. Thus, federal courts of 
appeals determine whether a group is “distinctive” by examining 
whether “(l) [ ] the group is defined by a limiting quality (i.e., 
th a t the group has a definite composition such as by race or sex); 
(2) a common thread or basic similarity in attitude, idea, or 
experience runs through the group,' and (3) [there is a] 
community of interests among members of the group such that 
the group's in terest cannot be adequately represented if the 
group is excluded from the jury selection process.” U nited S ta tes  
v. Green, 435 F.3d 1265, 1271 (10th Cir. 2006) (gathering cases).



7
in venires. See Alexander, 405 U.S. at 630 (noting, in 
an equal protection case, that “[t]his Court has never 
announced mathematical standards for the 
demonstration of ‘systemic’ exclusion” of distinctive 
groups). Variations in population figures and factual 
scenarios have led federal appellate courts to consider 
different measures of representativeness depending 
on the size of the distinctive group. See United 
States v. Weaver, 267 F.3d 231, 242-243 (3d Cir. 
2001) (considered facts to determine appropriate 
measure of representativeness); United States v. 
Jackman, 46 F.3d 1240, 1246-47 n.4 (2d Cir. 1995) 
(choosing measure of representativeness after 
considering facts).4

The most frequently used measures of 
representativeness are the absolute disparity and 
comparative disparity tests. “The absolute disparity 
test compares the members of a distinctive group that 
are jury eligible with those that appear on the

4 Many state supreme courts do the same. See W ashington v. 
People, 186 P.3d 594, 605 (Colo. 2008) (en banc) (advancing 
argum ent th a t “all the statistical evidence presented” should be 
examined by a court facing a fair cross-section claim); People v. 
Burgener, 62 P.3d 1, 22 (Cal. 2003) (noting th a t the California 
Supreme Court has declined to adopt a particular measure of 
underrepresentation); Azania v. State, 778 N.E.2d 1253, 1260 
(Ind. 2002) (applying both absolute and comparative disparity); 
S ta te  v. Gibbs, 758 A,2d 327, 336-37 (Conn. 2000) (explaining 
tha t the choice of proper statistical method is fact driven); State  
v. Lovell, 702 A.2d 261, 281 (Md. 1997) (advising M aryland state 
courts to use multiple measures and supplement absolute 
disparity figures with comparative disparity figures when the 
distinctive group population falls below 10 percent); Sta te  v. 
Williams, 525 N.W.2d 538, 543 (Minn. 1994) (validating reliance 
on multiple statistical measures of venire representation); Sta te  
v. Dixon, 593 A.2d 266, 271 (N.J. 1991) (looking to multiple tests 
of underrepresentation in examining a fair cross-section and 
equal protection challenge to jury selection process).



8
venire.” Pet. App. 18a. The difference between these 
two data ■ the distinctive group’s jury eligible 
population figure and presence on the venire - is the 
absolute disparity. The comparative disparity 
standard, in contrast, measures the percentage by 
which the probability of serving on a jury venire is 
reduced for the members of the distinctive group. 
Comparative disparity is calculated by dividing the 
absolute disparity by the population figure of the 
distinctive group at issue. David Kairys, Joseph 
Kadane & John Lehoczky, Jury Representativeness- 
A Mandate for Multiple Source Lists, 65 Cal. L. Rev. 
776, 790 (1977).

While the absolute disparity method may have 
some utility when the minority population is 
relatively large, the approach is of little value when 
applied to an underrepresented group that is a 
smaller percentage of the total population. Where, as 
here, the minority population comprises less than 10 
percent of the population, the absolute disparity test, 
with a maximum allowable disparity of 10 percent, 
yields meaningless results. Underrepresentations of 
small and medium-sized minority groups would be 
validated under this approach because, by definition, 
a smaller group can never have a large absolute 
disparity. See Kairys, supra, 65 Cal. L. Rev. at 795- 
96. In such circumstances, applying the absolute 
disparity measurement improperly allows total 
exclusion of the minority group. See United States v. 
Rogers, 73 F.3d 774, 776-77 (8th Cir. 1996); Jackman, 
46 F.3d at 1247. As the Sixth Circuit properly found 
below, the comparative disparity measurement is the 
“more appropriate measure of underrepresentation” 
where “the distinctive group alleged to have been 
underrepresented is small.” Pet. App. 20a.



9
3. Third Prong: Systematic Exclusion

When the underrepresentation of a distinctive 
group is “inherent in the particular jury-selection 
process” used, it is deemed “systematic.” Duren, 439 
U.S. at 366. Fair cross-section claims, unlike equal 
protection claims, do not require proof of intentional 
discrimination. See id. at 368 n.26. Consistent with 
their distinct purposes, the Sixth Amendment focuses 
on exclusion caused by the jury selection process, 
while the Equal Protection Clause is concerned with 
intentional discrimination by government officials 
involved in the jury selection process. See Castaneda 
v. Partida, 430 U.S. 482, 492-94 (1977) (addressing 
equal protection challenge to intentional racial 
discrimination in selection of grand jurors); Peters, 
407 U.S. at 499-500 (describing constitutional values 
offended by purposeful exclusion of African 
Americans from jury service). While racially-based 
exclusion from jury service may be relevant to finding 
a Sixth Amendment violation, intentional 
discrimination is not necessary to such a claim, and 
the “disproportion itself demonstrates an 
infringement of the defendant’s interest in a jury 
chosen from a fair community cross section.” Duren, 
439 U.S. at 368 n. 26. The Michigan Supreme 
Court’s decision must be evaluated within the context 
of this legal framework, which has been clearly 
established by this Court’s precedent.
II. THE MICHIGAN SUPREME COURT’S 

DECISION CONTRAVENES THIS COURT’S 
FAIR CROSS-SECTION PRECEDENTS.

The Michigan Supreme Court properly identified 
Duren s three-prong test as the standard for 
evaluating Mr. Smith’s fair cross-section claim. Pet. 
App. 144a. Its determination that Mr. Smith failed to



10
prove a Sixth Amendment violation cannot, however, 
be reconciled with this Court’s precedents.

A. The State Court Proceedings
Mr. Smith first raised his Sixth Amendment fair 

cross-section claim during jury selection for his Kent 
County Circuit Court trial. JA 3a-7a. The circuit 
court rejected Mr. Smith’s claim, but, on appeal, the 
Michigan Court of Appeals remanded his case for an 
evidentiary hearing to determine whether his jury 
was drawn from a fair cross-section of the 
community. Pet. App. 176a.

At the evidentiary hearing, Mr. Smith, an African 
American, presented evidence that, although African 
Americans constituted 7.28 percent of the potential 
jurors in Kent County, Pet. App. 179a, 214a, his 
venire panel of 60 to 100 prospective jurors included, 
at most, three African Americans. JA 6a-7a. The 
parties examined 37 potential jurors during voir dire, 
all of whom were white. Pet. App 86a.

Dr. Michael Stoline, testifying as an expert 
statistician in support of Mr. Smith, established that 
in September 1993, the month of Mr. Smith’s jury 
selection, African Americans were underrepresented 
by 34.8 percent, using a comparative disparity 
measure. JA 181a; Pet. App. 218a. In the six months 
preceding Mr. Smith’s trial, African Americans were 
underrepresented by an average of 18 percent.5 Id.

5 Dr. Stoline testified as to the number of African-American 
prospective jurors expected to appear on juror lists in each of the 
six months from April 1993 to October 1993. JA 217a. In the 
month tha t Mr. Smith’s jury was selected, an unbiased selection 
process should have produced 11.5 African-American 
prospective jurors on the juror lists. Id. a t 218a. Yet, under the 
County’s selection process, one could expect to see only 7.5 
African Americans for circuit court venires. This resulted in an



11
Through the uncontested testimony of Kim Foster, 

Kent County’s Circuit Court Administrator, and 
Richard Hillary, Director of Kent County’s Public 
Defender’s Office and Co-Chair of the Jury Minority 
Representation Committee of the Grand Rapids Bar, 
Mr. Smith also presented evidence about the method 
of allocating prospective jurors to Kent County’s 12 
district courts and one circuit court. The circuit court 
is a court of general jurisdiction that presides over 
felony criminal cases, like Mr. Smith’s case, 
throughout Kent County. By contrast, the district 
courts cover discrete geographic areas within the 
County and have limited criminal jurisdiction, 
hearing only misdemeanor criminal cases and 
preliminary criminal hearings.

Prior to October 1993, Kent County had a practice 
of filling the district courts’ juror needs before 
assigning any prospective jurors to the circuit court 
panels. JA 20a, 22a! Pet. App. 8a. Thus, it was only 
after the 12 district courts received their prospective 
jurors that the remaining prospective jurors were 
made available for service in the circuit court. Id. 
The district courts used only prospective jurors who 
resided in their districts, while the circuit court used 
prospective jurors from throughout the County. Pet. 
App. 176a. Prospective jurors who were not selected 
for district court service were not returned to the 
circuit court jury pools. JA 64a -  65a.

Kent County jury selection officials admitted that 
this practice of assigning prospective jurors to district 
courts before the circuit court effectively “siphoned”

estimated underrepresentation of four prospective African- 
American jurors (or 34.8 percent) from the 11.5 figure that 
would have occurred without the selection bias. Id. Further 
details on the statistical methods employed by Dr. Stoline can be 
found in Resp. Br. a t 6-8.



12
jurors away from circuit court venire panels. Pet. 
App. 176a-77a, 168a. The qualified jurors siphoned 
away from the circuit court included residents of the 
City of Grand Rapids, which, at the time, was 18.5 
percent African-American and contained 85 percent 
of Kent County’s African-American population. JA 
197a! Pet. App. 29a. As Mr. Foster, the circuit court 
administrator, testified, the “districts essentially 
swallowed up most of the minority jurors,” leaving 
the circuit court with a jury pool that “did not 
represent the entire county.” JA 22a.

Contrary to Petitioner’s contention, Pet. Br. at 61, 
Kent County jury officials were unequivocal in their 
testimony that the systematic siphoning significantly 
reduced the number of African-American prospective 
jurors available to the circuit court jury pool. 
Indeed, Mr. Foster testified that one month after 
Mr. Smith’s September 1993 jury selection, the 
County ended the siphoning practice because of the 
resulting racial disparities in the circuit court jury 
pool. JA 20a, 22a. He further noted that this change 
allowed the circuit court to access a “larger 
representation of the community [from which] to 
select [jurors],” thus remedying the “problem” of 
African-American underrepresentation. Id. at 22a. 
In addition, Mr. Hillary, the Co-Chair of the Jury 
Minority Representation Committee, testified that 
after “months of study,” his committee “made a 
determination that we were losing minorities by 
choosing the District Court jurors first and not 
returning the unused ones to . . . the pool that the 
Circuit Court was [choosing its jurors] from.” Id. at 
64a-65a.6

6 As detailed in Resp. Br. a t 48-50, Mr. Smith also presented 
evidence tha t Kent County’s system of automatically accepting



13
At the conclusion of the evidentiary hearing, the 

Kent County Circuit Court found 
underrepresentation of African Americans in the 
circuit court venires, but ruled that there was no 
systematic exclusion, thus denying Mr. Smith’s fair 
cross-section claim. Pet. App. 209a, 212a.

The Michigan Court of Appeals reversed the circuit 
court. In addition to holding that the “representation 
of blacks in venires from which juries were selected 
for Kent County Circuit Courts was at the time, i.e., 
before October 1, 1993, not fair and reasonable in 
relation to the number of such persons in the 
community,” the court also held that “the evidentiary 
record establishes that the juror allocation system in 
place before October 1993 [siphoning] drained the 
largest concentration of African-Americans from the 
master jury list.” Id. at 18la-182a. Thus, the court 
ruled that Mr. Smith was entitled to a new trial. Id. 
at 183a.

The Michigan Supreme Court reversed the Court of 
Appeals. Although the Michigan Supreme Court held 
that African Americans are a distinctive group under 
Dureiis first prong, id. at 159a, it ruled that 
Mr. Smith’s “statistical evidence failed to establish a 
legally significant disparity under either the absolute 
or comparative disparity tests.” Id. at 146a. In 
reaching its result, the court appeared to treat the 
absolute disparity test7 and other measurements of 
underrepresentation as equally valid, even though 
use of the absolute disparity test would have

non-statutory excuses systematically excluded African American 
prospective jurors.

7 The state court appears to have applied an absolute 
disparity test with a 10 percent threshold. This is consistent 
with Petitioner’s recommendation in this case.



14
sanctioned the complete exclusion of African 
Americans from the jury pool. See id. The court then 
proceeded to give Mr. Smith the “benefit of the doubt” 
as to his claim of underrepresentation, and held that 
Mr. Smith had not demonstrated that African 
Americans were systematically excluded from Kent 
County’s circuit court venires. Id. at 147a-49a, 169a- 
71a.

One concurring judge noted that the risks of using 
an absolute disparity measurement are “particularly 
acute when the percentage of the distinctive group 
eligible for jury service is small.” Id. at 161a. He 
explained that in the instant case, “a complete 
exclusion of black jurors would result in only a 7.28 
percent absolute disparity.” Id.

B. The Michigan Supreme Court Erred 
In Holding That Mr. Smith 
Failed To Demonstrate Constitutionally 
Significant African American
Underrepresentation.

In September 1993, the month of Mr. Smith’s jury 
selection, African Americans were underrepresented 
by 34.8 percent, using a comparative disparity 
measure. JA 181a-182a! Pet. App. 218a; see also 
supra at 10. Although in the six months preceding 
his trial, African Americans were underrepresented 
by an average of 18 percent, many individual 
defendants, including Mr. Smith, faced even higher 
underrepresentation on their venires. The extent of 
the systematic underrepresentation on any particular 
venire impermissibly depended on the extent of the 
siphoning to district court juries at the given time. 
Thus, for example, in June 1993, the circuit court 
disparity spiked to 42.2 percent. Id. This 
demonstrated that African Americans were not fairly



15
and reasonably represented on Kent County Circuit 
Court venires relative to their size in the community.

Despite this evidence, the Michigan Supreme Court 
held that while Mr. Smith “presented some evidence 
of a disparity between the number of jury-eligible 
African-Americans and the actual number of African- 
American prospective jurors selected to the Kent 
County Circuit Court jury pool list,” Pet. App. 146a, 
he failed to satisfy Dureris second requirement. Id. 
In reaching this conclusion, however, the court 
improperly relied on the absolute disparity test, 
which produces unreliable results when used to 
measure disparities in smaller distinctive group 
populations, such as the African-American 
community in Kent County. See Jackman, 46 F.3d at 
1247 (questioning validity of absolute disparity for 
smaller populations due to risk of selecting “a large 
number of venires in which members of the group are 
substantially underrepresented or even totally 
absent”); Kairys, supra, 65 Cal. L. Rev. at 793 
(criticizing absolute disparity as an inappropriate 
measure for smaller populations).

At the time of Mr. Smith’s trial, African Americans 
in Kent County comprised 7.28 percent of the adult 
population. JA 172aJ Pet. App. 179a, 214a. The 
absolute disparity test was, therefore, an 
inappropriate method of measuring the 
reasonableness of African-American representation 
because that test would condone the complete 
exclusion of Kent County’s African-American 
community in direct contravention of the Sixth 
Amendment’s fair cross-section requirement. This 
Court and federal courts of appeals have 
acknowledged that such an outcome cannot be 
reconciled with this Court’s fair cross-section 
precedents. See e.g., Taylor, 419 U.S. at 530-31



16
(explaining that the purposes of a jury cannot be 
fulfilled when distinctive groups are excluded from 
the venire); Rogers, 73 F.3d at 776-77 (recognizing 
that absolute disparity insulates wholesale exclusion 
from challenge because “the percentage disparity can 
never exceed the percentage of [a distinctive group] in 
the community.”); United States v. Butler, 615 F.2d 
685, 686 (5th Cir. 1980) (clarifying circuit precedent 
by stating that in cases where a “less-than-10- 
[percent] minority” was at issue, other statistical 
measures would be considered); United States v. 
Maskeny, 609 F.2d 183, 190 (5th Cir. 1980)
(considering argument that “reliance on absolute 
disparity could lead to approving the total exclusion 
from juries of a minority that comprised less than ten 
percent of the population of the community” and 
avoiding speculation on the appropriate measure to 
confront such a situation). The Michigan Supreme 
Court thus erred in considering an absolute disparity 
analysis when evaluating Mr. Smith’s evidence of 
African-American underrepresentation.

Given the size of Kent County’s African-American 
population, comparative disparity was the only 
measure that could effectively evaluate the extent to 
which African Americans were underrepresented on 
Kent County circuit court venires. See, e.g., Norris v. 
Alabama, 294 U.S. 587, 596-97 (1935) (finding an 
equal protection violation in the total exclusion of 
African Americans from jury service in a jurisdiction 
where African Americans constituted less than eight 
percent of the relevant population). The results of a 
comparative disparity analysis lead to the inexorable 
conclusion that the underrepresentation of African 
Americans in the Kent County Circuit Court jury pool 
was constitutionally significant and satisfied Dureiis 
second prong. See Rogers, 73 F.3d at 777



17
(comparative disparity of more than 30 percent 
proves underrepresentation under Duren’s second 
prong); Ramseur v. Beyer, 983 F.2d 1215, 1232 
(3d Cir. 1992) (comparative disparity of 40 percent 
establishes underrepresentation).8

C. Kent County’s Siphoning Of Prospective 
Jurors From The Circuit Court To The 
District Courts Systematically Excluded 
African Americans From Circuit Court 
Venires.

Although the Michigan Supreme Court recognized 
that “minority jury representation in Kent County 
has long been a problem,” Pet. App. 156A, it 
concluded that African Americans were not 
systematically excluded from Kent County Circuit

8 Petitioner suggests tha t reliance on the comparative 
disparity test prevents fair cross-section violations only if the 
county “creates a self-conscious process” th a t “identif[ies] the 
race and ethnicity of every prospective juror,” thereby 
“effectively requir[ing] a quota system.” Pet. Br. a t 43. This 
invocation of quotas is a red herring that this Court should 
ignore. First, as previously noted, comparative disparity is a 
legitimate and effective measure of representation endorsed by 
numerous courts and experts. Second, regardless of the method 
of proof, a quota is never a proper remedy for a Sixth 
Amendment fair cross-section violation. Quotas are applicable 
only to very limited circumstances involving demonstrated 
intentional discrimination. See Parents Involved  v. Seattle, 551 
U.S. 701, 795-796 (2007) (Kennedy, J., concurring) (describing 
the “allocation of benefits and burdens through individual racial 
classifications” as “sometimes permissible in the context of 
remedies” for a “de jure” wrong like “segregation”,). Finally, both 
Duren and Taylor expressly state that the fair cross-section 
mandate does not require “that petit juries actually chosen must 
m irror the community,” but rather m ust simply prevent the 
systematic underrepresentation of distinctive groups in jury 
pools. Duren, 439 U.S. a t 364 n.20 (quoting Taylor, 419 U.S. a t 
538). That requirem ent is not a quota, and will not lead to one.



18
Court jury pools. This conclusion cannot be 
reconciled with this Court’s precedents. Kent 
County’s jury selection process systematically 
excluded African Americans by siphoning prospective 
jurors away from the circuit court and to the district 
courts. Accordingly, Mr. Smith satisfied Dureiis 
third prong.

As detailed above, 85 percent of Kent County’s 
African-American population lived in Grand Rapids, 
which composes one of the 12 districts. As previously 
noted, it was Kent County ’s practice to prioritize 
district courts over the circuit court for the 
assignment of prospective jurors, allowing only 
residents of the district in which the district court 
sits to serve on juries in that court. Also, prospective 
jurors were excused after district court service rather 
than returned to the jury pool for circuit court 
service. As a consequence, the district courts 
“siphoned” African Americans away from circuit court 
jury service. This process was “inherent” to Kent 
County’s jury selection system, see Duren, 439 U.S. 
at 366, and resulted in the persistent 
underrepresentation of African Americans in circuit 
court venires, with 34 percent underrepresentation in 
the month of Mr. Smith’s voir dire, an average of 18 
percent underrepresentation over six months, and a 
high of 42 percent just two months before Mr. Smith’s 
jury selection. JA 181a-82al Pet. App. 218a.

Precise numbers documenting how many 
prospective African-American jurors were affected by 
this process were unavailable because the County 
chose not to collect race data.9 However, as the Sixth

9 Petitioner is incorrect to the extent tha t it suggests tha t 
underrepresentation can be shown only by actual race data, as 
opposed to extrapolations from a data set such as the 1990



19
Circuit explained, it logically follows that the jury 
selection process resulted in “fewer African 
Americans [being] available to serve as jurors for 
county circuit court[ ].” Pet. App. 29a.

Moreover, the Kent County Circuit Court 
Administrator readily admitted that the siphoning 
used by Kent County to select circuit court jury pools 
at the time of Mr. Smith’s trial systematically 
excluded African Americans. See supra at 12. 
Indeed, one month after Mr. Smith’s September 1993 
jury selection, the County ended the siphoning 
practice because it believed that siphoning caused 
unacceptable African-American underrepresentation 
in the circuit court jury pool. JA 20a, 22a. The 
Circuit Court Administrator noted that ending the 
siphoning practice allowed the circuit court to access 
a “larger representation of the community to select 
[jurors] from,” thus remedying the “problem” of 
African-American underrepresentation. Id. at 22a.

The record also reflects that once the County 
terminated its siphoning practice, the racial 
disparities began to fade. Dr. Stoline’s analysis 
included data covering the eleven months following 
Mr. Smith’s jury selection, beginning in October 1993 
when Kent County officials testified they 
discontinued siphoning African Americans away from 
the circuit court and towards the district courts. JA 
20a, 22a. Using a method consistent with that used 
for his analysis of the prior six months, Dr. Stoline 
found that African Americans were underrepresented

census. No court has so held; in fact, Duren expressly permitted 
the use of census data. See Duren, 439 U.S. at 364-65. 
Moreover, adoption of Petitioner’s position would allow court 
adm inistrators in communities with smaller, but sizeable, 
minority populations to foreclose fair cross-section challenges 
simply by choosing not to collect race data.



20
by 15.1 percent in the eleven-month post-siphoning 
period studied. Id. at 102a-03a. Extremely high 
levels of underrepresentation, similar to those 
occurring in the siphoning period, were concentrated 
in two of the first five months of the post-siphoning 
period: African Americans were underrepresented by
41.1 percent in the third month and 43.5 percent in 
the fifth month. Id. at 102a. These spikes appear to 
have faded away over time such that the highest 
African-American underrepresentation in the final 
six months of the eleven-month post-siphoning study 
was 22.2 percent, occurring in the penultimate 
month, and the average for the final six months was
15.1 percent. Id. These figures suggest that it took 
several months before the positive effects of the new 
non-siphoning system were reflected in the data. 
Petitioner’s contrary suggestion, Pet. Br. at 50, 61, 
that the post-siphoning data somehow proves that 
the systematic removal of over three-quarters of the 
county’s African Americans from the qualified pool 
never had any effect -  is therefore baseless.

Despite this undisputed evidence, the Michigan 
Supreme Court held that the third Duren prong was 
not satisfied. This conclusion is incorrect and 
unreasonable because the practice of allocating 
prospective jurors to the district courts before the 
circuit court necessarily “siphoned” African 
Americans away from the circuit court, leaving a 
persistent and serious underrepresentation in that 
court. This logical inference is supported by 
testimony from institutional actors within the jury 
selection system, see supra at 11-12, and by the 
recommendation of the study committee that ended 
the practice, JA 64a_65a. This evidence has 
dispositive weight because it reflects the most 
pertinent and knowledgeable observations of the



21
effect of siphoning in Kent County’s jury selection 
system. The Michigan Supreme Court’s decision that 
Mr. Smith failed to establish systematic exclusion of 
African Americans is incorrect.
III. AEDPA DID NOT REQUIRE THE

COURT OF APPEALS TO IGNORE
KENT COUNTY’S CONSTITUTIONAL
VIOLATION.

As shown above, the Michigan Supreme Court 
erred in rejecting Mr. Smith’s Sixth Amendment fair 
cross-section claim. Under AEDPA, however, a 
federal court may not grant habeas relief merely 
because it believes, in its own “independent 
judgment,” that the state court “erroneously or 
incorrectly” decided a federal constitutional claim. 
Williams v. Taylor, 529 U.S. 362, 411 (2000) 
(O’Connor J., concurring). Rather, a federal habeas 
court must determine that the decision was an 
“objectively unreasonable” application of “clearly 
established” United States Supreme Court law, as 
properly held by the Sixth Circuit below. Id. at 409 
(O’Connor, J., concurring).

The phrase “clearly established law” “refers to the 
holdings, as opposed to the dicta, of this Court’s 
decisions as of the time of the relevant state-court 
decision.” Id. at 365. “[Rjules of law may be 
sufficiently clear for habeas purposes even when they 
are expressed in terms of a generalized standard 
rather than a bright-line rule.” Id. at 382. States are 
bound by clearly established law regardless whether 
it is expressed in general or specific terms, and state 
court decisions applying such law must be 
reasonable.

AEDPA does not require state and federal courts to 
wait for some nearly identical factual pattern before a



22
legal rule must be applied.” Carey v. Musladin, 549 
U.S. 70, 80-81 (2006) (Kennedy, concurring). Nor 
does AEDPA prohibit a federal court from finding an 
application of a principle unreasonable when it 
involves a set of facts “different from those of the case 
in which the principle was announced.” Lockyer v. 
Andrade, 538 U.S. 63, 76 (2003). The statute 
recognizes, to the contrary, that even a general 
standard may be applied in an unreasonable manner. 
Panetti v. Quarterman, 551 U.S. 930, 953 (2007) 
(citing Williams, supra). Thus, “[i]f the state court’s 
adjudication is dependent on an antecedent 
unreasonable application of federal law, . . . the 
federal court must then resolve the claim without the 
deference AEDPA otherwise requires.” Id. at 931 
(citing Wiggins v. Smith, 539 U.S. 510, 534 (2003)); 
see also Williams, 529 U.S. at 412 (O’Connor, J„ 
concurring).

For the reasons detailed below, the Sixth Circuit 
properly applied this precedent and determined that 
while the Michigan Supreme Court “identified] the 
correct governing legal principle from this Court’s 
decisions,” it “unreasonably applie[d] that principle to 
the facts of [Mr. Smith’s] case.” Williams, 529 U.S. at 
412.



23
A. The Michigan Supreme Court Was 

Objectively Unreasonable In Determining 
That African Americans Were Not 
Underrepresented To A Constitutionally 
Significant Degree In Kent County’s Circuit 
Court Venire Pool.
1. The Sixth Amendment’s Fair Cross- 

Section Requirement Is Clearly 
Established Law.

In applying Dureiis first step, the Michigan 
Supreme Court correctly ruled that African
Americans constituted a distinctive group under the 
Sixth Amendment’s fair cross-section requirement. 
With respect to Dureiis second step, the Michigan 
Supreme Court concluded that because each of the 
various measures of representation had
shortcomings, it would “consider all these approaches 
to measuring whether representation was fair and 
reasonable.” Pet. App. 146a. This is an accurate 
recitation of this Court’s clearly established law.

Because the degree of underrepresentation 
necessary to implicate these Sixth Amendment 
concerns and to satisfy Dureiis second step will 
naturally vary with the size of the distinctive group 
and the size of the counties from which the venires 
are drawn, Duren requires exactly such a case- 
specific approach. Thus, the Michigan Supreme 
Court properly identified this Court’s clearly 
established law.

The fact that Dureiis assessment of “fair and 
reasonable representation” must be conducted on a 
case-bycase basis does not mean that the test lacks 
the specificity necessary to constitute clearly 
established law under AEDPA. To the contrary, as 
detailed in Respondent’s Brief at 55-57, Dureiis test



24
is analogous to the Strickland v. Washington test, see 
466 U.S. 668, 688-89 (1984), for assessing the 
effective assistance of counsel -  it is the type of 
clearly established rule “which of necessity requires a 
case-by-case examination of the evidence,” and can 
therefore “tolerate a number of specific applications 
without saying that those applications themselves 
create a new rule.” Williams, 529 U.S. at 382 
(quoting Wright v. West, 505 U.S. 277, 308-09 
(1992)). The Duren test is a clear standard that must 
be reasonably applied by the state courts.

In arguing that Duren’s test is too general to 
constitute clearly established law, Petitioner cites 
Wright v. Van Patten, 552 U.S. 120 (2008), and 
Carey, 549 U.S. at 70. See Pet. Br. at 35-36. These 
decisions are, however, inapplicable to Mr. Smith 
because they involve attempts to extend Supreme 
Court law to novel ground. In Carey, this Court 
rejected the Ninth Circuit’s attempt to extend the 
clearly established law governing state-sponsored 
courtroom practices to private spectator conduct, an 
application not previously contemplated by this 
Court. See 549 U.S. at 654. And in Wright, this 
Court found the state court’s application of federal 
law to be reasonable where “[n]o decision of this 
Court squarely addresse[d] the issue in this case.” 
Wright, 552 U.S. at 125, see also Knowles v. 
Mirzayance, -- U.S. ", 129 S.Ct. 1411, 1419 (2009) (no 
unreasonable application of clearly established law 
where “[t]his Court has never established anything 
akin to the Court of Appeals’ ‘nothing to lose’ 
standard for evaluating Strickland claims”); Schriro 
v. Landrigan, 550 U.S. 465, 478 (2007) (no
unreasonable application of clearly established 
federal law where “we have never addressed a 
situation like this.” (internal citation omitted)). This



25
authority is irrelevant to Mr. Smith because he did 
not seek to extend Supreme Court law to a new 
context. The fact that the federal courts of appeal 
and state supreme courts agree with Mr. Smith that 
Duren requires a case-specific approach to measuring 
underrepresentation further demonstrates that Mr. 
Smith is not seeking to apply Supreme Court law to a 
different, and unanticipated, circumstance. See Resp. 
Br. at 27-30.

2. The Michigan Supreme Court’s
Unreasonable Application Of Clearly 
Established Law.

After properly identifying this Court’s clearly 
established law, the Michigan Supreme Court 
unreasonably applied it to the facts of Mr. Smith’s 
case. The court concluded that Mr. Smith “presented 
some evidence of a disparity between the number of 
jury-eligible African Americans and the actual 
number of African-American prospective jurors 
selected to the Kent County Circuit Court jury pool 
list.” Pet. App. 146a. It then declared, however, that:

[Djefendant’s statistical evidence failed to 
establish a legally significant disparity under 
either the absolute or comparative disparity 
tests. Nevertheless, rather than leaving the 
possibility of systematic exclusion unreviewed 
solely on the basis of defendant’s failure to 
establish underrepresentation, we give the 
defendant the benefit of the doubt on 
underrepresentation and proceed to the third 
prong of the Duren analysis.

Id. at 146a-47a. Although this ruling is not a model 
of clarity, AEDPA controls here because the Michigan 
Supreme Court held that Mr. Smith failed to meet his 
prima facie burden under Duren’s second prong. See



26
28 U.S.C. § 2254(d). That decision was objectively 
unreasonable.

As previously noted, at the time of Mr. Smith’s 
trial, African Americans comprised 7.28 percent of 
Kent County’s adult population. Under these 
circumstances -  which are analogous to those 
prevailing in the vast majority of African-American 
communities throughout the country, see Brief for 
Social Scientists, Statisticians, and Law Professors, 
Jeffrey Fagan, et. al, as Amici Curiae Supporting 
Respondent -  the Sixth Amendment can tolerate only 
a limited degree of African-American exclusion and 
underrepresentation before implicating the Sixth 
Amendment’s fair cross-section requirement. Thus, 
in order to apply Duren and this Court’s other fair 
cross-section precedents reasonably, the state court 
should have recognized that relying on an absolute 
disparity measurement would have improperly 
allowed the total exclusion of Kent County’s African- 
American population from its jury venires.

The Michigan Supreme Court ignored this 
constitutionally significant fact. Although it properly 
acknowledged that Duren required a case-specific 
analysis and that the absolute disparity test 
“produces questionable results . . . where the 
members of the distinctive group comprise a small 
percentage of those eligible for jury service,” Pet. App. 
145a, the state court nonetheless gave equal 
consideration to the results of an absolute disparity 
analysis in evaluating Mr. Smith’s evidence of 
underrepresentation. Because the absolute disparity 
test would permit the total exclusion of Kent County’s 
African-American community from venires, the state 
court’s application of Duren was objectively 
unreasonable.



27
Furthermore, the state court unreasonably applied 

Duren by undervaluing Mr. Smith’s comparative 
disparity evidence. Mr. Smith demonstrated that 
African Americans were underrepresented by 34.8 
percent in the month of his jury selection. As the 
Sixth Circuit found, this is “sufficient to demonstrate 
that the representation of African-American 
veniremen in Kent County at the time of 
[Mr. Smith’s] trial was unfair and unreasonable.” 
Pet. App. 21a (citing Rogers, 73 F.3d at 777).

For these reasons, the Michigan Supreme Court’s 
judgment was objectively unreasonable under 
AEDPA in concluding that Mr. Smith failed to 
demonstrate that African Americans were 
underrepresented on Kent County Circuit Court 
venires.

B. The Michigan Supreme Court Was 
Objectively Unreasonable In Determining 
That African Americans Were Not 
Systematically Excluded From Kent County7s 
Circuit Court Venire Pool.

Duren’s third step requires proof that the 
underrepresentation is “due to systematic exclusion 
of the group in the jury-selection process.” Duren, 
439 U.S. at 364. The Sixth Circuit correctly held that 
the Michigan Supreme Court unreasonably applied 
Duren in finding no systematic exclusion in Kent 
County’s practice of siphoning jurors away from the 
circuit court to district courts. Pet. App. 41a_42a.

The “systematic exclusion” analysis focuses on the 
identification of processes integral to the jury 
selection system that may have caused persistent 
underrepresentation of distinctive groups, and must 
be conducted independent of Duren’s other steps. 
See, e.g., Weaver, 267 F.3d at 241 (“strength of the



28
evidence should be considered under the second 
prong, while the nature of the process and the length 
of time of underrepresentation should be considered 
under the third”)! Fordv. Seabold, 841 F.2d 677, 685 
& n.6 (6th Cir. 1988) (“Although the frequency of the 
selection of jury venires is not, of itself, helpful in 
establishing underrepresentation under the second 
prong of the prima facie case, it is of crucial 
significance in establishing that any existing 
exclusion was systematic.”). Accordingly, the 
persistence of African-American underrepresentation 
of 18 percent10 over a six-month period supports the 
systematic nature of their exclusion from the 
qualified circuit court jury pool. See Duren, 439 U.S. 
at 367 (“The resulting disproportionate and 
consistent exclusion of women from the jury wheel 
and at the venire stage was quite obviously due to the 
system  by which juries were selected.”).11

10 The persistence of the 18 percent underrepresentation over 
the course of six months, which satisfies D urens third step, is 
separate from the degree of the 34 percent underrepresentation 
during the month of Mr. Smith’s jury selection, which satisfies 
Duren s second step. To conflate these two analyses would make 
either the second or third step superfluous. See Paredes v. 
Quarterman, 574 F.3d 281, 290 (5th Cir. 2009) ( [Petitioner] 
repeats his position th a t a large discrepancy exists but does not 
explain how th a t second-prong statem ent alone satisfies the 
requirem ents of the third prong, which requires an independent 
showing th a t the discrepancy is systematic. If any discrepancy 
sufficient to satisfy the second prong would also satisfy the 
third, the th ird  inquiry would be superfluous.”).

11 Despite th is Court’s clearly established law th a t no 
discriminatory in tent is required in fair cross-section cases, See 
supra a t 9, Petitioner attempts to inject an in tent requirem ent 
into Duren s concern with the systematic underrepresentation of 
distinctive groups. Pet. Br. a t 44. In fact, Duren instructs only 
tha t the exclusion be “inherent in the particular jury-selection 
process utilized.” Duren, 439 U.S. a t 366. As explained in



29
The Michigan Supreme Court’s analysis of 

“systematic exclusion” was fundamentally flawed, 
and thus objectively unreasonable, because it ignored 
pertinent, uncontested evidence of Kent County’s 
siphoning procedure and imposed a burden of proof 
that exceeded Dureds mandate for establishing a 
prima facie case of a fair cross-section violation.

In rejecting Mr. Smith’s claim of systematic 
exclusion resulting from the siphoning procedure, the 
Michigan Supreme Court held that he had “failed to 
carry his burden of proof’ because the “record does 
not disclose whether the district court jury pools 
contained more, fewer, or approximately the same 
percentage of minority jurors as the circuit court jury 
pool.”12 Pet. App. 147a. The state court’s summary 
conclusion directly contradicts the evidence in the 
record. The uncontested testimony of Kent County 
jury officials and the results of a bar study committee 
both showed that the siphoning procedure was a 
significant source of African-American 
underrepresentation in circuit court venire panels, 
which led Kent County to terminate this practice in

Section 11(C), supra, Kent County’s siphoning practice meets 
th a t test because it was part of the jury selection system and 
caused the underrepresentation of African Americans in Kent 
County’s circuit court jury pool. Under this Court’s precedent, 
ignoring the discriminatory effect of the siphoning and 
demanding evidence of a discriminatory purpose would plainly 
be unreasonable.

12 The concurring opinion further commented th a t “after Kent 
County stopped selecting district court jurors first, the change in 
the representation of black jurors in circuit court was small, 
suggesting th a t the alleged systematic exclusion was not the 
cause of the underrepresentation.” Pet. App. 169a. However, as 
noted in Section 11(C), supra at 19-20, the record suggests th a t it 
took several months for the effects of term inating the siphoning 
procedure to be reflected in the jury pool data.



30
October 1993. There was no contrary evidence in the 
record. See supra at 18-20.

The state court’s understanding of Mr. Smith’s 
burden of proof under Duren was also objectively 
unreasonable. The State in Duren, like Petitioner 
here, argued that there was no proof “that the 
exemption for women had ‘any effect’ on or was 
responsible for the underrepresentation of women on 
venires.” 439 U.S. at 368. This Court, however, 
determined that Duren’s third step was satisfied by 
evidence that the exemption for women was “inherent 
in the particular jury-selection process utilized,” thus 
shifting the burden to Petitioner to “demonstrate that 
[other] exemptions caused the underrepresentation 
complained of.” Id. at 366, 368. The same response is 
warranted here, where the officials in the best 
position to provide information about Kent County’s 
jury selection system testified that the siphoning 
procedure systematically excluded African Americans 
from the qualified circuit court jury pool. See 
Section 11(C), supra. The Michigan Supreme Court’s 
imposition of a higher burden of proof is an 
objectively unreasonable application of Duren.13 See 
Davis v. Warden, 867 F.2d 1003, 1014 (7th Cir. 1989) 
(“The Supreme Court . . . and other courts have 
recognized that defendant should not be expected to

13 Contrary to Petitioner’s assertions (Pet. Br. a t 60), 
28 U.S.C. § 2254(e) has no relevance here because Mr. Smith 
seeks habeas relief under 28 U.S.C. § 2254(d)(1). Section 
2254(e) governs state court factual findings, and Mr. Smith does 
not take issue with the facts found by the Michigan Supreme 
Court. Instead, he challenges tha t Court’s application o f  law  to 
those facts and, thus, contends th a t the court’s ruling of no 
evidentiary support for “systematic exclusion” was an 
“unreasonable application of’ Dureris third prong under Section 
2254(d)(1).



31
carry a prohibitive burden in proving 
underrepresentation.” (citations omitted)); cf. 
Johnson, 545 U.S. at 170 (prima facie case of Batson 
violation established by “evidence sufficient to permit 
the trial judge to draw an inference that 
discrimination has occurred”).

CONCLUSION
The judgment of the Court of Appeals should be 

affirmed.
Respectfully submitted,

Virginia A. Seitz 
Gary Feinerman 
J effrey T. Green 
Rebecca K. Troth 
Anand H. Das 
Sidley Austin llp 
1501 K Street, NW 
Washington, DC 20005 
(202) 736-8000

J ohn Payton 
Director - Co unsel 

Debo P. Adegbile 
Christina Swarns* 
J ohanna B. Steinberg 
J in Hee Lee
Vincent M. Southerland 
Mary Hunter 
NAACP Legal Defense & 
Educational Fund, Inc. 
99 Hudson St., 16th Floor 
New York, NY 10013 
(212) 965-2200

Sarah O’Rourke Schrup 
Northwestern 
University Supreme 

Court Practicum 
357 East Chicago Avenue 
Chicago, IL 60611 
(312) 503-8576

Counsel for Amicus Curiae 
December 28, 2009 * Counsel o f Record