*1
have taken the role
of the federal
the rule established
Eddings
v. Okla
court,
district
refusing to allow that court
homa,
104,
869,
455 U.S.
102 S.Ct.
Stokley’s
deal
the first
with
instance
(1982).
L.Ed.2d 1
Assuming,
panel
as the
—
Thomas,
motion
Maples
under
v.
U.S.
does,
majority
that abandonment has been
-,
912,
132 S.Ct.
tural).
There is no reason for such haste. Stokley plausible has asserted claims un- America, UNITED STATES Maples der and Eddings. They may or Plaintiff-Appellee, may prove to be winning claims. But we should not allow the State of Arizona to
kill Stokley they before have proper- been HERNANDEZ-ESTRADA, Salvador ly considered. Defendant-Appellant. No. 11-50417.
WATFORD, Judge, joined Circuit PREGERSON, WARDLAW, W. United States of Appeals, Court FLETCHER, FISHER, PAEZ, Ninth Circuit. BERZON, CHRISTEN, NGUYEN, Argued Sept. and Submitted Judges, Circuit dissenting from the denial of en rehearing: banc Filed Dec.
I do not think any question there is here
that the Arizona Supreme Court violated *4 McKenzie, Federal Defend- A.
Michele CA, ers, Defendant-Appel- Diego, for San lant. Attorney; Duffy, States
Laura E. United Castetter, Curnow, and David Bruce R. (argued), Assistant United P. White Victor CA, for Attorneys, Diego, San States Plaintiff-Appellee. KOZINSKI, ALEX Chief
Before: Judge, PAUL J. WATFORD HURWITZ, Judges. D. Circuit ANDREW OPINION HURWITZ, Judge: Circuit appeal in this is whether question The District Court for the States United violated District California Southern Act of 1968 and Service Jury Selection (“JSSA”) compiling the Constitution or Although the jury wheel. its 2009 master from the re- departed District Southern in several re- of the JSSA quirements challenge no error in the to the speets, composition grand we find reversible underlying conviction. petit juries,” including challenges under the JSSA. United States v. Sanchez-Lo
I. (9th Cir.1989). pez, 879 F.2d in- Salvador Hernandez-Estrada was being deported alien
dicted for found II. in violation of 8 United States U.S.C. A. § 1326. Hernandez filed a motion to dis- indictment, arguing miss Ordinarily, we would consider stat Southern violated JSSA and utory reaching claims before constitutional by using the Fifth and Sixth Amendments arguments. Yamasaki, Califano juror consisting only regis- source list 682, 692, 99 S.Ct. He un- argued tered voters. the list (1979). L.Ed.2d 176 But here Hernandez’s derrepresented African-Americans arguments constitutional are intertwined Hispanics. alleged Hernandez also that with his JSSA claims. The JSSA contains *5 the Southern District violated the JSSA guarantee, a fair cross section 28 U.S.C. (1) jurors improperly disqualifying for hav- 1861, § which is coextensive with the fair ing English-language insufficient abilities requirement cross section of the Sixth juror ques- based on their answers on the Miller, Amendment. United States v. 771 (2) tionnaire; ju- improperly disqualifying (9th Cir.1985). 1219, 1227 F.2d According English-language rors whose levels of abil- ly, it makes more sense to address Her (3) unclear; failing ities were to return nandez’s constitutional claims first. questionnaires that omitted information on (4) race ethnicity; failing and and/or keep jury representativeness statistics. provides prospective ju- The JSSA that In response, the Government conceded reg- rors “shall be selected from the voter that the had Southern District violated the istration lists or the lists of actual voters of JSSA, disputed any but that of viola- political subdivisions within the district enough tions were substantial to warrant 1863(b)(2). § or division.” 28 U.S.C. Con- 1867(a) § (providing relief. See 28 U.S.C. sistent requirement, with this the South- only for relief for a failure “substantial jurors ern District selects at JSSA). comply” with the The Government registered random from the list of voters disputed also that the Southern District pro- the district. The further JSSA had violated the Constitution. vides, however, pre- that districts “shall The district court denied Hernandez’s scribe some other source or sources of dismiss, finding motion to no constitutional names in addition to voter lists where nec- violation and that JSSA violations section, essary to” ensure a fair af- cross technical, substantial, were not and did so opportunity ford all citizens the to be con- not warrant dismissal. The district court jury duty, sidered for and ensure that nevertheless recommended that the South- individuals are not excluded on the basis of significant changes ern District make to its “race, color, sex, origin, national religion, jury practices. selection Hernandez was 1861, §§ or economic status.” 28 U.S.C. charged. convicted as 1863(b)(2). 1862, The Southern District challenges only Hernandez’s not list. Her- appeal supplement does its source denial of the that failure to argues motion dismiss. “We re nandez its do so independently non-deferentially view the Fifth Amendments. violates and Sixth group at in the constitutionally percentage for a of the issue
“The test same, population subtracting total from it whether chal jury is the selected group repre of that percentage of the lenged the Sixth Amendment under sented on wheel.” San the master Jury under the Selection Constitution or 547; Miller, chez-Lopez, F.2d at see also at 1227. Act.” Service Rodriguez-Lara, United v. States prima In facie viola order to establish Cir.2005) (re-affirming 942-943 require tion fair-cross-section disparity our commitment to the absolute (1) ment, must the defendant show test); Smith, 314, 130 559 U.S. Berghuis alleged to be excluded is a group 1393-94, S.Ct. 176 L.Ed.2d community; in the group “distinctive” (2010) (neither requiring prohibiting nor (2) group of representation this of any particular addressing the use test juries are selected venires from underrepresentation).1 Although claims of fair and reasonable in relation to line, we have an exact we never drawn persons of such in the com number disparity have held that a of is ac 7.7% (3) underrepre munity; and ceptable. Rodriguez-Lara, F.3d at systematic due to exclusion sentation is 943-44. jury-selection pro cess. rely “must on the statistical Missouri, Duren approximates percent data best (1979). Hispan S.Ct. L.Ed.2d 579 age jury-eligible group] [members *6 ics African-Americans are each dis in the district.” United States Torres- the of groups prong tinctive under first (9th Hernandez, 699, 704 Cir. Cannady, 54 test. United States 2006). In 22.5% of the Southern Cir.1995). F.3d 547 District’s 18 population citizen and over prong second of the Duren
“The Hispanic was and 5.2% was African-Amer typically requires proof, test statistical ican.
data,
jury pool
adequate
does not
ly
in
represent
compare
percentages
the distinctive
rela
those
to
persons
to
in
the
percentages
Hispanics
tion
the number of such
the
of
African-
in
community.”
Esquivel,
Sanchez-Lopez,
United States v.
Americans
the wheel.
(9th Cir.1996).
determining
F.2d
In
analyz
per
726
In
879
at 547.
the
F.3d
centage
jury
use
in the
ing
prong,
Hispanics
we
the absolute
of
wheel we
the second
test,
disparity
requires
identify
us to meas
exclude those who did not
their
“by
ethnicity
ure
in de
underrepresentation
taking
questionnaire;
on the
require
assessing
precedents
of the
based on random selection when
Because our
use
test,
disparity
have
we
no occasion
absolute
qualified
By
constitutionality
of a
wheel.
today
methodologies
ana-
to consider other
to
definition,
qualified
prod
wheel is not
lyze
representativeness
the Southern
selection;
uct of
it entails
random
reasoned
difficulty
wheel. The
with use
disqualifications based on numerous factors.
disparity
dealing
test in
small
absolute
gauge
qualified
It is
to
wheel—
irrational
populations
recognized
Berghuis,
in
was
po
inherently
sample—by
an
non-random
its
But,
S.Ct. at 1393.
as
Court noted in
randomness.”);
Berghuis,
tential
see also
for
declining
any particular
to dictate the use
(noting that "no
...
uals, Hispanics up made 24.6% of the up African-Americans made wheel and III. Thus, Hispanics overrepresent- 3.5%. were rejection Because our of Hernandez’s ed 2.1% and African-Americans were constitutional claims fair dooms his cross underrepresented by 1.7%. Since these JSSA, Miller, section claim under the percentages begin approach do not 7.7% F.2d at we now turn to his remaining underrepresentation, Hernandez’s Sixth statutory claims. claim Amendment fails. in- urges
Hernandez
that we instead
A.
clude
our calculations individuals who
identify
failed to
their race
ethnici-
and/or
Congress enacted the
as
JSSA
matter,
ty.
way
As an initial
there is no
a response to concerns that racial discrimi
identify
know that those who failed to
their
frequently
jury
nation
infected the
selec
ethnicity
race or
were not members of a
process.
Esquivel,
tion
88 F.3d at
all, they
minority group—after
did not re-
In
order
combat such discrimina
if
ethnicity.
veal their race or
But even
tion, the
prescribes
variety
pro
JSSA
them,
we included
Hernandez’s claim still
in compiling
cedures
lists of
40,743 persons
fails. There were
jurors.
§§
See 28 U.S.C.
1861-69. “Con
wheel,
jury
non-re-
including
gress, recognizing that
there
un
would
sponders
ethnicity ques-
to the race and/or
doubtedly
be error
selection
these, 1,257
tions. Of
identified them-
process that should not result in the dis
6,625
selves as African-American and
as
indictment,
missal of an
left room for
*7
Thus,
Hispanic.
even if we assume that
by providing
harmless error
that dismissal
single
there was
a
African-American
not
only
should lie
when there was a substan
Hispanic among
or
Af-
non-responders,
comply
tial failure to
with the
United
Act.”
rican-Americans constituted 3.1% and His-
(5th
Evans,
701,
v.
526 F.2d
705
States
panics
qualified jury
16.3% of the
wheel. Cir.1976). Thus,
only
we will
dismiss Her
numbers,
Using these
African-Americans
nandez’s indictment
if he shows a “sub
underrepresented by 2.1% and His-
were
stantial” violation of the JSSA. 28 U.S.C.
panics by 6.2%. Neither clears the 7.7% 1867(a).
§
threshold.
“Technical violations are insub
they do not frustrate the
stantial where
Nelson,
equal
goals.”
To
a violation of the
Act’s
States v.
718
establish
United
(9th Cir.1983).
315,
protection guarantee
goals
of the Fifth Amend F.2d
Those
“
ment,
juror
only
a defendant must
are
‘random selection of
names
show
”
“
underrepresentation
pro
of a
from voter lists’ and ‘determination of
substantial
excuses, exemp
“discriminatory
juror disqualifications,
tected
but also
in
tions,
(citing
objec
at
exclusions on the basis of
Esquivel,
tent.”
88 F.3d
”
Partida,
494,
482,
only.’
v.
Castaneda v.
tive criteria
United States Good
(1977)).
(9th Cir.1979)
low,
159,
1272,
C. Quantitatively, a that violation does not are Hernandez’s other claims objectivity principle frustrate the Act’s 1968, prospec Before more substantive. must in a significant result number of if jurors disqualified were “unable to tive wrongful exclusions before it will be read, write, En speak, and understand the (1957). § deemed substantial. Id. glish language.” U.S.C. standard,
The JSSA amended There was no frustration of the JSSA’s juror a provides prospective now Indeed, objectivity principle here. it only if disqualified be he “is unable should objective to imagine hard more criteri read, write, English and understand the prospec on than the one used here. All degree proficiency a suffi language with jurors Question tive who answered “no” to juror satisfactorily fill out cient to disqualified; 4 were the clerk’s office exer speak form” or if “unable to qualification cised no discretion. See United States language.” 28 U.S.C. English Carmichael, Cir. (3). 1865(b)(2), § 2009). Question 4 of the Southern asks, the old questionnaire consistent with Hernandez must therefore show standard, “read, jurors statutory whether Question 4 in improper wording of resulted
write, speak
English
and understand the
a significant
wrongful
number of
exclu-
wheel,
language.”
assembling
In
the 2009
12,250
sions. Hernandez notes that of the
disqualified
prospec-
the clerk’s office
all
Hispanics
questionnaires,
who returned
jurors
tive
who answered “no” to that
1,420
disqualified solely
they
were
because
question. The Government concedes that
Question
“no” to
4. But
if
answered
even
jurors
prospective
the dismissal of
based
1,420
jurors
we assume that all
solely
question
on their answers to this
wrongfully disqualified,
were
that number
it
a
argues
violated the JSSA but
is not
does not
establish
substantial violation
substantial violation.
viewed,
be,
it
when
as must
the context
Hernandez bears the burden “to
jury pool.
of the entire
present
constituting
facts
a substantial vio
Bearden,
In
Fifth
Circuit found
Nelson,
lation.”
1025 tion, of legality could not have affected the the we affirm Hernandez’s conviction and sentence. resulting wheel. AFFIRMED. appropriate steps
The district must take response to rate these two increase the to Judge KOZINSKI, Chief with whom questions. we leave to district While the Judge joins, concurring: WATFORD accomplish how we exactly goal, to that join I Judge opinion (except Hurwitz’s suggested note that the district court three one) faithfully applies footnote because it remedies, all potential of which deserve the our law of circuit. United States careful consideration. 421 Rodriguez-Lara, F.3d First, each the district court noted that Cir.2005). But I do so without enthusiasm questionnaire jurors informed that federal because the rule we to apply are bound is required law them to answer the questions clearly It wrong. makes no sense to ethnicity help prevent on race and dis- disparity cross measure for fair section crimination that their would answers by purposes looking disparity, at absolute eligibility not affect their for service. jury up accept percent to 7.7 of the total But that information was in on print small permissible as a pool deviation. See questionnaire. the back district of the The Suttiswad, United States court moving recommended that instruc- (9th Cir.1982) (holding that 7.7 percent Second, tion to the front. the district disparity absolute is The acceptable). ab court of reversing recommended the order surdity brought of number is home ethnicity the questions race and so that the that a observing group that is less than 7.7 ethnicity question Many would come first. percent the total can population never jurors might ethnicity not answer underrepresented, be no matter how far question question when it follows the race pool the jury percentage deviates from they doing because that so unneces- feel total population. See Rodri all, sary or After redundant. the omission guez-Lara, F.3d at n. 10. ethnicity question rate for exceed- far anomaly larger This for disappears question. ed that for the race group groups. percent A that is 75 of the Finally, district court noted cognizable total population register could some permit districts online submission disparity only if it were of the percent districts, questionnaires. In such jury pool, percent while a that is 7.5 questionnaire be online cannot submitted register could even if disparity, never completing questions without at issue I have a entirely pool. absent from the here. The district should consider these accepting favors hard time a rule options, might as as well others that larger groups ignores groups smaller response increase the rate. altogether. “main Our cases have referred IV. as “com- disparity alternative” absolute rejected Despite parative disparity.” our conclusion that no Id. reversible here, approach thought error we because we it was un- exists caution the Southern others) (and to take with small numbers: His- “[I]f note workable community statutory panics violations we have identified and are 2% of the and 1% comparative disparity For practices jury pool, amend its the future. now, 50%, every jurors, only there is because no JSSA violation warrants but for Hispanic propor- viola- than be relief and there was no constitutional one fewer would *11 Id.; see States v. tional.” also United ORTIZ, Adilao Juan Petitioner- 541,
Sanchez-Lopez, 879 F.2d 547-48 Appellant, Cir.1989). dealing But we’re not here dealing with a people; a hundred we’re v. 40,000. For a jury pool of over YATES, Warden, A. James size, methods that can there are statistical Respondent-Appellee. easily sub-group tell us whether a is (blacks percent population 5.2 No. 11-56383. District) if underrepresented it Southern Appeals, United States Court percent jury pool. up only makes 3.5 Ninth Circuit. context, protection the Su- equal In the 9, Argued and Submitted Oct. 2012. has deviation preme Court used “standard see, Partida, analysis,” e.g., Castaneda 6, Filed Dec. 482, 1272, n. 430 U.S. 97 S.Ct. (1977),
L.Ed.2d 498 “seeks to deter- probability be- disparity
mine the group’s population jury-eligible
tween group’s percentage
and the
jury pool is attributable to random
chance,” Smith, Berghuis 559 U.S.
130 S.Ct. 1390 n. 176 L.Ed.2d
(2010). More than two or three standard hypothesis
deviations means that “the jury drawing was random would be Castaneda, a social
suspect to scientist.” Doing at 496 n. 97 S.Ct. math, I quick that the dis-
some calculate percent percent between 5.2 and 3.5
parity is more devi- this case than standard formula).
ations. id. So (detailing worry.
there’s cause to
I’m not sure whether standard deviation here, I
analysis appropriate suspect but laugh
that a at our cur- statistician would methodology. three-judge pan-
rent As a
el, to depart we’re not free from Rodri- could, an
guez-Lara, but en banc court should, perhaps take a fresh look at
the issue. notes Hernandez also omission, or error ambiguity, an there is Question 69.7% were swering “no” form, jury clerk or commission in a the 1,420 ju- Hispanic Hispanic, and instructions return the form with their shall the basis of solely on rors excluded or to make such additions person of to the up 4 make 25.2% Question answers to may necessary and to corrections as be excluded for 5,625 jurors Hispanic the jury or the form to the clerk However, do return these statistics any reason. days. commission within ten objec- relate to the randomness not JSSA; they relate to tivity goals of Id. goal. As ex- fair-cross-section the Act’s 1864(a) requires § assuming that Even earlier, has not estab- Hernandez plained omission every questionnaire of the JSSA violation lished a substantial returned, Hernandez has not demon- be fair- of the Act’s based on frustration strated that the Southern District substan- Notwithstanding any goal. cross-section requirements from the tially departed by the im- wrongful produced exclusions 1864(a) plainly de- the JSSA. Section Question Hispanics proper wording goals of assur- signed to serve JSSA’s substantially underrepresented are not juries are at random ing that “selected (and jury may even be qualified pool communi- from a fair cross section of the supra at slightly overrepresented). § dis- ty,” preventing 28 U.S.C. 1020-21. process, crimination the selection has failed to Simply because Hernandez § U.S.C. that this violation is sub- showing make a the 2009 representativeness The mean a future defendant stantial does not goal clear that neither was wheel makes District can will also fail. The Southern here. Even if we assume compromised remedy problem. Asking this and should His- non-responder that each was neither form is not itself Question 4 in its current African-American, panic nor JSSA, dismissing but a violation of fairly representa- was wheel nonetheless they an- prospective jurors solely because Moreover, prospec- of the district. no tive Thus, is. al- question “no” to that swer for failure to re- juror tive was excluded language ques- though changing questions, so there can be spond to these and most effective likely tion is the easiest thereby the district no contention violation, remedy may it not be way to subjective compiling criteria in used only way. leave remediation to Marcano, States v. wheel. See United district, is nec- emphasize change but (D.P.R.1980). F.Supp. other districts to essary. And we caution as this questionnaires, Nonetheless, evaluate their own may the Southern District unique to the problem appears not to be per- The lucky so the future. be Southern District. wheel centages of those ethnicity the race and who did not answer D. respective- and 33.81% questions—11.56% so, it significant. they If remain Finally, alleges ly—are Hernandez imagine years that in future failure to return not hard to the Southern District’s