Lopez et al v. Abbott et al

Filing 109

FINDINGS OF FACT AND CONCLUSIONS OF LAW.(Signed by Judge Nelva Gonzales Ramos) Parties notified.(mserpa, 2)

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 1 of 43
United States District Court
Southern District of Texas

ENTERED
UNITED STATES DISTRICT COURT
SOUTHERN DISTRICT OF TEXAS
CORPUS CHRISTI DIVISION
LIONEL LOPEZ, et al,
Plaintiffs,
VS.
GREG ABBOTT, et al,
Defendants.

September 12, 2018
David J. Bradley, Clerk

§
§
§
§ CIVIL ACTION NO. 2:16-CV-303
§
§
§
§

FINDINGS OF FACT AND CONCLUSIONS OF LAW
Plaintiffs challenge statewide, at-large elections of all justices to the Supreme
Court of Texas and judges to the Texas Court of Criminal Appeals under Section 2 of the
Voting Rights Act of 1965, 52 U.S.C. § 10301. They allege that at-large voting has
diluted the voting strength of registered voters who are Hispanic. To remedy this, they
request that the Court order the imposition of single member districts to be drawn up, in
the first instance, by the Texas legislature.
This action was tried to the bench from February 12 to 15, 2018. For the reasons
set out below, the Court holds that three of the individual Plaintiffs and Plaintiff La
Unión Del Pueblo Entero (LUPE) have standing. The Court holds that Plaintiffs have
satisfied the three preconditions set forth in Thornburg v. Gingles, 478 U.S. 30 (1986).
However, under the totality of the circumstances test, they have failed to satisfy their
burden of demonstrating that the lack of electoral success by Hispanic-preferred
candidates for high judicial office is on account of race rather than other factors,

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 2 of 43

including partisanship. Thus, they have not demonstrated a Section 2 violation and they
are not entitled to relief.
STANDING
The Court begins by addressing whether Plaintiffs have standing to bring their
suit. A plaintiff seeking to establish Article III standing must show “that he ‘(1) suffered
an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant,
and (3) that is likely to be redressed by a favorable judicial decision.’” Gill v. Whitford,
138 S. Ct. 1916, 1929 (2018) (quoting Spokeo, Inc. v. Robins, 136 S. Ct. 1540, 1547
(2016)). The injury in fact must involve the “‘invasion of a legally protected interest’
that is ‘concrete and particularized,’ i.e., which ‘affect[s] the plaintiff in a personal and
individual way.’” Id. (quoting Lujan v. Defenders of Wildlife, 504 U.S. 555, 560 & n.1
(1992)) (alterations in original).
Individual Plaintiffs. Beginning with the seven individual Plaintiffs,1 it is settled
that “voters who allege facts showing disadvantage to themselves as individuals have
standing to sue.”

Baker v. Carr, 369 U.S. 186, 206 (1962).

Here, three of the

Plaintiffs—Isabel Araiza, Lena Lorraine Lozano Solis, and Carmen Rodriguez—testified
that they are of Hispanic descent and that they vote regularly. They also all reside in
parts of Texas that are within the proposed Hispanic-majority districts under Plaintiffs’
illustrative redistricting schemes. Each of these Plaintiffs has therefore demonstrated an

1

Those Plaintiffs are Lionel Lopez, Isabel Araiza, Arlene Lira Easter, Alicia Benavidez, Andres Rosas, Lena
Lorraine Lozano Solis, and Carmen Rodriguez.

2

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 3 of 43

injury-in-fact, traceable to Texas’s method of selecting high-court justices and judges,
which may be remedied through this litigation. This is sufficient to establish standing.
The same cannot be said for the four Plaintiffs who did not testify,2 as “[t]he facts
necessary to establish standing . . . must not only be alleged at the pleading stage, but also
proved at trial.” See Gill, 138 S. Ct. at 1931. No such evidence was presented as to these
Plaintiffs. Plaintiffs cite Crawford v. Marion County Election Board, 553 U.S. 181, 189
n.7 (2008), for the proposition that the court need not examine whether all plaintiffs have
standing so long as one plaintiff demonstrates standing. Crawford, however, was decided
on summary judgment, where it makes sense to proceed to overarching questions of
liability after determining that any single plaintiff can establish standing. Here, the
burden at trial was on each Plaintiff to demonstrate his or her actual injury in fact. That
burden was not met for the non-testifying Plaintiffs, who for that reason are dismissed.
LUPE. Also joining as a Plaintiff is the organization LUPE. Founded by César
Chávez and Dolores Huerta in 1989 as a means to improve the lives of farmworkers,
LUPE is a non-profit, non-partisan organization of more than 7,000 members, many of
whom are Hispanic voters residing in Texas.
A three-part test is used to determine whether an organization such as LUPE may
establish standing. Under that test, an organization “has standing to bring suit on behalf
of its members when [1] its members would otherwise have standing to sue in their own
right, [2] the interests at stake are germane to the organization’s purpose, and [3] neither
the claim asserted nor the relief requested requires the participation of individual
2

Those Plaintiffs are Lionel Lopez, Arlene Lira Easter, Alicia Benavidez, and Andres Rosas.

3

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 4 of 43

members in the lawsuit.” Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc.,
528 U.S. 167, 181 (2000).
That test is met. According to testimony from LUPE executive director Juanita
Valdez-Cox, LUPE’s members include voters who reside in the heavily Hispanic areas of
South and West Texas where the alleged dilutive effects of statewide judicial elections
are felt most keenly. The interests at stake are also germane to LUPE’s mission of
promoting political engagement among its members, which LUPE furthers by
encouraging voter registration and hosting events at which candidates for elected office,
including judicial office, are invited to appear. Finally, the nature of the requested relief
does not require every LUPE member to participate in the case.
For these reasons, LUPE and Plaintiffs Isabel Araiza, Lena Lorraine Lozano Solis,
and Carmen Rodriguez have standing. The Court turns next to the merits of their claims.
STANDARD OF REVIEW
Recurring throughout this case are incorrect assertions—from both sides—as to
the parties’ relative burdens of proof. At the heart of the confusion are two issues: (1) a
failure to separate the different purposes of, and properly apply, the Gingles
preconditions analysis and the Gingles totality of circumstances test; and (2) a loose
extrapolation of Fifth Circuit opinions, LULAC3 and Teague4, beyond their respective
holdings. The Court has examined Gingles and its progeny at some length and will
adjudicate this case pursuant to the untangled burdens of proof described below. The

3
4

League of United Latin Am. Citizens (LULAC) v. Clements, 999 F.2d 831 (5th Cir. 1993) (en banc).
Teague v. Attala Cnty., Miss., 92 F.3d 283 (5th Cir. 1996).

4

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 5 of 43

Court will further interpret the holdings in LULAC and Teague in the context of their
respective factual records.
I. Burden of Proof on Plaintiffs’ Claims
Plaintiffs’ entitlement to Section 2 relief—as it applies to this case—requires
showing vote dilution on account of race. Vote dilution is a question of fact subject to
the clearly-erroneous appellate standard of review of Federal Rule of Civil Procedure
52(a). Gingles, 478 U.S. at 78; Teague, 92 F.3d at 287. The evidence relevant to its
determination “must be made over time and over the course of many elections. How
many elections must be studied to make this determination depends on the particular
circumstances of the locale.” Teague, 92 F.3d at 288-89 (citing Gingles, 478 U.S. at 57
n.25). The factors to be considered in a racially-correlated vote dilution case based on
the use of a multimember district are well-established and undisputed. They are derived
from Gingles, 478 U.S. at 50–51, which sets out three preconditions and a test based
upon the totality of the circumstances.
A. The Gingles Preconditions
Plaintiffs must first “offer evidence of the circumstances of the local political
landscape—evidence demonstrating that ‘its members have less opportunity than other
members of the electorate to participate in the political process and elect representatives
of their choice.’” Magnolia Bar Ass’n, Inc. v. Lee, 994 F.2d 1143, 1146 (5th Cir. 1993)
(quoting Section 2). The Supreme Court set out the three preconditions for proving vote
dilution as follows:

5

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 6 of 43

 “First, the minority group must be able to demonstrate that
it is sufficiently large and geographically compact to
constitute a majority in a single-member district.”
 “Second, the minority group must be able to show that it
is politically cohesive.”
 “Third, the minority must be able to demonstrate that the
white majority votes sufficiently as a bloc to enable it—in
the absence of special circumstances, such as the minority
candidate running unopposed—usually to defeat the
minority’s preferred candidate.”
Gingles, 478 U.S. at 50-51 (citations omitted). While the preconditions are not applied in
isolation, plaintiffs must prove each of them by a preponderance of the evidence.
Magnolia, 994 F.2d at 1146.
The First Circuit has summarized the function of the different elements of the
Gingles inquiries. “The first two Gingles preconditions look to whether, putting the
challenged practice, procedure, or structure to one side, minority voters within a given
constituency have the potential to elect representatives of their choice.” Uno v. City of
Holyoke, 72 F.3d 973, 979 (1st Cir. 1995). “The third Gingles precondition . . . addresses
whether the challenged practice, procedure, or structure is the cause of the minority
group’s inability to mobilize its potential voting power and elect its preferred
candidates.” Id. at 980. “Proof of all three preconditions creates an inference that
members of the minority are in fact harmed by the challenged electoral practice,
procedure, or structure. However, the inference is rebuttable.”

Id.

The totality of

circumstances test is the means by which the inference of vote dilution may be rebutted
or cemented. Id.; see also Teague, 92 F.3d at 292.

6

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 7 of 43

B. The Gingles Totality of Circumstances Test
Section 2 relief is justified when the voting methodology “results in a denial or
abridgement of the right of any citizen of the United States to vote on account of race or
color . . . .” 52 U.S.C. § 10301(a) (emphasis added). The court must balance a number
of considerations to determine whether there is unlawful vote dilution under the totality
of the circumstances. Those factors include:
1. The history of official discrimination in voting;
2. Racial polarization of elections;
3. Voting practices that enhance the potential for discrimination against
minorities;
4. Exclusive slating processes;
5. The social, educational, and financial legacy of discrimination that
hinders effective minority participation in the political process;
6. Racial appeals in elections;
7. The number of minorities elected to public office;
8. Elected representatives’ lack of responsiveness to the minority’s
particularized needs; and
9. Whether the reasoning for imposing the challenged voting methodology
is tenuous.
Gingles, 478 U.S. at 37-38. These issues present questions of fact for the court. Salas v.
Sw. Texas Jr. Coll. Dist., 964 F.2d 1542, 1555 (5th Cir. 1992).

7

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 8 of 43

C. Relationship Between Preconditions and Totality of Circumstances
With respect to the interaction between the three preconditions and the nine factors
relevant to the totality of circumstances test,
[I]t will be only the very unusual case in which the plaintiffs
can establish the existence of the three Gingles factors but
still have failed to establish a violation of § 2 under the
totality of circumstances. In such cases, the district court
must explain with particularity why it has concluded, under
the particular facts of that case, that an electoral system that
routinely results in white voters voting as a bloc to defeat the
candidate of choice of a politically cohesive minority group is
not violative of § 2 of the Voting Rights Act.
Clark v. Calhoun Cty., Miss., 21 F.3d 92, 97 (5th Cir. 1994) (quoting Jenkins v. Red Clay
Consol. Sch. Dist. Bd. of Educ., 4 F.3d 1103, 1135 (3d Cir.1993)). The Clark case notes
that it will be unusual for the fact of minority vote dilution to be unrelated to racial bias,
but separating the two tests—the three preconditions and the totality of circumstances—
ensures that the evidence supports both and that relief is not granted where they are not
both tethered. The “unusual case” that Clark refers to is a case, like LULAC, where the
racial minority is closely aligned with the losing political party and the evidence on the
totality of circumstances renders political partisanship the better explanation for the
defeat of minority-preferred candidates at the polls. LULAC, 999 F.2d at 859-60.
II. Burden of Proof on Defensive Issues
While offering evidence on several of the factors to be considered, the State has
defended this action on the basis of two main themes: (1) the electoral results represent
partisan defeats, not voter dilution; and (2) the State has an insurmountable linkage
interest in having all of the high court justices and judges elected on a statewide basis.
8

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 9 of 43

Because these two themes are so prominently featured in this case and because the parties
have advocated for the application of equally extreme, yet opposing, burdens of proof,
the Court must determine at the outset where those burdens on defensive issues lie and
what standard of proof must be met.
A. Partisan Politics
Plaintiffs argue that, if they have demonstrated racially polarized voting, “the
burden then shifts to the defendant to prove that a race-neutral reason is the cause of the
polarization” with a burden to demonstrate that “the record indisputably proves that
partisan affiliation, not race, best explains the divergent voting patterns.” D.E. 102, pp.
32-33 (quoting LULAC, 999 F.2d at 850 (emphasis Plaintiffs’) and Teague, 92 F.3d at
290). However, those cases did not articulate such an all-or-nothing test. And that test is
contrary to the nine-factor totality of the circumstances balancing test arising out of both
the Senate Report5 compiled when Congress amended Section 2 and the Gingles opinion
that enunciated the new test. An evaluation of the LULAC and Teague opinions in the
context of their respective evidentiary records makes this clear.
The LULAC opinion considered the issue of race versus partisanship at significant
length. 999 F.2d at 849-63. The trial court had refused to consider nonracial causes of
voting outcomes. On appeal, the defendants argued that—to support a vote dilution
case—white bloc voting must be analyzed with respect to evidence of considerations
alternative to racial bias, such as partisanship.

5

S. Rep. No. 417 (1982), reprinted in 1982 U.S. C.C.A.N. 177.

9

The Fifth Circuit agreed.

Citing

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 10 of 43

Whitcomb v. Chavis6 and White v. Regester,7 the LULAC court discussed the interaction
between polarized voting and the other factors affecting minority access to the political
system. Without diminished minority access, a vote dilution claim is a “mere euphemism
for political defeat at the polls.” 999 F.2d at 854 (quoting Whitcomb, 403 U.S. at 153).
The LULAC opinion does not support Plaintiffs’ representation that they have
shifted the burden of proof and that it is up to the State to prove that partisanship
indisputably explains the evidence that Plaintiffs interpret as demonstrating voter
dilution. The LULAC decision is inextricably tied to its evidence:
It is difficult to see how the record in this case could possibly
support a finding of liability under the approach outlined in
the Senate Report. Plaintiffs have not even attempted to
establish proof of racial bloc voting by demonstrating that
“race,” not, as defendants contend, partisan affiliation, “is
the predominant determinant of political preference.” They
have instead maintained in the very teeth of the Senate
Report, that such a showing is unnecessary. Because the
district court accepted this argument, the test employed at
trial enabled plaintiffs to prevail by proving little more than a
lack of success at the polls and a history of discrimination.
999 F.2d at 855 (emphasis added).
The Fifth Circuit discussion of the race-versus-partisanship issue includes some
support for the idea that plaintiffs have the burden to distinguish racially polarized voting
from partisan voting patterns by reference to the other factors in the totality of the
circumstances test. However, the court refused to articulate a burden of proof. “[W]e
need not resolve the debate today. Whether or not the burden of the plaintiffs to prove
6

403 U.S. 124 (1971).

7

412 U.S. 755 (1973).

10

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 11 of 43

bloc voting includes the burden to explain partisan influence, the result is the same. This
is so even if the partisan voting is viewed as a defensive parry.” 999 F.2d at 860.
In the end, the LULAC court found, as a matter of fact, “Because the evidence in
most instances unmistakably shows that divergent voting patterns among white and
minority voters are best explained by partisan affiliation, we conclude that plaintiffs have
failed to establish racial bloc voting in most, but not all, of the counties.” 999 F.2d at
861. The fact that the record made it “unmistakable” there does not mean that the burden
falls on a defendant to show that it is “unmistakable” in every case. By the same token,
“Plaintiffs are . . . entirely correct in maintaining that courts should not summarily
dismiss vote dilution claims in cases where racially divergent voting patterns correspond
with partisan affiliation as ‘political defeats’ not cognizable under § 2.” 999 F.2d at 86061. In sum, the LULAC opinion calls upon the court to look at all of the evidence
regarding each of the factors to determine whether racial bias or partisan politics better
explains the voting patterns.
The Teague opinion did not change the analysis. That opinion notes that plaintiffs
do not bear the burden in the first instance to eliminate factors other than race as
influencing voters. Teague, 92 F.3d at 290.
Plaintiffs are to present evidence of racial bias operating in
the electoral system by proving up the Gingles factors.
Defendants may then rebut the plaintiffs’ evidence by
showing that no such bias exists in the relevant voting
community. . . . The lack of evidence in the record on this
point favors the plaintiffs, not the defendants.

11

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 12 of 43

Id. In the absence of evidence that other issues explain minority electoral defeat, some
evidence of racial bias in voting will satisfy plaintiffs’ burden.
LULAC involved indisputable and unmistakable evidence of a factor other than
racial bias influencing voting. Teague involved the opposite—no evidence of any other
factor at all. Those are the two extremes, and as such they do not articulate the entirety of
the relevant burden of proof. Consistent with Gingles, the Court holds that Plaintiffs
have the duty, in the first instance, to demonstrate some evidence of racial bias through
the factors used in the preconditions and totality of circumstances test. Upon doing so,
the burden shifts to the State to demonstrate some evidence of partisan politics (or some
other issue) influencing voting patterns. If the State does so, then the Court must balance
the relative strength of the evidence directed to each of the totality of circumstances
factors to determine whether racial bias best explains the alleged vote dilution.
B. Linkage
A voting methodology that correlates the jurisdiction of the elected official with
his or her constituency is expressive of the State’s linkage interest. The reasoning behind
linkage is to promote the State’s substantial interest in judicial effectiveness by balancing
accountability and independence. LULAC, 999 F.2d at 868-69. The weight of such a
substantial state interest is a question of law. Id. at 868, 871. In a Section 2 case, the
court must balance that interest against localized evidence of racial vote dilution. Id.
The Fifth Circuit expressly rejected the extremes “that the linkage interest either
always or never defeats § 2 liability.” Id. at 871 (quoting Houston Lawyers’ Ass’n v.
Attorney Gen. of Texas, 501 U.S. 419, 427 (1991)). Thus linkage does not automatically
12

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 13 of 43

wield dispositive weight, but is part of the totality of circumstances balancing test. It will
fall only to substantial proof of racial vote dilution. Id. at 868-71. Plaintiffs must
produce evidence of racial dilution in voting pursuant to the factors of the totality of
circumstances test. If the state raises the linkage interest, the court determines the weight
of that interest as a question of law and then balances that, along with the other factors, to
determine if racial dilution, rather than an appropriate state interest in linking an elected
official with the office’s constituency, explains the alleged Section 2 violation.
III.

Summary

Despite the parties’ claims to the contrary, they have advanced no single
dispositive issue or burden of proof that bypasses the need to work through all of the
Gingles elements. Therefore, the Court will first determine whether Plaintiffs have met
the three Gingles preconditions. If so, the Court will then consider whether Plaintiffs
have produced some evidence of racial bias in the electoral system in the factors of the
totality of circumstances test. If so, the Court considers whether the State has advanced
alternative explanations. If so, the Court will find as a matter of fact or law—as required
for each interest—the weight to be accorded each issue, including partisan politics and
linkage. After doing so, the Court will balance the resulting factors against each other to
determine whether, on the whole, Plaintiffs as Hispanic voters have had their voting
power for electing judges and justices of Texas’s high courts diluted on account of race.

13

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 14 of 43

DISCUSSION
I. Facts
The record is largely comprised of public records related to Texas statewide
election contests and precinct voting patterns, as well as expert statistical analysis of
those records. Demographic data, extracted and compared against the election results,
reveal how different racial groups have voted in recent years. The underlying evidence,
methodologies for analysis, and ultimate statistical results are not disputed. Rather, the
controversy of this case relates to whether certain contests, particularly party primaries
and those in which the candidates are not racially diverse, are probative on the question
of voter dilution. That statistical information, the anecdotal accounts, and other evidence
will be discussed in the context of the legal issues to which they relate.
II. Gingles and Section 2
It is well understood that Section 2 does not provide an assurance of success at the
polls for minorities. Johnson v. De Grandy, 512 U.S. 997, 1013-14 n.11 (1994). What it
seeks to provide is an assurance of fairness. Uno, 72 F.3d at 979. “Thus, when ‘a certain
electoral law, practice, or structure interacts with social and historical conditions to cause
an inequality in the opportunities enjoyed by [minority] and white voters to elect their
preferred representatives,’ a section 2 claim lies.” Id. (citing Gingles, 478 U.S. at 47).
The Gingles preconditions are designed to demonstrate whether the electoral structure is
such that minority votes are consistently defeated. The totality of circumstances test is
designed to show whether that result is a matter of race-based dilution or some other
phenomenon.
14

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 15 of 43

A. Discussion of Preconditions
1. Minority-Majority District
The first Gingles precondition “requires submitting as evidence hypothetical
redistricting schemes in the form of illustrative plans.” Gonzalez v. Harris Cty., Tex., 601
F. App'x 255, 258 (5th Cir. 2015) (per curiam). Such a plan is not proposed as a final
remedy, but as a feasible alternative. Clark, 21 F.3d at 95. It must demonstrate that the
minority population exceeds fifty percent (50%) of Citizen Voting Age Population
(CVAP) in an illustrative single-member district. Valdespino v. Alamo Heights Indep.
Sch. Dist., 168 F.3d 848, 853 (5th Cir. 1999); Campos v. City of Houston, 113 F.3d 544,
548 (5th Cir. 1997).
Plaintiffs offered the testimony of Dr. Lisa Handley,8 who had drawn up two
hypothetical single-member district plans. One divided the state into nine districts on the
premise that each of the nine judges and justices for the respective courts would be
elected in a separate single-member district. PTX 128. The other plan divided the state
into eight districts, on the premise that only eight of the judges or justices would be
elected in single-member districts and the chief judge or justice would be elected at large.
PTX 129.
Each of Dr. Handley’s plans created two minority-majority districts along the
southern border of the state. With respect to CVAP, each reflects Hispanic CVAP

8

Dr. Handley is a political scientist who works as an election consultant in the United States and internationally.
In the United States, she concentrates her work on Section 2 issues, such as redistricting and minority vote dilution.
2 Tr. 135. She has utilized her expertise on behalf of state and county jurisdictions, along with the U.S. Department
of Justice. Id. at 136-37. The State did not object to Dr. Handley testifying as an expert witness. 2 Tr. 141-42.

15

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 16 of 43

comfortably over 50%. The nine-district map lists districts 2 and 3 with Hispanic CVAP
of 59.69% and 59.19%, respectively. 2 Tr. 146-47. The eight-district map lists districts 1
and 8 with Hispanic CVAPs of 55% and 58%. 2 Tr. 152.
The proposed districts also comply with the principle of one person, one vote
because the population of each illustrative district did not vary more than 2% from the
ideal number achieved by dividing total CVAP by the number of districts created. 2 Tr.
149. The Supreme Court has held that deviations up to 10% of mathematical equality are
permissible, as population counts are inexact snapshots and other considerations (such as
administrative integrity and contiguity) may be of sufficient importance to justify minor
deviations. Brown v. Thomson, 462 U.S. 835, 842 (1983). All of the districts are
contiguous and follow administrative (county) boundaries with the exception of the
districts that bisect Harris County. 2 Tr. 148-49. Dr. Handley explained that Harris
County was too populous to be placed in any single district. 2 Tr. 149. Thus, its
separation was a statistical necessity rather than a strategic choice.
The State argued that Plaintiffs’ proof was insufficient because they did not show
that the two majority-minority districts will, in fact, result in the election of minoritypreferred candidates. As Plaintiffs point out, no such requirement is stated in Gingles.
The plans need only be illustrative. See Valdespino, 168 F.3d at 852-53. Whether the
districts would actually perform effectively is a question that can be considered at the
remedial phase. Johnson, 512 U.S. at 1013-14 & n.11; Pope v. Cty. of Albany, 687 F.3d
565, 576 (2d Cir. 2012). It is not an element of proving liability. Dickinson v. Indiana
State Election Bd., 933 F.2d 497, 503 (7th Cir. 1991) (citing Gingles, 478 U.S. at 50–51);
16

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 17 of 43

Bone Shirt v. Hazeltine, 461 F.3d 1011, 1019 (8th Cir. 2006); Nipper v. Smith, 39 F.3d
1494, 1532-33 (11th Cir. 1994) (en banc).
The hypothetical plan must also show that the minority group is geographically
compact under traditional redistricting principles. Abrams v. Johnson, 521 U.S. 74, 92
(1997); Fairley v. Hattiesburg, Miss., 584 F.3d 660, 669 (5th Cir. 2009). Dr. Handley
testified to the compact nature of her plans. 2 Tr. 151:7-11. The single-member districts
are contiguous (each district is a single connected piece), follow existing administrative
boundaries (counties), and should take into consideration any known communities of
interest. League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 433 (2006).
The State did not offer a controverting expert analysis of Dr. Handley’s maps.
Instead, it challenged Dr. Handley’s analysis on three bases: (1) failure to consider
communities of interest; (2) failure to apply objective measures of compactness; and (3)
ipse dixit. The first two issues are components of an overall determination of whether the
plans comport with traditional redistricting principles: “Such principles include, inter
alia: compactness, contiguity, and respect for political subdivisions; avoiding contests
between incumbent representatives; not disrupting preexisting electoral minorityopportunity districts; and maintaining communities of interest and traditional
boundaries.” Gonzalez, 601 F. App’x at 259 (citations omitted).
Any variation in districting plans will, by necessity, value some principles over
others.

Thus, the objective of a challenge is to show that the legitimate political

objectives sought are comparably consistent with these principles. Id. It is the Court’s
unwelcome task to make this imprecise determination. The Court FINDS on this record
17

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 18 of 43

that the challenges to Plaintiffs’ illustrative plans are insufficient to defeat the first
precondition.
Communities of Interest. It is true that Dr. Handley did not testify as to whether
her illustrative single-member district plans interfered with, or accommodated, separate
communities of interest. And although the State questioned her about the size of the
districts and the various regions and cities that were combined or separated by the
districts drawn in the demonstrative map, there was no evidence regarding what
communities of interest might have existed and whether they were improperly combined
or separated. 3 Tr. 17-20. Neither has the State cited any authority for a holding that
silence on this consideration is fatal to proof of the first precondition.
Because it is only one factor to be considered on the issue of comportment with
traditional redistricting principles, the Court declines to find its omission on this record to
be a defect in prima facie proof. Dr. Handley testified credibly as to having successfully
accommodated interests of compactness, contiguity, and respect for political
subdivisions, all of which are legitimate considerations.

Absent evidence that the

illustrative plans affirmatively violate a redistricting principle related to communities of
interest, the Court rejects the challenge.
Objective Measures. The State suggests that Dr. Handley’s illustrative maps and
her testimony supporting them are also defective because she did not apply any objective
tests that provide a concrete measure of compactness. See Rodriguez v. Harris Cty., Tex.,
964 F. Supp. 2d 686, 740 (S.D. Tex. 2013) (discussing statistical measures of
compactness, referring to “area rubber band, perimeter-to-area, and population-rubber
18

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 19 of 43

band” metrics), aff’d sub nom., Gonzalez, 601 F. App’x at 255; see also Comm. for a Fair
& Balanced Map v. Illinois State Bd. of Elections, 835 F. Supp. 2d 563, 570 (N.D. Ill.
2011) (discussing the Polsby–Popper measure and the Reock indicator). Yet, the State
did not conduct its own objective statistical measure of compactness and has not
identified any feature of the illustrative maps that is particularly vulnerable to objections
on this basis.
As was the case with the State’s objection based on communities of interest, the
State faults Plaintiffs for failing to defend their demonstrative maps on a basis that has
not been attacked. And despite the existence of mathematical measures, there is still the
“eyeball test” by which the Court may make a visual inspection of the map to determine
whether the district is compact or unduly contorted. Commission, 835 F. Supp. 2d at 570.
Plaintiffs’ maps pass that test.
The Court has reviewed the maps and Dr. Handley’s testimony as to how she
created them. The Court is satisfied that the districts are reasonably compact without
application of mathematical metrics. Again, the compactness of the districts is only one
factor to be considered in determining whether the maps are consistent with traditional
redistricting principles. The Court rejects the challenge that the failure to calculate
objective metrics is dispositive of the first precondition.
Ipse Dixit. Dr. Handley testified that the Hispanic population groups in her
single-member districts are sufficiently large and geographically compact as to satisfy the
first Gingles precondition.

2 TR 151:7-11. The State objects to this testimony as

inadmissible ipse dixit, devoid of evidentiary value. Guile v. United States, 422 F.3d 221,
19

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 20 of 43

227 (5th Cir. 2005). This challenge suggests that her ultimate opinion, stated in the cited
reference, was unsupported in the rest of her testimony. Rather, this opinion testimony
was a summation of the evidence she had developed regarding Hispanic CVAP,
demographics, voting patterns, and the division of the state into geographic districts.
The Court rejects this challenge.
The Court FINDS that Plaintiffs have met the first precondition. The Hispanic
minority group is sufficiently large and geographically compact to constitute a majority
in a single-member district.
2. Politically Cohesive Minority
The second and third preconditions are components of the Supreme Court’s
definition of racially polarized voting, which in turn is—as a single concept—a factor of
the totality of circumstances test. “The purpose of inquiring into the existence of racially
polarized voting is twofold: to ascertain whether minority group members constitute a
politically cohesive unit and to determine whether whites vote sufficiently as a bloc
usually to defeat the minority’s preferred candidates.”

Gingles, 478 U.S. at 56

(emphasis added).
The [trial] court used the term “racial polarization” to
describe this [statistically significant] correlation [between
the race of the voter and the voter’s choice of certain
candidates]. It adopted Dr. Grofman’s definition—“racial
polarization” exists where there is “a consistent relationship
between [the] race of the voter and the way in which the voter
votes,” or to put it differently, where “black voters and white
voters vote differently.” We, too, adopt this definition of
“racial bloc” or “racially polarized” voting.
Id. at 54 n.21.
20

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 21 of 43

“If the minority group is not politically cohesive, it cannot be said that the
selection of a multimember electoral structure thwarts distinctive minority group
interests.” Id. at 51. “For purposes of § 2, the legal concept of racially polarized voting
incorporates neither causation nor intent.

It means simply that the race of voters

correlates with the selection of a certain candidate or candidates; that is, it refers to the
situation where different races (or minority language groups) vote in blocs for different
candidates.” Id. at 62. Correlation rather than causation is the test because “the reasons
black and white voters vote differently have no relevance to the central inquiry of § 2.
By contrast, the correlation between race of voter and the selection of certain candidates
is crucial to that inquiry.” Id. at 63.
Therefore, this precondition is proven by statistical evidence that “a significant
number of minority group members usually vote for the same candidates.” Id. at 56. The
preferred candidate need not be a member of the minority group. E. Jefferson Coal. for
Leadership & Dev. v. Par. of Jefferson, 926 F.2d 487, 493 (5th Cir. 1991). And the
analysis preferably covers a sufficient time to display a consistent pattern of racial bloc
voting. Gingles, 478 U.S. at 57 & n.25. A statistically significant level of cohesion to
satisfy this precondition has been found at a level as low as 51%. Patino v. City of
Pasadena, 230 F. Supp. 3d 667, 710-11 (S.D. Tex. 2017) (51%); Fabela v. City of
Farmers Branch, Tex., No. 3:10-CV-1425-D, 2012 WL 3135545, at *11 (N.D. Tex. Aug.
2, 2012) (54.1%, considered in context with a range of elections with higher cohesion
levels).

21

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 22 of 43

Dr. Handley performed the necessary statistical analysis using three established
techniques: homogenous precinct analysis, ecological regression, and King’s ecological
inference. 2 Tr. 156. See generally Teague, 92 F.3d at 290 (there is an “established
acceptance of regression analysis as a standard method for analyzing racially polarized
voting,” citing Gingles, 478 U.S. at 52-53 n.20). She used Texas’s Spanish surname
database to analyze Hispanic voting preferences in statewide judicial elections. 2 Tr.
160-61. There was no suggestion that these techniques were infirm. Rather, the State’s
expert, Dr. John Alford, used the same tools and reached the same results. 3 Tr. 207,
249-50.
Dr. Handley demonstrated that in 11 of the 13 relevant electoral contests, at least
70% of Latino voters preferred the same candidate. E.g., 2 Tr. 181. Dr. Alford produced
similar results, although he advocated a higher threshold for finding legally significant
minority political cohesion.

3 Tr. 235, 282 (recommending requiring 80 to 90%

cohesion). While Dr. Alford suggested that cohesion levels of 60-70% were too low to
be significant, he did not articulate any factual or methodological reason for his opinion
and he agreed that Hispanics voted cohesively for their preferred candidate. 3 Tr. 25860. His testimony that over 70% was required for compliance with Gingles is not
corroborated in the briefing. See 3 Tr. 236.
The Court FINDS that Plaintiffs have met the second Gingles precondition. The
record demonstrates that the minority group members—Latinos—constitute a politically
cohesive unit.

22

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 23 of 43

3. Majority Bloc Voting
As with the second precondition, the third Gingles precondition is demonstrated
through statistical evidence that the submergence of the minority group in a white
majority multimember district impedes the minority group’s ability to elect its chosen
representatives. See generally League of United Latin Am. Citizens #4552 (LULAC) v.
Roscoe Indep. Sch. Dist., 123 F.3d 843, 847 (5th Cir. 1997). While the results need not
show invariable white bloc voting success, white bloc success must be the usual result.
Teague, 92 F.3d at 288.
The amount of white bloc voting that can generally
“minimize or cancel” black voters’ ability to elect
representatives of their choice, however, will vary from
district to district according to a number of factors, including
the nature of the allegedly dilutive electoral mechanism; the
presence or absence of other potentially dilutive electoral
devices, such as majority vote requirements, designated posts,
and prohibitions against bullet voting; the percentage of
registered voters in the district who are members of the
minority group; the size of the district; and, in multimember
districts, the number of seats open and the number of
candidates in the field.
Gingles, 478 U.S. at 56 (citations omitted).
Dr. Handley used her three statistical techniques to analyze the data. To get an
even better understanding of the white bloc data, she used United States census results in
an effort to carve out black voters from the non-Hispanic group, rendering a better
comparison of Latinos to white voters. 2 Tr. 211-12. She again demonstrated that nonHispanic voters consistently preferred different candidates from those preferred by
Hispanics and did so at rates of over 60%. E.g., 2 Tr. 174. When she included the census
23

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 24 of 43

results, the white bloc voting rate was as high as 84.4% in an election involving
candidates from both major parties. PX 127.
Dr. Alford’s analysis did not negate the results Dr. Handley reached. The pattern
of voting demonstrated that white voters often voted as a bloc for a candidate that the
minority group opposed. 3 Tr. 212-16. The difference in Dr. Alford’s testimony is that
he focused on issues other than the ethnicity of the voters and their preferred
candidates—which are the issues relevant to bloc voting. He instead looked at elections
from the point of view of the ethnicity or race of the candidate, suggesting that there was
no racial polarization in contests where Hispanic voters voted for the white candidate
over the Hispanic candidate. 3 Tr. 213-16. And he attributed results strictly to partisan
preference rather than racial bloc voting. Id.
At this juncture, the Court is only concerned with whether there is a pattern of
white bloc voting that consistently defeats minority-preferred candidates. That analysis
requires a determination that the different groups prefer different candidates, as they do.
It does not require a determination of why particular candidates are preferred by the two
groups.
The Court FINDS that Plaintiffs have met the third Gingles precondition.
Plaintiffs have demonstrated that the white majority votes sufficiently as a bloc to enable
it—in the absence of special circumstances—usually to defeat the minority’s preferred
candidate.

24

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 25 of 43

B. The Three Preconditions Have Been Met
The State argues that Plaintiffs run afoul of direct Fifth Circuit precedent when
their proof on the three preconditions fails to relate their preferred candidates’ lack of
electoral success to racial bias in the electoral system. D.E. 108, pp. 8-9, 11 (citing
LULAC, 999 F.2d at 851, 857). As addressed in the section on the standard of review,
LULAC did not assign any racial bias issue to the three Gingles preconditions. The
preconditions require showing that there is a racial divide among the voters and that
single-member districts would provide the minority group with fair access in contrast to
the at-large elections in which white bloc voting consistently defeats their preferred
candidate.
The State’s argument treats the preconditions as the sum total of Plaintiffs’ case.
It is not. Rather, Plaintiffs must still demonstrate that the vote dilution of which they
complain is “on account of race.” And they still have the totality of circumstances test in
which to do so. Nothing in LULAC cabins racial bias issues inside the boundaries of the
proof required on the preconditions. Thus the State is correct that Plaintiffs’ case must
include evidence of racial bias, but it is not correct that that evidence is part of the
Gingles preconditions; it is part of the totality of the circumstances test.
Plaintiffs have met all three Gingles preconditions. Thus, they have demonstrated
that the at-large system of voting for justices of the Supreme Court of Texas and judges
on the Texas Court of Criminal Appeals works to their disadvantage and that it is
possible to remedy that disadvantage through single-member districts. However, they are

25

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 26 of 43

not entitled to relief unless they also show that, under the totality of the circumstances,
their electoral disadvantage is “on account of race,” as discussed below.
III.

Totality of the Circumstances

When the Gingles preconditions are met, the court then “considers whether, on the
totality of circumstances, minorities have been denied an equal opportunity to participate
in the political process and to elect representatives of their choice.” Abrams v. Johnson,
521 U.S. 74, 91 (1997) (internal quotation marks omitted; citing Section 2); N.A.A.C.P. v.
Fordice, 252 F.3d 361, 366 (5th Cir. 2001). After hearing the evidence set out in more
detail below, the Court FINDS that Plaintiffs have produced some evidence of racial bias
in the electoral system in the factors of the totality of circumstances test. Likewise, the
Court FINDS that the State has advanced evidence of alternative explanations.
Therefore, the Court proceeds to consider the weight to be accorded to the
evidence under each of the factors, and balances it to determine whether Plaintiffs, as
Hispanic voters or an organization advocating for the interests of Hispanic voters, have
had their voting power for electing justices and judges of Texas’s high courts diluted on
account of race. Plaintiffs need not prevail on all or a particular number of the factors.
Clark, 88 F.3d at 1401. However, before a remedy may be imposed, the Court must be
convinced that, on balance, the State has violated Section 2’s concern over discrimination
in elections on account of race.

26

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 27 of 43

A. Discussion and Weighing of Factors
1. History of Official Discrimination
Texas has a long history of official racial discrimination with respect to voting
rights. Veasey v. Abbott, 830 F.3d 216, 257 n.53 (5th Cir. 2016), cert. denied, 137 S. Ct.
612 (2017); 2 Tr. 11-15 (Dr. Henry Flores’s testimony regarding Texas’s poll tax, white
primaries, and at-large voting for legislators).

The vestiges of that discrimination

continue to be felt in racial disparities in socioeconomic conditions, which depress
political participation among minority voters. Veasey, supra.; Teague, 92 F.3d at 294; 2
Tr. 45-46. Noting this context, the Court focuses its analysis on recent evidence of
discrimination. See McCleskey v. Kemp, 481 U.S. 279, 298 n.20 (1987).
The Fifth Circuit, sitting en banc, upheld this Court’s finding that the 2011 Texas
voter identification law produced discriminatory results.

Veasey, supra.

The State

further defended a racial gerrymandering case addressing legislative districts, resulting in
a Supreme Court holding that one such district was racially gerrymandered. Abbott v.
Perez, 138 S. Ct. 2305, 2335 (2018). Plaintiffs have not, however, identified any specific
history of official discrimination with respect to establishing or maintaining the
multimember nature of voting for the State’s high courts. The Court thus FINDS that this
factor has only slight weight, favoring Plaintiffs.

27

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 28 of 43

2. Racially Polarized Elections
The bulk of the evidence at trial involved the analysis of Texas high court
elections since 2002,9 demonstrating significant polarization. While it is undeniable and
effectively conceded that Hispanic-preferred candidates for judicial office have
consistently fared poorly in statewide elections, the impetus behind that phenomenon is
the crux of this case. See 2 Tr. 180. Plaintiffs claim that it is a racial issue; the State
claims that it is a partisanship issue, which only happens to correlate with race. 10 Thus,
the decision requires parsing of closely aligned evidence of race and party.
Most of the election contests shine little light on this nuanced issue. Time and
again, when a general election race involved both major parties, the white-preferred
candidate won over the Latino-preferred candidate. Consistently throughout the election
cycles considered here, the white-preferred candidate ran as a Republican and the Latinopreferred candidate ran as a Democrat.11 The two-major-party general election contests
and their uniform results facially support equal inferences in favor of racial polarization
and partisan polarization. So the key lies somewhere outside of the macro-analysis of
Republican versus Democratic candidate or Latino versus non-Latino voter.

9

Dr. Handley testified that she used this time frame because it provided a sufficient number of elections to support
statistical conclusions and because 2002 was the first year the necessary data was available on the Texas legislative
website. 2 Tr. 163.
10

Dr. Handley agreed that most Hispanics are Democrats or vote for Democratic candidates. 2 Tr. 197.

11

In 11 out of 13 contested general elections for statewide judicial offices from 2002 to 2016, Hispanic voters
preferred the Democratic candidate at levels of 70 to 85%. PX 124; 2 Tr. 202-203. Most non-Hispanic voters
preferred the Republican candidates, although 30 to 40% of non-Hispanic voters supported the Democratic
candidates. PX 124; 2 TR. 220.

28

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 29 of 43

Race of the Candidate. Ordinarily, in considering vote dilution, the race of the
candidate is not the relevant issue. Instead, it is whether minority-preferred candidates
(of whatever race) are consistently defeated. Gingles, 478 U.S. at 68. However, when
faced with the challenge of determining whether race rather than partisanship explains
polarization, looking at the race of the candidate may offer some insight. LULAC, 999
F.2d at 854 (discussing the need to investigate built-in bias) and 861 (crediting evidence
that there was consistency in partisanship even where the race of the candidate varied); 2
Tr. 234.
The State took great care to point out that the data shows a highly consistent
correlation—across elections and election cycles—between racial group voting and
political parties. 2 Tr. 209-20. More to the point, the distribution of votes between
political parties remained at comparable levels even when the race of the candidate varied
or when the same candidate switched parties. Id.; 2 Tr. 262-64.
As an important case in point, in 2016, Justice Eva Guzman—a Latina incumbent
running as a Republican—did not get the Latino vote, which went to the Democratic
candidate.

Justice Guzman nonetheless won more non-Hispanic support than white

incumbent Justice Paul Green. PX 124, p. 4. In fact, Justice Guzman received more
votes than any other candidate for statewide judicial office in Texas history. DX 40, p. 4.
While Plaintiffs point out that this single example does not support a finding of a
pattern of a lack of racial bias, it is consistent with other evidence demonstrating that the
race of the candidate is less important to voters than the candidate’s party. In fact,
whether running as Democrats or Republicans, Hispanic candidates for high judicial
29

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 30 of 43

office have tended to slightly outperform non-Hispanic candidates with non-Hispanic
voters. 2 Tr. 223, 227; DX 25, DX 26. This evidence supports a finding that partisanship
is a better explanation for defeats of Hispanic-preferred candidates than racial vote
dilution.
Party Primaries.

Dr. Handley concluded that two of the three Democratic

primaries she analyzed were polarized and, between them, the Latino-preferred candidate
won once and lost once. 2 Tr. 266-68. These results do not reveal a pattern. For that
very reason, it is interesting to note that the premise for a Latino vote dilution claim
erodes when only Democratic candidates are before the voters. This is some evidence
that partisanship is the dominant factor in election results.
On the flip side, Dr. Handley opined that Republican primaries were not probative
because none of the three Republican primaries and the one Republican runoff she
analyzed drew more than 1% of registered Hispanic voters.

2 Tr. 165, 183-85.

Nonetheless, both she and Dr. Alford reviewed the data on those elections, using Dr.
Alford’s data because it more accurately controlled for low turnout. Using his data for
the Republican primaries and runoff she analyzed, Dr. Handley testified that the
Hispanic-preferred candidate won all four. 2 Tr. 279; PX 126. Expanding the analysis to
37 total Republican primaries or runoffs, half were racially polarized and half were not.
2 Tr. 188, 280; PX 132, 134.

Again, without the Democrat versus Republican

dichotomy, data show that the premise for vote dilution erodes.
However, Plaintiffs argue that simply analyzing voting data understates the
relative disenfranchisement of Hispanic voters because such analysis does not reflect the
30

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 31 of 43

phenomenon whereby Hispanic incumbents are more likely than white incumbents to
draw primary challengers. That is, using Republican primaries for Supreme Court of
Texas seats as an example, white candidates have challenged incumbent Latino justices at
a rate of 80%, compared to just 50% for white incumbents. 1 Tr. 136-37. And when
challenged, only 60% of Latino incumbents have prevailed, compared to 95% of white
incumbents. 1 Tr. 137. The only white incumbent to lose, Steven Wayne Smith, lost
multiple times against white opponents; his sole victory was against Hispanic Xavier
Rodriguez. 1 Tr. 141, 145-46. Judge Elsa Alcala testified that her vulnerability as a
Latina to this phenomenon influenced her decision not to seek reelection in 2018. 1 Tr.
265-67. This is some evidence that racial bias affects vote dilution.
Races without a Democratic Candidate. In approximately two-thirds of the
analyzed general elections in which there was no Democratic candidate, Latino voters
supported third-party candidates over Republican ones and the election was polarized. 2
Tr. 190 (11 out of 16 contests). However, with the exception of one race in which 60.4%
of Latinos voted for the Libertarian candidate, the Latino support for the losing candidate
did not exceed 60%, so those contests could not be considered cohesive. 2 Tr. 242, 24657.
Other times, Latino voters joined non-Latino voters in supporting the Republican
candidate. 2 Tr. 239-41, 247. As a result, those elections were not polarized. 2 Tr. 240241. For instance, at the time of the analysis, only one sitting judge, Judge Alcala, had
won election to a Texas high court as the Hispanic-preferred judge. 2 Tr. 199-200; PX
133. Judge Alcala ran in 2012 as a Republican and, with the support of a majority of
31

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 32 of 43

Hispanic voters, prevailed against a Libertarian candidate. 2 Tr. 199-200; PX 127. These
races support the conclusion that, once again, the evidence of vote dilution weakens when
the Democrat-Republican partisan divide is absent.
The State’s statistical analysis supporting the partisan explanation was compelling.
In each area where racial bias fails and partisanship appears to dominate voter decisions,
Plaintiffs have challenged the evidence as not probative. However, as explained, this is a
nuanced question and it depends not upon strong trends, but subtle cues. The Court
FINDS that partisan polarization better explains results in recent Texas statewide
elections for high judicial office. Because it is reflected only in non-dominant themes,
and because the primary challenge phenomenon is concerning, the Court gives it only
moderate weight. Therefore, this factor weighs moderately in favor of the State.
3. Practices that Enhance Discrimination in Voting
Under this factor, Plaintiffs cite three circumstances or practices in high court
elections that enhance discrimination in voting: (1) the unusually large election district;
(2) the majority-vote requirement; and (3) the numbered-place requirement. The Court
will consider each in turn.
Large District. The size of the election district is a function of the size of the
state.

And the State’s linkage interest will be addressed separately within.

While

discriminatory intent is not a requirement of the burden of proof, nothing in Plaintiffs’
evidence suggests that the State’s at-large system was chosen or maintained because of
its potential to dilute a racial minority’s vote.

32

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 33 of 43

Given the lower socioeconomic status of minority voters, Plaintiffs’ complaint is
that the size of the district presents particularly formidable obstacles for minoritypreferred candidates to campaign across the state and to raise sufficient funds to be
competitive. D.E. 102, pp. 22-23. However, the challenges of physically covering the
entire state are shared by all candidates. And to the extent that additional funds are
necessary, the socioeconomic legacy of discrimination is part of the fifth factor,
addressed below.
Fifty-Percent Threshold Majority Vote. The majority vote requirement ensures
a runoff between the two candidates receiving the highest number of votes if no candidate
receives over 50%. While minority-preferred candidates have occasionally garnered the
most votes when the white vote is split, a runoff often has the effect of pitting a minoritypreferred candidate against a majority-preferred candidate.

This precise scenario

occurred with respect to the 2012 Republican primary election that included Justice
David Medina, who earned the most votes cast in a three-way race, but lost the runoff to
the majority-preferred candidate who went on to win the general election. 2 Tr. 186-87.
The record contains no evidence of the rationale for imposing the majority vote
requirement. This works in Plaintiffs’ favor.
Numbered Place. According to Plaintiffs, the numbered-seat requirement offers
two features that work to disadvantage minority voters. First, it allows the majority to
use its dominant voting position to prevent a cohesive minority from succeeding in any of
the individual judicial races. Rogers v. Lodge, 458 U.S. 613, 627 (1982). Second, it
allows candidates to challenge specific incumbents, more often the incumbents preferred
33

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 34 of 43

by the minority. See generally 1 Tr. 247. These arguments are two sides of the same
coin. And the Supreme Court has held that the use of a numbered place system is not
infirm, merely because it has this effect. Whitcomb, 403 U.S. at 159. Indeed, Dr. Henry
Flores was unaware of any state electing multimember bodies that does not use a place
system. 2 Tr. 82.
The Court FINDS that the practices associated with a large multimember district
with a majority vote requirement and numbered places have little weight because there is
no evidence that racial bias rather than practical necessity motivated the adoption of these
practices. That weight slightly favors Plaintiffs
4. Exclusive Slating Processes
The Texas system of voting for justices and judges for the high courts does not
involve a formal slating process. Yet candidates can achieve an incumbency advantage if
first appointed by the governor to fill an unexpired term. This is a fairly common
scenario, as over half of the justices who have held seats on the Supreme Court of Texas
and a third of the judges on the Court of Criminal Appeals over the analyzed period
began their tenure with gubernatorial appointments. 2 Tr. 31, 33; 3 Tr. 28-30; PX 106,
106A.
While all of the governors responsible for selecting the appointees have been
white and the majority of appointees have been white, Latinos who have been appointed
to high judicial office have enjoyed some incumbency advantage, even if it is less of an
incumbency advantage than white judges and justices. For instance, Supreme Court
Justices Medina and Guzman and Court of Criminal Appeals Judge Alcala all won
34

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 35 of 43

reelection after their appointments. 1 Tr. 142; 2 Tr. 102. In addition, there is some
evidence that Republican governors have prioritized appointing Latinos to high judicial
office and have looked beyond members of their own party when considering candidates
for appointment. See 3 Tr. 74-75, 174. But for instance, Justices Linda Yanez and Gina
Benavides, both of whom are Democrats who have served on the Thirteenth Court of
Appeals, chose not to pursue potential appointments at least in part because they were
asked to pledge to run for reelection as Republicans. They declined the appointments
because they either identified more with the Democratic party or because they did not
believe that they could get elected as Republicans. Id.
Plaintiffs discount this evidence as anecdotal.

And it does not account for

eventual targeted challenges from white candidates in the Republican primaries, making
the advantage of incumbency less useful for Hispanic candidates in the long run.
In short, it appears that once appointed, Hispanic judges and justices enjoy less of
an incumbency advantage than non-Hispanic judges and justices and they face a greater
likelihood of drawing a primary challenger. Yet at least in some instances, partisanship
has interfered with Hispanic candidates accepting incumbency’s advantages. On the
whole, the competing forces involved in accepting and benefiting from appointments and
incumbency are such that the Court gives this, as a slating issue, little weight and neutral
benefit between the parties.
5. Legacy of Discrimination Hindering Minority Participation
As noted above, Texas is no stranger to discrimination against Hispanics. It is also
undisputed that Latinos, generally speaking, suffer a socioeconomic status and voter
35

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 36 of 43

registration rate significantly lower than those of whites. 2 Tr. 35-46; PTX 102, 103,
104, 104A, 105A, 109, 110. Therefore, Plaintiffs argue that the fifth factor weighs in
their favor.
The State argues that Plaintiffs have failed to establish a causal nexus between the
history of discrimination and Latino’s current weaker political power. 2 Tr. 97-98. But
the Fifth Circuit has held that such causal evidence is unnecessary when the history,
socioeconomic status, and political conditions are aligned as they are here. Teague, 92
F.3d at 294. The Court gives this factor heavy weight in favor of Plaintiffs.
6. Racial Appeals
The evidence contains a single instance of a patent racial appeal in a judicial race.
Justice Yanez testified that, during her supreme court candidacy, her white opponent’s
surrogates argued that she was and, if elected, would be a representative of the Mexican
government on the Texas court. 3 Tr. 170-71, 182. Because this is the only such
instance of racial appeals offered, this merits slight weight in Plaintiffs’ favor.
7. Number of Successful Minority Candidates
It is undisputed that no Latino candidate has ever been elected to the Supreme
Court of Texas or the Texas Court of Criminal Appeals without first having been
appointed to an unexpired term. Of 18 appointments to both courts from 1996 to 2016,
only five (27%) were given to Hispanics. DX 42. The evidence noted above shows that
there might have been more Hispanic appointments, but they were declined, either for
partisan reasons or a belief of inability to get elected as a Republican. And whatever the
number of appointments, the reduced incumbency advantage likely makes high judicial
36

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 37 of 43

office less attractive for Hispanic attorneys who might otherwise be inclined to pursue
appointments.
This is reflected in the overall underrepresentation of Hispanics on the high courts
which persists, despite the large number of Latino attorneys qualified to run for the high
court positions. DX 32, 33. See generally Johnson, 512 U.S. at 1014 n.11; LULAC, 999
F.2d at 865. The statistical gap between voting population and successful candidates is
not quite as extensive as Plaintiffs represent, because their statistics include a broad time
period going back to 1945 and 1966 for the high courts. PX 106rev, 106A. Yet their
quantitative analysis of racially polarized voting dates back only to 2002. The Court does
not credit statistical evidence of the alleged effect of vote dilution that predates the
evidence of vote dilution as the cause. See Uno, 72 F.3d at 990 (vote dilution must be
evaluated in the present tense).
As the State notes, Hispanic candidates for high judicial office running as
Democrats have suffered the same fate as nearly all Democrats running in statewide
elections since 2002.

At the same time, Justice Guzman, a Latina running as a

Republican, received the most votes of any Supreme Court of Texas candidate in history.
DX 40, p. 4. When the issue is whether race or partisanship explains the pattern of
electoral outcomes, rather than whether or not there is vote dilution, this factor’s
significance evaporates. The influences on the number of successful minority candidates
are better explained by other factors in the totality of the circumstances test. The Court
treats this factor as neutral.

37

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 38 of 43

8. Lack of Responsiveness to Minority Needs
Plaintiffs point out that the high courts have jurisdiction over many issues of
particular interest to Hispanics. However, they have not identified any examples of
decisions that are non-responsive to minority needs or wrongly decided in the overall
interest of equal justice. The Texas Office of Court Administration Public Trust and
Confidence in the Court Survey, upon which Dr. Jose Juarez relied, found that Hispanics,
as a racial group, expressed the highest degree of confidence in the courts. 1 Tr. 212.
And unlike other elected officials, those in judicial positions or candidates for such
positions, cannot espouse a special interest agenda. Thus, this factor has slight, if any,
weight in the balance of the totality of the circumstances and it is neutral here.
9. Tenuous Reasoning Behind Methodology—Linkage
The Court recognizes that its consideration of the State’s method of selecting
judges implicates important concerns of federalism. For instance, when considering
issues related to a state’s mandatory retirement age for judges, the Supreme Court wrote:
Through the structure of its government, and the character of
those who exercise government authority, a State defines
itself as a sovereign. “It is obviously essential to the
independence of the States, and to their peace and tranquility,
that their power to prescribe the qualifications of their own
officers . . . should be exclusive, and free from external
interference, except so far as plainly provided by the
Constitution of the United States.”
Gregory v. Ashcroft, 501 U.S. 452, 460 (1991) (quoting Taylor v. Beckham, 178 U.S.
548, 570-571 (1900), and citing Boyd v. Nebraska ex rel. Thayer, 143 U.S. 135, 161
(1892)).
38

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 39 of 43

At-large judicial districts have a salutary function.

“The distrust of judicial

subdistricts does not rest on paternalism. It recognizes Texas’ historic interest in having
district judges remain accountable to all voters in their district.” LULAC, 999 F.2d at
873.
The decision to make jurisdiction and electoral bases
coterminous is more than a decision about how to elect state
judges. It is a decision of what constitutes a state court judge.
Such a decision is as much a decision about the structure of
the judicial office as the office’s explicit qualifications such
as bar membership or the age of judges. The collective voice
of generations by their unswerving adherence to the principle
of linkage through times of extraordinary growth and change
speaks to us with power. Tradition, of course, does not make
right of wrong, but we must be cautious when asked to
embrace a new revelation that right has so long been wrong.
LULAC, 999 F.2d at 872. The Court thus proceeds with caution.
Plaintiffs criticize the State for not submitting evidence of its linkage interest.
Yet, as noted above, it is a question of law. “[T]he weight of a state’s interest has always
been a legal question, not a factual one.” Id. at 871. It is true that the LULAC case
involved trial court judges instead of high court appellate judges. Still, any differences in
the State’s interest associated with what constitutes high court appellate judges are
entitled to some deference.
Law professor and legal historian, Michael Ariens, testified that Texas tried
several alternate systems before settling on the statewide, at-large election model for its
high courts for purposes of promoting independence and accountability. Under the 1836
Constitution of the Republic of Texas, the legislature elected the justices. Associate
justices, assigned to divisions of the state, had both trial and appellate functions, while
39

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 40 of 43

the chief justice had exclusively appellate jurisdiction. Jointly, their jurisdiction was
coterminous with the geographical limits of the state. 3 Tr. 87-88. In 1845, Texas
adopted a gubernatorial appointment method. Id., p. 90. A referendum resulted in a
constitutional amendment approved by popular vote and ratified by the legislature,
effective in 1850, which provided for the popular election of supreme court justices. Id.,
pp. 90-91. In 1869, when Texas was under congressional reconstruction, supreme court
justices were nominated by the governor and confirmed by the senate. Id., p. 94.
Since 1876, the Texas constitution has required statewide popular election for all
justices for the Supreme Court of Texas. Id., pp. 95-96; Tex. Const. art. V, § 2. The
same is true with respect to the judges of the Court of Criminal Appeals, since 1891.
Tex. Const. art. V, § 4. Furthermore, the corresponding jurisdiction of both high courts
has been coterminous with the limits of the State. Tex. Const. art. V, §§ 3, 5. Just
because the State’s adoption of this method of judicial selection is at the constitutional
level and dates back to 1876 does not make it immune from attack.12

However, it is

indicative of the State’s deliberate choice, left undisturbed over significant time and with
substantial alternate experience, for governing its citizens.
Plaintiffs argue that linkage is not a compelling state interest at the level of the
high courts because all nine justices and judges on each court rule as a collegial body,
with each individual justice having an opportunity to affect case outcomes. See generally
LULAC, 999 F.2d at 873 (distinguishing the issue of single-member trial courts). “The
12

Plaintiffs point out that the 1876 Texas constitution had provisions that have been challenged and eliminated,
such as male-only juries, paupers not allowed to vote, poll taxes, and mandated school segregation. Tex. Const. of
1876, art. V, § 13; art. VI, §§ 1, 3; art. VII, § 7, available at: http://tarlton.law.utexas.edu/constitutions/texas1876.

40

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 41 of 43

ability to bring diverse perspectives to the court, not the prospect of outright dealmaking,
would be relevant in [an appellate court] context.” Nipper v. Smith, 39 F.3d 1494, 1535
n.78 (11th Cir. 1994). Dr. Juarez testified that, in the context of a multimember body,
single-member districts are preferable because they:

advance judicial effectiveness;

balance accountability and judicial effectiveness; ensure judicial independence; and avoid
prejudice and bias towards the parochial interest of a narrow constituency. 1 Tr. 152-62.
See also LULAC, 999 F.2d at 869 (quoting John L. Hill, Jr., Taking Texas Judges Out of
Politics: An Argument for Merit Election, 40 Baylor L. Rev. 339, 364 (1988) (emphasis
omitted)).
Even assuming some benefit, however, single-member districts have fundamental
drawbacks. For instance, the State notes that it would inject legislative control into the
drawing of judicial districts, contrary to the State’s sovereign interest in a governing
model that includes separation of powers. See generally Tex. Const. art. V, § 7a(g), (h)
(providing for legislative oversight and approval of the determinations of a Judicial
Districts Board created to reapportion the state’s judicial districts); Tex. Gov’t Code
§ 22.201 (setting out legislatively-approved composition of courts of appeals districts).
It would also reduce any given voter’s opportunity to have an impact on the
personnel on the courts from a vote for each of nine (the full composition of the courts) to
one justice or judge on each court (or one vote for the judge or justice from one’s own
district and one vote for the chief judge or justice to be elected at-large). This is
significant because the Supreme Court of Texas is empowered to determine cases by a
majority vote of five justices. Tex. Const. art. V, § 2. The Court of Criminal Appeals
41

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 42 of 43

decides cases in panels of three and by a vote of two judges, unless convened en banc, in
which case a vote of five judges will determine the case. Tex. Const. art. V, § 4. The
collegial nature of a multimember body does not prevent it from, in practice, running
roughshod over any given minority of judges. Single-member districts could have the
unintended effect of increasing the power of a majority of judges elected from districts
with wider polarization levels in favor of white voters.
The State also points out that the creation of single-member districts, as Plaintiffs
advocate, is not a foregone conclusion even if they prevail. Single-member districts are
only one remedy that might ultimately be imposed if liability is found. D.E. 106, p. 15.
See Clark, 21 F.3d at 95 (“If a § 2 violation is found, the [state] will be given the first
opportunity to develop a remedial plan.”). But Plaintiffs have predicated this entire case
on the argument that it is the at-large method of voting that is infirm. If that method has a
purpose, it is entirely appropriate to consider that purpose in determining whether
liability should be imposed. The Court gives this factor heavy, but not dispositive,
weight in favor of the State.
B. Summation of the Totality of the Circumstances
Having considered each of the established factors, the Court has assigned each its
weight and has determined which party is favored by each individual factor. When all
factors are considered together, the Court FINDS that, given the totality of the
circumstances, Plaintiffs have not satisfied their burden to show that the voting
methodology results in a denial or abridgement of the right of any citizen of the United
States to vote on account of race or color. Instead, the Court FINDS that partisanship
42

Case 2:16-cv-00303 Document 109 Filed in TXSD on 09/12/18 Page 43 of 43

rather than race better explains Hispanic defeat at the polls and the evidence was
insufficient to overcome that, combined with the State’s linkage interest.
CONCLUSION
Plaintiffs Araiza, Solis, Rodriguez, and LUPE have standing to bring their claims.
While Plaintiffs did satisfy the three Gingles preconditions for complaining of vote
dilution, the Court concludes that they did not also demonstrate that race rather than
partisanship better explains their preferred candidates’ lack of success at the polls and
they did not overcome the State’s interest in electing Supreme Court of Texas justices
and Court of Criminal Appeals judges in an at-large system, coterminous with the
geographic limits of the State. The State is entitled to judgment that Plaintiffs take
nothing by their claims.
ORDERED this 12th day of September, 2018.
___________________________________
NELVA GONZALES RAMOS
UNITED STATES DISTRICT JUDGE

43