Petteway v. Galveston Cnty.

Full title: Terry Petteway, et al., Plaintiffs, v. Galveston County, et al.…

Court: United States District Court, Southern District of Texas

Date published: Oct 13, 2023

698 F. Supp. 3d 952 (S.D. Tex. 2023)

Opinion

No. 3:22-cv-57

2023-10-13

Terry PETTEWAY, et al., Plaintiffs, v. GALVESTON COUNTY, et al., Defendants.

Bernadette Samson Reyes, Pro Hac Vice, Sonni Waknin, Pro Hac Vice, UCLA Voting Rights Project, Los Angeles, CA, Mark P. Gaber, Pro Hac Vice, Simone Leeper, Pro Hac Vice, Valencia Richardson, Pro Hac Vice, Alexandra Copper, Pro Hac Vice, Campaign Legal Center, Washington, DC, Neil G. Baron, Law Office of Neil G. Baron, League City, TX, Chad W. Dunn, Brazil & Dunn, Austin, TX, for Plaintiffs Honorable Terry Petteway, Honorable Derrick Rose, Honorable Penny Pope, Honorable Sonny James. Neil G. Baron, Law Office of Neil G. Baron, League City, TX, Chad W. Dunn, Brazil & Dunn, Austin, TX, for Plaintiff Roosevelt Henderson. J. Christian Adams, Pro Hac Vice, Maureen S. Riordan, Pro Hac Vice, Public Interest Legal Foundation, Alexandria, VA, Jason Brett Torchinsky, Pro Hac Vice, Shawn T. Sheehy, Pro Hac Vice, Dallin Brockbank Holt, Holtzman Vogel Baran Torchinsky & Josefiak PLLC, Haymarket, VA, Robert Barron Boemer, Galveston County Legal Department, Galveston, TX, Jordan Raschke Elton, Joseph R. Russo, Jr., Greer Herz Adams LLP, Galveston, TX, Angela Olalde, Greer Herz Adams LLP, League City, TX, Dalton L. Oldham, Austin, TX, James Edwin Trainor, III, Trainor Law Firm, PC, Austin, TX, Joseph M. Nixon, The Nixon Law Firm, PC, Houston, TX, for Defendants Texas Galveston County, Honorable Mark Henry. Dallin Brockbank Holt, Holtzman Vogel Baran Torchinsky & Josefiak PLLC, Haymarket, VA, Angela Olalde, Greer Herz Adams LLP, League City, TX, Jordan Raschke Elton, Joseph R. Russo, Jr., Greer Herz Adams LLP, Galveston, TX, for Defendant Dwight D. Sullivan. Randy Ray Howry, Howry Breen Herman LLP, Austin, TX, for Defendant Stephen Holmes.

Jeffrey Vincent Brown, United States District Judge

Bernadette Samson Reyes, Pro Hac Vice, Sonni Waknin, Pro Hac Vice, UCLA Voting Rights Project, Los Angeles, CA, Mark P. Gaber, Pro Hac Vice, Simone Leeper, Pro Hac Vice, Valencia Richardson, Pro Hac Vice, Alexandra Copper, Pro Hac Vice, Campaign Legal Center, Washington, DC, Neil G. Baron, Law Office of Neil G. Baron, League City, TX, Chad W. Dunn, Brazil & Dunn, Austin, TX, for Plaintiffs Honorable Terry Petteway, Honorable Derrick Rose, Honorable Penny Pope, Honorable Sonny James.

Neil G. Baron, Law Office of Neil G. Baron, League City, TX, Chad W. Dunn, Brazil & Dunn, Austin, TX, for Plaintiff Roosevelt Henderson.

J. Christian Adams, Pro Hac Vice, Maureen S. Riordan, Pro Hac Vice, Public Interest Legal Foundation, Alexandria, VA, Jason Brett Torchinsky, Pro Hac Vice, Shawn T. Sheehy, Pro Hac Vice, Dallin Brockbank Holt, Holtzman Vogel Baran Torchinsky & Josefiak PLLC, Haymarket, VA, Robert Barron Boemer, Galveston County Legal Department, Galveston, TX, Jordan Raschke Elton, Joseph R. Russo, Jr., Greer Herz Adams LLP, Galveston, TX, Angela Olalde, Greer Herz Adams LLP, League City, TX, Dalton L. Oldham, Austin, TX, James Edwin Trainor, III, Trainor Law Firm, PC, Austin, TX, Joseph M. Nixon, The Nixon Law Firm, PC, Houston, TX, for Defendants Texas Galveston County, Honorable Mark Henry.

Dallin Brockbank Holt, Holtzman Vogel Baran Torchinsky & Josefiak PLLC, Haymarket, VA, Angela Olalde, Greer Herz Adams LLP, League City, TX, Jordan Raschke Elton, Joseph R. Russo, Jr., Greer Herz Adams LLP, Galveston, TX, for Defendant Dwight D. Sullivan.

Randy Ray Howry, Howry Breen Herman LLP, Austin, TX, for Defendant Stephen Holmes.

FINDINGS OF FACT AND CONCLUSIONS OF LAW

Jeffrey Vincent Brown, United States District Judge.

Table of Contents

I. Introduction . . . 961

II. Findings of Fact . . . 962

A. Procedural History . . . 962

B. Parties . . . 964

1. Plaintiffs 964 2. Defendants . . . 965

C. Expert Witnesses' Credibility . . . 965

1. Gingles Precondition One . . . 965

2. Gingles Preconditions Two and Three . . . 967

3. Senate Factors and Arlington Heights Factors . . . 969

D. Galveston County Demographics and Voting Patterns . . . 970

1. Sufficiently Large and Geographically Compact . . . 971

a. Traditional Redistricting Criteria . . . 972

b. Geographically Compact . . . 974

2. Politically Cohesive . . . 975

3. Cannot Elect Candidate of Choice . . . 978

4. On Account of Race . . . 979

E. Discriminatory Impact of the Enacted Plan . . . 980

F. Galveston County Voting and Redistricting . . . 981

1. History of Discrimination in Voting Practices . . . 981

2. Attorney General's Objections . . . 982

3. Public Input and Transparency in Prior Redistricting . . . 984

G. The 2021 Redistricting Process . . . 985

1. Sequence of Events . . . 985

a. April 2021—Engaging Redistricting Counsel . . . 985

b. August 2021—Census Data Released . . . 986

c. October 14—Hiring a Demographer . . . 987

d. October 17—Bryan Creates Map 2 . . . 987

e. Late October—Finalizing and Announcing the Maps . . . 988

f. November 12—The Enacted Plan . . . 989

2. Deviations from Prior Redistricting Cycles . . . 990

a. No Redistricting Timeline . . . 990

b. No Redistricting Criteria . . . 991

c. Lack of Transparency in Engaging Counsel . . . 991

d. Lack of Public Notice and Comment . . . 991

e. The November 12 Special Meeting . . . 993

f. Disregard for Public Input from Minority Residents . . . 994

g. Excluding Commissioner Holmes . . . 995

3. Purported and Actual Redistricting Criteria . . . 995

H. Ongoing Discrimination Touching on Participation in Voting . . . 997

1. Contemporary Voting Barriers . . . 998

2. Lack of Electoral Success . . . 999

3. Responsiveness . . . 1000

4. Education . . . 1001

5. Employment and Poverty . . . 1001

6. Housing . . . 1002

7. Public Health . . . 1003

8. Criminal Justice . . . 1003

III. Conclusions of Law . . . 1003

A. Section 2 of the Voting Rights Act . . . 1004

1. Step One—Preconditions . . . 1005

a. Sufficiently Large and Geographically Compact . . . 1006

b. Political Cohesion . . . 1009

c. Cannot Elect Candidate of Choice . . . 1010

2. Step Two—Totality of the Circumstances . . . 1011

3. Strict Scrutiny . . . 1015

B. Remaining Constitutional Claims . . . 1016

IV. Relief . . . 1017

V. Conclusion . . . 1018 I. Introduction

This is a redistricting case brought under the Voting Rights Act and the Fourteenth and Fifteenth Amendments. It was tried to the bench from August 7-18.

On the third day of trial, William S. Cooper¹ — one of the experts for the NAACP plaintiffs² — perfectly described the heart of this case, which challenges the commissioners-precinct plan that the Galveston County Commissioners Court adopted in November 2021 ("the enacted plan") that dismantled Precinct 3 — the only Black-and-Latino-dominant³ precinct in the county:

1 As noted infra, Cooper has nearly four decades of experience drawing voting plans for about 750 United States jurisdictions. Dkt. 223 at 9-10. He has testified as an expert on redistricting and demographic analysis in federal court fifty-five times. Id. at 10; see also PX-341.

2 The NAACP plaintiffs include Dickinson Bay Area Branch NAACP, Galveston Branch NAACP, Mainland Branch NAACP, Galveston LULAC Council 151, Edna Courville, Joe A. Compian, and Leon Phillips. United States v. Galveston County, No. 3:22-cv-97, ECF No. 38 (May 31, 2022).

3 Unless otherwise specified, the court uses the term "Black" to refer to individuals who identify as Black or African American. It also uses "Latino" to refer to individuals who identify as Latino or Hispanic and "Anglo" for those who identify as White/Caucasian.

Q. What, if anything, do you observe about the differences between [the] benchmark and now the new 2021 enacted [plan]?

A. Well, if you look at the underlying census data, Precinct 3 went from being a Black plus Latino majority precinct to being a precinct with the lowest percentage of Blacks and Latinos in the county. . . . It's just a textbook example of a racial gerrymander.⁴ It's — it's egregious. I have never seen anything this bad. Because normally if a minority-majority district is in place, then you are not going to see a locality attempt to eliminate it unless [it] had no choice due to demographic changes.

Here there was absolutely no reason to make major changes to Precinct 3. It was just — it was mean-spirited. I've never — I mean, I'm just blown away by this. It's not fair, and . . . I am at a loss for words.

4 Although the Petteway plaintiffs challenged the enacted plan under the Constitution as a racial gerrymander, the court decided this matter under the Voting Rights Act. Accordingly, it does not reach the racial-gerrymandering claim.

Dkt. 223 at 42-43.⁵ The court finds the defendants' actions to be fundamentally inconsistent with § 2 of the Voting Rights Act. Although Galveston County is no longer subject to preclearance, the defendants still must comply with the edicts of § 2. They have not done so here. So the court has reached a grave conclusion: it must enjoin the defendants from using the enacted map in future elections.

5 Page citations refer to the PDF page number, not the document's internal pagination.

* * *

On June 1, 2022, the court consolidated Civil Action Nos. 3:22-cv-93 and 3:22-cv-117 with Civil Action No. 3:22-cv-57, resulting in one action under Civil Action No. 3:22-cv-57. Dkt. 45. All three sets of plaintiffs — the Petteway plaintiffs, NAACP plaintiffs, and the United States

challenge the enacted plan as violating § 2 of the Voting Rights Act. The Petteway and NAACP plaintiffs also challenge the enacted plan as (1) intentionally discriminatory against Galveston County's Black and Latino voters in violation of the Fourteenth and Fifteenth Amendments and (2) racially gerrymandered in violation of the Fourteenth Amendment.

The court convened a bench trial on August 7, 2023, which lasted until August 18. After thoroughly reviewing the entire record, the court finds that the enacted plan illegally dilutes the voting power of Galveston County's Black and Latino voters by dismantling Precinct 3, the county's historic and sole majority-minority commissioners precinct. The enacted plan distributes the county's Black and Latino voters, who comprise 38% of the county's eligible voter population, among all four newly drawn commissioners precincts. As a result, those minority voters have been subsumed in majority-Anglo precincts in a county with legally significant racially polarized voting. Under the enacted plan, Anglo voters will likely continue to vote as a bloc to usually elect candidates who are not the Black and Latino voters' candidates of choice, preventing Black and Latino voters from participating equally in county government.

The court finds in favor of the plaintiffs and enjoins the use of the enacted plan.

II. Findings of Fact

1. Findings of fact and conclusions of law are required in all actions "tried on the facts without a jury." Fed. R. Civ. P. 52(a)(1). A district court must "find the facts specially and state its conclusions of law separately." Id. "Rule 52(a) does not require that the district court set out findings on all factual questions that arise in a case." Valley v. Rapides Par. Sch. Bd., 118 F.3d 1047, 1054 (5th Cir. 1997); see also Century Marine Inc. v. United States, 153 F.3d 225, 231 (5th Cir. 1998) (quoting Burma Navigation Corp. v. Reliant Seahorse M/V, 99 F.3d 652, 656 (5th Cir. 1996)) (noting that Rule 52(a) "exacts neither punctilious detail nor slavish tracing of the claims issue by issue and witness by witness"). Instead, a court satisfies Rule 52 if it "afford[s] the reviewing court a clear understanding of the factual basis for [its] decision." Holman v. Life Ins. Co. of N. Am., 533 F. Supp. 3d 502, 506 (S.D. Tex. 2021) (quoting Interfirst Bank of Abilene, N.A. v. Lull Mfg., 778 F.2d 228, 234 (5th Cir. 1985)). And if the court fails to make a specific finding on a particular issue, the reviewing court "may assume that the court impliedly made a finding consistent with its general holding so long as the implied finding is supported by the evidence." Century Marine Inc., 153 F.3d at 231.

2. To the extent that any factual finding reflects or is better understood as a legal conclusion, it is also deemed a conclusion of law. Likewise, to the extent that any legal conclusion reflects or is better understood as a factual finding, it is also deemed a factual finding.

A. Procedural History

3. In February 2022, the Petteway plaintiffs challenged the enacted plan as discriminatory and in violation of § 2 of the Voting Rights Act, the Fourteenth Amendment, and the Fifteenth Amendment. Dkt. 1.

4. About one month later, the United States filed suit, alleging that the enacted plan violates § 2 of the Voting Rights Act. See United States v. Galveston County, No. 3:22-cv-93 (S.D. Tex. Mar. 24, 2022), Dkt. 1.

5. Three weeks later, the NAACP plaintiffs also filed suit challenging the enacted plan under the Fourteenth and Fifteenth Amendments and § 2 of the Voting Rights

Act. Dickinson Bay Area Branch NAACP v. Galveston County, No. 3:22-cv-117 (S.D. Tex. Apr. 14, 2022), Dkt. 1.

6. In May 2022, the NAACP plaintiffs moved to consolidate these three cases. Dkt. 37. The court granted the motion and ordered the cases consolidated in June 2022. Dkt. 45.

7. Also in June 2022, the defendants filed separate motions to dismiss the three consolidated complaints, arguing that the court lacked jurisdiction and the plaintiffs had failed to state a claim for relief. Dkts. 46, 47, 48. The court partially granted the defendants' motion related to the Petteway plaintiffs' complaint, dismissing Michael Montez based on lack of standing. Dkt. 125 at 12-13. The court otherwise denied the motions. Id.; Dkts. 123, 124.

8. In May 2023, the defendants moved for summary judgment on two grounds: (1) the three preconditions required to establish the § 2 claims under Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986); and (2) the racial predomination in map-drawing needed for the Petteway and NAACP plaintiffs' racial-gerrymandering claims. Dkt. 176. The court denied the motion two months later. Dkt. 200.

9. The court held a ten-day bench trial beginning on August 7, 2023. It heard live testimony from several of the individual plaintiffs — Constable Derrick Rose, the Honorable Penny Pope, Edna Courville, Joe A. Compian (in his individual capacity and on behalf of LULAC Council 151), and Lucretia Henderson-Lofton (on behalf of Dickinson Bay Area NAACP). Dkts. 221 at 55-104; 222 at 8-61, 211-79; 226 at 62-111, 188-215. The plaintiffs also presented live testimony from other county residents who are current and former elected officials: Lucille McGaskey, Robert Quintero, Sharon Lewis, Joe Jaworski, Pastor William Randall, Patrick Doyle, and Commissioner Stephen Holmes. Dkts. 221 at 105-79; 226 at 8-61, 112-87, 216-30; 228 at 11-64.

10. The court also heard expert testimony offered by the plaintiffs. William S. Cooper, Dr. Tye Rush, and Anthony E. Fairfax testified on the first Gingles precondition, illustrative map configurations, and redistricting principles. Dkts. 223 at 9-193; 224 at 9-162. Drs. Matthew A. Barreto, Jessica Trounstine, and Kassra A.R. Oskooii ("the quantitative experts") testified on the second and third Gingles preconditions. Dkts. 223 at 194-329; 224 at 163-349. Finally, Drs. Traci Burch, Rene R. Rocha, and Max Krochmal testified on the totality of the circumstances and indicia of discriminatory intent. Dkts. 222 at 62-210; 225 at 10-283.

11. By agreement, the parties presented live testimony from Judge Mark Henry, Commissioner Joe Giusti, Commissioner Darrell Apffel, redistricting consultant Dale Oldham, and mapping consultant Thomas Bryan. Dkts. 228 at 166-356; 231 at 8-308; 232 at 8-163, 289-379. After this testimony, the plaintiffs closed their case in chief.

12. The defendants presented live testimony from Commissioner Robin Armstrong and County Clerk Dwight Sullivan. Dkt. 230 at 191-219, 231-68. The court also heard expert testimony from Dr. Mark Owens, who addressed the first Gingles precondition, and Dr. John Alford, who addressed racially polarized voting. Id. at 10-190; Dkt. 232 at 164-288. Following this testimony, the defendants rested.

13. After resting, the defendants moved for judgment on partial findings under Fed. R. Civ. P. 52(c) on all claims, which the court denied. Dkt. 230 at 272-85.

14. About three weeks after trial, the parties filed post-trial closing-argument briefing, Dkts. 240, 241, 242, 244, along with proposed findings of fact and conclusions of law, Dkts. 239, 245. They then

filed response briefing one week after that. Dkts. 246, 247, 248, 249. The court reviewed these materials when preparing these findings and conclusions.

B. Parties

1. Plaintiffs

15. The Honorable Terry Petteway is a Black resident of the city of Galveston. PX-607 ¶¶ 2-3. Under the election plan adopted as part of the 2011 redistricting cycle ("the benchmark plan"), Petteway's home sat in Precinct 3, and Commissioner Holmes represented him. Id. ¶ 6. But under the enacted plan, Petteway now resides in commissioners Precinct 2, which Commissioner Giusti represents. Id. ¶ 7. Petteway is a registered voter who has voted in commissioners-court elections and intends to vote in these elections in the future. Id. ¶¶ 4-5.

16. The Honorable Penny Pope is a Black resident of the city of Galveston. Dkt. 222 at 8-9. She is a former justice of the peace who represented Justice of the Peace/Constable ("JP/constable") Precinct 3 for twenty-six years. Id. at 9, 12. Under the benchmark plan, Judge Pope's home sat in commissioners Precinct 3. Id. at 21. But under the enacted plan, she now resides in Precinct 2. Id. She is a registered voter who regularly votes in Galveston County elections. Id. at 10.

17. Constable Derrick Rose is a Black resident of Texas City. Dkt. 221 at 55. Since 2005, he has served as the elected constable for JP/constable Precinct 3. Id. at 57-59. Under the benchmark plan, Constable Rose's home sat in Precinct 3. Id. at 56, 60. But under the enacted plan, he now resides in commissioners Precinct 1, which Commissioner Apffel represents. Id. at 60. Constable Rose is a registered voter who regularly votes in Galveston County elections. Id. at 55-56.

18. Joe Compian is a Latino resident of La Marque. Dkt. 226 at 62. Under the benchmark plan, Compian's home sat in Precinct 3. Id. But under the enacted plan, he now resides in Precinct 4 and is represented by Dr. Armstrong. Id. Compian is a member of LULAC Council 151 and is a registered voter who votes religiously. Id. at 63-66. He intends to vote in the future. Id. at 63-64.

19. Edna Courville is a Black resident of Texas City. Dkt. 222 at 211. She used to reside in Precinct 3 under the benchmark plan but now lives in Precinct 4 under the enacted plan. Id. at 218. Courville is registered to vote, votes regularly, and intends to vote in the future. Id. at 219. She is a member of the Mainland Branch of the NAACP. Id. at 215-16.

20. Leon Phillips is a Black resident of the city of Galveston. DX-310 at 7. He is a member of the Galveston Branch of the NAACP. Id. at 21. Under the benchmark plan, he lived in Precinct 3. Id. at 32. But under the enacted plan, he now resides in Precinct 2. DX-34 at Row 31383. He is a registered voter and intends to vote in future elections.

21. The Dickinson Bay Area Branch of the NAACP is a nonprofit, nonpartisan membership organization and is an affiliate branch of the Texas State Conference of the NAACP. Dkt. 204-6 ¶ 27. The Dickinson Branch serves Dickinson and League City and has at least fifty members, including members who lived in benchmark Precinct 3. Dkt. 226 at 200-01. One such member includes Lucille McGaskey, who now lives in Precinct 4. Dkt. 221 at 106, 123; DX-34 at Row 95740; DX-115 at Row 133363. The Dickinson Branch's mission, consistent with the national NAACP and all other local NAACP units, includes educating people on discrimination and voting and helping people register to vote. Dkt. 226 at 199-200.

22. The Galveston Branch of the NAACP is a nonprofit, nonpartisan membership

organization and is an affiliate branch of the Texas State Conference of the NAACP. Dkt. 204-6 ¶ 27. The Galveston Branch has about sixty members, all living or working in Galveston County. PX-605 ¶¶ 4-5. After the 2021 redistricting cycle, at least one Galveston Branch member who was a resident of benchmark Precinct 3 has been redistricted into a different commissioners precinct. Id. ¶ 6. As a unit of the national NAACP, the Galveston Branch's mission includes educating people on discrimination and voting, as well as helping people register to vote. See Dkt. 226 at 199-200.

23. The Galveston LULAC Council 151 is a civic organization in Galveston County and an independent unit of the League of United Latin American Citizens. Dkts. 204-6 ¶ 28; 226 at 65-66. LULAC's goals include supporting and advocating for civil rights and improving Latinos' participation in the political system. Dkt. 226 at 66. After the 2021 redistricting cycle, at least one LULAC member who was a resident of benchmark Precinct 3 has been redistricted into a different commissioners precinct.

24. The Mainland Branch of the NAACP is a nonprofit, nonpartisan membership organization and is an affiliate branch of the Texas State Conference of the NAACP. Dkt. 204-6 ¶ 27. The Mainland Branch has over fifty members who live or work in Texas City, La Marque, and Hitchcock. PX-606 ¶¶ 7-8. As a unit of the national NAACP, the Mainland Branch's mission includes educating people on discrimination and voting, as well as helping people register to vote. See Dkt. 226 at 199-200; see also Dkt. 222 at 216. At least one Mainland Branch member in Galveston County was a resident of the benchmark Precinct 3 and has been redistricted into a different commissioners precinct. PX-606 ¶ 9.

25. The United States is represented by the Department of Justice. Congress has vested the Attorney General with the authority to enforce § 2 of the Voting Rights Act on the United States' behalf. See 52 U.S.C. § 10308(d).

2. Defendants

26. Galveston County is a political and geographical subdivision in southeast Texas on the Gulf of Mexico. Dkt. 204-6 ¶ 1.

27. Galveston County Commissioners Court is the county's governing body. Tex. Const. art. V, §§ 15, 16, 18(a)-(b). The commissioners court consists of a county judge elected at-large as the presiding officer and four county commissioners elected from single-member precincts, all serving four-year, staggered terms. Dkt. 204-6 ¶ 2. The commissioners court that adopted the enacted plan consisted of Judge Henry, Commissioner Apffel (Precinct 1), Commissioner Giusti (Precinct 2), Commissioner Holmes (Precinct 3), and Commissioner Ken Clark (Precinct 4). Id. ¶ 25.

28. Judge Mark Henry has been the county judge since 2010. Id. ¶ 4. The plaintiffs sued Judge Henry in his official capacity as Galveston County's chief officer. See, e.g., Dkt. 42 ¶ 33.

29. Defendant Dwight D. Sullivan is the incumbent county clerk for Galveston County. Dkt. 230 at 233. Sullivan's office oversees all county elections, which involves supervising poll workers, polling sites, ballot creation, and ballot tabulation. Id. at 233-34.

C. Expert Witnesses' Credibility⁷

7 The court's findings on the expert witnesses' qualifications, reliability, and credibility are limited to this case. They are not informed in any way by any testimony that may have been presented to the court in other cases and do not apply to future matters before this court.

1. Gingles Precondition One

30. The plaintiffs' experts William Cooper, Anthony Fairfax, and Dr. Tye Rush

testified about the first Gingles precondition for § 2 vote-dilution claims — whether Black and Latino residents in Galveston County are sufficiently large and geographically compact to constitute a majority in a single-member district. Dkts. 223 at 9-190; 224 at 9-162.

31. Each expert testified to forming their opinions by using publicly available data from the Census Bureau and applying standard and reliable redistricting methods in conducting their analyses and forming their opinions. Dkts. 223 at 15-17, 17; 224 at 22-26, 85-86, 92-93, 97; PXs-337 at 10-13; 342.

32. Cooper has nearly four decades of experience drawing voting plans for about 750 United States jurisdictions. Dkt. 223 at 9-10. He has testified on redistricting and demographic analysis in federal court fifty-five times. Id. at 10; see also PX-341. Cooper submitted, and the court received into evidence, principal and rebuttal reports addressing the first Gingles precondition. PXs-386, 438.

33. The court recognized Cooper as an expert on redistricting, demographic analysis, and the first Gingles precondition. Dkt. 223 at 11-12. After receiving Cooper's testimony and reviewing his reports, the court finds his analyses, opinions, and testimony credible.

34. Fairfax has over thirty years of map-drawing, demography, and redistricting experience. Dkt. 224 at 74-75. He testified that he has developed or helped develop hundreds of redistricting plans. Id. at 75-77. Fairfax has testified as an expert in redistricting matters nine times. Id. at 80. He submitted, and the court admitted into evidence, both his initial and rebuttal reports addressing the first Gingles precondition. PXs-337, 454.

35. The court recognized Fairfax as an expert on map-drawing, demography, redistricting, and census data as it applies to the first Gingles precondition. Dkt. 224 at 81. After receiving Fairfax's testimony and reviewing his reports, the court finds his analyses, opinions, and testimony credible.

36. Dr. Rush is the president's postdoctoral fellow at the University of California, San Diego, and has expertise in mapping and political geography. Id. at 12-14; PX-486. He holds a Bachelor of Arts in political science from the University of California, Riverside, and a Master of Arts and Ph.D. in political science from the University of California, Los Angeles. Dkt. 224 at 11-12. Dr. Rush previously was a senior policy fellow at the UCLA Voting Rights Project, where he led research projects, conducted mapping analyses, and taught mapping. Id. at 13. He also was a redistricting and voting fellow at Common Cause, where he taught mapping to lawyers and assisted with census research. Id. Dr. Rush has also taught mapping and political geography at the university level, and clients have hired him to perform political mapping. Id. at 13-14. He submitted, and the court admitted into evidence, both an initial and a rebuttal report as well as a supplemental declaration addressing the first Gingles precondition. PXs-485-487.

37. The court recognized Dr. Rush as an expert on political geography, mapping, and electoral behavior. Dkt. 224 at 14-15. After receiving Dr. Rush's testimony and reviewing his reports, the court finds his analyses, opinions, and testimony credible.

38. The defendants offered testimony from Dr. Mark Owens on the first Gingles precondition. Dkt. 232 at 164-288. Dr. Owens holds a Bachelor of Arts in political science from the University of Florida, a Master of Arts in government from Johns Hopkins University, and a Ph.D. in political

science from the University of Georgia. Id. at 165-66; DX-290 at 27. While working on his Ph.D. dissertation, he was also a visiting doctoral student at the University of Oxford. Dkt. 232 at 166.

39. After completing his Ph.D., Dr. Owens was a visiting assistant professor of American politics at Bates College. Id. at 168-69. After that, he joined the University of Texas at Tyler's faculty, where he taught classes and conducted research on American political institutions and elections. Id. at 169. At UT-Tyler, he developed expertise and published works on Texas politics and elections. Id. at 173. He recently accepted a position as a professor of political science at the Citadel, where he will continue his teaching and research. Id. at 175.

40. At trial, the court allowed Dr. Owens to proffer expert opinions on the first Gingles precondition and the population dispersion of minority groups in Galveston County. Id. at 198. But the court does not find his testimony on these topics credible. He neither describes himself as an expert nor even claims to focus any of his work on either redistricting or the first Gingles precondition. Id. at 189-90. Instead, he concentrates his work on the federal legislative process. Id. at 190. None of Dr. Owens's coursework included training on the technical aspects of drawing a voting plan, id. at 193, and he has not published any peer-reviewed work on any of the issues he opined on in his report, id. at 193-94. Dr. Owens has never taught a course on the technical aspects of drawing a voting plan. Id. at 196. Other than part of a single class on southern politics, he does not teach any specialized courses to graduate students on the Gingles standard. Id. at 197.

41. Before forming his opinions, Dr. Owens had reviewed fewer than ten voting plans for compactness — and only two of those professionally. Id. at 195. His only redistricting experience involved assisting a nonprofit by drawing statewide maps in Oklahoma that were neither considered by any court of law nor used in any election. Id. at 195-96. And his report revealed a fundamental misunderstanding of traditional redistricting principles. See, e.g., DX-290 at 16 (providing a table with an average population deviation instead of the maximum deviation); id. at 252-54.

42. Given the widespread shortcomings in Dr. Owens's testimony in this case, the court assigns little to no weight to Dr. Owens's opinions on traditional redistricting principles, the geographic dispersion of minority populations, and the first Gingles precondition.

2. Gingles Preconditions Two and Three

43. The plaintiffs' experts Drs. Matthew Barreto, Jessica Trounstine, and Kassra Oskooii testified about the second and third Gingles preconditions — whether (1) Black and Latino residents are politically cohesive and (2) Anglo voters sufficiently bloc vote to enable them to usually defeat their preferred candidate, respectively. See generally Dkts. 223 at 194-265; 224 at 163-349.

44. Drs. Barreto, Trounstine, and Oskooii base their opinions on quantitative analyses of demographic data and election results. PXs-356 ¶¶ 20-37; 384 ¶¶ 16-29; 476 ¶¶ 25-40; 501 ¶¶ 1-10.

45. Dr. Barreto is a political-science and Chicano-studies professor at the University of California, Los Angeles. Dkt. 223 at 196-97; PX-384 ¶ 2. He is a co-founder and faculty director of the Latino Policy and Politics Initiative at UCLA and the UCLA Voting Rights Project. Dkt. 223 at 196-197; PX-384 ¶ 2. Dr. Barreto has testified dozens of times in federal court on racially polarized voting, demographic change, map-making, and public polling. Dkt. 223 at 206-07; PX-384 ¶ 2. The court recognized

Dr. Barreto as an expert in mapping, racially polarized voting, demographic change, racial and ethnic politics, and Gingles preconditions two and three. Dkt. 223 at 209.

46. After receiving Dr. Barreto's testimony and reviewing his reports, the court finds his analyses, opinions, and testimony credible.

47. Dr. Trounstine is a political-science professor at Vanderbilt University. PX-604 at 1. Before Vanderbilt, she was the Foundation Board of Trustees Presidential Chair of Political Science at the University of California, Merced. Id. She also served as an assistant professor of politics and public affairs at Princeton University. Id. Dr. Trounstine holds a Ph.D. and a Master of Arts in political science from the University of California, San Diego, and a Bachelor of Arts in political science from the University of California, Berkeley. Id.

48. Dr. Trounstine has published several peer-reviewed publications, id. at 1-3, including two award-winning books published by university presses. Id. at 1. One of those books, Political Monopolies, "is about how local political coalitions get built," "how those coalitions end up electing officials to office," "how those officials keep themselves in power for multiple decades," and "the consequences those political monopolies . . . have for . . . representation." Dkt. 224 at 167.

49. As part of her academic work, Dr. Trounstine has analyzed "the building of political coalitions," "racial group representation," and "the political voting patterns of various racial, ethnic, and class groups, as well as other groups along gender lines." Id. at 168-69. As part of her academic work, Dr. Trounstine has also "looked at the various ways that coalitions are built over time." Id.

50. Dr. Trounstine is currently an Andrew Carnegie Fellow. PX-604 at 3. She was awarded the fellowship to "write a book on local political polarization in the United States." Dkt. 224 at 168.

51. The court recognized Dr. Trounstine as an expert in political science, particularly statistical analysis of group voting patterns and the ability of groups to elect their candidates of choice. Id. at 170-71. Having received Dr. Trounstine's testimony and reviewed her reports and declaration, the court credits her analyses, opinions, and testimony and grants them substantial weight.

52. Dr. Oskooii is a tenured associate professor of political science at the University of Delaware and is a faculty member at the university's Data Science Institute. Id. at 273-76. He has published peer-reviewed works on racially polarized voting analyses and served as an expert in Voting Rights Act cases nationwide. Id. The court recognized Dr. Oskooii as an expert on racially polarized voting analysis. Id. at 278.

53. The defendants' expert on the second and third Gingles preconditions, Dr. John Alford, testified that he greatly respects Dr. Oskooii as a methodologist. Dkt. 230 at 151. Dr. Alford agreed with the numerical accuracy of Dr. Barreto's and Dr. Oskooii's ecological-inference results and adopted their results for his analysis. Id. at 99-100. Having received Dr. Oskooii's testimony and reviewed his reports and declaration, the court credits his analyses, opinions, and testimony and grants them substantial weight.

54. Dr. Alford has been a professor in Rice University's political-science department for thirty-five years. Id. at 12. He teaches courses on elections and voting behavior, and has served as a testifying expert for about 30 years. Id. No court has ever declined to recognize him as an expert on the second and third Gingles preconditions. Id. 55. At trial, the parties stipulated to Dr. Alford's expertise on the second and third Gingles preconditions. Id. at 12-13. After receiving Dr. Alford's testimony and reviewing his reports, the court finds his analyses, opinions, and testimony credible.

3. Senate Factors and Arlington Heights Factors

56. Dr. Traci Burch is an associate professor of political science at Northwestern University and a research professor at the American Bar Foundation. Dkt. 222 at 64; PX-414 at 3, 52. She is an expert in political behavior, political participation, barriers to participating in politics, barriers to voting, race and ethnic politics, and criminal justice. Dkt. 222 at 64-65. She has been an expert in federal and state court on barriers to voting and felony disenfranchisement, as well as both the Senate and Arlington Heights factors. Id. at 66-67.

57. Dr. Burch testified about the racially discriminatory intent of the 2021 redistricting plan. Id. at 72-110. Dr. Burch also testified on the Senate factors. Id.

58. Reflecting a reliable application of Senate Factors 5 through 9 to the facts of this case, Dr. Burch based her opinions on a review of sources and methods that are standard for political scientists and social scientists, including the relevant political-science literature. Id. at 68-69. Dr. Burch also collected relevant data and analyzed publicly available information from websites, meeting records, newspaper articles, census data, and surveys. Id. at 69.

59. Reflecting a reliable application of the Arlington Heights factors, Dr. Burch's opinions are based on her analysis of relevant demographic data and county-specific primary sources, including statements by Judge Henry, the commissioners, and the public. See generally PX-414. She also bases her opinions on peer-reviewed political-science and sociological studies, which is standard practice for political scientists and social scientists. Dr. Burch "cast a fairly wide net," surveying public records and statements made by decision-makers and Galveston County residents. Dkt. 222 at 69-71.

60. The court qualified Dr. Burch as an expert in this case. Id. at 67. After receiving Dr. Burch's testimony and reviewing her reports, the court finds her analyses, opinions, and testimony credible.

61. The United States' expert, Dr. Rene Rocha, testified about the Senate factors. Dkt. 225 at 192-279. Dr. Rocha is the Herman J. and Eileen S. Schmidt Chair and Professor of Political Science and Latino Studies at the University of Iowa. Id. at 193; PX-336 at 1. He conducts research and teaches courses about race and ethnic politics, immigration policy, and voting rights. Dkt. 225 at 193. He has previously served as an expert in a § 2 case in federal court. Id. at 196-98.

62. Dr. Rocha's opinions are based on relevant demographic data, county-specific primary sources, and peer-reviewed political-science and sociological studies. His research included reviewing census and American Community Survey data, federal-and state-government documents, court decisions, peer-reviewed academic work, websites, and newspaper articles. He gathered evidence of incidents and events in Galveston County that fell within Senate Factors 1, 2, 3, 5, and 6. PX-335; id. at 192-283.

63. The court qualified Dr. Rocha as an expert in this case. Dkt. 225 at 199. After receiving Dr. Rocha's testimony and reviewing his reports, the court finds his analyses, opinions, and testimony credible.

64. Finally, the United States' last expert — Dr. Max Krochmal — is a history professor and the Czech Republic Endowed Professor and Director of Justice Studies at the University of New Orleans. Id. at 11; PX-317. Dr. Krochmal researches

and teaches courses on the history of the American South, African American history, Latino/Latina history, and multiracial coalitions, focusing on Texas history during the twentieth century. Dkt. 225 at 11.

65. Dr. Krochmal testified about the racially discriminatory intent of the 2021 redistricting plan. Id. at 36, 74-95. His testimony cataloged discriminatory events undertaken by local and state entities against Black and Latino residents in Galveston County that affected the right to vote. Id. at 52-67. Based on his research, Dr. Krochmal concluded that there was sufficient evidence to find a history of official, voting-related discrimination in Galveston County. Id. at 34-35.

66. Reflecting a reliable application of the Arlington Heights factors and the events leading up to the enacted plan, Dr. Krochmal bases his opinion on the historical method. Id. at 36-52; PX-412 at 8-9. To reach his conclusions here, Dr. Krochmal analyzed more than 300 newspaper articles, years of commissioners-court agendas and minutes, video streams, primary sources in archives, oral-history interviews, and multiple days of fieldwork. Dkt. 225 at 36-52. Dr. Krochmal examined Galveston County's past redistricting cycles, the specific sequence of events leading up to the 2021 redistricting plan, and the history of discrimination against the county's Black and Latino population. See generally PX-412.

67. The court recognized Dr. Krochmal as an expert. Dkt. 225 at 32-33. The court also noted Dr. Krochmal's advocacy within his community and how, at times, he provided legal and political opinions favorable to the plaintiffs. After hearing and observing Dr. Krochmal's testimony, reviewing his report, and considering the defendants' arguments and evidence proffered to show his bias, the court still finds his testimony credible — although less than that of Drs. Burch and Rocha.

D. Galveston County Demographics and Voting Patterns

68. According to the 2020 Census, Galveston County has a total population of 350,682-54.6% Anglo, 25.3% Latino, and 13.3% Black. Dkt. 204-6 ¶ 6. The combined Black and Latino population represents about 38.6% of the countywide population. PX-386 ¶ 26.⁸

8 Cooper explained there are several possible definitions of "Black" — such as non-Latino and any-part Black, non-Latino and DOJ Black, and single-race Black — in a demographic analysis. Dkt. 223 at 20-23. For Galveston County's population, the differences are "fairly insignificant" for overall population, and "de minimis" for citizen-voting-age population ("CVAP") calculations, and thus do not alter the court's analysis. Id.

69. Commissioners Precinct 3, which historically covered portions of Dickinson, La Marque, Texas City, and the city of Galveston, was the only majority-minority precinct in Galveston County from 1991 to 2021. Id. ¶ 38; PX-412 at 33-34.

70. The historic core of Precinct 3 was the product of advocacy by Black and Latino activists to create a majority-minority precinct in which they could elect a candidate of choice in the 1991 redistricting cycle. PX-412 at 32-37. This advocacy occurred shortly after the 1988 election of the first Black member of the commissioners court, Wayne Johnson, in a close campaign marked by racially polarized voting. Id. at 23-25; Dkt. 225 at 62-65.

71. Over time, Precinct 3 became an important political homebase for Black and Latino residents. "It was responsive. It was reflective of their priorities. And people took great pride and ownership in it." Dkt. 225 at 71; see also Dkts. 226 at 190-91; 228 at 46 (discussing how "different groups of people" take pride "not only in the precinct itself and the cohesiveness in

the precinct itself but even the pride they have in their elected official as the county commissioner").

72. By 2020, benchmark Precinct 3's CVAP was 58.31% Black and Latino. PX-386 ¶ 46. On the other hand, the enacted plan has no commissioners precinct with a Black and Latino CVAP larger than 35%. Id. ¶ 58. Ironically, Precinct 3 now has the smallest such population at 28%. Id.

1. Sufficiently Large and Geographically Compact

73. The Black community in Galveston County primarily resides in the center of the county — Texas City, La Marque, Dickinson, Hitchcock, and the city of Galveston. Meanwhile, the Latino community is evenly dispersed throughout the county.

74. Both parties agree that there is not a sufficiently large, compact, and separate Latino or Black population to constitute a majority-Latino or majority-Black precinct in Galveston County. They also agree, however, that when treated as a coalition, the Black and Latino populations are sufficiently large and compact to support a majority-minority commissioners precinct. The court finds both propositions to be true.

75. During the 2021 redistricting process, the commissioners court considered a proposed map — Map 1 — that featured a reasonably compact commissioners precinct with a majority Black and Latino population by CVAP. That precinct — Precinct 3 — was 30.86% Black and 24.28% Latino by CVAP. PX-487 ¶ 65.

76. The commissioners court's legal consultant for the redistricting process, Dale Oldham, testified that Map 1 was legally defensible. Dkt. 231 at 122-23. The plaintiffs' experts also testified that Map 1 met the first Gingles precondition. See Dkt. 224 at 73; cf. PX-386 ¶¶ 70-80; Dkt. 223 at 51-55.

77. The illustrative plans that the plaintiffs presented at trial demonstrate that Galveston County's Black and Latino population is sufficiently large to constitute a majority by CVAP in a single commissioners precinct. Cooper drafted four illustrative plans that all include a majority Black and Latino commissioners precinct by CVAP. See generally PXs-386 ¶¶ 81-96; 439 at 2. Cooper Plans 1, 2, 3, and 3A each include an illustrative commissioners precinct with 57.65%, 57.72%, 55.27%, and 54.52% Black and Latino CVAP, respectively, as calculated using the 5-Year 2017-2021 ACS Special Tabulation. PX-439.

78. Fairfax's illustrative plan likewise includes a majority Black and Latino commissioners precinct. According to the 2020 Census redistricting dataset and the 2016-2020 ACS 5-Year ACS Data, Fairfax's illustrative plan includes a commissioners precinct with 55.15% Black and Latino CVAP. Dkt. 224 at 109-11; PXs-337 ¶ 47; 551.

79. For his initial report, Dr. Rush created three illustrative plans containing Precinct 3 configurations wherein the Black and Latino communities together formed a majority by CVAP. PX-487 ¶¶ 34-54. The CVAP in Dr. Rush's plans was calculated using the 2020 Census redistricting dataset and the 2016-2020 5-year ACS Data. Id. The Precinct 3 configurations in his three initial plans exhibit a Black and Latino CVAP of 56.6%, 61.2%, and 57.5%, respectively. Id. Dr. Rush also presented a fourth plan with a Black and Latino CVAP of 57.92%. PX-486 at 19.

80. Dr. Rush subsequently created four additional plans containing coastal precincts, each unifying the county's entire county coastline into one commissioners precinct without fragmenting the mainland minority population. Id.; see also PXs-415-418. Texas Legislative Council-generated reports confirm that three of the coastal

precinct plans contain a Precinct 3 in which the combined Black and Latino CVAP is over 50%. PX-485 ¶ 8.

a. Traditional Redistricting Criteria

81. All the plaintiffs' experts on the first Gingles precondition credibly testified to applying traditional redistricting criteria in developing their illustrative maps.

82. Dr. Owens's criticisms of the plaintiffs' illustrative plans do not overcome their experts' testimony, leaving intact the plaintiffs' argument that each plan comports with traditional redistricting criteria.

83. NAACP Plaintiffs' Illustrative Plans (Cooper). Cooper developed Cooper Map 1 by shifting just two voting precincts from the benchmark plan, a "least-change" approach he deemed acceptable for Galveston County based on the characteristics of its population changes over the past decade. Dkt. 223 at 56-57; PX-386 ¶¶ 81-86. This least-change plan demonstrates the minimum number of changes necessary to eliminate malapportionment and brings the commissioners precincts within an "almost perfect deviation." Dkt. 223 at 58; PX-386 ¶ 31.

84. In Cooper Map 1, all commissioners precincts are contiguous, and Precinct 3 is reasonably compact given the county's complex geography. Dkt. 223 at 58. This plan keeps eleven municipalities whole and has fifteen populated municipal splits. PX-349 at 5-6. It respects municipal and political-subdivision boundaries better than the enacted plan, which keeps nine municipalities whole but has sixteen populated municipal splits as well as four populated voting-district splits. PX-346 at 5-6. Racial considerations did not predominate in drawing Cooper Map 1. Dkt. 223 at 58. Cooper Map 1 adheres to traditional redistricting principles and is reasonably configured. Id. at 62; PX-386 ¶ 86.

85. Cooper also developed Cooper Map 2 using a least-change strategy for equalizing populations while also including an entirely coastal Precinct 2. Dkt. 223 at 62-63; PX-386 ¶¶ 87-90. At 0.57%, the total population deviation is "even closer" to zero than that of the enacted plan. Dkt. 223 at 64; PX-350 at 3. All commissioners precincts are contiguous, and Cooper Map 2 keeps ten municipalities whole with fifteen populated splits. Dkt. 223 at 66; PX-350 at 4-5. Cooper Map 2 has nine populated voting-district splits, which Cooper explained were split to prioritize creating a coastal commissioners precinct that would be contiguous by driving. Dkt. 223 at 66; PX-350 at 6. Cooper testified that Precinct 3 in Cooper Map 2 is reasonably compact. Dkt. 223 at 67. Racial considerations did not predominate in drawing Cooper Map 2. Id. at 63. Cooper Map 2 adheres to traditional redistricting principles and is reasonably configured.

86. Cooper developed Cooper Maps 3 and 3A by attempting to unify all offshore islands in a single precinct. Dkt. 223 at 68; PXs-386 ¶¶ 92-96; 438 ¶¶ 35-38. The population deviations for both plans are below 5%. PXs-351 at 3; 443 at 3. Cooper included Cooper Map 3A as a slightly modified version of Cooper Map 3 to allow Precinct 1 to be contiguous by driving without requiring entry across the Moses Lake Floodgate. Dkt. 223 at 70-72; PX-438 ¶ 35.

87. Cooper Map 3 keeps nine municipalities whole and includes sixteen populated splits, while Cooper Map 3A keeps nine municipalities whole and includes fifteen populated splits. PXs-351 at 5; 443 at 5. Cooper Maps 3 and 3A have three voting-district splits, one less than the enacted plan. PXs-351 at 6; 443 at 6. Both are reasonably compact. Dkt. 223 at 68-69, 75. Race did not predominate in the development of either map. Id. at 70, 75-76. Cooper Maps 3 and 3A adhere to traditional redistricting principles and are reasonably configured. 88. In sum, all the Cooper illustrative plans adhere to traditional redistricting criteria without pairing any incumbents or predominating race. Cooper Maps 2, 3, and 3A prove that achieving these metrics and maintaining a majority-Black and Latino precinct is possible, even with a unified coastal precinct.

89. United States' Illustrative Plan (Fairfax). Fairfax developed an illustrative plan using the least-change approach to equalize the population among the commissioners precincts. Using this approach, he shifted only one voting district from Precinct 2 to Precinct 3 to bring the precinct's population deviations within the accepted guideline range of 5% and the total deviation under 10%. Dkt. 224 at 97-102; PX-337 ¶¶ 38-41. All commissioners precincts are contiguous. Dkt. 224 at 106-07; PX-340 at 9. Fairfax testified that his illustrative plan is reasonably compact and more compact than the benchmark plan. Dkt. 224 at 115-17. The illustrative plan is also similarly compact as compared to the enacted plan. PXs-454 ¶ 4; 557.

90. Fairfax testified that his illustrative plan adhered to traditional redistricting criteria, including equal population, contiguity, and compactness. Dkt. 224 at 104, 107. Fairfax's illustrative plan also maintained the same municipality and voting-district splits as the benchmark plan. Racial considerations did not predominate in drawing Fairfax's illustrative plan. Id. at 103. Therefore, Fairfax's illustrative plan adheres to traditional redistricting principles and is reasonably configured.

91. Petteway Plaintiffs' Illustrative Plans (Rush). Each precinct in Dr. Rush's eight illustrative plans is contiguous. PXs-415-418, 485, 486, 487.

92. Dr. Rush's illustrative plans have an overall plan deviation under 10%, and seven of his eight plans are within the 5% guideline.

93. Precinct 3 in Dr. Rush's illustrative plans is reasonably compact, as are the other three commissioners-court precincts in each of those plans. Dkt. 224 at 22-23; PXs-486, 487. Each precinct in Dr. Rush's illustrative plans is also comparatively compact when measured against the districts in the enacted and the benchmark plans. PXs-486, 487.

94. Dr. Rush's illustrative plans respect political and precinct boundaries. For example, Dr. Rush's Demonstrative Plans 1, 2, and 2b do not split any voting districts. PX-487 ¶¶ 38, 44. In addition to respecting political boundaries, Dr. Rush's illustrative plans keep together communities of interest. See Dkt. 224 at 21-22; PX-486 at 5-6.

95. Finally, racial considerations did not predominate in drawing Dr. Rush's illustrative plans, as he did not consider race or ethnicity while creating his maps. Dkt. 224 at 22, 27. Given this evidence, Dr. Rush's illustrative plans adhere to traditional redistricting principles and are reasonably configured.

96. Defendants' Assessment of Plaintiffs' Illustrative Plans (Owens). Although the court gives Dr. Owens's testimony almost no weight, Dr. Owens generally agreed that the plaintiffs' plans were "about as reasonably compact as the enacted plan." Dkt. 232 at 229, 276. He also agreed that it is common to use a least-change approach when rebalancing populations following a census. Id. at 259-60. Dr. Owens charged that the plaintiffs' illustrative plans used race as a predominating factor, but failed to explain what made him believe that, other than his work "comparing the outcomes" of the maps. Id. at 258. Nor did he dispute that the plaintiffs' experts used non-racial traditional redistricting criteria. Id. at 256-58.

97. Dr. Owens's opinions do not change the court's findings that the plaintiffs' illustrative plans exemplify several ways to

draw a reasonably compact commissioners precinct featuring a majority Black and Latino CVAP and comporting with traditional redistricting principles.

98. Conclusions Regarding Traditional Redistricting Principles. Overall, Cooper, Fairfax, and Dr. Rush's illustrative plans confirm that the combined Black and Latino population is sufficiently large and geographically compact. In their maps, Blacks and Latinos would constitute a majority by CVAP in a single commissioners precinct that is reasonably configured and adheres to traditional redistricting principles.

99. The illustrative plans would preserve Precinct 3 as a majority-minority precinct. Indeed, there are a "multitude of potential plans adhering to traditional redistricting principles that would result in maps that maintain a majority [Black and Latino] CVAP [c]ommissioners [p]recinct." PX-386 ¶ 97; see also Dkts. 223 at 52; 224 at 117-18; PX-337 ¶ 63.

100. The plaintiffs' illustrative plans further show that the commissioners court could retain a majority-minority precinct even if it prioritizes creating a unified coastal precinct.

b. Geographically Compact

101. Dr. Owens opined that Galveston County's minority population is neither geographically nor culturally compact. The court assigns no weight to these opinions.

102. Dr. Owens did not cite any academic literature to support his analysis. Dkt. 232 at 232. Concerning the Latino population, Dr. Owens based his conclusion on the distances between discrete concentrations of Latino residents, ranging from 305 people to 7,637. Id. at 237-40. He provided no authority or reference for the significance of those distances or even a definition for what would be considered "distant and disparate" in Galveston County. See id. at 237-40; DX-290.

103. When testifying that Blacks and Latinos are not "culturally compact," Dr. Owens had no basis for disputing that Black and Latino residents throughout Galveston County fare worse than their Anglo counterparts across most socio-economic measures. Dkt. 232 at 247-49. He analyzed only three of the twenty potential socio-economic factors available Id. at 245-46. When presented with factors that did not favor his opinion, he admitted to inconsistently choosing which factors to examine. Id. at 246. Additionally, he did not analyze how these groups compared to their Anglo counterparts. Id. at 247.

104. The court also does not credit the defendants' assertions that Blacks and Latinos are not culturally similar because minority residents in League City have higher standards of living than those in the rest of the county. Although League City is more affluent than other parts of the county, "disparities between Black and Latino residents as compared to their Anglo counterparts persist even in League City, which indicates that they share the common socio-economic challenges of Black and Latino residents in Galveston." PX-438 at 5; see also Dkt. 223 at 184. Blacks and Latinos are more affluent in League City than in the rest of the county, but that does not disprove the overwhelming evidence that they share similar socio-economic struggles countywide and in Precinct 3.

105. Although not nearly as probative as the quantitative socio-economic data, lay-witness testimony adduced at trial supports this conclusion. For example, Lucretia Henderson-Lofton is a former president of the Dickinson Bay Area NAACP and a Black resident of League City. Dkt. 226 at 188, 198-99. Born on Galveston Island and raised in Texas City, Henderson-Lofton moved to League City in 2016. Id. at 189. She testified to the

racial discrimination her family and others have experienced in League City and the significant contacts that she maintains in Texas City. Id. at 189-90, 193-98, 204-07.

106. Given this evidence, the court finds that Galveston County's Black and Latino population is sufficiently large and geographically compact to constitute a majority in a single commissioners precinct that is both reasonably configured and comports with traditional redistricting principles.

2. Politically Cohesive

107. Using ecological-inference methods, the plaintiffs' quantitative experts demonstrated that Black and Latino voters in Galveston County are cohesive in that a large majority of these voters have consistently favored the same candidates across a series of elections. PXs-356, 384, 476; Dkts. 223 at 226; 224 at 184, 188-89, 199, 279-82. These results were consistent across several data sources and in hundreds of statistical models. PXs-356, 384, 465, 476; Dkts. 223 at 221-32; 224 at 175.

108. Racially polarized voting "describes an electorate in which [Anglo] voters favor and vote for certain candidates. . . and minority voters vote for other candidates." Rodriguez v. Harris County, 964 F. Supp. 2d 686, 756 (S.D. Tex. 2013) (quoting LULAC, Council No. 4434 v. Clements, 986 F.2d 728, 744 (5th Cir. 1993)). The existence of racially polarized voting does not necessarily mean that voters are racist or harbor racial animus. See id. at 757 (noting that the correct question is "not whether [Anglo] voters demonstrate an unbending or unalterable hostility to whoever may be the minority group's representative of choice, but whether, as a practical matter, such bloc voting is legally significant").

109. Ecological inference is a reliable and standard method of measuring racially polarized voting. PXs-384 ¶¶ 18-21; 476 ¶ 25; Dkt. 223 at 216-17, 219. Two forms of ecological inference, King's Ecological Inference ("King's EI") and RxC EI, use aggregate data to identify voting patterns through statistical analysis of candidate choice and racial demographics within a precinct. PXs-384 ¶¶ 18-21; 476 ¶ 25; Dkt. 223 at 216-17, 219.

110. RxC EI is appropriate for analyzing elections with more than two candidates or more than two racial or ethnic groups. PXs-384 ¶ 18; 476 ¶ 25; Dkt. 224 at 188. The plaintiffs' quantitative experts produced estimates using both King's EI and RxC EI.

111. In addition to CVAP and Spanish Surname Turnout data used in King's EI and RxC EI, Dr. Barreto and his co-author, Dr. Michael Rios, conducted a Bayesian Improved Surname Geocoding ("BISG") analysis of Galveston County elections to more precisely assess voting patterns by race and ethnicity. PX-465.

112. BISG analysis creates a probability that a given voter who participated in an election is of a particular racial or ethnic group based on his or her surname and the racial composition of the census block. Id. ¶¶ 30-34. Because Latinos vote at lower rates than Anglo and Black voters, BISG is particularly useful for narrowing in on the vote choices of Latino voters who participate in elections. Dkt. 223 at 242-44. Studies have validated the reliability of using BISG for analyzing racially polarized voting. Id. at 236.

113. Dr. Oskooii replicated and reproduced Dr. Barreto's BISG results and achieved highly consistent results. PX-505. Dr. Oskooii testified that BISG is a reliable method and is widely employed across various industries and applications. Dkt. 224 at 305-06. Dr. Alford agreed that BISG is reliable for estimating Latino voting patterns in Texas. Dkt. 230 at 160. The court finds that BISG is a reliable methodology

for assessing racially polarized voting patterns.

114. The experts agree that there is no universal way to determine cohesion. Instead, they determine cohesion by analyzing elections that show a particular pattern within the relevant jurisdiction. Dkts. 224 at 301; 230 at 100-01; PX-476 ¶¶ 27-28, 30; DX-305 at 2.

115. The undisputed RxC EI analyses from Drs. Oskooii and Barreto show that, on average, over 85% of Black and Latino voters have voted for the same candidate countywide and within the illustrative Precinct 3 plans contained in those reports. PXs-356 at 14, 23; 465 ¶ 36; see also Dkt. 224 at 184, 188-89, 199.

116. The undisputed RxC EI analyses from Drs. Barreto and Oskooii show most Latinos and Blacks have separately voted for the same candidate in almost all general elections. PXs-372 at 2, 4; 384 ¶ 46. Drs. Barreto and Rios's BISG-based analysis shows even stronger cohesion among Latino voters, with over 75% favoring the same candidates in most of the twenty-nine elections they assessed. PX-465 ¶ 39. Dr. Oskooii's BISG analysis confirmed these results. PX-505 ¶¶ 1-8. These analyses show that Latino voters consistently supported the Black-preferred candidate and that Black voters consistently supported the Latino-preferred candidate. PX-465 ¶ 39. Dr. Alford did not dispute Drs. Barreto and Rios's BISG results. Dkt. 230 at 161.

117. Dr. Barreto agreed that the wider confidence interval for Latino voter cohesion stems from standard error. Dkt. 223 at 283. He observed that a lower standard error generates a tighter confidence interval, while a higher standard error generates a broader confidence interval for the exact point estimate. Id. at 288. Despite wider confidence intervals for Latino voters, Dr. Barreto had "equal faith" in the point estimates he reported in the BISG analysis. Id. at 289-92. Dr. Oskooii's estimates also had broad confidence intervals for Latino voters. Recognizing Dr. Alford's concerns about the reliability of the wide confidence intervals, the court still finds it to be probative evidence of Latino voter cohesion and attributable to the smaller sample sizes of Latino voters.

118. Dr. Alford considered the voting patterns of Anglos, Blacks, and Latinos separately, and testified that it would be hard to find "a more classic pattern of what polarization looks like in an election." Dkt. 230 at 17-18.

119. Here, the most probative election results demonstrate cohesion between Black and Latino voters in Galveston County. All experts agreed that recent elections are more probative and can more reliably confirm cohesion and polarization than more distant elections. PX-356 ¶ 22; Dkts. 223 at 247-48; 224 at 139, 176. Due to the limited number of contested endogenous⁹ elections, it was necessary to analyze exogenous elections. PXs-384 at 17-40 (analyzing twenty-eight exogenous elections across five election cycles); 476 at 33; Dkt. 224 at 281.

9 Endogenous elections are "contests within the jurisdiction and for the particular office that is at issue." Rodriguez, 964 F. Supp. 2d at 759. Exogenous elections are "elections in a district for positions that are not exclusively representative of that district." Id.

120. Exogenous elections encompassing Galveston County, such as those for Attorney General and Governor, are more probative than elections covering only portions of the county, such as municipal elections. Dkts. 224 at 181-82, 280; 230 at 144-45. The exogenous elections that cover the entire county show consistently high levels of cohesion. PXs-384 at 17-40; 356 at 14; 476 at 46-47. 121. All experts agreed that general elections are more probative than primary elections in this case; this includes determining inter-group cohesion, i.e., cohesion between Black and Latino voters in Galveston County. Dkt. 223 at 246-47; 224 at 181-87, 262-63; 230 at 145-46, 149; PXs-465 ¶ 27; 476 ¶ 34.

122. Primary elections have limited probative value in determining inter-group cohesion for several reasons. First, in the context of "racial[-] and ethnic[-]coalition building[,] . . . coalitions get built in the general election," not the primary election. Dkt. 224 at 181-87. Second, because primary elections generally have low turnouts, the resulting estimates are less robust and do not present a good picture of most voters for any demographic group. Id. at 292-93; PX-356 ¶ 24. Third, candidate preferences are not as likely to be as strong for any candidate given that candidates' ideological positions in the same party are likely closer than those in different parties in a general election. Dkt. 224 at 292-93; PX-356 ¶ 24.

123. Primary elections for the commissioners court are rarely contested, with lower levels of voter participation among all racial and ethnic groups — but especially Black and Latino voters. PXs-356 ¶ 24; 465 ¶ 27; Dkt. 224 at 292-93.

124. Considering their limited probative value, the primary elections that Dr. Oskooii analyzed show a steady presence of inter-group cohesion between Black and Latino voters. In nine out of the ten primary elections he studied, Black and Latino voters voted cohesively. PX-356 ¶¶ 64-65.

125. Between Drs. Oskooii and Alford, the analyzed results show that Blacks and Latinos usually support the same top-choice candidate in primary contests. Id. ¶ 64; DX-305 at 18-19; Dkt. 224 at 302-03.

126. The 2012 primary election for Precinct 3 provides very probative evidence because it is the most recent endogenous contest for Precinct 3. Dkt. 230 at 140-42. That election featured a highly cohesive Black and Latino electorate. Id. at 140.

127. Dr. Alford observed that several of the examined Democratic primary elections did not feature racially polarized voting because Anglo, Black, and Latino voters supported the same candidates. Id. at 30-31, 37-39, 47-48, 70. But on cross-examination, Dr. Alford admitted that this observation is irrelevant when determining racial cohesion between Black and Latino voters. Id. at 125-28, 130-31. And Dr. Alford acknowledged that when considering the third Gingles precondition, in general elections (in which voters can elect — rather than just nominate — a candidate of their choice), Anglo voting behavior is especially relevant. Id. at 131-32. The court thus does not credit Dr. Alford's observation about Anglo voting behavior in Democratic primaries for purposes of the second or third Gingles preconditions.

128. Two data limitations restrict the probative value of the local nonpartisan elections analyzed in this case. First, the local nonpartisan elections cover smaller geographic areas than any individual county-commissioners precincts. Dkt. 224 at 182. They often encompass very few election precincts, see, e.g., DX-287, thereby limiting the demographic information available to produce estimates, Dkts. 224 at 283-84; 230 at 67-68. Second, in many local nonpartisan elections there were multiple candidates and low voter turnout — two features that contribute to instability in ecological-inference estimates. Dkt. 224 at 294, 325-26. The local nonpartisan races also have less probative value than partisan general elections because commissioners-court races are partisan contests. PX-465 ¶ 25. The court therefore assigns little weight to the analyses of local nonpartisan elections. 129. Even so, local nonpartisan elections show cohesion between Black and Latino voters in Galveston County. PX-476 ¶ 56. Further, successful minority candidates in nonpartisan elections are primarily elected from majority-minority districts, see, e.g., Dkt. 230 at 265, which is consistent with racially polarized voting patterns, id. at 165.

130. Based on their analyses, the plaintiffs' quantitative experts concluded that Black and Latino voters in Galveston County are cohesive. PXs-356 ¶ 6; 384 ¶¶ 23-24; 476 ¶¶ 6, 34; Dkt. 224 at 184.

131. Although less probative than the quantitative evidence, lay testimony also shows political cohesion between Blacks and Latinos in Galveston County. Community leaders testified that Black and Latino voters in Galveston County vote cohesively. See, e.g., Dkts. 221 at 133-34; 226 at 15, 130. Several witnesses testified that the Black and Latino communities in Galveston County share interests and policy preferences, including those addressing education, housing, healthcare, and employment. Dkts. 221 at 65, 109-10, 133-34; 222 at 32-36; 226 at 67-68, 128-30, 156-57, 197-98, 204, 207-08. Galveston County's local LULAC and NAACP branches often collaborate, sharing services and resources. Dkt. 226 at 86, 117, 120-21, 204. Several witnesses are members of both organizations. Dkts. 222 at 217; 226 at 14, 65, 201-02, 204.

132. Thus, the court finds that Blacks and Latinos vote cohesively in Galveston County.

3. Cannot Elect Candidate of Choice

133. The evidence adduced at trial shows that Anglo voters in Galveston County engage in bloc voting such that a large majority of the county's Anglo voters favor their own candidates in both countywide and precinct-only elections. The high level of Anglo bloc voting usually prevents Black and Latino voters in Galveston County from electing their candidates of choice.

134. An electoral-performance/reconstituted-election analysis is a technique used to examine how candidates would have fared under different maps or precinct boundaries. PXs-356 ¶ 68; 476 ¶¶ 38-40; see also PX-384 ¶ 46. The plaintiffs' quantitative experts used this method to analyze elections encompassing the entirety of Galveston County for the enacted plan and the plaintiffs' illustrative maps. PXs-356 ¶¶ 67-75; 384 ¶¶ 44-48; 476 ¶¶ 38-40, 58.

135. Under the enacted plan, Anglo bloc voting defeated the candidate of choice of Black and Latino voters in every election in every commissioners precinct. PXs-356 ¶¶ 71-72; 384 ¶¶ 44-46; 476 ¶ 58; Dkt. 224 at 205, 288-89.

136. All three electoral-performance/reconstituted-election analyses from Drs. Barreto and Rios, Dr. Oskooii, and Dr. Trounstine establish that the candidate of choice of Black and Latino voters won in Precinct 3 in every election under the plaintiffs' illustrative maps. PXs-356 ¶¶ 72-75; 384 ¶¶ 44-46; 476 ¶ 58.

137. Dr. Alford also analyzed whether Anglo bloc voting is sufficient to defeat minority-preferred candidates in the enacted plan. Dkt. 230 at 123. He did not dispute the plaintiffs' quantitative experts' electoral-performance/reconstituted-election analyses. Id.; see generally DX-305.

138. A direct relationship exists between a precinct's demographic composition and a specific candidate's likelihood of success in any given election. As the minority percentage moves up or down, the performance of minority-preferred candidates moves in direct proportion. Dkt. 224 at 289-90. This relationship supports a finding of racially polarized voting and complements the ecological-inference estimates

the quantitative experts performed in this case. Id.; PX-356 ¶¶ 74-75.

139. In most of the recent general elections, over 85% of Anglos across Galveston County voted for candidates running against the minority-preferred candidates. PXs-356 ¶ 40; 384 ¶¶ 22-24. Similarly high levels of bloc voting are present at the individual-precinct level in the enacted commissioners precincts. PX-356 at 19.

140. The plaintiffs' quantitative experts concluded that these patterns at the county level also exist at the commissioner-precinct level. Dr. Oskooii found that there is Anglo bloc voting in the enacted plan's precincts and that there is cohesive minority voting in Cooper's illustrative maps. Id. ¶¶ 56-62. Dr. Barreto found that Anglo and non-Anglo voters are sharply polarized in their voting patterns in each of the four enacted precincts. PX-465 ¶¶ 44-46. Similarly, Dr. Trounstine found the same polarized voting pattern in Precinct 3 in Fairfax's illustrative map. PX-501 ¶ 2; see also Dkt. 224 at 198-99. The court credits the quantitative experts and agrees with their conclusions.

141. All experts agree that Anglo bloc voting usually defeats the Black and Latino candidate of choice in Galveston County elections in every precinct analyzed in the enacted plan.

142. The court finds that voting in Galveston County is racially polarized such that Anglo voters usually vote as a bloc to defeat the candidate of choice of Black and Latino voters.

4. On Account of Race

143. The defendants contend that partisanship alone explains the racially divergent voting patterns in Galveston County. To the extent that partisanship explains the voting patterns in the county, it still does not change the fact that the data unerringly points to racially polarized voting.

144. The parties agree that Anglo bloc voting exists in Galveston County such that Blacks and Latinos could not elect candidates of their choice. They also agree that Anglos in Galveston County, who comprise a supermajority, are mostly Republican and that Blacks and Latinos are mostly Democrats. The plaintiffs argue that race and politics are "inextricably intertwined," Dkt. 247 at 6 n.3, while the defendants and Dr. Alford contend that partisan affiliation is the "main driver of voter behavior," Dkt. 244 at 54-58.

145. Dr. Alford did not analyze whether any factors other than race or party identification explain the divergent racial voting patterns in partisan elections in Galveston County. Dkt. 230 at 107-08. He admits that assessing "partisan polarization" in addition to racial polarization is not standard practice among redistricting experts. Id. at 88. Characterizing the typical redistricting expert as being, unlike himself, an "advocate[ ] for a particular position," Dr. Alford defended his focus on the difference between racial and partisan polarization. Id. at 88-89.

146. Dr. Alford testified that political-issue attitudes are distinct from party identification and that party identification, unlike issue attitudes, is primarily the result of socialization. Id. at 77-79. Tellingly, he based his conclusions regarding the role of partisanship versus race primarily on one election: the 2018 Senate race between Senator Ted Cruz and Beto O'Rourke. Id. at 53, 166.

147. Although partisanship undoubtedly motivates voting behaviors in Galveston County, the defendants failed to show that a race-neutral explanation explains the racially divergent voting patterns. Dr. Oskooii testified that Black and Latino voters were cohesive behind their preferred candidate in about 93% of racially contested elections, while Anglo voters were cohesive

behind the Anglo-preferred candidate. Dkt. 224 at 298-300; PX-452 ¶ 7.

148. The racial composition of political parties in Galveston County, measured through participation in each party's primaries, further suggests that the county's electorate is racially polarized. All experts agree that relatively few Anglo voters in Galveston County participate in Democratic Party primaries. PX-465 ¶¶ 13-17; Dkt. 224 at 293; 300; see also PX-476 at A-12; Dkt. 230 at 109-10. Conversely, relatively few Black and Latino voters in Galveston County participate in the county's Republican primaries. PX-465 ¶¶ 17-19; Dkt. 224 at 183, 300; PX-476 ¶ 21. No Black or Latino Republican has ever won a primary election to be the Republican Party's nominee for county judge or a county commissioner. PX-465 ¶ 17. Commissioner Armstrong, who is Black, was appointed and did not participate in a Republican primary election. Dkt. 230 at 197. He ran uncontested in the general election. Dkt. 224 at 298-99.

149. In general elections in Galveston County, Anglos overwhelmingly vote for Republican candidates. PX-452 ¶ 8. Meanwhile, Blacks overwhelmingly vote for Democrat candidates, and Latinos very often support the same candidates. See PX-476 at 33.

150. Although it is not very probative, national scholarship on political trends helps to further explain the link between race and partisanship in Galveston County. See PX-384 ¶¶ 30-43. "The fact that Black and Latino voters tend to support candidates from one party is a reflection of their cohesion, not an alternative explanation for it." PX-476 ¶ 35. The history of discrimination resulting in ongoing socio-economic disparities and barriers to voting along racial lines also contributes to a finding that race, not partisanship alone, drives the voting patterns seen in Galveston County.

151. Moreover, Galveston County voters provided testimony of racially polarized voting based on their lengthy residences in the county, their elections to public office, or both. See Dkts. 221 at 128-29, 133-34; 222 at 17; 226 at 130. Although anecdotal and isolated, this evidence further supports that race provides a plausible explanation for voting patterns in Galveston County.

152. The court therefore finds that a partisan explanation for voting patterns in Galveston County does not overcome the weighty evidence of racially polarized voting on account of race.

E. Discriminatory Impact of the Enacted Plan

153. The enacted plan converted the benchmark Precinct 3 from the precinct with the highest percentage of Black and Latino residents to the one with the lowest. Dkt. 223 at 42-43. According to 2016-2020 ACS Special Tabulation data from the census, benchmark Precinct 3 is about 58% Black and Latino by CVAP. PX-386 ¶ 46. But after the 2021 redistricting, Precinct 3 now includes the lowest Black and Latino CVAP proportion of any precinct — about 28% — and the Black and Latino population is evenly distributed throughout the remaining precincts — with each one containing a range of 32% to 35% Black and Latino CVAP. Id. ¶ 58.

154. Accordingly, Black and Latino residents fail to comprise a majority in any new commissioners precinct — despite comprising about 38% of the overall population and 32% of the CVAP. Id. ¶ 31.

155. The plaintiffs' quantitative experts established that Black and Latino voters will usually fail to elect a candidate of their choice in any commissioners precinct within the enacted plan. PXs-356 ¶¶ 70-74; 384 ¶¶ 44-46; Dkt. 224 at 288-89. 156. Anglo voters comprise 64.1% of the county's voting age population but now control 100% of the electoral outcomes for Galveston County commissioners court. See PX-487 ¶ 14.

157. The county's redistricting counsel, Oldham, likewise acknowledged that the benchmark plan included a performing precinct for minority voters while the enacted plan no longer does. Dkt. 231 at 178. The enacted plan creates an evident and foreseeable impact on racial minorities in Galveston County by eliminating the sole majority-minority precinct. See Dkt. 222 at 110.¹⁰

10 Several witnesses testified that it was obvious on the face of the map that the enacted plan would fracture minority communities. Dkts. 221 at 62-63; 222 at 248-49; 226 at 21-22, 69, 77.

158. The court finds that the enacted plan disproportionately affects Galveston County's minority voters by depriving them of the only commissioners precinct where minority voters could elect a candidate of their choice. Likewise, the court finds that the commissioners court was aware of that fact when it adopted the enacted plan.

F. Galveston County Voting and Redistricting

159. For § 2 vote-dilution claims, a plaintiff must show under the "totality of circumstances" that the "challenged political process is not 'equally open' to minority voters." Allen v. Milligan, 599 U.S. 1, 18, 143 S.Ct. 1487, 216 L.Ed.2d 60 (2023). District courts use the Senate Judiciary Committee's Report accompanying the 1982 amendment to the Voting Rights Act to inform this determination, which provides several non-exhaustive factors to consider. S. Rep. No. 97-417, at 40 (1982) [hereinafter S. Rep.]. For intentional-discrimination claims, the Fifth Circuit follows the framework in Village of Arlington Heights v. Metropolitan Housing Development Corp. to determine whether a legislative body passed a redistricting plan with discriminatory purpose. 429 U.S. 252, 97 S.Ct. 555, 50 L.Ed.2d 450 (1977). The court will identify the factual findings that pertain to each framework as it presents those findings.

1. History of Discrimination in Voting Practices

160. The first Senate factor is the history of official voting-related discrimination in the political division. Similarly, the first Arlington Heights factor is the historical background of the decision.

161. Galveston County was a center for buying and selling enslaved Black people during the Antebellum era. Dkt. 225 at 52-53. After the Civil War, race relations in the county reflected those seen across much of the South, including segregation and Jim Crow laws. Id. 53-54; PX-412 at 13-14. At the same time, "state-supported practices and laws in a variety of different areas of life" came together to segregate Latinos in Galveston County, a system termed Juan Crow. Dkt. 225 at 57-58; PX-412 at 10-12.

162. The discrimination against Black and Latino residents in Galveston County extended to voting. Dkt. 225 at 58. For instance, the Texas legislature passed a poll tax in 1903, which required payment each January. PX-412 at 13; id. at 54-55. This affected many Black and Latino voters because many were agricultural laborers, and few had cash on hand in January due to the timing of the agricultural cycles. See Dkt. 225 at 54-55. During much of the twentieth century, the Texas Democratic Party allowed only Anglos to vote in its primary, preventing Black and Latino voters from participating in the "elections and

caucuses that really mattered." PX-412 at 13-15.

163. Restrictions on voting for minorities remained present in Texas even after the Voting Rights Act in 1965 and its 1975 extension. Before 2013, § 5 of the Voting Rights Act "required States to obtain federal permission before enacting any law related to voting." Shelby County v. Holder, 570 U.S. 529, 534, 133 S.Ct. 2612, 186 L.Ed.2d 651 (2013). Section 4(b) provided the coverage formula that defined the "covered jurisdiction" that must follow this preclearance process. Id. at 538-39, 133 S.Ct. 2612. From 1975 to 2013, Galveston County was subject to § 5 preclearance. PX-412 at 15; Dkt. 225 at 58, 75. Preclearance subjected Galveston County to multiple objection letters from the Attorney General. PX-335 ¶¶ 19-23.

164. Nevertheless, several witnesses acknowledged that it is easier to vote now than it has ever been in Galveston County. Dkts. 221 at 157; 222 at 58; 230 at 245. The county adopted countywide voting centers, which allow voters to "vote anywhere on election day or early voting." Dkt. 230 at 238. It is also relatively easy to register to vote in the county. See, e.g., Dkts. 221 at 82; 222 at 258-59; 230 at 202, 245. Early voting lasts two weeks in Galveston County. Dkt. 221 at 155-57.

165. Sullivan testified that if a mail-in ballot required postage and the voter failed to affix it, the clerk's office would pay for the postage because it "want[s] every vote to count." Dkt. 230 at 245-46.

166. The county provides election materials in English and Spanish for all elections. Dkt. 226 at 82.

167. Judge Henry has not heard any complaint in the last ten years that the county prevented someone from being able to vote. Dkt. 228 at 248.

168. The county collaborates with LULAC and allows them to use county property for its Cinco de Mayo event. Dkt. 230 at 236. The event is a blend of a cultural festival and a get-out-the-vote effort. Id. at 235-36.

2. Attorney General's Objections

169. Since 1976, Galveston County and its political subdivisions have been the subject of six objection letters from the Attorney General. PX-335 ¶ 19.

170. In 1976, the Attorney General objected to Texas City's proposal to adopt a numbered-post system for city-council elections. PXs-1; 335 ¶ 20; Dkt. 225 at 202-03. After examining the history of governmental discrimination, racial-bloc voting, and the city's responsiveness to minority concerns, the Attorney General could not conclude that the city's proposal would not have a racially discriminatory effect. PX-1.

171. In 1992, the Attorney General objected to Galveston County's redistricting plans for JP/constable districts. PXs-2; 335 ¶ 26; Dkt. 225 at 203. The Attorney General's letter noted that Black and Latino residents were not a majority in any of the eight districts despite comprising 31.4% of the county's population. PX-2 at 1. County officials had "rebuffed" multiple requests from minorities to create a district where they would have an equal opportunity to elect candidates of their choice. Id. Ultimately, the county entered a consent decree concerning the 1992 JP/constable redistricting plan. PX-563.

172. Also in 1992, the Attorney General objected to the city of Galveston's proposal to modify how city-council members are elected — from six at-large districts to four single-member districts, with two members elected at large to numbered posts. PXs-3; 335 ¶ 21; Dkt. 225 at 203-04. After noting that several minority candidates unsuccessfully ran for city council because of racially polarized voting, the Attorney General did not preclear this change. PX-3. Ultimately, the city entered a consent

decree to elect all city-council members from single-member districts. PX-335 ¶ 21.

173. In 1998, the city of Galveston again sought to change the method of electing its city council from six single-member districts to four single-member districts and two at-large posts — the same scheme to which the Attorney General filed an objection in 1992. PXs-4; 335 ¶ 22; Dkt. 225 at 204. Noting that two of the six single-member districts had elected minority officials, the Attorney General concluded that reverting to two at-large districts would retrogress minority voting strength. PX-4 at 2-3.

174. In 2001, the city of Galveston asked the Attorney General to reconsider the objection to four single-member and two at-large districts. But in 2002, he declined to withdraw the objection. PX-335 ¶ 22.

175. In 2011, the city of Galveston again sought to change the method of electing its city council from six single-member districts to four single-member districts, with two members elected at large to numbered posts. PXs-47; 335 ¶ 23; Dkt. 225 at 205. The Attorney General objected to this change, noting that racial-bloc voting played a significant role in city elections and that minority candidates could elect candidates of choice from three of the six single-member districts. PXs-47 at 3-4; 335 ¶ 23.

176. In 2012, the Attorney General objected to Galveston County's 2011 redistricting plans for the commissioners and JP/constable precincts. JX-6; PX-335 ¶ 26; Dkt. 225 at 205-06. The JP/constable-precinct plan proposed reducing the number of justices of the peace from nine to five and the number of constables from eight to five. JX-6 at 1-2. The Attorney General's letter noted that minority voters could elect candidates of choice in Precincts 2, 3, and 5. Id. at 4. For Precincts 2 and 3, this ability resulted from a court order in Hoskins v. Hannah, No. G-92-12 (S.D. Tex. Aug. 19, 1992), that created these precincts. Id. Under the proposed plan, minority voters' ability to elect a candidate of choice would be reduced to one precinct. Id.

177. In 2012, the Attorney General also concluded that the county had not met its burden of showing that the commissioners court did not adopt its proposed plan with a discriminatory purpose. JX-6. The Attorney General found that the county had failed to adopt, as it had in previous redistricting cycles, a set of criteria by which it would be guided in the redistricting process. Id. at 2. The Attorney General's letter noted that: (1) this procedural deviation was a deliberate decision by the county to avoid being held to a procedural or substantive standard of conduct; (2) the process may have been characterized by the deliberate exclusion from meaningful involvement of Commissioner Holmes; and (3) the proposed changes would reduce the overall minority electorate in Precinct 3 and lead to the loss of the ability of minority voters to elect a candidate of their choice. Id.

178. These efforts to reduce majority-minority districts are significant because research has shown that Blacks and Latinos are more likely to vote if they live in majority-minority districts. PX-335 ¶ 25. Former Justice of the Peace Penny Pope also observed that the results of the 2011, 2013, and 2021 redistricting processes created additional voting barriers for minority residents who felt less motivated to vote and participate politically. Dkt. 222 at 27-28.

179. On June 25, 2013, the Supreme Court held that § 4(b) of the Voting Rights Act is unconstitutional and that the coverage formula "can no longer be used as a basis for subjecting jurisdictions to preclearance." Shelby County, 570 U.S. at 557, 133 S.Ct. 2612. Yet the Court's ruling "in

no way affect[ed] the permanent, nationwide ban on racial discrimination in voting found in § 2." Id.

180. Galveston County adopted an electoral map for JP/constable precincts two months later. Petteway v. Galveston County, No. 3:13-cv-308, at 1 (S.D. Tex. Aug. 31, 2022). Before the change, the county had eight precincts, two of which were majority-minority precincts. Id. After the 2013 plan, the county had four precincts, one of which was majority-minority. Id. Six plaintiffs sued the county, alleging § 2 vote dilution and intentional discrimination under the Fourteenth Amendment. Id. Following a three-day bench trial, the court concluded that the plaintiffs failed to show vote dilution, as the 2013 plan "increased the percentage of Galveston County residents living in a majority-minority district" and therefore did not diminish the voting power of minority voters. Id. The court also ruled in the defendants' favor on the intentional-discrimination claim. Id. at 2-3.

3. Public Input and Transparency in Prior Redistricting

181. During the 1981 redistricting cycle, County Judge Ray Holbrook appointed a committee of about thirty citizens to make recommendations for redrawing the county's voting precincts. PX-412 at 24-25. This recommendation would be a basis for "remapping" the commissioners-court precincts. Id. at 24. The commissioners court ratified the public committee's work and adopted new commissioners-court precincts. Id. at 25. The 1981 commissioners precinct map reflected "minimal" change that only "[s]lightly increas[ed] the combined voting strength of the county's Black and Latin[o] residents," "stopping short of creating a precinct [within] which the 'total minority' vote would constitute a majority." Id. at 26; see also Dkt. 225 at 76. The Attorney General did not object to the county's 1981 redistricting plan. Dkt. 225 at 76.

182. During the 1991 redistricting cycle, the commissioners court adopted a set of criteria and a timeline before it held three public hearings where numerous members of the public, including minorities, participated in the process. Id. at 76-77; PX-412 at 32-34, 37, 67. The redistricting plan reflected input from local NAACP and LULAC chapters and created a majority-Black-and-Latino Precinct 3. PX-412 at 34; Dkt. 225 at 78. The Attorney General did not file any objection to the county's 1991 redistricting plan. Dkt. 225 at 187.

183. During the 2001 redistricting cycle, the commissioners court adopted redistricting criteria, created a schedule of public hearings, and held four public meetings across Galveston County. Id. at 78; PX-412 at 38-39, 67. Among the redistricting criteria the commissioners court adopted was that "[c]ommunities of interest should be maintained in a single district" and that the plan "should not fragment a geographically compact minority community or pack minority voters in the presence of polarized voting so as to create liability under section 2 of the Voting Rights Act." PX-539. The Attorney General did not file any objection to the county's 2001 redistricting plan. Dkt. 204-6 ¶ 17.

184. During the 2011 redistricting cycle, after consideration of several proposals for redistricting counsel, the defendants hired James E. "Trey" Trainor, III, Dale Oldham, and Joe Nixon of the law firm Beirne, Maynard & Parsons, L.L.P., to serve as redistricting consultants. JX-45.

185. The commissioners court did not adopt redistricting criteria in the 2011 redistricting cycle. See PX-23. Judge Henry later became aware that the Attorney General had objected to the 2011 commissioners map in part because the commissioners court failed to adopt criteria. Dkt. 228 at 274. 186. During the 2011 redistricting cycle, the commissioners court adopted a redistricting timeline that accounted for the preclearance process and the candidate-filing period. PX-412 at 44. This timeline included (1) an initial hearing to present draft maps and explain the census results in Galveston County and (2) five public hearings on redistricting in the evenings throughout the county. See PXs-45 at 9; 531-535.

187. The commissioners court adopted a redistricting plan in 2011 and submitted it to the Attorney General on October 14, 2011. JX-45. The Attorney General filed an objection. JX-6. In his letter, the Attorney General highlighted: (1) the county's decision not to adopt a set of criteria "to avoid being held to a procedural or substantive standard of conduct"; (2) "the deliberate exclusion from meaningful involvement in key deliberations of the only member of the commissioners court elected from a minority ability-to-elect precinct"; and (3) the retrogressive impact that the relocation of the Bolivar Peninsula from Precinct 1 had on Precinct 3. Id.

188. The plan after the 2011 redistricting process contained a Precinct 3 in which Black and Latino residents constituted a majority of the CVAP. This Precinct 3 was a continuation of a district that the commissioners court created in the 1991 redistricting cycle that allowed Black and Latino voters to elect a candidate of choice and which was maintained in the 2001 redistricting cycle. Over its decades of existence, this Precinct 3 has become "a political home of historical significance" to Galveston County's Black and Latino communities. PX-412 at 64.

G. The 2021 Redistricting Process

189. The second Arlington Heights factor is the specific sequence of events leading up to the challenged decision. Relatedly, the third and fourth Arlington Heights factors address procedural and substantive departures from the normal procedural sequence. This information also informs the court's totality-of-circumstances analysis for § 2 claims.

1. Sequence of Events

a. April 2021 — Engaging Redistricting Counsel

190. Judge Henry had the county's general counsel contact Oldham in November 2020 to retain him as redistricting counsel. JX-11 at 2. Henry understood that the commissioners court would have to complete the process by sometime in November 2021 and specifically wanted Oldham because of his prior redistricting experience in the county. Dkt. 228 at 181, 280-81, 283-84. Now a solo practitioner, Oldham required a law firm to assist him in his work, and the commissioners court retained Holtzman Vogel Baran Torchinsky & Josefiak PLLC for that purpose in April 2021. Dkt. 231 at 28; PX-138.

191. The commissioners court voted 4-1, with only Commissioner Holmes voting against, to hire Oldham and Holtzman Vogel as redistricting counsel. See generally PXs-140; 585 at 8. The commissioners court did not publicly consider any other counsel. The commissioners court provided neither information on the April 2021 meeting agenda nor accompanying backup materials about whom the commissioners court was considering hiring. See Dkt. 228 at 288.

192. Shortly after engaging Oldham, Judge Henry and the county's general counsel, Paul Ready, contacted Oldham to ask whether the county "had to draw a majority[-]minority district." PX-144 at 1. Oldham responded that it "may or may not need to draw a majority[-]minority district depending on census data." Id. b. August 2021 — Census Data Released

193. The release of the 2020 Census data necessary for redistricting was delayed due to the COVID-19 pandemic. The Census Bureau ultimately released the data in the "legacy format" in August 2021 followed by a more user-friendly format the following month. Dkt. 231 at 36; DX-175.

194. In the spring or summer of 2021, Judge Henry became aware of the census data's expected release date and of its actual release in August. Dkt. 228 at 290; see also PXs-568-569, 586.

195. At the time of the census data's initial release, Oldham lacked the technical ability to parse through it, and the defendants had not yet hired a demographer for the project. So Oldham contacted Adam Kincaid of the National Republican Redistricting Trust to interpret the census data about Galveston County. Dkt. 231 at 36-37, 68; PX-173 at 1.

196. On September 14, Kincaid emailed Oldham a chart reflecting each commissioners precinct's racial demographic changes from 2010 to 2020. PX-173 at 1, 3. Oldham then removed the logo of the National Republican Redistricting Trust from the document and sent it to Ready to distribute to the commissioners. Dkt. 231 at 51, 52.

197. Oldham reviewed the racial data Kincaid had sent and concluded that Galveston County's Black population had remained concentrated in Precinct 3; the Latino population, on the other hand, had grown throughout the county. Id. at 131-34.

198. Oldham was "pretty familiar" with "the population and demographic location of that population in Galveston County." Id. at 131. He knew that the Black population was centered in Precinct 3 in the 2011 plan. Id. at 133-34. Oldham reviewed racial-shading maps of Galveston County after the census-data release to identify where Black populations were concentrated. Id. at 134-36. Oldham's understanding was generally consistent with Judge Henry and Commissioners Apffel and Giusti's understanding that Galveston County's Black and Latino population was centered around Precinct 3, which had consistently elected Commissioner Holmes. See, e.g., Dkts. 228 at 271-73; 232 at 148-49, 370.

199. Oldham held a series of meetings in mid-September 2021 with the commissioners and Judge Henry to determine their priorities for redistricting. The first meeting on September 8 included both Judge Henry and Commissioner Apffel, followed by individual sessions with Commissioners Giusti and Clark, and ending with a meeting with Commissioner Holmes on September 20. Dkt. 231 at 38, 42-43, 45, 48.

200. In his meeting with Oldham and Commissioner Apffel, Judge Henry told Oldham that he wanted a map like the one he conceived in 2011 — the configuration that ultimately became Map 2. Id. at 39-40, 150-52.

201. Commissioner Apffel requested that a specific area be brought into his precinct so that it included a new home he and his wife had bought. Dkt. 228 at 189.

202. Commissioner Giusti asked Oldham to "level out" the commissioners precincts in population, "clean up" the lines, and keep his parents' home in his precinct. Giusti did not dispute Oldham's recollection that Giusti additionally requested his precinct lines not change more than necessary. Dkt. 232 at 124-26.

203. When Oldham first met with Commissioner Holmes on September 20, he was frustrated that Commissioner Holmes could not list his mapping priorities. Dkt. 231 at 49-51, 53. In the follow-up call on September 23, Commissioner Holmes provided detailed instructions on which areas he wanted to add to Precinct 3 to resolve population imbalances and increase the

district's compactness. See Dkt. 228 at 68-72; JX-23 at 4.

c. October 14 — Hiring a Demographer

204. Despite Oldham completing the meetings with the commissioners and Judge Henry by September 23, no one contacted a demographer until October 14, when Holtzman Vogel asked Thomas Bryan to start drafting maps. Dkt. 231 at 225. Bryan owns Bryan GeoDemo, a company that provides redistricting map-drawing services. Id. at 216, 219-20.

205. On an October 15 call between Bryan and Phil Gordon of Holtzman Vogel, Gordon instructed Bryan to create two plans: (1) a least-change plan and (2) a plan that created four Republican precincts, later titled a "Four R plan." Id. at 227-28, 233, 289-90; see also PX-188.

206. The purported motivation of Judge Henry — creating a "coastal precinct" — never arose during the hour-long phone call between Gordon and Bryan, and Bryan's initial draft plans included no coastal precinct. Dkt. 231 at 290-91; PX-516.

207. After that initial call, Bryan immediately understood that Oldham, not Gordon, was the lead person from whom he should take instructions about configuring plans. Dkt. 231 at 290. Bryan and Oldham spoke by phone for the first time on October 17. Id. at 68-69; PX-196.

208. The Four R plan was not the foundation upon which Bryan built Map 2. Dkt. 231 at 291. Oldham never told Bryan that Judge Henry wanted to create four Republican precincts, and Oldham denied any such partisan objective. Id. at 153-54.

d. October 17 — Bryan Creates Map 2

209. On October 17, Bryan drafted two map proposals that he shared via email with Oldham: (1) a "minimum change" plan that became Map 1, and (2) an "optimal" plan with an entirely coastal precinct and three mainland precincts — all of which fractured Precinct 3 — that became Map 2. See PX-197; Dkt. 231 at 145-50. Map 2 was "the visualization of the instructions" Judge Henry had provided Oldham. Dkt. 231 at 181.

210. Bryan did not exercise discretion in drawing Maps 1 or 2; Oldham told him where to place the lines. Id. at 296. Oldham gave Bryan "very specific instructions about how he wanted Map 2 to look," and Bryan did not know for what reason Oldham "was asking [him] to put [any] particular territory in each of the commissioner[s] precincts in Map 2." Id. at 291-93. Bryan could not speak to what motivated the drawing of Map 2. Dkt. 232 at 29.

211. Bryan testified credibly that he did not display or consult racial data while working on the Galveston County maps. Id. at 33. But he also credibly testified that he was "given no instruction one way or the other on racial and ethnic information." Id. at 19. The court credits Bryan — an eminently believable witness — and not Oldham in this regard.

212. The first draft of Map 2 represents a dramatic change in the commissioners-precinct lines, both on the coast and the mainland, in a way that distributes the population of benchmark Precinct 3 among all four new precincts and shifts Commissioner Holmes's precinct north: Image materials not available for display.

PX-197.

213. Oldham admitted that it was possible to retain a majority-minority precinct while also creating a coastal precinct without dismantling benchmark Precinct 3. Dkt. 231 at 164, 167-68, 171.

214. The court finds that a desire to create a coastal precinct cannot and does not explain or justify why Map 2, the "optimal" plan, was drawn the way it was — and especially does not explain its obliteration of benchmark Precinct 3.¹¹

11 The plaintiffs' experts presented at least five illustrative plans that included both a coastal precinct and a majority-minority Precinct 3. See PXs-386 ¶¶ 87-90; 415-418.

e. Late October — Finalizing and Announcing the Maps

215. After Bryan drew these maps, Oldham traveled to Galveston County to meet with Judge Henry and the commissioners. Id. at 79-80. Oldham met with Judge Henry on October 18, and Judge Henry told Oldham he preferred Map 2 because it was "essentially his criteria." Id. at 82-84; PX-199.

216. Commissioners Apffel, Giusti, and Clark initially told Oldham that they preferred Map 1. Dkt. 231 at 190.

217. Oldham knew Commissioner Holmes would be dissatisfied with Map 2 because it dramatically reduced the minority population in Precinct 3, resulting in Precinct 3 having the lowest minority population percentage of all four precincts. Id. at 177-78. Commissioner Holmes opposed Map 2 and insisted that Oldham inform the commissioners court that § 2 of the Voting Rights Act required a majority-minority precinct. Id. at 101-02.

218. Bryan also produced a spreadsheet for Maps 1 and 2. Id. at 268-69. The spreadsheet included racial data about the plans. The first tab included CVAP and voting-age population data by racial group for each census block within Galveston County. PX-528. The second tab, titled "Pop Pivot," provided the Black and Latino voting-age population percentages for each commissioners precinct in the benchmark plan, Map 1, and Map 2, as well as the two categories combined to identify the total majority-minority percentage share for each precinct. Id.

219. The court finds that the commissioners never expressly considered this spreadsheet information.

220. After Oldham's initial meeting with Judge Henry, Oldham met with the commissioners. Dkt. 228 at 202-03. He first met with Henry and Commissioner Apffel to review the map proposals. Id. at 301; Dkt. 231 at 194. Later, Henry also contacted Commissioner Giusti to ensure he was comfortable with his new coastal precinct because it was a dramatic change from his current precinct. Dkt. 228 at 305-06. Henry chose not to call Holmes to do the

same — to inquire whether he was comfortable with his new precinct. Id. at 306.

221. Ready set up a series of Zoom meetings between Oldham and Bryan on the one hand, and Commissioners Giusti, Clark, and Apffel on the other, to endeavor to accommodate the commissioners' wishes. Dkt. 231 at 191-92. Oldham met with Commissioners Giusti and Clark simultaneously to request and implement anything they wanted to see in Map 2. Id. at 191-92. Judge Henry also recalled meeting with Commissioner Apffel that same day. Dkt. 228 at 301; see also id. at 194-97.

222. Commissioner Holmes received a call from Constable Rose on October 21. Rose relayed a conversation he had with Commissioner Apffel, in which Apffel said, "There are a couple of maps floating out there, and it is not looking good for Holmes." Dkt. 228 at 81.

223. On October 28, Judge Henry's chief of staff, Tyler Drummond, emailed Oldham asking about the status of the "final maps" and stating the county "originally wanted to have a special meeting tomorrow to discuss and possibly adopt" them. JX-27 at 1. The commissioners court was "awaiting the final maps with split precincts so we can finalize everything and get a special meeting together for next week." Id.

224. Bryan finalized the maps and provided them to the commissioners court on October 29. Dkt. 231 at 118. The county publicly posted the two proposals, Map 1 and Map 2, on the county's website on October 29. See JX-29.

225. The web page provided an opportunity for public comment, but there were no instructions on when those comments had to be submitted for consideration. Id.; Dkt. 228 at 330.

226. The only evidence of the commissioners court announcing the creation of the redistricting web page or the release of proposed maps is a post on Judge Henry's campaign Facebook page encouraging the public to support Map 2, PX-588, which Commissioner Giusti reposted. PX-247.

227. Based on Oldham's assessment, Judge Henry believed that Maps 1 and 2 were both legally compliant. Dkt. 228 at 332.

228. Commissioner Apffel testified that he never witnessed anyone instruct Oldham to use or consider racial data when designing potential maps. Dkt. 232 at 310. Apffel also did not recall seeing, reviewing, evaluating, or using racial demographics when considering maps. Id. at 311.

f. November 12 — The Enacted Plan

229. On November 1, the Texas Secretary of State issued an election advisory confirming that the state's commissioners courts had to revise their commissioners precincts by November 13. JX-34 at 2. Judge Henry had mistakenly believed he had until December to complete the redistricting process. Dkt. 228 at 281, 283. He provided no credible explanation for this mistake.

230. Commissioner Apffel called Commissioner Holmes a few days before the November 12 special meeting. Id. at 86; JX-23 at 8. Apffel had known Holmes since 1989 and considered him a friend. Dkt. 232 at 318-19. Commissioner Holmes and Commissioner Apffel differ on their recollections of how this conversation proceeded. Dkts. 228 at 82-86; 232 at 326-32. The court credits Commissioner Holmes's recollection more than Commissioner Apffel's and finds that Apffel informed Holmes that the commissioners court would be adopting Map 2. Holmes also told Apffel that Map 2 was discriminatory and ran afoul of § 2. Dkt. 228 at 82. Apffel explained that Oldham told him Map 2 was a "legal map" and that he had concerns about what Harris County was doing to the Republican members of its commissioners

court through redistricting. Id. at 82-83. Holmes responded that "it was not about . . . Republican or Democrat but about the protections guaranteed to the minority groups in the Voting Rights Act." Id. at 83. Apffel ultimately told Holmes that Judge Henry planned to make a motion to adopt Map 2, that he would second the motion, and that the commissioners court would vote for Map 2. Id. at 86.

231. The commissioners court held a special meeting on November 12 to consider and vote for a new commissioners-court map. PX-591. Thirty-six members of the public spoke at the meeting — a fraction of those who actually attended — criticizing the redistricting process and the two map proposals. JX-41 at 2-3. Commissioner Holmes then spoke, noting the procedural irregularities in the 2021 redistricting cycle and opposing both map proposals. He offered two alternative maps that preserved Precinct 3 as a majority-minority precinct. The other commissioners refused to consider or vote on Commissioner Holmes's proposals. On a 3-1 vote,¹² with only Commissioner Holmes voting against, the commissioners court adopted Map 2, or the "enacted plan." Id. at 3.

12 Commissioner Ken Clark did not attend the meeting due to health issues. He passed away in 2022 and was succeeded on the commissioners court by Commissioner Robin Armstrong.

2. Deviations from Prior Redistricting Cycles

232. Drs. Burch and Krochmal surveyed the 2021 redistricting process and found several procedural anomalies as compared to previous redistricting cycles. These procedural departures included the: (1) failure to adopt a timeline, (2) failure to adopt any publicly available redistricting criteria to guide the process, (3) lack of transparency in engaging redistricting counsel, (4) lack of public notice and availability for comment, (5) conduct surrounding the November 12 special meeting, (6) disregard for minority input, and (7) exclusion of Commissioner Holmes from the process. The court credits these findings as evidence of departures from the typical procedural sequence. The record evidence and lay testimony adduced at trial substantiate these procedural deviations.

233. The Attorney General's 2012 objection letter noted several procedural deficiencies in the 2011 redistricting process that raised concerns of intentional discrimination. JX-6 at 2. These deficiencies included the failure to adopt redistricting criteria and the deliberate exclusion of Commissioner Holmes. Id. The 2012 objection letter put Judge Henry on notice of procedural defects that could raise concerns about the exclusion of minority stakeholders and lack of transparency — lapses that could be viewed as evidence of intentional discrimination.

234. During the 2021 redistricting process directed by Judge Henry, the county repeated these same procedural lapses. See generally Dkt. 222 at 122-24; PX-414 at 11, 18-19. The only alternative plan offered by Oldham during the 2021 redistricting cycle, Map 1, closely resembled the 2011 map, to which the Attorney General had objected. Compare JX-45 at 22, with JX-29. Oldham testified that the elimination of preclearance facilitated the dismantling of the majority-minority precinct. Dkt. 231 at 59-60.

a. No Redistricting Timeline

235. In contrast to past redistricting cycles, there is no evidence of any redistricting timeline established by the commissioners court in 2021.

236. The defendants have failed to provide any credible explanation for the lack of a redistricting timeline. Judge Henry, who was principally responsible for the

redistricting process, testified that he was always aware Galveston County would need to redraw the commissioners precincts and that he was aware this would need to be completed by the candidate-filing date. Dkt. 228 at 280-81. But Henry had no explanation for the commissioners court's failure to set a timeline publicly, or even privately, for redistricting. See id. at 295-97.

b. No Redistricting Criteria

237. Unlike in prior years, the commissioners court failed to adopt any public redistricting criteria in 2021. Dr. Burch testified that the absence of public redistricting criteria is notable because "redistricting criteria tend to guide the process and give people a sense of what the priorities are, and the [c]ounty saw fit to adopt them in previous years." Dkt. 222 at 192-93.

238. Judge Henry knew that the commissioners court's failure to adopt criteria in 2011 provided a basis for the Attorney General's objection to the 2011 map. Dkt. 228 at 274. He admitted that there was no way for anyone to know the commissioners court's preferences and propose alternative maps that would meet them. Id. at 310-11. The defendants have failed to provide any explanation for deciding not to publicly adopt redistricting criteria in 2021.

239. Overall, the commissioners court's failure to adopt redistricting criteria in 2021 is a deviation because the commissioners court had adopted criteria in prior years and other counties across the state have regularly adopted redistricting criteria. Dkt. 222 at 137-38.

c. Lack of Transparency in Engaging Counsel

240. The commissioners court deviated from past practice in engaging redistricting counsel. In prior cycles, the court publicly entertained bids from several prospective counsel. PXs-412 at 43; 414 at 17.

241. Judge Henry sought Oldham due to his "success" for the county the prior cycle. Dkt. 228 at 283-84. No other law firms besides Oldham's personal choice, Holtzman Vogel, were publicly considered during the process. Id. at 286.

242. During the April 2021 commissioners court meeting in which they voted to hire Oldham and Holtzman Vogel, the commissioners court failed to provide any advance notice in the meeting agenda that they would be hiring counsel. PXs-570 at 239-41; 585 at 2. The defendants have not offered any explanation for this lack of transparency.

d. Lack of Public Notice and Comment

243. Failure to disclose the data underlying the commissioners court's decision-making. At no point in the process did the commissioners court publicly disclose any quantitative data about the benchmark plan or proposed commissioners-court maps.

244. In 2011, before adopting a map, the commissioners court held public meetings after the census data came out. Dkt. 231 at 34-35; PX-414 at 17.

245. Judge Henry acknowledged that the commissioners court could have publicly announced the census results for Galveston County, Dkt. 228 at 291, as it had at a meeting in August 2011 in which "the [c]ounty's redistricting consultants presented a preliminary demographic report showing the results of the 2010 Census as they related to the existing commissioner[s] precincts," JX-45 at 9. Indeed, Judge Henry admitted to receiving a similar report from Oldham in September, including detailed information about why the commissioners-court lines needed to change. Dkt. 228 at 293-94.

246. When the commissioners court posted proposed Maps 1 and 2 on October 29, it provided no quantitative data by

which the public could assess the maps. See generally JX-29.

247. The failure to make quantitative data available "speaks to the lack of transparency," as "the public wasn't able to see underlying population and demographic data to fully understand exactly how these maps were changing." Dkt. 222 at 138.

248. Rushed redistricting process that prevented meaningful public comment. The commissioners court rushed the redistricting process in 2021 and failed to include any meaningful participation from the public and Commissioner Holmes. Id. at 126-27; PX-414 at 17-21; Dkt. 225 at 92-94.

249. The COVID-19 pandemic caused delays in the release of the data required to redistrict. Dkt. 222 at 127-28; PX-414 at 13-14; DX-175 at 2-3. Still, this delay does not account for the failure to include meaningful public participation or the rushed process.

250. Demographers for the parties agree that it was feasible to create timely redistricting plans despite the COVID-19 delay. Cooper testified that he had the 2020 Census data available "within a couple of days" of its release and thus had a "nationwide dataset breaking out the block-level census data for the whole country" around August 15. Dkt. 223 at 16. According to Cooper, this timing would be typical for anyone using standard demographic software such as Maptitude. Id. at 17. Fairfax likewise testified that "anybody with GIS skills" could access and use the 2020 Census data in the format provided by the Census Bureau on August 12. Dkt. 224 at 78-79. Fairfax also testified that the Census Bureau provided a database that could have been used to review the 2020 Census data released in August 2021. Id. at 79.

251. Bryan testified that he would have been able to download the Census Bureau's redistricting data immediately once it was released. Dkt. 231 at 297-99. Had the defendants retained him earlier, he could have prepared draft maps by the end of August. Id. at 298-99. The defendants offer no credible explanation for why the commissioners court did not begin drawing proposed maps until mid-October 2021.

252. Bryan testified to a rushed process in which he was made to draw maps on a flight back from a vacation in Hawaii and given only a few days to complete the project. Id. at 236, 298; Dkt. 232 at 36. It is Bryan's practice to visit a jurisdiction and study it before drawing a map for it. Dkt. 232 at 35. Bryan testified that his inability to research or visit Galveston County and the tight timeline he was given was unusual for his work. Id. at 35-36.

253. The commissioners court was aware that redistricting likely needed to occur no later than November 2021 due to the timing of the candidate-filing period. Judge Henry testified that he fully expected the county to need to redraw commissioners precincts even before the redistricting data came out, and that this redistricting would have to be completed before the candidate-filing period opened. Dkt. 228 at 280-81.

254. Commissioner Giusti also testified that he "was pretty sure" the candidate-filing period would be from November to mid-December in 2021 because there "was a lot of resistance to [the state] moving the election dates." Dkt. 232 at 106.

255. The commissioners court held several public meetings between retaining redistricting counsel and the November 12 special meeting adopting the enacted map. PX-129. However, none discussed redistricting. Id.

256. Decision to Hold Special Meeting on November 12. In past redistricting cycles, the commissioners court held several hearings at various locations around the county to solicit public input on map proposals, including seven public hearings

during the 2011 redistricting cycle. In 2021, Oldham advised the commissioners court to hold as many public meetings as possible and allow for supplementation of feedback after the meetings. Dkt. 231 at 201.

257. Judge Henry agreed the county had received initial map "proposals" by October 19 but did not want anything publicly disclosed until they were a "final product." Dkt. 228 at 310.

258. Even factoring in the COVID-19 pandemic, Dr. Burch opined that the dearth of public meetings in 2021 was unusual. Dkt. 222 at 191, 196. In both 2011 and 2021, there were two weeks between when the commissioners court first disclosed its proposed maps and when it actually enacted a map. Id. at 191. In the two weeks before it adopted the benchmark plan in 2011, the commissioners court held five meetings across the county. Id. In the same amount of time in 2021, it only held one. Id.

259. The only opportunities for public input in 2021 were an online public comment portal and the November 12 special meeting. JX-42.

260. Commissioner Holmes testified that he expressed concerns to Drummond that the online comment portal was inadequate to provide residents the opportunity to be heard because of the number of residents who lacked access to either the internet or a computer. Dkt. 228 at 135; JX-23 at 5; see also Dkt. 222 at 194 (noting the racial disparities in internet access).

e. The November 12 Special Meeting

261. The November 12 special meeting was unusual not only for its singularity during the redistricting cycle but also for its lack of accessibility for many of Galveston County's Black and Latino residents.

262. In past redistricting cycles, residents could choose among multiple meeting locations so that they could attend the most geographically accessible site. PX-412 at 60; Dkt. 225 at 75-82. Compounded with last-minute notice for the only meeting held about the maps, this factor "denied the public the opportunity to provide meaningful feedback on the maps." PX-414 at 19; see also Dkt. 226 at 211-12.

263. The commissioners court held the November 12 special meeting at the League City Annex on Calder Road. PX-412 at 55-60. League City is twenty-seven miles from the city of Galveston, the county seat and where the commissioners court holds its regular meetings.

264. The first meeting of the commissioners court at the League City Annex was in 2013. Dkt. 225 at 89-90. The custom of sometimes meeting away from the county courthouse, at locations termed "auxiliary courts," was initially conceived to occur only "in the event the County of Galveston becomes precluded from conducting business or judicial functions within the county seat due to meteorological or catastrophic events." PX-412 at 56. But meeting away from the county courthouse soon became more common. Even so, meetings at the League City Annex generally pertained to non-controversial routine business, such as payroll approvals. Id. at 56-57. Serious, non-run-of-the-mill county business continued to be conducted at the county courthouse in the county seat.

265. But in recent years, it became more common for topics involving race to be taken up at the League City Annex. Examples include: (1) an August 24, 2020 meeting on the removal of a Confederate statue; (2) a July 2, 2021 meeting when the commissioners court extended an immigration-related disaster declaration; and (3) the November 12 meeting on redistricting. Id. at 55-56.

266. The League City Annex is not spacious and will not accommodate as many people as the county courthouse. Lay testimony

and video evidence of the November 12 special meeting indicate that by holding the meeting at the League City Annex, the commissioners court failed to provide the adequate space needed to accommodate the number of persons who sought to attend.

267. The small size of the League City Annex and the foreseeably large crowd caused congestion and overcrowding. Constable Rose testified that the League City Annex "was under construction . . . . The parking was terrible. It's just not the place that you want to hold a meeting of that magnitude." Dkt. 221 at 75. Additionally, Constable Rose observed that "people were standing all along the walls in the hallways . . . . You have got people [in] wheelchairs, walkers, everything there, and the accommodation was very poor." Id.; see also Dkts. 221 at 134-41; 226 at 132-35, 226-27.

268. The defendants have not provided any credible explanation for their failure to hold the special meeting in a space that would accommodate the foreseeably sizable crowd. Judge Henry was responsible for scheduling the time and place of the November 12 meeting. Dkt. 228 at 257. Henry testified that after posting the draft maps on October 29, the commissioners court received "more comments . . . than [on] anything else we have ever posted" and "received more comments and feedback than [on] any other thing we had done." Id. at 213, 220-21. The commissioners court was well aware of how sensitive the issue was, and how interested the public was in how it would be dealt with.

f. Disregard for Public Input from Minority Residents

269. Conduct by Judge Henry and the county commissioners indicated a disregard for public input from the minority communities and those critical of the enacted plan's discriminatory effect.

270. Judge Henry admitted that he reviewed fewer than a dozen of the public comments. Dkt. 228 at 221, 330. Instead, he had his staff provide a breakdown of comments, which he then announced during the November 12 special meeting before making the motion to adopt Map 2:

Of the 440 [comments] that came in, 168 did not discuss a particular map, they just called me names mostly. Of the people who did choose a map preference, Map 1 — received 64 responses. Map 2 received 208 responses. So of those responding to a particular map, 76.4[%], Map 2. 23.5[%], Map 1. With that, I'm going to make the motion to approve Map 2.

PX-591 at 62.

271. A detailed look at the public comments, JX-42, indicates that Henry's summary during the November 12 meeting disregards public commentary expressing concern over the discriminatory impact of redistricting on Galveston County's minority community. Dr. Burch analyzed all 446 public comments that were submitted. Dkt. 222 at 145-46; PX-414 at 23. She found that Judge Henry "dismissed as devoid of meaningful content nearly every comment that did not support the maps and that expressed concerns about racial discrimination and minority[-]vote dilution." PX-414 at 23.

272. The county residents who appeared at the meeting on November 12 were predominantly Black and Latino and included many older residents. PX-412 at 60; PX-129.

273. When attendees informed the commissioners court that they could not hear the proceedings, Judge Henry reacted by threatening to have constables remove attendees:

I'm going to speak at this tone. That's all I can do. I'm not going to scream. I don't have a microphone. . . . I will clear you out. If you make a noise, I will clear you out of here. I've got constables here.

PX-591 at 3.

274. Witnesses testified that Judge Henry was "real ugly about clearing the room." Dkt. 221 at 77, 138-40. Commissioner Giusti believed his conduct was "aggressive." Dkt. 232 at 150-51.

275. Thirty-five of the thirty-six members of the public who spoke at that meeting opposed Map 2. See generally PX-591 at 4-57. The remaining comments noted the inconvenience of the meeting and the lack of public transparency in the process. See id. Commissioner Giusti acknowledged that only Commissioner Holmes attempted to respond to the audience's concerns. Dkt. 232 at 148-49.

276. The commissioners court adopted the enacted plan without addressing any public comments received at the meeting. They did not publicly debate the map proposals beyond Judge Henry's discussion of the online public commentary and Commissioner Holmes's remarks. PX-591 at 61-81.

g. Excluding Commissioner Holmes

277. The court finds that the 2021 redistricting process exhibited an exclusion of Commissioner Holmes. He was the only minority commissioner at the time. His district — Precinct 3 — was dramatically reshaped under the enacted plan, and there was otherwise a lack of opportunity for minority voters to participate. Commissioner Holmes testified to this exclusion: he was not notified when the maps were finalized, was not told why additional public meetings were not held, and was never sent the data underlying the map proposals as he requested. Dkt. 228 at 103, 111-12.

278. Because of his experience in the 2011 redistricting cycle, Commissioner Holmes took contemporaneous notes of his conversations concerning the 2021 redistricting. Id. at 61-62; JX-23. Commissioner Holmes requested specific changes to balance Precinct 3's population and to make the precinct lines more understandable to voters. Id. at 68-72; JX-23 at 3. Some of those changes were not reflected in any map proposal, including Map 1. Compare Dkt. 231 at 75-77 (Oldham testifying the "minimum change" map proposal was drafted to accommodate Commissioner Holmes but also to include predominantly Anglo Bolivar Peninsula in Precinct 3), with JX-23 at 3 (list of changes requested by Commissioner Holmes, including the addition of voting Precinct 142 to Precinct 3), and Dkt. 223 at 52-53 (confirming no portion of voting Precinct 142 was added to Precinct 3 in Map 1 even though it is roughly equal in population to the Bolivar Peninsula voting precincts that were added and would have been possible according to one-person, one-vote standards).

3. Purported and Actual Redistricting Criteria

279. One of the additional factors noted in the Senate Judiciary Report is whether the policy underlying the political subdivision's conduct was tenuous.

280. The defendants have disclaimed any consideration of race. They instead assert that they used seven factors in drafting and adopting the enacted plan, as described in their interrogatory responses:

(1) compliance with federal law,

(2) the creation of a coastal precinct,

(3) geographic compactness,

(4) minimizing voting precinct splits,

(5) incumbency protection,

(6) partisanship, and

(7) "adopt[ing] a map that would be clear and easy to understand by the public."

PX-593 at 6-8. The rationales stated by members of the commissioners court in public, in deposition testimony, and at trial

are inconsistent with these purported criteria.

281. No witness testified at trial to applying the criteria described in the defendants' interrogatory responses in either drawing or adopting the enacted plan. See generally Dkts. 228 at 312-25; 232 at 88 (Giusti testifying he considered the inclusion of his and his parents' residence in his precinct, population equalization, and that lines were "drawn in a way the people understood" during the 2021 redistricting process), 304-05, 307-08 (Apffel testifying that he had no requests other than "equalized population" and keeping his new home in his precinct). Judge Henry admitted that he did not know of or apply the criteria the commissioners court claimed in its interrogatory responses to have used in the redistricting process. Dkt. 228 at 323-24.

282. Notably, unlike the criteria used in the 2001 redistricting cycle, the criteria the county revealed in its interrogatory responses do not include such objectives as maintaining communities of interest, preventing the unnecessary fragmentation of minority populations, or adhering to historic boundaries. See PX-539. From 1981 until 2021, the commissioners court had at least one precinct that performed to elect the candidate of choice for Black and Latino voters; the criteria the county purports to have used in 2021 would have done little to preserve that longstanding, and long-performing, majority-minority precinct.

283. The plaintiffs have provided several illustrative map configurations that perform as well or better than the enacted plan under the disclosed criteria.

284. Cooper reviewed the criteria provided by the defendants and evaluated whether the enacted plan adhered to them. In his opinion, it did not. PX-386 at 23-26. As to the first factor — compliance with the Constitution and the Voting Rights Act — Oldham disclaimed any requirement to draw a precinct that conformed with § 2 of the Voting Rights Act. Dkt. 231 at 61-62. Commissioner Apffel knew of Commissioner Holmes's concerns about the potential Voting Rights Act violation. Dkt. 232 at 329-30.

285. Dr. Rush presented alternative maps that comply with federal law and maintain Precinct 3 as a majority-minority precinct. PX-487 at 9-17. Cooper explained that there "are many, many different ways to draw a majority Black plus Latino precinct. You can make [a] few changes. You can make lots of changes. It can look a lot of different ways." Dkt. 223 at 47. Similarly, Fairfax testified that "there are possibilities of different configurations [of illustrative maps that] still continue to create a majority Black and Latino district that satisfied the first precondition of Gingles and followed traditional redistricting criteria." Dkt. 224 at 117.

286. As to the second and third criteria — creating a coastal precinct and geographical compactness — the plaintiffs have provided multiple illustrative maps that would create a compact coastal precinct while maintaining a majority-minority Precinct 3. Creating a coastal precinct is not mutually exclusive with preserving Precinct 3 as a majority-minority district that allows Black and Latino voters to elect a candidate of choice. Dkt. 222 at 131.

287. Nor does the evidence support the need or any popular support for a single coastal precinct. Before the map's passage, "there weren't a bunch of people clamoring for a coastal precinct." Id. at 105-06; Dkt. 225 at 85. The documents that Dr. Krochmal examined had just one mention in a news article from the early 2010s. Dkt. 225 at 84-85. The commissioners did not engage the public on the need for a coastal precinct. See, e.g., Dkts. 228 at 315, 317; 232 at 142-43. Some even advocated

against a single-coastal precinct. Dkts. 228 at 315-17; 232 at 350-52.

288. As to the fourth and fifth criteria — to minimize the splitting of voting precincts before including incumbent residences — the defendants generally followed these criteria when drafting the enacted plan. But the alternative maps created by Dr. Rush protect the incumbency of the current commissioners while also preserving a majority-minority precinct. See, e.g., Dkt. 232 at 157-58.

289. The sixth criterion — partisanship — did not require the enacted plan's configuration, as all members of the commissioners court who voted for the enacted plan disclaimed partisanship as a predominating consideration. See Dkts. 228 at 197, 304; 232 at 98, 355-56. Consistent with this, Oldham testified that he never told Bryan that Judge Henry's purpose for Map 2 was to create four Republican districts, and Oldham denied there was any such partisan motivation. Dkt. 231 at 153-54.

290. As for the final criterion, that the map be "clear and easy to understand by the public," Cooper observed that the dramatic changes in the enacted plan do the opposite. "[B]ecause entire [voting] precincts are going to be shifted around into different districts," the likelihood of voter confusion — such as voters not knowing in which commissioner's precinct they reside — is high. Dkt. 223 at 85-88.

291. Any contention that the defendants adopted the enacted plan to achieve near-equal population deviation is unsupported by the record. See, e.g., PX-191. Indeed, there is no requirement to achieve a zero deviation for the commissioners-court map. Dkt. 223 at 186.

292. Judge Henry admitted he viewed benchmark Precinct 3 as a racial gerrymander and that any majority-minority Precinct 3 would have to look that way. Dkt. 228 at 319. Commissioner Apffel similarly testified that he believed and "ha[d] been told" that Precinct 3 had been "racially gerrymandered in favor of minorities." Dkt. 232 at 356. Judge Henry further admitted he "would not have asked for" a coastal precinct map that kept the core of benchmark Precinct 3. Dkt. 228 at 305.

H. Ongoing Discrimination Touching on Participation in Voting

293. Three critical Senate factors in this case are: (1) the extent to which minority-group members bear the effects of discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process; (2) the use of overt or subtle racial appeals in political campaigns; and (3) the extent to which members of the minority group have been elected to public office in the jurisdiction.

294. The 2021 redistricting process for commissioners precincts occurred within a climate of ongoing discrimination affecting Black and Latino voting participation. This climate has led to limited Black and Latino electoral success in Galveston County. Notably, the limited success of Black and Latino candidates for office is largely attributable to majority-minority districts like the one dismantled by the enacted plan.

295. Dr. Burch's and Dr. Rocha's qualitative and quantitative analyses show that Black and Latino residents of Galveston County bear the effects of discrimination in income, poverty, education, and health, all of which combine to increase the costs of voting and decrease political participation. PXs-335 ¶¶ 66-71; 414 at 23-32. The defendants did not call any fact witnesses to rebut these disparities between the minority and Anglo populations in Galveston County.

296. Historical disparities contribute to the contemporary inequalities that persist, not least because there are Galveston

County voters alive today who lived through the Jim Crow era. Dkt. 222 at 74. For example, Pope testified to living through desegregation in education and public accommodations and the difficulties the county faced during that time. Id. at 28-31. Similarly, Reverend Randall described "racial fights" that occurred "in the junior high and high school [that] spilled over into the community" as he grew up during school desegregation. Dkt. 226 at 218.

297. Testimony from Black and Latino residents confirms that the county's Black and Latino voters still suffer similarly from discrimination in income, poverty, education, and health as compared to Anglo residents. Dkts. 221 at 65, 133; 226 at 14.

298. Discrimination against minorities in Galveston County harms their ability to participate equally in the electoral process. PX-335 ¶¶ 9, 18, 27. Racial and ethnic disparities in education, income, housing, and public health are partly the result of past and present discrimination. Peer-reviewed academic research confirms that such disparities hinder Latinos and Blacks from participating in the political process. Id. ¶¶ 18, 27, 66-67.

299. Black and Latino voters, as measured by their consistently lower turnout rate than Anglo voters in Galveston County elections, have a depressed level of political participation. Dkt. 230 at 157.

1. Contemporary Voting Barriers

300. Residents testified how voter-identification requirements and voter-roll purging weigh more heavily on Black and Latino voters and constitute barriers to voting. Dkts. 222 at 226-28; 226 at 18; see also PX-414 at 11 (describing Texas passing stricter voter ID laws after Shelby County).

301. Closure of polling places has also made it more difficult for the county's Black and Latino residents to vote. The number of polling places in predominantly minority neighborhoods has decreased in recent decades. Dkt. 221 at 86-87.

302. The decline in polling places is partially attributable to the adoption of voting centers — locations where any voter can vote, regardless of the voter precinct in which he or she resides. But the convenience these centers provide is undercut by the fact that Galveston County has not established the mandatory minimum number of vote centers required under Texas law. Before the November 2022 general election, civil-rights organizations sent a letter to the Galveston County clerk and commissioners court informing them that the county had opened only twenty-eight voting centers, rather than the minimum required forty-one. PX-315. The letter described the disproportionate impact reduced polling places would have on minority residents with less access to transportation. Id. The county had also been warned in 2019 of this issue. Id.

303. Black and Latino residents have reported greater difficulty getting to polling places due to difficulties obtaining transportation. Dkts. 226 at 58, 209; 228 at 43-44. Studies have shown that polling-place distance affects voter turnout, and those effects are related to transportation availability. PX-414 at 28.

304. The county has recently closed or attempted to close specific polling places in predominantly Black and Latino neighborhoods. The historic Carver Park polling place was considered by many to be the "hub" of the benchmark Precinct 3. Dkt. 221 at 61-62. It was closed in 2020, leaving some people unable to vote because of a lack of transportation. Dkt. 222 at 278. Similarly, another polling place closed during the 2020 election, the Dickinson Senior Center, which was also predominantly used by Black and Latino voters. Dkt. 228 at 44. The county also attempted to eliminate the Alamo Elementary polling location,

which is located in a heavily Latino neighborhood. Dkt. 226 at 69.

305. Primary elections for commissioners court seats have a majority-vote requirement. Tex. Elec. Code § 172.003.

306. In recent years, candidates running for office in Galveston County have made implicit racial appeals in their campaigns. PX-335 ¶¶ 72-81; Dkt. 222 at 89. Racial appeals can "make racial attitudes and concerns more salient in the minds of voters, even without explicitly mentioning or referring to a particular race or group." PX-414 at 33; see also PX-335 ¶¶ 76-77; Dkt. 225 at 237-38.

307. For example, in the 2020 Republican primary for Galveston County tax assessor-collector, candidate Jackie Peden sent a mailer of a "Latino man covered in tattoos that indicates association with a violent gang (in this case, MS-13)." PXs-335 ¶ 78; 561; Dkts. 222 at 90-91; 225 at 240-41; Dkt. 226 at 17. Referring to her incumbent opponent, Peden's mailer stated, "Texans can thank Cheryl Johnson for having illegal immigrants vote in this November's Election!" PX-561. The mailer "use[d] text to associate her opposing candidate. . . with 'illegal immigrants' and appeals to race-based biases and fears regarding Latinos." PX-335 ¶ 78. The image of the tattooed man featured is not a Galveston County resident but an El Salvadoran man whose image has been featured in other political campaigns. Id. ¶ 79; PX-562; Dkt. 225 at 241. Other candidates in Galveston County have used anti-immigrant imagery and "invasion" language as an anti-minority appeal. PX-414 at 34.

308. In 2022, Julie Pickren, a candidate for District 7 of the Texas State Board of Education, shared a video showing Black students at a local high school vandalizing the school cafeteria. PX-335 ¶ 80. The video was accompanied by text stating, "Discipline in schools must be restored." Id. As an implicit racial appeal, the video contains images of Black youth engaged in a stereotype-confirming behavior: violent destruction of property. Id.

309. Other racial incidents surrounding campaigns have occurred in recent years. In 2019, the chairwoman of the Galveston County Republican Party referred to a particular Black Republican as a "typical nig." Dkt. 222 at 89-90; PX-414 at 34. Residents also heard racially derogatory language used toward Barack Obama during his presidential campaigns. Dkt. 226 at 17.

2. Lack of Electoral Success

310. Minorities have been underrepresented in electoral success relative to their share of Galveston County's population. PX-414 at 34. Until the 1992 consent decree that increased minority representation in the JP/constable precincts, few minorities were elected to county office. Dkt. 222 at 15-16.

311. There have been three Black members of the commissioners court: Wayne Johnson, Stephen Holmes, and Robin Armstrong. Dkt. 204-6 ¶¶ 14-16. Wayne Johnson was the first Black member of the commissioners court and was elected in 1988. Id. ¶ 15. Stephen Holmes was appointed in 1999 after Wayne Johnson's passing and has served continuously since then. Id. ¶ 16. Robin Armstrong was initially appointed in 2022 and ran unopposed in a majority-Anglo precinct in the November 2022 general election. Dkt. 230 at 195, 198, 211. There has been just one Latino member of the commissioners court, Frank Carmona, who served from 1971 to 1990. PX-335 ¶ 7.

312. The court heard testimony that Commissioner Armstrong holds "several [political] views that are outside the mainstream of Black Americans." PX-414 at 34-35. NAACP, LULAC, and other minority groups did not endorse him. Id. at 35. Commissioner Armstrong acknowledged

he does not have a basis to believe he is a candidate of choice for Black or Latino voters. Dkt. 230 at 210.

313. Commissioners Giusti and Apffel could not identify any minority candidates who successfully ran in a countywide Republican primary. Dkts. 232 at 153, 367. Latino candidates with Spanish surnames have had minimal success in the county's Republican primaries. Dkt. 226 at 15-17.

314. The limited number of successful Black and Latino elected officials within the county have tended to be members of city councils elected from majority-minority districts in cities with larger minority populations, such as Texas City and La Marque, or — in the case of the city of Galveston — elected from single-member districts created by court order to be majority-minority. See, e.g., Dkts. 222 at 16; 230 at 255-56; 232 at 151.

3. Responsiveness

315. In § 2 vote-dilution claims, courts must consider "whether there is a significant lack of responsiveness by elected officials to the minority group members' particularized needs." Patino v. City of Pasadena, 230 F. Supp. 3d 667, 715-16 (S.D. Tex. 2017). Lack of responsiveness to public policies important to minorities serves to create and maintain racial disparities. PX-414 at 35.

316. Anglo commissioners are evidently not actively engaged in specific outreach to Galveston County's minority residents. Dkt. 222 at 98-99; PX-414 at 35-36. Commissioner Apffel could not identify any wants, needs, or desires that African American and Latino constituents have. Dkt. 232 at 369. Judge Henry testified that he has never received an endorsement from leaders in the Black or Latino communities. Dkt. 228 at 254-55.

317. Minority residents have indicated that the commissioners court has become less responsive to the needs of minority constituents since the enacted plan went into effect. One resident testified that the lack of responsiveness makes minorities "not want to participate" in the political life of the county. Dkt. 222 at 28.

318. Recent votes by the commissioners court have been seen as unresponsive to the Black and Latino communities. Numerous witnesses cited the commissioners court's vote to allow $1.8 million to help pay for a U.S.-Mexico border wall and to send constables to the border as racially discriminatory and diverting much-needed resources from local Black and Latino residents. Id. at 103; Dkt. 226 at 19-20. In addition, Anglo commissioners opposed removing the Confederate statue in front of the county courthouse. PX-412 at 50.

319. The commissioners court's handling of the November 12 special meeting also portrayed a lack of responsiveness. Witnesses who attended the special meeting were taken aback by some of Judge Henry's comments. Dkts. 221 at 77; 226 at 137. They testified that his disinterested demeanor to public comments also suggested a lack of responsiveness. Dkts. 222 at 246; 226 at 133-34. Galveston City Councilwoman Lewis described it as "almost like a back-in-the-'60s environment." Dkt. 226 at 134.

320. Many residents reported that only Commissioner Holmes is responsive to Black and Latino residents' needs. See, e.g., Dkts. 221 at 93; 222 at 26-27, 235-36; 226 at 73-74, 192, 221 ("I don't know any important event that Stephen wasn't there and giving us what we needed.").

321. Commissioner Holmes was instrumental in reopening the Wayne Johnson Senior Center after Hurricane Ike. Dkt. 226 at 221 ("Even as far as Ike, Harvey, even Katrina. He led most of those efforts with housing and getting us back on our feet during tough times . . . ."). 322. During the COVID-19 pandemic, Commissioner Holmes participated in weekly calls with the community. He "would get doctors, the health district, epidemiologists" and arrange testing opportunities. Id. at 221-22. He helped set up a phone line for people without internet access to sign up for COVID-19 vaccines. Dkt. 221 at 118-19.

323. Commissioner Holmes also helped arrange transportation for senior citizens and students around the county. Dkt. 222 at 237-38.

324. Residents of benchmark Precinct 3 who now live in Commissioner Armstrong's precinct do not believe he would be responsive to his constituents from their neighborhoods due to his lack of involvement in minority community organizations and his absence from the Black and Latino community. Id. at 240-42; Dkt. 226 at 228. Commissioner Armstrong admitted that he has no basis to believe he is the candidate of choice among minority voters. Dkt. 230 at 210.

4. Education

325. Educational level is a dependable predictor of political participation. The higher one's education has advanced, the more likely he or she is to be politically active. PXs-335 ¶ 68; 414 at 22; Dkt. 225 at 233-34.

326. Minorities in Galveston County continue to bear the effects of discrimination in education. Dkts. 222 at 73; 225 at 220. Black and Latino residents are much less likely to have high-school diplomas than Anglo residents. PX-414 at 25-26; Dkt. 222 at 76-77. About 87.7% of Blacks and 75.9% of Latinos over twenty-five years old have completed high school — compared to 94.8% of Anglos. PXs-335 ¶ 14; 414 at 26. On the other hand, about 22.1% of Blacks and 17.5% of Latinos over twenty-five years old have earned a bachelor's degree or higher, while the Anglo rate is 37.5%. PX-335 ¶ 14.

327. Schools in Galveston County were under a court desegregation order from 1961 to 2009. Id. ¶ 28. During that time, the Galveston Independent School District struggled to achieve racially balanced enrollments in its elementary schools. Id.; Dkt. 225 at 209-10.

328. Racial inequality in K-12 educational achievement persists in Galveston County. PX-335 ¶¶ 28-39. Disproportionate numbers of Black and Latino students are not proficient in reading and math across all school districts in Galveston County. PX-414 at 24; Dkt. 222 at 75-76.

329. There is clear evidence in Galveston County of "practices that limit the integration of schools and deny minority students access to education." PX-335 ¶ 29. Minorities are less likely than Anglos to benefit from positive programs, such as Advanced Placement classes or gifted and talented programs, and are more likely to suffer disciplinary action, such as in-school and out-of-school suspension. Id. ¶¶ 29-39; PX-559; Dkts. 225 at 211-16; 226 at 118-19, 207-08.

330. These educational disparities have contributed to the lower likelihood that minorities in Galveston County will participate in the political process.

5. Employment and Poverty

331. Employment disparities are important to understanding the cost of voting; voters with lower-wage jobs are much more likely to be hindered from accessing the ballot box. PX-414 at 26. Research also shows that the workplace is an important site for recruitment into political participation. Id. at 27; see also PX-335 ¶ 67. So higher rates of employment and higher-wage jobs mean an increase in electoral participation.

332. Racial disparities in earnings are present in several employment sectors within Galveston County. PXs-335 ¶¶ 40-45;

414 at 26-28. In Galveston County, earnings for Blacks and Latinos are also lower in general than those for Anglo employees, partly because they are clustered in jobs that earn less. PXs-335 ¶¶ 40-45; 414 at 26-27; 559; Dkt. 225 at 222-23. Even when they hold the same types of jobs, Blacks and Latinos earn less than Anglo employees. PXs-335 ¶¶ 40-45; 559.

333. Black Galveston County households have a median income of $45,831, and Latino households have a median income of $60,297-markedly lower than Anglo households' median income of $86,165. PX-414 at 26; see also PX-386 ¶ 40.

334. Likewise, the unemployment rate in Galveston County is disproportionately higher for Black and Latino residents than for Anglo residents. The unemployment rates for Blacks and Latinos are 9.1% and 7.0%, respectively; the unemployment rate for Anglos is 4.8%. PXs-335 ¶ 15; 414 at 28; Dkt. 222 at 79; see also PX-386 ¶ 40.

335. About 29.2% of Black households and 15.1% of Latino households rely on the Supplemental Nutrition Assistance Program, whereas only 6.7% of Anglo households do. PX-386 ¶ 40.

336. Additionally, the child-poverty rate in Galveston County is disproportionately higher for Black children — nearly 25% — and Latino children-over 20%-than for Anglo children — under 10%. PX-414 at 27; Dkt. 222 at 78-79; see also id. ¶ 40.

337. Black and Latino residents in Galveston County have lower rates of car ownership than do Anglo residents. PX-414 at 28; Dkt. 222 at 79. Black households are four times less likely to have access to a car than Anglo households; Latinos are also less likely to have access to a vehicle. PX-414 at 28.

338. Together, these economic disparities hinder the ability of Galveston County's Black and Latino communities to participate effectively in the political process.

6. Housing

339. Renters move more frequently than do homeowners and so are less likely to vote, because changing residences frequently increases the administrative burden of maintaining voter registration. Id. at 30. Homeowners are also more likely to be mobilized by political campaigns, increasing their likelihood of voting. PX-335 ¶ 69.

340. In 1997, the Department of Housing and Urban Development found that the Galveston Housing Authority, the agency in charge of public-housing assistance and the management of Section 8 vouchers, had used public housing to reinforce patterns of segregation, in violation of Title VI of the Civil Rights Act of 1964. Id. ¶ 48.

341. Disparities in homeownership for Blacks and Latinos as compared to Anglos persist in Galveston County today and have not meaningfully decreased in recent years. Id. ¶ 49; PX-559; Dkt. 225 at 230-31. Black and Latino residents are less likely to live in owner-occupied housing than are Anglo residents. Dkt. 222 at 82; PX-414 at 31. About 47.5% of Blacks reside in owner-occupied housing units. PX-335 ¶ 16. Slightly more Latinos, 60.6%, reside in owner-occupied housing. Id. Anglo rates, though, are at 73.3%. Id.

342. In the aftermath of Hurricane Ike in 2008, disparities in housing grew even more pronounced. PXs-335 ¶¶ 50-59; 412 at 40-42; 414 at 31; Dkts. 221 at 111-13; 222 at 33-34.

343. The city of Galveston lost 16.5% of its overall population between 2000 and 2010, including an 11.4% loss in the Anglo population compared to a 36.7% loss in the Black population. PX-335 ¶ 51. Many of the predominantly minority Galveston Island residents displaced by Hurricane Ike moved to the mainland and have been unable to return due to the lack of affordable

housing. Dkt. 222 at 232-33. Rebuilding public housing after Ike also had a racialized component; predominantly Anglo residents and politicians opposed rebuilding efforts. PX-412 at 40-42.

344. These housing disparities hinder the ability of Galveston County's Black and Latino communities to participate effectively in the political process.

7. Public Health

345. Healthy individuals are more likely to be civically engaged. Poor health can reduce the odds of voting by 12%. PXs-335 ¶ 70; 414 at 31.

346. Black and Latino residents in Galveston County disproportionately suffer from public-health issues compared to Anglo residents and continue to bear the effects of discrimination in public health. PX-335 ¶ 58; Dkts. 222 at 83; 225 at 232.

347. Discrimination increases incidents of psychological distress, major depression, generalized anxiety disorder, and early initiation of substance abuse. These general patterns have been documented among minority residents in Galveston County. PX-335 ¶ 59. A study of 1,238 Latinos living in Texas City, published in the Journal of Social Science & Medicine in 2013, found a significant relationship between experiences with discrimination and poor mental-health outcomes. Id.

348. Significant disparities in infant health and life expectancy exist between Blacks and Anglos in Galveston County. Id. ¶¶ 60-61; Dkt. 225 at 232-33. Black infants are more than twice as likely to have low birth weight and have nearly double the infant mortality rate as Anglo infants. PX-335 ¶ 60.

349. In Galveston County, Blacks and Latinos suffer disparities in insurance coverage that also affect access to preventative health care. Id. ¶ 61. Latinos between the ages of nineteen and sixty-four are more than twice as likely as Anglos to be uninsured. Id. ¶ 17. About 12.5% of Blacks between the same ages do not have health insurance. Id.

350. Disparities in health outcomes for Blacks and Latinos in Galveston County decrease their level of political participation.

8. Criminal Justice

351. Black and Latino also residents face disparities in the criminal-justice system in Galveston County. Dkt. 222 at 84-85; PX-414 at 32-33. Criminal-justice interactions affect political behavior because higher arrest and incarceration rates can hinder one's ability to vote. Dkt. 222 at 85-86; PX-414 at 32.

352. Black residents in Galveston County are more likely to be arrested, and Black and Latino residents comprise a disproportionate percentage of jail and prison inmates compared to their share of the population. Dkt. 222 at 36, 85. Black and Latino residents also testified to over-policing and disparities in treatment by the criminal-justice system. See, e.g., Dkts. 221 at 67-70 (Constable Rose describing instances of being pulled over and treated aggressively); 226 at 13-14 (Quintero describing a complaint LULAC received after police severely damaged a Latino family's house), 194-95 (Henderson-Lofton describing being pulled over by police in League City).

353. A highly publicized 2019 incident in which police on horseback led a mentally disabled Black man, Donald Neely, by a rope led to widespread criticism, including by the police chief, who stated that the officers "exercised poor judgment." Dkt. 221 at 73; see also Dkt. 222 at 86-87.

III. Conclusions of Law

354. The Fifteenth Amendment was ratified in 1870 "amidst the struggles of Reconstruction to fully guarantee voting

rights to newly freed slaves." Johnson v. Waller County, 593 F. Supp. 3d 540, 592 (S.D. Tex. 2022). Section 1 protects citizens' right to vote from being "denied or abridged by the United States or by any State on account of race, color, or previous condition of servitude." U.S. Const. amend. XV, § 1. Section 2 grants Congress "the power to enforce this article by appropriate legislation." Id. amend. XV, § 2.

355. The first ninety-five years of congressional enforcement of the Fifteenth Amendment "can only be regarded as a failure," Nw. Austin Mun. Util. Dist. No. One v. Holder, 557 U.S. 193, 197, 129 S.Ct. 2504, 174 L.Ed.2d 140 (2009), marred by "Jim Crow laws like literacy tests, poll taxes, and 'good-morals' requirements," Milligan, 599 U.S. at 9, 143 S.Ct. 1487. Motivated by the Civil Rights movement, Congress passed the Voting Rights Act in 1965. Id. at 10, 143 S.Ct. 1487. The original text of § 2 of the Voting Rights Act "closely tracked the language" of the Fifteenth Amendment and was "'little-used' for more than a decade after its passage." Brnovich v. Democratic Nat'l Comm., 594 U.S. 647, 141 S. Ct. 2321, 2331, 210 L.Ed.2d 753 (2021). And in 1980 in City of Mobile v. Bolden, the Supreme Court held that § 2 prohibited only the discriminatory intent to dilute the voting strength of a minority group — not conduct that has the discriminatory effect of diluting its voting strength. 446 U.S. 55, 61-66, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980) (plurality opinion). As a result, § 2 was greatly weakened in its ability to protect minorities from voting practices producing discriminatory results.

356. Bolden "produced an avalanche of criticism." Milligan, 599 U.S. at 11, 143 S.Ct. 1487 (quoting Thomas M. Boyd & Stephen J. Markman, The 1982 Amendments to the Voting Rights Act: A Legislative History, 40 Wash. & Lee L. Rev. 1347, 1355 (1983)). After vigorous debate, Congress amended the Voting Rights Act in 1982, revising § 2 "to make clear that a violation could be proved by showing discriminatory effect alone." Gingles, 478 U.S. at 35, 106 S.Ct. 2752.

A. Section 2 of the Voting Rights Act

357. After its amendment, § 2 specifically prohibits any "voting qualification or prerequisite to voting or standard, practice, or procedure" that "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). Such a denial or abridgement occurs when "the totality of circumstances" shows that a state's "political processes . . . are not equally open to participation by" members of a minority group "in that [they] have less opportunity . . . to participate in the political process and to elect representatives of their choice." Id. § 10301(b).

358. Such claims are often called Gingles claims because Thornburg v. Gingles provides the "framework" for evaluating § 2 vote-dilution claims. Wis. Legislature v. Wis. Elections Comm'n, 595 U.S. 398, 402, 142 S.Ct. 1245, 212 L.Ed.2d 251 (2022) (per curiam).¹³ In Gingles, the Supreme Court construed § 2 to prohibit the "dispersal of [a minority group's members] into districts in which they constitute an ineffective minority of voters." Cooper v. Harris, 581 U.S. 285, 292, 137 S.Ct. 1455, 197 L.Ed.2d 837 (2017) (quoting Gingles, 478 U.S. at 46 n.11, 106 S.Ct. 2752). When "minority and majority voters consistently

13 Gingles itself involved § 2 challenges to multimember districts, 478 U.S. at 46, 106 S.Ct. 2752, but the Supreme Court later extended the analysis to apply to § 2 challenges to single-member districts like the ones at issue here. See Growe v. Emison, 507 U.S. 25, 40-41, 113 S.Ct. 1075, 122 L.Ed.2d 388 (1993).

prefer different candidates" in such districts, "the majority, by virtue of its numerical superiority, will regularly defeat the choices of minority voters," thus depriving minorities of an equal opportunity to elect representatives of their choice. Gingles, 478 U.S. at 48, 106 S.Ct. 2752. Today, § 2 still prohibits vote dilution in redistricting plans that "minimize or cancel out the voting strength of racial [minorities in] the voting population." Id. at 47, 106 S.Ct. 2752 (alteration in original) (internal citations omitted) (quoting Burns v. Richardson, 384 U.S. 73, 88, 86 S.Ct. 1286, 16 L.Ed.2d 376 (1966)); see also Milligan, 599 U.S. at 25, 143 S.Ct. 1487.

359. In Gingles, the Supreme Court "established a two-step analysis for vote-dilution claims." Patino, 230 F. Supp. 3d at 675 (citing Gingles, 478 U.S. at 50-51, 106 S.Ct. 2752). Plaintiffs must first establish three preconditions: "(1) [t]he minority group must be sufficiently large and compact to constitute a majority in a reasonably configured district, (2) the minority group must be politically cohesive, and (3) a majority group must vote sufficiently as a bloc to enable it to usually defeat the minority group's preferred candidate." Wis. Legislature, 595 U.S. at 402, 142 S.Ct. 1245. If plaintiffs establish the preconditions, they must then show that, under the "totality of circumstances," the "political process is [not] equally open to minority voters" without the proposed district. Id. (quoting Gingles, 478 U.S. at 79, 106 S.Ct. 2752). When a § 2 claim is successful, a court will require the creation of a majority-minority election district in which minority voters have an equal opportunity to elect their preferred candidates. See Bartlett v. Strickland, 556 U.S. 1, 13, 129 S.Ct. 1231, 173 L.Ed.2d 173 (2009).

360. Plaintiffs must prove § 2 claims by a preponderance of the evidence. LULAC v. Roscoe Indep. Sch. Dist., 123 F.3d 843, 846 (5th Cir. 1997).

361. In Allen v. Milligan , decided earlier this year, the Supreme Court upheld the Gingles framework. 599 U.S. 1, 143 S.Ct. 1487.

1. Step One — Preconditions

362. The first precondition is that the minority group "is sufficiently large and geographically compact to constitute a majority in a single-member district." Gingles, 478 U.S. at 50, 106 S.Ct. 2752; see also id. at 18, 143 S.Ct. 1487. This precondition is "needed to establish that the minority has the potential to elect a representative of its own choice." Milligan, 599 U.S. at 18, 143 S.Ct. 1487 (quoting Growe, 507 U.S. at 40, 113 S.Ct. 1075). Accordingly, the minority group must be able to constitute a majority by CVAP in the proposed district. Valdespino v. Alamo Heights Indep. Sch. Dist., 168 F.3d 848, 853 (5th Cir. 1999); see also LULAC v. Perry, 548 U.S. 399, 428-29, 126 S.Ct. 2594, 165 L.Ed.2d 609 (2006) (analyzing CVAP and noting that "only eligible voters affect a group's opportunity to elect candidates"). A plaintiff must also allege that its proposed majority-minority district "is consistent with 'traditional districting principles such as maintaining communities of interest and traditional boundaries.'" Robinson v. Ardoin, 37 F.4th 208, 218 (5th Cir. 2022) (per curiam) (quoting LULAC, 548 U.S. at 433, 126 S.Ct. 2594).

363. The Supreme Court has explained that a "district will be reasonably configured . . . if it comports with traditional districting criteria, such as being contiguous and reasonably compact." Milligan, 599 U.S. at 18, 143 S.Ct. 1487; see also Miller v. Johnson, 515 U.S. 900, 916, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995) (identifying traditional districting criteria such as "compactness, contiguity, and respect for political subdivisions or communities defined by actual shared interests"). Courts may also consider other traditional

redistricting criteria, including equal population, respect for political boundaries, and keeping together communities of interest. See, e.g., Milligan, 599 U.S. at 19-20, 143 S.Ct. 1487; Ala. Legis. Black Caucus v. Alabama, 575 U.S. 254, 259, 135 S.Ct. 1257, 191 L.Ed.2d 314 (2015); Shaw v. Reno, 509 U.S. 630, 651-52, 113 S.Ct. 2816, 125 L.Ed.2d 511 (1993).

364. Courts often discuss the second and third preconditions together. The second requires the minority group to be "politically cohesive." Gingles, 478 U.S. at 51, 106 S.Ct. 2752. Cohesiveness concerns whether "a representative of [a minority group's] choice would in fact be elected." Milligan, 599 U.S. at 18, 143 S.Ct. 1487 (quoting Growe, 507 U.S. at 40, 113 S.Ct. 1075). Relatedly, the third precondition is that "the [Anglo] majority votes sufficiently as a bloc to enable it . . . usually to defeat the minority's preferred candidate." Gingles, 478 U.S. at 51, 106 S.Ct. 2752 (citation omitted). The last precondition "'establish[es] that the challenged districting thwarts a distinctive minority vote' at least plausibly on account of race." Milligan, 599 U.S. at 19, 143 S.Ct. 1487 (quoting Growe, 507 U.S. at 40, 113 S.Ct. 1075). Unless both preconditions are met, "the challenged districting [does not] thwart[ ] a distinctive minority vote by submerging it in a larger [Anglo] voting population." Growe, 507 U.S. at 40, 113 S.Ct. 1075.

365. Plaintiffs usually demonstrate minority political cohesion by showing that "a significant number of minority group members usually vote for the same candidates." Gingles, 478 U.S. at 56, 106 S.Ct. 2752; see also Campos v. City of Baytown, 840 F.2d 1240, 1244 (5th Cir. 1988). That is described as "bloc voting"¹⁴ and typically means that a large majority of the group favors the same candidates. When minorities and Anglos vote in opposing blocs, courts conclude that voting is "racially polarized" and typically hold that a plaintiff has established the second and third preconditions.

14 E.g., Strickland, 556 U.S. at 19, 129 S.Ct. 1231 (plurality opinion); Fusilier v. Landry, 963 F.3d 447, 458 (5th Cir. 2020).

366. The second and third preconditions view the redistricting process from different vantages. A plaintiff must show the second precondition for the minority population in its proposed district. See Harris, 581 U.S. at 302, 137 S.Ct. 1455; LULAC, 548 U.S. at 427, 126 S.Ct. 2594; Growe, 507 U.S. at 40, 113 S.Ct. 1075. The third precondition must be established for the challenged district. See Harris, 581 U.S. at 302, 137 S.Ct. 1455; LULAC, 548 U.S. at 427, 126 S.Ct. 2594; Growe, 507 U.S. at 40, 113 S.Ct. 1075. And each of these preconditions must be shown on a district-by-district basis. See Wis. Legislature, 595 U.S. at 404-05, 142 S.Ct. 1245; Abbott v. Perez, 585 U.S. 579, 138 S. Ct. 2305, 2332, 201 L.Ed.2d 714 (2018); LULAC, 548 U.S. at 437, 126 S.Ct. 2594.

a. Sufficiently Large and Geographically Compact

367. The defendants do not dispute that Galveston County's Black and Latino communities, when considered as a coalition, are sufficiently large to satisfy the first Gingles precondition. Instead, they contend that coalition claims are per se unlawful and that the plaintiffs' illustrative plans are not reasonably configured. The court rejects these arguments.

368. Coalition districts are "districts in which minorities are together a CVAP majority, but no individual minority group is." LULAC v. Abbott, 604 F.

Supp. 3d 463, 500 (W.D. Tex. 2022). The Fifth Circuit permits coalition claims under § 2. See LULAC, Council No. 4434 v. Clements, 999 F.2d 831, 863-64 (5th Cir. 1993) (en banc); Campos, 840 F.2d at 1244. The cohesiveness of minority coalitions is "treated as a question of fact, allowing aggregation of different minority groups where the evidence suggests that they are politically cohesive." Clements, 999 F.2d at 864. In the Fifth Circuit, "[t]here is nothing in the law that prevents . . . plaintiffs from identifying the protected aggrieved minority to include both Blacks and [Latinos]." Campos, 840 F.2d at 1244.

369. "Precedent in the Fifth Circuit is governed by a strict rule of orderliness, such that later panels of that court, and much less district courts within the circuit, cannot overturn decisions of prior panels." Abbott, 604 F. Supp. 3d at 493. The court will follow well-established Fifth Circuit precedent and recognize that Blacks and Latinos together form to a coalition that satisfies the first Gingles precondition.

370. Additionally, the plaintiffs' illustrative plans — and Map 1 — demonstrate that Galveston County's Black and Latino population is geographically compact enough to form a majority of eligible voters within a reasonably configured commissioners-court precinct. Cooper, Fairfax, and Dr. Rush have provided several illustrative plans that contain one majority Black and Latino commissioners-court precinct and adhere to traditional redistricting criteria, including equal population, contiguity, and compactness. These illustrative maps are but a few examples of a multitude of potential districts that are reasonably configured and that contain a majority Black and Latino population by CVAP.

371. Additionally, the plaintiffs do not need to consider specific communities of interest when drawing illustrative maps to satisfy the first Gingles precondition. Kumar v. Frisco Indep. Sch. Dist., 476 F. Supp. 3d 439, 499 (E.D. Tex. 2020). The plaintiffs' illustrative maps still sufficiently preserve communities of interest — namely the Black and Latino communities in benchmark Precinct 3. As the Supreme Court recently affirmed in Milligan, a party satisfies the first Gingles precondition by showing that a majority-minority precinct "comports with traditional districting criteria, such as being contiguous and reasonably compact." 599 U.S. at 18, 143 S.Ct. 1487. The plaintiffs have done so here.

372. Furthermore, "cultural compactness" is neither an element of a § 2 claim nor a component of the first Gingles precondition. In LULAC v. Perry , the Supreme Court held that one of six Latino opportunity districts was not "reasonably compact." 548 U.S. at 430, 126 S.Ct. 2594. The district contained "a 300-mile gap between the Latino communities . . . and a similarly large gap between the needs and interests of the two groups." Id. at 432, 434, 126 S.Ct. 2594 (noting that "the different characteristics, needs, and interests of the Latino community near the Mexican border and the one in and around Austin are well supported and uncontested"). The Court noted that "in some cases members of a racial group in different areas — for example, rural and urban communities — could share similar interests and therefore form a compact district if the areas are in reasonably close proximity." Id. at 435, 126 S.Ct. 2594. The Court concluded its discussion with this critical caveat: "We emphasize it is the enormous geographical distance separating the [two] communities, coupled with the disparate needs and interests of these populations — not either factor alone — that renders [the district] noncompact for § 2 purposes." Id.

373. The Black and Latino areas joined in the plaintiffs' illustrative maps are marked by neither "enormous geographical

distance" nor "disparate needs and interests." See id. To the contrary, there is substantial quantitative evidence, supported by lay-witness testimony, that the needs and interests of communities included in the plaintiffs' illustrative plans are similar, including issues of ongoing discrimination. Galveston County's Black and Latino community, therefore, is reasonably compact.

374. The plaintiffs' illustrative plans satisfy the traditional redistricting principle of geographic compactness. Cooper, Fairfax, Dr. Rush, and Dr. Owens testified that the plaintiffs' illustrative maps are geographically compact. Indeed, their illustrative plans have compactness scores comparable to — and, in some cases, better than — the enacted plan. Even Dr. Owens agreed that the illustrative plans are as compact as the enacted plan. Dkt. 232 at 229, 276.

375. While district shape is relevant to determining whether a district satisfies the compactness inquiry, the first Gingles precondition "does not require some aesthetic ideal of compactness"; instead, it simply mandates "that the [minority] population be sufficiently compact to constitute a majority in a single-member district." Houston v. Lafayette County, 56 F.3d 606, 611 (5th Cir. 1995) (quoting Clark v. Calhoun County (Clark I), 21 F.3d 92, 95 (5th Cir. 1994)). Here, the plaintiffs' illustrative plans are reasonably compact, and the court does not need to weigh them against the enacted plan in a "beauty contest." Milligan, 599 U.S. at 21, 143 S.Ct. 1487.

376. All other traditional redistricting principles are satisfied in the plaintiffs' illustrative plans. They are all contiguous and satisfy the equal-population criterion. The Supreme Court has recognized that a redistricting plan for local jurisdictions with a maximum overall population deviation under 10% is consistent with the one-person, one-vote principle. See Evenwel v. Abbott, 578 U.S. 54, 60, 136 S.Ct. 1120, 194 L.Ed.2d 291 (2016). Cooper, Fairfax, and Dr. Rush applied this deviation measure appropriately when assessing their illustrative plans. The population deviation in the plaintiffs' illustrative plans is often lower than or like that of the enacted plan.

377. The plaintiffs' illustrative plans maintain "traditional boundaries" by minimizing municipal and voting-district splits. LULAC, 548 U.S. at 433, 126 S.Ct. 2594 (quotation omitted). The illustrative plans perform better than or similar to the enacted plan in maintaining traditional boundaries.

378. To the extent that the enacted plan had higher compactness scores or lower population deviation than some of the plaintiffs' illustrative plans, such evidence is insufficient to defeat a § 2 claim. See Milligan, 599 U.S. at 19-22, 143 S.Ct. 1487 (finding that the plaintiffs' illustrative plans were reasonably configured, even where the challenged plan arguably performed better on certain traditional redistricting criteria than the illustrative plans); Chen v. City of Houston, 206 F.3d 502, 519 (5th Cir. 2000). It is sufficient that Cooper, Fairfax, and Dr. Rush reliably testified that the illustrative plans comport with traditional redistricting principles such that they are reasonably configured.

379. Finally, the plaintiffs have demonstrated that race did not predominate in drawing the illustrative maps. Cooper, Fairfax, and Dr. Rush credibly testified that neither race nor any single criterion predominated when they drew their illustrative plans. The illustrative plans' compliance with neutral redistricting criteria confirms this, and the defendants have failed to provide any reliable evidence to the contrary.

380. Moreover, the Supreme Court has recently clarified that "there is a

difference between being aware of racial considerations and being motivated by them. The former is permissible; the latter is usually not. That is because redistricting legislatures will almost always be aware of racial demographics, but such race consciousness does not lead inevitably to impermissible race discrimination." Milligan, 599 U.S. at 30, 143 S.Ct. 1487 (cleaned up). Indeed, § 2 "demands consideration of race" because "[t]he question whether additional majority-minority districts can be drawn . . . involves a 'quintessentially race-conscious calculus.'" Id. at 31, 143 S.Ct. 1487 (quoting Abbott, 138 S. Ct. at 2315, Johnson v. De Grandy, 512 U.S. 997, 1020, 114 S.Ct. 2647, 129 L.Ed.2d 775 (1994)). Consideration is not the same as predominance, and none of the defendants' arguments or expert analyses provide any compelling evidence that race predominated in the plaintiffs' illustrative plans.

381. The court concludes that the Black and Latino population in Galveston County is sufficiently large and geographically compact to constitute a majority in a commissioners-court precinct, satisfying the first Gingles precondition.

b. Political Cohesion

382. "[T]here is no simple doctrinal test for the existence of legally significant racial bloc voting," Gingles, 478 U.S. at 58, 106 S.Ct. 2752. But "the most persuasive evidence of inter-minority political cohesion for Section 2 purposes is to be found in voting patterns." Brewer v. Ham, 876 F.2d 448, 453 (5th Cir. 1989) (discussing Campos, 840 F.2d at 1244-45). A 51% majority is "far short of the large majority typically required to show political cohesion." Abbott, 604 F. Supp. 3d at 499. Yet unless the evidence indicates that two groups vote for opposing candidates, the court assesses the cohesion of Black and Latino voters "as a whole" — i.e., as one "minority group" under Gingles — to determine "whether the minority group together votes in a cohesive manner." Campos, 840 F.2d at 1245 (emphasis added).

383. "[L]ay[-]witness testimony concerning cooperation between the minority groups and statistical evidence can be used to prove cohesion." Perez v. Abbott, 274 F. Supp. 3d 624, 669 (W.D. Tex. 2017). A court must undertake "a diligent inquiry into the political dynamics of the particular community" before treating multiple minority groups as a coalition, but "the determinative question is whether [B]lack-supported candidates receive a majority of the [Hispanic] vote [and] whether Hispanic-supported candidates receive a majority of the [Black] vote." Brewer, 876 F.2d at 453.

384. All experts have agreed that general elections are the most probative elections to consider for this case. They also agree that RxC ecological inference is an appropriate method for analyzing the voting patterns of different demographic groups in Galveston County. Using RxC ecological-inference analysis, the undisputed results show that Black and Latino voters frequently prefer the same candidates. When the voter file is used to refine analysis using the BISG method, the results show that even higher estimated percentages of the two groups vote together. The general-election results strongly support a conclusion that a supermajority of Black voters vote for Latino-preferred candidates and vice-versa.

385. Primary elections are relevant but "less probative than general elections for detecting racially polarized voting in an at-large district because general elections present the same candidate pool to every voter, while primary elections limit voters to one party's candidates." Patino, 230 F. Supp. 3d at 694. The court assigns significantly less weight to the statistical analysis of primary elections. Still, the combined results of Drs. Oskooii's and Alford's Democratic-primary

analyses show that Black and Latino voters shared a top-choice candidate in most Democratic primaries.

386. The plaintiffs produced significant evidence of non-statistical cohesion between the Black and Latino communities in Galveston County. This leads the court to conclude that there are distinctive minority interests that tie the two communities together.

387. The statistical analyses from general elections, statistical analyses from primary elections, and non-statistical evidence of cohesion all support the conclusion that Black and Latino voters in Galveston County act as a coalition for purposes of the second Gingles precondition because "[B]lack-supported candidates receive a majority of the [Hispanic] vote [and] Hispanic-supported candidates receive a majority of the [Black] vote." Brewer, 876 F.2d at 453.

388. The undisputed evidence shows that the combined Black and Latino coalition is highly cohesive. The undisputed RxC ecological-inference analysis shows that over 75% of Black and Latino voters have voted for the same candidates in numerous elections. This satisfies the Gingles standard that "a significant number of minority group members usually vote for the same candidates." Gingles, 478 U.S. at 56, 106 S.Ct. 2752.

389. Due to the limited usable data available for local non-partisan elections, the court affords very little weight to them.

390. The plaintiffs' experts' electoral-performance/reconstituted-election analyses show that if this cohesive group constitutes a majority of eligible voters in a county-commissioner precinct, it can elect a candidate of their choice.

391. The court concludes that the county's Black and Latino populations act as a coalition and are politically cohesive.

c. Cannot Elect Candidate of Choice

392. Generally, an Anglo "bloc vote that normally will defeat the combined strength of minority support . . . rises to the level of legally significant [Anglo] bloc voting." Rodriguez, 964 F. Supp. 2d at 757.

393. The defendants do not dispute the statistical evidence of Drs. Barreto and Oskooii showing that more than 85% of Anglos vote cohesively for candidates running in opposition to those supported by more than 85% of Black and Latino voters. They also do not dispute the plaintiffs' electoral-performance/reconstituted-election analyses, which show that the degree of Anglo bloc voting is sufficient to defeat a minority-preferred candidate in each commissioner precinct in the enacted plan.

394. The undisputed evidence shows that Anglo voters in Galveston County vote cohesively and for candidates opposing those supported by a majority of Black and Latino voters. Anglo voters do so at a rate sufficient to defeat the minority-preferred candidate consistently in each of the enacted commissioners-court precincts.

395. The plaintiffs must also show "'that the challenged districting thwarts a distinctive minority vote at least plausibly on account of race.'" Milligan, 599 U.S. at 19, 143 S.Ct. 1487 (quoting Growe, 507 U.S. at 40, 113 S.Ct. 1075). After plaintiffs present statistical evidence showing a racially divergent voting pattern, the burden shifts to defendants to show that there is a race-neutral explanation for the racially divergent voting pattern. Teague v. Attala County, 92 F.3d 283, 290 (5th Cir. 1996); see Gingles, 478 U.S. at 63, 106 S.Ct. 2752; Rodriguez, 964 F. Supp. 2d at 760. Whether the Anglo-preferred elected officials are responsive to minority communities "is intimately related" to the legal significance of bloc voting because bloc voting "'allows those

elected to ignore [minority] interests without fear of political consequences.'" Clements, 999 F.2d at 857 (quoting Gingles, 478 U.S. at 48 n.14, 106 S.Ct. 2752).

396. Contrary to the defendants' contentions, the plaintiffs do not need to initially show that partisan affiliation does not cause divergent voting patterns. See Rodriguez, 964 F. Supp. 2d at 760; Lopez v. Abbott, 339 F. Supp. 3d 589, 603 (S.D. Tex. 2018).

397. By establishing the second and third Gingles preconditions through acceptable statistical evidence and lay testimony, the plaintiffs have shown that racially polarized voting patterns exist in Galveston County elections.

398. The defendants have failed to present reliable or methodologically sound evidence sufficient to dispute that Anglo bloc voting "thwarts" the Black and Latino voting coalition in Galveston County for reasons wholly unconnected to race. The preponderance of the evidence supports the conclusion that the challenged plan "'thwarts a distinctive minority vote' at least plausibly on account of race." Milligan, 599 U.S. at 19, 143 S.Ct. 1487 (quoting Growe, 507 U.S. at 40, 113 S.Ct. 1075).

399. In reaching this conclusion, the court gives considerable weight to the facts that:

• there is a lack of successful minority candidates emerging from Republican primaries,

• there is an extreme degree of Anglo bloc voting for candidates running against minority-preferred candidates,

• minority candidates tend to only be elected from majority-minority areas,

• there are continued racial appeals in Galveston County politics,

• lay witnesses recounted instances of discrimination in Galveston County,

• there are persistent racial disparities across a wide range of measures in Galveston County, and

• Anglo voters in Galveston County overwhelmingly participate in Republican primaries, while Black and Latino voters in Galveston County overwhelmingly participate in Democratic primaries.

400. In sum, the court concludes that the plaintiffs have satisfied all three Gingles preconditions.

2. Step Two — Totality of the Circumstances

401. After examining the Gingles preconditions, courts must "adhere to the Supreme Court's instruction to examine challenged laws and practices in an intensely fact-based and local totality-of-the-circumstances analysis." Veasey v. Abbott, 830 F.3d 216, 261 (5th Cir. 2016) (citing Gingles, 478 U.S. at 36-38, 106 S.Ct. 2752). The totality-of-the-circumstances determination is "flexible" and "guided by factors drawn from the Senate Judiciary Committee report on the 1982 amendments to the Voting Rights Act." Teague, 92 F.3d at 292. These factors include:

(1) the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of a minority group to register, to vote, or otherwise to participate in the democratic process;

(2) the extent to which voting in the elections of the state or political subdivision is racially polarized;

(3) the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;

(4) if there is a candidate slating process, whether the members of the minority group have been denied access to that process;

(5) the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment[,] and health, which hinder their ability to participate effectively in the political process;

(6) whether political campaigns have been characterized by overt or subtle racial appeals; [and]

(7) the extent to which members of the minority group have been elected to public office in the jurisdiction.

S. Rep. at 28-29. Other factors include whether "there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group" and whether "the policy underlying the state or political subdivision's use of such . . . standard, practice, or procedure is tenuous." Id.

402. Ultimately, § 2 violations require "an intensely local appraisal of the electoral mechanism at issue, as well as a searching practical evaluation of the past and present reality." Milligan, 599 U.S. at 19, 143 S.Ct. 1487. Importantly, there is "no requirement that any particular number of factors be proved, or that a majority of them point one way or the other." S. Rep. at 29. The court may instead use its "overall judgment, based on the totality of circumstances and guided by those relevant factors in the particular case," to decide "whether the voting strength of minority voters is . . . 'minimized or canceled out.'" Id. at 29, 143 S.Ct. 1487 n.118. "In short, these factors simply suggest a framework for evidence to be presented at trial which is likely to aid a court's later consideration towards legal conclusions." Johnson, 593 F. Supp. 3d at 600.

403. "'[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 the circumstances.'" Clark I, 21 F.3d at 97 (quoting Jenkins v. Red Clay Consol. Sch. Dist. Bd. of Educ., 4 F.3d 1103, 1135 (3d Cir. 1993)).

404. Here, most of the Senate factors support § 2 liability. See generally PXs-335; 414. Substantial socio-economic differences between Black and Latino residents and Anglo residents in Galveston County create barriers to voting. The presence of racial appeals in recent local political campaigns, relative lack of Black and Latino electoral success, and lack of responsiveness on the part of Galveston County's officials to the needs of the Black and Latino communities further support this finding. Finally, the 2021 redistricting plan's justifications are tenuous and will prevent Galveston County's Black and Latino communities from electing a candidate of their choice.

405. Senate Factor 2: Racially Polarized Voting. The plaintiffs have shown extensive evidence of racially polarized voting in Galveston County. Racial-bloc voting "allows those elected to ignore [minority] interests without fear of political consequences." Rogers v. Lodge, 458 U.S. 613, 623, 102 S.Ct. 3272, 73 L.Ed.2d 1012 (1982). Racial-bloc voting continues to be a reality in Galveston County elections.

406. Senate Factor 3: Voting Practices Enhancing the Opportunity for Discrimination. The plaintiffs have also shown that voting practices exist that may "enhance the possibility that the [c]ounty's map has a dilutive effect." Rodriguez, 964 F. Supp. 2d at 785. Practices deemed to satisfy this factor exist in Galveston County, including voter purges and racially disparate access to polling places. Id. at 780-84. The majority-vote

requirement for primaries provides further support. See Jamison v. Tupelo, 471 F. Supp. 2d 706, 714 (N.D. Miss. 2007) ("Majority[-]vote primaries reduce the chance that a minority candidate will advance to a general election."). The court finds that this factor weighs slightly in the plaintiffs' favor.

407. Senate Factor 5: Effects of Discrimination Hindering Political Participation. The plaintiffs have demonstrated pervasive socio-economic disparities between Galveston County's Black and Latino communities on the one hand, and the Anglo population on the other. The defendants do not contest this evidence. See Dkt. 230 at 280.

408. In addition, Black and Latino voters participate in Galveston County elections at a lower rate than do Anglo voters. Because "courts have recognized that disproportionate educational[,] employment, income levels[,] and living conditions arising from past discrimination tend to depress minority political participation" plaintiffs "need not prove any further causal nexus between their disparate socio-economic status and the depressed level of political participation." S. Rep. at 29 n.114 (citing White v. Regester, 412 U.S. 755, 768, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973), Kirksey v. Bd. of Supervisors, 554 F.2d 139, 145 (5th Cir. 1977)); see also Clements, 999 F.2d at 867 (noting that the Senate Report does not "insist[ ] upon a causal nexus between socioeconomic status and depressed participation").

409. Nevertheless, the plaintiffs' experts have shown that the effects of the education, economic, housing, health, and other racially linked disparities in Galveston County negatively affect voter behavior.

410. Senate Factor 6: Racial Appeals. Evidence of racial appeals in political campaigns, even while "neither frequent nor routine," can "contribute" to a finding that minority voters lack equal opportunities to participate politically. See Patino, 230 F. Supp. 3d at 715. Here, the plaintiffs have demonstrated this factor by showing unrebutted evidence of racial appeals in recent political campaigns. These racial appeals contribute — albeit much less than other factors — to the court's finding that Black and Latino voters do not have equal opportunities for political participation.

411. Senate Factor 7: Minority Election to Public Office. Black and Latino candidates' success in elections "has been slow, slight, and disproportionately lower than" their population share in Galveston County. See Patino, 230 F. Supp. 3d at 715. In analyzing whether "minority voices are heard in a meaningful way during pertinent political decisions, versus being shut out of the process altogether," the court concludes that the enacted plan's elimination of Precinct 3 falls squarely within the latter category. Johnson, 593 F. Supp. 3d at 608; see also Clark v. Calhoun County, 88 F.3d 1393, 1397 (5th Cir. 1996) (Clark II) (holding that lack of minority electoral success in a relevant district has a significant effect on the evaluation of vote-dilution claims). The defendants' reliance on exogenous elections "not involving the particular office at issue" is "less probative than elections involving the specific office." Clark II, 88 F.3d at 1397. This factor strongly supports that Blacks and Latinos do not have an equal opportunity to participate in the political process.

412. Additional Senate Factor: Lack of Responsiveness. Beyond the Senate factors, the totality-of-the-circumstances inquiry "requires a court to ask whether there is a significant lack of responsiveness by elected officials to the minority group members' particularized needs." Patino, 230 F. Supp. 3d at 715-16. This factor weighs in the plaintiffs' favor. Numerous witnesses testified to the lack of responsiveness by the commissioners court in public housing — particularly after Hurricane

Ike — as well as in education and criminal justice. Additionally, the process by which the commissioners court adopted the 2021 redistricting plan demonstrates the county's pattern of "[i]gnoring clear and supported objections about the racially disparate impact of a proposed law," which is probative of a lack of responsiveness to minority concerns. Id. at 717 (citing Veasey, 830 F.3d at 262).

413. Additional Senate Factor: Tenuousness of Policy. Moreover, "[a]long with elected officials' lack of responsiveness to minority needs, a tenuous fit between the expressed policy and the provisions of the law bolsters the conclusion that minorities are not able to equally participate in the political process." Veasey, 830 F.3d at 262-63. Although a jurisdiction "is entitled to make policy choices about when and how it will address various priorities," a policy's rationales are tenuous when the enacted law "fail[s] to correspond in any meaningful way to the legitimate interests [it] claims to have been advancing." Id. at 263.

414. Here, very few members of the public advocated for creating a single coastal precinct. This criterion is further undermined by the existence of several maps that both create a single coastal precinct and maintain a majority-minority precinct. Id. Drawing a coastal precinct neither requires nor justifies cracking the county's minority population.

415. Additional Relevant Factor: Proportionality. Finally, "proportionality is 'a relevant fact in the totality of circumstances.'" LULAC, 548 U.S. at 436, 126 S.Ct. 2594 (quoting De Grandy, 512 U.S. at 1000, 114 S.Ct. 2647); see also De Grandy, 512 U.S. at 1025, 114 S.Ct. 2647 (O'Connor, J., concurring) ("Lack of proportionality is probative evidence of vote dilution."). In a vote-dilution claim, "it is the status of the candidate as the chosen representative of a particular racial group, not the race of the candidate, that is important." Citizens for a Better Gretna v. City of Gretna, 834 F.2d 496, 503 (5th Cir. 1987) (quoting Gingles, 478 U.S. at 69, 106 S.Ct. 2752).

416. For that reason, it is irrelevant that Commissioner Armstrong is Black. His precinct is predominantly Anglo, and several witnesses — including Commissioner Armstrong himself — testified that he would not be the candidate of choice of Black and Latino voters.

417. The county's plan precludes Black and Latino voters from electing a candidate of choice in any commissioners precinct. It does so even though these two groups comprise 38% of the total population in Galveston County. Moreover, it eliminated an existing commissioners precinct where such an opportunity had existed for decades.

418. "Shut Out of the Process Altogether." Another judge in this district recently noted in another Voting Rights Act case that "an underlying concern" in such cases "is whether minority voices are heard in a meaningful way" or are "shut out of the process altogether." Johnson, 593 F. Supp. 3d at 608 (Eskridge, J.). Looking — as this court must — at the totality of the circumstances, it is stunning how completely the county extinguished the Black and Latino communities' voice on its commissioners court during 2021's redistricting.

419. Galveston County was created in 1838. From its founding, it would be 133 years before a Latino, Frank Carmona, was elected to commissioners court. And it would be 150 years before a Black, Wayne Johnson, won a seat. Commissioner Johnson's district, old Precinct 3, would continue to elect the minority community's candidate of choice right up until 2021, when Precinct 3 was summarily carved up and wiped off the map. Blacks' and Latinos' commissioner of choice was always a lonely

voice on the court, but that commissioner's presence — whether it was Wayne Johnson or Stephen Holmes — meant that "minority voices [were] heard in a meaningful way." Id. The result of 2021's redistricting, however, has amounted to Black and Latino voters, as a coalition of like-minded citizens with shared concerns, "being shut out of the process altogether." Id.

420. This is not a typical redistricting case. What happened here was stark and jarring. The commissioners court transformed Precinct 3 from the precinct with the highest percentage of Black and Latino residents to that with the lowest percentage. Dkt. 223 at 42. The circumstances and effect of the enacted plan were "mean-spirited" and "egregious" given that "there was absolutely no reason to make major changes to Precinct 3." Id. at 42-43. Looking at the totality of the circumstances, it was a clear violation of § 2 of the Voting Rights Act. And it must be overturned.

421. The plaintiffs have demonstrated that the totality of the circumstances shows that Black and Latino voters in Galveston County have "less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice." Gingles, 478 U.S. at 63, 106 S.Ct. 2752.

3. Strict Scrutiny

422. Finally, the defendants argue in their post-trial briefing that § 2 is "no longer constitutional" because it is "too temporally distant from the wrongs it was built to remedy." Dkt. 244 at 66. According to them, "the lack of a temporal limit or termination mechanism" in § 2 "no longer satisfies strict scrutiny." Id. at 65. They rely on the Supreme Court's recent decision in Students for Fair Admissions, Inc. v. President & Fellows of Harvard College, which reasoned that race-based admissions programs must have reasonable durational limits. Id. at 65 (citing 600 U.S. 181, 223-28, 143 S.Ct. 2141, 216 L.Ed.2d 857 (2023)). Additionally, they cite obiter dictum from Shelby County, where the Court noted that the Voting Rights Act's coverage formula for preclearance used "decades-old data relevant to decades-old problems, rather than current data reflecting current needs." Id. (citing Shelby County, 570 U.S. at 553, 133 S.Ct. 2612). Finally, the defendants highlight Justice Kavanaugh's concurrence in Milligan, in which he expressly noted that he did not consider this temporal-limit argument. Id. (citing Milligan, 599 U.S. at 45, 143 S.Ct. 1487 (Kavanaugh, J., concurring)). From Kavanaugh's concurrence, the defendants surmise that "a [five-]justice majority might have reached a different result" if the parties preserved such an issue for appeal. Dkt. 244 at 66.

423. In Milligan, Justice Kavanaugh briefly discussed this temporal argument:

Justice [Thomas] notes, however, that even if Congress in 1982 could constitutionally authorize race-based redistricting under § 2, for some period of time, the authority to conduct race-based redistricting cannot extend indefinitely into the future. But Alabama did not raise that temporal argument in this Court, and therefore I would not consider it at this time.

599 U.S. at 45, 143 S.Ct. 1487 (internal citation omitted). Although the two dissenting opinions "raised arguments about the constitutionality of the Gingles framework, neither of them stated that Section 2 of the Voting Rights Act should be deemed unconstitutional." Alpha Phi Alpha Fraternity, Inc. v. Raffensperger, 2023 WL 5674599, at *20 (N.D. Ga. July 17, 2023) (citing Milligan, 599 U.S. at 45-94, 143 S.Ct. 1487 (Thomas, J., dissenting), id. at 95-109, 143 S.Ct. 1487 (Alito, J., dissenting)). As affirmed by a five-justice majority in Milligan, the Gingles framework remains controlling precedent. 424. The court is unconvinced by this temporal-limit argument. The "mere fact that race [is] given some consideration in the districting process, and even the fact that minority-majority districts were intentionally created, does not alone suffice in all circumstances to trigger strict scrutiny." Chen, 206 F.3d at 506 (citing Shaw v. Hunt, 517 U.S. 899, 904-05, 116 S.Ct. 1894, 135 L.Ed.2d 207 (1996)). The Supreme Court has assumed that compliance with § 2 can be narrowly tailored to serve a compelling interest. Bush v. Vera, 517 U.S. 952, 977, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996). And the Fifth Circuit has expressly held that compliance with § 2 "constitutes a compelling government interest" that may narrowly tailor the use of race in restricting plans "at the expense of traditional political concerns no more than is reasonably necessary to remedy the wrong." Clark II, 88 F.3d at 1405-06.

425. Although the defendants speculate that the Voting Rights Act has outlived its usefulness, they have not shown that § 2 does not narrowly remedy the current discriminatory effects in Galveston County's commissioners-court elections. Accordingly, § 2's lack of a temporal limit survives strict scrutiny.

B. Remaining Constitutional Claims

426. In the 1982 amendments to § 2 of the Voting Rights Act, Congress "'repudiated' a requirement that plaintiffs prove intentional discrimination to succeed on a claim that a challenged action violates the Voting Rights Act." Patino, 230 F. Supp. 3d at 718 (quoting Gingles, 478 U.S. at 44, 106 S.Ct. 2752). The amended § 2 "was designed to restore the 'results test' — the legal standard that governed voting discrimination cases" before the Supreme Court's decision in Bolden. Gingles, 478 U.S. at 44 n.8, 106 S.Ct. 2752. Thus, the "right" question following the amendment and Gingles is not whether the challenged mechanism "was adopted or maintained with the intent to discriminate against minority voters" but instead whether it left the plaintiffs without "an equal opportunity to participate in the political process and to elect candidates of their choice." Id. at 44, 106 S.Ct. 2752 (quoting S. Rep. at 28).

427. This court does not need to make findings on intentional discrimination or racial gerrymandering in this case. When plaintiffs succeed on their Gingles claims, the court need not determine the outcome of the intentional-discrimination or racial-gerrymandering claims unless "the remedy to which [the plaintiffs] would be entitled for a discriminatory intent violation is potentially broader than the remedy the district court may fashion for the discriminatory impact violation." Id. at 230 n.11 (citing City of Richmond v. United States, 422 U.S. 358, 378, 95 S.Ct. 2296, 45 L.Ed.2d 245 (1975)).

428. Here, the relief the plaintiffs seek is not broader than that which they are entitled to under § 2. They all seek: (1) declaratory judgments that the enacted plan violates the law; (2) preliminary and permanent injunctions preventing the defendants from calling, holding, supervising, or certifying any elections under the enacted plan; (3) procedures for the commissioners court to adopt a valid redistricting plan; and (4) attorneys' fees and costs. Dkt. 42 at 32-34; Galveston County, No. 3:22-cv-97, ECF No. 38 at 25-26 (May 31, 2022); Dickinson Bay Area Branch NAACP, No. 3:22-cv-117, ECF No. 38 at 38-39 (S.D. Tex. May 25, 2022); see also Dkts. 241, 242-1, 243-1. The requested relief is neither exclusive to intentional-discrimination or racial-gerrymandering claims nor broader than the relief allowed under § 2.

429. The court acknowledges that in their post-trial briefing the NAACP

plaintiffs have asked the court "to determine the appropriateness of retaining jurisdiction under Section 3(c) of the Voting Rights Act" — i.e., instituting a preclearance requirement on Galveston County. Dkt. 242 at 30. Section 3 permits this remedy if the court finds "that violations of the fourteenth or fifteenth amendment justifying equitable relief have occurred within the territory of" the defendant state or political division. 52 U.S.C. § 10302(c). Under § 2 alone, the court could not order § 3(c) preclearance. But the court sees this requested relief as akin to "special damages" — "those which, although resulting from the commission of [a] wrong, are neither such a necessary result that they will be implied by law nor will be deemed within the contemplation of the parties." Hycel, Inc. v. Am. Airlines, Inc., 328 F. Supp. 190, 193 (S.D. Tex. 1971). None of the plaintiffs pleaded for relief under § 3(c) — let alone with particularity — as required by Fed. R. Civ. P. 9(g). Because the plaintiffs never sought this relief with any specificity before or during trial, the court will not entertain such relief now.

430. Therefore, the court declines to reach the plaintiffs' remaining intentional-discrimination and racial-gerrymandering claims.

IV. Relief

431. "When devising a remedy to a § 2 violation, the district court's 'first and foremost obligation . . . is to correct the [§ 2] violation.'" United States v. Brown, 561 F.3d 420, 435 (5th Cir. 2009) (quoting Bone Shirt v. Hazeltine, 461 F.3d 1011, 1022 (8th Cir. 2006)). Any remedy "should be sufficiently tailored to the circumstances giving rise to the § 2 violation." Veasey, 830 F.3d at 269 (quoting Brown, 561 F.3d at 435). When possible, courts "should respect a legislature's policy objectives when crafting a remedy," even "when some aspect of the underlying law is unenforceable." Id.

432. The court recognizes that its review of the commissioners court's redistricting process "represents a serious intrusion on the most vital of local functions." Miller, 515 U.S. at 915, 115 S.Ct. 2475. "There are times when a court might give a . . . legislature an opportunity to cure the infirmities . . . before permitting the district court to fashion a remedy." Veasey, 830 F.3d at 269. Generally, courts should "offer governing bodies the first pass at devising" § 2 remedies. Id. (quoting Brown, 561 F.3d at 435); see also Wise v. Lipscomb, 437 U.S. 535, 540, 98 S.Ct. 2493, 57 L.Ed.2d 411 (1978) ("[R]edistricting and reapportioning legislative bodies is a legislative task which the federal courts should make every effort not to pre-empt."). But "when it is not practicable to permit a legislative body this opportunity because of an impending election, it becomes the unwelcome obligation of the federal court to devise and impose a remedy pending later legislative action." Veasey, 830 F.3d at 270 (cleaned up) (quoting Wise, 437 U.S. at 540, 98 S.Ct. 2493).

433. Galveston County's 2024 elections are imminent. The commissioners court must have an election map in place before the statutory opening date for candidate filing on November 11, 2023. See Tex. Elec. Code § 172.023(b). As established, the enacted plan violates § 2 of the Voting Rights Act, and so the county cannot use this map for future elections.

434. To balance the commissioners court's control over the redistricting process against the need for a plan that conforms with § 2 for the 2024 election, the court will allow the defendants until October 20, 2023, to file a redistricting plan with supporting expert analysis establishing that it adheres to § 2 and has at least one majority-minority precinct. The plaintiffs may file consolidated objections to the defendants' plan with proposed alternative

plans and supporting expert analysis by October 27, 2023. The court will conduct an in-person hearing on November 1, 2023, at 2 p.m. to decide which plan to order into effect.

435. If the defendants fail or prefer not to submit a revised plan, they are ordered to implement the illustrative plan presented by Anthony Fairfax on August 10, 2023 (PX-339), on or before November 1, 2023, and use that plan for all future elections until the commissioners court adopts a different plan.

V. Conclusion

The court finds that the 2021 commissioners-court precinct map adopted by the Galveston County Commissioners Court on November 12, 2021, violates § 2 of the Voting Rights Act. Regardless of the intent or motivation of the commissioners court, the enacted plan denies Black and Latino voters an equal opportunity to participate in the political process and to elect a candidate of their choice. The court will enter a separate order conforming to these findings and conclusions.