SINGLETON v. ALLEN (2025)

INJUNCTION AND ORDER FINDINGS OF FACT AND CONCLUSIONS OF LAW

These congressional redistricting cases returned to this Court for trial after the Supreme Court of the United States affirmed a preliminary injunction we entered. See Allen v. Milligan, 599 U.S. 1, 10, 17 (2023). These cases are three of four cases pending in the Northern District of Alabama that allege that Alabama's electoral maps are racially discriminatory in violation of the United States Constitution and/or dilute the votes of Black Alabamians in violation of the Voting Rights Act of 1965, 52 U.S.C. § 10301: Singleton v. Allen, Case No. 2:21-cv-1291-AMM (challenges the congressional map on constitutional and statutory grounds), Milligan v. Allen, Case No. 2:21-cv-1530-AMM (challenges the congressional map on constitutional and statutory grounds), and Caster v. Allen, Case No. 2:21-cv-1536-AMM (challenges the congressional map on statutory grounds).¹

Singleton and Milligan are before this three-judge Court, and Caster is before Judge Manasco sitting alone. Although there are differences in the Plaintiffs’ theories of liability, all Plaintiffs challenge districts in South and Central Alabama, with a focus on Alabama's Black Belt and Gulf Coast regions. Likewise, all Plaintiffs request an injunction barring Alabama Secretary of State Wes Allen from conducting elections according to the electoral map for Alabama's seven seats in the United States House of Representatives that the Alabama Legislature (“the Legislature”) passed after the Supreme Court's ruling in Allen (“the 2023 Plan”).

The preliminary injunction that the Supreme Court affirmed prohibited the use of Alabama's previous districting plan (“the 2021 Plan”). Milligan Doc. 107.² The 2021 Plan included only one majority-Black congressional district — District 7, which became a majority-Black district in 1992 when a federal court drew it that way in a ruling that was affirmed by the Supreme Court of the United States. Wesch v. Hunt, 785 F. Supp. 1491, 1497–1500 (S.D. Ala. 1992) (three-judge court), aff'd sub nom. Camp v. Wesch, 504 U.S. 902 (1992), and aff'd sub nom. Figures v. Hunt, 507 U.S. 901 (1993). In the 1992 election under the court-ordered map, District 7 elected Alabama's first Black Congressman in over 90 years. District 7 remains majority-Black to this day and in every election since 1992 has elected a Black Democrat. No other Alabama congressional district has elected a Black candidate in approximately 150 years, until District 2 elected Shomari Figures in 2024 under a court-ordered map that we imposed after the Legislature failed to remedy a likely violation of Section Two of the Voting Rights Act.

We issued that preliminary injunction with the benefit of a seven-day hearing and an extensive record: we heard live testimony from seventeen witnesses (including eleven experts); received more than 1,000 pages of briefing; reviewed more than 350 exhibits (including reports and rebuttal reports from every expert); and considered joint stipulations of fact that spanned seventy-five pages. Milligan Doc. 107 at 4.³ Forty-three able lawyers appeared in those proceedings, and the hearing transcript spanned nearly 2,000 pages. Id.

We found that the Milligan Plaintiffs were substantially likely to establish that the 2021 Plan violated Section Two of the Voting Rights Act (“Section Two”) by unlawfully diluting the votes of Black Alabamians, Judge Manasco found the same for the Caster Plaintiffs, and we said that the issue was not a close call. Milligan Doc. 107 at 4, Part V.B, 195; Caster Doc. 101 at 5. Because we granted relief under Section Two, we reserved ruling on the constitutional claims in Milligan and Singleton, invoking the longstanding canon of constitutional avoidance. Milligan Doc. 107 at 7; Lyng v. Nw. Indian Cemetery Protective Ass'n, 485 U.S. 439 (1988).

We ordered that under the Voting Rights Act and Supreme Court precedent, “the appropriate remedy [wa]s a congressional redistricting plan that includes either an additional majority-Black congressional district, or an additional district in which Black voters otherwise have an opportunity to elect a representative of their choice.” Milligan Doc. 107 at 5. And we ordered that as a practical reality, based on extensive evidence of intensely racially polarized voting in Alabama, any remedial plan would “need to include two districts in which Black voters either comprise a voting-age majority or something quite close to it.” Milligan Doc. 107 at 6.

The Secretary and legislative defendants, Senator Jim McClendon and Representative Chris Pringle, who co-chaired Alabama's Permanent Legislative Committee on Reapportionment (“the Legislators” and “the Committee”), appealed to the Supreme Court. Allen, 599 U.S. at 16–17. The Supreme Court stayed our injunction, so Alabama used the 2021 Plan for the 2022 congressional election.

In June 2023, the Supreme Court affirmed the preliminary injunction. See id. at 17. The Supreme Court “s[aw] no reason to disturb [our] careful factual findings” and found no “basis to upset [our] legal conclusions” because we “faithfully applied [Supreme Court] precedents and correctly determined that, under existing law, [the 2021 Plan] violated” Section Two. Id. at 23. And the Supreme Court rejected the State's request to overturn the legal standard for Section Two claims that the Supreme Court announced nearly forty years ago in Thornburg v. Gingles, 478 U.S. 30 (1986) and that federal courts have applied continuously since. Id. at 23–24.

Milligan and Caster thus returned to us for remedial proceedings. The Secretary and Legislators (together, “the State”) asked us to delay proceedings to allow the Legislature time to enact a new congressional districting plan. Milligan Doc. 166. The Court granted the State's request in deference to the role state legislatures play in redistricting. In July 2023, the Legislature passed and Governor Kay Ivey signed into law the 2023 Plan.

Just like the 2021 Plan that we enjoined, the 2023 Plan includes only one majority-Black district: District 7. The congressional district with the next highest Black share of the voting-age population (“BVAP”) in the 2023 Plan is District 2, with a BVAP of just 39.9%. All Plaintiffs requested another injunction, so we held another hearing. Singleton Doc. 147; Milligan Docs. 200, 265; Caster Doc. 179.

At that hearing, the State conceded that the 2023 Plan does not include an additional opportunity district. The State asserted that notwithstanding our order and the Supreme Court's affirmance, the Legislature was not required to include an additional opportunity district in the 2023 Plan. Aug. 14, 2023 Tr. 159–64. The Alabama Solicitor General argued that our statement in our order “that the appropriate remedy for the ․ likely violation that we found would be an additional opportunity district” did not have any relevance to the 2023 Plan. Id. at 75. Rather, the Solicitor General asserted that “the Legislature could enact a new map that was consistent with [our] findings and conclusions without adding a second opportunity district.” Id.

The Legislature's conduct and that concession thrust this case into an unusual posture: we are not aware of any other case in which a state legislature — faced with a federal court order declaring that its electoral plan unlawfully dilutes minority votes and requiring a remedial plan that provides an additional opportunity district — responded with a plan that the state concedes does not provide that district.

Based on that concession and the ample evidentiary record, we issued a second preliminary injunction. Milligan Doc. 272. We enjoined the Secretary from using the 2023 Plan because it does not remedy the likely Section Two violation that we found and the Supreme Court affirmed, and in the alternative, because we found the Milligan Plaintiffs were substantially likely to establish anew that the 2023 Plan violates Section Two, just as the 2021 Plan did. See generally id. Judge Manasco again ordered the same relief for the Caster Plaintiffs. Caster Doc. 223 at 4.

We again ordered that under the Voting Rights Act and Supreme Court precedent, the appropriate remedy is an additional district in which Black voters have an opportunity to elect a representative of their choice. Milligan Doc. 272 at 6–7. And we again invoked the canon of constitutional avoidance and reserved ruling on the constitutional claims. Id. at 8.

We directed the Special Master we had appointed to prepare and propose three remedial maps for us to consider. Milligan Doc. 273 at 6. We explained that we were “deeply troubled” that the State enacted a map that it readily admits does not provide the remedy we said the law requires, and “disturbed by the evidence that the State delayed remedial proceedings but ultimately did not even nurture the ambition to provide that remedy.” Milligan Doc. 272 at 8.

The Secretary (but not the Legislators ⁴ ) again appealed to the Supreme Court and sought a stay. Milligan Docs. 274, 275, 276, 281. We denied a stay because federal law required the creation of an additional opportunity district without further delay. Milligan Doc. 289 at 5. The Secretary sought a stay from the Supreme Court, which summarily denied the request with no noted dissents. See Allen v. Milligan, 144 S. Ct. 476 (2023) (mem.). The Secretary dismissed his appeals.⁵

The Special Master solicited proposals, prepared plans, and recommended three plans to us. See generally In re Redistricting 2023, No. 2:23-mc-01181-AMM (N.D. Ala.) (“Redistricting”); Milligan Docs. 295–96, 301–05; Caster Doc. 248; Redistricting Docs. 48–49. After another hearing, we ordered Secretary Allen to administer Alabama's 2024 election using the plan the Special Master recommended called “Remedial Plan 3” (“the Special Master Plan”). Milligan Doc. 311. The Special Master Plan satisfied all constitutional and statutory requirements while hewing as closely as possible to the Legislature's 2023 Plan. See id.

The Special Master Plan includes two districts in which Black Alabamians have a fair opportunity to elect a representative of their choice: District 7, where 51.9% of the voting-age population is Black, and District 2, where 48.7% of the voting-age population is Black. Id. at 41. In the 2024 election, District 7 voters re-elected Congresswoman Terri Sewell, and District 2 voters elected Congressman Figures, both of whom are Black.

Additionally, the Special Master Plan, which was prepared race-blind, provides compelling evidence that two reasonably configured Black-opportunity districts easily can be drawn in Alabama. The Special Master explained clearly that the Court's appointed cartographer, Mr. David Ely, did not consider race when he prepared plans:

The Special Master's proposed remedial plans are neither prohibited racial gerrymanders nor intentionally discriminatory․ [W]hile the Special Master confirmed that Black residents had an opportunity to elect candidates of their choice through an election performance analysis, the boundaries within the recommended remedial plans were not drawn on the basis of race. In fact, the Special Master's cartographer, Mr. Ely, did not display racial demographic data while drawing districts or examining others’ proposed remedial plans within the mapping software, Maptitude. Instead, Mr. Ely relied on other characteristics and criteria, such as preserving the Black Belt community of interest, restoring counties that had been split, and preserving precincts and municipalities to the extent possible.

Milligan Doc. 295 at 36.

The Plaintiffs now request a judgment that the 2023 Plan violates federal law and a permanent injunction barring Secretary Allen from conducting elections with that Plan and requiring him to conduct them with a court-ordered plan. Singleton Doc. 229 at 46; Milligan Doc. 329 ¶ 206; Caster Doc. 271 at 43; Milligan Doc. 485 at 425–27, ¶¶ 1150–52. Additionally, the Milligan Plaintiffs request under Section 3(c) of the Voting Rights Act that the Court bail Alabama back into federal preclearance for congressional redistricting “until 60 days after the Alabama Legislature enacts a congressional plan under the 2030 census or a period of approximately seven years.” Milligan Doc. 485 at 436, ¶ 1173; Milligan Doc. 329 at 77.

We again consider an extensive record. The coordinated trial of these cases consumed eleven trial days, and the transcript spans more than 2,600 pages. We heard live testimony from twenty-three witnesses (including thirteen experts); received reports and rebuttal reports from every expert; received testimony by designation for twenty-eight additional witnesses (either from depositions in these cases, or from live testimony in the state Senate redistricting trial that occurred before Judge Manasco in November 2024); considered stipulated facts spanning thirty-nine pages; processed more than 790 putative exhibits; and received more than 840 pages of proposed findings and conclusions after trial. We again have the assistance of numerous able counsel, with forty lawyers and eleven support staff participating in trial. And under Federal Rule of Civil Procedure 65(a)(2), we continue to have the benefit of evidence adduced in the first two preliminary injunction proceedings. It is difficult for us to imagine a more comprehensive record.

Based on the findings of fact and conclusions of law explained at length below, we see the same clear result on the Section Two claims now that we saw in 2022 and again in 2023. More particularly, we conclude that the Plaintiffs established that the 2023 Plan violates Section Two, and that they have established each part of the controlling legal standard, including that: (1) as a group, Black Alabamians are sufficiently numerous and geographically compact to constitute a voting-age majority in a second reasonably configured district; (2) voting in the challenged districts is intensely and extremely racially polarized, such that Black voters are (nearly always) politically cohesive and (3) White voters ordinarily (nearly invariably) vote as a bloc to defeat Black-preferred candidates; and (4) under the totality of the circumstances in Alabama today, including the factors that the Supreme Court has instructed us to consider, Black voters have less opportunity than other Alabamians to elect candidates of their choice to Congress. The long and short of it is that the 2023 Plan unlawfully dilutes Black voting strength by consigning it to one majority-Black district despite Alabama's Black population plainly being numerous and compact enough, and voting in Alabama racially polarized enough, to readily support an additional opportunity district under all the circumstances in Alabama today.

We repeat — now for the third time — that these Section Two claims are not a close call. Numerosity is undisputed, extensive evidence establishes reasonable compactness, and there is no serious dispute that voting is intensely racially polarized with extreme consequences: Black candidates have enjoyed zero success in statewide elections in Alabama since 1994 (when a single Black person was elected to the Alabama Supreme Court after a previous appointment), and only three Black candidates have ever been elected to any statewide office since Reconstruction. Similarly, Black candidates have enjoyed near-zero success in legislative elections outside of opportunity districts: thirty-two of the thirty-three Black Alabamians currently serving in the 140-person Legislature were elected from majority-Black districts created to comply with federal law. And as we explain below, substantial evidence establishes that under all the circumstances in Alabama today, Black Alabamians have less opportunity than other Alabamians to elect representatives of their choice.

We also repeat — for the third time — that because the Plaintiffs prevail under the Voting Rights Act, the appropriate remedy is a districting plan that includes either an additional majority-Black district, or an additional district in which Black voters otherwise have an opportunity to elect a representative of their choice. As the record makes abundantly clear, the necessary remedial district is not difficult to draw – it just requires splitting one of Alabama's 67 counties (Mobile County) that the Legislature would prefer to keep whole.

As we said once before in this litigation, “[t]he Voting Rights Act does not provide a leg up for Black voters — it merely prevents them from being kept down with regard to what is arguably the most ‘fundamental political right,’ in that it is ‘preservative of all rights’ — the right to vote.” Milligan Doc. 272 at 187 (quoting Democratic Exec. Comm. of Fla. v. Lee, 915 F.3d 1312, 1315 (11th Cir. 2019)).

Additionally, this time we decide the Milligan Plaintiffs’ constitutional claim that the Legislature intentionally discriminated against Black Alabamians when it passed the 2023 Plan. The canon of constitutional avoidance that previously compelled us to reserve on the claim of intentional discrimination now requires us to decide it: because the Milligan Plaintiffs request bail-in under Section 3(c) of the Voting Rights Act, and because that relief would be available only in connection with their constitutional claim, a decision on that claim could entitle them to relief beyond the relief to which they are entitled under Section Two. See Lyng, 485 U.S. at 446.

As we explain below, despite our searching review of all the evidence before us — much of it directly from the Legislators and Legislature, none of it in dispute — try as we might, we cannot understand the 2023 Plan as anything other than an intentional effort to dilute Black Alabamians’ voting strength and evade the unambiguous requirements of court orders standing in the way. After we and the Supreme Court ruled that the 2021 Plan, with only one majority-Black district, likely had the unlawful discriminatory effect of diluting Black Alabamians’ votes, the Legislature deliberately enacted another Plan that it concedes lacks the second Black-opportunity district we said was required. This amounted to intentional racial discrimination in violation of the Fourteenth Amendment's Equal Protection guarantee.

We are struck by the unusual corpus of undisputed evidence that confirms the obvious inference from the Legislature's conduct. We have found no other case that involves not only (1) a Legislature's admission that its remedial plan purposefully lacks the remedial district the court plainly required, but also (2) novel legislative findings enumerated in that plan that (a) are mathematically impossible to satisfy if the remedial district is drawn, (b) define and exalt one community of interest (the Gulf Coast) that plan serves at the expense of that remedial district and other longstanding communities of interest, (c) do not mention, let alone describe, any communities of interest in areas of the state that are not at issue in pending litigation, and (d) eliminate the express requirement that a plan not dilute minority voting strength; as well as (3) contemporaneous public statements by legislative leadership about their strategy to persuade the Supreme Court to change its view about the Legislature's Section Two violation, and (4) testimony by key legislators that the legislative findings were drafted surreptitiously, in the dead of night, at the very last minute, and without any input from either the Senate Co-Chair or House Co-Chair of the Legislature's Reapportionment Committee.

We also are struck by the candid admission at trial by diligent counsel for the State that when the Legislature passed the 2023 Plan without adding a second opportunity district, the Legislature “may have been hoping” to “find another argument” to persuade this Court and/or the Supreme Court that our orders were wrong. Tr. 2649. If we harbored any concern or doubt that we had misunderstood that the Legislature deliberately ignored our order because it wanted another bite at the apple in the Supreme Court, that acknowledgment resolved it.

This record thus leaves us in no doubt that the purpose of the design of the 2023 Plan was to crack Black voters across congressional districts in a manner that makes it impossible to create two districts in which they have an opportunity to elect candidates of their choice, and thereby intentionally perpetuate the discriminatory effects of the 2021 Plan. So we observe that although the success of the Milligan Plaintiffs’ claim of intentional discrimination is unusual, we also do not regard it as a particularly close call.

The Legislature protests that it acted in good faith, but if this record is insufficient to rebut the strong presumption of legislative good faith, then we doubt that the presumption is ever rebuttable. The Legislature knew what federal law required and purposefully refused to provide it, in a strategic attempt to checkmate the injunction that ordered it. It would be remarkable — indeed, unprecedented — for us to hold that a state legislature that purposefully ignored a federal court order acted in good faith. It would be shocking for us to hold that a state legislature that intentionally ignored a federal court order for the purpose of (again) diluting minority votes acted in good faith. And it would be unthinkable for us to hold that a state legislature that purposefully took calculated steps to make a court-required remedy impossible to provide, for the purpose of entrenching minority vote dilution, acted in good faith. Although it is robust, the legal presumption of legislative good faith cannot give the Legislature a free pass for its purposeful attempt to rob Black Alabamians of an equal opportunity under the law to elect candidates of their choice.

Because the Singleton Plaintiffs do not request bail-in, no decision is necessary on the constitutional claims of the Singleton Plaintiffs, so we apply the canon of constitutional avoidance and do not decide those claims. See Lyng, 485 U.S. at 446.

Accordingly, we ENJOIN Secretary Allen, and his successors in office, from conducting any elections according to Alabama's 2023 Plan, and we DECLARE that the 2023 Plan violates both Section Two of the Voting Rights Act and the Fourteenth Amendment to the United States Constitution.

Although we do not face the same time exigencies we did in 2022 and 2023, this Court must conduct remedial proceedings expeditiously in light of state-law deadlines applicable to Alabama's next congressional election — Alabama Code Section 17-13-5(a) effectively establishes a deadline of January 30, 2026 for candidates to qualify with major political parties to participate in the 2026 primary election for the United States Congress. We will address during remedial proceedings the Milligan Plaintiffs’ application to bail Alabama back into federal preclearance for future congressional districting under Section 3(c) of the Voting Rights Act. To facilitate the timely scheduling of remedial proceedings, a status conference is SET for all parties on Wednesday, May 28, 2025 at 12:00 p.m. Central Daylight Time. The conference will occur by Zoom and login information will be sent to the parties closer to that time.

***

We reach these conclusions with great reluctance and dismay and even greater restraint — only after another exhaustive analysis of another extensive record under well-developed legal standards, as Supreme Court precedent in these very cases instructs. We do not intrude lightly into a process ordinarily and properly reserved for the Alabama Legislature, but forty years’ worth of Supreme Court case law and forty years’ worth of statutory instructions from Congress compel this result in this case.

We emphasize that we remain deeply disturbed that the State purposefully enacted a map that the State readily admits does not provide the required remedy for the vote dilution that we clearly found. We also emphasize our concern about the State's assertion that in response to any injunction we may issue, it is free to repeat its checkmate move. We are troubled by the State's view that even if we enter judgment for the Plaintiffs after a full trial, the State remains free to make the same checkmate move yet again — and again, and again, and again.

We reject in the strongest possible terms the State's attempt to finish its intentional decision to dilute minority votes with a veneer of regular legislative process. On the rare occasion that federal law directs federal courts to intrude in a process ordinarily reserved for state politics, there is nothing customary or appropriate about a state legislature's deliberate decision to ignore, evade, and strategically frustrate requirements spelled out in a court order.

This is not the first time the Alabama Legislature has purposefully refused to satisfy a federal court order about redistricting even after the Supreme Court affirmed that order. See generally Sims v. Baggett, 247 F. Supp. 96 (M.D. Ala. 1965) (three-judge court: Rives, Thomas, and Johnson, JJ.) (per curiam). We hope it will be the last time.

The Legislature has raised the stakes of this litigation well beyond redistricting. In a case all too familiar to Alabama, the Supreme Court explained decades ago that decisions to ignore court orders are intolerable in our system of ordered liberty even when they are undertaken in unassailable good faith and for purely “righteous” purposes. See Walker v. City of Birmingham, 388 U.S. 307, 321 (1967). “[R]espect for judicial process,” the Supreme Court explained, “is a small price to pay for the civilizing hand of law, which alone can give abiding meaning to constitutional freedom.” Id.

Finally, we cannot help but observe the hazards of the Legislature's conduct that it apparently overlooked. We do not diminish the argument that race-based redistricting under Section Two cannot last forever. But it seems painfully obvious to us that the State's decision to purposefully dilute the votes of Black Alabamians, particularly after exhausting its appellate rights for a preliminary injunction entered under Section Two, flies in the face of its position that Section Two has outlived the purpose Congress intended.

Likewise, we do not diminish the substantial improvements Alabama has made in its official treatment of Black Alabamians in recent decades. Yet we cannot reconcile the State's intentional decision to discriminate in drawing its congressional districts with its position that Alabama has finally closed out its repugnant history of official discrimination involving voting rights.

The 2020 redistricting cycle in Alabama — the first cycle in 50 years that Alabama has been free of the strictures of federal preclearance — did not have to turn out this way. We wish it had not, but we have eyes to see the veritable mountain of evidence that it did.

TABLE OF CONTENTS

I. BACKGROUND․––––

A. Relevant Federal Laws․––––

B. The 2021 Plan and First Preliminary Injunction (2021–2022)․––––

C. The Supreme Court's Ruling (2023)․––––

1. Controlling Precedent․––––

2. Part III-B-1 of the Chief Justice's Opinion․––––

3. Justice Kavanaugh's Concurrence․––––

4. The Dissents․––––

D. The 2023 Plan․––––

E. Remedial Proceedings and the Second Preliminary Injunction (2023)․––––

F. The Special Master Plan․––––

G. The 2024 Election․––––

H. Trial (2025)․––––

I. Redistricting Litigation in Alabama․––––

1. State Legislative Redistricting․––––

2. Congressional Redistricting․––––

3. The 2023 Special Session․––––

J. Claims and Defenses․––––

1. Singleton Plaintiffs․––––

2. Milligan Plaintiffs․––––

3. Caster Plaintiffs․––––

4. The State․––––

II. STANDARD OF REVIEW․––––

III. APPLICABLE LAW․––––

A. Section Two Claims․––––

B. Constitutional Claims․––––

IV. ANALYSIS – VOTING RIGHTS ACT․––––

A. Milligan Plaintiffs’ Arguments․––––

1. Gingles I – Numerosity and Reasonable Configuration (Dr. Duchin)․––––

2. Gingles II and III – Racially Polarized Voting (Dr. Liu)․––––

3. The Senate Factors․––––

4. Fact Witnesses at Trial․––––

5. Designated Deposition Testimony․––––

6. Testimony from State Senate Redistricting Trial․––––

B. Caster Plaintiffs’ Arguments․––––

1. Gingles I – Numerosity and Reasonable Configuration (Mr. Cooper)․––––

2. Gingles II and III – Racially Polarized Voting (Dr. Palmer)․––––

3. The Senate Factors․––––

4. Fact Witnesses at Trial․––––

5. Designated Deposition Testimony․––––

C. Singleton Plaintiffs’ Arguments․––––

1. Expert Testimony about the Senate Factors․––––

2. Fact Witness at Trial․––––

3. Designated Deposition Testimony․––––

D. The State's Defenses․––––

1. Gingles I – Reasonable Configuration․––––

2. Gingles II and III – Racially Polarized Voting․––––

3. The Senate Factors․––––

4. Fact Witnesses Called at Trial․––––

5. Testimony from State Senate Redistricting Trial․––––

6. Designated Deposition Testimony․––––

E. Attacks on Section Two․––––

1. Constitutionality of Race-Based Redistricting․––––

2. Private Right of Action․––––

V. FINDINGS OF FACT AND CONCLUSIONS OF LAW – VOTING RIGHTS ACT․––––

A. The Plaintiffs establish a Section Two violation․––––

1. Gingles I – Numerosity․––––

2. Gingles I – Reasonable Configuration․––––

3. Gingles II and III – Racially Polarized Voting․––––

4. The Senate Factors․––––

B. Section Two Is Privately Enforceable․––––

1. Text of Section Two․––––

2. Section Two Precedents․––––

C. Section Two Is Constitutional․––––

VI. ANALYSIS – CONSTITUTIONAL CLAIMS․––––

A. The Singleton Plaintiffs’ Arguments․––––

1. Racial Gerrymandering․––––

2. Intentional Discrimination․––––

B. The Milligan Plaintiffs’ Arguments․––––

1. Historical Background of the 2023 Plan․––––

2. Sequence of Events Leading to the Passage of the 2023 Plan․––––

3. Substantive and Procedural Departures from the Norm․––––

4. Direct Evidence of Intent and Other Contemporaneous Statements․––––

C. The State's Arguments․––––

1. Singleton Plaintiffs’ Racial Gerrymandering Claim․––––

2. Singleton and Milligan Plaintiffs’ Intentional Discrimination Claims․––––

VII. FINDINGS OF FACT AND CONCLUSIONS OF LAW — CONSTITUTION․––––

A. Avoidance Canon․––––

B. Presumption of Legislative Good Faith․––––

C. Arlington Heights and the Eleventh Circuit's Additional Factors․––––

1. Historical Background of the 2023 Plan․––––

2. Sequence of Events Leading to the Passage of the 2023 Plan․––––

3. Substantive and Procedural Departures from the Norm, and Legislative History of the 2023 Plan․––––

4. Whether Disparate Impact Was the Natural and Foreseeable Consequence of the 2023 Plan․––––

5. Direct Evidence of Intent and Contemporaneous Statements․––––

6. Eleventh Circuit's Additional Factors: Knowledge of Disparate Impact and Availability of Less Discriminatory Alternatives․––––

D. The State's Efforts To Dispel the Inference․––––

VIII. REMEDY․––––

IX. EVIDENTIARY RULINGS․––––

APPENDIX A – COMMITTEE GUIDELINES (2021 and 2023)․––––

APPENDIX B – LEGISLATIVE FINDINGS (2023 PLAN/SB-5)․––––

APPENDIX C – CONGRESSIONAL MAP (2023 PLAN)․––––

APPENDIX D – COMMUNITY OF INTEREST PLAN and LIVINGSTON PLANS 1 & 2․––––

APPENDIX E – EXCERPTED FIGURES FROM MR. COOPER․––––

I. BACKGROUND

We divide our discussion of the background of these cases into ten parts.

A. Relevant Federal Laws

Article I, § 2, of the Constitution requires that Members of the House of Representatives “be apportioned among the several States ․ according to their respective Numbers” and “chosen every second Year by the People of the several States.” U.S. Const. art. I, § 2. Each state's population is counted every ten years in a national census, and state legislatures rely on census data to apportion each state's congressional seats into districts.

Redistricting “is primarily the duty and responsibility of the State[ ].” Allen, 599 U.S. at 29 (quoting Abbott v. Perez, 585 U.S. 579, 603 (2018)). “[F]ederal-court review of districting legislation represents a serious intrusion on the most vital of local functions,” and when “assessing the sufficiency of a challenge to a districting plan, a court must be sensitive to the complex interplay of forces that enter a legislature's redistricting calculus.” Abbott, 585 U.S. at 603 (quoting Miller v. Johnson, 515 U.S. 900, 915–16 (1995)) (internal quotation marks omitted).

Redistricting must comply with federal law as set forth in the Constitution and federal statutes. Reynolds v. Sims, 377 U.S. 533, 554–60 (1964); Wesberry v. Sanders, 376 U.S. 1, 6 (1964). “[F]ederal law impose[s] complex and delicately balanced requirements regarding the consideration of race” in redistricting. Abbott, 585 U.S. at 585.

On the one hand, the Equal Protection Clause “restrict[s] the use of race” in redistricting. Id. That Clause “forbids ‘racial gerrymandering,’ that is, intentionally assigning citizens to a district on the basis of race without sufficient justification.” Id. at 585–86 (quoting Shaw v. Reno, 509 U.S. 630, 641 (1993)). That Clause “also prohibits intentional ‘vote dilution,’ ” which is “invidiously ․ minimiz[ing] or cancel[ing] out the voting potential of racial or ethnic minorities.” Id. at 586 (quoting City of Mobile v. Bolden, 446 U.S. 55, 66–67 (1980) (plurality opinion) (alterations in original)).

On the other hand, “compliance with the Voting Rights Act of 1965 pulls in the opposite direction: It often insists that districts be created precisely because of race.” Id. (citation omitted). Section Two provides:

(a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 10303(f)(2) of this title, as provided in subsection (b).

(b) A violation of subsection (a) is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.

52 U.S.C. § 10301.

“The essence of a [Section Two] claim ․ is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by black and white voters.” Allen, 599 U.S. at 17 (quoting Thornburg v. Gingles, 478 U.S. 30, 47 (1986)). That occurs “when a State's electoral structure operates in a manner that ‘minimize[s] or cancel[s] out the[ir] voting strength,’ ” rendering “an individual ․ disabled from ‘enter[ing] into the political process in a reliable and meaningful manner’ ‘in the light of past and present reality, political and otherwise.’ ” Id. at 25 (first quoting Gingles, 478 U.S. at 47 and then quoting White v. Regester, 412 U.S. 755, 767 (1973) (alterations in original)). “A district is not equally open, in other words, when minority voters face—unlike their majority peers—bloc voting along racial lines, arising against the backdrop of substantial racial discrimination within the State, that renders a minority vote unequal to a vote by a nonminority voter.” Id.

“[A] plaintiff may allege a § 2 violation in a single-member district if the manipulation of districting lines fragments [cracks] politically cohesive minority voters among several districts or packs them into one district or a small number of districts, and thereby dilutes the voting strength of members of the minority population.” Shaw v. Hunt (Shaw II), 517 U.S. 899, 914 (1996).

Under Gingles, a plaintiff asserting a claim of vote dilution under Section Two “must satisfy three ‘preconditions.’ ” Allen, 599 U.S. at 18 (quoting Gingles, 478 U.S. at 50). “First, the ‘minority group must be sufficiently large and [geographically] compact to constitute a majority in a reasonably configured district.’ ” Id. (quoting Wis. Legislature v. Wis. Elections Comm'n, 595 U.S. 398, 402 (2022) (per curiam)). A district is “reasonably configured” when “it comports with traditional districting criteria, such as being contiguous and reasonably compact.” Id. “Second, the minority group must be able to show that it is politically cohesive.” Id. (quoting Gingles, 478 U.S. at 51). “And third, ‘the minority must be able to demonstrate that the white majority votes sufficiently as a bloc to enable it ․ to defeat the minority's preferred candidate.” Id. (quoting Gingles, 478 U.S. at 51).

If a plaintiff establishes the three preconditions, the plaintiff must then “show, under the ‘totality of circumstances,’ that the political process is not ‘equally open’ to minority voters.” Id. (quoting Gingles, 478 U.S. at 45–46); see Bartlett v. Strickland, 556 U.S. 1, 11–12 (2009) (plurality opinion). We have been instructed by the Supreme Court to use the factors identified in the Senate Judiciary Committee Report accompanying the 1982 amendments to the Voting Rights Act to assess the totality of the circumstances. See Allen, 599 U.S. at 18 (citing Gingles, 478 U.S. at 36–38, 45–46); see also infra Part III.A. “Another relevant consideration is whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area.” League of United Latin Am. Citizens v. Perry (“LULAC”), 548 U.S. 399, 426 (2006).

Notably, intent is not an element of a Section Two violation, and “proof that a contested electoral practice or mechanism was adopted or maintained with the intent to discriminate against minority voters, is not required.” City of Carrollton Branch of the NAACP v. Stallings, 829 F.2d 1547, 1553 (11th Cir. 1987).

B. The 2021 Plan and First Preliminary Injunction (2021–2022)

After the 2020 Census, the Alabama Legislature began the decennial redistricting process in May 2021 using population estimates from the Census Bureau. To guide the process, the Committee passed redistricting guidelines (“the 2021 guidelines”). Milligan Doc. 404-1 at 1–3 (Ex. MX-41).⁶ The 2021 guidelines are attached to this Order as Appendix A, and they provide (among other things) for how the Committee will consider and apply traditional redistricting principles. Traditional redistricting principles “includ[e] compactness, contiguity, ․ respect for political subdivisions or communities defined by actual shared interests, incumbency protection, and political affiliation.” Ala. Legis. Black Caucus v. Alabama, 575 U.S. 254, 272 (2015) (internal citation and quotation marks omitted).

The Milligan and Caster parties stipulated that under Alabama law, the Committee “was tasked with making a ‘continuous study of the reapportionment problems in Alabama seeking solutions thereto’ and reporting its investigations, findings, and recommendations to the Legislature as necessary for the ‘preparation and formulation’ of redistricting plans for the Senate, House, and congressional districts.” Milligan Doc. 436 ¶ 63 (quoting Ala. Code § 29-2-52). They also stipulated that the Committee can “prepare and propose the redistricting plan required for the State Board of Education.” Id. (citing Ala. Code § 16-3-3).

The Census Bureau released data to Alabama in August 2021, id. ¶ 70, and the Singleton Plaintiffs initiated this first redistricting lawsuit on September 27, 2021 against then-Alabama Secretary of State John Merrill, Singleton Doc. 1. The Singleton Plaintiffs are registered voters in Alabama's Second, Sixth, and Seventh Congressional Districts under the 2023 Plan and lead plaintiff Bobby Singleton is a Black Senator in the Legislature. Singleton Doc. 229 ¶¶ 9–12; Tr. 2362.⁷ The Singleton Plaintiffs asserted that holding Alabama's 2022 election under its 2011 map would violate the Constitution, so the Chief Judge of the United States Court of Appeals for the Eleventh Circuit convened this three-judge court. Singleton Doc. 1 at 30–36; Singleton Doc. 13.

Governor Ivey called a special legislative session on redistricting to begin in October 2021, and the 2021 Plan became state law in November 2021. Singleton Doc. 47 ¶¶ 35–37, 40. It is sometimes described as “HB1” and appears below:

Milligan Doc. 403-20 (Ex. MX-20). The Singleton Plaintiffs amended their complaint to assert constitutional claims based on the 2021 Plan and request a declaratory judgment and injunctive relief. Singleton Doc. 15 at 38–48. The Singleton Plaintiffs have since amended their complaint to include a Section Two claim. Singleton Doc. 229 ¶¶ 80–83.

The Caster Plaintiffs then filed their lawsuit against the Secretary. Caster Doc. 3. The Caster Plaintiffs are citizens of Alabama's First, Second, and Seventh Congressional Districts under the 2023 Plan. Caster Doc. 271 ¶¶ 10–18. They challenge the 2023 Plan, as they did the 2021 Plan, only under Section Two. Compare Caster Doc. 271 ¶¶ 123–129, with Caster Doc. 3 ¶¶ 89–95. Caster is pending before Judge Manasco sitting alone. The Caster Plaintiffs assert a single claim of vote dilution and request declaratory and injunctive relief. Caster Doc. 271 ¶ 129.

The Milligan Plaintiffs then filed their lawsuit against the Secretary and the Legislators. Milligan Doc. 1. The Milligan Plaintiffs are Black registered voters in Alabama's First and Second Congressional Districts and two organizations — Greater Birmingham Ministries and the Alabama State Conference of the National Association for the Advancement of Colored People (“NAACP”) — with members who are registered voters in those districts and the Seventh District. Milligan Doc. 329 ¶¶ 18–26. These plaintiffs assert claims of vote dilution under Section Two and racial gerrymandering and intentional discrimination under the Fourteenth Amendment. Id. ¶¶ 190–205. They request declaratory and injunctive relief and bail-in under Section 3(c) of the Voting Rights Act. Id. ¶ 206.

Judge Manasco ordered that Milligan was required to be heard by a three-judge court, and the Chief Judge of the Eleventh Circuit convened a court composed of the same judges who comprise the Singleton Court. Milligan Docs. 22, 23. The Legislators intervened in Singleton and Caster. Singleton Doc. 32; Caster Doc. 69.

Each set of Plaintiffs moved for a preliminary injunction. Singleton Docs. 42, 57; Milligan Doc. 69; Caster Doc. 56. The three-judge Singleton Court consolidated Singleton and Milligan “for the limited purposes” of preliminary injunction proceedings; set a hearing for January 4, 2022; and set prehearing deadlines. Milligan Doc. 40 at 3, 10–12. Judge Manasco set Caster for a hearing on the same date and set identical prehearing deadlines. Caster Doc. 40 at 2–5.

All parties agreed to a consolidated preliminary injunction proceeding that permitted consideration of evidence in a combined fashion. All parties also agreed that evidence admitted in any one of the three cases could be used in the other two cases absent a specific objection. See Singleton Doc. 72-1 at 2–3; Caster Doc. 74; Tr. Dec. 20, 2021 Hrg. at 14–17; Tr. Jan. 4, 2022 Hrg. 29.

The preliminary injunction hearing commenced on January 4 and concluded on January 12, 2022. Allen, 599 U.S. at 16. Later that month, we preliminarily enjoined the State from using the 2021 Plan because we concluded that it likely violated Section Two. See Milligan Doc. 107. In that order, we ruled:

Because the Milligan plaintiffs are substantially likely to prevail on their claim under the Voting Rights Act, under the statutory framework, Supreme Court precedent, and Eleventh Circuit precedent, the appropriate remedy is a congressional redistricting plan that includes either an additional majority-Black congressional district, or an additional district in which Black voters otherwise have an opportunity to elect a representative of their choice. See, e.g., Bartlett v. Strickland, 556 U.S. 1, 24 (2009); Cooper v. Harris, 137 S. Ct. 1455, 1470, 1472 (2017). Supreme Court precedent also dictates that the Alabama Legislature (“the Legislature”) should have the first opportunity to draw that plan. See, e.g., North Carolina v. Covington, 138 S. Ct. 2548, 2554 (2018); White v. Weiser, 412 U.S. 783, 794–95 (1973).

The Legislature enjoys broad discretion and may consider a wide range of remedial plans. As the Legislature considers such plans, it should be mindful of the practical reality, based on the ample evidence of intensely racially polarized voting adduced during the preliminary injunction proceedings, that any remedial plan will need to include two districts in which Black voters either comprise a voting-age majority or something quite close to it.

Id. at 5–6.

The State appealed and the Supreme Court stayed our injunction. Allen, 599 U.S. at 17. Wes Allen succeeded John Merrill as the Secretary. Milligan Doc. 161.

C. The Supreme Court's Ruling (2023)

On June 8, 2023, the Supreme Court affirmed the preliminary injunction. See Allen, 599 U.S. at 16–17. We divide our discussion of that ruling into four parts: we first discuss the opinion of the Court, we then turn to the part of the Chief Justice's opinion that is the opinion of four Justices, we then consider Justice Kavanaugh's concurrence, and last we discuss the dissents.

1. Controlling Precedent

Chief Justice Roberts delivered the opinion of the Court, joined by Justices Sotomayor, Kagan, Kavanaugh, and Jackson (except that Justice Kavanaugh did not join one portion of a Part of the opinion). Id. at 8, 30. The Supreme Court began by stating the ruling:

[A] three-judge District Court sitting in Alabama preliminarily enjoined the State from using the districting plan it had recently adopted for the 2022 congressional elections, finding that the plan likely violated Section 2 of the Voting Rights Act. This Court stayed the District Court's order pending further review. After conducting that review, we now affirm.

Id. at 9–10 (internal citations omitted).

Next, the Supreme Court discussed the history of the Fifteenth Amendment, the passage of the Voting Rights Act in 1965, and the congressional compromise behind the 1982 amendments to that statute. Id. at 10–14.

The Supreme Court explained that in the early 1980s, as the result of its decision in another Alabama case, City of Mobile, 446 U.S. 55 (1980), a “sharp debate” brewed in Congress about whether the legal test for relief under Section Two should focus on discriminatory effects or discriminatory intent. Allen, 599 U.S. at 11–13. In response to public concern that an effects test could produce “a quota system for electoral politics,” Congress ultimately compromised along lines proposed by Senator Bob Dole: “Section 2 would include the effects test that many desired but also a robust disclaimer against proportionality.” Id. at 13. That proportionality disclaimer endures in Section Two today. See 52 U.S.C. § 10301(b).

The Supreme Court then observed that “[f]or the first 115 years following Reconstruction, the State of Alabama elected no black Representatives to Congress.” Allen, 599 U.S. at 14. Only after a Section Two lawsuit did that change when District 7 elected Earl Hilliard in 1992 and Alabama's later maps largely resembled its 1992 map. Id. at 14–15; Wesch v. Hunt, 785 F. Supp. 1491 (S.D. Ala. 1992).

The Supreme Court then reiterated its ruling: “The District Court found that plaintiffs demonstrated a reasonable likelihood of success on their claim that [the 2021 Plan] violates § 2. We affirm that determination.” Allen, 599 U.S. at 17.

Next, the Supreme Court restated the controlling legal standard, as set forth in Gingles and applied by federal courts “[f]or the past forty years.” Id. at 17–19. The Court observed that “Congress has never disturbed [the Supreme Court's] understanding of [Section Two] as Gingles construed it,” and that Congress has remained silent despite decades of litigation under Gingles, as the Court has applied Gingles “in one [Section Two] case after another, to different kinds of electoral systems and to different jurisdictions in States all over the country.” Id. at 19 (citing cases from across the United States).

The Court then restated the ruling: “As noted, the District Court concluded that plaintiffs’ § 2 claim was likely to succeed under Gingles. Based on our review of the record, we agree.” Id. at 19 (internal citations omitted). The Supreme Court then reviewed our analysis of each Gingles requirement and agreed with our analysis as to each requirement. Id. at 19–23. It did not hold, suggest, or even hint that any aspect of our Gingles analysis was erroneous. See id.

“With respect to the first Gingles precondition,” the Supreme Court held that we “correctly found that black voters could constitute a majority in a second district that was reasonably configured.” Id. at 19 (internal quotation marks omitted). The Supreme Court ruled that “[t]he plaintiffs adduced eleven illustrative maps—that is, example districting maps that Alabama could enact—each of which contained two majority-black districts that comported with traditional districting criteria.” Id. at 20.

The Supreme Court then considered the illustrative plans prepared by Dr. Moon Duchin, one of the experts for the Milligan Plaintiffs (“the Duchin Plans”). The Supreme Court observed that we “explained that the maps submitted by [Dr. Duchin] performed generally better on average than did [the 2021 Plan].” Id. at 20 (internal quotation marks omitted). Likewise, the Supreme Court considered the illustrative plans prepared by Mr. Bill Cooper, one of the experts for the Caster Plaintiffs (“the Cooper Plans”). The Supreme Court observed that Mr. Cooper “produced districts roughly as compact as the existing plan,” and that “none of plaintiffs’ maps contained any tentacles, appendages, bizarre shapes, or any other obvious irregularities that would make it difficult to find them sufficiently compact.” Id. (internal quotation marks omitted).

Next, the Supreme Court held that the “Plaintiffs’ maps also satisfied other traditional districting criteria. They contained equal populations, were contiguous, and respected existing political subdivisions ․ Indeed, some of plaintiffs’ proposed maps split the same number of county lines as (or even fewer county lines than) the State's map.” Id. at 20. Accordingly, the Supreme Court “agree[d] with” us that “plaintiffs’ illustrative maps strongly suggested that Black voters in Alabama could constitute a majority in a second, reasonably configured, district.” Id. (internal quotation marks omitted).

Next, the Supreme Court turned to the State's argument “that plaintiffs’ maps were not reasonably configured because they failed to keep together a traditional community of interest within Alabama.” Id. The Supreme Court recited the State's argument that “the Gulf Coast region ․ is such a community of interest, and that plaintiffs’ maps erred by separating it into two different districts.” Id.

The Supreme Court “d[id] not find the State's argument persuasive.” Id. at 21. The Supreme Court reasoned that “[o]nly two witnesses testified that the Gulf Coast was a community of interest,” that “testimony provided by one of those witnesses was partial, selectively informed, and poorly supported,” and that “[t]he other witness, meanwhile, justified keeping the Gulf Coast together simply to preserve political advantage.” Id. (internal quotation marks omitted). The Supreme Court concluded that we “understandably found this testimony insufficient to sustain Alabama's overdrawn argument that there can be no legitimate reason to split the Gulf Coast region.” Id. (internal quotation marks omitted).

Next, the Supreme Court considered an alternative basis for its agreement with our Gingles I analysis: that “[e]ven if the Gulf Coast did constitute a community of interest ․ [we] found that plaintiffs’ maps would still be reasonably configured because they joined together a different community of interest called the Black Belt.” Id. at 21. The Supreme Court then described the reasons why the Black Belt is a community of interest — its “high proportion of black voters, who share a rural geography, concentrated poverty, unequal access to government services, ․ lack of adequate healthcare, and a lineal connection to the many enslaved people brought there to work in the antebellum period.” Id. (internal quotation marks omitted).⁸

The Supreme Court agreed with us again, ruling that we “concluded—correctly, under [Supreme Court] precedent—that [we] did not have to conduct a beauty contest between plaintiffs’ maps and the State's. There would be a split community of interest in both.” Id. at 21 (internal quotation marks omitted) (quoting Singleton v. Merrill, 582 F. Supp. 3d 924, 1012 (N.D. Ala. 2022)).

The Supreme Court then rejected the State's argument that the 2021 Plan satisfied Section Two because it performed better than Plaintiffs’ illustrative plans on a core retention metric — “a term that refers to the proportion of districts that remain when a State transitions from one districting plan to another.” Id. at 21–22. The Supreme Court rejected that metric on the ground that the Supreme Court “has never held that a State's adherence to a previously used districting plan can defeat a § 2 claim” because “[i]f that were the rule, a State could immunize from challenge a new racially discriminatory redistricting plan simply by claiming that it resembled an old racially discriminatory plan.” Id. at 22. “That is not the law,” the Supreme Court made clear: Section Two “does not permit a State to provide some voters less opportunity ․ to participate in the political process just because the State has done it before.” Id. (internal quotation marks omitted).

The Supreme Court next discussed the second and third Gingles requirements. The Supreme Court accepted our determination that “there was no serious dispute that Black voters are politically cohesive, nor that the challenged districts’ white majority votes sufficiently as a bloc to usually defeat Black voters’ preferred candidate.” Id. (internal quotation marks omitted). The Supreme Court recited the relevant statistics and noted that the State's expert “conceded that the candidates preferred by white voters in the areas that he looked at regularly defeat the candidates preferred by Black voters.” Id. (internal quotation marks omitted).

The Supreme Court next concluded that the plaintiffs “carried their burden at the totality of circumstances stage.” Id. The Supreme Court upheld our findings that “elections in Alabama were racially polarized; that Black Alabamians enjoy virtually zero success in statewide elections; that political campaigns in Alabama had been characterized by overt or subtle racial appeals; and that Alabama's extensive history of repugnant racial and voting-related discrimination is undeniable and well documented.” Id. (internal quotation marks omitted).

The Supreme Court concluded by again stating its ruling: “We see no reason to disturb the District Court's careful factual findings, which are subject to clear error review and have gone unchallenged by Alabama in any event. Nor is there a basis to upset the District Court's legal conclusions. The Court faithfully applied our precedents and correctly determined that, under existing law, [the 2021 Plan] violated § 2.” Id. at 23 (internal quotation marks and citation omitted).

We have carefully reviewed our Opinion and that of the Supreme Court and discern no basis to conclude that any aspect of our previous Section Two analysis was erroneous.

Next, the Supreme Court turned to arguments by the State urging the Supreme Court to “remake [its] § 2 jurisprudence anew,” which the Supreme Court described as “[t]he heart of these cases.” Id. The Supreme Court explained that the “centerpiece of the State's effort is what it calls the ‘race-neutral benchmark.’ ” Id. The Supreme Court then discussed problems with the argument, which it found “compelling neither in theory nor in practice.” Id. at 23–24.

The Supreme Court rejected the State's assertion that existing precedent “inevitably demands racial proportionality in districting, contrary to” Section Two. Id. at 26. “[P]roperly applied,” the Supreme Court explained, “the Gingles framework itself imposes meaningful constraints on proportionality, as [Supreme Court] decisions have frequently demonstrated.” Id. The Supreme Court then discussed three cases to illustrate how Gingles constrains proportionality: Shaw, 509 U.S. at 647, 655; Miller, 515 U.S. at 906, 910–11; and Bush v. Vera, 517 U.S. 952, 960 (1996) (plurality opinion). Allen, 599 U.S. at 27–29.

“Forcing proportional representation is unlawful,” the Supreme Court reiterated, and Section Two “never requires adoption of districts that violate traditional redistricting principles.” Allen, 599 U.S. at 28–30 (internal quotation marks omitted). Rather, its “exacting requirements ․ limit judicial intervention to those instances of intensive racial politics where the excessive role of race in the electoral process ․ denies minority voters equal opportunity to participate.” Id. at 30 (internal quotation marks omitted).

In Part III-B-1 of the opinion, the Supreme Court then discussed “how the race-neutral benchmark would operate in practice.” Id. Justice Kavanaugh did not join Part III-B-1, which is the only part of the Chief Justice's opinion that Justice Kavanaugh did not join. See id. at 8. We discuss it below. See infra Part I.C.2.

Finally, the Supreme Court rejected the State's arguments that the Supreme Court “should outright stop applying § 2 in cases like these” because it “does not apply to single-member redistricting” and “is unconstitutional as [we] applied it.” Allen, 599 U.S. at 38. The Supreme Court observed that it has “applied § 2 to States’ districting maps in an unbroken line of decisions stretching four decades” and has “unanimously held that § 2 and Gingles ‘[c]ertainly ․ apply’ to claims challenging single-member districts.” Id. (quoting Growe v. Emison, 507 U.S. 25, 40 (1993)). The Supreme Court reasoned that adopting the State's approach would require it to abandon this precedent and explained its refusal to do so: “Congress is undoubtedly aware of our construing § 2 to apply to districting challenges. It can change that if it likes. But until and unless it does, statutory stare decisis counsels our staying the course.” Id. at 39.

Ultimately, the Supreme Court acknowledged “[t]he concern that [Section Two] may impermissibly elevate race in the allocation of political power within the States,” but held that “a faithful application of our precedents and a fair reading of the record before us do not bear [that concern] out here.” Id. at 41–42.

The Supreme Court affirmed our judgments in Caster and Milligan. Id. at 42.

2. Part III-B-1 of the Chief Justice's Opinion

In Part III-B-1, the Chief Justice, in an opinion joined by three other Justices, explained why the State's race-neutral benchmark approach would “fare[ ] poorly” in practice.⁹ Id. at 30 (Roberts, C.J.). The four justices explained that Alabama's benchmark would “change existing law” by “prohibiting the illustrative maps that plaintiffs submit to satisfy the first Gingles precondition from being based on race.” Id. (internal quotation marks omitted). The four justices then explained why they saw “no reason to impose such a new rule.” Id. at 30–33. The four justices observed that on its face, Section Two “ ‘demands consideration of race,’ ” acknowledged that the “line between racial predominance and racial consciousness can be difficult to discern,” and explained their view that “it was not breached here.” Id. (quoting Abbott v. Perez, 585 U.S. 579, 587 (2018)).

The State has previously argued that Part III-B-1 tells us that only a plurality of Justices “concluded that at least some of the plans drawn by Bill Cooper did not breach the line between racial consciousness and racial predominance.” Milligan Doc. 267 ¶ 39 (internal quotation marks omitted). But the State overreads Part III-B-1 as leaving open for relitigation the question whether the Plaintiffs submitted at least one illustrative remedial plan in which race did not play an improper role.

The affirmance tells us that a majority of the Supreme Court concluded that the Plaintiffs satisfied their burden under the first Gingles precondition. This necessarily reflects a conclusion that the Plaintiffs submitted at least one illustrative map in which race did not play an improper role. Justice Kavanaugh's concurrence is to the same effect — Justice Kavanaugh did not suggest, let alone say, that he “vote[d] to affirm” despite finding that the Plaintiffs submitted no illustrative map that properly considered race. Allen, 599 U.S. at 45 (Kavanaugh, J., concurring in part). What Part III-B-1 tells us — and no more — is that only four Justices agreed with every statement in that Part.

3. Justice Kavanaugh's Concurrence

Justice Kavanaugh “agree[d] with the [Supreme] Court that Alabama's redistricting plan violates § 2,” and he “wr[o]te separately to emphasize four points.” Id. at 42. First, he rejected the State's request that the Supreme Court overrule Gingles because “the stare decisis standard for [the Supreme] Court to overrule a statutory precedent, as distinct from a constitutional precedent, is comparatively strict. Unlike with constitutional precedents, Congress and the President may enact new legislation to alter statutory precedents such as Gingles.” Id. Justice Kavanaugh observed that “[i]n the past 37 years ․ Congress and the President have not disturbed Gingles, even as they have made other changes to the Voting Rights Act.” Id.

Second, Justice Kavanaugh rejected the State's contention that “Gingles inevitably requires a proportional number of majority-minority districts, which in turn contravenes the proportionality disclaimer” in Section Two. Id. at 43. Justice Kavanaugh explained that the Supreme Court's precedents establish that “Gingles does not mandate a proportional number of majority-minority districts.” Id. Rather, “Gingles requires the creation of a majority-minority district only when, among other things, (i) a State's redistricting map cracks or packs a large and ‘geographically compact’ minority population and (ii) a plaintiff's proposed alternative map and proposed majority-minority district are ‘reasonably configured’—namely, by respecting compactness principles and other traditional districting criteria such as county, city, and town lines.” Id.

Justice Kavanaugh explained further that if “Gingles demanded a proportional number of majority-minority districts, States would be forced to group together geographically dispersed minority voters into unusually shaped districts, without concern for traditional districting criteria such as county, city, and town lines,” but “Gingles and [the Supreme] Court's later decisions have flatly rejected that approach.” Id.

Third, Justice Kavanaugh rejected Alabama's “race-neutral benchmark” because Section Two “requires in certain circumstances that courts account for the race of voters so as to prevent the cracking or packing—whether intentional or not—of large and geographically compact minority populations.” Id. at 44.

“Fourth,” Justice Kavanaugh emphasized, “Alabama asserts that § 2, as construed by Gingles to require race-based redistricting in certain circumstances, exceeds Congress's remedial or preventive authority,” but “the constitutional argument presented by Alabama is not persuasive in light of the Court's precedents.” Id. at 45. Justice Kavanaugh observed that “the authority to conduct race-based redistricting cannot extend indefinitely into the future,” but declined to consider that argument then because Alabama had not raised it. Id.

Justice Kavanaugh reiterated that he “vote[d] to affirm” and “concur[red] in all but Part III-B-1 of the Court's opinion.” Id. at 45.

4. The Dissents

Justice Thomas published a dissent. Justice Thomas, with Justice Gorsuch joining, first argued that Section Two does not apply to redistricting. Id. at 45–49 (Thomas, J., dissenting). Then, with Justices Gorsuch, Barrett, and Alito joining, he argued that Alabama “should prevail” even if Section Two were applicable because (1) there should be a race-neutral benchmark in Section Two cases and (2) race predominated in the drawing of the plaintiffs’ illustrative remedial plans. Id. at 50–65. Finally, with Justices Gorsuch and Barrett joining, Justice Thomas argued that the way we applied Section Two is unconstitutional. Id. at 67–88.

Justice Alito also published a dissent, joined by Justice Gorsuch. Id. at 94 (Alito, J., dissenting). Justice Alito wrote that he would reconfigure Gingles to “take constitutional requirements into account.” Id. at 95. He described his view that Dr. Duchin and Mr. Cooper assigned race a predominant role in their illustrative plans and argued that we gave “substantial weight” to proportionality, in violation of Section Two. Id. at 102. Finally, Justice Alito discussed his view that existing legal standards trap “States ‘between the competing hazards of liability’ imposed by the constitution and the [Voting Rights Act].” Id. at 109 (cleaned up).

D. The 2023 Plan

On return from the Supreme Court, Milligan came before this three-judge Court, and Caster before Judge Manasco, for remedial proceedings. The State requested that we delay remedial proceedings for approximately five weeks to allow the Legislature time to enact a new plan, and we did. Milligan Docs. 166, 168.

Governor Ivey called a legislative special session (the “2023 Special Session”) to consider congressional redistricting. Milligan Doc. 436 ¶¶ 119–120. Senator Livingston and Representative Pringle co-chaired the Committee, which had “22 members, including 7 Black legislators, who are all Democrats, and 15 White legislators, who are all Republicans.” Id. ¶ 121. Representative Pringle moved for the Committee to re-adopt the 2021 guidelines, and it did. Id. ¶ 123. The Committee's 2023 guidelines (“the 2023 guidelines”) are attached to this order as Appendix A.

The special session of the Legislature commenced on July 17, 2023. See Milligan Doc. 173-1. Ultimately, as we discuss at length below, see infra Part I.I.3, on July 20, 2023, the Alabama House of Representatives passed a congressional districting plan titled the “Community of Interest Plan.” Milligan Doc. 251 ¶¶ 16, 22. That same day, the Alabama Senate passed a different plan, titled the “Opportunity Plan.” Id. ¶¶ 19, 22. The next day, a six-person bicameral Conference Committee passed the 2023 Plan, which was a modified version of the Opportunity Plan. Id. ¶ 23. Later that day, the Legislature enacted the 2023 Plan (also known as “SB5”), and Governor Ivey signed it into law. Milligan Doc. 186.

Although neither the 2021 Plan, nor the Community of Interest Plan, nor the Opportunity Plan was accompanied by any legislative findings, when the Legislature enacted the 2023 Plan, it recited eight pages of legislative findings (“the 2023 legislative findings”). We attach those findings to this order as Appendix B.

The 2023 Plan keeps Alabama's two Gulf Coast counties (Mobile and Baldwin Counties) together in District 1 and combines much of the Black Belt in Districts 2 and 7:

Milligan Doc. 409-86 at 1 (Ex. DX-88).

The 2023 Plan, like the 2021 Plan we enjoined, has only one majority-Black district. Compare Milligan Doc. 186-1 at 2, with Milligan Doc. 107 at 2–3. In the 2023 Plan, District 7 has a BVAP of 50.65% (it was 55.3% in the 2021 Plan). Compare Milligan Doc. 186-1 at 2, with Milligan Doc. 53 ¶ 57. The district with the next largest BVAP is District 2, where Black Alabamians account for 39.93% of the voting age population. Milligan Docs. 186-1 at 2, 251 ¶ 3.

The inclusion of legislative findings in the 2023 Plan is novel; no such findings appear in Alabama's previous plans. (The Committee passed guidelines for Alabama's previous plans, as it did in 2023, but the Legislature did not enact findings. See Milligan Doc. 410-49 (Ex. DX-147); Milligan Doc. 410-50 (Ex. DX-148).) Additionally, as explained below, the 2023 legislative findings differ from the 2023 guidelines. Compare App. A, with App. B.

In the 2023 legislative findings, the Legislature “f[ound] and declare[d]” first that it “adheres to traditional redistricting principles when adopting congressional districts.” App. B at 1. The Legislature then quoted the Supreme Court's statement in these cases that Section Two “never requires adoption of districts that violate traditional redistricting principles.” Id. The 2023 legislative findings next provide that the Legislature's intent in adopting the 2023 Plan “is to comply with federal law,” including the Constitution and Voting Rights Act. Id. They further provide that the Legislature's intent is to give effect to several “traditional redistricting principles,” including: “minimal population deviation,” “contiguous geography,” “reasonably compact geography,” and that the plan “shall contain no more than six splits of county lines,” “keep together communities of interest,” and “not pair incumbent members of Congress.” Id. at 1–2.

The 2023 legislative findings provided that these “principles” are “non-negotiable,” defined “community of interest,” and “declare[d] that at least the three following regions are communities of interest that shall be kept together to the fullest extent possible in this congressional redistricting plan: the Black Belt, the Gulf Coast, and the Wiregrass.” Id. at 2–3. Although all of these communities of interest are located in South and Central Alabama, the 2023 legislative findings did not identify any other communities of interest in Alabama. See id.

The 2023 legislative findings described the Black Belt by listing the 23 counties it includes and providing three paragraphs describing that it is “characterized by its rural geography, fertile soil, and relative poverty, which have shaped its unique history and culture.” Id. at 3–4. The 2023 legislative findings described the Gulf Coast by listing the two counties it includes and providing nine paragraphs across nearly three pages detailing their economy and history, including their “French and Spanish Colonial heritage.” Id. at 4–7. The 2023 legislative findings described the Wiregrass by listing the 9 counties it includes (3 of which overlap with the Black Belt) and providing two sentences describing it. Id. at 7.

The 2023 legislative findings are unlike the 2021 and 2023 guidelines in several ways: first, the guidelines did not identify specific communities of interest, nor describe any community of interest, second, the guidelines did not impose a cap on the number of acceptable county splits, nor define that term, and third, the guidelines did not describe any traditional redistricting principle as “non-negotiable.” See App. A.

And finally, the 2023 guidelines (like their predecessors, see Milligan Doc. 410-49 (Ex. DX-147); Milligan Doc. 410-50 (Ex. DX-148)), specified both that the Legislature intended to comply with the Voting Rights Act and that the plan “shall have neither the purpose nor the effect of diluting minority voting strength.” See App. A. The 2023 legislative findings do not expressly prohibit a plan with the purpose or effect of diluting minority voting strength. See App. B.

Additionally, the 2023 legislative findings employed two definitions that differ from definitions that the State stipulated during the preliminary injunction proceedings and that we and the Supreme Court adopted. In their definition of “community of interest,” the 2023 legislative findings eliminate any reference to similarities based on ethnic, racial, or tribal identities that appeared in the definition that the State used during the preliminary injunction proceedings. The 2023 legislative findings state: “A community of interest is a defined area of the state that may be characterized by, among other commonalities, shared economic interests, geographic features, transportation infrastructure, broadcast and print media, institutions, and historical or cultural factors.” App. B at 3. By contrast, in affirming the stipulated definition used by this Court, the Supreme Court determined that a community of interest is an “area with recognized similarities of interests, including but not limited to ethnic, racial, economic, tribal, social, geographic, or historical identities.” Allen, 599 U.S. at 20 (quoting Singleton, 582 F. Supp. 3d at 1012).

And in their definition of the Black Belt, the 2023 legislative findings eliminate altogether the reference to race and slavery that was in the definition that the State previously stipulated. We adopted that stipulation in our order, and the Supreme Court quoted it. Compare Milligan Doc. 107 at 36–37 (quoting Milligan Doc. 53 ¶ 60) and Allen, 599 U.S. at 21, with App. B at 3–4.¹⁰ In our preliminary injunction order, we found that the Black Belt “is named for the region's fertile black soil. The region has a substantial Black population because of the many enslaved people brought there to work in the antebellum period. All the counties in the Black Belt are majority-or near majority-BVAP.” Milligan Doc. 107 at 36–37. We went on to find that: “The Black Belt is a collection of majority-Black counties that runs through the middle of Alabama. The Black voters in the Black Belt share a rural geography, concentrated poverty, unequal access to government services, and lack of adequate healthcare.” Id. at 165.

In closing argument, counsel for the State described the practical upshot of the 2023 legislative findings when he represented that he was “not aware of a way to draw two majority-Black districts without going against the legislature's priority of keeping Mobile and Baldwin County whole”:

JUDGE MANASCO: So is it possible to draw a map that satisfies the findings expressed in SB-5 with two opportunity districts?

[Counsel for the State]: I am not aware of a way to draw two majority-Black districts without going against the legislature's priority of keeping Mobile and Baldwin County whole.

JUDGE MANASCO: Okay.

JUDGE MARCUS: If that be the case, counsel, help me understand how we could infer anything other than that this map was drawn to avoid addressing and meeting the orders of this Court on remedy that had been affirmed by the Supreme Court? If you drew your findings in such a way as to make it mathematically impossible to comply with the order of this Court and to comply with the order of the Supreme Court of the United States, which affirmed our findings of fact, our conclusions of law, and the remedy that we said you had to adopt in order to comply -- if all of that's so, haven't you drawn a map in such a way as to simply say, we will checkmate the Court orders?

[Counsel for the State]: No, Your Honor. I do not believe –

JUDGE MARCUS: Well, how else can you read the inference from what you said, which is that this is nonnegotiable, effectively; that is to say, you must keep these counties together; you accept that if you keep them together, there is no way on God's green earth you can draw two majority-minority districts or anything quite close to it.

[Counsel for the State]: Because we think, Your Honor, that has to be read in light of the record as a whole and what was going on at the time. The legislature had before it, for example – you know, what the Milligan plaintiffs have said about not splitting the Black Belt so many ways, about what a race-neutral plan would look like. The legislature may have been hoping -- and we think the record supports this -- to find another argument that would lead the Court to believe a different remedy was appropriate or that there was no Section 2 violation at all. Obviously, those arguments did not work. They were made in good faith and they were made with the knowledge that this Court would have to look at those plans before they were used in an election.

JUDGE MARCUS: Gotcha.

Tr. 2647–49.

E. Remedial Proceedings and the Second Preliminary Injunction (2023)

All Plaintiffs objected to the 2023 Plan and requested another injunction. Singleton Doc. 147; Milligan Doc. 200; Caster Doc. 179. We adopted the parties’ joint proposed scheduling order for remedial proceedings. Milligan Docs. 193, 194.

The United States filed a Statement of Interest, Milligan Doc. 199, and we received three amicus briefs: one from Congresswoman Sewell and members of the Congressional Black Caucus in support of the Plaintiffs, see Milligan Docs. 208, 232, 236; another from the National Republican Redistricting Trust in support of the State, see Milligan Docs. 230, 232, 234; and another from certain elected officials in Alabama in support of the Plaintiffs, see Milligan Docs. 255, 258, 260.

At the request of the parties and after a prehearing conference, we clarified that remedial proceedings would be limited to the issue of whether the 2023 Plan complies with the order of this Court, affirmed by the Supreme Court, and Section Two. Milligan Doc. 203 at 3–4. We further clarified that because the scope of the remedial hearing would be limited, the constitutional claims of the Singleton Plaintiffs would not be at issue. Id. at 5. We set a remedial hearing in Milligan and Caster for August 14, 2023, id. at 3, and a preliminary injunction hearing in Singleton to commence immediately after the remedial hearing, id. at 6.

The State moved for further clarification about the remedial proceedings. Milligan Doc. 205. The State reiterated its position that because the 2021 Plan was repealed and replaced when the 2023 Plan became Alabama law, the 2023 Plan “remedies the likely § 2 violation unless Plaintiffs show that the 2023 Plan likely violates § 2.” Id. at 3. Put differently, the State's position was that these cases were not in a “purely remedial” posture because we needed to conduct a “preliminary injunction hearing related to a new law,” in which the Plaintiffs would be required to establish that the 2023 Plan violates Section Two “anew” to obtain relief. Id. at 2–3 (quoting Milligan Doc. 169). All Plaintiffs responded. Milligan Doc. 210; Caster Doc. 190; Singleton Doc. 161.

We again clarified the scope of the remedial proceedings and explained that the purpose of remedial proceedings would be to determine whether the 2023 Plan remedies the likely Section Two violation found by this Court, which the Supreme Court affirmed. Milligan Doc. 222 at 8–9. We emphasized that the plaintiffs “bear the burden to establish that the 2023 Plan does not remedy the likely Section Two violation that this Court found and the Supreme Court affirmed.” Id. at 9. We reiterated that the remedial proceedings would not relitigate the findings made in connection with the previous liability determination. Id. at 11. In our second preliminary injunction order, we not only addressed whether the 2023 Plan remedied the likely Section 2 violation, but in the alternative, we reviewed the 2023 Plan from scratch, starting anew. Milligan Doc. 272 at 139.

For purposes of the remedial hearing, all parties again agreed that we could consider all evidence admitted in either Milligan or Caster, including evidence admitted during the previous preliminary injunction hearing, in both cases unless counsel raised a specific objection. Milligan Doc. 203 at 5; Caster Doc. 182; Aug. 14, 2023 Tr. 61.

At the remedial hearing, the State maintained its position that despite the Supreme Court's affirmance of our injunction, the 2023 Plan had reset the proceedings to ground zero for the Plaintiffs, such that they had to establish anew their entitlement to injunctive relief. Invoking a baseball analogy, Judge Marcus asked the Alabama Solicitor General whether we were “in the first inning of the first [g]ame of this proceeding”; counsel responded, “I think we are.” Aug. 14, 2023 Tr. 61–62.

Also at that hearing, the State readily conceded that the 2023 Plan does not include an additional opportunity district. Indeed, the State asserted that notwithstanding our preliminary injunction order and the relief we granted, and the Supreme Court's affirmance, the Legislature was not required to include an additional opportunity district in the 2023 Plan. Id. at 75, 159–64.

We inquired extensively of the Solicitor General about the State's concession. First, the Solicitor General asserted that if we were again to order an additional opportunity district, we would violate the Supreme Court's affirmance of our preliminary injunction:

JUDGE MANASCO: ․ So in our previous order, we considered the tension between Section 2 compliance and racial gerrymandering. And we indicated following our liability finding what an appropriate remedy would be, that it would be a map that includes an additional opportunity district. I asked a question about that earlier with respect to the motion in limine, but now I'm asking a question with respect to the substance, not necessarily with respect to the evidence you think we ought to consider or ought not to. What role did our statement about the additional opportunity district play in what was necessary to comply with our order?

[Solicitor General]: I think your statement made clear that if we were going to move forward with the exact same priority given to communities of interest, compactness, and county lines as we gave in 2021, that we would likely need to have two majority-[B]lack districts or something quite close to it. But I don't think we were bound to stick to that same prioritization of those same legitimate principles, which the Supreme Court blessed in Allen and has blessed repeatedly as things that a state is allowed to do when it's doing the hard work of trying to draw congressional districting lines.

JUDGE MANASCO: All right. So where are we now? I take it that the state's position is that this is, although it's a remedial proceeding, sort of functionally very much like a preliminary injunction hearing, where if we were to grant the relief that the plaintiffs request, we would be entering an injunction against SB-5 instead of SB-1. So indulge a hypothetical for a moment. If we were to say again there is a violation and what has to happen is an additional opportunity district, what would be the impact in this context of the statement about an additional opportunity district?

[Solicitor General]: Your Honor, I think our position would be that that would be a violation of Allen vs. Milligan Supreme Court's order because they have not satisfied Gingles I. And so you would be requiring us to adopt a map that violates traditional principles which the Supreme Court declared to be unlawful.

JUDGE MANASCO: Well, at what point does the federal court in your view have the ability to comment on whether the appropriate remedy includes an additional opportunity district? On liability? On remedy? Both? Or never?

[Solicitor General]: I don't think there's any prohibition on the Court commenting on what it thinks an appropriate remedy would be, but I do think that that statement had to have been in the context of the 2021 plan and through traditional principles that were given effect in that plan, because again, this is again intensely local appraisal of -- it was an intensely local appraisal of that plan.

JUDGE MANASCO: You can appreciate the concern, though, that if all that's necessary to occur to avoid the additional opportunity district is to redefine the principles, that there never comes a moment where on the state's logic, which we're still in the hypothetical world -- there never comes a moment where the Court can say with force that there has to be an additional opportunity district, because all that's required is for the state to redefine the context every time.

[Solicitor General]: Your Honor, I would dispute that proposition. We couldn't rely on core retention. Allen made that clear. So if we said the new context is core retention, it is our number one priority, that would do us no good in a future challenge. But what we did rely on are those three principles that the Court has said are things that states can do and have always done.

JUDGE MANASCO: But for example, SB-5 pays attention to the Wiregrass. We weren't talking about the Wiregrass in January of 2022. Is there a point at which the context becomes somewhat fixed? We have a census every ten years. So the numerical features that -- the numerical demographics that we're dealing with are fixed at that point in time. But is there some point -- does the state acknowledge any point during the ten-year cycle where the ability to redefine the principles cuts off and the Court's ability to order an additional opportunity district attaches?

[Solicitor General]: Your Honor, I think it sounds a lot like a preclearance regime, which I don't think Section 2 –

JUDGE MANASCO: No. In this world, we've made a liability finding. It's not -- I mean, it's not preclearance. There's been a liability finding as to HB-1. I take it you are urging us to make a liability finding before we do anything, if we do, do anything with respect to HB-5. My question is: If we have to make the liability finding every time and you say that until we make the liability finding we can never comment on the appropriate remedy because the context can be redefined, when in the cycle does the loop cut off?

[Solicitor General]: Your Honor, there are obviously timing issues that we discussed earlier today. If you find that there is a problem with this map that it likely violates Section 2, as well, then our time has run out, and we will have a court drawn map for the 2024 election barring appellate review. But so I think that would address that concern. But -- and this is how federal courts work when it comes to any law that is challenged and is enjoined. If the new law that is enacted that repeals the law whether it's dealing with the First Amendment concern or dealing with -- with any other area of the law that is touched with potential federal interest, it's incumbent on the plaintiff to show that the new law is also violative of federal law. And if the new law looks identical or very, very close to the old law, that's an easy showing to make, the problem for the plaintiffs here is this is not the same map. This is –

JUDGE MANASCO: Let me ask it I guess a little more finely. With respect to HB-1 when we made the liability finding, is it the state's position that at that time this Court had no authority to comment on what the appropriate remedy would be because at that time the Legislature was free to redefine traditional districting principles?

[Solicitor General]: Of course, the Court could comment on it. And I think had the Legislature failed in its attempt to draw a new map, then we would have moved to a pure remedial proceeding, as Judge Marcus recognized on page 155 of Doc 172 in the Milligan case. But the Legislature did succeed in passing a new map that comports with Section 2.

JUDGE MANASCO: I guess that brings me back to my original question. The Legislature has drawn a new map. So what was the import according to the state of the original comment about the additional opportunity district?

[Solicitor General]: I think [it] let the Legislature know that if they were going forward with the exact same principles as they went forward with in 2021, which was refine splitting communities of interest, refine drawing really non-compact districts that might be harder to represent, then you are going to have to apply that in a way that ensures that there's not a [disparate] effect on the minority population, which is going to require two majority Black districts or something close to it. But I don't think we were locked in forever sticking with non-compact districts or sticking with an approach that violates or breaks up communities of interest. Now, we couldn't say it's really important to keep together these communities of interest while splitting the Black Belt. I think that much was made clear by this Court and the Supreme Court. That's why we have a plan now that does better on the Black Belt than every single one of the plaintiffs’ 11 plans. So now they are here asking you to split the Black Belt in order to hit racial goals. And the Supreme Court made clear that is unlawful, and it is unconstitutional.

Id. at 156–161.

Later, the Solicitor General repeated the State's position that another order requiring an additional opportunity district would violate the Supreme Court's affirmance of our preliminary injunction:

JUDGE MOORER: So ․ what I hear you saying is the state of Alabama deliberately chose to disregard our instructions to draw two majority-Black districts or one where minority candidates could be chosen.

[Solicitor General]: Your Honor, it's our position that the Legislature --

JUDGE MOORER: I am not asking you your position. Did they or did they not? Did they disregard it? Did they deliberately disregard it or not?

[Solicitor General]: Your Honor, District 2 I submit is as close as you are going to get to a second majority-Black district without violating Allen -- the Supreme Court's decision in Allen, which is the supreme law of the land when it comes to interpreting Section 2. So I think this is as close as you could get without violating the Constitution, without violating Allen vs. Milligan. So I do think –

JUDGE MOORER: In the view of the [S]tate?

[Solicitor General]: Yes, Your Honor.

Id. at 163–64. The Solicitor General reiterated the State's position that it could comply with Section Two without satisfying the requirement in our order of an additional opportunity district:

JUDGE MARCUS: Let me ask the question one more time. Can you draw a map that maintains three communities of interest, splits six or fewer counties, but that most likely if not almost certainly fails to create an opportunity district and still comply with Section 2?

[Solicitor General]: Yes. Absolutely.

JUDGE MARCUS: Thank you.

Id. at 164.

The Legislature's conduct and the State's concession put this case in an unusual posture. We are not aware of any other case in which a state legislature — faced with a federal court order declaring that its electoral plan unlawfully dilutes minority votes and requiring a plan that provides an additional opportunity district — responded with a plan that state officials concede does not provide that district.

Based on the State's concession and the evidentiary record, on September 5, 2023, we issued a second preliminary injunction, and Judge Manasco again issued a parallel preliminary injunction in Caster. Milligan Doc. 272, Caster Doc. 223.

In that injunction, we expressed concern about the State's position that “so long as the Legislature enacts a remedial map, we have no authority to craft a remedy without first repeating the entire liability analysis. But at the end of each liability determination, the argument goes, we have no authority to order a remedy if the Legislature plans and has time to enact a new map.” Milligan Doc. 272 at 126. “In essence,” we realized, “the State creates an endless paradox that only it can break, thereby depriving Plaintiffs of the ability to effectively challenge and the courts of the ability to remedy.” Id. We explained that the State's “infinity loop ․ terminated only by a new census” was a serious problem: “It cannot be that the equitable authority of a federal district court to order full relief for violations of federal law is always entirely at the mercy of a State electoral and legislative calendar.” Id. at 126–27.

The Secretary — but not the Legislators — appealed. Milligan Docs. 274, 275. After we and the Supreme Court denied the Secretary's requests for a stay, the Secretary dismissed his appeals. Milligan Docs. 276, 281, 307, Caster Doc. 251; Emergency Application for Stay, Allen v. Milligan, No. 23A231 (Sept. 11, 2023); Allen, 144 S. Ct. at 476.

F. The Special Master Plan

Also on September 5, 2023, we issued detailed instructions to the Special Master we appointed: Mr. Richard Allen, an “esteemed public servant with eminent knowledge of Alabama state government,” Milligan Doc. 130 at 3–4; see also Milligan Doc. 273. Mr. Allen served as Chief Deputy Attorney General under four Alabama Attorneys General, served as the Commissioner of the Alabama Department of Corrections, practiced law for many years in Montgomery, and retired from military service with the rank of Brigadier General. See Milligan Doc. 130 at 4. The Special Master was assisted by counsel we appointed for that purpose, Mr. Michael Scodro and the Mayer Brown LLP law firm, and the appointed cartographer, Mr. David Ely. See Milligan Doc. 226 at 4–5, Milligan Doc. 264. No party objected to these appointments. See id. Pursuant to Federal Rule of Civil Procedure 53(a)(2), Mr. Allen, Mr. Ely, and Mr. Scodro attested that they were aware of no grounds for their disqualification. Milligan Docs. 239, 240, 241.

In our detailed instructions, we directed the Special Master to file three proposed plans to remedy the likely Section Two violation we found in the 2023 Plan; include color maps and demographic data with each plan; and file a Report and Recommendation to explain “in some detail the choices made” in each plan and why each plan remedies the likely vote dilution we found. See Milligan Doc. 273 at 6. We directed the Special Master to discuss “the facts and legal analysis supporting the proposed districts’ compliance with the U.S. Constitution, the Voting Rights Act, traditional redistricting criteria, and the other criteria” we listed. See id. at 6–7.

We directed that each recommended plan must “[c]ompletely remedy the likely Section [Two] violation,” which required each plan to “include[ ] either an additional majority-Black congressional district, or an additional district in which Black voters otherwise have an opportunity to elect a representative of their choice.” Id. at 7 (second alteration in original). We further directed that each recommended plan must comply with the Constitution, the Voting Rights Act, and “the one-person, one-vote principle guaranteed by the Equal Protection Clause of the Fourteenth Amendment, based on data from the 2020 Census.” Id. at 7.

We also directed that each recommended plan must “[r]espect traditional redistricting principles to the extent reasonably practicable,” and we observed that “[o]rdinarily, these principles [i]nclud[e] compactness, contiguity, respect for political subdivisions or communities defined by actual shared interests, incumbency protection, and political affiliation.” Id. at 8–9 (quoting Ala. Legis. Black Caucus v. Alabama, 575 U.S. 254, 272 (2015) (internal quotation marks and citations omitted)). Because our Court is “forbidden to take into account the purely political considerations that might be appropriate for legislative bodies,” such as incumbency protection and political affiliation, id. at 9 (quoting Larios v. Cox, 306 F. Supp. 2d 1214, 1218 (N.D. Ga. 2004) (three-judge court)), we limited the Special Master's consideration of traditional districting criteria to compactness, contiguity, respect for political subdivisions, and communities of interest. Id.

We allowed the Special Master to consider the eleven illustrative plans submitted by the Milligan and Caster Plaintiffs, the Singleton Plaintiffs’ remedial maps, the 2021 Plan and 2023 Plan, and the 2023 guidelines and 2023 legislative findings. Id. at 9–10. We also said the Special Master could consider all the evidence before us, public proposals, and additional submissions by the parties. Id. at 10.

The Special Master prepared plans and solicited proposals and comments from the parties and public. Redistricting Doc. 2.

The Special Master observed that the proposals and comments were “necessarily done on an expedited basis but were nonetheless of extremely high quality and were clearly the product of extensive work and thoughtful analysis.” Redistricting Doc. 44 at 13. The Special Master “reviewed[ ] and carefully considered” each submission. Id.

The Special Master filed a 43-page Report and Recommendation that recommended three remedial plans and explained the care he took to limit his analysis as we directed and follow our instructions exactly. See Milligan Doc. 295. In each plan he recommended, the Special Master left Districts 3, 4, and 5 unchanged from the 2023 Plan and modified Districts 6 and 7 only minimally. Id. at 27. His plans were equipopulous and contained only contiguous districts. Id. at 35, 39.

The Special Master confirmed that his plans were not racial gerrymanders or intentionally discriminatory. See id. at 36. Indeed, they were prepared race-blind: the Special Master explained that Mr. Ely “did not display racial demographic data while drawing districts or examining others’ proposed remedial plans within the mapping software, Maptitude. Instead, Mr. Ely relied on other characteristics and criteria” related to communities of interest and political subdivisions. Id.

The Special Master provided core retention metrics, a performance analysis, compactness scores, and information about respect for political subdivisions and communities of interest, for each plan he recommended. See id. at 27–28 tbl. 2; 32 tbl. 4; 38 tbl. 6; 41–43. The Special Master also explained why he rejected other plans, which was principally because they proposed changes “beyond the minimum” changes to the 2023 Plan “needed to remedy the Section Two violation.” Id. at 29.

After we received the Special Master's Report and Recommendation, received objections, and held a hearing, we ordered Secretary Allen to conduct Alabama's 2024 congressional elections using the plan the Special Master titled “Remedial Plan 3” (the “Special Master Plan”). Milligan Docs. 295, 296, 301, 302, 303, 304, 305, 311; Caster Doc. 248, 253; Redistricting Docs. 48, 49; Singleton Doc. 210.

The Special Master Plan appears below:

Caster Doc. 319-18 (Caster Ex. PX-18).

This Court found in a detailed order that the Special Master Plan satisfied all constitutional and statutory requirements while hewing as closely as possible to the 2023 Plan. That order followed a process of elimination. “We beg[a]n by limiting our analysis to the proposed plans that d[id] not exceed our authority,” Milligan Doc. 311 at 36 (citing North Carolina v. Covington, 585 U.S. 969, 978 (2018)), which meant we eliminated any plans that redrew Districts 3, 4, and 5 because those districts were “not challenged in this litigation” and changes to them are “not necessary ․ to remedy the vote dilution we found.” Id. at 36–37. “This eliminate[d] all proposals other than the Special Master's plans and Grofman 2023 Plan.” Id. at 37. Next, we “limit[ed] our analysis to the proposed plans that satisfy the Legislature's limit of six county splits.” Id. We were not required to defer to this limit, but could “remedy the vote dilution we found without exceeding it, so we [did] not exceed it.” Id. (internal citation omitted). This eliminated the Special Master's Remedial Plan 1. Next, we considered how the remaining plans “respect political subdivisions other than counties.” Id. We eliminated the Grofman 2023 Plan because it “split[ ] substantially more voting districts than [was] necessary to remedy the vote dilution we found.” Id.

Although the two remaining plans (the Special Master's Remedial Plans 2 and 3) were similar in many ways, we found that “Remedial Plan 3 better respect[ed] municipal boundaries and the communities of interest that the Legislature identified.” Id. “Remedial Plan 3 [kept] 90.4% of the City of Mobile in a single district, whereas Remedial Plan 2 [kept] only 71.9% of that city in a single district.” Id. at 38. And “although the State [ ] introduced precious little evidence to establish the existence of the Wiregrass community of interest” at the preliminary injunction stage, we considered the fact that Remedial Plan 3 kept six of the Wiregrass counties together, whereas Remedial Plan 2 kept “only five of the Wiregrass counties together.” Id. at 38–39. We found “that of all the proposed remedial plans before us, Remedial Plan 3 ‘most closely approximate[d]’ the plan that the Legislature enacted and we enjoined.” Id. at 39 (quoting Upham v. Seamon, 456 U.S. 37, 42 (1982)).

Since we ordered the Secretary to use the Special Master Plan, the Milligan and Caster parties have executed the following stipulations about that plan:

• Mr. Ely, “drafted the [Special Master] Plan without reference to any illustrative or proposed plan.” Milligan Doc. 436 at 21, ¶ 140.

• “To prepare the [Special Master] Plan, Mr. Ely left CD 3, CD 4, and CD 5 unchanged from the 2023 Plan; preserved all 18 core counties in the Black Belt within CD 2 and CD 1 without splitting any of those counties; and minimized changes to CD 6 and CD 7.” Id.

• “The [Special Master] Plan splits six counties” and “places Henry County with the Wiregrass counties of Houston, Dale, Coffee, Geneva, and Covington in CD 1.” Id. at 21–22, ¶ 140.

• “[A]ccording to the Special Master's report, Mr. Ely sought to preserve the cities of Mobile and Birmingham within single districts and to follow municipal boundaries where possible” while “minimiz[ing] splitting voting districts (precincts) except where needed to equalize population.” Id. at 22, ¶ 141.

• “According to the Special Master's report, Mr. Ely did not display racial demographic data within the mapping software, Maptitude, while drawing his remedial proposals (including the [Special Master] Plan) or while he examined proposed remedies submitted by others.” Id. at 22, ¶ 143.

• “Instead, Mr. Ely drew his proposals ․ based on other nonracial characteristics and criteria related to communities of interest and political subdivisions.” Id.

• BVAP numbers for the 2023 Plan and the three plans filed by the Special Master are as follows (the Special Master Plan is identified as “SM3”):

Id. at 24, ¶ 147.

• The Special Master Plan “paired two incumbents in CD 1: Rep. Jerry Carl, then the CD 1 incumbent, and Rep. Barry Moore, then the CD 2 incumbent.” Id. at 24, ¶ 148.

Following the remedial proceedings and “lengthy negotiations between the parties,” the State agreed to pay $3 million to counsel for the Milligan Plaintiffs as reasonable costs and attorney's fees related to the preliminary injunction proceedings, appeal to the Supreme Court, and remedial proceedings. Milligan Doc. 383 at 6. We granted an unopposed motion by the State compelling it to pay that sum. Id. Judge Manasco granted a similar motion upon agreement by the parties in Caster, which required the State to pay $2,250,000 to counsel for the Caster Plaintiffs. Caster Doc. 297 at 5–6.

G. The 2024 Election

In the 2024 congressional election held under the Special Master Plan, a White candidate, Caroleene Dobson, won the Republican nomination for District 2 and now-Congressman Shomari Figures won the Democratic nomination. See Milligan Doc. 436 at 24, ¶ 149–50. Congressman Figures won the general election, receiving 54.6% of the vote to Ms. Dobson's 45.4%. Id. at 24, ¶ 151. For the first time in Alabama history, two of Alabama's seven Representatives are Black.

H. Trial (2025)

The Plaintiffs now request a final declaration that the 2023 Plan violates federal law; a permanent injunction barring Secretary Allen from conducting any elections pursuant to that Plan; and a permanent injunction under the Voting Rights Act ordering Secretary Allen to conduct Alabama's congressional elections according to a redistricting plan that complies with the Constitution and federal law. Singleton Doc. 229 at 46; Milligan Doc. 329 ¶ 206; Caster Doc. 271 at 43.

All Plaintiffs request that the remedial redistricting plan be court-ordered. Singleton Doc. 229 at 46; Milligan Doc. 485 at 425–27, ¶¶ 1150–52. And the Milligan Plaintiffs request that the Court “[r]etain jurisdiction over this matter and require all Defendants to subject future congressional redistricting plans for preclearance review from this Court or the U.S. Attorney General under Section 3(c) of the [Voting Rights Act], 52 U.S.C. § 10302(c).” Milligan Doc. 329 at 77.

In February 2024, we set a bench trial to commence on February 10, 2025, and set pretrial deadlines. Milligan Docs. 333, 391, 399, 429, 432. In January 2025, we entered the parties’ amended joint proposed trial order. Milligan Doc. 445. Although the cases were not consolidated, the trial proceeded on a coordinated basis that permitted the joint presentation of evidence and argument. The parties again agreed that evidence admitted in any one case could be used in any other case absent a specific objection. Milligan Docs. 444, 445.

Trial commenced on February 10, 2025, and ended on February 26, 2025. We emphasize that we have difficulty imagining a more extensive evidentiary record on the claims and defenses in these cases. After eleven days of trial, the length of the combined trial transcripts was 2,687 pages. Forty able lawyers tried these cases. We heard live testimony from twenty-three witnesses (including thirteen experts); reviewed reports and rebuttal reports from every expert; received testimony by designation for twenty-eight additional witnesses (either from depositions in these cases, or from live testimony in the state Senate redistricting trial that occurred before Judge Manasco in November 2024); considered stipulated facts spanning nearly 40 pages; processed more than 790 putative exhibits; and received more than 840 pages of proposed findings and conclusions after trial. And under Federal Rule of Civil Procedure 65(a)(2), we continue to have the benefit of evidence adduced in the preliminary injunction proceedings that the parties did not abandon at trial. We describe the relevant evidence and argument below. See infra Parts IV and VI.

I. Redistricting Litigation in Alabama

We briefly describe previous Alabama redistricting litigation, for context.

1. State Legislative Redistricting

In 1962, a federal court struck down Alabama's state legislative districting plans after the Legislature failed to redistrict following the decennial census for approximately 50 years. See Sims v. Frink, 208 F. Supp. 431 (M.D. Ala. 1962) (per curiam) (three-judge court). The Supreme Court affirmed. See Reynolds, 377 U.S. at 586–87. On remand, the district court gave the Legislature the opportunity to draw new maps. Sims, 247 F. Supp. at 99. The Legislature adopted new maps, and the district court found the state Senate districts constitutional, but that the state House maps “intentionally aggregated predominantly [Black] counties with predominantly white counties for the sole purpose of preventing the election of [Blacks].” Id. at 106–07, 109.

The district court ordered the State to use a court-drawn map for the next House election. See id. at 108–09; Sims v. Amos, 336 F. Supp. 924, 928 n.4, 931 (M.D. Ala. 1972). In that election, Fred Gray and Thomas Reed became the first Black members of the state House since Reconstruction.

The Legislature again failed to redistrict after the 1970 census, so the district court drew new districts for the state House and Senate. See Sims, 336 F. Supp. at 932, 936, 940. In the election held under that plan, Richmond Pearson and U.W. Clemon became the first Black members of the state Senate since Reconstruction.

Meanwhile, Congress passed the Voting Rights Act of 1965, which required (among other things) Alabama to receive “preclearance” from the Attorney General of the United States or a three-judge court before the State could change its voting procedures. 52 U.S.C. § 10304; Shelby County v. Holder, 570 U.S. 529, 537 (2013).

After the 1980 census, the Legislature passed two plans that did not receive preclearance, and then passed a constitutional plan. Burton v. Hobbie, 561 F. Supp. 1029, 1032–35 (M.D. Ala. 1983). Eleventh Circuit Judge Frank M. Johnson Jr. described the Legislature's previous failure to enact a plan that complied with federal court orders, and the “invidious discrimination existing in both houses of the Legislature.” See id. at 1030–32. He explained that after decades of litigation and judicial intervention, the Legislature, for “the first time in Alabama's history,” “provided an apportionment plan that is fair to all the people of Alabama.” Id. at 1030.

After the 1990 census, federal courts again invalidated the Legislature's redistricting plan, and a new plan was adopted in a state-court consent judgment. See Brooks v. Hobbie, 631 So. 2d 883, 884 (Ala. 1993). That plan, known as the Reed-Buskey Plan, included eight majority-Black state Senate districts. Montiel v. Davis, 215 F. Supp. 2d 1279, 1281–82 (S.D. Ala. 2002). The Supreme Court of the United States upheld the Reed-Buskey Plan. See Sinkfield v. Kelley, 531 U.S. 28, 30–31 (2000). After the 2000 census, the Legislature redistricted and maintained those eight majority-Black state Senate districts. Montiel, 215 F. Supp. 2d at 1281–82.

After the 2010 census, the Legislature again redistricted. Many majority-Black districts, including all eight of the majority-Black state Senate districts, were underpopulated for purposes of the one-person, one-vote requirement. Ala. Legis. Black Caucus v. Alabama, 231 F. Supp. 3d 1026, 1035 (M.D. Ala. 2017). A three-judge court found that the Legislature's plans were lawful, but the Supreme Court vacated that ruling on the ground that the district court misapplied the law on racial gerrymandering. Ala. Legis. Black Caucus v. Alabama, 989 F. Supp. 2d 1227, 1280–87 (M.D. Ala. 2013), vacated on other grounds, 575 U.S. 254 (2015). During the pendency of that litigation, the Supreme Court in a different case struck down the preclearance requirement of Section Five of the Voting Rights Act, so Alabama was no longer required to preclear redistricting plans. Shelby Cnty., 570 U.S. at 556–57. On remand, the district court determined that twelve districts were unconstitutional. Ala. Legis. Black Caucus, 231 F. Supp. 3d at 1140, 1348–49. The Legislature passed remedial plans. See 2017 Ala. Laws Act. Nos. 2017-347, 2017-348.

2. Congressional Redistricting

Unlike redistricting for state legislative seats, congressional redistricting in Alabama was a regular occurrence in the twentieth century. See Singleton Doc. 285-6 (Ex. S-6) (providing Alabama's congressional maps from 1822 to 2022). When congressional maps were drawn in 1933, 1940, and 1950, Alabama had nine districts. Milligan Doc. 436 ¶ 98. All maps between 1933 and 1965 placed Mobile and Baldwin Counties in separate districts. Id. ¶¶ 98–99. In fact, Mobile and Baldwin Counties were in separate congressional districts for almost 100 years, “from 1875 until the 1970s.” Tr. 1305. In every election held under those maps, Alabama elected all-White congressional delegations. Milligan Doc. 436 ¶ 103.

In 1961, Alabama enacted a law that “provide[d] for the manner of nominating candidates for Congress in primary elections and electing congressmen in statewide general elections.” Jansen v. State ex rel. Downing, 137 So. 2d 47, 48 (Ala. 1962). That at-large scheme was “referred to as the ‘9–8 Plan’ and [was] a legislative design, in lieu of redistricting, for meeting the reduction in the number of Alabama congressmen from nine to eight.” Id. at 48. This reduction was realized in 1963 when “the number of representatives from the State of Alabama in the House of Representatives of the United States Congress was reduced from 9 to 8.” Moore v. Moore, 229 F. Supp. 435, 437 (S.D. Ala. 1964). In 1962, both state and federal courts upheld the 9-8 Plan. Id.

In 1964, after the Supreme Court decided Wesberry v. Sanders, 376 U.S. 1 (1964) and Martin v. Bush, 376 U.S. 222 (1964), the federal district court found that it was its “duty to re-examine the [9-8 Plan], the two decisions of the Alabama Supreme Court, and [its own] decision [ ] in light of” Wesberry and Martin. Moore, 229 F. Supp. at 437. That district court struck down the 9-8 Plan, the Legislature again redistricted, and the district court struck down the revised plan because the revised districts were not sufficiently equipopulous. Moore v. Moore, 246 F. Supp. 578, 580 n.1 (S.D. Ala. 1965) (per curiam). The Legislature redistricted again and the district court upheld that plan upon finding that the Legislature “made a good faith effort which has resulted in the establishment of constitutional congressional districts.” Id. at 580–82.

Congressional redistricting following the 1970 census marked three major changes: (1) since 1973, Alabama has been apportioned seven seats in the United States House; (2) the 1970s redistricting cycle was the first cycle to fully occur under the preclearance regime established in Section 5 of the Voting Rights Act; and (3) Mobile and Baldwin Counties were both fully included in District 1 in this cycle, a practice that has continued in every congressional plan (but not every Board of Education plan) enacted since. Milligan Doc. 436 ¶¶ 97, 100–01.

In all the elections held under the maps drawn after the 1970 census and the 1980 census, Alabama elected all-White congressional delegations. See id. ¶ 103. After the 1990 census, the Legislature initially failed to redistrict. See Wesch, 785 F. Supp. at 1494–95. A voter sued and asserted that holding the 1992 election under the old map would violate the one person, one vote rule. Id. at 1492. Several Black voters intervened to assert a Section Two claim. Id. at 1493. The parties proposed plans, and the court retained its own expert. Id. at 1493, 1495.

The district court ultimately ordered that congressional elections be held according to a plan that closely tracked the original plaintiff's proposed plan. See Wesch v. Folsom, 6 F.3d 1465, 1467–68 (11th Cir. 1993). That plan created one “significant majority African–American district with an African–American population of 67.53%.” Id. at 1468; Wesch, 785 F. Supp. at 1498, 1581 app. A. That district, District 7, included Black communities in Jefferson, Tuscaloosa, and Montgomery Counties. Wesch, 785 F. Supp. at 1509, 1569 app. A (Jefferson); id. at 1510, 1581 app. A (Tuscaloosa); id. at 1510, 1575 app. A (Montgomery).

The Wesch court did not decide whether Section Two “require[d] the creation of such a district” because the parties stipulated that “the African American population in the State of Alabama is sufficiently compact and contiguous to comprise a single member significant majority (65% or more) African American Congressional district,” and “a significant majority African American Congressional district should be created.” Id. at 1498–99. The court found that the new plan “create[d] a majority African–American district that provide[d] African–Americans a reasonable opportunity to elect a candidate of their choice, and d[id] so without the need for extensive gerrymandering.” Id. at 1499. The new map was drawn by cartographer Randy Hinaman. Ala. Legis. Black Caucus, 231 F. Supp. 3d at 1038.

In the 1992 election held using the court-ordered map, voters in District 7 elected Alabama's first Black Congressman in over 90 years. See Milligan Doc. 436 ¶ 103. District 7 remains a majority-Black district to this day and in every election since 1992 has elected a Black Democrat. See id. ¶¶ 103, 106, 108, 113–14.

After the 2000 census, Alabama enacted a new congressional plan in which District 7 remained the only majority-Black district. Id. ¶ 104. The 2002 Plan took Montgomery County out of District 7 and split it two ways, between Districts 2 and 3. See Singleton Doc. 285-6 at 40. The 2002 Plan received federal preclearance. Milligan Doc. 436 ¶ 105. Mr. Hinaman also worked on the 2002 Plan. Ala. Legis. Black Caucus, 231 F. Supp. 3d at 1038. In 2002, 2004, 2006, and 2008, District 7 elected Artur Davis, a Black Democrat. Milligan Doc. 436 ¶ 106. In 2010, District 7 elected Congresswoman Sewell, also a Black Democrat. Id. ¶ 108.

After the 2010 census, the Legislature enacted the 2011 Plan. Id. ¶ 111. The 2011 Plan split Montgomery County three ways, placing parts in Districts 2, 3, and 7. See Singleton Doc. 285-6 at 41. Mr. Hinaman drew the 2011 Plan. Ala. Legis. Black Caucus, 231 F. Supp. 3d at 1038. Alabama submitted the 2011 Plan for preclearance, and that submission said: “As with the 1992 Wesch court plan and the plan in Act No. 2002-57, the new plan has one African-American majority district, District 7, which is located in the west central part of the state.” Milligan Doc. 436 ¶ 112. District 7 in the 2011 Plan had a BVAP of 60.91%.¹¹ Id. ¶ 111.

3. The 2023 Special Session

The Milligan and Caster parties have stipulated to much of what occurred in the 2023 Special Session (Milligan Doc. 436), including the following sequence of events:

Before the 2023 Special Session, the Committee held two pre-session hearings to receive public input. Id. ¶ 122. The only plans then available for comment were plans proposed by the plaintiffs in these cases. Id. ¶ 124.

On the first day of the 2023 Special Session, Representative Pringle introduced a plan he titled the “Community of Interest plan.” Id. ¶ 125. The Community of Interest Plan had one majority-Black district (District 7); the district with the next-highest BVAP was District 2, with a BVAP of 42.25%. Id. He advocated for that Plan because it “maintained the core of existing Congressional Districts.” Id. ¶ 126. The Committee passed that Plan on July 17, 2023. Id. ¶ 127. Under that Plan, the “Committee's performance analysis showed that Black-preferred candidates would have won two of the four” modeled elections. Id.

On July 17, 2023, Senator Livingston introduced a plan he titled the “Opportunity Plan” (or “Livingston 1” Plan). Id. ¶ 128. The Opportunity Plan included one majority-Black district (District 7); the district with the next-highest BVAP was District 2, with a BVAP of 38.31%. Id. ¶ 129.

On July 20, 2023, the Alabama House passed the Community of Interest Plan, and the Alabama Senate passed the Opportunity Plan. Id. ¶ 130. The next day, “a six-person bicameral Conference Committee passed Senate Bill 5” (a modified version of the Opportunity Plan that is sometimes referred to as the “Livingston 3 Plan”). Id. ¶ 131. Representative England, one of the Black Democrats on the Committee, “stated that, in his opinion, the Livingston 3 Plan was noncompliant with the Court's preliminary-injunction order and the Court would reject it.” Id. ¶ 132.

SB5 was then passed by both chambers of the Legislature and signed into law by Governor Ivey, at which point it became the 2023 Plan. Id. ¶ 133; Ala. Code § 17-14-70. “The 2023 Plan passed along party lines and almost entirely along racial lines, with all Black legislators except one—a Republican—voting against [it].” Milligan Doc. 436 ¶ 134.

In the 2023 Plan, District 7 has a BVAP of 50.65%; District 2 has a BVAP of 39.93%. Id. ¶¶ 135–37. The Legislature analyzed how the 2023 Plan would perform in seven elections; that performance analysis indicated that the Black-preferred candidate in District 2 “would not have received the most votes in any” of the modeled elections. Id. ¶ 138.

The parties have developed extensive testimony about the 2023 redistricting process from the Legislators and Mr. Hinaman.

a. Randy Hinaman

Mr. Hinaman testified that he drafted the Community of Interest Plan because Dorman Walker,¹² Senator Livingston, and Representative Pringle contacted him shortly after the Supreme Court ruled and asked him “to draw a new congressional map that took the Court's ruling into account and followed the guidelines,” which meant a map that “provided an opportunity for African Americans to elect the candidate of their choice in two districts.” Milligan Doc. 459-7 at 4–6.

He also testified that he was “specifically instructed to keep Mobile and Baldwin Counties together” as a community of interest by Representative Pringle, and possibly by Senator Livingston. Id. at 20.

Mr. Hinaman testified that Senators Livingston and Roberts requested that he review the Opportunity Plan before it was released to the public. Id. at 7–8. He testified that it had been his understanding that Senator Livingston and Representative Pringle would co-sponsor the Community of Interest Plan. Id. at 8. He also testified that he does not know who drafted the Opportunity Plan, that he also was asked to review the Livingston 2 Plan, and that he also did not know who drafted that plan. Id. at 8–10. Mr. Hinaman testified that Senator Livingston and another Senator (Scofield) directed modifications to the Livingston 2 Plan, including changes to District 2, while Mr. Hinaman operated the computer. Id. at 10–11. He said that the modified map was the Livingston 3 Plan and enacted the next day as the 2023 Plan. Id.

Mr. Hinaman testified about several other aspects of his work in June and July 2023. First, he said that Dr. Hood (one of the State's experts in these cases, see infra Part IV.D.3.a), prepared a performance analysis on various plans, and that in District 2 in the Community of Interest Plan, Democrats won two out of the four modeled races. Id. at 13–15. Mr. Hinaman also testified that the performance analysis for the Livingston 2 Plan was worse on this metric. Id. That performance analysis showed that without Dallas County (home to Selma) in District 2, Black-preferred candidates would have no chance of winning that District. Id. at 13. Indeed, according to Mr. Hinaman, the Black-preferred candidate lost every election Dr. Hood modeled in that District, if Selma was not in the District. Id. at 13–14. Mr. Hinaman said that he communicated the performance analysis to Senator Livingston and Representative Pringle. Id. at 15.

Second, Mr. Hinaman testified that he drew two other maps that he believes would have provided a second Black-opportunity district. Id. at 19.

Third, he testified that he chose District 2 as the potential second opportunity district in the Community of Interest Plan because “[i]t was an area of geography that had a compact enough African American population in the relevant counties to draw a district that could perform as an opportunity district.” Id. at 16–17.

Fourth, Mr. Hinaman testified that in drawing District 2 in that plan, he balanced various redistricting principles as follows:

I mean, I was looking at, you know, traditional redistricting principles in whole counties to the extent possible and not pairing incumbents, which, obviously, we had incumbents that could have been paired there in terms of Barry Moore in Coffee County and Jerry Carl in Mobile and taking those things into consideration. The BVAP number, you know, came out to what it came out to.

Id. at 17. Mr. Hinaman was also specifically instructed by the Legislators to keep Mobile and Baldwin Counties together. Milligan Doc. 459-7 at 20.

Fifth, Mr. Hinaman testified that the Alabama Solicitor General presented the “concept of [a] map” to him that was called the Whole Jefferson County Plan and would have connected Shelby County to Black Belt counties that are part of District 7 and moved District 2 into Chilton County. Id. at 23. Mr. Hinaman said that he got that plan “to zero deviation.” Id. at 22–23.

Sixth, upon reviewing the 2023 Plan during his deposition, Mr. Hinaman testified that he had never seen or been told about the 2023 legislative findings; and that as he performed his work, he was not aware of the requirements in those findings that differ from the 2023 guidelines. Id. at 23–24.

b. Senator Livingston

Senator Livingston testified about the development of the 2023 Plan. He explained that Mr. Hinaman was brought in to “help [them] with the maps” and told “to abide by the guidelines that were adopted by the committee.” Milligan Doc. 459-13 at 6. Senator Livingston testified that “it was expressed to [Mr. Hinaman] that the [courts] ordered us to look at an opportunity district -- districts.” Id. He said that “[a]s [he] underst[ood] it, the Courts have ordered [the State] to provide two opportunity districts.” Id.

When asked during his deposition what it means to provide two opportunity districts, Senator Livingston responded that it's “very vague. And I think it's to a matter of interpretation.” Id. However, when asked whether he had an interpretation of what that phrase meant, Senator Livingston replied, “I do not.” Id. at 7. He testified that any analysis to determine whether a district is an “opportunity district” would need to include a Black-preferred candidate who is “well-funded and well-known,” and there have not been Black-preferred candidates in statewide races “who have the funding and respect of their peers.” Id. at 7–8, 14.

Senator Livingston testified that the Legislature “tried to draw [the 2023 Plan] race neutral,” and that he was not looking at race as he evaluated potential plans. Id. at 12, 14. He testified that, although his deposition was the first time that he saw them, he was aware when drawing the 2023 Plan of the provisions in the Court's order about a second opportunity district. Id. at 13.¹³ He testified that the Committee accounted for our order by enacting “SB-5, which has a second congressional district approximately under 40 percent [B]lack voting age population” because that “qualifies as something quite close to a majority of [B]lack voting age population.” Id. at 13.

Senator Livingston said that he had “[v]ery little” involvement with the Community of Interest Plan Mr. Hinaman prepared. Id. at 15. He testified that the Community of Interest Plan “might have” “provided a fair opportunity for African American voters to elect preferred candidates” in District 2, but “[t]he committee members changed [their] focus” away from that plan, so he shifted with them because he “was going to be left behind.” Id. at 16–17. He explained that “the committee members had received some additional information they thought they should go in the direction of compactness, communities of interest, and making sure that congressmen are not paired against each other.” Id. at 17.

Senator Livingston testified that this “additional information” was a “large hiccup,” but he did not know what it was, where it had come from, or who received it. Id. He said he learned of the “information” in a “committee conversation,” but did not recall from whom and had no “idea at all” of its source. Id.

Senator Livingston testified that the Livingston 2 Plan is a modified version of the Opportunity Plan. Id. at 20. He further testified the Livingston 2 Plan “provided a better opportunity” for Black voters to elect candidates of their choice than the Opportunity Plan provided but could not fully explain the basis of this assertion. Id. Similarly, when Senator Livingston was asked about the decision to draw District 2 with a BVAP under 40 percent in SB5, he testified simply that “this is the plan that was brought forward in the end and was compromised upon.” Id. at 13. However, Senator Livingston did explain that the modifications made to the Livingston 2 Plan to yield the Livingston 3 Plan were to create “higher community of interest and compactness scores.” Id. at 21–22.

Senator Livingston acknowledged that he saw Dr. Hood's performance analysis of the 2023 Plan before its enactment, and it showed that the Black-preferred candidate would have lost all seven modeled races in District 2 by approximately seven points. Id. at 23. Senator Livingston testified that “some of those candidates were very weak,” but said that “despite being a well-known, a former incumbent, and well-funded, Senator Jones ¹⁴ would have lost to Senator Tuberville by 4 points” in one of the modeled races. Id. at 24–25.

Senator Livingston also testified about a conversation with Kevin McCarthy, then the Speaker of the United States House, in which Speaker McCarthy expressed his desire to keep a Republican majority in the House; Senator Livingston said that conversation and desire “really didn't play into [his] efforts.” Id. at 24.

Senator Livingston testified that he knew the Alabama Solicitor General drafted the legislative findings in the 2023 Plan but he did not have “any understanding” about why they were included. Id. at 26. He also testified that during the 2023 Special Session, he relied on talking points about the Livingston Plans that were prepared by the Solicitor General. Id. Dr. Joseph Bagley, an expert for the Milligan Plaintiffs, explained that those talking points emphasized the treatment of communities of interest. He quoted them as saying:

The Livingston Plan is a Compact, Communities of Interest Plan that applies the State's traditional districting principles fairly across the State. The 2023 Plan is a historic map that gives equal treatment to important communities of interest in the State, including three that have been the subject of litigation over the last several year – the Black Belt, the Gulf, and the Wiregrass.

Milligan Doc. 385-1 at 26. The talking points also said: “No map in the State's history, and no map proposed by any of the Plaintiffs who challenged the 2021 Plan, does better in promoting any one of these communities of interest, much less all three.” Id.

A lobbyist who worked with Senator Livingston during the 2023 Special Session, Christopher Brown, also testified about these matters. Mr. Brown testified that in 2023, he created plans on Maptitude for Senator Roberts and that a staffer for Congressman Moore communicated with him about redistricting. Milligan Doc. 459-3 at 16–18, 21. He also testified about text correspondence between him and Senator Livingston. See id. at 24–26 (reading from Milligan Doc. 404-23 (Ex. MX-63)). Mr. Brown said that he texted Senator Livingston on June 11, 2023 that he was running performance numbers on eight maps to determine whether they “were going to meet the standards of the Court” to create an opportunity district. Id. at 24. On June 28, 2023, Mr. Brown sent Senator Livingston a text message stating: “This map is workable. Not ideal for Moore. But win[n]able.” Milligan Doc. 404-23 at 2. Mr. Brown testified that he then texted Senator Livingston, “[w]ould 41.6 BVAP work?” Milligan Doc. 459-3 at 25–26. On September 19, 2023, Mr. Brown sent Senator Livingston an article about Representative Pringle's submission of the Community of Interest Plan to the Special Master. Milligan Doc. 404-23 at 2 (Ex. MX-63). Mr. Brown messaged Senator Livingston that Representative Pringle is “[n]ot a team player․ I read this article as an attack on you and the Senate,” and Senator Livingston responded, “[y]ou think[?]” Id. Additionally, in text messages between Senator Livingston and Mr. Brown, Senator Livingston referred to Montgomery as “monkey town.” Id. at 1.¹⁵

c. Representative Pringle

Representative Pringle testified that he understood the courts to require “[e]ither two majority minority districts or something close to it,” and that work began on a new map “within a matter of days” after the Supreme Court's ruling. Milligan Doc. 459-20 at 5–6. Like Senator Livingston, Representative Pringle testified about a phone call from former Speaker McCarthy in which Speaker McCarthy expressed his desire to keep a Republican majority in the U.S. House. Id. at 6. Representative Pringle explained that his own “overriding principle [was] complying with what the United States Supreme Court told me to do,” and that “the United States Supreme Court “told [the Legislature] to draw a map, and that's what [he] tried to do.” Id. But Representative Pringle could not “recall an example of any discussion with the legislature regarding what it means to create an opportunity district.” Id. at 5–6.

Representative Pringle testified that the Alabama Solicitor General “worked as a map drawer at some point in time” during the 2023 Special Session. Id. at 7. When asked what the Solicitor General did in that role, Representative Pringle replied that he “[d]rew maps.” Id. And when asked how he knew that the Solicitor General drew maps, Representative Pringle testified that the Solicitor General “was in the room with his computer across from [him] in reapportionment working on maps.” Id. Representative Pringle testified further that he “lost contact with [the Solicitor General] at the very beginning of the special session and never saw or communicated with him again. He was upstairs meeting with the senators in a different room working with them to draw what ultimately became the Livingston plans.” Id.

Representative Pringle described his instructions to Mr. Hinaman: “I just told him to follow the guidelines and comply with what the Supreme Court told us. And that was to draw two districts which had the ability to elect a Black candidate.” Id. at 8.

Representative Pringle testified that the Committee readopted the 2021 guidelines, which were largely the same since the 1990s because “they cover all the bases.” Id. at 12–13. Representative Pringle testified about the Committee's July 13, 2023 public hearing. Id. at 15. The Committee considered only the plaintiffs’ maps and not his plans because he “[did not] know how [he] could produce a plan until [he] finished hearing from the public.” Id. He testified that he “found it quite fascinating the [P]laintiffs turning on each other and fighting and to watch the democratic members of the committee fight amongst themselves over which plan they wanted based on their own personal political agendas.” Id. at 16.

Representative Pringle then testified about the July 17, 2023 Committee meeting, when he introduced the Community of Interest Plan. Id. at 17. He said that he knew that Dr. Hood's performance analysis predicted the Black-preferred candidate would win two out of four modeled elections, and this performance analysis is the reason why he supported the plan. Id. at 18.

Representative Pringle was aware of the Opportunity Plan before that July 17 meeting, and he described how he learned of the Opportunity Plan: “Senator Dan Roberts brought a thumb drive into the committee room claiming that all of our numbers were wrong, that his consultant, Chris Brown, had drawn a plan that had the right numbers, and we needed to use his plan because everything we had done was wrong.” Id. Representative Pringle testified that he did not know whether the Opportunity Plan “provides a fair opportunity to Black voters to elect preferred candidates” in District 2 because he “never looked [at] or studied” that Plan. Id. at 20. He did not know the differences between the Livingston 2 Plan and the Opportunity Plan. Id. at 22.

On the day the Community of Interest Plan reached the House floor, Representative Pringle “[g]athered [his] little file and went down to the floor.” Id. at 20. That file contained “the quote” about the Voting Rights Act (“what the Court ruled”), his map, “that analysis that was given to [him],” and the population analysis. Id. at 20–21.

Representative Pringle testified that the Livingston 3 Plan that was developed during the reconciliation process between the House and Senate split the difference between the District 2 BVAP in the Community of Interest Plan (42.4%) and Livingston 2 Plan (38.3%), to reach a BVAP of 39.9%. Id. at 25. When asked about that BVAP, Representative Pringle testified “that's what the [S]enate came up with, and they were not going to allow us to pass the [H]ouse plan.” Id. at 25–26. When Representative Pringle was asked why the Senate chose that number, he testified: “You're going to have to talk to Senator Livingston and [the Solicitor General].” Id. at 26.

Representative Pringle testified about his discussion with Senator Livingston:

Senator Livingston came to me towards the end and said, we're going to take your plan and substitute my bill and pass your plan with my map in it. And I said, no we're not. If you want to pass a [S]enate plan, you're going to pass the [S]enate on the [S]enate bill number, and you're not going to put my name on it. You're going – it's not going to be a [H]ouse bill number, it's going to be a [S]enate bill number, that's what we're going to pass.

Id. Representative Pringle testified that he did not want his name on the bill because he thought his plan “was a better plan” to comply with the Voting Rights Act. Id.

Representative Pringle testified that the first time he saw the legislative findings was “Friday morning on the floor of the [H]ouse when the [S]enate bill was brought up.” Id. at 23. Remarkably, he did not know who drafted the findings; he did not know they would be in the bill; and he did not know why they were in the bill. Id.

Representative Pringle testified that he saw the performance analysis for the 2023 Plan on the floor the morning it passed. Id. at 24. And he said that: (1) the performance analysis showed the Black-preferred candidate losing all seven modeled races in District 2, and (2) he believes that all members of the conference committee were aware of that analysis. Id. at 24–25.

Representative Pringle explained that he voted for SB5 because “it was necessary for [the Legislature] to pass a bill,” that “[t]he Senate made it perfectly clear they were not going to pass [the Community of Interest Plan], they were going to pass their plan,” and that the Legislature “had to pass something.” Id. at 24–26.

Representative Pringle testified about various newspaper articles he was shown at his deposition. He was asked about an article entitled “Alabama House Senate Approved Separate Congressional Maps,” which read in part:

Livingston said [S]enate [R]epublicans began working on their own map because the committee “got some information” that led them to prioritize “compactness and communities of interest being as important as the [B]lack voting age population.” Livingston, who did not say where the information came from, said that he had not heard concerns from senators about districts being over 40 percent Black.

Id. at 26. Representative Pringle testified that he did not know what “some new information” was referring to. Id. He testified that “[a]fter the initial meeting, [he] never met with the [R]epublican members of the committee from the [S]enate. They met in a different room on a different floor.” Id.

Representative Pringle was also shown a news article titled “Alabama shamelessly ignores U.S. Supreme Court” that reported that Alabama House Speaker Nathaniel Ledbetter said: “If you think about where we were, the Supreme Court ruling was five to four. So there's just one judge that needed to see something different. And I think the movement that we have and what we've come to compromise on today gives us a good shot.” Id. at 27–28. Representative Pringle testified that he did not want to speak on behalf of the Legislature to answer whether it was “attempting to get a justice to see something differently.” Id. at 28.

J. Claims and Defenses

1. Singleton Plaintiffs

The Singleton Plaintiffs allege that, in addition to violating Section Two of the Voting Rights Act, the 2023 Plan “intentionally perpetuates the unconstitutional racial gerrymandering” that occurred when the Wesch court created District 7 and again after the 2000 and 2010 censuses when the racial composition of that district was materially unchanged. Singleton Doc. 229 ¶¶ 1–2. The Singleton Plaintiffs also allege that the 2023 Plan violates the Fourteenth and Fifteenth Amendments to the Constitution because “the Legislature rejected a plan proposed by the [Singleton Plaintiffs] that more closely complies with the redistricting principles set out in [SB 5] because the [Singleton Plaintiffs’] plan contained two effective crossover districts that encouraged biracial political alliances in Jefferson County and ensures equal opportunity for Black voters in the Black Belt.” Id. ¶ 3.¹⁶

The Singleton Plaintiffs call their proposed remedial plan the “Whole County Plan.” Id. ¶ 40. They assert that “[t]hroughout the state's history, the most important traditional districting principle for drawing Alabama's Congressional districts has been preserving whole counties.” Id. ¶ 17. To that end, the Singleton Plaintiffs allege that the Whole County Plan “eliminated these racial gerrymanders” by drawing district lines solely on county lines without diminishing Black voters’ “opportunity to elect the candidates of their choice in two congressional districts” and “with only slight population deviations.” Id. ¶¶ 39–40, 51. In the Whole County Plan, District 7 would contain 49.9% registered Black voters, and District 6 would contain 42.3% registered Black voters. Id. ¶ 40. The Singleton Plaintiffs say that Black voters would “have an opportunity to elect the candidate of their choice in both districts” because recent election returns reflect “dependable biracial coalition voting” in both proposed districts. Id.

The Singleton Plaintiffs assert claims in three counts. In Count I, they allege that the 2023 Plan is racially gerrymandered, in violation of the Equal Protection Clause of the Fourteenth Amendment and Article I, § 2 of the Constitution. Id. ¶ 67. In Count II, they assert that the 2023 Plan violates the Fourteenth and Fifteenth Amendments because it was drawn (and the Whole County Plan was rejected) to intentionally discriminate against Black voters. Id. ¶¶ 75–79. In Count III, they assert that the 2023 Plan violates Section Two. Id. ¶¶ 80–83.

2. Milligan Plaintiffs

The Milligan Plaintiffs claim that the 2023 Plan was “designed with the intent to crack Black voters into congressional districts in a manner that prevents the creation of two congressional districts in which Black voters have an equal opportunity to elect candidates of their choice.” Milligan Doc. 329 ¶ 2.

The Milligan Plaintiffs allege that a significant number of Black Alabamians live in an area that begins in Jefferson County and extends south- and west-ward to Mobile County and then east- and north-ward to Montgomery and Macon counties. See id. ¶ 5 n.1. Much of that area is in the Black Belt. Id.¹⁷ According to the Milligan Plaintiffs, Black voters in congressional districts inclusive of the Black Belt tend to share common “history, political beliefs, cultural values, transportation, media, and economic interests.” Id. ¶ 97. Under the 2023 Plan, those Black voters are placed in three districts: Districts 1 and 2, where the Milligan Plaintiffs assert their votes are unlawfully diluted, and District 7, which these Plaintiffs assert is packed. Id. ¶ 5.

The Milligan Plaintiffs assert claims in two counts. Count One asserts a claim of vote dilution under Section Two. Id. ¶¶ 190–96. Count Two charges that the 2023 Plan was enacted intentionally to discriminate against Black people in violation of the Fourteenth Amendment, 42 U.S.C. § 1983, and Section Two. Id. ¶¶ 197–205.

The Milligan Plaintiffs did not include their claim for an intentional violation of Section Two in their proposed pretrial order. See Milligan Doc. 445. Because “ ‘a pretrial order supersedes the pleadings,’ thereby ‘eliminating’ any claims not preserved in the pretrial order,” FN Herstal SA v. Clyde Armory Inc., 838 F.3d 1071, 1089 (11th Cir. 2016) (quoting State Treasurer of Mich. v. Barry, 168 F.3d 8, 9–10 (11th Cir. 1999)), we do not consider that claim as part of Count Two. We do consider the Milligan Plaintiffs’ claim of intentional discrimination in violation of the Constitution which is the remainder of Count Two.

The Milligan Plaintiffs allege that the “2023 Plan represents Alabama's latest discriminatory scheme, designed with the intent to crack Black voters into congressional districts in a manner that prevents the creation of two congressional districts in which Black voters have an equal opportunity to elect candidates of their choice.” Milligan Doc. 445 at 9 (pretrial order). The Milligan Plaintiffs “contend that the direct and circumstantial evidence of the Legislature's intent will show that the 2023 Plan intentionally perpetuated the discriminatory effects of the 2021 Plan.” Id.

The Milligan Plaintiffs claim that the only proper remedy is a plan with two Black-opportunity districts. Milligan Doc. 329 at 76–77. To demonstrate that such relief is feasible, they rely on the Duchin Plans and Cooper Plans, the Special Master Plans, and “the VRA Plaintiffs’ Remedial Plan introduced in the Legislature.” Id. ¶ 96.

The Milligan Plaintiffs also urge that Alabama be bailed-in to preclearance review pursuant to Section 3(c) of the Voting Rights Act “until 60 days after the Alabama Legislature enacts a congressional plan under the 2030 census or a period of approximately seven years.” Milligan Doc. 485 at 436.

3. Caster Plaintiffs

The Caster Plaintiffs assert that the 2023 Plan violates Section Two because “it dilutes Black voting strength and confines Black voting power to one majority-Black district,” “despite Alabama's Black population being sufficiently numerous and geographically compact to support two majority-Black congressional districts.” Caster Doc. 271 ¶¶ 1, 2. They assert that the 2023 Plan cracks Black voters between Districts 1 and 2 and packs them into District 7. Id. ¶¶ 4, 57, 125. They also assert that “there is widespread racially polarized voting in Alabama, and when considered against the totality of the circumstances,” including Alabama's history of discrimination, unlawful redistricting, and racial appeals in political campaigns, the State's “failure to create two majority-Black districts dilutes the Black vote in violation of Section 2.” Id. ¶¶ 4, 76–122.

The Caster Plaintiffs assert only one count of vote dilution under Section Two and 42 U.S.C. § 1983. Id. ¶¶ 123–29. They request any remedy that declares that the 2023 Plan violates Section Two, enjoins the use of that Plan, and orders a plan that includes two majority-Black or Black-opportunity districts. Id. at 43.

4. The State

As to Section Two, the State argues that the “Plaintiffs have failed to produce an illustrative plan that is ‘reasonably configured,’—i.e., one that ‘comports with traditional districting criteria.’ ” Milligan Doc. 445 at 11 (pretrial order) (quoting Allen, 599 U.S. at 18). The State also argues that the totality of the circumstances reveals that “political processes in Alabama are open to all, and that ‘what appears to be bloc voting on account of race [is instead] ․ the result of political or personal affiliation of different racial groups with different candidates.’ ” Id. (quoting Ala. State Conf. of the NAACP v. Alabama, 612 F. Supp. 3d 1232, 1316 (M.D. Ala. 2020)). The State also asserts argues that Section Two cannot serve as the basis for a private suit, nor be constitutionally applied to redistricting plans. Id. at 11–12.

The State says that the Singleton and Milligan Plaintiffs cannot overcome the presumption that the Legislature acted in good faith, nor establish “that the 2023 Plan had ‘the purpose and effect of diluting the minority vote’ ” to support their claims of a constitutional violation. Id. at 12 (quoting Alexander v. S.C. State Conf. of the NAACP, 144 S. Ct. 1221, 1252 (2024)).

II. STANDARD OF REVIEW

“The usual standard of proof in civil litigation is preponderance of the evidence,” E.M.D. Sales, Inc. v. Carrera, 145 S. Ct. 34, 37 (2025), and redistricting cases do not require a higher threshold, see, e.g., Cooper v. Harris, 581 U.S. 285, 319 n.15 (2017). We consider whether the Plaintiffs have proven their claims by a preponderance of the evidence on a fresh slate because “the findings of fact and conclusions of law made by a court granting a preliminary injunction are not binding at trial on the merits.” Univ. of Tex. v. Camenisch, 451 U.S. 390, 395 (1981).

III. APPLICABLE LAW

A. Section Two Claims

“For the past forty years, [federal courts] have evaluated claims brought under [Section Two] using the three-part framework developed in [the Supreme Court] decision Thornburg v. Gingles.” Allen, 599 U.S. at 17 (citation omitted). The Supreme Court “ha[s] applied Gingles in one [Section Two] case after another, to different kinds of electoral systems and to different jurisdictions in States all over the country.” Id. at 19. “Congress has never disturbed [the] understanding of [Section Two] as Gingles construed it.” Id.; see also id. at 42 (Kavanaugh, J., concurring in part) (“In the past 37 years, ․ Congress and the President have not disturbed Gingles, even as they have made other changes to the Voting Rights Act.”).

Gingles requires district courts to conduct a two-step analysis for Section Two claims. First, we consider whether the Plaintiffs established the Gingles preconditions, including that: (1) as a group, Black voters in Alabama are “sufficiently large and [geographically] compact” to constitute a majority in an additional “reasonably configured district”; (2) Black voters are “politically cohesive”; and (3) each challenged district's “white majority votes sufficiently as a bloc to enable it ․ to defeat the [Black] preferred candidate.” Id. at 18 (majority opinion) (internal quotation marks and citations omitted).

“Each Gingles precondition serves a different purpose.” Id. “The ‘geographically compact majority’ and ‘minority political cohesion’ showings are needed to establish that the minority has the potential to elect a representative of its own choice in some single-member district. And the ‘minority political cohesion’ and ‘majority bloc voting’ showings are needed to establish that the challenged districting thwarts a distinctive minority vote by submerging it in a larger white voting population.” Growe, 507 U.S. at 40 (internal citations omitted).

As to the first Gingles requirement, “a party asserting [Section Two] liability must show by a preponderance of the evidence that the minority population in the potential election district is greater than 50 percent.” Bartlett, 556 U.S. at 19–20. As the Supreme Court has explained, “it is a special wrong when a minority group has 50 percent or more of the voting population and could constitute a compact voting majority but, despite racially polarized bloc voting, that group is not put into a district.” Id. at 19. Because “only eligible voters affect a group's opportunity to elect candidates,” LULAC, 548 U.S. at 429, the unit of analysis is the Black voting-age population (“BVAP”).

Any proposed majority-minority district must be reasonably configured. See Allen, 599 U.S. at 18. “A district will be reasonably configured ․ if it comports with traditional districting criteria, such as being contiguous and reasonably compact.” Id. The compactness analysis “refers to the compactness of the minority population, not to the compactness of the contested district.” LULAC, 548 U.S. at 433 (quoting Vera, 517 U.S. at 997 (Kennedy, J., concurring)).

Compactness “is critical to advancing the ultimate purposes of [Section Two], ensuring minority groups equal ‘opportunity ․ to participate in the political process and to elect representatives of their choice.’ ” Id. at 434 (alteration in original) (quoting 42 U.S.C. § 1973(b)). A “minority group [that] is spread evenly throughout” the relevant geographic area (i.e., “substantially integrated throughout” that area), is not compact enough to “maintain that they would have been able to elect representatives of their choice” in a single district. Gingles, 478 U.S. at 50 n.17.

“While no precise rule has emerged governing [Section Two] compactness, the ‘inquiry should take into account traditional districting principles such as maintaining communities of interest and traditional boundaries.’ ” LULAC, 548 U.S. at 433 (quoting Abrams v. Johnson, 521 U.S. 74, 92 (1997)). “A district that ‘reaches out to grab small and apparently isolated minority communities’ is not reasonably compact.” Id. (quoting Vera, 517 U.S. at 979). “[B]izarre shaping of” a district that, for example, “cut[s] across pre-existing precinct lines and other natural or traditional divisions,” suggests “a level of racial manipulation that exceeds what [Section Two] could justify.” Vera, 517 U.S. at 980–81.

“Community of interest” is a term of art. The 2023 legislative findings provide that it refers to “a defined area of the state that may be characterized by, among other commonalities, shared economic interests, geographic features, transportation infrastructure, broadcast and print media, educational institutions, and historical or cultural factors.” Milligan Doc. 403-31 at 4 (Ex. MX-31).

“[T]he first Gingles condition requires the possibility of creating more than the existing number of reasonably compact districts with a sufficiently large minority population to elect candidates of its choice.” Johnson v. De Grandy, 512 U.S. 997, 1008 (1994). So these Plaintiffs must establish that Black voters are sufficiently numerous and geographically compact to support two reasonably configured majority-Black districts. See id.; Voinovich v. Quilter, 507 U.S. 146, 153 (1993). Plaintiffs must “demonstrate the existence of a proper remedy.” Burton v. City of Belle Glade, 178 F.3d 1175, 1199 (11th Cir. 1999) (collecting cases).

Supreme Court precedents limit the role race may play in establishing the first Gingles precondition. The Supreme Court “ha[s] made clear that there is a difference ‘between being aware of racial considerations and being motivated by them.’ ” Allen, 599 U.S. at 30 (quoting Miller, 515 U.S. at 916). Because the Voting Rights Act “demands consideration of race,” Abbott, 585 U.S. at 587, Section Two plaintiffs and their map drawers will “be aware of racial demographics,” but this “race consciousness does not lead inevitably to impermissible race discrimination.” Allen, 599 U.S. at 30 (first quoting Miller, 515 U.S. at 916; and then quoting Shaw, 509 U.S. at 646). It's simply inherent: “The question whether additional majority-minority districts can be drawn, after all, involves a ‘quintessentially race-conscious calculus.’ ” Id. at 31 (quoting De Grandy, 512 U.S. at 1020).

That said, “race may not be ‘the predominant factor in drawing district lines unless [there is] a compelling reason.’ ” Id. (alteration in original) (quoting Cooper, 581 U.S. at 291). “Race predominates in the drawing of district lines ․ when ‘race-neutral considerations [come] into play only after the race-based decision had been made.’ ” Id. (second alteration in original) (quoting Bethune-Hill v. Va. State Bd. of Elections, 580 U.S. 178, 189 (2017)).

Accordingly, to determine whether the Plaintiffs satisfy the first Gingles precondition, we compare the 2023 Plan with each of the Plaintiffs’ illustrative remedial plans. See LULAC, 548 U.S. at 430 (quoting De Grandy, 512 U.S. at 1008). Further comparisons are not required; a Section Two “district that is reasonably compact and regular, taking into account traditional districting principles,” need not also “defeat [a] rival compact district[ ]” in a “beauty contest[ ].” Vera, 517 U.S. at 977 (internal quotation marks and emphasis omitted); see also Allen, 599 U.S. at 21 (noting that we correctly concluded that this Court did not have to conduct a beauty contest between Plaintiffs’ maps and the State's).

The second and third Gingles preconditions rise and fall on whether the Plaintiffs establish that voting in the challenged districts is racially polarized. See, e.g., LULAC, 548 U.S. at 427. As the Supreme Court has explained, “in the absence of significant white bloc voting it cannot be said that the ability of minority voters to elect their chosen representatives is inferior to that of white voters.” Voinovich, 507 U.S. at 158 (quoting Gingles, 478 U.S. at 49 n.15).

If the Plaintiffs establish all three Gingles requirements, the Court then must consider whether, “under the ‘totality of circumstances,’ ․ the political process is not ‘equally open’ to minority voters.’ ” Allen, 599 U.S. at 18 (quoting Gingles, 478 U.S. at 45–46). This “inquiry recognizes that application of the Gingles factors is ‘peculiarly dependent upon the facts of each case’ ” and requires the Court to “conduct ‘an intensely local appraisal’ of the electoral mechanism at issue, as well as a ‘searching practical evaluation of the past and present reality.’ ” Id. at 19 (quoting Gingles, 478 U.S. at 79).

“[I]t will be only the very unusual case in which the [P]laintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of [Section Two] under the totality of circumstances.” Ga. State Conf. of the NAACP v. Fayette Cnty. Bd. of Comm'rs, 775 F.3d 1336, 1342 (11th Cir. 2015) (quoting Jenkins v. Red Clay Consol. Sch. Dist. Bd. of Educ., 4 F.3d 1103, 1135 (3d Cir. 1993)).

“Courts use factors drawn from a report of the Senate Judiciary Committee accompanying the 1982 amendments to the [Voting Rights Act] (the Senate [F]actors) to make the totality-of-the-circumstances determination.” Id.; accord De Grandy, 512 U.S. at 1010 n.9. The Senate Factors include:

[(1)] the history of voting-related discrimination in the State or political subdivision; [(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 voting practices or procedures that tend to enhance the opportunity for discrimination against the minority group, such as unusually large election districts, majority vote requirements, and prohibitions against bullet voting; [(4)] the exclusion of members of the minority group from candidate slating processes; [(5)] the extent to which minority group members bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process; [(6)] the use of overt or subtle racial appeals in political campaigns; and [(7)] the extent to which members of the minority group have been elected to public office in the jurisdiction.

De Grandy, 512 U.S. at 1010 n.9 (quoting Gingles, 478 U.S. at 44–45). The Senate Factors also include (8) “evidence demonstrating that elected officials are unresponsive to the particularized needs of the members of the minority group,” and (9) “that the policy underlying the State's or the political subdivision's use of the contested practice or structure is tenuous.” Id.

The Senate Factors are not exhaustive. Under Supreme Court precedent, we may also consider whether the number of Black-majority districts in the 2023 Plan is roughly proportional to the Black share of the population in Alabama. See LULAC, 548 U.S. at 426; De Grandy, 512 U.S. at 1000. The Supreme Court has held that “whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area” is a “relevant consideration” in the totality-of-the-circumstances analysis. LULAC, 548 U.S. at 426; accord De Grandy, 512 U.S. at 1000. “[P]roportionality ․ is obviously an indication that minority voters have an equal opportunity, in spite of racial polarization to participate in the political process and to elect representatives of their choice.” De Grandy, 512 U.S. at 1020 (internal quotation marks omitted) (quoting 42 U.S.C. § 1973(b)); accord Ala. Legis. Black Caucus, 989 F. Supp. 2d at 1286–87 (concluding that the totality of the circumstances weighed against a finding that the state legislative map violated Section Two in part because the number of majority-Black districts in the Legislature is “roughly proportional to the black voting-age population”), vacated on other grounds, 575 U.S. 254 (2015).

But the proportionality evaluation is not and cannot be dispositive. Section Two expressly provides that “nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population,” 52 U.S.C. § 10301(b), and “[f]orcing proportional representation is unlawful and inconsistent with [the Supreme Court's] approach to implementing [Section Two],” Allen, 599 U.S. at 28. And “the Gingles framework itself imposes meaningful constraints on proportionality,” as its “exacting requirements ․ limit judicial intervention to ‘those instances of intensive racial politics’ where the ‘excessive role [of race] in the electoral process ․ den[ies] minority voters equal opportunity to participate.’ ” Id. at 26, 30 (quoting S. Rep. No. 97-417 at 33–34 (1982)).

We may also consider “any circumstance that has a logical bearing on whether” the challenged structure and its interaction with local, social, and historical conditions “afford[ ] equal ‘opportunity.’ ” Brnovich v. Democratic Nat'l Comm., 594 U.S. 647, 668–69 (2021). We are required to “consider ‘the whole picture’ ” and must not view each Senate Factor “in isolation.” District of Columbia v. Wesby, 583 U.S. 48, 60–61 (2018) (quoting United States v. Cortez, 449 U.S. 411, 417 (1981)).

Ultimately, we are required to “assess the impact of the contested structure or practice on minority electoral opportunities on the basis of objective factors.” Gingles, 478 U.S. at 44 (internal quotation marks omitted). Section Two protects against “electoral changes that are discriminatory in effect.” Allen, 599 U.S. at 41 (internal quotation marks omitted).

“[F]or the last four decades, [federal courts] have repeatedly applied the effects test of [Section Two] as interpreted in Gingles and, under certain circumstances, have authorized race-based redistricting as a remedy for state districting maps that violate [Section Two].” Id.

If we determine that the 2023 Plan violates Section Two, that would not amount to a determination that the Plaintiffs are entitled to a map of their choice, or to one of the remedial maps submitted to establish the first Gingles requirement: those maps are illustrative maps submitted for the purposes of establishing liability.

B. Constitutional Claims

“The Constitution entrusts States with the job of designing congressional districts.” Cooper, 581 U.S. at 291; accord Alexander, 144 S. Ct. at 1233. The Fourteenth Amendment mandates that no State shall “deny to any person within its jurisdiction the equal protection of the laws.” U.S. Const. amend. XIV, § 1. Under the Equal Protection Clause, “[a] State may not use race as the predominant factor in drawing district lines unless it has a compelling reason.” Cooper, 581 U.S. at 291.

Because we apply the canon of constitutional avoidance and do not decide the Singleton Plaintiffs’ claim of racial gerrymandering,¹⁸ we do not discuss the law applicable to that claim. A different line of authority applies to the “analytically distinct” claims of the Singleton and Milligan Plaintiffs about intentional discrimination. See Alexander, 144 S. Ct. at 1252.

The Equal Protection Clause “prohibits intentional vote dilution—invidiously ․ minimizing or canceling out the voting potential of racial or ethnic minorities.” Abbott, 585 U.S. at 585–86. A plaintiff alleging intentional discrimination “cannot prevail simply by showing that race played a predominant role in the districting process,” but “must show that the State ‘enacted a particular voting scheme as a purposeful device to minimize or cancel out the voting potential of racial or ethnic minorities.’ ” Alexander, 144 S. Ct. at 1252 (quoting Miller, 515 U.S. at 911). “[T]he plaintiff must show that the State's districting plan ‘has the purpose and effect’ of diluting the minority vote.” Id. (quoting Shaw, 509 U.S. at 649).

Proof of disparate impact is relevant but insufficient to establish a claim of intentional vote dilution. See Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266 (1977); Washington v. Davis, 426 U.S. 229, 242 (1976). So “where the character of a law is readily explainable on grounds apart from race ․ courts must look to ․ evidence [other than disproportionate impact] to support a finding of discriminatory purpose.” City of Mobile, 446 U.S. at 70.

“The task of assessing a jurisdiction's motivation ․ is not a simple matter; on the contrary, it is an inherently complex endeavor, one requiring the trial court to perform ‘a sensitive inquiry into such circumstantial and direct evidence of intent as may be available.’ ” Hunt v. Cromartie, 526 U.S. 541, 546 (1999) (quoting Arlington Heights, 429 U.S. at 266). Discriminatory purpose is “more than intent as volition or intent as awareness of consequences.” Pers. Adm'r of Mass. v. Feeney, 442 U.S. 256, 279 (1979). Rather, it means that “the decisionmaker, in this case a state legislature, selected or reaffirmed a particular course of action at least in part ‘because of,’ not merely ‘in spite of,’ its adverse effects upon an identifiable group.” Id.

To evaluate assertions of discriminatory purpose, the Supreme Court has directed us to consider (1) “[t]he historical background of the decision,” (2) “[t]he specific sequence of events leading up to the challenged decision,” (3) “[d]epartures from the normal procedural sequence,” “particularly if the factors usually considered important by the decisionmaker strongly favor a decision contrary to the one reached,” (4) “[t]he legislative or administrative history” of the decision or action, and (5) whether the “disparate impact” was “the natural and foreseeable consequence of the practices and policies” of the State. Arlington Heights, 429 U.S. at 267–68; accord, e.g., Abbott, 585 U.S. at 609–10; Rogers v. Lodge, 458 U.S. 613, 624–26 (1982), Columbus Bd. of Educ. v. Penick, 443 U.S. 449, 464 (1979). A full evaluation of these factors may require the court to consider “contemporary statements by members of the decisionmaking body, minutes of its meetings, or reports.” Arlington Heights, 429 U.S. at 267–68.

The Eleventh Circuit has recognized three additional evidentiary factors that courts may consider: (1) “the foreseeability of the disparate impact”; (2) “knowledge of that impact”; and (3) “the availability of less discriminatory alternatives.” Greater Birmingham Ministries v. Sec'y of State for Ala., 992 F.3d 1299, 1321–22 (11th Cir. 2021); Jean v. Nelson, 711 F.2d 1455, 1486 (11th Cir. 1983).

If a plaintiff establishes a prima facie claim of intentional discrimination, the burden shifts to the defendant “to dispel the inference.” Castaneda v. Partida, 430 U.S. 482, 497–98 (1977). A defendant must furnish more than “a simple protestation ․ that racial considerations played no part” in the decisionmaking. Id. at 498 n.19. “[D]iscriminatory intent can be rebutted only with evidence in the record about the way in which [the State] operated and their reasons for doing so.” Id. at 500.

Throughout the analysis, however, the “good faith of [the] state legislature must be presumed.” Abbott, 585 U.S. at 603 (quoting Miller, 515 U.S. at 915). The good faith presumption is necessary because it “reflects the Federal Judiciary's due respect for the judgment of state legislators”; it shows an appropriate hesitancy to “hurl ․ accusations” of “ ‘offensive and demeaning’ conduct” at a state legislature; and it evinces an appropriate “war[iness] of plaintiffs who seek to transform federal courts into ‘weapons of political warfare’ that will deliver victories that eluded them ‘in the political arena.’ ” Alexander, 602 U.S. at 11 (first quoting Miller, 515 U.S. at 912; and then quoting Cooper, 581 U.S. at 335 (Alito, J., concurring in judgment in part and dissenting in part)). “The allocation of the burden of proof and the presumption of legislative good faith are not changed by a finding of past discrimination” because “[p]ast discrimination cannot, in the manner of original sin, condemn governmental action that is not itself unlawful.” Abbott, 585 U.S. at 603 (quoting City of Mobile, 446 U.S. at 74).

A plaintiff asserting discriminatory intent may rely on direct and/or circumstantial evidence. Cooper, 581 U.S. at 291; Vera, 517 U.S. at 963 (finding that a mix of direct and circumstantial evidence showed that the legislature was motivated by race in its decisionmaking). Direct evidence may come in the form of a relevant state actor's express acknowledgment that race played a role in the drawing of district lines. See, e.g., Ala. Legis. Black Caucus, 575 U.S. at 259–60.

“[D]iscriminatory intent need not be proved by direct evidence. ‘Necessarily, an invidious discriminatory purpose may often be inferred from the totality of the relevant facts, including the fact, if it is true, that the law bears more heavily on one race than another.’ ” Rogers v. Lodge, 458 U.S. 613, 618 (1982) (quoting Davis, 426 U.S. at 242).

IV. ANALYSIS – VOTING RIGHTS ACT

A. Milligan Plaintiffs’ Arguments

We first consider the Milligan Plaintiffs’ arguments about the Gingles preconditions and then discuss what they say about the totality of the circumstances.

1. Gingles I – Numerosity and Reasonable Configuration (Dr. Duchin)

The Milligan parties stipulate that “[t]here is a numerically sufficient number of Black people of voting age in Alabama to draw to two majority-Black Congressional Districts.” Milligan Doc. 436 at ¶ 152. To establish that Black voters as a group are sufficiently geographically compact to constitute a majority in a second reasonably configured district, the Milligan Plaintiffs rely on Dr. Duchin.

Dr. Duchin's credentials include an undergraduate mathematics degree from Harvard University and two graduate mathematics degrees from the University of Chicago. Milligan Doc. 385-3 at 2; Tr. 279–80. Dr. Duchin now works as a Professor of Mathematics at Cornell University and uses metric geometry to understand redistricting. Tr. 279–81; see Milligan Doc. 385-3 at 1.

Dr. Duchin has published more than a dozen peer-reviewed papers about redistricting. Milligan Doc. 68-5 at 4;¹⁹ see Tr. 280–81. Since she last testified in this litigation, she has published “more than a dozen new publications and preprints about redistricting and elections.” Id. at 280. She was elected as a Fellow of the American Mathematical Society four years ago and has been both a Radcliffe Fellow and a Guggenheim Fellow. Milligan Doc. 68-5 at 4.

Dr. Duchin testified that she has served as “an expert in seven states in this cycle” and has “consulted with many independent commissions and governing bodies around the country in this cycle.” Tr. 280. Dr. Duchin was compensated at a rate of $300 per hour for her work on this case, and her compensation was not dependent on the substance of her testimony. Milligan Doc. 68-5 at 2.

At trial, Dr. Duchin was qualified with no objection as an expert in applied mathematics, quantitative redistricting analysis, demography, and use of census data. Tr. 281.

Dr. Duchin testified that her role in this case was “to study whether, and if so, how it's possible to draw a congressional plan in Alabama with a second majority-Black district.” Id. at 283. She testified that “[t]he work of a Gingles expert asks whether it's possible to do so in a reasonably configured plan,” and “reasonably configured [ ] is a reference to the suite of traditional redistricting principles.” Id. at 281–82. After conducting her work, Dr. Duchin “emphatically” concluded that “it is possible” to draw such a plan. Id. at 282.

Dr. Duchin offered the four illustrative remedial plans she offered at the preliminary injunction stage (Duchin Plans A–D), as well as a fifth plan she developed since (Duchin Plan E). Id. at 282–85, 289. She presented those plans alongside the 2023 Plan and all three Special Master Plans for reference, as reproduced below:

Milligan Doc. 68-5 at 7 fig. 2.

Milligan Doc. 385-3 at 3 fig. 1.

Dr. Duchin testified extensively about her mapmaking process. She testified that she began by “familiar[izing herself] with the census data and the physical geography of [Alabama]”; “stud[ying] some previous maps that had been issued by the state”; and reviewing guidelines “that had been introduced by the legislature in this cycle and in previous cycles to the extent that [she] could.” Tr. 285–86. Then, in “an exploratory phase,” she “us[ed] algorithms that had been developed by [her] lab and [her] research group in which you can randomize the creation of districting plans,” “to get ideas about the question of whether it was possible” to draw a reasonably configured second majority-Black district and “to get directional ideas about how it might be possible.” Id. at 286. She “learned that, yes; it is possible in many ways” and began drawing maps by hand. Id.

Mindful that “redistricting is always about trade-offs,” Dr. Duchin prepared Duchin Plans A–D to “illustrate some of the choices you face when elevating certain principles, when relaxing others, and when watching them trade-off in the creation of a holistically, reasonably configured map.” Id. at 286–87.

To decide which maps to submit, Dr. Duchin testified that she employed certain “nonnegotiable principles” at a “screening” stage. Id. at 287. She testified that “if you were to make a map that didn't meet, say, the one person, one vote requirement, you would ensure that it did so before submitting it to the Court.” Id. She further testified that “similarly, as a Gingles 1 expert, any map that you would submit would need to cross that majority Black threshold in a [s]econd [d]istrict.” Id. at 288. Dr. Duchin testified that “non-negotiable” did not mean “predominant”:

As a matter of process, race is a consideration that doesn't dominate over the others. You're holding many considerations in mind at the same time. Here, to say it was nonnegotiable means, if you are mapping, you might periodically check to see if you're crossing that threshold. And at the end, you need to cross that threshold in order to submit the map to the Court.

Id.

Dr. Duchin testified that she “just did not look at race.” Id. at 292. She “periodically checked to see if the plan, as a whole, had that property of two majority-Black districts.” Id. She clarified:

JUDGE MARCUS: Did you ever have occasion to take a look at the number, 50 plus one, discover that you fell below it, and thereby were nudged to alter or change the microscopic analysis in any way?

[DR. DUCHIN]: As I understand that question, it might ask: Did falling below 50 lead me to look for majority-Black precincts?

JUDGE MARCUS: Yes.

[DR. DUCHIN]: No. That's not my process.

JUDGE MARCUS: So did you not change the microscopic process ․ simply because the numbers in a particular district might have fallen beneath 50 plus one?

[DR. DUCHIN]: That's right. I would say the balancing of criteria looks the same when you're above 50 and below․ [T]he process is not one of going hunting for majority-Black precincts because you have fallen below the line.

JUDGE MARCUS: So let me ask the question again in my own words just to be sure that I have this right․ Did you ever change the microscopic part of your analysis to get up to [ ] 50 plus one?

[DR. DUCHIN] Best I understand the question, my answer would be no.

Id. at 365–66. Dr. Duchin testified that her “top priority” in drawing Plan E was compactness under the “new guidelines that were issued by the state and passed as part of SB-5,” and that “if you know something about the measurement of district compactness, you can see that Plan E is especially compact.” Id. at 289.

Dr. Duchin also testified about how her plans respect traditional redistricting principles. Dr. Duchin testified each of her plans nearly perfectly distributes Alabama's population into contiguous districts: each district in each plan is within a one-person deviation of the baseline of 717,754 people per district, and each district in each plan is contiguous. Milligan Doc. 68-5 at 8; Jan. 6, 2022 Tr. 586–90; Milligan Doc. 92-1; Milligan Doc. 385-3 at 4; see Tr. 287–88.

Dr. Duchin offered extensive testimony to support her opinion that her illustrative remedial districts are reasonably configured. She testified about how her plans perform on the metrics for scoring geographic compactness that the experts in these cases employ. At the preliminary injunction stage, Dr. Duchin explained three metrics. She explained the Polsby-Popper metric as follows: “Polsby-Popper is the name given in this setting to a metric from ancient mathematics: the isoperimetric ratio comparing a region's area to its perimeter via the formula 4πA/P2. Higher scores are considered more compact, with circles uniquely achieving the optimum score of 1.” Milligan Doc. 68-5 at 9.

Then, Dr. Duchin explained Reock scores: “Political scientist Ernest Reock created a different score based on the premise that circles were ideal: it is computed as the ratio of a region's area to that of its circumcircle, where the circumcircle is defined as the smallest circle in which the region can be circumscribed. Polsby-Popper is thought to be relevant as a measure of how erratically the geographical boundaries divide the districts, but this sometimes penalizes districts for natural features like coastlines of bays and rivers. Reock has a much weaker justification, since the primacy of circles is the goal rather than the consequence of the definition.” Milligan Doc. 68-5 at 9. Dr. Duchin explained that for both Polsby-Popper and Reock scores, a higher score is better than a lower score. Id.

Dr. Duchin also explained the cut-edges score: “Recently, some mathematicians have argued for using discrete compactness scores, taking into account the units of Census geography from which the district is built. The most commonly cited discrete score for districts is the cut edges score, which counts how many adjacent pairs of geographical units receive different district assignments. In other words, cut edges measures the ‘scissors complexity’ of the districting plan: how much work would have to be done to separate the districts from each other? Plans with a very intricate boundary would require many separations. Relative to the contour-based scores, this better controls for factors like coastline and other natural boundaries, and focuses on the units actually available to redistricters rather than treating districts like free-form Rorschach blots.” Id. A districting plan with a lower cut-edges score is considered more compact.

Dr. Duchin testified that the differences between her Plan E and the 2023 Plan in terms of compactness were minimal. See Tr. 357 (“My Plan E scores an average Polsby-Popper of a little over 27 percent while SB-5 a little over 28 percent. So there's less than a percentage point of difference.”). She provided a table:

Milligan Doc. 385-3 at 4 tbl. 1.²⁰

In her previous report, she offered a table about compactness scores for her Plans A–D to make the same comparative point, but with respect to the 2021 Plan:

Milligan Doc. 68-5 at 9 tbl. 2.

Dr. Duchin emphasized that these metrics are simply one part of a compactness evaluation, and that there is no “bright line” compactness score on any particular metric that guarantees (or forecloses) reasonableness. Tr. 324–25.

Accordingly, Dr. Duchin also testified about other metrics that are probative of reasonable compactness. She testified that each of her Plans A–D respects existing political subdivisions in the state. Jan. 6, 2022 Tr. 599. She opined that “to make seven finely population-tuned districts, it is necessary to split at least six of Alabama's 67 counties into two pieces, or to split some counties into more than two pieces.” Milligan Doc. 68-5 at 8; Jan. 6, 2022 Tr. 626. And she opined that Duchin Plans A–D “split nine counties or fewer, giving them high marks for respecting these major political subdivisions,” and Duchin Plan E splits six counties. Milligan Doc. 68-5 at 8; Milligan Doc. 385-3 at 4 tbl. 1.

Dr. Duchin also opined that all her plans “are comparable to the State's plan on locality splits, with [Duchin] Plan B splitting fewer localities” than the 2021 Plan, and Duchin Plan E splitting as many localities as the 2023 Plan. See Milligan Doc. 68-5 at 8 tbl. 1; Milligan Doc. 385-3 at 4 tbl. 1; Tr. 293–94.

Dr. Duchin also testified about how her plans treat communities of interest. She testified that different states define “community of interest” differently, and that it is important to respect communities of interest. Tr. 315. According to Dr. Duchin, “[r]espect for communities of interest can mean keeping them together. But there are times when respect or consideration for communities of interest, instead, might call for a split.” Id. She gave an example that she pulled from another mapmaker's preparation of a plan for New York State: “[W]hen he drew a district in the Buffalo area -- previously, Buffalo had been split, and he was able to draw a district that kept Buffalo together and thought he would be lauded as a local hero. And, instead, he was pilloried in the press for taking away a representative. Buffalo used to have two representatives and now only has one.” Id. at 316. She explained that this is “just an example that there are trade-offs that you make if you -- sometimes when you split a community of interest, you are doing so across two districts in which its residents will have a voice and creating more representation. It's really a very holistic situational consideration.” Id. at 316–17.

Dr. Duchin testified about how her plans respect the Black Belt. In the preliminary injunction proceedings, she observed that in the 2021 Plan, eight of the eighteen core Black Belt counties are “partially or fully excluded from majority-Black districts,” while “[e]ach of the 18 Black Belt counties is contained in majority-Black districts in at least some” of her plans. Milligan Doc. 68-5 at 13; see also Aug. 6, 2022 Tr. 666–68. At trial, Dr. Duchin explained that her Plan E respects the Black Belt because Plan E “meet[s] the requirements” of the 2023 legislative findings by keeping the “the number of districts touching [the Black Belt] to a maximum of two.” Milligan Doc. 385-3 at 8.

Dr. Duchin testified that she considered the Wiregrass when drawing her illustrative plans “once the new guidelines had been issue[d]. Previously in the preliminary injunction phase, it was primarily the Black Belt that was being discussed and analyzed.” Tr. 318–19.

All of Dr. Duchin's plans split Mobile and Baldwin Counties. In response to the State's assertion that such a split disrespects a community of interest in the Gulf Coast area, Dr. Duchin testified the Legislature has repeatedly split Mobile and Baldwin Counties in its maps for the Alabama State Board of Education districts, and the Legislature did so in 2020 at the very same time it drew the previous congressional plans. Id. at 325–27, 348. Dr. Duchin thus testified that as she decided where to split Mobile County in her illustrative plans, she drew boundaries based on “guidance from the state board of education map” because that map was “considered legitimate at some point in history by the state legislators.” Id. at 347–49.

Dr. Duchin also testified about the 2023 legislative findings, how they address traditional redistricting principles, and how she deferred to them in drawing her Plan E. Dr. Duchin first testified that the 2023 legislative guidelines were both “novel” and “mathematically impossible to satisfy.” Id. at 359–62. She testified that as far as she was aware, the Legislature had never before enacted requirements of (1) a precise limit on the acceptable number of county splits, or (2) zero incumbent pairings, both of which are enumerated as “non-negotiable” requirements in the 2023 legislative findings. Id. at 297; Milligan Doc. 385-3 at 7.

Dr. Duchin also testified that it is not “mathematically possible to keep together both the Black Belt counties and the Wiregrass counties ․ because those two communities of interest overlap and, taken together, they have more population than a congressional district can have.” Tr. 298; Milligan Doc. 385-3 at 7.

Additionally, Dr. Duchin testified that the requirement to keep Mobile and Baldwin Counties together as a community of interest “come[s] close to prescribing” a majority-White congressional district in the Gulf Coast because “together those [Counties] contain more than 90 percent of the population of a congressional district” and “as a matter of mathematical necessity,” a district that fully includes both Gulf Coast counties must be majority-White and would “submerge[ ]” the City of Mobile. Tr. 298–99, 314. On this point, no document, testimony, or lawyer disputes Dr. Duchin's opinion.

Indeed, in closing argument, counsel for the State represented that he was “not aware of a way to draw two majority-[B]lack districts without going against the [L]egislature's priority of keeping Mobile and Baldwin County whole”:

JUDGE MANASCO: So is it possible to draw a map that satisfies the findings expressed in SB-5 with two opportunity districts?

[Counsel for the State]: I am not aware of a way to draw two majority-Black districts without going against the legislature's priority of keeping Mobile and Baldwin County whole.

JUDGE MANASCO: Okay.

Id. at 2647–49.

Dr. Duchin also testified about the description in the 2023 legislative findings of specific communities of interest. She opined that “[i]t is notable” that all the enumerated communities of interest are located in the geographic area that is contested in this litigation, and that in her view “[i]t seems implausible that a good-faith list of important communities in Alabama would completely exclude the Northern and Northeastern areas of the state.” Milligan Doc. 385-3 at 8. At trial, Dr. Duchin testified that “if you put aside general considerations like [preserving urban cores], the only named communities of interest that [she] took into account were the ones listed by the [L]egislature,” but that she “[c]ertainly” would have considered any other communities of interest the Legislature enumerated. Tr. 362–63.

Despite her concerns that the 2023 legislative findings are “novel” and “mathematically impossible to satisfy,” Tr. 359–62, Dr. Duchin testified that she met their requirements as far as mathematics allowed. See Milligan Doc. 385-3 at 9. For instance, though the 2023 Plan “codifie[d] a way of measuring county preservation that has never before been used in Alabama,” Dr. Duchin split the same number of counties in her Plan E that the 2023 Plan split; she testified that her Plan D “passes this test while containing two majority-Black districts”; and she testified that Alabama's 2011 Plan would fail this test. Milligan Doc. 385-3 at 4 tbl. 1, 7 n.3.

Dr. Duchin also responded to a criticism of her Plan E from Dr. Sean Trende, an expert witness for the State. Dr. Trende opined that Dr. Duchin's Plan E “not only divvied up the districts [in Jefferson County] by [BVAP] but has also split precincts by [BVAP].” Milligan Doc. 384-5 at 74–75. In response, Dr. Duchin testified that “the numbers that underlie these decisions also make it quite clear that the splits are not made for racial reasons.” Tr. 310. She explained:

There's another clear piece of evidence here, which is I've drawn, you know, thousands of White residents on the north of Birmingham and in the northern suburbs into District 7, more than 10,000, in fact. And that makes it really completely implausible that, you know, on the level of dozens of people on that line to the south the decision would be made in a race-conscious way.

Id. at 311–12.

Ultimately, Dr. Duchin testified that it is “unambiguously” possible to draw “an additional reasonably configured majority-Black district,” and that “[i]t can be done in many different ways, which elevate various of the traditional districting principles” and do not prioritize race over such principles. Id. at 312–14.

2. Gingles II and III – Racially Polarized Voting (Dr. Liu)

To establish that Black voters are “politically cohesive” and that each challenged district's White majority votes “sufficiently as a bloc to usually defeat [Black voters’] preferred candidate,” Allen, 599 U.S. at 18 (internal quotation marks omitted) (quoting Gingles, 478 U.S. at 51), the Milligan Plaintiffs rely on a racial polarization analysis conducted by expert witness Dr. Baodong Liu.

Dr. Liu works as a tenured professor of political science at the University of Utah, where he focuses on the “relationship between election systems and the ability of minority voters to participate fully in the political process and to elect representatives of their choice.” Milligan Doc. 385-4 at 3; Tr. 557. Dr. Liu holds a doctoral degree in political science from the University of New Orleans, a graduate degree in political science from Oklahoma State University, and an undergraduate degree in law from East China University. Tr. 557, 618. Dr. Liu has written or edited nine books and published articles in many peer-reviewed journals. Milligan Doc. 385-4 at 3. He has served as an expert witness in vote dilution cases in seven states and has advised the United States Department of Justice on methodological issues concerning racially polarized voting. Milligan Doc. 385-4 at 3. Dr. Liu has been compensated at $300 per hour for his work on this case and his compensation does not depend on the substance of his testimony. Milligan Doc. 385-4 at 2–3.

At trial, Dr. Liu was qualified without objection as an expert in racial polarization analysis, American political behavior, and ecological inference. Tr. 560.

The Milligan Plaintiffs asked Dr. Liu to opine about whether racially polarized voting (“RPV”) occurs in Alabama and has resulted in the defeat of Black-preferred candidates in Alabama congressional elections. Milligan Doc. 385-4 at 2. Dr. Liu also responded to the opinions of Dr. Hood and Dr. Bonneau, experts for the State. Tr. 562. Finally, Dr. Liu performed an effectiveness analysis “to show to the Court how different redistricting plans may provide opportunities for minority voters to elect the candidate of their choice.” Id. at 578–79.

Dr. Liu examined ten biracial endogenous elections – congressional elections in the challenged districts that included a Black candidate and a White candidate – based on cases indicating that such elections are more probative of racially polarized voting than are other elections. See Milligan Doc. 385-4 at 5–6 & n.3; Tr. 571; Wright v. Sumter Cnty. Bd. of Elections & Registration, 979 F.3d 1282, 1301 (11th Cir. 2020); Davis v. Chiles, 139 F.3d 1414, 1417–18 & n.3 (11th Cir. 1998). He also considered sixteen biracial exogenous elections – elections for statewide offices that included a Black candidate and a White candidate. See Milligan Doc. 385-4 at 6.

Dr. Liu studied racially polarized voting in these twenty-six elections by using a statistical method known as ecological inference (“EI”), which he opined “has been widely used as the most-advanced and reliable statistical procedure for [racially polarized voting] estimates in not only academic research but also voting rights cases in the last two decades.” Id. at 6–7. Dr. Liu testified that ecological inference is “one of the best methods in the history of political science.” Tr. 560, 564.

Dr. Liu examined both census data and data from the American Community Survey, and he saw consistent results in both datasets. Id. at 566–67.²¹ Dr. Liu emphasized that he did not focus on voters’ “motivations” because his role is “simply to provide empirical evidence for the Court” by evaluating whether “Black voters voted cohesively for their candidates and whether White voters[, as the majority of voters,] agree with their choice.” Id. at 567–69.

Dr. Liu testified that he focused on biracial elections because in that setting, “we can see truly what's the preference of either side, of Black or White voters.” Id. at 569. He testified that reliance on such elections is “accepted by [a] supermajority of experts, including Dr. Hood.” Id. at 569–570.

Dr Liu also testified about how he assesses patterns:

So using just one or two elections in a given year or just two or three or even five years, one has no sufficient evidence about whether that is a consistent pattern.

So, for me, 15 years is a reasonable time span for me to establish whether or not there is a consistent pattern for White and Black voters.

Id. at 572. For Dr. Liu, using a fifteen-year window also meant that he could utilize two cycles of census data. Id. at 572–73.

In his report, Dr. Liu opined that the data “clearly demonstrates that in biracial elections in which Black voters had the opportunity to express a preference for Black candidates, that preference was not shared by a majority of [W]hite voters in Alabama,” and that “[d]espite the high degree of electoral cohesion among Black voters, the majority of [W]hite voters form a voting bloc to typically defeat all the Black preferred candidates in these elections.” Milligan Doc. 385-4 at 15–16.

On cross-examination, Dr. Liu testified that he did not look at candidates’ partisan affiliations. Tr. 626–28. In the general elections in the challenged districts Dr. Liu studied (excepting District 7), Black support for the Black-preferred candidate always exceeded 87% and White support for the Black-preferred candidate never exceeded 12.6%. Milligan Doc. 385-4 at 7–8. Dr. Liu observed that the “only Black candidate who was able to win any biracial Congressional election in Alabama was Terri Sewell[,] who ran in a Black-majority congressional district,” District 7.²² Id. at 8. Dr. Liu provided this table to demonstrate both the existence and the extent of the racially polarized voting that he observed:

Id. at 7–8 tbl. 1.

At trial, Dr. Liu emphasized the clarity and starkness of the pattern that he observed, particularly in what he regarded as the most probative data set – biracial endogenous elections. See Tr. 573–74. Dr. Liu explained that in those elections, Black support for the Black candidates produces “almost uniformly the same finding,” which is that “Black voters provided more than 90 percent or so support for the Black candidate involved in those biracial endogenous elections,” id. at 573, that “White voters supported the Black-preferred candidates with minimum level[s]” ranging from “single-digit support” to “the 20-percent range,” id., and that the Black-preferred candidate was defeated in every election except District 7, which is majority-Black, id. at 574. Dr. Liu testified that he observed a similar pattern in the exogenous elections he studied, id. at 575–76, which provides “supplemental evidence” of racially polarized voting, id. at 571, and ultimately that voting in Alabama is “highly, highly racially polarized,” id. at 576.

Dr. Liu testified that in more than twenty years of research, “this is arguably the highest level” of racially polarized voting that he has “ever seen.” Id. And he said the level of racially polarized voting in Alabama was nationally distinctive:

With the RPV analysis based on EI as well as the exit polls that I was able to collect data from, I am very confident to conclude that, in the State of Alabama, there is [a] consistent pattern of racially polarized voting as shown by the literature concerning American voters, especially the [S]outh.

And I would characterize that as one of the highest in the nation.

Id. at 578. Dr. Liu based this conclusion both upon the extremity of the pattern of racially polarized voting and the strength of the evidence of that pattern. Id. at 680–81.

In his effectiveness analysis, Dr. Liu compared the performance of the 2023 Plan to the performance of Duchin Plan E, with a focus on Districts 2 and 7. Id. at 579; see id. at 665. Dr. Liu explained that he considered only biracial elections for this analysis, id. at 665–66, which yielded this conclusion:

The [2023] Plan continues to dilute the Black voter strength in [District 2] to ensure the defeat of Black-preferred candidates there. The Plaintiffs’ Duchin-E Plan, however, increases the opportunity of Black voters to elect the candidates of their choice in 10 of the 11 biracial elections analyzed in [District 2], and 11 out of 11 in [District 7].

Milligan Doc. 385-4 at 16.

Dr. Liu testified that Black voters had “no chance” of electing their candidate of choice in District 2 under the 2023 Plan, Tr. 581–82, and he offered the following chart with data to support that that opinion ²³ :

Milligan Doc. 385-4 at 12 tbl. 3.

On cross-examination, Dr. Liu acknowledged that this analysis did not examine the money spent by campaigns or candidates’ previous political experiences, i.e., candidate quality. Tr. 661–62.

In his rebuttal report, Dr. Liu responded to the opinions of Dr. Hood and Dr. Bonneau. See infra Part IV.D.3.a & Part IV.D.3.b. In response to Dr. Hood, Dr. Liu opined that the election of a lone Black Republican to a seat in the Alabama House (Representative Kenneth Paschal of District 73) “is not instructive or representative” because, due to low White turnout in that election, “it says little about whether [W]hite voters in Alabama embrace Black Republican candidates.” Milligan Doc. 385-8 at 3 (discussing “extremely low” turnout of 5.3% of the voting age population, and only 1.7% of the White voting age population); see Tr. 669–670.

Dr. Liu further opined that the 2024 Republican congressional primary in District 2 offers a better estimate of White support for a Black candidate, and indicates low support because in an election with four White candidates and four Black candidates, “[t]he four Black candidates finished 5th, 6th, 7th and 8th places after the election results were announced and together received only 6.2% of the total vote.” Milligan Doc. 385-8 at 4. “[A]ccording to the voter file data, as many as 95.9% of the 2024 Republican primary participants in [District 2] were non-Hispanic [W]hite voters while only 2.44% of the [District 2] primary participants were Black.” Id.; Tr. 594–95. At trial, Dr. Liu clarified that this rebuttal analysis was not a formal racially polarized voting analysis. Tr. 667–68.

Dr. Liu offers several responses to the opinions of the State's experts that party, not race, is the primary driver of Alabama voters’ electoral choices. At trial, Dr. Liu testified about Dr. Hood's discussion of Dr. Ben Carson's candidacy in the 2016 Republican presidential primary; Dr. Liu pointed out that Dr. Carson received 10% of the primary vote in Alabama, whereas Donald Trump received “more than four times as much.” Id. at 593–94. Dr. Liu testified that Dr. Hood's opinion about the importance of partisanship “is not grounded in the true empirical data.” Id. at 597–99 (discussing Milligan Doc. 403-17 at 2–3), 666–67.

In his rebuttal report, Dr. Liu opined that Dr. Bonneau's conclusion that Black candidates’ lack of electoral success in Alabama is attributable to party rather than race is flawed because (1) “Alabama [has] no official record of party registration available,” and (2) Dr. Bonneau “conducted no ecological inference analysis to measure the extent to which Black voters voted for the Democratic candidates.” Milligan Doc. 385-8 at 4.

Dr. Liu also analyzed two nonpartisan Montgomery mayoral runoff elections (2019 and 2023) to control for party. He opined that those elections demonstrated that when the “party ‘cue’ [was] taken away and only the racial cue remain[ed],” “it [was] race, rather than party, that drove the election outcomes.” Milligan Doc. 385-4 at 9–10; Milligan Doc. 385-8 at 8.

Dr. Liu also responded to Dr. Bonneau's assertion that patterns of straight-ticket voting in Alabama indicate that party is a more forceful driver of electoral choices than race. Dr. Liu pointed out that “Dr. Bonneau does not explain whether he has any knowledge of these voters directly, nor the racial identities of these straight-ticket voters nor localities/precincts the voters resided in.” Milligan Doc. 385-8 at 4; see Tr. 605–08.

At trial, Dr. Liu testified at length that “race is more important than party” in Alabama elections. Tr. 584–87. Dr. Liu testified about the 2021 District 1 Democratic primary in which James Averhart, a Black candidate, faced two other candidates that included a White candidate. Id. at 587–88. In that race, over 50 percent of Black voters supported Mr. Averhart, whereas only 16.7 percent of White voters supported Mr. Averhart. Id. Dr. Liu also examined the 2008 Presidential Election. In that instance, Dr. Liu found that exit poll data from Alabama showed that 51 percent of White Democrats supported Senator John McCain over then-Senator Barack Obama, a Black man; Dr. Liu thus testified: “So the White Democrats showed in their vote choice that race mattered instead of party.” Id. at 588–90 (referring to Milligan Doc. 403-13 at 14). Dr. Liu testified that he drew a similar conclusion from his analysis of the 2008 Alabama Democratic primary for president (between then-Senator Hillary Clinton and then-Senator Obama). Id. at 592–93.

Dr. Liu also testified that Dr. Hood's opinions are consistent with his conclusions. At trial, Dr. Liu quoted this scholarly statement by Dr. Hood about the importance of race in the South: “Race, especially the Black-White dichotomy, is the largest dividing line between the Republican and Democratic parties in the region. In fact, in terms of party identification, race dwarfs the effects of religion and class.” Id. at 586–87 (reading Milligan Doc. 403-17 at 5 (Ex. MX-17)). Dr. Liu also testified that Dr. Hood (1) “had a significant publication explaining why biracial elections are necessary” to evaluate racially polarized voting, id. at 570, and (2) in any event, did not dispute his finding that voting in Alabama is racially polarized, id. at 584. Dr. Liu provided greater detail in his rebuttal report about the self-contradiction he alleged Dr. Hood made about biracial elections:

[Dr. Hood] also failed to consider the race of the candidates. This is in [sic] contrary to his own professional recommendation when it comes to empirical analysis of vote dilution claims. In Dr. Hood's published article ․ the appropriate approach to a[ ] [racially polarized voting] analysis, according to Dr. Hood and his two co-authors, “must also consider the race/ethnicity of the candidates running for election. Of the elections available for analysis, the more relevant are those that feature a minority candidate from the racial/ethnic group suing the jurisdiction in question. For example, in a vote dilution suit brought by Latino voters, one would seek election contests featuring Hispanic candidates, while also keeping in mind the other criteria previously discussed.” Using biracial elections in vote dilution litigation research is a widely held standard by experts. But Dr. Hood did not follow this longstanding practice he himself recommended in his publications, and did not conduct any racial polarization analysis whatsoever.

Milligan Doc. 403-17 at 2–3 (Ex. MX-17) (citing and quoting M.V. Hood III et al., From Legal Theory to Practical Application: A How-To for Performing Vote Dilution Analyses, 99 Soc. Sci. Q. 536, 546 (2017)) (internal citation and footnote omitted).

3. The Senate Factors

The Milligan Plaintiffs next turn to the totality of the circumstances. They rely on stipulations of fact and testimony from two experts and several fact witnesses to support their arguments. We first discuss the stipulations and experts (upon which the Caster Plaintiffs also rely), and we discuss the fact witnesses in the next sections.

Recall that the nine Senate Factors are:

1. “the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the 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”;

7. “the extent to which members of the minority group have been elected to public office in the jurisdiction”;

8. “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

9. “whether the policy underlying the state or political subdivision's use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.”

Gingles, 478 U.S. at 36–37 (quoting S. Rep. No. 97-417 at 28–29).

a. Stipulations

The Milligan and Caster parties stipulated to several facts about the totality of the circumstances in Alabama today. They stipulated that in 1992, Representative Earl Hilliard was the first Black Alabamian elected to Congress since Reconstruction, Milligan Doc. 436 ¶ 103; Representative Shomari Figures is the first Black Alabamian to be elected to Congress outside of District 7 since Reconstruction, see id. ¶¶ 103, 106, 108, 113–14, 151; and “[t]hirty-two (32) out of thirty-three (33) Black Alabamians currently serving in the Alabama Legislature were elected from majority-Black districts,” id. ¶ 155. They also stipulated about the lone Black statewide official currently occupying elected office in Alabama, Judge Bill Lewis, whom “Governor Kay Ivey appointed ․ to the Alabama Court of Civil Appeals in early 2024” and “will eventually need to stand for election to continue holding the seat.” See id. ¶ 153.

b. Dr. Joseph Bagley

The Milligan and Caster Plaintiffs also rely on the expert testimony of Dr. Joseph Bagley about the Senate Factors. Dr. Bagley testified at the preliminary injunction stage of this litigation, and he incorporated his prior reports into his expert report disclosed for trial. Milligan Doc. 385-1 at 2.

Dr. Bagley holds graduate degrees in history from Auburn University and Georgia State University. Id. at 34; Tr. 1277–78. He works as an Assistant Professor of History at Georgia State University, where he focuses on “United States constitutional and legal history, politics, and race relations, with a focus on Alabama and Georgia.” Milligan Doc. 68-2 at 1. He has published a book, numerous articles, and been accepted as an expert in other Alabama voting rights cases. Id.; Tr. 1278. Dr. Bagley was compensated at a rate of $150 per hour for his work and his compensation did not depend on the substance of his testimony. Milligan Doc. 385-1 at 2. At trial, Dr. Bagley was qualified without objection as an expert in Alabama history, political analysis, race relations, and historical methodology. Tr. 1279.

The Milligan Plaintiffs asked Dr. Bagley to analyze the Senate Factors, which he did according to “standards of historiography.” Milligan Doc. 385-1 at 3. In addition, Dr. Bagley testified that for his May 17, 2024 report, he was “asked to look at the passage and enactment of [the 2023 Plan] and to offer [his] opinion as a historian as to whether its passage and drafting were motivated by discriminatory intent,” which does not include “get[ting] at the motivation of any one specific legislator.” Tr. 1422–23.

At trial, Dr. Bagley explained his understanding of the Senate Factors and the methods and sources he used in his analysis. Id. at 1279–82. Dr. Bagley opined about Senate Factors 1, 5, 6, 7, 8, and 9, and he considered Factor 3 in connection with Factor 1. Milligan Doc. 68-2 at 3–31; Milligan Doc. 385-1 at 30–32; Tr. 1282–83.

Dr. Bagley summarized his trial opinion about the Senate Factors this way: “Events and elections that have taken place in the interceding years [since the preliminary injunction] only confirm my prior conclusion that the totality of the circumstances demonstrates that Black Alabamians lack an equal opportunity to participate in the political process and elect candidates of their choice.” Milligan Doc. 385-1 at 32. He testified that the Senate Factors “evidence that [he has] marshaled here point[s] strongly in the direction of vote dilution ․ [and] towards discriminatory intent.” Tr. 1368. And Dr. Bagley testified that it was the “presence of ․ multiple factors rather than the size of any one particular gap” that supports his conclusion about the Senate Factors. Id. at 1411–12.

 Senate Factor 1 – History of Official Discrimination Affecting Political Participation

Dr. Bagley opined about numerous examples of what he described as official discrimination that impacted Black Alabamians’ political participation. Dr. Bagley testified about these and other lawsuits. Id. at 1285–88, 1377–78, 1432–33. He also testified about objections to Alabama's attempts to obtain federal preclearance for redistricting plans before the State's preclearance requirement was eliminated, and orders bailing into federal preclearance several Alabama jurisdictions under Section 3(c) of the Voting Rights Act. Id.

Dr. Bagley also testified about Alabama's closure of several driver license offices, which he opined was done to serve political purposes and affected voters because of Alabama's requirement for voter identification. Id. at 1288–90; Milligan Doc. 385-5 at 5–6. Dr. Bagley cited statements by the U.S. Department of Transportation “that [the closures] would have a disparate impact on Black citizens, particularly in the Black Belt.” Tr. 1291.

Dr. Bagley also testified about school desegregation cases in several Alabama school districts (Jefferson County, Huntsville City, and Madison County) that remain ongoing in 2025. See id. at 1291–92.

Dr. Bagley opined at length about the history of redistricting in Alabama. Tr. 1297; Milligan Doc. 385-1 at 3–22. His reports tracked extensive federal judicial involvement in and supervision of Alabama redistricting efforts since the passage of the Voting Rights Act, Milligan Doc. 68-2 at 8–16; Milligan Doc. 385-1 at 3–22. He concluded that “Alabama has an undisputed history of discrimination against Black citizens, especially when it comes to registering to vote, voting, and enjoying an equitable chance to participate in the political process, and this has been recognized by numerous courts.” Milligan Doc. 68-2 at 3.

“In particular,” Dr. Bagley continued, “[W]hite legislators of both major political parties have, in the last 50 years, manipulated the redistricting process to prevent Black citizens from electing members of Congress or, in the last 30 years, to limit Black voters’ ability to elect members of Congress from more than one district.” Id.; see Tr. 1314.

Dr. Bagley described Alabama's history of official discrimination in voting before the passage of the Voting Rights Act. He testified about the State's decision to split Mobile and Baldwin Counties in 1875 “for the express purpose of unseating Representative Harrelson, the second Black individual elected to Congress from the State of Alabama.” Tr. 1303–04, 1309–11. Dr. Bagley testified that this split continued for “nearly 100 years” “from 1875 until the 1970s,” during which time Mobile was paired with stretches of the Black Belt in a congressional district. See id. at 1304–06, 1309–11. In 1972, the two counties were reunited in the Democrats’ “Cherner Plan,” which Dr. Bagley testified occurred to limit the political power of Republican Congressman Bill Dickinson. Milligan Doc. 385-1 at 8–9; Tr. 1306–08. Dr. Bagley explained that the Democratic map drawers removed the White-flight suburbs in Baldwin County from Dickinson's Congressional District 2 to reduce his odds of winning the election. Tr. 1307–08. Dr. Bagley testified that a consequence of the Cherner Plan was that the BVAP in Districts 1, 2, and 3 “goes down to where it was in the [1960s] from 40 [percent] down to around 30 [percent] in all three of those.” Id. at 1308.

In short, “[t]here was an effort to unseat Mr. Dickinson. And that was part of the motivation for putting those Eastern Shore votes over into [ ] CD1. And the result ․ was a reduction in a BVAP across the three southern districts.” Id. at 1445. Dr. Bagley summarized it this way: “Mobile and Baldwin were, first, united in order to prevent the reelection of a Black incumbent and, 100 years later, reunited in for similar racial reasons.” Milligan Doc. 385-1 at 8.

Dr. Bagley reiterated the point:

[COUNSEL]: Okay. So, to summarize, the legislature's purpose in splitting [S]outh Alabama into three districts in the 1870s was to crack the Black vote?

[DR. BAGLEY]: Yes, sir.

[COUNSEL]: In the 1972 plan, the state's purpose in maintaining that three-way split, was also to continue cracking the Black vote; is that correct?

[DR. BAGLEY]: Absolutely.

[COUNSEL]: Okay. Does the 2023 plan also divide [S]outh Alabama into three congressional districts?

[DR. BAGLEY]: It does.

Tr. 1311–12.

Dr. Bagley further opined that a plan to give Black Alabamians two opportunity districts had been introduced by Fred Gray and failed:

Fred Gray, newly elected as one of the first two Black members of the state legislature since Reconstruction, proposed a plan that would give Black voters “a fighting chance” to elect someone “responsive to their needs” in two [congressional districts] by giving them roughly half the population of each. The legislature never seriously considered that plan and could not agree on any of the other four. It adjourned in September [1971] with no plan passed.

Milligan Doc. 385-1 at 8; Tr. 1308–09.

Dr. Bagley testified about early calls for the creation of a majority-Black congressional district in the 1980s cycle, and that by the 1990s cycle, “it was generally understood” that “there would need to be one, at least, majority-Black district drawn.” Tr. 1316. Dr. Bagley testified that there was “broad disagreement as to what that would look like.” Id. at 1317. Dr. Bagley traced the origin of the resulting litigation to Mr. Hinaman, who worked as a staffer for an Alabama Congressman and “recruit[ed] a plaintiff, a Republican in Mobile County named Paul Charles Wesch to be a named plaintiff.” Id.

Dr. Bagley also testified about disagreement in the 1990s cycle over whether there would be one or two majority-Black congressional districts. See id. As Dr. Bagley explained, the United States Department of Justice rendered an objection to the Legislature's plan on the ground that it limited the state to “only one majority-minority district.” Id. Dr. Bagley described the advocacy of Black leaders, including Earl Hilliard and Michael Figures (the late father of Congressman Figures) for the creation of a second majority-Black district. Id. at 1317–18. Dr. Bagley described how several of the plans proposed to include a second district by pairing “portions of Mobile ․ with portions of the Black Belt; in some plans, Mobile was paired with Montgomery and portions of the Black Belt; in some of those plans, Mobile was included with Tuscaloosa, for example.” Id. Dr. Bagley described how others “were calling for a more robust singular majority-minority district,” and that “there was a feeling among some that if you were going to create this very first majority-minority district, it would need to be somewhere around, say, 65 percent of Black Voting Age Population, which is ultimately what you end up with.” Id. at 1318.

Dr. Bagley testified that two state legislators (Larry Dixon and Sam Pierce) proposed a map in the 1990 cycle, and “it's the Dixon-Pierce Plan that ultimately gets adopted by the Wesch Court.” Id. at 1319–20. Dr. Bagley further testified that years later, Larry Dixon “was among those who were caught on a wiretap during the [public corruption investigation of someone else] in the 2000, 2010s, roughly, making racist remarks.” Tr. 1320. Dr. Bagley testified that “in particular, Mr. Dixon was referring to an effort to keep a gambling referendum off the ballot in order to drive down Black turnout. I think he insisted that, quote, unquote, illiterates would be bused to the polls in HUD, Housing and Urban Development, busses.” Id. Dr. Bagley also testified that Sam Pierce “admitted” during election litigation in the 2000 redistricting cycle “that -- both at that time and more recently that he, quote, referred only to census data and attempted to minimize the number of Black persons residing in districts he was designing to favor Republican candidates.” Id. at 1322.

Although Dr. Bagley acknowledged that “very few” (if any) legislators from the 1960s, 1970s, and 1980s cycles remain in the Legislature today, he testified that those previous cycles are “a part of historical causation” and the “historical context.” Id. at 1376–77. And he testified that although Mr. Hinaman did not draw any maps for Alabama in the 1990s, “he subsequently would be in each successive decade,” id. at 1319, up to and including the Community of Interest Plan the Alabama House passed in 2023, id. at 1327.

 Senate Factor 3 – Voting Practices or Procedures that Enhance Discrimination

Dr. Bagley “discuss[ed] the kind of enhancing devices and schemes covered in [Senate] Factor 3 in [his] treatment of [Senate] Factor 1, and [he] contend[s] that the discriminatory redistricting plans discussed therein are as much exemplary of the devices highlighted by this factor as the at-large schemes and numbered place laws of the (somewhat) more distant past.” Milligan Doc. 68-2 at 3.

 Senate Factor 5 – Effects of Discrimination that Hinder Political Participation

Dr. Bagley opined in his report that “Black citizens in Alabama lag behind their [W]hite counterparts in nearly every statistical socioeconomic category, due largely to a history of discrimination,” and that these disparities adversely affect Black voters’ “ability to engage politically.” Id. at 17–26; see also Milligan Doc. 385-1 at 2. At trial, Dr. Bagley testified that such gaps are “substantially the result of past discrimination” because “as a historian, there's no other explanation for this -- these kinds of widespread myriad disparities other than the history of discrimination.” Tr. 1396–97. To assess this, Dr. Bagley explained that historians consider such questions as: “Did someone grow up during a time when schools were segregated or under a desegregation order? Did someone grow up in a community with endemic violence? Did someone face discrimination in trying to find a job or what have you?” Id. at 1398.

In his reports and at trial, Dr. Bagley opined about numerous socioeconomic categories and measures. These include racial disparities in poverty rates, living conditions, health outcomes, educational attainment, income, employment, and home ownership. To summarize, Dr. Bagley opined in his expert report:

Today, [W]hite Alabamians with more education and therefore higher income can afford a car, internet service, a personal computer, or a smart phone; they can take time off from work; they can afford to contribute to political campaigns; they can afford to run for office; they have access to better healthcare. Education has repeatedly been found to correlate with income [and] independently affects citizens’ ability to engage politically. Black people in Alabama are demonstrably poorer, less educated, less healthy, and far more likely to be incarcerated than [W]hite people as a consequence of past and continuing racism and discrimination.

Milligan Doc. 68-2 at 17.

In describing his opinions, we focus our attention on those disparities we ultimately regard as most relevant. See infra Part V.A.4.

Dr. Bagley opined at length about the extreme Black poverty in the Black Belt. Dr. Bagley described a 2019 United Nations report that “Black residents [in certain Black Belt counties] lacked proper sewage and drinking water systems and had unreliable electricity,” and that “[r]esidents had constructed homemade water delivery systems using PVC pipe, did not have consistent access to drinking water that had not been tainted by raw sewage, and often fell ill, entire households at a time, with E. Coli and hookworm.” Milligan Doc. 68-2 at 21. He also cited recent actions by the United States Department of Justice to explain that “Alabama was very recently found to have discriminated against Black residents by failing to address a chronic lack of access to clean drinking water not tainted by failed septic tank systems.” Milligan Doc. 200-15 at 2.

Dr. Bagley also opined about the schools that Alabama has evaluated as “failing” in reading and math proficiency. As Dr. Bagley explained, a 2013 Alabama statute establishes criteria for labeling “the bottom 6 percent of the state's schools, by proficiency in reading and math, as ‘failing,’ ” and that “[f]or 2020-2021, as in previous years, all 75 schools on the list of failing schools were majority Black, most overwhelmingly so.” Milligan Doc. 68-2 at 24–25 (footnotes omitted). Dr. Bagley testified that “[m]ost of the schools are in majority-Black school systems in or around Birmingham, Montgomery, and Mobile, or in the Black Belt.” Id. Dr. Bagley also opined that the COVID-19 pandemic had a disparate impact on Black school children in the Black Belt because they had limited internet access. Id. at 18.

Dr. Bagley testified about a discriminatory root cause of this problem in the Black Belt – that “[p]roperty tax laws prevent taxes on timber land from adequately funding the Black Belt's public schools, despite [m]any efforts of local Black leadership (elected thanks to enforcement of federal voting rights law) to raise millage rates,” and that the timber land “has been blanket-owned by [W]hite people since at least Indian Removal and the rise of the Cotton Kingdom in the early 19th century, if not the early 18th century when French colonists introduced African slave labor to the region.” Milligan Doc. 385-5 at 18.

 Senate Factor 6 – Racial Appeals in Political Campaigns

Dr. Bagley testified that he considers a racial appeal in a political campaign to occur when a candidate is making an appeal that would “drive to continue to racially polarize voting.” Tr. 1404–05. In other words, Dr. Bagley “think[s] the racial appeal is something that would tend to only motivate one race of voters.” Id. at 1433.

Dr. Bagley opined that White officials in Alabama “learned long ago to colormask their public statements,” that his analysis of campaign ads, speeches, and social media “reveal that direct invocations of race still appeal to [W]hite voters,” and that “campaigns and politicians’ public statements have recently trended back towards more overt racial appeals,” Milligan Doc. 68-2 at 3, 26–27.

Dr. Bagley offered several examples of racial appeals in Alabama campaigns. Id. at 26–28. These include: (1) former Congressman Mo Brooks's “repeated[ ] claim[s] that Democrats are waging a ‘war on Whites’ by ‘claiming that Whites hate everybody else’ ”; (2) former Alabama Supreme Court Chief Justice Roy Moore's 2017 acclamation of the antebellum period in the South (“I think it was great at the time when families were united – even though we had slavery. They cared for one another. People were strong in the families. Our families were strong. Our country had a direction.”); (3) former Chief Justice Moore's 2011 radio interview in which he stated that the amendments to the Constitution that follow the Tenth Amendment (including the Thirteenth Amendment, which abolished slavery, the Fourteenth Amendment, which requires States to provide equal protection under the law to all persons, and the Fifteenth Amendment, which provides that the right to vote shall not be denied or abridged on the basis of color or previous enslavement) have “completely tried to wreck the form of government that our forefathers intended”; and (4) former Congressman Bradley Byrne's ad depicting four public figures who are persons of color in a campfire. Id. at 27–28.

Dr. Bagley further opined that since we ordered the use of the Special Master Plan, “[W]hite candidates have also used racial appeals while running for the newly redrawn Second Congressional District.” Milligan Doc. 385-1 at 31. Dr. Bagley points to comments by two White Republican candidates. We highlight only one: candidate Dick Brewbaker running a campaign ad featuring “former Harvard president Claudine Gay, a Black woman, juxtaposed with images of [Mr. Brewbaker's] young relatives brandishing firearms, while he intones that ‘the media and woke corporations and liberal politicians sow division for their own profit.’ ” Id.

 Senate Factor 7 – Minority Electoral Success

Dr. Bagley testified that only three Black people have ever held any statewide office in Alabama, and that none holds statewide office presently ²⁴ or has held such office in the last twenty years. Milligan Doc. 68-2 at 29. At trial, Dr. Bagley testified about the 2024 Republican primary election in Congressional District 2 and how four Black candidates finished behind a White candidate despite all having more political experience than the White candidate. Tr. 1292–93. Dr. Bagley also testified that no Democrats hold statewide office in Alabama. Id. at 1406. Dr. Bagley opined that “Black candidates have had some success at the local level, thanks to litigation and federal government intervention.” Milligan Doc. 68-2 at 3.

 Senate Factor 8 – Elected Officials’ Responsiveness to Particularized Minority Needs

Dr. Bagley reasserted his previous testimony about occasions when he argued that elected leaders in Alabama failed to respond to the particularized needs of Black Alabamians. Milligan Doc. 385-5 at 6; see Tr. 1275–76. He opined that many of the discriminatory experiences he identified in connection with Senate Factor 5 evince Alabama's lack of responsiveness to the particularized needs of Black Alabamians, including the failure to ameliorate living conditions in the Black Belt or to improve healthcare coverage among Black households. Milligan Doc. 68-2 at 30–31.

Dr. Bagley also testified about three specific instances when elected leaders in Alabama failed to respond to the particularized needs of Black Alabamians. First, he testified that the state's response to the pandemic failed to respond to the needs of the Black community, and he argued that the state's distribution of vaccines was inequitable. Id. at 29; Tr. 1408–10. On cross examination, he conceded that Alabama did not choose its initial vaccine sites in a racially discriminatory manner, and that those choices were instead due to the “relative lack of viable hospital facilities in parts of the state like, say, the Black Belt.” Tr. 1408–410. Dr. Bagley clarified that his opinions about the pandemic were based on statements by Alabama's State Health Officer, Dr. Scott Harris, about the disparate impact of the pandemic on “people that already have other social determinants like chronic health problems or issues just related to education and income.” Tr. 1408–410. Dr. Bagley testified that he did not have “any reason” to dispute that “more Black citizens proportionately had received the vaccine than [W]hite Alabamians.” Id. at 1410.

Second, Dr. Bagley testified that elected leaders have failed to respond to the Black community's needs on environmental issues. He explained that in 2015, “Black residents of Uniontown, in Perry County, fought a decision by the state to allow 4 million tons of potentially toxic coal ash to be transferred from the site of a coal-fired electrical plant accident in Tennessee to a landfill in the town,” and the Black residents “met resistance from the state's Department of Environmental Management.” Milligan Doc. 68-2 at 21. He also testified that the Alabama Department of Environmental Management and Attorney General have “consistently [been] opposed” to “remov[ing] and replac[ing] soil laden with toxic materials from airborne and waterborne pollution emanating from nearby factories” in “the 35th Avenue area in North Birmingham,” which the Environmental Protection Agency has deemed a priority for cleanup. Milligan Doc. 68-2 at 21–22.

And third, Dr. Bagley opined that Alabama's lack of responsiveness to the needs of Black people is “exemplified” by the Legislature's failure to draw a second majority-Black congressional district after this Court's order to do so. Id. at 29; Tr. 1296–97, 1326–43; see also Milligan Doc. 385-1 at 30. Dr. Bagley opined that “[t]he 2023 events are an extension of the state's history of discrimination, especially as to redistricting.” Milligan Doc. 385-1 at 3. Dr. Bagley supported this opinion with various sources, including media articles. See e.g., Milligan Doc. 403-4 at 28–29; Tr. 1342–43. One of those articles (about which Representative Pringle also testified, see supra Part I.I.3.c) quoted Alabama House Speaker Nathaniel Ledbetter as saying about the 2023 Plan: “If you think about where we were, the Supreme Court ruling was 5-4, so there's just one judge that needed to see something different. And I think the movement that we have and what we've come to compromise on today gives us a good shot.”²⁵

Dr. Bagley opined about talking points that, according to Senator Livingston, the Solicitor General drafted. See supra Part I.I.3.b; Milligan Doc. 385-1 at 26. Dr. Bagley explained that those talking points emphasized the treatment of communities of interest. He quoted them as saying: “The Livingston Plan is a Compact, Communities of Interest Plan that applies the State's traditional districting principles fairly across the State. The 2023 Plan is a historic map that gives equal treatment to important communities of interest in the State, including three that have been the subject of litigation over the last several year – the Black Belt, the Gulf, and the Wiregrass.” Milligan Doc. 385-1 at 26. The talking points also said: “No map in the State's history, and no map proposed by any of the Plaintiffs who challenged the 2021 Plan, does better in promoting any one of these communities of interest, much less all three.” Id.

Dr. Bagley opined that the talking points about communities of interest (and their echoes in the State's litigation position) were “disingenuous” in their “touting of only splitting the Black Belt into two” districts because “Black voters did not appear, according to the analyses available to legislators, to have the ability to elect a candidate of choice in any [congressional district] other than [District 7].” Id.

Dr. Bagley testified about historical aspects of the communities of interest that the Legislature named in the 2023 Plan. He testified that the Legislature's reference in the 2023 legislative findings to the “French and Spanish Colonial heritage” of the Gulf Coast counties is a reference to White people. Tr. 1455–56. He also testified about the shared history of Mobile and the Black Belt:

If you study Mobile and the Black Belt both now and historically, you can see that there's a very real and significant shared history. If we are talking about historical cultural factors, socioeconomic factors, Mobile and the Black Belt have a shared history of, of course, chattel slavery, of emancipation, of Reconstruction, of Redemption, of the struggle for basic civil and voting rights in the 20th Century. And those are [a] very real connection, indeed, historically speaking.

Beyond that, I talk about very significant socioeconomic commonalities between the sort of urban core of Mobile and the Black Belt. And then, finally, also ties of migration. There have been waves of migration from the Black Belt to Mobile that I discuss. And then, also, you know, part and parcel with that, there are familial ties that even legislators discuss during the process of debating what would ultimately become the 2023 plan.

Id. at 1299–1300.

And he testified about the Wiregrass. He explained that historically the Wiregrass has “meant a region in [S]outheast Alabama that had relatively few Black folks relative to White folks with the exception of perhaps the City of Dothan itself or that metropolitan area.” Id. at 1312. He opined that “if you take a broader definition of [the Wiregrass] beyond just sort of Dothan, Ozark, Enterprise, it bleeds into or overlaps with the Black Belt.” Id. Dr. Bagley further testified about “the shared history, shared socioeconomic characteristics, history of migration, and familial ties” between Dothan and the Black Belt. Id. at 1313. Dr. Bagley also said that the Wiregrass is an “archaic term” to describe the region because the actual wiregrass “has long since ceased to exist there,” and “it has previously denoted a much larger region than it would tend to denote now.” Id. at 1312.

 Responses to the State's Arguments

Dr. Bagley also responded to three of the State's arguments. First, he responded to the opinions of one of the State's experts, Dr. Adam Carrington. Dr. Bagley opined that Dr. Carrington's conclusions “do not withstand historical and contemporary scrutiny.” Milligan Doc. 385-5 at 2. At trial, Dr. Bagley testified that Dr. Carrington does not discuss Alabama specifically and devoted “very little” analysis to Alabama even though partisan realignment[²⁶ ] did not occur “identically across the South.” Tr. 1344–46. Dr. Bagley testified that Dr. Carrington did not mention many key figures in Alabama political history (such as Fred Gray, Hank Sanders, Earl Hilliard, Vivian Figures, Richard Shelby, Fob James, Jeff Sessions, Bob Riley, Mike Hubbard, and Kay Ivey) and said that “you can't understand realignment in Alabama if you don't focus on those key figures.” Id. at 1346–47.

Dr. Bagley also testified that Dr. Carrington's failure to consider the experiences of Black voters in Alabama significantly undermined Dr. Carrington's analysis. Tr. 1347–49; see also Milligan Doc. 385-5 at 7–8. Dr. Bagley testified that “Black citizens fighting their way into the political process in the ‘60s and ‘70s, the formation of Black caucuses, formation of coalitions, the actual access to political power on the part of Black lawmakers -- all that is critical in understanding party realignment, in my opinion, in Alabama specifically.” Tr. 1348.

Ultimately, Dr. Bagley testified that Dr. Carrington's opinions are “not very helpful, I don't think, for us in terms of explaining the racial dynamics of partisan realignment in, specifically, Alabama.” Id. at 1362.

Second, Dr. Bagley responded to the opinions offered by one of the State's experts, Dr. Wilfred Reilly. Dr. Bagley opined that Dr. Reilly overlooked important context for his opinions. Id. at 1362–64. For example, Dr. Reilly opined that “the idea that Mobile has a natural link to the Black Belt region of Alabama because she was populated largely by African American refugees from that region – or, at very least, those fleeing shared abuse historically – seems at least debatable,” and he cited statistics showing that Mobile lost Black population to cities outside the South during the relevant time. Milligan Doc. 384-4 at 15–17. In response, Dr. Bagley explained that “Mobile was experiencing immigration from the Black Belt at the same time it was experiencing emigration to cities outside the South. The historian Wayne Flynt has described the former as a ‘hemorrhaging’ of people that, along with [W]hite flight from Mobile, left behind a ‘topography of despair’ in both Mobile and the Black Belt.” Milligan Doc. 385-5 at 19. At trial, Dr. Bagley testified that “[e]ven if you were to look at the numbers and say, okay, well, there were, on balance, more people leaving the state for cities in the [M]idwest or the [N]ortheast, that doesn't make it insignificant that, at the same time, there were people migrating out of the Black Belt to the City of Mobile.” Tr. 1364.

And Dr. Bagley cited the “riverine connections between Mobile and the Black Belt” to push back on Dr. Reilly's argument “that Mobile and Baldwin County are more intimately linked than Mobile and the Black Belt.” Milligan Doc. 385-5 at 17, 19; see Milligan Doc. 384-4 at 3–4. Dr. Bagley opined that “[r]ivers like the Alabama, Tombigbee, Black Warrior, and Mobile have provided transportation connecting Alabama's Black Belt to the Gulf of Mexico for centuries,” and that “[i]t remains an important entrepot today, exporting not just timber from the Black Belt, but also soybeans, livestock, cotton, and automobiles (manufactured at the Hyundai assembly plant in Montgomery), in contrast to Dr. Reilly's assertion that this connection is tenuous.” Milligan Doc. 385-5 at 19.

Third, Dr. Bagley responded to the State's argument that partisanship rather than race causes racially polarized voting in Alabama. He described evidence that many Black Alabamians identify as conservative Christians: “[I]f you look at what I cite to in this report, you would note that a lot of Black Alabamians are deeply religious, they're Christian, they consider themselves conservative, they consider themselves, perhaps, even fundamentalist conservatives or, more importantly, evangelical conservatives, and, yet, they do not vote for Republican candidates.” Tr. 1360. Dr. Bagley described evidence of Black Alabamians’ conservative Christian stances on abortion and same-sex marriage, and he testified that “if we were to try to say, well, it's not race; it's simply conservatism or it's moderation, it's I don't want to be associated with policies that are liberal, then these numbers don't bear that out either.” Tr. 1361–62.

c. Dr. Traci Burch

The Milligan and Caster Plaintiffs also rely on the testimony of Dr. Traci Burch about the totality of circumstances. Dr. Burch holds a doctoral degree in Government and Social Policy from Harvard University. Milligan Doc. 385-2 at 3; Tr. 920. She works as a political science professor at Northwestern University, where she has taught for seventeen years, and is a research professor for the American Bar Foundation. Tr. 920, 922; Milligan Doc. 385-2 at 39. Dr. Burch was compensated at a rate of $400 per hour for her work and her compensation did not depend on the substance of her testimony. Milligan Doc. 385-2 at 4.

Dr. Burch has published books, chapters, and peer-reviewed articles on race, political participation, and voter turnout. Milligan Doc. 385-2 at 39–41; Tr. 922–24. During trial, Dr. Burch testified about the peer-review process, wherein “experts in a field who would be well-equipped to evaluate the methods and the theoretical framing of a book or an article” are selected (based on reputation) to review an article or book and provide a report of their opinion, which is sent to the editorial board or publisher of the article or book. Tr. 923–24. This is an anonymous process, wherein the reviewer is not made aware of the author's identity. Id. Dr. Burch has “served as a peer reviewer for all of the major journals in political science.” Id. at 923.

Dr. Burch is currently the editor-in-chief of the Law and Social Inquiry journal, id. at 919, and she has qualified as an expert witness in fourteen lawsuits. See Milligan Doc. 385-2 at 48–49; Tr. 925. At trial, she was qualified “as an expert in political and social science and political behavior” without objection. Tr. 927.

 Senate Factor 5

Dr. Burch testified principally about socioeconomic disparities between White and Black Alabamians and how those affect political participation. See id. at 927. Dr. Burch testified that “socioeconomic variables have consistently been related to political participation and voting participation throughout the political science literature.” Id. at 929. She opined that racial disparities exist in educational attainment, income, unemployment, healthcare, access to transportation, and access to internet, all of which affect voting participation. Id. at 928, 943, 946; Milligan Doc. 385-2 at 7–37. She testified “that there is a racial gap in voter registration and turnout in Alabama,” Tr. 969, and that “socioeconomic disparities in Alabama ․ are either causally related to voter turnout or associated with voter turnout,” Tr. 972.

Dr. Burch testified that, on average, Black Alabamians have lower educational attainment than White Alabamians, “caused, in part, by historical and contemporary discrimination in elementary, secondary, and higher education that make Black Alabamians less likely to have graduated from high school and college.” Milligan Doc. 385-2 at 12; see Tr. 934–35. She testified that “educational attainment has been shown over and over again by political scientists to be the most important predicter of voting” and that “the relationship between education and voting isn't just associational; it's causal.” Tr. 929; Milligan Doc. 385-2 at 11.

More specifically, Dr. Burch testified about educational disparities among school-aged children. She testified that in 2022, “only 9 percent of Black Alabama 8th Graders were proficient in reading, compared with 30% of White Alabama 8th Graders. Likewise, only 7% of Black Alabama 8th Graders were proficient in Math, compared with 27% of White Alabama 8th Graders.” Milligan Doc. 385-2 at 12 (footnotes omitted). And she testified about the effects of such disparities, explaining that “in the [majority-Black] Black Belt especially, there are ․ disproportionately high illiteracy rates, as high as 30 percent.” Tr. 938, 998, 1049.

In her expert report, Dr. Burch observed that 17% of Black individuals aged 25 and older in the Black Belt counties do not have a high school diploma, compared to just 11% of White individuals in those counties. Milligan Doc. 385-2 at 15. Additionally, the percentage of Black individuals aged 25 and older in the Black Belt with a bachelor's degree is only 17%, compared to 27% of their White counterparts in the region having a bachelor's degree. Id. at 16.

Dr. Burch testified that Alabama's history of segregated public schools still impacts voting participation today. Tr. 936. She observed that “in 2020 ․ 38.6 percent of votes in the Alabama general election were cast by people age 60 and older. So those were people who were at least school age in 1970 when Alabama still maintained those separate and unequal schools for Black and White students.” Id. Dr. Burch testified that segregation resulted in fewer opportunities for Black people to attend college and access educational resources, which is why Black people today “are disproportionately concentrated in these lower educational attainment -- lower voter turnout groups.” Id. at 937–38.

Dr. Burch opined that lower educational attainment impacts other socioeconomic factors that also affect voting rates for Black Alabamians. See Tr. 938; Milligan Doc. 385-2 at 19–20. She testified that “[s]tatewide Black [un]employment is more than twice as high as White unemployment.” Tr. 952. Specifically, Dr. Burch noted that the unemployment rate for Black people aged 16 and older in the Black Belt counties was 10%, compared to only 4% for White people aged 16 and older in those counties. Milligan Doc. 385-2 at 24. Similarly, she testified that the median household income in Black households is lower than White households statewide and in every county she analyzed. Id. at 21; Tr. 950–51. Data from the American Community Survey shows that “the median household income for Black Alabama households is $36,104, compared with $62,545 for White Alabama households.” Milligan Doc. 385-2 at 21. She testified about “a gap of tens of thousands of dollars statewide” which contributes to racial disparities in family poverty, access to internet, and access to transportation, which in turn hamper voting participation due to an inability to read ballots, learn about candidates, absentee vote, locate voting information, and travel to polls. See Tr. 938, 950, 952, 956–57, 960–62.

Dr. Burch also opined about the socioeconomic commonalities between the Black Belt and Mobile. She testified that “the Black median income in Mobile City is similar to the Black Belt counties.” Id. at 951. Mobile City has a Black median income of $34,088, which is similar to the Black median income in many of the Black Belt counties that Dr. Burch analyzed. Id. Additionally, as Dr. Burch noted in her expert report, “significant proportions of people in neighboring Black Belt counties work in Mobile County and vice versa.” Milligan Doc. 385-6 at 8. Specifically, “34% of people who live in Washington County, 16% of people who live in Clarke County, and 11% percent of people who live in Monroe County work in Mobile County.” Id. Further, Mobile City's Black family poverty rate of 23% is close to the average Black poverty rate of 24% across all the Black Belt counties. Id. at 9.

Dr. Burch testified that in Alabama, “Black family poverty is nearly three times as high as White family poverty,” and that “about 24 percent of Black families live below the poverty line compared to seven percent of White families.” Tr. 957. She testified that in Greene County (in the Black Belt), “40 percent of Black families ․ live below the poverty lines compared with five percent of White families,” and that the situation is similar in Perry County (also in the Black Belt). Id. Dr. Burch also noted that 30% of Black households in the Black Belt receive SNAP or food stamps, compared to only 8% of White households. Milligan Doc. 385-2 at 26.

Additionally, Dr. Burch testified that “about 17 percent of Black families don't have a computer in the household” compared to 10 percent of White households without a computer, and “26 percent of Black households don't have Internet access at home compared with 14 percent of White households.” Tr. 960; Milligan Doc. 385-2 at 27–28. She explained that these disparities are heightened in some counties. She testified that in Escambia County (sometimes considered Black Belt), “43 [percent] of Black households don't have access to the Internet at home compared with 26 percent of White households.” Tr. 960. She added that the situation is similar in Crenshaw County (in the Black Belt). Id. at 960–61.

Dr. Burch testified that “statewide, Black Alabama households are more than twice as likely to lack access to a vehicle at home than White households,” id. at 961, and that in Hale County (in the Black Belt), “16 percent of Black families don't have access to a car at home compared with three percent of White families,” id. at 962. She observed that the situation is similar in Dallas County (also in the Black Belt and home to Selma). Id. Moreover, in the Black Belt counties she studied, 12% of Black households did not have vehicle access compared to just 4% of White households. Milligan Doc. 385-2 at 29.

Dr. Burch also testified that Black Alabamians are in demonstrably worse health than White Alabamians, and she gave as examples that (1) the infant mortality rate for Black infants is nearly three times higher than the rate for White infants, and (2) Black Alabamians have a shorter life expectancy rate than White Alabamians. Tr. 968. According to the Alabama Department of Public Health, in 2022, the Black infant mortality rate was 12.4 deaths per 1,000 births compared to a White infant mortality rate of 4.3 deaths per 1,000 births. Milligan Doc. 385-2 at 30 n.75. Regarding life expectancy in Alabama, Black women are expected to live 77.6 years while White women are expected to live 78.8 years, and Black men are expected to live 69.9 years while White men are expected to live 73.5 years. Id. at 30.

Dr. Burch further testified that Black Alabamians are more likely to be uninsured than White Alabamians, Tr. 964, and explained that even with insurance, “it can be difficult to find medical care in the Black Belt” because there are few hospitals, and “[s]ome counties ․ don't have a hospital at all,” id. at 967. In particular, 18% of Black Alabamians in the Black Belt lack health insurance whereas 14% of White Alabamians in the region lack health insurance. Milligan Doc. 385-2 at 31. Moreover, the Black Belt's “Lowndes, Perry, and Pickens Counties do not have a hospital at all.” Id. at 30. Dr. Burch testified about “several studies that ․ have associated poor health with lower voter turnout.” Tr. 968.

Dr. Burch also testified about how race shapes individual perspectives. See id. at 1043–44. She opined that how a White person and a Black person view the economy and choose to vote “is different based on racial considerations and racial attitudes.” Id. at 1044. She argued that “being concerned about the price of eggs and the way that you weight concerns about the price of eggs are related to ․ racial threat and racial anxiety.” Id.

 Responses to State's Arguments

Dr. Burch also responded to the opinions of Dr. Reilly and Dr. Carrington. Milligan Doc. 385-6 at 2. Dr. Burch addressed Dr. Reilly's assertion that “racial disparities between Black and White Alabamians are the result of cultural practices of Black people rather than systemic discrimination.” Tr. 974. She criticized Dr. Reilly for, among other things, failing to support his assertions with peer-reviewed evidence or address contrary literature. Id. at 978–80.

Dr. Burch also addressed Dr. Carrington's claim that “clear correlations between race and voting in Alabama are caused by differences in nonracial policy preferences among racial groups rather than racial attitudes or racial policy preferences.” Id. at 985. Dr. Burch testified that “racial identity and racial attitudes sha[p]e partisanship and party cohesion, and these two phenomen[a] have become increasingly linked together since 2008.” Id. Dr. Burch observed that “Dr. Carrington doesn't really engage with the literature examining this relationship in recent years.” Id. at 987. She testified that “partisan sorting in the electorate ․ was a reflection of racial attitudes rather than income or other non-race-related partisan preferences,” id. at 986, and that “the only factor that led to party realignment in the [S]outh” was race, id. at 1039–40. Dr. Burch cited a wide variety of literature in her report providing “support for the notion that the contemporary partisan alignment stems from the positioning of the two parties on the issue of race and civil rights.” Milligan Doc. 385-6 at 10–11. Dr. Burch opined that “[r]esearch that examines mass and elite partisanship from 2008 onward finds strong evidence of both partisan sorting and issue polarization along the lines of race in the electorate.” Id. at 11.

Dr. Burch also criticized “Dr. Carrington's arguments that trace[ ] differences in partisanship and vote choice to religiosity” because “Black people in Alabama traditionally are highly religious and even higher proportions of Black people in Alabama identify as evangelical or born-again.” Tr. 988. Rather than connecting partisanship to religiosity, “[r]ecent studies have shown that party and race are linked in the American mind.” Milligan Doc. 385-6 at 14. As Dr. Burch summed it up:

In conclusion, the literature clearly supports the point that party and candidate choice is shaped by racial identity and racial attitudes in the electorate. This relationship has been strengthening in recent years. To say that factors other than racial considerations explain the voting patterns along racial dimensions in Alabama, as Dr. Carrington argues, ignores the past fifteen years of evidence in the literature that race and racial attitudes drive partisanship and vote choice. Racial attitudes are becoming more salient to partisanship and vote choice and vote choice over time.

Id. at 14–15.

4. Fact Witnesses at Trial

The Milligan Plaintiffs offered five fact witnesses at trial.

a. Evan Milligan

The Milligan Plaintiffs offered testimony from named plaintiff Evan Milligan. Mr. Milligan is Black, lives in Montgomery County, and was forty-three years old during trial. Tr. 1152, 1238. He testified that he spent part of his youth in Birmingham and Montgomery and has lived in Montgomery off and on for approximately 32 years. Id. at 1152–53. The Montgomery neighborhoods where he grew up were “overwhelmingly Black.” Id. at 1154. He testified that Black and White people in Montgomery still live in different areas, a pattern he attributed to “the plantation economy” and “enslavement.” Id. at 1155.

Mr. Milligan testified about his “very vivid memory” of sitting on the steps of the Dexter Avenue King Memorial Baptist Church in Montgomery as a seven-year-old and watching a Ku Klux Klan rally. Id. at 1159–60. Looking back further, Mr. Milligan recounted his family's experience under Jim Crow segregation. Id. at 1175–76. He testified that his mother, born in 1952, remembers segregated drinking fountains. Id. at 1176. He also testified about Oak Park near her home in Montgomery – a segregated park that “contained a zoo, a swimming pool ․ a skating rink, [and] a merry-go-round.” Id. He explained that “after integration ․ the zoo was removed, the swimming pool was cemented, [and] the carousel and the merry-go-round” were removed. Id.

Mr. Milligan testified about his experience attending public magnet schools, which he described as schools where the curriculum is “one grade level ahead of where those student's peers would be in a nonmagnet program.” Id. at 1160–62. He testified that in the schools he attended, magnet programs were about “40 percent Black” while nonmagnet programs “were 90 to 92 percent Black.” Id. at 1161. Mr. Milligan also offered other observations about the stark racial differences in educational opportunities and attainment: he said that more Black peers than White peers began working as teenagers “not just for allowance money ․ [but] to contribute to household bills”; that his “Black peers were definitely more familiar with community-based violence, particularly gun violence”; and that “for students who exited the track towards traditional high school graduation ․ there was an overrepresentation of Black people.” Id. at 1167–68.

Mr. Milligan testified that as an adult, he has “primarily worked in the nonprofit sector in terms of full-time work here in Alabama,” but he has also worked “lots of part-time jobs to help supplement.” Id. at 1170. Mr. Milligan testified that he has worked for a tax referendum campaign, as an organizer with the Federation of Child Care Centers of Alabama, for the Equal Justice Initiative in Montgomery, as the founding executive director of Alabama Forward (a “coalition of private nonprofit organizations” that works on voting issues), and now directs the Jubilee Community Center in Montgomery (a nonprofit community-based arts organization). Id. at 1170–72. He explained that this experience gives him an understanding of the needs of Black communities in Alabama, he has been to all 67 counties in Alabama, and he has visited “similarly situated” Black communities in the Mississippi Delta, Georgia, Florida, and Louisiana. Id. at 1172–73.

Mr. Milligan testified that in his experience, there are people in the Black community who “are conservative on issues about sexuality, abortion, gender, and sometimes even in terms of government services.” Id. at 1174. These individuals have “an awareness” that Black communities “have been treated very differently by local, state, [and] federal governments,” “so they're very suspicious of government having overarching control of what people can do in their bedrooms, with their bodies, [and] in their faith houses.” Id.

Mr. Milligan testified about the relationship between Montgomery County and other Black Belt counties. Tr. 1175–76. He testified that “Montgomery is the Capit[a]l of the state, but ․ also ․ it's sort of the capital of the Black Belt.” Id. at 1175. “Montgomery has been a place that has drawn families from throughout ․ all the corners of the Black Belt and [S]outh Alabama.” Id. He also testified about connections between Montgomery and Mobile. Id. at 1177–78. For example, he described playing tuba in his high school marching band (the Sidney Lanier Marching Poets) and testified that “every year, [the band] looked forward to going to Mobile to participate in Battle of the Bands and in Mardi Gras parades.” Id. at 1178.

Mr. Milligan also testified about racial disparities in access to healthcare in Alabama. Id. at 1186–88. He offered his experience as a patient at a Montgomery hospital that has filed for bankruptcy, and explained that if it closes, Black Alabamians will be negatively impacted because it will be “yet another hospital in the Black Belt that has closed,” and such closures mean “less access to prenatal care, mental health services ․ [and] emergency care.” Id. at 1187–88.

Finally, Mr. Milligan testified about his involvement in this litigation and what he thinks it says about the State's responsiveness to the needs of Black Alabamians. Id. at 1196. He testified that he attended Committee hearings during the 2023 Special Session, “hoping that the state would set a different example as far as how the State of Alabama can comply with federal court orders, particularly around voting rights and Civil Rights issues.” Id. at 1199. He testified that the interests of the Black community “were manifested in what was presented to this Court. This Court made a ruling. The state appealed. The Supreme Court made a ruling. And the state didn't comply with that ruling.” Id. at 1203. He continued: “So not only did [the State] not take into account what we had asked for, it didn't take into account what this Court and the Supreme Court had actually ordered them to do.” Id.

Regarding the language of the 2023 legislative findings concerning the “French and Spanish colonial heritage” in the Gulf Coast, App. B, Mr. Milligan pointed out that he didn't “know of Mardi Gras existing in France and Spain” and that “one of the critical ingredients of [Mardi Gras] was the cultures and the traditions of the enslaved African people that also lived in” Alabama. Tr. 1206. In his view, “Mardi Gras was the real expression of those cultures coming together -- French, Spanish, West African, [indigenous] Alabamian.” Id. at 1178.

b. Shalela Dowdy

The Milligan Plaintiffs also rely on the testimony of Shalela Dowdy, one of the Milligan Plaintiffs. Ms. Dowdy is Black, was born in Mobile in 1989, and grew up in predominantly Black Prichard, north of Mobile. Id. at 16–17. She graduated from the United States Military Academy, still serves in the Army where she holds the rank of major, and works as a regional organizer for the nonprofit organization Black Voters Matter. Id. at 16–17, 23, 40.

Ms. Dowdy attended three elementary schools – two that were predominantly Black, and one that was not. Id. at 21. Ms. Dowdy testified that “[t]he infrastructure at [the White school] was better than the infrastructure at the Black schools within the Mobile city limits that [she] attended,” and “the interactions with the teachers and the time that the teachers were able to spend with the students” were different. Id. at 22.

Ms. Dowdy testified she has “ties to the Black Belt” through family and work. Id. at 27. She said that “in Mobile, many citizens do not have cars,” the “public bus transportation system ․ is very limited,” and that in the Black Belt, “a lot of citizens do not have transportation.” Id. at 28. Ms. Dowdy added that “Prichard deals ․ with a failing water infrastructure issue” while “Lowndes County and other particular areas of the Black Belt ․ hav[e] to live off of septic tanks.” Id. at 28–29. “Poverty is an issue in the Black Belt,” as are “food deserts.” Id. at 29. Ms. Dowdy also said “that there's a lack of hospitals in the Black Belt,” that there is one “hospital left that ․ [is] set to close, and so those citizens in the Black Belt are having to travel to Montgomery and Mobile for healthcare needs and for healthcare services.” Id. at 30–31.

Ms. Dowdy testified that Mardi Gras celebrations in Mobile “are segregated.” Id. at 39. She said that the Colored Carnival, a Mardi Gras association established in the 1930s, was created “because Black people were not allowed to participate in Mardi Gras festivities with the White citizens of Mobile.” Id.

Ms. Dowdy also offered her own experience working on political campaigns: “Fundraising for Black candidates typically tends to be an issue” because “a lot of Black people are living paycheck to paycheck, living in poverty, dealing with a state that has a minimum wage at 7.25.” Id. at 50.

Finally, Ms. Dowdy testified that her previous Congressman, Jerry Carl, who is White, was not responsive to the Black community. She attempted to meet him several times but did not see him campaigning in the Black community. Id. at 50–52. She is now represented by Congressman Figures, who did campaign in her community. Id. at 55. She explained that he had “signage throughout the community,” “[she] saw him at multiple community events” including at churches, and he “had a strong presence in our community.” Id.

c. Robert Clopton

The Milligan and Caster Plaintiffs also rely on the testimony of Robert Clopton, who is Black and was born in 1954 in Sipsey, Alabama. Id. at 236–37. Mr. Clopton testified that around 1958, his family moved to a community called Colony in Cullman County, Alabama, where they worked as sharecroppers. Id. at 237. They “lived in the house of” the White owner and worked on his farm. Id. at 238. Mr. Clopton said that “when the farms were harvested, they received wages predicated on the profits.” Id. These wages were not “enough to sustain the family,” so sharecroppers had to “make a loan” with “the crop owner,” which meant being “in debt to him” and being in “bondage.” Id. at 241. Mr. Clopton described sharecropping as “a vicious cycle.” Id.

Mr. Clopton testified that Cullman County had sundown towns. Id. at 237, 241. He said that he was told as a child “not [to] go to those towns” because “people were beaten, hung, et cetera et cetera.” Id. at 241. Mr. Clopton added that “[t]here were signs that warn[ed] against being in town after sundown.” Id.

Mr. Clopton attended segregated public elementary schools for five years. Id. at 239, 242. His school had a hose for drinking water and outhouses for restrooms and he explained that children at his school “had to make the fires at school in order to keep warm.” Id. at 243. He testified that they “had to go to the coal pile regardless of the weather to get coal to put on the fire throughout the course of the day.” Id. He also testified that students “had used books each and every year.” Id. Mr. Clopton added that “in the spare times,” the children worked and took “a week out of classes each year just to pick cotton.” Id. at 240. Mr. Clopton first attended school with White children in 1966. Id. at 242. At that school, Mr. Clopton testified that students had new books, indoor restrooms, an indoor gymnasium, and radiator heat. Id. at 243–45. As he put it, “there was no comparison” between the segregated school and integrated school. Id. at 244.

Mr. Clopton lived in the Black Belt for five years. Id. at 248–49. According to Mr. Clopton, “[t]here was no public transportation” in the Black Belt, id. at 253, or “public housing in Marengo County, Clarke County, [and] Wilcox County,” id. at 254. People from the Black Belt traveled to Mobile for healthcare because the available healthcare in the Black Belt was “poor” and was surrounded by “horror stories.” Id. at 255–56. He offered that in 1989, when his wife experienced a pregnancy complication and he took her to a Black Belt hospital, her pregnancy was nearly terminated by a doctor who did not even examine her. Id. at 256–57. He said that after he drove nearly 100 miles to a Mobile hospital, a doctor examined his wife and told her to resolve the complication by “stay[ing] off [her] feet for a week or two.” Id. at 257. Mr. Clopton testified that “every time [he] see[s] [his] 35-year-old daughter, [he] remember[s] that night when someone tried to take her life.” Id.

d. Letetia Jackson

The Milligan Plaintiffs also rely on the testimony of plaintiff Letetia Jackson, who is Black and grew up in Dothan (in the Wiregrass). Id. at 682–83. She testified that her “great grandmother was a slave on a plantation in Barbour County,” and her “grandfather is the slave owner's son through [her great] grandmother.” Id. at 683.

Ms. Jackson attended segregated public schools until “the eighth grade.” Id. at 684. She said that after Black schools were closed, Black students “had to be bused over” to the White public schools, and the White community “immediately created a private academy” for White students. Id. at 685. When she began attending integrated schools, Black students “were treated unfairly” and “weren't welcomed.” Id. at 686. “The teachers automatically assumed that ․ we came from a school that used hand-me-down books, that we were not smart, that we were inferior.” Id. She offered the view that the teachers “were very dismissive.” Id.

She added that “[e]verything was segregated” when she was growing up in Dothan, id. at 691, and that she grew up in “a public housing project[,]” id., where the racial demographics were “100-percent Black,” id. at 691–92. Her neighborhood “had a lot of dilapidated housing, boarded-up housing, [and] overgrown lots,” “had landfills ․ [and] railroad tracks where the train ran through the middle of the community,” and “[t]here was an actual acid plant that was located right in the neighborhood where you could just smell the fumes every day.” Id. at 692. In contrast, she said the other “side of town where most of the White citizens lived,” id. at 693, “was pristine” and had “good electrical grids,” id. at 692.

Ms. Jackson suggested that she still sees segregation in Dothan today and that “a lot of what we have is vestiges ․ of the old Jim Crow.” Id. at 693. She lives in an upper class, predominantly Black neighborhood, id. at 695, but “the same infrastructure exists in the neighborhoods today that existed then” — they “still have the same railroad tracks, still have the same landfills” and “have been in most recent years fighting against another landfill.” Id. at 696.

Ms. Jackson explained that “[t]here is no public transportation in Dothan,” and “if you don't have a car, you have to ․ know somebody with a car, you have to walk, or have to figure it out.” Id. at 698. She testified that the lack of transportation impacts Black residents’ political participation because “having access to transportation to get to your polling place, to be able to do the kinds of things that you need to do to register to vote” is vital. Id. at 726.

Ms. Jackson also testified about the lack of broadband and cell towers in the rural Black Belt. Id. at 715–16. She said that when she was in Butler County recently, “for about 25 minutes or so, [she] had no cell phone service.” Id. at 715. She explained that the lack of broadband impacts the ability of Black residents to participate in the political process because without service, voters cannot “look up [their] polling place,” “determine whether [they are] actually still registered to vote,” determine whether they “were being purged off of the voter list,” or “download a voter registration ․ form.” Id. at 727–28.

Ms. Jackson testified at some length about her participation in civic organizations. Id. at 705–06. She is “the convenor of the South Alabama Black Women's Roundtable,” a “life member of the NAACP Dothan chapter,” “the treasurer of the Downtown Dothan Redevelopment Authority,” and serves in several other organizations. Id. Ms. Jackson testified about her work on healthcare issues with South Alabama Black Women's Roundtable. Id. at 706, 709. She explained that “the lack of expansion of Medicaid” and the closure of “about 15 hospitals” primarily in “rural counties” has severely affected “[p]rimarily poor, Black residents.” Id. at 709–10. She said that this impacts political participation, because “if you don't have ․ the basic foundation of healthy living, it's hard for you to be able to do anything else.” Id. at 710.

Ms. Jackson also testified about her interactions with political campaigns and elected officials. See id. at 718–24. She said that her previous Congressman, Barry Moore, who is White, was unresponsive to the needs of the Black community. See id. at 721–23. She testified that “the majority of his town halls or his constituency service meetings are usually in a chamber of commerce or at a country club,” and that “to go to a country club for a meeting with your representative is probably not something most Black folks are going to do.” Id. at 722–23. Ms. Jackson also testified that she has “[n]ever” seen Congressman Moore attend a Black community event. Id. at 723.

Finally, Ms. Jackson testified that before the 2023 Special Session, she attended the Committee's hearings because she wanted to tell the Committee “that we are a better state when we have all of our residents participating.” Id. at 728–30. When Ms. Jackson spoke at the hearing, she testified that White Committee members “didn't pay any attention at all.” Id. at 730. “They were talking to each other or looking down ․ [their] body language was like hurry up and get it over with.” Id. She testified that “[t]hey didn't listen,” “[t]hey didn't care,” and “they had already made up their minds what they were going to do.” Id.

Ms. Jackson observed that this legislative hearing “was markedly different” and “almost hostile” compared to others that she had attended in the past. Id. at 742. In Ms. Jackson's view, the White legislators “did not give any respect whatsoever to the Black legislators that sat on that commission with them.” Id. Ms. Jackson said that the White legislators “didn't even allow their Black colleagues to even know what they were doing at the time․ [T]heir Black colleagues had never even seen the map and they're on the commission.” Id. at 730.

e. Janice Malone

Next, the Milligan and Caster Plaintiffs rely on the testimony of Janice Malone, who is Black and was born in Mobile in 1955. Id. at 1130–31. She has lived in the City of Mobile since 1992. Id. at 1131. She now lives in the Toulminville neighborhood, which is “98 percent Black.” Id. at 1132. Ms. Malone testified that her “family has lived in Alabama for five generations,” and that her husband's cousin is Vivian Malone Jones, who “defied segregationists and then Governor George Wallace, who was attempting to deny her an education” at the University of Alabama. Id. at 1132, 1136.

Ms. Malone grew up with segregation and attended segregated public schools until high school. Id. at 1134–35. She testified that “you had to go to the back of the restaurant, for the restaurants that would serve you, to get food.” Id. at 1135. There were certain water fountains that “you just were not allowed to drink out of” and “[t]here were stores that did not allow Black people to shop.” Id.

Ms. Malone testified about Vivian's Door, an organization she founded and now directs with a mission “to help minority businesses grow, scale, and reinvest in their marginalized communities and help residents that are not only marginalized but who face systemic poverty, who have low venture capital, and have faced racial segregation.” Id. at 1136. She explained that “[m]ost of the businesses [that Vivian's Door] serve[s] are Black.” Id. Ms. Malone described how her clients have been denied financial assistance from banks, sometimes with no stated reason. Id. at 1137. But when she reviewed loan applications for these clients, she saw no reason for the bank's denial because “[t]he paperwork ․ evenly matched.” Id. In contrast, she saw approximately twenty to twenty-five similar loan applications from White applicants who received bank approval. Id. at 1137–38.

Finally, Ms. Malone testified that her previous Congressman, Mr. Carl, was unresponsive to the needs of the Black community. See id. at 1141–42. She is currently represented by Congressman Figures, who is more involved and in touch with the needs of the Black community. See id. at 1142–43. She expressed the view “that Black Alabamians need to have someone who has lived experiences, who reflects them, knows the needs of their communities, knows what the community wants and deserves to represent them.” Id. at 1143.

5. Designated Deposition Testimony

The Milligan Plaintiffs offered deposition testimony from six witnesses. Milligan Doc. 259.

a. Representative Sam Jones

Alabama Representative Sam Jones ²⁷ is a member of Alabama's Permanent Legislative Committee on Reapportionment (the “Committee”). Milligan Doc. 459-9 at 5. Representative Jones lives in Mobile (in District 2 under the Special Master Plan and District 1 under the 2023 Plan). Id. at 7–8.

Representative Jones graduated from a segregated public high school in 1967, joined the Navy, later worked for Mobile Community Action (a nonprofit agency), and served on the Mobile County Commission for nearly 20 years, after which he was elected Mayor of Mobile. Id. at 9–11. He was then elected to the House and appointed to the Committee in 2018. Id. at 12–13.

Representative Jones testified that the 2021 redistricting process was “rushed” and that “information [given] to the members was limited.” Id. at 16. He added that out of 100 plans submitted to the Committee, some 23 of them were presented to all Committee members. Id. at 16, 18–19. Representative Jones described his efforts to determine how those 23 plans were chosen and his doubts that he could access the other plans even though he was not explicitly barred from doing so. Id. at 19–21.

Representative Jones testified that he voted against the 2021 Plan because it did not create “another seat in the Congress that represents the Black community of the State of Alabama.” Id. at 22. He offered the view that “unfortunately in our state, partisanship is related to race ․ not all over the country but in our state. So another way to make sure that Blacks don't get elected, you just draw a partisan district and you don't get elected.” Id. at 23.

Representative Jones testified about overlapping interests and communities in the Black Belt and Gulf Coast. He said that “probably the majority” of people in Baldwin County, and “some” people in Washington, Escambia, and Monroe Counties, work in Mobile because “Mobile County is the economic hub for the whole region.” Id. He explained that Mobile and Baldwin Counties are linked economically, and that “a lot of people migrate from Black Belt counties to Mobile County.” Id. at 29–31. He also testified that downtown Mobile has a stronger connection to Montgomery than to other parts of Mobile County. Id. at 32.

Representative Jones testified that he did not support the 2023 Plan and does not know what role (if any) race played in drawing it. Id. at 26.

b. Randy Hinaman

Randy Hinaman ²⁸ is a political consultant, lobbyist, and cartographer in Alabama. Milligan Doc. 459-6 at 6–7. He left Cornell University to work on former President Reagan's presidential campaign. Id. at 6. Mr. Hinaman has extensive experience redistricting for Alabama: he drew the 1992 congressional map adopted by the Wesch court, advised Republican Legislators in the 2000 cycle and worked on Alabama's 2002 plan, drew maps for Alabama in the 2010 cycle, drew the 2021 Plan, and drew the Community of Interest Plan in 2023. Milligan Docs. 459-6 at 6–9, 459-7 at 4; Ala. Legis. Black Caucus, 231 F. Supp. 3d at 1038.

Mr. Hinaman testified about Alabama's 1992 congressional map. Milligan Doc. 459-6 at 9. He said that he drew “District 7 with the intent to make it a majority Black district” by “includ[ing] areas of high concentration of African American voters” and relying on racial data down to the census block level. Id. Mr. Hinaman testified that race was a “major factor” in drawing the lines for District 7, in addition to geography, population deviation, and contiguity. Id. at 10. Mr. Hinaman said he drew the “finger” that extends District 7 into majority-Black areas of Jefferson County and that it was drawn “partially” because of “where the incumbent lived at that point. But also to create a majority[-]Black district.” Id. at 44.

Mr. Hinaman testified that Alabama's congressional maps from 2001 to 2021 can all be traced back to the 1992 map. Id. at 11. He testified that in drawing the 2011 Plan, he was “updating the 2001 map based on demographic changes” and utilizing new census data. Id. He used the 2011 Plan as a starting point for the 2021 Plan. Id. For the 2021 Plan, Mr. Hinaman employed census data, consulted Alabama's incumbent United States Representatives, referred to the 2021 guidelines, and reviewed election returns. Id. at 11–12.

Mr. Hinaman testified that each member of Alabama's United States House delegation agreed to “put in” $10,000 through their campaigns for him to draw the 2011 map. Id. at 12. This included Congresswoman Sewell, who “wanted to maintain” her majority-Black district. Id. Mr. Hinaman testified that Senator McClendon and Representative Pringle asked him to draw Alabama's 2020-cycle maps, and he was paid for that work by an organization that legislative leadership created, under a contract for him to receive $200,000. Id. at 14–15.

Mr. Hinaman further testified that although official census data was delayed until the “end of August” of 2021, work on the 2021 Plan began “in earnest” in May 2021 based on census estimates. Id. at 15–16. He said that he worked on the 2021 Plan in the Committee's office, and he described public feedback from various areas of the state, id. at 21, 25–26. Mr. Hinaman testified that he did not look at racial data in 2021 “until the week before” the Committee co-chairs, their counsel Mr. Walker, and Mr. Hinaman submitted plans to the Legislature, at which point those persons “did turn on race and look at the racial breakdowns in the various maps.” Id. at 26.

Mr. Hinaman testified that the proposed plan “completed the week before the [2021] special session is identical to the version of the map that was ultimately enacted.” Id. at 29. Mr. Hinaman testified about communications between him and Congresswoman Sewell about District 7, during which he told her that the BVAP in District 7 was 54.22%, and she requested that certain precincts, institutions, and areas be included in District 7. Id. at 29–31. For instance, Mr. Hinaman testified that Representative Sewell wanted to keep the University of Alabama (in Tuscaloosa) in her district and pick up Maxwell Air Force base and Alabama State University (in Montgomery), as well as some precincts in Homewood (in the Birmingham area). Id. at 30–31.

When asked about the factors he considered in 2021, Mr. Hinaman testified that “no plan is going to respect all” communities of interest, that “there are trade-offs,” and “you can't satisfy all communities of interest.” Id. at 40.

Mr. Hinaman also testified about his work on the 2023 Plan, and we already have described that testimony. See supra Part I.I.3.a.

c. Senator James McClendon

Senator McClendon ²⁹ is a White Republican born in 1943 in Mobile. Milligan Doc. 459-15 at 5. When Senator McClendon was a member of the Alabama House, he was a member of the Committee and co-chaired it in 2011. Id. at 8–9. He testified that “probably the single most important role of the attorney is to help the elected members of this committee know what the law is and ․ keep us up to date on recent court cases so we can do our best to be in compliance with what the law says.” Id. He testified that his duties as chair had “to do with making sure that [they] stay in compliance with the courts and the law and recent court cases.” Id. at 9.

Senator McClendon testified that during the 2011 redistricting process, “[t]he map drawer met with and talked to the members of the congressional delegation” about updating district lines. Id. The Legislature held 22 public hearings. Milligan Id. He testified that the Committee's role “was to take the map that was submitted ․ with the approval of the congressional delegation, and to approve or disapprove that map and submit it for introduction to the legislature.” Id. at 10.

Senator McClendon also testified about the 2011 guidelines. Id. He offered his “belief that [the Committee] followed the guidelines” by “consult[ing] with the attorney and with the person drawing the map to make sure that they were following the rules that [they] had before [them].” Id. at 11. He testified that Committee members “would talk about [the guidelines] from time to time” and that “it was just so well known that [they] followed the guidelines” because that was their “job.” Id.

Senator McClendon testified that he did not know why there was only one majority-Black district in 2011 and had not considered a plan with two such districts because the issue was not brought before the Committee. Id. at 12–13. Senator McClendon was then shown a 2011 news article in which he commented on a plan that would have created two majority-minority districts. Id. at 13. In the article, Senator McClendon commented that the plan “would lead to ‘retrogression,’ or a retreat from minority population benchmarks set by the [D]epartment of [J]ustice.” Id. Senator McClendon testified that he could not “recall making that statement,” and that if there were a plan “that complied with the redistricting guidelines and created two majority minority districts in 2011,” he “would certainly have considered it.” Id. at 14.

Senator McClendon testified that he did not know why Section 5 of the Voting Rights Act applied to Alabama in 2011. Id. at 12. And he testified that he does not know what it means for a map to comply with Section Two and did not “do any work to monitor whether [the map drawer's] work complied with the Voting Rights Act.” Milligan Doc. 459-16 at 10.

Senator McClendon also testified about the 2021 redistricting process, when he served as Senate chair of the Committee. Milligan Doc. 459-15 at 15. He explained that planning for that process began “two years ․ ahead of time,” and he had some involvement in coordinating public hearings. Id. at 15–16.

Senator McClendon testified that he complied with the 2021 guidelines, and that it was his “job to ensure compliance with the Voting Rights Act of 1965.” Id. at 17–18. He said that he accomplished this by “count[ing] on these experts[, the map drawer and attorney,] that were working for [him] and working for the committee to follow those guidelines and be familiar with the court cases and with the law and with the rulings.” Id. at 18. He testified that they discussed the Voting Rights Act “several times” and “don't use racial data” to draw lines. Id. at 18–19.

Senator McClendon further explained that at a Committee meeting on October 26, 2021, they discussed “racial polarization analysis.” Id. at 20. He described a racial polarization analysis as “an extra test tacked on to what we normally do to see if, in fact, we are in or out of compliance with the Voting Rights Act and our own guidelines and the court cases.” Id. He testified that he said at the meeting that “[t]he [BVAP in District 7] is sufficient[, 54%,] to where you don't need a” racial polarization analysis. Id. at 21. He also testified that Representative England asked what “the relationship between the 54 percent ․ and the actual results or potential results of a racial polarization study” was, to which Senator McClendon responded, “I got no clue.” Id. at 22.

Senator McClendon testified about his 2021 votes against maps introduced by Senators Singleton and Hatcher. Id. at 25. Senator Hatcher's plan contained two majority-minority districts, and Senator McClendon said he voted against it because it paired two incumbents. Id. He testified that the BVAP “had nothing to do with” his vote. Id. He voted against Senator Singleton's map because it contained two districts where “no minority candidate had a majority of the voters.” Id.

Finally, Senator McClendon testified that “Black voters in Alabama tend to vote for Black candidates,” and “that Black and [W]hite voters in Alabama in general have different views on the preservation of confederate monuments” and “about the prevalence of racial discrimination.” Milligan Doc. 459-16 at 20.

d. Senator Steve Livingston

Senator Livingston is a White Republican Senator from Scottsboro and co-chairs the Committee.³⁰ Milligan Doc. 459-13 at 4. He testified about the 2023 Plan, and we already have described that testimony. See supra Part I.I.3.b.

e. Christopher Brown

Christopher Brown ³¹ is a member of the Alabama Republican Party and has chaired the Jefferson County Republican Party since 2023. Milligan Doc. 459-3 at 6. Mr. Brown has extensive work history in Alabama Republican politics. See id. At the time of Mr. Brown's deposition, he worked as a political consultant and president of RedState Strategies, a political consulting firm that serves candidates for office in Alabama. Id. at 7–8. Mr. Brown testified that the firm offers advice about potential legislation and “general redistricting information,” and its clients include state Senators (he named Senators Livingston, Sam Givhan, Jabo Waggoner, and Dan Roberts). Milligan Doc. 459-3 at 8–9, 14.

Mr. Brown testified that he advised Senators Givhan, Roberts, and Livingston about redistricting after the 2020 Census. Id. at 10–11. He testified that he gave advice in both 2021 and 2023, did not work on any congressional maps in 2021, and worked on congressional maps in 2023. Id. at 11, 16–18. We have already described his testimony and text messages between him and Senator Livingston. See supra Part I.I.3.b.

f. Representative Chris Pringle

Representative Pringle serves as a Republican representative of Mobile in the Alabama House.³² Milligan Doc. 459-19 at 6. He initially served from 1994 to 2002 and then returned in 2014; he is Speaker Pro Tempore of the House and co-chaired the Committee in both the 2021 and 2023 Special Sessions. Id. at 6; Milligan Doc. 459-20 at 3–4.

Representative Pringle testified that in 2021, the Committee worked with members to draw state legislative districts, while Mr. Hinaman worked with Alabama's congressional delegation to draw the congressional map. Milligan Doc. 459-19 at 9. Representative Pringle testified about some of the public hearings the Committee held in 2021. See, e.g., id. at 23–24.

Representative Pringle described his understanding of the guidelines as “the parameters that we used in order to draw districts we thought complied with the Voting Rights Act and the 14th amendment to the Constitution and the court rulings.” Id. at 15. He described the guidelines as “a road map for everybody to follow when we're drawing lines.” Id.

Representative Pringle testified that in 2021, “Mr. Hinaman was directed by the [C]ommittee to follow the guidelines and to draw those plans race neutral, without looking at race until after he had developed a plan,” and that no racial polarization analysis was performed on maps prior to the Committee hearing on October 26, 2021. Id. at 28–29. He testified that in 2021 the Committee desired to maintain the “core” of districts, which meant keeping District 7 largely the same as was adopted in 1992. Id. at 31.

Representative Pringle agreed that Republican and Democratic views differ “when it comes to the view of whether there's a significant amount of discrimination against Black individuals in the state.” Id. at 32.

Representative Pringle also testified about the 2023 Special Session and 2023 Plan, and we already have described that testimony. See supra at Part I.I.3.c.

6. Testimony from State Senate Redistricting Trial

The parties stipulated that we “may consider trial transcript testimony from Alabama State Conference of the NAACP v. Secretary of State Allen, Case No. 2:21-cv-1531-AMM (N.D. Ala., pending), as to eight witnesses.” Milligan Doc. 441 at 2. Of those eight witnesses, the Milligan plaintiffs offer three. See Milligan Doc. 393.

a. Scott Douglas

Scott Douglas is the Executive Director of Greater Birmingham Ministries. Milligan Doc. 441-4 at 8. He testified that Greater Birmingham Ministries has two forms of membership: (1) organizational, which primarily includes churches, and (2) individual. Id. at 11. He identified members in Montgomery County who are Black registered voters. Id. at 14, 29–30. He testified that Greater Birmingham Ministries views its case as a success for establishing District 2 “as a Black opportunity district in the State of Alabama.” Id. at 42.

b. Bernard Simelton

Bernard Simelton is a Black registered voter in Limestone County who serves as president of the Alabama State Conference of the NAACP (“the State Conference”). Milligan Doc. 441-7 at 8–10. Mr. Simelton testified that Black members of the Legislature have met with the Huntsville NAACP to discuss Medicaid expansion and access to healthcare for Black Alabamians, but he has not seen any White legislators at those events. See id. at 36. Finally, he said that the only elected officials who have met with the NAACP about civil rights issues are Black. Id. at 20–29.³³

c. Tari Williams

As the organizing director at Greater Birmingham Ministries, Tari Williams works to restore voting rights to Alabamians. Milligan Doc. 441-8 at 9–10. She testified that Greater Birmingham Ministries disproportionately serves Black men to restore voting rights. Id. at 11–12. She explained that Greater Birmingham Ministries also offers workshops for individuals with literacy issues and that the majority of participants in those workshops are Black. Id. at 17. Ms. Williams testified that she has observed racial disparities in Alabama in educational opportunities, transportation, internet access, and ability to purchase food, clothing, or medications. Id. at 18–19, 21–23. She testified that in her opinion, the Legislature has not been responsive to the needs of Black Alabamians. Id. at 20–21.

B. Caster Plaintiffs’ Arguments

In addition to the evidence admitted in Milligan, the Caster Plaintiffs also rely on evidence they developed about their Section Two claim. See Milligan Doc. 444 at 2; Caster Doc. 356 at 2.

1. Gingles I – Numerosity and Reasonable Configuration (Mr. Cooper)

Numerosity is stipulated in Caster as it was in Milligan. See Milligan Doc. 436 ¶ 152 (joint stipulations in both cases).

To establish that it is possible to draw a second reasonably reconfigured majority-Black district, the Caster Plaintiffs rely on the expert testimony of Mr. Bill Cooper. See Caster Docs. 352-1, 352-2, 48, 65. The Caster Plaintiffs asked Mr. Cooper to “look at the population in Alabama statewide and determine whether the Black population is sufficiently numerous and geographically compact to allow for the creation of two majority-Black congressional districts.” Tr. 109.

Mr. Cooper holds an undergraduate degree in economics from Davidson College and has earned his living for the last 39 years by drawing maps, both for elections and demographic analysis. Caster Doc. 352-1 at 59; Tr. 105–07. He has extensive experience testifying in federal courts about redistricting and has been qualified in 57 voting rights cases, including two recent Alabama cases (Alabama Legislative Black Caucus, No. 12-cv-691, and Chestnut v. Merrill, No. 2:18-CV-00907). Caster Doc. 352-1 at 2–3, ¶¶ 3–4, 9. Since he filed his expert report, he has testified in two more cases. Tr. 106–07. Mr. Cooper was compensated at a rate of $150 per hour for his work in Caster, and his compensation did not depend on the substance of his testimony. Caster Doc. 352-1 at 1.

At trial, Mr. Cooper was qualified as an expert in redistricting, demographics, and census data with no objection. Tr. 107.

In Mr. Cooper's initial report, he provided statistics about Alabama and demographic changes that have occurred here since the 2010 census. See Caster Doc. 352-1 at 7–12, ¶¶ 18–27. Mr. Cooper reported that according to 2020 census data, Alabama's Black population increased by 83,618 residents, which is a 6.53% increase in Alabama's Black population since 2010, and 34% of the state's entire population increase since then. Id. at 9, ¶ 21. In the same period, Alabama's White population shrunk from 67.04% to 63.12% of its total population. Id. at 8 (In the 1990 census data used in Wesch, Alabama's White population was 73.65% of its total population. See Wesch, 785 F. Supp. at 1503 app. B.).

Mr. Cooper also offered eight illustrative remedial plans in his initial trial report: the seven plans he offered during the preliminary injunction proceedings, and one additional plan that he drew “to demonstrate that [he] could closely track the Special Master's plan.” Tr. 115; Caster Doc. 352-1 at 43, ¶ 105. Each of Cooper Plans 1–8 includes two congressional districts (Districts 2 and 7) with a BVAP over 50%. Caster Doc. 352-1 at 26–45, ¶¶ 62–107. In all the majority-Black districts in these plans, the BVAP is between 50% and 52%, except that in two plans, the District 7 BVAP is between 53% and 54%. See id. Cooper Plans 1–8 appear below:

Id. at 27 fig. 10.

Id. at 29 fig. 12.

Id. at 32 fig. 14.

Id. at 34 fig. 16.

Id. at 36 fig. 18.

Id. at 38 fig. 20.

Id. at 41 fig. 22.

Id. at 44 fig. 24.

Later, Mr. Cooper offered Cooper Plan 9, which he drew to rebut Dr. Trende's assertion that Cooper Plans 1–8 were drawn “strictly based on race and because of that one can't produce a plan that is reasonably compact,” and to establish “that a plan can be drawn that is as compact if not more so than the [2023 Plan] and the Special Master plan.” Tr. 116; Caster Doc. 352-2 at 6–7, ¶¶ 11–12.

Mr. Cooper opined that Cooper Plan 9 “place[s] greater emphasis on compactness, while still respecting other traditional redistricting criteria, including population equality, contiguity, preservation of political subdivision boundaries, and respect for communities of interest.” Caster Doc. 352-2 at 7, ¶ 11. Cooper Plan 9 appears below:

Id. at 8 fig. 1.

At trial, Mr. Cooper testified that he stood by his testimony from the preliminary injunction proceedings, Tr. 105, and he discussed the four reports he has offered throughout this litigation. See, e.g., Caster Docs. 48, 65, 352-1, 352-2. He described his trial report as “updat[ing] and expand[ing] on” his previous report. Caster Doc. 352-1 at 6, ¶ 14.

Ultimately, Mr. Cooper testified that “[t]here are any number of ways [the Legislature] can draw a seven-district plan that has two majority-Black districts,” Tr. 193, and he described his opinion as “unequivocal[ ],” id. at 109–10. Mr. Cooper's “take away” from these nine illustrative maps is “that the Gingles [I] inquiry can be answered [in the affirmative] with no question.” Id. at 117.

Mr. Cooper testified about how his plans satisfy various traditional redistricting principles. Like Dr. Duchin, Mr. Cooper testified that tradeoffs are part and parcel of mapmaking: he testified “no [traditional redistricting principle] reigns supreme” and that he is “constantly balancing these things.” Id. at 157.

Mr. Cooper testified that all his plans have minimal population deviation and contain only contiguous districts. Id. at 119, 125; Caster Doc. 352-1 at 31 n.18; id. at 24, ¶ 58. Mr. Cooper also testified about how his plans preserve core constituencies. For example, he explained that each of his illustrative plans keeps more of the City of Birmingham together than the 2023 Plan does. Tr. 147–48.

Mr. Cooper also testified that his illustrative plans fared similarly to the 2023 Plan when it came to splitting political subdivisions, id. at 138–42, and he provided tables of data comparing the splits in his plans to the splits found in the 2023 Plan (and in the Special Master Plan). See Caster Docs. 352-1 at 50 fig. 28; 352-2 at 12 fig. 3. According to Mr. Cooper's data, Cooper Plan 9 splits fewer counties (5) and fewer municipalities (29) than the 2023 Plan, which splits 6 counties and 31 municipalities. See Caster Doc. 352-2 at 12 fig. 3.

Mr. Cooper also testified that all his plans contain reasonably configured districts. Mr. Cooper testified that part of his evaluation of reasonableness focuses on geographic compactness. Tr. 124–25; see Caster Doc. 352-1 at 46–48. To evaluate compactness, Mr. Cooper testified that he first “make[s] a certain visual assessment,” which includes comparing his illustrative plans to previously enacted plans that “show the judgment of [Alabama]” as to compactness. Tr. 125–26.

At both the preliminary injunction hearing and trial, Mr. Cooper testified that this “eyeball test” has an important role to play in measuring geographic compactness. See id. at 178; Jan. 5, 2022 Tr. 444 (describing the eyeball test as the “most common” compactness metric). And he relied on eyeball test analysis to rebut some of Dr. Trende's assertions. For example, in response to Dr. Trende's criticism that District 2 in the Cooper Plans spans “too large an area from east to west,” Mr. Cooper pointed out that the distance east to west on his illustrative District 2 appears roughly the same as in the distance across District 4 in the 2023 Plan. See Tr. 131–34 (comparing Figure 4 and Figure 24 of Caster Doc. 319-1).

Mr. Cooper testified that he “also measure[s] the compactness of districts and the plan as a whole using the two most widely referenced compactness measures, the Reock score and the Polsby-Popper score.” Tr. 125; Caster Doc. 352-1 at 46. Mr. Cooper described these measures in the following manner:

Well, the Reock score just is basically looking at the area of a district as circumscribed by a circle. And there's a formula that will calculate from zero to one, with one being the highest you could possibly have, which would be a perfect circle, I suppose.

And then for the Polsby-Popper score, the same story, a circle around the perimeter of that district. And if a district is not at all compact, then it's going to be approaching zero.

Tr. 126–27. In other words, “the Reock is more about the area of the district, and the Polsby-Popper is more about the configuration of the perimeter.” Id. at 127. Mr. Cooper also described another perimeter measure, Convex Hull, as:

sometimes a useful way to look at the perimeter question because if you use that measure, it's -- does not penalize, say, some odd-shaped perimeters that can be justified, like following Mississippi River, which has lots of twists and turns.

So if you use a perimeter measure and you're working with Polsby-Popper, you can have incredibly low scores even though really there's nothing at all wrong with the district because ․ you have lots of twists and turns in the Mississippi River ․

Id. at 127–28.

On these metrics, Mr. Cooper testifies that his plans, both at a district level and at the plan-wide level, score “within the normal range.” Id. at 128–131; see also id. at 134–37. To support this assertion, Mr. Cooper offers detailed tables that compare the Reock and Polsby-Popper scores for the illustrative districts with scores for districts in previously enacted Alabama plans (including the 2023 Plan) to establish that the scores for Districts 2 and 7 in the Cooper Plans “fall well within the range of compactness scores for congressional districts Alabama has enacted since 1992.” Caster Doc. 352-2 at 18 fig. 9 (comparing Cooper Plan 9 with the 2023 Plan and the Special Master Plan); id. at 26–28 figs. 14 & 15 (comparing Reock and Polsby-Popper scores across illustrative districts and previously enacted Alabama plans). We have attached these tables to this order as Appendix E. During his cross-examination, Dr. Trende did not dispute these figures. Tr. 2049–51.

Mr. Cooper also testified about a fourth measure of geographic compactness: the Dave's Redistricting Application (“DRA”) compactness score. Id. at 135–36. Mr. Cooper described the DRA compactness score as “a composite score where the Reock score is scaled and the Polsby-Popper score is scaled or normalized on a zero-to-100 range and then that average of those scores are presented in the final composite score.” Id. at 136–37. Mr. Cooper explained that he consulted this metric “because Dr. Trende has suggested that it is an appropriate measure to use,” and Mr. Cooper employed it in his recent work in Arkansas. Id. at 137. Mr. Cooper provided tables containing the results of this analysis, see Caster Doc. 352-2 at 19 fig. 10 & 22 fig. 11, and he testified that the figures demonstrate that Cooper Plan 9 is more compact than both the 2023 Plan and the Special Master Plan. Tr. 135–38; Caster Doc. 352-2 at 18–22. Put differently, Mr. Cooper testified, Cooper Plan 9 is the “proof [ ] in the pudding” to establish that it is “possible to draw a congressional plan that contains two majority-Black districts that is more compact overall than any of Alabama's enacted plans over the last 30 years.” Tr. 138.

Mr. Cooper rested his testimony about the geographic compactness scores on his extensive experience across the country: “I'm just saying, based on my experience in Alabama and other states, the Reock scores I'm reporting for the illustrative plans are within the norm and basically within the same range as plans that have been developed by the State of Alabama over the past 35 years.” Id. at 190–91. He reiterated: “And you can look at other scores in other states and look at Reock scores and you'll see that things match up okay for the various plans. I mean, there are lower -- plans with lower scores and plans with higher scores, but it's within the norm.” Id. at 191. And he repeated: “I work in lots of states. So I am comfortable saying that the Reock and Polsby-Popper scores in this plan, even if in some instances they're lower than Alabama's mean average, they're still okay. They stack up well nationwide.” Id. at 191–92.

Although Mr. Cooper's eyeball test and analysis of geographic compactness scores told him his illustrative remedial districts were reasonably configured, Mr. Cooper testified that he regarded these indicators as insufficient:

My argument really is that you can't just look at the scores; you've got to look at the map; you've got to look at historical plans; you've got to look at historical demographics; you've got to look at water areas; you've got to look at highways.

I mean, there are so many factors involved. And then you come away with the subjective ruling or decision of which I cannot make as a mere expert for drawing illustrative plans. That's for the Court to decide.

Id. at 160.

Accordingly, Mr. Cooper testified that in his opinion, his illustrative plans are reasonably configured because they respect traditional districting principles:

It is my assessment that Illustrative Plan 9 and Illustrative Plan 8 and the other illustrative plans are sufficiently compact and are contiguous and are observant of political subdivisions to be considered plans that are following traditional redistricting principles.

But I can't make the final decision on whether or not it's acceptable. That is something for the Court to rule on ultimately, right?

I mean, and you can't just look at a score absent a map, absent demographics of the place you're examining and suddenly say, okay; this particular district has a low Reock score, therefore, the plan's no good. Because there could be good reasons for a low Reock score, given the shape of the county, given the shape of the jurisdictions and the shape of the VTDs.

So the Reock, Polsby-Popper scores are not the be all and the end all; they're an indicator, and they have to be taken into consideration with the multitude of other redistricting principles that one deals with when you're drawing a voting plan.

Id. at 175–76.

Mr. Cooper also testified that his plans respect communities of interest: he opined that Cooper Plan 9 maintains the Black Belt, Gulf Coast, and Wiregrass “in a similar number of districts as compared to the Special Master and 2023 Plans.” Caster Doc. 352-2 at 13. He opined that all of his plans “place significantly more of the Black Belt counties into a majority-Black district than the 2023 Plan․ [O]nly half (nine) of the Legislature's 18 identified Black Belt counties are in a majority-Black district in the 2023 Plan. Conversely, each of my illustrative plans place over 70% of the Black Belt counties in a majority-Black district, and four of my illustrative plans place all but one of the Black Belt counties in a majority-Black district.” Id. at 14. We attach the associated figure in Appendix E.

When asked about the Wiregrass, Mr. Cooper testified that he accepted it as a community of interest based on the Legislature's definition. Tr. 205. He further testified that though he did not keep the Wiregrass in a single district because of overlapping Black Belt counties, like Barbour County, his plans still respect the Wiregrass because “the counties are generally left intact.” Id. at 205–06.

Consistent with Dr. Duchin's testimony, when Mr. Cooper was asked whether “you have to keep a community of interest whole to respect it,” he responded: “No. Not necessarily. I mean, I would argue that I'm respecting Mobile County even though it's divided between two districts.” Id. at 195–96.

On cross-examination about the Gulf Coast and Wiregrass, Mr. Cooper observed: “[T]hey're not the only communities of interest in the entire state. And it is interesting that the legislature seems hyper focused on the Wiregrass but never talks about Appalachia, never talks about the Tennessee River area as a community of interest.” Id. at 209. He continued: “I mean, there's just nothing in the legislative findings that would suggest that there's anything based regionally other than the Black Belt, the Gulf Coast, and the Wiregrass.” Id.

Mr. Cooper also testified at length about how he considered race when he drew his illustrative plans. He testified that “applying traditional redistricting principles[,] you can naturally draw two majority-Black districts.” Id. at 124. When asked how he knew that “race [did] not have to predominate in order to accomplish two majority-Black districts,” Mr. Cooper responded that it was because he was able to draw two such districts while also honoring traditional districting principles:

Because I was looking at all the relevant factors. I was looking at the traditional redistricting principles, which would include compactness and minimizing political divisions, subdivision splits, minimizing splits of counties, contiguity, one person, one vote.

I was taking all of that into account while at the same time looking at whether or not you could fit a majority-Black district into that broader mix of traditional redistricting principles.

Id.

When Mr. Cooper was asked outright whether race predominated in his illustrative plans, he replied, “absolutely not.” Id. This echoed his testimony at the preliminary injunction hearing:

Q. So what specific traditional districting principles did you consider in drawing the illustrative plans in this case?

A. Well, I took all of them into consideration. I examined the document produced back in May by the Alabama Legislature outlining the guidelines for redistricting. But a lot of that just incorporates the general concept of traditional redistricting principles. So I didn't prioritize any of them. I tried to balance them.

․

Q. So was any one factor of the ones we just mentioned predominant, the predominant factor when you were preparing your illustrative plans in this case?

A. Not really. I feel like I gave them equal weighting. It would be possible to prioritize others and come up with different configurations, but perhaps at the expense of one of the key redistricting principles. So you could draw very compact districts, but they might split numerous counties because they're perfect squares. Or you draw a district that is -- two districts that are maybe 60 percent Black, but they wouldn't be contiguous. That, you know, so you have to balance it.

Q. And did race predominate in your development of any of the illustrative plans?

A. No. It was a consideration. This is a Section 2 lawsuit, after all. But it did not predominate or dominate.

Jan. 5, 2022 Tr. 439–41. Mr. Cooper explained at trial:

Q Is it fair to say that hitting 50 percent plus one was a nonnegotiable in your map-drawing process?

A Was a what?

Q Nonnegotiable.

A No. I don't -- I don't use that term. I would not have gone to 50 percent plus one for a second majority-Black district if I were not also balancing the other traditional redistricting principles.

So I don't know exactly what you mean, but I would not have produced these illustrative plans if I didn't think that they adhered to traditional redistricting plans – traditional redistricting principles.

Tr. 172–73.

Mr. Cooper opined that it would be obvious if he had allowed race to predominate in his illustrative plans. He explained:

[H]ad race been my overriding consideration, I could have drawn districts that consistently placed communities that have higher concentrations of Black Alabamians in majority-minority districts and communities with higher concentrations of White Alabamians in non-majority minority districts, resulting in majority-minority districts with higher BVAPs. But at no point have I been asked or have I attempted to prioritize BVAP in District 2 or District 7 (or prioritize the racial composition of any district) over other traditional redistricting principles.

Caster Doc. 352-2 at 6. At trial, he elaborated during cross-examination:

Q You also say in paragraph 10 that, had race been your overriding concern, you could have placed higher concentrations of Black Alabamians in majority-minority districts; is that correct?

A Yes. I believe that is true.

Q So none of your majority-minority districts maximize BVAP, correct?

A I don't think so. I think you could go significantly higher.

Q Why don't they maximize BVAP?

A Because I'm balancing traditional redistricting principles.

Q How would you go about maximizing the BVAP for a district?

A Well, you would just split lots of VTDs, split lots of municipalities, draw more irregular-looking districts, and you would get higher for sure.

Tr. 169–170.

In connection with his testimony about how he considered race, Mr. Cooper responded to Dr. Trende's assertion that Mr. Cooper “split Jefferson County along racial lines.” Id. at 142. Mr. Cooper testified that Dr. Trende's “color-coded” racial heat maps are “foreign” to Mr. Cooper because he “never look[s] at maps like these.” Id. at 143 (discussing Milligan Doc. 384-5 at 66 fig. 39). The following exchange occurred during that testimony:

Q Mr. Cooper, when you are drawing illustrative plans, do you ever see this kind of racial color coding of each precinct on your screen?

A Never ever. I do not employ color coding for by race at the district level.

Q When you are –

A At the precinct level.

Q When you are drawing your illustrative plans, do you have any knowledge of the five percentage point Black Voting Age Population range of each VTD in that map?

A No.

Id. (referring to the levels of shading on Dr. Trende's figure).

Further, Mr. Cooper testified that when he split voting districts, he was “following existing lines or at least existing demarcations by the Census Bureau.” Id. at 145–46. Such existing lines include “odd-shaped municipalities”; topographical features like mountains, ridges, and valleys; precinct lines; and primary roads. Id. at 144–46. Mr. Cooper explained:

I mean, [Dr. Trende is] treating Jefferson County like it's just a flat plain that has a bunch of precincts that are only identifiable by whether or not they're five percent Black versus 95 percent Black. I don't -- I mean, I just do not approach a redistricting plan drawing in that fashion.

Id. at 146–47; see also id. at 223 (opining that “municipal lines in and around Mobile are just as tricky as they are in Jefferson County in terms of irregular shapes, water areas”).

Mr. Cooper later reiterated that although he was “generally aware of where the municipalities that are predominantly Black are,” and that he “knew where the precincts with a BVAP above [30 percent were],” id. at 163–64, he never split a VTD for “the purpose of bolstering the Black Voting Age Population in a particular district” or “for the purpose of creating a majority-Black district,” id. at 153–54.

When asked whether he considered a race threshold as an outer limit on his respect for traditional districting principles, Mr. Cooper again resisted:

Q Would a fair restatement of “adhere to traditional redistricting principles” be that your plans comply with traditional districting principles as much as possible while retaining two majority-Black districts?

A I don't think that would be fair ․

Id. at 176–77.

Finally, Mr. Cooper also testified about matters going to his credibility. He testified that “routinely” during his career he consulted with potential Section Two plaintiffs and concluded they could not satisfy Gingles I. Id. at 121, 166–67. He also consulted with a jurisdiction and concluded the same. Id. at 121–22. Mr. Cooper gave an example from Alabama, when he was hired around 2009 “to determine whether a majority-Black district in the City of Calera[, Alabama] could be maintained in order to get Section 5 preclearance.” Id. at 122. After he reviewed the data, Mr. Cooper determined “that you just simply could not draw a reasonably compact district in the City of Calera based on the 2000 census which was the operative census at the time. So [he] told them no.” Id. Mr. Cooper testified that his services in Caster were not contingent on his ability to draw a second majority-Black district. Id. at 122–23.

2. Gingles II and III – Racially Polarized Voting (Dr. Palmer)

To satisfy the second and third Gingles preconditions—that Black voters are “politically cohesive” and that each challenged district's White majority votes “sufficiently as a bloc to usually defeat [Black voters’] preferred candidate,” Allen, 599 U.S. at 18 (quoting Gingles, 478 U.S. at 51)—the Caster Plaintiffs rely on the testimony of Dr. Maxwell Palmer.

Dr. Palmer works as a tenured Associate Professor of Political Science at Boston University, where he has been on the faculty since he earned his doctorate in political science at Harvard University in 2014. Caster Doc. 303-1 at 1, ¶ 1; Tr. 482–83. His work focuses on American politics, data science, and political methodology. Caster Doc. 303-1 at 1, ¶ 1; Tr. 483. He has published one book and numerous articles in peer-reviewed journals. Caster Doc. 303-1 at 1, ¶ 2. He has extensive experience as an expert witness in redistricting cases and served as an independent analyst of racially polarized voting for the Virginia Redistricting Commission in 2021. Id. at 1–2, ¶ 3. Dr. Palmer was compensated at a rate of $350 per hour for his work in Caster and his compensation did not depend on the substance of his testimony. Caster Doc. 49 at 2.

At trial, Dr. Palmer was qualified as an expert in redistricting, political science, and data analysis without objection. Tr. 483–84.

Dr. Palmer testified about two matters: (1) his analysis of whether and the extent to which voting is racially polarized in Alabama, Caster Doc. 303-1 at 2, ¶ 5; Tr. 487; and (2) his analysis of the performance of the majority-Black districts in the Cooper Plans and under the 2023 Plan. Caster Doc. 303-1 at 6–9; Tr. 487. At trial, he reaffirmed his previous opinions and testimony on these matters from the preliminary injunction proceedings in these cases. Tr. 484–85.

As to the first issue, about racially polarized voting, Dr. Palmer studied voting patterns in the 2016, 2018, 2020, and 2022 general elections in Alabama, as well as the 2017 special election for the United States Senate, in statewide elections for President, the United States Senate, Governor, Lieutenant Governor, Secretary of State, Attorney General, and several other offices. Caster Doc. 303-1 at 2, ¶ 9; see also Tr. 489–97 (explaining how he used precinct-level data and analyzed the results on a district-by-district basis). Dr. Palmer relied on publicly available data, including census data, that he ordinarily uses, and he relied on the same ecological inference statistical procedure that Dr. Liu employed, which “estimates group-level preferences based on aggregate data.” Caster Doc. 303-1 at 3, ¶¶ 10–11.

Dr. Palmer opined in his report that “Black voters are extremely cohesive, with a clear preferred candidate in all 17 elections [analyzed],” id. at 5, ¶ 14, and “White voters are highly cohesive, in voting in opposition to the Black-preferred candidate in every election [analyzed],” id. ¶ 15. Dr. Palmer concluded that “[o]n average, Black voters supported their candidates of choice with 93.0% of the vote[,]” and “[o]n average, White voters supported Black-preferred candidates with 14.3% of the vote, and in no election did this estimate exceed 26%.” Id. ¶¶ 14–15. He further opined that there is “strong evidence of racially polarized voting in each of the seven congressional districts under [the 2023 Plan].” Id. ¶ 16. He found “strong evidence of racially polarized voting across the state of Alabama.” Id. at 2, ¶ 6.

At trial, Dr. Palmer testified at length about his ecological inference methodology. Tr. 489–97. He opined that it is “the best available method for assessing racially polarized voting” and his “understanding is that [ecological inference] is regularly used in court testimony cases like this and has been the preferred method for estimating group-level behaviors for racially polarized voting analyses,” id. at 491–92. He described his analysis step-by-step. Id. at 492–98.

Dr. Palmer also testified that he “found strong evidence of racially polarized voting in Alabama. The Black and White voters consistently support different candidates both statewide and in the individual congressional districts.” Id. at 487, 498. He testified that this conclusion was consistent with his earlier opinions in these cases. See id. at 487. He offered several figures to provide visual depictions of his findings. See Caster Doc. 303-1 at 4 fig. 1 & 5 fig. 2. And he testified that “we can think of racially polarized voting as a matter of degree,” and that “it can be very sharply polarized, as I found here, where very large majorities of Black and White voters are supporting different candidates.” Tr. 526.

Dr. Palmer testified that he next examined whether the Black-preferred candidates were able to win elections in Alabama. Id. at 499. To do so, he “added up vote totals from the precinct-level data [provided by the state].” Id. Dr. Palmer testified that, at the statewide level, the Black-preferred candidate was able to win only one out of seventeen elections that he studied (when Doug Jones, a White Democrat, beat Roy Moore, a controversial Republican accused of sexual misconduct, in the special election for the United States Senate in 2017). Id. at 500; Caster Doc. 303-1 at 15 tbl. 10.

Dr. Palmer also testified that the district-level data produced similar results. More particularly, he explained that “there were five districts, the First, Third, Fourth, Fifth, and Sixth, where the White-preferred candidate defeated the Black-preferred candidate in every election.” Tr. 500. In the Second District, he testified, “the White-preferred candidate defeated the Black-preferred candidate in 16 of the 17 elections. And, in contrast, in the Seventh District, the Black-preferred candidate won all the elections.” Id. Dr. Palmer added that District Seven is a majority-Black district, while all the other districts in his analysis are majority-White. Id. Accordingly, Dr. Palmer concluded that “Black-preferred candidates are largely unable to win elections in Alabama and only regularly able to win elections within [District 7].” Id. at 487–88.³⁴

As to the second issue, Dr. Palmer's performance analysis of the majority-Black districts in the Cooper Plans and under the 2023 Plan, Dr. Palmer opined that, under all the Cooper Plans, “Black-preferred candidates would generally be able to win elections in both of the majority-minority districts [i.e., Districts 2 and 7].” Id. at 488. To reach this conclusion, Dr. Palmer testified that he overlaid the Cooper Plans “on to the election data, add[ed] up the votes and [saw] the vote shares that each candidate would have received under different sets of lines.” Id. at 501; Caster Doc. 303-1 at 8 fig. 4 (demonstrating performance of Cooper Plans 1–8). Dr. Palmer provided a performance analysis for Cooper Plan 9 in his reply report and opined that “[i]n both [majority-minority] districts the Black-preferred candidate won all 17 elections, with an average of 57% of the vote in [District 2] and 64% of the vote in [District 7].” Caster Doc. 303-2 at 7; Tr. 502–03.

Dr. Palmer also responded to the opinions of Dr. Hood. Tr. 503–04. First, Dr. Palmer opined that Dr. Hood overstated the importance of data about the performance of Dr. Ben Carson, who is Black, in the 2016 Alabama Republican presidential primary. Both Dr. Palmer and Dr. Hood observed that in that election, Dr. Carson “received 10.2% of the vote in Alabama.” Caster Docs. 302-3 at 20 (Hood), 303-2 at 1 (Palmer); Tr. 504. Dr. Palmer observed that “[Dr.] Carson ranked fourth in the primary, behind Donald Trump, Ted Cruz, and Marco Rubio; 90% of the voters in Alabama's Republican primary preferred a different candidate.” Caster Doc. 303-2 at 1. Accordingly, Dr. Palmer testified:

I don't think we can draw many conclusions from that single data point. And Dr. Hood did no other analysis of Ben Carson's performance in Alabama such as looking to see if he performed particularly well in certain parts of the state or if he was a preferred candidate for any group of voters.

Tr. 504; Caster Doc. 303-2 at 1–2.

Second, Dr. Palmer similarly criticized Dr. Hood's analysis of Senator Tim Scott of South Carolina, who is Black and “was appointed to his Senate seat by the governor of South Carolina to fill a vacancy and then subsequently won a Republican primary.” Tr. 505. Dr. Hood held out Senator Scott's 2014 election as an example of ideology rather than race driving election results, Caster Doc. 302-3 at 21, but Dr. Palmer testified that Senator Scott is “the exception but not the rule of Black candidates being successful in Republican primaries in South Carolina,” Tr. 505. Dr. Palmer further opined that there is “no general pattern of Black Republican candidates being successful in Republican primaries in South Carolina.” Tr. 506; Caster Doc. 303-2 at 2.

Third, Dr. Palmer similarly criticized Dr. Hood's discussion about the election of Alabama Representative Paschal, which Dr. Palmer described as “one example of [W]hite voters electing a minority candidate.” Caster Doc. 302-3 at 21. Dr. Palmer opined that “[t]he fact that one Black candidate was successful in one special Republican primary runoff is not evidence that White voters consistently support minority candidates in Republican primaries in Alabama.” Caster Doc. 303-2 at 2; Tr. 506–07. And he testified at trial:

This was a very low turnout special election, and it happened in two stages. First, there was a primary with several candidates where Representative Paschal came in second. And then he won the runoff election afterwards. This was an election with only three -- less than 3,000 votes. And Representative Paschal barely won this election by 63 votes.

I think the fact that we have one case where one Republican was successful is not evidence that we see any consistent support for minority candidates in Republican primaries in Alabama.

Tr. 506–07.

In the light of these criticisms, Dr. Palmer opined that a “more complete look at primary candidates shows that Black Republicans are rarely successful”:

There were ten Black candidates in the 2022 and 2024 Republican primaries in Alabama. Two ran for statewide office in 2022, two ran for state representative in 2022, and six ran for U.S. Congress in 2024. Nine of the ten Black Republican candidates lost to a White candidate in their primary elections. One Black candidate, Christian Horn, won the primary election for the majority-Black 7th Congressional District in 2024 against a White opponent with 58% of the vote.

Caster Doc. 303-2 at 2–3 (footnotes omitted); see also Tr. 507–08.

Dr. Palmer next responded to the opinions of Dr. Bonneau. Caster Doc. 303-2 at 3; Tr. 508–522. Dr. Palmer first took issue with Dr. Bonneau's opinion that “African American candidates either perform as well as or outperform White candidates of the same political party in judicial, state legislative, and congressional elections in Alabama.” Caster Doc. 303-2 at 3 (quoting Caster Doc. 302-1 at 20); Tr. 508–09. Dr. Palmer explained that “[t]his conclusion is based on incorrect data and misinterpretation of statistical results. After correcting Dr. Bonneau's data and reanalyzing his results, I find no evidence that Black candidates outperform White candidates of the same party in Alabama. Indeed, I find strong evidence that Black candidates receive fewer votes than White candidates of the same party.” Caster Doc. 303-2 at 3; see also Tr. 509.

Dr. Palmer testified that as he understood Dr. Bonneau's work, Dr. Bonneau “sought to use Alabama Supreme Court elections, which are a statewide contest, to compare the performance of Black Democratic candidates and White Democratic candidates. And he has county-level data about the vote shares that candidates received in a series of elections from 2010 to 2020. And he estimates a model where he's trying to look at ․ the difference between vote shares for Black and White Democratic candidates, conditional on the percentage of registered voters in each county that are Black.” Tr. 509.

But Dr. Palmer found a “serious error” in Dr. Bonneau's data – namely, that Dr. Bonneau “included mistakenly three uncontested contests for Supreme Court in this analysis; that is, elections where there was a Republican candidate but not a Democratic candidate. And in all three of those elections, he included the nonexistent losing Democratic candidate as a White Democrat.” Id. “In other words,” Dr. Palmer explained, “we have three White Democrats receiving zero percent of the vote in every county in this data.” Id. at 510.

Dr. Palmer further testified that when he corrected Dr. Bonneau's coding error, “his results completely flip. And now, using the same data and Dr. Bonneau's exact same model, his same analysis, [Dr. Palmer] estimate[d] that White candidates receive about ten percentage points more of the vote than Black Democrats do in the Supreme Court elections.” Id. at 509–10; Caster Doc. 303-2 at 3–4.

At trial, Dr. Bonneau acknowledged this data error and its material effect on his findings. Caster Doc. 304-1 at 2; see also Tr. 510.

Dr. Palmer then used Dr. Bonneau's data to “run a racially polarized voting analysis on these elections.” Id. at 510–11. Dr. Palmer testified that he found, “just as in the other elections that we have already talked about, these elections are sharply polarized, that Black voters have a clearly -- a clear preferred candidate in each election and White voters are opposing them.” Id. Dr. Palmer explained his additional findings: “[W]hat's more interesting is that when the Black-preferred candidate is a White Democrat, that candidate received, on average, 21.6 percent of the vote from White voters. And in the one case where the Black-preferred candidate was a Black Democrat, they received only 9.5 percent.” Id. at 511.

“So, in other words,” Dr. Palmer testified, “White voters supported the White Democratic candidate at twice the rate that they supported the Black Democratic candidate.” Id. Dr. Palmer testified that this analysis provides “some evidence that White Democrats outperform Black Democrats and some evidence that this difference can be attributed at least partially to different preferences or willingness to vote for Democratic candidates by race.” Id. On cross-examination, Dr. Palmer acknowledged that the analysis included only one Black candidate. Id. at 538.

Dr. Palmer next turned to Dr. Bonneau's analysis of Alabama legislative races in 2022, in which Dr. Bonneau compared “vote shares for Democrats who lost contested seats for the legislature by race” and found that “Black Democrats received higher vote shares than White Democrats for both the state house and the state senate.” Id. at 511–12.

Dr. Palmer identified “several issues with this analysis,” including that “these results are not statistically significant,” and that “it's really hard to make comparisons across different elections like this and different districts,” because “Black and White Democrats might be running in very different places with very different underlying demographics, potentially different turnout, different incumbency, status of their opponents, et cetera.” Id. Dr. Palmer also critiqued Dr. Bonneau's opinions about elections in Alabama House Districts 73 and 74. He objected to Dr. Bonneau's opinion about District 74 on the ground that it reflects “a single case where a White Democrat defeated a Black candidate in a primary” in a district that was “55-percent Black,” with no other analysis of turnout or vote choice in that primary. Id. at 515. Dr. Palmer elaborated that it “could be the case that the White Democrat was the Black-preferred candidate. It could be the case that the Black Democrat was the Black-preferred candidate but there was a variation, say, in cohesion and turnout that led to their defeat. So we can't rule out race as a factor in this election just because a White Democrat won.” Id.

And Dr. Palmer objected to Dr. Bonneau's opinion about Representative Paschal's election in District 73 on the ground that “[a]gain, we have the results of the election but no further analysis of the dynamics of the election. Especially here, where Dr. Bonneau is claiming that selections are based on candidate positions, Dr. Bonneau has no analysis or evidence about the role of candidates’ positions on how voters make decisions.” Id. at 515–16.

Dr. Palmer next criticized Dr. Bonneau's analysis of congressional elections, in which Dr. Bonneau examines “the correlation between the Democratic vote share and the percent of the population that's Black in each congressional district” for six elections. Id. at 516. Dr. Palmer testified that “this is simply showing a correlation between ․ the Black population and Democratic support,” and that it does not establish that “it's driven by party alone” because Dr. Bonneau “doesn't do any analysis of what factors cause Black voters to choose which candidates to support.” Id. Dr. Palmer further opined that “using a single correlation like this with only six observations is not really reaching the standard for empirical research in political science and ruling out alternative explanations.” Id.

Dr. Palmer ultimately attacked as artificial Dr. Bonneau's isolation of party from race:

Implicit in Dr. Bonneau's incorrect conclusion about the role of party is his assumption that the effects of race and party are separable. In other words, Dr. Bonneau assumes (without any evidence) that an individual's race and an individual's political party are two separate and independent factors that influence vote choice. A long literature in political science about how voters develop partisan attachments and make decisions about voting shows the opposite: an individual's background, including their race, is a key factor in their politics and party preferences. This means that even if members of a racial group strongly support candidates of a single party, race, as a key factor in driving their support for that party, is an inseparable part of their support for those candidates. If race causes party, then we can't find that party alone, without race, can cause vote choice. Due to the fundamental linkage of race and party, the effects of the two cannot be separated. In other words, the strong support of Democratic candidates by Black voters cannot be attributed to partisan preferences alone, but to a mix of personal and political factors and experiences of which race is an essential part.

Caster Doc. 303-2 at 6–7.

3. The Senate Factors

Like the Milligan Plaintiffs, the Caster Plaintiffs rely on joint stipulations and the expert testimony of Dr. Bagley and Dr. Burch about the Senate Factors. See supra Part IV.A.3.

4. Fact Witnesses at Trial

The Caster Plaintiffs offered trial testimony from four fact witnesses.

a. Dr. Marcus Caster

Plaintiff Marcus Caster is Black, was born in 1975, grew up in north Mobile County, and lives in Washington County. Tr. 369–71. Dr. Caster has family ties in the Black Belt and Mobile County. Id. at 369. Dr. Caster holds an M.B.A. and a doctoral degree, and he has been an educator for his entire career. Id. at 370–71. Dr. Caster testified that he has spent his entire life in southwestern Alabama (particularly Mobile, Clark, and Washington Counties), where he has worked at various educational institutions and he and his wife have founded several community organizations, most involving organized sports. Id. at 371–73, 376–77. He described how these organizations serve Black youth in the Black Belt and Mobile County, and he testified that children from Baldwin County do not participate because they “have their own” organizations. See, e.g., id. at 379. Dr. Caster testified about his campaigns when he twice ran for an Alabama House seat in Washington County. Id. at 382. And he testified that he affiliates with the Democratic Party because its “principles ․ as far as they help and support ․ the African-American community and Blacks align more with [his] core values.” Id. at 383.

Dr. Caster testified that based on his work and life experience, he understands the needs of Black Alabamians in the rural Black Belt. He testified about three issues. First, he testified that “Alabama is really experiencing a crisis when it comes to healthcare facilities.” Id. at 384. He testified that in “Thomasville, they had a healthcare facility that wasn't even open five years and it is closed.” Id. at 385. He said that “in Monroe County, women had to go deliver their babies elsewhere because they were no longer offering labor and delivery in that area.” Id. He observed that “Searcy Hospital in Mount Vernon” which “was a healthcare facility for mental health individuals ․ is no longer there.” Id. Second, he testified about the quality of school facilities. He described McIntosh High School, a predominantly Black school, where he testified that “they had to go inside ․ and remove bats,” and students are “subjected to different type of chemicals” because of the location of the school. Id. at 389. Third, he testified that internet service in the Black community is “slower than most” and there are “a lot of Internet outages.” Id. at 389–90. He gave the example of the last outage he experienced, when a technician arrived “after about two or three days” despite repeated calls by Dr. Caster. Id. at 390.

Dr. Caster also testified about his experience with elected officials. Id. at 391–96. He said that when he lived in District 1, he was represented by Congressman Byrne and then Congressman Carl. Id. at 391. Neither representative was responsive to the needs of the Black community, and he did not see either representative campaign in the Black community. Id. at 391–92. He explained that he supported Congressman Figures because “[h]e showed up,” attended “town hall meetings,” “came to the celebration parades in Jackson, Alabama,” and “assured [Dr. Caster] that he was going to do everything in his power to help us.” Id. at 394.

Dr. Caster testified that he was “proud” and “excited” when United States Senator Katie Britt, a White Republican from Alabama with ties to the Wiregrass and Montgomery, expressed opposition to a recent incident in which United States Air Force videos on the Tuskegee Airmen were “taken away.” Id. at 410–11.³⁵

Dr. Caster testified about his understanding of his case. He understood that “Alabama must redraw the maps to represent two ․ majority-minority ․ districts,” and “Alabama defied” the ruling. Id. at 400. He testified that Alabama's defiance was “disrespectful” and “a slap in the face.” Id. at 401.

b. Ronald Smith

Ronald Smith is Black and was born in 1954 in Union Springs, Alabama (in Bullock County, in the Black Belt).³⁶ Id. at 416. Mr. Smith served on the Union Springs City Council for six years, chaired the Bullock County Commission for 15 years, served in the Army for a decade, and worked nearly 20 years as a medical administrator for the Tuskegee VA Medical Center. Id. at 417–22. He is a member of the NAACP, the Alabama Democratic Conference, and several community organizations. Mr. Smith testified that he “had the opportunity to become familiar with the interests and needs of the Black community in Bullock County” and “across the Black Belt.” Id. at 427.

Mr. Smith testified that during his public service, the following issues were most frequently raised: “Jobs. Accessibility to quality and affordable healthcare. Affordable housing. Making sure we have our transportation corridors available so that we can adequately recruit industry to come into Bullock County. Better education by means of having much needed clean, healthy, environment and equipment for our children to be their best.” Id. at 427–28.

Mr. Smith also testified about his experience growing up during segregation. Id. at 428. He explained that “words can[not] adequately describe” the feeling “[w]hen you live on one side of the railroad track and you see your White counterparts enjoying some of the amenities that are government sponsored and it's taboo for you.” Id. He grew up with “signs that said colored water fountain, White only bathroom.” Id. Mr. Smith added that even though the signs ultimately came down, the practice remained the same. Id. at 437, 446. His “school only had two Bunsen burner[s] and two microscope[s] when, the White school, every student had a microscope and Bunsen burner.” Id. at 428. There were “only four available [typewriters] for the ․ 25 to 30 students,” so to learn how to type, the students “had to write the keyboard on a cardboard.” Id. at 428–29. Mr. Smith testified that “[t]he White kids [had] the swimming pool” while the Black kids “had a ditch.” Id. at 429. Mr. Smith testified that he once jumped into the pool and police responded. Id.

Mr. Smith testified extensively about educational disparities that affect Black Alabamians in the Black Belt. He said that “there's a shortage of qualified teachers today,” and that “when you live in a rural community like [Bullock County], recruitment of qualified teachers and instructors are limited” because “the salaries in the Black Belt are lower” and the lack of housing, recreational activities, and restaurants “impede ․ recruitment.” Id. at 434. He testified that “there is no equivalence between Black and White education in Bullock County and throughout the Black Belt” because of the White flight. Id. at 435. He testified that today, the Bullock County School System has approximately five White students. Id.

Mr. Smith also testified that “Black children face obstacles getting into college and higher education in universities” because of a lack of “qualified guidance counselors” and “parents ․ don't know how to do it.” Id. at 436. He added that Black students are ill-prepared for the ACT and SAT and because of low scores, they “can't receive a lot of financial assistance.” Id. at 436–37.

Mr. Smith observed that there “are pockets of Bullock County where they have no access to broadband.” Id. at 444. He testified about students receiving laptops during the pandemic, and that “[y]ou [can] give them a laptop, but a laptop without access to broadband is like having a car without tires.” Id.

Mr. Smith testified that “these educational disparities affect a person's ability to vote and participate in the political process.” Id. at 437. He explained that not having a quality education means that only certain jobs are available, “[a]nd you only have about 30 minutes to vote.” Id. Moreover, he said, even if you have a great job, the work is often out of town, requiring the voter to leave before the polls open and return after the polls close. Id. He opined that “there's all types of impediments that are there to strategically draw you out of the voting process.” Id. He “observed Black voters leaving polling places without voting” as recently as “[t]he last election” but he did not see White voters leaving. Id. at 449.

Mr. Smith testified “that the Black community in Bullock County and the Black Belt faces discrimination in healthcare.” Id. at 441. He “was born by midwife” because his family did not have access to the hospital in Union Springs. Tr. 441–42. Mr. Smith testified that there is not a single hospital in Bullock County; the nearest hospital is in Montgomery, “about an hour away.” Id. at 442. He testified that there is a “high rate of infant mortality,” unaffordable insurance, and a lack of Medicaid expansion, each with discriminatory impacts. Id.

Mr. Smith testified that he did not vote for the two White Republicans who previously represented District 2 before Congressman Figures. He offered that “[t]hey didn't campaign in [the Black] community” and “didn't come to ․ town hall meeting[s].” Id. at 452. He voted for Bobby Bright, a White Democrat who represented District 2 from 2009 to 2011 because Congressman Bright campaigned in the Black community. Id. Mr. Smith explained that he has voted for candidates based on how responsive they are to his community's needs. See id. at 473. He said that he voted for Ronald Reagan. Id. at 468.

c. Valtoria Jackson

Valtoria Jackson is Black, was born in 1961 in Montgomery, and works as a Pastor and a nurse. Id. at 1069, 1073, 1077. She grew up in an “an African-American community” as “neighbors to Rosa Parks.” Id. at 1069. Pastor Jackson testified that growing up, her family was “fortunate to be considered ․ upper middle class, they had purchased their own home, and so [her mother] was able to pay the poll tax” and “pass the test” to vote. Id. at 1070. She testified that “it was a family tradition to go to the polls” and that her mother “became a registrar” and “would always carry in the trunk of her car the paperwork to register everyone that she came upon.” Id.

Pastor Jackson testified that her family was “very involved in the Black church.” Id. at 1070–71. She testified that the churches she attended were “[p]redominantly Black,” but that the nuns and priests were “typically White.” Id. at 1071. She said that growing up, “church was the hub of [their] community meetings, [their] celebrations at each other's homes,” and “the center piece of [the Black] community.” Id. at 1071–72. She became a pastor in 2008 and pastors Saint Peter African Methodist Episcopal Church in Montgomery. Id. at 1073–74.

Pastor Jackson holds a nursing degree and a graduate divinity degree. Id. at 1076. She testified that she practiced full-time as a nurse for 39 years and worked in critical care, trauma care, and hospice nursing, and did administrative and consulting work. Id. at 1077–78. She also testified that she worked as a sexual assault nurse examiner (“SANE”) and coordinator, and that her team “developed a pediatric SANE nurse program that still exists in Montgomery.” Id. at 1079. She said that they “developed a program where [they] took mobile units into the rural areas, particularly in the Black Belt of Alabama to build sexual assault response teams and to help facilitate exams within the first 72 hours of sexual assault victims.” Id. She testified that she serves on the Board of the Alabama Coalition Against Rape and has “develop[ed] relationships in leadership throughout the state with domestic violence and sexual assault violence centers.” Id. at 1079–80.

Pastor Jackson testified that in her current pastoral role, she does administrative work, coordinates programs, collaborates with other churches in the area, works with the health ministry, and has coordinated statewide voter mobilization efforts on a nonpartisan basis. See id. at 1080–82. In her personal capacity, she has participated in phone and text banks, voter registration drives, and served as a poll worker. Id. at 1086–87. Her community engagement work has been concentrated in Montgomery, Dallas, Butler, Crenshaw, Bullock, and Mobile Counties in Black communities. Id. at 1087–88.

Pastor Jackson testified about her church's work with food security programs. Id. at 1082–83. She said that during the pandemic, “every week[, they] were distributing food to a wide, diverse community,” and that they continue to “serve over 200 seniors monthly.” Id. She explained that they “deliver 60 percent of [the food] boxes to the home because ․ transportation is an issue, [and] seniors are having to pay even family members to run errands for them.” Id. at 1083.

Pastor Jackson testified about her personal experiences with discrimination and disparities. Id. at 1088. She recounted that her parents were once “unable to purchase [a home] because of the color of their skin.” Id. And she said that her family was “harassed and not welcomed” when they moved into a White neighborhood. Id. She testified that “within two years, [her street] turned totally Black due to White flight.” Id. She offered that “due to redlining, [their] home was never valued,” which is an issue that they are “still fighting ․ today.” Id. at 1089.

Pastor Jackson testified that in her work in healthcare in Montgomery and the Black Belt, she has observed families living in unhealthy and unsafe conditions with “rodents and roaches,” “[in]adequate heating,” “[in]adequate plumbing,” and even without water. Id. at 1091–92. Despite her extensive professional experience, she thought she “was in a third-world country” when she witnessed some of the conditions. Id. at 1092. She described that during one home visit in the Black Belt, “[w]e could see the sky through the roof and dirt floors, and multiple children -- multiple generational living, poor transportation. But they were living off the land. And I just did not realize how poor some of our rural areas and the living conditions of people.” Id. She testified that in some houses, “[t]hey may only have just one room that did not get rained on and maybe had a tarp that would blow off when it was bad weather.” Id. at 1126. In fact, she observed that some houses “still had outhouses” and “didn't have indoor plumbing.” Id.

She testified that she witnessed these circumstances around 2010 or 2012, predominantly in the Black community, and that these sorts of issues are “[f]ive times greater in [the] Black community.” Id. at 1092–93, 1126. She explained that living in such conditions impacts political participation because “when you're struggling, can't put food on your table, can't afford transportation to even care for business ․ your concern is not getting to the polls.” Id. at 1093.

Pastor Jackson also testified about her observations of the disparities between predominantly Black and predominantly White schools in the Black Belt. Id. at 1095. She testified that “particularly in Bullock County,” the quality of education is lower because of overcrowded classrooms, poor building conditions, and low income. Id. She described as an example something she saw as a school nurse:

There was a young man, he was acting out ․ He was overheated ․ because he was wearing ․ a sweat hoodie in the summer time and then I had to unzip because he was about to have an episode. And I was like, Where is your shirt? And he's like, Well we don't have any running water, I don't have any clean clothes.

․ [H]e at least came to school. But I saw it repeatedly in our predominantly African-American schools where poverty is prevalent, that the children are fighting just to get food. Some come to school just to eat and other[s] do not come to school because they don't have clean clothes or clothes that fit.

Id.

Like many other witnesses, Pastor Jackson testified about the lack of internet access in the Black Belt. She testified that during the pandemic, “even though [students] had the laptops, they were unable to access the Internet until maybe like mid-COVID.” Id. at 1096.

Pastor Jackson testified that “[i]lliteracy is a reality” in the Black community that hinders political participation. Id. at 1097. She gave as an example a man “trying to get his [voting] rights restored” who she knew “couldn't read because he was holding [the form] around and adjusting the paper upside down.” Id. She also observed an illiterate person at church holding a hymnal upside down. Id. at 1098. Pastor Jackson testified that in all her years of nursing, pastoral, and community work in Montgomery and the Black Belt, she has never encountered an illiterate White adult. Id. at 1129.

Moreover, Pastor Jackson testified about the lack of access to transportation and employment in the Black Belt due to “[in]adequate infrastructure in ․ public transportation.” Id. at 1098. She observed that “being transported and getting to work is a problem.” Id. She continued: “They don't make enough to even afford gas to make it to work. And when they find a ride ․ they have to pay the person to get that ride. So it's very difficult to find quality jobs.” Id. at 1099. Pastor Jackson testified that inadequate transportation makes healthcare difficult to access as well. Id. at 1099–1100.

Ultimately, Pastor Jackson testified that these disparities limit political participation in Black communities in the Black Belt because “when you are in a mode of survival” voting is not a priority. Id. at 1101.

Pastor Jackson also testified about Black voters’ partisan alignment. She explained that most Black churchgoers “feel abortion is wrong,” and that on abortion as well as LGBTQ rights, the Republican Party aligns closely with the Black church. Id. at 1106–07. Nevertheless, she explained, most Black people do not support Republican candidates because Republicans do not campaign in Black communities. Id. at 1108–09. That said, Pastor Jackson acknowledged her “favor” for Senator Britt because, in Pastor Jackson's view, Senator Britt is “truly looking realistically at the issues of childcare and at the issues of healthcare.” Id. at 1108.

d. Janice Malone & Robert Clopton

The Caster Plaintiffs also rely on the testimony of Ms. Malone and Mr. Clopton. See supra Parts IV.A.4.c, IV.A.4.e; Caster Doc. 309 at 2.

5. Designated Deposition Testimony

The Caster Plaintiffs offered six witnesses by deposition. Caster Doc. 370.

a. William Carroll

William Carroll ³⁷ is Black and has served on the Mobile City Council for nearly 12 years. Milligan Doc. 459-4 at 7–9. He testified that, currently, his top priority is “under-served neighborhoods,” and that during his time on the Council, he has not talked to Congressman Carl “very much.” Id. at 9–10.

Mr. Carroll testified that “Mobile's economic base is huge” and that the main drivers of it are the Port, medical industry, Austal, Airbus, and schools. Id. at 12, 18. He said that people commute from counties such as Conecuh, Evergreen, and Escambia to Mobile for economic opportunities, but he does not know of anyone who commutes to work in Mobile from Butler, Pike, Barbour, Bullock, Macon, or Russell Counties. Id. at 12.

In his view, the City of Mobile and the Black Belt should be in the same congressional district because “Mobile has more influence politically than the Black Belt counties” and “if it wasn't for Mobile's strength, then the Black Belt doesn't get heard.” Id. at 15. Mr. Carroll testified that in District 1, Mobile, Baldwin, Escambia, and Covington Counties seem to have been grouped together “to create a Republican stronghold.” Id. at 16.³⁸ As he put it, “[i]f you're blue in these four counties, then you're lost,” and “[y]ou're not represented because you're going to get overpowered by the people that are represented on the other side.” Id.

Mr. Carroll offered the view that his city council district should be in District 2. Id. at 26. He testified that “most minorities are Democrats, and the voting block within the city is more purple than it is red.” Id. As a result, that assignment to District 2 would “lend[ ] itself to the ability of a Democrat or minority being able to select someone of their choice.” Id.

b. Bobby Lee DuBose

Plaintiff Bobby Lee DuBose ³⁹ is Black, was born in 1963, and grew up in Bullock County. Milligan Doc. 459-5 at 7; Milligan Doc. 436 ¶ 35. He lives in Birmingham. Milligan Doc. 459-5 at 7. He testified that he “grew up on a plantation” and “when you grow up on a plantation, you learn how to survive” and “how to take the bare minimum and you make it work for you.” Id. at 29.

Mr. DuBose testified that this lawsuit is important to him because “[t]here should be another African American that understands the African-American people[’s] needs, health insurance, job opportunity.” Id. at 10. In his view, Congresswoman Sewell alone is insufficient to care for the needs of all Black Alabamians. See id.

Like some of the other plaintiffs, Mr. DuBose testified that in the Black Belt, the “[e]ducation system is basically ruined,” and in Bullock County, “there's not a functional hospital.” Id. “When you hear the name Black Belt, you don't all of a sudden think of it being a thriving community.” Id. at 11. Rather, “low income, disadvantaged, [and] poor education” are what come to mind. Id. He testified that for those who live in the Black Belt, “[y]ou have to leave your home to find a better opportunity.” Id. “Someone needs to speak up on behalf of those counties that are underserved, under privileged,” and as he sees it, Black Alabamians are “truly the ones who've been there, who grew up there, understand the significance of trying to bring some improvement.” Id. at 10, 14.

Mr. DuBose testified that he has not “heard that there's racial discrimination in the voter registration process” and his family members have not experienced difficulty in registering to vote. Id. at 19. He said that his polling places “[o]pened on time, closed on time.” Id. at 20.

Mr. DuBose testified about how he votes: “If their thought line up with mine,” he testified, then he is “all in whether they're Black, [W]hite, green, or yellow” or whether they are “a Democrat or a Republican.” Id. at 22. He added that “economic opportunity” and “healthcare” are the issues that resonate the most. Id. Mr. DuBose also said that “[W]hite voters in Alabama usually prefer Republicans.” Id. at 33.

Finally, Mr. DuBose testified that there has been “a long history of racial discrimination in Alabama connected to voting in elections” and that racial discrimination in Alabama persists. Id. at 36.

c. Benjamin Jones

Plaintiff Benjamin Jones ⁴⁰ was born in 1965 in Barbour County. Milligan Doc. 459-8 at 6. He holds an engineering degree from Tuskegee University and attended law school at Faulkner University. Id. at 6. He now works as executive director at Montgomery Community Action and pastors St. James Missionary Baptist Church. Id. at 7–9. Montgomery Community Action is a nonprofit organization “that provides assistance to communities or the County of Montgomery residents who need assistance with utilities, rent, [and] medicine.” Id. at 10. He said that about seventy-five percent of the population it serves is Black. Id.

Mr. Jones testified that he joined this lawsuit to create a majority-minority district so that a “candidate [who] is responsive to the needs of the district” can be elected. Id. at 16. He testified that “low income [people], people who need Medicaid, Medicare ․ need assistance” need improved representation. Id. at 19. He said that it is not simply his goal “to get a second Democrat elected.” Id. at 22.

Mr. Jones detailed some of the most basic needs of Black communities in the Black Belt: he described that “[t]here are some areas in the state where people may be forty, fifty, maybe even more miles away from a hospital,” that there are “food deserts[,] where people may be miles and miles and miles away from a grocery store or a farmer's market,” and that “there are people who don't have access to the internet because of their location in the rural area.” Id. at 19.

Mr. Jones testified that he did not experience difficulties with registering to vote when he turned eighteen. Id. at 31–32. And he testified that apart from one incident (where he “had to vote a challenge ballot because [his] name didn't appear on the list”) he has not experienced “any barriers to fully participating in the political process in Alabama.” Id. at 42.

Finally, Mr. Jones testified that he experienced racial discrimination growing up in Alabama. Id. at 42–43. At the first preliminary injunction hearing, Mr. Jones said that his parents were active in civil rights marches in the 1960s and that “they went to jail on a number of occasions for voting.” Jan. 10, 2022 Tr. 1345. He added that they did not go to marches together because one of them had to be reliably out of jail to parent him and his fifteen siblings. See id.

d. Rodney Allen Love

Plaintiff Rodney Allen Love ⁴¹ was born in 1985, raised in Mobile, and moved to Birmingham before he graduated high school. Milligan Doc. 459-14 at 5–6. Mr. Love testified that he pursued this lawsuit because he and his community are “not getting ․ represented in Congress” because Congresswoman Sewell is the only one “fighting for us right now.” Id. at 9. Mr. Love testified that in his view, Alabama's enacted congressional plans are unfair because a “majority of Republicans [are] in one place” and “the majority of Democrats [are] in one place.” Id. at 10. This results in inequality because “the majority of the Blacks vote Democrat[ ] and the majority of the Republicans are [W]hite.” Id. He testified that to him, “candidate of choice” means “a candidate that's actually going to help [him].” Id. at 16.

Mr. Love said that his polling place is “probably five minutes away from the house,” that he has not experienced any problems voting, and that he does not feel voting is hard for him because he is Black. Id. at 18. He explained that he ordinarily votes Democrat because “they'll basically help me more than the Republicans will.” Id. at 19. He said that “the Democratic Party is more responsive to the needs of Black voters than the Republican Party.” Id. “Democrats ․ help out the poor.” Id.

Finally, Mr. Love testified that he has experienced discrimination growing up in Alabama and continues to experience discrimination today. Id. at 22.

e. Manasseh Powell

Plaintiff Manasseh Powell ⁴² is Black, was raised in Lowndes County (in the Black Belt), and moved to Montgomery when he was twelve years old. Milligan Doc. 459-18 at 8. Mr. Powell testified that “if the State of Alabama can draw a Black majority district, then, it should draw a Black majority district.” Id. at 21. Mr. Powell said that he “never had any problems with registering to vote or at [his] polling place.” Id. at 24, 25. Mr. Powell testified that when he decides how to vote, he regards issues as more important than race. Id. at 26. He testified that the issues most important to him are “rights for Black people” as well as “jobs, the economy, infrastructure, [and] our support to foreign countries.” Id.

f. C.J. Small

C.J. Small ⁴³ is Black, was born in 1978, and raised in Mobile. Milligan Doc. 459-23 at 3, 4. He has served on the Mobile City Council since 2012 and works as a funeral director. Id. at 5, 10. He testified that his council district is “around 68 percent” Black, and the issues most important to his constituents are “[b]light, infrastructure, [and that] kids don't have anything to do.” Id. at 6, 10. He said that the “vast majority” of his constituents, “Black and [W]hite, believe in all of those things.” Id. at 10.

Mr. Small opined that when interacting with constituents, “Black[ ] [constituents are] much warmer.” Id. at 7. He testified that some White constituents would not vote for him “no matter [his] policies” and “don't care for [him] because of [his] race.” Id.

Mr. Small testified about his family and work ties to the Black Belt. Id. at 11–13. He explained that the population in the Black Belt is “not as heavy as it was,” that “younger people are moving out,” and that “some African-Americans” are leaving the Black Belt “for education and for jobs.” Id. at 14. He testified that there are more opportunities in Mobile and Baldwin County than in the Black Belt. Id.

Mr. Small testified that he and former Congressman Carl have a “cordial” relationship, but Congressman Carl has not “provided adequate representation ․ for the needs of Mobile.” Id. at 21. He testified that Congressman Figures “would adequately represent all of [his] constituents.” Id. at 26.

C. Singleton Plaintiffs’ Arguments

The Singleton Plaintiffs rely on all the evidence adduced in Caster and Milligan for their Section Two claims, Singleton Doc. 288 at 13, as well as testimony from additional expert and fact witnesses. See generally Singleton Doc. 320.

1. Expert Testimony about the Senate Factors

a. Dr. R. Volney Riser

Dr. Riser holds an undergraduate degree from Florida State University and graduate degrees in American History from the University of Alabama. Singleton Doc. 253-1 at 1, ¶ 3. He has worked as a Professor of History at the University of West Alabama since 2005, where his scholarship focuses on the development and practical operation of political and constitutional systems in the Jim Crow-era South. Id. at 1–2, ¶¶ 3, 4. He has published two books: (1) Defying Disfranchisement: Black Voting Rights Activism in the Jim Crow South, 1890–1908, and (2) A Goodly Heritage: Judges and Historically Significant Decisions of the U.S. District Court for the Middle District of Alabama. Id. at 2, ¶ 4. He has also published numerous articles and contributed entries to reference volumes, including essays on disenfranchisement and various landmark episodes in United States legal, constitutional, and political history. Id.

At trial, Dr. Riser was qualified as an expert in the history of the role of race and politics and law in the South and Alabama, with no objection. Tr. 750. Dr. Riser was compensated at a rate of $200 per hour for his work, and his compensation did not depend on the substance of his testimony. Singleton Doc. 253-1 ¶ 2.

Dr. Riser opined in his report that in the late-nineteenth century, “Whites struggled to either believe or accept that African Americans could make political decisions for themselves, denigrating them as a ‘bloc’ to be manipulated rather than as a class of voters qualified to act in their own interest in partnerships and coalitions of their own design or choosing.” Id. at 3, ¶ 7. He further opined that both Democrats and Republicans were hostile toward African Americans and demonstrated a commitment “to limiting African Americans’ political opportunities, both on the ballot and at the ballot box.” Id. at 4, ¶ 8.

Dr. Riser testified that “from the moment of Reconstruction ․ Democratic partisans complain[ed] of Black domination or Black rule,” an idea that persisted “over the next quarter of the century.” Tr. 755. He testified that by the 1890s, Alabama's “Democratic leadership ․ resolved that they absolutely are going to move to disenfranchise all African-Americans.” Id. at 762. Dr. Riser detailed how disenfranchisement was rolled out: Black men were precluded from registering to vote “through all manner of little tricks and games that the registrars were empowered to play” including, among other things, understanding clauses, literacy tests, the grandfather clause, and poll taxes. Id. at 773–74. He testified that by the early 1900s, “roughly 98 percent” of Black men were disenfranchised in Alabama. Id. at 772, 774.

b. Dr. Kari Fredrickson

Dr. Frederickson holds a doctoral degree from Rutgers University and has worked as a Professor of History at the University of Alabama since 1999. Singleton Doc. 253-2 at 3. Her expertise is in twentieth-century American history with a focus on the South. Id. She has published four books, three of which won awards. Id. She has published numerous articles and essays in peer-reviewed publications and served as an expert for six documentaries. Id. She has served on the grants review committee for the National Endowment for the Humanities and serves on the Board of Directors of the Harry S. Truman Library Institute. Id. at 3–4.

At trial, Dr. Frederickson was qualified as an expert in American history with a focus on the role of race in the South and Alabama in the 20th century, with no objection from any party. Tr. 805.

In her report, Dr. Frederickson opined “[r]ace has served as a dividing line in political allegiance and activity since the period of Reconstruction” and that “the ability of first the Democratic Party and later, the Republican Party, to achieve viability and dominance has depended on each party's ability to secure the support of [W]hite voters through racial appeals.” Singleton Doc. 253-3 at 4. She further opined that “[t]oday, the parties are racially polarized; most [W]hites are Republicans and most Blacks are Democrats” and that “[W]hite identity politics occup[ies] the center of Republican politics, [which makes] creating effective and enduring bi-racial coalitions [ ] extremely difficult, if not impossible.” Id. at 5.

At trial, Dr. Frederickson testified that race is “the dominant factor for defining party identity in Alabama.” Tr. 807. She testified that during the early 1900s, “the Democratic Party achieved dominance through disenfranchisement” and in so doing “created itself as the White party and created the region as a one-party region with no viable competition.” Id. She testified that the Republican Party adopted the use of racial appeals from the 1960s onwards. See id. at 828–33, 841.

Dr. Frederickson unambiguously testified that in her “30-plus years as a historian specializing in the history of the American [S]outh since 1865” there has never “been a time in history when partisanship rather than race drove racially polarized voting in Alabama.” Id. at 843–44.

2. Fact Witness at Trial

The Singleton Plaintiffs also offer trial testimony from one fact witness, Leonette Slay. Ms. Slay is White and has lived in Jefferson County for 30 years. Tr. 864. Ms. Slay testified that she is a member of numerous community organizations, id. at 869–70, and that keeping Jefferson County whole in a congressional district “was a committed goal of [hers] because [she] see[s] Jefferson County as a community of interest with very unique issues and a capability to form a biracial coalition,” id. at 875. She testified that “there's a plethora of issues that really could be ameliorated, solved if we had one representative to combine with our county and city officials to work for the betterment of Jefferson County.” Id.

3. Designated Deposition Testimony

The Singleton Plaintiffs offered the testimony of Plaintiff Rodger Smitherman by deposition.⁴⁴ See Milligan Doc. 459-25; Singleton Doc. 260. Senator Smitherman is Black and represents part of Jefferson County in the Alabama Senate. Milligan Doc. 459-25 at 7. Senator Smitherman first ran for the Senate in 1994, and he is a member of the Committee. Milligan Doc. 459-25 at 20.

Senator Smitherman testified that when he discussed redistricting with members of the Senate Black Caucus and the Alabama Legislative Black Caucus, there “was a general feeling that the whole counties would be the best approach[—]keeping the counties together, all together, keep them whole as possible.” Id. at 13. He testified that when resources are distributed to the districts, they “are being split up among all these other counties.” Id. at 16. “Instead of getting them all, we're just getting a little fraction based on how many other counties are ․ sharing [those resources].” Id. He also offered the view that “rural people have different needs and concerns than people in the urban area.” Id. Essential issues regarding water quality, sewage, food deserts, roads, and schools are different for rural and urban communities. Id. He testified that Jefferson County should have been kept whole in the 2023 Plan because it is a community of interest. Id. at 31.

Senator Smitherman also testified that he supports the Special Master Plan “because it's an [additional] opportunity district.” Id. at 14. He said that if the Singleton Plaintiffs prevail, they request a plan that creates two opportunity districts and keeps Jefferson County whole. Id. at 18.

Senator Smitherman testified about his disappointment when the 2021 Plan was passed without creating two opportunity districts. Id. at 24. He testified that it is not “legitimate for the [L]egislature to consider political goals when it draws Congressional districts” when it negatively impacts “the rights of voters.” Id. at 29. He said that it is a “constitutional right of people ․ to have the opportunity to vote and get the person they want.” Id. Senator Smitherman added that he does not know what the intent of the Legislature was in drawing the 2023 Plan. Id. at 32.

Senator Smitherman testified that he votes “in all elections,” his polling place is “convenient,” and he has not had “trouble voting there.” Id. at 36. He said that he votes based on the issues. Id.

Finally, Senator Smitherman explained that he is “a product of segregation and integration.” Id. at 35. He testified that “in terms of political opportunities, we have not advanced to nowhere where we need to be.” Id. at 43.

D. The State's Defenses

The State argues that the Plaintiffs have not satisfied the Gingles preconditions and the 2023 Plan does not violate Section Two. Milligan Doc. 481 at 37. The State claims that “political processes in Alabama are open to all, and that ‘what appears to be bloc voting on account of race is instead the result of political or personal affiliation of different racial groups with different candidates.’ ” Milligan Doc. 445 at 11. It also asserts that Section Two cannot form the basis for a private lawsuit and that applying Section Two to redistricting plans is unconstitutional. Id. at 11–12.

1. Gingles I – Reasonable Configuration

Because numerosity is stipulated, the State's Gingles I arguments focus on whether an additional majority-Black congressional district can be reasonably configured. The State argues that “no illustrative plan presented in these cases demonstrates that there is a ‘reasonably configured’ alternative remedy that would respect the Legislature's neutral redistricting principles ‘at least as well as’ ” the 2023 Plan. Milligan Doc. 481 at 38, ¶ 84. The State asserts broadly that the Plaintiffs’ plans “chop up Alabama's important communities of interest in violation of Alabama's traditional principles.” Id. at 76, ¶ 196. But at its core, the State's position is that any district that splits Mobile County to join part of Mobile with the Black Belt is not reasonably configured.

The State argues that in our assessment of the 2023 Plan, we are required to defer to the 2023 legislative findings about communities of interest. The State cites Vance v. Bradley, 440 U.S. 93, 110–11 (1979) for the proposition that our “responsibility for making ‘findings of fact’ certainly does not authorize [us] ․ to reject the legislative judgment” reflected in the 2023 legislative findings. Id. at 41–44, ¶¶ 92, 94, 96–102. The State urges that “[t]he best evidence of [legislative] purpose is the statutory text adopted by both Houses ․ and submitted to the [Governor].” Id. at 41, ¶ 95 (quoting W. Va. Univ. Hosps., Inc. v. Casey, 499 U.S. 83, 98 (1991)).

The State argues that the 2023 Plan (including the findings) reflects a “policy of defining and uniting the State's regional communities” in the Black Belt, the Gulf Coast, and the Wiregrass. Id. at 64, ¶ 166. The State argues that the 2023 Plan “ended th[e] inconsistent treatment” of these three communities of interest in the 2021 Plan. Id. at 5. The State urges that it treated these three communities equally because “[t]he safest route” past the competing hazards of liability under the Constitution and Section Two “was for the Legislature to satisfy § 2 by answering Plaintiffs’ neutral call to ‘employ[ ] the same line-drawing standards in minority [communities of interest] as it used elsewhere.’ ” Id. 216, ¶ 582 (quoting the Milligan Plaintiffs’ brief in the Supreme Court).

The State also argues that “each of Plaintiffs’ alternative plans ‘are palpable racial gerrymanders,’ ” and that “[b]y connecting what the Plaintiffs call ‘Black Mobile’ ․ to the eastern Black Belt in [District 2], and by cramming the map's leftover counties into [District 1], Plaintiffs subordinate Alabama's ‘traditional race-neutral principles’ ‘to racial considerations.’ ” Id. at 38, ¶ 85.

The State argues that “Dr. Duchin's and Mr. Cooper's preliminary injunction hearing testimony and trial testimony are saturated with ‘express acknowledgement[s] that race played a role in the drawing of district lines.’ ” Id. at 77–80, ¶¶ 201–14. It claims that “Dr. Duchin and Mr. Cooper ‘purposefully established a racial target: African-Americans should make up no less than a majority of the voting age population’ in two congressional districts,” and that the evidence “confirms that the contours of Dr. Duchin's and Mr. Cooper's maps were ‘motivated by a predominant, overriding desire to assign [B]lack populations’ to [District 2] ‘and thereby permit the creation of a [second] majority-[B]lack district.” Id. at 80–81, ¶ 215.

The State offered testimony about compactness from one expert witness: Dr. Scott Trende. The State asked Dr. Trende to compare the Cooper Plans, Duchin Plans, 2021 Plan, 2023 Plan, and Special Master Plan. Milligan Doc. 384-5 at 7. Dr. Trende holds a law degree and graduate degrees in political science and statistics from Duke University and The Ohio State University. Id. at 5–6; Tr. 1971. He works as a visiting scholar at the American Enterprise Institute, a lecturer at The Ohio State University, and a Senior Elections Analyst for Real Clear Politics. Milligan Doc. 384-5 at 4; Tr. 1970–71. He has served as a retained expert and a court-appointed expert in redistricting litigation, Milligan Doc. 384-5 at 6–7; Tr. 1973–75, and a few courts have found his testimony unreliable, see Tr. 2084–86; Nairne v. Ardoin, 715 F. Supp. 3d 808, 850 (M.D. La. 2024); Matter of 2022 Legislative Districting of State, 282 A.3d 147, 185–86 (Md. 2022). Dr. Trende was compensated at a rate of $450 per hour for his work in these cases and his compensation did not depend on the substance of his testimony. Milligan Doc. 384-5 at 7.

Dr. Trende testified at the second preliminary injunction proceedings in 2023. In our September 2023 order, we observed that Dr. Trende “offer[ed] no opinion on what is reasonable or what is not reasonable in terms of compactness.” Milligan Doc. 272 at 151. At trial, Dr. Trende was admitted without objection “as an expert in redistricting.” Tr. 1975.

In his report, Dr. Trende opined that “by the standards of Alabama's recent history,” the remedial districts in the Duchin Plans and Cooper Plans “are some of the least compact districts drawn.” Milligan Doc. 384-5 at 93. He further opined that “[t]he illustrative districts carve up major population centers by race, and mostly function by stitching together two populations of Black residents in distinct metropolitan areas, with lightly populated, rural areas in between.” Id. He also opined that the 2023 Plan is more compact than the Duchin Plans and Cooper Plans. See id. at 28.

Like Dr. Duchin and Mr. Cooper, Dr. Trende “readily acknowledge[d] that ․ tradeoffs” of traditional redistricting principles “are built into the process.” Tr. 1982. Dr. Trende testified that population equality, contiguity, and compactness; respect for political subdivisions, natural boundaries, and communities of interest; and incumbent protection are all traditional redistricting principles. Id. at 2065. Dr. Trende testified that of those principles, he analyzed only geographic compactness, id., and did not consider any other factor enumerated by the Legislature, id. at 2082–83. Dr. Trende described the task of determining whether a district is reasonably configured as a “tough analysis,” and testified that he felt comfortable opining that a district is not reasonably configured even though he did not analyze “some of these traditional redistricting principles.” Id. at 1981–82.

At trial, Dr. Trende testified that the eyeball test has a role in measuring geographic compactness, but he cautioned against overreliance on such tests because “it's a little bit trickier to give an opinion with a reasonable degree of scientific certainty about what your eyes are telling you.” Id. at 1991. He described “[e]yeball tests, for better or for worse,” as “part of what we're asked to do ․” Id. at 2101.

Dr. Trende rested his compactness opinion on three measures of geographic compactness: Reock, Polsby-Popper, and Convex Hull scores. Id. at 1984. Dr. Trende described the Convex Hull metric as “look[ing] at the area of a convex polygon that would enclose a district.” Milligan Doc. 384-5 at 14–15. He likened the score to having a “rubber band snapped around a district,” and asking “what percentage of that rubber band the district would fill.” Id. He described how a square district would score on this metric: “That square will, by definition, fill ․ approximately 63.7% of the [minimum bounding] circle. Its Reock score would therefore be 0.637. That is still relatively high as far as Reock scores go, but many would consider a perfectly square district to be quite compact.” Id. (footnotes omitted). In his report and at trial, Dr. Trende offered data about the scores on these metrics for the maps he compared. Id. at 31 fig. 10, 32 fig. 11, 33 fig. 12, 36 figs. 15 & 16, 37 fig. 17; see also Tr. 1994–95.

Dr. Trende testified that there is no bright line, objective standard for reasonable scores, so he relies on relative comparisons. Tr. 2037–39. When asked whether he was testifying “that any of Plaintiffs’ illustrative plans are generally non-compact in an abstract sense,” he replied: “No. I try to ground it a little bit more than that in comparison to what's been drawn in Alabama and nationally.” Id. at 2039–40.

On direct examination, Dr. Trende cautioned against overreliance on plan-wide scores. He testified that looking at plan-wide scores can become “a problem” if you “draw a horribly shaped district and then sort of get a makeup call by drawing very compact districts with the remainder of the districts.” Id. at 1991. For that reason, he testified that “reporting district-by-district scores is more appropriate here than using plan-wide scores.” Id. at 1991–93.

On cross examination, Dr. Trende acknowledged that in a memorandum to the Supreme Court of Virginia regarding a map he drew, he wrote: “[H]owever, since we are drawing a whole map for the state, the most important compactness comparison is for the state as a whole. Dave's Redistricting app provides a composite compactness score for a whole map.” Id. at 2044 (quoting Caster Doc. 319-67 at 18). Dr. Trende testified that this memorandum was not prepared for purposes of a Gingles analysis. Id.

In any event, Dr. Trende testified that based on plan-wide scores, some Duchin Plans and some Cooper Plans are in the same “ballpark” as the 2023 Plan. For the Cooper Plans, Dr. Trende explained:

There are some that get in the ballpark, I guess you'd say, but even that does narrow -- a couple of them will get in the ballpark on certain discrete measures, but they tend to do it by drawing those box-shaped districts in the [N]orth to offset what's going on in the [S]outh.

Id. at 1999. In his report, he opined that Cooper Plan 7 “seems the closest to the 2023 map overall.” Milligan Doc. 384-5 at 32–33, 35 fig. 14. At trial, he testified that Cooper Plan 7 is “reasonably compact” on a plan-wide basis as compared to the 2023 Plan. Tr. 2043. He also testified that Cooper Plan 9 is more compact than the 2023 Plan on a plan-wide Reock score basis and is “in the same range of compactness” for Polsby-Popper scores. Id. at 2042–43. And he acknowledged that based on DRA scores that Mr. Cooper reported, Cooper Plan 9 is “more compact than any plan that Alabama has drawn or used in the last 40 years,” including the 2023 Plan. Id. at 2045.

For the Duchin Plans, Dr. Trende testified about the plan-wide Reock scores:

So it's the -- kind of the same story [as with the Cooper Plans]. The two enacted plans have ․ higher Reock scores, on average.

Some of Dr. Duchin's maps, I guess -- I don't know. You say they get in the ballpark, but, again, they do that by drawing those highly compact districts in the northern area.

Id. at 2000. Dr. Trende testified that on the Polsby-Popper metric, Duchin Plan “B performs better. [Duchin Plan] E, again, I guess you could say, gets in the ballpark.” Id. Dr. Trende also candidly acknowledged that “it's tough because you get into this splitting hairs of, well, the score for [Duchin Plan] E, on average, is nine-tenths of a point lower than the 2023 map.” Id.

Dr. Trende testified that the shape of a state can affect geographic compactness scores: for example, “[a] state like Virginia or Massachusetts just naturally has a more elongated shape to it. And so you're naturally going to get worse Reock scores in a state like Massachusetts.” Id. at 2003. Along the same lines, Dr. Trende cautioned that “a national comparison at the individual district level really gets fraught because you start to really run into the state-specific issues.” Id. at 2008. Regardless, Dr. Trende testified, comparisons are possible such that an expert can say “okay, it does or does not fall within these ranges.” Id. at 2003.

To that end, although Dr. Trende opined in his report that “it is difficult to see how [the Duchin Plans and Cooper Plans] could be considered within the normal range of maps in the United States,” Milligan Doc. 384-5 at 44, he repeatedly testified at trial that the plan-wide scores of the Duchin Plans are within the normal range for maps nationwide. He first testified:

I think it's literally -- the maps are literally within the range, but that's only because of how extreme the Illinois map is. When you take into account the nature of that Illinois map, the fact that Dr. Duchin and I both agree that the Texas map is a gerrymander, at the very least, a large number of Mr. Cooper's maps fall outside the normal range.

Tr. 2007–08. And he later testified on cross-examination that California's map is less compact on the Reock score than Duchin Plans A, B, D, and E, and that the Duchin Plans are more compact on the Polsby-Popper measure than both the California plan and the Special Master Plan. Id. at 2123–24. In addition, he testified Duchin Plans A, B, C, and D are more compact on Convex Hull than at least four of the plans that Alabama enacted since 1972. Id. at 2125.

Ultimately, Dr. Trende concluded that “Dr. Duchin has a stronger justification for the claim that her maps were in the normal range than Mr. Cooper did,” id. at 2124, and that he (Dr. Trende) should have written the conclusion in his report about the Duchin Plans “better” because his “focus was mostly on” the Cooper Plans, id. at 2125–26.

Dr. Trende also testified about the compactness scores of District 2 in the Cooper Plans compared to other ways Alabama has drawn District 2. He testified that “[As to Polsby-Popper, Cooper] Maps 6, 2, 8, and 1 are the least compact versions of District 2 drawn in Alabama in the last 50 years. [Cooper] Maps 3, 5, 7, and 4, the Special Master's version, is less compact.” Id. at 2012–13. But he admitted that “the Reock compactness scores for all of the majority-Black districts in Mr. Cooper's illustrative plans are within the range of compactness scores for congressional districts that Alabama has enacted since 1992.” Id. at 2050.

Dr. Trende also opined that the illustrative plans split political subdivisions based on race. At trial, he testified that “when you're talking about counties that were split between an illustrative majority-minority district and a non-majority-minority district, that split in the county often occurs in a way that appears to be based on race.” Id. at 2014–15. According to Dr. Trende, this is especially so for the Jefferson County split between Districts 6 and 7, and the Mobile County split between Districts 1 and 2. Id. at 2015–18. Dr. Trende specifically opined that Cooper Plans 6 and 7 are “particularly aggressive” in splitting Mobile County along racial lines and that Cooper Plans 1, 2, and 4 split Montgomery County three ways along racial lines. Milligan Doc. 384-5 at 69–72.

At trial, Dr. Trende acknowledged that he did not consider municipal boundaries, he did not consider transportation corridors, and he did not consider what the stated purpose was for making any given split. Tr. 2104–07. He acknowledged that the District 6 “finger” that extends into Jefferson County has existed since the 1992 decision in Wesch. See id. at 2055. And he agreed “that sort of split of Jefferson County has been a consistent feature of maps Alabama has passed” since 1992. Id. Ultimately, Dr. Trende testified that he did not offer an opinion as to whether race was the primary factor in any county split. Id. at 2113.

Dr. Trende also testified that the Plaintiffs’ illustrative plans appear to split voting districts based on race. Id. at 2100–02. To support this opinion, he testified that he examined only “the BVAP shading of the VTDs” and did not consider any factor or explanation other than race. Id. at 2101–02.

Dr. Trende also testified at trial about the district in the electoral plan for the Alabama State Board of Education that, according to Dr. Trende, was the first district to stretch from Montgomery to Mobile. Id. at 2024–30. Dr. Trende explained that the district was made in 2010 “by adding the Black population in Mobile” to Montgomery in order to address “geographic and racial constraints.” Milligan Doc. 384-5 at 61. In his expert report, Dr. Trende suggested that this may have been first done in order to comply with Section 5 of the Voting Rights Act before Shelby County v. Holder, was issued. Id. at 57. Dr. Trende opined that this district “appears to be a one-off configuration in Alabama” that appears to be “based upon the existence and understanding of what Section 5 of the [Voting Rights Act] required ․ and inertia.” Id. at 56–57. But Dr. Trende testified that in reaching that conclusion, he did not talk to legislators about the configuration, speak with a historian, consult an expert on legislative intent or legislative history, or talk to anyone in Mobile or Montgomery about relevant educational needs. Tr. 2095–97. Instead, he reviewed preclearance submissions, legal cases, shape files, and census data. Id. at 2095.

Moreover, Dr. Trende's explanation did not address that the Legislature continued to split Mobile County in the 2020 State Board of Education plan, well after the Shelby County opinion was issued in 2013, and well after preclearance could have provided any explanation for the subsequent split. Dr. Trende obliquely suggested that this was done in 2020 to maintain core retention, but did not elaborate on this hypothesis or provide any evidence to support it. Milligan Doc. 384-5 at 61.

Finally, Dr. Trende testified that he offers no opinions to dispute Dr. Palmer's conclusions regarding racially polarized voting, Dr. Liu's analysis and conclusions regarding racially polarized voting, Dr. Bagley's analysis and conclusions, or Dr. Burch's analysis and conclusions. Tr. 2032–33. In addition, Dr. Trende testified that he does not dispute that each of the Duchin Plans and Cooper Plans contains two majority-Black districts and takes no issue with the data underlying their reports. Id. at 2034. And he agreed that that “not every community of interest will be or can be kept together in a congressional district.” Id. at 2133.

2. Gingles II and III – Racially Polarized Voting

The State concedes that the Plaintiffs established the second Gingles precondition. See Milligan Doc. 481 at 99 (conceding that “Plaintiffs’ evidence shows that Black Alabamians in the challenged areas are politically cohesive”). And the State does not dispute the pattern of consistent (nearly invariant) electoral losses for Black-preferred candidates in Alabama that the Plaintiffs say establishes the third Gingles precondition. See generally id. The State argues that the Plaintiffs cannot establish the third Gingles precondition because their experts’ opinions are flawed, and the voting and loss patterns those experts opine about are based on party, not race. See id. at 99–114.

The State argues that while Dr. Liu and Dr. Palmer “opine that [W]hite voters typically do not support the Black-preferred candidate, and that the [W]hite-preferred candidates generally win elections,” “their ․ analyses did not reach the question of why voters voted the way they did.” Id. at 99, 101. The State contends that the experts’ analyses did not control for party affiliation, which “is no minor limitation ․ because every expert to offer an opinion on the subject agreed that [B]lack voters in Alabama and across the nation support the Democratic Party very strongly.” Id. at 101, ¶¶ 259, 261. The State suggests that “[a]s such, [Dr.] Liu's and [Dr.] Palmer's racial polarization analysis could just as easily be labeled a political polarization analysis.” Id. at 102, ¶ 263.

The State acknowledges that “courts have accepted” expert testimony about racially polarized voting that relies on ecological inference methods, but “agree[s] with Dr. Bonneau” that ecological inference is “a tool that comes with assumptions,” “not the only tool,” and should not be relied upon to the “exclusion of other possible relevant evidence.” Id. at 100, ¶ 257.

The State also argues that “[W]hite bloc voting in Alabama is not legally significant.” Id. at 99. The State relies on the recent opinion of the United States Court of Appeals for the Fourth Circuit in Pierce v. North Carolina State Board of Elections, 97 F.4th 194 (4th Cir. 2024). As the State describes Pierce, the Fourth Circuit distinguished legally significant White bloc voting in a challenged district from statistically significant White bloc voting at the district level. See Milligan Doc. 481 at 106–08. In the State's view of the Fourth Circuit's analysis, patterns of White bloc voting were not legally significant in a district unless they establish that without a Section Two remedy – a district with a BVAP greater than 50% – Black-preferred candidates would usually be defeated in that district. See id. The State argues that under Pierce, if a district-level performance analysis “yield[s] a ‘minority voting-age population level’ below 50% which provides ‘a realistic opportunity for ․ voters of that minority group to elect candidates of their choice,’ ” then “legally significant racially polarized voting does not exist” in that area. Id. at 108, ¶ 282.

On this line of reasoning, the State argues that “[d]ue to increased registration, turnout, and political participation among [B]lack voters (and crossover voting by [W]hite voters) in Alabama, the historic need for majority- or even large-majority-[B]lack districts in order to ensure [B]lack voters an ‘opportunity to elect’ has substantially lowered.” Id. at 110, ¶ 288. The State argues that “with that change, the point at which [W]hite bloc voting becomes ‘legally significant’ has risen.” Id. In this case, it argues “that [W]hite bloc voting in the challenged areas is not ‘legally significant’ because there is enough [W]hite crossover voting to obviate the need for court-ordered majority-minority districts.” Id. at 112, ¶ 294. And that “[s]o long as additional majority-minority districts are not ‘necessary for [B]lack-preferred candidates to win,’ legally significant [W]hite bloc voting is absent.” Id. at 113, ¶ 297.

Several of the State's experts opined that the pattern of consistent electoral losses for Black-preferred candidates in Alabama is attributable to party rather than race, and we discuss their testimony in connection with Senate Factor 2 below.

3. The Senate Factors

The State argues that “[t]he ‘totality of circumstances’ confirms that the 2023 Plan does not violate Section [Two].” Milligan Doc. 481 at 114. It says that the Plaintiffs have not established that “the political process is not equally open to Black Alabamians.” Id. at 118, ¶ 314. The State claims that “[a] ‘comprehensive, not limited, canvassing of relevant facts,’ ․ strongly suggests that Black Alabamians’ relative difficulty ‘elect[ing] representatives of their choice’ does not depend on ‘race or color,’ ” but on “the predictable result of bloc-voting for Democrat candidates in ‘one of the most Republican states in the entire South.” Id. at 118, ¶ 315.

 Senate Factor 2

As to Senate Factor 2, the State argues that “[r]acial polarization in Alabama is a product of political partisanship, not racial bias.” Milligan Doc. 481 at 119. It argues that “[i]f ‘Black-preferred candidates lose because they are running as Democrats in a red State,’ and not because they are Black, then there is no bloc voting on account of race and no illegal vote dilution for § 2 to remedy.” Id. at 120, ¶ 322.

The State claims that the opinions of Dr. Liu and Dr. Palmer “are almost entirely limited to identifying statistical existence of racially polarized voting, not its cause.” Id. at 124, ¶ 333. The State asserts that because “the second Senate factor supports a § 2 claim only insofar as racial polarization approximates racial bias in voting, and because correlation is not causation, Plaintiffs need more than ecological inferences to support their claims.” Id.

The State argues that “the evidence shows ․ that (1) [B]lack candidates face no penalty at the polls for being [B]lack; (2) [B]lack and [W]hite voters in Alabama tend to vote for parties, not for candidates; (3) the relative weakness of the Democratic Party contributes significantly to the failure of [B]lack Alabamians to elect their candidates of choice; and (4) [W]hite Alabamians, like [W]hite Southerners generally, vote overwhelmingly Republican for ideological, not racial, reasons.” Id. at 125, ¶ 336.

 Senate Factor 7

As to Senate Factor 7, the State argues that “Alabama has made remarkable progress in the election of people of color to public office in the last fifty years,” id. at 159, ¶ 425, and that “Black Democrats have achieved electoral success,” id. at 159. The State contends that “the fact that 32 of the 33 Black Alabamians currently serving in the Alabama Legislature were elected from majority-Black districts” is not a cause for concern “because every one of those 32 legislators ran as a Democrat.” Id. at 159–60, ¶ 426. The State argues that the evidence on this Senate Factor “suggest[s] nothing more than partisan politics.” Id. at 161, ¶ 430.

 Senate Factors 1, 3, and 5

As to Senate Factors 1, 3, and 5, the State argues that “Plaintiffs have not proven that Alabama's distant history of racial discrimination has made the political process less open for [B]lack Alabamians today.” Id. at 161. The State asserts that “Alabama has overcome its history.” Id. at 162.

The State argues that Dr. Bagley's “opinions on this topic are often irrelevant, overstated, missing significant context, or all of the above.” Id. at 163, ¶ 437. The State contends that his “proffered examples of State-sponsored discrimination after 1965 are incredibly sparse,” and his characterization of the legislature's discriminatory motives are unfounded, see id. 164–68, ¶¶ 439, 444, 447, 450.

The State also argues that “Plaintiffs have failed to prove that socioeconomic disparities experienced by Black Alabamians today are the effects of past racial discrimination.” Id. at 171, ¶ 458. It attacks Dr. Burch's testimony, arguing that “[a]lthough Dr. Burch identified ․ gaps” “in socioeconomic status between Black and White Alabamians that have been shown to affect voter registration and turnout,” “she provides no analysis demonstrating that past official discrimination caused these gaps.” Id. at 172, ¶ 460.

The State asserts that the “Plaintiffs did not present a lay witness whose political participation was hampered by past discrimination.” Id. at 184, ¶ 491. It argues that to the contrary, Plaintiffs’ “witnesses such as Ronald Smith, Letetia Jackson, [and] Janice Malone,” and the State's witness Bill McCollum, “each of whom is old enough to have attended segregated schools — are all extremely politically active.” Id. at 184, ¶ 491.

 Senate Factor 6

As to Senate Factor 6, the State argues that the “Plaintiffs have not shown that political campaigns in Alabama are characterized by racial appeals.” Id. at 188. It argues that the “Plaintiffs identify a few appeals that were (1) not made as part of a political campaign, (2) not from campaigns ‘in the area,’ ․ or (3) over which ‘reasonable people could disagree ․ whether they were racial appeals at all,’ ․ or some combination of the lot.” Id. at 189, ¶ 509. It argues that even “the best Plaintiffs can offer still does ‘not demonstrate a pattern, practice, or routine of racial appeals across the election landscape.’ ” Id. at 190, ¶ 512.

 Senate Factor 8

As to Senate Factor 8, the State argues that it is not true that “there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group.” Id. at 191, ¶ 515. The State claims that “the Legislature's good faith effort to remedy the defects observed in the 2021 Plan while complying with the Constitution” should not be considered unresponsive to the needs of Black Alabamians. Id. at 192, ¶ 517. And it says that any violations were remedied “by uniting Black Belt counties into two compact districts and keeping Montgomery whole—treating them as well as other communities of interest that were preserved in the 2021 Plan.” Id. at 192, ¶ 518.

The State argues that “[r]eplacing a plan that had split two majority-Black communities of interest with one in [which] ‘Black voters are no longer artificially denied electoral influence in a second district,’ ․ suggests responsiveness to Black voters.” Id. at 194, ¶ 523. It argues that “the Legislature could reasonably have thought that going beyond the maximum BVAP possible for [District 2] in a ‘race-neutral plan’ would constitute racial gerrymandering, or would at least invite a racial gerrymandering lawsuit.” Id. at 194, ¶ 524.

The State relies on the testimony of Dr. Landers, Colonel Jon Archer, and Kenneth Boswell to support their argument that the Legislature has been responsive to the needs of its citizens. See id. at 195, ¶ 528; 199, ¶ 539; 202, ¶ 546. The State offers as examples its broadband expansion efforts, id. at 204, ¶ 549; improving municipalities and counties through community development block grants, id. at 207, ¶ 558; and various federal initiatives, id. at 209, ¶ 563.

 Senate Factor 9

As to Senate Factor 9, the State argues that “[t]he policies underlying the 2023 Plan are not ‘tenuous.’ ” Id. at 214. It argues that the 2023 Plan advanced ‘traditional redistricting principles such as compactness,’ [and] ‘maintaining communities of interest and traditional boundaries.’ ” Id. at 214, ¶ 574. The State asserts that “the 2023 Plan reflects the Legislature's good faith efforts of complying simultaneously with § 2 of the Voting Rights Act and the Equal Protection Clause of the 14th Amendment.” Id. at 214, ¶ 576. The State posits that the “legislators had a desire to ‘protect ․ congressional incumbents,’ ․ which the 2023 Plan did for all incumbents—Republican and Democrat.” Id. at 216, ¶ 583. Ultimately, it contends that “[e]ven if that is viewed a partisan goal, ‘evidence that the [2023 Plan] was drawn to further partisan goals is a sufficient, non-tenuous justification for this Senate Factor.’ ” Id.

 Proportionality

Finally, the State argues that proportionality is “due very little weight.” Id. at 216, ¶ 585. The State asserts that “[p]roportionality will often elude a map drawer who adheres to traditional redistricting criteria; absent racial calibrations, maps reflect real-world geography and demography inconsistent with proportionality.” Id. at 217, ¶ 586. Moreover, the State argues that “the absence of proportionality does not, by itself, give rise to concern” because “even dramatic disproportionality may be ‘merely ․ a matter of ․ political geography.’ ” Id. at 217, ¶ 587. The State contends that “[w]ithout injecting race into the map drawing process, proportionality is nigh impossible to achieve.” Id. at 218, ¶ 589.

To support its arguments about partisan voting patterns and other aspects of the totality of the circumstances, the State offers four experts: Dr. M.V. Hood, III; Dr. Christopher Bonneau; Dr. Wilfred Riley; and Dr. Adam Carrington.

a. Dr. M.V. Hood, III

Dr. Hood holds graduate degrees in political science from Baylor University and Texas Tech University. Milligan Doc. 409-7 at 4; Tr. 1873. He has worked as a professor of political science at the University of Georgia for more than 20 years, and directs the Survey Research Center at the School of Public and International Affairs there. Milligan Doc. 409-7 at 2. His work focuses on electoral politics, racial politics, election administration, and Southern politics. Milligan Doc. 409-7 at 2; Tr. 1873–74. He has co-authored two books and published numerous articles in peer-reviewed journals. See Milligan Doc. 409-7 at 42–48. Dr. Hood has qualified as an expert in multiple redistricting cases, including in Alabama. Tr. 1875; Milligan Doc. 409-7 at 2. He was compensated at a rate of $400 an hour for his work and his compensation did not depend on the substance of his testimony. Milligan Doc. 409-7 at 2.

At trial, Dr. Hood was qualified with no objection as an expert “in political science, specifically in the areas of electoral politics, racial politics, election administration, and [S]outhern politics and empirical social science research and for the matters discussed in his report.” Tr. 1875–76.

The State asked Dr. Hood to answer five specific questions: (1) “How do Black voting patterns in Alabama compare to other states?”; (2) “Are racial disparities on various sociodemographic factors present outside of Alabama?”; (3) “How does 2016 Republican presidential primary candidate Ben Carson's vote share compare across states?”; (4) “Do [W]hite voters support minority Republican candidates?”; and (5) “How have Black political metrics changed over time in Alabama?”. Milligan Doc. 409-7 at 2; Tr. 1876, 1899. Dr. Hood testified that he “undertook no independent evaluation to determine whether these questions had any relevance to the case at hand.” Tr. 1899.

First, Dr. Hood testified about Black voting patterns. Id. at 1876–77. In his report, he compiled a list of comparison states that “had a Black population of ten percent or greater in 2020.” Tr. 1877; Milligan Doc. 409-7 at 3. He then considered Black voting patterns in those states in eight election cycles (2008–2022) in elections for President, Governor, and Congress. See Milligan Doc. 409-7 at 3–5 & tbls. 1–4; Tr. 1877. Dr. Hood testified that based on these data, in the comparison states and Alabama, Black voters are voting Democratic more than 90% of the time. Tr. 1881–82; Milligan Doc. 409-7 at 3, 5. He opined that “Black support for Democratic candidates across these jurisdictions could be characterized as being close to monolithic.” Milligan Doc. 409-7 at 6.

Dr. Hood acknowledged that his data does not address why Black voters support Democratic candidates, and he did not perform a racially polarized voting analysis. Tr. 1882. And he testified that Black voters in Alabama are politically cohesive, and that Black support for Democrats in Alabama “is slightly higher than Black support for Democratic candidates across the average” of the comparison states. Id. at 1901.

Second, Dr. Hood considered racial disparities on various sociodemographic factors. Id. at 1883–84. He analyzed disparity rates in Alabama and compared Alabama's rates with rates in twenty other states. Milligan Doc. 409-7 at 7; see Tr. 1883–84. He testified that racial disparities exist in Alabama and the comparison states on factors such as education, healthcare, poverty, Internet access, and incarceration. Tr. 1884–89; Milligan Doc. 409-7 at 7, 12, 15, 17–19. He opined that for ten of the thirteen measures he analyzed, “the disparity rate for Alabama is below the average disparity rate calculated for the comparison states” and never “constitute[d] the maximum value among the states analyzed.” Milligan Doc. 409-7 at 20; Tr. 1889–90, 1904–07.

On cross-examination, Dr. Hood testified that he did “not draw any conclusions based on this data,” or “offer any opinions as to why the disparity between Black and White residents exist.” Tr. 1903–04. He also agreed “that Black Alabamians fare worse than White Alabamians on socioeconomic factors regarding education, income, poverty, healthcare, unemployment, and Internet access,” but did not track these disparities across different regions of Alabama (such as the Black Belt). Id. at 1904. He emphasized that both Black and White Alabamians are worse off than the populations in the comparison states. Id. at 1947–48.

Third, Dr. Hood considered Dr. Carson's presidential campaign. Id. at 1890. Dr. Hood examined vote returns “from the first primary contest, the Iowa caucus, through the primaries held on Super Tuesday,” when Dr. Carson dropped out. Milligan Doc. 409-7 at 20. Dr. Hood found that Dr. Carson's “vote totals ranged from a low of 2.6% in Massachusetts to a high of 10.8% in Alaska. Carson earned his second highest vote total, at 10.2%, in Alabama.” Id.; Tr. 1890–91. Dr. Hood opined that this data “provides some evidence of Republican support for minority Republican candidates in Alabama.” Tr. 1891.

On cross-examination, Dr. Hood agreed that “[a] single anything is not a pattern, right. We can't denote a pattern from one case.” Id. at 1908. He also agreed that he “conducted no analysis of the vote shares for Ben Carson by race” (i.e., he did not analyze the preferences of White or Black voters, the racial demographics of the 2016 Republican presidential primary, the total turnout for that primary, or that information for any other state) because he was not asked to. Id. at 1909–10.

Fourth, Dr. Hood considered White support for minority Republican candidates. Dr. Hood testified that he examined peer-reviewed journal articles, id. at 1891, including a 2015 article in Public Opinion Quarterly that he co-authored — “True Colors: White Conservative Support for Minority Republican Candidates.” Milligan Doc. 409-7 at 21–22. That article examined White voter behavior in U.S. Senate and gubernatorial elections in 2006, 2010, and 2012. Milligan Doc. 409-7 at 21; Tr. 1891–92. Dr. Hood testified that he found that “[W]hite conservatives are more than willing to support minority Republican candidates,” Tr. 1892, and that “ideology trumps race in the case of [W]hite Republicans and their support for GOP minority nominees,” Milligan Doc. 409-7 at 21. Dr. Hood also cited other articles, id. at 21–22, and opined that “the result of elections is impacted by ideological congruence rather than race of the candidate,” Tr. 1895.

To examine these findings in Alabama, Dr. Hood discussed the election of Representative Paschal, a Black Republican, from a majority-White district. Tr. 1893–95. Dr. Hood testified that Representative Paschal was the first Black Republican elected to the Legislature since Reconstruction; that his district is in Shelby County, which does not overlap with the congressional districts at issue in this litigation; and that “you can't make a statewide generalization from a single state house election within Alabama.” Id. at 1921–24.

On cross-examination, Dr. Hood acknowledged that there were no Alabama races analyzed in his “True Colors” article; the article provides “no analysis of White voters’ willingness to vote for a minority candidate in Republican primaries”; and the article “make[s] no specific findings as to White voter support for Black Republican candidates.” Id. at 1912–15. He then read from a footnote in “True Colors” that described another scholarly finding that: “White Democratic voters were more supportive of Black Democratic candidates than White Democratic candidates, whereas White Republicans were less supportive of Black Democratic candidates as compared to White Democratic candidates, a decline of four to five percentage points.” Id. at 1918 (reading from Milligan Doc. 456-2 at 7). He was then asked to read into the record several other quotations from “True Colors,” including:

Unobtrusive survey methods under experimental designs reveal significantly more racial prejudice towards minority candidates.

For instance, not only does prejudice exist toward African-American candidates, but it is even more severe in the case of those with darker complexions -- a more nuanced analysis that goes beyond the simple reality of racial distinctions.

․

Challenging the notion of a more color-blind new [S]outh, [other scholars] employ an unobtrusive survey list experiment and find that racial prejudice among [S]outhern Whites is very high vis-a-vis their [N]orthern counterparts and especially among White [S]outhern males, a pillar of support for the contemporary GOP.

Finally, [another scholar] uses an experimental survey to conclude that, although some Whites not bold enough to express their disapproval of an African-American candidate that they otherwise are expected to support, cuing a racial issue, affirmative action, for example, leads to a large increase in the number of undecided White voters, and this is interpreted as racial prejudice.

In other words, most of these White voters are not really undecided but they just do not want to admit they oppose the Black candidate.

Id. at 1919–20 (reading from Milligan Doc. 456-2 at 7).

On redirect, Dr. Hood read into the transcript several quotations from the abstract of “True Colors,” including:

In this study, we assess the level of support that White conservative voters give to minority Republican candidates.

Controlling for various factors, we consistently find that White conservatives are either more supportive of minority Republicans or just as likely to vote for a minority as they are a White Republican.

In other words, a null result.

Although we hesitate to dismiss the presence of racial prejudice in voting behavior, in the case of White conservatives, our analyses suggest that the base of the GOP does not discriminate against minority nominees in high-profile contemporary general elections.

At a minimum, the level of [ideological] polarization in American politics masks racially prejudiced voting behavior and, at a maximum, it renders it inoperable because White conservatives view recent minority Republican nominees as at least as conservative as White GOP nominees, and their level of support reflects this.

Id. at 1954–56 (reading from Milligan Doc. 456-2 at 2). Dr. Hood further affirmed on redirect that when he discussed in “True Colors” other scholars’ work, he was not adopting that research as his opinion. Id. at 1956–57.

Fifth, Dr. Hood examined whether Black political metrics have changed over time in Alabama. See Milligan Doc. 409-7 at 2; Tr. 1895. He studied the number of Black elected officials in Alabama from the passage of the Voting Rights Act in 1965 to the present day. See Tr. 1896; Milligan Doc. 409-7 at 22. He testified that there were no Black members of the Legislature in 1965, three Black senators and thirteen Black representatives in 1981, and there are currently seven Black senators and twenty-six Black representatives. Milligan Doc. 409-7 at 22; see Tr. 1896.

Dr. Hood also studied Black voter registration rates. Tr. 1896. He observed that in 1965, 23.5 percent of eligible Black voters were registered to vote, and that number was 95.2 percent in 2024. Milligan Doc. 409-7 at 23; Tr. 1896–97. He thus opined that “there have been significant gains for Black Alabamians across the last six decades.” Milligan Doc. 409-7 at 23; see Tr. 1896–97.

On cross examination, Dr. Hood agreed that of the thirty-three Black Alabama legislators, thirty-two are from majority-Black districts (with Representative Paschal as the sole exception). Tr. 1926–27. And Dr. Hood acknowledged that those majority-minority districts were the product of and maintained by the Voting Rights Act and associated litigation. Id. at 1927–28.

On cross examination, Dr. Hood further testified that he offered no testimony to dispute Dr. Palmer's or Dr. Liu's findings of racially polarized voting; Mr. Cooper's or Dr. Duchin's conclusions about the first Gingles preconditions; Dr. Bagley's analysis or conclusions; or Dr. Burch's conclusions. Id. at 1898–99. He also testified that he did not dispute Dr. Palmer's opinion that “[a] more complete look at primary candidates shows that Black Republicans are rarely successful,” id. at 1924 (discussing Caster Doc 303-2 at 2–3, ¶ 9), and acknowledged that elections featuring a minority candidate would be more “probative” in a racially polarized voting analysis, id. at 1944. And Dr. Hood testified that he found racially polarized voting in Alabama. Id. at 1945.

Dr. Hood also testified during cross examination about two of his publications. First, he testified that in his book The Rational Southerner (2012), he wrote:

Simply put, we found, as the theory of relative advantage predicted, that the growth of [S]outhern Republicanism was primarily driven by racial dynamics, not class, demographic factors, or religion, as others have suggested.

․

at the midpoint of the last century, according to [another scholar], [S]outhern politics revolved around the issue of race. Southern politics in the early 21st Century still revolves around the issue of race.

․

Stated succinctly, the partisan and political transformation of the [S]outh over the past half-century has, most centrally, revolved around the issue of race.

․

What we can say is that the [S]outhern party system over the past half-century revolved around issues of race, not class. Much of the recent work on the American party system has clearly then underemphasized the crucial and distinctive role that race and racial dynamics have played.

․

Though we are not there yet and it is unclear when, or if, we ever will be, race has left an indelible imprint on the region, and it would certainly be a mistake to ignore the potential future role of racial dynamics in [S]outhern politics and, by implication, national politics.

․

there is considerable evidence that empowerment plays an important role in the extent to which African-Americans participate in a variety of political acts and activities. These range from donating money to political campaigns and attending campaign meetings to contacting local officials and participating in various community activities. But when the focal activity is registering to vote, it is difficult to see how empowerment can precede -- in a causal sense -- the focal mobilization activity. This is particularly true in the American [S]outh.

If empowerment is understood as significant representation and influence in political decisionmaking and it is measured by the prevalence of Black elected officials, including mayors, legislators, and members of Congress, then the manifestation of Black empowerment cannot logically precede the presence of a significant number of registered Black voters.

Id. at 1929–31, 1933–35 (reading excerpts from Caster Exhibits 157 and 159).

On redirect about this book, Dr. Hood was asked whether Black and White voters are “divided on account of racism or some sort of racial policy preferences,” and he answered: “I would say it's more policy preferences today. I'm not saying that racism has been completely stamped out. But in sort of a big picture scheme of things, I think it's more of policy-driven outcomes.” Id. at 1949.

Second, Dr. Hood testified about one of his articles published after these cases began called “Switching Sides but Still Fighting the Civil War in Southern Politics” (2022). Id. at 1936–38. There, he testified he wrote:

Not only does an overwhelmingly [W]hite electorate now favor the GOP in [S]outhern politics, but in this article we have also shown with an inventory of public opinion data that the party's adherents have reached back in time to defend the Lost Cause Myth.[45] Thus, in this regard, our findings support racially motivated explanations for partisan change in the South.

․

Hence, it stands to reason that contemporary debates over Confederate symbolism and public memory reflect ongoing conflict over racial inequality. Today's racialized partisan cleavage reflects a similar divide over views of a racially charged past.

․

It is true that the modern [S]outhern Republican Party stands for a host of things beyond being more racially conservative than its Democratic opponent. But it is also undeniable that the successful GOP strategy of attracting [S]outhern [W]hites by capturing the conservative position on African American civil rights has ultimately led to the reality that the Republican Party has now become the defender of the very flag that [W]hite [S]outherners once raised against the party of Lincoln on bloody battlefields and later in violent skirmishes over [B]lack equality. In addition, modern-day GOP adherents are also much more supportive of honoring the Confederate fallen, as we have shown with public opinion data on Confederate monuments. Finally, contemporary [S]outhern [W]hite Republicans are also the primary apologists for an almost universally disavowed historical argument that the “War Between the States” was mainly about states’ rights, as opposed to slavery. This development has come to fruition despite the fact that our data clearly show that [W]hite [S]outherners very much value the South's history and a large majority still think the Civil War remains relevant to American politics. Thus, the weight of the evidence shows that in “still fighting the Civil War,” [W]hite [S]outherners have rewritten history, at least with respect to switching partisan sides in their defense of the Lost Cause.

Caster Doc. 374-2 at 13–14 (internal citations omitted); Tr. 1939–41. On redirect, Dr. Hood opined that shifts in political parties and power in the South were not on “account of racism,” and that race was a descriptive, not causal, factor. Tr. 1950–53.

b. Dr. Christopher Bonneau

Dr. Bonneau holds graduate degrees in political science from Ball State University and Michigan State University and works as a professor of political science at the University of Pittsburgh. Milligan Doc. 384-1 at 3; Tr. 1660. He has co-authored or -edited three books and several chapters and articles on judicial elections. Milligan Doc. 384-1 at 3; see Tr. 1662–63. He has qualified as an expert witness in four other cases. Milligan Doc. 384-1 at 2; see Tr. 1663. He was compensated at a rate of $350 per hour in these cases and his compensation did not depend on the substance of his testimony. Milligan Doc. 384-1 at 2.

At trial, Dr. Bonneau was admitted with no objection as an expert in “American political science, election analysis, and political science research methodology.” Tr. 1665.

The State asked Dr. Bonneau to (1) “ascertain whether Black candidates in elections in Alabama perform worse than [W]hite candidates on account of their race,” and (2) respond to the opinions of Dr. Liu and Dr. Burch. Milligan Doc. 384-1 at 2, 13–20. On the first task, Dr. Bonneau testified that he limited his analysis and opinions to the question whether the race of the candidate matters to their success, and he did not consider the race of the voter. Tr. 1742–43, 1766, 1862. He also testified that although it can be difficult to reach conclusions with a small number of elections, “you go to war with the data you got, not the data you want.” Id. at 1819–20.

In his report, Dr. Bonneau opined that voting in Alabama is primarily based on political party, not race. See Milligan Doc. 384-1 at 5–13; Tr. 1666. He testified that he examined statewide judicial and legislative elections and that approximately two-thirds of Alabamians vote by “straight-ticket.” Milligan Doc. 384-1 at 5. He opined that that “[t]he prevalence of straight-ticket voting means that most voters are voting for a political party, not a candidate.” Id.; Tr. 1694–95. At trial, he testified that straight-ticket voting demonstrates that many Alabama voters vote for “teams,” not “candidates.” Tr. 1695–98.

Dr. Bonneau conceded that his initial report did not consider straight-ticket voting patterns by race of the voter, but opined that it would be likely that straight ticket voting is being used by both White and Black voters. Id. at 1699–708, 1829. Dr. Bonneau testified that “a significant predictor of how well Democrats do in Alabama is solely a result of the percentage of African-American voters in that county.” Id. at 1678. And he estimated in his supplemental report the straight-ticket voting attributable to race in the 2022 Alabama gubernatorial election. Id. at 1704–08; Milligan Doc. 387-1 at 4–7.

Dr. Bonneau also opined about two Alabama House elections: one when a White candidate (Philip Ensler) defeated a Black candidate (Malcolm Calhoun) in the Democratic primary in a majority-Black district in 2022, and another when a Black candidate (Kenneth Paschal) defeated a White candidate (Sheridan Black) in a majority-White district in 2021. Milligan Doc. 384-1 at 10–12; see Tr. 1685–89; Milligan Doc. 384-3 at 21. He opined that these two elections “indicat[e] that race is not the driving force behind vote choice” and that voters “make selections based on the candidate's positions as well as their political party affiliation.” Milligan Doc. 384-1 at 11–12.

At trial, Dr. Bonneau characterized Representative Paschal's election as a “unicorn” because “you have a Black Republican winning an election in Alabama,” and he acknowledged that Representative Paschal was “the first Black Republican to win election to the State House since Reconstruction.” Tr. 1688.

Dr. Bonneau also opined about Alabama Supreme Court elections between the 1980s and 2000. Tr. 1668–72; Milligan Doc. 384-1 at 4. He testified that both Black candidates and Democratic candidates enjoyed little success in Alabama judicial elections after the state became majority-Republican. Tr. 1667–76; see Milligan Doc. 384-1 at 4. And he attributed the lack of success for those candidates in part to lower campaign spending. See Tr. 1676–77; Milligan Doc. 384-1 at 6–9. On cross examination, Dr. Bonneau conceded that his campaign spending opinion is “not drawing any conclusions about the extent to which party is a better explanation for voting behavior than race,” nor “any conclusions about the extent to which party is a better explanation for the observed election results than race.” Tr. 1756.

In his report, Dr. Bonneau opined that “[i]n a multivariate regression model including both the percentage of the registered Black population and whether the losing [Alabama] state supreme court candidate was Black as independent variables,” Black candidates “perform 4.3 percentage points better than [W]hite candidates.” Milligan Doc. 384-1 at 10. At trial, Dr. Bonneau admitted a coding error in his data on this point (he coded certain uncontested elections as contested). Tr. 1679–80. He testified that when he corrected this error, his results flipped: the data indicated greater success (as defined by vote share) for White Democrats than Black Democrats. Id. Dr. Bonneau testified that after the correction, only one Black candidate would remain in the dataset for contested elections, and he “would not have done this kind of analysis with only one candidate because there's nothing to explain; it's just one election.” Id. Ultimately, he testified that with only one election to study, “[t]here's no way to distinguish between idiosyncratic factors and more systematic factors when you only have one case.” Id. at 1680.

At trial, Dr. Bonneau also testified that he examined the 2024 Republican primary in District 2 and the 2022 congressional districts statewide, and his review led him to conclude that there is “a strong relationship between African-American voters and votes received by the Democratic Party candidate.” Id. at 1690–94.

As to his second task, Dr. Bonneau criticized Dr. Liu for examining only biracial elections; Dr. Bonneau argued that approach “assumes that there are differences based on the race of the candidate” and fails to account for the role of political party. Milligan Doc. 384-1 at 18; Tr. 1717–22. Dr. Bonneau testified that Dr. Liu limited his focus to biracial elections based on an assumption “that voters always prefer to vote for candidates of their own race.” Tr. 1720. He further testified that Dr. Liu's analysis of nonpartisan elections may not effectively control for party because voters can know the partisan affiliation of a candidate even when the candidate does not run on a party platform. See id. at 1723–25. Dr. Bonneau testified that Dr. Liu's reports are “consistent with the story that political party is the most important factor here and not race.” Id. at 1727; see Milligan Doc. 384-1 at 18.

Dr. Bonneau further testified that Dr. Palmer's racially polarized voting analysis did not control for race or party because “all the Black-preferred candidates in Dr. Palmer's report [were] Democrats” and Dr. Palmer did not “attempt to analyze the role of political party.” Tr. 1725–26. According to Dr. Bonneau, Dr. Palmer observed racially polarized voting, but did not conduct a causal analysis. Id. at 1726.

Nevertheless, Dr. Bonneau testified that he “do[es] not dispute” the findings of Dr. Liu and Dr. Palmer that “Black voters vote differently than White voters,” id., nor contest that “Black voters vote cohesively in Alabama,” and “White voters ordinarily vote as a block sufficient to defeat those Black voters’ choices,” id. at 1862. And Dr. Bonneau conceded that Dr. Palmer's ecological inference analysis showed that “White voters in Alabama support White Democrats more than they support Black Democrats,” id. at 1789, and that if “Black voters [are] voting in higher numbers for Black Democratic candidates than White Democratic candidates, that [could] be an indication that race is the driving force behind vote choice,” id. at 1767.

Dr. Bonneau testified that his academic work found “that Black voters have historically faced unique impediments to registration and voting” and had lower turnout because of these impediments. Id. at 1731–32. And he testified “that race is likely a reason why some” Black candidates do not have success, but that it is “really, really difficult to disentangle race and party,” id. at 1792, 1794–95.

Finally, Dr. Bonneau testified that ecological inference is “the preferred method by the court” and “probably” the best method available, but that Dr. Liu's testimony that it is “one of the best methods in the history of political science” “is overstated significantly,” and other evidence should be considered. Id. at 1859–60.

c. Dr. Wilfred Reilly

Dr. Reilly holds a law degree from the University of Illinois College of Law and a doctoral degree in political science from Southern Illinois University. Milligan Doc. 384-4 at 2; Tr. 2161. He works as a professor of political science at Kentucky State University, where he has taught for approximately ten years. Tr. 2161; Milligan Doc. 384-4 at 2. His research focuses on race relations, public law, political theory, and the statistical examination of gaps between racial groups. Milligan Doc. 384-4 at 2; Tr. 2162. He has published four books, four book chapters, and numerous articles. Milligan Doc. 384-4 at 2–3, 32–38; see Tr. 2163. Dr. Reilly does not hold himself out as an expert in redistricting or communities of interest in Alabama. Tr. 2164.

Dr. Reilly has testified as an expert witness in only one previous case (the state Senate redistricting trial pending before Judge Manasco). Id. at 2211. He was compensated at a rate of $500 per hour for his work in these cases and his compensation did not depend on the substance of his testimony. Id. at 2328–29, 2332.

At trial, Dr. Reilly was admitted with no objection “as an expert on political science, with a focus on public law, international and race relations in political theory, statistics, group comparisons, methodology and research methods, and socioeconomic gaps and their causes.” Id. at 2164.

Dr. Reilly focused his opinions on two matters: first, the basis (or lack thereof) for joining portions of Mobile with portions of the Black Belt in a congressional district, and second, socioeconomic disparities between Black and White Alabamians.

As to the first issue, Dr. Reilly's report challenged the basis for joining portions of Mobile with portions of the Black Belt to create a majority-Black congressional district. Milligan Doc. 384-4 at 3–4. He opined that it “is at very best highly debatable” that the City of Mobile “shares more characteristics with rural Black Belt counties than with Mobile County itself and with closely neighboring and long-aligned Baldwin County.” Id. He opined that “ ‘work/live’ patterns in both directions indicates far more connectivity between Mobile, Mobile County, and Baldwin County than between Mobile County and any of the Black Belt counties.” Id. at 4. He suggested that “it is difficult to argue that an increasingly sophisticated city of 200,000 has more in common with a series of small agrarian counties hundreds of miles away ․ than with the large urban/suburban counties immediately adjacent to it.” Id. at 5. Ultimately, he opined that “the proposed majority-minority district likely to result from redistricting makes little sense in the context of any goal but securing more votes for the Democratic Party.” Id.

At trial, Dr. Reilly testified that there are more similarities between Mobile County and Baldwin County than there are between Mobile County and the Black Belt Counties (with the exception of Montgomery County). Tr. 2228. Dr. Reilly testified that based on his review of “commuting patterns,” “71 percent of the people who work in Mobile County live in Mobile County.” Id. at 2165–66. He further testified that “[a]bout 13 percent live in neighboring Baldwin County, which is metropolitan or suburban Mobile.” Id. at 2166. He testified that these commuting patterns establish “a very close Mobile-Baldwin relationship.” Id. Additionally, Dr. Reilly testified that “78.2 percent of the people that live in Mobile City or Mobile County work somewhere in Mobile County” and that “[e]ight percent work in Baldwin County.” Id. at 2168.

Dr. Reilly also testified that Mobile and Baldwin counties are both “sizeable, populated place[s]” while “Black Belt counties are almost all small, more agrarian, [and] semirural” with low populations. Id. at 2170. He also testified that “[t]he per capita income in Mobile, Mobile County, and Baldwin County was over $30,000,” which contrasts with “every single Black Belt County, except for Montgomery County.” Id. at 2171. Dr. Reilly also testified that jobs between Mobile and Baldwin counties are more similar than jobs between Mobile County and Black Belt counties. Id. at 2174–75.

As to the second issue, Dr. Reilly reported data about “disparities in performance or behavior that currently exist between Black and [W]hite Alabamians.” Milligan Doc. 384-4 at 4. In his report, he attributed those disparities to “cultural variables such as fatherlessness and family structure, structural-level variables such as Great Society welfare policy, and ․ the plain out-sourcing of millions of American jobs in the not-too-distant past.” Id. at 5 (cleaned up). He disclaimed the possibility of “contemporary bias, or even a still-lingering effect of past bias.” Id.

Dr. Reilly based his opinions about racial disparities in Alabama on national data. See id. at 17–29; Tr. 2183. Dr. Reilly conceded that socioeconomic disparities exist between Black and White Alabamians, but opined that such gaps “exist almost literally everywhere in the United States.” Milligan Doc. 384-4 at 4; Tr. 2183. Dr. Reilly repeatedly testified that racism does not explain these disparities, Tr. 2184, 2188, 2189, 2192, 2196, 2205, 2208, 2209, and he disavowed hereditary or genetic explanations, id. at 2185.

Dr. Reilly observed that “[W]hite students perform better educationally than Black students ․ in every single state,” Milligan Doc. 384-4 at 18, and Black people are more likely to be illiterate than other racial groups, Tr. 2252. But he opined that “the size of contemporary group gaps in SAT scoring and college attendance correlates only slightly with any sensible measure which might be used to document historical racism.” Milligan Doc. 384-4 at 18–19; see Tr. 2194–95. Dr. Reilly testified that Asian-American students outperform White students in SAT scores even though it is “completely implausible” that Whites are experiencing more racism than Asians. Tr. 2196; Milligan Doc. 384-4 at 19. Similarly, Dr. Reilly testified that “Nigerian-Americans ․ are the highest educated group in the USA” even though it is “completely implausible” that “Whites experience more racism than ․ Nigerians,” Tr. 2196. Dr. Reilly conceded that “new immigrants, like Nigerians, may not be experiencing the same generational effects of discrimination as descendants of enslaved Black Americans.” Id. at 2259.

Dr. Reilly testified that educational disparities are caused by cultural factors such as “time spent reading books,” “[s]tudy time,” “[p]arental expectations,” “[s]ocioeconomic class,” “[t]elevision time allowed,” and whether the individual is an athlete or varsity sportsman. Id. at 2196–97.

On cross-examination, Dr. Reilly admitted that “until [his] testimony in [the state Senate redistricting trial], [he was] not aware that federal courts in Alabama have recently entered orders requiring the school systems in Jefferson County, Huntsville, and Madison County to address their failure to offer Black students equal access to advanced courses.” Id. at 2257.

Dr. Reilly testified that he is “aware of testimony in this case or in expert reports from the Plaintiffs that Black voter turnout and registration has lagged behind White voters in some elections.” Id. at 2191. He opined that these gaps are not statistically significant, Milligan Doc. 384-4 at 27, or the result of racial discrimination, Tr. 2191; and he attributed them to age, fatherlessness, and felon disenfranchisement, id. at 2191–92. Dr. Reilly testified that “nationally, the average African-American is a little under 30” while the “[a]verage White American is over 50” and “young people are much less likely on average to vote than seniors.” Id. He also testified that “[f]atherlessness ․ bluntly has a negative impact on most forms of civic participation.” Id. at 2191. And he testified that “felon disenfranchisement” where “something like 15 percent of Black Alabamians can't vote because of previous felony crimes” is another reason for relatively lower Black voter turnout. Id. at 2192.

Dr. Reilly opined that “Black Americans are over-represented relative to [W]hite Americans in every state prison system in the country,” Milligan Doc. 384-4 at 22–23; see Tr. 2202–03, even though “[n]ationally and in Alabama ․ more than twice as many Whites as Blacks ․ commit crimes every year,” Tr. 2268. He testified that “offenses vary dramatically across population groups” and that “[d]ifferent groups have different pathologies.” Id. at 2206. He testified that “Black Americans generally commit the actual numerical majority of murders.” Milligan Doc. 384-4 at 29; Tr. 2205–06, 2268. He testified that “because the sentence handed down for murder is often life or closer,” Tr. 2268, “murderers make up the largest single bloc of inmates across the USA's federal and state prisons,” Milligan Doc. 384-4 at 29. Dr. Reilly opined that “realities like this largely explain racial disparities in long-term incarceration rates.” Milligan Doc. 384-4 at 29; see Tr. 2268–69.

On cross examination, Dr. Reilly conceded that he is “not a professional expert on [S]outhern politics” and that his academic work did not focus on Alabama politics. Tr. 2212. He testified that he has not “conducted academic research on the concept of communities of interest” or “on defining communities of interest” in Alabama “or anywhere else.” Id. He testified that he is “not a historian” or “an expert on Alabama history.” Id. at 2214. And he testified that he did not evaluate Alabama-specific data to form his opinions on social gaps, and his analysis of disparities was based on national data. Id. at 2249–50.

Dr. Reilly also testified about various matters concerning his credibility and the reliability of his opinions. He conceded that he relied on sources such as Zillow and Wikipedia in his scholarly works and expert report, id. at 2217, 2230–31, and that some data in his report is uncited, id. at 2306–07.⁴⁶ In response to Dr. Bagley's report, Dr. Reilly testified that he has “some questions about what food insecurity means in the United States, given obesity rates.” Tr. 2239. He also testified that “past history as a slave or conflict state does not predict the size of SAT or incarceration gaps,” even as he conceded he did not conduct analysis on that issue in this case. Id. at 2251. And when he was asked why one of his tables depicted “a thousand people liv[ing] in Mobile but work[ing] in Tuscaloosa” and “3,600 liv[ing] in Mobile and work[ing] in Jefferson County,” commutes of several hours each, Dr. Reilly responded, “I don't know exactly what the source of that is. I noticed that myself, in fact.” Id. at 2342.

Dr. Reilly described himself at trial as “an evocative writer.” Id. at 2232. Despite his disavowal of genetic explanations for racial gaps in educational attainment, he admitted that he posted on social media that “this whole debate illustrates why it is so silly to pretend IQ science does not exist” and that “U.S. Blacks at 92 [and] Whites at 103 ․ [is] correct.” Id. at 2312. He also posted that “[y]ou could literally pay smart Black people to have kids or boost Black merit immigration, to boost Black IQ, which, given what we do know about biracial scores, probably isn't low'ish for genetic reasons. This is really a very solvable problem.” Id. at 2353–55.

d. Dr. Adam Carrington

Dr. Carrington holds graduate degrees in political science from Baylor University and now works as an associate professor of political science at Ashland University (formerly, he was a professor at Hillsdale College for ten years). Milligan Doc. 384-2 at 1; Tr. 1546–47. His research focuses on “American political institutions in their historical context, including the judiciary, the presidency, and political parties” and he has published a book, book chapters, and articles. Milligan Doc. 384-2 at 1, 39–41; see Tr. 1548, 1583. Other than the recent state Senate redistricting trial before Judge Manasco, Dr. Carrington has never previously served as an expert witness in litigation. See Milligan Doc. 384-2 at 43. Dr. Carrington was compensated at a rate of $300 per hour for his work and his compensation did not depend on the substance of his testimony. Id. at 1.

The State offered Dr. Carrington “as an expert in political science, political parties, and the partisan shift in the American South,” and the Plaintiffs objected to his qualifications as to “identified partisan shift in the American South.” Tr. 1550–51. We received his testimony. Id. at 1553.

Dr. Carrington testified that his doctoral degree did not focus on the South and he has not published work about post-Reconstruction politics in the South (aside from work about the judiciary in the South). Id. at 1583–85. He testified that his academic work has not focused on post-1960s politics in the South, nor Alabama politics. Id. at 1583–84. Dr. Carrington testified that his scholarly work has not focused on partisan alignment post-1960s, nor the role of race in partisan alignment. Id. at 1585–86.

Nevertheless, Dr. Carrington opined that he “sought to provide a fuller context for how Alabamians in 2024 come to identify with and vote for one of the two major political parties,” Milligan Doc. 384-2 at 36, and he testified that Southern politics today is not dominated by race, Tr. 1556. At trial, Dr. Carrington testified that the campaigns of former Alabama Governor George Wallace show the diminishing power of race in Alabama politics as early as 1971. Dr. Carrington testified that in 1968, Wallace's “anti-integration viewpoints helped him attract support among White voters in Alabama when he ran for president,” id. at 1606, and that “as early as his 1971 inauguration, he is saying the government of Alabama is for all Alabamians, White and Black,” id. at 1610. Dr. Carrington acknowledged that Wallace might not have “meant” what he said in 1971, but said that nevertheless, those statements “show[ ] that, even among a staunch segregationist, he saw things he had to say and argue in an Alabama context that shows a diminishing power of race.” Id. at 1610.

Dr. Carrington testified that he does not dispute the existence of racially polarized voting in Alabama, only the reasons why voting is racially polarized. Id. at 1596–97. He testified that he does not “deny that race plays any factor whatsoever in the minds of any voters in Alabama in 2024.” Id. at 1581.

Dr. Carrington testified about the history of the realignment of the South from majority-Democrat to majority-Republican, and he focused on White voters. See id. at 1553–71. He testified that the shift was not solely or primarily caused by race, but by differences in factors such as economics, foreign policy, and social issues like religious ideology or abortion. Tr. 1557, 1571–72; see generally Milligan Doc. 384-2. Dr. Carrington testified that he analyzed White voting patterns, but not Black voting patterns. Tr. 1555, 1597, 1601. He testified that Alabama patterns aligned with Southern patterns, but he did not study Alabama elections. Id. at 1608, 1628–31.

Dr. Carrington also testified about shifts in Southern voters who identify as religious, opined that race does not trump religion among Alabama voters, and argued that voters’ positions on social issues drive their party affiliations. See id. at 1571–78. In response to criticisms from Dr. Bagley and Dr. Burch, Dr. Carrington conceded that he did not evaluate the religious beliefs or observance of Black voters, or its effect on their voting patterns. Id. at 1618–20. But he acknowledged that Black and White Christians in Alabama hold similar views on abortion. Id. at 1621.

At trial, Dr. Carrington testified about the use of racial appeals in several national campaigns. See id. at 1564, 1592–93, 1595–96, 1629–30. He conceded that, aside from his responses to Dr. Bagley's report, he did not evaluate any recent campaign advertisements of Alabama politicians and did not reach any conclusions about whether campaigns in Alabama are characterized by racial appeals. Id. at 1588. Dr. Carrington conceded that former Congressman Brooks's reference to a “war on Whites” may have been an “attempt[ ] to appeal to White voters.” Id. at 1594.

Dr. Carrington also testified about matters going to his credibility and the reliability of his opinions. Dr. Carrington was asked about three prominent figures in Alabama history: Fred Gray, U.W. Clemon, and Frank Johnson. See id. at 1622–23, 1645. In all three instances, Dr. Carrington did not know who the person was and could not offer a single sentence about them or their work. See id.⁴⁷ When asked whether he “regard[ed] Frank Johnson as beyond the scope of [his] report,” he testified that “it's obviously someone [he] did not look at.” Id. at 1645.

Dr. Carrington also testified about an opinion piece he published about the Supreme Court's affirmance of our first preliminary injunction, in which Dr. Carrington referred to the affirmance as a “missed opportunity” for the Supreme Court to follow pre-1982 Voting Rights Act precedents. Id. at 1626–28. Dr. Carrington testified that he wrote the piece before the State retained him as an expert. Id. at 1626.

4. Fact Witnesses Called at Trial

The State offered two fact witnesses at trial: Ms. Slay and Senator Bobby Singleton.

a. Leonette Slay

First, the State called Ms. Slay, who initially testified in the Singleton Plaintiffs’ case-in-chief (with an objection for consideration in Milligan and Caster cases). On examination by the State, Ms. Slay testified that she “cannot speak to the communities outside of” Jefferson and Tuscaloosa Counties and does “not purport to represent or speak on behalf of any Black individuals or communities within those two counties.” Id. at 2387. She testified that she does not believe the Special Master Plan is racially gerrymandered, and that her “aspirational goal” is to keep Jefferson County whole, but that “if that's not possible, [she] fully support[s] the Court-ordered plan that led to the ability for other counties to elect a candidate of their choice.” Id. at 2389–90.

b. Senator Bobby Singleton

Second, the State offered testimony from Singleton Plaintiff Bobby Singleton, who represents “parts of Tuscaloosa, Hale, Greene, Sumter, Marengo, and Choctaw Counties” in the Alabama Senate. Id. at 2362–63. Senator Singleton is a Black Democrat and has served in the Legislature since 2002. Id. at 2363. He testified that he thought it was important that “Jefferson County should be made whole” in a congressional district and that from an economic standpoint, “it works better for [the residents] to be in a whole county in a district alone.” Id. at 2369–70. He testified that he supports the Special Master Plan and does not believe it racially gerrymanders Jefferson County. Id. at 2370–71. Senator Singleton testified that he has known Senator Livingston “for the last six to eight years” and they interact together in the Senate. Id. at 2373. He said that he has never heard Senator Livingston make a remark that Senator Singleton considers racist, and that he gave Senator Livingston a bear hug at the end of the last special session. Id. at 2374.

5. Testimony from State Senate Redistricting Trial

Of the eight witnesses in the Alabama Senate redistricting trial from whom the parties stipulated we may consider testimony, the State offers five in these cases. See Milligan Docs. 441 at 2–3, 395 at 2–4.

a. Colonel Jonathan Archer

Colonel Archer serves as the Director of the Department of Public Safety at the Alabama Law Enforcement Agency (“ALEA”). Milligan Doc. 441-1 at 7. He previously served as the Chief of the Driver's License Division of ALEA. Id. at 9. Col. Archer testified about Dr. Bagley's assertion that the closures of certain driver's license offices in 2015 was a recent act of official discrimination against Black Alabamians. See id. at 23. He testified that certain offices were closed at that time due to financial and staffing concerns, when ALEA decided that “it would be better to suspend operations in those offices so th[e] examiners [at those locations] could remain at the district offices to serve more customers.” Id. at 16. He testified that the suspension lasted for thirty days and conceded that ALEA reopened the offices as part of a memorandum of understanding with the federal Department of Transportation that did not admit liability for discrimination. Id. at 22, 24, 38.

b. Valerie Branyon

Ms. Branyon is Black and a County Commissioner in Fayette County who represents a district that is half White and half Black. Milligan Doc. 441-2 at 8–9, 12, 28. She ran as a Republican in 2024 and previously ran as a Republican in 2020. Id. at 9–10, 20. In 2020, she defeated a White Republican in the primary but lost to a Democrat in the general election. Id. at 20–21. Ms. Branyon explained that she joined the Republican Party because of its stances on issues like abortion and same-sex marriage. Id. at 10. She testified that her campaign received support from the local and state Party, and the state Party invited her to a campaign training. Id. 16–20, 28; accord Milligan Docs. 459 at 2, 459-2.

c. Cedric Coley

Mr. Coley is a Black Republican voter in Montgomery. Milligan Doc. 441-3 at 7–8, 10, 17. He testified that he joined the Republican Party around 2016 and that members of the Republican party were “welcoming.” Id. at 18–19, 35. He is a member of the Montgomery County Republican Executive Committee and has held various positions in the Montgomery County Republican Party. See id. at 19–21, 23.

Mr. Coley is also involved in the Alabama Republican Party. He testified that he was appointed as regional director of the Alabama Outreach Coalition for the state Party, served as co-chair for Mo Brooks's federal senatorial campaign in Montgomery County, and “served as a field representative helping to consult candidates for ․ the State Senate.” Id. at 21–22. As a field representative, the state Party paid him to advise candidates. Id. at 23–24. Mr. Coley testified that he is also a member of the Alabama Minority GOP, a group that is “a launch pad for minority Alabamians.” Id. at 22.

Mr. Coley testified that he believes there is a “globalist network of international cartels that are deliberately destroying our nation,” and that these cartels are working through the education system, economy, and “sections of the judicial system and some sections of intelligence agencies.” Id. at 44. He testified that the COVID-19 “plandemic” was a bioweapon created by China. Id. at 43–44.

Mr. Coley testified that he does not believe that Republican candidates use racial appeals to attract voters. He also testified about his social media posts. Id. at 45–49. He was asked specifically about his post of an image that depicted two hand gestures. One gesture (that Mr. Coley acknowledged the FBI has described as indicating White supremacy) was by a White hand, above the text “Jobs, vote for civility, vote for prosperity, vote for unity, vote for patriotism, vote Republican.” Id. at 48–49. The other gesture (that Mr. Coley acknowledged has been associated with communism, uprisings, and “[B]lack power,”) was by a dark fist with text that read “Not mobs․ Walk away from violence, walk away from hypocrisy, walk away from globalist Democrats.” Id. at 47–48. At the state Senate redistricting trial, Mr. Coley testified that he never intended to advocate for White supremacy. Id. at 51.

d. Karen Landers, M.D.

Dr. Landers works as the Chief Medical Officer of the Alabama Department of Public Health (“the Department”); she joined the Department in 1982 and became its Chief Medical Officer in 2022. Milligan Doc. 441-5 at 9, 18, 29–30.

On direct examination, Dr. Landers testified that during the pandemic, the Department engaged in outreach to the minority community, offering testing and care in sixty-six out of sixty-seven counties, and engaged in further minority outreach after vaccines became available. See id. at 20–26.

On cross examination, Dr. Landers acknowledged that Black Alabamians were disproportionately hospitalized with and died from COVID-19, id. at 31–32; Black Alabamians are at a higher risk for underlying chronic health problems, such as diabetes or hypertension than White Alabamians, id. at 32–33; and Black Alabamians have less access to health care than White Alabamians, id. at 33. She testified that racial disparities in health care “result from barriers like a lack of access to education and information” that the Department is working to improve. Id. at 40, 47–48.

Dr. Landers testified that in 2022 the Department entered a resolution agreement with the federal government about Lowndes County residents who were without adequate sewage disposal infrastructure, and “no fault was found with the state of Alabama related to any discriminatory practices against the citizens of Lowndes County.” Id. at 36, 43, 49. She testified that progress has been made in the treatment of sewage in Lowndes County. Id. at 43. She testified that the Department initiated a community survey that could be used to rank individuals “to be eligible to apply to get a septic tank system that would be paid for through some funding ․ from the state of Alabama,” and its efforts to educate the community about how to maintain sewage systems after receiving a working system. Id. at 44.

e. Bill McCollum

Mr. McCollum is a Black registered voter who lives in Fayette County. Milligan Doc. 441-6 at 7–10, 22. Mr. McCollum testified that he joined the Republican Party because he “did[ ] [not] like a lot of the policies” advocated by the Democratic Party and liked conservative values. Id. at 9–10. He has served as vice-chairman of the Fayette County Republican Party for 15 years and has been a member of the Alabama Republican Party State Executive Committee for more than 15 years. Id. at 10–12.

He testified about his experience running for office in five elections. He testified that he experienced resistance to his candidacy in his first election in the 1970s, but became the first Black candidate to qualify in Fayette County. Id. at 18–19, 27. And he testified that he received support from the Fayette County Republican Party and Alabama Republican Party in his most recent campaign. See id. at 14–16.

6. Designated Deposition Testimony

The State offered seven witnesses by deposition. Milligan Doc. 259.

a. Kenneth Boswell

Mr. Boswell ⁴⁸ served the city of Enterprise, Alabama for seven years, first on the City Council and then as Mayor. Milligan Doc. 459-1 at 4. Since 2017, he has worked as Director of the Alabama Department of Economic and Community Affairs (“ADECA”). Id. at 5. He testified that ADECA is funded primarily through federal funds and works to expand access to broadband and healthcare. Id. at 6, 8, 13. He testified that its broadband efforts have “had a wonderful impact on all Alabamians that did not have access before.” Id. at 8. He testified that during the pandemic, ADECA “spen[t] dollars on Alabama broadband connectivity for students” and “focus[ed] on low to moderate income levels that did not and could not afford internet.” Id. He said that ADECA focuses on “all areas of the state,” particularly “rural areas.” Id. at 9.

b. Brad Kimbro

Mr. Kimbro ⁴⁹ works as the Chief Operating Officer of the Wiregrass Electric Cooperative. See Milligan Doc. 459-11 at 3, 9–10. He testified that the Wiregrass is “centered around Dothan,” but that “it's made up of a lot more smaller towns.” Id. at 7. He spoke about the infrastructure and economic challenges the Wiregrass faces as a rural farming community, and added that he has spoken with Congressman Moore about these issues. Id. at 7, 11–12, 18–19. He testified about his experience working with community partners to deliver broadband to the area. Id. at 9–11. He also described the importance of Fort Novosel (an Army base), cultural events like the Rattlesnake Rodeo, and Troy University. Id. at 12–14, 23–24.

Finally, Mr. Kimbro said that “he could see someone making a case” for Crenshaw and Pike Counties belonging to both the Wiregrass and the Black Belt. Id. at 15–16. He testified that he believes that the 2023 Plan “represents the Wiregrass region better” than other maps. Id. at 26–27.

c. Lee Lawson

Mr. Lawson ⁵⁰ lives in Baldwin County and serves as president of the Baldwin County Economic Development Alliance. Milligan Doc. 459-12 at 3–5. Mr. Lawson testified that “the City of Mobile is economic hub of South Alabama” and explained that his organization works with Mobile organizations and officials on economic projects. Id. at 5–6. He testified that approximately 60,000 people commute to work between Baldwin and Mobile Counties daily. Id. at 6, 8. He also testified that tourism is Baldwin's County's “largest economic driver” and it “is one of Alabama's fastest growing counties” with “unique” concerns. Id. at 9–10, 13.

Mr. Lawson testified that he does “not see the Black Belt and Mobile” as “being linked together” and that he has concerns about “sharing a large portion of [their] congressional district with Wiregrass counties because that focus and that representation will be diluted across other economic and political priorities.” Id. at 5, 11–12.

d. Gerald Nix

Mr. Nix ⁵¹ works as a statistician at the Alabama Department of Labor. Milligan Doc. 459-17 at 6–7. He “update[s] publications and reports that are on the workforce development portion of the Labor Market Information website within the Alabama Department of Labor website” and promotes use of that website across the state. Id. at 6. Mr. Nix testified that the website data in the county profiles about commuting patterns are not analyzed or confirmed by the Alabama Department of Labor, but are simply pulled from U.S. Census Bureau publications. Id. at 13–14.

e. Mike Schmitz

Mr. Schmitz ⁵² is a former mayor of Dothan who expressed concern about keeping the Wiregrass and Southeast Alabama together in a congressional district. Milligan Doc. 459-21 at 6–7; Milligan Doc. 459-22 at 9. He testified that the Wiregrass Counties have important ties to each other and offered as examples the Southeast Alabama Gas district and Fort Novosel. Milligan Doc. 459-22 at 7–9, 18.

Mr. Schmitz acknowledged that the Black Belt and Wiregrass, as the Legislature defined those areas in SB-5, overlap and might have shared interests because they are both rural. Id. at 11–12. He also testified that the Special Master map resolved “some” of his concerns “[b]ecause they did keep the Wiregrass together” and Dothan would remain in Congressman Moore's district. Id. at 12–14.

f. Derrick Turner

Mr. Turner ⁵³ is a Black resident of Baldwin County. Milligan Doc. 459-26 at 3, 7. Mr. Turner grew up in Prichard (north of Mobile) and holds a degree from Tuskegee University. Id. at 6. He previously worked at the Alabama Department of Industrial Relations ⁵⁴ and now works at the Mobile Career Center. Id. at 8–12. Mr. Turner testified about visiting other career centers in the state, estimated that there are 57 career centers in the state, and confirmed that a brochure listed centers in Choctaw, Clarke, Conecuh, Escambia, Monroe, Washington, and Wilcox Counties. Milligan Doc. 459-26 at 12, 18, 21.

Mr. Turner testified that the median income for Black Alabamians is lower than for White Alabamians because “the employment rate [is] lower.” Id. at 26. Mr. Turner testified that the “proportionality” of Black Alabama families living in poverty is greater than that of White families. Id. Mr. Turner testified that “based upon what [he has] heard from media in various stories, that the access to internet and technology is more a regional thing than a racial thing,” but acknowledged that Black Alabamians might proportionally have less access. Id. at 27.

Mr. Turner testified that he initially became a Democrat because of his family, but now affiliates with that party because it aligns with his beliefs about the middle class and civil rights. Id. at 15. Finally, he testified that he believes Black Alabamians have “traditionally” been underrepresented. Id. at 16–17.

g. Jeff Williams

Mr. Williams ⁵⁵ works as the regional president for SmartBank and serves in leadership positions at the Chamber of Commerce and Housing Authority in Dothan. Milligan Doc. 459-27 at 7; Milligan Doc. 459-28 at 4, 5–6. At the bank, Mr. Williams oversees mortgage operations and retail banking in Dothan, Auburn, Destin, Panama City, and Tallahassee. Milligan Doc. 459-28 at 4.

Mr. Williams testified that “the [W]iregrass is one cohesive unit, one cohesive area that works together.” Milligan Doc. 459-27 at 5. He testified that he would have concern about being placed in a congressional district with Mobile because it “is a completely different geography and culture” with “different types of industries.” Id. at 10. He testified that the Wiregrass lacks interstate access and a major airport and centers around a military base and agriculture. Milligan Doc. 459-28 at 6–7.

Mr. Williams testified that three counties overlap in the definitions of the Wiregrass and Black Belt in the 2023 Plan. Id. at 8–9, 16. And he acknowledged that the Special Master Plan kept the five counties around Dothan together, so his representation would be unlikely to change in the 2024 election. Id. at 10–11.

E. Attacks on Section Two

1. Constitutionality of Race-Based Redistricting

The State argues that even if Section Two “could have authorized race-based redistricting in the past, race-based redistricting justified by § 2 no longer passes Constitutional muster today.” Milligan Doc. 481 at 220–21, ¶ 597. It argues that since the Voting Rights Act was amended in 1982, “ ‘things have changed dramatically’ in the South ‘in large part because of the Voting Rights Act.” Id. at 222, ¶ 604. The State asserts that “voter turnout and registration rates now approach parity, blatant discriminatory evasions of federal decrees are rare, and minority candidates hold office at unprecedented levels.” Id. at 223, ¶ 605.

2. Private Right of Action

The State argues that “Congress has not expressly authorized private persons to sue under § 2” and “whether § 2 contains an implied private right of action” has been “an open question” unresolved by the courts. Id. at 226, ¶¶ 616–17. The State reasons that “Congress does not create substantive rights when enforcing the provisions of the Fourteenth and Fifteenth Amendments,” and because the Voting Rights Act “is Fifteenth Amendment enforcement legislation,” “it created only ‘new remedies,’ not new rights” that are privately enforceable. Id. at 227, ¶¶ 620, 621. The State further argues that Section Two “did not create the right to be free from racial vote dilution” or “dilutive effects in voting.” Id. at 228, ¶¶ 622–23. It asserts that “the right to an undiluted vote is a constitutional right recognized by the Supreme Court before the [Voting Rights Act] was enacted,” and “[p]rotecting an existing right is not creating a new one.” Id. at 228, ¶ 622.

V. FINDINGS OF FACT AND CONCLUSIONS OF LAW – VOTING RIGHTS ACT

We first consider whether the Plaintiffs have established their Section Two claims. We reiterate that we rely on evidence adduced by all Plaintiffs because all parties have stipulated that absent a specific objection, we may do so. Milligan Doc. 445 at 13 (pretrial order). We then address the State's attacks on Section Two.

A. The Plaintiffs establish a Section Two violation.

1. Gingles I – Numerosity

Based on the parties’ stipulation in Milligan and Caster, we find that “[t]here is a numerically sufficient number of Black people of voting age in Alabama to draw [ ] two majority-Black Congressional Districts.” Milligan Doc. 436 at 25.

2. Gingles I – Reasonable Configuration

We next find that the Plaintiffs have (again) established that, as a group, Black voters in Alabama are “sufficiently ․ compact to constitute a majority in a [second] reasonably configured district.” Allen, 599 U.S. at 18 (quoting Wis. Legislature, 595 U.S. at 402). We analyze this issue six ways, and we begin by making credibility determinations.

a. Expert Credibility Determinations

 Dr. Duchin

We find Dr. Duchin's testimony highly credible. Dr. Duchin is an eminently qualified expert – she has earned relevant degrees from some of the world's finest educational institutions, her research focused on redistricting is regularly reviewed by her peers and selected for publication in leading journals, and her work on redistricting includes both academic and litigation work. See supra Part IV.A.1.

Throughout Dr. Duchin's reports and testimony, her opinions were clear and consistent, and she explained the basis for each step of her analysis and every conclusion she drew. She explained a complex process in a manner that was sufficiently clear for non-mathematicians to understand it, evaluate it, and ask her questions about it. See Milligan Docs. 68-5, 76-4, 385-3, 385-7; Tr. 279–366.

Dr. Duchin subjected her work to very high standards and rigorous quality control. Every time she was asked whether she had reviewed relevant materials, she had. See, e.g., Tr. 300, 321, 285–86, 325. She was careful not to overstate her opinions or testify about matters outside their scope. See, e.g., Tr. 317–18, 320.

During Dr. Duchin's live testimony, we carefully observed her demeanor, particularly as she was cross-examined. She consistently defended her work with careful and deliberate explanations. Her testimony was internally consistent and supported. We find that her work is highly reliable and helpful to the Court.

We particularly credit Dr. Duchin's extensive testimony that in her experienced view, her remedial districts are reasonably configured. See, e.g., Milligan Doc. 385-3 at 4–5, 9; Tr. 282–313; Part IV.A.1. And we credit her testimony that race did not predominate in her map-drawing process. See Tr. 287–89, 292, 365 (testifying that “[a]s a matter of process, race is a consideration that doesn't dominate others”; that she “just did not look at race” as she placed district lines; that she “periodically checked to see if the plan, as a whole, had that property of two majority-Black districts”; and that when she ultimately decided which of her maps she would submit to the Court, she screened out any that did not include two majority-Black districts); see also supra Part IV.A.1; infra Part V.A.2.c.

 Mr. Cooper

We also find Mr. Cooper's testimony highly credible. Mr. Cooper has spent the majority of his professional life drawing maps for redistricting and demographic purposes, and he has accumulated extensive expertise (more so than any other Gingles I expert in the case) in redistricting cases, particularly in Alabama. See supra Part IV.B.1. His command of districting issues in Alabama is sufficiently strong that when he first became involved with these cases, he was immediately confident that he could draw an appropriate remedial plan, and he was able to sketch out a draft in less than a day. Tr. 229–31. We believe him when he says that “it is very obvious” to him that a reasonably configured remedial district is possible, and that “[i]t would just be a question of how you would draw them and how you improve what you may have initially sketched out.” Id. at 230–31.

Throughout Mr. Cooper's reports and his live testimony, his opinions were clear and consistent, and he had no difficulty articulating his basis for them, even on aggressive cross-examination. See Caster Doc. 48; Caster Doc. 65; Caster Doc. 352-1; Caster Doc. 352-2; Tr. 104–234. But he was not dogmatic: he took seriously Dr. Trende's criticism of the compactness of his first eight plans and prepared a ninth plan in response. See Caster Doc. 352-2 at 6–7.

Mr. Cooper's testimony demonstrated his independence of thought and integrity. See Tr. 121–22; 166–67 (testifying that he “routinely” consults with counsel or potential plaintiffs and advises them that in his opinion, they cannot satisfy the first Gingles precondition). This testimony enhances Mr. Cooper's trustworthiness and our regard for his opinions.

During Mr. Cooper's live testimony, we carefully observed his demeanor, particularly as he was cross-examined. He consistently defended his work with careful and deliberate explanations. His testimony demonstrated his respect for the role of the Court and his role as an expert. See supra Part IV.B.1. We observed no internal inconsistencies or other defects in his testimony. We find his work highly reliable and very helpful to the Court.

As with Dr. Duchin, we particularly credit Mr. Cooper's repeated testimony that in his considerably experienced opinion, his remedial districts are reasonably configured. See Caster Docs. 352-1, 352-2; Tr. 109–10, 115–17, 119–197; supra Part IV.B.1. And we credit his extensive testimony that race did not predominate in his map-drawing process. See, e.g., Tr. 124 (answering “[a]bsolutely not” to the question whether race predominated in his plans); id. at 153–54 (testifying that he never split a VTD for “the purpose of bolstering the Black Voting Age Population in a particular district,” nor “for the purpose of creating a majority-Black district”); id. at 172–73. (“I would not have gone to 50 percent plus one for a second majority-Black district if I were not also balancing the other traditional redistricting principles.”); id. at 124 (testifying that he knew race did not have to predominate to draw two majority-Black districts because he “was looking at the traditional redistricting principles” and “taking all of that into account”); see also supra Part IV.B.1; infra Part V.A.2.c. This testimony echoed Mr. Cooper's testimony at the preliminary injunction hearing:

Q. So what specific traditional districting principles did you consider in drawing the illustrative plans in this case?

A. Well, I took all of them into consideration. I examined the document produced back in May by the Alabama Legislature outlining the guidelines for redistricting. But a lot of that just incorporates the general concept of traditional redistricting principles. So I didn't prioritize any of them. I tried to balance them.

․

Q. So was any one factor of the ones we just mentioned predominant, the predominant factor when you were preparing your illustrative plans in this case?

A. Not really. I feel like I gave them equal weighting. It would be possible to prioritize others and come up with different configurations, but perhaps at the expense of one of the key redistricting principles. So you could draw very compact districts, but they might split numerous counties because they're perfect squares. Or you draw a district that is -- two districts that are maybe 60 percent Black, but they wouldn't be contiguous. That, you know, so you have to balance it.

Q. And did race predominate in your development of any of the illustrative plans?

A. No. It was a consideration. This is a Section 2 lawsuit, after all. But it did not predominate or dominate.

Jan. 5, 2022 Tr. 439–41.

This testimony strikes us as particularly trustworthy for two reasons: first, because Mr. Cooper explained the care he took not to alert himself to information about race that might allow it to become predominant. See id. at 143 (testifying that when he draws maps, he “[n]ever ever” “employ[s] color coding ․ by race at the district level,” and he “never look[s] at” color-coded racial heat maps like those Dr. Trende included in his analysis). Mr. Cooper identified with specificity the only information about race that he was aware of as he placed lines: (1) he was “generally aware of where the municipalities that are predominantly Black are,” and (2) he generally “knew where the precincts with a BVAP above [30 percent were].” Id. at 163–64.

Second, Mr. Cooper's testimony that race did not predominate in his mapdrawing strikes us as especially trustworthy because Mr. Cooper explained what did drive his decisions about where to place district lines: “[E]xisting lines or at least existing demarcations by the Census Bureau.” Id. at 145–46. Mr. Cooper explained that such “existing lines” could include “odd-shaped municipalities”; topographical features like mountains, ridges, and valleys; precinct lines; and primary roads. Id. at 144–46. In Mr. Cooper's view, this is the only way to do it properly, and others with a different view (such as Dr. Trende) have misunderstood. Id. at 146–47. He explained:

I mean, [Dr. Trende is] treating Jefferson County like it's just a flat plain that has a bunch of precincts that are only identifiable by whether or not they're five percent Black versus 95 percent Black. I don't -- I mean, I just do not approach a redistricting plan drawing in that fashion.

Id. at 146–47; see also id. at 223 (“municipal lines in and around Mobile are just as tricky as they are in Jefferson County in terms of irregular shapes, water areas”).

We do not have any doubt about the veracity of Mr. Cooper's testimony that race did not predominate, but if we did, it is resolved by his testimony that it would be obvious to us if he had allowed race to predominate. See Caster Doc. 352-2 at 6; Tr. 169–70 (opining and then testifying that he could have drawn majority-Black districts with higher BVAPs by placing communities with higher concentrations of Black Alabamians in majority-Black districts, and communities with higher concentrations of White Alabamians in majority-White districts).

 Dr. Trende

We assign less weight to the testimony of Dr. Trende, the State's only Gingles I expert, for two reasons. First, compared to the work of Dr. Duchin and Mr. Cooper, Dr. Trende's work was limited: Dr. Duchin and Mr. Cooper based their opinions on a wide-ranging consideration of the requirements of federal law and all or nearly all traditional districting principles, but Dr. Trende studied only geographic compactness scores and splits allegedly along racial lines. See supra Part IV.D.1; Milligan Doc. 384-5 at 7; Tr. 1980–82, 2065. Like Dr. Duchin and Mr. Cooper, Dr. Trende testified that there are “trade-offs built into the process,” and he described their role in his work on other cases. See Tr. 1982–83, 2133–35.

But in these cases, where Dr. Trende opines that the Plaintiffs’ illustrative districts “are not reasonably configured,” Milligan Doc. 384-5 at 93; accord Tr. 1980, he did not examine any trade-offs made by Dr. Duchin or Mr. Cooper. See e.g., Tr. 2134–35. It is difficult for us to assign substantial weight to his opinions knowing that he understands the necessity of tradeoffs, but still did not consider them.

We of course acknowledge that such considerations may have been outside the scope of the work the State asked Dr. Trende to perform, and we do not diminish Dr. Trende's qualification, experience, or skill. We simply observe that we give his work limited weight, in line with its limited scope.

Second, at times Dr. Trende's testimony was internally inconsistent. Despite his ultimate opinion that the Plaintiffs’ districts “are not reasonably configured,” Milligan Doc. 384-5 at 93; accord Tr. 1980, he was “reluctant” to say whether a plan is reasonably compact because “[i]t's unclear what the standard is to support that opinion,” Milligan Doc. 384-5 at 17. He explained that “[w]hile there may be extreme cases where no reasonable expert would dispute that a district is compact (e.g., a district with a Reock score of 0.8) or that a district is substantially similar to another district (e.g., a difference in Convex Hull scores of 0.00001), there's ultimately no clear way, at least from an expert perspective, to decide what percentage of a bounding circle a district must fill before it becomes reasonably compact.” Id. At trial, he reiterated that there are no bright-line standards for reasonable compactness, see Tr. 2001, 2014, 2037–40, and offered an analogy: “[I]n the same way that it's difficult to say that ․ your previous counsel has a beard; you're clean-shaven, I can't tell you exactly where in the line stubble turns into a beard or clean-shaven turns into stubble. But I can tell the ultimate differences in cases,” id. at 2001.

We accept the reality that there is not an objective numerical threshold for reasonable compactness. But we cannot reconcile Dr. Trende's hesitation to explain his understanding of reasonableness with his experience in redistricting and his willingness to opine that Plaintiffs’ “districts are not reasonably configured.” See Milligan Doc. 384-5 at 93.

We do not understand that reasonableness opinions have so eluded Dr. Trende in other cases: surely when he is a plaintiff's expert or consults for a jurisdiction, he does not propose maps that he regards as not reasonably compact. Reasonableness is ours to determine, and we would not accept from any expert an ipse dixit opinion about it. We carefully study all the evidence. But the reality that this task is not as simple as identifying a numerical threshold does not mean that we throw up our hands and say that the mapmaker got it all wrong. Without some explanation of what, in Dr. Trende's view, makes a district reasonably compact, we cannot assign much weight to his opinion that the illustrative districts are not reasonably configured.

Further to our concerns about internal inconsistency, as we explain below, at trial Dr. Trende was repeatedly forced to concede that on various scores, the Cooper Plans and Duchin Plans outperform not only the 2023 Plan, but also plans and districts that Alabama has enacted for the past thirty years. See infra Part V.A.2.b.

The corollary of our decision to credit Dr. Duchin and Mr. Cooper, and to give Dr. Trende's opinions less weight, is a finding that the Black population in the majority-Black districts in the Duchin Plans and Cooper Plans is reasonably compact.

b. The “Meet or Beat” Standard

We next explain how we assess the voluminous evidence about the configuration of the Plaintiffs’ illustrative plans. Our task is not to compare the Plaintiffs’ plans with the 2023 Plan to determine which plan would prevail in a “beauty contest.” Allen, 599 U.S. at 21. As the Supreme Court affirmed in these very cases, we do “not have to conduct a ‘beauty contest[ ]’ between plaintiffs’ maps and the State's.” Id.; see also Vera, 517 U.S. at 977 (plurality opinion) (“A § 2 district that is reasonably compact and regular, taking into account traditional districting principles such as maintaining communities of interest and traditional boundaries” is not required “to defeat rival compact districts designed by [the State] in endless ‘beauty contests.’ ”).

Nevertheless, the State urges us to reject the Plaintiffs’ maps in part because they “fail to respect” the 2023 Plan. See, e.g., Milligan Doc. 481 at 71. And the State attacks the Duchin Plans and Cooper Plans on the ground that they underperform against the 2023 Plan on various metrics. See id. at 72 (charting comparisons).

But “meet-or-beat” is not the controlling test. The essential question under Gingles I is and has always been whether the minority group is “sufficiently large and [geographically] compact to constitute a majority in a reasonably configured district.” Allen, 599 U.S. at 18 (quoting Wis. Legislature, 595 U.S. at 402). This standard does not require that an illustrative plan outperform the 2023 Plan by a prescribed distance on a prescribed number of prescribed metrics. An illustrative plan may be reasonably configured even if it does not outperform the 2023 Plan on every (or any particular) metric. The standard does not require the Plaintiffs to offer the best map; they must offer a reasonable one.

Indeed, requiring a plaintiff to meet or beat an enacted plan on every redistricting principle a State selects would allow the State to immunize from challenge a racially discriminatory redistricting plan simply by claiming that it best satisfied a particular principle the State defined as non-negotiable.

Accordingly, a finding that the 2023 Plan preserves communities of interest differently from the Plaintiffs’ maps, or splits counties or municipalities differently from the illustrative maps, does not automatically make the illustrative maps unreasonable. Different maps will necessarily prioritize traditional districting principles in different ways. This is why the maps offered by a Section Two plaintiff are only ever illustrative; states are free to prioritize the districting principles as they wish when they enact a remedial map, so long as they satisfy Section Two. The State has essentially conceded that it failed to do so here, maintaining that it can skirt Section Two by excelling at whatever traditional districting principle the Legislature deems most pertinent.

In any event, as we explain below, we find that the Plaintiffs’ illustrative plans often do meet or beat the 2023 Plan.

c. Visual Assessments

Because both Mr. Cooper and Dr. Trende acknowledged the role of the “eyeball test” to assess compactness, we begin with two visual assessments. First, we assess the geographic concentration of the Black population in Alabama. Dr. Duchin prepared a map that reflects the geographic distribution of Black residents across the state:

Milligan Doc. 68-5 at 12 fig. 3. She described the centers of Black population that are apparent on this map – several urban population centers and the Black Belt. See id. at 12–13. She reported that the Black population in the four largest cities (Birmingham, Huntsville, Montgomery, and Mobile) includes approximately 400,000 people and comprises approximately one-third of the Black population in Alabama. Id. at 12. She also reported that the Black population in the Black Belt, which stretches east to west across the state, includes approximately 300,000 people. Id. at 12–13. These aspects of Dr. Duchin's report are not in dispute.

Our visual assessment of the geographic dispersion of Black population in Alabama, together with statistics about Black population centers in the state, suggest to us that Black voters in Alabama are relatively geographically compact. It is obvious from the map that there are areas where much of Alabama's Black population is concentrated, and that many of these areas are in close proximity to each other. Just by looking at the map, we can see why Dr. Duchin and Mr. Cooper expected they could easily draw two reasonably configured majority-Black districts.

Second, we consider our visual assessment of the majority-Black districts in the Duchin and Cooper Plans. See Milligan Doc. 68-5 at 7 fig. 2; Milligan Doc. 385-3 at 3 fig. 1; Caster Doc. 352-1 at 27, 29, 32, 34, 36, 38, 41, 44 figs. 10, 12, 14, 16, 18, 20, 22, 24; Caster Doc. 352-2 at 8 fig. 1. We do not see tentacles, appendages, bizarre shapes, or any other obvious irregularities that would make it difficult to find that District 2 is reasonably configured in these illustrative plans. We do see that District 7 in all the illustrative plans has what has been referred to as a “finger” that reaches into Jefferson County. But that finger has been there (in some form, and basically the same form) in every congressional map since Wesch, including the 2023 Plan, so it cannot mean that the illustrative plans are any less compact than the 2023 Plan.

d. Geographic Compactness Scores

We next consider industry-standard geographic compactness scores for the Duchin Plans and Cooper Plans, and we find that these scores indicate that the majority-Black congressional districts in those plans are reasonably configured. Dr. Duchin and Mr. Cooper testified that the scores for their plans are reasonable, and Dr. Trende made multiple concessions to that effect.

Dr. Duchin testified that on a Polsby-Popper metric, Duchin Plans A, B, C, and D “are superior to” and “more compact than most of the recent Alabama plans.”. Milligan Doc. 68-5 at 9, 13; Tr. 304. Dr. Duchin made compactness her “top priority” in Plan E and testified that “Plan E is especially compact.” Tr. 289. She testified that the Polsby-Popper score differences between her Plan E and the 2023 Plan were minimal – less than one percentage point. Id. at 357; Milligan Doc. 385-3 at 4.

Dr. Trende opined that “the [Plaintiffs’] Illustrative Maps are all less compact than the [2023 Plan],” Milligan Doc. 384-5 at 28, but elsewhere in his report and at trial, he admitted that some Duchin Plans outperform the 2023 Plan. For example, he conceded that on a Polsby-Popper metric, Duchin Plan B outperforms the 2023 Plan, and that on a Convex Hull metric, Duchin Plans A, B, and C outperform the 2023 Plan. See Milligan Doc. 384-5 at 34. And he conceded that on a Polsby-Popper metric, Duchin Plan E “gets in the ballpark” of the 2023 Plan; he reasoned that “you get into this splitting hairs of, well, the score for E, on average, is nine-tenths of a point lower than the 2023 Map.” Tr. 2001. Dr. Trende also admitted that on a Convex Hull metric, Duchin Plans A, B, C, and D outperform at least four of the plans that Alabama enacted since 1972, and Duchin Plan E outperforms two such plans. Id. at 2125. And he agreed that all the Duchin Plans “are more compact under the Convex Hull measure than California, which [he] identified as a reasonable map.” Id.

For his part, Mr. Cooper repeatedly testified that his plans score “within the normal range.” See Tr. 128–31 (district level); id. at 134–37 (plan level). Based on his work “in lots of states,” he was “comfortable saying that [his scores] ․ stack up well nationwide,” and are “within the norm.” Id. at 191–92.

We particularly focus on Cooper Plan 9, which Mr. Cooper drew in response to criticisms about compactness and which he testified outperforms the 2023 Plan on the Reock metric. Id. at 135. But Cooper Plan 9 goes much further: on the DRA metric, Cooper Plan 9 outperforms not only the 2023 Plan, but also every plan Alabama has used going all the way back to 1992. Id. at 135–38 (Cooper testimony that on the DRA score, his Plan 9 is “significantly more compact” than Alabama's 1992, 2002, 2012, 2021, and 2023 congressional plans). Mr. Cooper described Cooper Plan 9 as “the proof ․ in the pudding” to establish that it is “possible to draw a congressional plan that contains two majority-Black districts that is more compact overall than any of Alabama's enacted plans over the last 30 years.” Id. at 138.

We find multiple instances where Cooper Plan 9 forced Dr. Trende to abandon his opinion that “[Plaintiffs’] Illustrative Maps are all less compact than the [2023 Plan],” Milligan Doc. 384-5 at 28. At trial, Dr. Trende conceded that Cooper Plan 9 is “more compact as measured by Reock score than the [2023 Plan],” and that it is “in the same range of compactness as the [2023 Plan] on the Polsby-Popper score.” Tr. 2042–43. He also agreed that on a DRA measure, Cooper Plan 9 is “more compact than any plan that Alabama has drawn or used in the last 40 years,” including the 2023 Plan. Id. at 2045.

Because Dr. Trende cautioned against overreliance on plan-wide scores (averages), id. at 1991, we next consider Mr. Cooper's testimony about district-level scores for his Districts 2 and 7.

There is no dispute that Districts 2 and 7 in all the Cooper Plans score within the range of what Alabama traditionally has considered acceptable. In the figures that we have attached as Appendix E, Mr. Cooper compared the Reock and Polsby-Popper scores for his Districts 2 and 7 with scores for districts in previously enacted Alabama plans, and he found that his Districts 2 and 7 “fall within the range [of compactness scores] one would find if you just examined” congressional districts Alabama has enacted since 1992. See id. at 131. Dr. Trende conceded this point at trial. See id. at 2049–51. He first agreed “that the Reock compactness scores for all of the majority-Black districts in Mr. Cooper's illustrative plans are within the range of compactness scores for congressional districts that Alabama has enacted since 1992,” and he next agreed that the data establish the same for Polsby-Popper scores. Id. at 2050–51.

Accordingly, we find that the Black population in the majority-Black districts in the Duchin Plans and the Cooper Plans is sufficiently compact that those plans and districts are reasonably configured according to industry-standard measures of geographic compactness. We emphasize that we have not based this finding exclusively on concessions by the State's expert, but we could have: Dr. Trende's concessions about plan-wide scores for the Duchin Plans and Cooper Plans (especially Cooper Plan 9), and district-level scores for the Cooper Plans, establish beyond debate that it is possible to draw a second majority-Black district in Alabama that scores reasonably well on measures of geographic compactness.

e. Reasonable Compactness and Traditional Districting Principles

Ultimately, reasonable compactness is about more than scores and eyeball tests. As Mr. Cooper explained:

I mean, and you can't just look at a score absent a map, absent demographics of the place you're examining and suddenly say, okay; this particular district has a low Reock score, therefore, the plan's no good. Because there could be good reasons for a low Reock score, given the shape of the county, given the shape of the jurisdictions and the shape of the VTDs.

So the Reock, Polsby-Popper scores are not the be all and the end all; they're an indicator, and they have to be taken into consideration with the multitude of other redistricting principles that one deals with when you're drawing a voting plan.

Id. at 175–76. Accordingly, we next evaluate whether the remedial districts in the Duchin Plans and Cooper Plans are reasonably configured by analyzing whether those Plans respect traditional districting principles.

Every Gingles I expert, along with the State's longtime cartographer, testified that redistricting always involves tradeoffs between traditional districting principles. See id. at 286–87 (Dr. Duchin); id. at 157–58 (Mr. Cooper); id. at 1982 (Dr. Trende); Milligan Doc. 459-6 at 40 (Mr. Hinaman). Both Dr. Duchin and Mr. Cooper testified about their extensive efforts to respect traditional districting principles, particularly as enumerated in the 2023 legislative findings (in Duchin Plan E and Cooper Plans 8 and 9), and that they did not ignore any principle. See Tr. 138–42, 147–48, 157–58, 160, 175–76 (Mr. Cooper); id. at 287–89, 293–94, 357, 359–63 (Dr. Duchin).

We discuss each principle in turn.

 Population Deviation & Contiguity

We find that the Duchin Plans and the Cooper Plans equalize population across districts because the parties agree, and the evidence makes clear that they do. See id. at 2066, 119; Caster Doc. 352-1 at 31 fn. 18; Milligan Docs. 385-3 at 4, 68-5 at 8. Likewise, we find that those Plans include only contiguous districts, which the parties agree they do (as the evidence makes clear). See Caster Doc. 352-1 at 24, ¶ 58; Tr. 125, 2066–67; Milligan Docs. 385-3 at 4, 68-5 at 8.

 Respect for Political Subdivisions

We next find that the Duchin Plans and the Cooper Plans respect political subdivisions such as counties, cities, and towns. Multiple Duchin Plans and Cooper Plans fall within the cap on county splits set in the 2023 legislative findings: Duchin Plans D and E, and Cooper Plans 1, 3, 4, 5, 7, 8, and 9 all split six counties or fewer. Milligan Doc. 68-5 at 8 tbl. 1; Milligan Doc 385-3 at 7 n.3, 4 tbl. 1; Caster Docs. 352-2 at 12 fig. 3, 352-1 at 50 fig. 28.

When we consider municipality splits in addition to county splits, we see that Duchin Plan E performs at least as well as the 2023 Plan. See Milligan Doc. 385-3 at 4 tbl. 1. Some other Duchin Plans underperform the 2023 Plan on this measure, but Dr. Duchin prepared those plans before the 2023 Plan was enacted, so she did not consider the 2023 legislative findings when she drew them. See Tr. 289–91.

As for the Cooper Plans, Cooper Plans 7 and 9 outperform the 2023 Plan on county splits and perform similarly on municipality splits. See Caster Doc. 352-1 at 50 fig. 28; Caster Doc. 352-2 at 12 fig. 3. And Cooper Plans 3, 7, 8, and 9 meet or beat the 2023 Plan on municipality splits. Id.; Tr. 140.

Additionally, we credit the testimony of both experts that when they decided where to place district lines, they often followed political subdivision boundaries. Dr. Duchin testified that as she decided where to split Mobile County, she took “guidance from the state board of education map” because that map was “considered legitimate at some point in history by the state legislators.” Tr. 348–50.

Mr. Cooper testified that when he split voting districts, he was “following existing lines or at least existing demarcations by the Census Bureau,” which include “odd-shaped municipalities” and precinct lines. Id. at 124, 144–46, 223. He described the challenges attendant to respecting political subdivisions in Jefferson County, home to much of the Birmingham metropolitan area and 29 separate municipalities that are “just really odd-shaped ․ as a result of annexations and also just because of the challenging topography in the county where you have lots of ridges and valleys and almost mountains.” Id. at 144. Likewise, Mr. Cooper testified that hewing to “municipal lines around the City of Mobile also is a way to split Mobile County and create two majority-Black districts,” and that municipal lines “in and around Mobile are just as tricky as they are in Jefferson County in terms of irregular shapes, water areas. There are a lot of complications in Mobile County.” Id. at 223.

Dr. Trende did not consider in his report whether the Duchin Plans or Cooper Plans respect political subdivisions, see Milligan Doc. 384-5, but he testified at trial that Mr. Cooper split some municipalities in Jefferson County, Tr. 2110. We are unmoved by this testimony because Mr. Cooper did not say he kept all 29 municipalities in Jefferson County whole – he simply said that when he had to draw a line, he tried to use municipal boundaries and other existing subdivisions as his guide. See Tr. 124, 144–46. And nowhere did Dr. Trende address Mr. Cooper's consideration of topography in connection with political subdivision boundaries.

Dr. Trende also opined that when the Duchin Plans and Cooper Plans split counties, they “typically do so on racial lines.” Milligan Doc. 384-5 at 64. As we explain fully below in our discussion of the State's race predominance argument, see infra Part V.A.2.f, we reject this accusation as unsupported. For now, we highlight Mr. Cooper's explanation of how Dr. Trende fundamentally missed the importance of political subdivisions. Mr. Cooper explained with understandable frustration:

I mean, [Dr. Trende is] treating Jefferson County like it's just a flat plain that has a bunch of precincts that are only identifiable by whether or not they're five percent Black versus 95 percent Black. I don't -- I mean, I just do not approach a redistricting plan drawing in that fashion. Perhaps he does. I don't know. But let him speak for himself on that.

Tr. 146–47.

For all these reasons, we find that the Duchin Plans and Cooper Plans respect political subdivisions.

 Communities of Interest

We next find that the Duchin Plans and the Cooper Plans respect communities of interest. We apply the Legislature's definition: “[A] defined area of the state that may be characterized by, among other commonalities, shared economic interests, geographic features, transportation infrastructure, broadcast and print media, educational institutions, and historical or cultural factors.” Milligan Doc. 403-31 at 4 (Ex. MX-31). And we focus on the communities the Legislature identified, all in South and Central Alabama: the Black Belt, Gulf Coast, and Wiregrass.⁵⁶

This issue was hotly disputed and consumed significant time at trial. The State objects to any plan that splits Mobile County (which is every Duchin Plan, every Cooper Plan, and the Special Master Plan), and insists that there can be no legitimate reason to split the Gulf Coast counties, particularly in the light of the 2023 legislative findings. Plaintiffs argue that historic and socioeconomic ties connect the City of Mobile to the Black Belt, stress the importance of the Black Belt, and argue that the State overstates the need to keep the Gulf Coast counties whole and together.

The record contains no map that includes two majority-Black districts without splitting Mobile County, and all agree that it is not possible to draw such a map without splitting Mobile County. Tr. 298–99, 340 (Dr. Duchin), 2648 (counsel for the State). In the simplest terms, these arguments require us to decide whether a district that splits Mobile County is (or can be) reasonably configured and respectful of communities of interest.

 Communities of Interest – the Black Belt

The Black Belt stands out to us as quite clearly a community of interest of substantial significance.⁵⁷ That the Black Belt is an important community of interest is common knowledge in Alabama; has been acknowledged in other redistricting cases, see Ala. Legis. Black Caucus, 231 F. Supp. 3d at 1222; and is clear from the record before us. The Milligan and Caster parties were able to stipulate what counties it includes, where it is located, and why it is known as the Black Belt. See Milligan Doc. 436 ¶¶ 71–73. They further stipulated that the Black Belt “has a substantial Black population because of the many enslaved people forcibly brought there to work before the Civil War.” Id. ¶ 72.

The 2023 legislative findings expressly designate the Black Belt as a community of interest and define it. See App. B. They list the 18 core counties and five “sometimes” counties, and further provide:

The Black Belt is characterized by its rural geography, fertile soil, and relative poverty, which have shaped its unique history and culture.

The Black Belt region spans the width of Alabama from the Mississippi boarder [sic] to the Georgia border.

App. B at 3–4 (enumeration omitted).

Dr. Bagley provided a fuller explanation of the tragic role that slavery played in the shared demographic heritage of the Black Belt:

White settlers began to flood into the state of Alabama when most of the remaining Creek Indians were forced out via the Indian Removal Act of 1830. By then, the United States government had banned the importation of slaves from abroad, so many settlers brought enslaved Black people with them from the older plantation areas of the Upper South. Others purchased them from slave markets in Montgomery, Mobile, Jackson, and other cities. American chattel slavery expanded dramatically between that time and the Civil War, giving rise to the “Cotton Kingdom” of the antebellum era when cotton was America's most valuable export and enslaved Black people were its most valuable commodity. The Black Belt of Alabama became home to not only the wealthiest [W]hite plantation owners in the state, but to some of the wealthiest individuals in the young nation, some of whom held hundreds of people in bondage.

Milligan Doc. 76-2 at 1.

Most Section Two experts testified about the Black Belt. They addressed a range of demographic, cultural, historical, and political issues about how the Black Belt became the Black Belt, how it has changed over time, and what shared experiences and needs there make it unique today. A slew of lay witnesses testified about their understanding of the Black Belt, their connections to it, and its significance to them, their political participation, and Alabama politics. We further heard from those witnesses about meaningful connections between the City of Mobile, the City of Montgomery, and the Black Belt.

This substantial body of evidence established the shared history and economy (or lack thereof) in the Black Belt; the overwhelmingly rural, agrarian experience; the extreme poverty; and major migrations and demographic shifts that impacted many Black Belt residents, just to name a few examples. See, e.g., Tr. 28–31 (Ms. Dowdy), id. at 1092, 1125–26 (Pastor Jackson), id. at Tr. 1188, 1214–15 (Mr. Milligan), Tr. 964–67 (Dr. Burch), Tr. 1299–300 (Dr. Bagley); Milligan Doc. 68-2 at 21. The Black Belt is overwhelmingly Black, but it blinks reality to say that it is a proxy for race – the reasons why it is a community of interest have many more dimensions.

Although the foregoing sufficiently describes why we find that the Black Belt is an important community of interest, it falls far short of adequately describing the shared experience of intense poverty in the rural Black Belt, which is extreme by any measure and so primitive that it often startles people. As just a few examples, we received evidence that:

• Certain Black Belt counties lack proper sewage disposal and drinking water systems. As a result, some rural Black Belt residents construct homemade systems and do not have consistent access to drinking water untainted by raw sewage. Milligan Doc. 68-2 at 21. A 2019 United Nations Report discussed how such residents “often fell ill, entire households at a time, with E. Coli and hookworm.” Id. The lack of proper sewage disposal also means that in some counties, children cannot play outside after it rains. Tr. 480. This problem continues today, and solutions are a “work in progress.” Id. at 481. Although this problem is well-known in Alabama, see id. at 480–81, it was “a new one” for Dr. Burch, a seasoned expert who makes her living studying racial disparities on socioeconomic indicators such as poverty, see id. at 1054.

• Squalid living conditions are not limited to sewage problems. Pastor Jackson testified that in her work in healthcare in Montgomery and the rural Black Belt, she has observed families living with “rodents and roaches,” “[in]adequate heating,” “[in]adequate plumbing,” and without water. Id. at 1091. She described a home visit in the Black Belt when she “could see the sky through the roof[,] and dirt floors,” with people “living off the land.” Id. at 1092. She testified that in some houses, “[t]hey may only have just one room that did not get rained on and maybe had a tarp that would blow off when it was bad weather.” Id. at 1126. She described how some homes “still had outhouses” and “didn't have indoor plumbing” when she made home visits in the Black Belt around 2010. Id. Despite her decades of hands-on service in Montgomery and the rural Black Belt, Pastor Jackson “just did not realize how poor some of our rural areas and the living conditions of people” were and she thought she was “in a third-world country” when she saw these things. Id. at 1092.

• Many rural Black Belt residents struggle with illiteracy. Illiteracy primarily burdens Black Alabamians, to such a degree that Pastor Jackson testified that in all her years of work in the rural Black Belt, although she observed illiterate Black people hold government forms or church hymnals upside down, she never encountered an illiterate White adult. Id. at 1097–98, 1129. Dr. Burch testified that “in the Black Belt especially, there are ․ disproportionately high illiteracy rates, as high as 30 percent.” Id. at 938; see also id. at 998, 1049.

• Basic communication infrastructure such as broadband internet access – and even cell service – are unavailable in many parts of the rural Black Belt, which isolates those who live there from the rest of the modern world. Id. at 715–16, 444. This common reality for the Black Belt is apparently poorly understood outside the Black Belt; Mr. Smith testified about Black Belt students receiving laptops during the pandemic, and that “[y]ou [can] give them a laptop, but a laptop without access to broadband is like having a car without tires.” Id. at 444.

Under the 2023 Plan, the Black Belt is split into three districts: Districts 1 and 2, which the Milligan Plaintiffs assert are cracked, and District 7, which the Milligan Plaintiffs assert is packed. The 2023 legislative findings provide that the 18 core Black Belt counties should be split into no more than two districts. See App. B at 4.

The Duchin Plans contain the overwhelming majority of the Black Belt in two districts, both of which are majority-Black, and Duchin Plans C and E keep all 18 core Black Belt counties in two districts, both of which are majority-Black. Milligan Docs. 385-3 at 8, 5 tbl. 2, 68-5 at 13, 7 fig. 2. The Cooper Plans “place over 70% of the Black Belt counties in a majority-Black district, and four of [his] plans place all but one of the Black Belt counties in a majority-Black district.” Caster Doc. 352-2 at 14. In contrast, the 2023 Plan places “only half (nine) of the Legislature's 18 identified Black Belt counties [ ] in a majority-Black district.” Id.; App. E at 3.

We thus have no difficulty finding that the Plaintiffs’ plans respect the Black Belt as an important community of interest, and that they respect it better – much better – than the 2023 Plan does. The State offers no rebuttal; their position is that it is per se unreasonable to split part of Mobile County away from the Gulf Coast to connect it with the Black Belt. See Milligan Doc. 481 at 57, ¶ 146; Tr. 2625–27.

Because we find that the illustrative plans respect the Black Belt, we need not consider how illustrative Districts 2 and 7 might perform in a beauty contest against other districts in other plans that respect other communities of interest. Together with our finding that the Duchin Plans and Cooper Plans respect political subdivisions, our finding that these plans respect the Black Belt supports a conclusion that they are reasonably configured. Nevertheless, we next consider carefully the other two communities of interest the Legislature specified in the 2023 legislative findings: the Gulf Coast and the Wiregrass.

 Communities of Interest – the Gulf Coast

The 2023 legislative findings designate Alabama's Gulf Coast counties (Mobile and Baldwin) as a community of interest and describe it at length, for more than two pages. See App. B at 4–7. The Legislature's description discusses the shared coastal geography (Mobile Bay and the Gulf of Mexico coastline) and associated industries (the Port of Mobile, fishing, and tourism). See id. It also refers to the “distinct culture” of the Gulf Coast “stemming from its French and Spanish colonial heritage” and mentions Mardi Gras. Id. at 6. At trial, we heard from both expert and lay witnesses about overlapping economic interests, commuting patterns, shared heritage, cultural events, and unique challenges that connect the Gulf Coast counties. See, e.g., Tr. 206–09, 405–06, 1417, 1467–68.

Because the Legislature found that the Gulf Coast is a community of interest, we find that it is a community of interest. We thus turn to the parties’ disputes about the Gulf Coast, which pertain to (1) the steps the Legislature took to prioritize it, (2) the weight it should be afforded as an inviolable community of interest, particularly relative to Section Two, and (3) whether Plaintiffs’ illustrative maps respect it.

As to the first issue, we discuss at length below the unusual lengths to which the Legislature went to prevent Alabama's congressional districting plan from splitting Mobile County. See infra Parts V.A.4 & VII. For present purposes, we find that the 2023 legislative findings had the practical effect of elevating the Gulf Coast as the most important community of interest in Alabama, decreeing that it may not be split, and prescribing a majority-White district there. As Dr. Duchin explained, the requirement to keep Mobile and Baldwin Counties whole and together “comes close to prescribing” a majority-White district on the Gulf Coast because “those [Counties] contain more than 90 percent of the population of a congressional district” and “as a matter of mathematical necessity,” a district that fully includes both Gulf Coast counties must be majority-White and would “submerge[ ]” the City of Mobile. Tr. 298–99, 314. No document, testimony, or lawyer disputes Dr. Duchin's opinion on this point. Indeed, counsel for the State conceded in closing argument that he is “not aware of a way to draw two majority-Black districts without going against the Legislature's priority of keeping Mobile and Baldwin County whole.” Tr. 2648. We further find that the 2023 legislative findings had the practical effect of elevating the Gulf Coast community of interest over all other traditional districting principles in Alabama – including compliance with federal law and nondilution of minority voting strength.

As to the second issue, we reject the State's argument that there can be no legitimate reason to split Mobile County because of the overriding importance of the Gulf Coast community of interest. As these cases make abundantly clear, the Black Belt and Gulf Coast communities of interest are in tension with one another, they pull in different directions, and in fact, they overlap in some ways. We thus cannot accept the suggestion that splitting one county in the Gulf Coast to better respect the Black Belt reflects a wholesale refusal to consider communities of interest, or dispositively establishes that the Plaintiffs’ plans are not reasonably configured. As the Supreme Court explained the last time we considered this issue, upon our finding that the Plaintiffs’ maps are reasonably configured because they join together the Black Belt as a community of interest, we need not conduct a “ ‘beauty contest[ ]’ between plaintiffs’ maps and the State's” as to the Gulf Coast — “[t]here would be a split community of interest in both.” Allen, 599 U.S. at 21.

To be clear, we accept that the Gulf Coast is a community of interest, but we cannot accept the Legislature's effective designation of it as unsplittable, nor its designation of it as superlative to all other traditional districting principles. We particularly cannot prioritize that effective designation above compliance with Section Two, or above the Black Belt for that matter. If evading the requirement of an additional opportunity district under Section Two were as easy as enacting a rule against splitting a specific majority-White community of interest, Section Two would have no meaning.

Likewise, neither the evidence nor the law supports a mandate that we or Dr. Duchin or Mr. Cooper must prioritize the Gulf Coast above all other communities of interest. The evidentiary record about the Gulf Coast (including the 2023 legislative findings) revolves primarily around shared economic interests in coastal industries, commuting patterns, and cultural events like Mardi Gras. See, e.g., Tr. 206–09 (Mr. Cooper); Tr. 405–06 (Dr. Caster); Tr. 1417, 1467–68 (Dr. Bagley). The evidentiary record about the Black Belt (including the 2023 legislative findings and the parties’ stipulation) is broader and deeper; it includes evidence about a shared life experience that is overwhelmingly rural, agrarian, and extremely impoverished, major migrations and demographic shifts that impacted many Black Belt residents, and a common heritage undeniably traceable to slavery. The record about the Gulf Coast is fuller now than it was in the preliminary injunction proceedings, but it still lacks a basis, if there could be one, for mandating the elevation of the Gulf Coast above the Black Belt (and every other community of interest in Alabama) or declaring it inviolable and therefore unsplittable. After all, Mobile and Baldwin Counties were split for “nearly 100 years” “from 1875 until the 1970s,” during which time Mobile was paired with large portions of the Black Belt to form a congressional district. Tr. 1305. And as Dr. Bagley summarized in his expert report: “Mobile and Baldwin were, first, united in order to prevent the reelection of a Black incumbent and, 100 years later, reunited in for similar racial reasons.” Milligan Doc. 385-1 at 9.

We are also mindful that the Legislature splits the Gulf Coast in the State Board of Education districting plans at present and that it decided to privilege the Gulf Coast in the 2021 Plan at the same time it decided to split it in the Board of Education map. Milligan Doc. 385-1 at 8–9; Tr. 1304–08.

As to the third issue, we find that the Plaintiffs’ plans did not fail to give due consideration to the Gulf Coast. The premise of the State's argument that the Plaintiffs’ plans are not reasonably configured because they split Mobile County and splitting Mobile County is intolerably harmful to the Gulf Coast. Yet splitting a community of interest does not always disrespect (or even disadvantage) that community. We received testimony from multiple witnesses that splitting a county in a community of interest does not necessarily harm the community — it may increase its representation. Dr. Duchin explained that “[r]espect for communities of interest can mean keeping them together. But there are times when respect or consideration for communities of interest, instead, might call for a split.” Tr. 315. She gave an example of when a fellow mapmaker revised a map to keep an area together and was “pilloried in the press for taking away a representative” because the area “used to have two representatives and now only has one.” Id. at 316. Mr. Cooper explained that it is not always necessary to keep a community of interest whole to respect it. Id. at 195–96. And Dr. Trende agreed that “not every community of interest will be or can be kept together in a congressional district.” Id. at 2133.

At trial, Mr. Cooper testified that his plans respect Mobile County “even though it's divided between two districts.” Id. at 195–96. He previously explained:

Well, in the illustrative plans, all of the illustrative plans include a significant portion of the city of Mobile, or in the case of District 6 and 7, all of Mobile. In illustrative plan 1, the only -- the primary area of Mobile that I excluded from District 2 is the waterfront area of Mobile, which is actually a grouping of precincts that are predominantly African-American and I put into District 1 so that there was a transportation route between District 1 and Mobile County and District 1 in Baldwin County. So you don't need to drive outside of District 1 to get from one part of District 1 to the other. You have a straight route going across U.S. 98 and Mobile Bay. And there are a few precincts that are split along that route I-10 area coming in to downtown Mobile. And that actually is a feature of most of my plans, except for illustrative Districts 6 and 7 -- illustrative plans 6 and 7, which keep all of Mobile whole, extending it right up to the waterfront.

Jan. 5, 2022 Tr. 451–52.

And Dr. Duchin testified that her plans respect Mobile County at least as much as the Legislature respected it in other districting maps. The Legislature has repeatedly split Mobile and Baldwin Counties in districting maps for the State Board of Education, and the Legislature did so at the very same time it drew the previous congressional plans. See, e.g., Tr. 325–27, 348, 2090–97, 2024–30; Milligan Doc. 384-5 at 56–57. At trial, Dr. Duchin explained that she took “some guidance from the state board of education map” because that was “considered legitimate at some point in history by the state legislators.” Tr. 348. Dr. Trende diminished the State Board of Education plan as “a one-off configuration” that may be based on previous federal preclearance requirements or “inertia,” but he conceded that in making that guess, he did not speak with legislators, a historian, an expert, or anyone in Mobile or Montgomery about potential educational needs or explanations. Milligan Doc. 384-5 at 93; Tr. 2094–97. We add that Dr. Trende's speculation in no way answers the fact that the Legislature split Mobile County from Baldwin County in drawing the 2020 State Board of Education plan, seven years after the preclearance regime had been eliminated in Shelby County.

Split communities of interest are inevitable in any plan, and we must evaluate them in an intensely local appraisal. We find that when Dr. Duchin and Mr. Cooper split Mobile County, they made just such an appraisal. We further find that when they explained it to us, they clearly and amply justified every districting decision the State challenges. We thus find that their decisions to split Mobile County did not violate communities of interest or produce unreasonably configured plans.

 Communities of Interest - The Wiregrass

The 2023 legislative findings identify the Wiregrass as a community of interest and offer a two-sentence definition that it “is characterized by rural geography, agriculture, and a major military base” and “home to Troy University's flagship campus in Troy and its campus in Dothan.” App. B at 7. The 2023 legislative findings identify nine counties that comprise the Wiregrass (Barbour, Coffee, Covington, Crenshaw, Dale, Geneva, Henry, Houston, and Pike), three of which overlap with the Black Belt (Barbour, Crenshaw, and Pike). Id. At trial, testimony from expert and lay witnesses connected the Wiregrass counties to each other. Tr. 1312–13 (Dr. Bagley); Milligan Doc. 459-11 at 7 (Mr. Kimbro); Milligan Doc. 459-21 at 6–7 (Mr. Schmitz); Milligan Doc. 459-28 at 9–10 (Mr. Williams).

Because three counties overlap the Black Belt and the Wiregrass, and together they have more population than one congressional district can accommodate, it is mathematically impossible to keep both communities of interest whole, and any districting decision will necessarily prioritize one over the other. See Tr. 298; App. B. The 2023 legislative findings address this reality by providing that the 2023 Plan keeps all Wiregrass counties together in District 2, except that it places Covington County “in District 1 so that the maximum number of Black Belt counties c[ould] be included within just two districts.” App. B at 7.

Although the State “introduced precious little evidence to establish the existence of the Wiregrass community of interest” at the preliminary injunction stage, we considered in our remedial order the fact that Remedial Plan 3 kept together six Wiregrass counties, whereas Remedial Plan 2 kept together only five. Milligan Doc. 311 at 38–39.

When asked about the Wiregrass, Mr. Cooper testified that he accepted it as a community of interest based on the Legislature's definition. Tr. 205. He further testified that though he did not keep the Wiregrass in a single district because of overlapping Black Belt counties (like Barbour), his plans still respect the Wiregrass because “the counties are generally left intact.” Id. at 205–06. And Dr. Duchin testified that she considered the Wiregrass when drawing her illustrative plans “once the new guidelines had been issue[d].” Id. at 318–19.

Because the Legislature found that the Wiregrass is a community of interest, we find that the Wiregrass is a community of interest. But we do not find that the failure of the Duchin Plans and Cooper Plans to keep the Wiregrass whole indicates that the remedial districts in those plans violated communities of interest or are not reasonably configured. Because the Duchin Plans and the Cooper Plans include remedial districts that we find reasonably compact (in part because they respect the Black Belt), we need not consider whether another plan could outperform them in a beauty contest by respecting instead the Wiregrass (or the Gulf Coast). Because of the demonstrable overlap between and the population of the three communities of interest that the Legislature identified, no map can equally respect all three.

 Communities of Interest – Equal Treatment

We specifically reject the State's argument that the 2023 Plan “ended th[e] inconsistent treatment” of these three communities of interest, Milligan Doc. 481 at 5, and treats them equally because it “employ[s] the same line-drawing standards in minority communities of interest as it used elsewhere,” id. at 216, ¶ 582; accord Milligan Doc. 220 at 27, 42 (the State's previous argument that the 2023 Plan “rectifies what Plaintiffs said was wrong with the 2021 Plan” by “unifying the Black Belt while also respecting the Gulf and Wiregrass communities of interest”); Aug. 14 Tr. 39 (the State's oral argument that the 2023 Plan “cures the cracking” of the Black Belt); Milligan Doc. 267 at ¶ 225 (the State's earlier argument that “there is no longer any need to split the Gulf” to respect the Black Belt, because the 2023 Plan keeps the Gulf Coast together and splits the Black Belt into only two districts).

The problem with this argument is the faulty premise that splitting the Black Belt into only two districts remedies the cracking problem found in the 2021 Plan. “Cracking” does not mean “divided,” and the finding of vote dilution in the 2021 Plan rested on a thorough analysis, not the bare fact that the 2021 Plan divided the Black Belt into three districts. See, e.g., Milligan Doc. 107 at 55, 147–74; Allen, 599 U.S. at 19–20, 22–23. As the Supreme Court has explained, “cracking” refers to “the dispersal of blacks into districts in which they constitute an ineffective minority of voters.” Bartlett, 556 U.S. at 14 (plurality opinion) (quoting Gingles, 478 U.S. at 46 n.11).

The Plaintiffs have established — and the State previously conceded — that in the new District 2, Black voters remain an ineffective minority of voters. Milligan Doc. 251 ¶¶ 5–9. This evidence — and concession — undermine the State's assertion that the 2023 plan remedies the cracking of Black voting strength in the Black Belt simply by splitting the Black Belt into fewer districts. In turn, it explains the reason why there remains a need to split the Gulf Coast: splitting the Black Belt as the 2023 Plan does (by placing half the core Black Belt counties in a majority-White district) dilutes Black voting strength, while splitting the Gulf Coast precipitates no such racially discriminatory harm.

 Incumbency Protection and Core Retention

The 2023 legislative findings provide that “[t]he congressional districting plan shall not pair incumbent members of Congress within the same district.” App. B at 2. We assign this principle little weight because after the Legislature enacted those findings, we ordered Alabama to conduct the 2024 election with the Special Master Plan, which paired two incumbents. In that election, one incumbent was not re-elected to Congress (Congressman Carl), and Congressman Figures was elected for the first time and remains the incumbent now. Moreover, while a state legislature may consider incumbency and attempt to protect incumbents when redistricting, that protection may not be used as a means to defeat the Voting Rights Act.

Core retention “refers to the proportion of districts that remain when a State transitions from one districting plan to another.” Allen, 599 U.S. at 21. The 2023 legislative findings provide that the plan shall “[p]reserve the cores of existing districts” to the extent that core retention can be given effect consistent with the non-negotiables enumerated in the findings. App. B at 2. We do not assign this principle substantial weight because, as the Supreme Court explained in these cases, it cannot defeat an otherwise-meritorious Section Two claim:

[The Supreme Court] has never held that a State's adherence to a previously used districting plan can defeat a § 2 claim. If that were the rule, a State could immunize from challenge a new racially discriminatory redistricting plan simply by claiming that it resembled an old racially discriminatory plan. That is not the law: § 2 does not permit a State to provide some voters “less opportunity ․ to participate in the political process” just because the State has done it before.

Allen, 599 U.S. at 22 (quoting 52 U.S.C. § 10301(b)).

In any event, although the 2023 Plan better maintains the cores of existing districts than do the Duchin Plans, Cooper Plans, and Special Master Plan, it does not follow that Dr. Duchin, Mr. Cooper, and the Special Master ignored core retention. Dr. Duchin demonstrated a similar level of retention between Duchin Plan E and the 2023 Plan as there was between the 2023 Plan and the 2011 Plan. See Milligan Doc. 385-3 at 4, tbl. 1. And she demonstrated a higher level of core retention for the Special Master Plan. See id. (Which is unsurprising because the Special Master Plan “meaningfully chang[ed]” only two congressional districts. Milligan Doc. 311 at 38–40.)

 Findings about Traditional Redistricting Principles

Accordingly, we make three findings. First, we reject the State's overdrawn assertion that the Plaintiffs’ illustrative maps “chop up Alabama's important communities of interest.” Milligan Doc. 481 at 76. There is no fair reading of the record that suggests that splitting a single county that the State would prefer to keep whole “chop[s] up” Alabama's important communities of interest. Indeed, the State does not deny that that split better serves a community of interest in the Black Belt that all agree is important.

Nor does the State deny that for many years, the Legislature has split the Gulf Coast counties in the districting plan for the State Board of Education, even in iterations of that plan enacted well after the ruling in Shelby County freed Alabama from federal preclearance requirements. See Milligan Doc. 384-5 at 64; Tr. 2097–98. The State's accusation that the Plaintiffs’ plans “chop[ped] up” important communities of interest by following the lead of a districting plan the Legislature is overdrawn and wrong. Milligan Doc. 481 at 76.

Second, we reject Dr. Trende's unsupported and largely abandoned opinion that “[Plaintiffs’] Illustrative Maps are all less compact than the [2023 Plan].” Milligan Doc. 384-5 at 28. The foregoing analysis illustrates why Section Two jurisprudence demands that our compactness analysis consider traditional districting principles: because it is impossible to understand whether a plan is reasonable or extreme without evaluating the extent to which the plan may be explained by traditional districting principles. It will ordinarily be difficult to draw an unreasonable district while respecting traditional districting principles, and it will ordinarily be difficult to draw a reasonably configured district while ignoring traditional principles. Because Dr. Trende considered none of this, we cannot make findings about reasonable compactness from his opinion.

Third, we find that the evidence clearly establishes that Dr. Duchin and Mr. Cooper studied the Legislature's redistricting guidelines and 2023 legislative findings, considered traditional districting principles, and made careful decisions about how to prioritize principles when circumstances forced tradeoffs. Dr. Duchin and Mr. Cooper not only respected traditional districting principles, but also explained the many ways and reasons they did so. The State does not give Plaintiffs enough credit on this traditional districting principle.

f. Race Predominance

Finally, we reject the State's assertion that the Plaintiffs’ illustrative plans are not reasonably configured because race predominated in their creation. The State delivers this argument at top volume, arguing that Plaintiffs “aim [to] segregat[e] voters on the basis of race,” “[d]irect evidence abounds that race was the predominant motivating factor for Dr. Duchin and Mr. Cooper,” their testimony is “saturated with express acknowledgment[ ]” that race predominated in their work, and “no one disputes that [they] purposefully established a racial target” of two majority-Black districts. See Milligan Doc. 481 at 77, 80 (internal citations and quotation marks omitted).

We reject these accusations for three separate and independent reasons. First, they badly misstate the record. As we have already explained, Dr. Duchin and Mr. Cooper repeatedly testified at trial that race did not predominate in their mapdrawing processes, and we credit that testimony. See supra Parts IV.A.1, IV.B.1, V.A.2.a. The State ignores it. See Milligan Doc. 481 at 77–79.

Rather, the State splices together Dr. Duchin's preliminary injunction testimony and trial testimony to suggest that she conceded at trial that she placed district lines in her plans based on racial targets. See id. But Dr. Duchin was asked about that at trial, expressly disclaimed it, and explained at length how she placed lines based on traditional districting criteria. And when she testified that she would “need to cross [the majority-Black] threshold in order to submit the map to the Court,” she was not admitting that she set a racial target as she drew maps; she was explaining that of the many maps she drew, she screened out of her expert report the maps that did not contain two Black-majority districts. See Tr. 288.

As for Mr. Cooper, the State impugns his belief that “the non-dilution of minority voting strength is a traditional redistricting principle,” see Milligan Doc. 481 at 80 (quoting Tr. 156), but the Legislature shared that belief as recently as when the Committee adopted the 2023 guidelines, see App. A. And the State describes his testimony about how he considered race as “increasingly telling,” insinuating impropriety, but each alleged admission is simply something Mr. Cooper explained he knew (for example, Mr. Cooper knows that much of the Black Belt is predominantly Black). See Milligan Doc. 481 at 80. The law does not require that a Gingles I expert be completely race-blind, and we will not infer from things Mr. Cooper knew that he assigned race a predominant role in his mapmaking process, particularly in the light of his testimony that he did not.

Dr. Duchin and Mr. Cooper very much dispute that they purposefully set a racial target for their work. Nowhere did they admit that they assigned race a predominant role. And nowhere did the State engage, let alone rebut, Mr. Cooper's explanation of how obvious it would have been if he had assigned race a predominant role. See generally id.

As Gingles I experts, Dr. Duchin and Mr. Cooper were tasked with determining whether it was possible to draw two reasonably configured majority-Black districts consistent with traditional districting principles, and they found more than a dozen ways to do it. They both testified that if a map did not include two remedial districts while also respecting traditional principles, they would not submit it to the Court. They both testified that they did not give race a predominant role as they drew. Having exhaustively studied the issue, we credit that testimony.

Second, we reject the State's assertions because, as we also have explained (see supra Part V.A.2.a), for every accusation of race predominance, Dr. Duchin and Mr. Cooper provided a specific explanation (and a non-race-based explanation at that) for the placement of their lines, which Dr. Trende either attacked without support or simply ignored, and which the State now ignores. In the State's argument that race predominated, it makes no mention of the oddly shaped municipalities and topography in Jefferson County that Mr. Cooper discussed, nor the district boundaries for the State Board of Education map in Mobile County that Dr. Duchin discussed. See Milligan Doc. 481 at 77–81.

In a later discussion about the oddly shaped municipalities in Jefferson County, the State asserts that we should not credit Mr. Cooper because he split some municipalities in Jefferson County, and Dr. Trende says those splits were along racial lines. See id. at 88. But as we previously explained, this is an unsupported swing at a straw man. Mr. Cooper did not testify that he kept all 29 municipalities in Jefferson County whole – he simply said that when he had to draw a line, he tried to use existing subdivisions as his guide. Dr. Trende offered no evidentiary basis for his bald assertion that the splits were along racial lines. And nowhere did Dr. Trende or the State address Mr. Cooper's consideration of topography. See Milligan Docs. 384-5, 481.

Similarly, in a later discussion about the State Board of Education map, the State repeats Dr. Trende's assertion that following that map is not a “race-neutral reason” to “split Mobile County along racial lines” because it may include lines drawn to satisfy the Voting Rights Act. See Milligan Doc. 481 at 92–96 (citing Milligan Doc. 384-5 at 56–57). But the State simply ignores Dr. Duchin's testimony that she was unaware of the reason (whatever it was) why the Legislature drew the State Board of Education districts as it did. Compare id., with Tr. 325. Because any potential racial considerations underlying the State Board of Education map (an issue we need not and do not decide) were unknown to Dr. Duchin, they cannot possibly have predominated in her map drawing process.

Third, we reject the State's assertion that race predominated in the Duchin Plans and Cooper Plans because even Dr. Trende's testimony does not support that conclusion. Dr. Trende is the only expert who testified for the State about this issue. At trial, Caster’s counsel repeatedly asked him whether his testimony was that race predominated in the preparation of the Cooper Plans. See Tr. 2112–13. He steadfastly refused to say. He repeated his accusations that Mr. Cooper's lines “follow the racial contours of Jefferson County pretty tightly,” and “follow the racial contours of the various counties,” and he asserted that it was “obvious” that “when [Mr. Cooper] splits municipalities, it's on race.” Id. But he insisted that he was “not offering a predominance opinion” because that was for us to determine after hearing Mr. Cooper's explanations for his lines. Id. And Dr. Trende either ignored or never offered evidence to rebut those explanations. Id.

We find no evidence that Dr. Duchin and Mr. Cooper allowed race to predominate, and extensive evidence that they took great care to avoid that fault.

***

Ultimately, all the arrows point in the same direction: the Duchin Plans and Cooper Plans, and the remedial districts in them, are reasonably configured. Regardless whether we credit expert testimony, make our own visual assessment, review statistical scores of geographic compactness, consider the extent to which those Plans and districts respect traditional districting principles, or do all these things, the result is the same: the Plaintiffs have offered at least one illustrative plan that contains only equipopulous and contiguous districts that are reasonably geographically compact; respects existing political subdivisions; protects important and overlapping communities of interest; protects all incumbents except one; and provides two majority-Black districts without allowing race to predominate in the drawing process. Although Plaintiffs were not required to meet or beat the 2023 Plan on these metrics, they quite often did.

We emphasize that the Plaintiffs have far surpassed their burden here. Federal law requires them to submit one map to show that it is possible to draw a reasonably configured remedial district, and they have submitted more than a dozen such maps. We have no doubt that a reasonably configured remedial district is achievable, and we find that the Plaintiffs have established the first Gingles precondition.

3. Gingles II and III – Racially Polarized Voting

We find (again) that there is no serious dispute that Black voters in Alabama are “politically cohesive,” nor that each challenged district's “white majority votes sufficiently as a bloc to enable it ․ to defeat the [Black] preferred candidate.” Allen, 599 U.S. at 18; see also Milligan Doc. 107 at Part V.B.3.

a. Expert Credibility Determinations

 Dr. Liu and Dr. Palmer

We credit the testimony of Dr. Liu and Dr. Palmer. Both experts have credentials that include substantial academic work in electoral politics and significant experience testifying in redistricting cases. See supra Parts IV.A.2, IV.B.2. Both witnesses consistently and thoroughly explained their work in these cases and the bases for the conclusions they reached; they employed commonly accepted methodologies; and we discern no reason to question their methods or conclusions. We carefully observed their demeanor, particularly on cross-examination, and their testimony was internally consistent, thorough, and well-supported. None of the State's experts conducted a racial polarization analysis to contradict their findings, and as we explain below, many of their conclusions are not disputed. Accordingly, we find their opinions highly credible, reliable, and helpful.

 Dr. Hood and Dr. Bonneau

Although we recite concessions by Dr. Hood and Dr. Bonneau in our analysis of the second and third Gingles preconditions, because their testimony focuses on the Senate Factors, we defer our credibility determination until that discussion.

b. Patterns of Racially Polarized Voting

In our first preliminary injunction, we explained that there was no serious dispute between the parties that Black voters in Alabama are “politically cohesive,” nor that each challenged district's “white majority votes sufficiently as a bloc to enable it ․ to defeat the [Black-]preferred candidate.” Allen, 599 U.S. at 18 (quoting Gingles, 478 U.S. at 51). The record is more voluminous now, but the reality remains the same.

We credit Dr. Liu's testimony that has consistently emphasized the clarity and extremity of the pattern of racially polarized voting he observes in Alabama. When he testified about biracial endogenous elections in the preliminary injunction proceedings, he testified that racially polarized voting is “very clear” in Alabama, Jan. 10, 2022 Tr. 1293; “Black support for Black candidates was almost universal” and “overwhelmingly in the 90[%] range,” id. at 1271; Black voters were “super cohesive,” id. at 1274; and the Black-preferred candidate was defeated in every election outside the majority-Black district, id. at 1275. His exogenous election data confirmed these findings. Id. at 1275–76.

At trial, Dr. Liu reiterated these statistics, Tr. 573–76, and gave us some perspective: he testified “in [his] more than 20 years [of] research, this is arguably the highest level” of racially polarized voting that he has “ever seen,” id. at 576, and that the level of racially polarized voting in Alabama (particularly in the challenged districts) is “one of the highest in the nation,” id. at 578.

And we credit Dr. Palmer's agreement. In the first preliminary injunction proceedings, Dr. Palmer repeatedly invoked adjectives and adverbs that indicate to us that voting in Alabama is clearly and intensely racially polarized: he opined that “Black voters are extremely cohesive,” Caster Doc. 49 ¶ 16; “White voters are highly cohesive,” id. ¶ 17; “[i]n every election, Black voters have a clear candidate of choice, and White voters are strongly opposed to this candidate,” id. ¶ 18; and he described the evidence of racially polarized voting across the districts he studied as “very strong,” Jan. 6, 2022 Tr. 701. Dr. Palmer based these adjectives and adverbs on statistical findings quite like Dr. Liu's: Dr. Palmer found that “[o]n average, Black voters supported their candidates of choice with 92.3% of the vote,” and “[o]n average, White voters supported Black-preferred candidates with 15.4% of the vote, and in no election did this estimate exceed 26%.” Caster Doc. 49 ¶¶ 16–17.

At trial, we again heard from Dr. Palmer the adjectives, adverbs, and statistical findings like Dr. Liu's. See Caster Doc. 303-1 ¶¶ 14–15 (opining that in the elections he studied, on average, “Black voters supported their preferred candidates with 93.0% of the vote,” and “White voters supported Black-preferred candidates with 14.3% of the vote, and in no election did this estimate exceed 26%”); Tr. 487.

We see no dispute about these patterns. At the preliminary injunction hearing, Dr. Hood conducted his own ecological inference analysis and repeatedly acknowledged that he either agreed with or did not dispute the findings of Dr. Liu and Dr. Palmer that voting in Alabama (and specifically in the challenged districts) is racially polarized. Jan. 11, 2022 Tr. 1421–22; Milligan Doc. 68-1 at 18; Caster Doc. 49 ¶¶ 6, 13–22. Dr. Hood testified that he and Dr. Liu “both found evidence of” racially polarized voting in Alabama, Jan. 11, 2022 Tr. 1421; he did not dispute “Dr. Palmer's conclusions that Black voters in the areas he examined vote for the same candidates cohesively,” Jan. 11, 2022 Tr. 1445; he did not dispute “Dr. Palmer's conclusion that Black Alabamians and [W]hite Alabamians in the areas he examined consistently preferred different candidates,” id.; he did not dispute “Dr. Palmer's conclusion that the candidates preferred by [W]hite voters in the areas that he looked at regularly defeat the candidates preferred by Black voters,” id.; and both he and Dr. Palmer found evidence of a “substantive pattern” of racially polarized voting in District 7, id. at 1448.

This testimony was unsurprising after Dr. Hood found in his report that “racially polarized voting is present with Black voters overwhelmingly supporting the Democratic candidate and more than a majority of [W]hite voters casting a ballot for the Republican candidate.” Milligan Doc. 66-4 at 14.

At trial, although Dr. Hood did not again perform an ecological inference analysis, see Tr. 1882, he again conceded that Black voters in Alabama are politically cohesive, id. at 1901. And he testified that when he compared Black voters in Alabama to Black voters elsewhere, he found that Black support for Democrats in Alabama “is slightly higher than Black support for Democratic candidates across” the comparison states. Id. And Dr. Hood again testified that he does not dispute the conclusions of Dr. Liu and Dr. Palmer about racially polarized voting. Id. at 1898.

At trial, Dr. Bonneau also testified that he does not dispute Dr. Liu or Dr. Palmer's findings, id. at 1726, and he agreed that Dr. Palmer's ecological inference analysis established “that White voters in Alabama support White Democrats more than they support Black Democrats,” id. at 1789.

After trial, the State conceded in its proposed order that the Plaintiffs have established the second Gingles precondition. See Milligan Doc. 481 at 99 (conceding that “Plaintiffs’ evidence shows that Black Alabamians in the challenged areas are politically cohesive.”) And the State did not dispute in that proposed order the pattern of consistent (nearly invariant) electoral losses for Black-preferred candidates in Alabama. See generally id.

Accordingly, we see a clear consensus among all parties that Black voters in Alabama (and particularly in the districts at issue) are “politically cohesive,” and that each challenged district's “white majority votes sufficiently as a bloc to enable it ․ to defeat the [Black-]preferred candidate.” Allen, 599 U.S. at 18 (quoting Gingles, 478 U.S. at 51).

Because of this consensus, we decline to decide the validity of the State's assertion that when Dr. Liu conducted his ecological inference analysis, he should not have considered the race of candidates. See generally Milligan Doc. 481 at 104–06. The agreement of all parties on the critical substantive issue obviates the need for us to resolve methodological quibbles about how the ecological inference analysis would best be conducted. And in any event, (1) like Dr. Liu, Dr. Bonneau focused his analysis on the race of the candidate, not the race of the voter, see Tr. 1742–43, 1766, 1862, and (2) Dr. Palmer's analysis is not vulnerable to this criticism, id. at 1713–14 (Dr. Bonneau testifying that Dr. Palmer “analyze[d] both biracial and uni-racial elections”); see Caster Doc. 303-1 at 3–4 (Dr. Palmer describing an analysis focused on the race of the voter, not the candidates).

Before we turn to the State's legal arguments about these findings, we pause to explain our confidence in the ecological inference method. Dr. Liu opined that ecological inference “has been widely used as the most-advanced and reliable statistical procedure for [racially polarized voting] estimates in not only academic research but also voting rights cases in the last two decades.” Milligan Doc. 385-4 at 6–7. Dr. Palmer agreed. Tr. 491–92. And Dr. Bonneau conceded that ecological inference is “the preferred method by the [c]ourts” and “probably” the best method available. Id. 1859–60. Nevertheless, Dr. Bonneau also testified that Dr. Liu's exaltation of ecological inference as “one of the best methods in the history of political science” “is overstated significantly.” Id. at 1860. We do not suggest, let alone hold, that ecological inference is perfect. We simply hold that because we have (1) consensus use of that methodology by experts on both sides, (2) concessions about its reliability, and (3) multiple analyses that all support the same findings, we are satisfied that our conclusions are well-founded.

c. Arguments About Legally Significant Racially Polarized Voting

The State makes a novel legal argument that “[W]hite bloc voting in Alabama is not legally significant,” even if it is “statistically significant,” Milligan Doc. 481 at 99, 106, ¶ 275 (internal quotation marks omitted). Citing Pierce v. North Carolina State Board of Elections, 97 F.4th 194 (4th Cir. 2024), the State argues that “[d]ue to increased registration, turnout, and political participation among Black voters (and crossover voting by [W]hite voters) in Alabama, the historic need for majority-or even large-majority-Black districts in order to ensure Black voters an ‘opportunity to elect’ has substantially lowered.” Milligan Doc. 481 at 110, ¶ 288. The State contends that “with that change, the point at which [W]hite bloc voting becomes ‘legally significant’ has risen.” Id. The State asserts “that [W]hite bloc voting in the challenged areas is not ‘legally significant’ because there is enough [W]hite crossover voting to obviate the need for court-ordered majority-minority districts.” Id. at 112, ¶ 294. It argues that “[s]o long as additional majority-minority districts are not ‘necessary for Black-preferred candidates to win,’ legally significant [W]hite bloc voting is absent.” Id. at 113, ¶ 297 (quoting Pierce, 97 F.4th at 217).

These cases are fundamentally unlike Pierce because the challenged districts are not functional crossover districts (as the challenged district was in Pierce),⁵⁸ and the Plaintiffs have developed more than a dozen illustrative plans with majority-Black remedial districts. Further, although we understand that Section Two liability is foreclosed when a challenged district functions as a successful crossover district (which did not occur here), we do not see that Section Two liability is foreclosed just because a functional remedial district may be drawn with a Black voting age population slightly below a majority threshold (which did occur here, in the Special Master Plan).

As copious amounts of evidence make clear, it denies reality to call District 2 in Alabama's previous redistricting cycle, under the 2021 Plan, or under the 2023 Plan, a “crossover district.” Indeed, Dr. Palmer's functionality analysis of the 2023 Plan predicted that the Black-preferred candidate would lose sixteen out of seventeen modeled elections in District 2, win all seventeen elections in District 7, and lose every other election in every other district. Caster Doc. 303-1 at 6.

Likewise, we cannot describe District 2 in the Special Master Plan as a crossover district: Black Alabamians comprise 48.69% of the voting-age population in that district, and we have no evidence that Congressman Figures won it with significant support from White voters. Depending on turnout, Congressman Figures may have won it with no support from White voters.

In any event, we do not view the concept of crossover districts as particularly relevant in these cases. Alabama's patterns of racially polarized voting are about as stark as they come. And Black candidates’ losing streak in statewide elections and legislative elections in Alabama (outside majority-Black or very nearly majority-Black districts) is about as bad as it comes. Although the Special Master configured a functional remedial district with a BVAP of slightly less than 50%, we have no evidence that a functional remedial district may depend on a Black-preferred candidate receiving significant White support. In some jurisdictions, evidence may establish that statistically observable differences in Black and White voting patterns are of little practical or legal significance. Not in Alabama.

d. Arguments About Party Politics

Finally, we turn to the State's argument that patterns of racially polarized voting in Alabama are attributable more to political party affiliations than to race. The State relies on Dr. Bonneau, but as we just discussed, he conceded at trial that he does not dispute that “Black voters vote cohesively in Alabama,” nor that “[W]hite voters ordinarily vote as a bloc sufficient to defeat those Black voters’ choices.” Tr. 1862. Likewise, the State relies on Dr. Hood, who conceded at trial that Black voters in Alabama are politically cohesive and that he does not dispute Dr. Liu's or Dr. Palmer's conclusions about voting patterns. Tr. 1898, 1901.

These concessions are dispositive of the second and third Gingles preconditions. Under controlling precedent, those preconditions do not require that we fully disentangle party and race. They direct us to assess only whether Black voters in Alabama are “politically cohesive,” and whether each challenged district's “white majority votes sufficiently as a bloc to enable it ․ to defeat the [Black-]preferred candidate.” Allen, 599 U.S. at 18 (quoting Gingles, 478 U.S. at 51). We see those patterns clearly from the evidence, a consensus of experts agrees, and that concludes our Gingles analysis.

We consider causation in our analysis of the totality of the circumstances (particularly Senate Factor 2). See infra Part V.A.4. We understand that the State agrees with this approach. See Milligan Doc. 481 at 119 ¶ 317 (State's proposed order, explaining that Senate Factor 2 is not “redundant with the second and third Gingles preconditions” because “[t]here, the inquiry focused solely on ‘how’ Black and [W]hite voters voted. The focus ․ at the totality-of-circumstances stage, is on evidence of causation ․”) (quoting Ala. State Conf. of the NAACP, 612 F. Supp. 3d at 1291).⁵⁹

***

This record supports only one finding: that voting in Alabama, particularly in the districts at issue in these cases, is intensely and extremely racially polarized for purposes of the second and third Gingles preconditions. We cannot imagine a more comprehensive record, and we really cannot imagine clearer proof.

4. The Senate Factors

We begin our analysis of the totality of the circumstances aware that “it will be only the very unusual case in which the [P]laintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of § 2 under the totality of circumstances.” Ga. State Conf. of the NAACP, 775 F.3d at 1342 (quoting Jenkins, 4 F.3d at 1135). Consistent with this reality, we find that the Plaintiffs (again) have established that the totality of the circumstances weighs decisively in their favor.

We first make credibility determinations, we next analyze the Senate Factors, and we then consider the proportionality arguments that the Plaintiffs have raised.

a. Expert and Legislator Credibility Determinations

 Dr. Bagley

We credit much of Dr. Bagley's testimony. The parties do not dispute that Dr. Bagley's training and experience qualify him as an expert. See Tr. 1279. Dr. Bagley's credentials and familiarity with Alabama clearly qualify him to opine on Alabama-specific matters. See Milligan Doc. 385-1 at 34; Tr. 1277–78. His research and writing have focused on Alabama, and he has experience testifying as an expert in voting rights cases, including in Alabama. See Milligan Doc. 68-2 at 1; Tr. 1278.

At trial, Dr. Bagley walked back several overstatements in his report. See, e.g., Tr. 1376–77, 1408–10, 1393–96. These do not cause us to regard his testimony as unreliable or assign it little weight. In general, we found Dr. Bagley's opinions well-supported, and he was able to explain his bases for his conclusions. When he was confronted with an imprecise statement or overstatement, he responded candidly and fairly, rather than dogmatically. See, e.g., Tr. 1410. We find all the statements we rely on credible, reliable, and helpful to the Court.

 Dr. Burch

Likewise, we credit much of Dr. Burch's testimony. The parties do not dispute that her training and experience qualify her as an expert. Tr. 927. Dr. Burch's opinions and testimony were thorough, consistent, and generally well-supported with applicable social science literature and Alabama-specific data. Throughout her testimony, including cross-examination, she had no difficulty articulating the basis for her opinions. Although the parties dispute the inferences we should draw from her data, her data is not in dispute. See Tr. 2252 (Dr. Reilly). We do not adopt or make findings about all of Dr. Burch's testimony, but that is not because we found her testimony unreliable – it is simply because we need not accept all of it to make relevant findings and draw conclusions. We find all the statements we rely on credible, reliable, and helpful to the Court.

 Dr. Hood

Dr. Hood's training and experience clearly qualify him to testify as an expert. His extensive published scholarship focuses on electoral politics, racial politics, election administration, and Southern politics, Milligan Doc. 409-7 at 2; Tr. 1873–74, and he has qualified as an expert in multiple redistricting cases, including in Alabama. Tr. 1875; Milligan Doc. 409-7 at 2.

We credit some aspects of Dr. Hood's testimony, but we cannot credit his testimony (about Senate Factor 2) that voting patterns and election results in Alabama are driven by party more than by race. Dr. Hood testified that “[W]hite conservatives are more than willing to support minority Republican candidates,” Tr. 1892, and “the result of elections is impacted by ide[o]logical congruence rather than race of the candidate,” id. at 1895; Milligan Doc. 409-7 at 21–22. These findings (1) improperly draw broad conclusions from very limited, atypical data, and (2) are widely inconsistent with Dr. Hood's own scholarly work.

Dr. Hood's findings are based primarily on the election of Representative Paschal, a Black Republican, from a majority-White district. See Tr. 1893–95. But it is a gross understatement to say that this election is atypical – as Dr. Bonneau explained: it's a “unicorn.” Id. at 1688. Representative Paschal was the first Black Republican elected to the Legislature since Reconstruction, and he remains the only Black Republican in the Legislature -- a Legislature that numbers 105 state Representatives and 35 state Senators. Even Dr. Hood admitted that “you can't make a statewide generalization from a single state house election within Alabama.” Id. at 1921–24. And Representative Paschal's district in Shelby County does not overlap with the challenged districts in these cases. Id. at 1921. Accordingly, although we do not diminish the importance of Representative Paschal's election, it does not support a finding that voting in Alabama, particularly in the districts at issue, is more about party than race.

Separately, we cannot reconcile Dr. Hood's testimony with his published scholarship. At trial, Dr. Hood testified about three of his publications that either do not support or directly refute his litigation opinions. Dr. Hood conceded that his 2015 article about White support for minority Republican candidates (“True Colors”) did not consider any Alabama races and “make[s] no specific findings as to [W]hite voter support for Black Republican candidates.” Tr. 1912–15. And he conceded that the article concludes that “[a]t a minimum, the level of [ideological] polarization in American politics masks racially prejudiced voting behavior and, at a maximum, it renders it inoperable because White conservatives view recent minority Republican nominees as at least as conservative as White GOP nominees, and their level of support reflects this.” Tr. 1954–56 (reading from Milligan Doc. 456-2 at 2).

Of far greater concern, Dr. Hood testified that in his 2012 book (The Rational Southerner), he wrote that “the growth of [S]outhern Republicanism was primarily driven by racial dynamics”; “Southern politics in the early 21st Century still revolves around the issue of race”; “the partisan and political transformation of the [S]outh over the past half-century has, most centrally, revolved around the issue of race”; “the [S]outhern party system over the past half-century revolved around issues of race”; “[m]uch of the recent work on the American party system has clearly then underemphasized the crucial and distinctive role that race and racial dynamics have played”; and “race has left an indelible imprint on the region, and it would certainly be a mistake to ignore the potential future role of racial dynamics in [S]outhern politics and, by implication, national politics.” Tr. 1929–31 (reading from Caster Exhibits 157 and 159).

And most recently, Dr. Hood testified that in his 2022 article about the role the Civil War plays in today's Southern politics (“Switching Sides”), he wrote extensively about the primacy of race in Southern party politics:

Not only does an overwhelmingly [W]hite electorate now favor the GOP in [S]outhern politics, but in this article we have also shown with an inventory of public opinion data that the party's adherents have reached back in time to defend the Lost Cause Myth. Thus, in this regard, our findings support racially motivated explanations for partisan change in the South.

․

Hence, it stands to reason that contemporary debates over Confederate symbolism and public memory reflect ongoing conflict over racial inequality. Today's racialized partisan cleavage reflects a similar divide over views of a racially charged past.

․

It is true that the modern [S]outhern Republican Party stands for a host of things beyond being more racially conservative than its Democratic opponent. But it is also undeniable that the successful GOP strategy of attracting [S]outhern [W]hites by capturing the conservative position on African American civil rights has ultimately led to the reality that the Republican Party has now become the defender of the very flag that [W]hite [S]outherners once raised against the party of Lincoln on bloody battlefields and later in violent skirmishes over [B]lack equality. In addition, modern-day GOP adherents are also much more supportive of honoring the Confederate fallen, as we have shown with public opinion data on Confederate monuments. Finally, contemporary [S]outhern [W]hite Republicans are also the primary apologists for an almost universally disavowed historical argument that the “War Between the States” was mainly about states’ rights, as opposed to slavery. This development has come to fruition despite the fact that our data clearly show that [W]hite [S]outherners very much value the South's history and a large majority still think the Civil War remains relevant to American politics. Thus, the weight of the evidence shows that in “still fighting the Civil War,” [W]hite [S]outherners have rewritten history, at least with respect to switching partisan sides in their defense of the Lost Cause.

Caster Doc. 374-2 at 13–14 (internal citations omitted); Tr. 1939–41.

We are not alone. At trial, Dr. Hood could not reconcile his litigation opinions with these published works. Compare Tr. 1929–31, 1933–35 (reading excerpts from Caster Exhibits 157 and 159), Caster Doc. 374-2 at 13–14, and Tr. 1939–41, with Tr. 1947–53. On redirect, he simply reiterated his litigation opinions, with no explanation for the glaringly obvious contradiction between the published analysis and the testimony. Under this circumstance, we cannot credit his testimony that voting in Alabama is polarized by party rather than race.

 Dr. Bonneau

We credit Dr. Bonneau's testimony. All parties agree that his training and experience qualify him as an expert, his opinions were clear and consistent, and (unlike some of the State's other experts) he relied on Alabama-specific data. Tr. 1665, Milligan Doc. 384-1 at 2. We observed his demeanor as he testified, he was careful not to overstate his opinions, and he acknowledged that the small number of elections he studied limited them. Tr. 1819–20 (“you go to war with the data you got, not the data you want”). When confronted with an error in his report, he acknowledged it and testified candidly about its effects on his conclusions. See, e.g., id. at 1679–80. And when he used data only for a limited purpose, he explained it. See id. at 1816–18. Accordingly, we find his testimony reliable and helpful.

 Dr. Reilly

We assign very little weight to Dr. Reilly's testimony for three reasons. First, most of Dr. Reilly's opinions do not focus on and are not about Alabama. Dr. Reilly admitted at trial that his expertise and academic research are not focused on Alabama, and that in his report about racial socioeconomic gaps, he chose not to examine Alabama-specific data. Tr. 2211–12, 2213–14, 2249–50, 2255–56.

Second, Dr. Reilly repeatedly offered opinion testimony without support. We distinguish these opinions from overstatements because their underlying support was unreliable or completely absent. For at least four assertions in his report, he cited only websites (including Wikipedia, Quora, and Reddit), with no scholarship or peer-reviewed backup. Milligan Doc. 384-4 at 3 n.7; 8 n.17; 13 n.25; 20 n.41. Indeed, these websites lack the reliability of scholarly work because of the ability for any person to edit, add, or remove information without subject matter expertise or verification. Dr. Reilly conceded at trial that some data in his report lacks any citation. Milligan Doc. 384-4 at 29; Tr. 2306–07. When asked to explain certain figures in his report, he was unsure about the source. Tr. 2342. And he conceded that he reached conclusions on matters for which he conducted no analysis. Id. at 2251. Standing alone, Dr. Reilly's refusal to limit himself to well-founded opinions forecloses our reliance on his testimony.

Third, we observed Dr. Reilly's demeanor at trial, particularly when he was cross-examined, and found that it was dogmatic, defensive, and deliberately confrontational. His manner of testifying left us with the impression that his goal was to be evocative (an adjective he used to describe himself, id. at 2232) rather than reliable and persuasive.

Dr. Reilly testified about some of his social media posts, and that testimony confirms this impression. For example, he posted that “it is so silly to pretend IQ science does not exist ․ U.S. Blacks at 92 [and] Whites at 103 ․ [is] correct.” Id. at 2312. And that “[y]ou could literally pay smart Black people to have kids or boost Black merit immigration, to boost Black IQ, which, given what we do know about biracial scores, probably isn't low'ish [sic] for genetic reasons. This is really a very solvable problem.” Id. 2353–55.

For all these reasons, we do not find Dr. Reilly's methods or conclusions reliable or helpful to the Court.

 Dr. Carrington

We also assign no weight to Dr. Carrington's testimony. Dr. Carrington opined about “the historical development of party affiliations among Alabama voters from comprising the core of the Democratic ‘Solid South’ to becoming a dependably Republican-voting state.” Milligan Doc. 384-2 at 1. But he conceded that his education and training “did not have a particular focus on the American [S]outh,” he has never taught courses about Alabama politics or history, and he is not an expert in Alabama politics or history. Tr. 1549–50, 1583, 1585. He has published two articles relating to Alabama in the nineteenth century, but no other work about Alabama. Id. at 1584–85.

At the outset, Dr. Carrington's very limited familiarity with Alabama history and politics greatly reduced the potential value of his testimony in our “ ‘intensely local appraisal’ of the electoral mechanism[s]” in Alabama. Allen, 599 U.S. at 19 (quoting Gingles, 478 U.S. at 79). He exacerbated this limitation by making little to no effort to learn about Alabama before opining about party affiliations here.

Dr. Carrington admitted that he did not study any Alabama Democrat political party platforms for his report. Tr. 1608. He testified that Alabama voting patterns aligned with Southern patterns, but he did not study Alabama elections. Id. at 1608, 1628–31. He opined about “the sixth [Senate] factor, which confronts the question of whether or not[ ] political campaigns have been characterized by overt or subtle racial appeals,” Milligan Doc. 384-2 at 3 (internal quotation marks omitted), but conceded that he did not evaluate any Alabama campaign materials other than those Dr. Bagley identified, Tr. 1588.

Dr. Carrington put forth so little effort to learn about Alabama that he opined at length about former Governor Wallace while having no idea who Judge Frank Johnson was. He likewise opined about segregationist viewpoints and party affiliations, but with no knowledge of relevant prominent civil rights figures. See id. at 1622–23, 1645.

Dr. Carrington's willingness to opine about Alabama without first learning about Alabama extends beyond the courtroom. Before he was retained as an expert in these cases (he did not participate in the preliminary injunction proceedings), he authored an opinion piece calling the Supreme Court's ruling in these cases a “missed opportunity.” Id. at 1625–28. On cross-examination about the piece, he distinguished his work as an op-ed columnist from his scholarly work. Id. at 1626.

Dr. Carrington's lack of expertise and carelessness foreclose our reliance on his testimony.

 Dr. Riser

We credit Dr. Riser's testimony. The parties do not dispute that his training and experience qualify him as an expert, and his credentials and familiarity with Alabama qualify him to opine on Alabama-specific matters. See Singleton Doc. 253-1 at 3–5; Tr. 747, 750. His training focused on “U.S. Constitutional History and Southern History, and [he] completed both a master's thesis and doctoral dissertation that examined Alabama's 1901 Constitutional Convention.” Singleton Doc. 253-1 at 3. And he has testified as an expert in voting rights cases, including in Alabama. See id. at 4.

 Dr. Frederickson

We credit Dr. Frederickson's testimony. All agree that her training and experience qualify her as an expert, and her credentials and extensive familiarity with Alabama qualify her to opine on Alabama-specific matters. See Singleton Doc. 253-2 at 3; Tr. 804–05.

 The Legislators

We have reviewed the videotaped depositions of Representative Pringle and Senator Livingston, so we have limited exposure to their demeanor and manner of testifying. From this limited exposure, we found Representative Pringle's testimony generally direct and frank, and we found Senator Livingston's testimony less helpful. We credit their testimony as explained below.

In our first preliminary injunction, we found that Senate Factors 1, 2, 3, 5, 6, and 7 weighed in favor of the Plaintiffs, and we made no findings about Factors 8 and 9. Milligan Doc. 107 at 178–93. In our second preliminary injunction, we adopted the earlier findings about Factors 1, 2, 3, 5, 6, and 7, and found (based on evidence about the 2023 Plan) that Factor 8 weighed in favor of the Plaintiffs. See Milligan Doc. 272 at 178–84. We again made no findings about Factor 9. See id. at 184.⁶⁰

We now have the benefit of full discovery and a trial, and the parties have thoroughly litigated both our previous findings and presented new experts, evidence, and arguments. We thus have a much fuller record now than we had before. Nevertheless, our findings remain the same: as we explain below, Senate Factors 1, 2, 3, 5, 6, 7, and 8 weigh in favor of the Plaintiffs. We analyze Senate Factor 9 for the first time and find that it also weighs in favor of the Plaintiffs.

We begin our analysis of the Senate Factors with Factors 2 and 7, which Gingles suggests are the “most important.” 478 U.S. at 48 n.15.

b. Senate Factor 2

“[T]he extent to which voting in the elections of the state or political subdivision is racially polarized.” Gingles, 478 U.S. at 37.

We find that Senate Factor 2 weighs heavily in favor of the Plaintiffs. We already have found that voting in the challenged districts is starkly and intensely racially polarized, and that finding is based on substantial evidence, concessions, and the material agreement of the State's experts. See supra Part V.A.3. In its Senate Factor 2 argument, the State urges us to examine the cause of that pattern and find that it is attributable to party politics, not racial causes. See Milligan Doc. 481 at 119–58. The State draws on case law warning courts that patterns do not tell the whole story of how voters vote because “what appears to be bloc voting on account of race may, instead, be the result of political or personal affiliation of different racial groups with different candidates.” Solomon v. Liberty Cnty. Comm'rs, 221 F.3d 1218, 1225 (11th Cir. 2000) (en banc); see Milligan Doc. 481 at 119–120.

But when we look past the pattern in these cases, we see no evidence that only party politics are at work. We consider in turn each evidentiary basis the State offers. First, the State offers Dr. Hood's testimony to suggest that “[r]acial polarization in Alabama is a product of political partisanship, not racial bias.” See id. at 119, 125–26. But as we have already explained, see supra Parts IV.D.3.a & V.A.4.a, Dr. Hood's published scholarship tells the opposite story: that race remains the dominant political influence in Southern politics today, and that race heavily influences the positions that political parties take on racial issues. Indeed, Dr. Hood's published scholarship not only tells this story, but repeats it in multiple peer-reviewed publications, some of them co-authored, spanning nearly a decade.

Further, although Dr. Hood did not perform a racially polarized voting analysis for his trial report, he performed one for his earlier report in these cases, and he found racially polarized voting in Alabama. Tr. 1945; Milligan Doc. 66-4 at 14. At trial, Dr. Hood testified he still agreed with that finding because “nothing's changed in regard to” the analysis he performed. Tr. 1945.

Ultimately, Dr. Hood's opinions support the Plaintiffs more than the State on the issue of racially polarized voting. Standing alone, his published scholarship and independent finding of racially polarized voting are sufficient to support our finding that Senate Factor 2 weighs in favor of the Plaintiffs. Nevertheless, we consider the State's other evidence and arguments.

Second, the State offers the testimony of Dr. Bonneau that the evidence he reviewed is “consistent with the story that political party is the most important factor here and not race.” Id. at 1727; see Milligan Doc. 384-1 at 18. Dr. Bonneau examined certain elections and straight-ticket voting. See Milligan Doc. 384-1.

But we find Dr. Bonneau's evidence limited and the State's arguments from it overdrawn. See supra Part IV.D.3.b. When Dr. Bonneau opined that “the better explanation for the data we observe is political party,” he candidly acknowledged that his limited opinion was based on a “subset” of Alabama elections. Tr. 1766. Dr. Bonneau's selected subset included certain judicial elections in the state (which he has studied before, and which analysis contained a material error that reversed his conclusions, see supra Part IV.D.3.b), one round of state legislative elections in 2022 (with a focus on the election of Representative Paschal, who Dr. Bonneau described as a “unicorn”), the 2018 election of Judge Lewis to a circuit judgeship in Alabama state court (which was flagged for him by counsel), and the 2024 primary elections in the new District 2. See Milligan Doc. 384-1 at 4–13, Tr. 1688, 1766. Between limitations and flaws, we do not see that this subset has the potential to tell us very much about how to view the relative influence of race and party in modern Alabama elections.

We also see significant limitations on Dr. Bonneau's opinion about straight-ticket voting – that approximately two-thirds of Alabamians vote by “straight-ticket,” and “[t]he prevalence of straight-ticket voting means that most voters are voting for a political party, not a candidate.” Milligan Doc. 384-1 at 5; Tr. 1694–95. As Dr. Liu pointed out, “Dr. Bonneau does not explain whether he has any knowledge of these voters directly, nor the racial identities of these straight-ticket voters nor localities/precincts the voters resided in.” Milligan Doc. 385-8 at 4; Tr. 605–08. And even Dr. Bonneau acknowledged that he could not rule out that Black candidates were penalized at the polls on account of race. See Tr. 1783–85. Further, when Dr. Bonneau was asked why “Black voters overwhelmingly identify as Democrats,” he repeatedly agreed that Black voters perceive the Democratic Party as better on race-based issues such as the Voting Rights Act, Civil Rights Act, and civil rights. Id. at 1788. Ultimately, Dr. Bonneau's limited evidence simply does not support the State's assertion that it has “presented substantial evidence that a majority of [W]hite voters in Alabama vote for someone other than the minority-preferred candidate not for racial reasons, but for partisan and ideological ones.” Milligan Doc. 481 at 125.

Third, in connection with the State's reliance on Dr. Bonneau, the State relies on a recent case involving a Section Two challenge to Alabama's at-large process for electing appellate judges: Alabama State Conference of the NAACP v. Alabama, 612 F. Supp. 3d 1232 (M.D. Ala. 2020). That court found that Alabama is a “ruby red” state, which has made it “virtually impossible for Democrats – of any race – to win statewide in Alabama in the past two decades.” Id. at 1291.

But that finding was based on an evidentiary record – trial testimony from two expert witnesses, one of whom (Dr. Bonneau) conducted a multivariate regression statistical analysis – that is absent here. And read in context, that finding does not stand for the broad proposition that racially polarized voting in Alabama is always simply party politics; rather, it supports the more limited proposition that in that case, “the notion that African-American candidates lose solely because of their skin color [wa]s not supported by the evidence.” Id. at 1293.

Further, we are not looking at a record about two decades’ worth of racially polarized voting in some judicial elections – we see a near-total absence of Black Alabamians in statewide office and legislative office (outside of Black-opportunity districts) that dates all the way back to Reconstruction. Accordingly, we cannot reach the same conclusion that the Alabama State Conference of the NAACP court reached, and we cannot assign the weight to its conclusion that the State urges us to assign.

Fourth, the State repeatedly relies on the recent election of Representative Paschal from a majority-White Alabama House district. See, e.g., Milligan Doc. 481 at 134, 138. We do not diminish the inherent significance of Representative Paschal's unusual election, but one election of one Black Republican from one majority-White district in 150 years is hardly a sufficient basis for us to find that patterns of racially polarized voting are caused by party more than race. Dr. Hood and Dr. Bonneau cannot help but agree. Tr. 1924 (Dr. Hood), 1688 (Dr. Bonneau). Dr. Bonneau was right — Representative Paschal's election is a “unicorn.” Id. at 1688.

Representative Paschal's unicorn status tells us that the State may be substantially overstating White voters’ willingness to support minority candidates, and Dr. Liu provides further insight. Dr. Liu testified about two elections that may allow us to estimate White support for Black candidates in both political parties – the 2008 presidential election, and the 2024 Republican congressional primary in District 2. In the 2008 presidential election, Dr. Liu found that exit poll data showed that 51 percent of White Alabama Democrats supported Senator John McCain over then-Senator Barack Obama; from this, Dr. Liu reasoned that those “White Democrats showed in their vote choice that race mattered instead of party.” Tr. 588–90 (referring to Milligan Doc. 403-13 at 14). In the 2024 Republican primary in District 2, Dr. Liu explained that the four Black candidates finished behind the four White candidates, and the four Black candidates “together received only 6.2% of the total vote,” which suggests that White Republicans are not willing to support minority candidates in large numbers. Milligan Doc. 385-8 at 4.

We cannot reconcile the State's assertion that White voters are willing to support minority candidates in large numbers with the political reality we see. If the State were right about this, Representative Paschal would not be a unicorn, and four Black Republican candidates would not have amassed only 6% of the vote in a primary election in a Black-opportunity district.

Ultimately, Dr. Palmer's testimony tells us that there is a fictional premise at the heart of the State's argument about party polarization — namely, that race and party are separate and independent factors that influence vote choice. See Caster Doc. 303-2 at 6–7. Tr. 519–20. In truth, Dr. Palmer opined, race is a critical factor in how and why voters form partisan attachments:

Implicit in Dr. Bonneau's incorrect conclusion about the role of party is his assumption that the effects of race and party are separable. In other words, Dr. Bonneau assumes (without any evidence) that an individual's race and an individual's political party are two separate and independent factors that influence vote choice. A long literature in political science about how voters develop partisan attachments and make decisions about voting shows the opposite: an individual's background, including their race, is a key factor in their politics and party preferences. This means that even if members of a racial group strongly support candidates of a single party, race, as a key factor in driving their support for that party, is an inseparable part of their support for those candidates. If race causes party, then we can't find that party alone, without race, can cause vote choice. Due to the fundamental linkage of race and party, the effects of the two cannot be separated. In other words, the strong support of Democratic candidates by Black voters cannot be attributed to partisan preferences alone, but to a mix of personal and political factors and experiences of which race is an essential part.

Caster Doc. 303-2 at 6–7; see also Tr. 519–20 (explaining why “we can't just isolate party from race alone because they're fundamentally linked”).

We credit Dr. Palmer's analysis on this issue for four reasons. First, no one disputes it. Indeed, Dr. Palmer's testimony that race drives party attachments fits better with Dr. Hood's published scholarship about how race remains the primary driver of party politics in the South than does Dr. Hood's litigation opinion that ideological and policy preferences drive Black voting patterns.

Second, Dr. Palmer's testimony that race drives party attachments is realistic. It allows us to hold in the same space the obvious truths that partisan affiliations drive voting patterns and issues of race drive Black voters’ choices at the polls. We have an overwhelming evidentiary record about the importance of race in Alabama politics, both historically and today. Again, it denies reality for us to say that at the end of the day, all of that is just party politics.

Third, Dr. Palmer's testimony fits with the lay testimony we heard from multiple Black voters. For example, Dr. Caster testified that racial concerns drive his affiliation with the Democratic Party: “The principles of the Democratic party as far as they help and support ․ the African-American community and Blacks align more with [his] core values,” he explained. Tr. 383. “And so that's why [he] ran with the Democratic party.” Id. Similarly, Mr. Love explained that racial concerns drive his affiliation with the Democratic Party. See Milligan Doc. 459-14 at 19.

And as other witnesses explained, the position of the Democratic Party on both racial issues and other issues that are important to Black Alabamians overrides the obvious alignment between these voters’ conservative Christian beliefs and the Republican Party. See, e.g., Tr. 1106–11 (Pastor Jackson); Tr. 1174 (Mr. Milligan).

We see no reason to think that these Black voters are unusual. If they were, or if their explanations were flawed, we would expect Black Alabamians to cast their votes for Republican candidates – particularly conservative Christian Republican candidates – far more than the evidence tells us that they do.

Fourth, acknowledging that race plays a central role in party attachments keeps the controlling legal standard honest and workable. It would be deeply contradictory for that standard to demand political cohesion in a minority group for the second and third Gingles preconditions, then deny Section Two relief based on that same cohesion because party politics tilt Factor 2 against the minority group.

Put differently, our analysis is not confounded by partisanship based on race. As we understand it, Gingles accounts for partisanship based on race in its demand for political cohesion among the minority group, which will be absent in times or places where party affiliations are driven primarily by something other than race.

We understand the statutory command about the totality of the circumstances as an instruction to look at the whole picture, not as permission (let alone a requirement) to carve it up into parts and examine each part in isolation from the others. When we consider the whole picture, we cannot understand the patterns we see as mere party politics. We acknowledge the well-known reality that party affiliations drive voting patterns, but we understand this evidentiary record as telling us that we cannot separate voters’ racial considerations from their party affiliations, and that we must not ignore the powerful role that voters’ race plays in their partisan attachments. Accordingly, we find that when we look at racial cleavages in voting patterns in Alabama, what we are seeing is appropriately described as racially polarized voting, and Senate Factor 2 weighs heavily in favor of the Plaintiffs.

c. Senate Factor 7

“[T]he extent to which members of the minority group have been elected to public office in the jurisdiction.” Gingles, 478 U.S. at 37.

Likewise, we find that Senate Factor 7 weighs heavily in favor of the Plaintiffs. Four stipulated or undisputed facts do the heavy lifting here:

(1) Since 1994, no Black Alabamian, regardless of party, has won a statewide race, see Milligan Doc. 403-1 at 29 61;

(2) in 1992, Representative Earl Hilliard was the first Black Alabamian elected to Congress since Reconstruction, Milligan Doc. 436 ¶ 103;

(3) Representative Shomari Figures is the first Black Alabamian to be elected to Congress outside of District 7 since Reconstruction, see Milligan Doc. 436 ¶¶ 103, 106, 108, 113–14, 151; and

(4) “Thirty-two (32) out of thirty-three (33) Black Alabamians currently serving in the Alabama Legislature were elected from majority-Black districts,” Milligan Doc. 436 ¶ 155, which were created to comply with the Voting Rights Act or the Constitution, Tr. 1927–28 (Dr. Hood acknowledging that those majority-minority districts were the product of and maintained by the Voting Rights Act and associated litigation).

The State is given “no pause” by these facts “because every one of those 32 legislators ran as a Democrat,” Milligan Doc. 481 at 159–60, so the State directs us back to arguments about party affiliations that we already have rejected, see supra Part V.A.4.b. But the State does not (because it cannot) rebut the reality that Black Alabamians enjoy zero success in statewide elections, and near-zero success in legislative elections outside of Black-opportunity districts protected by federal law. Indeed, it was the State's own expert who described the one-time election of a Black Republican from a majority-White state legislative district as a “unicorn” because “you have a Black Republican winning an election in Alabama.” Tr. 1688.

To be sure, Black Alabamians have made progress in electoral success. Dr. Hood reported that there were no Black senators or representatives in the Legislature in 1965, three Black senators and thirteen Black representatives in 1981, and there are currently seven Black senators and twenty-six Black representatives. Milligan Doc. 384-3 at 22; see Tr. 1896.

But just as we refused to evaluate Black voters’ partisan affiliations in a vacuum, likewise we refuse to evaluate their electoral gains in a vacuum. Every gain in congressional elections has come as a result of federal law (primarily Section Two), and even Dr. Hood acknowledges that the reality is much the same for the gains in state legislative elections. Senate Factor 7 weighs decidedly in favor of the Plaintiffs.

d. Senate Factors 1, 3, and 5

Senate Factor 1: “[T]he extent of any history of official discrimination in the state ․ that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process.” Gingles, 478 U.S. at 36–37.

Senate Factor 3: “[T]he extent to which the state ․ has used ․ voting practices or procedures that may enhance the opportunity for discrimination against the minority group.” Id. at 37.

Senate Factor 5: “[T]he extent to which members of the minority group in the state ․ bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process.” Id.

We analyze these Senate Factors together because much of the evidence that is probative of one of them is probative of more than one of them. Alabama's terrible history of racial and voting-related discrimination is undeniable and well documented. The State argues that Alabama has come a long way, but the question for us is more pointed: has it come far enough for these factors to be neutral or to weigh in favor of the State?

We are keenly aware of the instruction that “[p]ast discrimination cannot, in the manner of original sin, condemn governmental action that is not itself unlawful.” Abbott, 585 U.S. at 603 (quoting City of Mobile, 446 U.S. at 74). It should be apparent that we do not assign Alabama's shameful history dispositive weight, and we do not grant Section Two relief simply because we condemn past discrimination. It would not take us five hundred pages to explain that logic if we adopted it. We have carefully considered an extensive record about both past and present discrimination (some of it in these very cases), and a wealth of expert analysis of recent data about Black Alabamians’ lives and voting patterns, along with other evidence.

All the evidence about Senate Factors 1, 3, and 5 tells the same story: official discrimination on the basis of race has affected Black Alabamians’ lives and political participation for a long time, and it continues to affect Black Alabamians’ lives and political participation today. We first discuss findings we made in our first preliminary injunction, we then consider the lay testimony that offered firsthand recollections about official discrimination, we then consider the expert testimony about socioeconomic disparities and their impact on political participation, and we finally consider the lay testimony about the same.

 Findings from the Preliminary Injunction

In our first preliminary injunction, we made findings about Alabama's history of official discrimination based on stipulated facts and judicial findings. See Milligan Doc. 107 at 182–88. The parties revised their stipulations, but nothing has changed about the judicial precedents. So, we again find that:

• Prior to 1960, the Legislature failed to reapportion for 50 years. As a result, Alabama's entire legislative apportionment scheme was struck down for violating the principle of one person, one vote. Reynolds, 377 U.S. at 568. On remand, a three-judge court found that, in devising remedial maps to correct the malapportionment, the “Legislature intentionally aggregated predominantly Negro counties with predominantly white counties for the sole purpose of preventing the election of Negroes to [State] House membership.” Sims, 247 F. Supp. at 109.

• Following Reynolds and the 1970 Census, the Legislature again failed to redistrict and a three-judge federal court was forced to draw new district lines. Sims, 336 F. Supp. at 940. The court rejected the Alabama Secretary of State's proposed map because of its racially “discriminatory effect” on Black voters. Id. at 936.

• In the 1980s, the United States Attorney General denied preclearance under the Voting Rights Act to maps drawn by the Legislature to redistrict State House and Senate maps because of their discriminatory effect on Black voters in Jefferson County and the Black Belt. Letter from Wm. Bradford Reynolds, Assistant Att'y Gen., C.R. Div., U.S. Dep't of Just., Hon. Charles A. Graddick, Ala. Att'y Gen. (May 6, 1982), https://www.justice.gov/sites/default/files/crt/legacy/2014/05/30/AL-1520.pdf. Shortly thereafter, a three-judge court rejected Alabama's proposed interim remedial state maps in part because Alabama's maps “had the effect of reducing the number of ‘safe’ [B]lack districts” in and near Jefferson County. Burton v. Hobbie, 543 F. Supp. 235, 237 (M.D. Ala. 1982).

• After the 1990 census, the State entered a consent decree to resolve a Voting Rights Act lawsuit filed on behalf of Black voters. See Brooks, 631 So. 2d at 884.

• In 1986, a federal court found that the state laws requiring numbered posts for nearly every at-large voting system in Alabama had been intentionally enacted to dilute Black voting strength, and that numbered posts had the effect of diluting Black voting strength in at-large elections. Dillard v. Crenshaw County, 640 F. Supp. 1347, 1357 (M.D. Ala. 1986). The court also found that from the late 1800s to the 1980s, Alabama had purposefully manipulated the method of electing local governments as needed to prevent Black citizens from electing their preferred candidates. See id.

• Federal courts recently ruled against or altered local at-large voting systems with numbered posts created by the Legislature to address their alleged racially discriminatory purpose or effect. See, e.g., Jones v. Jefferson Cnty. Bd. of Educ., No. 19-CV-01821, 2019 WL 7500528, at *2, *4, 2019 U.S. Dist. LEXIS 223556, at *9 (N.D. Ala. Dec. 16, 2019); Ala. State Conf. of the NAACP v. City of Pleasant Grove, No. 18-cv-02056, 2019 WL 5172371, at *1, 2019 U.S. Dist. LEXIS 179206 (N.D. Ala. Oct. 11, 2019).

• The Supreme Court struck down Alabama's discriminatory misdemeanant disfranchisement law, Hunter v. Underwood, 471 U.S. 222, 225 (1985), and a state law permitting certain discriminatory annexations, City of Pleasant Grove v. United States, 479 U.S. 462, 466–67, 472 (1987).

• Since the decision in Shelby County v. Holder, federal courts have ordered more than one political subdivision in Alabama to be bailed back into preclearance review under Section 3(c) of the Voting Rights Act. See Jones, 2019 WL 7500528, at *4–5, 2019 U.S. Dist. LEXIS 223556, at *12; Allen v. City of Evergreen, No. 13-0107, 2014 WL 12607819, at *2, 2014 U.S. Dist. LEXIS 191739, at *3–4 (S.D. Ala. Jan. 13, 2014).

• In 2018, in a case challenging the attempt by the City of Gardendale, which is 85% White, to form a school district separate from Jefferson County's more racially diverse district, the Eleventh Circuit affirmed a finding that “race was a motivating factor” in the city's effort. Stout ex rel. Stout v. Jefferson Cnty. Bd. of Educ., 882 F.3d 988, 1000, 1009 (11th Cir. 2018).

• Alabama was subjected to a statewide injunction prohibiting the state from failing to disestablish its racially dual school system. Lee v. Macon Cnty. Bd. of Educ., 267 F. Supp. 458, 480 (M.D. Ala. 1967) (per curiam), aff'd sub nom. Wallace v. United States, 389 U.S. 215 (1967). The order resulted from the court's finding that the State Board of Education, through Governor George Wallace, had previously wielded its powers to maintain segregation across the state. Id. at 462. A trial court found that for decades, state officials ignored their duties under the statewide desegregation order. See Lee v. Lee Cnty. Bd. of Educ., 963 F. Supp. 1122, 1128–30 (M.D. Ala. 1997). A court also found that the state did not satisfy its obligations to remedy the vestiges of segregation under this order until as late as 2007. Lee v. Lee Cnty. Bd. of Educ., 476 F. Supp. 2d 1356, 1367–68 (M.D. Ala. 2007).

• In 1991, a trial court in Knight v. Alabama, 787 F. Supp. 1030 (N.D. Ala. 1991), found that Alabama had failed to eliminate the lingering and continued effects of segregation and discrimination in the University of Alabama and Auburn University, and at the state's public Historically Black Colleges and Universities. See id. at 1377–78. In 1995, the trial court issued a remedial decree analogous to the statewide injunction issued in Lee v. Macon, and the court oversaw implementation of that order for over a decade. Knight v. Alabama, 900 F. Supp. 272, 349–73 (N.D. Ala. 1995). Alabama did not satisfy its obligations under that order until 2006. Knight v. Alabama, 469 F. Supp. 2d 1016, 1039 (N.D. Ala. 2006).

• After the 2010 census, Black voters and legislators successfully challenged twelve state legislative districts as unconstitutional racial gerrymanders. See Ala. Legis. Black Caucus, 231 F. Supp. 3d at 1348–49.

• In United States v. McGregor, 824 F. Supp. 2d 1339, 1345–47 (M.D. Ala. 2011), a federal court found that Alabama State Senators conspired to depress Black voter turnout by keeping a referendum issue popular among Black voters (whom the Senators called “Aborigines”) off the ballot.

These judicial precedents illuminate a pervasive and protracted history of official discrimination in voting rights in Alabama. This history spans numerous electoral contexts, census cycles, and jurisdictions. In multiple cases it has run well into the present era: several of the decisions recited above were issued in the last ten years by federal judges who remain in service today. Against that backdrop, we turn to the evidentiary record in the cases before us.

 Lay Testimony About Firsthand Experiences of Official Discrimination

We heard at trial extensive and compelling testimony from Black Alabamians who personally experienced official discrimination, including several who attended segregated public schools. They described their experiences in detail:

• Robert Clopton, a Black Alabamian born in 1954, testified about his family's experiences as sharecroppers, his understanding from childhood that he should not go to sundown areas because “people were beaten [and] hung,” and his experiences in dilapidated segregated public schools, Tr. 236–38, 240–43.

• Letetia Jackson, a Black Alabamian born around the same time as Mr. Clopton, testified about her experiences attending segregated public schools and living with official segregation in public spaces. Id. at 684–92.

• Ronald Smith, a Black Alabamian born in 1954, testified about his experience attending segregated public schools and living with official segregation in public spaces. Id. at 416, 418, 428–31. He is an articulate person who testified that “words can[not] adequately describe” the feeling “[w]hen you live on one side of the railroad track and you see your White counterparts enjoying some of the amenities that are government sponsored and it's taboo for you.” Id. at 428.

• Janice Malone, a Black Alabamian born in 1955, testified about her experience attending segregated public schools and living with official segregation in public spaces. Id. at 1130–35.

• Valtoria Jackson, a Black Alabamian born in 1961, testified about her memories of her mother paying the poll tax to vote. Id. at 1069–70.

• Bobby Lee DuBose, a Black Alabamian born in 1963, testified about growing up on a plantation. Milligan Doc. 459-5 at 7, 10, 29.

• Benjamin Jones, a Black Alabamian born in 1965, testified about his childhood memories of his parents going to jail “on a number of occasions for voting,” and their strategy of not going to civil rights marches together because one of them had to be reliably out of jail to parent him and his fifteen siblings. Milligan Doc. 459-8 at 6; Jan. 10, 2022 Tr. 1345.

• Evan Milligan, a Black Alabamian who was forty-three years old at the time of trial, testified about witnessing demonstrations by the Ku Klux Klan while growing up in Montgomery. Tr. 1159, 1238.

The State does not dispute these firsthand recollections. Instead, the State asserts that this evidence cuts in its favor – that the Plaintiffs “did not present a lay witness whose political participation was hampered by past discrimination” because the witnesses who are “old enough to have attended segregated schools ․ are all extremely politically active.” Milligan Doc. 481 at 184.

We emphatically reject this assertion. We do not see political activism as evidence that these witnesses were not adversely affected by the official discrimination they experienced. We see that they are politically active despite that discrimination and because they experienced its harmful effects. Additionally, we refuse to give punitive effect to the political participation of Black Alabamians who have personally suffered the ill effects of official discrimination and responded with civic engagement in the democracy that discriminated against them.

 Expert Testimony

We also have the benefit of extensive expert testimony about these Senate Factors – from Dr. Bagley and Dr. Burch for the Plaintiffs, and Dr. Reilly and Dr. Carrington for the State. As an initial matter, we repeat our findings that both Dr. Bagley and Dr. Burch are credible experts (even though we do not adopt or rely on every aspect of their testimony), and that we assign less or no weight to the testimony of Dr. Reilly and Dr. Carrington. See supra Part V.A.4.a.

Dr. Bagley opined at length about Alabama's history of official discrimination, particularly with respect to voting rights and redistricting. See Milligan Docs. 68-2, 385-1; Tr. at 1282–92, 1297–98, 1307–24, 1376–77. We already made findings about that history based on extensive judicial precedents, and we regard those precedents as generally sufficient to establish the history.

But Dr. Bagley did tell us two additional details about the history that we find helpful to illuminate its scope and recency: that (1) Alabama is the only state to have more than one jurisdiction bailed back into federal preclearance requirements since Shelby County, id. at 1288; and (2) school desegregation litigation in three major school districts (Jefferson County, Huntsville City, and Madison County) remains ongoing in federal courts to this day, id. at 1291–92.

Dr. Bagley and Dr. Burch both opined about pressing socioeconomic disparities between Black Alabamians and White Alabamians on numerous dimensions: education, economics, housing, and health, among others. See Milligan Docs. 68-2, 385-2. We again find that many of these disparities are substantial and undeniable. See Milligan Doc. 107 at 185–87.

As one example, Dr. Bagley opined that “Black communities in the Black Belt continue to struggle in primitive conditions and suffer unusual health difficulties and lack of even the most basic services.” Milligan Doc. 68-2 at 21. He described the 2019 United Nations report that found that extreme poverty conditions in the Black Belt were “very uncommon in the First World,” reported that Black residents “lacked proper sewage and drinking water systems and had unreliable electricity,” and described instances in which whole households fell ill with infections contracted from drinking water contaminated with raw sewage. See id. As another example, Dr. Bagley reported that Black Alabamians are less likely to have access to a vehicle than are White Alabamians. Id. at 17. And as another example, he reported that for 2020–21, “the bottom 6 percent of the state's schools,” labeled as “failing” under Alabama law, “were majority Black, most overwhelmingly so” and “in or around Birmingham, Montgomery, and Mobile, or in the Black Belt.” Id. at 24–25.

Dr. Burch also identified substantial disparities from a systematic, statistical perspective. She testified that the unemployment rate for Black workers in Alabama (10%) is more than twice that of White workers (4%), Milligan Doc. 385-2 at 21, 24 tbl. 5; the family poverty rate for Black Alabamians (24%) is more than triple the rate for White Alabamians (7%), id. at 21, 25 tbl. 6; the infant mortality rate for Black infants in Alabama (12.4 deaths per 1,000 births) is nearly three times higher than the rate for White infants in Alabama (4.3 deaths per 1,000 births), id. at 30; Black Alabama households (12%) are more than twice as likely to lack access to a vehicle at home than White Alabama households (4%), id. at 21, 29 tbl. 10; and the percentage of Black households in Alabama without internet access at home (26%) is nearly double the percentage of White households in Alabama without access (14%), id. at 21, 28 tbl. 9; see also Tr. 928, 943, 946, 957–67.

We assign particularly substantial weight to Dr. Burch's explanation that many of these disparities are pronounced in the Black Belt. As to poverty, she testified that in Greene County “40 percent of Black families ․ live below the poverty lines compared with five percent of White families,” and that the situation is similar in Perry County. Tr. 957. As to communication infrastructure, she testified that in Escambia County, “43 [percent] of Black households don't have access to the Internet at home compared with 26 percent of White households,” and the situation is similar in Crenshaw County. Id. at 960–61. As to transportation access, she testified that in Hale County, “16 percent of Black families don't have access to a car at home compared with three percent of White families,” and the situation is similar in Dallas County. Id. at 962.

We also assign particularly substantial weight to Dr. Burch's testimony that Black Alabamians have significantly lower educational attainment than White Alabamians. She reported that “[s]tatewide and at the county level, Black adult Alabamians were less likely to have graduated from high school or to have attained a bachelor's degree than White Alabama adults.” Milligan Doc. 385-2 at 4. She also reported recent data establishing stark disparities among school-aged children: “[I]n 2022, only 9 percent of Black Alabama 8th Graders were proficient in reading, compared with 30% of White Alabama 8th Graders. Likewise, only 7% of Black Alabama 8th Graders were proficient in Math, compared with 27% of White Alabama 8th Graders.” Id. at 12. And she testified about the effects of such disparities, explaining that “in the Black Belt especially, there are ․ disproportionately high illiteracy rates, as high as 30 percent.” Tr. 938, 998, 1049.

Dr. Bagley and Dr. Burch both opined that these disparities are inseparable from (and in large part the result of) the state's history of official discrimination. See, e.g., Milligan Docs. 68-2 at 17, 385-2 at 12; Tr. 934–35, 1396–98. Dr. Bagley explained that from a historian's perspective, there is “no other explanation for this — these kinds of widespread myriad disparities other than the history of discrimination.” Tr. 1397. And Dr. Burch explained that Black Alabamians’ lower educational attainment in particular is “caused, in part, by historical and contemporary discrimination in elementary, secondary, and higher education,” including “separate-but-unequal” education. Milligan Doc. 385-2 at 12; see Tr. 934–35. And Dr. Burch linked educational attainment with “income, poverty, and employment,” meaning that Black Alabamians’ lower educational attainment in turn drives other socioeconomic disparities. Milligan Doc. 385-2 at 19–20.

We credit these explanations: it seems near-obvious to us that Black Alabamians’ lower educational attainment and higher rates of illiteracy are directly traceable to segregated public schools and dilapidated schools in predominantly Black areas. Likewise, it seems near-obvious to us that communities with lower educational attainment are at greater risk for widespread unemployment and poverty than communities with higher educational attainment.

Dr. Bagley and Dr. Burch also opined that many of these disparities hinder Black Alabamians’ opportunity to participate in the political process. See, e.g., Milligan Docs. 68-2 at 17, 385-2 at 19–20; Tr. 929, 936–38, 950, 952, 956–57, 960–62, 969, 972. Dr. Bagley explained (1) that because White Alabamians tend to have “more education and therefore higher income” than Black Alabamians, they tend to be better able than Black Alabamians to “afford a car, internet service, a personal computer, or a smart phone; ․ take time off from work; ․ afford to contribute to political campaigns; ․ afford to run for office; ․ [and to] have access to better healthcare,” and (2) that “[e]ducation has repeatedly been found to correlate with income [and] independently affects citizens’ ability to engage politically.” Milligan Doc. 68-2 at 17.

Dr. Burch relied on a well-established scholarly consensus linking critical disparities to political participation. She testified that “socioeconomic variables have consistently been related to political participation and voting participation throughout the political science literature”; that “educational attainment has been shown over and over again by political scientists to be the most important predictor of voting”; and that “the relationship between education and voting isn't just associational; it's causal.” Tr. 929; Milligan Doc. 385-2 at 11.

Dr. Burch specifically explained how Alabama's history of segregated public schools still impacts voting participation today: “[I]n 2020 ․ 38.6 percent of votes in the Alabama general election were cast by people age 60 and older. So those were people who were at least school age in 1970 when Alabama still maintained those separate and unequal schools for Black and White students.” Tr. 936. And she testified that segregation resulted in fewer opportunities for Black people to attend college and access educational resources, which is why Black people today “are disproportionately concentrated in these lower educational attainment -- lower voter turnout groups.” Id. at 937–38.

Dr. Burch further explained that lower educational attainment impacts other socioeconomic factors that also affect voting rates for Black Alabamians. See id. at 938; Milligan Doc. 385-2 at 19–20. She explained how racial disparities in family poverty, internet access, and access to transportation hamper voting participation due to an inability to read ballots, learn about candidates, absentee vote, locate voting information, and travel to polls. See Tr. 938, 950, 952, 956–57, 960–62.

We credit this testimony, which is not disputed, and we say again that it explains dynamics that strike us as near-obvious. That said, we do not make findings about all of Dr. Bagley's testimony, nor all of Dr. Burch's. We make only the findings that we have just described. We do not make findings about every instance of alleged official discrimination that was discussed in expert reports or at trial, nor every disparity that was discussed. For example, we make no findings about racial disparities in interactions with the criminal justice system. Further, we make no findings about the idea that Dr. Burch testified about, sometimes labeled as “structural racism,” that attributes most or all socioeconomic disparities or other differences in the lives of Black Alabamians and White Alabamians to discrimination. And we make no findings about whether or how Black Alabamians and White Alabamians worry differently about the price of eggs. We make only those findings necessary to reach a conclusion about these Senate Factors, and no more.

In that regard, we say simply that when we consider critical disparities – disparities in access to decent, integrated public schools and basic infrastructure in the modern world (water, sewage, electricity, communication, and transportation), we see stark racial disparities, particularly in the rural Black Belt, that (1) are clearly traceable to Alabama's lengthy and terrible history of official discrimination, and (2) unsurprisingly hinder Black Alabamians’ political participation.

 Lay Testimony About Disparities and Political Participation

Several fact witnesses corroborated the experts’ studies. Thematically, they testified that educational disparities, primitive conditions and extreme poverty, and lack of internet, transportation, and healthcare access make political participation difficult and unlikely for many Black Alabamians.

Pastor Jackson testified that economic destitution in the rural Black Belt, persistent effective segregation in public schools, lack of internet access, illiteracy, and living in economic survival mode are realities in the Black community that hinder political participation. Id. at 1091–99, 1101, 1126. Ms. Letetia Jackson testified about the lack of broadband and cell service in the Black Belt and explained that voters cannot “look up [their] polling place,” “determine whether [they are] actually still registered to vote,” determine whether they “were being purged off of the voter list,” or “download a voter registration ․ form,” Id. at 698, 715–16, 727–28. And Mr. Smith testified that “there is no equivalence between Black and White education in Bullock County and throughout the Black Belt,” and that “Black children face obstacles getting into college” because of “bottom-of-the-barrel instructors.” Id. at 435–37. He also testified about lack of access to broadband, health insurance, and healthcare in his community, and he too explained that these circumstances present real problems for political participation. See id. at 442–44. We see clearly from this testimony, as we did the expert testimony, how critical racial disparities with roots in official discrimination hinder Black Alabamians’ political participation.

Accordingly, based on the great weight of the evidence, we reject the State's accusation that Plaintiffs “simply assume” that Black Alabamians’ disparate socioeconomic status hinders their political participation, Milligan Doc. 481 at 11, and its assertion that racial parity in rates of voter registration and turnout means that Plaintiffs cannot demonstrate depressed political participation, id. at 173 ¶ 463; 178 ¶ 475; 187 ¶ 500; 223 ¶ 605. We regard those arguments as both unsupported and too formulaic. They are unsupported because Dr. Bagley and Dr. Burch provided evidence and analysis for their explanations, not mere assumptions. And they are too formulaic because the point of Factor 5 is for us to consider whether the lasting effects of official discrimination “hinder” the ability of Black Alabamians to participate in the political process, Gingles, 478 U.S. at 37, and a laser focus on parity in registration and turnout rates would overlook (1) other aspects of political participation, and (2) the question of whether the lasting effects of discrimination make it harder for Black Alabamians to participate at the levels that they do, even if those levels are nearly on par or on par with the levels of White participation.

We also reject the State's argument, based on the testimony of Dr. Hood and Dr. Reilly, that these kinds of racial disparities are everywhere in the United States, including places with “very different histories,” such that if they are assigned substantial weight, they will invariably drive a finding that the totality of the circumstances supports a Section Two plaintiff. Milligan Doc. 481 at 11–12. This is for two reasons. First, the State's assertion is overwrought: we do not consider socioeconomic disparities, nor their causes or effects, nor any other Senate Factor, in a vacuum. And we do not grant Section Two relief simply because Black Alabamians are worse off than White Alabamians on various metrics – we have analyzed substantial other evidence.

Federal law makes crystal clear that this is the way, as it has for forty years, so we harbor no concern that any other federal court will grant Section Two relief simply because of socioeconomic disparities across races. For example, when racially polarized voting is absent, socioeconomic disparities alone will not support Section Two relief. Likewise, when a reasonably configured remedial district cannot be drawn because the minority population is too geographically dispersed, socioeconomic disparities alone will not support Section Two relief.

Second, the State's assertion is too narrowly focused. We must do more than simply crunch numbers to analyze these Senate Factors properly. The bare fact of a statistical disparity is important, but insufficient, to generate a clear understanding of the presence or absence of the Factors.

The other intensely local information that we have considered tells us that the statistical disparities in Alabama (and their causes and effects) are not like everywhere else, even if some of Alabama's statistics might be similar in some ways to statistics from other places. As just a few examples: the conditions in Alabama's Black Belt are sufficiently unique to startle seasoned, leading national scholars; school desegregation cases are still pending in federal courts in Alabama; and Alabama is the only state to have multiple jurisdictions bailed back into federal preclearance in the last ten years. To repeat what we have already said in another context (about crossover districts): things are different here.

e. Senate Factor 6

Senate Factor 6: “[W]hether political campaigns have been characterized by overt or subtle racial appeals.” Gingles, 478 U.S. at 37.

We find that Senate Factor 6 weighs in favor of the Plaintiffs, but to a lesser degree than do Senate Factors 2, 7, 1, 3, and 5 (and 8 and 9, as we next explain). Dr. Bagley offered several examples of racial campaign appeals in his reports, see Milligan Docs. 68-2 at 26–28, 385-1 at 31–32, some of which he testified about at trial. We do not need to decide whether every example reflected a racial appeal, but at least three of them did, and all three were in recent congressional elections.

First, when a former Chief Justice of the Alabama Supreme Court, Roy Moore, ran for Senate in 2017, he won the Republican Party nomination. In 2011, the year before he was elected to the Alabama Supreme Court, he said during a radio interview that the amendments to the Constitution that follow the Tenth Amendment (including the Thirteenth Amendment, which abolished slavery, the Fourteenth Amendment, which requires States to provide equal protection under the law to all persons, and the Fifteenth Amendment, which provides that the right to vote shall not be denied or abridged on the basis of color or previous enslavement) have “completely tried to wreck the form of government that our forefathers intended.” See Milligan Doc. 68-2 at 27.

Later, during his 2017 Senate campaign, Mr. Moore acclaimed the antebellum period in the South: “I think it was great at the time when families were united – even though we had slavery. They cared for one another. People were strong in the families. Our families were strong. Our country had a direction.” See id.

Second, former Congressman Mo Brooks, a White Republican who represented District 5 and ran for the open Senate seat that Senator Katie Britt now occupies, has repeatedly claimed that Democrats are waging a “war on [W]hites.” See Milligan Doc. 68-2 at 27–28 & n.94.

Although the State suggests that the Plaintiffs have misunderstood other campaign ads that they claim are racial appeals, the State does not contest these two examples, which we find are obvious and overt appeals to race.

Third, we considered the campaign ad from former Congressman Bradley Byrne, a White Republican who represented District 1, that Dr. Bagley considered in in his opinion about Senate Factor 6. Id. at 28; Milligan Doc. 107 at 189–91.⁶² Even if Mr. Byrne did not intend his campfire commercial to be a racial appeal (a question that we need not and do not decide), a reasonable viewer might have perceived it as one. We do not disagree with the Plaintiffs that the video of a White man narrating as images of prominent persons of color (and only persons of color) are juxtaposed with images of the 9/11 terrorist attacks, in or on or hovering above a crackling fire, could be understood as a racial appeal.

Accordingly, we cannot accept the State's argument that we should find, as the court found in the case about judicial elections in Alabama, that “[t]here is no evidence that Alabama political campaigns generally ․ are characterized by racial appeals.” Milligan Doc. 481 at 191, ¶ 513 (quoting Ala. State Conf. of the NAACP, 612 F. Supp. 3d at 1311). That was a statement about a different record – a record that did not include testimony from Dr. Bagley, one that made no mention of Roy Moore's affection for slavery or a “war on [W]hites,” and one that primarily was focused on Alabama judicial elections (128 statewide judicial races over ad period of 38 years). See Ala. State Conf. of NAACP, 612 F. Supp. 3d at 1311.

But at the same time, we cannot find that this factor weighs as heavily in favor of the Plaintiffs as do the other factors we have discussed. Although the three examples we just described are prominent and recent, the record does not contain any systematic or statistical evaluation of the extent to which political campaigns are characterized by racial appeals, so we cannot determine whether these examples indicate that racial appeals occur frequently, regularly, occasionally, or rarely. Accordingly, we find that there is some evidence that political campaigns (more particularly, congressional campaigns) in Alabama are characterized by overt or subtle racial appeals.

f. Senate Factor 9

Senate Factor 9: Whether the policy underlying the Plan is “tenuous.”

In our first preliminary injunction, we made no finding about Senate Factor 9. See Milligan Doc. 107 at 193. In our second preliminary injunction, we again made no finding about Senate Factor 9. See Milligan Doc. 272 at 184. After full discovery and a trial, we find that Senate Factor 9 weighs strongly in favor of the Plaintiffs.

The Supreme Court has made clear that the policies underlying a districting plan may be tenuous if they entrench racial vote dilution, even if those policies might be legitimate in another context. See, e.g., LULAC, 548 U.S. at 441 (citing the tenuousness factor in explaining that using incumbency protection to exclude “voters from the district simply because they are likely to vote against the officeholder ․ cannot justify the effect on Latino voters”).

Here, the unusual genesis of the 2023 Plan, the novelty, substance, and effect of the embedded legislative findings, and the effect of the 2023 Plan support a finding of tenuousness. This is so even though some of the individual districting decisions reflected in the 2023 Plan, including some of the legislative findings, might be legitimate in another context.

We told the Legislature in 2022 that a districting plan with only one Black-opportunity district (the 2021 Plan) very likely violated Section Two. See Milligan Doc. 107. In that order, we explained that “any remedial plan will need to include two districts in which Black voters either comprise a voting-age majority or something quite close to it.” Id. at 6. The State appealed that decision to the Supreme Court, which affirmed our ruling in all respects. Allen v. Milligan, 599 U.S. at 16–17. When the case returned to us, we discerned no confusion or mystery about whether a plan with only one Black-opportunity district could comply with Section Two. Senator Livingston testified that he had the same understanding, as did Representative Pringle, who issued instructions about it to Mr. Hinaman. See Milligan Docs. 459-13 at 14, 459-20 at 8; supra Part I.I.3.a, b, c.

On remand, the Legislature asked for time to enact a new plan, and we allowed that time (approximately five weeks). The Legislature then purposefully enacted a plan with only one majority-Black or Black-opportunity district. Aug. 14, 2023 Tr. 75, 159–64; Milligan Doc. 409-85 (Ex. DX-87) (2023 Plan statistics).

If we had any doubt about what had just happened, the State resolved it when it sent a lawyer into court to concede that the 2023 Plan has only one Black-opportunity district and assert that notwithstanding our order and the Supreme Court's affirmance, the Legislature was not required to add another opportunity district. Aug. 14, 2023 Tr. 75, 159–64. The State told us that we would violate the Supreme Court's affirmance if we were to again require a second opportunity district, and that the Legislature's findings about communities of interest (which were Alabama law because they were embedded in the enacted plan) justified the State's decision to refuse to provide an additional opportunity district. Id.

It is a gross understatement for us to describe the Legislature's choices as tenuous. See infra Part VI (holding that they were intentionally discriminatory). After two federal court orders (one from us and one from the Supreme Court) explained in detail that the 2021 Plan, with only one majority-Black district, likely diluted Black Alabamians’ votes, the Legislature simply doubled down – it passed another map with only one Black-opportunity district.

But there is more. This time, the Legislature did not simply pass a new map. For the first time that anyone involved in these cases can remember or find, the Legislature included in the state law with the map extensive legislative findings -- the contents of which were unbeknownst to the two co-chairs of the Reapportionment Committee (Representative Pringle and Senator Livingston), who played no role in their drafting. Those novel findings had several notable features:

(1) they identified three communities of interest and went on for pages about only one of those (the Gulf Coast);

(2) that one community of interest was the community that would be served at the expense of the additional opportunity district;

(3) they identified no communities of interest in the northern half of the state;

(4) the cumulative effect of the “non-negotiable” provisions of the findings was to prescribe a majority-White district in an unsplittable community of interest, such that an additional opportunity district would be mathematically impossible to draw consistent with state law, see Tr. 298–99 (Dr. Duchin);

(5) although previous iterations of the Committee Guidelines (including the weeks-old 2023 guidelines) specified both that the Legislature intended to comply with the Voting Rights Act and that the plan “shall have neither the purpose nor the effect of diluting minority voting strength,” the 2023 legislative findings do not expressly prohibit a plan with the purpose or effect of diluting minority voting strength, see Apps. A, B.

See supra Parts I.D, I.I.3. This was not merely doubling down – this, the Legislature hoped, was checkmate.

But there is still more. Substantial evidence establishes that the Legislature's process to enact the 2023 Plan was anything but normal. The process began as normal: Governor Ivey called a special session, discussions between stakeholders began, and the Committee readopted the usual guidelines. Milligan Doc. 459-20 at 12–13. Then the Legislature took a series of unusual turns.

On the House side, Representative Pringle said that his own “overriding principle [was] complying with what the United States Supreme Court told me to do.” Milligan Doc. 459-20 at 6. According to Representative Pringle, that required a map with “[e]ither two majority minority districts or something close to it.” Id. at 5. Representative Pringle told Mr. Hinaman “to follow the guidelines and comply with what the Supreme Court told us. And that was to draw two districts which had the ability to elect a Black candidate.” Id. at 8. Eventually, Mr. Hinaman provided the map to Representative Pringle that he would introduce as the Community of Interest Plan. Id. at 9. In Representative Pringle's view, based on a performance analysis he reviewed, that Plan would “comply with what the Supreme Court ordered,” so he supported that Plan. Id. at 11, 18.

On the Senate side, things were different. Senator Livingston understood that federal courts had “ordered [the State] to provide two opportunity districts.” Milligan Doc. 459-13 at 6, 13–14. (He later testified that what it means to provide “two minority opportunity districts” is “vague,” and a “matter of interpretation,” and that he did not have an interpretation of his own. Id. at 6–7.) Despite that understanding, Senator Livingston did not support the Community of Interest Plan, which was the first plan reported out of the Committee during the 2023 Special Session. Id. at 16. He testified that the Community of Interest Plan “might have” “provided a fair opportunity for African American voters to elect preferred candidates in the second district” based on the performance analysis, but he shifted away from it because the Committee members did, and he “was going to be left behind.” Id. at 16–17.

The shift, Senator Livingston testified, was caused by “some additional information” Committee members received. Id. at 17. According to Senator Livingston, this information was the reason why “the committee moved” away from Representative Pringle's plan. Id. But Senator Livingston testified that he did not know what the “information” was, where it had come from, or even who received it. Id. at 17. He first learned of the “information” in a “committee conversation,” but he did not recall who told him about it and had no “idea at all” of its source. Id. This strains credulity.

Though the record does not tell us what that “additional information” was, it tells us that it was of great significance because it precipitated unusual political maneuvers from involved legislators, as well as an unusual result. Representative Pringle testified with apparent frustration about those unusual maneuvers. See supra Parts I.I.3.c, V.A.4.a; Milligan Doc. 459-20. While senators met with lawyers about what would eventually become the Livingston Plans, Representative Pringle was shut out, as was his plan. Milligan Doc. 459-20 at 7, 26. Then Senator Livingston told Representative Pringle that the Legislature would pass a Livingston Plan with a House bill number and Representative Pringle's name on it, and Representative Pringle refused out of a concern about compliance with federal law. Id. at 26. When Representative Pringle was later asked why the Senate chose the BVAP it did for District 2, he responded: “You're going to have to talk to Senator Livingston and [the Solicitor General].” Id.

At the end of all this, Representative Pringle saw the 2023 legislative findings for the first time the morning the 2023 Plan was passed; he did not know who drafted them, that they would be in the bill, or why they were in the bill. Id. at 23. Nevertheless, he voted for the bill.

Ultimately, when the Legislature passed the 2023 Plan, staked on an adamant refusal to split Mobile County to remedy unlawful vote dilution, the Legislature knew (1) that a plan with only one majority-Black district likely diluted Black Alabamians’ votes, in violation of the Voting Rights Act; and at least the conference committee also knew (2) that in a performance analysis of that plan, the Black-preferred candidate lost every race in District 2. Id. at 24–25. According to Senator Livingston, that analysis showed that the Black-preferred candidate would lose all seven races in District 2 by an average of seven points. Milligan Doc. 459-13 at 23.

But there is more. After all this, at least one legislator – an important one, the Speaker of the Alabama House – took to the media to explain that when the Legislature refused to provide an additional opportunity district, the Legislature was trying to get a different result at the Supreme Court. See, e.g., Milligan Doc. 403-4 at 28; Tr. 1342–43. (Nineteen months later, counsel for the State said at trial that when the Legislature passed the 2023 Plan without a second opportunity district, it “may have been hoping” to “find another argument” to persuade this Court and/or the Supreme Court that our orders were wrong. Tr. 2649.)

We observed the public aspects of these events as they unfolded and have now heard extensive testimony about them, and we are disturbed by them. The Legislature considered and rejected a map that might have provided the required remedy (an issue we need not and do not decide), in favor of a map that it knew in real time and later admitted in court certainly does not provide the required remedy. All to prescribe a majority-White congressional district in an exalted, unsplittable community of interest that was prioritized over every other districting principle, including compliance with federal law. This was not about compactness.

We thus find that the policy underlying the 2023 Plan is (at a minimum) tenuous. We do not make this finding because lawyers assisted with the mapmaking, nor because a lawyer drafted the 2023 legislative findings, nor because a lawyer made an argument in court. Lawyers regularly advise legislators about legislation and make arguments in court. It is the unusual process; novelty, substance, and effect of the 2023 legislative findings; and the effect of the legislation that compel this finding. The lawyer's concessions simply confirm it. United States v. Lopez, 590 F.3d 1238, 1256 (11th Cir. 2009) (“[S]tatements and arguments of counsel are not evidence ․” (quoting United States v. Smith, 918 F.2d 1551, 1562 (11th Cir. 1990))); United States v. Horn, 129 F.4th 1275, 1291 (11th Cir. 2025) (“[A] a statement made by [counsel] in argument ․ is not evidence.”).

g. Senate Factor 8

Senate Factor 8: “[W]hether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group.” Gingles, 478 U.S. at 37.

In our first preliminary injunction, we made no finding about Senate Factor 8. See Milligan Doc 107 at 192–93. The parties “vehemently dispute[d] whether the decisions that form the basis for the arguments of the Milligan plaintiffs and the Caster plaintiffs about this factor are political or race-based,” and on the record then before us, we could not “make a well-reasoned finding whether there is a lack of responsiveness on the part of elected officials in Alabama to the needs of the Black community, nor whether such lack of responsiveness (if it exists) is significant.” Id.

We again make no finding about Senate Factor 8 based on the Plaintiffs’ public policy arguments. See Milligan Doc. 485 at Part V.G.

In our second preliminary injunction, we found that “the circumstances surrounding the enactment of the 2023 Plan reflect ‘a significant lack of responsiveness on the part of elected officials to the particularized needs’ of Black voters in Alabama.” Milligan Doc. 272 at 179–80 (quoting Gingles, 478 U.S. at 37). We rested that finding on three undisputed facts, all of which we just discussed in connection with Senate Factor 9:

First, we discussed the unusual process in which the Legislature enacted the 2023 Plan, as well as its mysterious provenance. See id. at 180–81. We emphasized that the original source and cartographer for the map that eventually passed the Senate as a Livingston Plan and later (with the conference committee's adjustments) became the 2023 Plan were purportedly unknown to Senator Livingston when he voted on it. We discussed Representative Pringle's concern (after seeing a performance analysis) that the 2023 Plan did not or might not comply with Section Two. See id. at 181. And we discussed Representative Pringle's testimony that he “was not ‘attempting to get a justice to see something differently,’ but he did not ‘want to speak on behalf of 140’ Legislators.” Id. at 182 (quoting Milligan Doc. 261-5 at 109–10).

Second, we discussed the 2023 legislative findings. Id. at 182–83. We focused on the fact that “[a]lthough the findings eliminated the requirement of nondilution, they prioritized as ‘non-negotiable’ the principles that the 2023 Plan would ‘keep together communities of interest’ and ‘not pair incumbent[s].’ ” Id. at 183 (quoting 2023 legislative findings). We thus could not infer, let alone find, “that when the Legislature passed the 2023 Plan, it was trying to respond to the need that we identified for Black Alabamians not to have their voting strength diluted.” Id.

Third, we discussed the undisputed testimony of the Legislators, both of whom testified that they did not draft the 2023 legislative findings and did not know why they were included in the 2023 Plan. Id. at 183. We observed that Representative Pringle testified that he had not seen another redistricting bill contain similar (or any) findings. Id. In the light of this undisputed testimony, we “[could] not conclude that the findings weigh in favor of the 2023 Plan.” Id. at 183–84.

Ultimately, we “infer[red] from the Legislature's decision not to create an additional opportunity district that the Legislature was unwilling to respond to the well-documented needs of Black Alabamians in that way.” Id. at 184. We were clear that we had not deprived the Legislature of the presumption of good faith. See, e.g., Abbott, 585 U.S. at 603. We simply found “that on the undisputed evidence, Factor 8, like the other Factors, weighs in favor of the Plaintiffs.” Milligan Doc. 272 at 184.

For these same reasons, which remain undisputed after full discovery and a trial, we again find that Senate Factor 8 weighs heavily in favor of the Plaintiffs.

***

Accordingly, we find that every relevant Senate Factor weighs in favor of the Plaintiffs, and none is neutral or weighs in favor of the State. On balance, the analysis is not a close call.

h. Proportionality

Although Section Two expressly provides that “nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population,” 52 U.S.C. § 10301(b), the Supreme Court has held that “whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area” is a “relevant consideration” in the totality-of-the-circumstances analysis. LULAC, 548 U.S. at 426; accord De Grandy, 512 U.S. at 1000.

More particularly, “proportionality ․ is obviously an indication that minority voters have an equal opportunity, in spite of racial polarization, ‘to participate in the political process and to elect representatives of their choice.’ ” De Grandy, 512 U.S. at 1020 (quoting 42 U.S.C. § 1973(b) (current version at 52 U.S.C. § 10301(b))); accord Ala. Legis. Black Caucus, 989 F. Supp. 2d at 1286–87 (concluding that the totality of the circumstances weighed against a finding that the state legislative map violated Section Two in part because the number of majority-Black districts in the Legislature is “roughly proportional to the black voting-age population”), vacated on other grounds, 575 U.S. 254 (2015).

At the preliminary injunction stage, we did not resolve the Plaintiffs’ claims for relief solely (or even mainly) by conducting a proportionality analysis; consistent with LULAC and De Grandy, we considered proportionality as part and parcel of the totality of the circumstances, and we drew the limited and obvious conclusion that this consideration weighed in favor of the Plaintiffs.

On appeal, the Supreme Court explained the history of controversy tied to proportionality and Section Two. See Allen, 599 U.S. at 11–14. The Supreme Court further explained that “the Gingles framework itself imposes meaningful constraints on proportionality.” Id. at 26. The Supreme Court named cases in which it denied additional majority-minority districts on the ground that the proposed districts violated traditional districting criteria in service of proportional representation. Id. at 27–28 (citing Shaw v. Reno, Miller v. Johnson, and Bush v. Vera). The Supreme Court reiterated that “[f]orcing proportional representation is unlawful and inconsistent with [its] approach to implementing § 2.” Id. at 28.

The dissenting Justices saw the issue differently. Justice Thomas described the dispositive question before the Court as “whether [Section Two], as amended, requires the State of Alabama to intentionally redraw its longstanding congressional districts so that black voters can control a number of seats roughly proportional to the black share of the State's population.” Id. at 46 (Thomas, J., dissenting). In his view (shared by the joining Justices), vote dilution cases require a race-neutral benchmark for comparison, and the Supreme Court has “never succeeded in translating the Gingles framework into an objective and workable method of identifying the undiluted benchmark.” Id. at 69. This void, they reasoned, resulted in us applying “the decidedly nonneutral benchmark of proportional allocation of political power based on race.” Id. at 51.

Accordingly, we have not considered proportionality in this Order and do not rely on it for any purpose. We proceed in this manner (1) out of an abundance of caution, to avoid any risk of error, and (2) because although we discern a diversity of opinion on the Supreme Court about proportionality, we are aware of no such diversity on another question now in these cases, about a state legislature's intentional decision to refuse a remedy that a federal court order requires.

***

We thus find that Plaintiffs have established that: (1) as a group, Black Alabamians are sufficiently numerous and geographically compact to constitute a voting-age majority in a second reasonably configured district; (2) voting in the challenged districts is intensely racially polarized, such that Black voters are (nearly always) politically cohesive and (3) White voters ordinarily (nearly invariably) vote as a bloc to defeat Black-preferred candidates; and (4) under the totality of the circumstances in Alabama today, including all the relevant Senate Factors that we must consider, Black voters have less opportunity than other Alabamians to elect candidates of their choice to Congress. We turn to the State's two legal arguments.

B. Section Two Is Privately Enforceable

Since the passage of the Voting Rights Act, federal courts across the country, including both the Supreme Court and the Eleventh Circuit, have considered numerous Section Two cases brought by private plaintiffs. See, e.g., Allen, 599 U.S. 1; Brnovich, 594 U.S. 647; Bartlett, 556 U.S. 1; LULAC, 548 U.S. 399; Voinovich, 507 U.S. 146; Chisom v. Roemer, 501 U.S. 380 (1991); Hous. Laws.’ Ass'n v. Att'y Gen. of Tex., 501 U.S. 419 (1991); Gingles, 478 U.S. 30; Wright, 979 F.3d 1282. And on the other side of the scale, only one federal appellate court—the United States Court of Appeals for the Eighth Circuit—has held that private parties may not sue to enforce Section Two. See generally Ark. State Conf. NAACP v. Ark. Bd. of Apportionment, 86 F.4th 1204 (8th Cir. 2023).

Accordingly, if we were to accept the State's argument that private parties may not enforce Section Two, we would seriously disrupt longstanding and consistent federal law on this issue. We are not inclined to take that step.

We already rejected the State's argument that Section Two is not privately enforceable in our July 11, 2024 order during the preliminary injunction stage of these proceedings. Milligan Doc. 372. Because the State has repeated its argument, we repeat our answer here:

1. Text of Section Two

Federal law supplies two potential vehicles for private plaintiffs to sue under Section Two: either by way of a private right of action contained in Section Two itself, or pursuant to 42 U.S.C. § 1983 (“Section 1983”). Section Two contains no express private right of action, so the dispositive question is whether one is implied. To establish an implied private right of action, plaintiffs must show that Section Two confers both a private right and a private remedy. Alexander v. Sandoval, 532 U.S. 275, 286 (2001). If there is a private right, then private plaintiffs can presumptively sue under Section 1983, unless defendants show that Congress shut the door to a Section 1983 suit. Gonzaga Univ. v. Doe, 536 U.S. 273, 284 & n.4 (2002). Then–Chief Justice Rehnquist, writing for the majority in Gonzaga, reasoned this way:

Plaintiffs suing under § 1983 do not have the burden of showing an intent to create a private remedy because § 1983 generally supplies a remedy for the vindication of rights secured by federal statutes. Once a plaintiff demonstrates that a statute confers an individual right, the right is presumptively enforceable by § 1983.

Id. at 284 (internal citation omitted). And then, the State must “demonstrate that Congress shut the door to private enforcement.” Id. at 284 n.4.

The State concedes that Section Two created “new remedies,” but contends those remedies were only public, not private. See Milligan Doc. 481 at 228–29; Milligan Doc. 331 at 17; Caster Doc. 273 at 17. And the State has not given any reasons why it believes Section Two did not create a private remedy separate and apart from the reasons why it asserts Section Two did not create a private right. See Milligan Doc. 481 at 226–29; Milligan Doc. 331 at Part I.A; Caster Doc. 273 at Parts I, II.

All three sets of Plaintiffs have availed themselves of Section 1983, Singleton Doc. 229 ¶ 5; Milligan Doc. 329 ¶ 11; Caster Doc. 271 ¶ 129, and the State does not assert that Congress has shut the door to a remedy under Section 1983. See Milligan Doc. 481 at 226–29; Singleton Docs. 233, 239; Milligan Docs. 331, 342; Caster Docs. 273, 282. Accordingly, the essential question before us is whether Section Two creates a private right. If we conclude that it does, there is no basis to accept the State's argument that Section Two is not privately enforceable.

Although the task of determining whether Section Two contains a private right is ours, the creation of that right (if it exists) is an exclusively legislative authority. “Like substantive federal law itself, private rights of action to enforce federal law must be created by Congress. The judicial task is to interpret the statute Congress has passed to determine whether it displays an intent to create not just a private right but also a private remedy.” Sandoval, 532 U.S. at 286 (internal citation omitted). Accordingly, we examine at the threshold “whether Congress intended to create a federal right.” Gonzaga, 536 U.S. at 283.

A statute confers a private right “where the provision in question is phrased in terms of the persons benefitted and contains rights-creating, individual-centric language with an unmistakable focus on the benefited class.” Health & Hosp. Corp. of Marion Cnty. v. Talevski, 599 U.S. 166, 183 (2023) (internal quotation marks omitted) (quoting Gonzaga, 536 U.S. at 284, 287). A statute does not confer a private right when it contains no rights-creating language or focuses on persons or entities other than the benefited class. See, e.g., Sandoval, 532 U.S. at 288–89.

The most recent binding Supreme Court precedent about rights-creating language is Health & Hospital Corporation of Marion County, 599 U.S. 166 (2023), a case concerning two statutory provisions about the rights of nursing home residents. Id. at 171. We apply here the same methodology the Supreme Court used to decide that case, which can be summarized in this way:

• First, the Court began its analysis by observing that the statutory provisions at issue “reside in” a statutory section that “expressly concerns ‘[r]equirements relating to residents’ rights.’ ” Id. at 184 (emphasis omitted) (quoting 42 U.S.C. § 1396r(c)). In assigning weight to this observation, the Supreme Court relied on (1) the rule that “statutory provisions ‘must be read in their context,’ ” and (2) the recognition in Gonzaga that “[t]his framing is indicative of an individual ‘rights-creating’ focus.” Id. (first quoting West Virginia v. EPA, 597 U.S. 697, 721 (2022); and then quoting Gonzaga, 536 U.S. at 284).

• Next, the Court reviewed each statutory provision at issue and found that each one (1) discussed a specific right held by residents, with (2) a repeated focus on residents. See id. at 184–85.

• Then, the Court observed that the statutory provisions also discussed nursing homes, but found that this discussion did not undermine the focus of the provisions on residents’ rights. The Court reasoned that “it would be strange to hold that a statutory provision fails to secure rights simply because it considers, alongside the rights bearers, the actors that might threaten those rights.” Id. at 185.

• Finally, the Court distinguished the statutory provisions from the provisions in Gonzaga, which “lacked ‘rights-creating language,’ primarily directed the Federal Government's distribution of public funds, and had an aggregate, not individual, focus.” Id. at 185–86 (quoting Gonzaga, 536 U.S. at 290).

Like the provisions at issue in Health & Hospital Corporation of Marion County, Section Two resides in a statutory section that expressly concerns rights—in this case, voting rights for members of a class protected from discrimination based on race or color. The title of Section Two is “[d]enial or abridgement of right to vote on account of race or color through voting qualifications or prerequisites; establishment of violation.” 52 U.S.C. § 10301. Following the Supreme Court's lead, we take this context and framing as “indicative of an individual ‘rights-creating’ focus.” Health & Hosp. Corp. of Marion Cnty., 599 U.S. at 184 (quoting Gonzaga, 536 U.S. at 284).

Further, subsection (a) of Section Two expressly discusses “the right of any citizen of the United States to vote,” and it expressly prohibits voting practices that abridge voting rights based on race, color, or language-minority status. 52 U.S.C. § 10301(a) (incorporating by reference 52 U.S.C. § 10303(f)(2)). And subsection (b) expressly discusses the voting rights of persons who are “members of a class of citizens protected by subsection (a).” Id. § 10301(b). In the next sentence, subsection (b) refers twice to “members of a protected class.” Id. Together, these subsections protect citizens in the enumerated class from voting practices with discriminatory results, not just voting practices based on discriminatory intent (which the Fifteenth Amendment forbids based on race or color). See Reno v. Bossier Par. Sch. Bd., 520 U.S. 471, 482 (1997); U.S. Const. amend. XV. Because Section Two is comprised only of a title and three sentences of text, the upshot of the foregoing analysis is that every sentence of Section Two either refers to rights of the benefited class, contains rights-creating language that creates new rights for that specific class, or expressly focuses on the benefited class.

This precise and repetitive focus on the benefitted class distinguishes Section Two from the statutes at issue in Sandoval and Gonzaga, which the Supreme Court concluded did not confer implied private rights of action. In Sandoval, the statute at issue—Section 602 of Title VI of the Civil Rights Act, 42 U.S.C. § 2000d-1—did not even mention the benefited class: it said merely that “[e]ach Federal department and agency ․ is authorized and directed to effectuate the provisions of [Section 601].” 532 U.S. at 288–89 (quoting 42 U.S.C. § 2000d-1). Thus, the Court found that “the focus of § 602 is twice removed from the individuals who will ultimately benefit from Title VI's protection” because it “focuses neither on the individuals protected nor even on the funding recipients being regulated, but on the agencies that will do the regulating.” Id. at 289.

Likewise, Gonzaga considered provisions of the Family Educational Rights and Privacy Act of 1974, 20 U.S.C. § 1232g (“FERPA”). Gonzaga, 536 U.S. at 278. One such provision stated that: “No funds shall be made available under any applicable program to any educational agency or institution which has a policy or practice of permitting the release of education records ․ of students without the written consent of their parents ․,” id. at 279 (quoting 20 U.S.C. § 1232g(b)(1)), while another “direct[ed] the Secretary of Education to enforce this and other of the Act's spending conditions,” id. (citing 20 U.S.C. § 1232g(f)). The Court found that the focus of these provisions was also “two steps removed from the interests of” the benefited class because they “speak only to” the regulating agency. Id. at 287. The Court concluded that the provisions at issue did not imply a private right because they “contain no rights-creating language, they have an aggregate, not individual, focus, and they serve primarily to direct the [regulating agency's] distribution of public funds to educational institutions.” Id. at 290.

Unlike the statutes in Sandoval and Gonzaga, the language of Section Two “focuses ․ on the individuals protected.” Sandoval, 532 U.S. at 289. It explicitly protects “the right of any citizen of the United States to vote” without being discriminated against, and then refers repeatedly to “members of a protected class,” or some variation of that phrase. See 52 U.S.C. § 10301. It “serve[s] primarily” to protect citizens’ rights and to prevent states from interfering with those rights. See Gonzaga, 536 U.S. at 290. If all of this is not rights-creating language with an “unmistakable focus on the benefited class,” Cannon v. Univ. of Chi., 441 U.S. 677, 691 (1979), it is difficult to imagine what is.

Indeed, Section Fourteen of the Voting Rights Act reinforces the idea that Congress contemplated suits by private parties when it enacted Section Two. Section 14(e) provides: “In any action or proceeding to enforce the voting guarantees of the fourteenth or fifteenth amendment, the court, in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney's fee, reasonable expert fees, and other reasonable litigation expenses as part of the costs.” 52 U.S.C. § 10310(e). “[A]ny action or proceeding to enforce the voting guarantees of the fourteenth or fifteenth amendment” means all such actions or proceedings, because where Congress uses the word “any” and “ ‘did not add any language limiting the breadth of that word,’ ․ ‘any’ means all.” Merritt v. Dillard Paper Co., 120 F.3d 1181, 1186 (11th Cir. 1997) (quoting United States v. Gonzales, 520 U.S. 1, 5 (1997))); see also Deroy v. Carnival Corp., 963 F.3d 1302, 1316 (11th Cir. 2020) (recognizing that, in a statute, “ ‘any’ means ‘every’ or ‘all’ ” (citing United States v. Castro, 837 F.2d 441, 445 (11th Cir. 1988))). And Section Two is unambiguously an action or proceeding to “enforce the voting guarantees of the ․ fifteenth amendment.” 52 U.S.C. § 10310(e); see Brnovich, 594 U.S. at 656. Section Fourteen therefore anticipates that private litigants will sue to “enforce the guarantees of the ․ fifteenth amendment” alongside the United States. 52 U.S.C. § 10310(e).

The Eighth Circuit says, however, that the term “prevailing party” here refers only to defendants. Ark. State Conf. NAACP, 86 F.4th at 1213 n.4. As we see it, that offers too strained a reading of the statute. Congress specified that a “prevailing party, other than the United States” should receive attorneys’ fees, not that a “defendant” should receive attorneys’ fees—which would have been a much simpler and more direct way to prescribe that outcome, if that is what Congress had intended. In fact, the Supreme Court has construed identical language found in the attorney-fee provision of Title II of the Civil Rights Act of 1964, 42 U.S.C. § 2000a-3(a) (“CRA”),⁶³ to refer to private plaintiffs. See Newman v. Piggie Park Enters., Inc., 390 U.S. 400, 401 n.1, 402 (1968) (per curiam) (holding that the term “prevailing party, other than the United States” in Title II's attorney-fee provision refers to private plaintiffs); see also id. at 402 (“Congress ․ enacted the provision for counsel fees [in Title II of the CRA] ․ to encourage individuals injured by racial discrimination to seek judicial relief under Title II.”). Moreover, Congress has specifically told us that it intended private parties to be able to recover attorneys’ fees if they prevailed on Section Two claims: Congress explained that “[f]ee awards are a necessary means of enabling private citizens to vindicate these Federal rights.” See S. Rep. No. 94-295, at 40 (1975) (emphasis added); see also H. Rep. No. 97-227, at 32 (1981) (“It is intended that citizens have a private cause of action to enforce their rights under Section 2 ․ If they prevail they are entitled to attorneys’ fees under [Section 14(e)] and [42 U.S.C. §] 1988.”).

“[T]he words of a statute must be read in their context and with a view to their place in the overall statutory scheme.” West Virginia, 597 U.S. at 721. Thus, the reference in Section Fourteen of the Voting Rights Act to private plaintiffs suing to enforce their voting rights supports the determination that Section Two contains a private right of action. Viewing Section Two along with Section Fourteen reinforces Congress's intention to allow private parties to sue to enforce their right to vote free from discrimination. See Morse v. Republican Party of Va., 517 U.S. 186, 234 (1996) (reasoning that the language referring to a “prevailing party, other than the United States” in Section Fourteen indicates “the existence of a private right of action under § 10”).

As far as we can tell, no court has held under the first step of the analysis that Section Two does not create a private right. Rather, the one circuit court that has concluded that Section Two does not confer a private right of action, the Eighth Circuit, rested its decision on the second step of the analysis—a determination that Section Two does not create a private remedy. See Ark. State Conf. NAACP, 86 F.4th at 1216. Notably, the Eighth Circuit did not address the question whether private plaintiffs may sue under Section 1983 to enforce Section Two because the plaintiffs had not raised the issue. Id. at 1218.

The Eighth Circuit viewed the question whether Section Two creates a private right as an open one because, in addition to the rights-creating language we have described, Section Two also contains language that refers to states, and the court was unsure “what to do when a statute focuses on both.” Id. at 1209–10. But the Supreme Court has provided an unambiguous answer to that question that the Eighth Circuit did not consider.⁶⁴ In Health & Hospital Corporation of Marion County, the statutes at issue (like Section Two) referred to the rights of the benefitted class, but also directed requirements at “actors that might threaten those rights,” and the Supreme Court still held that the statutes created private rights. 599 U.S. at 185. That a statutory provision discussing the rights of a benefitted class “also establish[es] who it is that must respect and honor these statutory rights,” the Court explained, “is not a material diversion from the necessary focus on the [rights-holders].” Id. The Court further reasoned that “[t]he Fourteenth Amendment hardly fails to secure § 1983-enforceable rights because it directs state actors not to deny equal protection.” Id. at 185 n.12.

Based on case precedent and the text of Section Two, we see a clear answer to the question whether Section Two creates a private right: it does. Nevertheless, the State urged us in its earlier motion to hold that Section Two does not confer a private right for four reasons. We discuss each in turn.

First, the State argued in its earlier motion that for Section Two to create a private right of action, it must create a new right not found elsewhere in federal law. See Singleton Doc. 233 at 23–24; Milligan Doc. 331 at 17. The State claims that Section Two cannot do this because it was passed pursuant to Congress's power under Section Two of the Fifteenth Amendment, which gives Congress the power to enforce the rights guaranteed in the Fifteenth Amendment, but not the power to create new rights. See U.S. Const. amend. XV; Brnovich, 594 U.S. at 656; see Milligan Doc. 331 at 17–18.⁶⁵

The State is wrong that to create a private right of action, Section Two must create a new right not found elsewhere in federal law. That premise runs headlong into controlling precedent. For example, in Morse, 517 U.S. 186, the Court found an implied private right of action in Section Ten of the Voting Rights Act, which, on the State's logic, would also merely be protecting preexisting Fifteenth Amendment rights. See id. at 233 (holding that § 10 “established a right to vote without paying a fee”). And in Allen v. State Board of Elections, 393 U.S. 544 (1969), abrogated by Ziglar v. Abbasi, 582 U.S. 120, 132 (2017), the Supreme Court found an implied private right of action in Section Five of the Voting Rights Act. See id. at 557; cf. Schwier v. Cox, 340 F.3d 1284, 1297 (11th Cir. 2003) (finding an implied private right of action in the materiality provision of a similar statute passed under congressional Fifteenth Amendment enforcement power).

It is unsurprising, then, that the State has cited no precedent holding that Congress cannot imply a private right of action to enforce an existing federal right. It relies on language found in Sandoval (quoted later in Gonzaga) referring to “new rights,” but that language did not hold (or even suggest, in the context of those cases), that the protected right must be completely novel and found nowhere else in federal law. In Sandoval, the Court used the term “new rights” to explain that rights-creating language in one section of a statute did not necessarily imply a private right of action to enforce another section of the same statute. See 532 U.S. at 289 (cleaving a difference between Sections 601 and 602 of Title VI of the Civil Rights Act of 1964). Sandoval did not address the question whether Congress may grant a private right of action to enforce an existing federal right. Nor did Gonzaga, which merely quoted the sentence from Sandoval referring to “new rights” when explicating the general background principles for discovering congressional intent. See Gonzaga, 536 U.S. at 286–87. There was no discussion in Gonzaga of whether the rights referred to in the statute at issue were new or not. See id.

Second, the State argued in its earlier motion that Section Two does not unambiguously confer individual rights because there is ambiguity about its focus, which the State says one court has held is “unclear” because it includes both the conduct prohibited and the party regulated. Milligan Doc. 331 at 20 (citing Ark. State Conf. NAACP, 86 F.4th at 1209–10). But like the Eighth Circuit, the State does not account for the instructions found in Health & Hospital Corporation of Marion County. See 599 U.S. at 185. As we have already explained, see supra pp. 420–21, if the statutory text at issue in that case created private rights while also mentioning actors and conduct that could threaten those rights, then we can discern no principled basis to conclude that Section Two does not likewise create private rights.

Third, the State argued it is earlier motion that the mere use of the term “rights” is not enough to create a private cause of action, citing Pennhurst State School & Hospital v. Halderman, 451 U.S. 1 (1981). See Milligan Doc. 342 at 10. But our analysis doesn't rest exclusively on the use of the word “rights.” See supra pp. 418–24; infra Part V.B.2. In any event, Pennhurst State does not help the State. There, the Supreme Court declined to find an implied right in a statute that provided that mentally handicapped persons “have a right to appropriate treatment, services, and habilitation” in “the setting that is least restrictive of ․ personal liberty,” Pennhurst State, 451 U.S. at 13 (quoting 42 U.S.C. § 6010). The Court held that the reference to “a right” was precatory because it was found only in a “bill of rights” provision of the statute, while the enabling provisions of the statute were funding-related, and the bill of rights provision lacked “any language suggesting that [it] is a ‘condition’ for the receipt of federal funding” under the statute. Id. To the contrary, the Court reasoned, the language and structure of the statute “demonstrate[d] that it is a mere federal-state funding statute.” Id. at 18. Pennhurst State thus cautions that “[i]n expounding a statute, we must not be guided by a single sentence or member of a sentence, but look to the provisions of the whole law, and to its object and policy.” Id.

We have not looked at the word “rights” in a vacuum; rather, we have considered the word within the statutory provision and the statute taken as a whole, in order to see whether the statutory provision is using “rights-creating language.” Sandoval, 532 U.S. at 288 (quotation marks and citation omitted). And it is not merely the presence of the term “rights” in Section Two, but rather the entire provision's focus on the rights of “members of a protected class” and its place within the Voting Rights Act—a statute created, after all, for the sole purpose of enforcing a citizen's right to vote free from discrimination.

Fourth, the State asserted in its earlier motion that the “federal review mechanism” in the Voting Rights Act indicates that Congress did not mean to imply a private right of action in Section Two. Caster Doc. 273 at 27. The State relies on Gonzaga to argue that “where a statute provides a federal review mechanism, the Supreme Court has been less willing to identify individually enforceable private rights.” Id. (internal quotation marks omitted).

This argument fails at the gate because FERPA, the statute at issue in Gonzaga, is fundamentally unlike Section Two. In Gonzaga, the Supreme Court observed that its “conclusion that FERPA's nondisclosure provisions fail to confer enforceable rights [wa]s buttressed by the mechanism that Congress chose to provide for enforcing those provisions.” 536 U.S. at 289. FERPA “expressly authorized the Secretary of Education to ‘deal with violations’ of the Act,” and the Secretary did so by creating an office to field complaints from individuals and then initiate investigations, request a response from the institution subject to the complaint, find violations, and mandate steps to resolve them. Id. at 289–90 (emphasis omitted) (quoting 20 U.S.C. § 1232(g)(f)). But Congress chose no such extensive administrative procedures for Section Two, and they differ in kind from the Attorney General's prosecutorial discretion to bring Section Two lawsuits in court. Allowing the Attorney General to elect to bring a lawsuit is not the kind of detailed alternative “federal review mechanism” Congress created to enforce FERPA, which the Gonzaga Court was discussing. See id. at 290.

Even if the Attorney General's power to sue were like the elaborate federal review mechanism described in Gonzaga (and it is not), Gonzaga clarifies that the likeness is not “an independent basis for precluding private enforcement.” Id. at 290 n.8. This fits with other jurisprudence allowing both private and public lawsuits to enforce federal rights. See, e.g., Fitzgerald v. Barnstable Sch. Comm., 555 U.S. 246, 255 (2009) (finding a private right of action in Title IX of the Civil Rights Act despite it having an “express enforcement mechanism” in the form of “an administrative procedure”). Put simply, the reality that the Attorney General may bring a lawsuit in federal court does not compel, or even suggest, the conclusion that Congress meant to imply no right of action for private individuals also to bring enforcement actions pursuant to Section Two of the Voting Rights Act.

2. Section Two Precedents

Standing alone, our conclusion that the text of Section Two implies a private right of action is a sufficient reason to hold the statute privately enforceable. But there is more. Relevant precedent also supports our conclusion, including in particular two Supreme Court cases: Morse and Allen. And principles of congressional ratification and statutory stare decisis reinforce that result.

a. Relevant Precedent

As we previously explained, “[a] ruling that Section Two does not provide a private right of action would badly undermine the rationale offered by the Court in Morse.” Singleton, 582 F. Supp. 3d 924, 1031 (N.D. Ala. 2022). In Morse, the Supreme Court held that Section 10 of the Voting Rights Act contained a private right of action, reasoning that:

Although § 2, like § 5, provides no right to sue on its face, “the existence of the private right of action under Section 2 ․ has been clearly intended by Congress since 1965.” S. Rep. No. 97–417, at 30. We, in turn, have entertained cases brought by private litigants to enforce § 2. It would be anomalous, to say the least, to hold that both § 2 and § 5 are enforceable by private action but § 10 is not, when all lack the same express authorizing language.

517 U.S. at 232 (opinion of Stevens, J., with Ginsburg, J. joining) (some internal citations omitted); see id. at 240 (opinion of Breyer, J., with O'Connor, J. and Souter, J. joining) (agreeing that Section 10 confers a private right of action because Sections Two and Five do).

The Court's conclusion that Section 10 affords a private right of action turns in no small measure on its foundational observation that Section Two, like Section Five, is indeed enforceable by private right of action. See id. at 232. And the Court saw no reason for treating Section Ten any differently. Id. The very rationale for the Supreme Court's determination that Section Two affords a private right of action is that Congress has “clearly intended” that since 1965. Id. (quoting S. Rep. No. 97–417, at 30); see also Singleton, 582 F. Supp. 3d at 1031 (“[T]he understanding [in Morse] that Section Two provides a private right of action was necessary to reach the judgment that Section Ten provides a private right of action.”).⁶⁶

“When an opinion issues for the Court, it is not only the result but also those portions of the opinion necessary to that result by which we are bound.” Seminole Tribe of Fla. v. Florida, 517 U.S. 44, 67 (1996); see also Dana's R.R. Supply v. Att'y Gen., 807 F.3d 1235, 1240 n.3 (11th Cir. 2015) (noting that a statement is dicta only if it “could have been deleted without seriously impairing the analytical foundations of the holding” (quoting Denno v. Sch. Bd. of Volusia Cnty., 218 F.3d 1267, 1283 (11th Cir. 2000) (Forrester, J., concurring in part))); United States v. Crawley, 837 F.2d 291, 292 (7th Cir. 1988) (Posner, J.) (same). This holds true for any analysis that the court “explicat[es] and appl[ies],” even where the court “could have decided the case on other grounds.” United States v. Kaley, 579 F.3d 1246, 1253 n.10 (11th Cir. 2009).

However, even if we were to treat Morse’s statements as dicta, we are “obligated to respect [them].” Henderson v. McMurray, 987 F.3d 997, 1006 (11th Cir. 2021) (Pryor, C.J.). “[T]here is dicta and then there is dicta, and then there is Supreme Court dicta.” Schwab v. Crosby, 451 F.3d 1308, 1325 (11th Cir. 2006). As far as we see it, at the very least, this is Supreme Court dicta with the support of five justices; and if it is a holding, plainly it would be controlling, despite the fractured votes. See Marks v. United States, 430 U.S. 188, 193 (1977). We will not upend it.

In the 310-page proposed order the State submitted after trial, it did not mention Morse, either in connection with its argument that Section Two is not privately enforceable or otherwise. See Milligan Doc. 481 at 226–29. In its earlier motion, the State urged us to ignore the Morse language on the ground that it is gravely wounded by Sandoval. See Milligan Doc. 331 at 22–23; Caster Doc. 273 at 34–35. The Supreme Court has spurned some private-right-of-action cases that were decided before Sandoval, describing them as part of an “ancien regime” in which “the Court assumed it to be a proper judicial function to provide such remedies as are necessary to make effective a statute's purpose.” Ziglar, 582 U.S. at 131–32 (internal quotation marks and citation omitted). But Morse is not even mentioned in Sandoval and it is not part of the ancien regime that Sandoval criticized. As the Supreme Court explained in Sandoval, the headline case for abandoning the ancien regime was Cort v. Ash, 422 U.S. 66 (1975). See Sandoval, 532 U.S. at 287. Morse was decided twenty-one years after Cort. As an inferior federal court, we are required to “leav[e] to [the Supreme Court] the prerogative of overruling its own decisions.” Agostini v. Felton, 521 U.S. 203, 237 (1997) (citation omitted); see also United States v. Gibson, 434 F.3d 1234, 1246 (11th Cir. 2006) (“It is not given to us to overrule the decisions of the Supreme Court.”).

Furthermore, Shelby County v. Holder also suggested, albeit in dicta, that Section Two implies a private right of action, and Shelby County postdates Sandoval. In Shelby County, the Supreme Court invalidated Section Five's preclearance regime as unconstitutional. 570 U.S. at 537–38. In describing the statutory scheme, the Court explained that “[b]oth the Federal Government and individuals have sued to enforce § 2, and injunctive relief is available in appropriate cases to block voting laws from going into effect.” Id. at 537 (citations omitted). And in the final paragraph of the opinion, the Court ruled that its decision about Section Five “in no way affects the permanent, nationwide ban on racial discrimination in voting found in § 2.” Id. at 557. The State's earlier argument about Sandoval did not account for Shelby County either. See Milligan Doc. 331 at 22–23; Caster Doc. 273 at 34–35; Milligan Doc. 481 at 227.

Other federal circuits apparently share our understanding of Supreme Court jurisprudence, including the Eleventh Circuit. See Ala. State Conf. NAACP v. Alabama, 949 F.3d 647, 649 (11th Cir. 2020), vacated on other grounds sub nom. Alabama v. Ala. State Conf. NAACP, 141 S. Ct. 2618 (2021); see also Robinson v. Ardoin, 86 F.4th 574, 587–88 (5th Cir. 2023) (“We conclude that ․ there is a right for these [private] Plaintiffs to bring these [Section Two] claims.”); Mixon v. Ohio, 193 F.3d 389, 406 (6th Cir. 1999) (“An individual may bring a private cause of action under Section 2 of the Voting Rights Act.”).⁶⁷

In 2020, the Eleventh Circuit explained the history of private enforcement of Section Two this way:

The Voting Rights Act (VRA) is widely considered to be among the most effective civil rights statutes ever passed by Congress. Its success is largely due to the work of private litigants. For more than fifty years, private parties have sued states and localities under the VRA to enforce the substantive guarantees of the Civil War Amendments. Today, private parties remain the primary enforcers of § 2 of the VRA.

Ala. State Conf. NAACP, 949 F.3d at 649 (footnotes omitted). The Eleventh Circuit went on to observe that “[t]he Department of Justice has filed only 4 of the 61 enforcement actions under § 2 since 2013.” Id. n.2.⁶⁸ And the Circuit held that “[t]he VRA, as amended, clearly expresses an intent to allow private parties to sue the States. The language of § 2 and § 3, read together, imposes direct liability on States for discrimination in voting and explicitly provides remedies to private parties to address violations under the statute.” Id. at 652. Although we are not bound by this Circuit precedent because it was vacated on mootness grounds, the analysis is persuasive.

We next turn to the Supreme Court's decision in these very cases. Although Allen did not resolve the specific question whether Section Two provides a private right of action, it is nevertheless instructive. In Allen, the Supreme Court recognized that “[b]y 1981, ․ only sixteen years[ ] [after the VRA was passed in 1965], many considered the VRA ‘the most successful civil rights statute in the history of the Nation.’ ” Allen, 599 U.S. at 10 (quoting S. Rep. No. 97–417, at 111 (1982)). “The Act ‘create[d] stringent new remedies for voting discrimination,’ attempting to forever ‘banish the blight of racial discrimination in voting.’ ” Id. (quoting South Carolina v. Katzenbach, 383 U.S. 301, 308 (1966)). The Court described important amendments to Section Two enacted in 1982, and observed that since then, “[f]or the past forty years, [the Court has] evaluated claims brought under § 2 using the three-part framework developed in [its] decision in Thornburg v. Gingles, 478 U.S. 30 ․ (1986).” Allen, 599 U.S. at 17. That jurisprudence includes legions of Section Two claims asserted by private plaintiffs and adjudicated by the Supreme Court: Gingles, 478 U.S. 30; Voinovich, 507 U.S. 146; Growe, 507 U.S. 25; De Grandy, 512 U.S. 997; Holder v. Hall, 512 U.S. 874 (1994); Vera, 517 U.S. 952; Shaw II, 517 U.S. 89; Abrams, 521 U.S. 74; LULAC, 548 U.S. 399; Cooper, 581 U.S. 285; Abbott, 585 U.S. 579; Brnovich, 594 U.S. 647; Wis. Legislature, 595 U.S. 398; Allen, 599 U.S. 1.

b. Congressional Ratification

As the Allen Court explained repeatedly in the context of other attacks on Section Two, this long history of private plaintiffs bringing Section Two challenges means that Congress is “undoubtedly aware of [the Court's] constru[ction of] § 2,” and “Congress has never disturbed [the Court's] understanding of § 2 as Gingles construed it.” 599 U.S. at 19, 39. And Congress “can change that if it likes” Id. at 39.

It has long been the rule that “Congress is presumed to be aware of an administrative or judicial interpretation of a statute and to adopt that interpretation when it re-enacts a statute without change.” Forest Grove Sch. Dist. v. T.A., 557 U.S. 230, 239–40 (2009) (citation omitted). In none of its amendments to the Voting Rights Act has Congress ever questioned the then-unanimous view of the courts that Section Two was privately enforceable. See generally Pub. L. No. 91-285, 84 Stat. 314 (1970); Pub. L. No. 94-73, 89 Stat. 400 (1975); Pub. L. No. 97-205, 96 Stat. 131 (1982); Pub. L. No. 109-246, 120 Stat. 577 (2006). In its most recent amendment, in 2006, Congress expressly noted “the continued filing of section 2 cases that originated in covered jurisdictions” as “[e]vidence of continued discrimination” that supported the need to strengthen certain provisions of the Voting Rights Act. Pub. L. No. 109-246, 120 Stat. 577 (2006).

Indeed, the Senate Report to the 1982 amendment, which the Supreme Court has called the “authoritative source for legislative intent” behind Section Two, Gingles, 478 U.S. at 43 n.7, said that it “reiterates the existence of the private right of action under Section 2, as has been clearly intended by Congress since 1965,” S. Rep. No. 97-417, at 30 (citing Allen, 393 U.S. 544). The House Report to the 1982 amendment echoes precisely the same congressional intent. See H. Rep. No. 97-227, at 32 (1981) (“It is intended that citizens have a private cause of action to enforce their rights under Section 2.”). And the Senate Report to the 1975 amendment explains that fee awards under Section Fourteen of the Voting Rights Act “are a necessary means of enabling private citizens to vindicate these Federal rights.” S. Rep. No. 94-295, at 40 (1975) (emphasis added). Congress has not only ratified the federal courts’ longstanding interpretation that Section Two may be enforced by private plaintiffs through inaction by failing to change the law, but it has also explicitly stated that it agrees with this interpretation.

In its earlier motion, the State nevertheless urged us that because the Supreme Court has not definitively decided the issue, there is no interpretation for Congress to ratify. See Milligan Doc. 331 at 21, 23. This argument ignores the reality that (but for one very recent occasion), no federal court has ever closed its doors to a private plaintiff seeking to enforce Section Two on the ground that it implies no private right of action. The point is simple: if we have consistently misunderstood a congressional enactment in case after case, court after court, decade after decade, surely Congress would have told us so by now. Nearly forty years after Gingles—and nearly sixty years after the passage of the Voting Rights Act—it is appropriate to assign some degree of legal significance to this reality, even if only as a data point that confirms our reading of the text.

c. Statutory Stare Decisis

In addition, statutory stare decisis principles counsel that we should stay the course in allowing private plaintiffs to sue under Section Two. “[S]tare decisis carries enhanced force when a decision ․ interprets a statute” because “unlike in a constitutional case, critics of our ruling can take their objections” to Congress, which “can correct any mistake it sees.” Kimble v. Marvel Ent., LLC, 576 U.S. 446, 456 (2015); see also Bryan A. Garner et al., The Law of Judicial Precedent 333 (2016) (“Stare decisis applies with special force to questions of statutory construction. Although courts have power to overrule their decisions and change their interpretations, they do so only for the most compelling reasons — but almost never when the previous decision has been repeatedly followed, has long been acquiesced in, or has become a rule of property.”). We are guided by decades of unbroken controlling precedent suggesting that Section Two implies a private right of action, and we see no congressional effort to course correct. Accordingly, we think “statutory stare decisis counsels our staying the course.” Allen, 599 U.S. at 39.

The Supreme Court has “identified several factors to consider in deciding whether to overrule a past decision, including ․ the workability of the rule it established ․ and reliance on the decision.” Knick v. Twp. of Scott, 588 U.S. 180, 203 (2019) (internal quotation marks omitted) (quoting Janus v. State, Cnty., & Mun. Emps., 585 U.S. 878, 917 (2018)). Allowing private plaintiffs to bring Section Two claims has proven to be a workable rule—having gone unquestioned for decades in multiple Supreme Court decisions. In fact, the ability of private parties to bring Section Two claims has become “the sort of stable background rule that fosters meaningful reliance.” Loper Bright Enters. v. Raimondo, 144 S. Ct. 2244, 2272 (2024) (internal quotation marks and citation omitted). There has been no “tinkering” with the ability of private parties to bring Section Two claims by the Supreme Court, lower courts (with one, lone exception), or Congress. Id. And because “Congress has spurned multiple opportunities to reverse” the Supreme Court's and lower courts’ treatment of private-plaintiff Section Two actions, we think “a superspecial justification” would be necessary to reverse course, and we see none here. Kimble, 576 U.S. at 456, 458.

In its earlier motion, the State distinguished statutory stare decisis arguments on the same ground it attacks any suggestion of congressional ratification: it asserts that because the Supreme Court has not definitively decided the issue, the doctrine does not apply. See Milligan Doc. 331 at 23. We see the point and have taken care to rest our ruling on the statutory text. But we reject the argument that we must otherwise close our eyes to Congressional intent. The federal courts (including the Supreme Court) have consistently and uniformly allowed private plaintiffs to enforce a high-profile congressional enactment for nearly sixty years, and we see no indication in any congressional record that Congress believes all of that (or any of it) was mistaken.

***

In our view, the text of Section Two compels the conclusion that private plaintiffs may enforce it, either through an implied private right of action, Section 1983, or both. And other doctrines confirm our understanding of the text. It is difficult in the extreme for us to believe that for nearly sixty years, federal courts have consistently misunderstood one of the most important sections of one of the most important civil rights statutes in American history, and that Congress has steadfastly refused to correct our apparent error.

C. Section Two Is Constitutional

We reject the State's argument that “race-based redistricting justified by § 2 no longer passes Constitutional muster today,” Milligan Doc. 481 at 221, ¶ 597, for two reasons. First, by its own terms, Section Two is always rooted in today's circumstances – the totality of them. See 52 U.S.C. § 10301(b). The Senate Factors instruct us to consider some historical patterns and circumstances, but only insofar as they drive present patterns and circumstances, are on repeat, or otherwise tell us something of value about whether individuals are disabled from meaningful political participation at the present time.

It would be erroneous for us to grant Section Two relief solely based on historical patterns and circumstances, and we do not do so here. We have taken great care not to allow the terrible backdrop of Alabama's history of official racial discrimination to dictate the outcome of these cases, and we have attended to the State's presentation about present circumstances in Alabama at length and in comprehensive detail. In this regard, “what [we] did here is essentially no different from what many courts have done for decades under [the Supreme] Court's superintendence[.]” Allen, 599 U.S. at 90–91 (Thomas, J., dissenting). We cannot hold that the mere passage of time renders unconstitutional a statute that, on its face and by its terms, accounts for that passage of time.

Second, we reject the State's argument because the record compels us to conclude that even if Congress's justifications for Section Two may expire at some point, we are certainly not yet there. There can be no serious dispute that Alabama has made substantial progress away from its terrible history of official discrimination. But based on the evidence in these cases, much of which is not contested, there also can be no serious dispute that that terrible history still significantly affects the ability of many Black Alabamians, particularly in the rural Black Belt, to participate in the democratic process. Having heard testimony about extreme destitution and steep illiteracy rates among Black Alabamians in the rural Black Belt, and personal experiences in segregated public schools and spaces from Black leaders of those communities, we cannot hold that Alabama has yet outrun the effects of its past.

And as we have said previously and explain below, we were and are disturbed by the Legislature's deliberate decision to refuse the remedy we said was required. It has been nearly sixty years since the Alabama Legislature last purposefully refused to satisfy the requirements of a federal court order about redistricting, particularly one affirmed by the Supreme Court. In our view, that refusal precludes a finding that Congress's original justifications for Section Two no longer apply in today's Alabama.

We see clearly the practical realities. It is not lost on us that the Legislature's decision not to provide the required remedy came in Alabama's first opportunity to redistrict its congressional map free of federal preclearance. Nor is it lost on us that Section Two protects what few opportunities Black Alabamians have to select a candidate of their choice.

Nor is it lost on us that the Legislature's deliberate decision not to satisfy our order about Section Two was an avoidable and self-inflicted wound: although state legislatures may sometimes face the “competing hazards of liability” created by Section Two and the Equal Protection Clause, Abbott, 585 U.S. at 587 (quoting Vera, 517 U.S. at 977); see Louisiana v. Callais, No. 24-109 (U.S.), the Alabama Legislature was not facing that risk in 2023, after we and the Supreme Court had held that in these Section Two cases, the record contained lawful remedial maps. Nor is it lost on us that we have been warned by the State that in its view, if we enjoin the 2023 Plan under Section Two, we will be right back where we started: with the Legislature free to enact another plan that contains only one Black-opportunity district, on the reasoning that we may then (again) reject it.

In these circumstances, we cannot see how Section Two has outlived the purpose Congress intended. Accordingly, even if Section Two “cannot extend indefinitely into the future,” Allen, 599 U.S. at 45 (Kavanaugh, J., concurring in part), we hold that it extends at least past today.

VI. ANALYSIS – CONSTITUTIONAL CLAIMS

A. The Singleton Plaintiffs’ Arguments

The Singleton Plaintiffs raised two constitutional claims. Both claims focus on Jefferson County, which includes Districts 6 and 7.

1. Racial Gerrymandering

The Singleton Plaintiffs claim that the 2023 Plan is racially gerrymandered, in violation of the Equal Protection Clause and Article I, § 2 of the Constitution. Singleton Doc. 229 ¶ 67. They allege that the violation originated in the 1992 map that resulted from Wesch, which they say was a racial gerrymander because it split seven counties expressly “for the purpose of drawing one majority-Black district” that was “packed at 67.53% Black.” Id. ¶¶ 22, 25. They allege that “Alabama continued the 1992 racial gerrymander in the Congressional redistricting plans enacted after the 2000 and 2010 censuses,” which left District 7 “packed at 63.57% Black.” Id. ¶ 27. And they allege that the State conceded that the plan enacted in 2011 was racially gerrymandered, and “the Legislature preserved the racial gerrymander” when it enacted the 2021 Plan and the 2023 Plan. Id. ¶¶ 46, 55.

The Singleton Plaintiffs allege that “District 7 contains about 54% of Jefferson County's population, but more than 71% of its Black population, resulting in a thirty-point gap between the proportion of the population that is Black inside and outside the district (57% inside, compared to 27% outside).” Id. ¶ 55. They also allege that “[t]his is no accident: District 7 sharply separates majority-Black Birmingham from the relatively White ‘Over the Mountain’ suburbs like Mountain Brook and Vestavia Hills.” Id. They say that “[w]ith 2020 census data, it is practicable to end the 1992 racial gerrymander and draw a seven-district Congressional plan without splitting a single county and with only slight population deviations.” Id. ¶ 39. They offer their Whole County Plan as such a plan. And they claim that the Legislature acted with discriminatory purpose when it enacted the 2023 Plan that “intentionally perpetuates the unconstitutional racial gerrymandering of Jefferson County.” Id. ¶ 2.

2. Intentional Discrimination

The Singleton Plaintiffs also claim that the State violated the Fourteenth and Fifteenth Amendments by “enacting and carrying out a legislative plan for Alabama's congressional districts that intentionally discriminates against Black Alabamians.” Milligan Doc. 445 at 9 (pretrial order). They say that the drafters of the 2023 Plan “intentionally d[rew] Congressional District lines in order to destroy otherwise effective crossover Districts.” Singleton Doc. 229 ¶ 75. They allege that the drafter excluded Jefferson County from the communities of interest the 2023 Plan protects because it “is the one county in the state with a proven record of effective and persistent biracial politics.” Id. ¶ 64.

They further assert that the drafter “knew that White voters in Jefferson County are more likely to share the equal rights and progressive political agenda of Black voters than do White voters in the Wiregrass.” Id. ¶ 65. They allege that this is why the 2023 Plan “places Black voters in the eastern Black Belt in the same district with the Wiregrass counties, ensuring they would have no opportunity to elect a candidate of their choice.” Id. And they allege that by splitting Jefferson County and the Black Belt, the 2023 Plan “perpetuates Alabama's policy since Reconstruction of creating and maintaining election systems that are designed to encourage White electoral solidarity.” Id. ¶ 66.

At trial, the Singleton Plaintiffs offered two experts (Dr. Riser and Dr. Frederickson) and two lay witnesses (Ms. Slay, who testified live, and Senator Smitherman, who testified by deposition) to support these claims. In their proposed order, they also rely on the trial testimony of Senator Singleton. See Singleton Doc. 320. We already have discussed some of this testimony. See supra Part IV.C.

B. The Milligan Plaintiffs’ Arguments

The Milligan Plaintiffs assert one constitutional claim, which focuses on the Districts in the Black Belt and Gulf Coast and asserts that the 2023 Plan violates the Fourteenth Amendment because it intentionally discriminates against Black Alabamians. Milligan Doc. 445 at 9 (pretrial order). They allege that the 2023 Plan is “Alabama's latest discriminatory scheme, designed with the intent to crack Black voters into congressional districts in a manner that prevents the creation of two congressional districts in which Black voters have an equal opportunity to elect candidates of their choice.” Id. They claim that this racially discriminatory purpose motivated both the drawing and passage of the 2023 Plan. Milligan Doc. 329 ¶ 199. They allege that the 2023 Plan “was drafted and passed at least in part to minimize the political power of Black Alabamians by limiting their ability to influence congressional elections to a single district,” and to the part of the state that contains District 7, “despite substantial clusters of Black Alabamians living in concentrated areas of the State outside of [District 7].” Id. ¶¶ 199–200.

The Milligan Plaintiffs further claim that the “2023 Plan intentionally perpetuated the discriminatory effects of the 2021 Plan.” Milligan Doc. 445 at 9. They allege that in 2023, “the Legislature ignored this Court's orders and the repeated requests from Black legislators to draw two majority-Black, or opportunity, districts.” Milligan Doc. 329 ¶ 204. They allege that the State was “aware that Black Alabamians could elect candidates of their choice in two congressional districts in the state in a manner that complies with the U.S. Constitution and federal law, yet purposefully drew the congressional maps to prevent this.” Id. ¶ 200.

To support these allegations, the Milligan Plaintiffs rely on testimony from Dr. Bagley, Representative Pringle, Senator Livingston, Mr. Hinaman, Dr. Hood, Ms. Dowdy, Mr. Milligan, Mr. Clopton, and Dr. Duchin. See Milligan Doc. 485 ¶¶ 757, 780, 787, 810, 846, 847, 851, 875.

In their proposed order, the Milligan Plaintiffs discuss each part of the Arlington Heights test.

1. Historical Background of the 2023 Plan

The Milligan Plaintiffs begin with Alabama's history of discriminatory redistricting. See id. at 287–93. They argue that although we should not automatically assign the invidious intent of prior legislatures to the Legislature, under controlling precedent, the historical patterns “are relevant to the extent that they naturally give rise to—or tend to refute—inferences regarding the intent of” the current Legislature. Abbott, 585 U.S. at 607.

The Milligan Plaintiffs identify two salient histories: first, they argue that the Legislature's current insistence that Mobile and Baldwin Counties be kept whole and together in a majority-White district is “part of [a] historical pattern of manipulating the pairing or separation of these two counties for discriminatory reasons.” Milligan Doc. 485 ¶¶ 758, 760–76. And second, they argue that the Legislature's insistence on a majority-White district on the Gulf Coast is “part and parcel” of a broader, “continuous pattern of whichever party is in power targeting Black voters.” Id. ¶ 759. They agree with a public comment by Representative England to the effect that the Legislature's response to our order “proved that Alabama was still the ‘make me’ state when it came to affording Black citizens their rights.” Id. ¶ 757 (citing Milligan Doc. 385-1 at 2, 28).

The Milligan Plaintiffs discuss at length the Legislature's historical treatment of the Gulf Coast counties in congressional districting. They argue that although the Legislature now insists that Mobile and Baldwin Counties must be kept whole and together because they “comprise a well-known and well-defined community with a long history and unique interests,” “this has not always been the case.” Id. ¶ 760. They trace Dr. Bagley's exposition of the historical evidence that “the separation of Baldwin and Mobile counties in the 1870s and the unification of these counties in the 1970s were substantially motivated by race.” Id. ¶ 761.

Dr. Bagley reported that during Reconstruction, Alabama had two majority-Black congressional districts: Districts 1 and 2, based in the Black Belt and Mobile County. Id. ¶ 762. Three Black congressmen won elections from these districts, one each in 1870, 1872, and 1874. Id. ¶¶ 762–63. Dr. Bagley reported that these elections triggered a backlash from White legislators who set out to redraw the districts. Id. ¶ 764. In 1875, they created a “shoestring” district to pair two Black incumbents. Id. ¶ 765. That district split Baldwin County (which was then approximately half Black) and united Mobile County with several Black Belt counties. Id. It also eliminated one of Alabama's majority-Black districts. Id. The Milligan Plaintiffs discuss Dr. Bagley's explanation that the shoestring district was “widely understood as being done for the purpose of diluting the Black vote.” Id. ¶ 766. And they cite his analysis that from 1875 to 1970, with one brief exception, the Legislature separated Mobile and Baldwin Counties and continued to unite the City of Mobile with the western Black Belt. Id. ¶ 767.

The Milligan Plaintiffs then discuss Dr. Bagley's analysis of the 1972 Plan, in which the Legislature united Mobile and Baldwin Counties in one district. Id. ¶¶ 772–74. They rely on Dr. Bagley's explanation that in the 1972 Plan, the Legislature “did not act to unite a purported community of interest,” but acted “for racial reasons and with racial effects.” Id. ¶ 774. According to Dr. Bagley, the 1972 Plan reduced the BVAP in Districts 1, 2, and 3 “from around 40% Black under the prior plan to only around 30 percent in all three districts,” and the Legislature's decision to unite Mobile and Baldwin Counties “together in a way that significantly dropped the Black population in these districts occurred just as Black people began registering to vote” after the passage of the Voting Rights Act. Id.

The Milligan Plaintiffs thus argue from Dr. Bagley's testimony that “[i]n maintaining the split of southern Alabama into three congressional districts since 1972, pairing Mobile and Baldwin counties, the Legislature continued to crack the Black vote to prevent the election of a Black Congressperson,” id. ¶ 775, and that the 2021 Plan and 2023 Plan followed this pattern, id. ¶ 776.

2. Sequence of Events Leading to the Passage of the 2023 Plan

The Milligan Plaintiffs next argue that the events leading to the passage of the 2023 Plan evince discriminatory intent, and they discuss those events in great detail. Milligan Doc. 485 at 294–315. The Milligan Plaintiffs rely largely on stipulated facts, the Legislators’ testimony, and Mr. Hinaman's testimony. First, they cite Representative Pringle's and Senator Livingston's testimony that they understood what was required by Section Two and our preliminary injunction. Id. ¶¶ 780–86.

Second, they cite Mr. Hinaman's testimony that he understood he was tasked with drawing a second opportunity district, and that such a district was one where Black-preferred candidates “had a ‘50/50’ chance of winning an election.” Id. ¶¶ 790–91. They also cite Mr. Hinaman's testimony that no one instructed him to try to add a second majority-Black district, and he did not attempt to do so. Id. ¶ 792. And they cite his testimony that he drew three plans for the Committee chairs and understood that they preferred the Community of Interest Plan and would sponsor it in their respective chambers. Id. ¶¶ 793–96.

Third, the Milligan Plaintiffs discuss the Committee's pre-session hearings. Id. ¶¶ 797–805. They point out that the only plans available for public input at those hearings were plans prepared by the Plaintiffs, and that Representative Pringle “said the Community of Interest Plan was not yet done.” Id. ¶¶ 800–01. They recall that although Representative Pringle invited a historian to testify at one of the hearings about the historical connections between Mobile and Baldwin Counties, he “did not ask anyone to speak on behalf of the need for two districts in which Black voters could elect candidates of their choice.” Id. ¶ 802. And they describe his rejection of an amendment to the 2023 guidelines offered by Representative England (who is Black), about how to remedy the likely Section Two violation we found. Id. ¶ 803.

Next, the Milligan Plaintiffs argue that when the Committee passed the Community of Interest Plan, it ignored the objections of Black Legislators. Id. ¶¶ 806–15. The Milligan Plaintiffs argue (as is stipulated) that when Representative Pringle introduced the Community of Interest Plan in the Committee, he said that it preserved the cores of existing congressional districts. Id. ¶ 807. They also argue (as is stipulated) that Dr. Hood's performance analysis of that Plan showed that the Black-preferred candidate would have won two of four modeled races in District 2:

Id. ¶ 808; Milligan Doc. 436 at 19 (stipulations). The Milligan Plaintiffs argue that Dr. Hood's analysis was flawed because two of the four elections he analyzed involved unusually popular White Democratic candidates, Milligan Doc. 485 ¶¶ 811–12, but that Mr. Hinaman trusted the analysis anyway, id. ¶ 813. And they observe that every Black member of the Committee voted against the Community of Interest Plan. Id. ¶ 815.

Next, the Milligan Plaintiffs argue that Senators “turned against” the Community of Interest Plan to “pursue a more aggressive plan for preserving power at Black voters’ expense.” Id. ¶¶ 816–26. The Milligan Plaintiffs describe the “additional information” that Senator Livingston discussed in his deposition, its mysterious and allegedly unknown provenance, and the Opportunity Plan (in which District 2 had a BVAP of 38.31%) that Senator Livingston introduced in Committee and was drafted by Mr. Brown and his political consulting firm, Red State Strategies. Id. ¶¶ 816, 819. They cite text messages between Senator Livingston and Mr. Brown that (1) discuss the possibility of a higher BVAP (Mr. Brown asked Senator Livingston if a BVAP of 41.6% would “work”), and (2) contain a reference by Senator Livingston to “monkey town,” which the Plaintiffs assert is a racist nickname for Montgomery. Id. ¶ 820. The Milligan Plaintiffs make the point that according to Senator Livingston's testimony, when his Opportunity Plan (also known as the Livingston 2 Plan) passed the Senate, he had no belief one way or the other about whether it contained two opportunity districts. Id. ¶¶ 821–22. And they point out that according to Representative Pringle's testimony, he also did not have a view on that issue. Id.

The Milligan Plaintiffs argue that even though the configuration of District 2 in the Opportunity Plan was identical to its configuration in the predecessor plan (the Livingston 1 Plan, which was only slightly adjusted to become the Livingston 2 Plan), Senator Livingston testified that he believed that the Opportunity Plan provided a better opportunity for Black Alabamians to elect a representative of their choice than the Livingston 1 Plan provided. Id. ¶ 824. They point out that Senator Livingston had no explanation for this belief. Id.

The Milligan Plaintiffs cite Representative Pringle's refusal to pass the Opportunity Plan in the House, or to substitute it for the Community of Interest Plan, or to attach his name to it. Id. ¶ 825. And they describe that each plan passed its respective chamber entirely along racial lines, with one exception in the House (Representative Paschal – the sole Black Republican). Id. ¶ 826.

The Milligan Plaintiffs next argue that the Alabama Solicitor General worked with Senator Livingston and other Senators to draft the Livingston Plans, including the compromise plan that ultimately became the 2023 Plan. Id. ¶ 828. The Milligan Plaintiffs point out that when Senator Livingston was asked about the decision to draw District 2 with a BVAP under 40 percent in SB5, Senator Livingston said only that “this is the plan that was brought forward in the end and was compromised upon.” Id. ¶ 830. And that when Representative Pringle was asked about that decision, he responded: “You're going to have to talk to Senator Livingston and [the Solicitor General],” and “[t]hat's what the senate came up with, and they were not going to allow us to pass the house plan.” Id. ¶ 831.

Ultimately, the Milligan Plaintiffs say, the Legislature “knew that the 2023 Plan lacked a second opportunity district” and that the changes the Senate made to the Community of Interest Plan “would harm Black-preferred candidates’ chances.” Id. at 307. The Milligan Plaintiffs explain (as is stipulated) that before the Legislature passed the 2023 Plan, it conducted a performance analysis that modeled seven elections and predicted Black-preferred candidates would lose every election in District 2:

Id. ¶ 833; Milligan Doc. 436 at 21 (stipulations).

To support their assertion about the Legislature's knowledge, the Milligan Plaintiffs cite three pieces of evidence. First, they cite Representative Pringle's testimony that he saw the performance analysis before the Legislature voted on the 2023 Plan and that it was available to all members of the conference committee. Milligan Doc. 485 ¶ 834. Second, they explain (as is stipulated) that Representative England told legislators before the final vote that in his view, SB5 was noncompliant with our order and we would reject it. Id. ¶ 835. Third, they cite Mr. Hinaman's testimony that at least some legislators knew from Dr. Hood's earlier performance analysis of the Community of Interest Plan that without Dallas County (home to Selma) in District 2, Black-preferred candidates would have no chance of winning that District. Id. ¶ 837. According to Mr. Hinaman, the Black-preferred candidate lost every election Dr. Hood modeled in that District, if Selma was not in the District. Id. To explain the importance of Dallas County to District 2, the Milligan Plaintiffs describe the well-known history of political mobilization in Selma and explain that it is either the residence or hometown of several prominent Black Alabamians. Id. ¶ 838.

The Milligan Plaintiffs next turn to the 2023 legislative findings, which they assert the Solicitor General drafted and inserted into SB5. Id. at 309. They make several arguments from the findings:

• They identify several differences between the findings and the 2023 Committee guidelines, which were separated by only a week's time. Id. ¶ 839; see also supra Part I.D.

• They observe that in the findings, the Legislature described some traditional districting principles as “non-negotiable,” which the Legislature had never done before, and that neither compliance with the Voting Rights Act, nor nondilution of minority voting strength, were included on that list. Milligan Doc. 485 ¶ 840.

• They observe that in the findings, the Legislature recognized only three communities of interest in the state and removed from the definition of “community of interest” in the 2023 guidelines any reference to shared “ethnic, racial, tribal, [or] social ․ identities.” Id. ¶ 842.

• They observe that the findings reference the “French and Spanish colonial heritage” or “culture” of Mobile and Baldwin Counties four times in one paragraph but make no mention (anywhere) of the culture or heritage of the Black Belt, Mobile, or any other community of interest anywhere in Alabama. Id. ¶ 844. They contend that the reference to “French and Spanish colonial heritage” is a reference to White people. Id. ¶ 845.

• They observe that the findings were not requested by the Committee chairs, who did not know why they were in the bill and did not see them until the morning of the vote. Id. ¶¶ 848–50.

• They observe that Mr. Hinaman was never given the instructions in the findings when the Legislators asked him to draw maps. See id. at 30 (citing Milligan Doc. 459-7 at 94–95).

• They observe that the findings do not mention an additional opportunity district. See id. ¶ 874.

• And they observe that the findings eliminated the reference in the Committee guidelines to the nondilution of minority voting strength, and that the findings do not mention Black people at all. See id. at 320.

3. Substantive and Procedural Departures from the Norm

Next, the Milligan Plaintiffs discuss the evidence of substantive and procedural departures from the norm during the 2023 Special Session. Here again, they rely principally on stipulated facts, the Legislators’ testimony, and Mr. Hinaman's testimony:

• They observe that “[t]he Committee Co-Chairs failed to present any of their plans for input at the public hearings.” Id. ¶ 856.

• They observe that despite the Committee chairs asking Mr. Hinaman to draft the plans for consideration, “Senator Livingston had been communicating (apparently without knowledge of his Co-Chair, Committee counsel, or Mr. Hinaman), with another mapdrawer, Chris Brown.” Id. ¶ 857. Accordingly, the Milligan Plaintiffs contend that Senator Livingston's testimony that he was unsure why the Senate suddenly backed another plan (the one Brown was drafting) “blinks reality” and “raises questions about the import of then-U.S. House Speaker Kevin McCarthy's call to Senator Livingston asking him to consider his thin Republican majority in the House.” Id. ¶¶ 858–59.

• They contend that “the only two criteria that Mr. Brown raised with Senator Livingston” were the BVAP in his plan and whether a White incumbent (Congressman Moore) would still be able to beat a Black-preferred candidate. Id. ¶ 860. And that “[g]iven the use of racial targets in terms of BVAP, this either suggests that race was the real motive, or that Sen. Livingston was being untruthful about using race for partisan ends.” Id. ¶ 859.

• They assert that “the involvement of the Solicitor General both as a mapdrawer and drafter of legislative purpose also reveals a bizarre procedural and substantive departure.” Id. ¶ 861.

• They observe that “no redistricting bill in Alabama history contained similar legislative findings,” and that the Legislators and Mr. Hinaman “pointed to the unprecedented nature of the findings.” Id. ¶ 862. They cite Representative Pringle's testimony that the findings “were not debated by the Legislature and were not revealed until the members were asked to vote on the bill[.]” Id. ¶ 863. Likewise, Mr. Hinaman testified that he was never given the instructions in the findings when the Legislators asked him to draw the maps. Id. ¶ 864. And “Senator Livingston was also unaware of why the legislative findings were included.” Id. ¶ 865.

• And they argue that the findings reflect several substantive departures from the norm of redistricting in Alabama:

• They recite Representative Pringle's agreement that “some of [the Legislative findings] look like they are” in conflict with the guidelines adopted the week before. Id. ¶ 868.

• They reiterate that “the findings excluded the statement from the 2021 and 2023 guidelines that ‘[a] redistricting plan shall have neither the purpose nor the effect of diluting minority voting strength.’ ” Id. ¶ 869.

• They observe that the 2023 Plan enumerated communities of interest for the first time, and that it enumerated only three communities statewide: the Black Belt, the Gulf Coast, and the Wiregrass. Id. ¶ 870.

• They reiterate that the findings “altered the Guidelines’ definition of ‘community of interest’ to remove from the definition shared ‘ethnic, racial, tribal, social ․ identities,’ and add similarity of ‘transportation infrastructure, broadcast and print media, educational institutions.’ ” Id.

• They observe that “[w]hile several pages of findings are devoted to linking Mobile and Baldwin, including reference to its shared ‘French and Spanish colonial heritage,’ only five lines are provided to the Black Belt.” Id. ¶ 871.

4. Direct Evidence of Intent and Other Contemporaneous Statements

Finally, the Milligan Plaintiffs argue that the record contains direct evidence of discriminatory intent and other contemporaneous statements that suggest a discriminatory intent. See id. at 318–26. They identify the “unprecedented” 2023 legislative findings as direct evidence, and they repeat many of their earlier observations about the findings. See id. They recite Dr. Duchin's testimony that the findings “come close to prescribing” a majority-White congressional district in Mobile and Baldwin Counties and make it mathematically impossible to draw an additional Black-opportunity district, and they cite the State's counsel's concession to the same effect. See id. ¶¶ 879–82. The Milligan Plaintiffs argue that this evinces a purposeful effort by the Legislature to prevent the creation of an additional Black-opportunity district and require the dilution of Black votes. See id.

The Milligan Plaintiffs also rely on statements by the State's lawyers as indicative of the State's purpose to discriminate. Id. ¶¶ 885–93. They argue that the Solicitor General's arguments in court, as well as statements by other counsel for the State at trial, confirm that the Legislature's intent was to avoid creating an additional opportunity district, and its effect was to prevent the creation of an additional opportunity district. Id. ¶¶ 886–87.

The Milligan Plaintiffs claim that the Community of Interest Plan disproves the Solicitor General's argument in court that the 2023 Plan was “as close as you are going to get to a second majority-Black district” without violating traditional districting principles. Id. ¶¶ 890–92. The Milligan Plaintiffs do not concede that the Community of Interest Plan would remedy the likely Section Two violation we found, but they argue that it would have provided Black Alabamians greater opportunity than the 2023 Plan provides. Id.

Ultimately, the Milligan Plaintiffs argue that “White legislative leaders knew that absent a majority-minority district a Black candidate would lose and [W]hite voters would elect a [W]hite Republican instead” in District 2. Id. ¶ 894. They say that despite this understanding, the Legislature never even attempted to draw that district. Id. ¶ 895. And they point to Speaker Ledbetter's explanation why: to have a “good shot” at changing the mind of one Justice on the Supreme Court. Id. ¶ 897.

C. The State's Arguments

1. Singleton Plaintiffs’ Racial Gerrymandering Claim

The State argues that the Singleton Plaintiffs’ claim that the 2023 Plan is a racial gerrymander is unfounded because they “have not shown that the ‘statistical disparities’ are so stark that they are ‘tantamount for all practical purposes to a mathematical demonstration’ that the State acted with a discriminatory purpose.” Milligan Doc. 481 ¶ 885. It argues that “Plaintiffs have not attempted to demonstrate that dividing Jefferson County where the Legislature chose to is ‘unexplainable on grounds other than race.’ ” Id. ¶ 886. It also argues that “Plaintiffs have not proffered ‘an alternative map that would have allowed the State to achieve its districting goals’ ‘with greater racial balance.’ ” Id. ¶ 887.

2. Singleton and Milligan Plaintiffs’ Intentional Discrimination Claims

The State argues that the intentional discrimination claims of both the Singleton and Milligan Plaintiffs fail because they failed to establish that “the Legislature relied on race for an invidious reason: to harm a racial group's ability to elect the group's preferred candidates.” Milligan Doc. 481 ¶ 631. The State argues that “Plaintiffs ‘cannot prove this invidious reason merely by showing that the legislature knew that the revised map would have such harmful effects on the racial group.’ ” Id. ¶ 632. The State asserts that Plaintiffs must prove that “the [L]egislature must have drawn the map ‘because of, not merely in spite of, those adverse effects.’ ” Id.

The State claims that the Singleton and Milligan Plaintiffs fail to make this showing with direct or circumstantial evidence, and that we must presume the Legislature's good faith. Id. ¶¶ 636, 641. To support its argument about an evidentiary deficit, the State contends that the Plaintiffs “have not presented a single ‘express acknowledgement,’ ‘confession,’ ․ or statement showing that any Legislator who voted for the 2023 Plan ‘did so for a racist reason.’ ” Id. ¶ 656. The State argues that “the Alabama Legislature never exalted any ‘[W]hite community’ as ‘more important’ than a ‘cohesive [B]lack community,’ ” id. ¶ 666, and it “place[d] Mobile and Baldwin together [not] to make [District 1] ‘more Anglo,’ ” but to “respect ․ the counties’ shared cultural and economic ties, which are felt by Gulf Coast residents of all races,” id. ¶ 667.

To support its argument about good faith, the State takes two approaches: it challenges the Plaintiffs’ arguments about the Legislature's failure to satisfy the requirements of our preliminary injunction, and it defends the 2023 legislative findings as having been made in good faith. We consider each argument in turn.

a. Intentional Refusal to Satisfy Requirements of Federal Court Orders

The State argues that the Legislature acted in good faith when it enacted the 2023 Plan because the 2023 Plan did not defy our preliminary injunction. The State describes the Plaintiffs’ argument in this regard as an assertion that the 2023 Plan “defies this Court and violates the Equal Protection Clause because the Legislature enacted the plan after the Court's 2022 preliminary injunction order without including two districts likely to favor Democrats.” Id. at 14.

The State makes four points. First, the State argues that “this Court barred the Secretary from using the 2021 Plan; it did not order the Legislature to enact any particular map,” so “[t]he Court's order was not violated.” Id. Second, the State argues that it actually “respect[ed]” our order because the Legislature “did not simply pass the same plan with minor tweaks and try to evade federal court review,” and “the law was passed in time for this Court to preliminarily assess it before the 2024 elections.” Id.

Third, the State argues that preliminary injunctions are not final, and it “makes no sense to fault the State for curing the alleged ‘inconsistent treatment’ in [the 2021 Plan] ․ before going to trial.” Id. at 14–15. And fourth, the State argues that even if the Plaintiffs prevail, “the suggestion” that the law is “so clear” that any interpretation other than the Plaintiffs’ “could only be driven by racial animus is as baseless as it is divisive.” Id. at 15. On this last point, the State argues that it is “perfectly plausible” that the “Legislature believed it could remedy a likely § 2 violation by eliminat[ing] racial disparities – i.e., inconsistent treatment of the Gulf Coast and the Black Belt,” or that “Justice Kavanaugh was right to question whether” race-based redistricting can continue. Id. (internal quotation marks and citations omitted). The State reiterates that even if its legal arguments fail, “it is neither defiant nor racist to enact a law based on them and then test them in federal court.” Id.

b. The 2023 Legislative Findings

The State also challenges the Plaintiffs’ argument that the 2023 legislative findings rebut the presumption of legislative good faith. The State argues that the findings state the Legislature's intention to comply with the Voting Rights Act, and that the elimination of the provision about nondilution of minority voting strength simply resulted in an “absence of superfluous language.” Id. ¶¶ 778–79.

The State argues that we should not infer bad faith “from the Legislature's decision to include findings about the Black Belt, Wiregrass, and Gulf Coast, but not about communities lying elsewhere in the State.” Id. ¶ 668. It asserts that “[t]he ‘obvious alternative explanation’ ․ for this decision is that the 2021 Plan had been enjoined under § 2 in large part because of the Legislature's ‘inconsistent treatment’ of Black and White Communities.” Id. It argues that “[t]he 2021 Plan's treatment of communities north of Montgomery was not at issue, so the 2023 Legislature had no reason to articulate its intent with respect to those communities.” Id. ¶ 669.

And the State argues that the reference in the legislative findings to the shared “French and Spanish colonial heritage” in the Gulf Coast was not a reference to White people, but a reference to “cultural influences in a region that flow from its unique history,” which is not improper. Id. at 13.

c. Alternative Explanations for the 2023 Plan

The State next argues that “there are at least two obvious and broad reasons for the 2023 Plan other than racial animus: (1) avoiding a racial gerrymandering suit; and (2) achieving partisan goals.” Id. ¶ 815. To support the argument about a racial gerrymandering suit, the State claims that the “Legislature could have been aware that Plaintiffs’ expert, Dr. Imai” ran simulations in the preliminary injunction proceedings, and could have been aware of those results. Id. ¶ 825. The State cites no evidence that any legislator is aware of Dr. Imai's existence, let alone his simulations or testimony. See id.

The State argues that “[t]here's an obvious ‘damned if you do, damned if you don't’ nature to this endeavor” as the State “navigate[s] the precarious waters of vote dilution jurisprudence while avoiding racial gerrymandering claims.” Id. ¶ 841. And the State points out that the 2023 Plan, despite having a larger BVAP in District 2, drew a racial gerrymandering lawsuit (Singleton). Id. ¶ 828. But see Singleton Doc. 288 at 10–11 (describing Singleton gerrymandering claim as relating to Jefferson County, which is in Districts 6 and 7, not 1 and 2).

The State further argues that “at least eight Justices in Allen agreed that race cannot predominate in an illustrative map, while only four expressly held that race did not predominate in some of Mr. Cooper's maps,” and both concurring and dissenting Justices “questioned the constitutionality of continued race-based redistricting” under Section Two. Id. ¶ 833; but see Allen, 599 U.S. at 9, 17, 19, 21, 23 (a majority of Justices repeatedly affirming our ruling).

Thus, the State reasons, “[t]he Legislature's good faith belief that drawing a second majority-[B]lack district would unconstitutionally segregate voters based on race, and that ‘the authority to conduct race-based districting cannot extend indefinitely into the future,’ ․ are obvious justifications for the 2023 Plan other than invidious discrimination.” Id. ¶ 834.

To support its argument about partisan goals, the State asserts that the Plaintiffs “fail to disentangle race from the obvious alternative explanation of ‘securing partisan advantage.’ ” Id. ¶ 842. The State asserts that “finding legitimate ways to avoid the adoption of a map that would likely swing an additional congressional district to Democrats is a reasonable (and obvious) non-racial goal for Republican legislators to pursue,” as is the protection of Alabama's incumbent Representatives (including Congresswoman Sewell). Id. ¶¶ 847–48, 851. The State warns that we will clearly err if we simply “credit[ ] the less charitable conclusion that the legislature's real aim was racial.” Id. ¶ 854 (quoting Alexander, 602 U.S. at 22).

Finally, the State argues that even if the Singleton and Milligan Plaintiffs establish a prima facie case of intentional discrimination, their claim still fails because the 2023 Plan would have been enacted for two independent reasons: “so that Alabama's congressional delegation would retain its six to one Republican to Democratic composition,” id. ¶ 859, and “as an attempt to avoid a racial gerrymandering suit,” id. ¶ 860.

d. Singleton-only Rebuttal on Intentional Discrimination

Separately, the State argues that the Singleton Plaintiffs’ intentional discrimination claim fails because the Legislature's refusal to adopt their Whole County Plan was not the product of intentional discrimination. Milligan Doc. 481 ¶¶ 651–55. The State asserts that “the Singleton Plan does not achieve minimum population deviation and departs from the 2023 Plan's lines,” id. ¶ 652; that “[t]he ‘obvious alternative explanations’ of population equality and core retention better explain the Legislature's decision not to enact the Singleton Plan than an invidious racial motive,” id.; and that “[s]howing the availability of crossover districts does not establish the preconditions necessary for a claim of vote dilution,” because it “fails to establish discriminatory effects,” id. ¶ 654.

e. Milligan-only Rebuttal on Intentional Discrimination

Finally, the State argues that the Milligan Plaintiffs fail to establish intentional discrimination under Arlington Heights. Id. at 244. First, the State argues that “[p]ast discrimination is not evidence of a present-day intent to discriminate.” Id. at 245–53. The State asserts that these Plaintiffs cannot establish “that the 2023 Legislature intended to harm Black voters by ratifying elements of previous plans themselves imbued with racially discriminatory motives” because the “absence of any evidence that any member of the [Alabama] Legislature, much less a majority of its members, was actually motived by racial discrimination in passing the Enacted Map dooms the [P]laintiffs’ case for ratification.” Id. ¶¶ 686, 693. The State argues that Dr. Bagley's recitation of Alabama's history of discriminatory redistricting is “unconnected to the passage of the actual law in question” and “largely irrelevant,” and they describe it as a “time-traveling ‘cat's paw’ theory.” Id. ¶¶ 677–79.

Second, the State argues that “[t]he sequence of events leading to the passage of the 2023 Plan does not lead to the inference of discriminatory intent” because the State “complied with the orders of the Court, which were to refrain from using the 2021 and 2023 Plans in any election while those plans were enjoined.” Id. ¶¶ 707, 709. The State argues it did not “seek to evade this Court's jurisdiction.” Id. ¶ 711. The State observes that “the Parties were in the preliminary injunction phase” and argues that it was entitled to a trial and was “not required to waive that right in order to participate in the remedial phase.” Id. ¶ 712.

The State urges us not to “claim we know the mind of the Legislature,” “acknowledge[s] the possibility that the Legislature understood the language” of our first preliminary injunction differently than we intended, and asserts that the “record tells a more innocuous story reflecting the Legislature's good faith efforts.” Id. ¶¶ 713, 715. The State explains its view of the “ground rules” the Supreme Court set about traditional districting principles and the performance of the 2023 Plan on those principles. Id. ¶¶ 725–43. The State asserts that the Legislature “deemphasized the traditional principle of core retention” to “place the Black Belt counties into only two districts.” Id. ¶¶ 740–41. In this narrative the State makes no mention of whether those two districts are majority-Black districts – indeed, the State makes no mention of majority-Black or Black-opportunity districts at all. Id. ¶¶ 725–43.

The State urges us to be “sympathetic to the Legislature's plight of trying to make sense” of a confusing area of the law, and that the Legislature's actions reflect “a good faith effort to enact a redistricting plan in light of Allen’s discussion of Gingles 1 while simultaneously protecting the rights of all Alabamians not to be segregated on the basis of race.” Id. ¶¶ 744–45. And the State attempts to distinguish various precedents on which the Plaintiffs rely for their defiance argument. Id. ¶¶ 747–53. In sum, the State asserts that there is “no way” that the 2023 Plan “could conceivably be considered ‘open defiance.’ ” Id. ¶ 756.

Third, the State argues that the Legislature “did not depart from the usual process of enacting legislation,” and “followed roughly the same procedures” it followed in 2021, 2011, and 2001. Id. at 267 & ¶ 760. It also argues that the Plaintiffs “draw special attention to a single, insignificant difference between the Committee's Redistricting Guidelines and the legislative findings accompanying the 2023 Plan.” Id. ¶ 776. More particularly, the State argues that the failure of the findings to mention nondilution of minority voting strength was simply the elimination of “redundant,” “superfluous language.” Id. ¶¶ 777–79. The State argues that the Legislature may have decided to eliminate that language based on Dr. Duchin's testimony about it during the preliminary injunction proceedings. Id. ¶ 780. The State cites no evidence that any legislator knew anything about Dr. Duchin, let alone her testimony. See id.; Milligan Doc. 459-13 at 13 (Senator Livingston's testimony that although he was aware of what our order required before his deposition, he had not read it until then).

Fourth, the State argues that “[n]one of the ‘contemporary statements and actions of key legislators’ relied upon by Plaintiffs ‘show a racially discriminatory intent.’ ” Milligan Doc. 481 ¶ 785. The State argues that the Legislature passed the 2023 Plan for “traditional or partisan reasons.” Id. The State contends that Speaker Ledbetter's statement about having a “good shot” at changing the mind of a Supreme Court Justice “neither states nor suggests an invidious, racial motive.” Id. ¶ 786. Similarly, the State contends that other legislators’ statements about having “seven Republican congressman” and a “Republican opportunity plan” “speak to partisan, not racial, motives.” Id. at ¶ 787. And it argues that Senator Livingston's reference to Montgomery as “monkey town” is not racist because Senator Singleton did not interpret it that way. See id. ¶¶ 789–90.

Fifth, the State argues that the Legislature could not have foreseen that the 2023 Plan would have any discriminatory effects. Id. ¶ 796. The State argues that the “Legislature had every reason to believe the 2021 Plan would be ‘race-neutral’ in effect.” Id. ¶ 797. And the State asserts that “Dr. Hood's performance analysis of [District 2] in the 2023 Plan did not reveal to the Legislature a discriminatory effect on account of race.” Id. ¶ 800. Rather, the State says, “[a]ll it showed, on its face, is ‘political defeat at the polls.’ ” Id. The State also argues that there is no evidence that any legislator outside the conference committee saw Dr. Hood's analysis, and argues that “[b]ecause it is unclear how many legislators even had access to or considered the information, it cannot support a finding of discriminatory intent.” Id. ¶ 801.

Sixth, the State argues that “[t]he Legislature did not refuse to consider alternative plans that would lessen any potentially discriminatory impact.” Id. ¶ 807. The State asserts that the Milligan “Plaintiffs have ‘failed to identify viable alternatives to the [2023 Plan] that would have achieved the same objectives.’ ” Id. ¶ 808. The State argues that “[t]he unalarming fact that Democratic legislators were not brought into the ‘map drawing room,’ ․ suggests only that party politics were at play.” Id. ¶ 812.

VII. FINDINGS OF FACT AND CONCLUSIONS OF LAW — CONSTITUTION

A. Avoidance Canon

We first explain why we decide the constitutional claim of the Milligan Plaintiffs. Our previous decisions not to decide the Plaintiffs’ constitutional claims were based on the canon of constitutional avoidance, see Lyng, 485 U.S. at 445 (collecting cases dating back to Ashwander v. Tennessee Valley Auth., 297 U.S. 288, 341 (1936) (Brandeis, J., concurring)), which has particular salience when a court considers (as we did) a request for equitable relief, see id., and which is commonly applied by three-judge courts in redistricting cases that involve both constitutional and statutory claims, see, e.g., LULAC, 548 U.S. at 442; Gingles, 478 U.S. at 38. See Milligan Doc 107 at 223; Milligan Doc. 272 at 194–95.

Under the avoidance canon, “[c]onsiderations of propriety, as well as long-established practice, demand that [courts] refrain from passing upon the constitutionality of an act of Congress unless obliged to do so in the proper performance of our judicial function, when the question is raised by a party whose interests entitle him to raise it.” Ashwander, 297 U.S. at 341 (Brandeis, J., concurring) (quoting Blair v. United States, 250 U.S. 273, 279 (1919)). The avoidance canon recognizes the “great gravity and delicacy of [the courts’] function in passing upon the validity of an act of Congress” and restricts “exercise of this function by rigid insistence that the jurisdiction of federal courts is limited to actual cases and controversies; and that they have no power to give advisory opinions.” Id. at 345–46 (internal quotation marks omitted).

The canon has numerous substantial justifications:

The policy's ultimate foundations, some if not all of which also sustain the jurisdictional limitation, lie in all that goes to make up the unique place and character, in our scheme, of judicial review of governmental action for constitutionality. They are found in the delicacy of that function, particularly in view of possible consequences for others stemming also from constitutional roots; the comparative finality of those consequences; the consideration due to the judgment of other repositories of constitutional power concerning the scope of their authority; the necessity, if government is to function constitutionally, for each to keep within its power, including the courts; the inherent limitations of the judicial process, arising especially from its largely negative character and limited resources of enforcement; withal in the paramount importance of constitutional adjudication in our system.

Rescue Army v. Mun. Ct. of City of Los Angeles, 331 U.S. 549, 571 (1947).

In its simplest formulation, the canon holds that “[a] fundamental and longstanding principle of judicial restraint requires that courts avoid reaching constitutional questions in advance of the necessity of deciding them.” Lyng, 485 U.S. 439, 445–46.

Accordingly, we first determine whether a decision on the constitutional claims could entitle any Plaintiffs to relief beyond that to which they are entitled on their statutory claims. Id. “If no additional relief [is] warranted, a constitutional decision [is] unnecessary and therefore inappropriate.” Id.

Because the Milligan Plaintiffs request bail-in under Section 3(c) of the Voting Rights Act, which is available only upon a finding of a constitutional violation, see Milligan Doc. 329 at 77; 52 U.S.C. § 10302(c), we must decide the constitutional claims of the Milligan Plaintiffs. Because the Singleton Plaintiffs do not request bail-in and have different constitutional claims, none of which would involve relief beyond that to which they are entitled on their Section Two claims (declaratory and injunctive relief), we need not and therefore do not decide the constitutional claims of the Singleton Plaintiffs. Counsel for the Singleton Plaintiffs recognized as much at trial. See Tr. 2600–01.

B. Presumption of Legislative Good Faith

We begin our analysis of the Milligan Plaintiffs’ claim of intentional discrimination, as we must, with a heavy presumption of good faith in favor of the Legislature. The presumption of legislative good faith “directs district courts to draw the inference that cuts in the legislature's favor when confronted with evidence that could plausibly support multiple conclusions.” Alexander, 602 U.S. at 10 (citing Abbott, 585 U.S. at 610–12).

As the Supreme Court has explained, three reasons justify this presumption:

First, this presumption reflects the Federal Judiciary's due respect for the judgment of state legislators, who are similarly bound by an oath to follow the Constitution. Second, when a federal court finds that race drove a legislature's districting decisions, it is declaring that the legislature engaged in “offensive and demeaning” conduct that “bears an uncomfortable resemblance to political apartheid.” We should not be quick to hurl such accusations at the political branches. Third, we must be wary of plaintiffs who seek to transform federal courts into “weapons of political warfare” that will deliver victories that eluded them “in the political arena.”

Id. at 11 (internal citations omitted).

We understand the importance of this presumption, we agree with it, and we have tried to apply it. We have assumed the Legislature's good faith – and exercised restraint – throughout these proceedings. In our first preliminary injunction, we assumed the Legislature's good faith, conducted our Senate Factors analysis with restraint, and refused to consider arguments that the Legislature's conduct was likely unconstitutional. After we issued that injunction, the Singleton Plaintiffs urged us to act on their constitutional claims, even if we rejected them, and we refused. Singleton Docs. 98, 104, 114.

Perhaps most notably, we assumed the Legislature's good faith when we paused remedial proceedings (which all agreed were time-sensitive) for five weeks in June and July 2023 to allow the Legislature sufficient time to enact a new plan. In the second preliminary injunction, despite our serious concern about what the Legislature did, the posture of the case, and whether the requirements of our orders would be satisfied, we again assumed the Legislature's good faith, conducted our Senate Factors analysis with restraint, and refused to consider the Milligan Plaintiffs’ arguments that the Legislature's post-affirmance conduct was unconstitutional. Out of respect for the Legislature's role and our role, we proceeded in this manner even as we were acutely aware that any delay might eventually cause the Plaintiffs to suffer the further deprivation of their voting rights in another congressional election, even if they ultimately prevailed on the merits of their claim.

Put simply, for as long as the record in these cases allowed, we did not consider, comment on, or otherwise confront the allegation that the Legislature's intent was unlawful.

Even now, as we finally confront that claim, we begin our analysis with the presumption that the Legislature did act in good faith. As we discuss below, we draw every inference we can in the Legislature's favor. The problem for the Legislature is not the inferences we draw – it is that our analysis does not rest only, or in the main, on inferences. It rests on what the Legislature did, what it said about what it did in its enacted legislative findings, what key legislators and the legislators’ cartographer said, and the ultimate act of enacting a map that did not nurture any ambition to comply with what the Court said the law required. Thus, we have no occasion to try to read legislators’ minds, and we have not done so.

The current allegation of intentional discrimination in Milligan is fundamentally unlike the allegation in 2021 and unlike the typical allegation: it is not that the Legislature, navigating the twin hazards of constitutional and statutory liability, considered race too much when it placed district lines. See, e.g., Abbott, 585 U.S. at 607–14; Bethune-Hill, 580 U.S. at 181–86; Callais v. Landry, 732 F. Supp. 3d 574, 599 (W.D. La. 2024).

The essential charge is that the Legislature – knowing from both us and the Supreme Court, after extensive litigation, that a map with only one Black-opportunity district very likely unlawfully diluted the votes of Black Alabamians – intentionally passed just such a map. More particularly, it is that the Legislature did it again – fully knowing from our first preliminary injunction (consistent with more than a century of redistricting history in Alabama) that the way to crack the Black vote in South Alabama is to submerge much of the Black Belt in one majority-White district and submerge the Black Alabamians in Mobile in a different majority-White district. And it is that the Legislature – having been told in a court order the remedy that federal law requires – purposefully and admittedly refused to provide that remedy.

As we face a Legislature that admits it intentionally refused to provide the required remedy, and legislators who admit they understood the requirement, our work cannot fairly be reduced to uncharitable inferences and accusations.

Finally, we do not accuse any Legislator of being animated by racism. We see and credit the evidence that some legislators tried to persuade their colleagues to provide the remedy federal law requires. As to the lawyers, we well understand their professional obligations as zealous advocates. We simply review the Legislature's own actions, words, and enactments, measure those against the applicable legal test, and find that the Legislature purposefully diluted Black Alabamians’ opportunity to participate in the political process, knowing full well what it would accomplish if it succeeded and intending to do just that. We now turn to that test.

C. Arlington Heights and the Eleventh Circuit's Additional Factors

To evaluate the Milligan Plaintiffs’ assertion of discriminatory purpose, we first discuss the Arlington Heights factors, and then we discuss the additional factors the Eleventh Circuit has identified. See supra Part III.B.

1. Historical Background of the 2023 Plan

We start with “[t]he historical background” of the Legislature's decision to enact the 2023 Plan, and we ask whether the background reflects “a series of official actions taken for invidious purposes.” Arlington Heights, 429 U.S. at 267. We observe that in its discussion of this factor in Arlington Heights, the Supreme Court cited an Alabama voting rights case involving Mobile County: Davis v. Schnell, 81 F. Supp. 872, 880 (S.D. Ala.), aff'd, 336 U.S. 933 (1949), in which a three-judge court ruled an amendment to the Alabama Constitution unconstitutional in part because it was passed “in an attempt to obviate the consequences of” a Supreme Court decision. Id.

If we review the long-term history of redistricting and voting rights in Alabama, the record undeniably reflects a series of official actions taken for invidious purposes: between Dr. Bagley's undisputed testimony and the relevant judicial precedents we have cited, see supra Part V.A.4.d, the reality that Alabama has a long and repugnant history of purposefully discriminating against Black Alabamians in redistricting and voting rights is well-documented. That said, the more recent history of redistricting and voting rights in Alabama shows improvement: official actions in Alabama in the past 30 years have been less tainted by discriminatory purposes than they were in the 30 years before that, or the 30 years before that.

But if we review the immediate history of the Legislature's decision to enact the 2023 Plan, and we study carefully the 2023 Special Session, we cannot help but find that this background weighs in favor of a finding of intentional discrimination.

The immediate history of the 2023 Plan began in 2022, when we issued a preliminary injunction finding that the 2021 Plan likely violated Section Two and ordering that based on the extensive evidence of intensely racially polarized voting patterns in these cases, “any remedial plan will need to include two districts in which Black voters either comprise a voting-age majority or something quite close to it.” Milligan Doc. 107 at 6. That history continued in June 2023, when the Supreme Court affirmed our order with no hint, suggestion, or holding that we were mistaken either about our liability determination or about the remedy we said the law required. See supra Part I.C (discussing Allen). And that history continued in July 2023 when the State delayed remedial proceedings for five weeks; enacted a new plan that it admitted did not include an additional opportunity district; enacted novel legislative findings that made the additional opportunity district impossible to draw and materially reconfigured the State's definition of key terms (which we discuss in detail below); and then told us we were back at square one, needing to fully relitigate liability before we could determine anything about remedy or order the State to use a different map for the 2024 election. See supra Parts I.B–I.E.

Standing alone, this history weighs in favor of the Plaintiffs because it makes it difficult to understand the Legislature's decision to enact the 2023 Plan as anything other than a deliberate decision to double down on the dilution of Black Alabamians’ votes. After we explained and the Supreme Court affirmed the likely discriminatory effects of the 2021 Plan, and we unambiguously explained the remedy that the law requires, the Legislature “came back and [ ] did precisely what [we] said would not be a remedy,” Tr. 2669–70, and diluted minority voting strength on purpose. This history causes us grave concerns about the Legislature's good faith.

2. Sequence of Events Leading to the Passage of the 2023 Plan

The sequence of events leading to the passage of the 2023 Plan also strongly suggests that the Legislature intended to dilute the voting strength of Black Alabamians. We consider the Legislators’ real-time understanding of what the law required as the initial event in this sequence, and we credit their testimony that in the summer of 2023, they well understood what was required by Section Two of the Voting Rights Act and our preliminary injunction. See Milligan Docs. 459-13 at 6, 14 (Livingston), 459-20 at 5 (Pringle).

Likewise, we accept Mr. Hinaman's testimony that he understood that he needed to draw a second opportunity district, which meant one where Black-preferred candidates had an opportunity to elect a representative of their choice. See supra Part I.I.3.a. Mr. Hinaman understood the difference between a second opportunity district (which he said he attempted to draw) and a second majority-Black district (which he said he made no attempt to draw), as we would expect of someone with his level of redistricting cartography experience. See Milligan Doc. 459-7 at 16–17. At the outset then, we discern zero confusion among these key stakeholders about what would be necessary to satisfy the remedial requirements that our order laid out.

It thus troubles us that from the outset, the Legislators specifically instructed Mr. Hinaman to keep Mobile and Baldwin Counties whole and together. See id. at 20. As we have already explained at great length, and as all parties readily agreed, keeping those counties whole and together made it mathematically impossible to create a second opportunity district. This is particularly troubling since Mr. Hinaman explained that when he began his work in the summer of 2023, he had never seen the 2023 legislative findings (or even a draft of them), was told nothing about them, and ultimately performed all of his 2023 work completely in the dark about the “non-negotiable” rules and novel definitions they set out. See id. at 23–24. That Mr. Hinaman was not instructed about the findings and was instructed to keep Mobile and Baldwin Counties together, suggests that his work was never seriously intended to generate a map that the Legislature would pass, nor a map that could have provided a second Black-opportunity district.

Moreover, at the pre-session hearings, Representative Pringle invited a historian to testify about the historical connections between Mobile and Baldwin Counties, but “did not ask anyone to speak on behalf of the need for two districts in which Black voters could elect candidates of their choice.” Milligan Doc. 485 ¶ 802 (citing Milligan Doc. 459-20 at 12). He also rejected an amendment to the 2023 guidelines offered by Representative England about how to remedy the likely Section Two violation we found. Id. ¶ 803 (citing Milligan Doc. 403-4 at 23; Tr. 1328–29).

Our concerns were deepened by what followed. The parties stipulate that when Representative Pringle introduced the Community of Interest Plan in the Committee (with a BVAP of 42.4% in District 2), his explanation for adopting it was that it preserved the cores of existing districts, and he added that Dr. Hood's performance analysis of that Plan showed that the Black-preferred candidate would have won two of four modeled races. See Milligan Doc. 436 at 19. Mr. Hinaman understood that both Representative Pringle and Senator Livingston preferred that plan (out of all the plans he drafted) and would sponsor it in their respective chambers. See Milligan Doc. 459-7 at 8. Nevertheless, Senator Livingston, Alabama Senators, and ultimately the entire Legislature turned away from that plan.

According to Senator Livingston, “[t]he committee members changed [their] focus” away from the Community of Interest Plan based on “additional information” they received about “compactness, communities of interest, and making sure that congressmen are not paired against each other.” Milligan Doc. 459-13 at 17. Senator Livingston described this “additional information” as a “large hiccup,” but he did not know what it was, he did not know where it had come from, and he did not know who received it. Id. He said only that he learned of the “information” in a “committee conversation,” but he did not recall from whom and had no “idea at all” of its source. Id.

It strains credulity to say that the Committee co-chair was in the dark about all of this and cannot recall anything about it. Ultimately, the Senate turned to the Opportunity Plan, which Senator Livingston introduced in the Committee and which was drafted by Mr. Brown. In that plan, the BVAP in District 2 was 38.31%. Senator Livingston was a major stakeholder in congressional redistricting in Alabama, and was apparently so influential that he ultimately navigated the Legislature to the bill it eventually passed, even as Representative Pringle refused to lend his name to the bill or assist in its passage. We cannot easily accept that he turned away from a plan drafted by Mr. Hinaman and to a plan that conflicted with his understanding of the federal court orders for reasons unknown.

The record does not tell us what the missing “additional information” was, but the next sequence of events suggests that it was important – so important that it ultimately led to the demise of Representative Pringle's Community of Interest Plan. Representative Pringle and Mr. Hinaman were fully sidelined from the process of drafting the 2023 Plan because as they worked on their Community of Interest plan, Senator Livingston and other Senators met with the Solicitor General behind closed doors in a different room on a different floor on a different plan. See Milligan Doc. 459-20 at 7; Milligan Doc. 459-7 at 8–10.

The record tells us what the “additional information” was not: plainly it was not information about how to comply with Section Two or satisfy the requirements of this Court's orders. After Senator Livingston introduced the Opportunity Plan and it passed in the Senate, he delivered it to co-chair Representative Pringle, who refused to pass it in the House, or substitute it for his Community of Interest Plan, or even attach his name to it. We are particularly struck by his candid testimony that he broke with Senator Livingston over the Opportunity Plan because he believed it might not or did not satisfy the Voting Rights Act, and he believed the Community of Interest Plan had a better chance of meeting the obligations of the law. See Milligan Doc. 459-20 at 26. Representative Pringle's real-time doubt that the Opportunity Plan satisfied the requirements of the Voting Rights Act, which he communicated to Senator Livingston to no effect, is meaningful.

Still other evidence drawn from this stage of the sequence underscores Representative Pringle's concerns. Senator Livingston testified that during the 2023 Special Session, he relied on talking points about the Livingston Plans that were prepared by the Solicitor General. See Milligan Doc. 459-13 at 26. And Dr. Bagley explained that those talking points emphasized communities of interest. Dr. Bagley quoted the talking points as saying: “The Livingston Plan is a Compact, Communities of Interest Plan that applies the State's traditional districting principles fairly across the State. The 2023 Plan is a historic map that gives equal treatment to important communities of interest in the State, including three that have been the subject of litigation over the last several years – the Black Belt, the Gulf, and the Wiregrass.” Milligan Doc. 385-1 at 26. The talking points also said: “No map in the State's history, and no map proposed by any of the Plaintiffs who challenged the 2021 Plan, does better in promoting any one of these communities of interest, much less all three.” Id.

We take from the events in the sequence so far that the “additional information” was not about how to provide an additional opportunity district, and the purpose of the Livingston Plan was not to provide an additional opportunity district. Rather, that plan positioned communities of interest as a trump card to excuse the Legislature's refusal to provide an additional opportunity district.

Even if a State could use communities of interest to trump federal law (and it cannot, see supra Part V.A.2.e) Senator Livingston's promotion of the Livingston Plan as being equally respectful of three communities of interest is wholly unconvincing. The Livingston Plan homed in on Alabama's most well-known and nationally prominent Black community: Selma. In the Community of Interest Plan, Dallas County was placed entirely in District 2; in the Opportunity Plan, it was placed entirely in District 7. This made a significant difference to the District 2 BVAP and left Black Alabamians in District 2 in the Opportunity Plan utterly unable to elect a representative of their choice. Although we doubt that any legislator needed an expert to tell them that moving Selma out of District 2 destroyed any chance it might have had of performing as an opportunity district, the undisputed evidence from Dr. Hood's performance analysis told them exactly that. See supra Part I.I.3. Again, Dr. Hood said that the Black-preferred candidate would have lost all seven modeled races by approximately seven points in the 2023 Plan's District 2. Id.; see also Caster Doc. 352-2 at 14; App. E at 3; Tr. 298–99, 314 (explaining that the 2023 Plan placed nine of the eighteen Black Belt counties in majority-White districts).

The next event in the sequence came on the morning the 2023 Plan was enacted, and the Legislators saw for the first time the eight pages of legislative findings that would be embedded in the bill. We discuss in the next section the many substantive and procedural departures from the norm in connection with the findings. For present purposes, we focus on their last-minute appearance in the sequence of events.

In fact, the legislative findings came only a week after the Committee had readopted its 2021 guidelines, and they were materially different from those guidelines. See supra Part I.D; infra Part VII.C.3. Notably, the findings were not requested by the Committee chairs, who did not know why they were placed in the bill, had never seen them before (and thus, had never studied them), and had never seen anything like them in any redistricting legislation. See Milligan Docs. 459-13 at 26 (Livingston), 459-20 at 23 (Pringle). Based on these undisputed facts, we cannot infer that the findings were the ordinary result of the Legislature's deliberative process.

Standing alone, there is nothing wrong with legislative findings, nor with findings drafted by lawyers. But the last-minute, unsolicited arrival of these legislative findings forecloses any assertion that in real time, as the Legislature drafted its plans, there was anything ordinary or usual about the process.

Whatever else this sequence of events tells us, it leaves precious little doubt that the Legislature intentionally chose not to satisfy the remedial requirements found in our order. The Legislature claimed that the 2023 Plan treated the Black Belt and Gulf Coast equally, knowing full well that (1) half the Black Belt counties were placed in majority-White districts where those Black Alabamians had zero chance of electing a representative of their choice, and (2) the BVAP in one of those majority-White districts was reduced by moving Selma into District 7. This tells us that the Legislature did not accidentally stumble into a racially discriminatory districting plan for a second time.

3. Substantive and Procedural Departures from the Norm, and Legislative History of the 2023 Plan

But there is more. We discuss the next two Arlington Heights factors together because the evidence about departures from the norm dovetails nearly completely with the evidence about the legislative history of the 2023 Plan: it revolves around the 2023 legislative findings. Indeed, these legislative findings are at the heart of the case.

We observe that we give the findings substantially more weight than we ordinarily would assign to legislative history. The Legislature enacted the findings, so we do not have to wonder whether they reflect the sentiments of a majority of legislators. They are statutory text, and we are bound to accept that they do.

Standing alone, the fact of the findings is a severe departure from the norm. Although the Committee traditionally has passed redistricting guidelines, the Legislature has never previously enacted findings. Even when the Legislature last redistricted in connection with a court order (in the 2010 cycle, after the Alabama Legislative Black Caucus litigation), the Legislature did not enact findings.

But the fact of the findings is not the only departure from the norm or the most significant: in substance, the findings are replete with sharp departures from (and some outright conflicts with) Alabama's traditional districting guidelines:

• First, the findings describe some traditional districting principles as “non-negotiable,” which the guidelines never did.

• Second, although the findings elsewhere provide that the districting plan must comply with the Voting Rights Act, the Legislature did not include the nondilution of minority voting strength on its “non-negotiable” list. (And the only reason why the 2023 Plan even exists is because we enjoined the 2021 Plan on the ground that it likely diluted minority voting strength.)

• Third, the Legislature eliminated from the findings any reference to nondilution of minority voting strength, which the guidelines expressly addressed as a priority consideration.

• Fourth, the findings identify specific communities of interest, which the guidelines never did.

• Fifth, although the northern half of Alabama is home to numerous universities, a substantial military installation, various engines of economic growth, the Tennessee Valley, and two significant metropolitan areas (Huntsville and Birmingham), the findings identify zero communities of interest in that half of the state and focus exclusively on the areas discussed in the prior litigation here.

• Sixth, the findings materially revised the definition of “community of interest” that appeared in the guidelines: the findings stripped race (and ethnic, tribal, and social identities) out of the list of “similarities” that may support a community of interest. Compare App. A at 4, with App. B. This would be a sharp departure from the norm in any circumstance, but it is razor sharp in a case involving extensive testimony about a racial minority's shared experience of a long and sordid history of official race discrimination.

• Seventh, the findings exalt and extol one community of interest above others, describing the community of interest in the Gulf Coast for pages, while describing the longstanding and well-grounded community of interest in the Black Belt in a couple of short paragraphs. We are aware of no enacted redistricting document that ever has done this.

• Eighth, in those paragraphs, the Legislature eliminated from the definition of the Black Belt that the Legislators previously stipulated in these very cases the express recognition that the Black Belt “has a substantial Black population because of the many enslaved people brought there to work in the antebellum period.” Milligan Doc. 53 ¶ 60.

• Ninth, the findings describe the “French and Spanish colonial heritage” of one community of interest (the Gulf Coast) while remaining silent on the heritage of all other communities of interest in Alabama (including the Black Belt). App. A at 6. Here again, this would be unusual in any circumstance, but it is especially pointed in a voting rights case where one of the allegations is that the majority-White community of interest in the Gulf Coast is being used to entrench race-based discrimination against the majority-Black community of interest in the Black Belt, which shares a heritage of enslavement.

• Tenth, although the 2023 Plan exists only because we enjoined the 2021 Plan and ordered that a remedial plan would need to include an additional opportunity district, the findings say nothing about an additional opportunity district. See Milligan Doc. 385 ¶ 874.

We are not the only ones who are surprised by these departures from the norm. As we previously discussed, the Committee co-chairs had no idea they were coming.

The State has little to say about these differences, and it does not contest that any of them are substantial departures from the norm. It defends the elimination of the provision about nondilution of minority voting strength as an “absence of superfluous language.” See Milligan Doc. 481 ¶¶ 778–79. It defends its silence about communities of interest in the northern half of the State by reference to these lawsuits. See id. ¶¶ 668–69. And it says that the reference to the shared “French and Spanish colonial heritage” in the Gulf Coast was not a reference to White people, but a reference to “cultural influences in a region that flow from its unique history.” Id. at 13. To be clear, we do not hold that the reference to “French and Spanish colonial heritage” is a colormasked reference to White people. We simply recognize that it stands in stark contrast to the silence about the heritage of the Black Belt, which is one of enslavement.

We draw three conclusions about the findings. First, we observe at the threshold that they presented a rare opportunity for a litigant to make real-time evidence to bolster their arguments in court. We repeat that we see no issue with legislators receiving or following legal advice. But that doesn't accurately describe what happened here. The legislative findings appeared out of thin air in the final moments before the Legislature voted on the 2023 Plan, and we have no evidence that any legislator requested, anticipated, discussed, or debated them.

From the State's drumbeat reminders that we cannot draw an additional opportunity district without violating the Legislature's instruction about Mobile and Baldwin Counties, we gather that the findings were meant to prevent a federal court from drawing a remedial district. Any such district would run headlong into an argument that it violated the Legislature's provisions for communities of interest and therefore failed to respect traditional districting principles.

Put differently, we see the legislative findings as the centerpiece of the Legislature's effort to intentionally checkmate any remedial order designed to require a second opportunity district. Not only did the Legislature enact a plan that refused to provide for an additional opportunity district; what's more, when they embedded their findings in that refusal, they inserted a guardrail designed to prevent a federal court from providing that district. Quite simply, these legislative findings made it impossible to achieve what federal law required.

Second, in the context of these cases, we cannot understand the constellation of departures from the norm as anything other than an intentional official effort to entrench the likely vote dilution we found. The combination of (1) the elevation of communities of interest and incumbent protection to “non-negotiable” status, (2) the exclusion of nondilution of minority voting strength from not only the “non-negotiable” list, but also the entire findings, (3) the deletion of shared race-based experiences from the definition of “community of interest,” (4) the identification of a specific community of interest (the Gulf Coast) and the exaltation of this majority-White community of interest above all other communities of interest (and above all other traditional districting principles), (5) the utter failure even to acknowledge, let alone name, any community of interest not being litigated in these cases, and (6) the singular reference to the heritage of the majority-White community of interest, coupled with the deletion of any description of the race-based heritage of the very majority-Black community at issue in these cases, leaves little room to conclude that when the Legislature enacted these findings, it must have had some purpose other than minority vote dilution in mind.

We are mindful that the Legislators did not see the findings until the last minute. But the Legislature enacted the findings anyway, and by a large margin. See Milligan Doc. 436 at 20. If any Republican legislator objected to the findings in any way, there is no indication of that anywhere in our voluminous record. Even the Republican Legislators who might have been most likely to object voted for the findings. We would badly err if we discounted their meaning or otherwise failed to afford them the status they have as statutory text.

Third, we observe that even if, in extreme service of the presumption of legislative good faith, we were to discount our concerns about the first two Arlington Heights factors (historical background and sequence of events), and assume that perhaps the unusual sequence of events that led to the 2023 Plan was the result of pure political disagreements, personal grudges between legislators, or other considerations unknown to us, the extraordinary departures from the norm in the legislative findings would substantially undermine the presumption of good faith. As is clear by now, we do not have one data point from the findings, or two, or even three or four. The findings are inherently a departure from the norm and replete with departures from the norm, all of which point in the same direction: that the findings were intended to make the discriminatory vote dilution that we identified impossible to remedy.

Finally, we observe one additional departure from the norm in connection with the 2023 Plan: a purposeful refusal to provide the remedy that a court order requires is quite uncommon, and definitively not the norm. Occasionally, legislatures have been unable or unwilling to redistrict after courts enjoin electoral plans. See Vera v. Bush, 933 F. Supp. 1341, 1346 (S.D. Tex. 1996); Reynolds, 377 U.S. at 586; Bone Shirt v. Hazeltine, 461 F.3d 1011, 1022 (8th Cir. 2006). And sometimes, legislatures redistrict, but courts later determine that their map does not remedy the violation. See Burton, 561 F. Supp. at 1035; North Carolina v. Covington, 585 U.S. 969, 970–71 (2018); Jeffers v. Clinton, 756 F. Supp. 1195, 1196 (E.D. Ark. 1990), aff'd, 498 U.S. 1019 (1991). But we are aware of no other case, let alone a trend of other cases, that reflect a legislature's willingness to act coupled with its admitted, strategic, and express refusal to provide the required remedy.

4. Whether Disparate Impact Was the Natural and Foreseeable Consequence of the 2023 Plan

We next consider whether the disparate impact we found (see supra Part V) were the natural and foreseeable result of the 2023 Plan. We find that they were, for four separate, independent, and substantially undisputed reasons.

First, our first preliminary injunction made clear that because of the ample evidence of racially polarized voting in Alabama, the “practical reality” was that “any remedial plan will need to include two districts in which Black voters either comprise a voting-age majority or something quite close to it.” Milligan Doc. 107 at 6. That injunction, as affirmed by the Supreme Court, was the only reason the Legislature was in session to prepare and enact the 2023 Plan. It was therefore entirely foreseeable that a refusal to provide the additional opportunity district we ordered would perpetuate rather than remedy the likely vote dilution we found.

Second, no one disputes that before the Legislature passed the 2023 Plan, Dr. Hood conducted and shared with legislators a performance analysis that modeled seven elections and predicted Black-preferred candidates would lose every election in District 2 in that Plan. Milligan Doc. 436 at 21 (stipulations). That performance analysis would have given any legislator who reviewed it actual knowledge of what they already should have foreseen and expected: that as a plan without a second opportunity district, the 2023 Plan would perpetuate rather than remedy the likely vote dilution we found.

Third, the record establishes (without dispute) that key legislators were directly told that the 2023 Plan would not satisfy our order and would continue to dilute votes. We are reminded of Representative Pringle's conversation with Senator Livingston, in which Representative Pringle refused to put his name on the 2023 Plan based on his evaluation that his Community of Interest Plan was more likely to comply with the Voting Rights Act. And we credit Representative Pringle's testimony that he saw Dr. Hood's performance analysis before the Legislature voted on the 2023 Plan and that it was available to all members of the conference committee. And we credit (as we must) the parties’ stipulation that Representative England advised legislators before they voted on the 2023 Plan that in his view, it was noncompliant with our order and we would reject it. Milligan Doc. 485 ¶ 835.

Fourth, we credit Mr. Hinaman's testimony (which no one disputes) that at least some legislators actually knew from Dr. Hood's earlier performance analysis of the Community of Interest Plan that without Dallas County in District 2, Black-preferred candidates would have no chance of winning in that District. According to Mr. Hinaman, the Black-preferred candidate lost every election Dr. Hood modeled in that District, if Selma was not in the District. We thus find that it was reasonably foreseeable that the 2023 Plan, which left Dallas County out of District 2, would afford Black-preferred candidates zero chance of winning there.

Ultimately, perhaps the strongest indicator about this factor comes from the Legislature itself: in the light of the Legislators’ acknowledgement in court that the 2023 Plan does not include the additional opportunity district we ordered, we cannot find that it was unforeseeable that the 2023 Plan would perpetuate rather than remedy the likely vote dilution we found. No legislator can fairly be surprised that we again found disparate impact from a districting plan that includes only one majority-Black or Black-opportunity district.

5. Direct Evidence of Intent and Contemporaneous Statements

We now turn to the final Arlington Heights factor, which calls for us to analyze whether there is any direct evidence of the Legislature's intent and consider relevant contemporaneous statements by legislators.

At the outset, we reject the Milligan Plaintiffs’ arguments that statements by the Solicitor General in court are direct evidence (or evidence at all) of legislative intent. Arguments by lawyers can make concessions (which the Solicitor General did), but they are not evidence, and we do not treat them as evidence. See, e.g., Horn, 129 F.4th at 1291.

We have focused intensely on the question whether the record before us contains direct evidence of discriminatory legislative intent. We are mindful that although discriminatory intent need not be proved by direct evidence, the Supreme Court has “never invalidated an electoral map in a case in which the plaintiff failed to adduce any direct evidence.” Alexander, 602 U.S. at 8. We add, however, that counsel for all of the relevant parties, including counsel for the Secretary, agreed that a Fourteenth Amendment intentional vote dilution case can be established by circumstantial evidence alone. See Tr. 2574–77, 2636.

We regard both the 2023 legislative findings and the contemporaneous public comments by legislators as including direct evidence. The legislative findings are the most direct expressions available (to us and the public) of what the Legislature, as an institution that speaks through its enactments, set out to accomplish. Unlike the Committee guidelines and legislators’ public comments, the legislative findings are the product of a vote of the entire body. And they remain in force as legislative pronouncements unless and until they are amended, repealed, superseded, invalidated, or otherwise lawfully changed. The Legislature cannot escape reasonable inferences drawn from them (so long as the appropriate presumptions are applied). Accordingly, we realize that we draw inferences from the findings, but we decline to hold that this renders them indirect or otherwise diminishes their probative value.

As we have already explained, see supra Part VII.C.3, the findings are of immense probative value (as we understand they were intended to be). And they establish both that and how the Legislature intended the 2023 Plan to discriminate against Black Alabamians: by perpetuating vote dilution and making it impossible to remedy.

Contemporaneous public statements by legislators corroborate this conclusion. Representative Pringle testified about a media comment after the enactment of the 2023 Plan in which Speaker Ledbetter explained, “[i]f you think about where we were, the Supreme Court ruling was five to four. So there's just one judge that needed to see something different. And I think the movement that we have and what we've come to compromise on today gives us a good shot.” Milligan Doc. 459-20 at 27–28. We take from this comment that Speaker Ledbetter was not focused on trying to remedy likely vote dilution.

Likewise, Dr. Bagley reported numerous public comments from Black legislators expressing frustration about their belief that Republicans in the Legislature never intended to pass a map with an additional opportunity district. According to Dr. Bagley, Representative England told to the media, “[t]here was never any intent in this building to comply with their court order. There was never any intent in this building to comply with the Voting Rights Act.” See Milligan Doc. 403-4 at 27. And Representative Juandalynn Givan commented, “I'm ashamed of what we did here this week. We've chosen to outright, blatantly disobey the law and to further attempt and vote to bury the Voting Rights Act.” Id. And Senator Smitherman commented, “I think the process on the other side was set up so that you could make sure an African-American would not win it. I think it was intentionally set that way.” Id.

We accept that these comments are from legislators whose preferences did not prevail, and they reflect the frustration that naturally flows from that. Although we therefore do not assign them much weight, we recognize them as contemporaneous statements, consistent with each other and additional corroborating evidence, from stakeholders with knowledge of and experience in the workings of the Alabama Legislature.

Put differently, all the contemporaneous statements we have also point in the same direction: that the Legislature was not trying to remedy likely vote dilution, nor to give Black Alabamians an equal opportunity to elect a candidate of their choice, but to continue the likely dilution that we and the Supreme Court had found.

Even if we are wrong that the legislative findings or statements are direct evidence, our misclassification is not dispositive – we would interpret them in the same way and assign them the same weight upon their reclassification as circumstantial evidence. We take our lead from the Supreme Court. For more than two hundred years, the Supreme Court has recognized that circumstantial evidence may be as strong (if not stronger) than direct evidence:

Although [direct] proof may generally be desirable, we are not to shut our eyes on circumstances which sometimes carry with them a conviction which the most positive testimony will sometimes fail to produce. And if such circumstances cannot well consist with the innocence of the party, and arise out of her own conduct, and remain unexplained, she cannot complain if she be the victim of them.

The Robert Edwards, 19 U.S. 187, 190 (1821).⁶⁹ And for decades, the typical circumstantial-evidence-only redistricting case has been primarily about shapes and lines: the Supreme Court has, “at least in theory, kept the door open for those rare instances in which a district's shape is ‘so bizarre on its face that it discloses a racial design’ absent any alternative explanation.” Alexander, 602 U.S. at 9 (quoting Miller, 515 U.S. at 914).

Alabama is again in unusual territory. The circumstantial evidence in this case does arise out of the Legislature's own conduct, see The Robert Edwards, 19 U.S. at 190, and it is not about bizarre shapes, see Alexander, 602 U.S. at 9. In our view, the circumstantial evidence here (if it be that) is far more telling than shapes are: it establishes just how far the Legislature was willing to travel from the norm in service of its intention not only to refuse a remedy for the likely vote dilution we found, but to prevent a remedy altogether.

6. Eleventh Circuit's Additional Factors: Knowledge of Disparate Impact and Availability of Less Discriminatory Alternatives

We next turn to the two factors that the Eleventh Circuit has identified to supplement the Arlington Heights analysis. We do not suggest that an Arlington Heights analysis is insufficient to make findings of intentional discrimination. We consider these factors out of respect for the gravity of the issues before us and in an earnest effort to fully understand all relevant dimensions of the evidence. If there is something to discern about the additional factors that diminishes our findings about the Arlington Heights factors, we want to consider it.

When we discussed the foreseeability of disparate impact resulting from the 2023 Plan, see supra Part VII.C.4, we discussed some legislators’ actual knowledge that the 2023 Plan would disparately impact Black Alabamians. We thus consider whether the Legislature had a less discriminatory alternative available.

In the ordinary case, Gingles ensures that a legislature considering remedial plans has at least one lawful illustrative plan to consider: the first Gingles precondition requires a Section Two plaintiff to develop, offer, and substantiate that plan. In these cases, because we have multiple sets of plaintiffs and they offered multiple illustrative maps, the record contains at least eleven less discriminatory alternatives that the Legislature could have considered: the four Duchin Plans and seven Cooper Plans submitted in connection with the preliminary injunction proceedings.

Even if the Legislature refused to consider the Duchin Plans and Cooper Plans out of its view that race predominated in the preparation of those plans, the subsequent preparation of a race-blind plan by the Special Master on September 25, 2023 suggests to us how easy it would have been for the Legislature to consider another plan that complied with the requirements of Section Two. Milligan Doc. 295; see infra Part VIII (discussing the Special Master Plan). We selected the Special Master in large part because of his extensive Alabama expertise: he is a well-respected public servant who served alongside four Alabama Attorneys General as Chief Deputy Attorney General and has a deep familiarity with the local political landscape in Alabama. He and his team prepared three proposed remedial plans race-blind without any particular difficulty.

We acknowledge that although the Special Master Plan hews as closely as possible to the 2023 Plan while also respecting Alabama's traditional districting principles, see infra Part VIII, it still does not achieve all the political goals of the Legislature, particularly the goal of keeping Mobile and Baldwin Counties whole and together in one congressional district. We do not see that this aspect of the Special Master Plan is a failure on the metric before us: because the reason the Special Master Plan splits Mobile County is to remedy unlawful vote dilution, that split is the reason why the Special Master Plan is a less discriminatory alternative, not a perfect substitute. As a practical matter, the refusal of the 2023 Plan to split Mobile County is the reason why that Plan entrenches vote dilution, not a political or partisan goal we owe deference.

***

Accordingly, we find that every one of the Arlington Heights and Circuit factors suggests that the Plaintiffs have rebutted the presumption of legislative good faith, and that the Legislature acted with discriminatory intent when it passed the 2023 Plan. We see no factor, or even material part of a factor, that tilts in favor of the State.

D. The State's Efforts To Dispel the Inference

The State makes seven arguments to rebut a finding of intentional discrimination. As we see it, none succeed. For starters, we observe that the State badly misconstrues the Plaintiffs’ position as an assertion that the 2023 Plan is unconstitutional because the Legislature enacted it “without including two districts likely to favor Democrats.” Milligan Doc. 481 at 14. The Plaintiffs’ position is that the Legislature intentionally deprived them of the same chance at electing a representative of their choice that the Legislature affords White Alabamians, and that the Legislature did so because they are Black.

The State addresses its first four arguments to the effect of our preliminary injunctions.

 The State's Assertion that Our Order Was Not Violated

We reject the State's first argument that our order “was not violated” because we “did not order the Legislature to enact any particular map.” Id. This argument might have some relevance if these were contempt proceedings, but they are not. We have not suggested, and we do not find, that our order, strictly speaking, was “violated.” See Aug. 14, 2023 Tr. 75 (Judge Manasco: “I understand the face of the order did not order the Legislature to do anything.”). Upon a finding of likely liability for a Section Two violation, we ordered that a particular remedy was due, we described that remedy precisely, and the Legislature purposefully responded with an enactment that strategically, deliberately, and admittedly did not provide for that remedy though it could have done so in multiple ways.

Ultimately, this argument deepens rather than allays our concerns. As we explained in our second preliminary injunction, the State's position is that at the end of each liability determination, we have no authority to order a remedy if the Legislature plans and has time to enact a new map. If they enact a new map, the State says, we return to the first inning in the first ballgame of these proceedings and consider liability afresh. In essence, the State's argument that the Legislature was not ordered to provide a specific remedy creates an endless paradox that only the Legislature can break, thereby depriving Plaintiffs of the ability ever to effectively challenge and the courts of the ability to remedy. It cannot be that the equitable authority of a federal district court to order full relief for violations of federal law is always entirely at the mercy of a State electoral and legislative calendar.

Almost four years into these (not especially complex, although now exceedingly unusual) cases, we cannot accept that we are living in an infinity loop that no court order can break. And we refuse to accept that courts are powerless to effectuate relief under Section Two simply because the only named defendants in a lawsuit are the Secretary of State and two legislators. It is true both that we cannot hold the Legislature in contempt for violating our order, and that the Legislature purposefully and admittedly refused to provide the remedy our order said was required.

 The State's Assertion that It Respected Our Order

Second, we reject out of hand the State's assertion that it actually “respect[ed]” our order because the Legislature passed a new plan with time for us to assess it before it was used. Milligan Doc. 481 at 14. To be sure, it would have been worse for the State to jam the Court up on time before the 2024 election (although we did barely beat the deadline by which the Secretary said he needed a final map). See Caster Doc. 148 at 7. But the reality that the Legislature left us time to assess whether it satisfied the requirements of our order does not mean that the Legislature in any way satisfied the requirements of our order. This is particularly obvious when, in that limited window of time, the Legislature's lawyer told us the Legislature had not and would not provide the remedy our order said was required. Likewise, the reality that we would assess whether the Legislature satisfied the requirements of our order does not excuse the Legislature's purposeful refusal to do so.

 The State's Argument that Preliminary Injunctions Are Not Final

Third, we reject the State's assertion that because preliminary injunctions are not final, it “makes no sense to fault the State for curing the alleged ‘inconsistent treatment’ in [the 2021 Plan] ․ before going to trial.” Milligan Doc. 481 at 14–15. Preliminary injunctions are preliminary, but they are not advisory. Nor are they mysterious, especially when the Supreme Court affirms them and does so with great thoroughness and particularity. The State was of course free to enact a new plan before trial, and to extol the virtues of that plan as a complete remedy for the likely violations of federal law that we and the Supreme Court found. But the State had no basis to expect that it could enact a new plan, admit that the plan did not provide and could not provide (or even attempt to provide) the remedy we said federal law required, and still receive our blessing for that plan.

Furthermore, we reject the false premise of the State's assertion – that the 2023 Plan somehow “cure[d]” the alleged “inconsistent treatment” of the Black Belt and Gulf Coast in the 2021 Plan. Id. at 7 & ¶ 525. The 2023 Plan places “only half (nine) of the Legislature's 18 identified Black Belt counties [ ] in a majority-Black district,” and it places both Gulf Coast counties, whole and together, in a majority-White district that submerges the Black Alabamians who live in the City of Mobile. Caster Doc. 352-2 at 14; App. E at 3; Tr. 298–99, 314. This is just more cracking of the Black vote, not a cure for the underlying problem. In fact, the State never seriously contends that it is a cure, arguing instead that splitting the Gulf Coast intolerably “chop[s] up” Alabama's most important community of interest, indeed its so-called most important traditional redistricting criteria. Milligan Doc. 481 ¶¶ 146, 196; Tr. 2625–27.

 The State's Argument that in Summer 2023, the Law Was Unclear

We also reject the State's fourth argument, that even if the Milligan Plaintiffs prevail, “the suggestion” that the law is “so clear” that any interpretation other than Plaintiffs’ “could only be driven by racial animus” is “baseless” and “divisive.” Milligan Doc. 481 at 15. Here again, this is a swing at a straw man. When the Legislature enacted the 2023 Plan, there was no lack of clarity that an additional opportunity district was necessary in Alabama, nor what an additional opportunity district meant: we expressly said so, and the Supreme Court affirmed our order without suggesting or discussing, let alone holding, that we were wrong about that. And some members of the Legislature may well have believed that “Justice Kavanaugh was right to question whether” race-based redistricting can continue in perpetuity, but at that moment in time, that belief was of no moment: there was no basis for those legislators to believe that they could ignore our affirmed ruling just to receive a second bite at the apple only a few weeks later. Id.

There is a good reason why the law does not work this way: the State's suggestion that it could relitigate liability and return to the Supreme Court after each adjustment to its previous plan on some metric other than the one we ordered (here, county splits and communities of interest instead of an additional opportunity district) would thrust both our Court and the Supreme Court into the State's infinity loop, with only the 2030 Census as a terminus.

The State reiterates that even if its legal arguments fail, “it is neither defiant nor racist to enact a law based on them and then test them in federal court.” Id. We cannot construe the intentional and admitted refusal to provide the remedy we said was required as only being a normal and legitimate legal test balloon. If it were normal, there would be other cases like this one: where a state legislature — faced with a federal court order declaring that its electoral plan unlawfully dilutes minority votes and requiring a plan that provides an additional opportunity district — responded with a plan that the state concedes does not provide that district. The extremely unusual nature of this case is a clue that this is a novel trick, not a normal strategy. It is a warning that losing parties who are unwilling to comply with court orders will simply try to avoid them by changing the rules of the game.

The federal courts have long held that an attempt to evade a court order is not legitimate. In the context of vote dilution, even before the passage of the Voting Rights Act, the Supreme Court rejected attempts to employ a new minority vote dilution tactic after a prior one had been held unconstitutional. See, e.g., Lane v. Wilson, 307 U.S. 268 (1939); Guinn v. United States, 238 U.S. 347 (1915). And in the more modern era, federal courts have refused to allow jurisdictions to engineer slightly modified remedial plans to skirt the clear mandate of a previous court order. See, e.g., McMillan v. Escambia County, 559 F. Supp. 720 (N.D. Fla. 1983). As the Supreme Court explained most recently – in its discussion of how a litigant repackaged a forbidden argument for strategic gain in a congressional redistricting case – “[o]ur decisions cannot be evaded with such ease.” Alexander, 602 U.S. at 21.

We likewise reject the State's suggestion that we should not accept the Plaintiffs’ arguments because they are divisive. We take seriously the concern that Section Two “may impermissibly elevate race in the allocation of political power within the States,” and we do not “diminish or disregard these concerns.” Allen, 599 U.S. at 41–42. We simply say again – based on forty years’ worth of congressional instructions and controlling precedents, including precedent in this very case – that they are not borne out here, not on this record.

In its final three arguments, the State identifies purported reasons for the 2023 Plan that are not intentional discrimination.

 The State's Argument that It Was Trying To Avoid a Gerrymandering Claim

Fifth, we reject the State's assertion that its desire to avoid a racial gerrymandering claim explains the 2023 Plan. The State stakes this argument in part on the Legislature's possible supposed awareness of simulations run by one of the Plaintiffs’ experts in the preliminary injunction proceeding, Dr. Kosuke Imai. But the Legislature cites no evidence that any legislator is aware of Dr. Imai's existence, let alone his simulations or testimony. See Milligan Doc. 481 ¶ 825. The other piece of the State's argument in this regard is about Singleton, see id. ¶ 828, but Singleton does not argue that the remedial district designed to cure vote dilution (District 2) is racially gerrymandered; Singleton’s theory of liability is that Jefferson County, in Districts 6 and 7, is racially gerrymandered (and has been for a very long time). See Singleton Doc. 288 at 10–11 (pretrial order).

Further, we find it implausible that in the unique circumstances of this case, concerns about racial gerrymandering claims drove the 2023 Special Session. We had already ruled that there were eleven lawful ways to remedy the vote dilution that we found, all of which revolved around a common, critical feature: they split Mobile County to join Black Alabamians living in Mobile with parts of the Black Belt in a majority-Black district. Accordingly, although the Legislature could have been alert to potential gerrymandering liability in 2021, there was no basis for the Legislature in 2023 to have been concerned that splitting Mobile County exposed it to a racial gerrymandering claim.

In any event, we have no evidence that the Legislature was specifically concerned about potential gerrymandering liability when it enacted the 2023 Plan. The only evidence in the record suggests they were not: when the Legislators testified about the 2023 Special Session, they were both asked repeatedly about why the Legislature moved away from the Community of Interest Plan, and neither one of them cited a concern that it was a gerrymander. See Milligan Doc. 459-13 at 16–17, 21–23 (Livingston); Milligan Doc. 459-20 at 25–27 (Pringle). Even Representative Pringle, who testified that he would evaluate any draft map prepared by Mr. Hinaman to determine whether it was a gerrymander, cited no concern that the Community of Interest Plan, Opportunity Plan, or any other specific plan under consideration, was a gerrymander. See Milligan Doc. 459-20 at 10, 20, 24–27. We have scoured the record and found no evidence that during the 2023 Special Session, any legislator considered a map that would provide an additional Black-opportunity district and refused to support it out of a concern that it would trigger a lawsuit based on a gerrymandering claim.

 The State's Argument that It Was Merely Pursuing a Legal Strategy

Sixth, we reject the State's argument “[t]he Legislature's good faith belief that drawing a second majority-[B]lack district would unconstitutionally segregate voters based on race, and that ‘the authority to conduct race-based districting cannot extend indefinitely into the future,’ ․ are obvious justifications for the 2023 Plan other than invidious discrimination.” Milligan Doc. 481 ¶ 834. We reject these arguments for the same reason we earlier refused to construe the Legislature's response to our order as a normal and legitimate legal test balloon.

We also find an additional fact. Justice Kavanaugh's observation that “the authority to conduct race-based redistricting cannot extend indefinitely into the future,” was issued in June 2023 and paired with (1) an express refusal to consider that argument because Alabama had not raised it, and (2) a vote to affirm redistricting in this very case. Allen, 599 U.S. at 45 (Kavanaugh, J., concurring in part) (citations omitted). Justice Kavanaugh did not say anything to suggest that our authority would expire in a matter of weeks, almost immediately upon the return of the case to our Court. It is inconceivable to us that it could have worked that way, in this very case, with zero warning or instruction on remand. Accordingly, although we expect the State to make that legal argument as an attack on Section Two (as it does), we find it wholly unpersuasive as an explanation for why the Legislature intentionally passed a plan that it admitted did not provide the remedy we said was required.

 The State's Argument that This Was Party Politics

Seventh, we reject the State's assertion that partisan goals rather than racial animus motivated the 2023 Plan. The State asserts that “finding legitimate ways to avoid the adoption of a map that would likely swing an additional congressional district to Democrats is a reasonable (and obvious) non-racial goal for Republican legislators to pursue,” as is the protection of Alabama's incumbent representatives (including Congresswoman Sewell). Milligan Doc. 481 ¶¶ 847–48, 851. The State warns that we will err if we simply “credit[ ] the less charitable conclusion that the legislature's real aim was racial.” Id. ¶ 854 (quoting Alexander, 602 U.S. at 22).

As an initial matter, there is precious little evidence to support the State's claim. The only evidence in the record that any legislator considered partisan advantage in 2023 is the testimony from the Legislators about the calls they received from former Speaker McCarthy about the slim Republican majority in the United States House, and the testimony from Representative Pringle and Senator Livingston that they spoke with various congressional and political party staff. See Milligan Doc. 459-13 at 24 (Livingston); Milligan Doc. 459-3 at 25–26 (Livingston); Milligan Doc. 459-20 at 6 (Pringle). But neither Legislator testified that they acted on those conversations, or even that they seriously considered acting on them. See Milligan Doc. 459-13 at 24 (Livingston); Milligan Doc. 459-20 at 6 (Pringle).

Further, when the Legislators were asked why the Senators moved away from the Community of Interest Plan, both professed not to know, and neither testified that they did it (or even might have done it) to benefit Republicans. Senator Livingston testified that the Senators received some new information about the attention they should pay to incumbents, but the State is quick to say in its proposed order that by “incumbents,” the State means all incumbents, including Congresswoman Sewell, who is a Democrat. Milligan Doc. 459-13 at 17; Milligan Doc. 481 ¶¶ 583, 847. The State's suggestion of partisan advantage is just that – a suggestion of a familiar reason, but one that the evidence does not bear out here.

In any event, this argument might have served the Legislature well in 2022 during the earlier stage of this litigation, but it is of no moment now. The Legislature may well have drawn the 2021 Plan the way it did for partisan reasons – we did not decide that then, do not decide it now, and do not know. But even if it did, after we told the Legislature that the 2021 Plan likely diluted Black Alabamians’ votes in violation of Section Two of the Voting Rights Act and ordered the remedy that Section Two requires, the law did not allow the Legislature to use partisan gain as a free pass to evade the requirements of Section Two.

If such a free pass were available, partisan advantage would be the ultimate trump card, both against Section Two and the Constitution. A legislature could evade liability under Section Two by passing a plan that utterly refused to provide the required remedy but entrenched or improved the partisan advantage of the majority party, and a legislature could defend against liability under the Constitution by arguing that its refusal to provide the statutory remedy was based on party rather than race.

We do not hold that the Legislature could not consider partisan advantage as it deliberated about a remedial plan. Plainly it could. We simply hold under the circumstances of this case that it could not use those considerations as an excuse for its strategic, purposeful, and admitted failure to provide the remedy federal law requires for the likely vote dilution we found.

Accordingly, we can discern no inconsistency between our ruling and Alexander. Alexander was a round-one case (a case that involved a legislature's first attempt to redistrict after receiving new census data, free of judicial intervention), and this is a round-two case (a case that involves a legislature's second attempt to redistrict after federal court orders told it both that and why the first attempt was likely discriminatory, and that the required remedy is an additional opportunity district).

***

When a legislature both purposefully refuses to satisfy the remedial requirements unambiguously found in a federal court order and then intentionally takes steps to make them mathematically impossible to satisfy, reality overwhelms the presumption of good faith. If the reality of this case does not rebut the presumption, we seriously doubt that it is rebuttable absent a clear and direct expression of invidious discrimination in the statutory text of a bill or official arguments in support of its passage. See, e.g., Peterson v. City of Greenville, 373 U.S. 244 (1963) (statutory text); Davis, 81 F. Supp. at 878–79 (official arguments for passage of bill).

Ultimately, we do not regard the presumption of legislative good faith, nor the requirement to disentangle party and race as drivers of legislative action, as free passes for state legislatures to evade court orders or invisibility cloaks that obscure searching judicial review. They are appropriate guardrails on the work of federal courts that we understand and have not infringed.

We have examined each of the Plaintiffs’ arguments and the State's defenses exhaustively and as best we can, and from every angle we have been offered and can conceive, and we hold that when the Legislature enacted the 2023 Plan, it intentionally refused to create an additional Black-opportunity district for the purpose of entrenching what it knew from federal court orders was very likely discriminatory vote dilution. We further find that the Legislature's intentional refusal (1) purposefully deprived Black Alabamians of the same opportunity to elect a candidate of their choice that White Alabamians enjoy, and (2) attempted to prevent us from ordering an appropriate remedy for that discriminatory vote dilution, (3) on the basis of race.

Although we have approached this issue exhaustively, we acknowledge that our holding is a rare one in the modern era, and we are painfully aware of the gravity of our ruling, but we do not find the issue particularly complex or close.

Lest our rulings be assigned meaning and force we do not intend, we state clearly their limitations. We do not hold that if SB5 had been originally adopted in 2021, the Milligan Plaintiffs would have prevailed on a claim of intentional discrimination at that time. We have not considered that question. As the foregoing analysis makes clear, the Milligan Plaintiffs prevail on their claim now not merely because SB5 dilutes Black Alabamians’ votes, but also because of the substantial evidence that the Legislature knew as much when it passed the plan and enacted it for that reason.

Nor do we hold that when a state legislature works diligently to balance competing hazards of liability while redistricting, it violates the Constitution if the balancing act fails. That is not what happened here. Nor do we hold that when a legislature simply refuses to redistrict after a preliminary injunction, or tries and fails to pass a lawful remedial plan, that sequence of events reflects intentional discrimination. That too is not what happened here.

Nor do we hold that when a state legislature exercises its appellate rights after a preliminary injunction instead of immediately providing the remedy the injunction requires, that decision violates the Constitution. That does not even begin to cover what happened here. The State was free to exercise its appellate rights and free to make any arguments it likes at trial and in a successive appeal. But the State was not free to simply double down and try to checkmate our order after it appealed and lost.

Nor do we hold that when a state legislature redistricts after a preliminary injunction and includes in the remedial map a district that the legislature maintains is an additional opportunity district, but which is later determined by a court not to be an opportunity district, a constitutional violation occurs. The State admits that is not what happened here.

Nor do we even hold that when a Legislature redistricts after a preliminary injunction and expressly and admittedly fails to include the additional opportunity district the injunction requires, a constitutional violation necessarily occurs. That is closer to what happened here, but still not the same.

We hold that when (1) a Legislature redistricts a State's congressional electoral map; (2) a federal court enters a preliminary injunction ruling that the State's map likely dilutes minority votes and unambiguously requiring an additional opportunity district as a remedy; (3) the State exhausts its appellate rights as to that ruling; (4) the ruling is affirmed by the Supreme Court; (5) on remand, the Legislature enacts a new map that it admits does not include an additional opportunity district; (6) the Legislature enacts (as part of that map) legislative findings that all agree make an additional opportunity district mathematically impossible to draw; and (7) key legislators admit their understanding of the remedy that was required and refused, then the Legislature may subject itself to a finding that it has enacted a map for the purpose of discriminating against minority voters by diluting their voting strength. After an exhaustive review of all the evidence, we have made just that finding.

VIII. REMEDY

Remedial proceedings lie ahead, both as to the map(s) Alabama will use for the rest of this census cycle and the Milligan Plaintiffs’ request for bail-in through 2032.⁷⁰ We comment now about the record on remedy only to say that it tells us something important about our finding of liability. We had no doubt in 2021 or 2023, and we have no doubt now, that a lawful district easily may be drawn to remedy the Section Two violation we find. In addition to the fourteen maps offered by Dr. Duchin and Mr. Cooper, we have the Special Master Plan. See supra Part I.F.

Preparing the Special Master Plan was an unwelcome task for this Court because “reapportionment is primarily the duty and responsibility of the State.” Miller, 515 U.S. at 915 (quoting Chapman v. Meier, 420 U.S. 1, 27 (1975)). “Quite apart from the risk of acting without a legislature's expertise, and quite apart from the difficulties a court faces in drawing a map that is fair and rational, the obligation placed upon the Federal Judiciary is unwelcome because drawing lines for congressional districts is one of the most significant acts a State can perform to ensure citizen participation in republican self-governance.” LULAC, 548 U.S. at 415–16 (citation omitted). “That Congress is the federal body explicitly given constitutional power over elections is also a noteworthy statement of preference for the democratic process. As the Constitution vests redistricting responsibilities foremost in the legislatures of the States and in Congress, a lawful, legislatively enacted plan should be preferable to one drawn by the courts.” Id. at 416.

Preparing the Special Master Plan was a most unwelcome task for us because (1) the Legislature spent its opportunity to provide a lawful remedy trying to have a second bite at the apple on liability on arguments it had lost not only in our Court, but also in the Supreme Court; (2) the record in these cases provides not just one illustrative remedial map, but (at the time of the 2023 Special Session), eleven such maps, giving the Legislature extensive guidance on how to provide a lawful remedy; and (3) because we and the Supreme Court had found at least one of those maps lawful, and they all split Mobile County, the Legislature was not navigating confusion, uncertainty, or a legitimate concern about liability for racial gerrymandering if it split Mobile County. (It should be clear enough by now that all stakeholders know that all paths to cracking and packing minority votes in South Alabama run through Mobile County, and all paths to remedying such dilution revolve around Mobile County.)

Nevertheless, preparing the Special Master Plan served an important purpose above and beyond providing a lawful remedy for the 2024 election (which was important enough): it confirmed for us that a lawful remedial plan may be prepared race-blind.

Indeed, the Special Master prepared all three plans that he recommended race-blind. See Milligan Doc. 436 ¶ 143. And when the Special Master confirmed that the Special Master Plan satisfied all constitutional and statutory requirements while hewing as closely as reasonably possible to the 2023 Plan, see Milligan Doc. 295, supra Part I.F, the Special Master provided data for every available metric to support the point:

Milligan Doc. 436 ¶ 146.

The parties have since stipulated that: (1) Mr. Ely “drafted the [Special Master] Plan without reference to any illustrative or proposed plan,” id. ¶ 140; (2) “[a]ccording to the Special Master's report, Mr. Ely did not display racial demographic data within the mapping software, Maptitude, while drawing his remedial proposals (including the [Special Master] Plan) or while he examined proposed remedies submitted by others,” id. ¶ 143; and (3) according to the Special Master's report, “Mr. Ely drew his proposals ․ based on other nonracial characteristics and criteria related to communities of interest and political subdivisions,” id.

Although federal law does not require a Section Two remedial plan to be prepared race-blind, the ability of the Special Master to do it that way (on a very short timetable) confirms for us that it is not only possible, but relatively easy. We thus have no concern that race will predominate in the preparation of a remedial plan, nor that a remedial plan will segregate Alabama voters on the basis of race.

IX. EVIDENTIARY RULINGS

During the trial, the Court accepted into evidence the overwhelming majority of the exhibits that the parties offered; most were stipulated, and the Court ruled on some evidentiary objections and reserved ruling on others. We make explicit note of one ruling from trial. The Milligan and Caster Plaintiffs filed a joint motion in limine to exclude part of Dr. Reilly's testimony about communities of interest. Milligan Doc. 416; Caster Doc. 323. The Singleton Plaintiffs joined that motion, Milligan Doc. 421, and the State responded in writing, Milligan Doc. 423. We heard argument on the motion at trial. Tr. 1253–74. For the reasons stated on the record at trial, and because we afforded Dr. Reilly's testimony its due weight in our analysis, the joint motion in limine is DENIED.

All further pending objections are SUSTAINED.

DONE and ORDERED this 8th day of May, 2025.

/s/ Stanley Marcus

STANLEY MARCUS

UNITED STATES CIRCUIT JUDGE

/s/ Anna M. Manasco

ANNA M. MANASCO

UNITED STATES DISTRICT JUDGE

/s/ Terry F. Moorer

TERRY F. MOORER

UNITED STATES DISTRICT JUDGE

APPENDIX A – COMMITTEE GUIDELINES (2021 and 2023)

APPENDIX B – LEGISLATIVE FINDINGS (2023 PLAN/SB-5)

APPENDIX C – CONGRESSIONAL MAP (2023 PLAN)

APPENDIX D – COMMUNITY OF INTEREST PLAN and LIVINGSTON PLANS 1 & 2

APPENDIX E – EXCERPTED FIGURES FROM MR. COOPER

Caster Doc. 352-2 at 26–28 figs. 14 & 15.

Caster Doc. 352-2 at 18 fig. 9.

Caster Doc. 352-2 at 14 fig. 6.

FOOTNOTES

1.   The fourth case is Alabama State Conference of the NAACP v. Allen, Case No. 2:21-cv-1531-AMM which challenges the state Senate map on statutory grounds.

2.   When we cite a document that appears in more than one of these cases, we cite only the document filed in Milligan.

3.   Page number pincites are to the CM/ECF page number that appears in the top right-hand corner of each page, if available.

4.   In August 2023, Senator Steve Livingston (the new Senate chair of the Committee) was substituted for Senator McClendon as a defendant. Milligan Doc. 269 at 2.

5.   The Secretary had appealed Caster to the United States Court of Appeals for the Eleventh Circuit. See Allen v. Caster, No. 23-12923.

6.   Exhibits that are identified by a letter and number are trial exhibits.

7.   Citations to the trial transcript are identified by page number. Other transcripts are identified by date. The transcript for the trial may be found at Singleton Docs. 302, 304–305, 307–312, 318–319; Milligan Docs. 463, 465–466, 468–473, 479–480; and Caster Docs. 376, 378–379, 381–386, 392–393.

8.   The parties had stipulated that Alabama's Black Belt “is named for the region's fertile black soil. The region has a substantial Black population because of the many enslaved people brought there to work in the antebellum period. All the counties in the Black Belt are majority-or near majority-BVAP.” Milligan Doc. 53 ¶ 60. They further stipulated that the Black Belt includes eighteen “core counties” (Barbour, Bullock, Butler, Choctaw, Crenshaw, Dallas, Greene, Hale, Lowndes, Macon, Marengo, Montgomery, Perry, Pickens, Pike, Russell, Sumter, and Wilcox), and that five other counties (Clarke, Conecuh, Escambia, Monroe, and Washington) are “sometimes included.” Id. ¶ 61.

9.   We distinguish Part III-B-1, the opinion of four justices, from a plurality opinion. “A plurality opinion is one that doesn't garner enough appellate judges’ votes to constitute a majority, but has received the greatest number of votes of any of the opinions filed, among those opinions supporting the mandate.” Bryan A. Garner, et al., The Law of Judicial Precedent 195 (2016) (internal quotation marks and footnote omitted). All other parts of the Chief Justice's opinion garnered five votes.

10.   At trial, the State objected to the Caster Plaintiffs offering the parties’ previous stipulations (Caster Exs. 124–26) into evidence because “[t]hose stipulations were made for purposes of the preliminary injunction hearing, not for trial.” Tr. 1505. Judge Marcus replied to counsel for the State: “I understand you object to the substance of the stipulation because you have new stipulations. Let us know if you object to these as evidence that at one time they were stipulations.” Id. Later, when the issue was re-raised, counsel for the State provided a fuller objection:[W]e stipulated to those facts only for purposes of preliminary injunction and those earlier proceedings. We entered into a new round of stipulations where we had the benefit of all knowledge gained through discovery. We reset the clock, in other words. Earlier on, the proceedings were hurried. Everybody had a short time frame. And so, to make sure there weren't mistakes made or that they weren't based on incomplete knowledge, we have restarted the clock and entered into a new set of stipulations. We do not feel it's fair to hold [those against] us if we are no longer willing to enter into these stipulations.Tr. 2415–16. When provided the opportunity to respond, counsel for the Caster Plaintiffs stated: “We believe the [State] ha[s] agreed to these facts in the past and that, whether or not [it] stand[s] by them today, the fact is that [it] ha[s] previously stipulated that these were all facts.” Id. at 2416.Because the parties revised the stipulations, the Court does not rely on the previously stipulated definition of the Black Belt to understand the Black Belt. We simply observe that the Legislature did not mention race or slavery in its findings while, at the same time, the State stipulated to such facts in court.

11.   As we previously explained, Milligan Doc. 107 at 30 n.5, when we recite statistics about Black Alabamians from data collected in or after the 2000 census, we are referring to any census respondent who identified themselves as Black, regardless whether that respondent also identified as a member of another race or other races. To use the label that the parties supplied in the preliminary injunction proceedings, we employ the “any-part Black” metric rather than the “single-race Black” metric.

12.   Mr. Walker represented the Legislators in these cases.

13.   That order, as testified to by Senator Livingston, provided:The Legislature enjoys broad discretion and may consider a wide range of remedial plans. As the Legislature considers such plans, it should be mindful of the practical reality, based on the ample evidence of intensely racially polarized voting adduced during the preliminary injunction proceedings, that any remedial plan will need to include two districts in which Black voters either comprise a voting-age majority or something quite close to it.Milligan Doc. 459-13 at 13 (reading from Milligan Doc. 107 at 6).

14.   Former United States Senator Doug Jones is a White Democrat who was elected to represent Alabama in the United States Senate in a special election in 2017 after then-Senator Jeff Sessions resigned his seat to become Attorney General of the United States. Senator Jones is well-known for his work prosecuting members of the Ku Klux Klan who bombed 16th Street Baptist Church in Birmingham in 1963, killing four Black girls. Senator Jones was the first Democrat elected to any statewide office in Alabama in nearly a decade, and he lost the 2020 General Election to Senator Tommy Tuberville, a White Republican who remains in the Senate today.

15.   At trial, Dr. Bagley testified that “monkey town” is a name with a history as a racist pejorative. Tr. 1338. Senator Singleton testified that he did not consider it racist to refer to Montgomery as “monkey town” “in that context.” Id. at 2374.

16.   Most of the Birmingham metropolitan area is in Jefferson County.

17.   The Milligan and Caster parties stipulated for trial that “[t]he Black Belt is named for the region's fertile black soil”; “has a substantial Black population because of the many enslaved people forcibly brought there to work before the Civil War”; and “includes the core counties of Barbour, Bullock, Butler, Choctaw, Crenshaw, Dallas, Greene, Hale, Lowndes, Macon, Marengo, Montgomery, Perry, Pickens, Pike, Russell, Sumter, and Wilcox,” along with Clarke, Conecuh, Escambia, Monroe, and Washington counties, which “are sometimes included within the definition of the Black Belt.” Milligan Doc. 436 ¶¶ 71–73.This definition differs from the definition in the 2023 legislative findings in that the legislative findings do not mention race or slavery, and the stipulated definition does not mention “rural geography,” “relative poverty,” or it “span[ning] the width of Alabama.” Compare Milligan Doc. 436 ¶¶ 71–73, with App. B at 3–4.

18.   The Singleton Plaintiffs conceded this point in their closing arguments. Tr. 2600–02.

19.   Dr. Duchin incorporated her prior Milligan reports into her trial report. Milligan Doc. 385-3 at 1.

20.   “[T]he retention rows show the share of population (in the 2020 Census) whose address keeps them in the same district as in various benchmark plans.” Milligan Doc. 385-3 at 4.

21.   The Census is a count of the United States population. As one of the expert witnesses explained, the American Community Survey “is a survey that's fielded by the Census Bureau annually. It asks more detailed questions than the Decennial Census such as educational attainment.” Tr. 931. The American Community Survey often reports five-year estimates “because the American Community Survey gets asked every year but not every county is represented in every year. So you have to take five years of data to make sure you get every county.” Id. at 933.

22.   Dr. Liu issued his report before the November 2024 General Election.

23.   Dr. Liu testified that he made an error in this table (Greg Cook should be named in the 2022 Supreme Court Place 5 election instead of Bradley Byrne, Tr. 661), and that this typographical error had no substantive effect as the numbers, data, and percentages are correct. Tr. 680.

24.   Dr. Bagley issued his report before Governor Ivey appointed Judge Lewis.

25.   Milligan Doc. 403-4 at 28 (quoting Jeff Amy & Kim Chandler, Alabama Lawmakers Refuse to Create 2nd Majority-Black Congressional District, AP News (Jul. 21, 2023, 12:16 AM), https://apnews.com/article/alabama-legislature-redistricting-voting-rights-e2fc7c7550e10da353b72bafc3fb6604); Tr. 1342–43.

26.   In connection with the State's arguments that racially polarized voting patterns are attributable to partisanship rather than race, and all parties’ arguments about relevant Alabama history, the experts discuss the partisan realignment that occurred in Alabama and the South when, over a period of decades, White Democratic voters began primarily voting for Republican candidates. See Milligan Doc. 384-2 at 36–37 (Dr. Carrington); Milligan Doc. 385-5 at 2–5 (Dr. Bagley); Milligan Doc. 384-1 at 4, 20 (Dr. Bonneau).

27.   Representative Jones was deposed on August 29, 2024. Milligan Doc. 459-9.

28.   Mr. Hinaman was deposed on December 9, 2021, and August 9, 2023. Milligan Docs. 459-6 and 459-7.

29.   Senator McClendon was deposed on December 17, 2021, and April 18, 2024, Milligan Docs. 459-15 and 459-16.

30.   Senator Livingston was deposed on August 9, 2023. Milligan Doc. 459-13.

31.   Mr. Brown was deposed on June 18, 2024. Milligan Doc. 459-3.

32.   Representative Pringle was deposed on December 17, 2021, and August 9, 2023. Milligan Doc. 459-19 and 459-20.

33.   During Mr. Simelton's testimony, counsel raised various privilege objections. See Milligan Doc. 441-7 at 60–64, 66–69. We do not rely on any privileged information.

34.   Dr. Palmer's reports and analysis predate the 2024 General Election.

35.   The Tuskegee Airmen were Black pilots who joined “a new initiative launched in Alabama” that was “a first of its kind training program for Black pilots established [by the military] after the bombing of Pearl Harbor.” 171 Cong. Rec. S1307 (daily ed. Feb. 24, 2025) (statement of Sen. Britt). These pilots “could not live, work, eat, or drink alongside [W]hite countrymen or women” in the 1940s, “and yet still decided to risk everything to serve this Nation.” Id. Though one of the most accomplished units of World War II, “[i]t wasn't until 2007 that the Tuskegee Airmen received the Congressional Gold Medal for their valor.” Id. In the Act granting that award, Congress found that “[t]he Tuskegee Airmen inspired revolutionary reform in the Armed Forces, paving the way for full racial integration in the Armed Forces,” and noted that “[t]hese Black airmen came home with 150 Distinguished Flying Crosses, Bronze Stars, Silver Stars, and Legions of Merit, one Presidential Unit Citation, and the Red Star of Yugoslavia.” CONGRESSIONAL GOLD MEDAL AWARD—TUSKEGEE AIRMEN, PL 109–213, April 11, 2006, 120 Stat 322.

36.   The parties designated Mr. Smith's deposition testimony, Milligan Doc. 459 at 4, but Mr. Smith testified at trial. Accordingly, the Court recounts only his trial testimony.

37.   Mr. Carroll was deposed on September 5, 2024. Milligan Doc. 459-4.

38.   Mr. Carroll is referring to District 1 prior to the Special Master Plan.

39.   Mr. DuBose was deposed on August 7, 2024. Milligan Doc. 459-5.

40.   Mr. Jones was deposed on July 23, 2024. Milligan Doc. 459-8.

41.   Mr. Love was deposed on August 7, 2024. Milligan Doc. 459-14.

42.   Mr. Powell was deposed on July 25, 2024. Milligan Doc. 459-18.

43.   Mr. Small was deposed on September 5, 2024. Milligan Doc. 459-23 at 1.

44.   Senator Smitherman was deposed on July 29, 2024. Milligan Doc. 459-25 at 2.

45.   Dr. Hood and his co-authors explained the “Lost Cause Myth” as having “three tenets”: “the Confederacy's cause was noble and just and the war was fundamentally about states’ rights, not slavery. Second, slavery was benevolent and slaves were content in their station, so much so that the Civil War and Reconstruction upset a natural racial hierarchy. Third, Confederates were among the greatest soldiers in history and they were only defeated due to the Union's superior manpower and resources[.]” Caster Doc. 374-2 at 15 (quoting Adam H. Domby, The False Cause: Fraud, Fabrication, and White Supremacy in Confederate Memory 4 (2020)) (internal citation omitted).

46.   His report also cites Reddit and Quora. Milligan Doc. 384-4 at 21.

47.   As the reader is aware, Mr. Gray and Mr. Clemon were among Alabama's first Black state legislators. See supra Part I.I.1; Part IV.A.3.b. By way of further background, Mr. Gray is a Montgomery civil rights lawyer known for major civil rights litigation, including his representation of Rosa Parks, Martin Luther King, Jr., and the victims of the Tuskegee Syphilis Study. See Fred Gray, Encyclopedia of Alabama, https://encyclopediaofalabama.org/article/fred-gray/. After his time in the Legislature and practicing law, U.W. Clemon became the first Black federal judge in Alabama and served for nearly thirty years. Judge U.W. Clemon, United States District Court for the Middle District of Alabama, https://www.almd.uscourts.gov/oral-histories-profiles/judge-uw-clemon. Mr. Clemon is a lawyer in this case and was present during Dr. Carrington's testimony. Finally, Frank Johnson was a federal judge known for historic civil rights rulings, including in cases involving the Montgomery Bus Boycott and the march from Selma to Montgomery. About Judge Johnson, The Judge Frank M. Johnson, Jr. Institute, https://www.thejohnsoninstitute.org/. Judge Johnson's rulings “repeatedly defied racist Alabama Governor George Wallace, a law school classmate,” and the Ku Klux Klan called him “the most hated man in Alabama.” Judge Frank Johnson – International Civil Rights Walk of Fame, National Park Service, https://www.nps.gov/features/malu/feat0002/wof/Frank_Johnson.htm.

48.   Mr. Boswell was deposed on August 12, 2024. Milligan Doc. 459-1.

49.   Mr. Kimbro was deposed on August 11, 2023, and August 29, 2024. Milligan Docs. 459-10, 459-11.

50.   Mr. Lawson was deposed on September 4, 2024. Milligan Doc. 459-12.

51.   Mr. Nix was deposed on July 17, 2024. Milligan Doc. 459-17.

52.   Mr. Schmitz was deposed on August 10, 2023. Milligan Doc. 459-21.

53.   Mr. Turner was deposed on August 26, 2024. Milligan Doc. 459-26.

54.   That agency later became the Department of Labor, Milligan Doc. 459-26 at 10, and is now the Department of Workforce, see Ala. Code § 25-2-1.2.

55.   Mr. Williams was deposed on September 6, 2024. Milligan Doc. 459-28.

56.   Obviously, these are not the only three communities of interest in Alabama. Yet, in an apparent reference to this ongoing litigation, these are the only communities of interest that the Legislature chose to define.

57.   We refresh the reader's recollection about the parties’ stipulated definition of the Black Belt, which lists the 18 core counties and five “sometimes included” counties and provides: “The Black Belt is named for the region's fertile black soil,” and “has a substantial Black population because of the many enslaved people forcibly brought there to work before the Civil War.” Milligan Doc. 436 ¶¶ 71–73.

58.   In Pierce, the Fourth Circuit explained that a “crossover district” is one in which “the minority population is not a majority but is large enough to elect the candidate of its choice with help from voters who are members of the majority and who cross over to support the minority's preferred candidate.” 97 F.4th at 204.

59.   See also, e.g., Pierce, 97 F.4th at 223; United States v. Charleston County, 365 F.3d 341, 347–49 (4th Cir. 2004); Solomon v. Liberty Cnty. Comm'rs, 221 F.3d 1218, 1225 (11th Cir. 2000) (en banc); Lewis v. Alamance County, 99 F. 3d 600, 615 n.12 (4th Cir. 1996); Nipper v. Smith, 39 F.3d 1494, 1536 (11th Cir. 1994).

60.   Because there is not a slating process for Alabama's congressional elections, Senate Factor 4 is not relevant. See Milligan Doc. 481 ¶¶ 504–05.

61.   There is one Black statewide official in Alabama today who occupies an elected office – Judge Lewis. Governor Ivey appointed Judge Lewis in 2024 and he will need to run for election to keep his position. Milligan Doc. 436 ¶ 153.

62.   The Caster Plaintiffs provided the ad, which is entitled “Dale,” on a USB drive for the Court's viewing. Caster Doc. 319-133.

63.   The CRA's attorney-fee provision reads as follows: “In any action commenced pursuant to this subchapter, the court, in its discretion, may allow the prevailing party, other than the United States, a reasonable attorney's fee as part of the costs, and the United States shall be liable for costs the same as a private person.” 42 U.S.C. § 2000a-3(b).

64.   The Supreme Court issued Health & Hospital Corporation of Marion County after the Eighth Circuit heard oral argument but before the Eighth Circuit issued its decision. See Health and Hosp. Corp. of Marion Cnty., 599 U.S. at 166; Ark. State Conf. NAACP, 86 F.4th at 1204.

65.   The Supreme Court already has rejected, in this very case, the argument that Section Two exceeds congressional authority under the Fifteenth Amendment. See Allen, 599 U.S. at 41.

66.   In addition to observing that Sections Two and Five conferred private rights of action, the Court in Morse supported its conclusion that Section Ten confers a private right of action by reasoning that: the achievement of the Voting Rights Act's goals would be severely hampered if only the Attorney General could sue to enforce Section Ten; the Attorney General had urged the Court to find a private right of action; and other sections of the Voting Rights Act (specifically, Sections Three and Fourteen) contain language recognizing that private persons can sue to enforce their rights under the Voting Rights Act. See Morse, 517 U.S. at 231–34.

67.   Most recently, a three-judge district court in the Southern District of Mississippi has followed Robinson and relevant Supreme Court precedent in holding that Section Two confers a private cause of action. See Miss. State Conf. NAACP v. State Bd. of Election Comm'rs., No. 3:22-cv-00734-DPJ-HSO-LHS (July 2, 2024) (per curiam).

68.   Indeed, the Department of Justice has previously observed that private plaintiffs have brought over 400 Section Two cases resulting in judicial decisions since 1982, while the Department of Justice itself has brought just 44 cases. See Brief of United States as Amicus Curiae at 1–2, Turtle Mountain Band of Chippewa Indians v. Howe, No. 23-3655, 2024 WL 1417744 (8th Cir. Mar. 25, 2024) (citing Ellen D. Katz et al., To Participate and Elect: Section 2 of the Voting Rights Act 40, Univ. Mich. L. Sch. Voting Rts. Initiative (2024), https://voting.law.umich.edu; Voting Section Litigation, U.S. Dep't of Just. (2024), https://perma.cc/V5XK-Z7L8).

69.   In criminal cases, where the standard requires proof beyond a reasonable doubt, courts regularly instruct jurors about direct and circumstantial evidence, and circumstantial evidence is sufficient to convict. See 11th Cir. Crim. Pattern Instr. B4; accord, e.g., United States v. Mapson, 96 F.4th 1323, 1336 (11th Cir. 2024). It cannot be that in a civil case, where the standard requires lesser proof, circumstantial evidence is less reliable or probative.

70.   We are mindful that as a matter of state law, the Legislature may redraw Alabama's congressional districting plan at any time. The Alabama Constitution prohibits mid-decade redistricting for state legislative seats, but makes no such provision for congressional seats. See Ala. Const. art. IX, §§ 198, 200.

PER CURIAM: