BOBBY SINGLETON v. WES ALLEN (2026)

ORDER

BY THE COURT:

These long-running congressional redistricting cases have returned to us after another trip to the Supreme Court. In earlier rulings, the Supreme Court affirmed our first preliminary injunction and declined to stay our second. Allen v. Milligan, 599 U.S. 1, 10, 17 (2023); Allen v. Milligan, 144 S. Ct. 476 (2023) (mem.). More recently, the Supreme Court received appeals of our order permanently enjoining Alabama from implementing its 2023 congressional districting plan, which we concluded violated Section Two of the Voting Rights Act of 1965 and intentionally discriminated against Black voters based on race in violation of the Fourteenth Amendment to the United States Constitution. 52 U.S.C. § 10301; U.S. Const. amend. XIV. While those appeals were pending, the Supreme Court issued Louisiana v. Callais, 146 S. Ct. 1131 (2026), which updated the standard for proving a Section Two claim. It then vacated our permanent injunction and instructed us to give these cases further consideration in light of Callais.

We now face a critical decision on a very tight timeline. We can either allow the Secretary of State to administer Alabama's 2026 elections with a legislatively enacted districting plan that we found (after a full trial) intentionally discriminated against Black voters based on race in violation of the Constitution, or we can issue a preliminary injunction two and a half months ahead of Alabama's scheduled special primaries (and some five months before the general election), requiring the Secretary to administer the 2026 elections with the race-blind plan that he used on orders from us and the Supreme Court for Alabama's 2024 elections and May 19, 2026, primary elections. We emphasize that because of the exceptional public importance of this matter, we carefully reviewed the extensive evidentiary record in these cases with fresh eyes in light of Callais.

After that exacting review, we conclude that a preliminary injunction must issue. Ultimately, we cannot see our way clear to requiring Alabamians to cast their votes in the 2026 elections under a districting plan tainted by intentional race-based discrimination. And under the unusual circumstances of this case, we conclude that a limited order requiring the Secretary to continue using this Court's race-blind map will not disrupt Alabama's elections (all candidates ran under the race-blind map until fifteen days ago, and all voters remain districted under the race-blind map in electoral computer systems).

We do not lightly intrude in state affairs, but our previous review of the undisputed evidence left us in no doubt that Alabama's legislatively enacted plan (the “2023 Plan”) intentionally discriminated based on race in violation of the Constitution. Our re-examination in light of Callais yields the same conclusion.

We again cannot understand the 2023 Plan as anything other than intentionally discriminatory. When the Legislature enacted the 2023 Plan, it made a calculated, purposeful decision to refuse to provide the remedy for discriminatory vote dilution that our order (affirmed by the Supreme Court) required. The Legislature well knew that a plan without an additional Black-opportunity district would dilute Black Alabamians' opportunity to participate in the political process, and it intentionally enacted that very plan. Further, the Legislature well knew what dilutive mechanisms would prevent Black voters in Alabama's Black Belt and Gulf Coast communities from having any opportunity to elect representatives of their choice, and the Legislature employed precisely those mechanisms. The Legislature also took a series of highly unusual steps along the way. Those unprecedented steps culminated in the enactment of novel legislative findings (“the 2023 legislative findings”) that departed sharply from the Legislature's norms and made it impossible not only to remediate the vote dilution we identified, but also to respect the longstanding community of interest the Legislature identified in Alabama's Black Belt. These events, along with legislators' contemporaneous statements about race, support only one inference: the purpose of the 2023 Plan was to distribute Black voters across districts to dilute their votes, at least in part because they are Black.

Counsel argues mightily that the Legislature's partisan motives drove the creation of the 2023 Plan, but this enormous record contains no evidence of a partisan motive. And the only evidence on the issue cuts against one: Alabama's legislative leadership testified that overtures from national party leaders did not affect their work.

We reach this conclusion with great reluctance and dismay and even greater restraint — only after another exhaustive analysis of an extensive record, as the Supreme Court's remand order and its precedent instructs us. We reject in the strongest possible terms the State's attempt to finish its intentional decision to dilute minority votes with a veneer of legislative regularity.

Alabama charges that Callais upends our finding of intentional discrimination because it is fully derivative of our Section Two finding, but that charge is wrong. And in any event, after again reviewing the extensive evidence regarding Plaintiffs' Section Two claims, we find that the Plaintiffs have likely established their Section Two claims when measured against the Callais standard. We explain below.

Timing issues have been ever-present in this case, and we must consider them again here. We issued preliminary injunctions ahead of the 2022 and 2024 elections and then a permanent injunction ahead of the 2026 election. Each injunction traveled to the Supreme Court. The Court ruled that our first preliminary injunction came too late, and Alabama's 2022 elections occurred under a legislatively enacted map the Supreme Court later declared unlawful under Section Two (“the 2021 Plan”). But the Court declined to stay our second injunction (that prohibited use of the 2023 Plan), and Alabama's 2024 elections occurred under a court-ordered map (“the Special Master Plan”). Now we must consider if time is too short to rule again for the 2026 election.

We conclude that it is not. On the unique record before us, we determine that enjoining the 2023 Plan will not disrupt Alabama's elections. Requiring the use of the Special Master Plan will forestall an expensive, aggressive, and perhaps logistically impossible voter reassignment effort. We take extremely seriously the Supreme Court's command that federal district courts ordinarily should not intervene on the eve of an election, for risk of causing administrative challenges and confusion. But the record here is clear: enjoining the unconstitutional 2023 Plan will improve the administrative situation in Alabama, not worsen it.

As we see it, the irreducible minimum is that federal law requires that all Alabamians have an opportunity to vote under districting plans untainted by intentional race-based discrimination. Accordingly, we are duty-bound to preliminarily enjoin the Secretary from conducting any 2026 congressional elections according to the 2023 Plan, and we further order the Secretary to administer all remaining events comprising Alabama's 2026 elections according to the Special Master's race-blind plan. We also issue a scheduling order for timely proceedings before qualifying opens for the 2028 primaries in November 2027, and we will amend it as necessary upon any further districting by the Legislature.

I. TIMING - PURCELL

We first consider, as we must, whether the proximity to Alabama's 2026 elections categorically bars us from issuing a preliminary injunction requiring the Secretary to administer those elections under the Special Master Plan. Under Alabama law, the May 19, 2026, primaries for Districts 1, 2, 6, and 7 became ineffective on May 12, 2026, when the Governor set special primaries for those districts for August 11, 2026. Ala. Code § 17-13-3.1(d); Milligan Doc. 529-1.

The Supreme Court clearly understood two weeks ago that Alabama's 2026 primaries were underway. By expediting its proceedings and issuing judgment forthwith, we understand that the Court expected us to give these cases further consideration in connection with the 2026 elections. If the Court did not mean for us to conduct our work expeditiously, we trust that it would have said so, not expedited its proceedings, issued judgment in the ordinary course, and/or stayed our injunction.

Nevertheless, we are obliged to carefully consider anew the principle that “federal district courts ordinarily should not enjoin state election laws in the period close to an election.” Merrill v. Milligan, 142 S. Ct. 879 (2022) (mem.) (Kavanaugh, J., concurring) (citing Purcell v. Gonzalez, 549 U.S. 1 (2006) (per curiam)). The Purcell principle is weighty, but “not absolute.” Id. at 881. It can be overcome if the Plaintiffs establish that “(i) the underlying merits are entirely clearcut in [their] favor; (ii) [they] would suffer irreparable harm absent the injunction; (iii) [they have] not unduly delayed ․; and (iv) the changes in question are at least feasible before the election without significant cost, confusion, or hardship.” Id. at 881 (citation modified); see also League of Women Voters of Fla., Inc. v. Fla. Sec'y of State, 32 F.4th 1363, 1371 (11th Cir. 2022) (per curiam).

It is rare that a plaintiff can make this showing. But the facts of this case are extremely unusual, perhaps unique. As we explain below, although the 2023 Plan became the legal status quo after the Supreme Court vacated our permanent injunction, the Special Master Plan remains the practical status quo on the ground in Alabama. And because the Special Master Plan remains administratively in place, the Plaintiffs can carry their burden.

We examine each Purcell factor in reverse order and enter these findings:

4. We are satisfied after examining this record and having just taken testimony from the Alabama Director of Elections, Mr. Jeff Elrod, on Friday, May 22, that the Secretary can administer Alabama's 2026 elections according to the Special Master Plan without significant cost, confusion, or hardship.

Cost and hardship. As the Secretary's office has testified, voters are currently districted in county electoral systems across the State according to the Special Master Plan. See Milligan Doc. 530-1 ¶¶ 6, 30–31, 41. An order requiring him to continue using the Special Master Plan would require him only to leave that system as it already is. Absent such an order, we understand from Director Elrod, it will take a chaotic, decentralized, and Herculean effort for officials in his office and fourteen counties to reassign voters according to the 2023 Plan. That reassignment process has not yet begun because Alabama's voter records are currently locked due to the May 19 primaries and will remain locked until May 27. Moreover, Director Elrod testified that the voter reassignment process must conclude by June 2, when the voter records will lock again. Id. ¶¶ 1, 6–7, 18, 41–63; May 22, 2026 Tr. at 37, 78.¹ This timeline gives local officials a maximum of seven days to complete a process that, on previous occasions, has taken several months. May 22, 2026 Tr. at 36, 56, 61, 80.

Director Elrod testified, and counsel for the State acknowledged, that the largest problem associated with implementing the 2023 Plan (a plan that has never been used before) is the urgent need to reassign voters who must be moved from one congressional district in the Special Master's Plan to another district under the 2023 Plan. This is costly: Director Elrod's predecessor (Mr. Clay Helms) previously testified in these cases that changing congressional district lines comes with significant additional expenses for the State, Milligan Doc. 82-7 at ¶¶ 2, 8, 18, and Director Elrod testified about the State's coordination with GIS systems vendors, Milligan Doc. 530-1 at ¶¶ 58–60. We think it plain that as between a warp-speed reassignment process and the administrative status quo, the former will cost the State far more money and hardship than the latter.

The record is crystal clear: administering Alabama's 2026 congressional elections under the Special Master Plan is simpler by an order of magnitude than administering the elections under the 2023 Plan.

Confusion. Candidates and voters in Alabama were not confused before Callais: Alabama's 2024 congressional elections were administered statewide without any apparent confusion, and the 2026 elections ran for months under the Special Master Plan as well. In the two weeks between Callais and the vacatur of our injunction, confusion emerged, especially during the Special Session. Since vacatur, the confusion has only grown.

Candidate and voter confusion is troublesome and warrants significant consideration, but we do not see that a preliminary injunction will worsen it. To the contrary, we expect a preliminary injunction to lessen it. Until May 11, the Special Master Plan was the well-established status quo in Alabama. As Director Elrod testified, an injunction requiring the use of the Special Master Plan simply keeps the candidates and the voters in the districts they have been in for nearly three years, minus a two-week interregnum. See May 22, 2026 Tr. at 32.

Director Elrod acknowledged that an injunction would avoid the confusion and error risk attendant to pivoting computer systems from the Special Master Plan to the 2023 Plan this close to the scheduled special primaries. See, e.g., id. at 80–81. And he acknowledged that when the voter reassignment process is rushed, voters may be accidentally assigned to the wrong district and provided the wrong ballot. Milligan Doc. 530-1 at ¶ 62; May 22, 2026 Tr. at 107–08. That happened before, Milligan Doc. 530-1 at ¶ 62, even without the unprecedented speed associated with the State's reassignment efforts.

In any event, federal law supplies no basis for confusion to override a clearcut right to relief from an intentionally discriminatory map. We trust the Secretary will issue appropriate information and credit the testimony of Director Elrod that the Secretary's office will “direct counties to implement whatever is approved for use.” May 22, 2026 Tr. at 81.

3. No one has delayed. The Milligan Plaintiffs moved this Court for a temporary restraining order just over six hours after the Supreme Court vacated our permanent injunction. Milligan Doc. 527. The Singleton and Caster Plaintiffs followed suit the next morning. Singleton Doc. 354; Caster Doc. 432.

2. It is beyond any real doubt that the Plaintiffs would suffer irreparable harm if required to vote under a map that is tainted by intentional race-based discrimination. “Courts routinely deem restrictions on fundamental voting rights irreparable injury. And discriminatory voting procedures in particular are the kind of serious violation of the Constitution and the Voting Rights Act for which courts have granted immediate relief.” Singleton v. Merrill, 582 F. Supp. 3d 924, 1026 (N.D. Ala. 2022) (per curiam) (quoting League of Women Voters of N.C. v. North Carolina, 769 F.3d 224, 247 (4th Cir. 2014)). “[O]nce the election occurs, there can be no do-over and no redress.” League of Women Voters, 769 F.3d at 247; Robinson v. Ardoin, 86 F.4th 574, 600 (5th Cir. 2023) (same).

1. Finally, we have re-examined this voluminous record and measured it anew against the standard the Supreme Court has just announced. Plaintiffs' right to relief remains entirely clearcut after Callais. This is for two separate and independent reasons that we explain below.

First and of primary importance, Callais does not disturb our finding that the Legislature intentionally discriminated based on race when it passed the 2023 Plan. Standing alone, this finding is sufficient for a preliminary injunction. See infra Part III. Second and in any event, based on our extensive record, we find it likely that these Section Two claims satisfy the standard announced in Callais. See infra Part IV.

Accordingly, Purcell does not bar the preliminary injunction we issue. This conclusion is consonant with other precedents. Reynolds v. Sims, 377 U.S. 533 (1964), explained that “once a State's legislative apportionment scheme has been found to be unconstitutional, it would be the unusual case in which a court would be justified in not taking appropriate action to insure that no further elections are conducted under the invalid plan.” Id. at 585. More recently, the Supreme Court has “authorized District Courts to order or to permit elections to be held pursuant to apportionment plans that do not in all respects measure up to the legal requirements, even constitutional requirements. Necessity has been the motivating factor in these situations.” Upham v. Seamon, 456 U.S. 37, 44 (1982) (per curiam). Because the only evidence on this issue establishes that it would be enormously disruptive, clearly unlawful, and not necessary for the Secretary to administer the 2026 elections with the 2023 Plan, Purcell does not bar relief. We find it difficult to imagine a case where the facts more clearly counsel this conclusion.

II. BACKGROUND

These cases allege that Alabama's electoral maps are intentionally racially discriminatory in violation of the Constitution and/or dilute the votes of Black Alabamians in violation of Section Two: Singleton and Milligan assert constitutional and statutory claims, and Caster v. Allen, No. 2:21-cv-1536-AMM, asserts statutory claims. Singleton and Milligan are before this three-judge Court, and Caster is before Judge Manasco sitting alone. All Plaintiffs challenge districts in South and Central Alabama, with a focus on Alabama's Black Belt and Gulf Coast regions.² Likewise, all Plaintiffs request an injunction (this time, another preliminary injunction) barring the Secretary from conducting congressional elections according to the map the Legislature passed after the Supreme Court's affirmance of our first preliminary injunction — the 2023 Plan (appended to this Order as Appendix C). We briefly explain the history and posture of these coordinated and related cases.

A. Two Rounds of Preliminary Injunction Proceedings

The preliminary injunction that we entered in 2022 and the Supreme Court affirmed in 2023 prohibited the use of Alabama's 2021 Plan. Milligan Doc. 107.³ 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 affirmed by the Supreme Court. Wesch v. Hunt, 785 F. Supp. 1491, 1497–1500 (S.D. Ala. 1992), 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, District 7 elected Alabama's first Black Congressman in over ninety years. District 7 remains majority-Black and in every election since 1992 has elected a Black Democrat. No other Alabama district has elected a Black candidate in approximately 150 years, until District 2 elected Shomari Figures in 2024 under the race-blind map that we imposed and the Supreme Court declined to stay.

We issued that preliminary injunction after a seven-day hearing and on an extensive record. Milligan Doc. 107 at 4.⁴ We found that the Milligan Plaintiffs were substantially likely to establish that the 2021 Plan violated Section Two, Judge Manasco found the same in Caster, and we said it was not close. Milligan Doc. 107 at 4, 195–96; Caster Doc. 101 at 5. We reserved ruling on the constitutional claims and ordered that “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, 7.

The Secretary and legislative defendants (together, “the State”), Senator Jim McClendon and Representative Chris Pringle (“the Legislators”), co-chairs of the Permanent Legislative Committee on Reapportionment (“the Committee”), appealed; the Supreme Court stayed our injunction. Allen, 599 U.S. at 16–17.

In June 2023, the Supreme Court affirmed the preliminary injunction in all respects. 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 announced in Thornburg v. Gingles, 478 U.S. 30 (1986). Allen, 599 U.S. at 23–24.

The Legislature then passed the 2023 Plan and Governor Kay Ivey signed it into law. Just like the 2021 Plan, the 2023 Plan includes only one majority-Black district: District 7. The district with the next highest Black share of the voting-age population (“BVAP”) in the 2023 Plan is District 2, with a BVAP of 39.9%. All Plaintiffs sought another injunction. Singleton Doc. 147; Milligan Docs. 200, 265; Caster Doc. 179.

The State then 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 add an opportunity district. Aug. 14, 2023 Tr. at 159–64. Alabama argued that our order about “the appropriate remedy for the ․ likely violation that we found” did not have any relevance to the 2023 Plan. Id. at 75. It 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 made this case unique: we were not (and still are not) aware of any other case in which a legislature — faced with a federal court order declaring that its electoral plan unlawfully dilutes minority votes and requiring a remedy that provides an additional opportunity district — responded with a plan that the state conceded did not provide that district.

We issued a second preliminary injunction in August 2023 because we found that the 2023 Plan did not remedy the likely Section Two violation that we found and the Supreme Court affirmed, and alternatively because we found the Milligan Plaintiffs were substantially likely to establish anew that the 2023 Plan violates Section Two. See generally Milligan Doc. 272. Judge Manasco again ordered the same relief in Caster. Caster Doc. 223 at 4. We again ordered that the appropriate remedy was 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 reserved ruling on the constitutional claims, invoking constitutional avoidance. Id. at 8.

We directed a Special Master to prepare and propose remedial maps. Milligan Doc. 273 at 6. The Secretary (but not the Legislators ⁵ ) again appealed and sought a stay. Milligan Docs. 274–76, 281. We denied a stay and the Secretary sought a stay from the Supreme Court, which denied the request with no noted dissents. See Allen v. Milligan, 144 S. Ct. 476 (2023) (mem.).⁶

The Special Master prepared and recommended three plans. See 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 the Secretary to administer the 2024 elections using the Special Master's “Remedial Plan 3,” which is 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 one majority-Black district (District 7, where 51.9% of the voting-age population is Black) and one Black-opportunity district (District 2, where 48.7% of the voting-age population is Black). Id. at 41.

The Special Master Plan was drawn race-blind. The Special Master explained that the Court's cartographer, Mr. David Ely, did not consider race:

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 (emphasis added).

The Secretary conducted Alabama's 2024 elections using the Special Master Plan.

B. Trial and Final Judgment

The Plaintiffs then requested a judgment that the 2023 Plan violates federal law and a permanent injunction barring the Secretary from using that Plan and requiring him to use a court-ordered plan for the rest of this census cycle. Singleton Doc. 229 at 46; Milligan Doc. 329 ¶ 206; Caster Doc. 271 at 43; Milligan Doc. 485 at 425–27, ¶¶ 1150–52. And the Milligan Plaintiffs requested under Section 3(c) of the Voting Rights Act that the Court bail Alabama back into federal preclearance for congressional redistricting. Milligan Docs. 485 at 436, ¶ 1173; 329 at 77.

In February 2025, we held a coordinated trial of all three cases that consumed eleven trial days. We heard live testimony from twenty-three witnesses (including thirteen experts); received reports and rebuttal reports from every expert; received testimony by designation from twenty-eight additional witnesses; 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. Under Federal Rule of Civil Procedure 65(a)(2), we continued to have the benefit of evidence adduced in the first two preliminary injunction proceedings.

Based on findings and conclusions that we explained at length in a permanent injunction in May 2025, we saw the same clear result on the Section Two claims that we saw in 2022 and 2023. See Milligan Doc. 490 at 323–436. We concluded that the Plaintiffs established each part of the then-controlling legal standard from Gingles, 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 nearly invariably vote as a bloc to defeat Black-preferred candidates; finally (4) under the totality of the circumstances in Alabama today, Black voters have less opportunity than other Alabamians to elect candidates of their choice. We concluded that the 2023 Plan unlawfully dilutes Black votes. Id. at 436.

Numerosity was undisputed, extensive evidence established reasonable compactness, and there was no serious dispute that intensely racially polarized voting carried extreme consequences: Black candidates have won zero statewide elections in Alabama since 1994 (when one Black person was elected after a previous appointment to office), and only three Black candidates have ever been elected to statewide office since Reconstruction. Black candidates have enjoyed near-zero success in legislative elections outside of opportunity districts: thirty-two of the thirty-three Black members of the 140-person Legislature represent majority-Black districts. We found substantial evidence that Black Alabamians have less opportunity than other Alabamians to elect representatives of their choice.

We reiterated our understanding that “[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. 490 at 12 (quoting Democratic Exec. Comm. of Fla. v. Lee, 915 F.3d 1312, 1315 (11th Cir. 2019)).

Additionally, we ruled that the 2023 Plan intentionally discriminated against Black voters based on race. Milligan Doc. 459 at 494–542. We conducted a searching review of extensive undisputed evidence from legislators and the Legislature and could not understand the 2023 Plan as anything other than an intentional effort to dilute votes based on race. Id. at 496–542. We concluded that if this record did not rebut the strong presumption of legislative good faith, we doubted the presumption is ever rebuttable.⁷

Before remedial proceedings, the Legislators conceded that subject to appellate rights, legislative leadership would “voluntarily forego any rights that [it] may have to attempt to draw an additional congressional district map,” so long as the Special Master Plan remained in place for this census cycle. See Milligan Doc. 493 at 2–3. All Plaintiffs agreed that an injunction barring the Secretary from administering elections under the 2023 Plan and ordering him to administer them with the Special Master Plan was “a full remedy to the Section 2 violation,” Milligan Doc. 497 at 2, but the Milligan parties argued it could not resolve the request for bail-in.

Before our remedial hearing, the Supreme Court re-listed Callais and indicated that it would consider whether a legislature's intentional creation of a majority-minority district to satisfy Section Two violates the Fourteenth Amendment. See Louisiana v. Callais, 45 S. Ct. 2608 (2025) (mem.).

After a remedial hearing, we issued a final judgment in August 2025 that ordered the Secretary (consistent with the parties' agreement) to administer elections under the Special Master Plan for the rest of this census cycle, denied bail-in relief, and maintained enforcement jurisdiction. Milligan Doc. 509. The Secretary appealed a third time. Milligan Doc. 511; Singleton Doc. 339. He filed jurisdictional statements and petitioned the Supreme Court for a writ of certiorari before judgment in Caster, the Plaintiffs sought a summary affirmance, and the Supreme Court conferenced the appeals in November 2025.

C. Callais

Following the 2020 census, Louisiana enacted a congressional map that was successfully challenged under Section Two by a group of Black voters. See generally Robinson v. Ardoin, 605 F. Supp. 3d 759 (M.D. La. 2022), vacated and remanded, 86 F.4th 574 (5th Cir. 2023). In connection with remedial proceedings, the Louisiana Legislature drew a district that “stretche[d] some 250 miles from Shreveport in the northwest corner of the state to Baton Rouge in southeast Louisiana, slicing through metropolitan areas to scoop up pockets of predominantly Black populations from Shreveport, Alexandria, Lafayette, and Baton Rouge.” Callais v. Landry, 732 F. Supp. 3d 574, 588 (W.D. La. 2024). Legislators believed they were “ordered to draw a new [B]lack district” but found it “difficult” because Louisiana does not “have concentrated populations of” Black people. Id. at 587–88. So Louisiana enacted the new map after “primarily consider[ing] race” as the “driving force.” Id. at 588. The Callais plaintiffs then sued, arguing that Louisiana impermissibly used race to draw the remedial district, and a three-judge district court agreed. See generally id. at 606–13.

On April 29, 2026, the Supreme Court ruled in Callais and adjusted the controlling legal standard for Section Two claims. Callais clarified that Section Two “imposes liability only when the circumstances give rise to a strong inference that intentional discrimination occurred.” 146 S. Ct. at 1156. Callais explained that “[o]nly when understood this way does § 2 of the Voting Rights Act properly fit within Congress's Fifteenth Amendment Enforcement Power.” Id.

Callais did not overrule Gingles. Id. at 1157. It instead “updated” each part of the Gingles framework to align with the proper construction of Section Two. For the first part of Gingles (about numerosity and compactness), plaintiffs now must produce an illustrative map with an additional remedial district without using race “as a districting criterion” and while “meet[ing] all the State's legitimate districting objectives, including traditional districting criteria and the State's specified political goals.” Id. at 1159. So “[i]f the State's aims in drawing a map include a target partisan distribution of voters, a specific margin of victory for certain incumbents, or any other goal not prohibited by the Constitution, the plaintiffs' illustrative maps must achieve these goals just as well.” Id. This “help[s] to ‘disentangle race' from politics and other constitutionally permissible considerations.” Id. (quoting Alexander v. S.C. State Conf. of the NAACP, 602 U.S. 1, 6 (2024)).

To satisfy the second and third parts of Gingles (about racially polarized voting), and to serve the broader aim of disentangling race from politics, plaintiffs now must provide an analysis that “show[s] that voters engage in racial bloc voting that cannot be explained by partisan affiliation.” Id. It is no longer enough that a Section Two plaintiff identifies a statistically significant pattern of voters supporting candidates along racial lines. Evidence of racially polarized voting must “control[ ] for party affiliation” to demonstrate that race is an independent driver of voter activity. Id. That showing is best made with evidence of intra-party polarized voting — that is, evidence that “black and white voters ha[ve] dramatically different voting patterns within” a party such that “the minority plaintiffs have ‘less opportunity’ than their majority counterparts because of race, not just because of partisan affiliation.” Id. (quoting 52 U.S.C.A. § 10301(b)).

Finally, Callais held that in evaluating all the circumstances, “[d]iscrimination that occurred some time ago, as well as present-day disparities that are characterized as the ongoing effects of societal discrimination, are entitled to much less weight” than “current data and current political conditions that shed light on current intentional discrimination.” Id. at 1160 (internal quotation marks omitted). In addition, any totality of the circumstances analysis must account for partisan explanations. To the extent “a State defends a districting scheme on the ground that it was drawn for partisan purposes,” a successful Section Two claim must “rul[e] out the competing explanation that political considerations dominated the legislature's redistricting efforts.” Id. at 1156–57 (quoting Alexander, 602 U.S. at 9).

D. Alabama's 2026 Elections

When Callais was re-argued, the Secretary did not seek a stay (from us or the Supreme Court) of our permanent injunction. So when Callais issued, the Secretary had already commenced the 2026 elections according to the Special Master Plan: candidates qualified in January 2026, were certified in March 2026, and absentee balloting had begun. See Ala. Code § 17-13-5(a), (b); 52 U.S.C. § 20302(a)(8). Campaigns were nearly over, with in-person voting only twenty days ahead.

Callais unleashed a flurry of proceedings. On the litigation front, the Secretary immediately moved to expedite his appeals and later sought a stay from us and the Supreme Court. Milligan Doc. 520. On the legislative front, Governor Ivey called a Special Session to begin on May 4, 2026. See Milligan Doc. 518-1. Her Proclamation allowed the Legislature to “consider legislation to provide for a special primary election for ․ members of the United States House ․ in districts whose boundary lines are altered by a court issuing a judgment, vacating an injunction, or otherwise ordering or permitting an alteration in the boundaries of such districts.” Id. at 1.

Representative Pringle introduced House Bill 1 during the 2026 Special Session to provide for special primaries. It became law on May 8, 2026. It provides that (1) “the state shall hold a new special primary election” if “a federal court, by issuing a judgment or by vacating an injunction, permits the reinstatement of the last legislatively enacted Congressional districts,” and (2) any certified results from the May 19, 2026 congressional primary would become “void for purposes of determining the party nominee once a new special primary election is required.” Ala. Code § 17-13-3.1(b), (d). It does not require the use of a particular districting plan.

On May 11, 2026, the Supreme Court summarily vacated our permanent injunction and remanded these cases. All Plaintiffs sought a temporary restraining order, and Governor Ivey called special primary elections to occur on August 11, 2026, for congressional Districts 1, 2, 6, and 7. Milligan Doc. 529-1. She required candidates to qualify between May 20 and May 22, 2026, and parties to certify candidates by May 26, 2026. Id. at 1. The special primaries affect forty of Alabama's sixty-seven counties.⁸

We set expeditious deadlines for preliminary injunction motions and directed the Secretary to file an affidavit advising us in detail of all practical considerations relevant to the administration of the 2026 congressional elections. Milligan Doc. 528 at 5. These cases are now before us on fully briefed preliminary injunction motions. Milligan Docs. 531, 534–35; Singleton Docs. 357, 360–61; Caster Docs. 437, 440–41. We convened a hearing on May 22, 2026.⁹

We do not have before us a request to enjoin the state law that allows the August special primaries. We consider only challenges to the districting plan. Following vacatur of our permanent injunction, these cases are now before us as though it never issued. Upon vacatur, although voters remained districted under the Special Master Plan and the Secretary was using it to administer the May 19 primary election, the 2023 Plan became the legal status quo ante. Hewitt v. United States, 606 U.S. 419, 431 (2025).

We measure the motions for a preliminary injunction against Callais and other binding precedent, based on the extensive evidence from trial and, under Rule 65(a)(2), the evidence from preliminary injunction proceedings that was not abandoned at trial.¹⁰

We fully revisit the merits of each claim, about which the summary vacatur expressed no view and “merely requires further consideration in light of [the] new Supreme Court decision.” United States v. M.C.C. of Fla., Inc., 967 F.2d 1559, 1561–62 (11th Cir. 1992).

At the outset, we make the same credibility findings that we made in our permanent injunction. See Milligan Doc. 490 at 324–33, 370, 380–92. Part of the applicable law has changed, but the evidence has not, so we continue to believe and disbelieve witness testimony as we did before. Additionally, we adopt our previous recitation of the evidence before us, which remains unchanged after Callais. See Milligan Doc. 490 at 24–103, 117–323, 466–494.

III. PRELIMINARY INJUNCTION MERITS – CONSTITUTIONAL VIOLATION (MILLIGAN)

The Milligan Plaintiffs' constitutional claim is that in 2023, the Legislature knew from court orders that a map without an additional Black-opportunity district would unlawfully dilute Black Alabamians' opportunity to participate in the political process, intentionally enacted just such a map, and purposefully tried to prevent a remedy.

More particularly, the claim is that the Legislature knew that to dilute the Black vote, it could crack the Black population in South Alabama by submerging much of the Black Belt in one majority-White district while simultaneously submerging Black voters in Mobile in a separate majority-White district (that keeps the Gulf Coast whole). In addition, the Legislature knew from our preliminary injunction and the Supreme Court's affirmance that this was unlawful vote dilution, and did it anyway. Accordingly, these Plaintiffs allege 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. 445 at 9.

Our earlier constitutional analysis and conclusion, on the same record that is before us now, are undisturbed by Callais. See Milligan Doc 490 at 466–542. The Supreme Court had no occasion to consider in Callais the standard for intentional discrimination under the Fourteenth Amendment, nor did it opine on how a state's purported goal in drawing electoral maps may be unconstitutional.

To obtain a preliminary injunction, the Plaintiffs must establish that they are likely to succeed on this claim. Benisek v. Lamone, 585 U.S. 155, 158 (2018) (per curiam). A plaintiff alleging intentional vote dilution (distinct from racial gerrymandering) “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, 602 U.S. at 38–39 (quoting Miller v. Johnson, 515 U.S. 900, 911 (1995)). The plaintiff “must show that the State's districting plan ‘has the purpose and effect' of diluting the minority vote.” Id. at 39 (quoting Shaw v. Reno, 509 U.S. 630, 649 (1993)).

To assess a legislature's motives, a court must consider all available evidence of intent. Hunt v. Cromartie, 526 U.S. 541, 546 (1999) (citing, inter alia, Vill. of Arlington Heights v. Metro. Hous. Dev. Corp., 429 U.S. 252, 266 (1977)). “Discriminatory purpose” means that a 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.” Pers. Adm'r of Mass. v. Feeney, 442 U.S. 256, 279 (1979). Showing “discriminatory intent ․ ‘does not require a plaintiff to prove that the challenged action rested solely on racially discriminatory purpose.’ ” Allen, 599 U.S. at 37 (quoting Arlington Heights, 429 U.S. at 265).

The Supreme Court has directed us to consider six factors: (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; (5) whether the “disparate impact” was “the natural and foreseeable consequence of the practices and policies” of the State; and (6) “contemporary statements by members of the decisionmaking body.” Arlington Heights, 429 U.S. at 267–68; accord, e.g., Abbott v. Perez, 585 U.S. 579, 609–10 (2018); Brnovich v. Democratic Nat'l Comm., 594 U.S. 647, 687–88 (2021).

If the Milligan Plaintiffs establish a prima facie case of intentional discrimination, the State must “dispel the inference” with “evidence in the record.” Castaneda v. Partida, 430 U.S. 482, 497–98, 500 (1977).

We presume, as we must, that the Legislature acted in good faith. This command “reflects [our] due respect for the judgment of state legislators”; shows an appropriate hesitancy to “hurl ․ accusations” of “ ‘offensive and demeaning’ conduct” at a legislature; and 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; then quoting Cooper v. Harris, 581 U.S. 285, 335 (2017) (Alito, J., concurring in judgment in part and dissenting in part)).

We draw every inference we can in the Legislature's favor, make no effort to read anyone's mind, and accuse no legislator of racism.¹¹ We review what the Legislature did, what it said about why, and testimony and statements by key legislators.¹² We find as a fact the following:

Arlington Heights 1: Historical Background of the 2023 Plan. Alabama's history of discriminating against Black Alabamians is well-documented. To give the Legislature every benefit of the doubt, we do not rely on that history. We focus on the immediate history of the 2023 Plan: the 2023 Special Session.

Arlington Heights 2: Sequence of Events Leading to the 2023 Plan. We begin with the Legislators' real-time understanding of what federal law required of the 2023 Plan. We credit their testimony (and that of Mr. Randy Hinaman, the Legislature's longtime cartographer) that they understood that, as set forth in our affirmed order, Section Two required an additional opportunity district. See Milligan Docs. 459-13 at 6, 14 (Livingston), 459-20 at 5 (Pringle); 459-7 at 5 (Hinaman).

Undisputed evidence shows that nevertheless, events unfolded this way:

• Legislators instructed Mr. Hinaman from the outset to keep the Gulf Coast Counties (Mobile and Baldwin) whole and together. See Milligan Doc. 459-7 at 20. As all parties agree, keeping the Gulf Coast whole and together made it mathematically impossible to create a second majority-Black district. Tr. at 298–99 (Dr. Duchin); 2648 (counsel for the State).13 It was yet unknown whether it was possible to draw a second opportunity district while keeping the Gulf Coast whole and together.

• Mr. Hinaman performed his work completely unaware of the 2023 legislative findings: he never saw them, was told nothing about them, and was in the dark about the “non-negotiable” rules and novel definitions they set out. See Milligan Doc. 459-7 at 23–24.

• 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).

• When Representative Pringle introduced his proposed plan (“the Community of Interest Plan,” with a BVAP of 42.4% in District 2), he argued that it preserved the cores of existing districts and supplied a performance analysis to show that the Black-preferred candidate would have won two of four modeled races. See Milligan Docs. 459-20 at 25; 436 at 19. Mr. Hinaman understood that both Representative Pringle and Senator Livingston preferred that plan and would sponsor it. See Milligan Doc. 459-7 at 8. But the Legislature ultimately turned away from that plan.

• Senator Livingston testified that the change of mind was based on “additional information” Committee members 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, notably, did not know what it was, where it had come from, or who received it. Id.

• Ultimately, the Senate passed a plan called the Opportunity Plan, and the 2023 Plan was a slight variant of the Opportunity Plan. Both these plans were variants of plans Senator Livingston proposed.14 In the Opportunity Plan, the BVAP in District 2 was 38.31%. Milligan Doc. 436 ¶¶ 128–29.

• After the Opportunity Plan passed in the Senate, Representative Pringle refused to pass it in the House or even attach his name to it. Representative Pringle testified that he believed the Opportunity Plan might not or did not satisfy the Voting Rights Act, and the Community of Interest Plan was more likely lawful. See Milligan Doc. 459-20 at 26.

• Senator Livingston testified that he relied on talking points prepared by the Solicitor General. See Milligan Doc. 459-13 at 26. Those talking points extolled how Senator Livingston's plans treated three communities of interest (the Black Belt, the Gulf Coast, and the Wiregrass), not partisanship or providing an opportunity district. Milligan Doc. 385-1 at 26.

• Senator Livingston's plans, the Opportunity Plan, and the 2023 Plan homed in on Alabama's most well-known and nationally prominent Black community: the City of Selma, in Dallas County. 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 left Black Alabamians in District 2 in the Opportunity Plan utterly unable to elect a representative of their choice.

• The Legislature had in hand a performance analysis from its own expert, Dr. Trey Hood, that told them that moving Dallas County out of District 2 destroyed any chance it might have of performing as an opportunity district. See Milligan Doc. 459-13 at 23 (Senator Livingston, acknowledging that, according to Dr. Hood's performance analysis, the Black-preferred candidate would lose all seven modeled races by seven points in the 2023 Plan's District 2); Milligan Doc. 459-7 at 13 (Mr. Hinaman, discussing that Dr. Hood's analysis showed Dallas County affecting the performance of District 2); see also Caster Doc. 352-2 at 14; App. E at 3; Tr. at 298–99, 314 (explaining that the 2023 Plan placed nine of the eighteen Black Belt counties in majority-White districts).

• On the morning the 2023 Plan was enacted, the Legislators saw for the first time the eight pages of embedded legislative findings. These novel findings came a week after the Committee had readopted its 2021 guidelines, were materially different from those guidelines, and were not requested by the Committee chairs (who did not know why they were placed in the bill, had never seen 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); see also infra Part III pp. 35–38 (discussing substance of 2023 legislative findings).

Standing alone, this evidence forecloses a finding that the Legislature accidentally stumbled into a second round of discriminatory vote dilution. It establishes that in addition to purposefully refusing to remedy the vote dilution we and the Supreme Court found, the Legislature touted the 2023 Plan with a claim that it 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 Black Alabamians had zero opportunity to elect a candidate of their choice, and (2) the key mechanism for ensuring that result was placing Selma in District 7. It also establishes that the 2023 legislative findings were (at a minimum) not the ordinary result of the Legislature's usual process.

Arlington Heights 3 & 4: Departures from the Norm and Legislative History. At the outset, we observe that a legislature's calculated, purposeful refusal to provide the remedy that a court order requires, after the Supreme Court affirms the order, is definitively not the norm. Occasionally, legislatures have been unable or unwilling to redistrict after courts enjoin electoral plans. See, e.g., Vera v. Bush, 933 F. Supp. 1341, 1346 (S.D. Tex. 1996); Reynolds, 377 U.S. at 586. Sometimes legislatures redistrict, but courts determine that the new map is not a remedy. See, e.g., North Carolina v. Covington, 585 U.S. 969, 970–71 (2018) (per curiam). But we are aware of no other case, let alone a norm, that involves a legislature's willingness to act coupled with its admitted and strategic refusal to provide the required remedy.

Separately, the 2023 legislative findings reflect severe departures from the Alabama Legislature's norm. That the Legislature produced and issued those findings is itself a full departure: there is no evidence the Legislature ever has made such findings in previous districting legislation. Further, the substance of the 2023 legislative findings is replete with sharp departures from the Committee's longstanding districting guidelines, which the Committee re-adopted only a week before the Legislature enacted the findings:

• First, the findings describe some traditional districting principles as “non-negotiable,” which the guidelines did not. Compare App. B at 2, with App. A. For example, the guidelines did not cap the number of acceptable county splits, nor define that term, but the legislative findings did. As another example, the guidelines describe the “polic[y]” that “[c]ontests between incumbents will be avoided whenever possible” as one among several “redistricting policies” that “shall be observed to the extent they do not violate” the Constitution, but the legislative findings state simply that a districting plan “shall not pair incumbent[s].” Compare App. A. at 2 with App. B at 2.

• Second, although the findings provide that the districting plan must comply with the Voting Rights Act, the Legislature did not designate the nondilution of minority voting strength as “non-negotiable.” Compare App. A at 1, with App. B at 2–3. Indeed, the 2023 guidelines (like previous iterations), 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. This departure from the norm came even though the only reason the 2023 Special Session occurred is because we enjoined the 2021 Plan on the ground that it likely diluted minority votes.

• Third, the findings name and define specific communities of interest; the guidelines did not. Compare App. B at 3–7, with App. A at 2–3.

• Fourth, 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 at issue in this pending litigation. See App. B at 3–7.

• Fifth, the findings revise the definition of “community of interest” from the guidelines: the findings strip 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 was a razor-sharp departure from the norm during pending litigation involving extensive testimony about a minority's shared experience of race discrimination.

• Sixth, the findings exalt and extol one community of interest, describing the Gulf Coast for pages, while describing the longstanding community of interest in the Black Belt in a couple of short paragraphs. Compare App. B at 4–7, with id. at 3–4. We are aware of no enacted redistricting document in Alabama that ever has done this. In addition, while the Legislature claimed it was “non-negotiable” that Mobile and Baldwin Counties be in the same congressional district, the Legislature repeatedly split those Counties in the State Board of Education districting plans, and did so as recently as 2020. Tr. at 325–27, 348, 2024–30, 2090–97.

• Seventh, 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. At the same time, the findings eliminated from the definition of the Black Belt the express recognition that it “has a substantial Black population because of the many enslaved people brought there to work in the antebellum period.” Milligan Doc. 53 ¶ 60. This departure was pointed because the Legislators had previously stipulated that definition in these cases.

The sixth and seventh departures are especially pointed because they came during this litigation, where one of the claims is that the majority-White community of interest in the Gulf Coast is being used to entrench discrimination against Black voters in the majority-Black community of interest in the Black Belt.

• Eighth, 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.

The 2023 Special Session provided a unique opportunity for the Legislature to develop real-time evidence during the litigation of these cases. The evidence they developed in the findings, together with their drumbeat reminders in court that it is not possible to draw an additional majority-Black district without violating their novel and exacting instructions, makes clear to us that they meant both (1) not to provide a remedial district and (2) to prevent a court-ordered remedial district, even though a remedial district was required by the Supreme Court's affirmance of our order.

Arlington Heights 5: Foreseeability of Disparate Impact. Undisputed evidence establishes that in 2023, the Legislature should have foreseen, and in fact actually foresaw, the disparate impact of the 2023 Plan on Black Alabamians, and enacted it anyway:

• First, our preliminary injunction, as affirmed by the Supreme Court, was the only reason the Legislature was in session for redistricting, and it was entirely foreseeable from the injunction alone that a refusal to provide an additional opportunity district would perpetuate the likely vote dilution we found. See Milligan Docs. 166, 168, 174, 186.

• Second, undisputed evidence establishes that before the Legislature passed the 2023 Plan, legislators considered a performance analysis supplied by the State's expert that modeled seven elections and predicted Black-preferred candidates would lose every election in District 2 in that Plan. Milligan Doc. 436 at 21.

• Third, Representative Pringle testified that he refused to put his name on the 2023 Plan based on his personal understanding that it might not or would not comply with the Voting Rights Act. Milligan Doc. 459-20 at 26.

• 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. See Milligan Doc. 459-7 at 13–14. According to Mr. Hinaman, the Black-preferred candidate lost every election Dr. Hood modeled in that District, if Dallas County was not in the District. See id. This made it foreseeable that the 2023 Plan, which left Dallas County out of District 2, would afford Black-preferred candidates zero chance of winning there.

Contemporaneous Statements About Race. Finally, contemporaneous public statements by legislators shed light on how they considered race. As an initial matter, the contemporaneous-statement evidence furnishes no basis for a finding that legislators were focused on remedying vote dilution. The enormous record before us contains no contemporaneous-statement evidence that the Legislature originally passed the 2023 Plan to remedy vote dilution or to secure a partisan advantage.

Additionally, other contemporaneous-statement evidence indicates that in real time, key legislators were focused on race. During the 2023 Special Session, after one of Senator Livingston's consultants texted him to ask outright, “Would 41.6[%] BVAP work?”, Senator Livingston supported a plan with a substantially lower BVAP in District 2. As the State's own performance analysis advised legislators, that plan did not elect a Black-preferred candidate in any of the seven modeled elections. Milligan Doc. 459-3 at 26; see Milligan Doc. 459-13 at 23 (conceding that in the State's performance analysis of the 2023 Plan, the Black-preferred candidate would lose all seven modeled races by approximately seven points in District 2).

Similarly, Representative Pringle testified that his limited knowledge of the reconciliation process between the Alabama House and Senate to reach the 2023 Plan is that the process was focused on the BVAP of District 2: the 2023 Plan “split the difference” on that between the plans the House and Senate had passed. Milligan Doc. 459-20.

Together with the evidence about the State's performance analysis, the contemporaneous-statement evidence suggests that the Legislature was focused on race — more particularly, keeping the BVAP in District 2 low enough to foreclose the election of a Black-preferred candidate — when it passed the 2023 Plan.¹⁵

Considering all the evidence Arlington Heights instructs us to consider, we find that the Legislature intended to discriminate against Black voters based on race when it passed the 2023 Plan. This evidence is the Legislature's own, see The Robert Edwards, 19 U.S. 187, 190 (1821), is undisputed, and does not rest on uncharitable inferences, see Alexander, 602 U.S. at 9. No Arlington Heights factor favors the State or cuts against our finding, even in part.

In the simplest terms, the sequence and substance of extraordinary legislative events, against the backdrop of the Legislature's knowledge, compels us to conclude that the Legislature doubled down on racially discriminatory vote dilution after we and the Supreme Court found that it was racially discriminatory vote dilution.

That same evidence tells us that when the Legislature doubled down, it was not content simply to refuse to remedy the vote dilution that we and the Supreme Court found. The Legislature took the additional step — highly unusual, strategically calculated, and traveling far from legislative norms — of trying to prevent a court-drawn remedy. And the same evidence leaves us no room to conclude that when the Legislature did all this, it had party politics in mind. The only available intent evidence tells us that considerations of race were the key reason.

We thus turn to the State's attempt to dispel the inference of intentional discrimination. The record includes extensive argument against the inference, see Milligan Doc. 534 at 52–75, but we see no dispelling evidence. We reject the State's charge that there is no direct evidence of discriminatory intent, id. at 53–55, because all the Arlington Heights evidence comes from the Legislature or a legislator, and it is undisputed. We discuss in turn each of the State's other arguments:

The State's arguments about the Gulf Coast. The State argues that “treating the Legislature's emphasis on the Gulf Coast” as evidence of discriminatory intent is inconsistent with Callais and Alexander, id. at 55–60, but this grossly misstates both the law and our finding. Callais and Alexander are expressly limited to “the State's legitimate districting objectives,” Callais, 146 S. Ct. at 1159, Alexander, 602 U.S. at 34, and do not foreclose a court from finding that an objective is illegitimate because it was used for an unconstitutional purpose. And our finding that the Legislature (among other things) exalted the Gulf Coast in the 2023 legislative findings for an unconstitutional purpose is based on all the steps the Legislature took to dilute minority votes on purpose, not solely the bare fact of the Legislature's overdrawn exaltation of the Gulf Coast. See supra Part III pp. 35–38. We previously found that the Gulf Coast is a community of interest, Milligan Doc. 490 at 353, and our finding that the Legislature misused it (among other things) for an unconstitutional, and therefore illegitimate, purpose does not disturb that finding.

The State's arguments about the Black Belt. The State defends the short-shrift discussion of the Black Belt in the 2023 legislative findings on the ground that “there was little need for the Legislature to elaborate on it.” Milligan Doc. 534 at 58. This too ignores all our other findings. See supra Part III pp. 35–38.

The State's arguments about Callais. The State argues that Callais vindicates its concerns about racial gerrymandering, that had Alabama “adopted one of the Plaintiffs' maps in 2023, it likely would have adopted an unconstitutional racial gerrymander,” and that we erred by finding “racial animus behind the State's effort to avoid the noxious practice of racial sorting.” Milligan Doc. 534 at 60–61 (citation modified). The State asserts that Callais shows that Louisiana should have done just what Alabama did, and “doubled down on its traditional districting criteria and constitutional defenses.” Id. at 62, 71.

These arguments ignore the Supreme Court's affirmance of our first preliminary injunction. That affirmance would not have occurred if we or it demanded something unconstitutional; there is no world in which we or the Supreme Court demanded the State do anything noxious. Likewise, the arguments ignore that Callais expressly did not overrule that affirmance, 146 S. Ct. at 1160–61. Callais thus tells us that remediating discriminatory vote dilution in Alabama did not demand anything unconstitutional in 2023 and still does not demand anything unconstitutional now. Finally, the arguments conveniently ignore that the Special Master easily drew a race-blind remedial map in these cases. That reality disproves that anything unconstitutional was required then or is required now.

These realities follow naturally from a critical factual difference between these cases and Callais: although we have consistently held that as a group, Black Alabamians are sufficiently geographically compact to constitute a voting-age majority in an additional, reasonably configured district, Milligan Doc. 490 at 324–69, the “unusual shape” of the remedial district in Callais reflected an effort to gather up a dispersed Black population from the four corners of Louisiana to create an additional district, see Callais v. Landry, 732 F. Supp. 3d 574, 604 (W.D. La. 2024), aff'd sub nom. Callais, 146 S. Ct. 1131.

Alabama cannot use Callais to legitimize its pre-Callais decision to double down on the discriminatory vote dilution that we and the Supreme Court found. And it cannot use Callais to legitimize the series of specific and unusual decisions it made to entrench that dilution. If such retroactive validation strategies were available, States would be encouraged to govern themselves according to what they think federal law ought to be, not what it is.

Separately and in any event, there is zero evidence that racial gerrymandering fears drove the 2023 Special Session. When the Legislators testified about why the Legislature moved away from the Community of Interest Plan, they did not so much as mention a gerrymandering concern. See Milligan Doc. 459-13 at 16–17, 21–23 (Livingston); Milligan Doc. 459-20 at 25–27 (Pringle). And we have scoured the record and see no evidence that any legislator considered a map that provided 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. This deficit of proof makes sense: the Legislature knew that the Supreme Court had affirmed our ruling that there were lawful ways to remedy the vote dilution that we found.

The State's arguments about partisanship. The State now argues that its legislatively enacted plan is “better explained by partisan motives than by racial animus,” and that Plaintiffs “fail to disentangle race from politics.” Milligan Doc. 534 at 66–67, 73. As the Supreme Court observed in Callais, Alabama “did not cite partisan goals in defending” the 2021 Plan. 146 S. Ct. at 1161. The record now includes extensive lawyer argument about purportedly partisan motives for the 2023 Plan, see Milligan Doc. 534 at 52–75, but there is zero evidence the Legislature enacted the 2023 Plan for partisan purposes. The evidence is as follows:

• The 2023 legislative findings state with great particularity many goals and omit entirely party politics. See App. B.

• The State cites two statements by legislators about having “seven Republican congressmen” and “a Republican opportunity plan,” see id. at 67 (citing Milligan Doc. 481 at ¶ 787), but those were not statements about the 2023 Plan.

• The Legislators testified about (1) calls they received from the then-Speaker of the U.S. House about Republicans' slim congressional majority, and (2) conversations they had with congressional and 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: “[The Speaker's phone call] really didn't play into our efforts.”); Milligan Doc. 459-20 at 6 (Pringle).

• When the Legislators were asked in depositions about why the Legislature 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. See Milligan Doc. 459-13 at 15–17; Milligan Doc. 459-20 at 25–26.

• The evidence that the Legislature acted to protect incumbents is not evidence that it acted for Republican partisan gain; the State has urged us that by “incumbents,” it means all incumbents, including Congresswoman Sewell, who is a Democrat. Milligan Doc. 459-13 at 17; Milligan Doc. 481 ¶¶ 583, 847.

This evidence supplies no basis for us to accept the State's retroactive assertions about the Legislature's intent. We must determine whether the Legislature violated the Constitution by assessing the evidence of its own contemporaneous statements and reasons. The State cannot avoid liability by way of revisionist history.

The State's arguments about Section Two as a predicate. The State argues that the Milligan Plaintiffs must establish their Section Two claim as a predicate for their claim under the Fourteenth Amendment, because they must show that vote dilution occurred to establish that it was intentional. Milligan Doc. 534 at 47–51. But because of the posture of these cases, that water was over the dam before the 2023 Special Session. By then we had already held (and the Supreme Court had affirmed) that a districting plan without an additional opportunity district would likely dilute Black votes in violation of Section Two, and the litigation had involved extensive evidence about the specific dilutive mechanisms at issue. And the State has conceded that the 2023 Plan purposefully did not provide an additional opportunity district. Simply put, we know that the Legislature knew what it was doing because we and the Supreme Court had already seen it and called it out.

The State also suggests that the Legislature cannot be faulted for the vote dilution we and the Supreme Court found because those findings were about the “since-repealed” 2021 Plan. Milligan Doc. 534 at 71. But this is window dressing. The Legislature's decision not to include an additional opportunity district in the 2023 Plan made that plan, for vote-dilution purposes, materially identical to the 2021 Plan. And even before the passage of the Voting Rights Act, the Supreme Court rejected attempts to employ a new vote dilution tactic after a prior one had been held unlawful. See, e.g., Lane v. Wilson, 307 U.S. 268 (1939); Guinn v. United States, 238 U.S. 347 (1915). As the Supreme Court more recently said in a discussion of how a litigant repackaged a forbidden argument for strategic gain in a congressional redistricting case — judicial “decisions cannot be evaded with such ease.” Alexander, 602 U.S. at 21.

The State's arguments about the cleansing effect of the 2026 Special Session. The State urges us to consider evidence from the 2026 Special Session about party politics and the “reenactment of the 2023 Plan by a new Legislature under new circumstances after a new deliberative process that clearly involved partisan and political motives.” Milligan Doc. 534 at 66–71. But we do not see anything from 2026 that can cure the original discriminatory intent of the 2023 Plan. The State supplies no basis in federal law for a finding that after a Legislature passes legislation for an unlawful discriminatory purpose, it can later absolve itself of that intent or otherwise cure it by making post-hoc statements that cast the history of the legislation, or the legislation itself, in a better light.

And in any event, the Legislature did not enact a districting plan during the 2026 Special Session: it simply provided a mechanism for special primary elections to occur if an opportunity to use the 2023 Plan arose. See Ala. Code § 17-13-3.1. By themselves then, statements during the 2026 Special Session do not tell us anything about what the Legislature meant or did not mean when it enacted the 2023 Plan.

***

In our view, the unique reality of this evidentiary record overwhelms the strong presumption of legislative good faith. If this record does not rebut the presumption, we seriously doubt that it is rebuttable absent a clear and direct expression of invidious discrimination in the text of a bill or official arguments in support of its passage.

We find that the Milligan Plaintiffs likely will prevail on the merits of their claim under the Fourteenth Amendment. Specifically, we find that those Plaintiffs likely will establish that (1) 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 was discriminatory vote dilution, and (2) that the Legislature did this at least in part because of race, and not party politics.

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 in this unusual posture and on this extensive record, we do not find the issue particularly complex or close.

IV. PRELIMINARY INJUNCTION MERITS – SECTION TWO

Although our finding of a likely constitutional violation is sufficient for us to find that one set of Plaintiffs will prevail on the merits, we also discuss all Plaintiffs' request for a preliminary injunction under Section Two. All Plaintiffs argue that although Callais made changes to the applicable legal standard, it did not change the outcome of the analysis in these cases. Milligan Doc. 531 at 22–33; Singleton Doc. 357 at 15–17; Caster Doc. 437 at 7–27.

We begin with what Callais instructs must be “the focus” of Section Two: “enforcement of the Fifteenth Amendment's prohibition on intentional racial discrimination.” 146 S. Ct. at 1155. As Callais explained, although a proper interpretation of Section Two “does not demand a finding of intentional discrimination, it imposes liability only when the circumstances give rise to a strong inference that intentional discrimination occurred.” Id. at 1156. The circumstances here give rise to more than a strong inference — they support our separate and independent finding that intentional discrimination occurred. Accordingly, we are unsurprised to conclude at the end of the fresh analysis below that the Plaintiffs are likely to establish a Section Two violation under Callais.

Gingles/Callais I. First, a Section Two plaintiff must establish by a preponderance of the evidence that as a group, Black voters are sufficiently numerous and geographically compact to constitute a voting-age majority in a reasonably configured district. To do this, the plaintiff must produce an illustrative map with an additional remedial district that does not use race “as a districting criterion” and “meet[s] all the State's legitimate districting objectives, including traditional districting criteria and the State's specified political goals.” Callais, 146 S. Ct. at 1159.

Because the evidentiary record has not changed since we tried these cases, there is no basis for us to revisit our extensive findings of fact about numerosity (which was stipulated, Milligan Doc. 436 at 25), the geographic compactness of Alabama's Black population, and the reasonable configuration of the remedial districts in all of the Plaintiffs' illustrative maps in these cases. See Milligan Doc. 490 at 340–64. Accordingly, we adopt those findings now as though they were fully set forth herein.

Consistent with our instructions from the Supreme Court, we focus on what Callais added to Gingles. Callais has first instructed that “in drawing illustrative maps, plaintiffs cannot use race as a districting criterion.” Callais, 146 S. Ct. at 1159. At the outset, we reject the State's erroneous shorthand that Callais means that a cartographer simply may not “use race” at all. Milligan Doc. 534 at 31. Under longstanding precedent that Callais does not disturb, and as the State conceded at argument, a cartographer may consider “relevant racial data” so long as he “generate[s] that data solely for a lawful purpose, namely, to check that the maps he produced complied with ․ Voting Rights Act precedent.” Alexander, 602 U.S. at 22; May 22, 2026 Tr. at 171. Under Voting Rights Act precedent, “a community of minority voters must be sufficiently numerous and compact to constitute a majority in a reasonably configured district.” Callais, 146 S. Ct. at 1159; Bartlett v. Strickland, 556 U.S. 1, 19–20 (2009) (plurality op.). Indeed, “[t]he very reason a plaintiff adduces a map at the first step of Gingles is precisely because of its racial composition — that is, because it creates an additional majority-minority district that does not then exist.” Allen, 599 U.S. at 34 n.7. If a cartographer was unable to consider race even to determine whether a map he drew complies with Section Two precedent, no cartographer could comply with that precedent.

As we understand it, Callais forbids a particular use of race — namely, using “race as a districting criterion” “in drawing illustrative maps,” which “would be unconstitutional if a State engaged in such mapmaking.” 146 S. Ct. at 1159 (emphasis added). To support this bar, Callais cites the part of Alexander where the Supreme Court explained that “if a legislature gives race a predominant role in redistricting decisions, the resulting map is subjected to strict scrutiny and may be held unconstitutional.” 602 U.S. at 6.

On a careful re-examination of Dr. Duchin's testimony, we are satisfied that she did not “use race” unlawfully under Callais. We first found in 2022 that Dr. Duchin did not allow race to predominate in her work, which included four illustrative plans for these cases. Milligan Doc. 107 at 149, 206. When the Supreme Court affirmed that finding, it accepted and cited her testimony that her randomized algorithms “found plans with two majority-Black districts in literally thousands of different ways,” and that “ ‘it is certainly possible’ to draw the illustrative maps she produced in a race-blind manner.” Allen, 599 U.S. at 34 n.7. We made the same finding in 2025, when she submitted an additional illustrative map for trial. Milligan Doc. 490 at 364–69.

We have re-examined Dr. Duchin's trial testimony carefully, and we again credit her testimony 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,” in order to comply with the numerosity requirement of Bartlett; 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, as federal law requires. See Tr. at 287–89, 292, 365. We also credit her trial testimony that it was “not [her] process” to “look for majority-Black precincts” whenever she discovered that a map she was drawing had fallen below a 50%-Black threshold. Id. 365–66. We asked her no less than three times whether race informed the granular process of drawing maps, and she testified without contradiction that it did not. Id.; Milligan Doc. 490 at 122–24. Accordingly, we find that when Dr. Duchin considered race, it was for a purpose that remains lawful after Callais.

Separately, because of the unique posture of these cases, the record also includes a race-blind map that provides an additional opportunity district: the Special Master Plan. Mr. Ely prepared the Special Master Plan race-blind and “without reference to any other illustrative” or proposed plan. Milligan Doc. 311 at 19–20, 23, 44. As we said in 2023:

We well understand the legitimate concern about the role that considerations of race have in redistricting, but as we have found and the Supreme Court has affirmed, the record simply does not bear out that concern in this case. Nor can one fairly assert that the Special Master conducted his work in a way that runs afoul of the Equal Protection Clause.

Id. at 40 (internal citation to Allen omitted). As we also explained then, the Special Master Plan satisfies all constitutional and statutory requirements while hewing as closely as possible to the 2023 Plan (including the cap on county splits). See id. at 37–38.

We realize that the Special Master Plan includes an additional Black-opportunity district, not an additional majority-Black district,¹⁶ so we do not suggest that it satisfies Gingles I. We simply say that it further convinces us that Dr. Duchin was correct when she testified that it is possible to draw many race-blind illustrative maps, and that we correctly find that her own maps did not improperly use race.

In light of the powerful factual differences between Callais and this case, we are unsurprised that race-blind relief is available here but was unavailable there. Here, it has been consistently obvious to us from (1) our visual assessment of the geographic dispersion of Alabama's Black population and (2) statistics about Black population centers in the state, that Black voters in Alabama are relatively geographically compact. Milligan Docs. 107 at 160–61; 272 at 140–78; 490 at 336. There are areas where much of Alabama's Black population is concentrated, and many of these areas are in close proximity to each other. Milligan Docs. 107 at 161; 272 at 152–55; 490 at 336. Accordingly, we have consistently held that the additional majority-Black districts in the Plaintiffs' illustrative maps are reasonably configured, Milligan Docs. 107 at 173–74; 272 at 177–78; 490 at 369; Dr. Duchin has consistently testified that it is relatively easy to draw an additional reasonably configured majority-Black district, Milligan Docs. 107 at 54; 272 at 152–55; 490 at 324–64, and the Special Master prepared three remedial plans with reasonably-configured remedial districts with no apparent difficulty, Milligan Doc. 490 at 64–72.

This is not the case in Louisiana, where the geographic dispersion of the Black population required a remedial district to “slic[e] through” multiple “metropolitan areas to scoop up pockets of predominantly Black populations,” even though Louisiana does not “have concentrated populations of” Black voters. Callais, 732 F. Supp. 3d at 587–88.

Callais also requires that Plaintiffs' illustrative maps “meet all the State's legitimate [re]districting objectives, including traditional [re]districting criteria and the State's specified political goals.” 146 S. Ct. at 1159. The State contends that no illustrative map in this record satisfies this standard because no such map follows the Legislature's “non-negotiable” goal of keeping the Gulf Coast Counties whole and together while pairing zero incumbents. Milligan Doc. 534 at 33–34. But the State never seriously grapples with our finding that the novel “non-negotiable goal” about the Gulf Coast was the linchpin of a calculated effort to discriminate based on race that also included a novel categorical bar on pairing incumbents. See supra Part III pp. 45–46. Accordingly, on this record, we cannot treat the “non-negotiable goal” about the Gulf Coast as a “legitimate” objective that an illustrative plan must meet under Callais. 146 S. Ct. at 1159.

The State urges that we must blindly accept the “non-negotiable” goals expressed in the 2023 legislative findings, but this is badly wrong. Callais does not require that, and there is no other legal basis for it. If the findings had articulated partisan goals (which they did not), attacks on those goals would be non-justiciable in federal court. See Rucho v. Common Cause, 588 U.S. 684, 718 (2019). Otherwise, federal courts are duty-bound to consider a plaintiff's argument that a legislature adopted a particular goal or set of goals for unconstitutional purposes. An otherwise legitimate objective cannot stand if it is no more than pretext for an unconstitutional end. Cf. Griffin v. Cnty. Sch. Bd. of Prince Edward Cnty., 377 U.S. 218, 231 (1964) (holding that although a county has the legal power to close public schools, it could not do so for the purpose of preventing children from attending desegregated schools). Callais recognized this when it included an express limitation for goals “not prohibited by the Constitution,” 146 S. Ct. at 1159, and nothing in Callais alters the reality that the Constitution forbids legislatures from intentionally discriminating based on race in districting. The State's blind-acceptance theory would demolish constitutional guardrails on purposeful official discrimination, and we do not accept it.¹⁷

After trial, we found that many of the Plaintiffs' illustrative plans meet or beat the 2023 Plan on the legitimate goals expressed in the 2023 legislative findings. After a full review of voluminous evidence, we found that “all the arrows point in the same direction” because the illustrative 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[.]” Milligan Doc. 490 at 369.

We have carefully re-examined each component of that finding, and we adopt it again now. The supporting evidence is unchanged, and when we measure it against Callais, we arrive at the same findings and conclusions we did before.

Gingles/Callais II and III. Second and third, a Section Two plaintiff must establish that voting in the challenged districts is racially polarized, in that (2) Black voters are politically cohesive and (3) White voters ordinarily vote as a bloc to defeat Black-preferred candidates. To do this, a plaintiff must provide an analysis that “show[s] that voters engage in racial bloc voting that cannot be explained by partisan affiliation.” Callais, 146 S. Ct. at 1159. It must show that “black and white voters ha[ve] dramatically different voting patterns within” a party such that “the minority plaintiffs have ‘less opportunity’ than their majority counterparts because of race, not just because of partisan affiliation.” Id.

There has been no serious dispute throughout this litigation that voting in Alabama, particularly in the districts at issue, is intensely and extremely racially polarized, with extreme electoral results for Black-preferred candidates. We adhere to our earlier findings about the evidence of these patterns. See Milligan Doc. 490 at 369–78.

The dispute has been whether party or race drives the dynamic. Although controlling precedent did not previously require that we fully disentangle party and race, we understand that it now requires exactly that.

We have carefully re-examined all the evidence. Unlike in Callais, where there was no evidence that race specifically drove the polarized voting patterns identified by the Section Two plaintiffs, the record before us contains ample evidence that remains pertinent under Callais, including evidence of intra-party racial bloc voting. 146 S. Ct. at 1159, 1162.

Specifically, the following evidence in these cases is probative of disentanglement:

• Dr. Baodong Liu (an expert for the Milligan Plaintiffs) testified at length that “race is more important than party” in Alabama elections. Tr. at 584–87. In support of that assertion, he provided analysis showing: (1) that White Democrats in Alabama supported Senator John McCain over then-Senator Barack Obama in the 2008 Presidential Election, id. at 588–90 (referring to Milligan Doc. 403-13 at 14); (2) in 2024, the four Black candidates in the District 2 Republican primary together amassed only six percent of the primary vote, all of them finishing behind the four White candidates, Milligan Doc. 385-8 at 4; (3) in two nonpartisan Montgomery mayoral runoff elections (2019 and 2023), race drove the outcome when the party cue was taken away, Milligan Docs. 385-4 at 9–10, 385-8 at 8; and (4) in the 2021 District 1 Democratic primary, over 50% of Black voters supported the Black candidate, compared to only 16.7% of White voters, Tr. at 587–88.

• Dr. Christopher Bonneau (an expert for the State) testified at trial that he agreed with the ecological inference analysis prepared by one of the Caster Plaintiffs' experts, Dr. Maxwell Palmer, which established “that White voters in Alabama support White Democrats more than they support Black Democrats.” Id. at 1789.

• Dr. Joseph Bagley (an expert for the Milligan and Caster Plaintiffs) described substantial 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.” Id. at 1361–62.

• Dr. Adam Carrington (an expert for the State) agreed that Black and White Christians in Alabama hold similar views on abortion. Id. at 1621. Fact witnesses corroborated these assertions. Id. at 1106–07 (Pastor Valtoria Jackson), 1174 (Plaintiff Evan Milligan).

• Fact witnesses testified about a general lack of responsiveness to the Black community, regardless of party affiliation. For example, Bernard 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 (regardless of party) at those events. See Milligan Doc. 441-7 at 36.

• Alabamians have elected only two Black Republicans to represent a majority-White legislative district in the past 150 years,18 and no Black Alabamian (regardless of party) has won a statewide race in the past thirty years. These electoral results defeat the State's assertion that “White voters are willing to support minority candidates in large numbers.”

Additionally, the trial record includes evidence from the peer-reviewed scholarship of one of the State's own experts (Dr. Hood, who prepared the performance analysis we discussed in our finding of intentional discrimination) that race drives Southern voters' party attachments. According to Dr. Hood's scholarship, “[r]ace, 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.” Tr. at 587 (Dr. Liu, reading Milligan Doc. 403-17 at 5 (Ex. MX-17)).

The distinction between these facts and those of Callais underscores the point. In Callais, the Court found that the evidence “failed to disentangle race from politics” in part because in Louisiana, “black voters have been aligned with the Democratic party for decades and ․ issues discussed by that party appealed to black voters.” 146 S. Ct. at 1162. Here, as Dr. Bagley testified, Black Alabamians “consider themselves conservative, ․ perhaps, even fundamentalist conservatives or, more importantly, evangelical conservatives,” and yet “they do not vote for Republican candidates.” Milligan Doc. 490 at 167. Rather, Dr. Bonneau noted 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 395 (emphasis added). A clearer test for disentangling race from politics could hardly be asked for — if party politics drove voting patterns in Alabama, it is unclear why Black voters don't support the party that aligns more closely with their values.

The reality is that most White Alabamians are Republicans, most Black Alabamians are Democrats, and the Republican Party is Alabama's dominant political party. Dr. Hood's scholarship tells us that race drives these dynamics, and the other disentanglement evidence makes clear that the Black-White dichotomy he describes exists within both political parties in Alabama and transcends voters' partisan affiliations. We thus find from all the evidence that race is the core driver of voting behavior and election results in Alabama. And we cannot attribute Alabama's stark patterns of racial bloc voting, nor the extreme electoral failures of Black-preferred candidates to party politics. Accordingly, we find that the Plaintiffs have established that Black Alabamians in the districts at issue in these cases “have ‘less opportunity’ than their majority counterparts because of race, not just because of partisan affiliation.” Callais, 146 S. Ct. at 1159.

Finally, although we do not stake our findings about racially polarized voting on our finding of intentional discrimination, we observe that the evidentiary basis for our finding of intentional discrimination is consonant with our findings about racially polarized voting. The absence of any evidence that the Legislature refused to remedy racially discriminatory vote dilution because of party politics fits with our finding that considerations of race, not party politics, drives electoral behavior in Alabama.

Totality of the Circumstances. We thus turn to the evidence about the totality of the circumstances in today's Alabama. We have re-examined this evidence in light of the instruction in Callais that “[d]iscrimination that occurred some time ago, as well as present-day disparities that are characterized as the ongoing effects of societal discrimination, are entitled to much less weight” than “current data and current political conditions that shed light on current intentional discrimination.” 146 S. Ct. at 1160 (internal quotation marks omitted).

We find for each Senate Factor ¹⁹ as follows:

• Senate Factor 2: We continue to find that there is no serious dispute that voting in Alabama is intensely racially polarized. We credit Dr. Liu's testimony, which has consistently emphasized the clarity and extremity of the pattern of racially polarized voting he observes in Alabama. He has testified that racially polarized voting is “very clear” in Alabama, Jan. 10, 2022 Tr. at 1293; in the elections he studied, “Black support for [B]lack 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; accord Tr. at 573–78. We also credit Dr. Palmer's testimony about this, which was materially the same. Caster Doc. 49 ¶¶ 16–18; Caster Doc. 303-1 ¶¶ 14–15; Tr. at 487.

The State's experts do not dispute these patterns. See Jan. 11, 2022 Tr. at 1421–22; Tr. at 1726 (Dr. Bonneau).

We further credit Dr. Liu's trial testimony giving us perspective: that “in [his] more than 20 years [of] research, this is arguably the highest level” of racially polarized voting that he has “ever seen,” Tr. at 576; Milligan Doc. 490 at 371–72, and that the level of racially polarized voting in Alabama (particularly in the challenged districts) is “one of the highest in the nation,” Tr. at 578.

The findings of Dr. Liu and Dr. Palmer are based on recent data: the oldest election Dr. Liu analyzed occurred in 2010, and the most recent occurred in 2022, id. at 580, and the oldest election Dr. Palmer analyzed occurred in 2016, and the most recent occurred in 2022, id. at 490.

• Senate Factor 7: We find that Black candidates, regardless of party, are almost never elected to statewide or legislative public office in Alabama, based on the parties' stipulations and the undisputed facts that “(1) [s]ince 1994, no Black Alabamian, regardless of party, has won a statewide race,” Milligan Doc. 490 at 401 (citing Milligan Doc. 403-1 at 29)20; “(2) in 1992, Representative Earl Hilliard was the first Black Alabamian elected to Congress since Reconstruction,” id. (citing Milligan Doc. 436 ¶ 103); “(3) Representative Shomari Figures is the first Black Alabamian to be elected to Congress outside of District 7 since Reconstruction,” id. (citing Milligan Doc. 436 ¶¶ 103, 106, 108, 113–14, 151); and (4) “[t]hirty-two (32) out of thirty-three (33) Black Alabamians currently serving in the Alabama Legislature were elected from majority-Black districts,” id. (citing Milligan Doc. 436 ¶ 155), which were created to comply with the Voting Rights Act or the Constitution, id. (citing Tr. at 1927–28).

We note again that unlike in Callais, where the evidence of the “low number of Black Louisianans who have been elected to Congress in recent decades ․ failed to disentangle race from politics,” Callais, 146 S. Ct. at 1162, here we have pointed to substantial evidence that race, not politics, is at work. See supra Part IV pp. 57–61. The result is a lack of electoral success for Black candidates “regardless of party.” Milligan Doc. 403-1 at 29.

• Senate Factors 1 and 3: We find substantial evidence of official race discrimination in Alabama in the past twenty years.

○ First, 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 (N.D. Ala. Dec. 16, 2019); Ala. State Conf. of the NAACP v. City of Pleasant Grove, No. 18-cv-02056, 2019 WL 5172371 (N.D. Ala. Oct. 11, 2019).

○ Second, 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; Allen v. City of Evergreen, No. 13-0107, 2014 WL 12607819, at *2 (S.D. Ala. Jan. 13, 2014).

○ Third, 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 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).

○ Fourth, after the 2010 census, Black voters and legislators successfully challenged twelve state legislative districts as unconstitutional racial gerrymanders. See Ala. Legis. Black Caucus, v. Alabama, 231 F. Supp. 3d 1026, 1348–49 (M.D. Ala. 2017).

○ Fifth, in United States v. McGregor, 824 F. Supp. 2d 1339 (M.D. Ala. 2011), a federal court found that Alabama State Senators conspired in 2010 to depress Black voter turnout by keeping a referendum issue popular among Black voters (whom the Senators called “Aborigines”) off the ballot. Id. at 1345–47.

○ Sixth, we credit Dr. Bagley's testimony that (1) Alabama is the only state to have more than one jurisdiction bailed back into federal preclearance requirements since Shelby County v. Holder, 570 U.S. 529 (2013), see Tr. 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.

• Senate Factor 5. In light of Callais, this Factor receives less weight than the other Factors. We credit the evidence that based on recent data, Black Alabamians “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,” Gingles, 478 U.S. at 45, as follows:

○ We credit Dr. Bagley's testimony 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 a 2019 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.

○ We also credit Dr. Bagley's testimony that for the 2020–21 school year, “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.

○ We credit the testimony by another Plaintiffs' expert, Dr. Traci Burch, that 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), Milligan Doc. 385-2 at 30; Black Alabama households are more than twice as likely to lack access to a vehicle at home (12%) 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. at 928, 943, 946, 957–67.

○ We also credit Dr. Burch's testimony that Black Alabamians have significantly lower educational attainment than White Alabamians, Milligan Doc. 385-2 at 4, particularly among school-aged children, id. at 12. We also credit her testimony that “in the Black Belt especially, there are ․ disproportionately high illiteracy rates, as high as 30 percent.” Tr. at 938, 998, 1049. Likewise, we credit the expert testimony tracing current disparities in educational attainment to Alabama's pattern of “historical and contemporary discrimination in elementary, secondary, and higher education,” including “separate-but-unequal” education. Milligan Doc. 490 at 413; Milligan Docs. 68-2 at 17, 385-2 at 12; Tr. at 934–35, 1396–98. Their testimony fits with testimony we heard from fact witnesses who attended segregated public schools as teenagers. See, e.g., Milligan Docs. 463 at 239; 467 at 35.

○ Finally, we credit Dr. Burch's testimony (which no one disputes) linking Alabama's history of segregated public schools to present-day racial disparities in voting participation: she explained that in “[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. at 936. And she testified that segregation resulted in fewer opportunities for Black people to attend college which is why Black people today “are disproportionately concentrated in these lower educational attainment — lower voter turnout groups.” Id. at 937– 38.

We emphasize that we are faithful to the instruction in Callais to assign this factor far less weight than other factors. We simply say that the record indicates that racial disparities in access to decent public schools and basic modern infrastructure (water, sewage, communication, and transportation) affect Black Alabamians' participation in the political process.

• Senate Factor 6. We find that some recent political campaigns have been characterized by overt or subtle racial appeals, but that this Factor weighs in favor of the Plaintiffs to a lesser degree than do other Factors. We credit Dr. Bagley's testimony about racial appeals by candidates in three recent congressional elections, see Milligan Docs. 68-2 at 26–28, 385-1 at 31–32, as follows:

○ First, when a former Chief Justice of the Alabama Supreme Court, Roy Moore, ran for U.S. Senate in 2017, he won the Republican Party nomination. In 2011, the year before he was elected to that Court, he said during an interview that the amendments to the Constitution that follow the Tenth Amendment have “completely tried to wreck the form of government that our forefathers intended.” See Milligan Doc. 68-2 at 27. During his 2017 Senate campaign, he acclaimed the antebellum period in the South: “I think it was great at the time when families were united — even though we had slavery․ Our families were strong. Our country had a direction.” See id.

○ Second, former Congressman Mo Brooks repeatedly claimed (in 2014, 2017, and 2020) that Democrats are waging a “war on [W]hites.” See Milligan Doc. 68-2 at 27–28 & n.94.

○ Third, we credit Dr. Bagley's testimony about the 2020 campaign ad about a campfire from former Congressman Bradley Byrne. Milligan Docs. 68-2 at 28, 107 at 190–91.21 We credit the evidence that a reasonable viewer might have perceived the commercial as a racial appeal because it shows a White man narrating as images of prominent people of color (and only people of color) are juxtaposed with images of the 9/11 terrorist attacks, in or on or hovering above a crackling fire.

The record does not contain any systematic evaluation of the extent to which campaigns in Alabama involve racial appeals, so we cannot determine whether these examples indicate that they occur frequently, regularly, occasionally, or rarely.

• Senate Factors 8 and 9. In our first two preliminary injunctions, we made no findings about whether the policy underlying the 2021 and 2023 Plans was tenuous. However, we found that Senate Factor 9 weighed strongly in favor of the Plaintiffs when issuing our permanent injunction. Milligan Doc. 490 at 423. We make that same finding here.

○ This finding rests on evidence the Legislature made during the pendency of these cases. We find that the unusual genesis of the 2023 Plan; the Legislature's departures from the norm in the 2023 Special Session; the novelty, substance, and effect of the 2023 legislative findings; our ruling about the intended purpose of the 2023 Plan; and the absence of evidence that the Legislature acted for partisan reasons, all support a finding of tenuousness. “This is so even though some of the individual districting decisions reflected in the 2023 Plan may be legitimate in another context.” Milligan Doc. 490 at 423.

○ Ultimately, the 2023 Special Session saw the Legislature consider and reject a map that might have provided the required remedy (an issue we do not decide), in favor of a map that it knew in real time and later admitted in court certainly does not provide the remedy we said was required, 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. We cannot find that this was about compactness.

In our second preliminary injunction, we found that Senate Factor 8 tilts strongly in favor of the Plaintiffs because “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 adhere to that finding now. In the simplest terms, the provenance of the 2023 Plan means that we cannot find that when the Legislature passed that Plan, it was trying in earnest to respond to the need we identified for Black Alabamians not to have their votes diluted.

Accordingly, we find that every Senate Factor we examine favors the Plaintiffs. We further find that every piece of this intensely local corpus of evidence tells us the same thing: things are still different here in Alabama.

Finally, we observe that we have already found in this case what the totality of the circumstances analysis is meant to demonstrate: “an objective likelihood of intentional discrimination.” Callais, 146 S. Ct. at 1162. Under the unaltered standard of Arlington Heights, we have found that the 2023 Plan represents an intentional effort to crack the Black population in Alabama. See supra Part III. We have also considered and ruled out alternative, non-discriminatory explanations, including purely partisan motives. See supra Part III pp. 41–47; Callais, 146 S. Ct. at 1156–57. Unlike in Callais, Alabama has not been “forthright” about any partisan objective that it may have had in drawing the 2023 Plan — indeed, we find no evidence from when the 2023 Plan was actually drafted to indicate that the Legislature was moved by a desire for partisan advantage. Callais, 146 S. Ct. at 1163.

***

For the foregoing reasons, we find that under Gingles as updated by Callais, the Plaintiffs are likely to prevail on the merits of their claim that under the totality of the circumstances in Alabama today, Black voters have less opportunity than other Alabamians to elect candidates of their choice.

V. REMAINING ELEMENTS FOR PRELIMINARY INJUNCTION

To obtain a preliminary injunction, the Plaintiffs also must establish “that [they are] likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in [their] favor, and that an injunction is in the public interest.” Benisek, 585 U.S. at 158 (quoting Winter v. Nat. Res. Def. Council, Inc., 555 U.S. 7, 20 (2008)). We discuss each element in turn:

Irreparable Harm. As discussed in connection with Purcell, the Plaintiffs will suffer irreparable harm if we allow the Secretary to administer Alabama's 2026 congressional elections under a plan that is unconstitutional because it purposefully discriminates based on race. “Courts routinely deem restrictions on fundamental voting rights irreparable injury. And discriminatory voting procedures in particular are the kind of serious violation of the Constitution and the Voting Rights Act for which courts have granted immediate relief.” League of Women Voters, 769 F.3d at 247 (citation modified) (citing, inter alia, Obama for Am. v. Husted, 697 F.3d 423, 436 (6th Cir. 2012)). “Voting is the beating heart of democracy,” and a “fundamental political right, because it is preservative of all rights.” Democratic Exec. Comm. of Fla., 915 F.3d at 1315 (quoting Yick Wo v. Hopkins, 118 U.S. 356, 370 (1886)). And “once the election occurs, there can be no do-over and no redress” for constitutional violations. League of Women Voters, 769 F.3d at 247.

The State does not argue that if an election occurs under an intentionally discriminatory map, the injury to voters is reparable. Rather, the State argues that if an injunction issues, it faces an irreparable injury to its sovereign interest in “enforc[ing] duly enacted plans.” See Milligan Doc. 534 at 91 (quoting Abbott, 585 U.S. at 602 n.17). This argument ignores the likely constitutional and statutory defects we find in the 2023 Plan. And in any event, in a contest between irreparable harm to voters' right to participate in the political process under a constitutional districting plan, and irreparable harm to the State's interest in using its legislatively enacted plan, we must give greater weight to preventing the irreparable injury to voters' constitutional rights.

Public Interest and Equities. We analyze these elements together because many of the parties' arguments about them overlap. See Gonzalez v. Governor of Georgia, 978 F.3d 1266 (11th Cir. 2020) (balance of equities and public interest factors merge when the government is a party); Texas v. United States, 809 F.3d 134, 187 (5th Cir. 2015) (“[The public interest] factor overlaps considerably with the [balance of equities factor], and most of the same analysis applies.”).

At the outset, we reject the State's argument that the Supreme Court “sided with” the State on the equities. Milligan Doc. 534 at 79–80. The summary vacatur did not analyze the merits of any arguments, and we examine the parties' contentions about balancing the equities on a clean slate. See United States v. Ardley, 242 F.3d 989, 990 (11th Cir. 2001) (per curiam) (explaining that a vacate-and-remand order “implies nothing about what a [lower court] should do on remand, except that it is to reconsider the case in light of the intervening decision of the Supreme Court” (citation modified)).

We find that a preliminary injunction is in the public interest and that the equities tip decidedly in favor of a preliminary injunction for five reasons. First, our preliminary injunction prevents voters from being forced to participate in the electoral process under a districting plan tainted by intentional race-based discrimination. Second, our preliminary injunction allows voters to participate in that process under a race-blind plan that satisfies all applicable constitutional and statutory requirements while hewing as closely as possible to the State's legislatively enacted plan.

Third, there is an unmeasurable risk of error attendant to allowing elections to proceed under the 2023 Plan. Director Elrod testified that if the 2026 elections proceed under the 2023 Plan, he and affected county officials will have just seven days to successfully redistrict voters after voter rolls “unlock” on May 27, 2026. May 22, 2026 Tr. at 37. He was unaware of any Alabama redistricting effort accomplished on such a breakneck schedule, admitted that he was unsure if the counties could successfully complete their task, and testified about previous sworn statements by the Secretary's office (in these cases and others) to the effect that reassignment efforts ordinarily take weeks or months, not days. See id. at 36, 56, 79, 96. Director Elrod explained that many Alabama counties have only three registrars, who are sometimes not full-time employees. Id. at 62. He candidly acknowledged that redistricting voters for the 2026 elections to occur under the 2023 Plan would require an immense effort that required “all hands on deck” working hours during late nights or on weekends. Id. at 77. And he acknowledged that errors could include balloting errors including voters receiving the wrong ballots. May 22, 2026 Tr. at 35, 39–40, 58, 81. It is obvious to us that if the Secretary administers the remainder of Alabama's 2026 elections with the 2023 Plan, there is a significant risk of large-scale election mismanagement and error.

Fourth, these practical exigencies necessitate a familiar race-blind remedy. As explained more fully below, we impose a race-blind remedy that alleviates the need for extensive redistricting by the Secretary. The State and its electorate conducted all events in the 2026 elections until just two weeks ago on the expectation that candidates would run and votes would be cast under the Special Master Plan. The Special Master Plan has been the practical status quo in Alabama since we imposed it in 2023, it was used for the 2024 congressional elections, and it was upended during an ongoing primary election. Candidates staked campaigns on the Special Master Plan, voters cast votes for candidates under this Plan, and Alabama's election machinery was in full motion before the Supreme Court's ruling. Rather than spawning electoral chaos and serious error potential, the Special Master Plan provides for certainty.

And fifth, we impose a limited remedy. This preliminary injunction (our third) is time-limited and respects the Legislature's authority to redistrict. We have continuously regarded our task as most unwelcome, and this remedy reflects this restraint.

As we understand the law, when a plaintiff asserts a meritorious claim of intentional race-based discrimination under the Fourteenth Amendment, voters should be forced to cast a vote based on an unconstitutional plan only if absolutely necessary. There is no convincing evidence that it is necessary for us to allow Alabama to pivot to the 2023 Plan in the middle of an election, and substantial evidence that it is not.

VI. REMEDY

A State has a “sovereign interest in implementing its redistricting plan.” Bush v. Vera, 517 U.S. 952, 978 (1996). Even when a federal court finds a redistricting plan unlawful, the Supreme Court “has repeatedly held that redistricting ․ is a legislative task which the federal courts should make every effort not to pre-empt.” Wise v. Lipscomb, 437 U.S. 535, 539–40 (1978) (opinion of White, J.) (collecting cases). Upon such a finding, it is “appropriate, whenever practicable, to afford a reasonable opportunity for the legislature to meet [applicable federal legal] requirements by adopting a substitute measure rather than for the federal court to devise and order into effect its own plan.” Id. at 540.

Only when a state legislature fails to adopt a lawful remedial map in time for an upcoming election does the job of drawing an interim map fall to the courts. “Legislative bodies should not leave their reapportionment tasks to the federal courts; but when those with legislative responsibilities do not respond, or the imminence of a state election makes it impractical for them to do so, it becomes the unwelcome obligation of the federal court to devise and impose a reapportionment plan pending later legislative action.” Id. (internal quotation marks and citation omitted); accord Growe v. Emison, 507 U.S. 25, 36–37 (1993).

“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.” League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 415–16 (2006) (citation omitted).

In this iteration of these cases, for all the foregoing reasons, we find ourselves in a position where the Legislature has not adopted a lawful remedial map in time for Alabama's 2026 elections. Accordingly, it falls to us to order a remedy. This is the case regardless of whether we are right about our constitutional findings, our Section Two findings, or both.

The Special Master Plan is the obvious remedy. We have re-examined our earlier detailed finding that it is race-blind and satisfies all applicable constitutional and statutory requirements, while hewing as closely as possible to the 2023 Plan, see Milligan Doc. 311, and we adhere to that finding. And, Alabama has used the Special Master Plan not only for the last congressional elections, but also for the 2026 congressional elections until just fifteen days ago. It is the only available remedy for taint of intentional discrimination and/or likely vote dilution in the 2023 Plan that addresses the serious exigencies described to us by Director Elrod in connection with the 2026 congressional elections. And we require only a limited use of it – one that, until May 11, the State expected to use anyway.

VII. PRELIMINARY INJUNCTION

For the foregoing reasons, the motions for preliminary injunctive relief are GRANTED IN PART AND DENIED IN PART. They are DENIED to the extent that they request we require the Secretary to certify and canvass the results from Alabama's May 19 primary elections. Singleton Doc. 357 at 19; Milligan Doc. 531 at 37. As we understand it, those election results (and/or any such certification) became void under an Alabama law that no party asks us to enjoin. See Ala. Code § 17-13-3.1(d). The motions also are DENIED to the extent that they request preliminary relief to preserve the status quo, because under controlling precedent the 2023 Plan is the status quo ante.

The motions are GRANTED only as follows, and pursuant to Federal Rule of Civil Procedure 65(d), this Court ORDERS the following:

1. The Court PRELIMINARILY ENJOINS the Alabama Secretary of State from conducting the 2026 congressional elections according to the 2023 Plan.

2. The Court further ORDERS the Alabama Secretary of State to administer all remaining events comprising Alabama's 2026 congressional elections according to the Special Master Plan (appended to this Order as Appendix E). The part of the Court's preliminary injunction set forth in this paragraph EXPIRES upon Alabama's legislative enactment of a new congressional districting plan.

We expressly leave to the State's discretion (as we must) how it addresses candidate qualification and other matters for the remaining events in Alabama's 2026 congressional elections.

We RESERVE RULING on the grounds for a preliminary injunction asserted by the Singleton Plaintiffs that we have not discussed in this Order, including but not limited to the “Singleton Plan.” Singleton Doc. 357.

VIII. STAY AND SCHEDULING ORDER

In the interests of time and judicial economy, and at the State's request, Milligan Doc. 534 at 92, we have considered whether to stay our preliminary injunction. For all the same reasons we found that the Plaintiffs are likely to prevail on the merits of their claims and the equities tip in their favor, and in recognition of the scarcity of time, we DENY the State's motion for a stay.

We are more than halfway through this ten-year census cycle, and these cases must proceed without delay. We therefore enter the following scheduling order with leave for the parties to propose amendments for good cause:

1. Discovery: Fact and expert discovery are open, and the parties SHALL meet, confer, and jointly propose on or before June 22, 2026, applicable limits on depositions, interrogatories, and requests for production and admission, as well as deadlines for the exchange of expert reports.

2. Trial: The parties SHALL make every effort to ensure this case will be trial ready not later than January 11, 2027, and must timely advise the Court how many days trial is expected to consume.

DONE and ORDERED this 26th day of May, 2026.

APPENDIX A – COMMITTEE GUIDELINES (2021 and 2023)

REAPPORTIONMENT COMMITTEE REDISTRICTING GUIDELINES

May 5, 2021

I. POPULATION

The total Alabama state population, and the population of defined subunits thereof, as reported by the 2020 Census, shall be the permissible data base used for the development, evaluation, and analysis of proposed redistricting plans. It is the intention of this provision to exclude from use any census data, for the purpose of determining compliance with the one person, one vote requirement, other than that provided by the United States Census Bureau.

II. CRITERIA FOR REDISTRICTING

a. Districts shall comply with the United States Constitution, including the requirement that they equalize total population.

b. Congressional districts shall have minimal population deviation.

c. Legislative and state board of education districts shall be drawn to achieve substantial equality of population among the districts and shall not exceed an overall population deviation range of ± plusmn;5%.

d. A redistricting plan considered by the Reapportionment Committee shall comply with the one person, one vote principle of the Equal Protection Clause of the 14th Amendment of the United States Constitution.

e. The Reapportionment Committee shall not approve a redistricting plan that does not comply with these population requirements.

f. Districts shall be drawn in compliance with the Voting Rights Act of 1965, as amended. A redistricting plan shall have neither the purpose nor the effect of diluting minority voting strength, and shall comply with Section 2 of the Voting Rights Act and the United States Constitution.

g. No district will be drawn in a manner that subordinates race-neutral districting criteria to considerations of race, color, or membership in a language-minority group, except that race, color, or membership in a language-minority group may predominate over race-neutral districting criteria to comply with Section 2 of the Voting Rights Act, provided there is a strong basis in evidence in support of such a race-based choice. A strong basis in evidence exists when there is good reason to believe that race must be used in order to satisfy the Voting Rights Act.

h. Districts will be composed of contiguous and reasonably compact geography.

i. The following requirements of the Alabama Constitution shall be complied with:

(i) Sovereignty resides in the people of Alabama, and all districts should be drawn to reflect the democratic will of all the people concerning how their governments should be restructured.

(ii) Districts shall be drawn on the basis of total population, except that voting age population may be considered, as necessary to comply with Section 2 of the Voting Rights Act or other federal or state law.

(iii) The number of Alabama Senate districts is set by statute at 35 and, under the Alabama Constitution, may not exceed 35.

(iv) The number of Alabama Senate districts shall be not less than one-fourth or more than one-third of the number of House districts.

(v) The number of Alabama House districts is set by statute at 105 and, under the Alabama Constitution, may not exceed 106.

(vi) The number of Alabama House districts shall not be less than 67.

(vii) All districts will be single-member districts.

(viii) Every part of every district shall be contiguous with every other part of the district.

j. The following redistricting policies are embedded in the political values, traditions, customs, and usages of the State of Alabama and shall be observed to the extent that they do not violate or subordinate the foregoing policies prescribed by the Constitution and laws of the United States and of the State of Alabama:

(i) Contests between incumbents will be avoided whenever possible.

(ii) Contiguity by water is allowed, but point-to-point contiguity and long-lasso contiguity is not.

(iii) Districts shall respect communities of interest, neighborhoods, and political subdivisions to the extent practicable and in compliance with paragraphs a through i. A community of interest is defined as an area with recognized similarities of interests, including but not limited to ethnic, racial, economic, tribal, social, geographic, or historical identities. The term communities of interest may, in certain circumstances, include political subdivisions such as counties, voting precincts, municipalities, tribal lands and reservations, or school districts. The discernment, weighing, and balancing of the varied factors that contribute to communities of interest is an intensely political process best carried out by elected representatives of the people.

(iv) The Legislature shall try to minimize the number of counties in each district.

(v) The Legislature shall try to preserve the cores of existing districts.

(vi) In establishing legislative districts, the Reapportionment Committee shall give due consideration to all the criteria herein. However, priority is to be given to the compelling State interests requiring equality of population among districts and compliance with the Voting Rights Act of 1965, as amended, should the requirements of those criteria conflict with any other criteria.

g. The criteria identified in paragraphs j(i)-(vi) are not listed in order of precedence, and in each instance where they conflict, the Legislature shall at its discretion determine which takes priority.

III. PLANS PRODUCED BY LEGISLATORS

1. The confidentiality of any Legislator developing plans or portions thereof will be respected. The Reapportionment Office staff will not release any information on any Legislator's work without written permission of the Legislator developing the plan, subject to paragraph two below.

2. A proposed redistricting plan will become public information upon its introduction as a bill in the legislative process, or upon presentation for consideration by the Reapportionment Committee.

3. Access to the Legislative Reapportionment Office Computer System, census population data, and redistricting work maps will be available to all members of the Legislature upon request. Reapportionment Office staff will provide technical assistance to all Legislators who wish to develop proposals.

4. In accordance with Rule 23 of the Joint Rules of the Alabama Legislature “[a]ll amendments or revisions to redistricting plans, following introduction as a bill, shall be drafted by the Reapportionment Office.” Amendments or revisions must be part of a whole plan. Partial plans are not allowed.

5. In accordance with Rule 24 of the Joint Rules of the Alabama Legislature, “[d]rafts of all redistricting plans which are for introduction at any session of the Legislature, and which are not prepared by the Reapportionment Office, shall be presented to the Reapportionment Office for review of proper form and for entry into the Legislative Data System at least ten (10) days prior to introduction.”

IV. REAPPORTIONMENT COMMITTEE MEETINGS AND PUBLIC HEARINGS

1. All meetings of the Reapportionment Committee and its sub-committees will be open to the public and all plans presented at committee meetings will be made available to the public.

2. Minutes of all Reapportionment Committee meetings shall be taken and maintained as part of the public record. Copies of all minutes shall be made available to the public.

3. Transcripts of any public hearings shall be made and maintained as part of the public record, and shall be available to the public.

4. All interested persons are encouraged to appear before the Reapportionment Committee and to give their comments and input regarding legislative redistricting. Reasonable opportunity will be given to such persons, consistent with the criteria herein established, to present plans or amendments redistricting plans to the Reapportionment Committee, if desired, unless such plans or amendments fail to meet the minimal criteria herein established.

5. Notice of all Reapportionment Committee meetings will be posted on monitors throughout the Alabama State House, the Reapportionment Committee's website, and on the Secretary of State's website. Individual notice of Reapportionment Committee meetings will be sent by email to any citizen or organization who requests individual notice and provides the necessary information to the Reapportionment Committee staff. Persons or organizations who want to receive this information should contact the Reapportionment Office.

V. PUBLIC ACCESS

1. The Reapportionment Committee seeks active and informed public participation in all activities of the Committee and the widest range of public information and citizen input into its deliberations. Public access to the Reapportionment Office computer system is available every Friday from 8:30 a.m. to 4:30 p.m. Please contact the Reapportionment Office to schedule an appointment.

2. A redistricting plan may be presented to the Reapportionment Committee by any individual citizen or organization by written presentation at a public meeting or by submission in writing to the Committee. All plans submitted to the Reapportionment Committee will be made part of the public record and made available in the same manner as other public records of the Committee.

3. Any proposed redistricting plan drafted into legislation must be offered by a member of the Legislature for introduction into the legislative process.

4. A redistricting plan developed outside the Legislature or a redistricting plan developed without Reapportionment Office assistance which is to be presented for consideration by the Reapportionment Committee must:

a. Be clearly depicted on maps which follow 2020 Census geographic boundaries;

b. Be accompanied by a statistical sheet listing total population for each district and listing the census geography making up each proposed district;

c. Stand as a complete statewide plan for redistricting.

d. Comply with the guidelines adopted by the Reapportionment Committee.

5. Electronic Submissions

a. Electronic submissions of redistricting plans will be accepted by the Reapportionment Committee,

b. Plans submitted electronically must also be accompanied by the paper materials referenced in this section.

c. See the Appendix for the technical documentation for the electronic submission of redistricting plans.

6. Census Data and Redistricting Materials

a. Census population data and census maps will be made available through the Reapportionment Office at a cost determined by the Permanent Legislative Committee on Reapportionment.

b. Summary population data at the precinct level and a statewide work maps will be made available to the public through the Reapportionment Office at a cost determined by the Permanent Legislative Committee on Reapportionment.

c. All such fees shall be deposited in the state treasury to the credit of the general fund and shall be used to cover the expenses of the Legislature.

Appendix.

ELECTRONIC SUBMISSION OF REDISTRICTING PLANS REAPPORTIONMENT COMMITTEE -STATE OF ALABAMA

The Legislative Reapportionment Computer System supports the electronic submission of redistricting plans. The electronic submission of these plans must be via email or a flash drive. The software used by the Reapportionment Office is Maptitude.

The electronic file should be in DOJ format (Block, district # or district #, Block). This should be a two column, comma delimited file containing the FIPS code for each block, and the district number. Maptitude has an automated plan import that creates a new plan from the block/district assignment list.

Web services that can be accessed directly with a URL and ArcView Shapefiles can be viewed as overlays. A new plan would have to be built using this overlay as a guide to assign units into a blank Maptitude plan. In order to analyze the plans with our attribute data, edit, and report on, a new plan will have to be built in Maptitude.

In order for plans to be analyzed with our attribute data, to be able to edit, report on, and produce maps in the most efficient, accurate and time saving procedure, electronic submissions are REQUIRED to be in DOJ format.

Example: (DOJ FORMAT BLOCK, DISTRICT #)

SSCCCTTTTTTBBBBDDDD

SS is the 2 digit state FIPS code

CCC is the 3 digit county FIPS code

TTTTTT is the 6 digit census tract code

BBBB is the 4 digit census block code

DDDD is the district number, right adjusted

Contact Information:

Legislative Reapportionment Office

Room 317, State House

11 South Union Street

Montgomery, Alabama 36130

(334) 261-0706

For questions relating to reapportionment and redistricting, please contact:

Donna Overton Loftin, Supervisor

Legislative Reapportionment Office donna. overton@ alsenate.gov

Please Note: The above e-mail address is to be used only for the purposes of obtaining information regarding redistricting. Political messages, including those relative to specific legislation or other political matters, cannot be answered or disseminated via this email to members of the Legislature. Members of the Permanent Legislative Committee on Reapportionment may be contacted through information contained on their Member pages of the Official Website of the Alabama Legislature, legislature.state.al.us/aliswww/default.aspx.

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

Enrolled, An Act,

To amend Section 17-14-70, Code of Alabama 1975, to provide for the reapportionment and redistricting of the state's United States Congressional districts for the purpose of electing members at the General Election in 2024 and thereafter, until the release of the next federal census; and to add Section 17-40-70.1 to the Code of Alabama 1975, to provide legislative findings.

BE IT ENACTED BY THE LEGISLATURE OF ALABAMA:

Section 1. Section 17-14-70.1 is added to the Code of Alabama 1975, to read as follows.

§ 17-14-70.1

The Legislature finds and declares the following:

(1) The Legislature adheres to traditional redistricting principles when adopting congressional districts. Such principles are the product of history, tradition, bipartisan consensus, and legal precedent. The Supreme Court of the United States recently clarified that Section 2 of the Voting Rights Act “never requires adoption of districts that violate traditional redistricting principles.”

(2) The Legislature's intent in adopting the congressional plan in this act described in Section 17-14-70.1 is to comply with federal law, including the U.S. Constitution and the Voting Rights Act of 1965, as amended.

(3) The Legislature's intent is also to promote the following traditional redistricting principles, which are given effect in the plan created by this act:

a. Districts shall be based on total population as reported by the federal decennial census and shall have minimal population deviation.

b. Districts shall be composed of contiguous geography, meaning that every part of every district is contiguous with every other part of the same district.

c. Districts shall be composed of reasonably compact geography.

d. The congressional districting plan shall contain no more than six splits of county lines, which is the minimum number necessary to achieve minimal population deviation among the districts. Two splits within one county is considered two splits of county lines.

e. The congressional districting plan shall keep together communities of interest, as further provided for in subdivision (4).

f. The congressional districting plan shall not pair incumbent members of Congress within the same district.

g. The principles described in this subdivision are non-negotiable for the Legislature. To the extent the following principles can be given effect consistent with the principles above, the congressional districting plan shall also do all of the following:

1. Preserve the cores of existing districts.

2. Minimize the number of counties in each district.

3. Minimize splits of neighborhoods and other political subdivisions in addition to minimizing the splits of counties and communities of interest.

(4)a. 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, educational institutions, and historical or cultural factors.

b. The discernment, weighing, and balancing of the varied factors that contribute to communities of interest is an intensely political process best carried out by elected representatives of the people.

c. If it is necessary to divide a community of interest between congressional districts to promote other traditional districting principles like compactness, contiguity, or equal population, division into two districts is preferable to division into three or more districts. Because each community of interest is different, the division of one community among multiple districts may be more or less significant to the community than the division of another community.

d. The Legislature declares 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.

e. l. Alabama's Black Belt region is a community of interest composed of the following 18 core counties: Barbour, Bullock, Butler, Choctaw, Crenshaw, Dallas, Greene, Hale, Lowndes, Macon, Marengo, Montgomery, Perry, Pickens, Pike, Russell, Sumter, and Wilcox. Moreover, the following five counties are sometimes considered part of the Black Belt: Clarke, Conecuh, Escambia, Monroe, and Washington.

2. The Black Belt is characterized by its rural geography, fertile soil, and relative poverty, which have shaped its unique history and culture.

3. The Black Belt region spans the width of Alabama from the Mississippi boarder to the Georgia border.

4. Because the Black Belt counties cannot be combined within one district without causing other districts to violate the principle of equal population among districts, the 18 core Black Belt counties shall be placed into two reasonably compact districts, the fewest number of districts in which this community of interest can be placed. Moreover, of the five other counties sometimes considered part of the Black Belt, four of those counties are included within the two Black Belt districts -Districts 2 and 7.

f.l. Alabama's Gulf Coast region is a community of interest composed of Mobile and Baldwin Counties.

2. Owing to Mobile Bay and the Gulf of Mexico coastline, these counties also comprise a well-known and well-defined community with a long history and unique interests. Over the past half-century, Baldwin and Mobile Counties have grown even more alike as the tourism industry has grown and the development of highways and bay-crossing bridges have made it easier to commute between the two counties.

3. The Gulf Coast community has a shared interest in tourism, which is a multi-billion-dollar industry and a significant and unique economic driver for the region.

4. Unlike other regions in the state, the Gulf Coast community is home to major fishing, port, and ship-building industries. Mobile has a Navy shipyard and the only deep-water port in the state. The port is essential for the international export of goods produced in Alabama,

5. The Port of Mobile is the economic hub for the Gulf counties. Its maintenance and further development are critical for the Gulf counties in particular but also for many other parts of the state. The Port of Mobile handles over 55 million tons of international and domestic cargo for exporters and importers, delivering eighty-five billion dollars ($85,000,000,000) in economic value to the state each year. Activity at the port's public and private terminals directly and indirectly generates nearly 313,000 jobs each year.

6. Among the over 21,000 direct jobs generated by the Port of Mobile, about 42% of the direct jobholders reside in the City of Mobile, another 39% reside in Mobile County but outside of the City of Mobile, and another 13% reside in Baldwin County.

7. The University of South Alabama serves the Gulf Coast community of interest both through its flagship campus in Mobile and its campus in Baldwin County.

8. Federal appropriations have been critical to ensuring the port's continued growth and maintenance. In 2020, the Army Corps of Engineers allocated over two hundred seventy-four million dollars ($274,000,000) for the Port of Mobile to allow the dredging and expansion of the port. Federal appropriations have also been critical for expanding bridge projects to further benefit the shared interests of the region.

9. The Gulf Coast community has a distinct culture stemming from its French and Spanish colonial heritage. That heritage is reflected in the celebration of shared social occasions, such as Mardi Gras, which began in Mobile. This shared culture is reflected in Section 1-3-8(c), Code of Alabama 1975, which provides that “Mardi Gras shall be deemed a holiday in Mobile and Baldwin Counties and all state offices shall be closed in those counties on Mardi Gras.” Mardi Gras is observed as a state holiday only in Mobile and Baldwin Counties.

10. Mobile and Baldwin Counties also work together as part of the South Alabama Regional Planning Commission, a regional planning commission recognized by the state for more than 50 years. The local governments of Mobile, Baldwin, and Escambia Counties, as well as 29 municipalities within those counties, work together through the commission with the Congressional Representative from District 1 to carry out comprehensive economic development planning for the region in conjunction with the U.S. Economic Development Administration. Under Section 11-85-51(b), factors the Governor considers when creating such a regional planning commission include “community of interest and homogeneity; geographic features and natural boundaries; patterns of communication and transportation; patterns of urban development; total population and population density; [and] similarity of social and economic problems.”

g.l. Alabama's Wiregrass region is a community of interest composed of the following nine counties: Barbour, Coffee, Covington, Crenshaw, Dale, Geneva, Henry, Houston, and Pike.

2. The Wiregrass region is characterized by rural geography, agriculture, and a major military base. The Wiregrass region is home to Troy University's flagship campus in Troy and its campus in Dothan.

3. All of the Wiregrass counties are included in District 2, with the exception of Covington County, which is placed in District 1 so that the maximum number of Black Belt counties can be included within just two districts.

Section 2. Section 17-14-70, Code of Alabama 1975, is amended to read as follows:

“§ 17-14-70

(a) The State of Alabama is divided into seven congressional districts as provided in subsection (b).

(b) The numbers and boundaries of the districts are designated and established by the map prepared by the Permanent Legislative Committee on Reapportionrnent and identified and labeled as Pringla Congressional—Plan 1 Livingston Congressional Plan 3-2023, including the corresponding boundary description provided by the census tracts, blocks, and counties, and are incorporated by reference as part of this section.

(c) The Legislature shall post for viewing on its public website the map referenced in subsection (b), including the corresponding boundary description provided by the census tracts, blocks, and counties, and any alternative map, including the corresponding boundary description provided by the census tracts, blocks, and counties, introduced by any member of the Legislature during the legislative session in which this section is added or amended.

(d) Upon enactment of Act 2021–555, adding the act amending this section and adopting the map identified in subsection (b), the Clerk of the House of Representatives or the Secretary of the Senate, as appropriate, shall transmit the map and the corresponding boundary description provided by the census tracts, blocks, and counties identified in subsection (b) for certification and posting on the public website of the Secretary of State.

(e) The boundary descriptions provided by the certified map referenced in subsection (b) shall prevail over the boundary descriptions provided by the census tracts, blocks, and counties generated for the map.”

Section 3. The provisions of this act are severable. If any part of this act is declared invalid or unconstitutional, that declaration shall not affect the part which remains.

Section 4. This act shall be effective for the election of members of the state's U.S. Congressional districts at the General Election of 2024 and thereafter, until the state's U.S. Congressional districts are reapportioned and redistricted after the 2030 decennial census.

Section 5. This act shall become effective immediately upon its passage and approval by the Governor, or upon its otherwise becoming law.

President and Presiding Officer of the Senate

Speaker Of the House of Representatives

SB5

Senate 19-Jul-23

I hereby certify that the within Act originated in and passed the Senate, as amended.

Senate 21-Jul-23

I hereby certify that the within Act originated in and passed the Senate, as amended by Conference Committee Report.

Patrick Harris, Secretary.

House of Representatives Amended and passed: 21-Jul-23

House of Representatives

Passed 21-JuI-23, as amended by Conference Committee Report.

By: Senator Livingston

Tabular or graphical material not displayable at this time.

APPENDIX C – CONGRESSIONAL MAP (2023 PLAN)

Tabular or graphical material not displayable at this time.

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

Tabular or graphical material not displayable at this time.

Tabular or graphical material not displayable at this time.

Tabular or graphical material not displayable at this time.

APPENDIX E – SPECIAL MASTER PLAN

Tabular or graphical material not displayable at this time.

FOOTNOTES

1.   Citations to the trial transcript are identified by page number. Other transcripts are identified by date. The trial transcript may be found at Milligan Docs. 463, 465–66, 468–73, 479–80.

2.   In the first preliminary injunction proceedings, the parties 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” and five other counties are “sometimes included.” Id. ¶ 61. The Gulf Coast includes two counties (Mobile and Baldwin).

3.   When we cite a document from more than one case, we cite the Milligan document.

4.   Page number pincites are to the CM/ECF page number that appears in the top right-hand corner of each page, if available.

5.   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.

6.   The Secretary then dismissed his appeals, including Caster (in the United States Court of Appeals for the Eleventh Circuit. Allen v. Caster, No. 23-12923).

7.   The Singleton Plaintiffs did not request bail-in, so we did not decide their constitutional claims.

8.   Autauga, Baldwin, Barbour, Bibb, Bullock, Butler, Chilton, Choctaw, Clarke, Coffee, Conecuh, Coosa, Covington, Crenshaw, Dale, Dallas, Elmore, Escambia, Geneva, Greene, Hale, Henry, Houston, Jefferson, Lowndes, Macon, Marengo, Mobile, Monroe, Montgomery, Perry, Pickens, Pike, Russell, Shelby, Sumter, Talladega, Tuscaloosa, Washington, and Wilcox.

9.   After the hearing, the Milligan and Caster Plaintiffs, joined by the Singleton Plaintiffs, filed data from the May 19, 2026, primary elections. Milligan Doc. 536; Caster Doc. 442; Singleton Doc. 362. We did not consider these filings.

10.   All counsel agreed before trial that all evidence admitted in any of these cases, including evidence adduced in the preliminary injunction proceedings, was admitted in all three cases unless counsel raised a specific objection. See Milligan Docs. 444, 445. Accordingly, we have re-examined all evidence admitted in Singleton, Milligan, and Caster.

11.   We have elsewhere explained our restraint and refusal for four years to consider whether the Legislature acted unconstitutionally. See Milligan Doc. 490 at 494–96.

12.   One unusual detail about these cases is that the Legislators waived their legislative immunity and have testified. See, e.g., Milligan Docs. 459-13, 459-14, 459-20.

13.   As Dr. Moon Duchin (an expert for the Milligan Plaintiffs) explained, the requirement to keep Mobile and Baldwin Counties together “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[ ]” Black voters in the City of Mobile. Tr. at 298–99, 314. On this point, no document, testimony, or lawyer disputes Dr. Duchin's opinion.

14.   The Community of Interest Plan and Livingston Plans are appended to this Order as Appendix D.

15.   There is also contemporaneous-statement evidence from three Black legislators that they believe that the Legislature never intended to pass a map to remedy race-based vote dilution. See Milligan Doc. 403-4 at 27. In service of the presumption of good faith, we assign these comments no weight.

16.   In the additional Black-opportunity district (District 2), 48.7% of the voting-age population is Black. Milligan Doc. 311 at 41.

17.   To be clear, we do not hold that a legislature runs afoul of the Constitution when it sets goals that prioritize a particular community of interest or protects incumbents. We simply hold that on these unique facts and in the unique posture of these cases, this Legislature did.

18.   At trial, the evidence established that Alabamians had elected only one Black Republican, state legislator Mr. Kenneth Pascal, who we described as a “unicorn.” Milligan Doc. 490 at 396–97. Preliminary results of the May 19 Republican primary indicate that Mr. Maurice McCaney, who is Black, was elected the Republican nominee in House District 1, which is majority-White; Mr. McCaney will not have Democratic opposition in the general election. See May 22, 2026 Tr. at 215.

19.   “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.” Ga. State Conf. of the NAACP v. Fayette Cnty. Bd. of Comm'rs, 775 F.3d 1336, 1342 (11th Cir. 2015); accord Johnson v. De Grandy, 512 U.S. 997, 1010 n.9 (1994). 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.

20.   There was recently one Black statewide official in elected office in Alabama: Bill Lewis, who was appointed by Governor Ivey to the Alabama Supreme Court in 2025 and who became a federal district judge later that same year. Milligan Doc. 436 ¶ 153. Judge Lewis never stood for statewide election.

21.   The Caster Plaintiffs provided the ad, which is entitled “Dale,” on a USB drive for the Court's viewing. Caster Doc. 319-133.

STANLEY MARCUS UNITED STATES CIRCUIT JUDGE