Reply Brief — Wes Allen, Alabama Secretary of State, et al., Appellants v. Bobby Singleton, et al.

No. 25-273

Supreme Court of the United States

──────────────────────────

WES ALLEN, SEC’Y OF STATE, ET AL.,

Appellants,

v.

BOBBY SINGLETON, ET AL.,

Appellees.

──────────────────────────

On Appeal from the United States District Court

for the Northern District of Alabama

──────────────────────────

REPLY BRIEF

Michael P. Taunton

Riley Kate Lancaster

BALCH & BINGHAM LLP

1901 Sixth Ave. N., Ste. 1500

Birmingham, AL 35203

Taylor A.R. Meehan

CONSOVOY MCCARTHY PLLC

1600 Wilson Blvd., Ste. 700

Arlington, VA 22209

Steve Marshall

Alabama Att’y Gen.

A. Barrett Bowdre

Solicitor General

Counsel of Record

Robert M. Overing

Dylan Mauldin

George Muirhead

James W. Davis

Misty Messick

Brenton M. Smith

Benjamin M. Seiss

OFFICE OF THE ATT’Y GEN.

501 Washington Ave.

Montgomery, AL 36130

(334) 242-7300

Barrett.Bowdre@

AlabamaAG.gov

i

TABLE OF CONTENTS

TABLE OF AUTHORITIES ................................. ii

REPLY BRIEF .......................................................1

I. The

District

Court’s

§2

Holding

Contravened Allen by Ordering the State to

Sacrifice a Community of Interest to

Achieve a Racial Goal. ......................................3

II. Requiring Race-Based Districts is Racial

Discrimination That Cannot Survive Strict

Scrutiny. ............................................................8

A. Race-based redistricting lacks an exact

connection between means and ends. .........8

B. Race-based redistricting will not end on

its own or “any time soon.”........................ 10

CONCLUSION ..................................................... 12

ii

TABLE OF AUTHORITIES

Cases

ALBC v. Alabama,

989 F.Supp.2d 1227 (M.D. Ala. 2013) ...................7

Ala. NAACP v. Allen,

No. 2:21-cv-1531, 2025 WL 2451166

(N.D. Ala. Aug. 22, 2025) ............................... 11-12

Alexander v. S.C. NAACP,

602 U.S. 1 (2024) ................................................ 2, 4

Allen v. Milligan,

599 U.S. 1 (2023) .......................................... 1, 3, 10

Banerian v. Benson,

589 F.Supp.3d 735 (W.D. Mich. 2022)............. 9-10

Bartlett v. Strickland,

556 U.S. 1 (2009) ....................................................7

Bethune-Hill v. Va. State Bd. of Elections,

580 U.S. 178 (2017) ................................................4

Black Voters Matter v. Sec’y, Fla. Dep’t of State,

415 So.3d 180 (2025) ..............................................2

Cooper v. Harris,

581 U.S. 285 (2017) ................................................5

Grutter v. Bollinger,

539 U.S. 306 (2003) .............................................. 10

Johnson v. De Grandy,

512 U.S. 997 (1994) ............................................ 2-3

iii

LULAC v. Perry,

548 U.S. 399 (2006) ............................................ 2, 3

Merrill v. Milligan,

142 S.Ct. 879 (2022) ..............................................8

Rucho v. Common Cause,

588 U.S. 684 (2019) ...................................... 2, 8, 10

SFFA v. Harvard,

600 U.S. 181 (2023) .............................2, 6, 8, 10-12

Shaw v. Reno,

509 U.S. 630 (1993) ........................................ 1, 2, 4

Thornburg v. Gingles,

478 U.S. 30 (1986) ..................................... 1, 3, 7, 8

Other Authorities

Br. of Alabama and 15 Other States,

Louisiana v. Callais, No. 24-109

(U.S. Sept. 24, 2025) .................................. 8, 10-12

Motion to Affirm of Milligan Appellees,

Allen v. Milligan,

No. 25-274 (U.S. Oct. 20, 2025) .............................9

Response to Petition for Writ of Caster Plaintiffs,

Allen v. Caster,

No. 25-243 (U.S. Oct. 20, 2025) ............... 2, 5, 8, 11

Transcript of Oral Reargument,

Louisiana v. Callais,

No. 24-109 (U.S. Oct. 15, 2025) ......................... 2, 8

1

REPLY BRIEF

After Allen v. Milligan, 599 U.S. 1 (2023), Alabama

went back to the drawing board rather than head to

trial on its 2021 Plan. Unlike the 2021 Plan, the

State’s 2023 Plan preserved both the Black Belt and

Gulf Coast regions in as few districts as possible. Not

enough, said the district court. Ordering the State to

split the Gulf Coast for the first time in fifty years, the

court required what §2 “never” requires: the “adoption

of districts that violate traditional redistricting principles.” Id. at 30. Alabama’s court-drawn congressional districts are “unexplainable on grounds other

than race.” Shaw v. Reno, 509 U.S. 630, 644 (1993).

The court below went far beyond Allen, such that

principles that make it “impossible to draw [another]

reasonably configured majority-Black district[]” must

give way, Mot.15, not just for purposes of Gingles-1

but for the actual congressional district lines that will

govern 5 million Alabamians. It is not enough to treat

communities of interest equally—as earlier iterations

of this case had suggested. App.937; Allen, 599 U.S. at

21. What matters now is the priority of drawing an

additional majority-minority district, or something

quite close. See Mot.2, 15; accord, e.g., App.329. Whatever the State’s actual race-neutral priorities, those

must come second. Under that version of §2, race is

always the criterion that cannot be compromised.

This Court must reverse, even assuming that race

can be used in districting in a way that would not otherwise be allowed. J.S.10-12. But the Court should

question that assumption too. For decades, States

have had “good reason … to recognize that explicit

race-based districting embarks us on a most dangerous course.” Johnson v. De Grandy, 512 U.S. 997, 1031

2

(1994) (Kennedy, J., concurring). Officials recognize

the dangers, but they have no way to navigate them.

Try to satisfy courts with a race-based map; get

enjoined. Try to satisfy courts with a race-neutral

map; get enjoined. Alabama and Louisiana took two

different paths that ended up in the same place. Callais v. Louisiana, No. 24-109 (U.S.); see also Black Voters Matter v. Sec’y, Fla. Dep’t of State, 415 So.3d 180,

184 (Fla. 2025) (“two-hundred-mile-long congressional district encompassing several communities of

black voters across North Florida” would be unconstitutional).

If erring on the side of declining to discriminate is

now deemed racial discrimination, J.S.27-35, a State

can never decide that enough is enough. That holding

is proof positive that §2 is not “ridding our electoral

process of race,” Callais.Rearg.Tr.41, but prolonging

“fixat[ion]” on it, LULAC v. Perry, 548 U.S. 399, 434

(2006); see also De Grandy, 512 U.S. at 1030 (Kennedy, J., concurring); Shaw, 509 U.S. at 657. As long

as the judiciary is balancing a “fair share of political

power” among racial groups, cf. Rucho v. Common

Cause, 588 U.S. 684, 709 (2019), courts will be treated

as “weapons of political warfare,” Alexander v. S.C.

NAACP, 602 U.S. 1, 11 (2024). “That is a feature, not

a bug, of § 2.” Caster.BIO.32. It cannot go on forever.

SFFA v. Harvard, 600 U.S. 181, 221-26 (2023).

3

I.

The

District

Court’s

§2

Holding

Contravened Allen by Ordering the State to

Sacrifice a Community of Interest to

Achieve a Racial Goal.

A. After Alabama united the Black Belt into two

districts—the smallest number possible—the district

court moved the goalposts. Now, §2 requires the map

that places the most “Black Belt counties in a majority-Black district.” App.345. According to the court,

because some of the illustrative plans placed “all but

one of the Black Belt counties in a majority-Black district,” they respected the Black Belt “better – much

better – than the 2023 Plan.” Id.; contra Mot.2 (denying that “the district court fault[ed] the State for failing ‘to place more Black voters … in a majority-Black

district’”).

But this Court already rejected that move in Allen.

The “minority population in and of itself” cannot be

the community of interest; the Black Belt is “a ‘historical feature’ of the State, not a demographic one,” and

must be “treated … as a community of interest for

[that] reason.” Allen, 599 U.S. at 32 n.5 (plurality).

Thus, the notion that the 2023 Plan limited black voting strength could be the conclusion of a successful §2

challenge, but not the beginning proof of it. That

would be circular.

Even if Gingles-1 maps can join together “farflung

segments of a racial group,” contra LULAC, 548 U.S.

at 433, race cannot predominate in the actual districts

that govern Alabama without satisfying strict scrutiny. Now at the end of this case, a district court has

concluded that Alabama’s only means of complying

with §2 was a race-first district combining “Black Mobile” with black voters 250 miles away. App.708. That

4

remedy uses race “explicit[ly]” by demanding the

State place more Black Belt counties in a race-based

district. Shaw, 509 U.S. at 642. As the district court

observed, “all paths” to another majority-black district

“require[] splitting” white voters from black voters in

the Gulf Coast. App.7, 531. Admitting that no §2-compliant map “achieve[s] all the political goals” of the

2023 Plan, App.514; see App.492, the court thus

“subordinated” neutral criteria to race. Alexander, 602

U.S. at 7.

B. The rejoinder that the court-drawn districts

were “race-blind,” Mot.10, or “drawn without regard

to race,” Mot.8, is preposterous. The court ordered a

“district[] in which Black voters either comprise a voting-age majority or something quite close to it.”

App.13. It insisted that CD2 have enough black voters

that a Democrat would likely win. E.g., Milligan,

DE311:3, 41. These are “racial target[s],” contra

Mot.2, even without a bright line “threshold[] for racial demographics” in the remedial districts, Mot.13.

Cf. Ricci v. DeStefano, 557 U.S. 557, 594 (2009)

(Scalia, J., concurring) (“intentionally design[ing] …

hiring practices to achieve the same end” as a quota is

equally discriminatory).

We know that “race-neutral considerations ‘came

into play only after the race-based decision had been

made,’” Bethune-Hill v. Va. State Bd. of Elections, 580

U.S. 178, 189 (2017), because the court treated Alabama’s “redistricting principle[s]” as “[]negotiable,”

App.329. Traditional criteria were to be “consider[ed]”

by the special master, Milligan, DE273:9, but followed

only “to the extent reasonably practicable,” id. at 8.

Traditional principles could be compromised; racial

targets could not. See App.329 (no “non-negotiable”

principles); App.514 (no “deference” to principle that

5

“entrenches vote dilution”); App.719 (communities of

interest not a “trump card”).

Drawing maps without displaying racial data, as

the special master said he did, does not erase racepredominance. Cf. Cooper v. Harris, 581 U.S. 285,

313-17 (2017).1 The court required that he remedy

“ineffective” “Black voting strength.” App.715; see Milligan, DE273:7. He “confirmed” he hit his target before presenting his plans. Milligan, DE295:36. His

plans sacrificed compactness, paired incumbents, and

split a major city, a county, and a community of interest. See Milligan, DE295:14, 17, 23, 25, 38; id. at 42

(acknowledging “need to split the Gulf Coast” to increase “Black voting strength”).

At bottom, Plaintiffs play word games to deny the

obvious. Alabama, they say, “need not meet any racial

target,” Mot.2; “form a majority-Black district,” Mot.9;

“place more Black voters [from the Black Belt] in a

majority-Black district,” Mot.2; or “draw a race-based

plan,” Mot.9-10. It just needed to draw a district that

gave black-preferred candidates sufficient “opportunity to elect candidates of their choice.” Mot.2.

There is no difference.

Under the district court’s interpretation of §2, the

bottom line is that if the wrong race wins too many

1 The special master had a roadmap from the district court to

hit the racial target without explicitly relying on racial data:

“split the Gulf Coast,” App.715, “split Mobile County,” App.947,

and connect Mobile’s urban core with Montgomery and Black

Belt counties on the Georgia border, id.; see Milligan, DE295:13

(special master had “the eleven illustrative plans”). Small wonder he “grouped together the same” farflung populations as

Plaintiffs had. Caster.BIO.6. Alabama stipulated, contra Mot.12,

19, only that his report states that he did not “display racial demographic data … while drawing.” Milligan, DE436:22.

6

elections, the State must draw new lines to enable another race to win. By definition, the remedy has everything to do with race. When the government “picks

winners and losers based on the color of their skin,”

SFFA, 600 U.S. at 229, it is no less odious whether it

draws district lines randomly until it likes the balance

of racial power, draws them with awareness of racial

data, or draws them to meet racial quotas. It is all

racial discrimination. And when neutral goals are

sacrificed to better allocate political power among

races, strict scrutiny should apply.

C. Only the Singleton Plaintiffs insist that compliance with §2 does not require splitting the Gulf Coast,

Mot.20-21, citing the Singleton Plan as evidence. The

district court did not instruct the special master to

draw anything like it. And Plaintiffs’ insistence that

Alabama depart from its non-racial priorities simply

confirms the need for this Court’s review.

First, though the Singleton Plaintiffs support this

plan, the Caster and Milligan Plaintiffs objected to it

out of fear that its second opportunity district was

“unlikely to … perform[.]” Milligan, DE295:33. They

argued that given previous elections, it failed to create

“reliable opportunities to elect.” In re Redistricting

2023, No. 2:23-mc-1181 (N.D. Ala.), DE23:6-7. “The

uncertainty” was “especially” troubling to them because one of the opportunity districts in the Singleton

Plan would have paired Alabama’s only black representative at the time—Terri Sewell—“against a white

incumbent.” Id. at 6.2 Unsurprisingly then, the special

2 Had Sewell faced that Representative in 2022, the other

Plaintiffs argued, she “likely would have lost.” Id. at 7. Sewell

also objected to this plan because she thought it “fail[ed] to

7

master split the Gulf Coast in all three of his remedial

maps. See Milligan, DE295:17, 23, 25; see also App.7;

530-31.

Second, the Singleton Plan fails Gingles-1 because

it does not create two majority-black districts. As

Plaintiffs admit, drawing two majority-black districts

requires splitting the Gulf Coast. Mot.15. The two

“opportunity districts” in the Singleton Plan are CD6

(39.61% BVAP) and CD7 (49.38%). In re Redistricting

2023, DE5:7. CD6 is a crossover district, but because

§2 “does not require crossover districts,” Bartlett v.

Strickland, 556 U.S. 1, 23 (2009) (plurality), Alabama

could not defend this district as “narrowly tailored to

achieve compliance with section 2,” ALBC v. Alabama,

231 F.Supp.3d 1026, 1320 (M.D. Ala. 2017).

Third, race predominates in the Singleton Plan

too. By modifying all seven congressional districts, it

“made significant changes across the state,” Milligan,

DE295:29, 44, paired incumbents (Mot.23), and added

strange shapes—all to craft a new race-based district.

The map “satisfies every relevant redistricting principle” (Mot.22) only if “relevant” is defined to exclude

principles standing in the way of another opportunity

district. That is why Plaintiffs must concede on the

very next page that their plan does not, in fact, satisfy

all the State’s redistricting criteria. Mot.23. If the

State had adopted this map, sacrificing non-racial criteria in pursuit of a racial goal, its decision would have

been “driven by race.” Mot.24.

provide Black voters with a reasonable opportunity to elect.” In

re Redistricting 2023, DE21:5.

8

II.

Requiring Race-Based Districts is Racial

Discrimination That Cannot Survive Strict

Scrutiny.

The district court addressed the State’s constitutional argument in just three pages, did not cite

SFFA, and refused to apply strict scrutiny. App.45456. Bereft of any serious analysis of the Equal Protection Clause, its judgment must be reversed. See

J.S.16-25; Br. of Alabama and 15 States, Louisiana v.

Callais, No. 24-109 (U.S. Sept. 24, 2025).

A.

Race-based redistricting lacks an exact

connection between means and ends.

No one really thinks that “few legal tests are as

clear as the Gingles inquiry.” Caster.BIO.29. The law

of vote dilution “is notoriously unclear.” Merrill v. Milligan, 142 S. Ct. 879, 881 (2022) (Kavanaugh, J., concurring). Even after forty years, there remains

“considerable disagreement and uncertainty” about

the “nature” of vote dilution. Id. at 883 (Roberts, C.J.,

dissenting). The “essence of a §2 claim” is easy to recite (Caster.BIO.28) but hard to explain. See, e.g., Callais.Rearg.Tr.14 (when “race is playing a role to

contaminate” democracy), 17 (when “race is playing

an outsized role”), 37 (when maps “use race in excessive fashion”). If no one can articulate the very “concept of injury,” then this area of law is not safe enough

for courts to wield the “dangerous” tool of racial sorting. SFFA, 600 U.S. at 209, 212; see Ala.Callais.Br.69 & nn.1-4.

Experience has proven that allotting a “fair share

of political power” among racial groups is not possible.

Cf. Rucho, 588 U.S. at 709. In the same breath that

Plaintiffs say traditional redistricting principles are

“important” and must be “take[n] … seriously,” they

9

add that criteria “incompatible” with a “majorityblack district[]” must give way. Mot.15, 16. That

standard does not discipline the use of race.

This case exemplifies why. The Legislature’s judgment is that congressional districting plans respect

the Gulf Coast by maintaining it in a single district.

Rather than accept that judgment, and despite finding the Gulf is its own community of interest, the district court considered for itself whether the “harm[]”

from the split was “intolerabl[e].” App.350. The State

offered “expert and lay witnesses about overlapping

economic interests, commuting patterns, shared heritage, cultural events, and unique challenges that connect the Gulf Coast counties.” App.346. But that

evidence stood no chance against speculation that

some other communities of interest might prefer a

split. App.350. Nor did it stop the district court from

concluding that because the State placed the Gulf

Coast in two districts in a completely different context, splitting it in a congressional map “respected”

the community well enough. App.351-52.3

Determining whether a map respects a State’s

neutral principles, especially communities of interest,

is a political judgment for legislatures. Because many

of the districting principles are “malleable” and “ethereal,” Milligan.Mot.27, courts can easily discount or

discard race-neutral principles because there is “no

3 Plaintiffs and the district court latched onto the Alabama

State Board of Education Plan, which has one more district than

the congressional map and originally split Mobile County for preclearance reasons and to avoid retrogression, as the State’s expert testified unrebutted. Tr.2024-29. It has nothing to do with

the Legislature’s valid policy prerogatives for maintaining one

Representative of the Gulf in Congress.

10

legal answer” to the “question of ‘how much deviation

from each [criterion] to allow.’” Banerian v. Benson,

589 F.Supp.3d 735, 738 (W.D. Mich. 2022) (Kethledge,

J.) (quoting Rucho, 588 U.S. at 708); see Ala.Callais.Br.15-17; Allen, 599 U.S. at 35.4 And when districting principles conflict with each other, courts

have no neutral way to strike a balance. Ala.Callais.Br.11-14. As a result, States cannot predict how

courts may balance their priorities, which no longer

seem to constrain the use of race but invite its elevation above traditional redistricting principles. Without a “reliable way to determine who wins, or even

where the finish line is,” Allen, 599 U.S. at 37, §2’s use

of race is not narrowly tailored but “inescapably imponderable,” SFFA, 600 U.S. at 215. This “is no business of the courts.” Banerian, 589 F.Supp.3d at 738.

B.

Race-based redistricting will not end

on its own or “any time soon.”

The district court held that race-based redistricting could not be “render[ed] unconstitutional” by “the

mere passage of time,” App.454, but that’s not the argument, see J.S.23-25. Rather, time is “the acid test of

[the] justification” for using race. Grutter v. Bollinger,

539 U.S. 306, 343 (2003). If race-based districting had

any “efficacy,” Plaintiffs would be able to say when it

will “no longer be necessary.” Id.

Under the current regime, States can never stop

using race. Each census restarts redistricting anew.

4 At the preliminary-injunction stage, a Milligan expert

wrongly interpreted Alabama’s non-dilution principle to mean

prioritizing race above bedrock criteria like contiguity. App.833.

Plaintiffs criticize the State for removing the guideline, Mot.5-7,

but never address the chance that the Legislature removed it to

avoid such manipulation.

11

But that currently only guarantees that lawmakers

will conduct a “periodic review” of how they use (or do

not use) race every decade in perpetuity, little different than Harvard’s use of race every admissions cycle.

SFFA, 600 U.S. at 225.

The theory that §2 will sunset on its own is

divorced from reality. See Ala.Callais.Br.19-25. According to the district courts, if a 250-mile-wide majority-black district can be drawn in Alabama (or

Louisiana), it must be drawn. As for the totality-ofcircumstances test, States can win repeatedly one cycle and lose in the next, even with overlapping evidence. J.S.16-17. Anything that has ever happened in

the State is fair game; which facts courts will deem

relevant is anyone’s guess. Ben Carson’s finish in the

2016 primary counted against Alabama in 2022.

App.844. After Alabama proved that Carson’s tally in

Alabama was one of his best in the country, App.281,

Plaintiffs dug up a CNN exit poll from two cycles

prior, and the State was faulted for President

Obama’s unpopularity in 2007, App.389. Likewise,

when Alabama proved that it had the second-smallest

racial gap in incarceration in the Nation, Tr.2203:2425, the district court turned to gaps in infant mortality, App.405. The test is ever evolving, and the sun

seems only to be rising. Contra Caster.BIO.34-35 &

n.5; see Ala.Callais.Br.7-8 & n.5.

Thus, States can make “substantial progress”

across every vector, App.455, without ever knowing

whether they have satisfied the test. One more racebased district will not end myriad “disparities … in

the modern world.” App.404-09. It will not stop candidates from making politics about race, especially if it

helps their party in court, Ala. NAACP v. Allen, No.

2:21-CV-1531, 2025 WL 2451166, at *81 (N.D. Ala.

12

Aug. 22, 2025); see Ala.Callais.Br.29. Whatever §2 diagnoses, it will not cure anything “any time soon.”

SFFA, 600 U.S. at 225.

CONCLUSION

These cases present every issue raised in the jurisdictional statement. The Court should note probable

jurisdiction and reverse.

Respectfully submitted,

Michael P. Taunton

Riley Kate Lancaster

BALCH & BINGHAM LLP

1901 Sixth Ave. N., Ste. 1500

Birmingham, AL 35203

Taylor A.R. Meehan

CONSOVOY MCCARTHY PLLC

1600 Wilson Blvd., Ste 700

Arlington, VA 22209

November 4, 2025

Steve Marshall

Attorney General

A. Barrett Bowdre

Solicitor General

Counsel of Record

Robert M. Overing

Dylan Mauldin

George Muirhead

James W. Davis

Misty S. F. Messick

Brenton M. Smith

Benjamin M. Seiss

STATE OF ALABAMA

OFFICE OF THE ATT’Y GEN.

501 Washington Ave.

Montgomery, AL 36130

(334) 242-7300

Barrett.Bowdre@

AlabamaAG.gov