The Voting Rights Act of 19651 (VRA), “the [legislative] crown jewel of the civil rights movement,”2 was enacted to serve as a bulwark against racial voter suppression. Section 2 of the VRA, which implemented a nationwide ban on racial discrimination in voting,3 provides remedies — including, but not limited to, redistricting4 — that ensure that minority voters can effectively aggregate their political power and elect representatives of their choice. But in Petteway v. Galveston County,5 the Fifth Circuit, sitting en banc, overturned decades of circuit precedent by holding that Section 2 does not allow plaintiffs to bring coalition-based districting claims.6 Beyond resting on an erroneously narrow reading of the VRA, the Fifth Circuit’s holding in Petteway weakened the federal judiciary’s ability to ensure minority voting power in an increasingly heterogeneous society.7
To bring a Section 2 claim, plaintiffs must meet the three-pronged threshold inquiry established in Thornburg v. Gingles.8 The first Gingles prong requires that “minority group[s]” be “sufficiently large and geographically compact to constitute a majority in a single-member district.”9 In a coalition district, two or more minority groups combine to constitute a majority of the population in a single-member district in order to collectively meet the first Gingles prong.10 Coalition districts are distinct because of their multiracial composition; other majority-minority districts11 aggregate members of a singular racial minority group. For decades, courts have adjudicated Section 2 vote dilution claims by adopting coalition districts.12 The Supreme Court has never directly addressed the permissibility of coalition claims,13 but federal circuits unanimously accepted the practice until 1996.14 In Nixon v. Kent County,15 the Sixth Circuit was the first to bar coalition districts, creating a circuit split.16 In August 2024, the Fifth Circuit became the second.
In 2021, for the first time in three decades, the Galveston County Commissioners Court adopted a redistricting plan without any majority-minority districts.17 Precinct 3 was the county’s sole majority-minority precinct under prior maps.18 Under the new map, there were none.19 A group of Black and Hispanic voters sued.20 They alleged that the new map diluted their votes in violation of Section 2 of the VRA because minorities were deprived of voting power when Galveston County broke up their coalition district.21 The county responded by disputing the plaintiffs’ factual allegations, arguing that Section 2 does not permit coalition claims, and positing that Section 2 itself has outlived its constitutionality.22
Following trial,23 the district court ruled for the plaintiffs.24 Judge Brown held that “well-established Fifth Circuit precedent”25 sanctioned coalition claims and noted the absence of any Fifth Circuit precedent holding otherwise.26 He also rejected the county’s “speculat[ion] that the [VRA] has outlived its usefulness.”27 Judge Brown then applied the Gingles framework for a vote dilution inquiry, finding that the preconditions had been met and that the totality of the circumstances evinced a Section 2 violation.28 He ordered the county to submit a revised map that complied with Section 2 and contained “at least one” majority-minority district.29
The Fifth Circuit affirmed but requested a poll to determine whether the case should be reheard en banc.30 In a per curiam opinion, Judges Jones, Barksdale, and Elrod held that the district court had “appropriately applied precedent” from Campos v. City of Baytown,31 which permitted coalition-based Section 2 vote dilution claims.32 Furthermore, the panel held that the county had not met its burden of proving that the district court had committed a clear error in its Gingles analysis, and had “failed to show that Section 2 is unconstitutional under existing precedent.”33 But although the panel agreed with the district court’s application of binding precedent, it also opined that the Fifth Circuit precedent established in Campos was “wrong as a matter of law.”34 The panel called for the case to be reheard to consider overruling Campos.35 The Fifth Circuit vacated the panel opinion and reheard the case en banc.36
In a split decision, the en banc Fifth Circuit accepted the panel’s invitation by overruling Campos, reversing the district court’s judgment, and remanding the case for further proceedings.37 The court’s opinion, written by Judge Jones,38 held that coalition claims are incompatible with the VRA’s prohibition on vote dilution.39 The court read the text of Section 2 — which refers to “a class” of voters — to allow only dilution claims brought by members of individual racial groups, not multiracial coalitions.40 The court reasoned that individuals of different racial groups would be considered members of “two distinct classes” rather than one shared “class,” compelling separate inquiries into whether each group had experienced a distinct dilution injury.41 The Fifth Circuit rejected plaintiffs’ reliance on the Supreme Court’s instruction in Chisom v. Roemer42 that courts should interpret the VRA “in a manner that provides ‘the broadest possible scope’ in combating racial discrimination”43 to accord with the “broad remedial purpose” of the Act.44 The Fifth Circuit pointed to more recent Supreme Court decisions outside of the VRA context instructing that “vague notions of a statute’s ‘basic purpose’ are . . . inadequate to overcome the words of its text regarding the specific issue under consideration.”45 The court posited that “Congress’s failure to expressly authorize coalition claims” compelled its holding.46
The majority also rejected the plaintiffs’ argument that, when Congress enacted the pivotal 1982 VRA amendments,47 “Congress was aware of cases involving coalition claims” and thus tacitly consented to them through its intentionally broad statutory language.48 Because “[n]one of the pre-1982 cases . . . decided, as a matter of law, whether coalition claims are permissible,” the majority found it improper to conclude that Congress’s awareness of these cases implied such legislative authorization.49 The Fifth Circuit also extended Supreme Court precedent, arguing that the logic of Bartlett v. Strickland50 — which had held that Section 2 did not compel the drawing of “crossover” districts wherein minority voters can influence electoral outcomes with sufficient support from white voters51 — “applies with equal force to coalition claims.”52 Finally, the majority concluded that stare decisis did not require adherence to the Fifth Circuit’s prior holdings in League of United Latin American Citizens v. Clements53 and Campos.54
Judge Ho concurred in part and in the judgment.55 He reasoned that the Supreme Court’s rejection of “crossover” districts necessarily foreclosed coalition claims.56 Moreover, he noted that Justices Scalia, Thomas, and Gorsuch “have suggested that courts should not decide vote dilution claims under Section 2 of the Voting Rights Act at all.”57
Judge Haynes dissented.58 She argued that Campos was correctly decided.59 Since she could not “identify any reversible error of fact or law in the district court’s reasoning,” she would have affirmed the district court’s ruling.60
Judge Douglas dissented,61 sharply criticizing the majority’s “atextual and ahistorical” decision as having “dismantled the effectiveness of the Voting Rights Act in this circuit.”62 First, she recounted the origins of the VRA and described the history of racial voting discrimination nationwide and in Galveston County.63 Second, Judge Douglas argued that the majority did not correctly contend with stare decisis, and thus was not on solid footing to overturn its en banc precedent.64 Judge Douglas also critiqued the majority’s interpretation of the VRA and argued that the majority had strayed from the Chisom Court’s suggestion that the Act be read broadly.65 Her dissent rebuked the majority’s reading of the Act’s legislative history,66 and cautioned that the majority’s holding would have negative consequences for a diverse and multiracial society.67
The Fifth Circuit’s erroneously narrow reading of Section 2 contravenes the statute’s broad language and expansive remedial intent. This error will diminish the protection that the Act was intended to provide in the context of our heterogeneous and diversifying electorate.
Since the 1960s, Congress consistently drafted the VRA and its subsequent amendments in a manner that affirms the statute’s expansive scope. The text of Section 2 includes intentionally broad-sweeping language, proscribing any “voting qualification or prerequisite to voting or standard, practice, or procedure” that burdens “the right . . . to vote on account of race.”68 This language came from lawmakers’ efforts, alongside those of the Department of Justice, to design the VRA to effectuate the promises of the Fifteenth Amendment69 by addressing the ever-evolving array of state voter suppression tactics.70 Congress has repeatedly rebuffed judicial attempts to limit the VRA’s scope: In response to the Supreme Court’s narrowing construction in City of Mobile v. Bolden,71 Congress amended Section 272 to restore and enlarge its coverage. “The 1982 amendment further expanded the protection afforded by § 2”73 and affirmed Congress’s intent to give the VRA a broad remedial scope.74 The Supreme Court accordingly embraced Congress’s directive to interpret the VRA broadly and instructed other courts to do the same.75
The Fifth Circuit’s rejection of this directive is both unpersuasive and troubling. Its best response to Chisom’s instruction is that “vague notions of a statute’s ‘basic purpose’ are . . . inadequate to overcome the words of its text regarding the specific issue under consideration.”76 There are two problems with this response. First, the VRA’s “basic purpose” of broadly remediating discrimination in voting is far from “vague.” Here, the Fifth Circuit conflates vagueness with breadth. The statute’s purpose of precluding all racially discriminatory voter suppression practices has been clearly expressed since 1965.77 A law is not vague simply because its purpose is capacious or maximalist.78 Second, the Fifth Circuit’s reliance on the “words of [the VRA’s] text regarding the specific issue” is misplaced.79 The Supreme Court has considered and rejected the structure of the majority’s argument — that is, if coalition claims were permissible, Congress would have expressly written them into the law — as an illegitimate interpretative strategy for analyzing the VRA.80 By demanding express authorization of coalition districts, the Fifth Circuit disregarded the capacious legislative purpose undergirding the VRA81 and ignored the “super-strong presumption” of statutory stare decisis,82 which would have compelled an alternative result.
The Fifth Circuit’s disregard for the VRA’s primary purpose limits minority access to the political process. In the 1980s and 1990s, Black and Hispanic residents of Galveston County organized to create Precinct 3.83 Though neither group alone was sufficiently numerous and compact to comprise a majority within one precinct, these communities recognized that their interests would be better represented by forming a coalition district.84 They transformed what was once “an epicenter of the nation’s slave trade”85 into “an important political homebase for Black and Latino residents.”86 Without a coalition district, they have been dispersed into other precincts, and lacking a district containing sufficient numerosity and compactness to satisfy the Gingles preconditions, may become “locked out of the political process.”87 The new Galveston County map dilutes Black and Hispanic votes, but the Fifth Circuit denies them a remedy. They suffer in solidarity but must seek relief separately.
The Fifth Circuit’s rejection of coalition claims risks undermining cross-racial solidarity and rewarding residential segregation. The United States is becoming more diverse and less residentially segregated for certain minority groups — two positive developments.88 But the Fifth Circuit’s ruling risks making residential heterogeneity an impediment to Section 2 relief.89 As a community diversifies, it becomes less likely that households of any single race will be sufficiently concentrated to meet the Gingles compactness threshold that plaintiffs must satisfy to bring a Section 2 claim,90 despite high levels of racially polarized voting in many jurisdictions.91 Minorities seeking representation will be disincentivized from building ties with residents of different racial communities.92 Such a development could perversely incentivize a zero-sum approach to addressing racial voting discrimination and discourage cross-racial coalition-building.93
The Fifth Circuit’s approach removes a tool from the toolboxes of litigators seeking to protect racial minorities from vote discrimination — but alternative strategies remain on the table if coalition districts are no longer a viable litigation tool. Alternative systems of electoral design can build minority political power by changing the rules under which elections are conducted, rather than the districts from which representatives are elected.94 Systems that do not rely on geography, such as proportional representation,95 may be better suited than districting to our contemporary sociopolitical landscape, wherein voters of different races increasingly live side-by-side.96 These alternative systems could be enacted as litigation remedies97 or through legislative or administrative change at the state and local levels, as exemplified by state voting rights acts that explicitly contemplate alternative methods of election.98 Thus, the Fifth Circuit’s holding — though damaging — need not be fatal to the goal of ensuring the effective aggregation of cross-racial groups’ political power. By seeking to enact structural changes, advocates can still strive to ensure that the political process remains open to all, regardless of race — the foundational purpose of the VRA of which courts today, including the Fifth Circuit, have lost sight.