In the Supreme Court of the United States
GARY BARTLETT, EXECUTIVE DIRECTOR OF THE NORTH CAROLINA STATE BOARD OF ELECTIONS, ET AL., PETITIONERS
DWIGHT STRICKLAND, ET AL.
ON WRIT OF CERTIORARI
TO THE SUPREME COURT OF NORTH CAROLINA
BRIEF FOR THE UNITED STATES
AS AMICUS CURIAE SUPPORTING AFFIRMANCE
GREGORY G. GARRE
Acting Solicitor General
Counsel of Record
GRACE CHUNG BECKER
Acting Assistant Attorney
KANNON K. SHANMUGAM
Assistant to the Solicitor
DIANA K. FLYNN
ANGELA M. MILLER
Department of Justice
Washington, D.C. 20530-0001
Whether Section 2 of the Voting Rights Act of 1965, 42 U.S.C. 1973, obligated North Carolina to draw House District 18, in which African Americans constituted 39.36% of the voting-age population.
In the Supreme Court of the United States
GARY BARTLETT, EXECUTIVE DIRECTOR OF THE
NORTH CAROLINA STATE BOARD OF ELECTIONS,
ET AL., PETITIONERS
DWIGHT STRICKLAND, ET AL.
ON WRIT OF CERTIORARI
TO THE SUPREME COURT OF NORTH CAROLINA
BRIEF FOR THE UNITED STATES
AS AMICUS CURIAE SUPPORTING AFFIRMANCE
INTEREST OF THE UNITED STATES
Petitioners claim that Section 2 of the Voting Rights Act of 1965, 42 U.S.C. 1973, required North Carolina to draw the legislative district at issue in this case, in which African Americans constituted 39.36% of the voting-age population. Because the United States has primary responsibility for enforcing Section 2, see 42 U.S.C. 1973j(d), the Court's decision in this case could impact federal enforcement efforts. The United States has participated in all of the Court's cases involving the interpretation of amended Section 2.
1. Section 2(a) of the Voting Rights Act of 1965 (Vot ing Rights Act), as amended by Congress in 1982, prohibits any "standard, practice, or procedure * * * which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color" or membership in a language minority group. 42 U.S.C. 1973(a). Section 2(b), which was added in 1982, provides that Section 2(a) is violated when, "based on the totality of circumstances, it is shown that * * * members of a class of citizens protected by sub section (a) of this section * * * have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice." 42 U.S.C. 1973(b). Section 2(b) also provides that "nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population." Ibid.
Vote dilution occurs when a voting practice "oper ate[s] to minimize or cancel out the voting strength of * * * elements of the voting population." Fortson v. Dorsey, 379 U.S. 433, 439 (1965). In Thornburg v. Gingles, 478 U.S. 30 (1986), this Court held that, as "necessary preconditions" to proving a Section 2 vote- dilution claim, a plaintiff must show (1) that a minority group is "sufficiently large and geographically compact to constitute a majority in a single-member district"; (2) that the minority group is "politically cohesive"; and (3) that "the white majority votes sufficiently as a bloc to enable it * * * usually to defeat the minority's pre ferred candidate." Id. at 50-51. If a plaintiff satisfies the Gingles preconditions, a court must then consider the totality of circumstances, "based upon a searching practical evaluation of the past and present reality," to determine whether Section 2 has been violated. Id. at 79 (internal quotation marks and citation omitted).
2. Following the 2000 census, the North Carolina General Assembly adopted a series of redistricting plans for its two houses, the Senate and the House of Repre sentatives. In 2003, after its initial plans were invali dated under state law, the General Assembly adopted the plan at issue in this case. The North Carolina Con stitution provides that "[n]o county shall be divided" in the formation of legislative districts. See N.C. Const. Art. II, §§ 3(3), 5(3). The North Carolina Supreme Court has held, however, that the General Assembly may divide a county into multiple districts where it is necessary to do so in order to comply with federal law, including the Voting Rights Act. See Stephenson v. Bartlett, 562 S.E.2d 377, 396-397 (2002).
As part of the 2003 redistricting plan, the General Assembly divided Pender County, which is located in southeastern North Carolina, into two House districts. The district at issue in this case, House District 18, in cludes portions of Pender County and of a larger adjoin ing county to the south, New Hanover County. The evi dence presented below indicates that African Americans constitute 39.36% of the voting-age population of House District 18. Pet. App. 69a. The evidence also indicates that it would be possible to draw a House district that keeps Pender County together with an African Ameri can voting-age population of 35.33%. J.A. 73.1
3. Petitioners are the Governor of North Carolina and other state officials; respondents are three members of the Pender County Board of Commissioners. On May 14, 2004, respondents and others brought suit in Wake County Superior Court, contending that, by splitting Pender County into two House districts, the 2003 redis tricting plan violated the whole-county provisions of the North Carolina Constitution. J.A. 5-14. Petitioners moved for summary judgment, asserting that it was nec essary for the General Assembly to draw House District 18 as it did in order to avoid violating Section 2 of the Voting Rights Act. Respondents, in turn, moved for summary judgment, contending, inter alia, that House District 18 failed to satisfy the first Gingles precondition because it did not contain a majority African American voting-age population. Pet. App. 72a-74a.
The trial court granted partial summary judgment to petitioners. Pet. App. 51a-105a. The court concluded that, while "House District 18 is not a majority-minority district because of the number of African American vot ers located therein," id. at 90a, it was "a de facto black majority district * * * sufficient to meet the first prong of Gingles." Id. at 96a-97a. The court also con cluded that the second Gingles precondition-that the minority group be politically cohesive-was satisfied. Id. at 99a. Respondents subsequently stipulated that the third Gingles precondition-that the majority group vote sufficiently as a bloc to defeat the minority group's preferred candidate-was met. Id. at 130a-131a. The trial court concluded that the ultimate totality-of-the- circumstances test was also satisfied, and accordingly entered judgment for petitioners. Id. at 106a-120a.
4. The North Carolina Supreme Court reversed. Pet. App. 1a-50a. The court held that "a minority group must constitute a numerical majority of the voting-age population in the area under consideration before Sec tion 2 of the [Voting Rights Act] requires the creation of a legislative district to prevent dilution of the votes of that minority group." Id. at 19a. Because "the African- American minority group in House District 18 does not constitute a numerical majority of citizens of voting age," the court determined that it was unnecessary for the General Assembly to draw House District 18 in the manner it did in order to comply with Section 2. Id. at 27a. The court therefore ordered the General Assembly to redraw the district. Id. at 34a.
SUMMARY OF ARGUMENT
Section 2 of the Voting Rights Act did not obligate North Carolina to draw House District 18, in which Afri can Americans constitute approximately 39% of the voting-age population.
A. To establish a Section 2 claim, a plaintiff must generally show that members of a minority group can constitute a majority of the relevant population in a pro posed district. That requirement follows from this Court's decision in Thornburg v. Gingles, 478 U.S. 30 (1986), as well as the text of Section 2. Where a politi cally cohesive minority group constitutes a numerical majority in a proposed district, the group will have the "opportunity to elect" a candidate of its own choice. And where a jurisdiction fails to adopt such a proposed dis trict, it can fairly be presumed that the jurisdiction is not providing an equal opportunity to the minority group to elect the representative of its choice-at least where, in the absence of such a district, majority-bloc voting would defeat the minority group's preferred can didate. That general rule provides legislatures drawing districts, and courts reviewing them, with a clear and administrable standard for applying Section 2.
B. While it correctly recognized the majority-minor ity requirement, the North Carolina Supreme Court went further than necessary in this case by stating that the requirement is absolute. The United States has pre viously explained that the general majority-minority requirement may be relaxed in two situations. The first is where a plaintiff shows that a challenged voting prac tice was adopted with discriminatory intent. Such a showing tends to suggest that a jurisdiction is not pro viding an equal opportunity to minority voters to elect the representative of their choice. There is no allegation in this case of such an intent. The second is where the minority group is substantial in size yet just short of a majority. That adjustment accounts for the fact that census data may be imprecise and that population is a fluid, not static, concept. The minority group at issue here-at 39.36% of the voting-age population- consti tutes far less than a majority in the proposed district. To decide this case, therefore, it was not necessary for the North Carolina Supreme Court to state that a nu merical majority is always required.
C. Petitioners, who bear the burden of persuasion on the Section 2 issue in this case, have not shown that House District 18 is required by Section 2. In particu lar, petitioners' proposed "functional majority" approach is unsustainable. Such an approach is inconsistent not only with the Gingles framework, but with the text of Section 2, which grants minority groups the right to equal electoral opportunity, not the right to maximize electoral opportunity. In addition, such a "functional majority" test would make race a predominant factor in many redistricting decisions, thus raising constitutional concerns identified by this Court. Likewise, it would require district lines to be drawn based on a complex, predictive inquiry into how voters in a proposed district would vote in future elections with unknown candidates. That inquiry not only would place legislatures in an un tenable position in deciding whether a given district was required by Section 2, but also trigger perpetual cycles of Section 2 litigation. In the absence of any indication that Congress intended to adopt such a regime, the Court should decline to impose it.
D. If this Court nevertheless adopts petitioners' "functional majority" approach, it may wish to remand for further proceedings, including factfinding as to the likelihood that African Americans would be able to elect a candidate of choice in the proposed 39% African Amer ican district, as opposed to the 35% district that could have been drawn in compliance with state law.
SECTION 2 OF THE VOTING RIGHTS ACT DID NOT COM PEL NORTH CAROLINA TO DRAW HOUSE DISTRICT 18
Nothing in the Voting Rights Act of 1965 prevented North Carolina from drawing House District 18 with an approximately 39% African American voting-age popula tion. And the United States certainly supports efforts to draw legislative districts in a manner that will provide equal electoral opportunities for all voters, regardless of race, including districts in which minority voters consti tute less than a majority but nonetheless may carry sig nificant political clout because of coalitions with voters from other minority groups or "crossover" voting by members of the majority group. This case arises, how ever, because North Carolina's own law against splitting counties in redistricting prevented the state legislature from creating the district that it desired. Rather than complying with that law and drawing a district with a 35% African American voting-age population, North Carolina invoked Section 2 of the Voting Rights Act and argued that Section 2 required it to draw House District 18 with a 39% African American voting-age population. That position should be rejected.2
A. To Assert A Valid Claim, A Section 2 Plaintiff Ordi narily Must Show That Members Of A Minority Group Would Constitute A Majority In A Proposed District
1. In Thornburg v. Gingles, 478 U.S. 30 (1986), this Court held that, as "necessary preconditions" to proving that the use of multi-member districts constituted im permissible vote dilution under Section 2, a plaintiff must show (1) that a minority group is "sufficiently large and geographically compact to constitute a majority in a single-member district"; (2) that the minority group is "politically cohesive"; and (3) that "the white majority votes sufficiently as a bloc to enable it * * * usually to defeat the minority's preferred candidate." Id. at 50-51. The Court has held that the Gingles preconditions apply with at least as much force to a claim that a single-mem ber districting plan gives rise to impermissible vote dilu tion. Growe v. Emison, 507 U.S. 25, 40 (1993).
Both in Gingles itself and in numerous later cases, this Court has reserved the question whether there are any circumstances in which a plaintiff may assert a Sec tion 2 vote-dilution claim without showing that a district could be drawn in which members of a minority group would constitute a numerical majority. See Gingles, 478 U.S. at 46 n.12; Growe, 507 U.S. at 41 n.5; Voinovich v. Quilter, 507 U.S. 146, 154 (1993); Johnson v. De Grandy, 512 U.S. 997, 1009 (1994); League of United Latin American Citizens v. Perry, 548 U.S. 399, 443 (2006) (LULAC) (opinion of Kennedy, J.). The federal appel late courts that have considered the issue, however, have consistently held that a plaintiff must show that a majority-minority district could be drawn in order to satisfy the first Gingles precondition and assert a valid Section 2 vote-dilution claim. See, e.g., Hall v. Virginia, 385 F.3d 421, 427-430 (4th Cir. 2004), cert. denied, 544 U.S. 961 (2005); Valdespino v. Alamo Heights Indep. Sch. Dist., 168 F.3d 848, 852-853 (5th Cir. 1999), cert. denied, 528 U.S. 1114 (2000); Cousin v. Sundquist, 145 F.3d 818, 828-829 (6th Cir. 1998), cert. denied, 525 U.S. 1138 (1999); Sanchez v. Colorado, 97 F.3d 1303, 1311- 1312 (10th Cir. 1996), cert. denied, 520 U.S. 1229 (1997); McNeil v. Springfield Park Dist., 851 F.2d 937, 947 (7th Cir. 1988), cert. denied, 490 U.S. 1031 (1989).
In the decades since the Voting Rights Act was passed, no federal appellate court has held that Section 2 requires the creation of a district like the one at issue here, in which the minority group constitutes substan tially less than a majority of the voting-age population. In Metts v. Murphy, 363 F.3d 8 (2004) (en banc) (per curiam), the First Circuit refused, at the motion-to-dis miss stage, to "foreclose the possibility that a section 2 claim can ever be made out" where a majority-minority district could not be drawn. Id. at 11 (emphasis added). The court stopped short, however, of holding that Sec tion 2 required the creation of such a district.
2. In prior cases before this Court, the United States has "agree[d] with those courts that have re jected" the assumption "that Section 2 requires creation of districts in which minorities are demonstrably not a majority of the voting age population." U.S. Br. at 16, Voinovich, supra (No. 91-1618); see U.S. Br. at 7, Growe, supra (No. 91-1420); U.S. Br. at 18-20, LULAC, supra (Nos. 05-204, 05-276 & 05-439). The general re quirement that a plaintiff bringing a Section 2 vote-dilu tion claim must show that members of a minority group would constitute a numerical majority in a proposed dis trict is supported by two principal considerations.
a. The majority-minority requirement squares with the text of Section 2. If a minority group constitutes a numerical majority in a proposed district (and is politi cally cohesive), the group will have an "opportunity * * * to elect" a representative of its own choice with out support from other groups. See Growe, 507 U.S. at 40. If the preferred candidate of voters from a minority group can be elected only with "crossover" votes from the majority group, however, members of the minority group-i.e., "members of a class of citizens protected by [Section 2]"-would not, in ordinary parlance, be de prived of the opportunity to "elect representatives of their choice." 42 U.S.C. 1973(b) (emphases added). Rather, as Justice Brennan observed for the Court in Gingles, when the minority group "could not constitute a majority in a single-member district, these minority voters cannot maintain that they would have been able to elect representatives of their choice in the absence of the [challenged voting practice]." 478 U.S. at 50 n.17.
When a minority group must rely on "crossover" votes, the prevailing candidate would be elected not by a majority composed of voters from that group, but by a coalition of voters from that group and other voters- and that coalition (as opposed to the minority group) would be the only group deprived of the opportunity to elect the representative of its choice if a proposed "crossover" district were not adopted. Such a coalition of minority and non-minority voters is not "a class of citizens protected by [Section 2]." 42 U.S.C. 1973(b). Thus, this Court has recognized that, where the majority-minority requirement is not satisfied, minority voters "cannot claim to have been injured by [the chal lenged] structure or practice," Gingles, 478 U.S. at 50 n.17, and "there neither has been a wrong nor can be a remedy," Growe, 507 U.S. at 41.
In addition, because it is highly likely that voters from a politically cohesive minority group would elect a representative of their choice from a proposed district in which they constitute a majority, it can fairly be pre sumed that the failure to adopt such a district would deprive those voters of an equal opportunity to elect the representative of their choice-at least where, in the absence of such a district, the majority has voted suffi ciently as a bloc to enable it to defeat the minority's pre ferred candidate. Together with the other Gingles pre conditions, therefore, the threshold majority-minority requirement helps courts to winnow out those cases in which there is no substantial likelihood that a jurisdic tion is affording minority voters "less opportunity [than majority voters] * * * to elect representatives of their choice." 42 U.S.C. 1973(b).3
b. Like the other Gingles preconditions, the ma jority-minority requirement also provides a clear and administrable threshold standard for applying Section 2. As lower courts have noted, such a standard furthers "interests in clarity and uniformity" in the application of Section 2. McNeil, 851 F.2d at 944. That is true not only for the courts, but also for legislatures that must draw districts in compliance with Section 2 and thus must look to this Court's precedents for guidance. Even those members of the Court who have criticized the majority-minority requirement have acknowledged the need for a "clear-edged rule" at the threshold of the Sec tion 2 inquiry. LULAC, 548 U.S. at 485 (Souter, J., con curring in part and dissenting in part).
Under the majority-minority requirement, one need only determine whether members of a minority group would constitute more than 50% of the relevant popula tion in a proposed district. See LULAC, 548 U.S. at 429 (suggesting that the relevant population is the citizen voting-age population). In the vast majority of cases, that inquiry will be straightforward. Petitioners sug gest (Br. 41-42) that such a requirement may be difficult to administer because, insofar as the relevant population is the citizen voting-age population, data concerning citizenship status may not be readily available. Where that is the case, however, a plaintiff may simply present evidence that members of a minority group could consti tute a numerical majority of the overall voting-age population-as, indeed, petitioners did here. See Pet. App. 27a; Barnett v. City of Chicago, 141 F.3d 699, 705 (7th Cir.), cert. denied, 524 U.S. 954 (1998).
B. This Case Does Not Fit Into Either Of The Situations In Which The United States Has Stated That The General Majority-Minority Requirement May Be Relaxed
The United States has previously explained that the majority-minority requirement may be relaxed in two specific situations. See U.S. Br. at 19-20, LULAC, supra (Nos. 05-204, 05-276 & 05-439). Neither of those situa tions, however, is presented here. Accordingly, while the North Carolina Supreme Court properly concluded that the State was not obligated under Section 2 to draw the proposed 39% minority district at issue, it went fur ther than necessary to decide this case by stating that a proposed district must always "satisfy the numerical majority requirement as defined herein." Pet. App. 33a.
1. The United States has explained that "the re quirement that the minority group be sufficiently nu merous and compact to constitute a majority of a single- member district may be relaxed where intentional racial discrimination has been shown." U.S. Br. at 12 n.6, Growe, supra (No. 91-1420); see Garza v. County of Los Angeles, 918 F.2d 763, 770-771 (9th Cir. 1990), cert. de nied, 498 U.S. 1028 (1991); U.S. Br. in Opp. at 21-23, Garza, supra (No. 90-849). Where a plaintiff shows that the challenged redistricting plan was adopted with dis criminatory intent, such evidence tends to suggest that the jurisdiction is not providing an equal opportunity to minority voters to elect the representative of their choice, and it is therefore unnecessary to consider the majority-minority requirement before proceeding to the ultimate totality-of-the-circumstances analysis. And where intentional racial discrimination is shown, relax ation of the majority-minority requirement ensures that a Section 2 claim will be available whenever such a claim is also cognizable under the Fourteenth or Fifteenth Amendment of the Constitution (each of which requires a showing of discriminatory intent). See Reno v. Bossier Parish Sch. Bd., 520 U.S. 471, 481-482 (1997).
2. The United States has explained that the ma jority-minority requirement may also be relaxed where the minority group is "compact * * * and substantial in size yet just short of a majority." U.S. Br. at 19, LULAC, supra (Nos. 05-204, 05-276, and 05-439) (cita tion omitted). In Valdespino v. Alamo Heights Inde pendent School District (No. 98-1987), for example, the minority group at issue represented 48% of the citizen voting-age population. The United States filed a brief as amicus curiae at the certiorari stage, repeatedly refer ring to the fact that the minority group was "just short of a majority." U.S. Br. at 11, Valdespino, supra; see id. at 10 (noting that the minority population was "slightly less than 50%"). In Perez v. Pasadena Independent School District (No. 98-1747), the United States filed an amicus brief reiterating the position it took in Valde spino. See U.S. Br. at 7-8, Perez, supra.
This qualification accounts for the fact that there may be difficulties in determining the precise percent age of minority members in a proposed district's popula tion, and statistical discrepancies could therefore make the difference in borderline cases. For example, as amici note, see, e.g., Persily Br. 7-15; LWV Br. 19-28; Illinois Br. 27-29, the Census Bureau's data on citizen voting-age population rely to some extent on estimates. As a result, to the extent that the citizen voting-age pop ulation is the relevant population, there may be situa tions in which the figures for a proposed district are sub ject to sampling error (or other types of error, such as undercounting). See, e.g., Garza v. County of Los An geles, 756 F. Supp. 1298, 1322 (C.D. Cal.), aff'd, 918 F.2d 763 (9th Cir. 1990), cert. denied, 498 U.S. 1028 (1991). Where the plaintiff presents data indicating that the minority group at issue represents nearly 50% of the relevant population in a proposed district, it is thus pos sible that the true figure is in fact above 50%.
Moreover, any data on population merely represent a snapshot in time, and migration to and from a pro posed district, as well as changes in citizenship rates (to the extent that citizenship status is relevant), may result in an actual minority population that is higher or lower than statistical estimates based on preexisting data sug gest. In Garza, for example, the minority group at issue did not constitute more than 50% of the population in the first year after the 1980 census, but evidence suggested that it had grown to a majority in the following years. See U.S. Br. in Opp. at 12-13, Garza, supra (No. 90-849). For that reason as well, it is reasonable to conclude that the first Gingles precondition may be satisfied where the data indicate that a minority group is just short of 50% of the relevant population in a proposed district.4
3. The proposed district at issue in this case does not fall into either category. Petitioners do not contend that House District 18 was adopted with discriminatory intent.5 Likewise, African Americans constitute only 39.36% of the voting-age population of House District 18 (and no evidence suggests that African Americans con stitute a different percentage of the citizen voting-age population). Pet. App. 69a. This case is therefore far removed from Valdespino (where the minority group represented 48% of the citizen voting-age population), or a case where, based on a discrepancy in data or recent population changes, it is plausible to conclude that a mi nority group may in fact constitute more than 50% of the relevant population in the proposed district. As a result, in order to decide this case, it was not necessary for the North Carolina Supreme Court to state that a numerical majority is always required. See id. at 33a.
C. Petitioners Have Not Sustained Their Burden Of Estab lishing That Section 2 Required North Carolina To Draw The 39% Minority District At Issue
At a minimum, the further that a minority group in a proposed legislative district is from a numerical major ity, the more difficult it is for the group to maintain a valid Section 2 claim. Petitioners, as the parties effec tively raising the Section 2 claim in this case, bear the burden of persuasion on the Section 2 issue. See Voino vich, 507 U.S. at 155. To meet that burden in sustaining the 39% minority district here, petitioners urge this Court to adopt a "functional majority" test that goes far beyond the threshold precondition originally articulated in Gingles. That contention should be rejected.
1. To begin with, adopting petitioners' proposed interpretation of Section 2 would require retooling Gingles-if not, as petitioners' own amici suggest, rele gating Gingles to the status of an alternative (and fallback) standard. See LULAC, 548 U.S. at 490 n.8 (Souter, J., concurring in part and dissenting in part) (recognizing that "[a]ll aspects of our established analy sis for majority-minority districts in Gingles and its progeny may have to be rethought in analyzing ostensi ble ['crossover'] districts"); see also, e.g., Lawyers' Com mittee Br. 22-28; MALDEF Br. 4. Although it is true that this Court has reserved the question presented by petitioners, the Court should be reluctant to adopt a position that would require it to overhaul, if not effec tively dismantle, the framework used by it and the fed eral appellate courts for more than two decades in evalu ating Section 2 vote-dilution claims.6
2. Petitioners' "functional majority" test also is diffi cult to reconcile with the text of Section 2. As discussed above, see pp. 10-11, when a minority group constitutes substantially less than a majority in a proposed district, "minority voters cannot maintain that they would have been able to elect representatives of their choice in the absence of the [challenged voting practice]." Gingles, 478 U.S. at 50 n.17. And such voters therefore cannot claim that they were denied an equal opportunity to elect the representative of their choice because the pro posed district was not drawn. See 42 U.S.C. 1973(b).
Petitioners' interpretation of Section 2 also is at odds with the Court's admonition that the "[f]ailure to maxi mize [minority electoral opportunity] cannot be the mea sure of § 2." De Grandy, 512 U.S. at 1017; see id. at 1026 (Kennedy, J., concurring in part and concurring in the judgment). As this Court has stressed, "the ultimate right of § 2 is equality of opportunity, not a guarantee of electoral success for minority-preferred candidates." Id. at 1014 n.11. In that regard, petitioners err in stat ing that "[t]he text of Section imposes an opportunity- to-elect precondition." Br. 17 (emphasis added); see, e.g., Br. 28 (contending that Section 2 "asks whether a minority group, of whatever size, has been deprived of an opportunity to elect candidates of choice"). To be clear, Section 2 grants minority groups the right to an equal opportunity to elect, see 42 U.S.C. 1973(b), not the right simply to maximize electoral opportunity.
Petitioners' position is fundamentally inconsistent with that principle. Petitioners contend that Section 2 effectively requires the State to draw a district when a minority group can put together a winning coalition, even if the minority group itself is substantially less than a majority. Under petitioners' standard, a plaintiff presumably could pursue a Section 2 vote-dilution claim on behalf of a minority group whose members consti tuted only a fraction of the relevant population, where the plaintiff could show that the rest of the population was divided in such a way that the preferred candidate of the minority group would be unlikely to be elected (absent the proposed district). Such a standard would potentially give one group not merely an equal opportu nity, but a greater opportunity than another group to elect the candidate of its choice.
Even petitioners appear to recognize that there must be some limit to the "functional majority" test. Thus, they suggest that a Section 2 vote-dilution claim could not proceed where the amount of "crossover" voting re quired to render it likely that the preferred candidate of voters from a minority group would be elected exceeds a certain numerical threshold. See Br. 16 (suggesting that a "functional majority" claim can proceed only where there is a "limited number" of "crossover" vot ers). Such a requirement is no more grounded in the text of Section 2 than an unqualified "functional major ity" test. Moreover, it would in effect simply substitute a lower percentage for the majority-minority require ment of the first Gingles precondition. And ultimately -no doubt following significant litigation and uncer tainty-the "crossover" line on which courts ultimately settle would be no less "unbending" (ibid.) than the 50% requirement of which petitioners complain.
Petitioners' amici alternatively suggest that a court could allow a Section 2 vote-dilution claim to proceed to the totality-of-the-circumstances inquiry based on satis faction of the "functional majority" test, but reject such a claim under the totality of circumstances where the resulting number of "functional majority" districts would be disproportionate to the minority group's share of the population. See, e.g., Lawyers' Committee Br. 14- 15. This Court, however, has consistently held that pro portionality is not a requirement of Section 2, but in stead merely one of many factors to be considered in the totality-of-the circumstances inquiry. See LULAC, 548 U.S. at 436; De Grandy, 512 U.S. at 1017-1021. That approach, moreover, would likewise simply be a recipe for more, and more protracted, Section 2 litigation.
3. The constitutional concerns and potentially dras tic practical consequences of petitioners' proposed inter pretation of Section 2 also counsel against adopting that interpretation, particularly in the absence of any indica tion that Congress intended it. See, e.g., Koons Buick Pontiac GMC, Inc. v. Nigh, 543 U.S. 50, 63 (2004); Chisom v. Roemer, 501 U.S. 380, 396 & n.23 (1991).7
a. The Voting Rights Act was intended to "hasten the waning of racism in American politics," De Grandy, 512 U.S. at 1020, not to ensure that "race predominates in the redistricting process," Miller v. Johnson, 515 U.S. 900, 916 (1995); see Shaw v. Reno, 509 U.S. 630, 657 (1993) ("Racial classifications with respect to voting carry particular dangers."). Indeed, as this Court ob served in Georgia v. Ashcroft, 539 U.S. 461 (2003), "the Voting Rights Act, as properly interpreted, should en courage the transition to a society where race no longer matters: a society where integration and color-blind ness are not just qualities to be proud of, but are simple facts of life." Id. at 490-491; see De Grandy, 512 U.S. at 1029-1031 (Kennedy, J., concurring in part and concur ring in the judgment). Petitioners' proposed standard runs counter to that goal.
Notwithstanding petitioners' assurances to the con trary, see, e.g., Br. 37-39, their proposed "functional ma jority" test will inevitably magnify the consideration of race in redistricting decisions. Members of this Court have already recognized that extending Section 2 to "in fluence" districts would have that effect. See LULAC, 548 U.S. at 446 (opinion of Kennedy, J.) ("If § 2 were interpreted to protect ['influence' districts], it would unnecessarily infuse race into virtually every redistrict ing, raising serious constitutional questions"). Extend ing Section 2 to "functional majority" districts would raise similar concerns. Indeed, under a "functional ma jority" standard, a redistricting record might frequently (if not typically) resemble the record in Georgia, where the State sought to maximize the number of districts in which minority groups, even if not a majority of the voting-age population, would control electoral races. See 539 U.S. at 470. As Justice Kennedy observed in his concurring opinion, while the issue was not before the Court, the facts of Georgia indicated that "race was a predominant factor in drawing the lines of Georgia's * * * redistricting map." Id. at 491.
This Court has interpreted the Voting Rights Act to avoid such constitutional concerns, see, e.g., Miller, 515 U.S. at 926, and it should do so here as well. Indeed, as discussed next, the inevitable administrative difficulties and litigation that petitioners' approach invites would not only magnify the legislatures' use of race in redis tricting, but enlist the courts in making predictive judg ments about the role of race in the electoral process. That cannot help but reenforce "the very racial stereo typing the Fourteenth Amendment forbids." Id. at 928; see Shaw, 509 U.S. at 647-648.
b. A "functional majority" test would also generate considerable uncertainty and administrative difficulties for legislatures responsible for drawing district lines. As petitioners recognize (Br. 34), the Gingles precondi tions focus on "the actual voting behavior of the elector ate" (e.g., whether the minority group is politically cohe sive or whether the majority group votes sufficiently as a bloc to enable it to defeat the minority's preferred can didate). By contrast, a "functional majority" test would mandate a much more complicated, predictive inquiry into likely voting behavior. See McNeil, 851 F.2d at 944 (noting that "[m]ovement away from the Gingles stan dard invites courts to build castles in the air, based on quite speculative foundations"). That inquiry would greatly complicate the task faced by legislatures in de termining whether Section 2 requires the creation of a particular district following new census data, a new elec tion cycle, or simply a new political trend.
For example, a legislature considering whether a proposed district is required, or a court engaging in a threshold "functional majority" inquiry, would have to determine, inter alia, (1) how many voters in the pro posed district would be registered and turn out from each group; (2) how many voters would "cross over" to support the (hypothetical) preferred candidate of a minority group; (3) how many voters from the minority group would "cross over" in the opposite direction to support a (hypothetical) alternative candidate; (4) whether an incumbent would run for reelection in future cycles (and, if so, whether the presence of the incumbent would affect the likelihood that the preferred candidate of the minority group would be elected); and (5) whether there would be "coattail" effects in future cycles based on the results in other elections. See, e.g., Pet. App. 25a; Persily Br. 15-16; Bishop Br. 13-17.8
As petitioners' own amici concede, therefore, a "func tional majority" test would require "a more complicated analysis" involving "development of additional evidence that is not part of the existing § 2 calculus." Lawyers' Committee Br. 5, 7, 24-28; cf. Georgia, 539 U.S. at 480 ("The ability of minority voters to elect a candidate of their choice is * * * often complex in practice to de termine."). Moreover, as petitioners' amici further ac knowledge, "[a] functional approach is likely to be espe cially problematic in cases involving local elections for governing bodies"-which represent a "substantial amount of § 2 litigation"-given that relevant data for such districts will be even more difficult, if not impossi ble, to obtain. Lawyers' Committee Br. 24.9
The upshot is that, by presenting considerably greater difficulties in application, the threshold Gingles (or Gingles-like) inquiry would cease to perform any useful screening function, and legislatures would face enormous uncertainty in drawing districts in compliance with Section 2. This Court has repeatedly stressed that "[e]lectoral districting is a most difficult subject for leg islatures"; that "reapportionment is primarily the duty and responsibility of the State"; and that "[f]ederal- court review of districting legislation represents a seri ous intrusion on the most vital of local functions." Miller, 515 U.S. at 915 (citation omitted); see Voinovich, 507 U.S. at 156-157; Growe, 507 U.S. at 34. Redistrict ing is already hard enough. Petitioners' proposed stan dard would greatly complicate the task faced by the States; eliminate a significant degree of discretion that States currently possess in deciding how to draw district lines; and, as discussed next, subject the States to pro tracted litigation over new plans. Given the absence of any indication that Congress intended to subject the States to such a regime, this Court should decline to interpret Section 2 in that fashion.
c. Adopting petitioners' "functional majority" test would not only transform the nature of Section 2 litiga tion, but increase its volume as well. As the North Carolina Supreme Court observed, under petitioners' proposed approach, "each legislative district is exposed to a potential legal challenge by a numerically modest minority group with claims that its voting power has been diluted and that a district therefore must be con figured to give it control over the election of candi dates," creating a "Pandora's box of marginal Voting Rights Act claims by minority groups of all sizes." Pet. App. 24a (citation omitted). That concern is magnified by the fact that, unlike Section 5 of the Voting Rights Act, which applies only to certain covered jurisdictions, Section 2 applies nationwide.
Moreover, if a plaintiff could pursue a Section 2 claim by proposing a district with as little as a 39% minority population, then it would be difficult, if not impossible, to establish a reasoned endpoint or judicially manage able standards for evaluating vote-dilution claims by minority groups constituting less than a majority. Cf. Holder v. Hall, 512 U.S. 874, 885 (1994) (plurality opin ion) (rejecting a Section 2 challenge to the size of a gov erning authority on the ground that such claims are "inherently standardless"); id. at 890 (O'Connor, J., con curring in part and concurring in the judgment) (same). The result would not only make Section 2 claims even more difficult to adjudicate, but-given the breadth and complexity of petitioner's proposed standard-trigger potentially never-ending cycles of Section 2 litigation.
4. A "functional majority" test is not, as petitioners suggest (Br. 34-36), necessary to align Section 2 with Section 5. As this Court has repeatedly emphasized, "the § 2 inquiry differs in significant respects from a § 5 inquiry." Georgia, 539 U.S. at 478. The basic premise of petitioners' argument is therefore fundamentally un sound. Moreover, in Georgia, the Court rejected the argument that Section 5 requires the States to maximize the number of majority-minority districts and, instead, held that "Section 5 leaves room for States to use * * * influence and coalitional districts" as well. Id. at 483. That decision is inconsistent with petitioners' position that Section 2 required North Carolina to draw a "cross over" district like House District 18, because such an interpretation of Section 2 would deprive the States of the "flexibility" that Georgia held the States retained under Section 5 "to choose one theory of effective repre sentation over the other." Id. at 482.
As part of the most recent reenactment of the Voting Rights Act in 2006, Congress added new Section 5(b), which provides that, for purposes of Section 5, "[a]ny voting * * * practice * * * that * * * will have the effect of diminishing the ability of any citizens of the United States on account of race or color, or [language minority status], to elect their preferred candidates of choice denies or abridges the right to vote." Fannie Lou Hamer, Rosa Parks, and Coretta Scott King Voting Rights Act Reauthorization and Amendments Act of 2006, Pub. L. No. 109-246, § 5(b), 120 Stat. 580-581 (to be codified at 42 U.S.C. 1973c(b)). In adopting that pro vision, however, Congress did not amend Section 2 (or its equal-opportunity-to-elect standard). In addition, although it is clear that new Section 5(b) was intended to overrule Georgia at least in part, it is unclear whether Congress intended to preclude consideration of only "influence" districts in the Section 5 inquiry, see, e.g., H.R. Rep. No. 478, 109th Cong., 2d Sess. 70-71 (2006), or "crossover" districts as well, see, e.g., S. Rep. No. 295, 109th Cong., 2d Sess. 20-21 (2006).
In any event, even assuming that the amended Sec tion 5 mandates consideration of "crossover" districts as part of the retrogression inquiry, that is not incompati ble with the longstanding interpretation of Section 2 discussed above. The majority-minority requirement of the first Gingles precondition does not preclude consid eration of "crossover" districts altogether, because it does not require a jurisdiction to draw a majority-minor ity district whenever such a district could be drawn. Instead, a jurisdiction remains free to draw a "cross over" district, provided that, under the totality of cir cumstances, the resulting district does not deny minor ity voters an equal opportunity to elect the representa tive of their choice. See, e.g., Uno v. City of Holyoke, 72 F.3d 973, 990-991 (1st Cir. 1995). So too, assuming that Section 5 mandates consideration of "crossover" dis tricts as part of the retrogression inquiry, a jurisdiction would remain free to draw a "crossover" district, pro vided that there is no discriminatory purpose or retro gressive effect. Accordingly, there is no "disharmony" (Pet. Br. 34) between Section 2 and Section 5 that a "functional majority" test is needed to resolve.
5. As noted, this Court's refusal to adopt a "func tional majority" test would not preclude state legisla tures from drawing "crossover" districts, consistent with the other restraints imposed by the Voting Rights Act. Indeed, a number of legislatures did so in the wake of the 2000 census, notwithstanding the fact that no court of appeals had embraced the "functional majority" test. See, e.g., Bishop Br. 4-5. Many have argued that such districts are desirable as a policy matter, because "mi nority voters are not immune from the obligation to pull, haul, and trade to find common political ground." De Grandy, 512 U.S. at 1020; see generally Georgia, 539 U.S. at 480-483. Section 2, however, leaves States a measure of discretion to reach their own conclusion on such matters of political theory; it does not require States to create such districts whenever feasible.
Certainly, there have been significant gains since the Voting Rights Act was enacted in eliminating the invidi ous discrimination targeted by the Act. See Bishop Br. 4-5. Petitioners' amici argue that, "[o]ver time, as voters become increasingly willing to judge candidates on their merits, rather than on the color of their skin, the de mands of the Voting Rights Act must evolve." LWV Br. 4. The proper forum for such evolution, however, is Congress, not the courts. As the recent amendments to (and reauthorization of) Section 5 underscore, Congress has not hesitated to adapt the Voting Rights Act to new trends or concerns as it sees fit. The policy arguments raised by petitioners and their amici are therefore better addressed to Congress than to this Court.
D. If This Court Adopts A "Functional Majority" Standard For Section 2 Vote-Dilution Claims, It May Wish To Remand For Further Proceedings
If this Court adopts a "functional majority" test or otherwise alters the longstanding Gingles framework, it may wish to remand this case for further proceedings and, in particular, for consideration of two issues that the lower courts did not address.
First, it may be appropriate for the lower courts to determine whether the "crossover" required to ensure the election of the preferred candidate of minority vot ers in House District 18 is simply too large to permit a Section 2 vote-dilution claim challenging that district to proceed. In this case, with African Americans constitut ing 39.36% of the voting-age population of House Dis trict 18, almost 18% of the remaining voting-age popula tion would have to "cross over" in a given election-and perhaps more, depending on the extent to which African American voters support other candidates.10 "Cross over" voting of that magnitude may exceed the appropri ate threshold, even under a "functional majority" test. Cf. Abrams v. Johnson, 521 U.S. 74, 92-93 (1997).11
Second, it may be appropriate for the lower courts to determine whether, under the totality of circumstances, it was necessary for the General Assembly to draw House District 18 in order to avoid violating Section 2 of the Voting Rights Act. While the evidence presented below indicates that African Americans constitute 39.36% of the voting-age population in House District 18, Pet. App. 69a, it also indicates that the General As sembly could have drawn an alternative district that kept Pender County together with an African American voting-age population of 35.33%, J.A. 73. It may be pos sible to argue that a 4% difference in the African Ameri can voting-age population would not materially affect the likelihood that the preferred candidate of African American voters would be elected-and that, without a showing that the 4% difference is likely to be dispositive for purposes of ensuring an equal opportunity to elect, Section 2 should not be read to require North Carolina to create the district at issue.12 Of course, the difficul-
ties of the foregoing inquiries simply underscore the doctrinal and practical problems raised more generally by petitioners' proposed "functional majority" rule.
The judgment of the North Carolina Supreme Court should be affirmed.
GREGORY G. GARRE
Acting Solicitor General
GRACE CHUNG BECKER
Acting Assistant Attorney
KANNON K. SHANMUGAM
Assistant to the Solicitor
DIANA K. FLYNN
ANGELA M. MILLER
1 Forty North Carolina counties, but not Pender County or New Hanover County, are covered jurisdictions subject to the preclearance requirement in Section 5 of the Voting Rights Act, 42 U.S.C. 1973c. See 28 C.F.R. Pt. 51, App. North Carolina submitted the 2003 redistricting plan for preclearance, and the Attorney General declined to interpose an objection. Preclearance under Section 5 does not preclude a sub sequent action under Section 2. See 28 C.F.R. 51.55(b); Reno v. Bossier Parish Sch. Bd., 520 U.S. 471, 477 (1997).
2 Districts in which members of a minority group do not constitute a numerical majority have variously been described as "crossover," "coalition," and "influence" districts. Like the decision below, see Pet. App. 18a-19a, this brief uses "crossover" district to refer to a district in which the preferred candidate of voters from one minority group can be elected only with "crossover" votes from other voters. A "coalition" district, by contrast, is one in which voters from two minority groups could be said to form a "coalition" to elect the representative of their choice. There is no claim in this case that House District 18 was a "coalition" district. An "influence" district is one in which the preferred candidate of voters from a minority group cannot be elected, but voters from the minority group can nevertheless influence the outcome of the election. In League of United Latin American Citizens v. Perry, 548 U.S. 399 (2006) (LULAC), a majority of this Court agreed that Section 2 vote-dilution claims involving "influence" districts are not cognizable. See id. at 445 (opinion of Kennedy, J.); id. at 512 (Scalia, J., concurring in the judgment in part and dissenting in part).
3 Petitioners contend (Br. 32-34) that the majority-minority require ment is inconsistent with Section 2's directive that a court should con sider "the totality of circumstances" in determining whether members of a minority group have been deprived of an equal opportunity to elect the representative of their choice. 42 U.S.C. 1973(b). This Court has made clear, however, that courts should assess the Gingles precondi tions before analyzing the totality of circumstances. See, e.g., LULAC, 548 U.S. at 425-427. Thus, Section 2's "totality of circumstances" langu age imposes an additional, not alternative, requirement. In addition, as discussed below, the United States has previously explained that the majority-minority requirement may be relaxed in particular situations. See pp. 13-17, infra. Those exceptions permit the consideration of additional factors, depending on the "nature of the claim," and thus prevent a "mechanical" application of the majority-minority rule. Johnson v. De Grandy, 512 U.S. 997, 1007 (1994) (citation omitted).
4 In some of the near-50% borderline cases discussed in the text, the United States has justified relaxation of the majority-minority require ment in part on the ground that, in those cases, only a "small amount" of "crossover" voting would be necessary to ensure that the preferred candidate of voters from the minority group could be elected. U.S. Br. at 11, Valdespino, supra (No. 98-1987); see id. at 10-11 n.2 (observing that only "a minimal amount of crossover voting * * * would enable minority voters to elect representatives of their choice"); id. at 12 n.3 (referring to "a small but consistent amount of crossover voting" and stating that, "[a]t some point, of course, the amount of crossover voting" may be too substantial to support a Section 2 claim). The amount of "crossover" voting required to ensure that a candidate preferred by African Americans is elected in the proposed district at issue in this case, however, is significantly larger-almost 18% or more, see pp. 29- 30 & n.10, infra-than in previous cases in which the United States recognized that a Section 2 claim may lie, where the minority group was "substantial in size yet just short of a majority." U.S. Br. at 19, LULAC, supra (Nos. 05-204, 05-276, and 05-439).
5 Amici NAACP et al. suggest (Br. 13-15) that the whole-county provisions of the North Carolina Constitution, which prohibit the splitt ing of counties in legislative redistricting, themselves raise "serious concerns" under the Voting Rights Act. While a proper plaintiff would certainly be free to bring that claim, petitioners did not raise such a claim below and do not make it here.
6 As several courts (and petitioners' own amici, see, e.g., Lawyers' Committee Br. 13-14) have recognized, a "functional majority" test also could create tension between the first and third Gingles preconditions. See, e.g., Metts, 363 F.3d at 12 ("To the extent that African-American voters have to rely on cross-over voting to prove they have the 'ability to elect' a candidate of their choosing, their argument that the majority votes as a bloc against their preferred candidate is undercut."); id. at 14 (Selya, J., joined by Torruella, J., dissenting) (noting that "[a] showing of majoritarian bloc voting is structurally inconsistent with" a minority group's "reliance on a high level of crossover voting" to assert electoral control); see also U.S. Br. 12 n.3, Valdespino, supra (No. 98-1987).
7 Far from establishing that Congress intended to permit vote- dilution claims under Section 2 on a "functional majority" theory, the legislative history of the 1982 amendments to the Voting Rights Act suggests that Congress was focused on voting practices that reduced the minority population in electoral districts below 50%. See, e.g., S. Rep. No. 417, 97th Cong., 2d Sess. 120-121 (1982). Moreover, when it came to "interpreting the definitional parameters of districts which give blacks an opportunity to elect the candidate of their own choice," the legislative history indicates that Congress was working off a base line rule that "a 65 percent level of minority population in a given district is * * * one which will 'give blacks an opportunity to elect a candidate of their choice.'" Id. at 121 n.40 (quoting testimony of Professor George C. Cochran).
8 In LULAC, Justice Souter, joined by Justice Ginsburg, suggested that the first Gingles precondition could be satisfied where "minority voters in a [proposed] district constitute a majority of those voting in the primary of the dominant party, that is, the party tending to win in the general election." 548 U.S. at 485-486 (opinion concurring in part and dissenting in part). But Section 2 was designed to protect minority groups, not the "dominant party" in place. See 42 U.S.C. 1973(b). In addition, leaving aside the question of how the other Gingles precondi tions would operate in that context, see 548 U.S. at 490, such a standard would potentially present difficulties in application similar to those of petitioners' "functional majority" standard. For example, it would require a legislature or court to predict how many voters would be registered and turn out from each group and which party voters would support in the general election-all with regard to a proposed district as to which precise actual data likely do not exist.
9 In some cases, a legislature or court may determine that it is possible, but not certain, that the preferred candidate of minority voters would be elected on the basis of "crossover" voting. Under petitioners' apparent standard, it is unclear how probable the election of a preferred candidate would have to be (or how confident a legisla ture or court would have to be about the candidate's election) in order to satisfy the "functional majority" test. See Pet. Br. 40 (suggesting that the plaintiff must show that minority voters are "likely" to have the "opportunity" to elect candidates of their choice); Lawyers' Committee Br. 4 (suggesting that the plaintiff must show that "minority voters * * * have a realistic potential to elect candidates of their choice").
10 Petitioners suggest (Br. 31) that the necessary amount of "cross over" voting is 11%. That is misleading. Because African American voters constituted 39.36% of the voting-age population in House District 18, other voters constitute 60.64% of that population. Assuming that every African American voter supported the candidate of choice, it is true that another 10.64% of the total voting-age population would have to support that candidate for the candidate to be elected. The critical fact, however, is that 17.55% of the remaining voting-age population (i.e., 10.64% divided by 60.64%) would have to "cross over" to support the minority group's candidate of choice.
11 Although respondents stipulated, with regard to the third Gingles precondition, that "the racial difference in the preference of voters results in the white majority voting sufficiently as a block [sic] to usually enable it to defeat the minority's preferred candidate," Pet. App. 130a, they made no stipulation concerning the distinct question of the extent to which white voters in House District 18 would "cross over" to support the minority group's preferred candidate (or the extent to which members of the minority group would support other candidates).
12 In making this determination, the lower courts would need to consider petitioners' evidence that, in the wake of the 1990 census, no House district elsewhere in North Carolina with an African American voting-age population of less than 38.37% had a history of electing African American representatives. J.A. 40, 45. The lower courts would also need to consider evidence concerning how the alternative district proposed by respondents would perform. See Resp. Br. 8-9, 49-50.