Johnson v. De Grandy , 114 S. Ct. 2647 ( 1994 )

Justice Souter

delivered the opinion of the Court.

These consolidated cases are about the meaning of vote dilution and the facts required to show it, when §2 of the Voting Rights Act of 1965 is applied to challenges to single-member legislative districts. See 79 Stat. 437, as amended, 42 U. S. C. § 1973. We hold that no violation of § 2 can be found here, where, in spite of continuing discrimination and racial bloc voting, minority voters form effective voting majorities in a number of districts roughly proportional to the minority voters’ respective shares in the voting-age population. While such proportionality is not dispositive in a challenge to single-member districting, it is a relevant fact in the totality of circumstances to be analyzed when determining whether members of a minority group have “less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.” Ibid.

I

On the first day of Florida’s 1992 legislative session, a group of Hispanic voters including Miguel De Grandy (De Grandy plaintiffs) complained in the United States District Court against the speaker of Florida’s House of Representatives, the president of its Senate, the Governor, and other state officials (State). The complainants alleged that the districts from which Florida voters had chosen their state senators and representatives since 1982 were malapportioned, failing to reflect changes in the State’s population during the ensuing decade. The State Conference of NAACP Branches and individual black voters (NAACP plaintiffs) filed a similar suit, which the three-judge District Court consolidated with the De Grandy case.¹

Several months after the first complaint was filed, on April 10, 1992, the state legislature adopted Senate Joint Resolution 2-G (SJR 2-G), providing the reapportionment plan currently at issue. The plan called for dividing Florida into 40 single-member Senate, and 120 single-member House, districts based on population data from the 1990 census. As the Constitution of Florida required, the state attorney general then petitioned the Supreme Court of Florida for a declaratory judgment that the legislature’s apportionment plan was valid under federal and state law. See Fla. Const., Art. Ill, § 16(c). The court so declared, while acknowledging that state constitutional time constraints precluded full review for conformity with §2 of the Voting Rights Act and recognizing the right of any interested party to bring a §2 challenge to the plan in the Supreme Court of Florida. See In re Constitutionality of Senate Joint Resolution 2G, Special Apportionment Session 1992, 597 So. 2d 276, 285-286 (1992).²

The De Grandy and NAACP plaintiffs responded to SJR 2-G by amending their federal complaints to charge the new reapportionment plan with violating § 2.³ They claimed that SJR 2-G “ ‘unlawfully fragments cohesive minority communities and otherwise impermissibly submerges their right to vote and to participate in the electoral process/” and they pointed to areas around the State where black or Hispanic populations could have formed a voting majority in a politically cohesive, reasonably compact district (or in more than one), if SJR 2-G had not fragmented each group among several districts or packed it into just a few. De Grandy v. Wetherell, 815 F. Supp. 1550, 1559-1560 (ND Fla. 1992).

The Department of Justice filed a similar complaint, naming the State of florida and several elected officials as defendants and claiming that SJR 2-G diluted the voting strength of blacks and Hispanics in two parts of the State in violation of § 2. The Government alleged that SJR 2-G diluted the votes of the Hispanic population in an area largely covered by Dade County (including Miami) and the black population in an area covering much of Escambia County (including Pensacola).⁴ App. 75. The District Court consolidated this action with the other two and held a 5-day trial, followed immediately by an hours-long hearing on remedy.

At the end of the hearing, on July 1, 1992, the District Court ruled from the bench. It held the plan’s provisions for state House districts to be in violation of §2 because “more than [SJR 2-G’s] nine Hispanic districts may be drawn without having or creating a regressive effect upon black voters,” and it imposed a remedial plan offered by the De Grandy plaintiffs calling for 11 majority-Hispanic House districts. App. to Juris. Statement 2a, 203a. As to the Senate, the court found that a fourth majority-Hispanic district could be drawn in addition to the three provided by SJR 2-G, but only at the expense of black voters in the area. Id., at 202a; 815 F. Supp., at 1560. The court was of two minds about the implication of this finding, once observing that it meant the legislature’s plan for the Senate was a violation of §2 but without a remedy, once saying the plan did not violate §2 at all.⁵ In any event, it ordered elections to be held using SJR 2-G’s senatorial districts.

In a later, expanded opinion the court reviewed the totality of circumstances as required by § 2 and Thornburg v. Gingles, 478 U. S. 30 (1986). In explaining Dade County’s “tripartite politics,” in which “ethnic factors . . . predominate over all other[s] .. . ,” 815 F. Supp., at 1572, the court found political cohesion within each of the Hispanic and black populations but none between the two, id., at 1569, and a tendency of non-Hispanic whites to vote as a bloc to bar minority groups from electing their chosen candidates except in a district where a given minority makes up a voting majority,⁶ id., at 1572. The court further found that the nearly one million Hispanics in the Dade County area could be combined into 4 Senate and 11 House districts, each one relatively compact and with a functional majority of Hispanic voters, id., at 1568-1569, whereas SJR 2-G created fewer majority-Hispanic districts; and that one more Senate district with a black voting majority could have been drawn, id., at 1576. Noting that Florida’s minorities bore the social, economic, and political effects of past discrimination, the court concluded that SJR 2-G impermissibly diluted the voting strength of Hispanics in its House districts and of both Hispamos and blacks in its Senate districts. Id., at 1574. The findings of vote dilution in the senatorial districts had no practical effect, however, because the court held that remedies for the blacks and the Hispanics were mutually exclusive; it consequently deferred to the state legislature’s work as the “fairest” accommodation of all the ethnic communities in south Florida. Id., at 1580.

We stayed the judgment of the District Court, 505 U. S. 1232 (1992), and noted probable jurisdiction, 507 U. S. 907 (1993).

II

Before going to the issue at the heart of these cases, we need to consider the District Court’s refusal to give preclusive effect to the decision of the State Supreme Court validating SJR 2-G. The State argues that the claims of the De Grandy plaintiffs should have been dismissed as res judicata because they had a full and fair opportunity to litigate vote dilution before the State Supreme Court, see In re Constitutionality of Senate Joint Resolution 2G, Special Apportionment Session 1992, 597 So. 2d, at 285. The premise, however, is false, exaggerating the review afforded the De Grandy plaintiffs in the state court and ignoring that court’s own opinion of its judgment’s limited scope. Given the state constitutional mandate to review apportionment resolutions within 30 days, see Fla. Const., Art. Ill, § 16(c), the Supreme Court of Florida accepted briefs and evidentiary submissions, but held no trial. In that court’s own words, it was “impossible ... to conduct the complete factual analysis contemplated by the Voting Rights Act... within the time constraints of article III,” and its holding was accordingly “without prejudice to the right of any protestor to question the validity of the plan by filing a petition in this Court alleging how the plan violates the Voting Rights Act.” 597 So. 2d, at 282, 285-286.

The State balks at recognizing this express reservation by blaming the De Grandy plaintiffs for not returning to the State Supreme Court with the § 2 claims. But the plaintiffs are free to litigate in any court with jurisdiction, and their choice to forgo further, optional state review hardly converted the state constitutional judgment into a decision following “full and fair opportunity to litigate,” Allen v. McCurry, 449 U. S. 90, 104 (1980), as res judicata would require. For that matter, a federal court gives no greater preclusive effect to a state-court judgment than the state court itself would do, Marrese v. American Academy of Orthopaedic Surgeons, 470 U. S. 373, 384-386 (1985), and the Supreme Court of Florida made it plain that its preliminary look at the vote dilution claims would have no preclusive effect under Florida law.

The State does not, of course, argue that res judicata bars the claims of the United States, which was not a party in the Florida Supreme Court action. It contends instead that the Federal Government’s §2 challenge deserved dismissal under this Court’s Rooker!Feldman abstention doctrine, under which a party losing in state court is barred from seeking what in substance would be appellate review of the state judgment in a United States district court, based on the losing party’s claim that the state judgment itself violates the loser’s federal rights. See District of Columbia Court of Appeals v. Feldman, 460 U. S. 462, 482 (1983); Rooker v. Fidelity Trust Co., 263 U. S. 413, 416 (1923). But the invocation of Rooker/Feldman is just as inapt here, for unlike Rooker or Feldman, the United States was not a party in the state court. It was in no position to ask this Court to review the state court’s judgment and has not directly attacked it in this proceeding. Cf. Feldman, supra, at 468, and n. 2, 472, and n. 8 (suing District of Columbia Court of Appeals); Rooker, supra, at 414 (seeking to have state court’s judgment declared null and void). The United States merely seeks to litigate its § 2 case for the first time, and the Government’s claims, like those of the private plaintiffs, are properly before the federal courts.-

Ill

On the merits of the vote dilution claims covering the House districts, the crux of the State’s argument is the power of Hispanics under SJR 2-G to elect candidates of their choice in a number of districts that mirrors their share of the Dade County area’s voting-age population (i. e., 9 out of 20 House districts); this power, according to the State, bars any finding that the plan dilutes Hispanic voting strength. The District Court is said to have missed that conclusion by mistaking our precedents to require the plan to maximize the number of Hispanic-controlled districts.

The State’s argument takes us back to ground covered last Term in two cases challenging single-member districts. See Voinovich v. Quilter, 507 U. S. 146 (1993); Growe v. Emison, 507 U. S. 25 (1993). In Growe, we held that a claim of vote dilution in a single-member district requires proof meeting the same three threshold conditions for a dilution challenge to a multimember district: that a minority group be “ ‘sufficiently large and geographically compact to constitute a majority in a single-member district’”; that it be “‘politically cohesive’ ”; and that “ ‘the white majority vot[e] sufficiently as a bloc to enable it . . . usually to defeat the minority’s preferred candidate.’ ” Id., at 40 (quoting Thornburg v. Gingles, 478 U. S., at 50-51). Of course, as we reflected in Voinovich and amplify later in this opinion, “the Gingles factors cannot be applied mechanically and without regard to the nature of the claim.” 507 U. S., at 158.

In Voinovich we explained how manipulation of district lines can dilute the voting strength of politically cohesive minority group members, whether by fragmenting the minority voters among several districts where a bloc-voting majority can routinely outvote them, or by packing them into one or a small number of districts to minimize their influence in the districts next door. See id., at 153-154. Section 2 prohibits either sort of line-drawing where its result, “ ‘inter-netting] with social and historical conditions,’ impairs the ability of a protected class to elect its candidate of choice on an equal basis with other voters.” Ibid, (quoting Gingles, supra, at 47).⁷

Plaintiffs in Growe and Voinovich failed to show vote dilution because the former did not prove political cohesiveness of the minority group, Growe, supra, at 41-42, and the latter showed no significant white bloc voting, Voinovich, supra, at 158. Here, on the contrary, the District Court found, and the' State does not challenge, the presence of both these Gingles preconditions. The dispute in this litigation centers on two quite different questions: whether Hispanics are sufficiently numerous and geographically compact to be a majority in additional single-member districts, as required by the first Gingles factor; and whether, even with all three Gingles conditions satisfied, the circumstances in totality support a finding of vote dilution when Hispanics can be expected to elect their chosen representatives in substantial proportion to their percentage of the area’s population.

A

When applied to a claim that single-member districts dilute minority votes, the first Gingles condition requires the possibility of creating more than the existing number of reasonably compact districts with a sufficiently large minority population to elect candidates of its choice. The District Court found the condition satisfied by contrasting SJR 2-G with the De Grandy plan for the Dade County area, which provided for 11 reasonably compact districts, each with a voting-age population at least 64 percent Hispanic. 815 F. Supp., at 1580. While the percentage figures are not disputed, the parties disagree about the sufficiency of these supermajorities to allow Hispanics to elect representatives of their choice in all 11 districts. The District Court agreed with plaintiffs that the supermajorities would compensate for the number of voting-age Hispanics who did not vote, most commonly because they were recent immigrants who had not become citizens of the United States. Id., at 1567-1568. The State protests that fully half of the Hispanic voting-age residents of the region are not citizens, with the result that several districts in the De Grandy plan lack enough Hispanic voters to elect candidates of their choice without cross-over votes from other ethnic groups. On these assumptions, the State argues that the condition necessary to justify tinkering with the State’s plan disappears.

We can leave this dispute without a winner. The parties’ ostensibly factual disagreement raises an issue of law about which characteristic of minority populations (e. g., age, citizenship) ought to be the touchstone for proving a dilution claim and devising a sound remedy. These cases may be resolved, however, without reaching this issue or the related question whether the first Gingles condition can be satisfied by proof that a so-called influence district may be created (that is, by proof that plaintiffs can devise an additional district in which members of a minority group are a minority of the voters, but a potentially influential one). As in the past, we will assume without deciding that even if Hispanics are not an absolute majority of the relevant population in the additional districts, the first Gingles condition has been satisfied in these cases. See Voinovich, supra, at 154; see also Growe, supra, at 41-42, n. 5 (declining to reach the issue); Gingles, supra, at 46-47, n. 12 (same).

B

We do, however, part company from the District Court in assessing the totality of circumstances. The District Court found that the three Gingles preconditions were satisfied, and that Hispanics had suffered historically from official discrimination, the social, economic, and political effects of which they generally continued to feel, 815 F. Supp., at 1573-1574. Without more, and on the apparent assumption that what could have been done to create additional Hispanic supermajority districts should have been done, the District Court found a violation of § 2. But the assumption was erroneous, and more is required, as a review of Gingles will show.

1

Thornburg v. Gingles, 478 U. S. 30 (1986), prompted this Court’s first reading of §2 of the Voting Rights Act of 1965 after its 1982 amendment.⁸ Section 2(a) of the amended Act 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]. ...” Section 2(b) provides that a denial or abridgment occurs where,

“based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.” 42 U. S. C. § 1973(b).

Gingles provided some structure to the statute’s “totality of circumstances” test in a case challenging multimember legislative districts. See 478 U. S., at 46-51. The Court listed the factors put forward as relevant in the Senate Report treating the 1982 amendments,⁹ and held that

“[w]hile many or all of [them] may be relevant to a claim of vote dilution through submergence in multimember districts, unless there is a conjunction of the following circumstances, the use of multimember districts generally will not impede the ability of minority voters to elect representatives of their choice. Stated succinctly, a bloc voting majority must usually be able to defeat candidates supported by a politically cohesive, geographically insular minority group.” Id., at 48-49 (footnote omitted) (emphasis in original).

The Court thus summarized the three now-familiar Gingles factors (compactness/numerousness, minority cohesion or bloc voting, and majority bloc voting) as “necessary preconditions,” id., at 50, for establishing vote dilution by use of a multimember district.

But if Gingles so clearly identified the three as generally necessary to prove a §2 claim, it just as clearly declined to hold them sufficient in combination, either in the sense that a court’s examination of relevant circumstances was complete once the three factors were found to exist, or in the sense that the three in combination necessarily and in all circumstances demonstrated dilution. This was true not only because bloc voting was a matter of degree, with a variable legal significance depending on other facts, id., at 55-58, but also because the ultimate conclusions about equality or inequality of opportunity were intended by Congress to be judgments resting on comprehensive, not limited, canvassing of relevant facts. Lack of electoral success is evidence of vote dilution, but courts must also examine other evidence in the totality of circumstances, including the extent of the opportunities minority voters enjoy to participate in the political processes. Id., at 46, 79-80; id., at 98-99 (O’Connor, J., concurring in judgment). To be sure, some § 2 plaintiffs may have easy cases, but although lack of equal electoral opportunity may be readily imagined and unsurprising when demonstrated under circumstances that include the three essential Gingles factors, that conclusion must still be addressed explicitly, and without isolating any other arguably relevant facts from the act of judgment.¹⁰

2

If the three Gingles factors may not be isolated as sufficient, standing alone, to prove dilution in every multimember district challenge, a fortiori they must not be when the challenge goes to a series of single-member districts, where dilution may be more difficult to grasp. Plaintiffs challenging single-member districts may claim, not total submergence, but partial submergence; not the chance for some electoral success in place of none, but the chance for more success in place of some. When the question thus comes down to the reasonableness of drawing a series of district lines in one combination of places rather than another, judgments about inequality may become closer calls. As facts beyond the ambit of the three Gingles factors loom correspondingly larger, factfinders cannot rest uncritically on assumptions about the force of the Gingles factors in pointing to dilution.

The cases now before us, of course, fall on this more complex side of the divide, requiring a court to determine whether provision for somewhat fewer majority-minority districts than the number sought by the plaintiffs was dilution of the minority votes. The District Court was accordingly required to assess the probative significance of the Gingles factors critically after considering the further circumstances with arguable bearing on the issue of equal political opportunity. We think that in finding dilution here the District Court misjudged the relative importance of the Gingles factors and of historical discrimination, measured against evidence tending to show that in spite of these facts, SJR 2-G would provide minority voters with an equal measure of political and electoral opportunity.

The District Court did not, to be sure, commit the error of treating the three Gingles conditions as exhausting the, enquiry required by § 2. Consistently with Gingles, the court received evidence of racial relations outside the immediate confines of voting behavior and found a history of discrimination against Hispanic voters continuing in society generally to the present day. But the District Court was not critical enough in asking whether a history of persistent discrimination reflected in the larger society and its bloc-voting behavior portended any dilutive effect from a newly proposed districting scheme, whose pertinent features were majority-minority districts in substantial proportion to the minority’s share of voting-age population. The court failed to ask whether the totality of facts, including those pointing to proportionality,¹¹ showed that the new scheme would deny minority voters equal political opportunity.

Treating equal political opportunity as the focus of the enquiry, we do not see how these district lines, apparently providing political effectiveness in proportion to voting-age numbers, deny equal political opportunity. The récord establishes that Hispanics constitute 50 percent of the voting-age population in Dade County and under SJR 2-G would make up supermajorities in 9 of the 18 House districts located primarily within the county. Likewise, if one considers the 20 House districts located at least in part within Dade County, the record indicates that Hispanics would be an effective voting majority in 45 percent of them (i e., nine), and would constitute 47 percent of the voting-age population in the area. 815 F. Supp., at 1580; App. to Juris. Statement 180a-183a. In other words, under SJR 2-G Hispanics in the Dade County area would enjoy substantial proportionality. On this evidence, we think the State’s scheme would thwart the historical tendency to exclude Hispanics, not encourage or perpetuate it. Thus in spite of that history and its legacy, including the racial cleavages that characterize Dade County politics today, we see no grounds for holding in these cases that SJR 2-G’s district lines diluted the votes cast by Hispanic voters.

The De Grandy plaintiffs urge us to put more weight on the District Court’s findings of packing and fragmentation, allegedly accomplished by the way the State drew certain specific lines: “[T]he line of District 116 separates heavily Hispanic neighborhoods in District 112 from the rest of the heavily Hispanic Kendall Lakes area and the Kendall area,” so that the line divides “neighbors making up the . . . same housing development in Kendall Lakes,” and District 114 “packs” Hispanic voters, while Districts 102 and 109 “fragment” them. 815 F. Supp., at 1569 (internal quotation marks omitted). We would agree that where a State has split (or lumped) minority neighborhoods that would have been grouped into a single district (or spread among several) if the State had employed the same line-drawing standards in minority neighborhoods as it used elsewhere in the jurisdiction, the inconsistent treatment might be significant evidence of a §2 violation, even in the face of proportionality. The District Court, however, made no such finding. Indeed, the propositions the court recites on this point are not even phrased as factual findings, but merely as recitations of testimony offered by plaintiffs’ expert witness. While the District Court may well have credited the testimony, the court was apparently wary of adopting the witness’s conclusions as findings. But even if one imputed a greater significance to the accounts of testimony, they would boil down to findings that several of SJR 2-G’s district lines separate portions of Hispanic neighborhoods, while another district line draws several Hispanic neighborhoods into a single district. This, however, would be to say only that lines could have been drawn elsewhere, nothing more. But some dividing by district lines and combining within them is virtually inevitable and befalls any population group of substantial size. Attaching the labels “packing” and “fragmenting” to these phenomena, without more, does not make the result vote dilution when the minority group enjoys substantial proportionality.

3

It may be that the significance of the facts under § 2 was obscured by the rule of thumb apparently adopted by the District Court, that anything short of the maximum number of majority-minority districts consistent with the Gingles conditions would violate §2, at least where societal discrimination against the minority had occurred and continued to occur. But reading the first Gingles condition in effect to define dilution as a failure to maximize in the face of bloc voting (plus some other incidents of societal bias to be expected where bloc voting occurs) causes its own dangers, and they are not to be courted.

Assume a hypothetical jurisdiction of 1,000 voters divided into 10 districts of 100 each, where members of a minority group make up 40 percent of the voting population and voting is totally polarized along racial lines. With the right geographic dispersion to satisfy the compactness requirement, and with careful manipulation of district lines, the minority voters might be placed in control of as many as 7 of the 10 districts. Each such district could be drawn with at least 51 members of the minority group, and whether the remaining minority voters were added to the groupings of 51 for safety or scattered in the other three districts, minority voters would be able to elect candidates of their choice in all seven districts.¹² The point of the hypothetical is not, of course, that any given district is likely to be open to such extreme manipulation, or that bare majorities are likely to vote in full force and strictly along racial lines, but that reading § 2 to define dilution as any failure to maximize tends to obscure the very object of the statute and to run counter to its textually stated purpose. One may suspect vote dilution from political famine, but one is not entitled to suspect (much less infer) dilution from mere failure to guarantee a political feast. However prejudiced a society might be, it would be absurd to suggest that the failure of a districting scheme to provide a minority group with effective political power 75 percent above its numerical strength¹³ indicates a denial of equal participation in the political process. Failure to maximize cannot be the measure of § 2.

4

While, for obvious reasons, the State agrees that a failure to leverage minority political strength to the maximum possible point of power is not definitive of dilution in bloc-voting societies, it seeks to impart a measure of determinacy by applying a definitive rule of its own: that as a matter of law no dilution occurs whenever the percentage of single-member districts in which minority voters form an effective majority mirrors the minority voters’ percentage of the relevant population.¹⁴ Proportionality so defined, see n. 11, supra, would thus be a safe harbor for any districting scheme.

The safety would be in derogation of the statutory text and its considered purpose, however, and of the ideal that the Voting Rights Act of 1965 attempts to foster. An inflexible rule would run counter to the textual command of § 2, that the presence or absence of a violation be assessed “based on the totality of circumstances.” 42 U. S. C. § 1973(b). The need for such “totality” review springs from the demonstrated ingenuity of state and local governments in hobbling minority voting power, McCain v. Lybrand, 465 U. S. 236, 243-246 (1984), a point recognized by Congress when it amended the statute in 1982: “[S]ince the adoption of the Voting Rights Act, [some] jurisdictions have substantially moved from direct, over[t] impediments to the right to vote to more sophisticated devices that dilute minority voting strength,” Senate Report 10 (discussing §5). In modifying §2, Congress thus endorsed our view in White v. Regester, 412 U. S. 755 (1973), that “whether the political processes are ‘equally open’ depends upon a searching practical evaluation of the ‘past and present reality,’” Senate Report 30 (quoting 412 U. S., at 766, 770). In a substantial number of voting jurisdictions, that past reality has included such reprehensible practices as ballot box stuffing, outright violence, discretionary registration, property requirements, the poll tax, and the white primary; and other practices censurable when the object of their use is discriminatory, such as at-large elections, runoff requirements, anti-single-shot devices, gerrymandering, the impeachment of officeholders, the annexation or deannexation of territory, and the creation or elimination of elective offices.¹⁵ Some of those expedients could occur even in a jurisdiction with numerically demonstrable proportionality; the harbor safe for States would thus not be safe for voters.¹⁶ It is, in short, for good reason that we have been, and remain, chary of entertaining a simplification of the sort the State now urges upon us. Cf. Gingles, 478 U. S., at 77 (“[Pjersistent proportional representation ... [may] not accurately reflect the minority group’s ability to elect its preferred representatives”).

Even if the State’s safe harbor were open only in cases of alleged dilution by the manipulation of district lines, however, it would rest on an unexplored premise of highly suspect validity: that in any given voting jurisdiction (or portion of that jurisdiction under consideration), the rights of some minority voters under §2 may be traded off against the rights of other members of the same minority class. Under the State’s view, the most blatant racial gerrymandering in half of a county’s single-member districts would be irrelevant under §2 if offset by political gerrymandering in the other half, so long as proportionality was the bottom line. But see Baird v. Consolidated City of Indianapolis, 976 F. 2d 357, 359 (CA7 1992) (“A balanced bottom line does not foreclose proof of discrimination along the way”); Richmond v. United States, 422 U. S. 358, 378-379 (1975) (territorial annexation aimed at diluting black votes forbidden by § 5, regardless of its actual effect).

Finally, we reject the safe harbor rule because of a tendency the State would itself certainly condemn, a tendency to promote and perpetuate efforts to devise majority-minority districts even in circumstances where they may not be necessary to achieve equal political and electoral opportunity. Because in its simplest form the State’s rule would shield from §2 challenge a districting scheme in which the number of majority-minority districts reflected the minority’s share of the relevant population, the conclusiveness of the rule might be an irresistible inducement to create such districts. It bears recalling, however, that for all the virtues of majority-minority districts as remedial devices, they rely on a quintessentially race-conscious calculus aptly described as the “politics of second best,” see B. Grofman, L. Handley, & R. Niemi, Minority Representation and the Quest for Voting Equality 136 (1992). If the lesson of Gingles is that society’s racial and ethnic cleavages sometimes necessitate majority-minority districts to ensure equal political and electoral opportunity, that should not obscure the fact that there are communities in which minority citizens are able to form coalitions with voters from other racial and ethnic groups, having no need to be a majority within a single district in order to elect candidates of their choice. Those candidates may not represent perfection to every minority voter, but minority voters are not immune from the obligation to pull, haul, and trade to find common political ground, the virtue of which is not to be slighted in applying a statute meant to hasten the waning of racism in American politics.

It is enough to say that, while proportionality in the sense used here is obviously an indication that minority voters have an equal opportunity, in spite of racial polarization, “to participate in the political process and to elect representatives of their choice,” 42 U. S. C. § 1973(b), the degree of probative value assigned to proportionality may vary with other facts.¹⁷ No single statistic provides courts with a shortcut to determine whether a set of single-member districts unlawfully dilutes minority voting strength.

5

While the United States concedes the relevance of proportionality to a § 2 claim, it would confine proportionality to an affirmative defense, and one to be made only on a statewide basis in cases that challenge districts for electing a body with statewide jurisdiction. In this litigation, the United States would have us treat any claim that evidence of proportionality supports the State’s plan as having been waived because the State made no argument in the District Court that the proportion of districts statewide in which Hispanics constitute an effective voting majority mirrors the proportion of statewide Hispanic population.¹⁸

The argument has two flaws. There is, first, no textual reason to segregate some circumstances from the statutory totality, to be rendered insignificant unless the defendant pleads them by way of affirmative defense. Second, and just as importantly, the argument would recast these cases as they come to us, in order to bar consideration of proportionality except on statewide scope, whereas up until now the dilution claims have been litigated on a smaller geographical scale. It is, indeed, the plaintiffs themselves, including the United States, who passed up the opportunity to frame their dilution claim in statewide terms. While the United States points to language in its complaint alleging that the redistricting plans dilute the votes of “Hispanic citizens and black citizens in the State of Florida,” App. 77, the complaint identifies “several areas of the State” where such violations of § 2 are said to occur, and then speaks in terms of Hispanics in the Dade County area (and blacks in the area of Escambia County), id., at 75-76. Nowhere do the allegations indicate that claims of dilution “in the State of Florida” are not to be considered in terms of the areas specifically mentioned. The complaint alleges no facts at all about the contours, demographics, or voting patterns of any districts outside the Dade County or Escambia County areas, and neither the evidence at trial nor the opinion of the District Court addressed white bloc voting and political cohesion of minorities statewide. The De Grandy plaintiffs even voluntarily dismissed their claims of Hispanic vote dilution outside the Dade County area. See 815 F. Supp., at 1559, n. 13. Thus we have no occasion to decide which frame of reference should have been used if the parties had not apparently agreed in the District Court on the appropriate geographical scope for analyzing the alleged §2 violation and devising its remedy.

6

In sum, the District Court’s finding of dilution did not address the statutory standard of unequal political and electoral opportunity, and reflected instead a misconstruction of § 2 that equated dilution with failure to maximize the number of reasonably compact majority-minority districts. Because the ultimate finding of dilution in districting for the Florida House was based on a misreading of the governing law, we hold it to be clearly erroneous. See Gingles, 478 U. S., at 79.

IV

Having found insufficient evidence of vote dilution in the drawing of House districts in the Dade County area, we look now to the comparable districts for the state Senate. As in the case of House districts, we understand the District Court to have misapprehended the legal test for vote dilution when it found a violation of § 2 in the location of the Senate district lines. Because the court did not modify the State’s plan, however, we hold the ultimate result correct in this instance.

SJR 2-G creates 40 single-member Senate districts, 5 of them wholly within Dade County. Of these five, three have Hispanic supermajorities of at least 64 percent, and one has a clear majority of black voters. Two more Senate districts crossing county lines include substantial numbers of Dade County voters, and in one of these, black voters, although not close to a majority, are able to elect representatives of their choice with the aid of cross-over votes. 815 F. Supp., at 1574, 1579.

Within this seven-district Dade County area, both minority groups enjoy rough proportionality. The voting-age population in the seven-district area is 44.8 percent Hispanic and 15.8 percent black. Record, U. S. Exh. 7. Hispanics predominate in 42.9 percent of the districts (three out of seven), as do blacks in 14.3 percent of them (one out of seven). While these numbers indicate something just short of perfect proportionality (42.9 percent against 44.8; 14.3 percent against 15.8), the opposite is true of the five districts located wholly within Dade County.¹⁹

The District Court concentrated not on these facts but on whether additional districts could be drawn in which either Hispanics or blacks would constitute an effective majority. The court found that indeed a fourth senatorial district with a Hispanic supermajority could be drawn, or that an additional district could be created with a black majority, in each case employing reasonably compact districts. Having previously established that each minority group was politically cohesive, that each labored under a legacy of official discrimination, and that whites voted as a bloc, the District Court believed it faced “two independent, viable Section 2 claims.” 815 F. Supp., at 1577. Because the court did not, however, think it was possible to create both another Hispanic district and another black district on the same map, it concluded that no remedy for either violation was practical and, deferring to the State’s plan as a compromise policy, imposed SJR 2-G’s senatorial districts. Id., at 1580.

We affirm the District Court’s decision to leave the State’s plan for Florida State Senate districts undisturbed. As in the case of the House districts, the totality of circumstances appears not to support a finding of vote dilution here, where both minority groups constitute effective voting majorities in a number of state Senate districts substantially proportional to their share in the population, and where plaintiffs have not produced evidence otherwise indicating that under SJR 2-G voters in either minority group 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).

V

There being no violation of the Voting Rights Act shown, we have no occasion to review the District Court’s decisions going to remedy. The judgment of the District Court is accordingly affirmed in part and reversed in part.

It is so ordered.

1 The complaints also challenged Florida’s congressional districts, but that element of the litigation has been resolved separately, see De Grandy v. Wetherell, 794 F. Supp. 1076 (ND Fla. 1992) (three-judge court), and without appeal.

2 In an additional step not directly relevant to this appeal, the State submitted SJR 2-G to the Department of Justice for preclearance pursuant to 42 U. S. C. § 1973c (§5 of the Voting Rights Act of 1965). Five Florida counties, but not Dade County, are subject to preclearance. De Grandy v. Wetherell, 815 F. Supp. 1550, 1574 (ND Fla. 1992). When the Attorney General of the United States refused to preclear the plan’s Senate districts for the Hillsborough County area and the state legislature refused to revise the plan, the Supreme Court of Florida ordered the adjustments necessary to obtain preclearance, 601 So. 2d 543 (1992); it is the version of SJR 2-G so adjusted that is at issue in this litigation. 815 F. Supp., at 1557-1558.

3 The complaints also alleged violation of Art. I, § 2, and the Fourteenth and Fifteenth Amendments of the United States Constitution, but these claims were later dismissed voluntarily.

4 The Voting Rights Act of 1965 and constitutional claims as to the Escambia County area were settled by the parties and are not at issue in this appeal.

5 The court’s judgment filed July 2, 1992, App. to Juris. Statement 5a, said SJR 2-G’s state senatorial districts “do not violate Section 2,” but its subsequent opinion explaining the judgment said the senatorial districts do indeed violate §2, and that its earlier language “should be read as holding that the Florida Senate plan does not violate Section 2 such that a different remedy must be imposed.” 815 F. Supp., at 1582 (emphasis added).

Any conflict in these two formulations is of no consequence here. “This Court ‘reviews judgments, not statements in opinions,”’ California v. Rooney, 483 U. S. 307, 311 (1987) (per curiam) (quoting Black v. Cutter Laboratories, 351 U. S. 292, 297 (1956)), and the De Grandy plaintiffs and the United States have appealed the failure of the District Court to provide relief for alleged § 2 violations in SJR 2-G’s senatorial districts. The State is entitled to “urge any grounds which would lend support to the judgment below,” Dayton Bd. of Ed. v. Brinkman, 433 U. S. 406, 419 (1977), including the argument it makes here that the District Court was correct not to impose a remedy different from SJR 2-G because the State’s reapportionment plan did not violate §2.

6 The Court recognizes that the terms “black,” “Hispanic,” and “white” are neither mutually exclusive nor collectively exhaustive. We follow the practice of the District Court in using them as rough indicators of south Florida’s three largest racial and linguistic minority groups.

7 See also 478 U. S., at 50, n. 16 (discussing vote dilution through gerrymandering district lines). For earlier precedents recognizing that racial gerrymanders have played a central role in discrimination against minority groups, see Gomillion v. Lightfoot, 364 U. S. 339 (1960); Perkins v. Matthews, 400 U. S. 379 (1971); Connor v. Finch, 431 U. S. 407 (1977).

8 Congress amended the statute to reach cases in which discriminatory intent is not identified, adding new language designed to codify White v. Regester, 412 U. S. 755, 766 (1973). S. Rep. No. 97-417, p. 2 (1982) (hereinafter Senate Report).

9 As summarized in Gingles, 478 U. S., at 44-45: “The Senate Report specifies factors which typically may be relevant to a § 2 claim: the history of voting-related discrimination in the State or political subdivision; the extent to which voting in the elections of the State or political subdivision is racially polarized; the extent to which the State or political subdivision has used voting practices or procedures that tend to enhance the opportunity for discrimination against the minority group, such as unusually large election districts, majority vote requirements, and prohibitions against bullet voting; the exclusion of members of the minority group from candidate slating processes; the extent to which minority group members bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process; the use of overt or subtle racial appeals in political campaigns; and the extent to which members of the minority group have been elected to public office in the jurisdiction. [Senate Report 28-29.] The Report notes also that evidence demonstrating that elected officials are unresponsive to the particularized needs of the members of the minority group and that the policy underlying the State’s or the political subdivision's use of the contested practice or structure is tenuous may have probative value. Id., at 29.”

10 If challenges to multimember districts are likely to be the easier plaintiffs’ cases, it is worth remembering that even in multimember district challenges, proof of the Gingles factors has not always portended liability under §2. In Baird v. Consolidated City of Indianapolis, 976 F. 2d 357 (1992), the Seventh Circuit confronted a scheme for electing a City-County Council of 29 members. Voters chose 25 of their representatives from single-member districts and 4 at large, from a district representing the entire area. Black plaintiffs brought a vote dilution claim challenging the lines for single-member districts and the existence of the four-member at-large district. After the Council had redrawn its single-member districts to rectify dilution there, the District Court held, and the Seventh Circuit affirmed, that the four-member district did not dilute black voting strength because proof of the three Gingles factors was not enough “if other considerations show that the minority has an undiminished right to participate in the political process.” 976 F. 2d, at 359. The “other considerations” in Baird included the fact that the new single-member districts were so drawn that blacks formed a voting majority in seven of them (28 percent of the single-member districts and 24 percent of the entire council) while blacks constituted 21 percent of the local population; and that while the four at-large seats tended to go to Republicans, one of the Republicans elected in 1991 was black. Id., at 358, 361.

11 “Proportionality” as the term is used here links the number of majority-minority voting districts to minority members’ share of the relevant population. The concept is distinct from the subject of the proportional representation clause of §2, which 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.” 42 U. S. C. § 1973(b). This proviso speaks to the success of minority candidates, as distinct from the political or electoral power of minority voters. Of. Senate Report 29, n. 115 (minority candidates’ success at the polls is not conclusive proof of minority voters’ access to the political process). And the proviso also confirms what is otherwise clear from the text of the statute, namely, that the ultimate right of §2 is equality of opportunity, not a guarantee of electoral success for minority-preferred candidates of whatever race.

12 Minority voters might instead be denied control over a single seat, of course. Each district would need to include merely 51 members of the majority group; minority voters fragmented among the 10 districts could be denied power to affect the result in any district.

13 When 40 percent of the population determines electoral outcomes in 7 out of 10 districts, the minority group can be said to enjoy effective political power 75 percent above its numerical strength.

14 See Brief for Appellees in Nos. 92-593, 92-767, p. 20 (“If the statutory prohibition against providing minorities ‘less opportunity than other members of the electorate ... to elect representatives of their choice’ is given its natural meaning, it cannot be violated by a single-member district plan that assures minority groups voting control over numbers of districts that are numerically proportional to their population in the area where presence of the three Gingles preconditions has been established”).

The parties dispute whether the relevant figure is the minority group’s share of the population, or of some subset of the population, such as those who are eligible to vote, in that they are United States citizens, over 18 years of age, and not registered at another address (as students and members of the military often are). Because we do not elevate this proportion to the status of a magic parameter, and because it is not dispositive here, we do not resolve that dispute. See supra, at 1008-1009.

15 See generally J. M. Kousser, The Shaping of Southern Politics: Suffrage Restriction and the Establishment of the One-Party South, 1880-1910 (1974); Kousser, The Undermining of the First Reconstruction, Lessons for the Second, in Minority Vote Dilution 27 (C. Davidson ed. 1984); Hearings on the Extension of the Voting Rights Act before the Subcommittee on Civil and Constitutional Rights of the House Committee on the Judiciary, 97th Cong., 1st Sess., 1999-2022, 2115-2120 (1981).

16 The State might say, of course, that ostensibly “proportional” districting schemes that were nonetheless subject to diluting practices would not “assur[e]” minority voters their apparent voting power. But this answer would take us right back to a searching review of the factual totality, leaving the State’s defensive rule without any particular utility.

17 So, too, the degree of probative value assigned to disproportionality, in a case where it is shown, will vary not only with the degree of disproportionality but with other factors as well. “[TJhere is no indication that Congress intended to mandate a single, universally applicable standard for measuring undiluted minority voting strength, regardless of local conditions and regardless of the extent of past discrimination against minority voters in a particular State or political subdivision.” Gingles, 478 U. S., at 94-95 (O’Connor, J., concurring in judgment).

18 The argument for proportionality statewide favors the State if it is based on the proportion of Hispanic citizens of voting age statewide. According to census data not available at the time of trial and thus not in the record, Hispanics constitute 7.15 percent of the citizen voting-age population of Florida, which corresponds to eight or nine Hispanic-majority House districts (120 x 7.15% = 8.58).

If instead one calculates the proportion of statewide Hispanic-majority House districts on the basis of total population or voting-age population, the result favors plaintiffs. Hispanics constitute 12.2 percent of the State’s total population and 11.7 percent of the State’s voting-age population, corresponding to 14 or 15 seats (120 X 12.2% = 14.64; 120 X 11.7% = 14.04). We need not choose among these calculations to decide these cases.

19 In the five districts wholly within Dade County, where Hispanics are concentrated, the voting-age population is 53.9 percent Hispanic and 13.5 percent black. Sixty percent of the districts are Hispanic majority (three out of five), and 20 percent are black majority (one out of five), so that each minority group protected by §2 enjoys an effective voting majority in marginally more districts than proportionality would indicate (60 percent over 53.9; 20 percent over 13.5).