Since 1969, Charleston County, South Carolina has been governed by a County Council composed of nine members elected in county-wide, partisan elections. Despite the County's substantial minority population, few minority-preferred candidates, and very few minority candidates, have ever been elected to the Council. The United States brought this suit, alleging that the County's at-large election of its Council diluted minority voting strength in violation of Section 2 of the Voting Rights Act of 1965. The district court agreed, finding that the County's severe voting polarization, its particular electoral structure, and its sheer size combined to deny minority voters an equal opportunity to elect their preferred representatives. Because the district court's finding is not clearly erroneous, we affirm.

Located in the southeastern corner of South Carolina, where the Ashley and Cooper Rivers converge on the Atlantic, Charleston County covers over nine hundred square miles. It includes tiny municipalities like Awendaw and McClellanville; islands like Kiawah and Seabrook; and of course cities like Charleston, the state's second largest. The County is ethnically as well as geographically diverse. The third most populous of the state's forty-six counties, it has the second highest total number of black residents. Of its roughly 310,000 residents, 188,542 (60.8%) are white; 106,337 (34.3%) are black; and 15,090 (4.9%) are of other racial or ethnic descent.1

The numbers of registered and actual voters, however, are more disparate. As of November 2000, 177,279 people were registered to vote in Charleston County, 122,557 (69.1%) of whom were white and 54,722 (30.9%) of whom were nonwhite. Of those registered voters, 114,166 actually voted in the November 2000 general election, 82,395 (72.2%) of whom were white and 31,771 (27.8%) of whom were nonwhite. According to the evidence of voter turnout presented by the County's own expert witness, Dr. Ronald Weber, minority registered voters have consistently participated at a lower rate than white registered voters in Charleston County Council general elections.

The County Council oversees local governance on issues ranging from economic development to public safety, and it is composed of nine members elected to staggered terms in at-large, partisan elections. Candidates for the Council run from four residency districts: three Council seats are reserved for residents of the City of Charleston, three for residents of North Charleston, two for residents of West Ashley, and one for a resident of East Cooper. Charleston County is one of only three counties in South Carolina that elects its entire county council at-large, and it is the only county with a majority white population to do so. The other two counties that elect their county councils at-large, Hampton and Jasper, are less populated, rural counties with roughly equal numbers of minority and white residents.

The County's modified at-large system, in which all of the County's residents may vote for candidates residing in specific areas of the County, was created in 1969, and it was precleared by the Attorney General under Section 5 of the Voting Rights Act. See 42 U.S.C. § 1973c (2003). In 1989, the County's residents narrowly rejected a referendum to switch from the at-large electoral system to a single-member district system. Both the County and the United States agree that voting on the referendum was extremely polarized: at least 98% of minority voters approved the switch to single-member districts, while at least 75% of white voters wanted to retain at-large elections.

Since 1970, 41 people have been elected to the County Council, only three of whom are minorities: Lonnie Hamilton III, who was elected six times between 1970 and 1990, serving a total of twenty-four years; Marjorie Amos-Frazier, who was elected twice in 1974 and 1978; and Timothy Scott, who has remained the Council's only minority member since 1995. While minority candidates preferred by minority voters have had great difficulty winning election to the Council, white candidates who were preferred by minority voters have been somewhat more successful, according to evidence of recent Council elections presented by the United States' expert witness, Dr. Theodore Arrington.

Section 2(a) of the Voting Rights Act prohibits a State or its political subdivision from imposing any voting practice that "results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color...." 42 U.S.C. § 1973(a). Section 2(b), as amended in 1982, further provides that a violation of § 2(a) occurs if, based on the totality of the 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 [§ 2(a)] 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.

Id. § 1973(b). The 1982 amendment made clear that Section 2 condemns not only voting practices borne of a discriminatory intent, but also voting practices that "operate, designedly or otherwise," to deny "equal access to any phase of the electoral process for minority group members." S.Rep. No. 97-417, at 28, 30 (1982), reprinted in 1982 U.S.C.C.A.N. 177, 205, 207 (hereinafter "Senate Report"); see also Chisom v. Roemer, 501 U.S. 380, 393-95 & nn. 20-21, 111 S. Ct. 2354, 115 L. Ed. 2d 348 (1991).

In Thornburg v. Gingles, 478 U.S. 30, 106 S. Ct. 2752, 92 L. Ed. 2d 25 (1986), the Supreme Court established the framework for claims that an at-large voting system dilutes minority voting strength in violation of § 2 of the Voting Rights Act. According to the Court, three preconditions are necessary to a finding of vote dilution:

First, the minority group must be able to demonstrate that it is sufficiently large and geographically compact to constitute a majority in a single member district....

Second, the minority group must be able to show that it is politically cohesive....

Third, the minority must be able to demonstrate that the white majority votes sufficiently as a bloc to enable it ... usually to defeat the minority's preferred candidate.

Gingles, 478 U.S. at 50-51, 106 S. Ct. 2752 (citations and footnotes omitted). If these three preconditions are satisfied, then the trier of fact must determine whether, based on the totality of the circumstances, there has been a violation of Section 2. Johnson v. DeGrandy, 512 U.S. 997, 1011-12, 114 S. Ct. 2647, 129 L. Ed. 2d 775 (1994); Cane v. Worcester County, 35 F.3d 921, 925 (4th Cir.1994).

In determining which circumstances courts should take care to consider, the Supreme Court has turned for guidance to the Senate Report that accompanied Congress's 1982 amendments to the Voting Rights Act. See Gingles, 478 U.S. at 43-45, 106 S. Ct. 2752. According to Congress and the Court, the most important factors in the inquiry into the totality of the circumstances are the "extent to which minority group members have been elected to public office in the jurisdiction," and "the extent to which voting in the elections of the state or political subdivision is racially polarized." Id. at 48-49 n. 15, 106 S. Ct. 2752 (quoting Senate Report at 28-29) (internal quotation omitted). If these factors are present, other considerations "are supportive of, but not essential to" a § 2 claim: the history of voting-related discrimination in the State or political subdivision; the extent to which the State or political subdivision has used voting practices that enhance the opportunity for discrimination against the minority group, such as unusually large election districts or anti-single shot provisions; the exclusion of minorities from the candidate slating process; the use of racial appeals in political campaigns; and the degree to which past discrimination in such areas as education, employment and health has hindered the ability of minorities to participate effectively in the political process. Id. (emphasis in original).

However, this list is "neither comprehensive nor exclusive." Gingles, 478 U.S. at 45, 106 S. Ct. 2752; see Senate Report at 29 ("[The Senate Report's factors] will often be the most relevant ones, [but] in some cases other factors will be indicative of the alleged dilution."). Congress intended "the ultimate conclusions about equality or inequality of opportunity ... to be judgments resting on comprehensive, not limited, canvassing of relevant facts." DeGrandy, 512 U.S. at 1011, 114 S. Ct. 2647.

Before the district court, Charleston County's at-large method for electing its County Council was challenged as a violation of § 2 by both the United States and a group of Charleston County voters. The United States and the private plaintiffs claimed that the County's at-large method resulted in the unlawful dilution of minority voting strength. In addition, the private plaintiffs — but not the United States — claimed that the County's at-large system intentionally discriminated against minority voters. Both the United States and the private plaintiffs moved for partial summary judgment as to the three Gingles preconditions necessary to establish a § 2 violation.

On April 26, 2002, based on the reports and testimony of both Dr. Arrington, the United States' expert; and Dr. Weber, the County's expert, the magistrate judge recommended that the plaintiffs' motion be granted. On the first Gingles precondition — numerosity and compactness — the magistrate judge found that the minority population of Charleston County is sufficiently large and geographically compact to constitute a majority in a single-member district. On the second Gingles precondition — minority cohesion — the magistrate judge reviewed evidence presented by both parties indicating that minority voters in Charleston County are politically cohesive. For instance, according to Dr. Weber, minority voters have been politically cohesive in 28 of 33(85%) contested Council elections since 1988.

Finally, on the third Gingles precondition — majority bloc voting — the magistrate judge again reviewed evidence from both parties that minority voters' preferred Council candidates are usually defeated by cohesive white voting. Pursuant to our decision in Lewis v. Alamance County, 99 F.3d 600, 615-16 n. 12 (4th Cir.1996), the magistrate judge declined to consider Charleston County's argument that minority-preferred candidates were usually defeated on account of partisanship, not race. According to the magistrate judge, Alamance County renders causation irrelevant to the third Gingles pre-condition: the question for purposes of the third Gingles precondition is whether whites vote as a bloc, and not why they do so. Under Alamance County, according to the magistrate judge, causation is instead relevant to the inquiry into the totality of the circumstances, once the Gingles preconditions have been satisfied.

On July 10, 2002, the United States District Court for the District of South Carolina adopted the magistrate's report, and granted summary judgment to the United States and the private plaintiffs as to the three Gingles preconditions. The district court then held a trial on the remaining issue: whether, based on the totality of the circumstances, the County's at-large electoral method violated § 2. On March 6, 2003, the district court agreed with the United States that the County's at-large system diluted minority voting strength, depriving minority voters of an equal opportunity to participate in the political process and elect representatives of their choice. However, the district court rejected the private plaintiffs' claim that the County's at-large system was adopted with the intent to discriminate against minority voters.

Charleston County now appeals the district court's decision.2 The County does not challenge the district court's finding that the United States had established the first two Gingles preconditions: the County concedes not only that its minority voters are sufficiently numerous and geographically compact to constitute a majority in a single-member district, but also that they are politically cohesive. The County, however, does contend that the district court erred in granting summary judgment to the plaintiffs on the third Gingles precondition — the presence of white racial bloc voting. The County also contends that, even if the three Gingles preconditions were satisfied, the district court clearly erred in determining that the at-large voting scheme diluted minority voting strength in violation of § 2. We address the County's claims in turn.

The crux of the County's argument, from the outset of this litigation, has been that voting in Charleston County is polarized as a result of partisanship rather than race. According to the County, whites vote as a bloc — but for Republican candidates, whatever their race — and blacks vote as a bloc — but for Democratic candidates, whatever their race. In the County's view, the third Gingles precondition requires more than a showing that white bloc voting usually defeats the minority-preferred candidate. In order to demonstrate Gingles' "legally significant" white bloc voting, the County claims, the United States must prove that race rather than partisanship is the cause of the polarized voting. See 478 U.S. at 56, 106 S. Ct. 2752. For its part, the United States responds that evidence of partisanship as the cause of the racially divergent voting should be considered in the totality of the circumstances inquiry, after the Gingles preconditions have been satisfied.

The parties thus rightly agree that an inquiry into causation is relevant. See Alamance County, 99 F.3d at 615-16 n. 12; see also Gingles, 478 U.S. at 100, 106 S. Ct. 2752 (O'Connor, J., concurring in judgment) ("Evidence that a candidate preferred by the minority group in a particular election was rejected by white voters for reasons other than those which made that candidate the preferred choice of the minority group would seem clearly relevant in answering the question whether bloc voting by white voters will consistently defeat minority candidates."); Uno v. City of Holyoke, 72 F.3d 973, 982 (1st Cir.1995) ("[Section 2] does not require courts to ignore evidence that factors other than race are the real obstacles to the political success of a minority group."); League of United Latin Am. Citizens (LULAC) v. Clements, 999 F.2d 831, 850 (5th Cir.1993) (en banc) ("[The] rigorous protections [of the Voting Rights Act], as the text of § 2 suggests, extend only to defeats experienced by voters `on account of race or color.'"). The parties simply disagree over the stage in the vote dilution inquiry at which causation evidence is appropriate.

As we explained in Alamance County, the approach most faithful to the Supreme Court's case law "is one that treats causation as irrelevant in the inquiry into the three Gingles preconditions, but relevant in the totality of circumstances inquiry." Alamance County, 99 F.3d at 615-16 n. 12 (citations omitted); see also Gingles, 478 U.S. at 100, 106 S. Ct. 2752 (O'Connor, J., concurring in judgment) (explaining that causation evidence should be part of "the overall vote dilution inquiry"). The inquiry into the Gingles preconditions is a preliminary one, designed to determine whether an at-large system potentially violates § 2. See Growe v. Emison, 507 U.S. 25, 40-41, 113 S. Ct. 1075, 122 L. Ed. 2d 388 (1993). An at-large system cannot be responsible for diluting minority voting strength unless minority voters cohesively support particular candidates, the minority-preferred candidates are being systematically defeated by white bloc voting, and those defeats would not be occurring under a system of single-member districts. See id.

But simply clearing the Gingles hurdles, while necessary to prove a possible violation of § 2, is not sufficient to establish an actual violation. DeGrandy, 512 U.S. at 1011-12, 114 S. Ct. 2647. To demonstrate an actual violation, a plaintiff asserting vote dilution must show "that, under the totality of the circumstances, the State's [challenged electoral scheme] has the effect of diminishing or abridging the voting strength of the protected class." Voinovich v. Quilter, 507 U.S. 146, 157, 113 S. Ct. 1149, 122 L. Ed. 2d 500 (1993). Accordingly, a court must undertake "a searching practical evaluation of the past and present reality," Gingles, 478 U.S. at 45, 106 S. Ct. 2752 (quoting Senate Report at 30) (internal quotation omitted), which demands a "comprehensive, not limited, canvassing of relevant facts," DeGrandy, 512 U.S. at 1011, 114 S. Ct. 2647. It is this inclusive examination of the totality of the circumstances that is tailor-made for considering why voting patterns differ along racial lines.

By expanding the inquiry into the third Gingles precondition to ask not merely whether, but also why, voters are racially polarized, the County would convert the threshold test into precisely the wide-ranging, fact-intensive examination it is meant to precede. We have rejected this approach, Alamance County, 99 F.3d at 615-16 n. 12, as has the majority of our sister circuits, compare Goosby v. Town Bd., 180 F.3d 476, 493 (2d Cir.1999) (treating causation under the totality of circumstances analysis rather than the third Gingles precondition); Milwaukee Branch of the N.A.A.C.P. v. Thompson, 116 F.3d 1194, 1199 (7th Cir.1997); Sanchez v. Colorado, 97 F.3d 1303, 1313 (10th Cir.1996); Uno, 72 F.3d at 980-81; Nipper v. Smith, 39 F.3d 1494, 1524-25 & n. 60 (11th Cir.1994) (en banc), with LULAC, 999 F.2d at 891-92 (finding third Gingles precondition unsatisfied because "partisan affiliation, not race, caused the defeat of the minority-preferred candidate").

The County also misfires when it argues that by referring to the third Gingles precondition as "legally significant" white bloc voting, what the Gingles Court meant was white bloc voting that occurs on account of racial animus or bias. Gingles, 478 U.S. at 56, 106 S. Ct. 2752. Rather, what the Gingles Court meant was the degree of racially polarized voting that matters in the context of § 2. According to the Court, for racially polarized voting to be "legally significant," minority voters must "usually" vote for the same candidates, and white bloc voting must "normally" or "generally" lead to the defeat of minority-preferred candidates. Id. "Legally significant" white bloc voting thus refers to the frequency with which, and not the reason why, whites vote cohesively for candidates who are not backed by minority voters. The County does not even attempt to argue that its racially polarized voting is legally insignificant in this sense. Indeed, the County's own expert testified that minority-preferred candidates are usually defeated by white bloc voting. The district court therefore properly found the third Gingles precondition satisfied.

The County next argues that, even if the Gingles preconditions were satisfied, the district court clearly erred in determining that the County's at-large voting scheme violated § 2. According to the County, the district court improperly discounted substantial evidence that party affiliation, not racial animus, drives voting in the County, which has an increasingly strong Republican base. In the County's view, even if the district court was right not to consider this evidence of partisanship in connection with the third Gingles precondition, it still should have been swayed by this same evidence when it assessed the totality of the circumstances.

Certainly the reason for polarized voting is a critical factor in the totality analysis and Charleston County has presented evidence of partisanship. For instance, since 1995 Timothy Scott has repeatedly been elected as a minority Republican (Scott recently finished his term as Chairman of the Council), with overwhelming support from white voters and only minimal support from minority voters. Although Scott is the Council's only minority member in the last decade, the County claims that Scott's success shows minorities can win election to the Council, so long as they share the political philosophy that prevails among the majority of Charleston County's residents. In addition, two white Council members, Dr. Charles Wallace and Charlie Lybrand, ran as minority-preferred Democrats and lost, only to switch parties and win as Republicans with little minority support. According to the County, all of this demonstrates that minority-preferred candidates have suffered electoral defeat on account of partisan politics, not race or color.

We need not inquire whether we would have been persuaded by the County's evidence in the first instance, for our function is not to reweigh the evidence presented to the district court. Claims of vote dilution require trial courts to immerse themselves in the facts of each case, and to engage in "an intensely local appraisal of the design and impact of the contested electoral mechanisms." Gingles, 478 U.S. at 79, 106 S. Ct. 2752 (internal quotation omitted); see also Goosby, 180 F.3d at 492 ("[R]esolution of the question of vote dilution is a fact intensive enterprise to be undertaken by the district court."). The Supreme Court has been explicit that in order to "preserve[] the benefit of the trial court's particular familiarity with the indigenous political reality," we may set aside a trial court's finding of vote dilution only if it is clearly erroneous. Gingles, 478 U.S. at 79, 106 S. Ct. 2752; see also Cane v. Worcester County, 35 F.3d 921, 925 (4th Cir.1994). While the district court's finding that the County's at-large voting scheme violated § 2 is certainly disputable, it is not clearly mistaken. Rather, the district court's finding rested on substantial, credible evidence, much of it presented by the County's own expert, Dr. Weber.

First, both Congress and the Supreme Court have made clear that among the most important factors in assessing a vote dilution claim are "the extent to which voting in the elections of the state or political subdivision is racially polarized," and "the extent to which minority group members have been elected to public office in the jurisdiction." Gingles, 478 U.S. at 48-49 n. 15, 106 S. Ct. 2752 (quoting Senate Report at 28-29). Neither of these factors was in any serious dispute before the district court. The United States presented uncontroverted evidence of racial polarization and minimal minority electoral success.

According to Dr. Weber, there was racially polarized voting in 25 of the 33 (75.8%) contested general elections for the County Council between 1988 and 2000. In the 10 general elections that involved at least one minority candidate, Dr. Weber found that white and minority voters were polarized 100% of the time. See Alamance County, 99 F.3d at 610 n. 8 (suggesting that elections involving minority candidates may be more probative "on the question of whether racial polarization exists"); accord Ruiz v. City of Santa Maria, 160 F.3d 543, 552-53 (9th Cir.1998); Uno, 72 F.3d at 988 n. 8; N.A.A.C.P. v. City of Niagara Falls, 65 F.3d 1002, 1016-18 (2d Cir.1995); Nipper, 39 F.3d at 1540; Jenkins v. Red Clay Consol. Sch. Dist. Bd. of Educ., 4 F.3d 1103, 1128 (3d Cir.1993); LULAC, 999 F.2d at 864.

Notably, the United States' expert, Dr. Arrington, found an even higher overall rate of racially polarized voting in contested Council elections between 1984 and 2000 (94% to Weber's 75.8%). But whatever the actual level of polarization, evidence presented by both parties supported the district court's conclusion "that voting in Charleston County Council elections is severely and characteristically polarized along racial lines." Indeed, Charleston County voters are racially divided not only over the desirability of particular candidates, but over the desirability of the at-large system itself. In 1989 the County held a referendum on switching from the at-large system for County Council elections to a system of single-member districts: at least 98% of minority voters endorsed the change, while at least 75% of white voters wanted to retain at-large elections.

As for minority electoral success, of the 41 people elected to the nine-member Council since 1970, only 3 have been minorities: Lonnie Hamilton III, Marjorie Amos-Frazier, and Timothy Scott. Of those only Scott has served on the Council in the last decade. As the district court recognized, the electoral successes of Hamilton and Amos-Frazier in the 1970s and 1980s were of marginal relevance to whether minorities currently enjoy equal access to the electoral process. In fact, Dr. Weber declined to consider pre-1988 elections, recognizing that recent elections are the most probative in determining vote dilution. See, e.g., Ruiz, 160 F.3d at 555; Uno, 72 F.3d at 990; Meek v. Metropolitan Dade County, 985 F.2d 1471, 1482-83 (11th Cir.1993). And the rarity with which minorities are elected is not unique to the County Council: disproportionately few minorities have ever won any of the at-large elections in Charleston County.

However, the County repeatedly asserts that the district court mistakenly considered only minority candidates' success, rather than minority-preferred candidates' success, as part of the totality of the circumstances. The County is certainly right that"[t]he extent to which members of a protected class have been elected to office" is but "one circumstance which may be considered" in assessing whether minority voters have been denied an equal opportunity "to participate in the political process and to elect representatives of their choice." 42 U.S.C. § 1973(b); see also id. ("[N]othing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.").

Yet the district court did not treat the lack of minority electoral success as conclusive proof of vote dilution. Rather, pursuant to the clear command of Congress and the Supreme Court, the district court treated minimal minority electoral success as but one factor in the totality of the circumstances inquiry. See Gingles, 478 U.S. at 48-49 n. 15, 106 S. Ct. 2752. The district court recognized that the number of minorities elected to office, while relevant to vote dilution, was not dispositive, and it went on to analyze a host of other factors that weighed in its decision. The court was aware that the ultimate question remained whether minority voters were able to elect their preferred candidates, whatever the candidates' race. See Alamance County, 99 F.3d at 606-07.

Neither did the district court overlook the success of minority-preferred candidates in assessing the totality of the circumstances. According to the County, the district court ignored dozens of Council elections during the past thirty-plus years in which minority-preferred candidates prevailed. As an initial matter, the County overstates its case: two-thirds of the elections to which the County points occurred prior to 1988, and yet Dr. Weber did not consider pre-1988 Council elections. In other words, the County faults the district court for not examining election results that the County's own expert viewed as marginally relevant.

As for the post-1988 elections, the County is correct to note that minority-preferred candidates met with some limited success, which is surely relevant to the inquiry into the totality of the circumstances. But Dr. Weber himself testified that minority-preferred candidates, whatever their race, generally were defeated by white bloc voting. Dr. Weber's testimony, along with other evidence that we discuss below, permitted the district court to fairly conclude that the at-large system was diluting minority vote strength, even if minority-preferred candidates still occasionally managed to prevail. And it is not as if the district court ignored altogether the post-1988 elections. Rather, the district court discussed at length Dr. Weber's and Dr. Arrington's reports, which analyzed the very post-1988 elections that the district court is supposed to have neglected. The County cannot credibly claim that the district court's focus was too narrow, or its analysis too slipshod.

Second, the district court found that County Council's electoral structure, along with the County's sheer size, diminished minority voters' ability to elect their preferred representatives. As we explained earlier, candidates for the County Council must qualify from one of four residency districts, and they are then elected in partisan contests to staggered four-year terms. Both the parties' experts agree that the County's use of staggered terms, residency districts, and a primary nominating system makes it more difficult for minority-preferred candidates to prevail. The residency districts and staggered terms confine County Council elections to either single-seat or two-seat contests, and the primary nominating system produces only two viable candidates for each seat. As even the County's expert testified, this creates a de facto majority vote requirement, and limits the opportunity for minority voters to engage in single-shot voting.3

Further, Charleston County is the largest county in South Carolina, with a nearly 100-mile stretch of coastline along the Atlantic. Several witnesses testified that the County's size works to the detriment of minority candidates, who typically have fewer financial resources, in particular because costly television advertising and direct mail have proven important in recent Council elections. See Goosby, 180 F.3d at 494 ("Campaign financing is especially difficult in such a large district for black candidates, who have been able to campaign more effectively in smaller districts.").

Third, although the County argued that partisanship rather than race drives the County's racially polarized voting patterns, there was no systematic proof to support its claim. Dr. Weber, the County's expert, acknowledged that he could not assess the extent to which racial bias has caused polarized voting in Charleston County, and he agreed with other expert witnesses that partisanship and race as determinants of voting are "inextricably intertwined." Even assuming that the effects of partisanship and race on voting could have been isolated and measured, no such evidence was before the district court. As both parties' experts testified, and as the district court explained, party registration information and survey research are the primary data relied on by political scientists in determining the effect of political partisanship on electoral outcomes. Neither datum is available for Charleston County Council elections, because the County does not require that voters register by party for general elections and neither the County nor the United States has conducted any survey research.

The County did present anecdotal evidence of partisanship, such as that some candidates (Wallace and Lybrand) had switched parties and won election to the Council, or that a minority Republican (Scott) has been elected to Council several times. The County also presented evidence of Charleston County School Board elections, which are non-partisan and which the County claims are less severely polarized along racial lines. However, the district court thoroughly examined all of the County's evidence, and deemed it insufficiently comprehensive or persuasive. For instance, the district court found the evidence pertaining to the School Board elections of questionable value, in part because the parties' experts disagreed over whether the School Board elections were even less racially polarized. But even to the extent that minority-preferred candidates fared better in School Board than in Council elections, the district court recognized that this was due at least in part to special circumstances like minority candidates' running unopposed, single-shot voting, or a split in the white vote among several white candidates. Likewise, the district court struggled in weighing the County's evidence of party-switching, because the court could not determine whether minority voters abandoned their support of Wallace and Lybrand solely for partisan reasons. In the end, the district court was faced with inconclusive evidence of partisanship as the determinant of voting, but "decisive[]" evidence of "severe voting polarization, minimal minority electoral success, and an uncommonly large voting district." The County's evidence of partisanship in this case was also far from persuasive on its own terms. Dr. Weber conceded that minority preferred minority candidates are defeated more often than minority-preferred white candidates. To be more precise, looking at County Council elections since the early 1990s, white Democrats have at least occasionally won, while minority Democrats have invariably lost. Although minority voters give more cohesive support to minority Democratic candidates than to white Democratic candidates, the opposite is true among white voters. This is consistent with the parties' evidence that white and minority voters are more often racially polarized in Council general elections involving at least one minority candidate. Thus even controlling for partisanship in Council elections, race still appears to play a role in the voting patterns of white and minority voters in Charleston County. Or at least it was not clearly erroneous for the district court to so conclude.

The result here is required by the framework Congress established for vote-dilution claims in § 2; the proof scheme the Supreme Court set forth for such actions in Gingles; and, most significantly, by the findings of fact the district court made in this case, findings that were not clearly erroneous.

We need not decide whether any of the factors on which the district court relied — the County's severe voting polarization and minimal minority electoral success, its hybrid electoral structure, or its sheer size — would have been enough in isolation to prove a violation of § 2. See Gingles, 478 U.S. at 45, 106 S. Ct. 2752 ("[T]here is no requirement that any particular number of factors be proved, or that a majority of them point one way or the other.") (quoting Senate Report at 29). Taken in combination, these factors were sufficient to prove a § 2 violation.4 Indeed, the County's own expert appeared to support, rather than undermine, the district court's conclusions. Based on evidence submitted by all parties, the district court conducted a "searching practical evaluation" of local electoral conditions in the County, and its conclusion that Charleston County's at-large system violated § 2 of the Voting Rights Act is not clearly erroneous. Id. (quoting Senate Report at 30). The judgment of the district court is therefore

AFFIRMED.