No. 96-1695B IN THE SUPREME COURT OF THE UNITED STATES OCTOBER TERM, 1996 ATTALA COUNTY, ET AL., PETITIONER v. BERNARD TEAGUE ,ET AL. ON PETITION FOR A WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT BRIEF FOR THE UNITED STATES IN OPPOSITION WALTER DELLINGER Acting Solicitor General ISABELLE KATZ PINZLER Acting Assistant Attorney General MARK L. GROSS MICHELLE ARONOWITZ Attorneys Department of Justice Washington, D.C. 20530-0001 (202) 514-2217 ---------------------------------------- Page Break ---------------------------------------- QUESTIONS PRESENTED 1. Whether the court of appeals held that proof of the three Gingles factors in a case under Section 2 of the Voting Rights Act. establishes vote dilution. 2. Whether the court of appeals misapplied the clearly erroneous standard in rejecting the district court's finding that blacks and whites prefer different candidates for reasons other than race. (I) ---------------------------------------- Page Break ---------------------------------------- TABLE OF CONTENTS Page Opinions below . . . . 1 Jurisdiction . . . . 1 Statement . . . . 2 Argument . . . . 10 Conclusion . . . . 20 TABLE OF AUTHORITIES Cases: Abrams v. Johnson, N0. 95-1425, 1997 WL 331,802 (June 19, 1997) . . . . 10-11,19 Anderson v. City of Bessemer City, 470 U.S. 564(1985) . . . . 15 Clark v. Calhoun County, 21 F.3d 92(5th Cir. 1994) . . . . 12 Johnson v. De Grandy, 512 U.S. 997 (1994) . . . . 10, 13 League of United Latin American Citizens v. Clements, 999 F.2d 831(5th Cir.1993), cert. denied, 510 U.S. 1071 (1994) . . . . 16 Nipper v. Smith, 39 F.3d 1494 (llth Cir. 1994), cert. denied. l15 S. Ct. 1795 (1995) . . . . 8, 15 Thornburg v. Gingles, 478 U.S. 30 (1986) . . . . 2, 8, 10, 13, 14,15,20 Statutes: Voting Rights Act of 1965 2, 42 U.S.C. 1973 . . . . 2,4,6, 10, 11, 13, 14 (III) ---------------------------------------- Page Break ---------------------------------------- In the Supreme Court of the United States OCTOBER TERM, 1996 No. 96-1695 ATTALA COUNTY, ET AL., PETITIONERS v. BERNARD TEAGUE, ET AL. ON PETITION FOR A WRIT OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT BRIEF FOR THE UNITED STATES IN OPPOSITION OPINIONS BELOW The opinion of the court of appeals (Pet. App. 58-85) is reported at 92 F.3d 283. The district court's opin- ion after remand (Pet. App. 32-57) is unreported. The opinion of the court of appeals (Pet. App. 26-31) re- manding to the district court is reported at 17 F.3d 796. The initial opinion of the district court (Pet. App. 1-25) is reported at 807 F. Supp. 392. JURISDICTION The judgment of the court of appeals was entered on August 8, 1996. A petition for rehearing was denied on January 22, 1997. Pet. App. 86. The petition for a writ of certiorari was filed on April 22, 1997. The (1) ---------------------------------------- Page Break ---------------------------------------- 2 jurisdiction of this Court is invoked under 28 U.S.C. 1254(1). STATEMENT Respondents challenge the redistricting plans of Attala County, Mississippi, for the election of its five county supervisors and five election commissioners, and of its two county constables and two justice court judges, alleging that the plans dilute minority voting strength in violation of Section 2 of the Voting Rights Act of 1965, 42 U.S.C. 1973. Each candidate is elected from a single-member district by majority vote. Although African-Americans comprise 39.5% of the population in Attala County, and 35.4% of the voting age population, no black candidate has ever won a contested election from a white majority district, despite a number of attempts. Pet. App. 8, 35, 39, 41, 55 & n.16, 59. African-Americans are a minority in four of five supervisor-election commissioner dis- tricts and in both constable-justice court districts. Pet. App. 60. 1. In its first decision, the district court found that respondents failed to prove a Section 2 violation. The district court found that respondents met, the first Thornburg v. Gingles, 478 U.S. 30 (1986), precondition in that African-Americans are sufficiently numerous and geographically compact to constitute a majority in two of five supervisor districts and one of two constable districts. Pet. App. 16-17. But it found that respondents had not met the second and third Gingles preconditions of black political cohesion and white bloc voting. The district court rejected respondents' statistical analysis of black-white elections from 1979) to 1991 in favor of election returns from two precincts in two elections which it found showed some cross- ---------------------------------------- Page Break ---------------------------------------- 3 over voting, and anecdotal testimony of one politician that race is not a factor in county elections. Id. at 17- 19. The district court found respondents' ecological regression analysis "nondemonstrative of a minority vote dilution claim," id. at 22, and rejected respon- dents' extreme case analysis because it analyzed only white voter preferences. Id. at 18-19; see also id. at 29. Further, the court found that even if the three Gingles preconditions had been satisfied, the totality of the circumstances demonstrates that blacks and whites have an equal opportunity to elect candidates of their choice, citing the success of black candidates in black majority districts, 1. equal statewide voter registration rates for whites and blacks, improvement in the socioeconomic status of blacks over the past ten years, the lack of a candidate slating process, and no evidence of racial appeals in election campaigns. Pet. App. 7-9,21-23. Although the district court found that "blacks continued to trail the levels of attain- ment for whites in every significant area," i.e., income, education, housing, and employment, id. at 9- 10, it identified voter apathy, rather than discrimina- tion or lower socioeconomic status, as the cause of minority candidate defeat. Id. at 23. 2. The court of appeals vacated the district court's judgment and remanded for a more thorough examina- tion of the evidence, particularly of respondents' ___________________(footnotes) 1 In addition to the sole black county supervisor district, there were instances of black candidate success in black majority districts of local government bodies not involved in this case-for example, aldermanic districts for towns within Attala County. See Pet. App. 7-8, 59-60. ---------------------------------------- Page Break ---------------------------------------- 4 statistical evidence on racially polarized voting. The court of appeals held: The district court is not obliged to accept sta- tistical evidence as conclusive on the question whether racially polarized voting exists. * * * But in making its intensely fact-specific inquiry here, the district court ought to have discussed appellants' statistical evidence more thoroughly because that was the principal evidence they of- fered and because their statistics had at least surface plausibility. Further, the district court findings on the subjects of racial polarization and minority political cohesion are broad and general and not explicitly tied to the testimony, although many witnesses were called in the case. Pet. App. 30. 3. On remand, the district court again concluded that the redistricting plans do not violate Section 2. 2. In reviewing the statistical evidence, the district court found that respondents' original expert, using ecological regression analysis of eight black-white elections, showed an average of 15% white support and 87% black support for black candidates, and 85% white support and 13% black support for white candidates. Pet. App. 45. Respondents' post-remand expert, using ecological regression analysis of the three post- remand elections, found that between 86.3% and 92.8% of black voters supported black candidates, while be- tween 12% and 25.2% of white voters supported black candidates. Pet. App. 45. The United States' expert, using ecological regression analysis of six elections, ___________________(footnotes) 2 The United States intervened on remand in support of respondents. ---------------------------------------- Page Break ---------------------------------------- 5 found on average 89.5% of black voters and 17.1% of white voters supported black candidates, while 82.9% of white voters and 10.5% of black voters supported white candidates. Id. at 45-46; C.A.R.E. tab 12, at 11. Respondents' experts confirmed these findings using extreme case analysis. Pet. App. 47. 3 The court found the statistical findings of peti- tioners' expert, who used only ecological regression analysis, "similar to the above findings of [respon- dents' experts." Pet. App. 46; see also id. at 44 ("[The] statistical evidence produced by both sides * * * was somewhat similar."). It concluded that "there can be no doubt that the statistical evidence in the record favors a finding of racial bloc voting in Attala County." Id. at 48. Nonetheless, the district court failed to find ra- cially polarized voting. It found that "[a]lthough sta- tistical evidence may weigh in favor of a finding of racial polarization, the court is of the opinion that this evidence alone is not sufficient. Plaintiffs have failed to establish * * * that voters in Attala County cast their vote on the basis of race, to the exclusion of other nonracial factors." Pet. App. 51. It came to this conclusion in spite of finding "no proof in the record that in any way compares the candidates on any basis other than race." Id. at 50. The court relied instead on the testimony of two local politicians who stated that factors other than race influence voters and ___________________(footnotes) 3 Because there are no homogeneous black precincts in Attala County, extreme case analysis directly confirms only the white voting patterns. However, the United States' expert directly confirmed both black and white voting patterns using complementary overlapping percentages analysis, which does not require homogeneous precincts. Pet. App. 62; C.A.R.E. tab 12 at 13-14 & n.14. ---------------------------------------- Page Break ---------------------------------------- 6 determine election outcomes in Attala County. Id. at 51. It also relied on the same two instances of black crossover voting it identified in Teague I: The 1991 primary election for Sheriff, Sallis precinct, where at least 232 black voters (41% of black voters) voted for white candidates, and the 1987 primary election for constable, McCool precinct, where the black candidate received only 96 votes, which, the court held, sug- gested that many blacks in that precinct voted for white candidates. Id. at 19-20; 48. The district court went on to find that, even if respondents had proven the three Gingles factors, under the totality of the circumstances blacks none- theless have an equal opportunity to participate in the political process and elect representatives of their choice. Pet. App. 52. The court made a number of findings relating to the Gingles factors that would suggest a Section 2 violation. It found that "no black candidate has ever won a contested election for a county office in a majority-white election district." Id. at 39, 55. It acknowledged the long history of discrimination against blacks in Mississippi and At- tala County, id. at 52-53, and found that " [t]here can be no question of the socioeconomic disparities be- tween whites and blacks in Attala County." Id. at 52- 53. It also found that past discrimination and socio- economic disparities have "a significant impact on present day politics in Attala County," id. at 55, and noted the depressed levels of black voter registration and turnout in the County. 4. Id. at 54. The court, how- ___________________(footnotes) 4 Notably, the court did not reiterate its finding from Teague I that there was no evidence of racial appeals in election campaigns. Indeed, the record contains testimony of segrega- tionist campaign appeals in Attala County and derogatory ---------------------------------------- Page Break ---------------------------------------- 7 ever, discounted these circumstances because it found that voter apathy also has a significant impact on Attala County politics. Id. at 55. The court cited comparable statewide voter registration for blacks and whites, and expert testimony by a retired political science professor that national and statewide voter apathy led to a sharp decline in voter participation among both blacks and whites from 1988-1990. Id. at 54-55,63. 4. On a second appeal, a different panel of the court of appeals reversed the district court, ordered entry of judgment for respondents, and remanded for reme- dial proceedings. The court concluded that the dis- trict court's failure to find racially polarized voting was clearly erroneous because the only outcome permitted by the record was a finding of racially polarized voting. It held that two isolated instances of black crossover voting and the contested testimony of two politicians that race does not affect county elections does not "overcome the admittedly compel- ling statistical evidence of racial bloc voting." See Pet. App. 69-70, 76. The court also held that the dis- trict court erred by "plac[ing] upon the plaintiffs the insurmountable burden of coming forward with evi- dence disproving all nonracial reasons that can ex- plain election results in spite of the fact that the defendant had itself produced no real evidence that factors other than race were at work." Id. at 75. Instead, the court of appeals held that " [a] defendant may try to rebut plaintiffs' claim of vote dilution via evidence of 'objective, nonracial factors under the totality of the circumstances standard.'" Id. at 76 ___________________(footnotes) racial comments by county supervisors, including the board president. Tr. 171-173, 193-195, C.A.R.E. 13. ---------------------------------------- Page Break ---------------------------------------- 8 (quoting Nipper v. Smith, 39 F.3d 1494, 1513 (11th Cir, 1994) (en banc) (Tjoflat, C.J.), cert. denied, 115 S. Ct. 1795 (1995)). The court overturned the district court's finding that there is no racial bloc voting. The court stated that racial bloc voting need not be "absolute" to satisfy Gingles; it need only be "the usual pattern for a significant portion of voters." Pet. App. 68-69 (citing Gingles, 478 U.S. at 57). The court stated that "[i]n the face of a strong statistical case that defen- dants were unable to shake, general statements that race played no role at the polls carry little weight," id. at 75, and that the lay testimony in this case was "ambiguous at best and certainly not dispositive." Id. at 76. The court stated that one of the two politicians on whose testimony the district court relied to support its finding that Attala County voters are not influenced by race also testified that whites in Attala County do not campaign for black candidates. Id. at 73. The court of appeals noted that a third local politi- cian testified that there were racially hostile encoun- ters between whites and black politicians, and that a fourth, who was a witness for petitioners, admitted that it is harder for a candidate to become popular with voters of a different race because social life in Attala County is racially segregated. Id. at 73-74. As in the first appeal, the court of appeals recognized that [t]he district court is not obliged to accept sta- tistical evidence as conclusive on the question whether racially polarized voting exists. Teague, 17 F.3d at 798. Still, the district court does have an obligation to give particularized findings when it discredits the statistical evidence, especially ---------------------------------------- Page Break ---------------------------------------- 9 where the evidence is weighty. And it is weighty here. Indeed, the district court did not even have to choose among competing opinions based on the statistics. The results of the regression analy- ses of both plaintiffs' and defendants' experts were consistent. And when the district court looked to lay testimony, even the anecdotal evi- dence was conflicting. Id. at 74. The court agreed with the district court that statistics do not capture all the reasons motivat- ing a voter, but faulted the district court for refusing to give any weight at all to the admittedly strong statistical evidence. Id. at 74-75. The court of appeals also held that the district court clearly erred in finding that, even if the three Gingles preconditions were present, the totality of the circumstances demonstrates that blacks have an equal opportunity to participate in the political process. Pet. App. 80. The court of appeals noted that "[a]lthough there maybe cases in which even proof of the three Gingles factors may not sustain a vote dilution claim, this is certainly not such a case." Id. at 76. In particular, the court of appeals found "unten- able" the district court's ascription of depressed levels of black political participation and black elec- toral losses to voter apathy, particularly in light of the "long and dubious history of discriminati[on] against blacks," the "grave socioeconomic dispari- ties" between whites and blacks, and the 20% dispar- ity between black and white voter registration in Attala County. Pet. App. 59, 81, 82. The court held that "[t]he record before us does not contain evidence to support the district court's conclusion that voter ---------------------------------------- Page Break ---------------------------------------- 10 apathy is the reason for the failure of blacks in Attala County to elect the candidates of their choice to political office." Id. at 84. Moreover, the court con- sidered it "counterintuitive" to attribute the inability of blacks to elect candidates of their choice to apathy, when apathy affects all voters, both black and white. Id. at 83-84. As the court held, "[t]he fact that blacks and whites in Attala County are going to the polls in decreasing proportions does not explain why blacks alone are essentially shut out of the political process of the county." Id. at 84. ARGUMENT The court of appeals correctly applied Section 2 of the Voting Rights Act of 1965, 42 U.S.C. 1973, to the record before it and correctly determined that certain key factual findings by the district court were clearly erroneous. The decision of the court of appeals does not conflict with any decision of this Court or any other court of appeals. Further review is therefore not warranted. 1. Petitioners contend (Pet. 16-18) that the court of appeals' decision conflicts with Johnson v. De Grandy, 512 U.S. 997 (1994), Thornburg v. Gingles, 478 U.S. 30 (1986), and seven federal appellate deci- sions because it applies a presumption that the pres- ence of the three Gingles preconditions establishes a violation of Section 2. The court of appeals did not apply any such presumption, and its decision is therefore consistent with all of the cited cases. De Grandy held that the three Gingles factors do not "necessarily and in all circumstances demon- strate [ ] dilution." 512 U.S. at 1011. "[C]ourts must also examine other evidence in the totality of circum- stances." Ibid. See also Abrams v. Johnson, No. 95- ---------------------------------------- Page Break ---------------------------------------- 11 1425, 1997 WL 331802 (June 19, 1997), slip op. 15-16. Citing De Grandy, the court of appeals explicitly held that "a finding of the three Gingles preconditions does not end the inquiry. Reviewing courts are to look beyond the Gingles threshold factors when evaluating vote dilution claims." Pet. App. 76; see also id. at 65 ("Even if plaintiffs satisfy [the Gingles] prerequisites, the court must still look to the 'totality of the circumstances' to determine whether the challenged electoral system is equally open to minority voters."). Petitioners construe (Pet. 15-17) the court of ap- peals' statement that it would be "unusual" for a plaintiff to establish the three Gingles factors with- out also establishing a violation of Section 2 as creat- ing a legal presumption that proof of the Gingles factors establishes a Section 2 violation. That con- struction misreads the court's holding. The court of appeals stated: [I]t will be only the very unusual case in which the plaintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of 2 under the totality of the circumstances. In such cases, the district court must explain with particularity why it has concluded, under the particular facts of that case, that an electoral system that routinely results in white voters voting as a bloc to defeat the candi- date of choice of a politically cohesive minority group is not violative of 2 of the Voting Rights Act. ---------------------------------------- Page Break ---------------------------------------- 12 Pet. App. 80-81 (quoting Clark v. Calhoun County, 21 F.3d 92,97 (5th Cir. 1994)). 5 The court of appeals' comment clearly did not place any limits upon the district court's totality of the circumstances review, nor did it lessen the respon- dents' burden of proof under that standard. Instead, the court of appeals simply acknowledged that the Gingles preconditions typically encompass the major elements of a vote dilution claim. The court of appeals correctly noted that when plaintiffs already have established that white bloc voting routinely defeats the candidates of choice of a politically cohesive mi- nority, a district court, before ruling against plain- tiffs, must point to specific and credible evidence pro- viding a nonracial explanation for this fact, or at least evidence weighing heavily against plaintiffs on other relevant matters (e.g., historical discrimination and socioeconomic disparities) unrelated to racial bloc voting and minority electoral success. Pet. App. 72- 76. In this case, the district court (after two opportu- nities) did neither, relying instead on arbitrary assumptions and speculative conclusions as evidence. Thus, the court of appeals' conclusion that the dis- trict court erred in its totality of the circumstances analysis was a straightforward application of estab- lished law. The court of appeals' decision is entirely consistent with De Grandy, Gingles, and the seven court of appeals decisions cited in petitioners' footnote. See Pet. 16-17 n.7. After examining the three Gingles fac- tors, the court of appeals examined other relevant ___________________(footnotes) 5 The three court of appeals decisions that petitioners cite in a footnote (Pet. 16 n.6) as similarly in conflict with De Grandy contain the same language. ---------------------------------------- Page Break ---------------------------------------- 13 considerations to determine whether the totality of the circumstances established vote dilution in viola- tion of Section 2. Pet. App. 76-84. It did not limit the "canvassing of relevant facts," De Grandy, 512 U.S. at 1011, as petitioners contend. Pet. 16. Rather, it held that blacks do not have an equal opportunity to participate in the political process and elect candi- dates of their choice only after a "searching practical evaluation of the past and present reality." Gingles, 478 U.S. at 79 (internal citation omitted). In addition to the three Gingles factors, the court cited Missis- sippi's "long and dubious history of discriminati[on] against blacks," the current "grave socioeconomic disparities" between blacks and whites, Pet. App. 81, the depressed levels of black political participation in Attala County, and the fact that no black has ever been elected to office in a contested election from a majority white district. Id. at 59-60, 82-84. The court also evaluated, and rejected as without evidentiary support, the district court's finding of voter apathy as the cause of black electoral losses, and noted the lack of evidence in the record to support any other non- racial explanation of the statistical evidence. Id. at 83-84. Indeed, it was the district court that limited its totality of the circumstances review. As the court of appeals pointed out: [t]he district court did not take adequate notice of the considerable statistical evidence of vote dilu- tion. The conclusion it drew in its totality dis- cussion is similarly dubious, relying entirely on the denials of a few lay witnesses and ignoring the contrary testimony of others. The district court did not place this testimony in the context ---------------------------------------- Page Break ---------------------------------------- 14 of the area's history of voter exclusion nor give voice to any plausible nonracial explanation for Attala County's voting patterns. Pet. App. 81. Thus, contrary to petitioners' conten- tion, the court of appeals followed clearly established standards for determining whether Section 2 has been violated. 6 2. Petitioners advance two related contentions regarding the court of appeals' treatment of the racial bloc voting issues in the case. They contend that the court of appeals "confined Petitioners' proof of non- racial explanations [for the different voting patterns of black and white voters] to evidence that voters choose candidates based on partisanship, not race," Pet. 23, and that the court of appeals "improperly substitut[ed] its view of the facts for that of the District Court" regarding whether "county voters cast their ballots not on the basis of race but, rather, on the candidates' qualifications, education, contact with the electorate, and experience," Pet. 18. Those contentions are wrong. The court of appeals did not limit the evidence that could be considered on these issues, and it correctly articulated and applied the well-established clearly erroneous standard. As the court stated, under that standard "a finding is `clearly ___________________(footnotes) 6 Petitioners' contention (Pet. 17-18) that the court of appeals collapsed the second and third Gingles factors into a single statistical analysis is wrong. The court of appeals sepa- rately considered black political cohesion and white bloc voting. See Pet. App. 69-71. Black political cohesion was evidenced by ecological regression analysis of black voting preferences. White bloc voting was evidenced by ecological regression analy- sis and extreme case analysis of white voter preferences. That is the standard method of proving racially polarized voting by blacks and whites. See Gingles, 478 U.S. at 52, 55-61. ---------------------------------------- Page Break ---------------------------------------- 15 erroneous' when although there is evidence to support it, the reviewing court on the entire evidence is left with the definite and firm conviction that a mistake has been committed." Pet. App. 64 (quoting Anderson v. City of Bessemer City, 470 U.S. 564, 573 (1985)). a. The court of appeals did not improperly confine petitioners' proof. In reviewing the district court's findings, the court recognized that "[t]he analysis the district court must make to evaluate a 2 claim is [a] fact-intensive [inquiry], requiring 'an intensely local appraisal of the design and impact of the contested electoral mechanisms.'" Pet. App. 64 (quoting Gin- gles, 478 U.S. at 79). Further, the court of appeals held that under the totality of the circumstances "[a] defendant may try to rebut plaintiffs' claim of vote dilution via evidence of `objective, nonracial factors'" that explain voting patterns. Id. at 76 (quoting Nipper v. Smith, 39 F.3d 1494, 1513 (llth Cir. 1994) (en banc) (Tjoflat, C.J.), cert. denied, 115 S. Ct. 1795 (1995)). However, in this case, the court of appeals stated that the record "is devoid of any proof that [nonracial] factors * * * are as or more important than race in predicting voting behavior in Attala County." Id. at 72. Even the district court found "no proof in the record that in any way compares the candidates on any basis other than race." Id. at 50. The court of appeals thus could not have limited petitioners' proof of nonracial factors because there was no such evidence in the record. Petitioners select (Pet. 23) one sentence from the court of appeals' opinion to support their claim that the court limited its review of nonracial explanations to partisan politics. The court of appeals stated: ---------------------------------------- Page Break ---------------------------------------- 16 We are not denying that the ultimate inquiry of Section 2 is racial discrimination; nor are we holding that a defendant cannot offer evidence of the nonracial reasons for the voting patterns. In LULAC we did not hold that the plaintiff has the burden of negating all nonracial reasons possibly explaining plaintiffs' statistical case. That case concerned a different problem-the role of partisan politics. Pet. App. 84 (emphasis added). This discussion of League of United Latin American Citizens v. Clem- ents, 995 F.2d 831 (5th Cir. 1993) (en banc), cert. denied, 510 U.S. 1071 (1994), does not suggest that the court has improperly limited the proof of nonracial explanations to partisan polities. The court simply contrasted LULAC, in which the court of appeals found that defendants offered substantial, objective evidence that partisan politics better explained the electoral outcomes, to this case, in which there was no objective evidence of any nonracial explanation and the district court improperly required plaintiffs to disprove all possible nonracial explanations. See Pet. App. 84. b. The court of appeals correctly applied the clearly erroneous standard in rejecting petitioners' claim that black and white voters prefer different candidates for reasons other than race. This is not a case in which the court of appeals substituted its judgment for the district court's by choosing between two plausible views of the evidence, as petitioners allege. See Pet. 19. The district court's analysis mis- applied the burden of proof. See Pet. App. 72, 74-75. In addition, the district court disregarded the standard methodology for proving racial bloc voting approved ---------------------------------------- Page Break ---------------------------------------- 17 by this Court, see id. at 69,72,75, and disregarded the court of appeals' remand order that it should base its conclusions on particularized findings of fact, see id. at 75. As the court of appeals held, the evidence of racial polarization was "overwhelming," id. at 70, 75, and the district court's failure to find racial polariza- tion was not supported by the evidence. See id. at 69, 72,74,76. (i). The district court misapplied the burden of proof by improperly "plac[ing] upon the [respondents] the insurmountable burden of coming forward with evidence disproving all nonracial reasons that can ex- plain election results in spite of the fact that the defendant had itself produced no real evidence that factors other than race were at work." Pet. App. 75; see also id. at 72, 74. Petitioners admit that it would be error to require such a showing, but claim that the district court committed no such error. Pet. 24 n.8. To the contrary, the district court's error is clear. The district court found that respondents "failed to establish that voters in Attala County cast their vote on the basis of race, to the exclusion of other non- racial factors, such as a candidate's experience, quali- fications, education and contact with the electorate." Pet. App. 51. It made this finding even though it ac- knowledged that "[conspicuously missing from the record in this [case] is any proof to the contrary. Of the black-white elections analyzed, there is no proof in the record that in any way compares the candidates on any basis other than race." Id. at 50. And as the court of appeals pointed out," [t]he lack of evidence in the record on this point favors the [respondents], not the [petitioners]." Id. at 72. (ii). In weighing the evidence, moreover, the dis- trict court clearly erred by placing undue weight on ---------------------------------------- Page Break ---------------------------------------- 18 isolated instances of black crossover voting in two precincts in two elections and on controverted, con- clusory testimony of two politicians, and by disre- garding the consistent evidence of racially polarized voting spanning 13 years that the district court itself found to be strong. See Pet. App. 47-48, 51, 69-74. Pe- titioners state that "[t]he District Court found that the expert testimony presented by the parties `on racial bloc voting patterns varied considerably.'" Pet. 18 (quoting Pet. App. 44). But this partial quotation is incomplete. The district court's full finding is: "At trial, expert testimony on racial bloc voting patterns varied considerably, although statistical evidence produced by both sides on the same was somewhat similar." Pet. App. 44 (emphasis added). The district court reiterated that the statistical findings of peti- tioners' expert "were similar to the above findings of [respondents'] experts," id, at 46, 7. and found that "there can be no doubt that the statistical evidence in the record favors a finding of racial bloc voting in Attala County." Id. at 48; see also id. at 51. The court of appeals found that the meager, con- jectural and controverted rebuttal evidence identified by the district court simply did not support the dis- ___________________(footnotes) 7 Petitioners' expert report shows levels of estimated white and black support for black candidates similar to the levels found by respondents' experts. See Defs.' Response to Pls.' Post-Remand Evidentiary Proffer and Defs.' Proffer, Attach. B, Tables 1 & 2 (filed July 8, 1994). Petitioners' expert also conceded that racial polarization exists in exogenous elections. Id., Attach. A, Weber Decl. "Par" 11 (analyses of congressional elections "reveal a pattern of some racial differences in voting preferences"); id. at 12 (finding "[r]acial polarization in voting" in analyses of exogenous county-wide or state-wide elections). ---------------------------------------- Page Break ---------------------------------------- 19 trict court's conclusion that "[f]actors such as a candidate's experience, qualifications, education and contact with the electorate are of much greater significance than race to voters in Attala County." Pet. App. 50. The court of appeals acknowledged that "while the [district] court does have discretion in the manner in which it weighs the facts, the statistical evidence here is so one-sided as to require a more convincing discussion of the lay testimony before refuting the objective results." Id. at 76. 8. As the court of appeals held, "[i]n the face of a strong statis- tical case that [petitioners] were unable to shake, general statements that race played no role at the polls carry little weight." Id. at 75. Accordingly, the court of appeals correctly concluded that the record ___________________(footnotes) 8 In Abrams v. Johnson, No. 95-1425, 1997 WL 331,802 (June 19, 1997), this Court reiterated the legal principles applied by the court of appeals in this case. Abrams, slip op. 15-16. The Court held in Abrams that the three-judge district court's findings of lack of racially polarized voting were adequately supported by the record in that case. That conclu- sion is entirely consistent with the court of appeals' finding that, on the much stronger evidence in this case, the district court had clearly erred in finding a lack of racially polarized voting. In Abrams, "the statistical evidence was for the most part inconclusive and conflicting." Abrams, slip op. 17. Here, by contrast, the statistical evidence presented by both sides was consistent and demonstrated the existence of racially polarized voting. See Pet. App. 48, 69. Finally, the Abrams district court found "a general willingness of white voters to vote for black candidates," slip op. 18, and it noted that candidates of choice of black voters had "achieved many electoral victories in local and statewide elections and have received significant- occasionally overwhelming-support from both black and white voters" in the district at issue there. Slip op. 17. No such findings were made-or could have been made-on the record in this case. ---------------------------------------- Page Break ---------------------------------------- 20 in this case did not support the district court's failure to give proper weight to the consistent results of regression and extreme case analyses in disregard of the racial bloc voting methodology approved by this Court in Gingles, 478 U.S. at 52, 55-61. The court of appeals' record-specific application of established legal principles warrants no further review. CONCLUSION The petition for a writ of certiorari should be denied. Respectfully submitted. WALTER DELLINGER Acting Solicitor General ISABELLE KATZ PINZLER Acting Assistant Attorney General MARK L. GROSS MICHELLE ARONOWITZ Attorneys JUNE 1997