LACY H. THORNBURG, ET AL., APPELLANTS V. RALPH GINGLES, ET AL. No. 83-1968 In the Supreme Court of the United States October Term, 1985 On Appeal from the United States District Court for the Eastern District of North Carolina Brief for the United States as Amicus Curiae Supporting Appellants TABLE OF CONTENTS Interest of the United States Statement Introduction and summary of argument Argument: The district court erroneously held that the redistricting plan at issue violates amended Section 2 of the Voting Rights Act of 1965 A. Amended Section 2 guarantees every citizen the right to an equal opportunity to participate in the political process B. The district court misapplied the factors appropriate to an analysis of appellees' claim of unlawful vote dilution Conclusion QUESTIONS PRESENTED 1. Whether the district court correctly construed amended Section 2 of the Voting Rights Act of 1965, 42 U.S.C. 1973, as invalidating certain multi-member legislative districts in which minority candidates had a proven opportunity to participate in the electoral process, on the ground that there was no guarantee that minorities would enjoy the continued electoral success guaranteed by "safe" districts. 2. Whether racial bloc voting exists as a matter of law whenever less than 50% of the white voters cast ballots for a minority candidate. INTEREST OF THE UNITED STATES On October 1, 1984, the Court entered an order inviting the Solicitor General to express the views of the United States in this case. We responded in a brief urging summary affirmance on two questions and plenary review on two others, and the Court noted probable jurisdiction on the latter two questions on April 29, 1985. This case presents several questions concerning the proper construction of the 1982 amendment to Section 2 of the Voting Rights Act of 1965, 42 U.S.C. 1973. The United States has the primary responsibility for enforcing the Voting Rights Act and thus has a substantial interest in ensuring that the Act is construed in a manner that advances, rather than impedes, its objectives. STATEMENT 1. In July 1981, as a result of the 1980 census, North Carolina enacted redistricting plans for the state's House of Representatives and Senate. In September 1981, appellees filed this suit, alleging that the plans had been enacted pursuant to provisions of the North Carolina constitution that required, but had not received, preclearance pursuant to Section 5 of the Voting Rights Act of 1965, 42 U.S.C. (1976 ed.) 1973c, and that the use of large multi-member districts submerged concentrations of black voters and diluted minority voting strength in violation of the Constitution and Section 2 of the Voting Rights Act of 1965, 42 U.S.C. (1976 ed.) 1973 (J.S. App. 3a-4a). /1/ After the plants were ultimately adopted by the state legislature, /2/ appellees amended their pleadings to challenge five House Districts (Nos. 8, 21, 23, 26, and 39) and two Senate Districts (Nos. 2 and 22) and to conform their pleadings to the newly-amended Section 2 of the Voting Rights Act. /3/ The "gravamen" of appellees' claim with reference to these multimember districts was that the State's plan "makes use of multi-member districts with substantial white voting majorities in some areas of the state in which there are sufficient concentrations of black voters to form majority black single-member districts * * *" (J.S. App. 4a). The plan was in this respect claimed to violate amended Section 2 of the Voting Rights Act. 2. The case was tried before a three-judge court on the basis of extensive stipulations of fact, documentary evidence, and oral testimony (J.S. App. 8a). The court entered an order and opinion containing extensive findings on the various factors identified in the legislative history of amended Section 2 and case law as relevant to a vote dilution claim. J.S. App. 21a-51a. The court held that "it has now become possible for black citizens to be elected to all levels of state government in North Carolina" (id. at 37a). However, the court further held that, under the totality of the relevant circumstances, the redistricting plan in all seven challenged districts diluted minority votes in violation of amended Section 2 and enjoined elections in the challenged districts (ibid.). /4/ The district court also reviewed at length the racial demographics and voting history of each challenged multi-member district. House District 21. House District 21, in Wake County, elects six members to the General Assembly on an at-large basis (J.S. App. 19a). The population of the district is 21.8% black, and black voters constitute 15.1% of all registered voters (ibid.). /5/ 72% of the white voting age population is registered to vote, and 49.7% of the black voting age population is registered to vote (id. at 24a n.22). The black population is so situated that one single-member legislative district could be drawn within the present boundaries, with a black population of 67% (id. at 20a). Under the challenged plan and its predecessor, /6/ one black legislator was elected in 1980 and reelected in 1982 (id. at 35a, 44a). In those elections, respectively, he received the votes of 31% and 39% of the white voters in the primary, and the votes of 44% and 45% of the white voters in the general election (id. at 44a). House District 23. House District 23, in Durham County, elects three members at-large to serve two-year terms in the General Assembly (J.S. App. 19a). The black population is 36.3% of the total, and blacks constitute 28.6% of the registered voters (ibid.). 66% of the white voting age population is registered to vote, and 52.9% of the black voting age population is registered (id. at 24a n.22). The black population is so situated that one single-member district could be drawn within the present boundaries, with a black population of 70.9% (id. at 20a). Under the challenged plan and its predecessor, this district has elected one black representative in every election since 1973 (id. at 35a). The black legislator was unopposed in the general election in 1978, and in both the primary and general elections in 1980. In 1978, he was elected with 16% of the white vote in the primary, and in 1982 he received 37% of the white vote in the primary and 43% of the white vote in the general election. /7/ A second black candidate also garnered 26% of the white vote in the 1982 primary (id. at 43a-44a). House District 36. House District 36, in Mecklenburg County, has an eight-member House delegation, elected at-large (J.S. App. 19a). Blacks constitute 26.5% of the district's population and 18% of its registered voters (ibid.). 73% of the white voting age population is registered to vote, and 50.8% of the black voting age population is registered (id. at 24a n.22). The black population of the district is so situated that two single-member legislative districts could be drawn that would be 66.1% and 71.2% black (id. at 20a). Under the present plan, one black representative was elected in 1982; he is the first black citizen to be elected to the House from Mecklenburg County in this century (id. at 43a). He received 50% of the white vote in the primary and 42% of the white vote in the general election (id. at 41a). /8/ A second, unsuccessful, black candidate received 39% of the white vote in the 1982 primary and 29% in the general election (ibid.). /9/ House District 39. House District 39, in a part of Forsyth County, has five at-large seats in the General Assembly (J.S. App. 19a). The population of the district is 25.1% black, and blacks constitute 20.8% of the registered voters (ibid.). 69.4% of the white voting age population is registered to vote, and 64.1% of the black voting age population is also registered (id. at 24a n.22). The black population is so situated that one single-member legislative district, with a 70.0% concentration of black voters, could be drawn (id. at 20a). Under the present plan, two of the five representatives elected in 1982 were black; under the predecessor plan, a black representative was elected in 1974 and reelected in 1976 (id. at 35a). The two black representatives elected in 1982 received 25% and 36% of the white vote in the primary election, and 42% and 46% in the general election (id. at 43a). One of these representatives had previously won the Democratic nomination in 1978 and 1980 (with 28% of the white vote in 1978 and 40% of the white vote in 1978 and 40% of the white vote in 1980), but lost the general election in those years, (id. at 42a-43a). Senate District 22. Senate District 22, in Mecklenburg and Cabarrus Counties, is a four member district (J.S. App. 19a). The population is 24.3% black, and 16.8% of the registered voters are black (ibid.). In Mecklenburg County, 73% of the white voting age population is registered to vote, as is 50.8% of the black voting age population (id. at 24a n.22). /10/ The black population is so situated that one single-member district could be created with a 70.0% black population (id. at 20a). Under the present plan, no black Senator is part of the delegation; however, a black citizen was elected from 1975-1980 (id. at 34a). The black senatorial incumbent (Alexander) received 47% of the white vote in the 1978 primary, and 41% of the white vote in the general election; his share of the white vote dropped to 23% in the 1980 primary (id. at 42a). A second black candidate (Polk), running in 1982, garnered 32% of the white votes in the primary and 33% in the general election. Ibid. INTRODUCTION AND SUMMARY OF ARGUMENT This is the first case in this Court to accord plenary appellate review to a trial court's finding of a violation of the 1982 amendment to Section 2 of the Voting Rights Act. That provision, enacted after an intense legislative struggle, represents a studied compromise that condemns only those electoral procedures that "result" in a denial of an "equal opportunity to participate in the electoral process." That conclusion is a matter of law, the proper conception of which must be established and assured by this Court. This ultimate determination requires delicate judgments that can hardly be reached or reviewed by any mechanical standard. If the integrity of the Section 2 compromise is to be maintained, moreover, an appellate court must assure itself not only that a trial court has considered the appropriate evidence in reaching its conclusion, but also that this evidence, taken as a whole and properly balanced, supports the trial court's answer to the ultimate question that Congress has prescribed. The district court considered all of the evidence, but it reached an ultimate conclusion at odds with the correct legal standard. If left undisturbed, that decision means that wherever there has been discrimination in the past and some measure of racial polarization in voting in the present, district courts will be free to strike down virtually any scheme that does not -- or even those that do -- deliver electoral successes proportional to minority voting strength. That is not what Congress intended. Specifically, we shall argue that the trial court, by ignoring recent minority electoral successes in the districts in issue, could not reasonably have found a violation under the proper "equal opportunity to participate" standard, but rather must implicitly have sought to guarantee continued minority electoral success. Further, the court below adopted and made dispositive a definition of racial block voting that, taken literally, might justify finding this factor present in virtually any district with a racially mixed electorate and thus could justify requiring proportional representation in all such districts. Congress crafted a precise standard for intervention in the electoral process, and fidelity to that standard requires that this judgment be set aside. /11/ ARGUMENT THE DISTRICT COURT ERRONEOUSLY HELD THAT THE REDISTRICTING PLAN AT ISSUE VIOLATES AMENDED SECTION 2 OF THE VOTING RIGHTS ACT OF 1965 A. Amended Section 2 Guarantees Every Citizen The Right To An Equal Opportunity To Participate In The Political Process 1. The legislative background of amended Section 2 underscores the centrality of the principles noted above to the compromise enacted into that law. Amended Section 2 reflects the consensus of an overwhelming majority of the Congress, reached only after an intensive and devisive debate, whether to endorse or reject the holding in City of Mobile v. Bolden, 446 U.S. 55 (1980). The product of that debate was a provision that does not require proof of racial animus to establish a violation of amended Section 2 and does not allow proof of electoral failure solely or even preponderantly to establish a violation under the Act. Congress chose an altogether different approach: As adopted, Section 2 guarantees every citizen equal access to the electoral process and thus focuses upon that process itself. /12/ a. Amended Section 2 originated in the 97th Congress when H.R. 3112 was introduced to extend certain features of the 1965 Voting Rights Act and to modify Section 2 of the Act because of the decision in City of Mobile. H.R. 3112 would have eliminated an intent standard by forbidding any jurisdiction from imposing or applying any electoral practice "in a manner which results in a denial or abridgement of the right * * * to vote on account of race or color * * *," /13/ a test claimed by its supporters to stem from White v. Regester, 412 U.S. 755 (1973), and Whitcomb v. Chavis, 403 U.S. 124 (1971). Most of the discussion in the House regarding H.R. 3112 was devoted to other aspects of the bill; the proposal to amend Section 2 attracted little debate. /14/ As passed by the House, H.R. 3112 contained the results test in the original bill and a disclaimer that numerical underrepresentation itself violated Section 2. /15/ b. After the House passed H.R. 3112, the Senate Subcommitte on the Constitution began hearings on two bills, one that contained the results test in H.R. 3112 (S. 3112) and one that would have retained the City of Mobile standard (S. 1975). /16/ The ensuing debate focused on the proper standard for Section 2. Proponents of a results test chiefly argued that the Court's holding in City of Mobile insulated discriminatory practices from review because of the difficulty of obtaining evidence regarding the subjective motivations of legislators, especially when the practices in question were adopted long ago. /17/ They proposed that the analysis should be based upon the various so-called "objective" factors identified in White v. Regester and pre-City of Mobile lower court cases applying that standard. Critics of the results test agreed, in essence, that a finding of unlawful vote dilution could and should be made on the strength of objection evidence, but were concerned with, among other things, the potentially-limitless scope of the test. /18/ A principal concern was the implication left by the disclaimer: given its limited terms -- that numerical underrepresentation of minorities would not amount "in and of itself" to a violation of Section 2 -- opponents of the results test maintained that proportional representation would ineluctably follow simply from proof of some additional factor identified in White or elsewhere. /19/ Another major criticism was that the House version lacked a "core value" or an "ultimate or threshold criterion" other than proportional representation for evaluating vote dilution claims. /20/ Supporters of the results test repeatedly assured its critics that it was not a mandate for proportional representation, /21/ that it was merely a means of ensuring that minorities were not effectively "shut out" of the electoral process, /22/ and that, given the heavy burden the test placed on a plaintiff -- one supporter described it as "incredibly difficult" /23/ -- the test would invalidate only those electoral practices that denied minorities an equal opportunity to participate in the political process. /24/ As Armand Derfner, head of the Voting Rights Project, put it, the "goal" of amended Section 2 "is to create an opportunity -- nothing more than an opportunity -- to participate in the political system." 1 Senate Hearings 821 (Prepared Statement). /25/ Nonetheless, the Constitution Subcommittee rejected the House effects test in favor of the City of Mobile standard. 2 Senate Hearings 10. c. To break the deadlock, Senator Dole, with the backing of the President, offered a compromise version of Section 2 that responded to criticisms of the effects test by introducing "additional language" incorporated from White v. Regester "delineating what legal standard should apply under the results test" and "clarifying that (this test) is not a mandate for proportional representation." 2 Senate Hearings 60 (statement of Sen. Dole); id. at 58-59. The most significant feature of the compromise was to modify and expand the language of the House-passed bill to ensure that "equal opportunity," not "proportional results," would be the legal test. Senate Report 193-194 (Additional Views of Sen. Dole); id. at 199 (Supplemental Views of Sen. Grassley). As Senator Dole put it, because his version of amended Section 2 "focus(es) on access to the process, not election results" (2 Senate Hearings 61-62), the question to be answered is "not whether (minorities) have achieved proportional election results," but "whether members of a protected class enjoy equal access. I think that is the thrust of our compromise: equal access, whether it is open; equal access to the politial process" (id. at 60; see also 2 Senate Hearings 46 (Sen. Leahy) ("(i)t is the opportunity to participate, not the actual use of that right, which is crucial * * *")). The Committee adopted Senate (128 Cong. Rec. S7139 (daily ed. June 18, 1982)). Although the Senate bill differed from the House version, the House dispensed with a conference and adopted the Senate bill (id. at H3846 (daily ed. June 23, 1982)). /26/ 2. The legislative history thus reveals that the compromise encompassed three key areas of consensus. First, there was widespread agreement that direct evidence of intent to discriminate should not be necessary to establish a violation under Section 2. House Report 29; Senate Report 193 (Additional Views of Sen. Dole). Second, during the course of the debate, a consensus -- Senator Dole described it as "a unanimous consensus" -- developed against permitting Section 2 claims to be based upon the inability of a group to achieve representation in proportion to its population within the jurisdiction. /27/ Rather, members of Congress who favored /28/ or opposed /29/ the original results test and the compromise version of amended Section 2, as well as private supporters of the bill, /30/ agreed that proof of minority underrepresentation was a necessary but not a sufficient element of a successful vote dilution claim, as the Court's decision in White and Whitcomb had held. /31/ Third, both sides in the controversy agreed that the concepts of unconstitutional vote dilution developed by this Court in White and Whitcomb and as applied by the lower courts prior to City of Mobile /32/ should govern amended Section 2 cases. /33/ Amended Section 2, as the text itself makes clear, thus focuses not on guaranteeing election results, but instead on securing to every citizen the right to an equal "opportunity * * * to participate in the political process * * *" (42 U.S.C. 1973). As Senator Dole, whose views, as principal sponsor of the compromise Section 2 that passed the Congress, provide an authoritative guide to the statute's construction, /34/ stated in explanation of his proposal, "(c)itizens of all races are entitled to have an equal chance of electing candidates of their choice, but if they are fairly afforded that opportunity, and lose, the law should offer no redress." Senato report 193 (Additional Views of Sen. Dole). Senator Dole made the same point during the floor debate on his compromise (128 Cong. Rec. S6961 (daily ed. June 17, 1982)) and added that (ibid.): (T)he standard is whether or not the political processes are equally "open," whether there is access, whether they are open in that members of a protected class have the same opportunity as others to participate in the political process and to elect candidates of their choice. In response to a question from Senator Thurmond whether "the focus on the section 2 standard (is) on equal access to the political process or is * * * on whether a minority group has achieved equal election results?" (id. at S6962), Senator Dole replied (ibid.): The focus in section 2 is on equal access, as it should be. I thank the Senator for directing the question. I know of no one in this Chamber and I heard no one anywhere else indicate that it should be otherwise. It should be on access. Is the system open to people in Kansas, South Carolina, North Carolina, California, New York, wherever? Do they have access and an opportunity to cast their vote? It is not a right to elect someone of their race but it is equal access and having their vote counted. Amended Section 2, Senator Dole further explained, would "(a)bsolutely not" provide any redress "if the process is open, if there is equal access, if there are no barriers, direct or indirect, thrown up to keep someone from voting or having their vote counted, or registering, whatever the process may include" (ibid.) In his view, so long as "(t)he political process leading to nomination or election (is) * * * equally open to participation by members of a class of citizens without regard to race, color, or language minority" there could not be "a denial or abridgement of the right to vote under the amendment" (128 Cong. Rec. S7120 (daily ed. June 18, 1982)(colloquy between Sen. Dole and Sen. Gorton); see also id. at S7119 (Sen. Dole)); cf. Whitcomb v. Chavis, 403 U.S. at 153. Supporters of amended Section 2 in the Senate echoed Senator Dole's understanding of his compromise amendment to Section 2. They repeatedly emphasized that the provision guaranteed "equal access" /35/ or "an equal opportunity to participate," /36/ but that it did not apply where minority voters or candidates "failed to participate given an equal opportunity" /37/ to do so. /38/ These statements demonstrate that the supporters of Senator Dole's compromise version of amended Section 2 shared his construction of its terms. Accordingly, the central issue under amended Section 2, as all participants in the Senate floor debate agreed, is whether a challenged electoral practice "result(s) in the denial of equal access to any phase of the electoral process for minority group members" (Senate Report 30 (emphasis added)). /39/ 3. The foregoing discussion makes clear that appellees err in claiming that district court's finding that the multimember district plan dilutes black votes is subject to Fed. R. Civ. P. 52(a). Mot. to Dis. 21, 35-36. Like proximate cause in the law of torts, in therm "results" requires an evaluation of the facts in light of the purposes of the policy being served. Cf. Metropolitan Edison Co. v. PANE, 460 U.S. 766, 774 (1983) (construing terms "'environmental effect'" and "'environmental impact'" in light of the "the congressional concerns that led to the enactment of NEPA"). The question under amended Section 2 -- whether a particular electoral practice "results" in the denial of "equal access" to the political process -- thus calls for more than a factual conclusion not only because Congress eschewed reliance upon a "mechanical 'point counting' device" to resolve Section 2 claims (Senate Report 29 n.118; see 128 Cong. Rec. S6648 (daily ed. June 17, 1982) (Sen. Grassley)); but also because the undertaking requires a careful analysis of the challenged electoral process, as informed by its actual operation, including the nonquantifiable, but undeniable, fact that a numerical minority may exercise substantial, and sometimes decisive, influence upon the process. See Whitcomb, 403 U.S. at 149-155. /40/ The Court has recognized in a variety of other situations that a conclusion based largely upon the application of a rule of law to a particular set of facts is a legal, not a factual conclusion. /41/ In addition, for plaintiffs as well as defendants "the stakes -- in terms of impact on future cases and future conduct -- are too great to entrust them finally to the judgment of the trier of fact" (Bose Corp. v. Consumer Union of United States, Inc., No. 82-1246 (Apr. 30, 1984), slip op. 15 n.17; see id. at 15-25). Were the ultimate issue under amended Section 2 simply a question of fact, plaintiffs would be disabled from effectively challenging decisions where, on an essentially standardless basis, the court determined that the "totality of the circumstances" did not support their case. Accordingly, because is it clear that an appellate court must independently resolve mixed questions of fact and law (Bose Corp., slip op. 15), this Court is not bound by Rule 52(a) in determining whether the multi-member districts in the 1982 reapportionment plan violates Section 2. /42/ B. The District Court Misapplied The Factors Appropriate To An Analysis Of Appellees' Claim Of Unlawful Vote Dilution In voiding the use of multi-member districts in the 1982 reapportionment plan, the district court made two fundamental errors in construing and applying amended Section 2, either of which is sufficient to require reversal. First, the court found a violation of the statute in the absence of evidence that the "results" of the multi-member legislative districts challenged here denied minority voters an equal opportunity to participate in the electoral process. Second, the court adopted an erroneous definition of racially polarized voting, one that misconceives the proper force of that criterion as an element of a successful Section 2 claim. 1. a. Each of the districts is a multimember district. However, it is firmly settled that multimember districts are not inherently unlawful. Senate Report 33; White, 412 U.S. at 765; Whitcomb v. Chavis, supra; see also 2 Senate Hearings 81 (statement of Sen. Dole). While it is true that in each of the districts at issue here it would be possible to create one or more single-member districts with effective black voting majorities (see pages 3-6, supra), this point cannot be dispositive. Minority voters have no right to the creation of safe electoral districts merely because they could feasibly be drawn. Whitcomb v. Chavis established that principle prior to the 1982 amendment to Section 2, and the Court's recent summary affirmances in Brooks v. Allain, No. 83-1865 (Nov. 13, 1984), and Strake v. Seamon, No. 83-1823 (Oct. 1, 1984), have reaffirmed that principle under amended Section 2. /43/ Nor can it be presumed without more that "safe" seats for minority officeholders would necessarily be in the interests of minority voters. See United States v. Board of Supervisors, 571 F.2d 951, 956 (5th Cir. 1978). Accordingly, if the "gravamen" of appellees' claim is simply that North Carolina chose to use multimember districts where "there are sufficient concentrations of black voters to form majority black single-member districts" (J.S. App. 4a), their claim necessarily falls short of establishing a violation. /44/ b. Moreover, in three of the challenged districts, black candidates supported by the black community have been elected under the challenged plan in numbers as great as or greater than would be expected under a single-member plan, and black voters have wielded influence over other seats as well. Even since 1973, black voters in House District 23, who make up 36.3% of the population and 28.6% of the registered voters, have elected a black member of the three-person delegation. Page 4, supra. /45/ In House District 21, the 21.8% black minority, constituting 15.1% of the registered voters, elected a black representative to its six-member delegation in 1980 under a substantially-identical predecessor to the challenged plan (J.S. App. 19a) and reelected him in 1982 under the challenged plan. Pages 3-4, supra. The district court's error is even clearer in House District 39. In that district, where 25.1% of the population is black and 20.8% of the registered voters are black, a black candidate was elected to the five-member delegation in 1974 and reelected in 1976. In 1982, under the challenged plan, two black representatives, or 40% of the delegation, were elected. Page 5, supra. By contrast, under the alternative plan favored by appellees, in each of these districts black voters would be relegated to one single-member district with a large black majority. The ability of black voters to contest the remaining seats would be lessened -- indeed, in House District 39 minority voters could have a reduced number of delegates -- and (more importantly) their electoral influence on the other representatives would be reduced. /46/ Accordingly, judged simply on the basis of recent electoral "results" the multimember plans in these districts have apparently enhanced -- not diluted -- minority voting strength. In the remaining districts -- House District 36 and Senate District 22 -- black candidates have been less successful. Even there, however, the 26.5% black minority in the House district, constituting 18% of the registered voters, elected a black member to the eight-member delegation in 1982, and a second black candidate (who lost in the general election) received 39% of the white vote in the primary. Pages 4-5, supra. In Senate District 22, although the 24.3% black minority, constituting 16.8% of the registered voters, has not been able to elect a black Senator in the 1980s, a black candidate prevailed throughout the period 1975-1980. Pages 5-6, supra. /47/ This experience cannot be reconciled with the district court's holding that the challenged plan results in vote dilution. /48/ Indeed, the district court never articulated a standard under which "results" such as these could support a conclusion that the multi-member electoral system in these districts -- which is the procedure under challenge -- is "not equally open to participation" by black voters. The court only stated -- without reference to actual results in any of the challenged districts -- that "the success that has been achieved by black candidates to date" is "too minimal in total number and too recent" to support a finding that a black candidate's race is no longer "a significant adverse factor" (J.S. App. 37a-38a). /49/ However, the election of representatives in numbers as great as or greater than the approximate black proportion of the population, as in House Districts 21, 23, and 39, is surely not "minimal." And in House District 36 and Senate District 22, while the results admittedly fall short of a standard of "proportional representation" -- which Congress rejected as the governing legal criterion -- minority candidates either are or have been successful and plainly are competitive. /50/ In fact, the district court itself concluded that "(t)hirty-five years after the first successful candidacies for public office by black citizens in this century, it has now become possible for black citizens to be elected to office at all levels of state government in North Carolina" (J.S. App. 37a). The district court also erred by discounting the proven minority electoral success on the ground that it was "too recent in relation to the long history of complete denial of any elective opportunities" to support the conclusion that "a black candidate's race is no longer a significant adverse factor" (J.S. App. 37a-38a). That ruling is overbroad. To the extent that the court held that past discrimination cannot be overcome by providing minorities with contemporary access to the process, that ruling is in error. The lower court decisions prior to City of Mobile repeatedly emphasized that the key question is not whether there was past discrimination but whether that discrimination prevents minorities from currently participating in the political process. See, e.g., Hendrix, 559 F.2d at 1270; David, 553 F.2d at 930; Bradas, 508 F.2d at 1112; Zimmer, 485 F.2d at 1306; accord, McCarty v. Henderson, 749 F.2d 1134, 1137 (5th Cir. 1984). /51/ Historical discrimination that has resulted in a current lower minority registration rate, for instance, as the district court found to be the case here (J.S. App. 22a-26a & n.22), is an entirely appropriate consideration under amended Section 2. /52/ But past discrimination that does not deny minorities current access to the political process cannot support a violation of the Act. /53/ And to the extent that the district court held that past discrimination persists in the form of racial bloc voting, the court relied upon an erroneous definition of that concept, as we will later explain. Congress could not have expressed more clearly its intention not to invalidate multimember districting plans where minorities have had an equal opportunity to participate in the electoral process, even if minority candidates did not win a proportionate share of the seats. /54/ Congress adopted Senator Dole's compromise precisely to ensure that Section 2 would guarantee minority voters access to the electoral process -- not ensure victories for minority candidates -- as the Senate floor debate plainly demonstrates. Pages 15-17, supra. See also Rogers, 458 U.S. at 616; Whitcomb, 403 U.S. at 158-159 (multimember districts challenged for "their winner-take-all aspects"). /55/ The pre-City of Mobile decisions of this and other courts bear out that multimember districts are not unlawful where, as here, minority candidates are not effectively shut out of the electoral process. The closest analogy to this case is Dove v. Moore, supra, in which the court of appeals upheld the validity of an at-large system under which the 40% black minority elected one member to an eight-member city council. Indeed, in many cases where the aggregate of factors was unquestionably less favorable to minority voters than in this case -- most particularly, where no black citizen had ever been elected under the system -- challenges to the voting plans were nonetheless held to be insufficient. See, e.g., Black Voters v. McDonough, supra; Hendrix v. Joseph, supra; David v. Garrison, supra; McGill v. Gadsden County Comm'n, 535 F.2d 277 (5th Cir. 1976). And it is significant that the Senate majority and other supporters of amended Section 2 pointed to these cases -- including Dove v. Moore -- as indications of the way in which the new provision would operate. See, e.g., Senate Report 33; 1 Senate Hearings 795-796, 797 (testimony of Armand Derfner); id. at 1701-1702 (colloquy between Sen. Mathias and Assistant Attorney General Reynolds regarding Dove). Accordingly, given the proven electoral success that black candidates have had under the multimember system, the district court erred by concluding that use of that system "results" in a denial of "equal access" to the electoral process for minorities. 2. The district court correctly held (J.S. App. 15a) that proof of racial bloc voting is the "linchpin" of a successful vote dilution claim. See Senate Report 33. /56/ However, the district court adopted a definition of racial bloc voting under which racial polarization is "substantively significant" or "severe" whenever "the results of the individual election would have been different depending upon whether it had been held among only the white voters or only the black voters in the election" (J.S. App. 39a-40a (footnote omitted)). This means that even a minor degree of racial bloc voting would be sufficient to make out a violation, regardless of whether it actually results in black electoral defeats. For instance, in a two-person election where there is a small white voting majority, if the white candidate receives 51% of the vote in the white community and 49% of the vote in the black community, and the black candidate receives the reverse, the district court would hold that the community is severely racially polarized. That definition is unacceptable because "'there will almost always be a raw correlation with race in any failing candidacy of a minority whose racial or ethnic group is (a) small percentage of the total voting population'" (Lee County Branch of NAACP v. City of Opelika, 748 F.2d 1473, 1482 n.15 (11th Cir. 1984) (quoting Jones v. City of Lubbock, 730 F.2d 233, 234 (5th Cir. 1984) (Higginbotham, J., specially concurring)); See Terrazas v. Clements, 581 F.Supp. 1329, 1351-1352 (N.D. Tex. 1984) (three-judge court)(test is whether "such bloc voting as many exist" operates so as to "persistently defeat (minority) candidates"); accord, Seamon v. Upham, slip op. 10 n.4. /57/ Under the district court's definition, virtually an electoral district in the country might be deemed to suffer "substantively significant" racial bloc voting. Congress believed that the contrary was true, however. See Senate Report 33 (in "most communities" minority candidates "receive substantial support from white voters"). /58/ If white voters are willing to cross racial lines in sufficient numbers that "minority candidates (do) not lose elections solely because of their race" (Rogers, 458 U.S. at 623), then it is largely irrelevant whether the black candidate would have won even if the election "had been held among only the white voters" (J.S. App. 40a). In that case, racially polarized voting, to the extent that it exists, is not "the overriding criterion in voting" (Dove, 539 F.2d at 1156). It was firmly settled prior to 1982 that no person had the right to be represented by members of any particular group to which he belongs or to participate in an electoral process that maximizes his chances of success, either as a voter or a candidate. Rather, the principle repeatedly endorsed was the right to participate in an electoral process -- to vote, first and foremost, but also to join a political party, to participate in its affairs, to become a candidate (Whitcomb, 403 U.S. at 149-150) -- in which there is no "built-in bias" against the opportunity to participate (id. at 153). /59/ Amended Section 2 reaffirmed these principles. See Senate Report 23-24, 30. It thus follows that where "blacks and white alike have rejected race as the overriding criterion in voting" (Dove, 539 F.2d at 1155-1156), then, since no such "built in bias" exists, "minority candidates (will) not lose elections solely because of their race" (Rogers, 458 U.S. at 628), and the political process is, by definition, "equally open to participation" by minorities (White, 412 U.S. at 766; see Whitcomb, 403 U.S. at 153). In other words, the relevant inquiry is not simply into the existence of bloc voting by race; the court must assess the effect of racial polarization on the opportunity for blacks to participate in the political process. Only where the impact of racial bloc voting in combination with the challenged procedure -- here, multi-member districts -- deprives black voters of equal access to the electoral process is Section 2 offended. 3. Given the electoral success that black candidates have attained with substantial white support in House Districts 21, 23, and 39 -- success equal to or greater than could be expected under single-member districts -- it is difficult to imagine any basis for invalidating these districts under Section 2. /60/ And while black candidates have been less successful in House District 36 and Senate District 22, the district court's findings as to those districts warrant no different result. They show that black candidates have received substantial white voting support. /61/ In one case, a black candidate ran unopposed for a delegate seat, which is significant because the make-up of the candidate slate is indicative of the voting strength of a district's constituencies. Page 22 note 46, supra. It is also significant, as the court's opinion reveals, that there are no present barriers to minority registration, party affiliation, or candidacy; no anti-single shot voting or equivalent requirement has been employed; candidate slating has not been dominated by white voters; and there is no majority vote requirement in general elections. Some or all of these factors were usually present in pre-City of Mobile cases in which multi-member districts were invalidated or were expressed during Congress' consideration of the 1982 amendements as a justification for their enactment. See, e.g., White, 412 U.S. at 623-624; Wallace v. House, 515 F.2d 619, 623-624 (5th Cir. 1975), vacated and remanded on other grounds, 425 U.S. 947 (1976); Zimmer, 485 F.2d at 1305-1306; cf. Whitcomb v. Chavis, supra; Black Voters, 565 F.2d at 6; Bradas, 508 F.2d at 1112; Senate Report 10 n.22; House Report 31 n.105. The absence of such barriers to participation in the electoral process, coupled with the findings made by the court regarding the success that black candidates have had and the white voting support that these candidates have received in House District 36 and Senate District 22, supports the conclusion that the multi-member system has not deprived blacks of the opportunity to participate in the electoral process in these two districts. /62/ CONCLUSION The judgment of the district court should be reversed. Respectfully submitted. CHARLES FRIED Acting Solicitor General WM. BRADFORD REYNOLDS Assistant Attorney General CHARLES J. COOPER Deputy Assistant Attorney General PAUL J. LARKIN, JR. Assistant to the Solicitor General JULY 1985 /1/The state constitutional provision to which the suit referred was a provision adopted in 1968 prohibiting the division of counties for the purpose of creating electoral districts. /2/ The proceedings are described in our earlier brief (at 1-2). /3/ Only two of these districts -- House District 8 and Senate District 2 -- were subject to and had received preclearance under Section 5 of the Voting Rights Act. /4/ The district court found (J.S. App. 51a-52a) that the totality of the following circumstances, in combination with the use of large multi-member districts, diluted minority votes in each of the challenged districts: (1) "the lingering effects of seventy years of official discrimination against black citizens in matters touching registration and voting," (2) "substantial to severe racial polarization in voting," (3) "the effects of thirty years of persistent racial appeals in political campaigns," (4) "a relatively depressed socioeconomic status resulting in significant degree from a century of de jure and de facto segregation," and (5) "the continuing effect of a majority vote requirement." The court also found that in creating the sole single-member district challenged -- Senate District 2 -- the State had diluted black voting strength by fracturing the black community into two districts containing black voting minorities (J.S. App. 52a). Subsequent proceedings are described in our earlier brief (at 3 n.1). /5/ The court did not make a finding for any of the districts regarding voting age population, which is the preferred measure. See City of Rome v. United States, 446 U.S. 156, 186 n.22 (1980); Wyche v. Madison Parish Police Jury, 635 F.2d 1151, 1161-1162 (5th Cir. 1981). /6/ The challenged multi-member districts continued pre-existing districts and their apportionments (J.S. App. 19a). Thus, it is possible to evaluate the plan's dilutive impact, if any, by looking at results from more than one election. /7/ In the 1982 primary election there were only four candidates, two of whom were black, for three positions (J.S. App. 44a). /8/ There were only seven white candidates for eight positions in the primary (J.S. App. 42a). /9/ In addition, the district court observed that a black citizen has been elected mayor of the City of Charlotte, receiving 38% of the white vote in the general election against a white Republican (J.S. App. 35a). /10/ The district court did not make a finding for Carrabus County (see J.S. App. 24a-25a n.22). /11/ We will not discuss House District 8 and Senate District 2, because appellants' challenge to the district court's conclusion as to those districts is not within the scope of the Court's notation of probable jurisdiction. /12/ At the same time, the legislative history of amended Section 2 is complicated, variegated, and, on occasion, contradictory. The language ultimately incorporated into this provision was proposed by Senator Dole as a means of resolving a deadlock in the Senate Judiciary Committee that arose after the Senate Constitution Subcommittee had rejected the House version of Section 2. In this setting, undue emphasis must not be given to the views of any one faction in the controversy. The legislative history must be understood in terms of its dominant movement and fundamental purposes. Statements of the majority in the Senate Report, while illuminative on many issues, must be evaluated against the record established before the Congress as a whole and particularly against statements of the additional views of individual members who insisted upon and supported the compromise. The statements of Senator Dole, the sponsor of the compromise, must also be given particular weight. /13/ H.R. 3198, 97th Cong. 1st Sess. (1981). See Extension of the Voting Rights Act: Hearings Before the Subcomm. on Civil and Constitutional Rights of the House Comm. on the Judiciary, 97th Cong., 1st Sess. Pt. 1, at 2 (1981) (hereinafter cited as House hearings). /14/ See generally Boyd & Markman, The 1982 Amendments to the Voting Rights Act: A Legislative History, 40 Wash. & Lee L. Rev. 1347, 1354-1379 (1983) (hereinafter cited as Boyd & Markman). /15/ The disclaimer provided: The fact that members of a minority group have not been elected in numbers equal to the group's proportion of the population shall not, in and of itself, constitute a violation of this section." H.R. Rep. 97-227, 97th Cong., 1st Sess. 48 (1981) (emphasis added)(hereinafter cited as House Report). Although he had sponsored the disclaimer, Representative Hyde later concluded that it failed to achieve its purposes. See 1 Voting Rights Act: Hearings on S.53, et al. Before the Subcomm. on the Constitution of the Senate Comm. on the Judiciary, 97th Cong., 2d Sess. 407-408 (1982) (testimony of Rep. Hyde) (hereinafter cited as Senate Hearings); id. at 886-887 (letter from Rep. Hyde to Sen. Hatch). /16/ Senators Kennedy and Mathias (and more than 60 co-sponsors) introduced S.1992, 97th Cong., 1st Sess. (1981), which was identical to H.R. 3112. 127 Cong. Rec. S15694 (daily ed. Dec. 16, 1981). Senator Grassley introduced S.1975, 97th Cong., 1st Sess (1981). 127 Cong. Rec. S15372 (daily ed. Dec. 15, 1981). /17/ See, e.g., 1 Senate Hearings 199 (statement by Sen. Mathias); id. at 256, 265 (testimony of Benjamin L. Hooks, Exec. Dir., NAACP); id. at 290-291 (testimony of Vilma Martinez, Pres., MALDEF); id. at 831-819 (Prepared Statement of Armand Derfner). Another criticism was that the intent test fostered racial divisiveness by requiring a person to be branded as a racist before a violation could be found. See id. at 1181 (Prepared Statement of Arthur Fleming, Chairman, U.S. Comm'n on Civil Rights). /18/ A complete discussion of the objections to the results test is contained in the Subcommittee's Report. See S.Rep. 97-417, 97th Cong., 2d Sess. 108-111, 127-158, 169-173 (1982) (hereinafter cited as Senate Report) (Voting Rights Act: Report of the Subcomm. on the Constitution of the Senate Judiciary Comm.) (hereinafter cited as Subcomm. Report); see also Boyd & Markman 1396-1406 (discussing Subcommittee's objections). /19/ See, e.g., 1 Senate Hearings 516 (statement of Sen. Hatch); id. at 229-231 (testimony of Prof. Walter Berns); id. at 407-408 (testimony of Rep. Hyde); id. at 424-432 (testimony of Prof. Barry R. Gross); id. at 655 (testimony of Prof. John Bunzel); id. at 1438 (testimony of Prof. Irving Younger). See generally Subcomm. Report 142-146. /20/ Subcomm. Report 137. /21/ See, e.g., 1 Senate Hearings 200 (Prepared Statement of Sen. Kennedy) ("The courts have made clear that under the standard in our bill there is no right to a quota to or proportional representation, even in the context of at large elections"); id. at 243 (Benjamin L. Hooks, Exec. Dir., NAACP); id. at 283, 287 (Memorandum of Ralph G. Neas, Exec. Dir., Leadership Conf. on Civil Rights); id. at 796 (testimony of Armand Derfner, Voting Rights Project). /22/ As Armand Derfner, head of the Voting Rights Project, put it, "(t)he precise proof might vary, but the essential element of proving that the racial minority was 'shut out,' i.e., denied access -- not simply to winning offices but to the opportunity to participate in the electoral system -- was always required (under pre-City of Mobile cases)." 1 Senate Hearings 810 (Prepared Statement); see also, e.g., id. at 223 (Prepared Statement of Sen. Kennedy); id. at 626 (testimony of David Walbert). /23/ 1 Senate Hearings 368 (testimony of Laughlin McDonald, Southern Regional Dir., ACLU). /24/ See, e.g., 1 Senate Hearings 201 (testimony of Sen. Mathias); id. at 223 (Prepared Statement of Sen. Kennedy) ("effectively shut out of a fair opportunity (to) participate in the election"); id. at 810, 819-820 (Prepared Statement of Armand Derfner). /25/ Other supporters of the results standard made the same point. See, e.g., 1 Senate Hearings 305 (Prepared Statement of Vilma S. Martinez, President, MALDEF) ("The issue then, is not proportional representation, but equal access to the political process"); id. at 372 (Laughlin McDonald, Southern Regional Dir., ACLU) ("What those (pre-City of Mobile) cases do is establish equality of access"). See also id. at 223 (Prepared Statement of Sen. Kennedy); id. at 275-276 (Prepared Statement of Benjamin L. Hooks, Pres. NAACP); id. at 283, 286-287 (Memorandum form Ralph G. Neas, Exec. Dir., Leadership Conf. on Civil Rights); id. at 305 (Prepared Statement of Vilma S. Martinez, Pres., MALDEF); id. at 706 (Memorandum from Frank R. Parker, Lawyers' Comm. for Civil Rights Under Law). /26/ There was little debate in the House, and, with one exception, no one disagreed with the thrust of Senator Dole's position that "equal access" and an "equal opportunity to participate" was the standard for amended Section 2. See 128 Cong. Rec. H3840-H3841 (daily ed. June 23, 1982) (Rep. Edwards); id. at H3841 (Rep. Sensenbrenner); id. at H3842 (Rep. Hyde); id. at H3846 (Rep. Butler). But see id. at H3844 (Rep. Lungren) (describing standard in terms of intent). /27/ Senate Report 193 (Additional Views of Sen. Dole); Senate Report 33; House Report 30; 128 Cong. Rec. S6647 (daily ed. June 10, 1982) (Sen. Grassley); id. at S6920 (daily ed. June 17, 1982) (Sen. Hatch); id. at S6961 (Sen. Dole); 18 Weekly Comp. Pres. Doc 846 (June 19, 1982) (President's signing statement). /28/ As Senator Kennedy explained his version: "Section 2, as amended would not make mere failure of minorities to win proportional representation a violation, even if that came as the result of at large elections. Plaintiffs would have to prove additional factors establishing that, in the total circumstances minority voters not only failed 'to win' but were effectively shut out of a fair opportunity (to) participate in the election." 1 Senate Hearings 223 (emphasis in original) (Prepared Statement). /29/ See Subcomm. Report 139-147. /30/ Benjamin Hooks, Executive Director of the NAACP, made this point during the Senate hearings: "I would say that -- and let me be very frank -- simply proven results would not be enough to trigger the mechanism of Section 2. It would only trigger it if the results were caused by some practice. Results simply trigger looking at the practices; that is all." 1 Senate Hearings 267; see also, e.g., id. at 283 (Memorandum of Ralph G. Neas, Exec. Dir., Leadership Conf. on Civil Rights); id. at 420 (Laughlin McDonald, Southern Regional Dir., ACLU) ("I do not know of a single case * * * that says the mere absence of blacks from office is ever enough to violate either section 2 of the 14th or the 15th amendment. Not only are there no cases that have ever said that, but every case says precisely the opposite"); id. at 957 (Prof. Norman Dorsen); id. at 987 (Prepared Statement of Joseph L. Rauh, Jr.) /31/ Because the Senate endorsed this principle as well as the Court's decisions in Whitcomb and White which had enunciated it, the statement in the House report that the consistent defeat of minority or minority-backed candidates in at at-large system itself would establish a violation of amended Section 2 (House Report 30-31) does not express Congress' intent. See also page 17 note 39, infra. /32/ See, e.g., Black Voters v. McDonough, 565 F.2d 1 (1st Cir. 1977); Hendrix V. Joseph, 559 F.2d 1265 (5th Cir. 1977); Dove v. Moore, 539 F.2d 1152 (8th Cir. 1976; Zimmer v. McKeithen, 485 F.2d 1297 (5th Cir. 1973) (en banc), aff'd on other grounds sub nom. East Carroll Parish School Bd. v. Marshall, 424 U.S. 636 (1976); see also 1 Senate Hearings 1216-1226 (Appendix to Prepared Statement of Frank R. Parker, Lawyers' Comm. for Civil Rights Under Law) (collecting cases). The Court discussed these factors in Rogers v. Lodge, 458 U.S. 613, 619-620 n.8 (1982). /33/ See House Report 30 & n.104; Senate Report 27-30; id. at 104 n.24 Paragraph 6 (Additional Views of Sen. Hatch); id. at 194 (Additional Views of Sen. Dole); id. at 198 (Supplemental Views of Sen. Grassley); 128 Cong. Rec. S6941 (daily ed. June 17, 1982) (Sen. Mathias); id. at S6961 (Sen. Dole); id. at H3841 (daily ed. June 23, 1982) (Rep. Edwards). /34/ See, e.g., Grove City College v. Bell, No. 82-792 (Feb. 28, 1984), slip op. 11; North Haven Bd. of Educ. v. Bell, 456 U.S. 512, 527 (1982). This is particularly true given Senator Dole's pivotal role in the adoption of amended Section 2 and the absence of a conference report on the Act. See North Haven, 456 U.S. at 527. /35/ E.g., 128 Cong. Rec. S6655 (daily ed. June 10, 192) (Sen. Boren)); accord, id. at S6500 (daily ed. June 9, 1982) (Sen. Stevens) ("the issue to be decided under the results test is whether the political processes are equally open to minority voters"). /36/ 128 Cong. Rec. S6560 (daily ed. June 9, 1982) (Sen. Kennedy); id. at S6557 (Sen. Stevens). /37/ E.g., id. at S6779 (daily ed. June 15, 1982) (Sen. Specter). /38/ Accord, e.g., id. at S6647 (daily ed. June 10, 1982) (Sen. Grassley); id. at S6717 (daily ed. June 14, 1982)(Sen. Tower); id. at S6717-S6718 (daily ed. June 17, 1982) (Sen. Moynihan); id. at S6964 (Sen. Kennedy); ibid. (Sen. Heflin(; id. at S7110 (daily ed. June 18, 1982) (Sen. Metzenbaum); id. at S7118 (Sen. Sasser); id. at S7138 (Sen. Robert Byrd). As Senator Robert Byrd put it, "(t)he law seeks to protect the right to vote, not the ability to be guaranteed election." Ibid. /39/ The legislative background to amended Section 2 also makes this point clear in another way. Under amended Section 5 of the Act, jurisdictions with a history of discrimination touching upon voting may not obtain approval to enforce changes in their election laws that have the effect of causing a retrogression in the position of minorities with respect to their exercise of the franchise. City of Lockhart v. United States, 460 U.S. 125, 133-136 (1983); Beer v. United States, 425 U.S. 130, 137 (1976). The legislative history of amended Section 2, however, conclusively shows that the Section 5 retrogression standard was not incorporated into Section 2. Senate Report 68; id. at 104 n.24 Paragraph 8 (Supplemental Views on Sen. Hatch); 128 Cong. Rec. H3841 (daily ed. June 23, 1982) (remarks of Rep. Sensenbrenner with Rep. Edwards concurring); id. at S7095 (daily ed. June 18, 1982) (Sen. Kennedy); id. at S6930 (daily ed. June 17, 1982) (Sen. DeConcini); 2 Senate Hearings 80 (Statement of Sen. Dole); 1 Senate Hearings 414 (testimony of Laughlin McDonald, Southern Regional Dir., ACLUE); id. at 449 (testimony of Mayor Henry L. Marsh); id. at 801 (testimony of Armand Derfner); id. at 891-892 (colloquy between Rep. Sensenbrenner and Sen. Grassley); id. at 1254 (colloquy between Subcomm. Counsel Markman and Julius L. Chambers, Pres., NAACP Legal Defense Fund); id. at 1575-1576 (Prepared Statement of Nathan Z. Dershowitz, Amer. Jewish Congress). The Senate report expressly states that "(p)laintiffs could not establish a Section 2 violation merely by showing that a challenged reapportionment * * * involved a retrogressive effect on the political strength of a minority group" (Senate Report 68 n.224). In other words, while a retrogressive effect may be relevant evidence, access, not effect, is the touchstone of a Section 2 inquiry. While some courts have said that retrogression alone may violate amended Section 2, those courts have failed to consider the above legislative history. See Ketchum v. Byrne, 740 F.2d 1398, 1407 (7th Cir. 1984), cert. denied, No. 84-627 (June 3, 1985); Buskey v. Oliver, 565 F.Supp. 1473, 1482 (M.D. Ala. 1983). /40/ See, e.g., Whitcomb, 403 U.S. at 150 (footnote omitted)(where "ghetto votes were critical to Democratic Party success * * * it seems unlikely that the Democratic Party could afford to overlook the ghetto in slating its candidates"); Dove v. Moore, 533 F.2d at 1153, 1155 n.4 (noting that local voters "have a strong affinity for incumbents" and that each candidate's 40% black constituency "cannot be ignored with impunity"). See also Seamon v. Upham, No. P-81-49-CA (E.D. Tex. Jan. 30, 1984), aff'd sub nom. Strake v. Seamon, No. 83-1823 (Oct. 1, 1984); page 22 note 46, infra. /41/ Compare, e.g., Harper & Row, Publishers, Inc. v. Nation Enterprises, No. 83-1632 (May 20, 1985), slip op. 20, and Strickland v. Washington, No. 82-1554 (May 14, 1984), slip op. 27-28, with Wainwright v. Witt, No. 83-1427 (Jan 21, 1985), slip op. 15-17, and Patton v. Yount, No. 83-95 (June 26, 1984), slip op. 11-13 & n.12. See generally Bose Corp. v. Consumers Union of United States, Inc., No. 82-1246 (Apr. 30, 1984). In this respect, the inquiry under amended Section 2 is similar to the type of analysis that appellate courts follow in determining whether a particular adjudicative procedure is consistent with due process (e.g., Walters v. National Ass'n of Radiation Survivors, No. 84-571 (June 28, 1985)) and whether a state law violates the First Amendment Establishment Clause (e.g., Grand Rapids School Dist. v. Ball, No. 83-990 (July 1, 1985)). /42/ The decisions of this Court and the lower courts both before City of Mobile and after passage of amended Section 2 also make this point clear. These decisions have engaged in a far more searching review of a district court's analysis than application of Rule 52(a), which appellees advocate here, would permit. See, e.g., Whitcomb v. Chavis, 403 U.S. at 144-155; Jones v. City of Lubbock, 737 F.2d 364, 383-386 (5th Cir. 1984); Hendrix, 559 F.2d at 1268-1271; David v. Garrison, 553 F.2d 923, 929-931 (5th Cir. 1977); Bradas v. Rapides Parish Police Jury, 508 F.2d 1109, 1112-1113 (5th Cir. 1975). The court in Velasquez v. City of Abilene, 725 F.2d 1017, 1021 (5th Cir. 1984), thus erred in stating that an ultimate Section 2 finding is a question of fact. The court was mistaken as to the central question to be answered under the statute. Pages 14-17, supra. /43/ In Seamon, the district court rejected a Section 2 claim that minority voters were entitled to a "'safe' district in which the minority population approaches 65% of the overall population" (No. P-81-49-CA (E.D. Tex. Jan. 30, 1984), slip op. 11-12). Under the challenged plan, minority voters, while not guaranteed the ability to elect one a representative of their choice, were found to "exert a significant impact" and to "play pivotal roles in key elections" in two high minority impact districts (id. at 15). Similarly, in Brooks, the plaintiffs urged the three-judge district court to create a congressional district with a 64% black population minimum on the ground that, because of low voter registration and turnout among blacks, they would be unable to elect candidates of their choice with a lesser percentage. In rejecting the super-majority plans proposed by the plaintiffs, the court noted that "(a)mended Section 2 * * * does not guarantee or insure desired results, and it goes no further than to afford black citizens an equal opportunity to participate in the political process" (No. GC82-80-WK-O) (N.D. Miss. Apr. 16, 1984), slip op 15). This Court's summary affirmances in Seamon and Brooks establish that minorities do not have a right under Section 2 to the creation of "safe" minority-controlled districts, even where other objective factors contribute to the finding of a violation of Section 2 under the "totality of the circumstances." Moreover, as we explained in our brief (at 8-19) in City Council v. Ketchum, cert. denied, No. 84-627 (June 3, 1985) (a copy of which has been served upon the parties), creation of super-majority districts as a matter of law is inappropriate to remedy a Section 2 violation. /44/ The district court correctly recognized that a lawful state policy regarding a particular electoral practice is entitled to weight (Whitcomb, 403 U.S. at 149; see Upham v. Seamon, 456 U.S. 37 (1982)), but erred by disregarding the North Carolina policy against splitting legislative districts (J.S. App. 49a-50a). The court acknowledged that "the state adduced fairly persuasive evidence that the 'whole-county' policy was well-established historically, had legitimate functional purposes, and was in its origins completely without racial foundation" (id. at 50a). But the court held that "that all became irrelevant as matters developed in this particular legislative plan" (ibid.) because the legislature chose to split counties "only when necessary to meet population deviation requirements or to obtain Section 5 preclearance" (ibid. (emphasis added)). That reasoning is plainly in error. The fact that the state adhered to its policy except where necessary to ensure that each voter -- black and white -- had his vote counted equally and to ensure that the reapportionment plan did not cause a retrogression in the political strength of black voters (see page 17 note 39, supra) surely counts in the state's favor. /45/ The population percentages in the five counties may overestimate the actual voting strength of minorities, because the percentage voting age population in these districts may be less than the population percentage. See page 3 note 5, supra. /46/ As Prof. Archibald Cox informed the Senate Subcommittee, "(v)oters in a minority group may have exactly the same opportunities for participation as other voters, even though no members of the group are elected to office. The minority may not vote as a bloc. The minority may vote as a bloc but make its influence felt in the selection of non-minority candidates for election, in framing their programs and policies, and in support of one or more candidates against their opponents." 1 Senate Hearings 1428 (Prepared Statement). Indeed, in the 1982 primaries in House Districts 23 and 36, whites did not field a candidate for each of the open positions. Pages 4-5 notes 7-8, supra. That fact reinforces the conclusion that blacks have not been denied equal access to the electoral process in these districts by virtue of the multi-member plan, because the make up of the candidate slate is itself a reflection of and a response to the voting strength of the various constituencies in a district. /47/ Appellees seek to minimize the significance of this electoral success on the ground (Mot. to Dis. 26-27) that the 1982 election year was "obviously aberrational" -- attributing this conclusion to the district court. However, the district court's words have been taken out of context. The court's finding (J.S. App. 37a (footnote omitted)) was as follows: There are intimations from recent history, particularly from the 1982 elections, that a more substantial breakthrough of success could be imminent -- but there were enough obviously aberrational aspects present in the most recent elections to make that a matter of sheer speculation. In a footnote, the court observed that both parties had offered evidence to establish either that the 1982 elections presaged a "breakthrough" or that they were "aberrational." The court stated that its "finding" in text (quoted above) "reflects our weighing of these conflicting inferences" (id. at 37a n.27). It is thus inaccurate for appellees to assert that the district court adopted their view that the 1982 elections should be disregarded as "aberrational." In fact, the most that can be said is that the court rejected the opposing view -- that the 1982 election results should be deemed evidence that black candidates would achieve even greater success in the "imminent" future. /48/ It is inappropriate to conclude, as some courts have done, that the state must prove that the existence of past discrimination has not reduced the current potential electoral success of black candidates. McMillan v. Escambia County, 748 F.2d 1037, 1045 (5th Cir. 1984). That approach misconstrues the governing legal standard, improperly shifts the burden of proof, and requires proven and continued minority electoral success to avoid Section 2 liability. Neither Congress nor Senator Dole had any such requirement in mind. Pages 12-17, supra, and pages 27-28, infra. /49/ Appellees claim (Mot. to Dis. 27, 41; Supp. Br. 10 & n.9) that the district court's disparagement of black electoral success in the challenged districts is supported by language in the Senate majority report, a document which, we have argued, cannot be taken as determinative on all counts. In any event, the report simply notes (Senate Report 29 n.115) that the election of a "few" minority candidates should not be deemed conclusive because it would enable election officials to evade amended Section 2 by engineering the election of "a 'safe' minority candidate." The case cited by the report to illustrate this caveat, Zimmer, arose in a context "where the multi-member system was devised, despite historic policy and a state statute forbidding it, in reaction to a dramatic voter registration drive directed at blacks, who, although comprising 58 per cent of the parish's population, had not been permitted to vote until 1962" (Black Voters, 565 F.2d at 4). Given these circumstances, an "abrupt change in policy -- which coincided with increased black voter registration" (Wallace, 515 F.2d at 631), Zimmer declined to treat recent black electoral success as dispositive. Appellees have failed to prove that black electoral success in these districts is attributable to 11th hour efforts by the General Assembly to engineer the election of "safe" minority candidates to thwart a Section 2 claim. Indeed, the district court made no mention of any evidence that would tend to support such a claim. Moreover, the district court noted that "in recent years there has been a measurable increase in the ability and willingness of black citizens to participate in the state's political processes and in its government at state and local levels" (J.S. App. 47a). The district court discounted this increased participation because of its finding of racial polarization (ibid.), but that finding is flawed in several respects (see pages 28-33, infra). /50/ The court's reasoning is also flawed in another respect. Although the district court made factual findings on a district-by-district basis, it drew its ultimate legal inferences regarding racial bloc voting and the effect on minority electoral opportunities on the basis of "(t)he overall results achieved to date at all levels of elective office" (J.S. App. 37a). It is only on such a basis that the court could have held that black electoral success is "minimal" in a district such as House District 39, where the 25.1% black minority has, with substantial white support, elected 40% of the at-large representatives. To invalidate a specific district on the basis of generalized statewide results at "all levels of elective office" is a clear legal error. See White v. Regester, 412 U.S. at 769 (requiring an "intensely local appraisal" of the electoral scheme). /51/ As Senator Heflin stated, "(t)he Dole compromise has a now application but allows for a consideration of yesterday factors as well as present day good faith efforts to remedy past mistakes if the yesterday factors touch on the new result." 128 Cong. Rec. S6964 (daily ed. June 17, 1982). /52/ This history may have had an effect in House District 36 and Senate District 22, given the electoral results in those districts; but, viewed in combination with other factors, it appears not to have shut blacks out of the electoral process there (see pages 32-34, infra). Given the fact that minorities have been elected to office in House Districts 21, 23, and 39 in numbers at least as great as would be expected under a single-member system, the historical discrimination found by the district court does not appear to have affected the electoral opportunities that black voters enjoy in those districts. /53/ The district court thus plainly erred by relying (J.S. App. 29a) upon inoperative numbered seat and anti-single shot voting requirements of state law. As the court itself noted (id. at 23a-24a), those requirements were invalidated more than a decade ago (Dunston v. Scott, 336 F.Supp. 206 (E.D.N.C. 1972)), and there is no basis in amended Section 2 (or logic) for concluding that these now-repealed legal measures could have any current effect on the multimember system. See pages 12-17, supra (discussing Sen. Dole's compromise). /54/ The district court plainly misconstrued the significance of Congress' rejection of the proportional representation standard. The court dismissed the "proportional representation" disclaimer in Section 2(b), 42 U.S.C. 1973(b), as meaning no more than that the fact that blacks have not been elected in numbers proportional to their percentage of the population "does not alone establish that vote dilution has resulted" (J.S. App. 15a & n.13 (emphasis added)). As discussed above (pages 9-17), the disclaimer was expressly drafted to avoid any such narrow interpretation. In effect, the district court has interpreted the Act as imposing a "proportional representation plus" standard, rather than an "equal opportunity" standard, as Congress intended. /55/ As Armand Derfner explained to the Senate Subcommittee (1 Senate Hearings 803): "the at-large elections that I * * * have been focusing on are those in which the result of those at-large elections is basically to shut out the minority voters. It is not a question of whether they will get more or less or whether the majority voters will get more or less. It is a question of some versus nothing." /56/ As the Court explained in Whitcomb (403 U.S. at 153), where "the failure of the ghetto to have legislative seats in proportion to its population emerges more as a function of losing elections than of built-in bias against poor Negroes * * * (t)he voting power of ghetto residents may have been 'cancelled out' * * * but this seems a mere euphemism of political defeat at the polls." See also United Jewish Orgs. v. Carey, 430 U.S. 144, 166 n.24 (1977) (plurality opinion) ("if voting does not follow racial lines, the white (or black) voter has little reason to complain that the percentage of nonwhites (or whites) in his district has been increased"). It is erroneous, however, to conclude that proof of racial bloc voting atop numerical underrepresentation together are sufficient to establish a violation of amended Section 2, as some courts have said. See McMillan, 748 F.2d at 1043; United States v. Marengo County Comm'n, 731 F.2d 1546, 1566 (11th Cir. 1984). Supporters of the Act stated that proof of more than numerical underrepresentation and racial bloc voting is essential to establish a Section 2 violation. See 1 Senate Hearings 819-820 (Prepared Statement of Armand Derfner) (emphasis in original) ("amended section 2, like White v. Regester, applies only in that small category of places where there is no functioning system of politics for minority voters, where there is already severe racial division, and where it is simply impossible for minority voters to have any significant opportunity under the election system as it is"); accord, e.g., id. at 287 (Memorandum of Ralph G. Neas, Exec. Dir., Leadership Conf. on Civil Rights); id. at 564 (testimony of Joaquin G. Avila, Assoc. Gen. Counsel, MALDEF); id. at 1184 (testimony of Frank Parker, Dir., Voting Rights Project, Lawyers' Comm. for Civil Rights Under Law). /57/ In most vote dilution cases, a plaintiff can establish a prima facie case of racial bloc voting by using a statitiscal analysis of voting patterns that compares the race of a candidate with the race of the voters. A defendant can then introduce its own study, which takes other factors into account, to rebut a plaintiff's prima facie case. For a discussion, in a different context, of the type of statitical studies that can be used, see McCleskey, v. Zant, 580 F.Supp. 338, 352-379 (N.D. Ga. 1984), affD, 753 F.2d 877 (11th Cir. 1985) (en banc). Resort to such analyses has been approved. As Judges Higginbotham and Wisdom have cogently observed, "race or national origin may mask a host of other explanatory variables" including "explanatory factors * * * as intuitively obvious as campaign expenditures, party identification, income, media use measured by cost, religion, name, identification, or distance that a candidate lived from a particular precinct" (Jones, 730 F.2d at 235 (Higginbotham, J., specially concurring); Lee County, 748 F.2d at 1482 (Wisdom, J.)). /58/ See 1 Senate Hearings 821 (emphasis added) (Prepared Statement of Armand Derfner) ("Section 2, of course, will apply only in those places where there is already an extraordinary amount of (racial) division"). Other witnesses also described racial bloc voting in less absolute terms than the district court. See id. at 306 (Prepared Statement of Vilma S. Martinez, President, MALDEF) (emphasis added) ("'It is a situation where, when candidates of different races are running for the same office, the voters will by and large vote for the candidate of their own race'") (citation omitted); id. at 543 (testimony of Prof. Susan A. MacManus (emphasis added) ("racial polarization * * * occurs when citizens of one racial group uniformly vote for one candidate and citizens of another racial group uniformly vote for another. * * * (T)he basic purpose of the test (for calculating racial polarization) is to determine whether race is the primary and exclusive determinant of individual voting decisions across time in any given community"). /59/ See City of Mobile, 446 U.S. at 75-80 (plurality opinion); id. at 86 (Stevens, J., concurring in the judgment); id. at 111 n.7 (Marshall, J., dissenting); United Jewish Orgs., 430 U.S. at 165-168 (plurality opinion); Beer, 425 U.S. at 136 n.8; Whitcomb, 403 U.S. at 149-160; Taylor v. McKeithen, 499 F.2d 893, 905 (5th Cir. 1974) (Wisdom, J.); Turner v. McKeithen, 490 F.2d 191, 197 (5th Cir. 1973) (Brown, C.J.). /60/ In House Districts 21, 23, and 39, where black candidates have been elected in numbers at least as great as would be expected under a single-member plan, black candidates have recieved substantial white support. In House District 21, the black candidate in the 1978 primary (Blue) received 21% of the white vote, but he later increased his share of the white vote in 1980 to 31% in the primary and 44% in the general election and was elected; in 1982 he again increased his share to, respectively, 39% and 45% of the white vote in the primary and general election and was reelected. J.S. App. 44a. In House District 23, a black Republican ran in the 1978 general election and received more white votes (17%) than black votes (5%). The black candidate was unopposed in the 1978 general election, and in the 1980 primary and general election. Nontheless, he received 16% and 37% of the white vote in the 1978 primary and general election, 40% of the white vote in the 1980 general election, and 37% and 43% of the vote in the 1982 primary and general election, respectively. J.S. App. 43a-44a. In House District 39, the two black representatives elected in 1982 received 25% and 36% of the white vote in the primary, and 42% and 46% of the white vote in the general election. One of those representatives had previously received 40% and 32% of the white vote in the 1980 primary and general election, respectively. In 1978, a black republican candidate received more white votes (33%) than black votes (25%) in the general election. J.S. App. 42a-43a. /61/ In House District 36, the black representative elected in 1982 received 50% of the white vote in the primary and 42% in the general election. Another, unsuccessful black candidate in that race received 39% and 29% of the white vote in the primary and general election, respectively. This was an increase from 1980, when a different black candidate received 22% and 28% of the white vote in the primary and general election, respectively. In Senate District 22, the black member of the four-person delegation from 1975-1980 received 47% of the white votes in the 1978 primary and 41% in the general election. A second black candidate (Polk) ran in 1982 and garnered 32% of the white vote in the primary and 33% in the general election. J.S. App. 42a. Moreover, while blacks form only 31% of the population of the city of Charlotte, a black Democratic candidate was elected mayor with 38% of the white vote against a white Republican. J.S. App. 35a. This figure is significant because it shows that in a head-to-head contest more than one-third of the white voters were willing to vote for a black candidate in Charlotte. Blacks also held 28.6% of the district and 16.7% of the at-large city counsel seats from 1977-1981. J.S. App. 34a. /62/ Should the Court nonetheless conclude that there is an insufficient basis in the record for finding no violation of amended Section 2 with respect to these two districts, then, given the district court's reliance upon an incorrect legal standard, the appropriate disposition would be to remand the case to the district court for further proceedings under the correct legal standard. See Pullman-Standard v. Swint, 456 U.S. 273, 291-292 (1982).