[PUBLISH]




                  IN THE UNITED STATES COURT OF APPEALS

                           FOR THE ELEVENTH CIRCUIT
                                                                       FILED
                                      No. 96-9284         U.S. COURT OF APPEALS
                                                            ELEVENTH CIRCUIT
                         D. C. Docket No. 1:90-CV-1001-RCF       10/30/98
                                                             THOMAS K. KAHN
                                                                  CLERK


TYRONE BROOKS, LANETT STANLEY, et al.,

                                                                     Plaintiffs-Appellants,

                                         versus

ZELL MILLER, Governor of Georgia, GEORGIA STATE
BOARD OF ELECTIONS, et al.,

                                                                   Defendants-Appellees.



                      Appeal from the United States District Court
                         for the Northern District of Georgia

                                   (October 30, 1998)

Before DUBINA and MARCUS, Circuit Judges, and PROPST*, Senior District Judge.

DUBINA, Circuit Judge:

__________________________
*Honorable Robert B. Propst, Senior U.S. District Judge for the Northern District of
Alabama, sitting by designation:
       In this voting rights action, Plaintiffs, 27 black Georgia residents and voters,

challenge the majority vote requirement for primary elections in Georgia, set forth in

O.C.G.A. § 21-2-501. The complaint was certified as a class action on behalf of all

present and future black registered voters in Georgia. Plaintiffs contend that the majority

vote requirement, also known as the primary runoff requirement, violates § 2 of the

Voting Rights Act, 42 U.S.C. § 1973, as well as the First, Fourteenth, and Fifteenth

Amendments to the United States Constitution. After a bench trial, the district court ruled

that Georgia’s majority vote provision for primary elections is constitutional and does not

violate § 2.



                                         I. FACTS

       O.C.G.A. § 21-2-501 originally required that a candidate in a primary or general

election receive a majority of the votes cast in order to be nominated or elected and

provided for a runoff election in the event that no candidate received a majority. In 1994,

after the filing of this action, Georgia’s General Assembly repealed the majority vote

requirement for all general elections, except those for certain constitutional offices not at

issue here, and replaced it with a 45 percent plurality rule. See O.C.G.A. § 21-2-501(b),

as amended by Ga. L. 1994, p. 279, § 11; O.C.G.A. § 21-2-2(18.1) (defining plurality as

45 percent of the total votes). In this action, Plaintiffs do not challenge the 45 percent

plurality rule for general elections or majority voting as it affects multi-member offices

such as school boards and county commissions. What Plaintiffs do challenge is the

                                              2
majority vote requirement for primary elections for single member, county-level offices,

members of the General Assembly, superior court judges and district attorneys, and

judges and justices of the Georgia Court of Appeals and Georgia Supreme Court.1

       In 1990, after conducting a five-day hearing on Plaintiffs’ motion for preliminary

injunctive relief, the district court denied the Plaintiffs’ motion. Shortly thereafter, the

United States Department of Justice filed a parallel case which was consolidated with this

action. Before trial, the United States filed a motion to voluntarily dismiss its complaint

which the district court granted.

       The district court made extensive findings of fact in this case. The district court

relied on testimony presented at the four-day bench trial in 1996 and testimony presented

at the hearing held in 1990 on Plaintiffs’ motion for preliminary injunctive relief, along

with other exhibits and reports submitted by the parties. In order to resolve the Plaintiffs’

claims, the district court focused its findings of fact on two key areas. First, the district

court examined evidence relating to the process by which the majority vote provision was

enacted into law to determine whether it was passed for a racially discriminatory purpose.

Second, the district court considered evidence presented on the issue of whether the




       1
        Georgia is one of nine states with a majority vote requirement for primary
elections. See Ala. Code § 17-16-36 (1995); Ark. Code Ann. § 7-7-102 (Michie 1993);
Fla. Stat. Ann. § 100.061 (West 1982); O.C.G.A. § 21-2-501 (1993 & Supp. 1997); La.
Rev. Stat. Ann. § 18:511 (West 1979 & Supp. 1998); Miss. Code Ann. 23-15-191 (1990);
Okla. Stat. Ann. tit. 26, § 1-103 (West 1991); S.C. Code Ann. § 7-17-600 (Law. Co-op.
1977); Tex. Elec. Code Ann. § 172.003 (West 1986).

                                               3
majority vote law has a discriminatory impact. The following is a summary of the district

court’s factual findings. A.       Discriminatory Purpose

       The district court acknowledged Georgia’s long history of racial discrimination at

all levels of government. The voting strength of blacks has historically been diminished

in Georgia in numerous ways, including property ownership requirements, literacy tests,

and the use of the county unit system which undermined the voting power of counties

with large black populations. The county unit system gave each county twice as many

unit votes for the purpose of nominating candidates for statewide office as the county had

representatives in the Georgia House. This system had the effect of increasing the voting

power of rural counties and diluting that of urban areas. In 1962, a three-judge federal

panel struck down Georgia’s county unit system as unconstitutional. See Sanders v.

Gray, 

(N.D. Ga. 1962).

       The majority vote provision at issue in this case was enacted as part of a sweeping

election reform bill signed into law in 1964 by then-Governor Carl E. Sanders (“Governor

Sanders” or “Sanders”). This law was Georgia’s first election code. Prior to 1964,

Georgia’s election system was chaotic. The rules for primary and general elections varied

from county to county and election to election. Corruption and manipulation were

commonplace. Throughout the state, and particularly in rural counties, groups known as

“courthouse crowds” controlled local elections and often manipulated election practices

to maintain their own power. Georgia was essentially a one-party state, and a victory in a

Democratic primary was tantamount to election.

                                             4
       One form of election manipulation used under the plurality system consisted of

entering a “stalking horse” into a local race to split the opposition vote and assure victory

to courthouse crowd candidates. Governor Sanders experienced this tactic first hand

when he ran against Peter Zack Greer in the 1962 lieutenant governor’s race. Sanders

withdrew from the race after hearing that Greer planned to enter a candidate named Carl

F. Sanders in the race to confuse voters and ensure his own victory. Sanders ran for

governor instead and won.

       When the Georgia legislature created the first Election Laws Study Committee

(“ELSC”) in 1957 to examine election practices and propose legislation, members of this

committee were interested in maintaining the discriminatory status quo through measures

such as literacy tests. The first ELSC did not produce any comprehensive election

reform. A second ELSC was formed in 1961 but dissolved without taking any action.

       In 1963, after the county unit system had been struck down as unconstitutional,

Denmark Groover (“Groover”), a state representative known for his staunch

segregationist views, attempted to pass a majority vote requirement. There is no doubt

that Groover’s majority vote bill was the product of racial animus. He opposed “bloc

voting,” a euphemism for black citizens voting in a bloc. Groover’s bill passed in the

House but died in committee in the Georgia Senate.

       Governor Sanders called for the creation of a third ELSC in 1963 for the purpose

of drafting an election code. There were more racially moderate members on this

committee than on the previous ELSCs. The third ELSC recommended a majority vote

                                              5
requirement, in addition to numerous other measures, as part of a comprehensive election

code. The General Assembly adopted most of these recommendations, including the

majority vote requirement. The ELSC’s recommendations included some discriminatory

measures such as a scaled-down version of a literacy test. However, some of the

proposals of the third ELSC were either not related to race or promoted increased black

participation in elections. For example, the 1964 code required that a voting registrar be

present at the county courthouse during all business hours and that voter registration

remain open until 50 days before an election.

       Many supporters of the 1964 election code’s majority vote requirement had

legitimate “good government” motives. These backers wanted to reduce the power of the

courthouse crowds and to eliminate the use of stalking horses and dummy candidates as a

method of election manipulation. Governor Sanders was one of the key supporters of the

measure. In the context of Georgia politics in the early 1960's, Sanders was regarded as a

moderate on racial issues. He opposed the civil rights movement, yet he was not a

militant segregationist. Sanders had considerable power over the General Assembly,

enabling him to influence the agenda of the legislature. His power in the legislature was

so great that he was able to install his friend and supporter, George T. Smith (“Smith”), as

Speaker of the House.

       Plaintiffs emphasize the racial antipathy of Groover and attempt to impute the

improper motivations behind his proposal of the majority vote in 1963 to the supporters

of the 1964 code’s majority vote requirement. The district court found that Groover’s

                                             6
racist motives were not attributable to the supporters of the 1964 law. Groover supported

Sanders’ opponent in the governor’s race and was not involved in the Sanders

administration. He was not a member of the third ELSC, and he had no influence over

that committee.

       The district court also acknowledged that Sanders favored at-large elections in

1962, when he was president pro tem of the Senate, for racially discriminatory reasons as

well as some sound government reasons. Again, the district court chose not to impute

those discriminatory motives to Sanders’s support of the majority vote law in 1964 in the

absence of evidence indicating that the 1964 majority vote provision itself was racially

motivated.

B.     Discriminatory Impact

       On the question of the effect of the majority vote law, the district court considered

testimony as to the law’s impact on individuals considering a run for office as well as

statistical analyses of data showing how the law affected actual elections. The district

court found that though there was some evidence minimally supporting the theory that

black candidates were discouraged from running for office by the prospect of a runoff, the

evidence did not enable the court to ascertain how pervasive any deterrent effect had

been. In other words, the Plaintiffs failed to prove that the majority vote law improperly

discourages potential black candidates. In addition, there was no proof that the law would

act as a greater deterrent for black candidates than for white candidates.



                                              7
       Defendants’ expert witness supervised the compilation of a database containing

information on the effect of the majority vote requirement on Georgia primary elections

from 1970 to 1995. During this period, there were a total of 2,798 runoff sequences, and

complete data was available for 2,773 of these. Over 90 percent of these runoffs were in

Democratic party primaries. Of these 2,773 runoff sequences, there were 278 which

involved a black candidate and a white candidate in both the primary election and the

runoff. In 85 of the runoffs between a black candidate and a white candidate, the

candidate who won a plurality of votes in the initial primary election lost in the runoff.

The district court described these 85 runoff elections as “flip” sequences.

       In 56 of the flip sequences, the black candidate lost the runoff after receiving a

plurality of the votes in the initial primary. In 29 of the flip elections, the white

candidate lost the runoff after receiving a plurality of the votes in the initial primary.

Thus, the majority vote requirement for primary elections yielded a net result of 27 fewer

black candidates than would have been nominated under a pure plurality scheme.2

       Most of the 27 flip elections adverse to black candidates occurred at the county

and local levels. A net loss of 30 black candidates in local and county elections resulted

from the majority vote requirement. In state legislature and superior court elections, the

primary runoff requirement resulted in a net gain of three black candidates. In state-wide



       2
        Significantly, it does not follow that Georgia would have had 27 additional black
office holders because a victory in a primary election does not ensure a win in the general
election.

                                               8
and federal primaries, the net results would have been the same under a plurality scheme

as they were under the majority vote requirement.

       The district court found that the disparity in outcomes in primary runoff elections

was attributable to the relative strength of individual candidates and not to any alleged

discriminatory impact of the majority vote requirement. The Defendants’ political science

expert presented a theory under which a candidate who receives 40 percent or more of the

votes in the initial election and who leads his or her opponent by at least five percent is

considered a “strong leader.” In other words, a “strong leader” is a candidate who

soundly wins an election. Runoff winners were overwhelmingly “strong leaders” in their

initial primary races. The runoff winners averaged above 40 percent of the initial vote

and generally led their nearest opponent by 10 to 12 percentage points. Black “strong

leaders” defeated white opponents in runoffs 71 percent of the time. The black

candidates who lost in runoffs typically entered the runoff elections in a much weaker

position. The district court concluded that black leaders who lost runoffs were on average

weaker candidates than black or white leaders who won runoff elections. On the question

of racial polarization, the district court found that the Plaintiffs failed to prove that voting

in Georgia was racially polarized.

       The district court concluded that, overall, Plaintiffs failed to demonstrate that the

majority vote requirement in primary elections had a significant adverse effect on black

voters and candidates.



                                               9
                               II. STANDARD OF REVIEW

       The issue of whether an election system was established or maintained for a

discriminatory purpose is a question of fact subject to the clearly erroneous standard of

review. Rogers v. Lodge, 

, 622-23 (1982).        Rule 52(a) of the Federal Rules

of Civil Procedure is particularly relevant to our review of the district court’s findings on

purpose. Rule 52(a) requires that in reviewing the district court’s factual findings, we

must give due regard to the court’s opportunity to judge the credibility of witnesses. Fed.

R. Civ. P. 52(a).

       We review the district court’s determination of whether the majority vote

requirement has had a discriminatory impact for clear error. Thornburg v. Gingles, 

, 79 (1986) (stating that “application of the clearly-erroneous standard to ultimate

findings of vote dilution preserves the benefit of the trial court’s particular familiarity

with the indigenous political reality without endangering the rule of law”). Thus, while

we review the district court’s conclusions of law de novo, Davis v. Chiles, 

,

1420 (11th Cir. 1998), the Supreme Court has clarified that the clearly erroneous standard

applies to the district court’s ultimate conclusion as to whether an election mechanism

leads to discriminatory results. 

. The trial court “is to consider the

totality of the circumstances and to determine, based upon a searching practical

evaluation of the past and present reality, whether the political process is equally open” to

minorities. 

quotations omitted). In conducting our review of the

district court’s decision, we have the “power to correct errors of law, including those that

                                              10
may infect a so-called mixed finding of law and fact, or a finding of fact that is predicated

on a misunderstanding of the governing rule of law.” 

                       III. ANALYSIS OF FACTUAL FINDINGS

A.     Discriminatory Purpose

       The district court found that the majority vote requirement was not passed for a

discriminatory purpose and that the key supporters of the requirement had good

government reasons for their support. There is adequate support in the record for this

finding, and we hold that it was not clearly erroneous. The district court credited the

testimony of Governor Sanders who explained that prior to the passage of the 1964

election code, Georgia elections were “in a state of chaos.” (1st Supp. R., vol. 4, 167).

Some counties used a plurality system, while others had a majority vote requirement. 

was pervasive throughout the state, including the use of “stalking horses”

in elections where only a plurality was required for victory. 

In addition,

according to Governor Sanders, incumbents frequently attempted to “muddy the water”

by getting as many people as possible to run in their races to dilute the opposition vote

and win a plurality. 

       The district court also credited the testimony of other witnesses to the effect that

the majority vote requirement was not motivated by an intent to discriminate on the basis

of race. Eugene Patterson, an anti-segregationist who was editor of The Atlanta

                                             11
Constitution in the early sixties testified that Sanders was not racist in his attitudes and

that the 1964 code, including the majority vote provision, was passed for legitimate

government reasons and not to discriminate against black voters and candidates. (R., vol.

15, 447-49). George T. Smith, Speaker of the House during the Sanders administration,

testified that there was no discussion of discrimination against black candidates in

connection with the Sanders administration’s support for majority voting. (1st Supp. R.,

vol. 5, 77). Melba Williams, a member of the third ELSC, whom no one suggests was

racially motivated in her support for majority voting, testified that the committee’s

decision to include a majority voting provision in the ELSC’s proposed code had nothing

to do with race. (1st Supp. R., vol. 5, 137-40).

       Plaintiffs contend that the district court clearly erred in failing to impute the

discriminatory motives behind Groover’s 1963 majority vote bill to the supporters of the

1964 election code’s majority vote provision. However, Governor Sanders testified that

Groover, who had supported his opponent in the governor’s race, had “nothing to do

with” the comprehensive 1964 election reform law. (1st Supp. R., vol. 4, 195). This

testimony as to Groover’s lack of input in the 1964 bill was corroborated by Smith. (1st

Supp. R., vol. 5, 74-77). It appears that Plaintiffs’ emphasis on the discriminatory intent

of Groover in connection with majority voting is necessitated by the lack of evidence that

the proponents of the 1964 election code had racially discriminatory motives for their

support of the majority vote provision. One of the Plaintiffs’ experts admitted on cross

examination that in his review of all of the records of the third ELSC that produced the

                                              12
1964 election code, he did not find any evidence of a “smoking gun,” meaning evidence

providing a clear link between the majority vote provision in that code and race

discrimination. (R., vol. 14, p. 187). The Plaintiffs’ theory was that a discriminatory

purpose behind the 1964 provision could be inferred circumstantially from the improper

motives behind Groover’s previous majority vote bill, from the third ELSC’s proposal of

a modified literacy test, and from evidence that Sanders’ support of previous measures,

such as at-large voting, was racially motivated. The district court did not commit clear

error by rejecting this theory and focusing instead on specific evidence concerning how

and why the majority vote rule was included in the 1964 election code.

B.     Discriminatory Impact

       With respect to Plaintiffs’ allegation that the challenged provision discouraged

blacks from running for office, the district court heard testimony from three witnesses

who had held office in Georgia, including Plaintiff Tyrone Brooks, a state legislator from

Atlanta. These witnesses testified to the effect that the majority vote law was a deterrent

for many potential black candidates. Although two of these witnesses gave a few specific

examples of black individuals they had talked to who said that they were discouraged by

the prospect of a primary runoff, the district court found that this evidence gave only

minimal support for the theory that black candidates were deterred by the possibility of a

runoff before entering an election. This evidence did nothing to bolster the theory that the

potential for a runoff was more of a deterrent for black candidates than for white

candidates.

                                             13
       On the question of whether the Plaintiffs proved that the majority vote requirement

had a significant, adverse effect on black candidates, the district court found that the

racial impact of the majority vote requirement has been negligible at most. The court

considered the fact that the requirement caused a net loss of 27 black nominees over 25

years and that there was no negative impact on black candidates in roughly 99 percent of

all runoffs during this period. The district court’s finding that the majority vote law does

not have a discriminatory effect on black candidates has strong support in the record and

therefore is not clearly erroneous.



                                  IV. LEGAL ANALYSIS

A.     Section 2 of the Voting Rights Act

       With respect to the Plaintiffs’ claim under § 2 of the Voting Rights Act, 42 U.S.C.

§ 1973, the district court concluded that Georgia’s majority vote requirement for primary

elections does not violate § 2.

       1.     Discriminatory Purpose

       Plaintiffs argue that a showing of discriminatory purpose alone is sufficient for a

violation of § 2.   The Plaintiffs’ claim under this theory fails for two reasons. First, as

discussed above, we affirm the district court’s factual finding that the majority vote

requirement contained in the 1964 election code was not motivated by a discriminatory

purpose. Second, even if we found clear error in the district court’s finding on

discriminatory purpose, we are bound by Johnson v. DeSoto County Bd. of Comm’rs, 72

                                              

(11th Cir. 1996), which held that discriminatory intent alone, in the absence of a

showing of discriminatory effect, is insufficient to establish a violation of § 2. 

We will, therefore, focus our discussion on the “results test” for § 2 violations.

       2.     Discriminatory Results

       Congress amended § 2 in 1982 to clarify that a violation of this statute may be

proved by a showing of discriminatory results alone, thereby superceding City of Mobile

v. Bolden, 

(1980), which held that a plaintiff had to show both a

discriminatory purpose and a discriminatory effect to establish a violation. Section 2, as

amended, provides:

       (a) No voting qualification or prerequisite to voting or standard, practice, or
       procedure shall be imposed or applied by any State or political subdivision
       in a manner which results in a denial or abridgement of the right of any
       citizen of the United States to vote on account of race or color, or in
       contravention of the guarantees set forth in section 1973b(f)(2), as provided
       in subsection (b).
       (b) A violation of subsection (a) of this section is established if, based on
       the totality of circumstances, it is shown that the political processes leading
       to nomination or election in the State or political subdivision are not
       equally open to participation by members of a class of citizens protected by
       subsection (a) in that its members have less opportunity than other members
       of the electorate to participate in the political process and to elect
       representatives of their choice. The extent to which members of a protected
       class have been elected to office in the State or political subdivision is one
       circumstance which may be considered: Provided, That nothing in this
       section establishes a right to have members of a protected class elected in
       numbers equal to their proportion in the population.




                                              15
42 U.S.C.A. § 1973. The Senate Judiciary Committee Majority Report accompanying the

1982 amendment listed a number of factors that may show a § 2 violation.3




      3
       The “typical factors” listed in the Senate Report are as follows:
      1. the extent of any history of official discrimination in the state or political
      subdivision that touched the right of the members of the minority group to
      register, to vote, or otherwise to participate in the democratic process;
      2. the extent to which voting in the elections of the state or political
      subdivision is racially polarized;
      3. the extent to which the state or political subdivision has used unusually
      large election districts, majority vote requirements, anti-single shot
      provisions, or other voting practices or procedures that may enhance the
      opportunity for discrimination against the minority group;
      4. if there is a candidate slating process, whether the members of the
      minority group have been denied access to that process;
      5. the extent to which members of the minority group in the state or
      political subdivision bear the effects of discrimination in such areas as
      education, employment and health, which hinder their ability to participate
      effectively in the political process;
      6. whether political campaigns have been characterized by overt or subtle
      racial appeals;[and]
      7. the extent to which members of the minority group have been elected to
      public office in the jurisdiction.
       Additional factors that in some cases have had probative value as part of
      plaintiffs' evidence to establish a violation are:
       whether there is a significant lack of responsiveness on the part of elected
      officials to the particularized needs of the members of the minority group
      [and]
       whether the policy underlying the state or political subdivision's use of
      such voting qualification, prerequisite to voting, or standard, practice or
      procedure is tenuous.

S.Rep. No. 97-417, 97th Cong. 2nd Sess. 28, 28-29 (1982), U.S.Code Cong. &
Admin.News 1982, pp. 206-207. The Senate Committee clarified that “‘there is no
requirement that any particular number of factors be shown, or that a majority of them
point one way or the other.’” 

(quoting S.Rep. No. 97-417 at 29).

                                             16
       Though the test for § 2 violations is generally flexible and fact-intensive, the

Senate Report placed three limits on the way § 2 violations may be proved. 

. First, electoral devices such as at-large elections, or in this case, a majority

vote requirement, do not violate § 2 per se. See 

an electoral

device must prove that under the totality of the circumstance, the device “result[s] in

unequal access to the electoral process.” 

violation of § 2 cannot be

established by showing “the conjunction of an allegedly dilutive electoral mechanism and

the lack of proportional representation alone.” 

must prove the

existence of racial bloc voting. 

test does not assume it. 

of the Gingles prerequisites

       The Gingles case involved a vote dilution claim against the use of multi-member

districts. In this context, the Supreme Court developed three prerequisites for a claim of

vote dilution. 

. The Supreme Court has not heard any challenges

to majority vote requirements, though such challenges have been presented in lower

federal courts.4 Because the requirement of a majority of the votes in a primary election


       4
          In Whitfield v. Democratic Party of the State of Arkansas, 

(E.D. Ark. 1988), aff’d by equally divided court without opinion, 

(8th Cir.
1990) (en banc), a district court held that Arkansas’ majority vote requirement for
primary elections did not violate § 2 of the Voting Rights or the Fourteenth and Fifteenth
Amendments to the United States Constitution. In another challenge, a three-judge panel
held that Arkansas’ majority vote provision for general elections was unconstitutional, but
rejected the constitutional attack on the majority vote requirement as applied to primary
elections. Jeffers v. Clinton, 

, 594-95 (E.D. Ark. 1990) (reasoning that
the primary runoff represents a legitimate means to ensure that the nominee has a
majority of party support and that a runoff is not necessary in a general election because

                                              17
is distinct from the establishment of a multi-member district such as that at issue in the

Gingles case, Georgia’s majority vote provision does not fit neatly into the analytical

framework set out in Gingles. See 

(stating that it “is

doubtful” that the Gingles prerequisites would be emphasized in a challenge to a runoff

requirement).

       Nonetheless, we agree with the district court that, with slight modification, the

three prerequisites for the results test set out in Gingles are applicable to this case.

Gingles calls for a flexible, fact-intensive inquiry into whether an electoral mechanism

results in the dilution of minority votes, and the Court in Gingles recognized the

“potentially dilutive” effect of majority vote requirements. 

, 56.

                b.    Analysis under Gingles

       The Court in Gingles developed the following three threshold requirements for

establishing a violation of § 2: (1) “the minority group must be able to demonstrate that it

is sufficiently large and geographically compact to constitute a majority in a single-

member district;” (2) “the minority group must be able to show that it is politically

cohesive;” and (3) “the minority group must be able to demonstrate that the white

majority votes sufficiently as a bloc to enable it . . . to defeat the minority’s preferred

candidate.” 

. These three circumstances are “necessary

preconditions” for a showing that an electoral scheme impaired minority voters’ ability to


there are almost never more than two substantial candidates).


                                               18
elect their chosen representatives. 

establish the three prerequisites,

they must go on to show that, under the totality of the circumstances, they have been

denied an equal opportunity to elect representatives of their choice. Nipper v. Smith, 

, 1512 (11th Cir. 1994).

                     i.      The first Gingles prerequisite

       The first prerequisite essentially “asks whether the court can fashion a remedy for

a demonstrated abridgement.” 

. If the plaintiffs in a § 2 case

cannot show the existence of an adequate alternative electoral system under which the

minority group’s rights will be protected, then the case ends on the first prerequisite. See

n.34. In its analysis under the first prerequisite, the district court determined

that “[t]he analogous question in the case at bar is whether black voting strength would be

‘less diluted’ under some workable regime other than a strict majority vote requirement.”

(R., vol. 12, Tab 165, 22). This is a proper interpretation of the first Gingles prerequisite

as applied to Georgia’s majority vote provision.

       The district court found that the Plaintiffs failed to establish the first Gingles

prerequisite because they did not prove that any alternative to the majority vote

requirement would result in a net increase of black elected officials. Under a 45 percent

plurality rule, like that in place in Georgia’s general elections, there would have been a

net loss of one black primary winner between 1970 and 1995. Although the evidence

demonstrated that under a pure plurality system in Georgia primaries, there would have

been a net increase of 27 black nominees for political office during the same period, “[i]n

                                              19
assessing a plaintiff’s proposed remedy, a court must look to the totality of the

circumstances, weighing both the state’s interest in maintaining its election system and

the plaintiff’s interest in the adoption of his suggested remedial plan.” Davis v. Chiles,

, 1419-20 (11th Cir. 1998) (citing Houston Lawyers’ Ass’n v. Attorney

Gen’l of Tex., 

(1991)). As Defendants pointed out, a pure plurality system

such as that suggested by Plaintiffs could theoretically result in a candidate’s winning

with 1% of the vote if enough candidates entered the race. Obviously, this type of result

would seriously undermine the legitimacy of the government, and the state has a

substantially compelling interest in preventing this from occurring. Particularly when

balanced against the average of only one additional nominee per year statewide over the

past twenty-five years that would have resulted from a pure plurality system, the district

court could properly conclude that the harm resulting from Plaintiffs’ proposed remedy is

simply too great to justify ordering such a system. Accordingly we hold that the district

court did not err in concluding that Plaintiffs failed to establish an adequate remedy.

Because Plaintiffs fail on the first prerequisite, the case under § 2 could end here. 

n. 34. Nevertheless, we will review the remainder of the district court’s §

2 analysis.

                     ii.    The second and third Gingles prerequisites

       The district court dealt with the second and third Gingles factors together,

concluding that the Plaintiffs failed to meet either requirement. The second prerequisite

requires a showing that the minority group is politically cohesive, while the third calls for

                                             20
proof of majority white bloc voting sufficient to defeat the minority’s preferred

candidates. 

. In other words, these prerequisites ask whether

voting is racially polarized and, if so, whether the white majority is usually able to defeat

the minority bloc’s candidates. In the absence of minority political cohesion and

significant white bloc voting, it would be difficult, if not impossible, to prove dilution of

minority votes in violation of § 2. See 

.

       The district court, while acknowledging Georgia’s notorious history of official

discrimination against black voters as well as the possibility “that racially polarized

voting is the norm in Georgia,” found that the Plaintiffs failed to carry their burden of

proof on the question of racial polarization. (R., vol. 12, Tab 165, 23). The district court

went on to state that even if the Plaintiffs had proven racially polarized voting, they failed

to show that the purported white voting bloc “usually” defeated the purported minority

bloc’s candidates under the majority vote system. 

       In discussing the proper legal standard for analysis of racial polarization, the

Gingles Court explained that “in general, a white bloc vote that normally will defeat the

combined strength of minority support plus white ‘crossover’ votes rises to the level of

legally significant white bloc 

. We agree with the district court

that the statistical evidence offered by the Plaintiffs did not support the proposition that

the white voting bloc normally defeated the candidate supported by most black voters.

Runoffs caused by Georgia’s majority vote law occur in only a small fraction of all

primaries in the state. Thus, even if there were a significant racial impact in the context

                                              21
of all runoffs, it could not be said that the white majority usually defeats the minority’s

preferred candidate under Georgia’s primary runoff system. Moreover, the majority vote

law had no net adverse racial impact on black candidates in roughly 99 percent of all

runoffs over a 25 year period. We therefore affirm the district court’s conclusion that

Plaintiffs failed to prove that white bloc voting “usually” leads to the defeat of black

voters’ preferred candidates under the majority vote scheme.

       Plaintiffs criticize the district court’s analysis under the second and third Gingles

factors as containing “an extraordinary inconsistency.” (Plaintiffs’ brief at 36.) The

alleged inconsistency lies in the following passage of the district court’s order:

       Rather than demonstrating to the court that black primary winners would be
       capable of winning general elections, plaintiffs have made every effort to
       convince the court that voting in Georgia is so polarized that racial
       minorities cannot win in majority white districts. Accepting their
       arguments as true, a plurality primary system would not increase “the
       potential [of blacks] to elect representatives in the absence of the challenged
       structure.”

(R., vol. 12, Tab. 165, 23 (quoting 

n. 17)). The district court did

make a factual finding that the Plaintiffs failed to prove racially polarized voting. We do

not, however, agree with the Plaintiffs that the district court’s reasoning was logically

flawed merely because it assumed, arguendo, that voting was as racially polarized as the

Plaintiffs attempted to demonstrate for purposes of the second and third Gingles

prerequisites.

       Rather, the Plaintiffs themselves have advocated conflicting positions on the

question of racially polarized voting. Georgia maintains the majority vote requirement

                                              22
only for primary elections. To succeed on the first Gingles prerequisite, the Plaintiffs

must demonstrate that if an alternative system such as a pure plurality were in place, more

black candidates would in fact be elected to office. This, however, would require proof of

substantial white crossover voting in general elections which is very difficult to achieve

in conjunction with a showing, required by the second and third Gingles prerequisites,

that voting is so racially polarized that the white majority, voting as a bloc, has the ability

to defeat the minority’s preferred candidates. In an effort to walk that fine line, the

Plaintiffs point to the fact that there was “a minimal level of ” white crossover voting, to

the success of some black candidates in runoffs in majority white jurisdictions, and to the

testimony of former Atlanta mayor Maynard Jackson. (Plaintiffs’ brief at 36). Jackson

testified that in his opinion, if a black candidate could win the primary, he or she would

gain the support of the Democratic party, obtaining an advantage in the general election.

(R., vol 13, 79-80.) The district court did not consider this evidence sufficient proof of

Plaintiffs’ theory that those black candidates who are only capable of winning a primary

under a pure plurality system would in fact go on to defeat their opponents in the general

election.

       Overall, the district court concluded that under the totality of the evidence

presented by the parties, Georgia’s majority vote requirement does not “eviscerate[] the

ability of minority voters to elect their candidates of choice.” (R., vol. 12, Tab 165, 24

(citing 

)). Detecting no errors of law in the district court’s

analysis, we affirm under the clearly erroneous standard the district court’s ultimate

                                              23
finding that the majority vote law does not yield racially discriminatory results. See

(stating that clearly erroneous standard applies to ultimate finding

of vote dilution).




B.     Constitutional Claims

       To make out a constitutional claim against the majority vote requirement, which is

racially neutral on its face, Plaintiffs must demonstrate that the provision was motivated

by a discriminatory purpose under the analysis used by the Supreme Court in Hunter v.

Underwood, 

, 225 (1985).       The Court in Hunter held that the test set out in

Mt. Healthy City Bd. of Ed. v. Doyle, 

(1979) applies to the determination of

whether voting statutes are unconstitutional because of a discriminatory purpose. 

. Under this test, a court must first determine whether the discriminatory

motive was a “substantial factor” or a “motivating factor” behind a governmental

decision. Mt. 

. “Once racial discrimination is shown to have

been a ‘substantial’ or ‘motivating’ factor behind enactment of the law, the burden shifts

to the law’s defenders to demonstrate that the law would have been enacted without this

factor.” 

(quoting Mt. 

).

       The district court found that discrimination was not a substantial or motivating

factor behind enactment of the majority vote provision. As previously discussed in our

factual analysis, we conclude that this finding is not clearly erroneous. The district court

                                             24
explained that although it was evident that “the virus of race-consciousness was in the

air,” Plaintiffs failed to prove that “the specific measure at issue here–the majority vote

requirement in the 1964 election code–was infected thusly.” (R., vol. 12, Tab 165, 26).

       The court went on to state that even if the Plaintiffs had proven that race-conscious

reasons were a substantial or motivating factor behind the challenged law, Defendants

have shown that it would have been enacted even in the absence of those reasons. We

agree that the Defendants demonstrated that the supporters of the 1964 majority vote

provisions had ample “good government” reasons and that the law would have been

passed in the absence of any discriminatory motive. Plaintiffs argue that the district court

placed too much emphasis on the motivations of Governor Sanders and his

administration, in light of the fact that the entire Georgia legislature enacted the code

which contained the majority vote provision. However, the Defendants’ evidence

persuaded the district court that Sanders wielded enough power in the legislature not only

to install one of his friends and allies as Speaker of the House, but also to effectuate the

passage of the 1964 election code. The Defendants proved that legitimate “good

government” reasons were the primary motivating factor behind the majority vote

provision’s key proponents, including Sanders, and that the Sanders administration had

sufficient power over the legislature to have the 1964 election code enacted. Even if

some of the legislators who eventually voted the 1964 code into law had discriminatory

reasons for their support, the Plaintiffs’ constitutional claims fail under the “but for” test

set out in Hunter and Mt. Healthy.

                                              25
       Plaintiffs critique the district court’s fact finding on the question of discriminatory

purpose, citing Arlington Heights for the proposition that the district court erred in relying

heavily on the testimony of decision makers involved in the passage of the majority vote

provision. See Village of Arlington Heights v. Metropolitan Housing Dev. Corp., 

, 268 (1977) (stating that in “extraordinary instances the members [of a decision

making body] might be called to the stand at trial to testify concerning the purpose of the

official action”); Citizens to Preserve Overton Park, Inc. v. Volpe, 

, 420

(1971) (expressing concern that after-the-fact reconstructions of legislative purpose can

be self-serving and unreliable and cautioning that such recollections should be viewed

critically). Plaintiffs contend that the district court should have placed more emphasis on

the contemporaneous record relating to the legislation. See 

-29;

Overton 

.

       While the Plaintiffs are correct that the contemporaneous record should factor

heavily into a trial court’s determination of legislative purpose, the district court did in

fact consider evidence of the legislative history of the 1964 majority vote provision. To

the extent the Plaintiffs contend that newspaper evidence is part of the contemporaneous

record and should, therefore, be the primary source for ascertaining legislative intent, we

reject this theory. News articles often contain multiple layers of hearsay and do not trump

the sworn testimony of eyewitnesses. In ascertaining legislative purpose, a trial court

operates under the same rules of evidence that control in any case. The Arlington Heights

Court explained that “[d]etermining whether invidious discriminatory purpose was a

                                              26
motivating factor demands a sensitive inquiry into such circumstantial and direct evidence

of intent as may be 

. In our view, the district court’s review of

the available evidence constituted the “sensitive inquiry” required by Arlington Heights.

The court analyzed the history of the 1964 majority vote provision and considered the

ELSC reports, specifically acknowledging that relevant considerations of discriminatory

purpose “include the historical background of the challenged act, the ‘specific sequence

of events’ leading up to the act’s passage, and the legislative history of the act, ‘especially

where there are contemporaneous statements by members of the decisionmaking body,

minutes of its meeting, or reports’” (R. vol. 12, Tab 165, 25 (quoting Arlington 

)).

       Although Plaintiffs offered evidence of racially discriminatory intent from the first

two ELSCs appointed by Governor Sanders’ predecessors, this evidence did not address

why the 1964 code’s majority vote provision was passed. The evidence included the

unpurged records of the ELSC that proposed the 1964 code, (R. vol. 9, Tab 151, 2), and

the Plaintiffs’ own expert admitted on the stand that nothing in these records provides a

clear link between race and the majority vote provision. (R., vol. 14, p. 187). There was

evidence of explicit racial considerations on other issues in the third ELSC’s records, as

well as in the records of the previous ELSC’s. However, the clear historical trail of racial

purpose on other issues and in previous committees stands in stark contrast with the

absence of evidence of racial purpose in connection with the majority vote proposal in the



                                              27
third ELSC’s records. This contrast supports the district court’s finding that the

challenged provision was not racially motivated.

       Because we affirm the district court’s finding that race was not a substantial

purpose behind the majority vote law, we need not address the Plaintiffs’ contention that

even if the racial impact of the majority vote provision is insignificant, it can be

invalidated as unconstitutional and in violation of § 2 of the Voting Rights Act if race was

a predominant purpose behind its adoption.



                                     V. CONCLUSION

       For the foregoing reasons, we affirm the judgment of the district court.

       AFFIRMED.




                                              28
29