Amicus Curiae Brief — City of Gretna v. Citizens for a Better Gretna

oo

Supreme Court, U.S.

(..) FILED

No. 88-597 ~~” JUN 5 1989

|

TEE TN “|

CLERK

In the Supreme Court of the Unitedrsstgtes —

OCTOBER TERM, 1988

CiTY OF GRETNA LOUISIANA, ET AL., PETITIONERS

v.

CITIZENS FoR A BETTER GRETNA, ET AL.

ON PETITION FOR A WRIT OF CERTIORARI

TO THE UNITED STATES COURT OF APPEALS

FOR THE FIFTH CIRCUIT

BRIEF FOR THE UNITED STATES AS AMICUS CURIAE

KENNETH W. STARR

Solicitor General

JAMES P. TURNER

Acting Assistant Attorney General

THOMAS W. MERRILL

Deputy Solicitor General

CHRISTOPHER J. WRIGHT

Assistant to the Solicitor General

JESSICA DUNSAY SILVER

MIRIAM R. EISENSTEIN

Attorneys

Department of Justice

Washington, D.C. 20530

(202) 633-2217

QUESTION PRESENTED

Whether, in this vote dilution case brought under Sec-

tion 2 of the Voting Rights Act, 42 U.S.C. 1973, the

district court properly prohibited the practice of filling all

positions on the Board of Aldermen of the City of Gretna,

Louisiana, through at-large elections, where the evidence

showed that black candidates supported by substantial

numbers of black voters received almost no votes from

white voters and hence could not be elected to the Board

under the at-large system.

(1)

FABLE OF CONTENTS

Page

Statement ........ Da ih aay as Sy ere ae l

Discussion ...

I cd mae 6a ; 16

TABLE OF AUTHORITIES

Cases:

City of Carrollton Branch of the NAACP vy. Stallings,

829 F.2d 1547 (11th Cir. 1987), cert. denied, 108 S. Ct.

I Rr i Sas aD as ge i-3 Rens 14

Smith v. Clinton, 687 F. Supp. 1310 (E.D. Ark.), aff'd

Sooke 2) ae Bere et 14

Solomon v. Liberty County, 865 F.2d 1566 (11th Cir.

ty i Ra Sin el a na a eee 14

Thornburg v. Gingles, 478 U.S. 30 (1986) ........0......5, 6, 7,

Statutes:

Voting Rights Act of 1965, 42 U.S.C. 1971 ef seq.:

> ea EOS ok Na 2, 5, 6, 9, 13, 14

§ 2(b), 42 U.S.C. 1973(b) 2

§ 5, 42 U.S.C. 1973c ; a 6

La. Rev. Stat. Ann. (West 1979 & Supp. 1989):

8 SEN Pea ee Fe ek ee 2

B, SB

Miscellaneous:

H. Blalock, Social Statistics (Rev. 2d ed. 1979) 1]

(111)

Jn the Supreme Court of the Anited States

OCTOBER TERM, 1988

No. 88-597

City OF GRETNA LOUISIANA, ET AL., PETITIONERS

Vv.

CITIZENS FoR A BETTER GRETNA, ET AL.

ON PETITION FOR A WRIT OF CERTIORARI

. TO THE UNITED STATES COURT OF APPEALS

FOR THE FIFTH CIRCUIT

BRIEF FOR THE UNITED STATES AS AMICUS CURIAE

This brief is submitted in response to the Court’s invita-

tion to the Solicitor General to express the views of the

United States.

STATEMENT

1. The City of Gretna, located in Jefferson Parish,

Louisiana, has a population of 20,615. Pet. App. A4l.

Approximately one quarter of the voting age population 1s

black. PX 56, at 12. Before the district court’s decision in

this case, Gretna was governed by a five-member Board of

Aldermen who were elected at-large. It was the largest city

in the State to use an at-large system. /d. at A41. Voters

could vote for five candidates in an election for the Board.

“Single-shot” or “bullet” voting (/.e., voting for fewer than

five candidates) also was permitted. /d. at A42 n.27. To be

elected, candidates had to obtain a “majority” under La.

Rev. Stat. Ann. § 18:511 (West 1979 & Supp. 1989). Pet.

(1)

tt

App. A41-A42.' No black person was ever elected to the

Board of Aldermen under this system since Gretna’s incor-

poration in 1913. /d. at A34.

The plaintiffs, eleven black voters in Gretna, challenged

the at-large system of electing the Board of Aldermen.

They alleged that it effectively denied blacks the oppor-

tunity to elect their own representatives to the Board even

though blacks comprise a significant portion of the total

population of the City, in violation, inter alia, of Section 2

of the Voting Rights Act of 1965, 42 U.S.C. 1973. Section

2(b) provides that the Act is violated where members of

racial minorities “have less opportunity than other

members of the electorate * * * to elect representatives of

their choice.”

2. After trial, the district court ruled that Gretna’s at-

large system of electing the Board of Aldermen violates

Section 2 of the Voting Rights Act. The court found that,

since 1877, the white majority in Louisiana has resisted at-

tempts by blacks to participate in the political process.

Pet. App. A25-A26. In the 1950s and 1960s, the Jefferson

Parish registrar of voters required blacks, but not whites,

“to state their age exactly in terms of number of years,

months, and days, and to recite the preamble to the United

States Constitution in order to register to vote.” /d.at

' Respondents’ expert explained that a “majority” was calculated by

taking the entire number of votes cast in an election, dividing by five

(the number of seats), and then dividing by two and adding one. If

fewer than five candidates received a majority (as was especially likely

if there were a large number of serious candidates), those who ob-

tained a majority were declared elected and a runoff election was held

in which no more than twice the number of candidates as there were

untilled positions could compete (see La. Rev. Stat. Ann. § 18:482

(West 1979). If more than five candidates received a majority (as was

especially likely if there were few serious candidates), the five with the

most votes won. I] Tr. 77-78.

A26. “Schools in Jefferson Parish remained segregated

until 1971,” after which they were desegregated pursuant

to a federal court order. /d. at A27. With respect to hous-

ing, the court found that “[v]irtually all residential neigh-

borhoods are racially identifiable.” Jd. at A29.

The court further found that, while “blacks are

registered to vote at a rate equivalent to white residents of

Gretna,” their turnout rate is substantially lower than that

of white voters. Pet. App. A30. It suggested that blacks

may believe that participation in local elections is futile

because, for many years, there has been an informal

slating process for city offices in Gretna from which

blacks have been excluded. Jd. at A31 & n.12. The slating

process has been dominated by the chief of police and the

mayor, whose supporters are known as “the Miller-White

faction.”? The candidates that the Miller-White faction

endorses consistently win seats on the Board of Aldermen.

Id. at A38-A41.

Only two blacks have ever run for city office in Gretna,

Edwin Romain and Leo Jones. Both ran for alder-

man — Romain in 1973, and Jones in 1977 and 1979. Pet.

App. A33. Romain sought the endorsement of several of

the incumbent aldermen in 1973, but was told to talk to

Chief of Police Miller. Miller, in turn, told him “that he

could not be on the slate, that he was not needed to get the

black vote.” Jd. at A40. Romain was the only non-

incumbent candidate in 1973 who was not invited to a

meeting held for the purpose of selecting candidates for

the Miller-White faction to endorse for the two seats on

the Board that were being vacated by incumbents. Two

? Beauregard Miller, Sr., was chief of police from 1925 until 1976,

when he was succeeded by his son, Beauregard, Jr. William White was

the mayor of Gretna from 1951 until recenily. Pet. App. A38.

Eee

4

candidates were endorsed by the Miller-White faction

following the meeting, and they were elected. Jd. at

A39-A40.

Romain was supported by the Westside Citizens Com-

mittee, a black political organization. Pet. App. A40.3 He

campaigned in both the black and white communities. Ro-

main, “a light-skinned black” (id. at A56), testified that

many of the voters did not initially know that he was

biack. In the first round, he finished sixth in a field of 17

candidates. Between the first round and the runoff, a

newspaper article noted that Romain might become the

City’s first black alderman. Jd. at A56-A57. “The evidence

Suggests,” the district court found, that the revelation that

Romain is black “had severe consequences.” Jd. at A56. In

the runoff election, Romain finished last in all but the two

most heavily black precincts (where he came in first and

second), and last in the total vote. Jd. at AS6-A57.

Leo Jones ran for one of five open seats in 1977, and for

a single open seat in the 1979 special election. Unlike the

1973 election, data was available for the 1977 and 1979

elections with respect to turnout by race and precinct. The

district court found ihat both of those races manifested

high degrees of racially polarized voting. Respondents’ ex-

pert calculated that, in 1977, Jones won approximately 67

percent of the black vote and, in 1979, 66 percent, but only

12 percent and one percent of the white vote in the same

elections. Pet. App. A46.4 The court concluded that the

’ The record retlects that the Westside Citizens Committee general-

ly supported the Miller-White ticket. It made an exception and sup-

»orted Romain in 1973, but it did not support Leo Jones either time he

I PP

ran. | Tr. 168; INI] Tr. 33, 46, 48-54.

* Respondents’ expert performed bivariate regression analyses of

the election returns in the 1977 and 1979 elections, a statistical analysis

that seeks to determine whether there is a direct and close relationship

between race and voting behavior. The regression analyses showed

4)

results in the 1977 and 1979 elections showed a high degree

of racial polarization. /d. at A51.

Because there were so few races in which blacks had run

for office in Gretna, the court also considered how

Gretna’s electorate had voted in two other races involving

black candidates, and found extreme polarization in those

races as well. In 1979, Ben Jeffers ran for Louisiana

Secretary of State, and won about 70 percent of the black

vote in Gretna but only about eight percent of the white

vote. Pet. App. A46. Similarly, in Jesse Jackson’s run in

the 1984 Democratic presidential primary, he obtained 95

percent of the black vote in Gretna, but only two percent

of the white vote. /bid. Two other blacks had run in

statewide elections for which data was available and drew

littke support from either blacks or whites. The court

noted, however, that one of those candidates was a “fringe

candidate,” while the other withdrew before the election

was held. Jd. at A46-A47 n.31.

After reviewing the factors set forth in the Senate

Report on the 1982 amendment of Section 2, the district

court noted that “racially polarized voting will ordinarily

that in both the 1977 and 1979 elections there was a very high correla-

tion between the number of blacks in a precinct and the percentage ol

the votes for Jones. Pet. App. ASO. The expert calculated the percent

age of the blacks who voted for Jones by estimating, based on the ex

isting data, what the vote would be in a hypothetical precinct that was

100 percent black. See Pet. App. ASI n.33; Il Tr. 87-95.

An extreme case analysis corroborated the regression analysis

lhree of Gretna’s precincts have fewer than ten black voters. Analysis

of the vote in these homogeneously white precincts showed that vit

tually no one in those precincts voted for Jones. Pet. App. A45, AS2

Gretna has no homogenous black precincts.

These two techniques —bivariate regression analysis and extreme

case analysis—were also used by the plaintiffs in Thornburg v

Gineles, 478 U.S. 30 (1986), and provided the basis for the tinding ot

a Section 2 violation that was affirmed by this Court

be the keystone of a dilution case.” Pet. App. A62. Given

its findings of fact, the district court then concluded that

the use of the at-large system should be prohibited under

Section 2. Pet. App. A67. The court ordered the defend-

ants to propose a new election plan, and subsequently

authorized elections in which four aldermen are elected

trom single-member districts, and one is elected at-large.

Id. at Al7 n.20.° Leo Jones was elected to represent a

predominantly black district in the first election held

under that system. Br. in Opp. 20 n.4.

3. The court of appeals affirmed. Pet. App. Al-Al7.

From the language of the statute and this Court’s decision

in Thornburg v. Gingles, 478 U.S. 30 (1986), the court of

appeals concluded that “[a]t the heart of a § 2 vote dilu-

tion claim lies the issue of whether minorities have an

equal opportunity to elect their candidates of choice.” Pet.

App. A3. “Racial bloc voting,” the court further con-

cluded, “is the linchpin of a § 2 vote dilution claim.” Jd. at

A7. The court of appeals held that the district court had

not erred in finding strong evidence of racial bloc voting.

Id. at Al7.

Petitioners attacked the district court’s findings by

offering a factual argument based upon calculations not

presented to the disirict court that were set forth for the

first time in a lengthy appendix to their brief in the court

of appeals. Petitioners urged that the district court had

erred in finding that blacks could not elect candidates of

their choice because, they argued, Jones never was the

black voters’ preferred candidate. On the contrary, they

claimed, he received only 49 percent of the votes of black

voters in 1979, and, in 1977, two white candidates out-

polled him among blacks. They further argued (Pet. C.A.

> This plan was precleared by the Attorney General under Section 5

of the Voting Rights Act of 1965, as amended, 42 U.S.C. 1973c.

Br. 30) that in 1981, when none of the candidates was

black, four of the five winners received biracial support

and only one was elected without substantial black sup-

port. See Pet. App. Al2.

The court of appeals first held, contrary to petitioners’

new factual allegations, that the district court's findings

were not clearly erroneous. Pet. App. A7. In addition, the

court concluded that even if petitioners’ new factual

allegations were correct, that would not undermine the

district court’s finding of racial bloc voting. If Jones had

obtained 49 percent of the black vote in the 1979 special

election rather than the 66 percent the plaintiffs claimed,

the court explained, that would still be a sufficiently

significant proportion of the black vote to qualify Jones as

the preferred candidate of the black voters. Jd. at Al12.

The court noted that, according to petitioners’ new

analysis, no white candidate in the eight-way race for one

seat received more than 15 percent of the black vote. Thus,

even under petitioners’ figures Jones received three times

more support among black voters than any other can-

didate. In addition, according to petitioners’ new figures,

Jones received only 1.5 percent of the white vote. /d. at

Al2 n.12.

As for the 1977 and 1981 elections, the court of appeals

concluded that since five seats were at issue and there was

only one black candidate in 1977 and none in 1981, “it was

virtually unavoidable that certain white candidates would

be supported by a large percentage of Gretna’s black

voters.” Pet. App. Al3. More significant, the court found,

was that the black candidate lost in 1977, even though he

received 67 percent of the black vote, because very few

whites voted for him. /bid. Moreover, the court held, it

was proper for the district court to focus, in its analysis of

racially polarized voting, upon those elections in which

there was a black candidate. Jd. at Al5-A17. The court

noted that in Gingles this Court upheld a finding of racial

bloc voting based on analysis of elections in which blacks

ran for office. Jd. at Al7.

Finally, the court of appeals addressed petitioner’s claim

that the district court erred in considering the non-

aldermanic elections involving Jeffers and Jackson. The

court held that, while such so-called “exogenous” elections

could not, alone, make a case of racial bloc voting as to

elections for Gretna’s Board of Aldermen, “the district

court properly considered them as additional evidence of

bloc voting—particularly in light of the sparsity of

available data.” Pet. App. Al3.

DISCUSSION

This case involves a straightforward application of Sec-

tion 2 of the Voting Rights Act and this Court’s decision in

Thornburg v. Gingles, 478 U.S. 30 (1986). The courts

below correctly applied established law to the facts o1 the

case, and there is no conflict between the decision below

and the decision of any other court of appeals. Petitioners’

arguments —that the courts erred in finding Leo Jones to

be the blacks’ preferred candidate and by considering the

elections in which Ben Jeffers ran for statewide office and

Jesse Jackson ran for president —are based principally on

factual allegations that were never brought to the attention

© In Gingles, this Court held that there are three preconditions to a

finding that a multimember electoral district operates to impair

minority voters’ ability to elect representatives of their choice: “First,

the minority group must be able to demonstrate that it is sufficiently

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

member district. * * * Second, the minority group must be able to

show that it is politically cohesive. * * * Third, the minority must be

able to demonstrate that the white majority votes sufficiently as a bloc

to enable it —in the absence of special circumstances * * * —usually to

defeat the minority’s preferred candidate.” 478 U.S. at 50-51.

of the district court and were not accepted by the court of

appeals. Accordingly, further review is not warranted.

1. The district court found racially polarized bloc

voting on the basis of five elections, including three elec-

tions for the Board of Aldermen. The district court

analyzed each of those three cases (1973, 1977, 1979) in

detail, and found that black candidates Romain and

Jones, who received a substantial proportion of the vote

among blacks, were defeated by whites voting as a bloc.

The court of appeals concluded that the district court’s

findings were not clearly erroneous. That evidence is suffi-

cient to meet the requirement of Ging/es that “the minority

must be able to demonstrate that the white majority votes

sufficiently as a bloc to enable it —in the absence of special

circumstances * * *—usually to defeat the minority’s pre-

ferred candidate.” 478 U.S. at 51. Indeed, the evidence

showed that whites never supported a black candidate for

the Board of Aldermen— Romain finished last in all but

the two most heavily black precincts in 1973 after the

newspaper pointed out that he was black, and Jones

received 12 percent and one percent of the white vote in

1977 and 1979—although they received substantial sup-

port from black voters.’

Ignoring the findings of the two courts below, peti-

tioners present the case as if it were devoid of any elections

in which whites, voting as a bloc, defeated the blacks’ pre-

ferred candidate. Specifically, petitioners contend (see,

e.g., Pet. 21) that the courts below misconstrued Section 2

to protect black candidates, as such, regardless of whether

they are the preferred candidates of black voters. Were

’ Petitioners raise no challenge to the sufficiency of the evidence of

vote dilution by offering nonracial “explanations of the reasons why

white voters rejected minority candidates.” Gingles, 478 U.S. at 100

(opinion of O’Connor, J.); see also id. at 83 (opinion of White, J.).

10

this the case we would agree that there had been error,

since the Voting Rights Act is aimed at protecting minority

voters rather than minority candidates. But no such ques-

tion is in fact presented because petitioners’ allegations are

not consistent with the facts found below and are unsup-

ported by the record.

The centerpiece of petitioners’ argument is their claim

that Leo Jones was never the preferred candidate of the

black community and that, therefore, the 1977 and 1979

elections are not examples of racially polarized bloc

voting. Petitioners do not dispute that Jones received vir-

tually no support among white voters. They deny,

however, that blacks supported Jones in sufficient

numbers to support the finding of racially polarized

voting. But the district court’s contrary findings, which are

subject to review under the clearly erroneous standard

(Gingles, 478 U.S. at 79), are plainly correct.

The district court found that Jones had received about

67 percent of the black vote in 1977 (Pet. App. A46), and

petitioners do not question that finding. The district court

made no finding below regarding the proportion of black

votes received by any white candidate, and the court of ap-

peals declined to make its own finding on the basis of

allegations made for the first time on appeal. /d. at Al0.

Petitioners nevertheless contend that Jones was not the

preferred candidate of black voters in 1977 because, ac-

cording to petitioners, two of the white candidates who

won seats in that same election also received a ma-

jority —indeed, a large majority —of the black vote.*

’ Petitioners cite to nothing but their own brief in the court of ap-

peals for the proposition that two white aldermanic candidates out-

polled Jones among blacks in 1977. Pet. 10. To estimate the percent-

age of the black voters who voted for white candidates Le Boeuf and

Bush, petitioners performed the same kind of regression analysis that

11

The court of appeals correctly found that factual allega-

tion unavailing. Assuming that some of the white can-

didates in the 1977 election won with black support, that

does not vitiate the fact that the black candidate lost

because of racially polarized voting. As the court of ap-

peals explained: “Gingles detines bloc voting as white ma-

jority bloc voting that defeats minority preferred can-

didates. * * * In a multiple seat election such as Gretna’s

aldermanic race, the minority necessarily will have more

than one preferred candidate. In the Gretna elections

studied, blacks exercised their right to vote to fill all at-

large positions, but only one available candidate was

black. Thus, it was virtually unavoidable that certain

white candidates would be supported by a large percentage

of Gretna’s black voters. Significance lies in the fact that

the black candidate preferred by the minority was defeated

by white bloc voting. That blacks also support white can-

didates acceptable to the majority does not negate in-

stances in which white votes defeat a black preference.”

Pet. App. Al3. Thus, the 1977 election shows that while

blacks in Gretna supported some white candidates, whites

in Gretna refused then (and every other time a black ran)

to support a black candidate who was widely supported by

black voters. As the court below stated, that is evidence of

racial bloc voting.

In the 1979 special election for a single seat on the Board

of Aldermen, the district court found that Jones received

respondents’ expert had done with respect to Jones. As the scat-

tergram in the appendix to their court of appeals brief (at A22) shows,

however, there was almost no correlation between the percentage of

black voters in a precinct and the vote for LeBoeuf or Bush. While it is

theoretically possible to extrapolate to an intercept and estimate the

percentage of blacks who voted for the white candidates in that cir-

cumstance, the result is unreliable. See H. Blalock, Social Statistics

396-397 (Rev. 2d ed. 1979).

12

about 66 percent of the vote among blacks but was again

defeated because he received virtually no votes from

whites. Once more, petitioners do not contend that Jones

received the support of white voters, but they maintain

that he was not the blacks’ preferred candidate because he

actually received only 49 percent of the votes of blacks

who voted. The court of appeals did not overturn the

district court’s finding that Jones received 66 percent of

the black vote, which is the correct figure.? In any event,

even if Jones had received “only” 49 percent of the black

vote in the eight-way race for one seat in 1979, he was

plainly the heavy favorite of the black voters. By peti-

tioners’ Own estimates, Jones received far more support

among blacks than did any other candidate; the white can-

didate receiving the next highest proportion of the black

vote (Weigel) received only 15 percent. Pet. App. Al2

n.iZ.

% Petitioners estimated the percentage of blacks who voted for

Jones in 1979 by performing a bivariate regression analysis. They

used, as One variable, the percentage of voters who voted for Jones in

each precinct, just as the respondents had—but using different

figures. See Appendix to Appellants’ Brief in the Court of Appeals

A6. For example, in Precinct 1-3, 620 persons turned out to vote

((bid.), but only 547 of them voted in the election for the Board of

Aldermen. See PX 57. Jones received 234 of the 547 votes cast for

alderman. Petitioners, however, calculated that Jones had won 234

out of 620 votes. Petitioners’ calculation, in other words, credited

Jones with a lower percentage of the total vote than he actually re-

ceived in that (and every other) precinct. Not surprisingly, when peti-

tioners extrapolated from these figures to find the percentage Jones

would have received in an hypothetical all-black precinct, they came

out with a depressed figure of 49 percent. That petitioners’ estimate is

faulty is illustrated by the fact that the sum of the black vote for all

candidates as calculated by their method is only about 86 percent

rather than 100 percent. See Pet. App. Al2 n.12 (sum of “black

preference” column).

13

Thus, the lower courts correctly concluded that the elec-

tions in which Jones ran, together with the evidence re-

garding the candidacies of Romain, Jeffers, and Jackson,

offered sufficient evidence of racial bloc voting to support

a finding of dilution. The lower courts also relied, of

course, On a wide variety of other evidence in concluding

that, under the totality of the circumstances, there was a

Section 2 violation. See Pet. App. A7, Al7, A22-A43,

A55-A58, A64-A67. Petitioners’ contention that the court

of appeals’ decision protects black candidates who are un-

supported by black voters has no basis in either the opin-

ions below or the record.!®

2. There is also no merit in petitioners’ contention that

the district court was obliged to ignore the evidence of

racially polarized bloc voting in the 1984 Democratic

presidential primary and the 1979 election for Louisiana

Secretary of State. To be sure, if blacks had succeeded in

electing their preferred candidates to Gretna’s Board of

Aldermen, there would be no violation of Section 2 here

even if whites voted as a bloc against black candidates for

other offices. There is no reason to conclude, however,

that an election which reflects the racial voting patterns of

the constituency of the jurisdiction at issue—in this case,

the City of Gretna —is irrelevant.'' As the court of appeals

'© Petitioners purport to raise the broader question (see Questions

Presented 6 and 8) whether respondents could prove racial bloc voting

it the record reflected a pattern, in elections having only white can-

didates, of the black-supported candidates winning. Petitioners make

no argument to this effect in the body of their petition. Nor could

they. No evidence was presented to the district court, by either side,

regarding black voting patterns in elections for the Board of Aldermen

other than those in 1973, 1977, and 1979.

'! Petitioners consistently refer to the 1984 Democratic presidential

primary and the 1979 election of Louisiana Secretary of State as “ex-

ovenous” elections. See, e.g., Pet. 20. Contrary to the suggestion that

14

said, while such “elections alone could not prove racially

polarized voting in Gretna aldermanic elections,” they

may be “properly considered * * * as additional evidence

of bloc voting.” Pet. App. Al3. Indeed, such elections

have typically been considered in the analysis of racial bloc

voting in Section 2 cases. See, e.g., Solomon vy. Liberty

County, 865 F.2d 1566, 1575-1576 & n.15 (11th Cir. 1988);

City of Carrollton Branch of the NAACP v. Stallings, 829

F.2d 1547, 1556 (11th Cir. 1987), cert. denied, 108 5. Ct.

1111 (1988); Smith v. Clinton, 687 F. Supp. 1310, 1315

(E.D. Ark.), aff'd, 109 S. Ct. 548 (1988).

Contrary to petitioners’ assertion (Pet. 17-18), there is

no conflict between the lower courts’ analysis of the Jef-

fers and Jackson races and the Eleventh Circuit’s decision

in Carrollton. In that case, the court of appeals held only

that the district court did not err when it omitted two elec-

tions from its analysis of racial bloc voting as atypical.

Specifically, the district court there had not considered a

city election where the constituency was substantially

smaller than the county commission election system under

attack, and a statewide judicial election in which a black

ran as an incumbent and won. 829 F.2d at 1558-1559.

Nothing in the decision in Carrollton is inconsistent with

the court’s decision in the present case to consider other

elections, featuring prominent black candidates, in which

every Gretna precinct participated, and in which there was

a clear pattern of racially polarized bloc voting. Indeed,

the absence of a conflict is shown by the fact that the court

of appeals in Carrellton approved the district court’s con-

sideration of the votes cast for Jackson in that jurisdiction

in the 1984 Democratic primary. /d. at 1556.

“exogenous” implies, voters in Gretna participated in those elections,

and the only figures that were studied concerned how Gretna residents

voted. No votes cast outside Gretna were analyzed.

15

Finally, there is no merit to petitioners’ contradictory

complaint (Pet. 16-17) that the district court erred by not

considering in detail the statewide elections in which black

candidates received little support from anyone. As the

district court explained, one of those candidates withdrew

before the election, so little could be learned from an

analysis of the support that he received on election day

despite his withdrawal. Pet. App. A46-A47 n.31. And in

the case of the other candidate, described by the district

court as a “fringe candidate” (ibid.), it could not be said

that whites voting as a bloc did or did not defeat the choice

of the black voters in the two elections in which he ran,

since he was not the black voters’ choice. The courts below

properly decided that those elections were not relevant to

determining whether “the white majority votes sufficiently

as a bloc to enable it * * * usually to defeat the minority’s

preferred candidate.” Gingles, 478 U.S. at 51. In arguing

that those elections are relevant, petitioners make the very

error they incorrectly accuse respondents of making — con-

fusing candidates who are black with candidates who are

supported by black voters.