Hall v. Holder Reply Brief of Appellants

    IN THE UNITED STATES COURT OF APPEALS
FOR THE

ELEVENTH CIRCUIT

CASE NO. 91-8306

REV. E.K. HALL, SR., DAVID WALKER, U.S. DONALDSON, RICHARD 
HARRIS, WILLIE ATES, REV. WILSON C. ROBERSON, and NAACP Chapter 
of Cochran/Bleckley County

vs.
Plaintiffs-Appellants,

JACKIE HOLDER, individually and in his official capacity as 
County Commissioner for Bleckley County, Georgia; and ROBERT 
JOHNSON, individually and in his official capacity as 
Superintendent of Elections for Bleckley County, Georgia,

Defendants-Appellees.

ON APPEAL FROM THE UNITED STATES DISTRICT COURT 
FOR THE MIDDLE DISTRICT OF GEORGIA 

MACON DIVISION

REPLY BRIEF OF APPELLANTS

Christopher Coates 
Georgia Bar No. 170980 
111 West Washington Street 
Milledgeville, Georgia 31061 
(912) 453-9512
Laughlin McDonald 
Neil Bradley 
Kathy Wilde 
Mary Wyckoff
American Civil Liberties Union 
44 Forsyth Street, N.W. 
Atlanta, Georgia 30303 
(404) 523-2721

COUNSEL FOR APPELLANTS



t ab le of c o n t e n t s

TABLE OF CONTENTS....................................... 1
TABLE OF AUTHORITIES...................................  11
STATEMENT OF THE ISSUES.................................  1
ARGUMENT AND CITATIONS OF AUTHORITY....................  3

CONCLUSION..............................................  25
CERTIFICATE OF SERVICE



TABLE OF AUTHORITIES

Avery v . Georgia, 345 U.S. 559 (1953).................  6
Anderson v . Martin, 375 U.S. 399 (1964)...............  6
Bell v^ Southwell, 376 F.2d 659 (5th Cir. 1967) .....  6
Campos v. City of Baytownj Texas, 840 F.2d 1240 (5th 
Cir. 1988) .........................................  13> 24
'^Carrollton Branch NAACP v. Stallings, 829 F . 2d 1547 
(11th Cir. 1987) . ................................  9, 10, 14
Citizens for a Better Gretna v . City of Gretna, 636 
F.Supp 1113 (E.D.La. 1986) . ........................  13, 14
Gingles v . Edmisten, 590 F.Supp. 345 (E.D.N.C. 1984).. 18

Harris v. S iegelman, 695 F.Supp. 517 (M.D. Ala.
1988)  ................................................  7 > 8
Harris v. Graddic_k, 593 F. Supp. 128 (M.D. Ala.
1984)  ................................................  7 > 8
Jackson v. Edge field County, South Carolina School 
District,~~650 F. Supp. 1176 (D.S .C. 1986).............  18
Lodge v . Buxton, 639 F.2d 1358 (5th Cir. 1981) .....  18
Lucas v. Townsend, 908 F .2d 851 (11th Cir. 1990)....  7
NAACP of Cochran/Bleckley County v . Bleckley County,
Civ. No. 88-32-MAC (M. D . Ga . ) . . . ........................  5
Over ton v . Austin, 871 F.2d 529 (5th Cir. 1989)......  10
*Rodger s v . Lodge, 458 U.S. 613 (1982).......  18, 21, 22

*Thornburg v. Gingles, 478 U.S. 30 (1986) ...............
........ 7.77.. 6, 9, 10, 12, 13, 14, 15, 16, 17, 18, 20

^United States v. Marengo County Commission, 731 F.2d_ 
1546 (1 1th Cir. 1984) . 777......................... 5, 7, 8

ii



Wh i tus v . Georgia, 385 U.S. 545 (1967)................  °
Upqtweeo Citizens for Better Government Westwego, 906
F . 2d 1042 (5th Cir . 1990) . ..........................  17

OTHER AUTHORITIES

S. Rep. No. 417, 97th Cong., 2d Sess. 28-9 (1982).... 5, 24

H.R. Rep. No. 227, 97th Cong., 1st Sess. 30 (1981).....  24
Wildgen, "Adding Thornburg to the Thicket: The Ecological 
Fallacy and Parameter Control in Vote Dilution Cases,
20 Urban Lawyer 155, 169 (1988).......................... 15



STATEMENT OF THE ISSUES

1. Whether the district court erred as a matter of law by 

refusing to consider evidence other than that drawn from prior 

elections, such as socio-economic or other barriers to voting and 
political participation, the history of segregation, racial 
campaign appeals, and the testimony of experienced local politi­
cians, in holding that plaintiffs' failed to prove that voting in 

Bleckley County, Georgia was racially polarized?
2. Whether the district court was clearly erroneous in 

finding that plaintiffs failed to prove the existence of racially 

polarized voting?
3. Whether the district court erred as a matter of law in 

holding that plaintiffs failed to prove that the black community 

was politically cohesive because election returns did not show "a 
pattern of bloc voting" or "a pattern of unified support" for black 

candidates, and by failing to consider other evidence of cohesive­
ness, such as a distinctive socio-economic status and a history of 

segregation and discriminatory treatment of the black community?
4. Whether the district court was clearly erroneous in 

holding that plaintiffs failed to prove that the black community 

was politically cohesive?
5. Whether the district court erred as a matter of law in 

holding that the issue of whether the majority vote requirement for 

the Bleckley County Commissioner had a discriminatory purpose or 

effect was not properly before the court because (a) the majority

1



vote statute was a part of state law which the county lacked power 

to change, and (b) the claim that the majority vote statute 
violated the Voting Rights Act was the subject of pending litiga­
tion in another district?

6. Whether the district court erred as a matter of law in 
holding that plaintiffs failed to produce "any evidence" that the 

at-large elected, sole commissioner form of government was adopted 
or was being maintained with a racially discriminatory purpose 
because they failed to present specific, smoking gun evidence, and 
in refusing to consider circumstantial evidence of. racial purpose?

7. Whether the district court was clearly erroneous in 
finding that the at-large elected, sole commissioner form of 

government was not adopted or being maintained with a racially 
discriminatory purpose?

8. Whether the district court erred in concluding that black 
voters were not denied the equal opportunity to participate in the 
political process and elect candidates of their choice by the at- 
large elected, sole commissioner form of government in violation of 
Section 2 of the Voting Rights Act?

2



ARGUMENT AND CITATIONS OF AUTHORITY

I. Past Discrimination and Its Continuing Effects
The defendants argue that "any alleged past discrimination in 

Bleckley County or disparities in education, housing or income does 
[sic ] not presently hinder blacks from registering and

participating in Bleckley County politics." Brief of Appellees, p.

9. First, this argument is expressly contrary to the findings of 
the district court, findings which the defendants do not contend 

are clearly erroneous.
According to the district court, the depressed socio-economic 

status of blacks hinders their ability "to participate in the 
Bleckley County political process because... less educated people 

are more difficult to mobilize to vote even if they are registered 

to vote." (R2-59-5) The court further found that socio-economic 
barriers to minority participation in the political process "are 
today compounded by the fact that Bleckley County now has only one 
voting precinct for the entire 219 square-mile area." (R2-59-6 
n.3) Given the facts that blacks are poorer than whites and own 
fewer automobiles and telephones, and because of the lack of public 

transportation, it is much more difficult to organize and mobilize 

the black community for effective political activity, including 

campaigning, soliciting contributions, and get-out-the-vote drives.
Second, the defendants are incorrect that plaintiffs' 

witnesses established "there have been no barriers to

3



blacks... voting in Bleckley County from at least 1964 to the 

present." Brief of Appellees, p. 8. To the contrary, registration 

was permitted only at the courthouse until 1984, which discouraged 
many blacks from registering. (R4-334, 338) In addition,

plaintiff Roberson testified that because voting in the county is 
conducted at the Jaycee Barn, a facility owned by an all white 

organization, "a lot of people will not go there to vote." (R3-83) 

Third, the defendants ignore the fact that black voter turn 
out is consistently lower than white voter turn out.^ For

example, at the 1986 referendum election concerning whether the at- 
large elected sole commissioner system would be retained, black 
participation therein was "a small percentage of participation." 

(R4-321) The depressed level- of black voter participation is 
further reflected in the 1988 . exit poll conducted by the 

plaintiffs. A total of 794 whites (13% of age eligibles and 19% of 

registered voters) participated in the exit poll, but only 164 
blacks (11% of age eligibles and 17% of registered voters). (P.
Exh. 390) Socio-economic factors continue to have a present day

1 While blacks and whites are registered at approximately the 
same rates, the defendants are incorrect that blacks are registered 
at a higher rate than whites in Bleckley County. Brief of 
Appellees, p. 8. As of 1988, there were 5,327 registered voters in 
Bleckley County, 4,303 (81%) white and 1,024 (19%) black. 
According to the 1990 Census, there are 7,719 residents of voting 
age in the county, 6,245 (81%) white and other, and 1,474 (19%) 
black. Thus, 69% of the age eligible white population is 
registered, and 69% of the age eligible black population.

4



effect on minority political participation in Bleckley County 2

II. Discrimination in the Appointment Of Election Officials Is
Injury in Fact and Law
The defendants breezily dismiss plaintiffs' evidence of 

discrimination in the appointment of poll workers and managers on 
the grounds that "[n]o evidence whatsoever was presented that 

indicated black citizens have had a problem voting due to the 
number of black poll workers and managers or that the number of 
black workers and managers is attributable to race discrimination. 

Brief of Appellees, p. 14. Plaintiffs do, of course, have a 

"problem" with discrimination in the appointment of poll workers 

and managers in Bleckley County and are attempting to remedy that 
problem now in Federal District Court. See, NAACP of
Cochran/Bleckley County v. Bleckley County, Civ. No. 88-32-MAC 

(M.D.Ga.). More to the present point, the failure to appoint
blacks as poll workers and managers is a "problem" as a matter of 
fact and law, while plaintiffs are not required under the results 

test of Section 2 to prove that a challenged voting practice is

^While the district court found as a matter of fact that there 
was a causal connection between the depressed socio-economic status 
of blacks and their depressed political participation, where such 
disparities are found a causal connection is presumed as a matter 
of law. S. Rep. No. 417, 97th Cong., 2d Sess. 29 n.114 (1982); 
United States v. Marengo County Commission, 731 F.2d 1546, 1568-69 
(11th Cir. 1984).

5



motivated by a discriminatory purpose. Thornburg v. Gingles, 478 

U.S. 30, 50-1 (1986).
The courts have frequently held that some forms of 

discrimination are so egregious that they are invalid apart from 
any injury they may inflict. Thus, in jury discrimination cases 
once a significant statistical racial disparity is shown, the law 

requires relief even though the accused is unable to show any 

injury in fact. Whitus v. Georgia, 385 U.S. 545, 552 & n.2 (1967). 
It is not even necessary to prove actual discrimination if a 
racially conscious selection system affords a ready opportunity for 

its practice. Avery v. Georgia, 345 U.S. 559 (1953). In cases 
involving the electoral process, the courts have fully applied the 

rationale of these decisions.
In Anderson v. Martin, 375 U.S. 399 (1964), the Court struck 

down a Louisiana law requiring the designation of the race of each 

candidate on the ballot. In reaching its decision the Court held 
that the "vice-lies not in the resulting injury but in the placing 
of the power of the State behind a racial classification that 

induces racial prejudice at the polls." 375 U.S. at 456. Applying 
the reasoning of Anderson, the Fifth Circuit invalidated local 
elections in Sumter County, Georgia because they had been held on 
a racially segregated basis and without requiring the plaintiffs to 

show that the outcome would have been different. According to the 
court, "there are certain discriminatory practices which, apart 

from demonstrated injury or the inability to do so, so infect the 

processes of the law as to be stricken down as invalid. Bell v .

6



Southwell, 376 F.2d 659, 662 (5th Cir. 1967). The exclusionary 

practices of Bleckley County in this case are stark and flagrant, 
and they induce racial prejudice at the polls and in the elective 
process. Plaintiffs are injured by these practices as a matter of 

law, and their voting strength as a consequence is diluted.

In addition, Congress expressly provided when it amended 
Section 2 in 1982 that whether or not a minority can register and 

vote does not end the statutory inquiry. Section 2 can also be 
violated "if a protected class has 'less opportunity than other 
members of the electorate to participate in the political 
process.'" United States v. Marengo County Commission, 731 F.2d 

1546, 1556 (11th Cir. 1984)(quoting 42 U.S.C. 1973 (b)). 
According to the senate report that accompanied the 1982 
amendments, "Section 2...also prohibits practices which, while 
episodic and not involving permanent structural barriers, result in 

the denial of equal access to any phase of the electoral process 
for minority group members." S. Rep. No. 417, 97th Cong., 2d Sess. 

30 (1982). The house report is even more specific and provides 

that Section 2 extends to the "refusal to appoint minority 

registration and election officials." H.R. Rep. No. 227 , 97th 
Cong., 1st Sess. 14 (1981). The courts have, accordingly, 
consistently held that the results standard of Section 2 can be 
violated by the refusal to appoint minority election officials. 

Harris v. Graddick, 593 F.Supp. 128, 133 (M.D.Ala. 1984); Harr is v . 
Siegelman, 695 F.Supp. 517, 526 (M.D.Ala. 1988); United States v. 
Marengo County Commission, supra, 731 F.2d at 1570; Lucas v.

7



Townsend. 908 F.2d 851, 857 (lUh C i, 1990)C".a 'standard,
practice, or procedure' within the meaning of section 2 can include

the appointment of poll officials ).
The legal injury to black citizens caused by their exclusion

from poll manager and worker positions is not limited to the mere 

act of casting a ballot, although the lower turn out rate of blacks 
indicates that they are in fact intimidated or deterred from voting 
by the county's discriminatory practices, just as they are deterred 
from voting at the "all white" Jaycee Barn. The actions of the 
county also deny blacks the equal opportunity to serve as election 
officials and participate in the electoral process. They 

stigmatize blacks by stamping them as inferior and not worthy to 
participate in the conduct of elections. They erode the confidence 

blacks have in the system's fairness. See, llnited States n  

M-reneo County Commission, supra, 731 F.2d at 1570 ("lb]y having 
few black poll officials...county officials impaired black access 

to the political system and the confidence of blacks in the 

system's openness ).
The courts have recognized that discrimination in the 

appointment of poll workers and managers has a direct impact 
-political socialization." Harris v. Siegelman, su£ra, 695 F.Supp. 
at 526. That is, such discrimination fosters an official attitud 
that minority voters cannot participate in the electoral process on 
the same terms and to the same extent as non-minority voters. 

HarriS v. Graddick, supra, 593 F.Supp. at 133. This attitude in 

turn has a far reaching impact on political behavior by enhancing

8



polarized

■ r. if/tf jfvi S ^ L ^ * /v 3 » 5 X .i i

racial isolation, exacerbating racially 
depressing black political participation.

Blacks in Bleckley County are injured by 
exclusion as poll managers and workers. Such 
evidence of racial polarization and that the 
elections dilutes minority voting strength.

voting and

their discriminatory 
exclusion is strong 

existing system of'

III. Blacks Are Geographically Compact

The defendants acknowledge, as they must, that Carroll ton 
Branch of NAACP v. Stallings, 829 F.2d 1547 (11th Cir. 1987), is 
controlling, and that an at-large elected sole commissioner system 
of county government can be challenged under the vote dilution 

principles of Section 2. Brief of Appellees, pp. 18, 21. In
Carrollton Branch of NAACP, a challenge to the sole commissioner 

system in Carroll County, Georgia, the district court rejected the 

plaintiffs' evidence of geographic compactness in proposed three 
member and five member forms of commission government. The court 

of appeals reversed, holding that the exclusion of such evidence 
was error. According to the court:

The demographic evidence proffered by plaintiffs is 
highly relevant to a showing under Gingles: whether black 
voters in Carroll County are sufficiently numerous and 
compact to constitute a majority in a single member 
district, if single member districts were created out of 
the one county district. In this case, it is the at- 
large nature of the districting'scheme which is alleged 
to operate in the same fashion, as multi-member districts 
in diluting black voting strength. Clearly, the district 
court erred in disallowing such evidence under the 
Supreme Court's approach in Gingles. On remand, such 
evidence should be admitted on the question of the 
geographic concentration of the voters by race.

9 -



829 F .2d at 1563.

Despite the ruling in Carrollton Branch of NAACP, and without 

arguing that the decision was wrongly decided or should be 
overruled, defendants nevertheless claim that plaintiffs failed to 
meet the geographic compactness requirement of Gingles because 
"Bleckley County is itself a single-member district. Brief of 
Appellees, p. 21 n.21. The argument is not only foreclosed by 

Carrollton Branch of NAACP, but it is also undercut by Overton,̂  
Austin, 871 F.2d 529 (5th Cir. 1989), the only authority cited by 

the defendants in support_of their contention.
In Over ton, the district court held that Gingles permitted it 

to order an expansion of the Austin city council from six to eight, 
members to satisfy the geographic compactness standard. The 
findings by the district court, including the finding on other- 

grounds of no Section 2 violation, were affirmed by a majority of 

the panel on appeal. The opinion of the majority, rather than the 
concurring opinion of a single judge, establishes the meaning and 

precedential value of Overton. Sole commissioner systems are 

challengeable under Section 2.

^The geographic compactness requirement of Gingles, moreover, 
arose in the context of a complaint that alleged a Section 2 
violation based on "the legislative decision to^employ multimember 
rather than single-member, districts in the contes e 
jurisdictions." 478 U.S. at 46 (emphasis added) . Under the 
circumstances, it was appropriate to require plaintiffs to PJ-0 e 
geographic compactness, i.e. that single member districts could 
indeed be created. But where, as here and in Carroll County, the 
violation is alleged to be the use of an at-large elected sole 
commissioner form of government, the imposition of a geographic 
compactness requirement based on the county as a whole would be 
nonsensical. The vote dilution in this case could be easily and 
reasonably remedied by the creation of a five member board of

10



;  -t.'-.j-u. 'J: - S v .. sA J*. •' . . . . . . . . .  . , . -:u * ........ I r v . . - - . '™ I ' . v  .....g

IV. Voting Is Racially Polarized
Defendants make various claims regarding the evidence of 

racially polarized voting in Bleckley County. They are discussed 

below.

A . Plaintiffs Analyzed all the Relevant Elections 
Defendants say that plaintiffs "did not analyze the results of 

any county commissioner elections or any other state or local 
elections to determine if there was evidence of racial bloc voting 
for the challenged mechanism." Brief of Appellees, p. 9 (emphasis 
in original). Although they could not always conduct a regression 

or other statistical analysis, plaintiffs did in fact "analyze" 

state, county and municipal elections in Bleckley County to 

determine the existence of racially polarized voting. That 

analysis, which is recited in Brief of Appellants, pp. 7-13, and is 
summarized in Exhibit A attached hereto, showed: few minority
elected officials; the consistent defeat of minority candidates; a 

depressed level of black candidates for at-large elected, county 
wide offices; racial bloc voting in county wide contests involving 
viable black candidates;̂  the strong showing of candidates (Lester

commissioners similar to those which exist in the majority of 
Georgia counties today.

4 One of the elections, the 1984 Democratic presidential 
preference primary, could be analyzed by ecological regression. 
The regression estimates, which were virtually identical to the 
results reported state wide by CBS News/New York Times exit polls, 
showed whites voting white, i.e. for white candidates, at the level 
of 991, and blacks voting black at the rate of 61%. Plaintiffs

11



Maddox, George Wallace and J. B. Stoner) identified with white 
supremacy; and an increased level, and success, of black candidates 

in non or less dilutive election systems utilizing majority black 

districts (board of education and city council), and in elections 
with no majority vote requirement (city council). 5

Plaintiffs did not "chose not to comment" on the results of 
the four statewide elections involving C. B. King, Hosea Williams, 
Mildred Glover and Otis Smith. .Brief of Appellees, p. 26. To the 

contrary, plaintiffs discussed the contests in their brief, Brief 
of Appellants, p. 12, noted that they involved minor black 
candidates who received little support from either black or white 
voters, and agreed with the district court that the elections were 
"simply a nullity." (R2-59-39)

County wide elections involving all white candidates are 

similarly of little probative value in this case. While Gingles

conducted an exit poll at a second election, the 1988 Democratic 
presidential preference primary, which showed 98% of whites voting 
white, and 90% of blacks voting black. In a third election for 
judge of probate, in which blacks were 13.5% of actual voters, 
plaintiff Hall lost, getting 15% of the total vote. In a fourth 
contest involving two black candidates, plaintiff Hall was elected 
to the board of education from a 66% black district.

5 For example, during the time elections were conducted at- 
large for the city council of Cochran, in which blacks are 38% of 
the population, black candidates regularly got 30-40% of the vote. 
(R3-85-87, R4-304, P. Exh. 26, 27, 28, 170, 174, 183, 205, 215, 
227, 317, 354) However, with the exception of Basby, who was 
elected initially by a plurality (39%) in a contest with two 
whites, and who subsequently ran with the advantages of incumbency 
or having been an incumbent, no black was ever able to win a seat 
on the city council. After the method of elections was changed to 
district voting, blacks were regularly elected to the two majority 
black seats, a scenario strongly indicating polarized voting among 
both black and white voters.

12



. .. - -' .~ ■

indicated that "only the race of the voter, not the race of the 

candidate, is relevant to vote dilution analysis," 478 U.S. at 68, 
that portion of the opinion was joined by only a plurality of the 

Court. In addition, all the elections relied upon by the Gingles 
Court in finding racially polarized voting were in fact black-white 

contests. 478 U.S. at 80-3. Accordingly, courts have interpreted 
Gingles as meaning that "the race of the candidate is in general of 

less significance than the race of the voter - but only within the 
context of an election that offers voters the choice of supporting 
a viable minority candidate." Citizens for a Better Gretna v, City

_Gretna , La . , 834 F.2d 496, 502 (5th Cir. 1987). Accord, Campos

v. City of Baytown, Texas. 840 F.2d 1240, 1245 (5th Cir. 1988)("the 
district court properly focused only on those races that had a 
minority member as a candidate"). The all white contests in 

Bleckley county that did not offer voters the choice of supporting 

a viable minority candidate are simply of little use in determining 
the existence of racially polarized voting.

B. Plaintiffs' Methodology Was Standard and Has Been 
Approved by the Courts

The defendants, relying upon the testimony of their expert Dr. 
Wildgen, claim that plaintiffs' regression analysis of the 1984 

Democratic presidential primary was defective because plaintiffs' 

expert Dr. Engstrom failed to apply "confidence bands." Brief of

i

-  13 -

• ? r .



Appellees, p. 28.6 It should be noted that Dr. Wildgen did the 

same analysis as Dr. Engstrom - and got the same results. (R5-674) 
The defendants' real quarrel, apparently, is not with the numbers 
Dr. Engstrom computed but with the ecological regression method of 
analysis he used. As the defendants' counsel advised the court, 
"[w]e're challenging the validity...of using the ecological 
regression analysis...I do not challenge the figures." (R5-464) 

Also— see, the testimony of Dr. Wildgen that "I don't think it's 
[ecological regression analysis] the best approach." (R5-664)

The defendants challenge to plaintiffs' methodology cannot be 
seriously credited, for regression analysis has been approved as a 
technique for showing racially polarized voting by both the Supreme 

Court in Gingles, and this court in Carrollton Branch of NAACP.7 

Moreover, Dr. Wildgen has himself used regression analysis in 
concluding that racial polarization existed among voters in New 

Orleans. See, Citizens for a Better Gretna v. City of Gretna, La..

• ^9?t}fidence bands are a technique for determining the 
reliability of regression estimates. Another test for reliability 
is statistical significance.^ See, e.g. Thornburg v. Gingles, 
supra, 478 U.S. at 53. In this case Dr. Engstrom found that "the 

[between the race of voters and the race of 
candidates] is so consistent and pronounced that it is
statistically significant at the .0001 level, meaning that there's 
one in ten thousand chances that a relationship that consistent and 
pronounced would occur by chance." (R4-430) The reliability of 
the regression ̂ estimates was further supported by homogeneous 
P^fcinct: analysis, which showed whites voting white at the level of 
y/X. These estimates were also virtually identical to those 
reported^state wide by CBS News/New York Times exit polls. Dr. 
refill111 S est:‘'mat:es voting behavior were in fact extremely

" ^Ecological regression and homogeneous precinct analysis are 
standard in the literature for the analysis of racially polarized 

voting. Thornburg v. Gingles, supra, 478 U.S. at 53 n.20.
-  14 -

!»

I



& J 5

636 F.Supp. 1113, 1130 (E.D.La. 1986). Dr. Wildgen also previously 

published a study on ecological regression in which he stated that 

significance tests, and therefore confidence bands, "are just plain 

insignificant" when applied to ecological regression analysis based 

upon a population of precincts, as in this case, as distinguished 
from a sample of precincts. See, Wildgen, "Adding Thornburg to 
the Thicket: The Ecological Fallacy and Parameter Control in Vote 
Dilution Cases," 20 Urban Lawyer 155, 169 (1988). Dr. Wildgen's 
testimony in this case is contradicted by his previous publications 
and testimony, Citizens for a Better Gretna v. City of Gretna. La.. 

supr a, 663 F.Supp. at 1130, and is inherently unreliable. It is 
nothing more than a red herring and should simply be ignored.

The defendants' attack on plaintiffs' homogenous precinct 
analysis is also off the mark. Brief of Appellees, p. 29. While 
there may be no homogeneous (90% or more) black precincts in the 

county, there have been five homogeneous white precincts, and they 
clearly reveal significant polarized voting by whites. In the 1984 

Democratic presidential preference primary, the black candidate got 
at most only 3.4% of the white vote in the five homogeneous white 
precincts. The fact that whites vote as a bloc at such a high 
level is extremely relevant to the Gingles formulation.

C. Plaintiffs' Exit Poll Is Reliable

The defendants claim that the exit poll conducted by 
plaintiffs was unreliable because it overestimated the votes

15



received by Rev. Jesse Jackson. Brief of Appellees, p. 30-1.8 

While it is true that proportionately more Jackson supporters 
participated in the poll, the poll was extremely reliable in that 
it virtually mirrored the order of finish of the candidates in the 
election. The only differences were that the poll listed Gephardt 
7th and Robertson 6th, and Haigh 13th and DuPont 12th, while 

Gephart finished 6th in the election and Robertson 7th, and Haigh 

12th and DuPont 13th. (P.Exh. 390) The poll is reliable and very 
relevant evidence of voting patterns in Bleckley County, and it was 
error for the district court to discount it.

D . A Trial Court Can Consider Evidence other than 
Election Data

The defendants erroneously contend that Gingles prohibits a 
trial court from considering evidence other than election data in 

determining the existence of racial bloc voting. Brief of 
Appellees, p. 22-3. Gingles held that polarized voting and the
extent of minority electoral success were "primary" evidence of 
vote dilution, and that proof of other factors identified in the 
legislative history of amended Section 2, such as the effects of 
past discrimination, racial appeals in campaigns and the use of 

electoral devices which enhance the opportunity for discrimination, 

are supportive of, but not essential to, a minority voter's 
claim. 478 U.S. at 48 n.15. Whatever the relevance of the

®The defendants claim that the poll overestimated Jackson 
support by 150%. To the contrary, the absolute disparity between 
the percent of the vote for Jackson in the election and in the poll 
was only 7.5%. r

16



effects of past discrimination, etc. in proving a Section 2 
violation, the Court never said or implied that they could not be 
considered in proving polarized voting or that polarized voting 
could only be proved through election data. Indeed, the Court

expressly held to the contrary.
Gingles recognized that in some cases a minority group may 

never have been able to sponsor a candidate. Under such 
circumstances, "courts must rely on other factors [than elections] 
that tend to prove unequal access to the electoral process. 478 
U.S. at 57 n .25. And, where the minority has begun just recently 

to sponsor candidates, "the fact that statistics from only one or 
a few elections are available for examination does not foreclose a

vote dilution claim." Id. Accord, Westwego Citizens for Be11er

Government v. Westwego, 906 F.2d 1042, 1045 (5th Cir. 1990).
Moreover, the district court in Gingles, in determining the extent 

to which voting in the challenged districts was racially polarized, 
considered, among other things, "the testimony of lay witnesses. 

478 U.S. at 41. The district court's analysis, as well as its 
findings of polarized voting (with the exception of one district), 

were affirmed by the Supreme Court. There is no basis for 
contending that Gingles precludes a court from considering evidence 
other than election data in determining the existence of polarized 

voting.

E . Testimony of Experienced Local Politicians Is Very 
Relevant Evidence of Polarized Voting

The defendants are simply wrong in saying that no case law
17



~ c *£ 5 . i r . f k  ^  _  r.

provides that polarized voting can be shown by the testimony of 
experienced local politicians or lay witnesses. Brief of 
Appellees, p. 32. Plaintiffs have cited numerous such cases in 
their brief, Brief of Appellants, p. 34, e .g . , Gingles v. Edmisten, 

590 F.Supp. 345, 367 (E.D.N.C. 1984), aff'd sub nom. Thornburg v. 
Gingles, supra; Jackson v. Edgefield County, South Carolina School 

District, 650 F.Supp. 1 176, 1 198 (D.S.C. 1986)("[e ]ven more 

persuasive to the Court than the experts' quantitative analysis of 
polarization on voting behavior is the testimony by the local 
politicians who, through their participation in the political 

processes, have the direct observation and are familiar with the 

voting practices and voting patterns in Edgefield County").
The defendants' error is in thinking that polarized voting can 

only be shown through statistical analysis and that evidence of how 
elections actually work is irrelevant. The fact that blacks 

consistently lose in county and city at-large elections but win in 
county and city elections from majority black single member 

districts is highly relevant evidence that voting is along racial 
lines and that at-large elections dilute minority voting strength. 
See, e .g . , Lodge v. Buxton, 639 F.2d 1358, 1378 (5th Cir. 

1981)(finding racially polarized voting and "of particular 
signi f icance. . . the fact that in the one city election in which city 
councilmen were elected from single-member districts, a Black was 
elected"), aff'd sub nom. Rogers v. Lodge, 458 U.S. 613, 623 
(1982); and Jackson v. Edgefield County, South Carolina, supra, 650

F.Supp. at 1198 (the success of black candidates in related single

- 18 -



member district elections "confirmed the political unity of each 
racial group and the cohesiveness of its voting behavior"). To 
conclude otherwise would be to ignore common sense and experience.

F . The District Court Considered only Election Data in 
Determining if Voting Was Polarized

The defendants claim that plaintiffs are wrong in saying that 
the district court refused to consider evidence other than election 
data in determining the question of polarized voting, and that the 
court in fact considered whether "'additional' evidence established 

a violation of Section 2." Brief of Appellees, p. 24. The 

defendants are in errot on two counts. First, plaintiffs d id point 
out in their brief that the district court considered "brief[ly ] " 
(the court's phrase) the various factors listed in the senate 

report that were probative of vote dilution, such as the socio­
economic status of blacks, etc., in connection with its 
determination of Section 2 liability. Brief of Appellants, p. 28. 

Second, the plaintiffs' claim was not that the district court 

refused to consider the additional senate factors in connection 
with Section 2 liability, but that the court refused to consider 
them as evidence of racially polarized voting. The distinction is 

critical, for having concluded that there was no polarized voting, 

the court's consideration of the additional senate factors on the 
issue of liability was an empty and meaningless exercise.

19



G. Plaintiffs are not: Required to Replicate the 
Data Base and Statistical Analysis in Gingles

The data base and statistical analysis- in Gingles cannot, 

and need not, be replicated in a case such as this involving a 
small rural county in which the political participation of the 
minority has been substantially repressed by a highly restrictive 

system of elections.
The district court found "insufficient available 

evidence" of polarized voting to meet the conditions for relief set 

out by Gingles. (R2-59-46 n.48) Gingles, it should be remembered, 
was a challenge to six multimember house and senate districts for 
the North Carolina general assembly and involved 14 counties, 
including the major metropolitan counties of Mecklenburg, Durham 

and Wake. Given the large populations and number of jurisdictions 
involved, the plaintiffs were able to collect and evaluate data 
from 53 general assembly primary and general elections involving 

black candidacies held over a period of three different election 
years. 478 U.S. at 35 ns.1 & 2, 52-3. Because of the number of 

jurisdictions and precincts, the Gingles plaintiffs were also able 

to make extensive use of homogeneous precinct and ecological 
regression analysis in examining the elections for polarized 

voting.
By contrast, in a small rural county such as Bleckley, 

election data and the opportunities for purely statistical analysis 

are simply not available on the same scale as in Gingles. But that 
does not foreclose a vote dilution challenge. As Gingles provides, 

where election data is sparse or unavailable, a court is required

20



to look at other factors in determining the existence of polarized 
voting and unequal access to the electoral process. 487 U.S. at 57 
n.25. To hold otherwise would be to impose a dual standard for 
Section 2 liability - one for large counties, or collections of 
counties, where data was abundant and statistical analysis was 
readily available, which would be subject to the statute, and 

another for small counties, where data and the opportunities for 
statistical analysis were less abundant or where minority political 

participation had been particularly stiffled, which would be 

exempted from the statute. Neither Congress nor the courts 
envisioned or have sanctioned such a result.

In any event, the relevant evidence of polarized voting in

this case, despite the fact that most of it was completely ignored
by the district court, was extensive. In Rogers v. Lodge, supra,

for example, which involved a challenge to at-large elections in

Burke County, Georgia, a rural county comparable in size to
Bleckley, the Supreme Court found racially polarized voting based
upon evidence that was considerably more sparse and less compelling
than here. Unlike as in this case, no regression analysis or exit
polling of any kind was done in Rogers v. Lodge. The evidence of

polarized voting consisted entirely of: one black candidate carried

the four majority black precincts while losing in the others;

another black candidate carried three of the majority black
precincts and lost in the others; a white candidate sympathetic to *
black concerns was defeated; and, a black was elected to the city 
council from a heavily black district. See, Brief of Appellants,

21



p. 36. The finding of Che district court in this case that there 
was insufficient evidence of polarized voting cannot be squared 
with the holding in Rogers v. Lodge, supra, 458 U.S. at 623, that 
"[tjhere was overwhelming evidence of bloc voting along racial 

lines" in Burke County.

V . Other Erroneous Factual and Legal Assertions
The defendants make a number of other erroneous factual and 

legal assertions which are discussed briefly below.
A . Blacks Do not Prefer the Sole Commissioner System 

The defendants contend that "the citizens of Bleckley County
prefer the sole commissioner form of government." Brief of 
Appellees, p. 5.9 The contention is clearly not true for the 

black citizens of the county, whose voting strength is diluted by 

the at-large elected sole commissioner system and who seek the 

implementation of a board of commissioners elected from districts.
B. The Black Community Is Politically Cohesive

In claiming that blacks in Bleckley County are not politically 

cohesive, the defendants cite the testimony of their expert Dr. 
Loftin for the proposition that there is "no evidence of political 

organization among black citizens." Brief of Appellees, p. 34. 
What Dr. Loftin actually said was that she had found no evidence of 
"[£] political organization" among black citizens. (R5-

601)(emphasis added) Of course, Dr. Loftin and the defendants are 
wrong. There is not only "political organization" among blacks, as

^The defendants' record cite for this proposition, R5-618, 
makes no reference at all to the alleged preferences of the 
citizens of Bleckley County.

22



this litigation and the record in this case demonstrate, but there 
is "a political organization" in the black community - the 
Concerned Citizens Committee. (R4-286, 296)

The defendants also take Dr. Engstrom's testimony out of 

context when they claim he said "no determination could be made 
regarding political cohesiveness in plaintiff Hall's successful 
campaign for the school board." Brief of Appellees, p. 35. What 
Dr. Engstrom actually said was that "if there are two black 
candidates in the race, you can conclude that one is the preference 

of black voters." (R4-438) What he could not determine without 
"sign-in" data was which one of the blacks was most preferred by 
black voters. (R4-439)10

C . Responsiveness Is not an Issue in this Case 

The defendants claim that responsiveness of white elected 
officials "was a valid consideration for the court" in resolving 

the Section 2 claim. Brief of Appellees, p. 37. To the contrary, 

the legislative history of Section 2 provides that evidence of

Although it is not possible to do regression analysis, it 
is sometimes possible to estimate by simple arithmetic the minimum 
level of bloc voting by blacks in the majority black districts and 
thus the political cohesiveness of the black community. For 
example, in the 1986 election for city council Post II, Roberson, 
a black, got 492 (841) of the votes, and his white opponent got 96 
(161) of the votes. (R3-85-87) According to Roberson, the 
registration in the district was approximately 800 (731) black and 
300 (271) white. ■(R3-85-7) A total of 588 votes were cast in the 
election, meaning that some 53.51 of the registered voters turned 
out (588 - 1100 = 53.51). Assuming the same level of turn out for 
black and white voters, approximately 428 blacks (800 x 53.51 = 
428) and 161 whites (300 x 53.5 = 161) voted. Further, assuming 
that all the white voters voted for Roberson, Roberson got a 
minimum of 331 black votes (492 - 161 = 331), or 671 of the black 
vote. Even this unrealistically conservative estimate reveals 
substantial political cohesion in the black community.

23



i.'' v ’v >  . jiV TL-.'i-> i n t u ' .

responsiveness is to be avoided in Section 2 cases because at as a
highly subjective factor which "creates inconsistencies among court

decisions on the same or similar facts and confusion about the law
nr officials and voters.” H.R. Rep. No. 227, su£ra, among government officials

at 30. in addition, evidence of responsiveness cannot rebut a 

plaintiffs showing of dilution under Section 2. S. Rep. No. 417, 

supra, at 29 n.116. Accord, Campos v. Catv of Baytown, Texa^, 
supra. 840 F.2d at 1250. Plaintiffs made no claim that defendants 
were unresponsive. Accordingly, responsiveness is not an issue in

this case,

j c t-hflt- the sole commissioner system hasThe claim by defendants that tne
been used without change from the creation of Bleckley County to 

the present" is incorrect. Brief of Appellees, p. 39. The 
adoption of a majority vote requirement in 1964, first by the 

county Democratic party and then by the general assembly, was a 
major change in the sole commissioner system.!' It put it beyond 
the power of the black minority to elect a candidate of its choice 

by a plurality of the votes, and thus insured absolute control by 

the bloc voting white majority. The adoption of the majority vote 
requirement significantly enhanced the dilution of minority voting 

strength, and has deterred or discouraged blacks from running for

county wide office.

” " M le  “J ' L°£ m 8teStthereed m V
requirement1 ŵ is u V U  to 1964.

24



The defendants also say that plaintiffs' challenge to the use 

of the majority vote requirement "makes no sense because 
plaintiffs did not state a separate claim that the requirement 
violated Section 2 or the Constitution. Brief of Appellees, p. 49. 
Plaintiffs contend that by enacting the majority vote requirement 
in 1964, the general assembly reshaped at-large election systems 
wherever they existed in the state, including in Bleckley County, 
into more secure mechanisms for discrimination. It would be
possible, however, to conduct elections for a board of
commissioners utilizing a majority vote requirement, e _ ^  by using 
fairly drawn single member districts. Properly understood, 

plaintiffs' challenge to the majority vote requirement makes 

perfect sense.
Conclusion

The evidence in this case meets the three part test in 
Gingles. It also shows that blacks in Bleckley County are denied 

equal access to the electoral process. The judgment below should 
be reversed and the case remanded for the implementation of an 
appropriate remedy utilizing a board of commissioners elected from 
districts, or in a manner that does not dilute minority voting

strength.
ChTfstoph^r Coates 
Georgia Bar No. 170980 
111 West Washington Street 
Milledgeville, Georgia 31061
Laughlin MacDonald, Neil Bradley 
Kathleen Wilde, Mary Wyckoff 
American Civil Liberties Union 
44 Forsyth Street N.W.
Atlanta, Georgia 30303

Attorneys for Appellants

25



I.
EXHIBIT A

NON-MUNICIPAL ELECTIONS INVOLVING VIABLE BLACK CANDIDATES

CONTEST YEAR CANDIDATE METHOD OF 
ELECTION

RESULT ANALYSIS

Judge of 
Probate

1984 Hall Countywide L03t Got 15% of vote. 
Blacks 13.5% of 
actual voters.

Democratic 
Pres iden- 
tial
Preference
Primary

1984 Jackson Countyvide Finished 
fourth in 
field of 
nine
candidates

Regression and 
Homogeneous 
Precinct Analysis 
showed Jackson got 
61 % of black vote 
and less than 1% 
of white vote.1

Board of 
Education

1986 Hall Single
member
district

Won Blacks were 66% of 
District 
Population.

Democratic 
Pres iden 
tial
Preference
Primary

1988 Jackson Countywide Finished 
second in 
field of 
seven
candidates

Exit Poll showed 
Jackson got 90% of 
Black Vote and 
less than 8% of 
white vote.

1 These estimates 
results of the 1984 CBS were virtually identical to the statewide 

News/New York Times exit polls.



EXHIBIT A
II. CONTESTS IN WHICH CANDIDATES WERE STRONGLY IDENTIFIED WITH WHITE

SUPREMACY

CONTEST YEAR CANDIDATE METHOD OF 
ELECTION

RESULT ANALYSIS

Governor 
(Genera 1 
election)

1966 Maddox Countyvide Finished
first

Got 69% of 
tota 1 
vote .

Democratic 
Presidential 
Preference 
Pr imary

1968 Wallace Countyvide Finished
first

Won a
"solid"
majority.

Governor 
(Primary 
election)

1974 Maddox Countyvide Finished 
first in 
field of ten 
candidates

Got 44% of 
total 
vote .

Lt. Governor 
(Primary 
Election)

1974 Stoner Countyvide Finished 
second in 
field of ten 
candldates

Got 20% of 
total 
vote .



EXHIBIT A
III. MUNICIPAL ELECTIONS INVOLVING BLACK CANDIDATES

YEAR BLACK
CANDIDATE

METHOD OF 
ELECTION

RESULT ANALYSIS

1972 Basby At-large Lost Got 39% of vote in contest 
with one white.

1973 Basby At-large Won Got a plurality of vote (39%) 
in a contest with two whites.

1975 Basby At-large L03t Got 48% of vote in a contest 
with one white.

1976 Harris At-large L03t Got 22% of vote in contest 
.with three whites.

1976 Howard At-large Lost Got-34% of vote in contest 
with one white.

1977 Black3hear At-large L03t Got 40% of vote in contest 
with one white.

1977 Basby At-large Won Got 50.5% of vote in contest 
with two whites. Ran as 
former incumbent.

1978 Pitts At-large Lost Got 32% of vote in contest 
with one white.

1978 Hall 'At-large Lost Got 43% of vote in contest 
with one white.

1979 Ba3by At-large Won Got 70% of vote in contest 
with one white. Was an 
incumbent.

1980 Pi t t 3 At-large Lost Got 7% of vote in contest with 
two whites.

1980 Walker At-large Lost Got 20% of vote in contest 
with two whites.

1981 —Basby At-large Won Unopposed, Incumbent.
1982 Walker At-large Lost Got 23% of vote in contest 

with two white3.
1982 McDonald - At-large Lost Got 37% of vote in contest 

with one white.
1983 Basby At-large Won Unopposed, incumbent.



YEAR
■

BLACK
CANDIDATE

—

METHOD OF 
ELECTION

RESULT ANALYSIS

1984 Harr is At-large Lost Got 39% of vote in contest 
vi th one white .

1985 Basby At-large "Won Unopposed, incumbent.
1986 Roberson.__" Single

member
district

Won Got 84% of vote in contest 
with one white. District is 
71% black.

1987 Basby Single
member
district

Won Unopposed, incumbent. 
District is 71% black.

1988 Roberson Single
member
district

Won Unopposed, incumbent. 
District is 71% black.

1989 Basby Single
member
district

Won Unopposed, incumbent. 
District i3 71% black.



■ «v». U'-.'ris’w .  *. j ‘  i  O. «i'*- «I x  U

!L
t.

f
(
!)

CERTIFICATE OF SERVICE

I hereby certify that I have this day mailed a true and correct

copy of the Reply Brief of Appellants to counsel for Appellees in

envelopes addressed to them and having affixed thereto sufficient
postage prepaid thereon to ensure delivery as follows:

Mr. R. Napier Murphy 
Mr. John C. Daniel, III 
240 Third Street 
P.0, box 1606
Macon, Georgia 31202-1606
Mr. W. Lonnie Barlow 
P.0. Box 515 
Cochran, Georgia 31014

This 27th day of August, 1991.

.a. .ChristO/pher Coates 
Attorney for Plaintiffs