Appeal from the United States District Court for the Eastern District of North Carolina

    (Slip Opinion) 

NOTE: Where it is feasible, a syllabus (headnote) will be released, as is 
being done in connection with this case, at the time the opinion is issued. 
The syllabus constitutes no part of the opinion of the Court but has been pre­
pared by the Reporter of Decisions for the convenience of the reader. See 
United States v. Detroit Lumber Co., 200 U. S. 321, 337. 

SUPREME COURT OF THE UNITED STATES 

Syllabus 

THORNBURG, ATTORNEY GENERAL OF NORTH 
CAROLINA, ET AL. v. GINGLES ET AL. 

APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE 
EASTERN DISTRICT OF NORTH CAROLINA 

No. 83-1968. Argued December 4, 1985--Decided June 30, 1986 

In 1982, the North Carolina General Assembly enacted a legislative re­
districting plan for the State's Senate and House of Representatives. 
Appellees, black citizens of North Carolina who are registered to vote, 
brought suit in Federal District Court, challenging one single-member 
district and six multimember districts on the ground, inter alia, that the 
redistricting plan impaired black citizens' ability to elect representatives 
of their choice in violation of§ 2 of the Voting Rights Act of 1965. After 
appellees brought suit, but before trial, § 2 was amended, largely in re­
sponse to Mobile v. Bolden, 446 U. S. 55, to make clear that a violation 
of § 2 could be proved by showing discriminatory effect alone, rather 
than having to show a discriminatory purpose, and to establish as the 
relevant legal standard the "results test." Section 2(a), as amended, 
prohibits a State or political subdivison from imposing any voting quali­
fications or prerequisites to voting, or any standards, practices, or pro­
cedures that result in the denial or abridgment of the right of any citizen 
to vote on account of race or color. Section 2(b), as amended, provides 
that § 2(a) is violated where the "totality of the circumstances" reveals 
that "the political processes leading to nomination or election . . . are not 
equally open to participation by members of a [protected class] ... in 
that its members have less opportunity than other members of the elec­
torate to participate in the political process and to elect representatives 
of their choice," and that the extent to which members of a protected 
class have been elected to office is one circumstance that may be consid­
ered. · The District Court applied the "totality of the circumstances" test 
set- forth in § 2(b) and held that the redistricting plan violated § 2(a) be­
cause it resulted in the dilution of black citizens' votes in all of the dis­
puted districts. Appellants, the Attorney General of North Carolina 



II THORNBURG v. GINGLES 

Syllabus 

and others, took a direct appeal to this Court with respect to five of the 
multimember districts. 

Held: The judgment is affirmed in part and reversed in part. 

590 F. Supp. 345, affirmed in part and reversed in part. 
JUSTICE BRENNAN delivered the opinion of the Court with respect to 

Parts I, II, III-A, III-B, IV-A, and V, concluding that: 
1. Minority voters who contend that the multimember form of district­

ing violates § 2 must prove that the use of a multimember electoral struc­
ture operates to minimize or cancel out their ability to elect their pre­
ferred candidates. While many or all of the factors listed in the Senate 
Report may be relevant to a claim gf vote dilution through submergence 
in multimember districts, unless there is a conjunction of the following 
circumstances, the use of multimember districts generally will not im­
pede the ability of minority voters to elect representatives of their 
choice. Stated succinctly, a bloc voting majority must usually be able 
to defeat candidates supported by a politically cohesive, geographically 
insular minority group. The relevance of the existence of racial bloc 
voting to a vote dilution claim is twofold: to ascertain whether minority 
group members constitute a politically cohesive unit and to determine 
whether whites vote sufficiently as a bloc usually to defeat the minority's 
preferred candidate. Thus, the question whether a given district ex­
periences legally significant racial bloc voting requires discrete inquiries 
into minority and white voting practices. A showing that a significant 
number of minority group members usually vote for the same candidates 
is one way of proving the political cohesiveness necessary to a vote dilu­
tion claim, and consequently establishes minority bloc voting within the 
meaning of § 2. And, 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 voting. Because 
loss of political power through vote dilution is distinct from the mere in­
ability to win a particular election, a pattern of racial bloc voting that 
extends over a period of time is more probative of a claim that a district 
experiences significant polarization than are the results of a single elec­
tion. In a district where elections are shown usually to be polarized, the 
fact that racially polarized voting is not present in one election or a few 
elections does not necessarily negate the conclusion that the district ex­
periences legally significant bloc voting. Furthermore, the success of a 
minority candidate in a particular election does not necessarily prove 
that the district did not experience polarized voting in that election. 
Here, the District Court's approach, which tested data derived from 
three election years in each district in question, and which revealed that 
blacks strongly supported black candidates, while, to the black candi-



THORNBURG v. GINGLES III 

Syllabus 

dates' usual detriment, whites rarely did, satisfactorily addresses each 
facet of the proper standard for legally significant racial bloc voting. 
Pp. 19-27. 

2. The language of § 2 and its legislative history plainly demonstrate 
that proof that some minority candidates have been elected does not 
foreclose a§ 2 claim. Thus, the District Court did not err, as a matter of 
law, in refusing to treat the fact that some black candidates have suc­
ceeded as dispositive of appellees' § 2 claims. Where multimember dis­
tricting generally works to dilute the minority vote, it cannot be de­
fended on the ground that it sporadically and serendipitously benefits 
minority voters. Pp. 41-43. 

3. The clearly-erroneous test of Federal Rule of Civil Procedure 52( a) 
is the appropriate standard for appellate review of ultimate findings of 
vote dilution. As both amended § 2 and its legislative history make 
clear, in evaluating a statutory claim of vote dilution through districting, 
the trial court is to consider the "totality of the circumstances" and to 
determine, based upon a practical evaluation of the past and present re­
alities, whether the political process is equally open to minority voters. 
In this case, the District Court carefully considered the totality of the 
circumstances and found that in each district racially polarized voting; 
the legacy of official discrimination in voting matters, education, hous­
ing, employment, and health services; and the persistence of campaign 
appeals to racial prejudice acted in concert with the multimember dis­
tricting scheme to impair the ability of geographically insular and politi­
cally cohesive groups of black voters to participate equally in the political 
process and to elect candidates of their choice. Pp. 44-47. 

JUSTICE BRENNAN, joined by JUSTICE MARSHALL, JUSTICE BLACK­
MUN, and JUSTICE STEVENS, concluded in Part III-C that for purposes 
of§ 2, the legal concept of racially polarized voting, as it relates to claims 
of vote dilution- that is, when it is used to prove that the minority group 
is politically cohesive and that white voters will usually be able to defeat 
the minority's preferred candidates-refers only to the existence of a 
correlation between the race of voters and the selection of certain candi­
dates. Plaintiffs need not prove causation or intent in order to prove a 
prima facie case of racial bloc voting, and defendants may not rebut that 
case with evidence of causation or intent. Pp. 28-40. 

JUSTICE BRENNAN, joined by JUSTICE WHITE, concluded in Part 
IV-B, that the District Court erred, as a matter of law, in ignoring the 
significance of the sustained success black voters have experienced in 
House District 23. The persistent proportional representation for black 
residents in that district in the last six elections is inconsistent with ap­
pellees' allegation that black voters' ability in that district to elect repre­
sentatives of their choice is not equal to that enjoyed by the white major-



IV THORNBURG v. GINGLES 

Syllabus 

ity. Pp. 43-44. 
JUSTICE O'CONNOR, joined by THE CHIEF JUSTICE, JUSTICE POWELL, 

and JUSTICE REHNQUIST, concluded that: 
1. Insofar as statistical evidence of divergent racial voting patterns is 

admitted solely to establish that the minority group is politically cohe­
sive and to assess its prospects for electoral success, such a showing can­
not be rebutted by evidence that the divergent voting patterns may be 
explained by causes other than race. However, evidence of the reasons 
for divergent voting patterns can in some circumstances be relevant to 
the overall vote dilution inquiry, and there is no rule against consider­
ation of all evidence concerning voting preferences other than statistical 
evidence of racial voting patterns. - Pp. 17-18. 

2. Consistent and sustained success by candidates preferred by minor­
ity voters is presumptively inconsistent with the existence of a § 2 viola­
tion. The District Court erred in assessing the extent of black electoral 
success in House District 39 and Senate District 22, as well as in House 
District 23. Except in House District 23, despite these errors the Dis­
trict Court's ultimate conclusion of vote dilution is not clearly erroneous. 
But in House District 23 appellees failed to establish a violation of § 2. 
Pp. 18-22. 

BRENNAN, J., announced the judgment of the Court and delivered the 
opinion of the Court with respect to Parts I, II, III-A, III-B, IV-A, and 
V, in which WHITE, MARSHALL, BLACKMUN, and STEVENS, JJ., joined, an 
opinion with respect to Part III-C, in which MARSHALL, BLACKMUN, and 
STEVENS, JJ., joined, and an opinion with respect to Part IV-B, in which 
WHITE, J. , joined. WHITE, J., filed a concurring opinion. O'CONNOR, J., 
filed an opinion concurring in the judgment, in which BURGER, C. J., 
POWELL, and REHNQUIST, JJ., joined. STEVENS, J., filed an opinion 
concurring in part and dissenting in part, in which MARSHALL and 
BLACKMUN, JJ., joined. 



NOTICE: This opinion is subject to formal revision before publication in the 
preliminary print of the United States Reports. Readers are requested to 
notify the Reporter of Decisions, Supreme Court ofthe United States, Wash­
ington, D. C. 20543, of any typographical or other formal errors, in order 
that corrections may be made before the preliminary print goes to press. 

SUPREME COURT OF THE UNITED STATES 

No. 83-1968 

LACY H. THORNBURG, ET AL., APPELLANTS v. 
RALPH GINGLES ET AL. 

ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR 
THE EASTERN DISTRICT OF NORTH CAROLINA 

[June 30, 1986] 

JUSTICE BRENNAN announced the judgment of the Court 
and delivered the opinion of the Court with respect to Parts 
I, II, III-A, III-B, IV- A, and V, and an opinion with respect 
to Part III- C, in which JUSTICE MARSHALL, JUSTICE 
BLACKMUN, and JUSTICE STEVENS join, and an opinion with 
respect to Part IV- B, in which JUSTICE WHITE joins. 

This case requires that we construe for the first time § 2 of 
the Voting Rights Act of 1965, as amended June 29, 1982. 
42 U. S. C. § 1973. The specific question to be decided is 
whether the three-judge District Court, convened in the 
Eastern District of North Carolina pursuant to 28 U. S. C. 
§ 2284(a) and 42 U. S. C. § 1973(c), correctly held that the 
use in a legislative redistricting plan of multimember districts 
in five North Carolina legislative districts violated § 2 by 
impairing the opportunity of black voters "to participate in 
the political process and to elect representatives of their 
choice." § 2(b), 96 Stat. 134. 

I 

BACKGROUND 

In April 1982, the North Carolina General Assembly en­
acted a legislative redistricting plan for the State's Senate 
and House of Representatives. Appellees, black citizens 
of North Carolina who are registered to vote, challenged 



2 THORNBURG v. GINGLES 

seven districts, one single-member 1 and six multimember 2 

districts, alleging that the redistricting scheme impaired 
black citizens' ability to elect representatives of their choice 
in violation of the Fourteenth and Fifteenth Amendments to 
the United States Constitution and of § 2 of the Voting 
Rights Act. 3 

After appellees brought suit, but before trial, Congress 
amended § 2. The amendment was largely a response to this 
Court's plurality opinion in Mobile v. Bolden, 446 U. S. 55 
(1980), which had declared that, in order to establish a viola­
tion of either § 2 or of the Fourteenth or Fifteenth Amend­
ments, minority voters must prove that a contested electoral 
mechanism was intentionally adopted or maintained by state 
officials for a discriminatory purpose. Congress substan­
tially revised § 2 to make clear that a violation could be 
proven by showing discriminatory effect alone and to estab­
lish as the relevant legal standard the "results test," applied 
by this Court in White v. Regester, 412 U. S. 755 (1973), and 
by other federal courts before Bolden, supra. S. Rep. 
No. 97-417, p. 28 (1982) (hereinafter S. Rep.). 

Section 2, as amended, 96 Stat. 134, reads as follows: 

1 Appellees challenged Senate District No. 2, which consisted of the 
whole of Northampton, Hertford, Gates, Bertie, and Chowan Counties, 
and parts of Washington, Martin, Halifax, and Edgecombe Counties. 

2 Appellees challenged the following multimember districts: Senate 
No. 22 (Mecklenburg and Cabarrus Counties-4 members), House No. 36 
(Mecklenburg County-8 members), House No. 39 (part of Forsyth 
County-5 members), House No. 23 (Durham County-3 members), 
House No. 21 (Wake County-6 members), and House No. 8 (Wilson, 
Nash and Edgecombe Counties-4 members). 

3 Appellants initiated this action in September 1981, challenging the 
North Carolina General Assembly's July 1981 redistricting. The history 
of this action is recounted in greater detail in the District Court's opinion in 
this case, Gingles v. Edmisten, 590 F. Supp. 345, 350-358 (EDNC 1984). 
It suffices here to note that the General Assembly revised the 1981 plan in 
April 1982 and that the plan at issue in this case is the 1982 plan. 



THORNBURG v. GINGLES 3 

"(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 4(f)(2), as provided in subsection (b). 
"(b) A violation of subsection (a) is established if, based 
on the totality of the 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 opportu­
nity 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 pro­
tected 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." Codified at 42 U. S. C. § 1973. 

The Senate Judiciary Committee majority Report ac­
companying the bill that amended § 2, elaborates on the 
circumstances that might be probative of a § 2 violation, not­
ing the following "typical factors:" 4 

"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; 

'These factors were derived from the analytical framework of White v. 
Regester, 412 U. S. 755 (1973), as refined and developed by the lower 
courts, in particular by the Fifth Circuit in Zimmer v. McKeithen, 485 
F . 2d 1297 (1973), aff'd sub nom. East Carroll Parish School Board v. 
Marshall, 424 U. S. 636 (1976) (per curiam). S. Rep. 28, n. 113. 



4 THORNBURG v. GINGLES 

"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; 
"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 proba­
tive 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. 
"whether the policy underlying the state or political sub­
division's use of such voting qualification, prerequisite to 
voting, or standard, practice or procedure is tenuous." 
S. Rep. 28- 29. 

The District Court applied the "totality of the circum­
stances" test set forth in § 2(b) to appellees' statutory claim, 
and, relying principally on the factors outlined in the Senate 
Report, held that the redistricting scheme violated § 2 
because it resulted in the dilution of black citizens' votes in all 



THORNBURG v. GINGLES 5 

seven disputed districts. In light of this conclusion, the 
court did not reach appellees' constitutio'nal claims. Gingles 
v. Edmisten, 590 F. Supp. 345 (EDNC 1984). 

Preliminarily, the court found that black citizens consti­
tuted a distinct population and registered-voter minority in 
each challenged district. The court noted that at the time 
the multimember districts were created, there wer~ con­
centrations of black citizens within the boundaries of each 
that were sufficiently large and contiguous to constitute 
effective voting majorities in single-member districts lying 
wholly within the boundaries of the multimember districts. 
With respect to the challenged single-member district, Sen­
ate District No. 2, the court also found that there existed a 
concentration of black citizens within its boundaries and 
within those of adjoining Senate District No. 6 that was suffi­
cient in numbers and in contiguity to constitute an effective 
voting majority in a single-member district. The District 
Court then proceeded to find that the following circum­
stances combined with the multimember districting scheme 
to result in the dilution of black citizens' votes. 

First, the court found that North Carolina had officially 
discriminated against its black citizens with respect to their 
exercise of the voting franchise from approximately 1900 to 
1970 by employing at different times a poll tax, a literacy 
test, a prohibition against bullet (single-shot) voting 5 and 

5 Bullet (single-shot) voting has been described as follows: 
"'Consider [a] town of 600 whites and 400 blacks with an at-large election 

to choose four council members. Each voter is able to cast four votes. 
Suppose there are eight white candidates, with the votes of the whites split 
among them approximately equally, and one black candidate, with all the 
blacks voting for him and no one else. The result is that each white candi­
date receives about 300 votes and the black candidate receives 400 votes. 
The black has probably won a seat. This technique is called single-shot 
voting. Single-shot voting enables a minority group to win some at-large 
seats if it concentrates its vote behind a limited number of candidates and if 
the vote of the majority is divided among a number of candidates.'" City 
of Rome v. United States, 446 U. S. 156, 184, n. 19 (1980), quoting U. S. 



6 THORNBURG v. GINGLES 

designated seat plans 6 for multimember districts. The 
court observed that even after the removal of direct barriers 
to black voter registration, such as the poll tax and literacy 
test, black voter registration remained relatively depressed; 
in 1982 only 52.7% of age-qualified blacks statewide were 
registered to vote, whereas 66.7% of whites were registered. 
The District Court found these statewide depressed levels of 
black voter registration to be present in all of the disputed 
districts and to be traceable, at least in part, to the historical 
pattern of statewide official discrimination. 

Second, the court found that historic discrimination in 
education, housing, employment, and health services had 
resulted in a lower socioeconomic status for North Carolina 
blacks as a group than for whites. The court concluded that 
this lower status both gives rise to special group interests 
and hinders blacks' ability to participate effectively in the 
political process and to elect representatives of their choice. 

Third, the court considered other voting procedures that 
may operate to lessen the opportunity of black voters to elect 
candidates of their choice. It noted that North Carolina has 
a majority vote requirement for primary elections and, while 
acknowledging that no black candidate for election to the 
State General Assembly had failed to win solely because of 
this requirement, the court concluded that it nonetheless 
presents a continuing practical impediment to the opportu­
nity of black voting minorities to elect candidates of their 
choice. The court also remarked on the fact that North 
Carolina does not have a subdistrict residency requirement 
for members of the General Assembly elected from 
multimember districts, a requirement which the court found 

Commission on Civil Rights, The Voting Rights Act: Ten Years After, 
pp. 206- 207 (1975). 

6 Designated (or numbered) seat schemes require a candidate for elec­
tion in multimember districts to run for specific seats, and can, under 
certain circumstances, frustrate bullet voting. See, e. g., City of Rome, 
supra, at 185, n. 21. 



THORNBURG v. GINGLES 7 

could offset to some extent the disadvantages minority voters 
often experience in multimember districts. 

Fourth, the court found that white candidates in North 
Carolina have encouraged voting along color lines by appeal­
ing to racial prejudice. It noted that the record is replete 
with specific examples of racial appeals, ranging in style from 
overt and blatant to subtle and furtive, and in date from the 
1890's to the 1984 campaign for a seat in the United States 
Senate. The court .determined that the use of racial appeals 
in political campaigns in North Carolina persists to the 
present day and that its current effect is to lessen to some 
degree the opportunity of black citizens to participate effec­
tively in the political processes and to elect candidates of 
their choice. 

Fifth, the court examined the extent to which blacks have 
been elected to office in North Carolina, both statewide and 
in the challenged districts. It found, among other things, 
that prior to World War II, only one black had been elected 
to public office in this century. While recognizing that "it 
has now become possible for black citizens to be elected to of­
fice at all levels of state government in North Carolina," 590 
F. Supp., at 367, the court found that, in comparison to white 
candidates running for the same office, black candidates are 
at a disadvantage in terms of relative probability of success. 
It also found that the overall rate of black electoral success 
has been minimal in relation to the percentage of blacks in the 
total state population. For example, the court noted, from 
1971 and 1982 there were at any given time only two-to-four 
blacks in the 120-member House of Representatives-that 
is, only 1.6% to 3.3% of House members were black. From 
1975 to 1983 there were at any one time only one or two 
blacks in the 50-member State Senate-that is, only 2% to 4% 
of State Senators were black. By contrast, at the time of the 
District Court's opinion, blacks constituted about 22.4% of 
the total state population. 



8 THORNBURG v. GINGLES 

With respect to the success in this century of black candi­
dates in the contested districts, see also Appendix B, infra, 
the court found that only one black had been elected to House 
District 3&--after this lawsuit began. Similarly, only one 
black had served in the Senate from District 22, from 
1975-1980. Before the 1982 election, a black was elected 
only twice to the House from District 39 (part of Forsyth 
County); in the 1982 contest two blacks were elected. Since 
1973 a black citizen had been elected each 2-year term to the 
House from District 23 (Durham County), but no black had 
been elected to the Senate from Durham County. In House 
District 21 (Wake County), a black had been elected twice to 
the House, and another black served two terms in the State 
Senate. No black had ever been elected to the House or 
Senate from the area covered by House District No. 8 and no 
black person had ever been elected to the Senate from the 
area covered by Senate District No. 2. 

The court did acknowledge the improved success of black 
candidates in the 1982 elections, in which 11 blacks were 
elected to the State House of Representatives, including 5 
blacks from the multimember districts at issue here. How­
ever, the court pointed out that the 1982 election was con­
ducted after the commencement of this litigation. The court 
found the circumstances of the 1982 election sufficiently ab­
errational and the success by black candidates too minimal 
and too recent in relation to the long history of complete 
denial of elective opportunities to support the conclusion that 
black voters' opportunities to elect representatives of their 
choice were not impaired. 

Finally, the court considered the extent to which voting in 
the challenged districts was racially polarized. Based on 
statistical evidence presented by expert witnesses, supple­
mented to some degree by the testimony of lay witnesses, the 
court found that all of the challenged districts exhibit severe 
and persistent racially polarized voting. 



THORNBURG v. GINGLES 9 

Based on these findings, the court declared the contested 
portions of the 1982 redistricting plan violative of § 2 and en­
joined appellants from conducting elections pursuant to those 
portions of the plan. Appellants, the Attorney General of 
North Carolina and others, took a direct appeal to this Court, 
pursuant to 28 U. S. C. § 1253, with respect to five of the 
multimember districts-House Districts 21, 23, 36, and 39, 
and Senate District 22. Appellants argue, first, that the 
District Court utilized a legally incorrect standard in deter­
mining whether the contested districts exhibit racial bloc vot­
ing to an extent that is cognizable under § 2. Second, they 
contend that the court used an incorrect definition of racially 
polarized voting and thus erroneously relied on statistical 
evidence that was not probative of polarized voting. Third, 
they maintain that the · court assigned the wrong weight 
to evidence of some black candidates' electoral success. Fi­
nally, they argue that the trial court erred in concluding that 
these multimember districts result in black citizens having 
less opportunity than their white counterparts to participate 
in the political process and to elect representatives of their 
choice. We noted probable jurisdiction, 471 U. S. -­
(1985), and now affirm with respect to all -of the districts 
except House District 23. With regard to District 23, the 
judgment of the District Court is reversed and remanded for 
further proceedings. 

II 

SECTION 2 AND VOTE DILUTION THROUGH USE 
OF MULTIMEMBER DISTRICTS 

An understanding both of § 2 and of the way in which 
multimember districts can operate to impair blacks' ability to 
elect representatives of their choice is prerequisite to an 
evaluation of appellants' contentions. First, then, we re­
view amended § 2 and its legislative history in some detail. 
Second, we explain the theoretical basis for appellees' claim 
of vote dilution. 



10 THORNBURG v. GINGLES 

A 
SECTION 2 AND ITS LEGISLATIVE HISTORY 

Subsection 2(a) prohibits all States and political subdi­
visions from imposing any voting qualifications or prereq­
uisites to voting, or any standards, practices, or procedures 
which result in the denial or abridgment of the right to vote 
of any citizen who is a member of a protected class of racial 
and language minorities. Subsection 2(b) establishes that§ 2 
has been violated where the "totality of the circumstances" 
reveal that "the political processes leading to nomination or 
election . . . are not equally open to participation by mem­
bers of a [protected class] . .. in that its members have less 
opportunity than other members of the electorate to partici­
pate in the political process and to elect representatives of 
their choice." While explaining that "[t]he 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" in evaluating an alleged violation, § 2(b) 
cautions that "nothing in [§ 2] establishes a right to have 
members of a protected class elected in numbers equal to 
their proportion in the population." 

The Senate Report which accompanied the 1982 amend­
ments elaborates on the nature of § 2 violations and on the 
proof required to establish these violations. 7 First and fore­
most, the Report dispositively rejects the position of the 
plurality in Mobile v. Bolden, 446 U. S. 55 (1980), which 

7 The Solicitor General urges this Court to give little weight to the Sen­
ate Report, arguing that it represents a compromise among conflicting 
"factions," and thus is somehow less authoritative than most Committee 
Reports. Brief for United States as Amicus Curiae 8, n. 12, 24, n. 49. 
We are not persuaded that the legislative history of amended § 2 contains 
anything to lead us to conclude that this Senate Report should be accorded 
little weight. We have repeatedly recognized that the authoritative 
source for legislative intent lies in the committee reports on the bill. See, 
e. g. Garcia v. United States , 469 U. S. --, -- (1984); Zuber v. Allen, 
396 u. s. 168, 186 (1969). 



THORNBURG v. GINGLES 11 

required proof that the contested electoral practice or mecha­
nism was adopted or maintained with the intent to discrimi­
nate against minority voters. 8 See, e. g., S. Rep. 2, 15-16, 
27. The intent test was repudiated for three principal rea­
sons-it is "unnecessarily divisive because it involves charges 
of racism on the part of individual officials or entire communi­
ties," it places an "inordinately difficult" burden of proof on 
plaintiffs, and it "asks the wrong question." I d., at 36. The 
"right" question, as the Report emphasizes repeatedly, is 
whether "as a result of the challenged practice or structure 
plaintiffs do not have an equal opportunity· to participate in 
the political processes and to elect candidates of their 
choice." 9 Id., at 28. See also id., at 2, 27, 29, n. 118, 36. 

In order to answer this question, a court must assess the 
impact of the contested structure or practice on minority 
electoral opportunities "on the basis of objective factors." 
Id., at 27. The Senate Report specifies factors which typi­
cally may be relevant to a § 2 claim: the history of voting­
related discrimination in the State or political subdivision; the 
extent to which voting in the elections of the State or political 

8 The Senate Report states that amended § 2 was _ designed to restore 
the "results test"-the legal standard that governed voting discrimination 
cases prior to our decision in Mobile v. Bolden, 446 U. S. 55 (1980). 
S. Rep. 15-16. The Report notes that in pre-Bolden cases such as 
Regester, 412 U. S. 755 (1973), and Zimmer, 485 F. 2d 1297 (1973), plain­
tiffs could prevail by showing that, under the totality of the circumstances, 
a challenged election law or procedure had the effect of denying a protected 
minority an equal chance to participate in the electoral process. Under 
the "results test," plaintiffs are not required to demonstrate that the 
challenged electoral law or structure was designed or maintained for a 
discriminatory purpose. S. Rep. 16. 

9 The Senate Committee found that "voting practices and procedures 
that have discriminatory results perpetuate the effects of past purposeful 
discrimination." S. Rep. 40 (footnote omitted). As the Senate Report 
notes, the purpose of the Voting Rights Act was" 'not only to correct an 
active history of discrimination, the denying to Negroes of the right to reg­
ister and vote, but also to deal with the accumulation of discrimination.' " 
S. Rep. 5 (quoting 111 Cong. Rec. 8295 (1965) (remarks of Sen. Javits)) .. . 

/ : 

I• 



12 THORNBURG v. GINGLES 

subdivision is racially polarized; the extent to which the State 
or political subdivision has used voting practices or proce­
dures that tend to enhance the opportunity for discrimination 
against the minority group, such as unusually large election 
districts, majority vote requirements, and prohibitions 
against bullet voting; the exclusion of members of the minor­
ity group from candidate slating processes; the extent to 
which minority group members bear the effects of past dis­
crimination in areas such as education, employment, and 
health, which hinder their ability to participate effectively in 
the political process; the use of overt or subtle racial appeals 
in political campaigns; and the extent to which members of 
the minority group have been elected to public office in the 
jurisdiction. S. Rep. 28-29; see also supra, at - -. The 
Report notes also that evidence demonstrating that elected 
officials are unresponsive to the particularized needs of the 
members of the minority group and that the policy underly­
ing the State's or the political subdivision's use of the con­
tested practice or structure is tenuous may have probative 
value. S. Rep. 29. The Report stresses, however, that this 
list of typical factors is neither comprehensive nor exclusive. 
While the enumerated factors will often be pertinent to cer­
tain types of § 2 violations, particularly to vote dilution 
claims, 10 other factors may also be relevant and may be con­
sidered. S. Rep. 29-30. Furthermore, the Senate Commit­
tee observed that "there is no requirement that any particu­
lar number of factors be proved, or that a majority of them 
point one way or the other." Id. , at 29. Rather, the Com­
mittee determined that "the question whether the political 
processes are 'equally open' depends upon a searching practi­
cal evaluation of the 'past and present reality,"' id., at 30 
(footnote omitted), and on a "functional" view of the political 
process. Id., at 30, n. 120. 

10 Section 2 prohibits all forms of voting discrimination, not just vote 
dilution. S. Rep. 30. 



THORNBURG v. GINGLES 13 

Although the Senate Report espouses a flexible, fact­
intensive test for § 2 violations, it limits the circumstances 
under which § 2 violations may be proved in three ways. 
First, electoral devices, such as at-large elections, may not 
be considered per se violative of § 2. Plaintiffs must demon­
strate that, under the totality of the circumstances, the 
devices result in unequal access to the electoral process. 
I d., at 16~ Second, the conjunction of an allegedly dilutive 
electoral mechanism and the lack of proportional representa­
tion alone does not establish a violation. Ibid. Third, the 
results test does not assume the existence of racial bloc vot­
ing; plaintiffs must prove it. I d., at 33. 

B 

VOTE DILUTION THROUGH THE USE OF 
MULTIMEMBER DISTRICTS 

Appellees contend that the legislative decision to employ 
multimember, rather than single-member, districts in the 
contested jurisdictions dilutes their votes by submerging 
them in a white majority/1 thus impairing their ability to 
elect representatives of their choice. 12 

11 Dilution of racial minority group voting strength may be caused by the 
dispersal of blacks into districts in which they constitute an ineffective 
minority of voters or from the concentration of blacks into districts where 
they constitute an excessive majority. Engstrom & Wildgen, Pruning 
Thorns from the Thicket: An Empirical Test of the Existence of Racial Ger­
rymandering, 2 Legis. Stud. Q. 465, 465-466 (1977) (hereinafter Engstrom 
& Wildgen). See also Derfner, Racial Discrimination and the Right to 
Vote, 26 Vand. L. Rev. 523, 553 (1973) (hereinafter Derfner); F. Parker, 
Racial Gerrymandering and Legislative Reapportionment (hereinafter 
Parker), in Minority Vote Dilution 86-100 (Davidson ed. , 1984) (hereinafter 
Minority Vote Dilution). 

12 The claim we address in this opinion is one in which the plaintiffs al­
leged and attempted to prove that their ability to elect the representatives 
of their choice was impaired by the selection of a multimember electoral 
structure. We have no occasion to consider whether § 2 permits, and if 
it does, what standards should pertain to, a claim brought by a minority 
group, which is not sufficiently large and compact to constitute a majority 



14 THORNBURG v. GINGLES 

The essence of a § 2 claim is that a certain electoral law, 
practice or structure interacts with social and historical con­
ditions to cause an inequality in the opportunities enjoyed by 
black and white voters to elect their preferred represent­
atives. This Court has long recognized that multimember 
districts and at-large voting schemes may "'operate to mini­
mize or cancel out the voting strength of racial [minorities in] 
the voting population."' 13 Burns v. Richardson, 384 U. S. 

in a single-member district, alleging that the use of a multimember district 
impairs its ability to influence elections. 

We note also that we have no occasion to consider whether the standards 
we apply to respondents' claim that multimember districts operate to 
dilute the vote of geographically cohesive minority groups, which are large 
enough to constitute majorities in single-member districts and which are 
contained within the boundaries of the challenged multimember districts, 
are fully pertinent to other sorts of vote dilution claims, such as a claim 
alleging that the splitting of a large and geographically cohesive minority 
between two or more multimember or single-member districts resulted in 
the dilution of the minority vote. 

'
3 Commentators are in widespread agreement with this conclusion. 

See, e. g., Berry & Dye, The Discriminatory Effects of At-Large Elec­
tions, 7 Fla. St. U. L. Rev. 85 (1979) (hereinafter Berry & Dye); Blacksher 
& Menefee, From Reynolds v. Sims to City of Mobile v. Bolden, 34 
Hastings L. J. 1 (1982) (hereinafter Blacksher & Menefee); Bonapfel, 
Minority Challenges to At-Large Elections: The Dilution Problem, 10 Ga. 
L. Rev. 353 (1976) (hereinafter Bonapfel); Butler, Constitutional and Stat­
utory Challenges to Election Structures: Dilution and the Value of the 
Right to Vote, 42 La. L. Rev. 851 (1982) (hereinafter Butler); Carpeneti, 
Legislative Apportionment: Multimember Districts and Fair Representa­
tion, 120 Pa. L. Rev. 666 (1972) (hereinafter Carpeneti); Davidson & 
Korbel, At-Large Elections and Minority Group Representation, in Minor­
ity Vote Dilution 65; Derfner; B. Grofman, Alternatives to Single-Member 
Plurality Districts: Legal and Empirical Issues (hereinafter Grofman, 
Alternatives), in Representation and Redistricting Issues 107 (B. Grof­
man, R. Lijphart, H. McKay, & H. Scarrow eds., 1982) (hereinafter 
Representation and Redistricting Issues); Hartman, Racial Vote Dilution 
and Separation of Powers, 50 Geo. Wash. L. Rev. 689 (1982); Jewell, The 
Consequences of Single- and Multimember Districting, in Representation 
and Redistricting Issues 129 (1982) (hereinafter Jewell); Jones, The Impact 
of Local Election Systems on Political Representation, 11 Urb. Aff. Q. 345 



THORNBURG v. GINGLES 15 

73, 88 (1966) (quoting Fortson v. Dorsey, 379 U. S. 433, 439 
(1965)). See also Rogers v. Lodge, 458 U. S. 613, 617 (1982); 
White v. Regester, 412 U. S., at 765; Whitcomb v. Chavis, 
403 U. S. 124, 143 (1971). The theoretical basis for this type 
of impairment is that where minority and majority voters 
consistently prefer different candidates, the majority, by 
virtue of its numerical superiority, will regularly defeat the 
choices of minority voters. 14 See, e. g., Grofman, Alterna­
tives, in Representation and Redistricting Issues 113-114. 
Multimember districts and at-large election schemes, how­
ever, are not per se violative of minority voters' rights. 
S. Rep. 16. Cf. Rogers v. Lodge, supra, at 617; Regester, 
supra, at 765; Whitcomb, supra, at 142. Minority voters 
who contend that the multimember form of districting vio­
lates § 2, must prove that the use of a multimember electoral 
structure operates to minimize or cancel out their ability to 
elect their preferred candidates. See, e. g., S. Rep. 16. 

While many or all of the factors listed in the Senate Report 
may be relevant to a claim of vote dilution through submer­
gence in multimember districts, unless there is a conjunction 
of the following circumstances, the use of multimember dis­
tricts generally will not impede the ability of minority voters 
to elect representatives of their choice. 15 Stated succinctly, 

(1976); Karnig, Black Resources and City Council Representation, 41 J . 
Pol. 134 (1979); Karnig, Black Representation on City Councils, 12 Urb. 
Mf. Q. 223 (1976); Parker 87-88. 

14 Not only does "[v]oting along racial lines" deprive minority voters of 
their preferred representative in these circumstances, it also "allows those 
elected to ignore [minority] interests without fear of political conse­
quences," Rogers v. Lodge, 458 U. S. 613, 623 (1982), leaving the minority 
effectively unrepresented. See, e. g., Grofman, Should Representatives 
Be Typical of Their Constituents?, in Representation and Redistricting 
Issues 97 (hereinafter Grofman, Should Representatives be Typical?); 
Parker 108. 

15 Under a "functional" view of the political process mandated by § 2, 
S. Rep. 30, n. 120, the most important Senate Report factors bearing on 
§ 2 challenges to multimember districts are the "extent to which minority 
group members have been elected to public office in the jurisdiction" and 



16 THORNBURG v. GINGLES 

a bloc voting majority must usually be able to defeat candi­
dates supported by a politically cohesive, geographically in­
sular minority group. Bonapfel 355; Blacksher & Menefee 
34; Butler 903; Carpeneti 696-699; Davidson, Minority Vote 
Dilution: An Overview (hereinafter Davidson), in Minority 
Vote Dilution 4; Grofman, Alternatives 117. Cf. Bolden, 446 
U. S., at 105, n. 3 (MARSHALL, J., dissenting) ("It is obvious 

the "extent to which voting in the elections of the state or political subdi­
vision is racially polarized." Id., 28-29. If present, the other factors, 
such as the lingering effects of past discrimination, the use of appeals to 
racial bias in election campaigns, and the use of electoral devices which 
enhance the dilutive effects of multimember districts when substantial 
white bloc voting exists-for example anti bullet voting laws and majority 
vote requirements, are supportive of, but not essential to, a minority 
voter's claim. 

In recognizing that some Senate Report factors are more important to 
multimember district vote dilution claims than others, the Court effectu­
ates the intent of Congress. It is obvious that unless minority group 
members experience substantial difficulty electing representatives of their 
choice, they cannot prove that a challenged electoral mechanism impairs 
their ability "to elect." § 2(b). And, where the contested electoral 
structure is a multimember district, commentators and courts agree that in 
the absence of significant white bloc voting it cannot be said that the ability 
of minority voters to elect their chosen representatives is inferior to that of 
white voters. See, e. g., McMillan v. Escambia County, 748 F. 2d 1037, 
1043 (CA5 1984); United States v. Marengo County Comm'n, 731 F. 2d 
1546, 1566 (CAll 1984) appeal dismissed, cert. denied, 469 U. S. -­
(1984); Nevett v. Sides, 571 F. 2d 209, 223 (CA5 1978), cert. denied, 446 
U. S. 951 (1980); Johnson v. Halifax County, 594 F. Supp. 161, 170 
(EDNC 1984); Blacksher & Menefee; Engstrom & Wildgen 469; Parker 
107. Consequently, if difficulty in electing and white bloc voting are not 
proven, minority voters have not established that the multimember struc­
ture interferes with their ability to elect their preferred candidates. Mi­
nority voters may be able to prove that they still suffer social and economic 
effects of past discrimination, that appeals to racial bias are employed in 
election campaigns, and that a majority vote is required to win a seat, but 
they have not demonstrated a substantial inability to elect caused by the 
use of a multimember district. By recognizing the primacy of the history 
and extent of minority electoral success and of racial bloc voting, the Court 
simply requires that § 2 plaintiffs prove their claim before they may be 
awarded relief. 



THORNBURG v. GINGLES 17 

that the greater the degree to which the electoral minority is 
homogeneous and insular and the greater the degree that 
bloc voting occurs along majority-minority lines, the greater 
will be the extent to which the minority's voting power is 
diluted by multimember districting"). These circumstances 
are necessary preconditions for multimember districts to 
operate to impair minority voters' ability to elect represent­
atives of their choice for the following reasons. First, the 
minority group must be able to demonstrate that it is suffi­
ciently large and geographically compact to constitute a ma­
jority in a single-member district. 16 If it is not, as would be 
the case in a substantially integrated district, the multi­
member form of the district cannot be responsible for minor­
ity voters' inability to elect its candidates. 17 Cf. Rogers, 

16 In this case respondents allege that within each contested multi­
member district there exists a minority group that is sufficiently large and 
compact to constitute a single-member district. In a different kind of 
case, for example a gerrymander case, plaintiffs might allege that the 
minority group that is sufficiently large and compact to constitute a 
single-member district has been split between two or more multimember 
or single-member districts, with the effect of diluting the potential 
strength of the minority vote. 

17 The reason that a minority group making such a challenge must show, 
as a threshold matter, that it is sufficiently large and geographically 
compact to constitute a majority in a single-member district is this: Unless 
minority voters possess the potential to elect representatives in the 
absence of the challenged structure or practice, they cannot claim to have 
been injured by that structure or practice. The single-member district is 
generally the appropriate standard against which to measure minority 
group potential to elect because it is the smallest political unit from which 
representatives are elected. Thus, if the minority group is spread evenly 
throughout a multimember district, or if, although geographically compact, 
the minority group is so small in relation to the surrounding white popula­
tion that it could not constitute a majority in a single-member district, 
these minority voters cannot maintain that they would have been able to 
elect representatives of their choice in the absence of the multimember 
electoral structure. As two commentators have explained, 
"To demonstrate [that minority voters are injured by at-large elections], 
the minority voters must be sufficiently concentrated and politically 



18 THORNBURG v. GINGLES 

supra, at 616. See also, Blacksher & Menefee 51-56, 58; 
Bonapfel 355; Carpeneti 696; Davidson 4; Jewell 130. Sec­
ond, the minority group must be able to show that it is politi­
cally cohesive. If the minority group is not politically co­
hesive, it cannot be said that the selection of a multimember 
electoral structure thwarts distinctive minority group in­
terests. Blacksher & Menefee 51-55, 58-60, and n. 344; 
Carpeneti 696- 697; Davidson 4. 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 circum­
stances, such as the minority candidate running unopposed, 
see, infra, at usually to defeat the minority's preferred 
candidate. See, e. g., Blacksher & Menefee 51, 53, 56-57, 
60. Cf. Rogers, supra, at 616-617; Whitcomb, supra, at 
158-159; McMillan v. Escambia County, Fla., 748 F. 2d 
1037, 1043 (CA5 1984). In establishing this last circum­
stance, the minority group demonstrates that submergence 
in a white multimember district impedes its ability to elect its 
chosen representatives. 

Finally, we observe that the usual predictability of the 
majority's success distinguishes structural dilution from the 
mere loss of an occasional election. Cf. Davis v. Bandemer, 
-- U. S. - - , -- (1986) (opinion of WHITE, J.); Bolden, 
supra, at 111, n. 7 (MARSHALL, J., dissenting); Whitcomb, 
supra, at 153. See also Blacksher & Menefee 57, n. 333; 
Note, Geometry and Geography: Racial Gerrymandering and 
the Voting Rights Act, 94 Yale L. J. 189, 200, n. 66 (1984) 
(hereinafter Note, Geometry and Geography). 

cohesive that a putative districting plan would result in districts in which 
members of a racial minority would constitute a majority of the voters, 
whose clear electoral choices are in fact defeated by at-large voting. If 
minority voters' residences are substantially integrated throughout the ju­
risdiction, the at-large district cannot be blamed for the defeat of minority­
supported candidates . . . . [This standard] thus would only protect racial 
minority votes from diminution proximately caused by the districting plan; 
it would not assure racial minorities proportional representation." 
Blacksher & Menefee 55-56 (footnotes omitted) (emphasis added). 



( 

THORNBURG v. GINGLES 19 

III 
RACIALLY POLARIZED VOTING 

Having stated the general legal principles relevant to 
claims that § 2 has been violated through the use of 
multimember districts, we turn to · the arguments of appel­
lants and amicus curiae the United States addressing 
racially polarized voting. 18 First we describe the District 
Court's treatment of racially polarized voting. Next, we 
consider appellants' claim that the District Court used an 
incorrect legal standard to determine whether racial bloc 
voting in the contested districts was sufficiently severe to be 
cognizable as an element of a § 2 claim. Finally, we consider 
appellants' contention that the trial court employed an incor­
rect definition of racially polarized voting and thus errone­
ously relied on statistical evidence that was not probative qf 
racial bloc voting. 

A 
THE DISTRICT COURT'S TREATMENT OF RACIALLY 

POLARIZED VOTING 

The investigation conducted by the District Court into the 
question of racial bloc voting credited some testimony of 
lay witnesses, but relied principally on statistical evidence 
presented by appellees' expert witnesses, in particular that 
offered by Dr. Bernard Grofman. Dr. Grofman collected 
and evaluated data from 53 General Assembly primary and 
general elections involving black candidacies. These elec­
tions were held over a period of three different election years 
in the six originally challenged multimember districts. 19 Dr. 
Grofman subjected the data to two complementary methods 
of analysis- extreme case analysis and bivariate ecological 

'"The terms "racially polarized voting" and "racial bloc voting" are used 
interchangeably throughout this opinion. 

19 The 1982 reapportionment plan left essentially undisturbed the 1971 
plan for five of the original six contested multimember districts. House 
District 39 alone was slightly modified. Brief for Appellees 8. 



20 THORNBURG v. GINGLES 

regression analysis 20-in order to determine whether blacks 
and whites in these districts differed in their voting behavior. 
These analytic techniques yielded data concerning the voting 
patterns of the two races, including estimates of the percent­
ages of members of each race who voted for black candidates. 

The court's initial consideration of these data took the form 
of a three-part inquiry: did the data reveal any correlation be­
tween the race of the voter and the selection of certain candi­
dates; was the revealed correlation statistically significant; 
and was the difference in black and white voting patterns 
"substantively significant?" The District Court found that 
blacks and whites generally preferred different candidates 
and, on that basis, found voting in the districts to be racially 
correlated. 21 The court accepted Dr. Grofman's expert opin­
ion that the correlation between the race of the voter and the 
voter's choice of certain candidates was statistically signifi­
cant. 22 Finally, adopting Dr. Grofman's terminology, see Tr. 

20 The District Court found both methods standard in the literature for 
the analysis of racially polarized voting. 590 F. Supp., at 367- 378, nn. 28, 
and 32. See also Engstrom & McDonald, Quantitative Evidence in Vote 
Dilution Litigation: Political Participation and Polarized Voting, 17 Urban 
Lawyer 369 (Summer 1985); Grofman, Migalski & Noviello, The "Totality 
of Circumstances Test" in Section 2 of the 1982 Extension of the Voting 
Rights Act: A Social Science Perspective, 7 Law & Policy 199 (Apr. 1985) 
(hereinafter Grofman, Migalski, & Noviello). 

21 The court used the term "racial polarization" to describe this correla­
tion. It adopted Dr. Grofman's definition-"racial polarization" exists 
where there is "a consistent relationship between [the] race of the voter 
and the way in which the voter votes," Tr. 160, or to put it differently, 
where "black voters and white voters vote differently." ld., at 203. We, 
too, adopt this definition of "racial bloc" or "racially polarized" voting. 
See, irifra, at --. 

22 The court found that the data reflected positive relationships and that 
the correlations did not happen by chance. 590 F. Supp. 368, and n. 30. 
See also D. Barnes & J . Conley, Statistical Evidence in Litigation 32-34 
(1986); Fisher, Multiple Regression in Legal Proceedings, 80 Colum. L. 
Rev. 702, 716-720 (1980); Grofman, Migalski, & Noviello 206. 



THORNBURG v. GINGLES 21 

195, the court found that in all but two of the 53 elections 23 

the degree of racial bloc voting was "so marked as to be 
substantively significant, in the sense that the results of the 
individual election would have been different depending upon 
whether it had been held among only the white voters or only 
the black voters." 590 F. Supp., at 368. 

The court also reported its findings, both in tabulated nu­
merical form and in written form, that a high percentage of 
black voters regularly supported black candidates and that 
most white voters were extremely reluctant to vote for black 
candidates. The court then considered the relevance to the 
existence of legally significant white bloc voting of the fact 
that black candidates have won some elections. It deter­
mined that in most instances, special circumstances, such as 
incumbency and lack of opposition, rather than an a diminu­
tion in usually severe white bloc voting, accounted for these 
candidates' success. The court also suggested that black 
voters' reliance on bullet voting was a significant factor in 
their successful efforts to elect candidates of their choice. 
Based on all of the evidence before it, the trial court con­
cluded that each of the districts experiences racially polarized 
voting "in a persistent and severe degree." 590 F. Supp., 
at 367. 

B 

THE DEGREE OF BLOC VOTING THAT IS LEGALLY 
SIGNIFICANT UNDER § 2 

1 

Appellants' Arguments 

North Carolina and the United States argue that the test 
used by the District Court to determine whether voting 
patterns in the disputed districts are racially polarized to an 
extent cognizable under § 2 will lead to results that are incon­
sistent with congressional intent. North Carolina maintains 

23 The two exceptions were the 1982 State House elections in Districts 21 
and 23. 590 F. Supp., at 368, n. 31. 



22 THORNBURG v. GINGLES 

that the court considered legally significant racially polarized 
voting to occur whenever "less than 50% of the white voters 
cast a ballot for the black candidate." Brief for Appellants 
36. Appellants also argue that racially polarized voting is 
legally significant only when it always results in the defeat of 
black candidates. ld., at 39-40. 

The United States, on the other hand, isolates a single line 
in the court's opinion and identifies it as the court's complete 
test. According to the Solicitor General, the District Court 
adopted a standard under which legally significant racial bloc 
voting is deemed to exist whenever "'the results of the 
individual election would have been different depending upon 
whether it had been held among only the white voters or only 
the black voters in the election.' " Brief for United States as 
Amicus Curiae 29 (quoting Gingles, 590 F. Supp., at 368). 
We read the District Court opinion differently. 

2 

The Standard for Legally Significant Racial Bloc Voting 

The Senate Report states that the "extent to which voting 
in the elections of the state or political subdivision is racially 
polarized," S. Rep. 29, is relevant to a vote dilution claim. 
Further, courts and commentators agree that racial bloc 
voting is a key element of a vote dilution claim. See, e. g., 
Escambia County, 748 F. 2d, at 1043; United States v. 
Marengo County Comm'n, 731 F. 2d 1546, 1566 (CA111984); 
Nevett v. Sides, 571 F. 2d 209, 223 (CA5 1978), cert. denied, 
446 U. S. 951 (1980); Johnson v. Halifax County, 594 
F. Supp. 161, 170 (EDNC 1984); Blacksher & Menefee; 
Engstrom & Wildgen, Pruning Thorns from the Thicket: An 
Empirical Test for the Existence of Racial Gerrymandering, 
2 Legis. Stud. Q. 465, 469 (1977) (hereafter Enstrom & 
Wildgen); Parker 107; Note, Geometry and Geography 199. 
Because, as we explain below, the extent of bloc voting 
necessary to demonstrate that a minority's ability to elect its 
preferred representatives is impaired varies according to 



THORNBURG v. GINGLES 23 

several factual circumstances, the degree of bloc voting 
which constitutes the threshold of legal significance will vary 
from district to district. Nonetheless, it is possible to state 
some general principles and we proceed to do so. 

The purpose of inquiring into the existence of racially po­
larized voting is twofold: to ascertain whether minority group 
members constitute a politically cohesive unit and to deter­
mine whether whites vote sufficiently as a bloc usually to de­
feat the minority's preferred candidates. See, supra, at 

Thus, the question whether a given district experi­
ences legally significant racially polarized voting requires dis­
crete inquiries into minority and white voting practices. A 
showing that a significant number of minority group mem­
bers usually vote for the same candidates is one way of prov­
ing the political cohesiveness necessary to a vote dilution 
claim, Blacksher & Menefee 59-60, and n. 344, and, conse­
quently, establishes minority bloc voting ·within the context 
of § 2. And, 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 voting. Id., at 60. The amount of white bloc voting 
that can generally "minimize or cancel," S. Rep. 28; Regester, 
412 U. S., at 765, black voters' ability to elect represent­
atives of their choice, however, will vary from district to dis­
trict according to a number of factors, including the nature of 
the allegedly dilutive electoral mechanism; the presence or 
absence of other potentially dilutive electoral devices, such as 
majority vote requirements, designated posts, and prohi­
bitions against bullet voting; the percentage of registered 
voters in the district who are members of the minority group; 
the size of the district; and, in multimember districts, the 
number of seats open and the number of candidates in the 
field. 24 See, e. g., Butler 874-876; Davidson 5; Jones, The 
Impact of Local Election Systems on Black Political Repre-

24 This list of factors is illustrative, not comprehensive. 



24 THORNBURG v. GINGLES 

sentation, 11 Urb. Aff. Q. 345 (1976); U. S. Commission on 
Civil Rights, The Voting Rights Act: Unfulfilled Goals 38-41 
(1981). 

Because loss of political power through vote dilution is dis­
tinct from the mere inability to win a particular election, 
Whitcomb, 403 U. S., at 153, a pattern of racial bloc voting 
that extends over a period of time is more probative of a 
claim that a district experiences legally significant polariza­
tion than are the results of a single election. 25 Blacksher 
& Menefee 61; Note, Geometry and Geography 200, n. 66 
("Racial polarization should be seen as an attribute not of a 
single election, but rather of a polity viewed over time. The 
concern is necessarily temporal and the analysis historical 
because the evil to be avoided is the subordination of minor­
ity groups in American politics, not the defeat of individuals 
in particular electoral contests"). Also for this reason, in a 
district where elections are shown usually to be polarized, 
the fact that racially polarized voting is not present in one or 
a few individual elections does not necessarily negate the con­
clusion that the district experiences legally significant bloc 
voting. Furthermore, the success of a minority candidate in 
a particular election does not necessarily prove that the 
district did not experience polarized voting in that election; 
special circumstances, such as the absence of an opponent, 
incumbency, or the utilization of bullet voting, may explain 
minority electoral success in a polarized contest. 26 

25 The number of elections that must be studied in order to determine 
whether voting is polarized will vary according to pertinent circumstances. 
One important circumstance is the number of elections in which the minor­
ity group has sponsored candidates. Where a minority group has never 
been able to sponsor a candidate, courts must rely on other factors that 
tend to prove unequal access to the electoral process. Similarly, where a 
minority group has begun to sponsor candidates just recently, the fact that 
statistics from only one or a few elections are available for examination 
does not foreclose a vote dilution claim. 

25 This list of special circumstances is illustrative, not exclusive. 



THORNBURG v. GINGLES 25 

As must be apparent, the degree of racial bloc voting that 
is cognizable as an element of a § 2 vote dilution claim will 
vary according to a variety of factual circumstances. Conse­
quently, there is no simple doctrinal test for the existence of 
legally significant racial bloc voting. However, the forego­
ing general principles should provide courts with substantial 
guidance in determining whether evidence that black and 
white voters generally prefer different candidates rises to the 
level of legal significance under § 2. 

3 
Standard Utilized by the District Court 

The District Court clearly did not employ the simplistic 
standard identified by North Carolina- legally significant 
bloc voting occurs whenever less than 50% of the white vot­
ers cast a ballot for the black candidate. Brief for Appel­
lants 36. And, although the District Court did utilize the 
measure of "substantive significance" that the United States 
ascribes to it-"the results of the individual election would 
have been different depending on whether it had been held 
among only the white voters or only the black voters,"' Brief 
for United States as Amicus Curiae 29 (quoting Gingles, 590 
F. Supp., at 368)-the court did not reach its ultimate conclu­
sion that the degree of racial bloc voting present in each dis­
trict is legally significant through mechanical reliance on this 
standard. 27 While the court did not phrase the standard for 
legally significant racial bloc voting exactly as we do, a fair 
reading of the court's opinion reveals that the court's analysis 
conforms to our view of the proper legal standard. 

27 The trial court did not actually employ the term "legally significant." 
At times it seems to have used "substantive significance" as Dr. Grofman 
did, to describe polarization severe enough to result in the selection of 
different candidates in racially separate electorates. At other times, 
however, the court used the term "substantively significant" to refer to its 
ultimate determination that racially polarized voting in these districts is 
sufficiently severe to be relevant to a § 2 claim. 



26 THORNBURG v. GINGLES 

The District Court's findings concerning black support for 
black candidates in the five multimember districts at issue 
here clearly establish the political cohesiveness of black vot­
ers. As is apparent from the District Court's tabulated find­
ings, reproduced in Appendix A, infra, black voters' support 
for black candidates was overwhelming in almost every elec­
tion. In all but 5 of 16 primary elections, black support for 
black candidates ranged between 71% and 92%; and in the 
general elections, black support for black Democrat candi­
dates ranged between 87% and 96%. 

In sharp contrast to its findings of strong black support for 
black candidates, the District Court found that a substantial 
majority of white voters would rarely, if ever, vote for a 
black candidate. In the primary elections, white support for 
black candidates ranged between 8% and 50%, and in the 
general elections it ranged between 28% and 49%. See 
Appendix A, infra. The court also determined that, on 
average, 81.7% of white voters did not vote for any black can­
didate in the primary elections. In the general elections, 
white voters almost always ranked black candidates either 
last or next to last in the multicandidate field, except in 
heavily Democratic areas where white voters consistently 
ranked black candidates last among the Democrats, if not last 
or next to last among all candidates. The court further ob­
served that approximately two-thirds of white voters did not 
vote for black candidates in general elections, even after the 
candidate had won the Democratic primary and the choice 
was to vote for a Republican or for no one. 28 

28 In stating that 81.7% of white voters did not vote for any black candi­
dates in the primary election and that two-thirds of white voters did not 
vote for black candidates in general elections, the District Court aggre­
gated data from all six challenged multimember districts, apparently for 
ease of reporting. The inquiry into the existence of vote dilution caused 
by submergence in a multimember district is district-specific. When con­
sidering several separate vote dilution claims in a single case, courts must 
not rely on data aggregated from all the challenged districts in concluding 
that racially polarized voting exists in each district. In the instant case, 



THORNBURG v. GINGLES 27 

While the District Court did not state expressly that the 
percentage of whites who refused to vote for black candidates 
in the contested districts would, in the usual course of events, 
result in the defeat of the minority's candidates, that conclu­
sion is apparent both from the court's factual findings and 
from the rest of its analysis. First, with the exception of 
House District 23, see infra, at--, the trial court's findings 
clearly show that black voters have enjoyed only minimal and 
sporadic success in electing representatives of their choice. 
See Appendix B, infra. Second, where black candidates 
won election, the court closely examined the circumstances of 
those elections before concluding that the success of these 
blacks did not negate other evidence, derived from all of the 
elections studied in each district, that legally significant ra­
cially polarized voting exists in each district. For example, 
the court took account of the benefits incumbency and run­
ning essentially unopposed conferred on some of the success­
ful black candidates, 29 as well as of the very different order of 

however, it is clear from the trial court's tabulated findings and from the 
exhibits that were before it, 1 App., Exs. 2- 10, that the court relied on 
data that was specific to each individual district in concluding that each dis­
trict experienced legally significant racially polarized voting. 

29 For example, the court found that incumbency aided a successful black 
candidate in the 1978 primary in Senate District 22. The court also noted 
that in House District 23, a black candidate who gained election in 1978, 
1980, and 1982, ran uncontested in the 1978 general election and in both the 
primary and general elections in 1980. In 1982 there was no Republican 
opposition, a fact the trial court interpreted to mean that the general 
election was for all practical purposes unopposed. Moreover, in the 1982 
primary, there were only two white candidates for three seats, so that one 
black candidate had to succeed. Even under this condition, the court 
remarked, 63% of white voters still refused to vote for the black incum­
bent-who was the choice of 90% of the blacks. In House District 21, 
where a black won election to the six-member delegation in 1980 and 1982, 
the court found that in the relevant primaries approximately 60% to 70% of 
white voters did not vote for the black candidate, whereas approximately 
80% of blacks did. The court additionally observed that although winning 
the Democratic primary in this district is historically tantamount to 



28 THORNBURG v. GINGLES 

preference blacks and whites assigned black candidates, 30 in 
reaching its conclusion that legally significant racial polariza­
tion exists in each district. 

We conclude that the District Court's approach, which 
tested data derived from three election years in each district, 
and which revealed that blacks strongly supported black can­
didates, while, to the black candidates' usual detriment, 
whites rarely did, satisfactorily addresses each facet of the 
proper legal standard. 

c 
EVIDENCE OF RACIALLY POLARIZED VOTING 

1 
Appellants' Argument 

North Carolina and the United States also contest the evi­
dence upon which the District Court relied in finding that 
voting patterns in the challenged districts were racially po­
larized. They argue that the term "racially polarized voting" 
must, as a matter of law, refer to voting patterns for which 
the principal cause is race. They contend that the District 
Court utilized a legally incorrect definition of racially polar­
ized voting by relying on bivariate statistical analyses which 
merely demonstrated a correlation between the race of the 
voter and the level of voter support for certain candidates, 
but which did not prove that race was the primary determi­
nant of voters' choices. According to appellants and the 
United States, only multiple regression analysis, which can 
take account of other variables which might also explain vot­
ers' choices, such as "party affiliation, age, religion, income[,] 

election, 55% of whites declined to vote for the Democratic black candidate 
in the general election. 

30 The court noted that in the 1982 primary held in House District 36, out 
of a field of eight, the successful black candidate was ranked first by black 
voters, but seventh by whites. Similarly, the court found that the two 
blacks who won seats in the five-member delegation from House District 39 
were ranked first and second by black voters, but seventh and eighth by 
white voters. 



THORNBURG v. GINGLES 29 

incumbency, education, campaign expenditures," Brief for 
Appellants 42, "media use measured by cost, . . . name, iden­
tification, or distance that a candidate lived from a particular 
precinct," Brief for United States as Amicus Curiae 30, 
n. 57, can prove that race was the primary determinant of 
voter behavior. 31 

Whether appellants and the United States believe that it is 
the voter's race or the candidate's race that must be the pri­
mary determinant of the voter's choice is unclear; indeed, 
their catalogs of relevant variables suggest both. 32 Age, reli­
gion, income, and education seem most relevant to the voter; 
incumbency, campaign expenditures, name identification, 
and media use are pertinent to the candidate; and party affili­
ation could refer both to the voter and the candidate. In 
either case, we disagree: For purposes of § 2, the legal con­
cept of racially polarized voting incorporates neither causa­
tion nor intent. It means simply that the race of voters cor­
relates with the selection of a certain candidate or candidates; 
that is, it refers to the situation where different races (or 
minority language groups) vote in blocs for different candi­
dates. Grofman, Migalski & N oviello 203. As we demon­
strate below, appellants' theory of racially polarized voting 
would thwart the goals Congress sought to achieve when it 
amended § 2 and would prevent courts from performing the 
"functional" analysis of the political process, S. Rep. 30, 
n. 119, and the "searching practical evaluation of the 'past 

31 Appellants argue that plaintiffs must establish that race was the 
primary determinant of voter behavior as part of their prima facie showing 
of polarized voting; the Solicitor General suggests that plaintiffs make out a 
prima facie case merely by showing a correlation between race and the 
selection of certain candidates, but that defendants should be able to rebut 
by showing that factors other than race were the principal causes of voters' 
choices. We reject both arguments. 

32 The Fifth Circuit cases on which North Carolina and the Solicitor 
General rely for their position are equally ambiguous. See Lee County 
Branch of NAACP v. City of Opelika, 748 F. 2d 1473, 1482 (1984); Jones v. 
City of Lubbock, 730 F. 2d 233, 234 (1984) (Higginbotham, J., concurring). 



30 THORNBURG v. GINGLES 

and present reality,"' id., at 30 (footnote omitted), mandated 
by the Senate Report. 

2 

Causation Irrelevant to Section 2 Inquiry 
The first reason we reject appellants' argument that ra­

cially polarized voting refers to voting patterns that are in 
some way caused by race, rather than to voting patterns that 
are merely correlated with the race of the voter, is that the 
reasons black and white voters vote differently have no rele­
vance to the central inquiry of § 2. By contrast, the correla­
tion between race of voter and the selection of certain candi­
dates is crucial to that inquiry. 

Both § 2 itself and the Senate Report make clear that the 
critical question in a § 2 claim is whether the use of a con­
tested electoral practice or structure results in members of a 
protected group having less opportunity than other members 
of the electorate to participate in the political process and to 
elect representatives of their choice. See, e. g., S. Rep. 2, 
27, 28, 29, n. 118, 36. As we explained, supra, at --, 
multimember districts may impair the ability of blacks to 
elect representatives of their choice where blacks vote suffi­
ciently as a bloc as to be able to elect their preferred candi­
dates in a black majority, single-member district and where a 
white majority votes sufficiently as a bloc usually to defeat 
the candidates chosen by blacks. It is the difference be­
tween the choices made by blacks and whites-not the rea­
sons for that difference-that results in blacks having less 
opportunity than whites to elect their preferred represent­
atives. Consequently, we conclude that under the "res'..!lts 
test" of § 2, only the correlation between race of voter and se­
lection of. certain candidates, not the causes of the correla­
tion, matters. 

The irrelevance to a § 2 inquiry of the reasons why black 
and white voters vote differently supports, by itself, our 
rejection of appellants' theory of racially polarized voting. 
However, their theory contains other equally serious flaws 



THORNBURG v. GINGLES 31 

that merit further attention. As we demonstrate below, the 
addition of irrelevant variables distorts the equation and 
yields results that are indisputably incorrect under § 2 and 
the Senate Report. 

3 

Race of Voter as Primary Determinant of Voter Behavior 
Appellants and the United States contend that the legal 

concept of "racially polarized voting" refers not to voting 
patterns that are merely correlated with the voter's race, but 
to voting patterns that are determined primarily by the 
voter's race, rather than by the voter's other socioeconomic 
characteristics. 

The first problem with this argument is that it ignores the 
fact that members of geographically insular racial and ethnic 
groups frequently share socioeconomic characteristics, such 
as income level, employment status, amount of educatio!l, 
housing and other living conditions, religion, language, and 
so forth. See, e. g., Butler 902 (Minority group "members' 
shared concerns, including political ones, are ... a function 
of group status, and as such are largely involuntary .... As a 
group blacks are concerned, for example, with police brutal­
ity, substandard housing, unemployment, etc., because these 
problems fall disproportionately upon the group"); S. Verba 
& N. Nie, Participation in America 151-152 (1972) (hereinaf­
ter Verba & Nie) ("Socioeconomic status ... is closely re­
lated to race. Blacks in American society are likely to be in 
lower-status jobs than whites, to have less education, and to 
have lower incomes.") Where such characteristics are 
shared, race or ethnic group not only denotes color or place of 
origin, it also functions as a shorthand notation for common 
social and economic characteristics. Appellants' definition of 
racially polarized voting is even more pernicious where 
shared characteristics are causally related to race or ethnic­
ity. The opportunity to achieve high employment status and 
income, for example, is often influenced by the presence or 
absence of racial or ethnic discrimination. A definition of ra-



32 THORNBURG v. GINGLES 

cially polarized voting which holds that black bloc voting does 
not exist when black voters' choice of certain candidates is 
most strongly influenced by the fact that the voters have low 
incomes and menial jobs -when the reason most of those 
voters have menial jobs and low incomes is attributable to 
past or present racial discrimination-runs counter to the 
Senate Report's instruction to conduct a searching and prac­
tical evaluation of past and present reality, S. Rep. 30, and 
interferes with the purpose of the Voting Rights Act to elimi­
nate the negative effects of past discrimination on the elec­
toral opportunities of minorities. I d., at 5, 40. 

Furthermore, under appellants' theory of racially polarized 
voting, even uncontrovertible evidence that candidates 
strongly preferred by black voters are always defeated by a 
bloc voting white majority would be dismissed for failure to 
prove racial polarization whenever the black and white popu­
lations could be described in terms of other socioeconomic 
characteristics. 

To illustrate, assume a racially-mixed, urban multimember 
district in which blacks and whites possess the same socioeco­
nomic characteristics that the record in this case attributes to 
blacks and whites in Halifax County, a part of Senate District 
2. The annual mean income for blacks in this district is 
$10,465, and 47.8% of the black community lives in poverty. 
More than half-51.5o/cr-of black adults over the age of 25 
have only an eighth grade education or less. Just over half 
of black citizens reside in their own homes; 48.9% live in 
rental units. And, almost a third of all black households are 
without a car. In contrast, only 12.6% of the whites in the 
district live below the poverty line. Whites enjoy a mean in­
come of $19,042. White residents are better educated than 
blacks-only 25.6% of whites over the age of 25 have only an 
eighth grade education or less. Furthermore, only 26.2% of 
whites live in rental units, and only 10.2% live in households 
with no vehicle available. 1 App., Ex-44. As is the case in 



THORNBURG v. GINGLES 33 

Senate District 2, blacks in this hypothetical urban district 
have never been able to elect a representative of their choice. 

According to appellants' theory of racially polarized voting, 
proof that black and white voters in this hypothetical district 
regularly choose different candidates and that the blacks' 
preferred candidates regularly lose could be rejected as not 
probative of racial bloc voting. The basis for the rejection 
would be that blacks chose a certain candidate, not princi­
pally because of their race, but principally because this candi­
date best represented the interests of residents who, because 
of their low incomes, are particularly interested in govern­
ment subsidized health and welfare services; who are gener­
ally poorly educated, and thus share an interest in job train­
ing programs; who are, to a greater extent than the white 
community, concerned with rent control issues; and who 
favor major public transportation expenditures. Similarly, 
whites would be found to have voted for a different candi­
date, not principally because of their race, but primarily 
because that candidate best represented the interests of 
residents who, due to their education and income levels, and 
to their property and vehicle ownership, favor gentrification, 
low residential property taxes, and extensive expenditures 
for street and highway improvements. 

Congress could not have intended that courts employ this 
definition of racial bloc voting. First, this definition leads to 
results that are inconsistent with the effects test adopted by 
Congress when it amended § 2 and with the Senate Report's 
admonition that courts take a "functional" view of the politi­
cal process, S. Rep. 30, n. 119, and conduct a searching and 
practical evaluation of reality. I d., at 30. A test for racially 
polarized voting that denies the fact that race and socioeco­
nomic characteristics are often closely correlated permits nei­
ther a practical evaluation of reality nor a functional analysis 
of vote dilution. And, contrary to Congress' intent in adopt­
ing the "results test," appellants' proposed definition could 
result in the inability of minority voters to establish a critical 



34 THORNBURG v. GINGLES 

element of a vote dilution claim, even though both races 
engage in "monolithic" bloc voting, id., at 33, and generations 
of black voters have been unable to elect a representative of 
their choice. 

Second, appellants' interpretation of "racially polarized 
voting" creates an irreconcilable tension between their pro­
posed treatment of socioeconomic characteristics in the bloc 
voting context and the Senate Report's statement that "the 
extent to which members of the minority group . . . bear the 
effects of discrimination in such areas as education, employ­
ment and health" may be relevant to a § 2 claim. I d., at 29. 
We can find no support in either logic or the legislative his­
tory for the anomalous conclusion to which appellants' posi­
tion leads-that Congress intended, on the one hand, that 
proof that a minority group is predominately poor, unedu­
cated, and unhealthy should be considered a factor tending to 
prove a § 2 violation; but that Congress intended, on the 
other hand, that proof that the same socioeconomic charac­
teristics greatly influence black voters' choice of candidates 
should destroy these voters' ability to establish one of the 
most important elements of a vote dilution claim. 

4 

Race of Candidate as Primary Determinant of 
Voter Behavior 

North Carolina's and the United State's suggestion that 
racially polarized voting means that voters select or reject 
candidates principally on the basis of the candidate's race is 
also misplaced. 

First, both the language of§ 2 and a functional understand­
ing of the phenomenon of vote dilution mandate the conclu­
sion that the race of the candidate per se is irrelevant to racial 
bloc voting analysis. Section 2(b) states that a violation is 
established if it can be shown that members of a protected 
minority group "have less opportunity than other members of 
the electorate to . . . elect representatives of their choice." 



THORNBURG v. GINGLES 35 

(Emphasis added). Because both minority and majority vot­
ers often select members of their own race as their preferred 
representatives, it will frequently be the case that a black 
candidate is the choice of blacks, while a white candidate is 
the choice of whites. Cf. Letter to the Editor from Chandler 
Davidson, 17 New Perspectives 38 (Fall1985). Indeed, the 
facts of this case illustrate that tendency-blacks preferred 
black candidates, whites preferred white candidates. Thus, 
as a matter of convenience, we and the District Court may 
refer to the preferred representative of black voters as the 
"black candidate" and to the preferred representative of 
white voters as the "white candidate." Nonetheless, the fact 
that race of voter and race of candidate is often correlated is 
not directly pertinent to a § 2 inquiry. Under § 2, it is the 
status of the candidate as the chosen representative of a 
particular racial group, not the .race of the candidate, that is 
important. 

An understanding of how vote dilution through submer­
_gence in a white majority works leads to the same conclusion. 
The essence of a submergence claim is that minority group 
members prefer certain candidates whom they could elect 
were it not for the interaction of the challenged electoral law 
or structure with a white majority that votes as a significant 
bloc for different candidates. Thus, as we explained in Part 
III, supra, the existence of racial bloc voting is relevant to a 
vote dilution claim in two ways. Bloc voting by blacks tends 
to prove that the black community is politically cohesive, that 
is, it shows that blacks prefer certain candidates whom they 
could elect in a single-member, black majority district. Bloc 
voting by a white majority tends to prove that blacks will 
generally be unable to elect representatives of their choice. 
Clearly, only the race of the voter, not the race of the candi­
date, is relevant to vote dilution analysis. See, e. g., 
Blacksher & Menefee 59- 60; Grofman, Should Represent­
atives be Typical? 98; Note, Geometry and Geography 207. 



36 THORNBURG v. GINGLES 

Second, appellants' suggestion that racially polarized vot­
ing refers to voting patterns where whites vote for white 
candidates because they prefer members of their own race or 
are hostile to blacks, as opposed to voting patterns where 
whites vote for white candidates because the white candi­
dates spent more on their campaigns, utilized more media 
coverage, and thus enjoyed greater name recognition than 
the black candidates, fails for another, independent reason. 
This argument, like the argument that the race of the voter 
must be the primary determinant of the voter's ballot, is 
inconsistent with the purposes of § 2 and would render mean­
ingless the Senate Report factor that addresses the impact of 
low socioeconomic status on a minority group's level of politi­
cal participation. 

Congress intended that the Voting Rights Act eradicate in­
equalities in political opportunities that exist due to the vesti­
gial effects of past purposeful discrimination. S. Rep. 5, 40; 
H. R. Rep. No. 97- 227, p. 31 (1981). Both this Court and 
other federal courts have recognized that political participa­
tion by minorities tends to be depressed where minority 
group members suffer effects of prior discrimination such as 
inferior education, poor employment opportunities, and low 
incomes. See, e. g., White v. Regester, 412 U. S., at 
768-769; Kirksey v. Board of Supervisors of Hinds County, 
Miss., 554 F. 2d 139, 145-146 (CA5) (en bane), cert. denied, 
434 U. S. 968 (1977). See also Verba & Nie 152. The Sen­
ate Report acknowledges this tendency and instructs that 
"the extent to which members of the minority group ... bear 
the effects of discrimination in such areas as education, em­
ployment and health, which hinder their ability to participate 
effectively in the political process," S. Rep. 29 (footnote 
omitted), is a factor which may be probative of unequal 
opportunity to participate in the political process and to elect 
representatives. Courts and commentators have recognized 
further that candidates generally must spend more money in 
order to win election in a multimember district than in a 



THORNBURG v. GINGLES 37 

single-member district. See, e. g., Graves v. Barnes, 343 F. 
Supp. 704, 720- 721 (WD Tex. 1972), aff'd in part, rev'd in 
part sub nom. White v. Regester, supra. Berry & Dye 88; 
Davidson & Fraga, Non partisan Slating Groups in an At­
Large Setting, in Minority Vote Dilution 122-123; Derfner 
554, n. 126; Jewell131; Karnig, Black Representation on City 
Councils, 12 Urb. Aff. Q. 223, 230 (1976). If, because of in­
ferior education and poor employment opportunities, blacks 
earn less than whites, they will not be able to provide the 
candidates of their choice with the same level of financial sup­
port that whites can provide theirs. Thus, electoral losses 
by candidates preferred by the black community may well be 
attributable in part to the fact that their white opponents 
outspent them. But, the fact is that, in this instance, the 
economic effects of prior discrimination have combined with 
the multimember electoral structure to afford blacks less 
opportunity than whites to participate in the political process 
and to elect representatives of their choice. It would be 
both anomalous and inconsistent with congressional intent to 
hold that, on the one hand, the effects of past discrimination 
which hinder blacks' ability to participate in the political 
process tend to prove a § 2 violation, while holding on the 
other hand that, where these same effects of past discrimina­
tion deter whites from voting for blacks, blacks cannot make 
out a crucial element of a vote dilution claim. Accord, 
Escambia County, 748 F. 2d, at 1043 ("'the failure of the 
blacks to solicit white votes may be caused by the effects of 
past discrimination"') (quoting United States v. Dallas 
County Comm'n, 739 F. 2d 1529, 1536 (CA111984)); United 
States v. Marengo County Comm'n, 731 F. 2d 1546, 1567 
(CAll 1984), appeal dismissed, cert. denied, 469 U. S. -­
(1984). 

5 
Racial Animosity as Primary Determinant of 

Voter Behavior 
Finally, we reject the suggestion that racially polarized 



38 THORNBURG v. GINGLES 

voting refers only to white bloc voting which is caused by 
white voters' racial hostility toward black candidates. 33 To 
accept this theory would frustrate the goals Congress sought 
to achieve by repudiating the intent test of Mobile v. Bolden, 
446 U. S. 55 (1980), and would prevent minority voters who 
have clearly been denied an opportunity to elect represent­
atives of their choice from establishing a critical element of a 
vote dilution claim. 

In amending § 2, Congress rejected the requirement an­
nounced by this Court in Bolden, supra, that § 2 plaintiffs 
must prove the discriminatory intent of state or local govern­
ments in adopting or maintaining the challenged electoral 
mechanism. 34 Appellants' suggestion that the discriminatory 
intent of individual white voters must be proven in order to 
make out a § 2 claim must fail for the very reasons Congress 
rejected the intent test with respect to governmental bodies. 
See Engstrom, The Reincarnation of the Intent Standard: 
Federal Judges and At-Large Election Cases, 28 How. L. J. 
495 (1985). 

The Senate Report states that one reason the Senate Com­
mittee abandoned the intent test was that "the Committee 
... heard persuasive testimony that the intent test is unnec­
essarily divisive because it involves charges of racism on the 

33 It is true, as we have recognized previously, that racial hostility may 
often fuel racial bloc voting. United Jewish Organizations v. Carey, 430 
U. S. 144, 166 (1977); Rogers v. Lodge, 458 U. S., at 623. But, as we 
explain in this decision, the actual motivation of the voter has no relevance 
to a vote dilution claim. This is not to suggest that racial bloc voting is 
race-neutral; because voter behavior correlates with race, obviously 
it is not. It should be remembered, though, as one commentator has ob­
served, that "[t]he absence of racial animus is but one element of race 
neutrality." Note, Geometry and Geography 208. 

34 The Senate Report rejected the argument that the words "on account 
of race," contained in § 2(a), create any requirement of purposeful 
discrimination. "[I]t is patently [clear] that Congress has used the words 
'on account of race or color' in the Act to mean 'with respect to' race or 
color, and not to connote any required purpose of racial discrimination." 
S. Rep. 27-28, n. 109. 



THORNBURG v. GINGLES 39 

part of individual officials or entire communities." S. Rep. 
36. The Committee found the testimony of Dr. Arthur S. 
Flemming, Chairman of the United States Commission on 
Civil Rights particularly persuasive. He testified: 

"'[Under an intent test] [l]itigators representing ex­
cluded minorities will have to explore the motivations of 
individual council members, mayors, and other citizens. 
The question would be whether their decisions were 
motivated by invidious racial considerations. Such 
inquiries can only be divisive, threatening to destroy any 
existing racial progress in a community. It is the intent 
test, not the results test, that would make it necessary 
to brand individuals as racist in order to obtain judicial 
relief.'" 

Ibid. (footnote omitted). 
The grave threat to racial progress and harmony which 

Congress perceived from requiring proof that racism caused 
the adoption or maintenance of a challenged electoral mecha­
nism is present to a much greater degree in the proposed re­
quirement that plaintiffs demonstrate that racial animosity 
determined white voting patterns. Under the old intent 
test, plaintiffs might succeed by proving only that a limited 
number of elected officials were racist; under the new intent 
test plaintiffs would be required to prove that most of the 
white community is racist in order to obtain judicial relief. 
It is difficult to imagine a more racially divisive requirement. 

A second reason Congress rejected the old intent test was 
that in most cases it placed an "inordinately difficult burden" 
on § 2 plaintiffs. Ibid. The new intent test would be 
equally, if not more, burdensome. In order to prove that a 
specific factor-racial hostility-determined white voters' 
ballots, it would be necessary to demonstrate that other 
potentially relevant causal factors, such as socioeconomic 
characteristics and candidate expenditures, do not correlate 
better than racial animosity with white voting behavior. As 
one commentator has explained: 



40 THORNBURG v. GINGLES 

"Many of the[se] independent variables ... would be 
all but impossible for a social scientist to operationalize 
as interval-level independent variables for use in a multi­
ple regression equation, whether on a step-wise basis or 
not. To conduct such an extensive statistical analysis 
as this implies, moreover, can become prohibitively 
expensive. 

"Compared to this sort of effort, proving discrimina­
tory intent in the adoption of an at-large system is both 
simple and inexpensive." McCrary, Discriminatory In­
tent: The Continuing Relevance of "Purpose" Evidence 
in Vote-Dilution Lawsuits, 28 How. L. J. 463, 492 
(1985). 

The final and most dispositive reason the Senate Report 
repudiated the old intent test was that it "asks the wrong 
question." S. Rep. 36. Amended § 2 asks instead "whether 
minorities have equal access to the process of electing their 
representatives." Ibid. 

Focusing on the discriminatory intent of the voters, rather 
than the behavior of the voters, also asks the wrong question. 
All that matters under § 2 and under a functional theory of 
vote dilution is voter behavior, not its explanations. More­
over, as we have explained in detail, supra, requiring proof 
that racial considerations actually caused voter behavior will 
result-contrary to congressional intent-in situations where 
a black minority that functionally has been totally excluded 
from the political process will be unable to establish a § 2 
violation. The Senate Report's remark concerning the old 
intent test thus is pertinent to the new test: The requirement 
that a "court ... make a separate ... finding of intent, after 
accepting the proof of the factors involved in the White [ v. 
Regester, 412 U. S. 755] analysis . . . [would] seriously 
clou[d] the prospects of eradicating the remaining instances 
of racial discrimination in American elections." I d., at 37. 
We therefore decline to adopt such a requirement. 



THORNBURG v. GINGLES 

6 

Summary 

41 

In sum, we would hold that the legal concept of racially po­
larized voting, as it relates to claims of vote dilution, refers 
only to the existence of a correlation between the race of vot­
ers and the selection of certain candidates. Plaintiffs need 
not prove causation or intent in order to prove a prima facie 
case of racial bloc voting and defendants may not rebut that 
case with evidence of causation or intent. 

IV 
THE LEGAL SIGNIFICANCE OF SOME BLACK 

CANDIDATES' SUCCESS 

A 

North Carolina and the United States maintain that the 
District Court failed to accord the proper weight to the 
success of some black candidates in the challenged districts. 
Black residents of these districts, they point out, achieved 
improved representation in the 1982 General Assembly elec­
tion. 35 They also note that blacks in House District 23 have 
enjoyed proportional representation consistently since 1973 
and that blacks in the other districts have occasionally 
enjoyed nearly proportional representation. 35 This electoral 

36 The relevant results of the 1982 General Assembly election are as 
follows. House District 21, in which blacks make up 21.8% of the popula­
tion, elected one black to the six-person House delegation. House District 
23, in which blacks constitute 36.3% of the population, elected one black to 
the three-person House delegation. In House District 36, where blacks 
constitute 26.5% of the population, one black was elected to the eight­
member delegation. In House District 39, where 25.1% of the population 
is black, two blacks were elected to the five-member delegation. In Sen­
ate District 22, where blacks constitute 24.3% of the population, no black 
was elected to the Senate in 1982. 

36 The United States points .out that, under a substantially identical 
predecessor to the challenged plan, see n. 15, supra, House District 21 
elected a black to its six-member delegation in 1980, House District 39 



42 THORNBURG v. GINGLES 

success demonstrates conclusively, appellants and the United 
States argue, that blacks in those districts do not have "less 
opportunity than other members of the electorate to partici­
pate in the political process and to elect representatives 
of their choice." 42 U. S. C. § 1973(b). Essentially, appel­
lants and the United States contend that if a racial minority 
gains proportional or nearly proportional representation in a 
single election, that fact alone precludes, as a matter of law, 
finding a § 2 violation. 

Section 2 provides that "[t]he extent to which members of 
a protected class have been elected to office ... is one cir­
cumstance which may be considered." 42 U. S. C. § 1973(b). 
The Senate Committee Report also identifies the extent to 
which minority candidates have succeeded as a pertinent fac­
tor. S. Rep. 29. However, the Senate Report expressly 
states that "the election of a few minority candidates does not 
'necessarily foreclose the possibility of dilution of the black 
vote,"' noting that if it did, "the possibility exists that the 
majority citizens might evade [§ 2] by manipulating the elec­
tion of a 'safe' minority candidate." !d., at 29, n. 115, quot­
ing Zimmer v. McKeithen, 485 F. 2d 1297, 1307 (CA5 1973) 
(en bane), aff'd sub nom. East Carroll Parish School Board 
v. Marshall, 424 U. S. 636 (1976) (per curiam). The Senate 
Committee decided, instead, to "'require an independent 
consideration of the record."' S. Rep. 29, n. 115. The Sen­
ate Report also emphasizes that the question whether "the 
political processes are 'equally open' depends upon a search­
ing practical evaluation of the 'past and present reality."' 
I d., at 30 (footnote omitted). Thus, the language of § 2 and 
its legislative history plainly demonstrate that proof that 
some minority candidates have been elected does not fore­
close a § 2 claim. 

Moreover, in conducting its "independent consideration of 
the record" and its "searching practical evaluation of the 'past 

elected a black to its five-member delegation in 1974 and 1976, and Senate 
District 22 had a black Senator between 1975 and 1980. 



THORNBURG v. GINGLES 43 

and present reality,'" the District Court could appropriately 
take account of the circumstances surrounding recent black 
electoral success in deciding its significance to appellees' 
claim. In particular, as the Senate Report makes clear, !d., 
at 29, n. 115, the court could properly notice the fact that 
black electoral success increased markedly in the 1982 elec­
tion -an election that occurred after the instant lawsuit had 
been filed-and could properly consider to what extent "the 
pendency of this very litigation [might have] worked a one­
time advantage for black candidates in the form of unusual 
organized political support by white leaders concerned to 
forestall single-member districting." 37 590 F . Supp., at 367, 
n. 27. 

Nothing in the statute or its legislative history prohibited 
the court from viewing with some caution black candidates' 
success in the 1982 election, and from deciding on the basis of 
all the relevant circumstances to accord greater weight to 
blacks' relative lack of success over the course of several 
recent elections. Consequently, we hold that the District 
Court did not err, as a matter of law, in refusing to treat the 
fact that some black candidates have succeeded as dispositive 
of appellees' § 2 claim. Where multimember districting gen­
erally works to dilute the minority vote, it cannot be de­
fended on the ground that it sporadically and serendipitously 
benefits minority voters. 

37 See also Zimmer v. McKeithen, 485 F. 2d 1297, 1307 (CA5 1973) 
("[W]e cannot endorse the view that the success of black candidates at the 
polls necessarily forecloses the possibility of dilution of the black vote. 
Such success might, on occasion, be attributable to the work of politicians, 
who, apprehending that the support of a black candidate would be politi­
cally expedient, campaign to insure his election. Or such success might be 
attributable to political support motivated by different considerations­
namely that election of a black candidate will thwart successful challenges 
to electoral schemes on dilution grounds. In either situation, a candidate 
could be elected despite the relative political backwardness of black 
residents in the electoral district"). 



I' 

44 THORNBURG v. GINGLES 

B 

The District Court did err, however, in ignoring the 
significance of the sustained success black voters have ex­
perienced in House District 23. In that district, the last six 
elections have resulted in proportional representation for 
black residents. This persistent proportional representation 
is inconsistent with appellees' allegation that the ability of 
black voters in District 23 to elect representatives of their 
choice is not equal to that enjoyed by the white majority. 

In some situations, it may be possible for § 2 plaintiffs to 
demonstrate that such sustained success does not accurately 
reflect the minority group's ability to elect its preferred 
representatives, 38 but appellees have not done so here. Ap­
pellees presented evidence relating to black electoral success 
in the last three election; they failed utterly, though, to offer 
any explanation for the success of black candidates in the 
previous three elections. Consequently, we believe that the 
District Court erred, as a matter of law, in ignoring the 
sustained success black voters have enjoyed in House Dis­
trict 23, and would reverse with respect to that District. 

v 
ULTIMATE DETERMINATION OF VOTE DILUTION 
Finally, appellants and the United States dispute the 

District Court's ultimate conclusion that the multimember 
districting scheme at issue in this case deprived black voters 
of an equal opportunity to participate in the political process 
and to elect representatives of their choice. 

A 
As an initial matter, both North Carolina and the United 

States contend that the District Court's ultimate conclusion 
that the challenged multimember districts operate to dilute 

38 We have no occasion in this case to decide what types of special circum­
stances could satisfactorily demonstrate that sustained success does not ac­
curately reflect the minority's ability to elect its preferred representatives. 



THORNBURG v. GINGLES 45 

black citizens' votes is a mixed question of law and fact 
subject to de novo review on appeal. In support of their 
proposed standard of review, they rely primarily on Bose 
Corporation v. Consumers Union of U. S., Inc., 466 U. S. 
485 (1984), a case in which we reconfirmed that, as a matter 
of constitutional law, there must be independent appellate 
review of evidence of "actual malice" in defamation cases. 
Appellants and the United States argue that because a find­
ing of vote dilution under amended § 2 requires the applica­
tion of a rule of law to a particular set of facts it constitutes a 
legal, rather than factual, determination. Reply Brief for 
Appellants 7; Brief for United States as Amicus Curiae 
18-19. Neither appellants nor the United States cite our 
several precedents in which we have treated the ultimate 
finding of vote dilution as a question of fact subject to the 
clearly-erroneous standard of Rule 52(a). See, e. g., Rogers 
v. Lodge, 458 U. S., at 622-627; City of Rome v. United 
States, 446 U. S. 156, 183 (1980); White v. Regester, 412 
U. S., at 765-770. Cf. Anderson v. Bessemer City, 470 
u. s. -, - (1985). 

In Regester, supra, we noted that the District Court had 
based its conclusion that minority voters in two multimember 
districts in Texas had less opportunity to participate in the 
political process than majority voters on the totality of the 
circumstances and stated that 

"we are not inclined to overturn these findings, repre­
senting as they do a blend of history and an intensely 
local appraisal of the design and impact of the . .. 
multimember district in the light of past and present 
reality, political and otherwise." Id., at 769-770. 

Quoting this passage from Regester with approval, we 
expressly held in Rogers v. Lodge, supra, that the question 
whether an at-large election system was maintained for dis­
criminatory purposes and subsidiary issues, which include 
whether that system had the effect of diluting the minority 
vote, were questions of fact, reviewable under Rule 52(a)'s 



46 THORNBURG v. GINGLES 

clearly-erroneous standard. 458 U. S., at 622-623. Simi­
larly, in City of Rome v. United States, we declared that the 
question whether certain electoral structures had a "discrimi­
natory effect," in the sense of diluting the minority vote, was 
a question of fact subject to clearly-erroneous review. 446 
U. S., at 183. 

We reaffirm our view that the clearly-erroneous test of 
Rule 52(a) is the appropriate standard for appellate review of 
a finding of vote dilution. As both amended § 2 and its legis­
lative history make clear, in evaluating a statutory claim of 
vote dilution through districting, 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,'" S. Rep. 30 (footnote omitted), whether the 
political process is equally open to minority voters. "'This 
determination is peculiarly dependent upon the facts of each 
case,"' Rogers, supra, at 621, quoting Nevett v. Sides, 571 F. 
2d 209, 224 (CA5 1978), and requires "an intensely local ap­
praisal of the design and impact" of the contested electoral 
mechanisms. 458 U. S., at 622. The fact that amended § 2 
and its legislative history provide legal standards which a 
court must apply to the facts in order to determine whether 
§ 2 has been violated does not alter the standard of review. 
As we explained in Bose, Rule 52(a) "does not inhibit an ap­
pellate court's power to correct errors of law, including those 
that 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." 466 U. S., at 501, citing Pullman­
Standard v. Swint, 456 U. S. 273, 287 (1982); Inwood Lab­
oratories, Inc. v. Ives Laboratories, Inc., 456 U. S. 844, 855, 
n. 15 (1982). Thus, the 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. 



THORNBURG v. GINGLES 47 

B 

The District Court in this case carefully considered the 
totality of the circumstances and found that in each district 
racially polarized voting; the legacy of official discrimination 
in voting matters, education, housing, employment, and 
health services; and the persistence of campaign appeals to 
racial prejudice acted in concert with the multimember dis­
tricting scheme to impair the ability of geographically insular 
and politically cohesive groups of black voters to participate 
equally in the political process and to elect candidates of their 
choice. It found that the success a few black candidates 
have enjoyed in these districts is too recent, too limited, and, 
with regard to the 1982 elections, perhaps too aberrational, 
to disprove its conclusion. Excepting House District 23, 
with respect to which the District Court committed legal 
error, see supra, at--, we affirm the District Court's judg­
ment. We cannot say that the District Court, composed of 
local judges who are well-acquainted with the political reali­
ties of the State, clearly erred in concluding that use of a 
multimember electoral structure has caused black voters in 
the districts other than House District 23 to have less oppor­
tunity than white voters to elect representatives of their 
choice. 

The judgment of the District Court is 

Affirmed in part, reversed in part. 



48 THORNBURG v. GINGLES 

APPENDIX A 

Percentages of Votes Cast by Black and White Voters for 
Black Candidates in the. Five Contested Districts 

Senate District 22 

Primary General 
White Black White Black 

1978 (Alexander) 47 87 41 94 
1980 (Alexander) 23 78 n/a n/a 
1982 (Polk) 32 83 33 94 

House District 21 

Primary General 
White Black White Black 

1978 (Blue) 21 76 n/a n/a 
1980 (Blue) 31 81 44 90 
1982 (Blue) 39 82 45 91 

House District 23 

Primary General 
White Black White Black 

1978 Senate 
Barns (Repub.) n/a n/a 17 5 
1978 House 
Clement 10 89 n/a n/a 
Spaulding 16 92 37 89 
1980 House 
Spaulding n/a n/a 49 90 
1982 House 
Clement 26 32 n/a n/a 
Spaulding 37 90 43 89 



THORNBURG v. GINGLES 49 

House District 36 

Primary General 
White Black White Black 

1980 (Maxwell) 22 71 28 92 
1982 (Berry) 50 79 42 92 
1982 (Richardson) 39 71 29 88 

House District 39 

Primary General 
White Black White Black 

1978 House 
Kennedy, H. 28 76 32 93 
Norman 8 29 n/a n/a 
Ross 17 53 n/a n/a 
Sumter (Repub) n/a n/a 33 25 
1980 House 
Kennedy, A. 40 86 32 96 
Norman 18 36 n/a n/a 

1980 Senate 
Small 12 61 n/a n/a 

1982 House 
Hauser 25 80 42 87 
Kennedy, A. 36 87 46 94 

590 F. Supp., at 369-371. 



50 THORNBURG v. GINGLES 

APPENDIX B 

Black Candidates Elected From 7 Originally Contested 
Districts 

District Prior to 
(No. Seats) 1972 1972 1974 1976 1978 1980 1982 
House 8 (4) 0 0 0 0 0 0 0 
House 21 (6) 0 0 0 0 0 1 1 
House 23 (3) 0 1 1 1 1 1 1 
House 36 (8) 0 0 0 0 0 0 1 
House 39 (5) 0 0 1 1 0 0 2 
Senate 2 (2) 0 0 0 0 0 0 0 
Senate 22 (4) 0 0 1 1 1 0 0 

See Brief for Appellees, table printed between pages 8 and 9; 
App. 93-94. 



SUPREME COURT OF THE UNITED STATES 

No. 83-1968 

LACY H. THORNBURG, ET AL., APPELLANTS v. 
RALPH GINGLES ET AL. 

ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR 
THE EASTERN DISTRICT OF NORTH CAROLINA 

[June 30, 1986] 

JUSTICE WHITE, concurring. 
I join Parts I, II, III-A, III-B, IV- A and V of the Court's 

opinion and agree with JUSTICE BRENNAN's opinion as to 
Part IV-B. I disagree with Part III-C of JUSTICE BREN­
NAN's opinion. 

JUSTICE BRENNAN states in Part III-C that the crucial 
factor in identifying polarized voting is the race of the voter 
and that the race of the candidate is irrelevant. Under this 
test, there is polarized voting if the majority of white voters 
vote for different candidates than the majority of the blacks, 
regardless of the race of the candidates. I do not agree. 
Suppose an 8-member multimember district that is 60% 
white and 40% black, the blacks being geographically located 
so that 2 safe black single-member districts could be drawn. 
Suppose further that there are 6 white and 2 black Demo­
crats running against 6 white and 2 black Republicans. 
Under JuSTICE BRENNAN's test, there would be polarized 
voting and a likely § 2 violation if all the Republicans, includ­
ing the 2 blacks, are elected, and 80% of the blacks in the pre­
dominantly black areas vote Democratic. I take it that there 
would also be a violation in a single-member district that is 
60% black, but enough of the blacks vote with the whites to 
elect a black candidate who is not the choice of the majority of 
black voters. This is interest-group politics rather than a 
rule hedging against racial discrimination. I doubt that this 



2 THORNBURG v. GINGLES 

is what Congress had in mind in amending § 2 as it did, and it 
seems quite at odds with the discussion in Whitcomb v. 
Chavis, 403 U.S 124, 149-160 (1971). Furthermore, on 
the facts of this case, there is no need to draw the 
voter/candidate distinction. The District Court did not and 
reached the correct result except, in my view, with respect to 
District 23. 



SUPREME COURT OF THE UNITED STATES 

No. 83- 1968 

LACY H. THORNBURG, ET AL., APPELLANTS v. 
RALPH GINGLES ET AL. 

ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR 
THE EASTERN DISTRICT OF NORTH CAROLINA 

[June 30, 1986] 

JusTICE O'CONNOR, with whom THE CHIEF JUSTICE, Jus­
TICE POWELL, and JUSTICE REHNQUIST join, concurring in 
the judgment. 

In this case, we are called upon to construe § 2 of the Vot­
ing Rights Act of 1965, as amended June 29, 1982. Amended 
§ 2 is intended to codify the "results" test employed in 
Whitcomb v. Chavis, 403 U. S. 124 (1971), and White v. 
Regester, 412 U. S. 755 (1973), and to reject the "intent" test 
propounded in the plurality opinion in Mobile v. Bolden, 446 
U. S. 55 (1980) . S. Rep. No. 97-417, pp. 27-28 (1982) (here­
inafter S. Rep.). Whereas Bolden required members of a 
racial minority who alleged impairment of their voting 
strength to prove that the challenged electoral system was 
created or maintained with a discriminatory purpose and led 
to discriminatory results, under the results test, "plaintiffs 
may choose to establish discriminatory results without prov­
ing any kind of discriminatory purpose." S. Rep. 28. At 
the same time, however, § 2 unequivocally disclaims the . 
creation of a right to proportional representation. This 
disclaimer was essential to the compromise that resulted in 
passage of the amendment. See id., at 193- 194 (Additional 
Views of Sen. Dole). 

In construing this compromise legislation, we must make 
every effort to be faithful to the balance Congress struck. 
This is not an easy task. We know that Congress intended 



2 THORNBURG v. GINGLES 

to allow vote dilution claims to be brought under § 2, but we 
also know that Congress did not intend to create a right to 
proportional representation for minority voters. There is an 
inherent tension between what Congress wished to do and 
what it wished to avoid, because any theory of vote dilution 
must necessarily rely to some extent on a measure of minor­
ity voting strength that makes some reference to the propor­
tion between the minority group and the electorate at large. 
In addition, several important aspects of the "results" test 
had received little attention in this Court's cases or in the de­
cisions of the Courts of Appeals employing that test on which 
Congress also relied. See id., at 32. Specifically, the legal 
meaning to be given to the concepts of "racial bloc voting" 
and "minority voting strength" had been left largely un­
addressed by the courts when § 2 was amended. 

The Court attempts to resolve all these difficulties today. 
First, the Court supplies definitions of racial bloc voting and 
minority voting strength that will apparently be applicable in 
all cases and that will dictate the structure of vote dilution 
litigation. Second, the Court adopts a test, based on the 
level of minority electoral success, for determining when an 
electoral scheme has sufficiently diminished minority voting 
strength to constitute vote ·dilution. Third, although the 
Court does not acknowledge it expressly, the combination of 
the Court's definition of minority voting strength and its test 
for vote dilution results in the creation of a right to a form of 
proportional representation in favor of all geographically and 
politically cohesive minority groups that are large enough to 
constitute majorities if concentrated within one or more sin­
·gle-member districts. In so doing, the Court has disre­
garded the balance struck by Congress in amending § 2 and 
has failed to apply the results test as described by this Court 
in Whitcomb and White. 

I 

In order to explain my disagreement with the Court's in­
terpretation of § 2, it is useful to illustrate the impact that al-



THORNBURG v. GINGLES 3 

ternative districting plans or types of districts typically have 
on the likelihood that a minority group will be able to elect 
candidates it prefers, and then to set out the critical elements 
of a vote dilution claim as they emerge in the Court's opinion. 

Consider a town of 1,000 voters that is governed by a coun­
cil of four representatives, in which 30% of the voters are 
black, and in which the black voters are concentrated in one 
section of the city and tend to vote as a bloc. It would be 
possible to draw four single-member districts, in one of which 
blacks would constitute an overwhelming majority. The 
black voters in this district would be assured of electing a 
representative of their choice, while any remaining black vot­
ers in the other districts would be submerged in large white 
majorities. This option would give the minority group 
roughly proportional representation. 

Alternatively, it would usually be possible to draw four 
single-member districts in two of which black voters con­
stituted much narrower majorities of about 60%. The black 
voters in these districts would often be able to elect the 
representative of their choice in each of these two districts, 
but if even 20% of the black voters supported the candidate 
favored by the white minority in those districts the candi­
dates preferred by the majority of black voters might lose. 
This option would, depending on the circumstances of a par­
ticular election, sometimes give the minority group more 
than proportional representation, but would increase the risk 
that the group would not achieve even roughly proportional 
representation. 

It would also usually be possible to draw four single-mem­
ber districts in each of which black voters constituted a mi­
nority. In the extreme case, black voters would constitute 
30% of the voters in each district. Unless approximately 
30% of the white voters in this extreme case backed the mi­
nority candidate, black voters in such a district would be 
unable to elect the candidate of their choice in an. election 
between only two candidates even if they unanimously sup-



4 THORNBURG v. GINGLES 

ported him. This option would make it difficult for black 
voters to elect candidates of their choice even with significant 
white support, and all but impossible without such support. 

Finally, it would be possible to elect all four represent­
atives in a single at-large election in which each voter could 
vote for four candidates. Under this scheme, white voters 
could elect all the representatives even if black voters turned 
out in large numbers and voted for one and only one candi­
date. To illustrate, if only four white candidates ran, and 
each received approximately equal support from white vot­
ers, each would receive about 700 votes, whereas black vot­
ers could cast no more than 300 votes for any one candidate. 
If, on the other hand, eight white candidates ran, and white 
votes were distributed less evenly, so that the five least fa­
vored white candidates received fewer than 300 votes while 
three others received 400 or more, it would be feasible for 
blacks to elect one representative with 300 votes even with­
out substantial white support. If even 25% of the white vot­
ers backed a particular minority candidate, and black voters 
voted only for that candidate, the candidate would receive a 
total of 475 votes, which would ensure victory unless white 
voters also concentrated their votes on four of the eight 
remaining candidates, so that each received the support of 
almost 70% of white voters. As these variations show, the 
at-large or multimember district has an inherent tendency to 
submerge the votes of the minority. The minority group's 
prospects for electoral success under such a district heavily 
depend on a variety of factors such as voter turnout, how 
many candidates run, how evenly white support is spread, 
how much white support is given to a candidate or candidates 
preferred by the minority group, and the extent to which mi­
nority voters engage in "bullet voting" (which occurs when 
voters refrain from casting all their votes to avoid the risk 
that by voting for their lower-ranked choices they may give 
those candidates enough votes to defeat their higher-ranked 
choices, see ante, at --, n. 5). 



THORNBURG v. GINGLES 5 

There is no difference in principle between the varying 
effects of the alternatives outlined above and the varying 
effects of alternative single-district plans and multimember 
districts. The type of districting selected and the way in 
which district lines are drawn can have a powerful effect on 
the likelihood that members of a geographically and politi­
cally cohesive minority group will be able to elect candidates 
of their choice. 

Although § 2 does not speak in terms of "vote dilution," I 
agree with the Court that proof of vote dilution can establish 
a violation of § 2 as amended. The phrase "vote dilution," in 
the legal sense, simply refers to the impermissible discrimi­
natory effect that a multimember or other districting plan 
has when it operates "to cancel out or minimize the voting 
strength of racial groups." White, 412 U. S., at 765. See 
also Fortson v. Dorsey, 379 U. S. 433, 439 (1965). This defi­
nition, however, conceals some very formidable difficulties. 
Is the "voting strength" of a racial group to be assessed solely 
with reference to its prospects for electoral success, or should 
courts look at other avenues of political influence open to the 
racial group? Insofar as minority voting strength is as­
sessed with reference to electoral success, how should undi­
luted minority voting strength be measured? How much of 
an impairment of minority voting strength is necessary to 
prove a violation of § 2? What constitutes racial bloc voting 
and how is it proved? What weight is to be given to evi­
dence of actual electoral success by minority candidates in 
the face of evidence of racial bloc voting? 

The Court resolves the first question summarily: minority 
voting strength is to be assessed solely in terms of the minor­
ity group's ability to elect candidates it prefers. Ante, at 
----- Under this approach, the essence of a vote di­
lution claim is that the State has created single-member or 
multimember districts that unacceptably impair the minority 
group's ability to elect the candidates its members prefer. 



6 THORNBURG v. GINGLES 

In order to evaluate a claim that a particular multimember 
district or single-member district has diluted the minority 
group's voting strength to a degree that violates § 2, how­
ever, it is also necessary to construct a measure of "undi­
luted" minority voting strength. "[T]he phrase [vote dilu­
tion] itself suggests a norm with respect to which the fact of 
dilution may be ascertained." Mississippi Republican Exec­
utive Committee v. Brooks, 469 U. S. 1002, 1012 (1984) 
(REHNQUIST, J., dissenting from summary affirmance). Put 
simply, in order to decide whether an electoral system has 
made it harder for minority voters to elect the candidates 
they prefer, a court must have an idea in mind of how hard it 
"should" be for minority voters to elect their preferred candi­
dates under an acceptable system. 

Several possible measures of "undiluted" minority voting 
strength suggest themselves. First, a court could simply 
use proportionality as its guide: if the minority group consti­
tuted 30% of the voters in a given area, the court would re­
gard the minority group as having the potential to elect 30% 
of the representatives in that area. Second, a court could 
posit some alternative districting plan as a "normal" or "fair" 
electoral scheme and attempt to calculate how many candi­
dates preferred by the minority group would probably be 
elected under that scheme. There are, as we have seen, a 
variety of ways in which even single-member districts could 
be drawn, and each will present the minority group with its 
own array of electoral risks and benefits; the court might, 
therefore, consider a range of acceptable plans in attempting 
to estimate "undiluted" minority voting strength by this 
method. Third, the court could attempt to arrive at a plan 
that would maximize feasible minority electoral success, and 
use this degree of predicted success as its measure of "undi­
luted" minority voting strength. If a court were to employ 
this third alternative, it would often face hard choices about 
what would truly "maximize" minority electoral success. An 
example is the scenario described above, in which a minority 



THORNBURG v. GINGLES 7 

group could be concentrated in one completely safe district or 
divided among two districts in each of which its members 
would constitute a somewhat precarious majority. 

The Court today has adopted a variant of the third ap­
proach, to wit, undiluted minority voting strength means the 
maximum feasible minority voting strength. In explaining 
the elements of a vote dilution claim, the Court first states 
that "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." Ante, at--. If 
not, apparently the minority group has no cognizable claim 
that its ability to elect the representatives of its choice has 
been impaired. 1 Second, "the minority group must be able 
to show that it is politically cohesive, that is, that a signifi­
cant proportion of the minority group supports the same can­
didates." Ante, at - -. Third, the Court requires the mi-

' I express no view as to whether the ability of a minority group to con­
stitute a majority in a single-member district should constitute a threshold 
requirment for a claim that the use of multimember districts impairs the 
ability of minority voters to participate in the political processes and to 
elect representatives of their choice. Because the plaintiffs in this case 
would meet that requirement, if indeed it exists, I need not decide whether 
it is imposed by § 2. I note, however, the artificiality of the Court's dis­
tinction between claims that a minority group's "ability to elect the repre­
sentatives of [its] choice" has been impaired and claims that "its ability to 
influence elections" has been impaired. Ante, at --, n. 12. It is true 
that a minority group that could constitute a majority in a single-member 
district ordinarily has the potential ability to elect representatives without 
white support, and that a minority that could not constitute such a majority 
ordinarily does not. But the Court recognizes that when the candidates 
preferred by a minority group are elected in a multimember district, the 
minority group has elected those candidates, even if white support was in­
dispensable to these victories. On the same reasoning, if a minority group 
that is not large enough to constitute a voting majority in a single-member 
district can show that white support would probably be forthcoming in 
some such district to an extent that would enable the election of the candi­
dates its members prefer, that minority group would appear to have dem­
onstrated that, at least under this measure of its voting strength, it would 
be able to elect some candidates of its choice. 



8 THORNBURG v. GINGLES 

nority group 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." Ibid. If these three requirements are 
met, "the minority group demonstrates that submergence in 
a white multimember district impedes its ability to elect its 
chosen representatives." Ibid. That is to say, the minority 
group has proved vote dilution in violation of § 2. 

The Court's definition of the elements of a vote dilution 
claim is simple and invariable: a court should calculate minor­
ity voting strength by assuming that the minority group is 
concentrated in a single-member district in which it consti­
tutes a voting majority. Where the minority group is not 
large enough, geographically concentrated enough, or politi­
cally cohesive enough for this to be possible, the minority 
group's claim fails. Where the minority group meets these 
requirements, the representatives that it could elect in the 
hypothetical district or districts in which it constitutes a 
majority will serve as the measure of its undiluted voting 
strength. Whatever plan the State actually adopts must be 
assessed in terms of the effect it has on this undiluted voting 
strength. If this is indeed the single, universal standard for 
evaluating undiluted minority voting strength for vote dilu­
tion purposes, the standard is applicable whether what is 
challenged is a multimember district or a particular single­
member districting scheme. 

The Court's statement of the elements of a vote dilution 
claim also supplies an answer to another question posed 
above: how much of an impairment of undiluted minority vot­
ing strength is necessary to prove vote dilution. The Court 
requires the minority group that satisfies the threshold re­
quirements of size and cohesiveness to prove that it will 
usually be unable to elect as many representatives of its 
choice under the challenged districting scheme as its un­
diluted voting strength would permit. This requirement, 
then, constitutes the true test of vote dilution. Again, no 



THORNBURG v. GINGLES 9 

reason appears why this test would not be applicable to a 
vote dilution claim challenging single-member as well as 
multimember districts. 

This measure of vote dilution, taken in conjunction with 
the Court's standard for measuring undiluted minority voting 
strength, creates what amounts to a right to usual, roughly 
proportional representation on the part of sizeable, compact, 
cohesive minority groups. If, under a particular multi­
member or single-member district plan, qualified minority 
groups usually cannot elect the representatives they would 
be likely to elect under the most favorable single-member 
districting plan, then § 2 is violated. Unless minority 
success under the challenged electoral system regularly ap­
proximates this rough version of proportional representation, 
that system dilutes minority voting strength and violates § 2. 

To appreciate the implications of this approach, it is useful 
to return to the illustration of a town with four council repre­
sentatives given above. Under the Court's approach, if the 
black voters who comprise 30% of the town's voting popula­
tion do not usually succeed in electing one representative of 
their choice, then regardless of whether the town employs at­
large elections or is divided into four single-member districts, 
its electoral system violates § 2. Moreover, if the town had a 
black voting population of 40%, on the Court's reasoning the 
black minority, so long as it was geographically and politi­
cally cohesive, would be entitled usually to elect two of the 
four representatives, since it would normally be possible to 
create two districts in which black voters constituted safe 
majorities of approximately 80%. 

To be sure, the Court also requires that plaintiffs prove 
that racial bloc voting by the white majority interacts with 
the challenged districting plan so as usually to defeat the mi­
nority's preferred candidate. In fact, however, this require­
ment adds little that is not already contained in the Court's 
requirements that the minority group be politically cohesive 
and that its preferred candidates usually lose. As the Court 



10 THORNBURG v. GINGLES 

acknowledges, under its approach, "in general, a white bloc 
vote that normally will defeat the combined strength of mi­
nority support plus white 'crossover' votes rises to the level 
oflegally significant white bloc voting." Ante, at--. But 
this is to define legally significant bloc voting by the racial 
majority in terms of the extent of the racial minority's elec­
toral success. If the minority can prove that it could consti­
tute a majority in a single-member district, that it supported 
certain candidates and that those candidates have not usually 
been elected, then a finding that there is "legally significant 
white bloc voting" will necessarily follow. Otherwise, by 
definition, those candidates would usually have won rather 
than lost. 

As shaped by the Court today, then, the basic contours of a 
vote dilution claim require no reference to most of the "Zim­
mer factors" that were developed by the Fifth Circuit to 

. implement White's results test and which were highlighted 
in the Senate Report. S. Rep. 28-29; see Zimmer v. 
MC'Keithen, 485 F. 2d 1297 (1973), aff'd sub nom. East 
Carroll Parish School Board v. Marshall, 424 U. S. 636 
(1976) (per curiam). If a minority group is politically and 
geographically cohesive and large enough to constitute a vot­
ing majority in one or more single-member districts, then un­
less white voters usually support the minority's preferred 
candidates in sufficient numbers to enable the minority group 
to elect as many of those candidates as it could elect in such 
hypothetical districts, it will routinely follow that a vote dilu­
tion claim can be made out, and the multimember district will 
be invalidated. There is simply no need for plaintiffs to es­
tablish "the history of voting-related discrimination in the 
State or political subdivision," ante, at--, or "the extent to 
which the State or political subdivision has used voting prac­
tices or procedures that tend to enhance the opportunity for 
discrimination against the minority group," ibid., or "the ex­
clusion of members ofthe minority group from candidate slat­
ing processes," ibid., or "the extent to which minority group 



THORNBURG v. GINGLES 11 

members bear the effects of past discrimination in areas such 
as education, employment, and health," ibid. , or "the use of 
overt or subtle racial appeals in political campaigns," ibid., or 
that "elected officials are unresponsive to the particularized 
needs of the members ofthe minority group." Ante, at--. 
Of course, these other factors may be supportive of such a 
claim, because they may strengthen a court's confidence that 
minority voters will be unable to overcome the relative disad­
vantage at which they are placed by a particular districting 
plan, or suggest a more general lack of opportunity to partici­
pate in the political process. But the fact remains that elec­
toral success has now emerged, under the Court's standard, 
as the linchpin of vote dilution claims, and that the elements 
of a vote dilution claim create an entitlement to roughly pro­
portional representation within the framework of single­
member districts. 

II 

In my view, the Court's test for measuring minority voting 
strength and its test for vote dilution, operating in tandem, 
come closer to an absolute requirement of proportional repre­
sentation than Congress intended when it codified the results 
test in § 2. It is not necessary or appropriate to decide in 
this case whether § 2 requires a uniform measure of undiluted 
minority voting strength in every case, nor have appellants 
challenged the standard employed by the District Court for 
assessing undiluted minority voting strength. 

In this case, the District Court seems to have taken an ap­
proach quite similar to the Court's in making its preliminary 
assessment of undiluted minority voting strength: 

"At the time of the creation of these multi-member dis­
tricts, there were concentrations of black citizens within 
the boundaries of each that were sufficient in numbers 
and contiguity to constitute effective voting majorities in 
single-member districts lying wholly within the bound­
aries of the multi-member districts, which single-mem­
ber districts would satisfy all constitutional require-



12 THORNBURG v. GINGLES 

ments of population and geographical configuration." 
Gingles v. Edmisten, 590 F. Supp. 345, 358-359 (ED 
N. C. 1984). 

The Court goes well beyond simply sustaining the District 
Court's decision to employ this measure of undiluted minority 
voting strength as a reasonable one that is consistent with 
§ 2. In my view, we should refrain from deciding in this case 
whether a court must invariably posit as its measure of "un­
diluted" minority voting strength single-member districts in 
which minority group members constitute a majority. 
There is substantial doubt that Congress intended "undiluted 
minority voting strength" to mean "maximum feasible minor­
ity voting strength." Even if that is the appropriate defini­
tion in some circumstances, there is no indication that Con­
gress intended to mandate a single, universally applicable 
standard for measuring undiluted minority voting strength, 
regardless of local conditions and regardless of the extent of 
past discrimination against minority voters in a particular 
State or political subdivision. Since appellants have not 
raised the issue, I would assume that what the District Court 
did here was permissible under § 2, and leave open the 
broader question whether § 2 requires this approach. 

What appellants do contest is the propriety of the District 
Court's standard for vote dilution. Appellants claim that the 
District Court held that "[a]lthough blacks had achieved con­
siderable success in winning state legislative seats in the 
challenged districts, their failure to consistently attain the 
number of seats that numbers alone would presumptively 
give them (i. e., in proportion to their presence in the popula­
tion)," standing alone, constituted a violation of § 2. Brief 
for Appellants 20 (emphasis in original). This holding, ap­
pellants argue, clearly contravenes § 2's proviso that "nothing 
in this section establishes a right to have members of a pro­
tected class elected in numbers equal to their proportion in 
the population." 42 U. S. C. § 1973. 



THORNBURG v. GINGLES 13 

I believe appellants' characterization of the District Court's 
holding is incorrect. In my view, the District Court con­
cluded that there was a severe diminution in the prospects 
for black electoral success in each of the challenged districts, 
as compared to single-member districts in which blacks could 
constitute a majority, and that this severe diminution was in 
large part attributable to the interaction of the multimember 
form of the district with persistent racial bloc voting on the 
part of the white majorities in those districts. See 590 F. 
Supp., at 372.2 The District Court attached great weight to 
this circumstance as one part of its ultimate finding that "the 
creation of each of the multi-member districts challenged in 
this action results in the black registered voters of that dis­
trict being submerged as a voting minority in the district and 
thereby having less opportunity than do other members of 
the electorate to participate in the political process and to 
elect representatives of their choice." I d., at 37 4. But the 
District Court's extensive opinion clearly relies as well on a 
variety of the other Zimmer factors, as the Court's thorough 
summary of the District Court's findings indicates. See 
ante, at -- - --. 

If the District Court had held that the challenged multi­
member districts violated § 2 solely because blacks had not 
consistently attained seats in proportion to their presence in 
the population, its holding would clearly have been inconsist­
ent with § 2's disclaimer of a right to proportional representa­
tion. Surely Congress did not intend to say, on the one 
hand, that members of a protected class have no right to pro­
portional representation, and on the other, that any consist­
ent failure to achieve proportional representation, without 

'At times, the District Court seems to have looked to simple proportion­
ality rather than to hypothetical single-member districts in which black 
voters would constitute a majority. See, e. g., 590 F. Supp., at 367. No­
where in its opinion, however, did the District Court state that§ 2 requires 
that minority groups consistently attain the level of electoral success that 
would correspond with their proportion of the total or voting population. 



14 THORNBURG v. GINGLES 

more, violates § 2. A requirement that minority representa­
tion usually be proportional to the minority group's propor­
tion in the population is not quite the same as a right to strict 
proportional representation, but it comes so close to such a 
right as to be inconsistent with § 2's disclaimer and with the 
results test that is codified in § 2. In the words of Senator 
Dole, the architect of the compromise that resulted in pas­
sage of the amendments to § 2: 

"The language of the subsection explicitly rejects, as did 
White and its progeny, the notion that members of a pro­
tected class have a right to be elected in numbers equal 
to their proportion of the population. The extent to 
which members of a protected class have been elected 
under the challenged practice or structure is just one fac­
tor, among the totality of circumstances to be consid­
ered, and is not dispositive." S. Rep. 194 (Additional 
Views of Sen. Dole). 

On the same reasoning, I would reject the Court's test for 
vote dilution. The Court measures undiluted minority vot­
ing strength by reference to the possibility of creating single­
member districts in which the minority group would consti­
tute a majority, rather than by looking to raw proportionality 
alone. The Court's standard for vote dilution, when com­
bined with its test for undiluted minority voting strength, 
makes actionable every deviation from usual, rough propor­
tionality in representation for any cohesive minority group as 
to which this degree of proportionality is feasible within the 
framework of single-member districts. Requiring that 
every minority group that could possibly constitute a major­
ity in a single-member district be assigned to such a district 
would approach a requirement of proportional representation 
as nearly as is possible within the framework of single-mem­
ber districts. Since the Court's analysis entitles every such 
minority group usually to elect as many representatives 
under a multimember district as it could elect under the most 
favorable single-member district scheme, it follows that the 



THORNBURG v. GINGLES 15 

Court is requiring a form of proportional representation. 
This approach is inconsistent with the results test and with 
§ 2's disclaimer of a right to proportional representation. 

In enacting § 2, Congress codified the "results" test this 
Court had employed, as an interpretation of the Fourteenth 
Amendment, in White and Whitcomb. The factors devel­
oped by the Fifth Circuit and relied on by the Senate Report 
simply fill in the contours of the "results" test as described in 
those decisions, and do not purport to redefine or alter the 
ultimate showing of discriminatory effect required by Whit­
comb and White. In my view, therefore, it is to Whitcomb 
and White that we should look in the first instance in deter­
mining how great an impairment of minority voting strength 
is required to establish vote dilution in violation of § 2. 

The "results" test as reflected in Whitcomb and White re­
quires an inquiry into the extent of the minority group's 
opportunities to participate in the political processes. See -
White, supra, at 766. While electoral success is a central 
part of the vote dilution inquiry, White held that to prove 
vote dilution, "it is not enough that the racial group allegedly 
discriminated against has not had legislative seats in propor­
tion to its voting potential," 412 U. S., at 765- 766, and 
Whitcomb flatly rejected the proposition that "any group 
with distinctive interests must be represented in legislative 
halls if it is numerous enough to command at least one seat 
and represents a minority living in an area sufficiently com­
pact to constitute a single member district." 403 U. S., at 
156. To the contrary, the results test as described in White 
requires plaintiffs to establish "that the political processes 
leading to nomination and election were not equally open to 
participation by the group in question-that its members had 
less opportunity than did other residents in the district to 
participate in the political processes and to elect legislators of 
their choice." 412 U. S., at 766. By showing both "a his­
tory of disproportionate results" and "strong indicia of lack of 
political power and the denial of fair representation," the 



16 THORNBURG v. GINGLES 

plaintiffs in White met this standard, which, as emphasized 
just this Term, requires "a substantially greater showing of 
adverse effects than a mere lack of proportional representa­
tion to support a finding of unconstitutional vote dilution." 
Davis v. Bandemer,-- U.S. - -,--,-- (1986) (plu­
rality opinion). 

When Congress amended § 2 it intended to adopt this "re­
sults" test, while abandoning the additional showing of dis­
criminatory intent required by Bolden. The vote dilution 
analysis adopted by the Court today clearly bears little 
resemblence to the "results" test that emerged in Whitcomb 
and White. The Court's test for vote dilution, combined 
with its standard for evaluating "voting potential," White, 
supra, at 766, means that any racial minority with distinctive 
interests must usually "be represented in legislative halls if 
it is numerous enough to command at least one seat and rep­
resents a minority living in an area sufficiently compact to 
constitute" a voting majority in "a single member district." 
Whitcomb, 403 U. S., at 156. Nothing in Whitcomb, White, 
or the language and legislative history of § 2 supports the 
Court's creation of this right to usual, roughly proportional 
representation on the part of every geographically compact, 
politically cohesive minority group that is large enough to 
form a majority in one or more single-member districts. 

I would adhere to the approach outlined in Whitcomb and 
White and followed, with some elaboration, in Zimmer and 
other cases in the Courts of Appeals prior to Bolden. Under 
that approach, a court should consider all relevant factors 
bearing on whether the minority group has "less opportunity 
than other members of the electorate to participate in the po­
litical process and to elect representatives of their choice." 
42 U. S. C. § 1973 (emphasis added). The court should not 
focus solely on the minority group's ability to elect represent­
atives of its choice. Whatever measure of undiluted minor­
ity voting strength the court employs in connection with eval­
uating the presence or absence of minority electoral success, 



THORNBURG v. GINGLES 17 

it should also bear in mind that "the power to influence the 
political process is not limited to winning elections." Davis 
v. Bandemer, supra, at--. Of course, the relative lack of 
minority electoral success under a challenged plan, when 
compared with the success that would be predicted under the 
measure of undiluted minority voting strength the court is 
employing, can constitute powerful evidence of vote dilution. 
Moreover, the minority group may in fact lack access to or 
influence upon representatives it did not support as candi­
dates. Cf. Davis v. Bandemer, supra, at-- (POWELL, J., 
concurring in part and dissenting in part). Nonetheless, a 
reviewing court should be required to find more than simply 
that the minority group does not usually attain an undiluted 
measure of electoral success. The court must find that even 
substantial minority success will be highly infrequent under 
the challenged plan before it may conclude, on this basis 
alone, that the plan operates "to cancel out or minimize the 
voting strength of [the] racial grou[p]." White, 412 U. S., 
at 765. 

III 

Only a plurality of the Court joins Part III-C of JUSTICE 
BRENNAN's opinion, which addresses the validity of the sta­
tistical evidence on which the District Court relied in finding 
racially polarized voting in each of the challenged districts. 
Insofar as statistical evidence of divergent racial voting pat­
terns is admitted solely to establish that the minority group 
is politically cohesive and to assess its prospects for electoral 
success, I agree with the plurality that defendants cannot re­
but this showing by offering evidence that the divergent ra­
cial voting patterns may be explained in part by causes other 
than race, such as an underlying divergence in the interests 
of minority and white voters. I do not agree, however, that 
such evidence can never affect the overall vote dilution in­
quiry. Evidence that a candidate preferred by the minority 
group in a particular election was rejected by white voters 
for reasons other than those which made that candidate the 



18 THORNBURG v. GINGLES 

preferred choice of the minority group would seem clearly 
relevant in answering the question whether bloc voting by 
white voters will consistently defeat minority candidates. 
Such evidence would suggest that another candidate, equally 
preferred by the minority group, might be able to attract 
greater white support in future elections. 

I believe Congress also intended that explanations of the 
reasons why white voters rejected minority candidates would 
be probative of the likelihood that candidates elected without 
decisive minority support would be willing to take the minor­
ity's interests into account. In a community that is polarized 
along racial lines, racial hostility may bar these and other in­
direct avenues of political influence to a much greater extent 
than in a community where racial animosity is absent al­
though the interests of racial groups diverge. Indeed, the 
Senate Report clearly stated that one factor that could have 
probative value in § 2 cases was "whether there is a signifi­
cant lack of responsiveness on the part of elected officials to 
the particularized needs of the members of the minority 
group." S. Rep. 29. The overall vote dilution inquiry nei­
ther requires nor permits an arbitrary rule against consider­
ation of all evidence concerning voting preferences other than 
statistical evidence of racial voting patterns. Such a rule 
would give no effect whatever to the Senate Report's re­
peated emphasis on "intensive racial politics," on "racial 
political considerations," and on whether "racial politics ... 
dominate the electoral process" as one aspect of the "racial 
bloc voting" that Congress deemed relevant to showing a § 2 
violation. S. Rep. 33-34. Similarly, I agree with JuSTICE 
WHITE that the plurality's conclusion that the race of the can­
didate is always irrelevant in identifying racially polarized 
voting conflicts with Whitcomb and is not necessary to the 
disposition of this case. Ante, at -- (WHITE, J., concur­
ring in part and in the judgment and dissenting in part). 

In this case, as the Court grudgingly acknowledges, the 
District Court clearly erred in aggregating data from all of 

:..~' ,, 



THORNBURG v. GINGLES 19 

the challenged districts, and then relying on the fact that on 
average, 81.7% of white voters did not vote for any black 
candidate in the primary elections selected for study. Ante, 
at - - , n. 28. Although Senate District 22 encompasses 
House District 36, with that exception the districts at issue in 
this case are distributed throughout the State of North Caro­
lina. White calls for "an intensely local appraisal of the de­
sign and impact of the ... multimember district," 412 U. S., 
at 769-770, and racial voting statistics from one district are 
ordinarily irrelevant in assessing the totality of the circum­
stances in another district. In view of the specific evidence 
from each district that the District Court also considered, 
however, I cannot say that its conclusion that there was se­
vere racial bloc voting was clearly erroneous with regard to 
any of the challenged districts. Except in House District 23, 
where racial bloc voting did not prevent sustained and virtu­
ally proportional minority electoral success, I would accord­
ingly leave undisturbed the District Court's decision to give 
great weight to racial bloc voting in each of the challenged 
districts. 

IV 

Having made usual, roughly proportional success the sole 
focus of its vote dilution analysis, the Court goes on to hold 
that proof that an occasional minority candidate has been 
elected does not foreclose a § 2 claim. But JUSTICE BREN­
NAN, joined by JuSTICE WHITE, concludes that "persistent 
proportional representation" will foreclose a § 2 claim unless 
the plaintiffs prove that this "sustained success does not ac­
curately reflect the minority group's ability to elect its pre­
ferred representatives." Ante, at--. I agree with Jus­
TICE BRENNAN that consistent and sustained success by 
candidates preferred by minority voters is presumptively in­
consistent with the existence of a § 2 violation. Moreover, I 
agree that this case presents no occasion for determining 
what would constitute proof that such success did not accu-



20 THORNBURG v. GINGLES 

rately reflect the minority group's actual voting strength in a 
challenged district or districts. 

In my view, the District Court erred in assessing the ex­
tent of black electoral success in House District 39 and Sen­
ate District 22, as well as in House District 23, where the 
Court acknowledges error. As the evidence summarized by 
the Court in table form shows, ante, at--, App. B, the de­
gree of black electoral success differed widely in the seven 
originally contested districts. In House District 8 and Sen­
ate District 2, neither of which is contested in this Court, no 
black candidate had ever been elected to the offices in ques­
tion. In House District 21 and House District 36, the only 
instances of black electoral success came in the two most re­
cent elections, one of which took place during the pendency of 
this litigation. By contrast, in House District 39 and Senate 
District 22, black successes, although intermittent, dated 
back to 1974, and a black candidate had been elected in each 
of these districts in three of the last five elections. Finally, 
in House District 23 a black candidate had been elected in 
each of the last six elections. 

The District Court, drawing no distinctions among these 
districts for purposes of its findings, concluded that "[t]he 
overall results achieved to date at all levels of elective office 
are minimal in relation to the percentage of blacks in the total 
population." 590 F. Supp., at 367. The District Court 
clearly erred to the extent that it considered electoral success 
in the aggregate, rather than in each of the challenged dis­
tricts, since, as the Court states, "[t]he inquiry into the exist­
ence of vote dilution is district-specific." Ante, at --, 
n. 28. The Court asserts that the District Court was free to 
regard the results of the 1982 elections with suspicion and to 
decide "on the basis of all the relevant circumstances to ac­
cord greater weight to blacks' relative lack of success over 
the course of several recent elections," ante, at--, but the 
Court does not explain how this technique would apply in 
Senate District 22, where a black candidate was elected in 



THORNBURG v. GINGLES 21 

three consecutive elections from 1974 to 1978, but no black 
candidate was elected in 1982, or in House District 39, where 
black candidates were elected in 1974 and 1976 as well as in 
1982. Contrary to what the District Court thought, see 590 
F. Supp., at 367, these pre-1982 successes, which were pro­
portional or nearly proportional to black population in these 
three multimember districts, certainly lend some support for 
a finding that black voters in these districts enjoy an equal 
opportunity to participate in the political process and to elect 
representatives of their choice. 

Despite this error, I agree with the Court's conclusion 
that, except in House District 23, minority electoral success 
was not sufficiently frequent to compel a finding of equal 
opportunity to participate and elect. The District Court 
found that "in each of the challenged districts racial polariza­
tion in voting presently exists to a substantial or severe de­
gree, and ... in each district it presently operates to mini­
mize the voting strength of black voters." I d., at 372. I 
cannot say that this finding was clearly erroneous with re­
spect to House District 39 or Senate District 22, particularly 
when taken together with the District Court's findings con­
cerning the other Zimmer factors, and hence that court's ul­
timate conclusion of vote dilution in these districts is ade­
quately supported. 

This finding, however, is clearly erroneous with respect to 
House District 23. Blacks comprise 36.3% of the population 
in that district, and constitute 28.6% of the registered voters. 
In each of the six elections since 1970 one of the three repre­
sentatives from this district has been a black. There is no 
finding, or any reason even to suspect, that the successful 
black candidates in District 23 did not in fact represent the 
interests of black voters, and the District Court did not find 
that black success in previous elections was aberrant. 

Zimmers caveat against necessarily foreclosing a vote di­
lution claim on the basis of isolated black successes, 485 
F. 2d, at 1307; see S. Rep. 29, n. 115, cannot be pressed this 



22 THORNBURG v. GINGLES 

far. Indeed, the 23 Court of Appeals decisions on which the 
Senate Report relied, and which are the best evidence of the 
scope of this caveat, contain no example of minority electoral 
success that even"remotely approximates the consistent, dec-

. ade-long pattern in District 23. See, e. g., Turner v. 
McKeithen, 490 F. 2d 191 (CA5 1973) (no black candidates 
elected); Wallace v. House, 515 F. 2d 619 (CA5 1975), va­
cated on other grounds, 425 U. S. 947 (1976) (one black candi­
date elected). 

I do not propose that consistent and virtually proportional 
minority electoral success should always, as a matter of law, 
bar finding a § 2 violation. But, as a general rule, such suc­
cess is entitled to great weight in evaluating whether a chal­
lenged electoral mechanism has, on the totality of the circum­
stances, operated to deny black voters an equal opportunity 
to participate in the political process and to elect represent­
atives of their choice. With respect to House District 23, the 
District Court's failure to accord black electoral success such 
weight was clearly erroneous, and the District Court identi­
fied no reason for not giving this degree of success preclusive 
effect. Accordingly, I agree with JUSTICE BRENNAN that 
appellees failed to establish a violation of § 2 in District 23. 

v 
When members of a racial minority challenge a multi­

member district on the grounds that it dilutes their voting 
strength, I agree with the Court that they must show that 
they possess such strength and that the multimember district 
impairs it. A court must therefore appraise the minority 
group's undiluted voting strength in order to assess the 
effects of the multimember district. I would reserve the 
question of the proper method or methods for making this as­
sessment. But once such an assessment is made, in my view 
the evaluation of an alleged impairment of voting strength 
requires consideration of the minority group's access to the 
political processes generally, not solely consideration of the 



THORNBURG v. GINGLES 23 

chances that its preferred candidates will actually be elected. 
Proof that white voters withhold their support from minor­
ity-preferred candidates to an extent that consistently en­
sures their defeat is entitled to significant weight in plaintiffs' 
favor. However, if plaintiffs direct their proof solely to­
wards the minority group's prospects for electoral success, 
they must show that substantial minority success will be 
highly infrequent under the challenged plan in order to estab­
lish that the plan operates to "cancel out or minimize" their 
voting strength. White, 412 U. S., at 765. 

Compromise is essential to much if not most major federal 
legislation, and confidence that the federal courts will enforce 
such compromises is indispensable to their creation. I be­
lieve that the Court today strikes a different balance than 
Congress intended to when it codified the results test and 
disclaimed any right to proportional representation under § 2. 
For that reason, I join the Court's judgment but not its 
opinion. 





SUPREME COURT OF THE UNITED STATES 

No. 83-1968 

LACY H. THORNBURG, ET AL., APPELLANTS v. 
RALPH GINGLES ET AL. 

ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR 
THE EASTERN DISTRICT OF NORTH CAROLINA 

[June 30, 1986] 

JUSTICE STEVENS, with whom JUSTICE MARSHALL and 
JUSTICE BLACKMUN join, concurring in part and dissenting in 
part. 

In my opinion, the findings of the District Court, which the 
Court fairly summarizes, ante, at 4-8; 19-20 and n. 23; 26-27 
and nn. 28 and 29, adequately support the District Court's 
judgment concerning House District 23 as well as the balance 
of that judgment. 

I, of course, agree that the election of one black candidate 
in each election since 1972 provides significant support for 
the State's position. The notion that this evidence creates 
some sort of a conclusive, legal presumption, ante, at 43-44, 
is not, however, supported by the language of the statute or 
by its legislative history. 1 I therefore cannot agree with the 
Court's view that the District Court committed error by fail­
ing to apply a rule of law that emerges today without statu-

'See ante, at 42 ("Section 2 provides that '[t]he extent to which mem-
bers of a protected class have been elected to office .. . is one circumstance 
which may be considered.' 42 U. S. C. § 1973(b) .... However, the Sen-
ate Report expressly states that 'the election of a few minority candidates 
does not "necessarily foreclose the possibility of dilution of the black vote," 
noting that if it did, 'the possibility exists that the majority citizens might 
evade [§ 2] by manipulating the election of a "safe" minority candi­
date.' . . . The Senate Committee decided, instead, to 'require an inde­
pendent consideration of the record'") (internal citations omitted). 



2 THORNBURG v. GINGLES 

tory support. The evidence of candidate success in District 
23 is merely one part of an extremely large record which the 
District Court carefully considered before making its ulti­
mate findings of fact, all of which should be upheld under a 
normal application of the "clearly erroneous" standard that 
the Court traditionally applies. 2 

The Court identifies the reason why the success of one 
black candidate in the elections in 1978, 1980, and 1982 is not 
inconsistent with the District Court's ultimate finding con­
cerning House District 23.3 The fact that one black candi­
date was also elected in the 1972, 1974, and 1976 elections, 
ante, App. B, is not sufficient, in my opinion, to overcome the 
additional findings that apply to House District 23, as well as 
to ' other districts in the State for each of those years. The 
:Court·accurately summarizes those findings: 

"The.District Court in this case carefully considered the 
totality of the circumstances and found that in each dis­
trict racially polarized voting; the legacy of official dis­
crimination in voting matters, education, housing, em­
ployment, and health services; and the persistence of 
campaign appeals to racial prejudice acted in concert 
with the multimember districting scheme to impair the 
ability of geographically insular and politically cohesive 
groups of black voters to participate equally in the politi­
cal process and to elect candidates .of. their choice. It 
found that the success a few black candidates have en­
joyed in these districts is too recent, too limited, and, 
with regard to the 1982 elections, perhaps too aberra­
tional, to disprove its conclusion." Ante, at 46-47. 

To paraphrase the Court's conclusion about the other dis­
tricts, i bid., I cannot say that the District Court, composed of 

2 See ante, at 46 ("[T]he 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 endan­
gering the rule of law"). 

3 See ante, at 19-20 and n. 23; 27, n. 29; 42- 43. 



THORNBURG v. GINGLES 3 

local judges who are well-acquainted with the political reali­
ties of the State, clearly erred in concluding that use of a 
multimember electoral structure has caused black voters in 
House District 23 to have less opportunity than white voters 
to elect representatives of their choice. 4 Accordingly, I con­
cur in the Court's opinion except Part IV- B and except inso­
far as it explains why it reverses the judgment respecting 
House District 23. 

4 Even under the Court's analysis, the decision simply to reverse--with­
out a remand-is mystifying. It is also extremely unfair. First, the 
Court does not give respondents an opportunity to address the new legal 
standard that the Court finds decisive. Second, the Court does not even 
bother to explain the contours of that standard, and why it was not satis­
fied in this case. Cf. ante, at 44, n. 38 ("we have no occasion in this case to 
decide what types of special circumstances could satisfactorily demonstrate 
that sustained success does not accurately reflect the minority's ability to 
elect its preferred representatives"). Finally, though couched as a conclu­
sion about a "matter of law," id., at 44, the Court's abrupt entry of judg­
ment for petitioners on District 23 reflects an unwillingness to give the Dis­
trict Court the respect it is due, particularly when, as in this case, the 
District Court has a demonstrated knowledge and expertise of the entire 
context that Congress directed it to consider. 












	NAACP1016
	NAACP1017
	NAACP1018
	NAACP1019
	NAACP1020
	NAACP1021
	NAACP1022
	NAACP1023
	NAACP1024
	NAACP1025
	NAACP1026
	NAACP1027
	NAACP1028
	NAACP1029
	NAACP1030
	NAACP1031
	NAACP1032
	NAACP1033
	NAACP1034
	NAACP1035
	NAACP1036
	NAACP1037
	NAACP1038
	NAACP1039
	NAACP1040
	NAACP1041
	NAACP1042
	NAACP1043
	NAACP1044
	NAACP1045
	NAACP1046
	NAACP1047
	NAACP1048
	NAACP1049
	NAACP1050
	NAACP1051
	NAACP1052
	NAACP1053
	NAACP1054
	NAACP1055
	NAACP1056
	NAACP1057
	NAACP1058
	NAACP1059
	NAACP1060
	NAACP1061
	NAACP1062
	NAACP1063
	NAACP1064
	NAACP1065
	NAACP1066
	NAACP1067
	NAACP1068
	NAACP1069
	NAACP1070
	NAACP1071
	NAACP1072
	NAACP1073
	NAACP1074
	NAACP1075
	NAACP1076
	NAACP1077
	NAACP1078
	NAACP1079
	NAACP1080
	NAACP1081
	NAACP1082
	NAACP1083
	NAACP1084
	NAACP1085
	NAACP1086
	NAACP1087
	NAACP1088
	NAACP1089
	NAACP1090
	NAACP1091
	NAACP1092
	NAACP1093
	NAACP1094
	NAACP1095
	NAACP1096
	NAACP1097
	NAACP1098
	NAACP1099
	NAACP1100
	NAACP1101
	NAACP1102
	NAACP1103