Eugene Baten v. Henry McMaster ( 2020 )

PUBLISHED
UNITED STATES COURT OF APPEALS
FOR THE FOURTH CIRCUIT
No. 19-1297
EUGENE BATEN; CHESTER WILLS; CHARLETTE PLUMMER-WOOLEY;
BAKARI SELLERS; CORY C. ALPERT; BENJAMIN HORNE,
Plaintiffs - Appellants,
v.
HENRY MCMASTER, in his official capacity as Governor of the State of South
Carolina; MARK HAMMOND, in his official capacity as Secretary of the State of
South Carolina; SOUTH CAROLINA ELECTION COMMISSION; BILLY WAY,
JR., in his official capacity as a Chair of the Election Commission; MARK
BENSON, in his official capacity as a Commission Member of the Election
Commission; MARILYN BOWER; E. ALLEN DAWSON, in his official capacity
as a Commissioner Member of the Election Commission; NICOLE SPAIN WHITE,
in her official capacity as a Commission Member of the Election Commission,
Defendants - Appellees.
--------------------------------------
EDWARD FOLEY,
Amicus Curiae.
Appeal from the United States District Court for the District of South Carolina, at
Charleston. David C. Norton, District Judge. (2:18-cv-00510-DCN)
Argued: May 26, 2020 Decided: July 21, 2020
Amended: July 28, 2020
Before NIEMEYER, WYNN, and FLOYD, Circuit Judges.
Affirmed by published opinion. Judge Niemeyer wrote the opinion, in which Judge Floyd
joined. Judge Wynn wrote a dissenting opinion.
ARGUED: Jacob Max Rosen, MUNGER, TOLLES & OLSON, LLP, San Francisco,
California, for Appellants. Thomas Ashley Limehouse, Jr., OFFICE OF THE
GOVERNOR OF SOUTH CAROLINA, Columbia, South Carolina, for Appellees. ON
BRIEF: David Boies, Armonk, New York, James P. Denvir, III, Amy J. Mauser, Karen
L. Dunn, Lisa Barclay, Amy L. Neuhardt, Hamish P.M. Hume, Melissa Shube, BOIES,
SCHILLER & FLEXNER LLP, Washington, D.C.; Randall L. Allen, B. Parker Miller,
Max Marks, Cassandra K. Johnson, ALSTON & BIRD LLP, Atlanta, Georgia; Richard
Harpootlian, Christopher P. Kenney, RICHARD A. HARPOOTLIAN, PA, Columbia,
South Carolina, for Appellants. Vordman Carlisle Traywick, III, ROBINSON GRAY
STEPP & LAFFITTE, LLC, Columbia, South Carolina, for Appellees. Jeffrey I. Pasek,
Philadelphia, Pennsylvania, for Amicus Curiae.
2
NIEMEYER, Circuit Judge:
A group of South Carolina voters commenced this action to challenge the “winner-
take-all” aspect of South Carolina’s process for appointing its nine Electors to the Electoral
College.
Under South Carolina’s winner-take-all appointment process, every eligible voter
in South Carolina is given the right to cast a vote for candidates for President and Vice
President, and every such vote is given the same weight and dignity. The vote so cast is,
under South Carolina law, an indirect vote for a slate of nine Electors committed to a
presidential ticket, and the Electors then vote for those candidates in the Electoral College.
Thus, the candidates receiving the most votes secure the State’s appointment of Electors
committed to vote for them in the Electoral College. The losing candidates have no
Electors appointed to vote for them in the Electoral College, as the method for appointing
Electors is a unity method in which the entire slate of Electors is awarded to the winning
candidates.
In their complaint, the plaintiffs alleged that this winner-take-all process dilutes the
voting power of the State’s political minority and burdens their ability to advocate
effectively for their preferred political candidates, in violation of the First and Fourteenth
Amendments. They also alleged that the process prevents Black citizens from exercising
electoral power in presidential elections commensurate with the size of their population, in
violation of the Voting Rights Act (“VRA”).
3
The district court granted South Carolina’s motion to dismiss the complaint,
concluding that South Carolina’s winner-take-all process complies with the Constitution
and the requirements of the VRA. We agree and affirm.
I
The U.S. Constitution provides that the President and Vice President shall be elected
by a College * of “Electors” appointed by the States “in such Manner as the Legislature[s]
thereof may direct.” U.S. Const. art. II, § 1, cl. 2. The number of Electors allocated to
each State for appointment is equal to the “Number of Senators and Representatives to
which the State may be entitled in the Congress.” Id. Once appointed, the Electors are
directed to “meet in their respective states” on a day determined by Congress and vote “in
distinct ballots” for President and Vice President. Id. amend. XII. The votes are then
transmitted to Washington, D.C., where the President of the Senate counts them. See id.
“The person having the greatest number of votes for President, shall be the President, if
such number be a majority of the whole number of Electors appointed,” and the same is so
for the Vice President. Id.
In exercising their authority to appoint Electors, the States have, over the years,
adopted different methods. For example, in the first presidential election in 1788–89, the
legislatures in five States appointed the State’s Electors directly, and in other States, the
Electors were elected by the people in districts created by the State. Pennsylvania provided
*
The body of Electors is traditionally referred to as a “College,” although the word
does not appear in the Constitution. It does appear in the implementing statute. See 3
U.S.C. § 4.
4
for a statewide election of Electors — Electors “on a general ticket” pledged to a
presidential candidate. See McPherson v. Blacker, 146 U.S. 1, 29–30 (1892). In the
decades that followed, the States continued to exercise their appointment authority
variously, as their legislatures determined. Some legislatures continued to appoint Electors
directly; some authorized the voters to select a slate of Electors by a statewide election;
some authorized voters to select individual Electors by districts; and some created hybrid
systems. But eventually, as it became understood that a unified slate of Electors would
give the States the greatest influence in the Electoral College, States made the political
decision that the selection of a “general ticket” of Electors pledged to the winning candidate
— the winner-take-all approach — was advantageous. As of 1836, all States except South
Carolina appointed their Electors by statewide popular vote. See id. at 32. And following
the Civil War, South Carolina followed suit. At the present time, every State but Maine
and Nebraska awards all of its electoral votes to the presidential ticket that received a
plurality of the votes statewide. And in Maine and Nebraska, two Electors are selected by
statewide election and the remainder are selected by districts. See Chiafalo v. Washington,
140 S. Ct. 2316, 2321 n.1 (2020).
South Carolina’s current statutory scheme implements a rather typical winner-take-
all process. See Chiafalo, 140 S. Ct. at 2321–22 (describing in general terms the standard
process used by States to select Electors and to ensure their votes for pledged candidates).
Under its scheme, each political party recognized in South Carolina submits a list of Elector
candidates to the Secretary of State prior to Election Day. See S.C. Code Ann. § 7-19-70.
Those Elector candidates are required to “declare which candidate for president and vice-
5
president he will vote for if elected” and, if elected, are directed to vote in the Electoral
College “for the president and vice-president candidates for whom they declared.” Id. § 7-
19-80. The names of the Elector candidates submitted by the parties to the Secretary of
State, however, do not appear on the ballot. Instead, the voters are presented with “the
names of the candidates for President and Vice President” nominated by each party, and a
“vote for the candidates named on the ballot [constitutes] a vote for the electors of the party
by which those candidates were nominated.” Id. § 7-19-70. After the polls close, South
Carolina’s Secretary of State receives the vote tally and appoints the slate of Electors
submitted by the party whose presidential and vice-presidential candidates received the
most votes. Id. The appointed Electors are then required, on the day designated by
Congress, to cast their votes for those candidates. See id. § 7-19-80. This winner-take-all
process thus involves the statewide, indirect election of a slate of Electors pledged to the
candidates on the ballot who receive a plurality of the votes. The Electors so appointed
then vote for their candidates for President and Vice President at the Electoral College
election held at the time specified by Congress — currently, the first Monday after the
second Wednesday in the December following the election. See 3 U.S.C. § 7.
II
The six plaintiffs, three of whom are African-American, are voting-age residents of
various South Carolina counties who have voted for the Democratic presidential candidate
in past elections and plan to do so again in the future. They commenced this action to
challenge “the decision of South Carolina to award and select Electors on a [winner-take-
6
all] basis,” and they named as defendants South Carolina’s Governor, Secretary of State,
and Election Commission, including its members — all sued in their official capacity and
collectively referred to herein as “South Carolina.” The complaint explicitly stated that it
was not filed to challenge the Electoral College, “which is mandated by the Constitution,”
but rather to challenge the South Carolina statutory scheme adopted to select electors to
the College on a winner-take-all basis. The plaintiffs alleged that “the political party of the
leading candidate among South Carolina’s voters selects every Elector, with a vote of every
other South Carolina citizen rendered meaningless by receiving no Elector directly or
through a political party.” (Emphasis added). They explained that in the 2016 election,
Donald Trump received roughly 55% of the popular vote in South Carolina but yet received
all nine of the State’s electoral votes, whereas Hillary Clinton received roughly 41% of the
popular vote “but received none of the electoral votes from South Carolina.” The
complaint also asserted that under the scheme, African-American voters, who “represent
over a quarter of the voting age population of the [S]tate,” are effectively disenfranchised.
“In each of the last ten presidential elections the candidate who won South Carolina and
received all of South Carolina’s Electors has been a Republican and has not been the
preferred candidate for African-American voters.”
Thus, as framed in the complaint, with use of the winner-take-all process, the votes
of those who did not vote for the winning candidate were “cancelled when the final direct
election for President [took] place,” including “as many as 855,373 South Carolina citizens
who voted for Hillary Clinton in 2016.” For that reason, the plaintiffs claimed that the
7
process violates the Fourteenth Amendment, the First Amendment, and § 2 of the VRA.
With respect to the alleged Fourteenth Amendment violation, the complaint asserted:
[The winner-take-all] scheme violates the Fourteenth Amendment because it
counts votes for a losing presidential candidate in South Carolina only to
discard them in determining Electors who cast votes directly for the
presidency. Put differently, the [winner-take-all] system unconstitutionally
magnifies the votes of a bare plurality of voters by translating those votes
into an entire slate of presidential Electors, all of whom support the nominee
of a single political party — while, at the same time, the votes cast for all
other candidates are given no effect.
This, according to the plaintiffs, violates the principle of “one person, one vote.” With
respect to the alleged First Amendment violation, the complaint asserted that the system
“burdens . . . the right of association and . . . the right to have a voice in presidential
elections through casting a vote.” And finally, with respect to the alleged VRA violation,
the complaint asserted that, “[d]espite the fact that South Carolina has nine Electors, and
African Americans represent over a quarter of the voting age population of the state, the
[winner-take-all] system allows white voters to usually — if not always — defeat all
Electors slated for African-American preferred candidates.”
For relief, the complaint sought a declaratory judgment that South Carolina’s
winner-take-all process violates the referenced provisions and an injunction prohibiting the
use of the process in the future.
The district court granted South Carolina’s motion to dismiss the complaint,
concluding that South Carolina’s winner-take-all system “complies with equal protection
because it does not inherently favor or disfavor a particular group of voters.” Baten v.
McMaster, 374 F. Supp. 3d 563, 569 (D.S.C. 2019) (quoting Lyman v. Baker, 352 F. Supp.
83d 81, 89 (D. Mass. 2018)). The court likewise rejected the plaintiffs’ First Amendment
claim, as the plaintiffs had not “sufficiently alleged that the rights of voters to associate in
any manner or to engage in any political activity or association ha[d] actually been
burdened.” Id. at 570. And finally, the court dismissed the plaintiffs’ VRA claim, noting
that the winner-take-all system “does not mean that the political process is not equally open
to participation by Democratic voters, whether they be African-American or of another
race.” Id. at 571 (cleaned up).
From the district court’s judgment dated March 11, 2019, the plaintiffs filed this
appeal. South Carolina filed a motion to dismiss the appeal for lack of subject matter
jurisdiction, arguing (1) that this case implicates non-justiciable political questions and (2)
that the plaintiffs lack Article III standing. We deferred ruling on that motion pending
briefing on the merits and oral argument.
III
At the threshold, we address South Carolina’s motion to dismiss this appeal for lack
of subject matter jurisdiction. It contends that the plaintiffs’ claims present non-justiciable
political questions, as “the state legislatures’ power to choose the manner of appointing
presidential electors is plenary,” and “partisan proportionality claims” are non-justiciable
under the Supreme Court’s recent decision in Rucho v. Common Cause, 139 S. Ct. 2484(2019). It also contends that the plaintiffs’ claims are simply “generalized partisan
grievances” and, therefore, that the plaintiffs lack Article III standing.
9
On the issue of justiciability, it is true that Article II of the Constitution gives state
legislatures the power to appoint Electors in the manner they see fit. But it is also well
settled that Article II does not vest the States with unreviewable authority. See, e.g., Bush
v. Gore, 531 U.S. 98, 104 (2000) (per curiam) (recognizing that although “the state
legislature’s power to select the manner for appointing electors is plenary,” “[w]hen the
state legislature vests the right to vote for President in its people, the right to vote as the
legislature has prescribed is fundamental”); see also Chiafalo, 140 S. Ct. at 2324 & n.4
(explaining that “Article II, § 1’s appointments power gives the States far-reaching
authority over presidential electors, absent some other constitutional constraint,” such as
the Equal Protection Clause (emphasis added)). Moreover, the Supreme Court has
confirmed that whether a state legislature has exercised its delegated power to appoint
Electors in a way that violates another provision of the Constitution presents a justiciable
question. See, e.g., McPherson, 146 U.S. at 23–24 (“It is argued that the subject-matter of
the controversy is not of judicial cognizance, because it is said that all questions connected
with the election of a presidential elector are political in their nature. . . . But the judicial
power of the United States extends to all cases in law or equity arising under the
constitution and laws of the United States, and this is a case so arising, since the validity
of the state law was drawn in question as repugnant to such constitution and laws . . . .”);
Williams v. Rhodes, 393 U.S. 23, 28 (1968) (rejecting Ohio’s argument that “the political-
question doctrine precludes judicial consideration” of challenges to its laws regulating
access to the state ballot to choose electors, concluding unequivocally that these types of
cases “do raise a justiciable controversy under the Constitution and cannot be relegated to
10
the political arena”). We do not believe that the Court’s subsequent decision in Rucho
undermined these principles, as South Carolina argues.
In Rucho, the Supreme Court held that partisan gerrymandering claims present
nonjusticiable political questions. But it recognized that partisan gerrymandering claims
were distinct from other similar claims, noting that “[p]artisan gerrymandering claims have
proved far more difficult to adjudicate” than other challenges to the electoral system, such
as those alleging racial gerrymandering or violations of the one-person, one-vote principle.
Rucho, 139 S. Ct. at 2497. We thus agree with those courts that have considered the
question before us and determined that challenges to winner-take-all systems for
appointing presidential Electors are indeed justiciable. See Lyman v. Baker, 954 F.3d 351,
378 n.16 (1st Cir. 2020); Williams v. Va. State Bd. of Elections, 288 F. Supp. 622, 625
(E.D. Va. 1968) (holding that the court “ha[d] jurisdiction of the complaint,” which
challenged Virginia’s use of the winner-take-all system to appoint presidential Electors),
aff’d per curiam, 393 U.S. 320 (mem.), reh’g denied, 393 U.S. 1112 (1969).
With respect to Article III standing, we conclude that the plaintiffs adequately
alleged that they “had the requisite stake in the outcome [of the case] when the suit was
filed.” Davis v. FEC, 554 U.S. 724, 734 (2008). They were required to allege (1) that they
suffered a concrete and particularized “injury in fact” that was more than merely
hypothetical or conjectural; (2) that the injury is “fairly traceable to the challenged action
of the defendant”; and (3) that it is likely “that the injury will be redressed by a favorable
decision.” Lujan v. Defs. of Wildlife, 504 U.S. 555, 560–61 (1992) (cleaned up).
11
In their complaint, the plaintiffs have fairly satisfied these elements. First, they
alleged that they are voters whose votes for Democratic presidential candidates were, in
effect, discarded under South Carolina’s winner-take-all process. They argued that,
because all of South Carolina’s electors were pledged to vote for the plurality winner, their
preferred candidates received zero votes in the Electoral College. Thus, the system
“disadvantage[d] [them] as individuals.” Gill v. Whitford, 138 S. Ct. 1916, 1920–21 (2018)
(quoting Baker v. Carr, 369 U.S. 186, 206 (1962)) (contrasting the individual harm felt by
a voter who casts his ballot in a gerrymandered district with the “generalized grievance” of
one who disapproves of gerrymandering in his state but does not live in a gerrymandered
district). This is the type of concrete, particularized injury that Article III contemplates.
See Gray v. Sanders, 372 U.S. 368, 375 (1963) (“[A]ny person whose right to vote is
impaired has standing to sue” (citations omitted)). Second, the plaintiffs alleged that the
injury is caused by the winner-take-all system, which, even though enacted by the State’s
legislature, was implemented and enforced by the state officials who are named as
defendants. And third, they claimed that it is within the judicial power to enjoin the use of
an unlawful system, even though a court could not order South Carolina specifically to
adopt the proportional system of appointing Electors, as the plaintiffs might prefer. In
short, the plaintiffs alleged that they suffered individualized injuries that are traceable to
South Carolina’s implementation of the challenged winner-take-all system and that the
injuries can be redressed by a court order enjoining use of the system.
We conclude therefore that we have subject matter jurisdiction to address the merits
of the plaintiffs’ claims on appeal.
12
IV
For their main argument on appeal, the plaintiffs contend that South Carolina’s
winner-take-all method of selecting presidential Electors violates the Equal Protection
Clause. They explain that at the first stage of the two-step presidential election process,
use of winner-take-all “dilutes votes” by not giving a voice to the political minority in
selecting Electors and, at the second stage, the process “discards” the minority votes
because only the Electors chosen by the plurality cast votes in the Electoral College. As
they summarize,
By awarding all of a State’s Electors to whoever wins a plurality of the vote
at the first stage, South Carolina’s [winner-take-all] rules dilute and discard
minority votes in two ways. It dilutes votes for the Electors themselves, using
an at-large election for nine Electors to ensure minority voters never have
any representation in that delegation. And it discards their votes for president
at the second stage, ensuring that only Electors selected by the plurality can
ever affect the presidential vote.
To support their argument that the first stage as implemented in South Carolina is
unconstitutional, the plaintiffs rely on the Supreme Court’s decision in White v. Regester,
412 U.S. 755 (1973), which, they argue, holds that “states may not use at-large, slate
elections for multi-member bodies to ensure minority voters receive no representatives in
those bodies.” And to support their argument that the second stage as implemented in
South Carolina also violates the Constitution, they rely on Gray v. Sanders, which, they
argue, supports their position, noting that South Carolina’s use of winner-take-all to award
all of its Electoral votes to the plurality winner, impermissibly “ensures that the plaintiffs’
13
‘votes for a different candidate are worth nothing and counted only for the purpose of being
discarded.’” (cleaned up) (quoting Gray, 372 U.S. 381 n.12).
The untoward result, according to the plaintiffs, is that the system improperly
magnifies the influence of the political plurality, thereby undermining the principle of one-
person, one-vote, in violation of the Equal Protection Clause, which safeguards the “equal
weight accorded to each vote and the equal dignity owed to each voter.” Bush, 531 U.S.
at 104.
In response, South Carolina contends that the plaintiffs’ argument “ignores
longstanding electoral practice and is foreclosed by binding Supreme Court precedent” in
Williams v. Virginia State Board of Elections, 393 U.S. 320 (1969), which summarily
affirmed a district court’s rejection of a constitutional challenge to Virginia’s use of the
winner-take-all approach to select presidential Electors. It also argues that even if the case
is not controlled by Williams, the winner-take-all system “comports with the Equal
Protection Clause because it gives all who participate an equal vote and does not inherently
favor or disfavor any particular group of voters.”
The Electoral College system established in Article II and the Twelfth Amendment
of the Constitution for the election of the President and Vice President manifests a complex
and carefully balanced division of power and interests between the national government
and state governments; between the large States and the smaller States; and between
representational democracy and direct democracy. And it resulted from a lengthy debate
of various proposals at the Constitutional Convention. See, e.g., Chiafalo, 140 S. Ct. at
2320 (quoting a delegate to the Convention who remarked that the issue of determining the
14
method for presidential selection was “the most difficult of all [that] we have had to
decide”). Among the various methods considered by the delegates at that Convention were
a nationwide popular vote and an election by Congress. Ultimately, however, they settled
on a hybrid method involving nation-based and state-based aspects. Article II thus
provides that the President and Vice President are to be elected by Electors appointed by
the States; that the number of Electors to be appointed from each State is to equal the
number of Senators and House members representing the State in the Congress; and that
the States’ legislatures are to determine the manner of appointing the Electors.
The subtlety of this structure was eloquently described by James Madison in The
Federalist. He explained that the House of Representatives derives its power directly from
the people; that the Senate derives its powers from the States as “political and coequal
societies” and therefore indirectly from the people; and the presidency derives its power
from a “compound source,” through the Electoral College. See The Federalist No. 39, at
197 (George W. Carey & James McClellan eds. 1990) (emphases added). Madison
continued, “[the Electoral College] appears [in the Constitution] to be of a mixed character
presenting at least as many federal as national features.” Id. Thus, in creating an Electoral
College comprising Electors equal to the number of Senators and Representatives of each
State, the Constitution takes into account the States as political and coequal societies by
giving each two Electors simply by virtue of their statehood, and, at the same time, accounts
for population of each State by giving each a number of Electors directly related to the
State’s House members. And while the Constitution creates this body of Electors to serve
the national function of electing a President and Vice President, it directs that the States
15
appoint the Electors in the manner determined by their legislatures. At bottom, the system
reflects a considered balance between national and state power.
The plaintiffs’ arguments are narrowly focused on two distinct aspects of this
structure. They argue first that South Carolina, in selecting Electors, unconstitutionally
conducts a statewide winner-take-all election by selecting a “general ticket” of Electors,
all of whom are pledged to their party’s candidates. In this manner, they argue, South
Carolina fails to give effect to the votes of its citizens who did not cast votes with the
plurality. Second, they argue, when the Electors so selected uniformly vote for the
candidate of their party, the system in effect “discards” the votes cast by the minority.
As to the plaintiffs’ first argument, it cannot be disputed that each State has plenary
authority to determine through its legislature how to appoint Electors. And the power
includes the authority to require Electors to vote for the presidential ticket that received the
plurality of votes in the State. As the Supreme Court noted recently,
Article II, § 1’s appointments power gives the States far-reaching authority
over presidential electors. . . . [A] State can insist (as Ray [v. Blair, 343 U.S.
214 (1952)] allowed) that the elector pledge to cast his Electoral College
ballot for his party’s presidential nominee, thus tracking the State’s popular
vote.
Chiafalo, 140 S. Ct. at 2324 (emphasis added). Thus, nearly every State, including South
Carolina, has validly exercised its authority in adopting a system for appointment of
Electors that involves a statewide election for a slate of electors pledged to vote for the
presidential ticket that wins the plurality vote — the very system that the plaintiffs contend
is constitutionally barred because it fails to give effect to votes cast for other candidates.
It is plaintiffs’ position, nonetheless, that a State may not adopt, as a political calculation,
16
a plan that maximizes the State’s influence at the Electoral College election, as the winner-
take-all plan does. And if the plaintiffs are correct, the only permissible methods of
appointment would necessarily incorporate proportional or district-level allocation,
methods that are not prescribed by the Constitution. The plaintiffs argue that only in this
way may the State “give effect” to votes cast for the losing candidates.
The plaintiffs’ argument, however, runs headlong into the fundamental democratic
principle that the one who receives the most votes wins, and the others lose, thus leaving
them with no voice. While this argument is obviously untenable, it does not, of course,
follow that an electoral mechanism can trample the fundamental electoral requirements
that everyone entitled to vote be given the vote and that each vote be given equal weight.
See Bush, 531 U.S. at 104–05. But the plaintiffs have not made the case that those
fundamental requirements are not met by South Carolina’s chosen system. Despite the
plaintiffs’ argument to the contrary, no vote in the South Carolina system is diluted. Every
qualified person gets one vote and each vote is counted equally in determining the final
tally. In the end, the presidential ticket that receives the most votes wins. That the system
results in both winners and losers is inherent in our electoral process and does not give rise
to a constitutional violation. This was precisely the position articulated by the three-judge
district court in Williams while rejecting a challenge to the winner-take-all system, which
we find persuasive:
[I]t is difficult to equate the deprivations imposed by the unit [winner-take-
all] rule with the denial of privileges outlawed by the one-person, one-vote
doctrine or banned by Constitutional mandates of protection. In the selection
of electors the rule does not in any way denigrate the power of one citizen’s
ballot and heighten the influence of another’s vote. Admittedly, once the
17
electoral slate is chosen, it speaks only for the element with the largest
number of votes. This in a sense is discrimination against the minority
voters, but in a democratic society the majority must rule, unless the
discrimination is invidious. No such evil has been made manifest here.
Every citizen is offered equal suffrage and no deprivation of the franchise is
suffered by anyone.
Id. Recognizing that the winner-take-all method was merely an example of the “unit rule,”
the Williams district court stated that “nothing in the unit rule [is] offensive to the
Constitution.” Id. And the Supreme Court summarily affirmed the district court’s ruling
on appeal. Williams, 393 U.S. 320, reh’g denied, 393 U.S. 1112 (1969). To be sure, the
Supreme Court’s summary affirmance of the district court’s decision in Williams “affirm[s]
the judgment but not necessarily the reasoning by which it was reached,” Mandel v.
Bradley, 432 U.S. 173, 176 (1977) (per curiam) (quoting Fusari v. Steinberg, 419 U.S. 379,
391–92 (1975) (Burger, C.J., concurring )), but it does “prevent lower courts from coming
to opposite conclusions on the precise issues presented and necessarily decided by those
actions,” id.The plaintiffs contend nonetheless that the Supreme Court’s subsequent decision in
White v. Regester precludes use of an at-large, statewide election that dilutes minority
votes. In White, the Court held that the reapportionment plan for the Texas House of
Representatives, which included both single-member districts and multi-member districts,
was, as a whole, not invidiously discriminatory. 412 U.S. at 764. But it did affirm the
district court’s finding that in two multi-member districts, racial minority groups were
effectively excluded from the political process to the extent that they were unable to elect
representatives. Id. at 765. It also affirmed the district court’s judgment to redraw those
18
multi-member districts as single-member districts. Id. The Court’s justification for
affirming these aspects was based on the extensive factual findings made by the district
court of invidious discrimination against the minorities in those multi-member districts.
Thus, the Supreme Court’s holding is limited to where “multimember districts are being
used invidiously to cancel out or minimize the voting strength of racial groups,” id.,requiring a showing that members of minority groups have, as a result of a districting plan,
“less opportunity than [do] other residents in the district to participate in the political
processes and to elect legislators of their choice,” id. at 766 (emphasis added).
Thus understood, White’s holding does nothing to undermine the reasoning of the
district court in Williams, as White addresses the dilutive effects of creating multi-member
legislative districts and not of appointing Electors as a slate. Moreover, the plaintiffs in
this case have not purported to claim invidious discrimination on the part of South Carolina.
Indeed, they seem to agree that all persons entitled to vote have had the opportunity to vote
and that their votes have been counted with equal weight, regardless of political affiliation.
The essential problem that they identify is one that we cannot remedy — the fact that they
did not have enough votes to achieve the outcome they desired. As the Supreme Court has
observed, “we have not yet deemed it a denial of equal protection to deny legislative seats
to losing candidates, even in those so-called ‘safe’ districts where the same party wins year
after year.” Whitcomb v. Chavis, 403 U.S. 124, 153 (1971).
The plaintiffs also argue that their votes are impermissibly “discarded” prior to the
second stage of the electoral process, when the Electors cast the votes for the presidential
candidates chosen by the plurality, as no Electors vote in accordance with the political
19
minority. According to the plaintiffs, the Supreme Court’s decision in Gray v. Sanders
precludes such “discarding” of votes after the first stage of a two-stage election. In Gray,
a Georgia voter challenged the method by which Georgia’s Democratic party tabulated
votes to determine the primary winner in statewide elections. See 372 U.S. at 370. The
Georgia system allocated to each county a certain number of “units,” all of which were
awarded to the popular vote winner of that county. Because of the way county units were
allocated, small rural counties had disproportionate influence compared to larger or urban
counties; “counties having population of one-third of the total in the state [had] a clear
majority of county units.” Id. at 373. Following an election, both the popular vote and the
county unit vote were tabulated, and the candidate receiving a majority of both secured the
nomination without a runoff. See id. at 372. If the tabulation of the popular vote and the
county unit vote were split, the candidate with the greatest number of county unit votes in
a runoff election would prevail. See id. The Court held that Georgia’s system involved
impermissible differential weighing of votes based on voter location. Id. at 379–81. It also
suggested, in a footnote, that using a winner-take-all approach to allocate county units was
impermissible even if the units were allocated proportionately by population:
The county unit system . . . would allow the candidate winning the popular
vote in the county to have the entire unit vote of that county. Hence the
weighting of votes would continue, even if unit votes were allocated strictly
in proportion to population. Thus if a candidate won 6,000 of 10,000 votes
in a particular county, he would get the entire unit vote, the 4,000 other votes
for a different candidate being worth nothing and being counted only for the
purpose of being discarded.
Id. at 381 n.12.
20
The plaintiffs argue that the Gray Court’s language in footnote 12 reveals the
constitutional flaw in South Carolina’s winner-take-all system. The Georgia
circumstances, however, are materially distinguishable from those before us. First, in
Gray, using winner-take-all at the county level had the potential to change the outcome of
a statewide election by discarding minority votes in each county, possibly leading to the
nomination at the second stage of a political representative that failed to secure even
plurality support in the State in lieu of one who had such support. But in the presidential
context, there is no statewide elected seat up for grabs — the second stage of the election
occurs at the national level, and the two-step system is dictated by the Constitution. While
using winner-take-all at the State level leaves open the possibility that a President could,
on a nationwide basis, be elected despite losing the national popular vote, this very
possibility is baked into the Constitution. Indeed, the Gray Court was careful to recognize
this distinction:
We think the analogies to the electoral college, to districting and redistricting,
and to other phases of the problems of representation in state or federal
legislatures or conventions are inapposite. The inclusion of the electoral
college in the Constitution, as the result of specific historical concerns,
validated the collegiate principle despite its inherent numerical inequality,
but implied nothing about the use of an analogous system by a State in a
statewide election. No such specific accommodation of the latter was ever
undertaken, and therefore no validation of its numerical inequality ensued.
372 U.S. at 378 (footnotes omitted). Thus, the Gray Court recognized that the type of
“numerical inequality” that may render a state electoral process unconstitutional is a feature
— not a flaw — of the Electoral College system.
21
Moreover, the central concern in Gray — the differential treatment of votes
depending on the county in which they were cast — is not at issue here. All votes cast in
presidential elections in South Carolina are treated the same, and the candidate with the
most support across the State gets the State’s allocation of electoral votes in the Electoral
College. Thus, there is no risk, as there was in Gray, that votes are treated differently based
on geography.
The plaintiffs’ criticism of the unitary system — where a slate of Electors from a
State votes for a single ticket — is moreover dubious in light of the fact that the
Constitution itself explicitly embraces such an approach, albeit at a different stage of the
electoral process. In addressing how a tie vote in the Electoral College is to be resolved,
the Constitution provides that the election of the President is then committed to the House
of Representatives, and in carrying out the election, “the votes [in the House] shall be taken
by states, the representation from each state having one vote.” U.S. Const. amend. XII
(emphasis added). Of course, the constitutionally prescribed method of tie-breaking
“dilutes” or “discards” the vote of the minority House members from each State, in effect
silencing the minority to allow each State to speak with one voice — the very process that
the plaintiffs decry on the state level.
At bottom, South Carolina’s winner-take-all system “does not treat any particular
group of [voters in the State] differently at all — it does not inherently favor or disfavor
voters from any particular group (political or otherwise).” Lyman, 954 F.3d at 371
(affirming dismissal of the complaint in a parallel case brought by Republican voters in
Massachusetts). To be sure, when the plurality wins, it has the effect of rejecting the
22
outcome sought by voters supporting minority parties. But that is the reality of any
democratic system. Absent some invidious discrimination that infects the process, it is
difficult to comprehend a challenge to the various roles exercised in the selection of a
President that does not also challenge the Constitution itself. We affirm the district court’s
conclusion that plaintiffs did not state a claim for relief under the Equal Protection Clause.
V
The plaintiffs next contend that the winner-take-all method employed by South
Carolina burdens their right to freedom of association, as protected by the First Amendment
and applicable to the States through the Fourteenth Amendment. In their view, the winner-
take-all system “burdens Plaintiffs’ ability to associate with like-minded voters — all of
whom know that any such association, fundraising, or activities would be functionally
useless.” (Citing Gill, 138 S. Ct. at 1938 (Kagan, J., concurring)).
South Carolina maintains, as the district court concluded, that this argument
“conflates a diminishing motivation to participate with a severe burden on the actual ability
of people to participate in the voting process.” (Quoting Baten, 374 F. Supp. 3d at 570).
According to South Carolina, the plaintiffs have not alleged any actual burden, and merely
“losing an election does not give rise to an unconstitutional infringement upon the
associational rights of the losing candidate’s supporters.”
The Supreme Court has indeed recognized that “[t]he freedom of association
protected by the First and Fourteenth Amendments includes partisan political
organization. . . . The right to associate with the political party of one’s choice is an integral
23
part of this basic constitutional freedom.” Tashjian v. Republican Party of Ct., 479 U.S.
208, 214 (1986) (cleaned up). Thus, States are prohibited from unduly interfering with a
political party’s ability to encourage participation, see id. at 217, or from placing
unnecessary and unequal burdens on a political party seeking a place on the ballot, see
Rhodes, 393 U.S. at 30–31.
The plaintiffs’ claim of interference, however, is far more tenuous than those
recognized as First Amendment violations. They do not claim that South Carolina
prevented their preferred candidates from appearing on the ballot or interfered with their
ability to convince potential voters of their candidates’ advantages. Rather, the plaintiffs
contend that their inability to secure the selection of any Democratic-affiliated Electors in
presidential elections dampens enthusiasm for Democratic politics in South Carolina and
thereby burdens their right to associate freely with the political party of their choice. But
this effect stems not from any action by South Carolina to inhibit Democratic participation
in presidential elections, but rather from the Democrats’ inability to muster a majority of
South Carolina’s votes. And though the First Amendment protects the right to associate
for the purposes of advocacy, it “provides no guarantee that a speech will persuade or that
advocacy will be effective.” Smith v. Ark. State Highway Emp., Local 1315, 441 U.S. 463,
465 (1979) (citation omitted). In other words, “[t]hat one’s candidate of choice does not
prevail at the ballot box simply does not translate into an associational rights violation.”
Lyman, 954 F.3d. at 377. The winner-take-all system may “raise[] the stakes of victory,”
but it does not interfere with the plaintiffs’ opportunity to associate for the purposes of
advocating for such victory. Id.
24
Similarly, the plaintiffs’ suggestion that the winner-take-all system serves as a
disincentive for political candidates to campaign in South Carolina, thereby impeding their
ability to participate effectively in the political process, is too tenuous to support their
freedom of association claim. The First Amendment guarantees the right to associate with
the political party of one’s choice to participate in the political process; it does not
guarantee to the residents of South Carolina attention from candidates for nationwide
office. Moreover, many considerations other than the likelihood that the candidate will
receive needed Electoral College votes from a given State influence a presidential
candidate’s campaign event route, and therefore it is simply impossible to predict whether
presidential candidates would be more likely to campaign in South Carolina if the winner-
take-all system were abandoned. Indeed, if, as the plaintiffs maintain, a reliable portion of
the eligible voting population in the State consistently votes for Democratic candidates and
would continue to do so if South Carolina adopted a proportional system, such candidates
might nonetheless make the strategic assessment that those votes were “safe” even without
campaign stops in the State. At bottom, this argument does not support a cognizable First
Amendment claim.
We thus affirm the district court’s order also dismissing the plaintiffs’ freedom of
association claim.
VI
Finally, the plaintiffs contend that the district court erred in dismissing their claim
under § 2 of the VRA, which prohibits standards, practices, or procedures imposed “in a
25
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.” 52 U.S.C. § 10301(a). A violation of § 2 is
established if, “based on the totality of circumstances, it is shown that the political
processes leading to nomination or election in the State or political subdivision are not
equally open to participation by members of a class of citizens protected by subsection (a)
in that its members have less opportunity than other members of the electorate to
participate in the political process and to elect representatives of their choice.” 52 U.S.C.
§ 10301(b) (emphasis added). The Act specifies that it does not “establish[] a right to have
members of a protected class elected in numbers equal to their proportion in the
population.” Id.The plaintiffs maintain that “the vote dilution caused by South Carolina’s [winner-
take-all] rules is a mathematical certainty,” arguing that the process prevents Black voters
from choosing their preferred electors. Given the demographic breakdown of South
Carolina’s electorate and the partisan preferences of Black voters, the plaintiffs maintain
that “South Carolina’s [B]lack voters would be able to appoint two electors with no help
from [W]hite voters if they had ‘the opportunity to exercise an electoral power that is
commensurate with [their] population in the relevant jurisdiction.’” (Quoting Hall v.
Virginia, 385 F.3d 421, 429 (4th Cir. 2004)).
South Carolina contends simply that “the ultimate right of § 2 [of the VRA] is
equality of opportunity, not a guarantee of electoral success for minority preferred
candidates of whatever race,” (emphasis added) (quoting Johnson v. DeGrandy, 512 U.S.
997, 1014 n.11 (1994)), and that Black voters in South Carolina have been unsuccessful in
26
having their preferred electors appointed simply because they vote for the less popular
political party, not because minorities are excluded from the political process.
As the district court correctly stated, plaintiffs seeking to bring a claim under § 2 of
the VRA to challenge the dilutive effects of a multi-member district must be able to show
that (1) “the minority group . . . is sufficiently large and geographically compact to
constitute a majority in a single-member district;” (2) “the minority group . . . is politically
cohesive;” and (3) “the [W]hite majority votes sufficiently as a bloc to enable it . . . usually
to defeat the minority’s preferred candidate.” Thornburg v. Gingles, 478 U.S. 30, 50–51
(1986). But these requirements do not map onto the type of challenge that the plaintiffs
have mounted here. Gingles involved a challenge to the use of multimember districts in
North Carolina’s legislative apportionment, which the plaintiffs argued diluted minority
voting blocs that would form “effective voting majorities in single-member districts.” Id.
at 38. Whereas, here, by contrast, there is no alternative of a “single-member district”
because the selection of electors in South Carolina does not involve districts at all. While
the plaintiffs indeed allege that Black voters in South Carolina are a minority group
“sufficiently large and geographically compact to constitute a majority in a single-member
district,” they fail to address what this means in the context of a statewide election. The
relevant geographic area for the selection of presidential Electors is the entire State, and
Black voters do not constitute a majority statewide. Moreover, even though the plaintiffs
assert that the White majority votes sufficiently as a bloc to enable it to defeat the minority-
preferred candidate, they elide the distinction between the candidates the minorities prefer
— i.e. the Democratic presidential and vice-presidential candidates — and the “candidates”
27
that, in their view, they have sufficient political power to elect — i.e., two out of nine of
the presidential Electors. Section 2 refers to the election of “representatives,” not the
appointment of Electors. And while it is true that White voters may have, in recent decades,
preferred Republican candidates and voted accordingly, Black voters are not sufficiently
numerous to change the outcome. This does not mean, however, that Black voters, who
tend to prefer Democratic candidates, “have less opportunity than other members of the
electorate to participate in the political process and to elect representatives of their choice.”
52 U.S.C. § 10301(b) (emphasis added), as required to establish a violation of the VRA.
We therefore affirm the district court’s order dismissing the plaintiffs’ VRA claim.
* * *
In sum, we deny South Carolina’s motion to dismiss the appeal, and we affirm the
judgment of the district court.
AFFIRMED
28
WYNN, Circuit Judge, dissenting 1:
This matter arises from one simple yet remarkable fact: In every presidential
election since 1980, the votes by South Carolinians for Democratic presidential candidates
have been counted only for the purpose of being discarded because, despite representing a
significant share of the state’s total votes, South Carolinians’ votes for Democratic
candidates have translated into no votes in the Electoral College. Incredibly, South
Carolina’s state-wide, winner-take-all method of selecting presidential electors has
rendered more than 4 million votes for Democratic candidates over the last five presidential
elections worthless.
So, six Democratic voters from South Carolina brought this case alleging that this
system unconstitutionally dilutes and discards their votes. The majority opinion affirms the
district court in dismissing the case, putting Plaintiffs’ claims to bed at an early stage.
But Plaintiffs deserve more than this short shrift dismissal of their claims. Viewing
Plaintiffs’ allegations with an open mind and accepting them as true, as we are required to
do, Plaintiffs have stated plausible violations of the First and Fourteenth Amendments and
the Voting Rights Act. In essence, they deserve, at least, an opportunity to be heard. With
great respect for the differing view of my colleagues, I must dissent.
Having said that, the majority opinion must not be read as deterring future plaintiffs
from questioning longstanding election practices that run counter to this country’s
1
I concur in the majority’s denial of South Carolina’s motion to dismiss the appeal for lack
of subject matter jurisdiction.
foundational principles of democracy. At best, the majority opinion reflects a limited view
of the judiciary’s role in addressing fairness in elections. Yet, the Electoral College itself
is an undemocratic process. And states like South Carolina exacerbate its undemocratic
aspects by allocating all of their presidential electors to the state-wide popular vote winner.
This is not a question of partisanship—South Carolina discards votes for
Democratic presidential candidates, but other states discard votes for Republican
presidential candidates. 2 This is a question of the openness and fairness of the winner-take-
all method of selecting presidential electors for the Electoral College.
I.
On appeal, South Carolina argues that the Supreme Court’s summary affirmance in
Williams v. Virginia State Board of Elections, 288 F. Supp. 622 (E.D. Va. 1968), aff’d, 393
U.S. 320 (1969), forecloses Plaintiffs’ claims. That argument largely carried the day in the
district court and in the other courts that have recently considered parallel litigation. 3 The
2
For example, in Massachusetts, more than one million votes were cast for the Republican
presidential candidate in 2016. Despite constituting almost 33% of the total votes cast in
the state, the Republican candidate received none of Massachusetts’s eleven electoral
votes. Lyman v. Baker, 954 F.3d 351, 356 (1st Cir. 2020). A similar result occurred in
Massachusetts in the seven previous presidential elections. And the same pattern holds in
other states like California and New York.
3
At the time Plaintiffs brought this suit, their counsel filed parallel litigation on behalf of
voters in Texas, Massachusetts, and California. See Lyman v. Baker, 954 F.3d 351 (1st Cir.
2020); League of United Latin Am. Citizens v. Abbott, 951 F.3d 311 (5th Cir. 2020);
Rodriguez v. Brown, No. 2:18-cv-001422, 2018 WL 6136140 (C.D. Cal. Sept. 21, 2018),
appeal filed, No. 18-56281 (9th Cir. 2018). The First and Fifth Circuits have affirmed the
dismissal of the Massachusetts and Texas suits, respectively, although neither court
considered a Voting Rights Act claim. The California case is still pending before the Ninth
Circuit.
30
majority here describes Williams as “persuasive” but, apparently, does not give it the
controlling weight afforded by other courts. Majority Op. at 17. Nonetheless, before
addressing the approach taken by the majority in this matter, it is appropriate to address
South Carolina’s arguments and the district court’s conclusion regarding Williams.
The district court’s conclusion rested on the simple fact that Williams, like this case,
involved a challenge to a state’s winner-take-all method of selecting presidential electors.
But Williams was a summary affirmance by the Supreme Court without an opinion
explaining its reasoning. Thus, we must consider the reach and controlling effect of that
summary affirmance. Mandel v. Bradley, 432 U.S. 173, 176 (1977); Hicks v. Miranda, 422
U.S. 332, 345 n.14 (1975) (“Ascertaining the reach and content of summary actions may
itself present issues of real substance.”).
A summary affirmance resolves the issues for the parties, and lower courts may not
come “to opposite conclusions on the precise issues presented and necessarily decided.”
Mandel, 432 U.S. at 176. But at the same time, “a summary affirmance is an affirmance of
the judgment only,” not the rationale of the lower court, and “should not be understood as
breaking new ground.” Id. And a summary affirmance “is not to be read as a renunciation
. . . of doctrines previously announced in [the Court’s] opinions after full argument.” Id.(quotation marks omitted).
Indeed, “summary affirmances have considerably less precedential value than an
opinion on the merits.” Ill. State Bd. of Elections v. Socialist Workers Party, 440 U.S. 173,
180–81 (1979). And “doctrinal developments” may, in some circumstances, undermine the
31
controlling effect of a summary disposition. Hicks, 422 U.S. at 344 (quoting Port Auth.
Bondholders Protective Comm. v. Port of N.Y. Auth., 387 F.2d 259, 263 n.3 (2d Cir. 1967)).
So, to assess the controlling effect of Williams, we must first identify the “precise
issues presented and necessarily decided” in that case. Mandel, 432 U.S. at 176. And then,
even if the precise issues presented and decided in Williams are again presented in this
case, we must consider whether doctrinal developments since the Williams decision
indicate that its controlling influence has waned.
In Williams, the district court addressed three claims made by the plaintiffs:
(1) the intendment of Article II, Section 1, providing for the appointment of
electors is that they be chosen in the same manner as Senators and
Representatives, that is two at large and the remainder by Congressional or
other equal districts; (2) the general ticket method violates the ‘one-person,
one-vote’ principle of the Equal Protection Clause of the Fourteenth
Amendment, i.e., the weight of each citizen’s vote must be substantially
equal to that of every other citizen; and (3) the general ticket system gives a
citizen in a State having a larger number of electors than Virginia the
opportunity to effectuate by his vote the selection of more electors than can
the Virginian.
Williams, 288 F. Supp. at 624 (citations omitted) (citing Wesberry v. Sanders, 376 U.S. 1,
18 (1964); Gray v. Sanders, 372 U.S. 368, 381 (1963)).
Here, Plaintiffs do not argue that South Carolina must allocate electors in the same
manner as senators or representatives, nor do they argue that South Carolina’s current
system somehow affects the weight of Plaintiffs’ votes relative to that of voters in other
states. And so, we are left to consider what precise issues were necessarily decided in
resolving the Williams plaintiffs’ one person, one vote claim.
32
A.
To begin, Williams did not address or resolve any issues related to a claim of vote
dilution. In claims of racial vote dilution, “[t]he basic concept, broadly stated, is that racial
minorities may not have their group voting power impermissibly ‘diluted’ by multimember
districting or at-large electoral processes which ‘submerge’ the minority voting group in a
voting constituency in which the voting power of a racially ‘bloc-voting’ white majority
always insures defeat for the candidates of the minority group’s choice.” McGhee v.
Granville County, 860 F.2d 110, 116 (4th Cir. 1988). The one person, one vote principle
at issue in Williams, in contrast, implicates the “requirement that all citizens’ votes be
weighted equally,” and usually arises in challenges to disparities in population among
districts. Raleigh Wake Citizens Ass’n v. Wake Cty. Bd. of Elections, 827 F.3d 333, 340
(4th Cir. 2016).
In Republican Party of North Carolina v. Martin, this Court concluded that Wells
v. Edwards, 347 F. Supp. 453 (M.D. La. 1972), aff’d mem., 409 U.S. 1095 (1973)—a
summary affirmance of a one person, one vote claim—did not control in a case involving
a claim of vote dilution. 980 F.2d 943, 954 (4th Cir. 1992). The same is true here. The
Williams court’s analysis of the one person, one vote argument does not foreclose the vote
dilution claim in this case. And moreover, Williams could not have “necessarily decided”
the “precise issue[ ]” of a vote dilution claim because the doctrine supporting such a claim
did not develop in any significant sense until the 1970s, that is, after the Williams summary
affirmance in January 1969. Mandel, 432 U.S. at 176.
33
Although “[t]he question of the constitutional validity of multi-member districts
ha[d] been pressed in [the] Court since the first of the modern reapportionment cases,” the
Supreme Court’s first substantial vote dilution case was its 1971 decision in Whitcomb v.
Chavis. 403 U.S. 124, 142 (1971); see id. at 142 n.22 (discussing previous cases in which
multi-member districts were involved but where vote dilution claims were not squarely
presented). There, the Supreme Court considered a challenge to Indiana’s multi-member
state legislative districts, holding that multi-member districts are not categorically
unconstitutional but that a challenger could show unconstitutional vote dilution by proving
“that multi-member districts unconstitutionally operate to dilute or cancel the voting
strength of racial or political elements.” Id. at 144.
Then, two years later, the Court for the first time struck down a multi-member
district scheme as an unconstitutional vote dilution. White v. Regester, 412 U.S. 755 (1973).
The Court concluded that in bringing these types of claims, “it is not enough that the racial
group allegedly discriminated against has not had legislative seats in proportion to its
voting potential.” Id. at 765–66. Instead, a challenger’s “burden is to produce evidence to
support findings 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.” Id. at 766. Subsequent cases reaffirmed and refined
the doctrine. See City of Mobile v. Bolden, 446 U.S. 55 (1980); Rogers v. Lodge, 458 U.S.
613 (1982).
34
And so, the issue of vote dilution was not “presented and necessarily decided” in
Williams because the Williams court had neither the arguments nor the relevant precedent
before it. Mandel, 432 U.S. at 176. Because Williams does not speak to the issue of vote
dilution, it does not foreclose Plaintiffs’ arguments on that issue.
B.
But even as to Plaintiffs’ one person, one vote claim—a claim that alleges that South
Carolina unconstitutionally “discards” Plaintiffs’ votes after the first stage of the two-stage
presidential election—we must take seriously our obligation to ascertain the reach of
Williams. See Mandel, 432 U.S. at 176. Inasmuch as vote dilution claims post-date
Williams, similarly, the one person, one vote principle has developed significantly since
the Williams summary affirmance.
This Court has recognized that doctrinal developments—especially decades’
worth—can call into question the continued force of summary dispositions. For example,
in Bostic v. Schaefer, we considered whether the Supreme Court’s summary dismissal in
Baker v. Nelson, 409 U.S. 810 (1972), settled a challenge to Virginia’s prohibition on same-
sex marriage. 760 F.3d 352, 373 (4th Cir. 2014). We noted that an intervening Supreme
Court case—United States v. Windsor, 570 U.S. 744 (2013)—had considered a related
issue and did not mention Baker. Id. at 373–74. We said that such an omission “speaks
volumes.” Id. at 374. We then went on to say that “[t]he Court’s development of its due
process and equal protection jurisprudence in the four decades following Baker is even
more instructive.” Id.35
Accordingly, an understanding of the historical development of the Court’s one
person, one vote jurisprudence is essential to determining the controlling effect of
Williams.
In 1946 in Colegrove v. Green, the Supreme Court rejected a challenge to Illinois’s
unequally populated congressional districts. 328 U.S. 549 (1946). Justice Frankfurter’s
plurality opinion concluded that such a challenge was beyond the competence of the federal
judiciary because of its “peculiarly political nature.” Id. at 552. Although the Constitution
“gives ample power to provide against [the] evils” of malapportionment, Justice
Frankfurter concluded “[a]uthority for dealing with such problems resides elsewhere”—
not with the judiciary. Id. at 554. He warned, “Courts ought not to enter this political
thicket.” Id. at 556.
Not even twenty years later, in Baker v. Carr, the Supreme Court rejected Justice
Frankfurter’s conclusion that claims of malapportioned districts that diluted voting strength
were beyond the competence of federal courts. 369 U.S. 186 (1962). Baker considered a
challenge to the malapportionment of the state legislature of Tennessee. Tennessee’s
legislative districts, like the districts of many other states at the time, did not share equal
populations. Id. at 319–21 (Frankfurter, J., dissenting). Diverging from Colegrove, the
Court in Baker held that the judiciary was equipped to resolve such a challenge: “Judicial
standards under the Equal Protection Clause are well developed and familiar, and it has
been open to courts since the enactment of the Fourteenth Amendment to determine, if on
the particular facts they must, that a discrimination reflects no policy, but simply arbitrary
and capricious action.” Id. at 226 (majority opinion). And so even where, much like the
36
case before us now, previous decisions had been read to condone the state’s practice, and
even where many states used a system suffering from the same alleged defect, the Court in
Baker remanded the case for consideration on the merits. Id. at 237.
Then, in Gray v. Sanders, the Court further developed this jurisprudence, using the
phrase “one person, one vote.” 372 U.S. 368, 381 (1963). Gray involved a challenge to
Georgia’s “county unit system” of allocating and counting votes in a state-wide primary.
Id. at 370. Each county received a number of votes based on its number of representatives
in the lower house of the state’s general assembly. Id. at 371. Because the general assembly
was malapportioned, there were significant disparities in the voting power of citizens of
different counties. Id. at 370–71 & n.1. The Supreme Court rejected Georgia’s justification
of the system based on analogy to the United States Senate or Electoral College. Id. at 377–
380. In so doing, the Court characterized the philosophy underlying the Electoral College
as “belong[ing] to a bygone day.” Id. at 376 n.8. Instead, the Court concluded “once the
class of voters is chosen and their qualifications specified, we see no constitutional way by
which equality of voting power may be evaded.” Id. at 381. “The conception of political
equality from the Declaration of Independence, to Lincoln’s Gettysburg Address, to the
Fifteenth, Seventeenth, and Nineteenth Amendments can mean only one thing—one
person, one vote.” Id.The Court continued to refine the one person, one vote doctrine in subsequent cases.
In Wesberry v. Sanders, it held that the one person, one vote, equal population principle
applied to congressional districts. 376 U.S. 1, 18 (1964). In doing so, the Court’s majority
rejected the arguments raised in Justice Harlan’s dissent, including that the Elections
37
Clause in Article I, Section 4 gives states the power to prescribe the “Manner” of
conducting congressional elections. See id. at 23–24 (Harlan, J., dissenting). 4
And in Reynolds v. Sims, the Court held that “as a basic constitutional standard, the
Equal Protection Clause requires that the seats in both houses of a bicameral state
legislature must be apportioned on a population basis.” 377 U.S. 533, 568 (1964).
“Legislators represent people, not trees or acres.” Id. at 562. And “each citizen [should]
have an equally effective voice in the election of members of his state legislature.” Id. at
565. Reynolds therefore marked the expansion of the one person, one vote doctrine from a
rule requiring equally populated districts (as in Baker and Gray), to a theory of equally
effective voting power. And Reynolds rejected the view that the courts could turn a blind
eye to constitutional violations because “a denial of constitutionally protected rights
demands judicial protection; our oath and our office require no less.” Id. at 566.
“Over the ensuing decades, the Court has several times elaborated on the scope of
the one-person, one-vote rule.” Evenwel v. Abbott, 136 S. Ct. 1120, 1124 (2016). For
example, after Reynolds and after the summary affirmance in Williams, the Court clarified
that states may deviate from equal population among districts to a greater degree in drawing
state and local legislative districts than in drawing congressional districts. See Brown v.
Thomson, 462 U.S. 835, 842–43 (1983); see also id. at 850 n.2 (O’Connor, J., concurring).
4
Justice Harlan’s comments in dissent illustrate the significant changes the Court’s one
person, one vote jurisprudence brought: “I had not expected to witness the day when the
Supreme Court of the United States would render a decision which casts grave doubt on
the constitutionality of the composition of the House of Representatives. It is not an
exaggeration to say that such is the effect of today’s decision.” Wesberry, 376 U.S. at 20(Harlan, J., dissenting).
38
And the Court addressed which types of governmental bodies are subject to the doctrine.
See, e.g., Salyer Land Co. v. Tulare Lake Basin Water Storage Dist., 410 U.S. 719, 733–
34 (1973) (permitting water storage district board votes to be apportioned based on land
value rather than population); Hadley v. Junior Coll. Dist. of Metro. Kan. City, 397 U.S.
50, 52 (1970) (applying one person, one vote to elections for trustees of junior college).
And the Court has held the doctrine applies to other democratic processes, like the
nominations and selection of presidential electors. See, e.g., Bush v. Gore, 531 U.S. 98,
104–05 (2000) (applying one person, one vote to recounting of votes in an election of
presidential electors); Moore v. Ogilvie, 394 U.S. 814, 819 (1969) (striking down an Illinois
law requiring signatures from at least 50 of the state’s 102 counties to qualify independent
candidates for president).
Thus, in the more than fifty years since the Court’s summary affirmance in Williams,
the Supreme Court has clarified and elaborated on the early one person, one vote cases,
particularly in cases moving beyond strict numerical equality of district population and into
theories of equally effective voting power. In the same way that this Court in Bostic
concluded that Baker was no longer controlling, the intervening decades of precedent on
the one person, one vote principle undermine Williams’s controlling effect on that issue.
Therefore, we must fully consider each of Plaintiffs’ arguments.
Plaintiffs’ First Amendment and Voting Rights Act arguments were clearly not
raised in Williams. And this Court has previously held that a summary decision in a one
person, one vote case does not control in a subsequent vote dilution case. Martin, 980 F.2d
at 954. And because voting rights jurisprudence developed so substantially over the fifty
39
years between the Williams court’s consideration of a one person, one vote argument and
the Plaintiffs’ challenge here, we are not bound by Williams with respect to that claim,
either.
When courts—like the district court and like our sister circuits who have heard and
rejected related cases—fail to seriously consider claims like those raised by Plaintiffs, they
risk returning to Justice Frankfurter’s approach in Colegrove—turning a blind eye to
constitutional wrongs out of fear of a political thicket. By stepping away from our
responsibility to resolve the questions presented to us, we ignore that Baker v. Carr rejected
that very approach and ushered in an era of more democratic governance. “When faced
with such constitutional wrongs, courts must intervene: ‘It is emphatically the province and
duty of the judicial department to say what the law is.’” Rucho v. Common Cause, 139 S.
Ct. 2484, 2525 (2019) (Kagan, J., dissenting) (quoting Marbury v. Madison, 5 U.S. (1
Cranch) 137, 177 (1803)).
Judicial abdication in a case such as the one before us causes particularly insidious
harm to our democracy. Courts play an essential role in protecting constitutional rights in
cases that question allocations of political power. “[T]he courts were designed to be an
intermediate body between the people and the legislature, in order, among other things, to
keep the latter within the limits assigned to their authority.” The Federalist No. 78
(Alexander Hamilton).
When those who wield political power use mechanisms of government to entrench
and enlarge their power, “politicians’ incentives conflict with voters’ interests, leaving
citizens without any political remedy for their constitutional harms.” Gill v. Whitford, 138
40
S. Ct. 1916, 1941 (2018) (Kagan, J., concurring). Members of a legislature may derive
personal political benefit from a particular choice of electoral system—for example,
members of one party of a state legislature may prefer a system that will favor their party
on a national level, thus increasing that party’s power and, in turn, increasing their own
power. Courts will be, in some instances, the only real avenue for redress.
Accordingly, it undermines our government of and for the people when courts
choose to leave unconstitutional electoral mechanisms uncorrected. “The courts cannot
reject as ‘no law suit’ a bona fide controversy as to whether some action denominated
‘political’ exceeds constitutional authority.” Baker, 369 U.S. at 217.
II.
The majority apparently agrees with me that our review is not controlled by
Williams. Thus, this matter involves considering Plaintiffs’ Fourteenth Amendment
challenges to South Carolina’s state-wide, winner-take-all system of allocating presidential
electors.
The parties—and the majority—agree that although the Constitution gives state
legislatures power to choose the method of selecting presidential electors, that power is not
without limits. It is “always subject to the limitation that [it] may not be exercised in a way
that violates other specific provisions of the Constitution.” Williams v. Rhodes, 393 U.S.
23, 29 (1968). In the election context, the Equal Protection Clause requires that “[h]aving
once granted the right to vote on equal terms, the State may not, by later arbitrary and
disparate treatment, value one person’s vote over that of another.” Bush, 531 U.S. at 104–
05. And so, states must “[e]nsure that each person’s vote counts as much, insofar as it is
41
practicable, as any other person’s” and protect each person’s vote “against dilution or
debasement.” Hadley, 397 U.S. at 54.
Plaintiffs argue their Fourteenth Amendment claim through the lens of vote dilution
and through the lens of one person, one vote, as articulated in Gray v. Sanders, which
Plaintiffs label as a claim of vote “discarding.” At this early stage, Plaintiffs’ burden is, of
course, relatively low. And where Plaintiffs offer “a novel legal theory that can best be
assessed after factual development,” dismissals at such an early stage are disfavored.
Wright v. North Carolina, 787 F.3d 256, 263 (4th Cir. 2015) (internal quotation marks
omitted). The complaint must allege facts sufficient to support a plausible claim that South
Carolina’s state-wide, winner-take-all system violates the Fourteenth Amendment. See
Ashcroft v. Iqbal, 556 U.S. 662, 678 (2009). For the reasons below, I believe it does.
A.
Claims of vote dilution allege that members of a minority have had “their group
voting power impermissibly ‘diluted’ by multimember districting or at-large electoral
processes which ‘submerge’ the minority voting group” within a larger majority group.
McGhee, 860 F.2d at 116. For example, “[a] distinct minority, whether it be a racial, ethnic,
economic, or political group, may be unable to elect any representatives in an at-large
election, yet may be able to elect several representatives if the political unit is divided into
single-member districts.” Rogers, 458 U.S. at 616; see also Whitcomb, 403 U.S. at 143(noting that the Court has “recogniz[ed] . . . that [multi-member districts] may be subject
to challenge where the circumstances of a particular case may ‘operate to minimize or
42
cancel out the voting strength of racial or political elements of the voting population’”
(quoting Fortson v. Dorsey, 379 U.S. 433, 439 (1965))).
Multi-member districts and at-large systems are not inherently unconstitutional, but
they violate the Constitution when they dilute minority votes. See White, 412 U.S. at 765;
Whitcomb, 403 U.S. at 142–43. Generally, if plaintiffs can demonstrate “members [of a
group] had less opportunity than did other residents in the district to participate in the
political processes and to elect legislators of their choice,” then plaintiffs have shown “the
political processes leading to nomination and election were not equally open to
participation” in violation of the Equal Protection Clause. White, 412 U.S. at 766.
Specifically, in the context of state legislatures, the Supreme Court has identified
several considerations relevant to the question of whether a minority’s votes are
impermissibly diluted. The risk of vote dilution “is enhanced when the district is large and
elects a substantial proportion of the seats in either house of a bicameral legislature, if it is
multi-member for both houses of the legislature or if it lacks provision for at-large
candidates running from particular geographical subdistricts.” Whitcomb, 403 U.S. at 143–
44. And in White v. Regester, where the Supreme Court affirmed the district court’s
invalidation of dilutive districts, the Court approvingly discussed the district court’s
consideration of historical and ongoing racial discrimination and the minority’s inability to
elect representatives of its choice in the districts at issue. 412 U.S. at 765–70.
Here, South Carolina’s system of selecting presidential electors has many of the
features that risk creating the vote dilution identified in Whitcomb: the “district” from
which presidential electors are elected is large (indeed, it is the entire state); every one of
43
South Carolina’s electors is selected through this method; and no electors run from
particular geographical subdistricts. Whitcomb, 403 U.S. at 143–44.
The “past and present reality, political and otherwise” similarly shows vote dilution.
White, 412 U.S. at 770. For example, in the 2016 presidential election, 40.67% of South
Carolina voters cast their ballots for the Democratic candidate. But despite this support,
South Carolina selected no Democratic electors.
A similar pattern has occurred in previous elections. In every presidential election
since 2000, the Democratic candidate for president received at least 40% of the vote in
South Carolina, but South Carolina has selected no Democratic electors. Although the
Democratic candidate earns a significant number of votes, none of South Carolina’s
electors are the electors supporting that candidate.
Plaintiffs allege that here, by submerging their votes in the state-wide total and then
allocating electors only on the basis of the state-wide plurality winner, South Carolina has
subjected them to arbitrary and disparate treatment and created a system that is not equally
open to participation by South Carolina’s Democratic voters. This is a viable claim of vote
dilution in violation of the Fourteenth Amendment. 5
5
The parties disagree whether an allegation of invidious intent is required to make out a
claim of vote dilution. In Bolden, a plurality of the Supreme Court concluded that
constitutional claims of vote dilution require proof of invidious intent, not just harmful
effects. 446 U.S. at 66 (plurality opinion). But in Bush v. Gore, the Court held that equal
protection prohibits a state from, through “arbitrary and disparate treatment, valu[ing] one
person’s vote over that of another.” 531 U.S. at 104–05. Plaintiffs argue the Bush Court’s
prohibition on “arbitrary and disparate treatment” removed any requirement of invidious
intent. And although South Carolina argues that Bush did not abrogate any existing
requirement of invidiousness, it takes the position that either invidious discrimination or
44
As discussed above, the one person, one vote principle can apply to governmental
bodies other than Congress and state legislatures. See, e.g., Hadley, 397 U.S. at 52(applying one person, one vote to elections for trustees of a junior college); Avery v.
Midland County, 390 U.S. 474, 482 (1968) (applying equal population principle to five-
member county commissioners court where the court had administrative, executive, and
judicial functions); Martin, 980 F.2d at 953 (concluding that the plaintiffs could bring a
vote dilution claim challenging the method of electing judges).
In Avery, the Court observed that the commissioners court had the “power to make
a large number of decisions having a broad range of impacts on all the citizens of the
county.” 390 U.S. at 483. The function of South Carolina’s electors is limited, particularly
because South Carolina requires its electors to vote for their declared candidates. See S.C.
Code Ann. § 7-19-80. But South Carolina’s electors nonetheless exercise significant
power. The electors express South Carolina’s preferences in national presidential elections
and they, in fact, cast South Carolina’s only constitutionally effective votes for president.
And South Carolina’s practices in selecting and allocating its presidential electors
affect the nation as a whole. “[I]n the context of a Presidential election, state-imposed
restrictions implicate a uniquely important national interest. For the President and the Vice
arbitrary and disparate treatment of voters can support a one person, one vote challenge.
Plaintiffs allege that South Carolina’s system arbitrarily treats their votes differently—
diminishing as much as 48% support for Democratic candidates into zero presidential
electors and magnifying as little as 38% support for Republican candidates into 100% of
the state’s electors.
45
President of the United States are the only elected officials who represent all the voters in
the Nation.” Anderson v. Celebrezze, 460 U.S. 780, 794–95 (1983) (footnote omitted).
To be clear, there is little case law applying the one person, one vote principle to the
selection of presidential electors. 6 And to some degree, the Electoral College is already
“districted” by state by the Constitution. See U.S. Const. art. II, § 1, cl. 2. But the Supreme
Court has long held that although the Constitution gives states the power to set the method
for selecting their electors, states may not select a method that violates other constitutional
provisions. See Chiafalo v. Washington, No. 19-465, 2020 WL 3633779, at *6 n.4 (U.S.
July 6, 2020) (“A State, for example, cannot select its electors in a way that violates the
Equal Protection Clause.”); Rhodes, 393 U.S. at 29 (“Nor can it be thought that the power
to select electors could be exercised in such a way as to violate express constitutional
commands that specifically bar States from passing certain kinds of laws.”).
And although the Constitution allocates electors on a state-by-state basis—and
permits each state one presidential vote in the House of Representatives in the case of an
Electoral College tie—those allocations predate and are not governed by the Fourteenth
Amendment. See U.S. Const. amend. XII. The Fourteenth Amendment requires that states,
like South Carolina, afford every “person within [their] jurisdiction the equal protection of
the laws.” U.S. Const. amend. XIV. And so the Constitution’s allocation of electors to the
6
The majority cites Chiafalo v. Washington, No. 19-465, 2020 WL 3633779, at *6 (U.S.
July 6, 2020) for the proposition that states’ authority to determine how to appoint electors
includes the authority to require electors to vote for the plurality vote winner in a state.
Majority Op. at 16. Chiafalo did not consider whether a state-wide winner-take-all system
like South Carolina’s violates the Equal Protection Clause.
46
states does not mean the states, limited as they are by the Fourteenth Amendment, are free
to replicate that allocation on a smaller scale if that system would violate the Equal
Protection Clause.
For example, if South Carolina were to allocate its nine electors by district and draw
those districts with population disparities that violated the one person, one vote principle,
I suspect, and hope, courts would take a constitutional challenge to that system seriously.
We should take Plaintiffs’ arguments just as seriously, and not excuse unconstitutional vote
dilution simply on the basis that the Constitution—prior to the adoption of the Fourteenth
Amendment—allocated electors on a state-by-state basis or gave each state one vote in the
House following an Electoral College tie.
Because Plaintiffs’ allegations make out a plausible claim of vote dilution in
violation of the Fourteenth Amendment, I would allow Plaintiffs to proceed in this
litigation with this claim.
B.
Plaintiffs also allege South Carolina’s state-wide, winner-take-all system violates
the Fourteenth Amendment by discarding their votes before the operative stage of the two-
stage presidential election. 7 In support of this position, Plaintiffs rely on the one person,
one vote principle, particularly as discussed in Gray, Reynolds, and Bush.
7
South Carolina describes Plaintiffs as “seek[ing] to complicate this case” by describing
presidential elections as having two stages. Response Br. at 39. But there is no legitimate
disagreement about how presidential elections occur or that there are multiple stages. First,
states select presidential electors. U.S. Const. art. II, § 1, cl. 2. Today, every state uses a
system in which citizens vote for electors. Bush, 531 U.S. at 104. In South Carolina, the
47
As previously addressed, the principle of one person, one vote does not simply
require mathematical equality of population among districts. It “requires . . . that each
citizen have an equally effective voice in the election” of government. Reynolds, 377 U.S.
at 565. And “the right of suffrage can be denied by a debasement or dilution of the weight
of a citizen’s vote just as effectively as by wholly prohibiting the free exercise of the
franchise.” Id. at 555. So South Carolina does not insulate itself from an equal protection
claim by merely allowing Plaintiffs to vote in the election for presidential electors. “Equal
protection applies as well to the manner of [the franchise’s] exercise. Having once granted
the right to vote on equal terms, the State may not, by later arbitrary and disparate
treatment, value one person’s vote over that of another.” Bush, 531 U.S. at 104–05.
Here, Plaintiffs allege that “[t]he consequence of [South Carolina’s] system is to
give no effect to the votes of citizens who voted for a losing candidate in South Carolina
in the tabulation of the final vote for President.” J.A. 23–24. 8 “[V]otes for a losing
presidential candidate are counted in South Carolina only to be discarded when another
candidate wins more votes in South Carolina.” J.A. 24.
names of electors do not appear on the ballot. S.C. Code Ann. § 7-19-70. Instead, the names
of presidential and vice-presidential candidates appear, and “[a] vote for the candidates
named on the ballot shall be a vote for the electors of the party by which those candidates
were nominated.” Id. The electors of each state cast their ballots for president and vice
president, and if a presidential candidate receives a majority of the electoral votes, he or
she is elected president. U.S. Const. amend. XII. Plaintiffs refer to the vote by citizens for
electors as the first stage, and the vote by electors for president as the second.
8
Citations to ‘J.A. __’ refer to the Joint Appendix filed by the parties in this appeal.
48
The Gray Court identified the problem with South Carolina’s state-wide, winner-
take-all elections in describing a problem with Georgia’s county unit system: “if a
candidate won 6,000 of 10,000 votes in a particular county, he would get the entire unit
vote, the 4,000 other votes for a different candidate being worth nothing and being counted
only for the purpose of being discarded.” 372 U.S. at 381 n.12; see also Gordon v. Lance,
403 U.S. 1, 4 (1971) (describing Gray as holding “that the county-unit system would have
been defective even if unit votes were allocated strictly in proportion to population”). In
every presidential election since 1980, South Carolinians’ votes for Democratic candidates
for president have been “counted only for the purpose of being discarded” because, despite
representing a significant share of the votes, these votes have translated into no votes in the
Electoral College.
South Carolina argues that states have used the winner-take-all system since the first
presidential election and that today, 47 other states and the District of Columbia use a
similar system. But the first presidential elections in this country were not governed by the
Fourteenth Amendment and its Equal Protection Clause, which was not ratified until 1868.
And as was demonstrated in Baker v. Carr, longstanding or widespread use of a particular
election practice does not mean the practice complies with constitutional demands. See
also McGirt v. Oklahoma, No. 18-9526, 2020 WL 3848063, at *20 (U.S. July 9, 2020)
(rejecting an argument that an adverse ruling would be transformative because “the
magnitude of a legal wrong is no reason to perpetuate it.”).
Hundreds of thousands of South Carolinians vote in each presidential election, only
to have their votes disregarded weeks later when South Carolina’s electors cast their votes
49
for president—the only votes that matter for constitutional purposes. Through this system,
these South Carolina voters are denied an “equally effective voice in the election” and their
votes are not accorded equal value. Reynolds, 377 U.S. at 565.
Although this practice has infected generations of presidential elections, I
nonetheless believe Plaintiffs have plausibly alleged this system unconstitutionally
discards their votes and I would allow Plaintiffs to pursue this claim.
III.
In addition to their Fourteenth Amendment claims, Plaintiffs also claim South
Carolina’s state-wide, winner-take-all system unconstitutionally burdens their First
Amendment rights. Again, Williams v. Virginia Board of Elections did not involve a First
Amendment claim, and so we must consider this claim on its own terms, without regard to
anything we believe the Williams court may have thought but left unsaid. Plaintiffs allege
South Carolina’s system burdens their right to express their political preference, their right
to associate, and their right to petition major party candidates. This is sufficient to state a
plausible claim of a First Amendment violation, and I do not believe Plaintiffs’ claim
should be dismissed at this early stage.
“[P]olitical belief and association constitute the core of those activities protected by
the First Amendment.” Elrod v. Burns, 427 U.S. 347, 356 (1976). And the First
Amendment “has its fullest and most urgent application to speech uttered during a
campaign for political office.” Citizens United v. Fed. Election Comm’n, 558 U.S. 310, 339
(2010) (internal quotation marks omitted). And of course, the First Amendment protects
“the right of individuals to associate for the advancement of political beliefs, and the right
50
of qualified voters, regardless of their political persuasion, to cast their votes effectively.”
Rhodes, 393 U.S. at 30.
Therefore, “[t]he First Amendment demands judicial scrutiny of state election
regulations because regulations that ‘govern[ ] the registration and qualifications of voters,
the selection and eligibility of candidates, or the voting process itself, inevitably affect[ ]—
at least to some degree—the individual’s right to vote and his right to associate with others
for political ends.’” Common Cause v. Rucho, 318 F. Supp. 3d 777, 926 (M.D.N.C. 2018)
(second and third alterations in original) (quoting Anderson, 460 U.S. at 788), vacated and
remanded with instructions, 139 S. Ct. 2484; see also Benisek v. Lamone, 348 F. Supp. 3d
493, 523 (D. Md. 2018) (majority opinion by Niemeyer, J.) (concluding the plaintiffs had
established associational harms with evidence of “a lack of enthusiasm, indifference to
voting, a sense of disenfranchisement, a sense of disconnection, and confusion”), vacated
and remanded with instructions sub nom. Rucho v. Common Cause, 139 S. Ct. 2484.
States may regulate elections, and important regulatory interests generally justify
“reasonable, nondiscriminatory restrictions.” Anderson, 460 U.S. at 788. But because
election regulations affect important First Amendment interests, courts apply a sliding
scale of scrutiny to state election regulations. See Burdick v. Takushi, 504 U.S. 428, 433–
34 (1992). Severe burdens are subject to strict scrutiny and must be narrowly tailored to
serve a compelling state interest; “reasonable, nondiscriminatory restrictions” are subject
to less searching review. Anderson, 460 U.S. at 788; see also Burdick, 504 U.S. at 433–34.
The Supreme Court has struck down even facially neutral electoral regulations that had the
effect of burdening particular political parties, candidates, or groups of voters. See,
51
e.g., Tashjian v. Republican Party of Conn., 479 U.S. 208, 225 (1986) (concluding that a
state’s enforcement of a statute requiring closed primaries, against the will of the
Republican Party, violated the First Amendment); Anderson, 460 U.S. at 806 (striking
down a state candidate filing deadline because it imposed an unjustified burden on third-
party candidates and their supporters, with the “interests of the voters who chose to
associate together” for political ends constituting the Court’s “primary concern”).
In their complaint, Plaintiffs allege South Carolina’s state-wide, winner-take-all
system burdens their First Amendment rights because their “voices are not heard” in the
presidential election. J.A. 28. Generally, “the function of the election process is to winnow
out and finally reject all but the chosen candidates.” Burdick, 504 U.S. at 438 (internal
quotation marks omitted). And “[a]ttributing to elections a more generalized expressive
function would undermine the ability of States to operate elections fairly and efficiently.”
Id. So the Court has “repeatedly upheld reasonable, politically neutral regulations that have
the effect of channeling expressive activity at the polls.” Id. But Plaintiffs do not complain
that their political expression is channeled; they complain that it is silenced because
Plaintiffs’ votes are discarded before the critical stage of the election.
Plaintiffs also allege their associational rights are burdened because they cannot
translate shared principles into “concerted action” and “political power in the community.”
J.A. 29. As a result of the system, “candidates from major political parties rarely hold
campaign events in South Carolina once they are selected by their parties in the primary.
This results in a reduced opportunity for all South Carolinians to interface with and petition
the candidates for major political parties in person[.]” J.A. 29.
52
This is a cognizable First Amendment burden. “[D]ifficulties fundraising,
registering voters, attracting volunteers, generating support from independents, and
recruiting candidates to run for office (not to mention eventually accomplishing their policy
objectives)” all implicate the First Amendment. Gill, 138 S. Ct. at 1938 (Kagan, J.,
concurring); see Anderson, 460 U.S. at 791 n.12 (concluding that similar harms imposed
by a state election law amounted to a “burden imposed on . . . associational rights”);
Borough of Duryea v. Guarnieri, 564 U.S. 379, 388 (2011) (describing the right to petition
as “allow[ing] citizens to express their ideas, hopes, and concerns to their government and
their elected representatives”).
The district court concluded that “[P]laintiffs merely allege that the inability of
Democratic voters in South Carolina to succeed in having any electoral votes distributed
to a Democratic presidential candidate for the past forty years has dampened the likelihood
that they will engage in political activity, as it appears useless.” Baten v. McMaster, 374 F.
Supp. 3d 563, 570 (D.S.C. 2019). The district court believed (and the majority appears to
agree) that Plaintiffs “conflate[ ] a diminishing motivation to participate with a severe
burden on the actual ability of people to participate in the voting process.” Id. (emphasis
omitted). And because “[t]he parties [did] not provide[ ] any cases in which a court has
questioned the [winner-take-all] system under a freedom of association theory,” the district
court dismissed this claim. Id. at 571. In my view, that was error.
First, of course, just because a court has not previously addressed the First
Amendment implications of a state-wide, winner-take-all system of allocating presidential
electors does not mean that such a system cannot create First Amendment burdens. But
53
second, and more importantly, the district court misunderstood Plaintiffs’ allegations.
Plaintiffs’ allegation that the system burdens their right to petition is that “candidates from
major political parties rarely hold campaign events in South Carolina once they are selected
by their parties in the primary.” J.A. 29. This is not an allegation that Democrats do not
come to South Carolina because Democrats are likely to lose. It is an allegation that all
major party candidates do not come to South Carolina because of the state’s winner-take-
all system.
And finally, these alleged burdens are not just a result of Plaintiffs’ “inability to
muster a majority of South Carolina’s votes.” Majority Op. at 24. Republican candidates
have earned all nine of South Carolina’s electoral votes with only a plurality of South
Carolinians’ votes. Democrats in South Carolina may turn out ever increasing numbers of
voters for presidential elections, but because of the system South Carolina has chosen, such
an increase is unlikely to make any difference in the selection of South Carolina’s nine
electors.
At a later point in this litigation, a court might measure Plaintiffs’ burdens and
consider South Carolina’s interest in its state-wide, winner-take-all system in light of those
burdens. Anderson, 460 U.S. at 788. And indeed, at that later point, a court may conclude
Plaintiffs have been unable to prove the burdens they allege. E.g., Gill, 138 S. Ct. at 1939(Kagan, J., concurring) (concluding that during the course of the litigation, the “plaintiffs
did not sufficiently advance a First Amendment associational theory” to establish standing
based on a burden to their associational interests). But they should have an opportunity to
develop those claims. Indeed, Plaintiffs’ alleged harms have been recognized as First
54
Amendment burdens before. Yet today the Court affirms the district court’s conclusion that
Plaintiffs cannot even pursue their claims. I believe that to be a grave error.
Moreover, because Plaintiffs’ claims, if allowed to proceed, would eventually be
weighed against South Carolina’s interest in its state-wide, winner-take-all system, it is
worth considering for a moment whether such an interest even exists. South Carolina
argues that, even if Plaintiffs have alleged cognizable First Amendment burdens, its interest
in exerting the greatest possible influence in the Electoral College justifies those burdens.
South Carolina has offered little explanation for how this is a legitimate state interest.
Burdening some South Carolinians’ First Amendment rights so that other South
Carolinians—those who voted for the presidential candidate garnering a plurality of
votes—may speak with more force on a national stage is not, on its face, a legitimate state
interest. Accordingly, if Plaintiffs were allowed to pursue their claims, there would be little
to weigh their burden against.
Tellingly, South Carolina attaches much authority to the fact that Thomas Jefferson
advocated for Virginia to adopt a winner-take-all system for selecting its electors. South
Carolina argues Jefferson’s “advice sprang from a desire to protect his State against the use
of [the] general ticket by other States.” Response Br. at 46 (quoting Williams, 288 F. Supp.
at 626). Jefferson’s “advice” to Virginia also gave him a better chance to win the next
presidential election, by directing all of Virginia’s Electoral College votes to the state’s
winner—likely to be Jefferson himself. Jefferson even recognized that he had a personal
stake in the question, noting that “[p]erhaps it will be thought I ought in delicacy to be
55
silent on this subject.” Letter from Thomas Jefferson to James Monroe (Jan. 12, 1800),
available at https://www.loc.gov/item/mtjbib009254.
Thus, some people—people other than the state of South Carolina—do have an
interest in South Carolina’s state-wide, winner-take-all system. As illustrated by the
historical example that South Carolina offers—Jefferson’s preference for winner-take-all
systems to shore up his Electoral College support—candidates who believe they can win a
plurality of votes in South Carolina (and the voters and elected officials who support them)
have a strong interest in maintaining this system. But preservation of political power for
the people who currently hold power is not a legitimate state interest. And, of course,
Jefferson’s advocacy for a particular practice does not mean the practice does not offend
the Constitution. See, e.g., Extracts from Letter from Thomas Jefferson to Jean Nicolas
Démeunier (Apr. 29, 1795), available at http://tjrs.monticello.org/letter/2352#X3184736
(discussing Jefferson’s enslavement of “a dozen little boys from 10[ ] to 16[ ] years of age”
as nail-makers). Our nation’s shared ideas of democracy and equality have evolved from
Jefferson’s day.
IV.
Finally, I turn with special concern to Plaintiffs’ claim under the Voting Rights Act.
Section 2 of the Voting Rights Act prohibits a state from imposing any voting practice that
“results in a denial or abridgement of the right of any citizen of the United States to vote
on account of race or color . . . .” 52 U.S.C. § 10301(a). A § 2 violation does not require
discriminatory intent. The statute prohibits “voting practices that ‘operate, designedly or
otherwise,’ to deny ‘equal access to any phase of the electoral process for minority group
56
members.’” United States v. Charleston County, 365 F.3d 341, 345 (4th Cir. 2004) (quoting
S. Rep. No. 97-417, at 28, 30 (1982)).
As the majority recognizes, courts evaluate § 2 vote dilution claims through the
familiar framework of Thornburg v. Gingles, 478 U.S. 30 (1986). To make out a successful
claim, a plaintiff must establish three prerequisites: the minority group must be sufficiently
large and geographically compact to constitute a majority in a single member district; the
minority group must be politically cohesive; and the white majority must vote sufficiently
as a bloc so as usually to defeat the minority’s preferred candidate. Gingles, 478 U.S. at50–51; see also Charleston Cty., 365 F. 3d at 345. After a plaintiff establishes the three
prerequisites, “the trier of fact must determine whether, based on the totality of the
circumstances, there has been a violation of Section 2.” Charleston Cty., 365 F.3d at 345. 9
9
The majority asserts that the Gingles prerequisites “do not map onto the type of challenge
that the plaintiffs have mounted here.” Majority Op. at 27. I disagree and discuss each
prerequisite in turn. However, even if Plaintiffs’ challenge were not a perfect fit with the
Gingles prerequisites, that would not mean Plaintiffs cannot make out a § 2 claim. Other
frameworks for understanding § 2 exist, most notably the vote-denial framework discussed
in League of Women Voters of N.C. v. North Carolina, 769 F.3d 224, 239–40 (4th Cir.
2014). A § 2 vote-denial claim requires two elements: first, the challenged practice must
impose a discriminatory burden on members of a protected class such that members of the
protected class have less opportunity than other members of the electorate to participate in
the political process; and second, the burden must in part be caused by or linked to social
and historical conditions that have or currently produce discrimination against members of
that protected class. Id. at 240. Moreover, the text of § 2 of the Voting Rights Act is broad
and not limited to only those violations that perfectly fit the framework described in
Gingles. See id. at 238 (“Section 2 ‘prohibits all forms of voting discrimination’ that lessen
opportunity for minority voters.” (quoting Gingles, 478 U.S. at 45 n.10)).
57
A.
As a preliminary matter, the district court seemed to cast doubt on whether the
Voting Rights Act applies to the election of presidential electors: “Regardless of whether
plaintiffs can meet these three provisions [of Gingles], the fact remains that this claim
operates on the same premise as their [one person, one vote] claim—that courts should be
able to consider the constitutionality of a state’s [winner-take-all] electoral college system
using the same legal tools and concepts of constitutional fairness that the courts have relied
on in assessing state-level voting procedures.” Baten, 374 F. Supp. 3d at 571. Of course,
the Voting Rights Act claim does not challenge the constitutionality of South Carolina’s
state-wide, winner-take-all system. It brings a statutory claim under the Voting Rights Act.
But more importantly, this statement by the district court may be read to suggest the Act
does not apply to states’ systems of allocating presidential electors. The majority corrects
that error, if only implicitly.
And to its credit, South Carolina did not take the position that the Voting Rights Act
does not apply to presidential elections in its briefing. However, at oral argument, South
Carolina was asked if it took the position that the Voting Rights Act does not apply to
presidential elections. South Carolina responded that it “ha[d] not taken that position
affirmatively” but described the issue as “certainly an open question.” Oral Arg. at 22:16–
22:50.
There can be no serious debate over whether the Voting Rights Act applies to
presidential elections, and this is not an “open question.” For example, the original Voting
Rights Act of 1965 provided penalties for anyone who knowingly gave false information
58
in order to establish his or her eligibility to vote and specified that “this provision shall be
applicable only to general, special, or primary elections held solely or in part for the
purpose of selecting or electing any candidate for the office of President, Vice President,
presidential elector” or other specified federal office. Pub. L. No. 89-110, § 11(a), 79 Stat.
437, 443 (emphasis added). And the Voting Rights Act Amendments of 1970 included
provisions that specifically addressed residency requirements in elections for presidential
electors. Pub. L. No. 91-285, § 202(d), 84 Stat. 314, 316–17. Although several states
challenged those amendments, the Supreme Court upheld the provisions, recognizing
Congress’s “prerogative . . . to oversee the conduct of presidential and vice-presidential
elections.” Oregon v. Mitchell, 400 U.S. 112, 124 (1970).
And nothing about § 2 of the Voting Rights Act indicates that it was meant to have
a narrower reach than the rest of the Act. Section 2 prohibits any “voting qualification or
prerequisite to voting or standard, practice, or procedure” to be “imposed or applied by any
State or political subdivision in a matter which results in a denial or abridgement of the
right” to vote on account of race. 52 U.S.C. § 10301(a). In striking down or enjoining
various election procedures under § 2, courts of appeals (including this one) have not
differentiated presidential elections or prohibited the challenged practices in all elections
other than presidential elections. See, e.g., Democratic Nat’l Comm. v. Hobbs, 948 F.3d
989, 1016 (9th Cir. 2020); Veasey v. Abbott, 830 F.3d 216, 268–72 (5th Cir. 2016); League
of Women Voters of N.C. v. North Carolina, 769 F.3d 224, 248–49 (4th Cir. 2014); see also
Hous. Lawyers’ Ass’n v. Attorney Gen. of Tex., 501 U.S. 419, 425 (1991) (“The term
‘representatives’ [in § 2] is not a word of limitation.”). And the only other district court
59
that considered a Voting Rights Act claim in parallel litigation unambiguously applied the
Voting Rights Act to presidential elections. See League of United Latin Am. Citizens v.
Abbott, 369 F. Supp. 3d 768, 784 (W.D. Tex. 2019) (“As a threshold matter, the Court
firmly agrees with Plaintiffs that Section 2 [of] the VRA applies to Presidential
Elections.”), aff’d, 951 F.3d 311 (5th Cir. 2020).
South Carolina’s position that it is “certainly an open question” whether the Voting
Rights Act applies to presidential elections is concerning. South Carolina must comply
with the Voting Rights Act in presidential elections. And if a South Carolina voter brings
a claim alleging South Carolina has violated the Voting Rights Act in how it selects
presidential electors, the Court has an obligation to hear that claim.
B.
Turning to Plaintiffs’ specific allegations and the question of whether Plaintiffs’
Voting Rights Act claim can survive a motion to dismiss, I believe the district court erred.
The district court’s basis for dismissing this claim was unclear, but it seemed to find that
Plaintiffs failed to establish the third Gingles prerequisite: that the white majority votes
sufficiently as a bloc to usually defeat the minority-preferred candidate. Baten, 374 F.
Supp. 3d at 571. The district court did not explain this conclusion, and under controlling
Fourth Circuit authority, Plaintiffs sufficiently established all three Gingles prerequisites.
Plaintiffs have therefore stated a plausible claim of a violation of § 2 of the Voting Rights
Act and this claim should not have been dismissed.
The first Gingles prerequisite requires that Plaintiffs establish that South Carolina’s
African American population is “sufficiently large and geographically compact to
60
constitute a majority in a single-member district.” Gingles, 478 U.S. at 50. In United States
v. Charleston County, this Court upheld the district court’s conclusion that the plaintiffs
had satisfied this prerequisite where African Americans were 34.3% of the population of a
900-square-mile county and where non-white voters represented 30.9% of registered voters
in the county. 365 F.3d at 343–44. Based on data from the South Carolina Election
Commission, Plaintiffs allege that African American voters made up 27.6% of the
registered voters in South Carolina in 2016. African Americans are a majority of the
population in the state’s Sixth Congressional District, as well as in eight South Carolina
counties. And in ten South Carolina counties, African Americans are a majority of
registered voters. There are several ways South Carolina’s nine presidential electors could
be allocated such that African Americans could be the majority in a subdivision of the state.
Plaintiffs’ allegations are sufficient to establish the first Gingles prerequisite. 10
The second prerequisite requires that Plaintiffs establish that African American
voters are “politically cohesive.” Gingles, 478 U.S. at 51. This Court has previously noted
that African American voters in one of South Carolina’s most populous counties are
politically cohesive. Charleston Cty., 365 F.3d at 347. And in their complaint, Plaintiffs
allege that in 2008 and 2016, African American South Carolinians voted for the
Democratic presidential candidate at rates of 96% and 94%, respectively. This is sufficient
10
The majority concludes “there is no alternative of a ‘single-member district’ because the
selection of electors in South Carolina does not involve districts at all.” Majority Op. at 27.
But at-large systems that do not use districts are commonly challenged under the Voting
Rights Act. And indeed, the lack of any subdivision of South Carolina’s nine electoral votes
is the very practice Plaintiffs challenge.
61
to satisfy the second Gingles prerequisite. Cf. League of United Latin Am. Citizens v. Perry,
548 U.S. 399, 427 (2006) (second prerequisite established where 92% of Latinx voters
voted against a candidate).
Finally, the third Gingles prerequisite requires that Plaintiffs establish that white
voters in South Carolina vote sufficiently as a bloc to usually defeat the minority-preferred
candidate. Gingles, 478 U.S. at 51. 11 Although the district court concluded that Plaintiffs
failed to establish this factor, it offered no analysis as to why the complaint’s allegations
failed. Baten, 374 F. Supp. 3d at 571. This Court has previously noted, in considering this
prerequisite, that “[t]he terms used by the Gingles Court are ‘usually,’ ‘normally,’ and
‘generally.’” Lewis v. Alamance County, 99 F.3d 600, 606 n.4 (4th Cir. 1996). And
although we did not specifically define those terms, we observed, “suffice it to say that
they mean something more than just 51%.” Id.Plaintiffs allege that, according to census estimates, in 2016 white people accounted
for 69.7% of the voting age population in South Carolina. And in exit polls from the 2008
and 2016 presidential elections, “white voters in South Carolina supported the Republican
candidate at rates of 73% and 70%, respectively.” J.A. 38. In the last ten presidential
11
The majority finds that Plaintiffs’ “elide the distinction between the candidates the
minorities prefer . . . and the ‘candidates’ that, in their view, they have sufficient political
power to elect.” Majority Op. at 27–28. But when South Carolina voters go the polls to
vote for president, they are, in fact, selecting South Carolina’s nine presidential electors.
And so, as a general matter, when African American South Carolina voters prefer the
Democratic nominees for president and vice president, they prefer the electors who will
cast their votes in the Electoral College for those nominees. The “candidates” they support
are, therefore, the ones they allege they have political power to elect.
62
elections, not one of South Carolina’s presidential electors has been aligned with the
minority-preferred candidate.
South Carolina seems to argue that Plaintiffs must allege that the minority-preferred
candidate is defeated because of race and not partisan preference, relying on League of
United Latin American Citizens, Council No. 4434 v. Clements, 999 F.2d 831, 853–54 (5th
Cir. 1993) (en banc). But in the Fourth Circuit, “[l]egally significant white bloc voting . . .
refers to the frequency with which, and not the reason why, whites vote cohesively for
candidates who are not backed by minority voters.” Charleston Cty., 365 F.3d at 348–49
(quotation marks omitted). Because Plaintiffs sufficiently allege that white bloc voting
usually defeats the minority-preferred candidate in statewide elections in South Carolina,
Plaintiffs have established the third Gingles prerequisite.
Although this Court has not explicitly held that allegations establishing the Gingles
prerequisites are sufficient to survive a motion to dismiss, it is difficult to see how a court
could reach a different conclusion. Once the three Gingles prerequisites are established,
courts evaluate the totality of the circumstances, with special attention to the non-
exhaustive list of factors identified in Gingles, including: the extent to which members of
a protected class are elected; any history of official discrimination in voting practices;
discriminatory housing, education, and employment practices; and the existence of racial
appeals in campaigning. Gingles, 478 U.S. at 38–40; see id. at 36–37 (citing S. Rep. No.
97-417, at 28–29 (1982)).
This is, of course, a fact-intensive inquiry. And several of our sister circuits have
acknowledged that where a plaintiff establishes the Gingles prerequisites, that plaintiff is
63
likely to succeed under the totality of the circumstances. See Mo. State Conference of the
NAACP v. Ferguson-Florissant Sch. Dist., 894 F.3d 924, 930 (8th Cir. 2018); Sanchez v.
Colorado, 97 F.3d 1303, 1322 (10th Cir. 1996); Clark v. Calhoun County, 88 F.3d 1393,
1396 (5th Cir. 1996); Uno v. City of Holyoke, 72 F.3d 973, 983 (1st Cir. 1995); Nipper v.
Smith, 39 F.3d 1494, 1524 (11th Cir. 1994); Jenkins v. Red Clay Consol. Sch. Dist. Bd. of
Educ., 4 F.3d 1103, 1116 n.6 (3d Cir. 1993) (“[I]t would be a highly unusual case in which
a plaintiff successfully proved the existence of the three Gingles factors and still failed to
establish a violation.”).
Here, Plaintiffs sufficiently allege the three Gingles prerequisites. And even if we
were to consider allegations about the totality of the circumstances at this early stage,
Plaintiffs sufficiently make the necessary, plausible allegations. Plaintiffs allege that South
Carolina has a history of judicially recognized discriminatory voting practices; there are
significant racial disparities in education, employment, health, housing, income,
transportation, and incarceration in the state; the racial disparities in these categories
indicate a lack of responsiveness to the needs of minority African Americans; there
continue to be racially charged campaign and election practices; and voters have elected
only one African American to a state-wide office. Given these allegations, it is plausible
that South Carolina’s state-wide, winner-take-all system violates § 2 of the Voting Rights
Act based on the totality of the circumstances.
On appeal, South Carolina argues that Plaintiffs must establish an “undiluted”
baseline to state a claim, relying on Reno v. Bossier Parish School Board, 520 U.S. 471(1997). Response Br. at 48, 51. But Reno is inapposite. That case involved preclearance
64
under the Voting Rights Act; it is simply not relevant to Plaintiffs’ burdens at the pleading
stage, and cannot be read to impose a requirement on Plaintiffs to offer an undiluted
baseline this early in the litigation. Reno, 520 U.S. at 477.
Plaintiffs’ factual allegations are sufficient to establish the Gingles prerequisites.
And even assuming their allegations must show a plausible violation under the totality of
the circumstances at this early stage, Plaintiffs have met that burden. The district court
erred in dismissing this claim.
V.
In this case, Plaintiffs challenge a longstanding and widespread method of selecting
presidential electors. But the method’s ubiquity does not mean it is the best method, or the
most democratic method, or even a constitutional method. I would allow Plaintiffs to
pursue their claims so that the courts can consider their arguments on a developed record
and in light of modern voting rights jurisprudence. That is the legally correct and fair way
to address this matter.
65