Harper v. Hall

IN THE SUPREME COURT OF NORTH CAROLINA
No. 413PA21-2
Filed 28 April 2023
REBECCA HARPER; AMY CLARE OSEROFF; DONALD RUMPH; JOHN
ANTHONY BALLA; RICHARD R. CREWS; LILY NICOLE QUICK; GETTYS
COHEN, JR.; SHAWN RUSH; JACKSON THOMAS DUNN, JR.; MARK S.
PETERS; KATHLEEN BARNES; VIRGINIA WALTERS BRIEN; DAVID DWIGHT
BROWN
v.
REPRESENTATIVE DESTIN HALL, in his official capacity as Chair of the House
Standing Committee on Redistricting; SENATOR WARREN DANIEL, in his
official capacity as Co-Chair of the Senate Standing Committee on Redistricting
and Elections; SENATOR RALPH HISE, in his official capacity as Co-Chair of the
Senate Standing Committee on Redistricting and Elections; SENATOR PAUL
NEWTON, in his official capacity as Co-Chair of the Senate Standing Committee
on Redistricting and Elections; SPEAKER OF THE NORTH CAROLINA HOUSE
OF REPRESENTATIVES TIMOTHY K. MOORE; PRESIDENT PRO TEMPORE
OF THE NORTH CAROLINA SENATE PHILIP E. BERGER; THE NORTH
CAROLINA STATE BOARD OF ELECTIONS; and DAMON CIRCOSTA, in his
official capacity
NORTH CAROLINA LEAGUE OF CONSERVATION VOTERS, INC.; HENRY M.
MICHAUX, JR.; DANDRIELLE LEWIS; TIMOTHY CHARTIER; TALIA FERNÓS;
KATHERINE NEWHALL; R. JASON PARSLEY; EDNA SCOTT; ROBERTA
SCOTT; YVETTE ROBERTS; JEREANN KING JOHNSON; REVEREND
REGINALD WELLS; YARBROUGH WILLIAMS, JR.; REVEREND DELORIS L.
JERMAN; VIOLA RYALS FIGUEROA; and COSMOS GEORGE
v.
REPRESENTATIVE DESTIN HALL, in his official capacity as Chair of the House
Standing Committee on Redistricting; SENATOR WARREN DANIEL, in his
official capacity as Co-Chair of the Senate Standing Committee on Redistricting
and Elections; SENATOR RALPH E. HISE, JR., in his official capacity as Co-Chair
of the Senate Standing Committee on Redistricting and Elections; SENATOR
PAUL NEWTON, in his official capacity as Co-Chair of the Senate Standing
Committee on Redistricting and Elections; REPRESENTATIVE TIMOTHY K.
MOORE, in his official capacity as Speaker of the North Carolina House of
HARPER V. HALL
Opinion of the Court
Representatives; SENATOR PHILIP E. BERGER, in his official capacity as
President Pro Tempore of the North Carolina Senate; THE STATE OF NORTH
CAROLINA; THE NORTH CAROLINA STATE BOARD OF ELECTIONS;
DAMON CIRCOSTA, in his official capacity as Chairman of the North Carolina
State Board of Elections; STELLA ANDERSON, in her official capacity as
Secretary of the North Carolina State Board of Elections; JEFF CARMON III, in
his official capacity as Member of the North Carolina State Board of Elections;
STACY EGGERS IV, in his official capacity as Member of the North Carolina State
Board of Elections; TOMMY TUCKER, in his official capacity as Member of the
North Carolina State Board of Elections; and KAREN BRINSON BELL, in her
official capacity as Executive Director of the North Carolina State Board of
Elections
On direct appeal pursuant to Rule 3 of the North Carolina Rules of Appellate
Procedure from the unanimous decision of a three-judge panel entered on 23
February 2022 in the Superior Court, Wake County, approving Legislative
Defendants’ Remedial House Plan and Remedial Senate Plan, rejecting their
Remedial Congressional Plan, and adopting an Interim Congressional Plan. Heard in
the Historic 1767 Chowan County Courthouse in Edenton, North Carolina on 4
October 2022, and opinion filed on 16 December 2022. Subsequently, this Court
allowed Legislative Defendants’ petition for rehearing pursuant to Rule 31(a) of the
North Carolina Rules of Appellate Procedure. Heard in the Supreme Court on 14
March 2023.
Patterson Harkavy LLP, by Burton Craige, Narendra K. Ghosh, and Paul E.
Smith; Elias Law Group LLP, by Lalitha D. Madduri, Jacob D. Shelly, and
-2-
HARPER V. HALL
Opinion of the Court
Abha Khanna; and Arnold and Porter Kaye Scholer LLP, by Elisabeth S.
Theodore, R. Stanton Jones, and Samuel F. Callahan, for Harper Plaintiffs.
Robinson, Bradshaw & Hinson, P.A., by John R. Wester, Adam K. Doerr,
Stephen D. Feldman, and Erik R. Zimmerman; and Jenner & Block LLP, by
Sam Hirsch, pro hac vice, and Jessica Ring Amunson, pro hac vice, for Plaintiff
North Carolina League of Conservation Voters.
Southern Coalition for Social Justice, by Hilary H. Klein, Mitchell Brown,
Katelin Kaiser, Jeffrey Loperfido, and Noor Taj; and Hogan Lovells US LLP,
by J. Tom Boer, pro hac vice, and Olivia T. Molodanof, pro hac vice, for Plaintiff
Common Cause.
Nelson Mullins Riley & Scarborough LLP, by Phillip J. Strach, Thomas A.
Farr, John E. Branch, III, D. Martin Warf, Nathaniel J. Pencook, and Alyssa
M. Riggins; and Baker Hostetler LLP, by Mark E. Braden, pro hac vice,
Katherine McKnight, pro hac vice, and Richard Raile, pro hac vice, for
Legislative Defendants.
North Carolina Department of Justice, by Amar Majmundar, Senior Deputy
Attorney General, Terence Steed, Special Deputy Attorney General, Mary Carla
Babb, Special Deputy Attorney General, and Stephanie Brennan, Special
Deputy Attorney General, for State Defendants.
NEWBY, Chief Justice.
“A frequent recurrence to fundamental principles is absolutely necessary to
preserve the blessings of liberty.” N.C. Const. art. I, § 35. Since our founding in 1776
almost 250 years ago, this provision in our state constitution has reminded us of the
critical importance of remembering fundamental principles. This case now invites us
to return to those principles.
-3-
HARPER V. HALL
Opinion of the Court
The constitution is our foundational social contract and an agreement among
the people regarding fundamental principles. It is for everyone, not just lawyers and
judges. The state constitution is different from the Federal Constitution: the Federal
Constitution is a limited grant of power while the state constitution is a limitation on
power. The state constitution declares that all political power resides in the people.
N.C. Const. art. I, § 2. The people exercise that power through the legislative branch,
which is closest to the people and most accountable through the most frequent
elections. See id. art. I, § 9. In the constitutional text, the people have assigned specific
tasks to, and expressly limited the powers of, each branch of government. The state
constitution is detailed and specific. The people speak through the express language
of their constitution, and only the people can amend it. See id. art. XIII.
The constitution is interpreted based on its plain language. The people used
that plain language to express their intended meaning of the text when they adopted
it. The historical context of our constitution confirms this plain meaning. As the
courts apply the constitutional text, judicial interpretations of that text should
consistently reflect what the people agreed the text meant when they adopted it.
There are no hidden meanings or opaque understandings—the kind that can only be
found by the most astute justice or academic. The constitution was written to be
understood by everyone, not just a select few.
-4-
HARPER V. HALL
Opinion of the Court
The state constitution establishes three branches of government: legislative,
executive, and judicial. It assigns specific roles to each branch. Since its inception,
the constitution has provided for separation of powers: in other words, each branch is
directed to perform its assigned duties and avoid encroaching on the duties of another
branch. Separation of powers protects individual freedoms. The will of the people is
achieved when each branch of government performs its assigned duties. When,
however, one branch grasps a task of another, that action violates separation of
powers.
The judicial branch is designed to resolve legal disputes and to ensure that the
other branches do not violate the constitution. Our power of judicial review, however,
is not unlimited. Since the first articulation of the doctrine of judicial review in
Bayard v. Singleton, 1 N.C. (Mart.) 5 (1787), courts have refused to exercise that
power if the constitution assigns the matter to another branch, or the constitution
does not provide a judicially discoverable or manageable standard, or resolution of
the matter involves policy choices. Such matters are deemed political questions and
are nonjusticiable. The Supreme Court of the United States recognized these
limitations in its seminal case, Marbury v. Madison, in which it first adopted the
concept of judicial review:
It is scarcely necessary for the court to disclaim all
pretensions to [intermeddle with the prerogatives of
-5-
HARPER V. HALL
Opinion of the Court
another branch]. An extravagance, so absurd and
excessive, could not have been entertained for a moment.
The province of the court is, solely, to decide on the rights
of individuals, not to enquire how [other branches] perform
duties in which they have a discretion. Questions, in their
nature political, or which are, by the constitution and laws,
submitted to [another branch], can never be made in this
court.
5 U.S. (1 Cranch) 137, 170 (1803).
Historically, North Carolina courts have respected their significant but
restrained role of judicial review by adhering to a standard of review that sets the
most demanding requirements for reviewing legislative action: courts presume that
an act of the General Assembly is constitutional, and any challenge alleging that an
act of the General Assembly is unconstitutional must identify an express provision of
the constitution and demonstrate that the General Assembly violated the provision
beyond a reasonable doubt.
Giving a fixed meaning to the constitution and using a deferential standard to
review legislation ensures that courts will perform their assigned role, stay within
their lane of authority, and refrain from becoming policymakers. Courts are not
designed to be thrust into the midst of various political disputes. Such engagement
in policy issues forces courts to take sides in political battles and undermines public
trust and confidence in the judiciary. Choosing political winners and losers creates a
perception that courts are another political branch. The people did not intend their
-6-
HARPER V. HALL
Opinion of the Court
courts to serve as the public square for policy debates and political decisions. Instead,
the people act and decide policy matters through their representatives in the General
Assembly. We are designed to be a government of the people, not of the judges. At its
heart, this case is about recognizing the proper limits of judicial power.
This matter is before this Court on rehearing. The North Carolina Rules of
Appellate Procedure authorize rehearing a case when “the court has overlooked or
misapprehended” a point “of fact or law.” N.C. R. App. P. 31(a). In their petition for
rehearing, Legislative Defendants ask the Court to revisit the crucial issue in this
case: whether claims of partisan gerrymandering are justiciable under the state
constitution. They assert that such claims are not justiciable. Legislative Defendants
maintain that “[t]he Harper experiment” has failed: “Harper II failed . . . because
Harper I set this Court up to fail.” In support of this argument, Legislative
Defendants argue that Harper I “fell short in concrete guidance” and “declined to
disclose what standard applies.” They assert that “Harper II reaffirms the non-
justiciable and unprecedented standard set forth in Harper I” and, therefore, “a
necessary consequence of correcting the errors in Harper II is to overrule Harper I.”
Legislative Defendants argue that their rehearing petition “gives this Court a much[-]
needed opportunity to address the root of the problem: Harper I was based on
-7-
HARPER V. HALL
Opinion of the Court
profoundly flawed legal principles.” Accordingly, they ask this Court to withdraw its
Harper II opinion and overrule Harper I.
In this case plaintiffs claim that the General Assembly violated the state
constitution by drawing legislative districts that unfairly benefited one political party
at the expense of another, in other words, partisan gerrymandering.1 Partisan
gerrymandering is the practice of dividing a geographical or jurisdictional area into
political units or election districts to give a particular political party or group “a
special advantage.” See Gerrymandering, Black’s Law Dictionary (11th ed. 2019).
In the first opinion in this matter, four justices held that partisan
gerrymandering presents a justiciable claim, Harper v. Hall (Harper I), 380 N.C. 317,
390, 868 S.E.2d 499, 551 (2022), and violates several provisions of the Declaration of
Rights of our constitution, id. at 383, 868 S.E.2d at 546. The four justices then
discussed certain political science tests that they claimed were judicially discoverable
and manageable. Id. at 384–85, 868 S.E.2d at 547–48. They maintained that these
political science tests could reliably identify unconstitutional partisan
1 In their complaints, plaintiffs allege that “partisan gerrymandering” violates the
state constitution. Sometimes they modify this phrase with words like “extreme” or “severe.”
In Rucho v. Common Cause, the Supreme Court of the United States referred to this concept
as “excessive partisan gerrymandering.” Rucho v. Common Cause, 139 S. Ct. 2484, 2507
(2019). In this opinion we will generally use the term “partisan gerrymandering” to refer to
these claims.
-8-
HARPER V. HALL
Opinion of the Court
gerrymandering, id., but they did not define how much partisan gerrymandering is
too much, id. at 384, 868 S.E.2d at 547. In the most recent opinion in this matter, the
same four members of this Court said that the General Assembly, three former jurists
serving as Special Masters, the three-judge panel, and three members of this Court—
in total, nine current and former jurists—all wrongly applied the approach set out in
Harper I. See Harper v. Hall (Harper II), 383 N.C. 89, 94, 881 S.E.2d 156, 162 (2022).
Thus, we must now reconsider whether a standard that only four justices know and
understand, that is riddled with policy choices, and that is not mentioned in our
constitution is truly judicially discoverable and manageable. That inquiry requires
us to revisit the fundamental premises underlying the decisions in both Harper II
and Harper I.
The issue presented in this case is whether the North Carolina Constitution
prohibits partisan gerrymandering. Specifically, plaintiffs allege that legislative and
congressional redistricting plans drawn by the General Assembly in 2021 and then
again in 2022 on remand are partisan gerrymanders in violation of specific provisions
of the constitution.
Our constitution expressly assigns the redistricting authority to the General
Assembly subject to explicit limitations in the text. Those limitations do not address
partisan gerrymandering. It is not within the authority of this Court to amend the
-9-
HARPER V. HALL
Opinion of the Court
constitution to create such limitations on a responsibility that is textually assigned
to another branch. Furthermore, were this Court to create such a limitation, there is
no judicially discoverable or manageable standard for adjudicating such claims. The
constitution does not require or permit a standard known only to four justices.
Finally, creating partisan redistricting standards is rife with policy decisions. Policy
decisions belong to the legislative branch, not the judiciary.
Recently, the Supreme Court of the United States reviewed similar claims
under the Federal Constitution and determined that “excessive” partisan
gerrymandering claims involve nonjusticiable, political questions. See Rucho v.
Common Cause, 139 S. Ct. 2484, 2491, 2507 (2019). We find the Supreme Court’s
analysis in Rucho insightful and persuasive.
For all these reasons, we hold that partisan gerrymandering claims present a
political question that is nonjusticiable under the North Carolina Constitution.
Accordingly, the decision of this Court in Harper I is overruled. We affirm the three-
judge panel’s 11 January 2022 Judgment concluding, inter alia, that partisan
gerrymandering claims are nonjusticiable, political questions and dismissing all of
plaintiffs’ claims with prejudice. This Court’s opinion in Harper II is withdrawn and
superseded by this opinion. The three-judge panel’s 23 February 2022 order is
vacated. Plaintiffs’ claims are dismissed with prejudice.
-10-
HARPER V. HALL
Opinion of the Court
I. Procedural History
A. Initial Litigation
As required by both our state constitution and the Federal Constitution, the
General Assembly, following the 2020 census, enacted redistricting plans for the
North Carolina Senate and House of Representatives and for the United States House
of Representatives (2021 Plans).2 The General Assembly enacted the 2021 Plans on 4
November 2021. The North Carolina League of Conservation Voters and a group of
individual North Carolina voters (NCLCV plaintiffs), along with another group of
2 Before drawing any maps, the General Assembly’s Senate Committee on
Redistricting and Elections convened a Joint Meeting of the Senate Redistricting and
Elections Committee and the House Redistricting Committee on 5 August 2021 to discuss
the criteria that would govern the redistricting process. Following this initial meeting, a
General Assembly staff member distributed to the joint committee members a list of the
legislative redistricting criteria that had been previously mandated by a three-judge panel in
Common Cause v. Lewis—a case decided just a few years earlier in 2019. See Common Cause
v. Lewis, No. 18 CVS 014001, 2019 WL 4569584 (N.C. Super. Ct. Wake County Sept. 3, 2019).
One week after its first meeting, the Joint Redistricting Committee adopted final
redistricting criteria that would govern its 2021 map drawing process (Adopted Criteria). In
many respects, the Adopted Criteria were nearly identical to the criteria ordered by the court
in Common Cause v. Lewis in 2019. Notably, just like the Lewis criteria, the Adopted Criteria
mandated that no “[p]artisan considerations [or] election results data” would be used in
drawing the 2021 Plans. It appears that the Joint Redistricting Committee incorporated the
criteria from Common Cause v. Lewis into its Adopted Criteria for the 2021 redistricting
process because it believed that compliance with the Common Cause v. Lewis criteria was
necessary to create constitutionally compliant redistricting plans. See Legislative
Defendants-Appellees’ Brief at 20−21, Harper v. Hall, 380 N.C. 317 (2022) (No. 413PA21-1)
(“To avoid violations identified in the 2010 [redistricting] cycle,” including those identified in
the Lewis order, the General Assembly included a prohibition on the consideration of partisan
election data in its Adopted Criteria.).
-11-
HARPER V. HALL
Opinion of the Court
individual North Carolina voters (Harper plaintiffs) each filed suit against the
President Pro Tempore of the North Carolina Senate, the Speaker of the North
Carolina House, and the Chairs of the House Standing Committee on Redistricting
and the Senate Standing Committee on Redistricting and Elections (Legislative
Defendants).3 NCLCV plaintiffs and Harper plaintiffs challenged the legality of these
plans, arguing they were unconstitutional partisan gerrymanders. Additionally,
NCLCV plaintiffs alleged that the 2021 Plans “engag[ed] in racial vote dilution” in
violation of the free elections clause and the equal protection clause of the North
Carolina Constitution and that the 2021 Plans violated the Whole County Provisions
(WCP) of the North Carolina Constitution. See N.C. Const. art. I, §§ 10, 19, 14, 12; id.
art. II, §§ 3(3), 5(3). Both groups of plaintiffs also sought a preliminary injunction to
enjoin use of the 2021 Plans.
The NCLCV and Harper actions were assigned to a three-judge panel of the
Superior Court in Wake County and then consolidated. On 3 December 2021, the
three-judge panel denied both NCLCV plaintiffs’ and Harper plaintiffs’ motions for
preliminary injunction. Both sets of plaintiffs filed a notice of appeal with the North
3 NCLCV plaintiffs and Harper plaintiffs also collectively named the State of North
Carolina, the North Carolina State Board of Elections, and the Chairman, Secretary, and
Members of the State Board of Elections. These defendants took “no position on the merits”
of this case. State Defendants’ Brief at 2, Harper v. Hall, 380 N.C. 317 (2022) (No. 413PA21-
1).
-12-
HARPER V. HALL
Opinion of the Court
Carolina Court of Appeals.
The Court of Appeals denied NCLCV plaintiffs’ and Harper plaintiffs’ requests
for a temporary stay on 6 December 2021. NCLCV plaintiffs and Harper plaintiffs
then filed several documents with this Court, including two petitions for discretionary
review prior to determination by the Court of Appeals, a motion to suspend appellate
rules to expedite a decision, and a motion to suspend appellate rules and expedite
briefing and argument. On 8 December 2021, this Court allowed both petitions for
discretionary review, granted a preliminary injunction, and temporarily stayed the
candidate filing period for the 2022 election cycle until “a final judgment on the
merits . . . including any appeals, is entered and a remedy, if any is required, has
been ordered.” In the same order, this Court expedited the matter, directing the
three-judge panel to hold proceedings on the merits of plaintiffs’ claims “and to
provide a written ruling” on or before 11 January 2022.
Subsequently, Common Cause moved to intervene as a plaintiff in the
consolidated proceedings, and the three-judge panel granted the motion on 15
December 2021. Like the NCLCV and Harper plaintiffs, Common Cause filed a
complaint alleging that the 2021 Plans were unconstitutional partisan gerrymanders
in violation of the free elections clause, the equal protection clause, and the free
speech and freedom of assembly clauses of the North Carolina Constitution. Common
-13-
HARPER V. HALL
Opinion of the Court
Cause also alleged that the 2021 Plans violated North Carolina’s equal protection
clause by “purposefully discriminat[ing] against” African American voters through
“intentional destruction of functioning crossover districts.” Finally, Common Cause
brought a declaratory judgment claim asking the three-judge panel to declare that
the North Carolina Constitution requires the General Assembly to undertake a
racially polarized voting (RPV) analysis prior to drawing any legislative districts.
Hereinafter, NCLCV plaintiffs, Harper plaintiffs, and Common Cause are collectively
referred to as “plaintiffs.”
Legislative Defendants filed their answers on 17 December 2021, and the
parties then engaged in an “expedited” two-and-one-half-week discovery period
culminating in rulings on over ten discovery-related motions, designation of ten
expert witnesses, and submission of over 1000 pages of expert reports and rebuttal
materials. After the discovery period closed on 31 December 2021, the three-judge
panel commenced a three-and-one-half-day trial on 3 January 2022 during which it
received approximately 1000 exhibits into evidence and testimony from numerous
fact and expert witnesses.
On 11 January 2022, the three-judge panel entered a judgment (11 January
2022 Judgment) concluding that plaintiffs’ partisan gerrymandering claims
presented nonjusticiable, political questions because redistricting “is one of the purest
-14-
HARPER V. HALL
Opinion of the Court
political questions which the legislature alone is allowed to answer.” The three-judge
panel reached this conclusion because “satisfactory and manageable criteria or
standards do not exist for judicial determination” of partisan gerrymandering claims.
Specifically, the three-judge panel noted that this Court already addressed the
justiciability of similar claims based on North Carolina’s Declaration of Rights in
Dickson v. Rucho and concluded there was no manageable standard to assess such
claims:
Finally, plaintiffs argue that the enacted plans violate the
“Good of the Whole” clause found in Article I, Section 2 of
the Constitution of North Carolina. We do not doubt that
plaintiffs’ proffered maps represent their good faith
understanding of a plan that they believe best for our State
as a whole. However, the maps enacted by the duly elected
General Assembly also represent an equally legitimate
understanding of legislative districts that will function for
the good of the whole. Because plaintiffs’ argument is not
based upon a justiciable standard, and because acts of the
General Assembly enjoy “a strong presumption of
constitutionality,” Pope v. Easley, 354 N.C. 544, 546, 556
S.E.2d 265, 267 (2001) (per curiam) (citation omitted),
plaintiffs’ claims fail.
(Quoting Dickson v. Rucho (Dickson I), 367 N.C. 542, 575, 766 S.E.2d 238, 260 (2014),
vacated on federal grounds, 137 S. Ct. 2186 (2017) (mem.) (emphasis added).) As a
result, the three-judge panel concluded that “[w]ere we as a [c]ourt to insert ourselves
in the manner requested, we would be usurping the political power and prerogatives
of an equal branch of government. Once we embark on that slippery slope, there
-15-
HARPER V. HALL
Opinion of the Court
would be no corner of legislative or executive power that we could not reach.”
Additionally, the three-judge panel concluded that the 2021 Plans did not
violate the North Carolina Declaration of Rights because “[t]he objective
constitutional constraints that the people of North Carolina have imposed on
legislative redistricting are found in Article II, Sections 3 and 5 of the 1971
Constitution and not in the Free Elections, Equal Protection, Freedom of Speech or
Freedom of Assembly Clauses found in Article I of the 1971 Constitution.” Finally,
the three-judge panel considered NCLCV plaintiffs’ and Common Cause’s additional
claims of racial vote dilution, racial discrimination, violation of the WCP, and request
for a declaratory judgment. Specifically, the three-judge panel concluded that NCLCV
plaintiffs and Common Cause “failed to satisfy” their burdens for both the racial vote
dilution and racial discrimination claims under the equal protection clause and that
the free elections clause is “inapplicable” to vote dilution claims. The three judge-
panel then concluded that the evidence did not support NCLCV’s WCP claim and that
the North Carolina Constitution does not, as Common Cause alleged, require the
General Assembly to undertake an RPV analysis prior to drawing legislative districts.
Accordingly, the three-judge panel dismissed plaintiffs’ claims with prejudice.
Pursuant to this Court’s 8 December 2021 order certifying the case for review
prior to determination by the Court of Appeals, all plaintiffs filed notices of appeal to
-16-
HARPER V. HALL
Opinion of the Court
this Court from the three-judge panel’s 11 January 2022 Judgment. The case was
argued before this Court on 2 February 2022. On 4 February 2022, in a four-to-three
decision, this Court entered an order (Remedial Order) adopting the findings of fact
from the 11 January 2022 Judgment but concluding that the 2021 Plans were
“unconstitutional beyond a reasonable doubt under the free elections clause, the
equal protection clause, the free speech clause, and the freedom of assembly clause of
the North Carolina Constitution.” The Remedial Order specifically enjoined the use
of the 2021 Plans “in any future elections.” The Remedial Order also required that,
in drawing new redistricting plans, the General Assembly must first conduct an RPV
analysis. The Remedial Order remanded the matter to the three-judge panel for
remedial proceedings and noted that a full opinion would follow. Three justices
dissented to the Remedial Order.
B. Harper I
Ten days later, the four-justice majority issued its full opinion. See Harper I,
380 N.C. at 317, 404, 868 S.E.2d at 499, 558–60. The Harper I opinion first held that
“partisan gerrymandering claims are justiciable in North Carolina courts under
the . . . [North Carolina] Declaration of Rights” because the right to aggregate votes
based on partisan affiliation is a fundamental right and there are “several
manageable standards for evaluating the extent to which districting plans dilute
-17-
HARPER V. HALL
Opinion of the Court
votes on the basis of partisan affiliation.” Id. at 390, 868 S.E.2d at 551. Specifically,
the majority determined that various political science metrics could serve as a
sufficient standard. See id. at 384–85, 868 S.E.2d at 547–48. It indicated that two
tests in particular—the Mean-Median Difference and the Efficiency Gap—could
demonstrate whether a redistricting map “is presumptively constitutional.”4 See id.
at 386, 868 S.E.2d at 548. According to the Harper I majority, a 1% or less Mean-
Median Difference score and a 7% or less Efficiency Gap score could serve as
thresholds of constitutionality. See id.
Nevertheless, the Harper I majority refused to delineate a precise standard.
Id. at 384, 868 S.E.2d at 547 (“We do not believe it prudent or necessary to, at this
time, identify an exhaustive set of metrics or precise mathematical thresholds which
conclusively demonstrate or disprove the existence of an unconstitutional partisan
gerrymander.”). Instead, the majority insisted that the three-judge panel—and future
trial courts adjudicating redistricting cases—would “work out more concrete and
4 The Mean-Median Difference and Efficiency Gap tests are statistical metrics that
purport to forecast partisan success under a particular redistricting plan in hypothetical,
future elections. See id. at 385−87, 868 S.E.2d at 548−49. The Mean-Median Difference
compares a party’s mean vote share with its median vote share in each district and assumes
that if the mean and median are equal, then the map contains no partisan skew. See id. at
386, 868 S.E.2d at 548. As explained in the filings before the three-judge panel, the Efficiency
Gap purports to compare each political parties’ “wasted votes.” According to Harper I, a 7%
Efficiency Gap score serves as a “workable . . . threshold” of constitutionality. Id.
-18-
HARPER V. HALL
Opinion of the Court
specific standards for evaluating state legislative apportionment schemes in the
context of actual litigation.” Id. at 384, 868 S.E.2d at 547 (quoting Reynolds v. Sims,
377 U.S. 533, 578, 84 S. Ct. 1362, 1390 (1964)).
The Harper I majority held that “[p]artisan gerrymandering of legislative and
congressional districts violates the free elections clause, the equal protection clause,
the free speech clause, and the freedom of assembly clause” of the North Carolina
Constitution. Id. at 383, 868 S.E.2d at 546. Specifically, the majority reasoned that
these provisions reflect “the principle of political equality,” id. at 382, 868 S.E.2d at
546, which in turn requires that “the channeling of ‘political power’ from the people
to their representatives in government through the democratic processes . . . must be
done on equal terms,” id. at 382, 868 S.E.2d at 546. Accordingly, the majority
concluded that to comport with these provisions in the Declaration of Rights, “the
General Assembly must not diminish or dilute on the basis of partisan affiliation any
individual’s vote” because “[t]he fundamental right to vote includes the right to enjoy
‘substantially equal voting power and substantially equal legislative
representation.’ ” Id. at 383, 868 S.E.2d at 546 (quoting Stephenson v. Bartlett
(Stephenson I), 355 N.C. 354, 382, 562 S.E.2d 377, 396 (2002)). In turn, the majority
concluded that “[t]he right to equal voting power encompasses the opportunity to
aggregate one’s vote with likeminded citizens to elect a governing majority of elected
-19-
HARPER V. HALL
Opinion of the Court
officials who reflect those citizens’ views.” Id. Thus, ironically, the Harper I majority
held that the constitution requires consideration of partisanship to remedy the
perceived use of partisanship.
The majority determined that because “[t]he right to vote on equal terms is a
fundamental right in this state,” strict scrutiny must apply once a party demonstrates
that a redistricting plan “infringes upon his or her fundamental right to substantially
equal voting power” based on partisan affiliation. Id. at 392–93, 868 S.E.2d at 553.
The majority held that to trigger strict scrutiny a party must demonstrate that a
redistricting plan “makes it systematically more difficult for a voter to aggregate his
or her vote with other likeminded voters.” Id. at 392, 868 S.E.2d at 552. A party may
make this demonstration using a variety of political science-based tests such as
median-mean difference analysis; efficiency gap analysis;
close-votes-close seats analysis[;] partisan symmetry
analysis; comparing the number of representatives that a
group of voters of one partisan affiliation can plausibly
elect with the number of representatives that a group of
voters of the same size of another partisan affiliation can
plausibly elect; and comparing the relative chances of
groups of voters of equal size who support each party of
electing a supermajority or majority of representatives
under various possible electoral conditions. Evidence that
traditional neutral redistricting criteria were subordinated
to considerations of partisan advantage may be
particularly salient in demonstrating an infringement of
this right.
-20-
HARPER V. HALL
Opinion of the Court
Id. at 392, 868 S.E.2d at 552–53. Once a party makes this initial demonstration, the
challenged redistricting plan is “unconstitutional [unless] the State [can] establish
that it is narrowly tailored to advance a compelling governmental interest.” Id. at
393, 868 S.E.2d at 553 (quoting Stephenson I, 355 N.C. at 377, 562 S.E.2d at 393).
The majority opined that “compliance with traditional neutral districting principles,
including those enumerated in [the WCP] of the North Carolina Constitution,” might
“constitute a compelling governmental interest” that would overcome strict scrutiny,
but “[p]artisan advantage” does not. Id. at 393, 868 S.E.2d at 553.
The majority then applied these ideas to the three-judge panel’s factual
findings and determined that the evidence at trial demonstrated that all of the 2021
Plans were partisan gerrymanders. Id. at 391−92, 868 S.E.2d at 552. The majority
then applied strict scrutiny to each map and concluded that the 2021 Plans were not
“carefully calibrated toward advancing some compelling neutral priority.” Id. at 396,
398, 401, 868 S.E.2d at 555, 556, 558.
The three dissenting justices concluded that plaintiffs’ claims were non-
justiciable. See id. at 413–34, 868 S.E.2d at 566–78 (Newby, C.J., dissenting). The
dissent noted that our state constitution expressly assigns the redistricting
responsibility to the General Assembly and that the majority failed to identify a
-21-
HARPER V. HALL
Opinion of the Court
judicially discernable, manageable standard by which to adjudicate the partisan
gerrymandering claims at issue. Id. at 424, 868 S.E.2d at 572.
C. Remedial Process
1. Three-Judge Panel’s Initial Orders
On remand, this Court’s 4 February 2022 Remedial Order required the General
Assembly to submit new congressional and state legislative redistricting plans “that
satisfy all provisions of the North Carolina Constitution” by 18 February 2022. The
Remedial Order also permitted plaintiffs to submit proposed remedial districting
plans by the same deadline and allowed all parties to file comments on any of the
submitted plans by 21 February 2022. The Remedial Order mandated that the three-
judge panel “approve or adopt compliant congressional and state legislative
districting plans no later than noon on 23 February 2022.”
In an 8 February 2022 order, the three-judge panel informed the parties of its
intent to appoint Special Masters to assist in reviewing the parties’ proposed remedial
plans and, if needed, in developing alternative remedial plans. Pursuant to the three-
judge panel’s order, each party submitted suggested individuals to serve as Special
Masters, but the three-judge panel appointed three other individuals of its own
choosing—former jurists Robert F. Orr, Robert H. Edmunds, Jr., and Thomas W.
Ross.
-22-
HARPER V. HALL
Opinion of the Court
The three-judge panel authorized the Special Masters to hire advisors
“reasonably necessary to facilitate their work.” The Special Masters hired four
advisors to assist in evaluating the General Assembly’s new remedial redistricting
plans: Dr. Bernard Grofman, Dr. Tyler Jarvis, Dr. Eric McGhee, and Dr. Samuel
Wang.
2. The General Assembly’s Remedial Process
The General Assembly understood Harper I as requiring it “to intentionally
create more Democratic districts in the [Remedial Plans].” To accomplish this task,
the General Assembly started with a blank slate and followed the same process to
create each map. Each redistricting committee kept the county groupings used for
the 2021 Plans as base maps. Accordingly, any single district county groupings from
each of the 2021 Plans were carried over to the Remedial Plans, but otherwise, each
map was entirely new.
Next, each redistricting committee “dr[e]w new districts and ma[d]e
adjustments tailored to legitimate criteria.” To do so, the General Assembly chose to
utilize Caliper’s Maptitude redistricting software, a “widely accepted districting
program.” Although expressly prohibited by its previous redistricting criteria and the
court-ordered criteria from Common Cause v. Lewis, the General Assembly “used
partisan election data as directed by the Supreme Court’s Remedial Order” to achieve
-23-
HARPER V. HALL
Opinion of the Court
its goal of “intentionally creat[ing] more Democratic districts.” Specifically, the
General Assembly chose to utilize partisan data from the set of twelve statewide
elections that plaintiffs’ expert, Dr. Mattingly, used to analyze the 2021 Plans
(Mattingly Election Set).
After Maptitude produced an initial set of House, Senate, and congressional
maps, the General Assembly analyzed the partisan fairness of each map using two
political science metrics—the Mean-Median Difference and the Efficiency Gap. The
General Assembly chose these two metrics because “they have been peer-reviewed in
numerous articles by numerous scholars, and because there is some (but not uniform)
agreement among scholars regarding thresholds for measuring partisanship.”
Additionally, the General Assembly selected these metrics because the Harper I
majority identified them as two of the “multiple reliable ways of demonstrating the
existence of an unconstitutional partisan gerrymander.” Harper I, 380 N.C. at 384,
868 S.E.2d at 547 (majority opinion). For each of these metrics, the General Assembly
selected threshold scores that, if achieved, would indicate that the relevant map
contained an acceptable level of partisan fairness under Harper I. Specifically, the
General Assembly selected a 1% threshold score for the Mean-Median Difference
metric and a 7% threshold score for the Efficiency Gap metric.
The General Assembly selected these threshold scores based on general
-24-
HARPER V. HALL
Opinion of the Court
agreement among political scientists that a redistricting plan with a Mean-Median
Difference less than 1% and an Efficiency Gap less than 7% is “presumptively
constitutional.” Additionally, the General Assembly selected these threshold scores
because the Harper I majority opined that they were “possible bright-line standards”
that could indicate a presumptively constitutional level of partisanship:
[U]sing the actual mean-median difference measure, from
1972 to 2016 the average mean-median difference in North
Carolina’s congressional redistricting plans was 1%.
Common Cause [v. Rucho], 318 F. Supp. 3d [777,] 893
[(M.D.N.C. 2018)]. That measure instead could be a
threshold standard such that any plan with a mean-
median difference of 1% or less when analyzed using a
representative sample of past elections is presumptively
constitutional.
With regard to the efficiency gap measure, courts
have found “that an efficiency gap above 7% in any
districting plan’s first election year will continue to favor
that party for the life of the plan.” Whitford v. Gill, 218 F.
Supp. 3d 837, 905 (W.D. Wis. 2016), rev’d on other grounds,
138 S. Ct. 1916 (2018). It is entirely workable to consider
the seven percent efficiency gap threshold as a
presumption of constitutionality, such that absent other
evidence, any plan falling within that limit is
presumptively constitutional.
Id. at 385, 386, 868 S.E.2d at 548.
After selecting its political science metrics and corresponding threshold scores,
the General Assembly then adjusted each of the Remedial Plans until their Mean-
Median Difference and Efficiency Gap scores were at or below the selected thresholds.
-25-
HARPER V. HALL
Opinion of the Court
Along with prioritizing the creation of more “purportedly Democratic leaning
districts” and ensuring the Remedial Plans scored well on the selected metrics, the
General Assembly also focused on the “neutral and traditional redistricting criteria”
used in creating the 2021 Plans unless those criteria conflicted with Harper I.
After drawing their respective plans, each chamber presented its plan to the
relevant redistricting committee. The General Assembly enacted the Remedial Plans
on 17 February 2022 and submitted them to the three-judge panel on 18 February
2022. Plaintiffs then offered comments and objections to the Remedial Plans. The
Special Masters transmitted a report on the Remedial Plans that was based primarily
on four reports written by the advisors. Notably, in crafting their reports, none of the
advisors used the General Assembly’s chosen redistricting program, Maptitude, nor
did they use the General Assembly’s chosen Mattingly Election Set. Instead, each
advisor used his own preferred data and methods.
The Special Masters’ Report found that the Remedial House Plan (RHP) and
Remedial Senate Plan (RSP) met the requirements of Harper I, but that the Remedial
Congressional Plan (RCP) did not. Because the Special Masters concluded that the
RCP was unconstitutional, they developed and submitted an alternative plan
(Interim Congressional Plan) in consultation with one of the advisors, Dr. Bernard
Grofman, for the three-judge panel to consider.
-26-
HARPER V. HALL
Opinion of the Court
In reviewing the Remedial Plans, the three-judge panel “adopt[ed] in full the
findings of the Special Masters.” Like the Special Masters, the three-judge panel
concluded that the RHP and RSP complied with the requirements of Harper I but
that the RCP was “not presumptively constitutional,” was “subject to strict scrutiny,”
and was not “narrowly tailored to a compelling governmental interest.” Accordingly,
the three-judge panel concluded that the RCP was unconstitutional. To support its
conclusion, the three-judge panel relied primarily on “the analysis performed by the
Special Masters and their advisors” and its conclusion that the RHP and RSP scored
below the relevant thresholds for the Mean-Median Difference and Efficiency Gap
metrics, but the RCP did not. The three-judge panel did not point to any other
evidence regarding the purported level of partisan bias in the Remedial Plans.
Finally, because the three-judge panel rejected the General Assembly’s RCP, it
adopted the Interim Congressional Plan recommended by the Special Masters.
Following the three-judge panel’s remedial order, all parties appealed to this
Court. The parties petitioned this Court to stay the three-judge panel’s remedial
ruling, but this Court denied those petitions. Accordingly, the RSP, RHP, and Interim
Congressional Plan were used in the 2022 elections.
D. Harper II
In June 2022, Common Cause filed a motion for expedited hearing and
-27-
HARPER V. HALL
Opinion of the Court
consideration of the three-judge panel’s remedial order. On 13 July 2022, Legislative
Defendants moved to dismiss their appeal of the three-judge panel’s rejection of the
RCP because the Interim Congressional Plan “ordered by [the three-judge panel] is
only applicable to the 2022 election, and that map will apply to the 2022 election
regardless of” this Court’s holding on the three-judge panel’s remedial order.
Legislative Defs.’ Mot. to Dismiss Appeal 3, Harper v. Hall, 380 N.C. 317 (2022) (No.
413PA21-1). Accordingly, Legislative Defendants sought to dismiss their appeal “in
an effort to avoid further cost and confusion to the taxpayers and voters of North
Carolina.” Id.In July 2022, the same four-justice majority from Harper I granted Common
Cause’s motion for expedited hearing and consideration and set oral argument for
October 2022. Harper v. Hall, 382 N.C. 314, 315–16, 874 S.E.2d 902, 904 (2022) (order
allowing motion to expedite hearing and consideration). Notably, in the same order,
the Court expressly declined to address Legislative Defendants’ motion to dismiss
their appeal. Id. at 316, 874 S.E.2d at 904. The three dissenting justices from Harper
I dissented from this order. Id. at 317–24, 874 S.E.2d at 904–09 (Barringer, J.,
dissenting) (noting that no jurisprudential reason existed to expedite consideration
of the appeal).
Ultimately, the same four-justice majority from Harper I affirmed the three-
-28-
HARPER V. HALL
Opinion of the Court
judge panel’s rejection of the RCP and its approval of the RHP and reversed the three-
judge panel’s approval of the RSP.5 Harper II, 383 N.C. at 94, 881 S.E.2d at 162. First,
the majority attempted “to clarify and reaffirm” its “constitutional standard” from
Harper I. Id. at 114, 881 S.E.2d at 174. In Harper I the majority stated that “some
combination” of political science metrics could demonstrate that “there is a significant
likelihood” that a redistricting plan “is presumptively constitutional.” 380 N.C. at
384–85, 868 S.E.2d at 547–48. Specifically, the majority opined that a 1% Mean-
Median Difference and a 7% Efficiency Gap could serve as “possible bright-line
standards” for identifying a plan that “will give the voters of all political parties
substantially equal opportunity to translate votes into seats.” Id. at 385, 868 S.E.2d
at 548.
In Harper II, however, the same majority reversed course and declared that no
combination of political science tests or analysis could adequately identify a
redistricting plan that meets their standard:
Constitutional compliance is not grounded in narrow
statistical measures, but in broad fundamental rights.
Therefore, a trial court reviewing the constitutionality of a
challenged proposed districting plan must assess whether
that plan upholds the fundamental right of the people to
5 The four-justice majority issued its Harper II opinion on 16 December 2022 when it
knew that two members of its majority would complete their terms on this Court just fifteen
days later.
-29-
HARPER V. HALL
Opinion of the Court
vote on equal terms and to substantially equal voting
power. This fundamental right “encompasses the
opportunity to aggregate one’s vote with likeminded
citizens to elect a governing majority of elected officials
who reflect those citizens’ views.” Put differently, it
requires that “voters of all political parties [have]
substantially equal opportunity to translate votes into
seats.”. . .
Although Harper [I] mentions several potential
datapoints that may be used in assessing the
constitutionality of a proposed districting plan, those
measures are not substitutes for the ultimate
constitutional standard noted above. That is, a trial court
may not simply find that a districting plan meets certain
factual, statistical measures and therefore dispositively,
legally conclude based on those measures alone that the
plan is constitutionally compliant. Constitutional
compliance has no magic number. Rather, the trial court
may consider certain datapoints within its wider
consideration of the ultimate legal conclusion: whether the
plan upholds the fundamental right of the people to vote on
equal terms and to substantially equal voting power.
Harper II, 383 N.C. at 114, 881 S.E.2d at 174 (first alteration in original) (citations
omitted). The majority insisted that it could not delineate a particular set of metrics
that would identify a constitutional redistricting map “because our constitution
speaks in broad foundational principles, not narrow statistical calculations.” Id. at
115, 881 S.E.2d at 174.
As a result, the majority implied that the three-judge panel relied too heavily
on its findings regarding the Mean-Median Difference and Efficiency Gap in reaching
-30-
HARPER V. HALL
Opinion of the Court
its ultimate legal conclusions and then “encourage[d] future trial courts . . . to specify
how the evidence does or does not support the plan’s alignment with the broader
constitutional standard of upholding the fundamental right to vote on equal terms.”
Id. at 116, 881 S.E.2d at 175. The majority, however, provided no guidance regarding
what sorts of concrete evidence might assist future trial courts in this endeavor, nor
did the majority explain how to recognize and weigh it.
The Harper II majority then reviewed the three-judge panel’s findings of fact
and conclusions of law for each of the Remedial Plans. First, the majority affirmed
the three-judge panel’s rejection of the RCP and adoption of the Interim
Congressional Plan, holding that the three-judge panel’s conclusions of law were
supported by the relevant findings of fact, which were in turn supported by competent
evidence. Id. at 116−19, 881 S.E.2d at 175−77. Similarly, the majority then affirmed
the three-judge panel’s approval of the RHP, determining that the panel’s conclusions
of law were supported by the relevant findings of fact, which were in turn supported
by competent evidence. Id. at 119−20, 881 S.E.2d at 177−78.
Lastly, the majority reversed the three-judge panel’s approval of the RSP
because, “unlike for the RHP,” the pertinent conclusions of law were not supported
by the relevant findings of fact, and some “findings of fact regarding the RSP . . .
[we]re unsupported by competent evidence.” Id. at 120–21, 881 S.E.2d at 178. As the
-31-
HARPER V. HALL
Opinion of the Court
dissent noted, however, this result was puzzling because on remand, the General
Assembly “made the exact same policy choices and followed the exact same redrawing
process for the RSP as it did for the RHP”; “the Special Masters made almost identical
findings regarding the RHP and the RSP”; and the three-judge-panel made “specific
findings regarding the RSP and RHP [that] were nearly identical.” Id. at 150, 881
S.E.2d at 195−96 (Newby, C.J., dissenting). The dissent highlighted how this
conflicting result, along with other contradictions throughout the Harper II opinion,
demonstrated that the Harper I principles are not grounded in a judicially
discoverable and manageable standard. See id. at 169−70, 881 S.E.2d at 208. The
dissent concluded that in both Harper I and Harper II, the majority “intentionally
stat[ed] vague standards” so that it could remain entrenched in the General
Assembly’s redistricting process and enthrone itself as the final authority over which
plans will be used in North Carolina elections. Id. at 128, 881 S.E.2d at 183.
E. Legislative Defendants’ Petition for Rehearing
This Court filed its Harper II opinion on 16 December 2022, and the mandate
issued on 5 January 2023. On 20 January 2023, Legislative Defendants timely filed
a petition for rehearing under Rule 31 of the North Carolina Rules of Appellate
Procedure. Legislative Defs.’ Pet. for Reh’g, Harper v. Hall, 383 N.C. 89 (2022) (No.
413PA21). Specifically, Legislative Defendants asked this Court to rehear Harper II
-32-
HARPER V. HALL
Opinion of the Court
because it confirms, inter alia, that the standards set forth in both Harper I and
Harper II are unmanageable. As a result, Legislative Defendants requested that this
Court, in rehearing Harper II also revisit Harper I and the issue of whether partisan
gerrymandering claims are justiciable under the North Carolina Constitution. This
Court granted the petition for rehearing on 3 February 2023. Harper v. Hall, ___ N.C.
___, ___, 882 S.E.2d 548, 549−50 (2023) (order granting Legislative Defendants’
petition for rehearing).
II. Rucho v. Common Cause
We begin our analysis with the Supreme Court of the United States’ insightful
and persuasive opinion in Rucho v. Common Cause. In that case the Supreme Court
considered claims that “excessive” partisan gerrymandering violated various
provisions of the Federal Constitution. Rucho, 139 S. Ct. at 2491. There some of the
same plaintiffs in this case challenged North Carolina’s congressional redistricting
map and brought similar claims to those presented here. Specifically, the Rucho
plaintiffs alleged that the challenged plan violated the Equal Protection Clause of the
Fourteenth Amendment by “intentionally diluting the electoral strength of
Democratic voters,” violated their rights to free speech and freedom of association
guaranteed under the First Amendment, exceeded the state legislature’s delegated
authority to prescribe the “Times, Places and Manner of holding Elections,” U.S.
-33-
HARPER V. HALL
Opinion of the Court
Const. art. I, § 4, cl. 1, and “usurped the right of ‘the People’ to elect their preferred
candidates for Congress, in violation of the requirement in Article I, § 2, of the
Constitution that Members of the House of Representatives be chosen ‘by the People
of the several States.’ ”6 Id. at 2492. Accordingly, the Supreme Court was tasked with
deciding whether partisan gerrymandering claims are “ ‘justiciable’—that is, properly
suited for resolution by the federal courts.” Id. at 2491. Ultimately, the Supreme
Court held that partisan gerrymandering claims present nonjusticiable, political
questions. Id. at 2506–07.
The Supreme Court first considered the historical background of partisan
gerrymandering during the formation of our country. Id. at 2494–96. The Supreme
Court noted that partisan gerrymandering existed at the time of our nation’s
founding and that the framers of our Constitution affirmatively considered how to
address it. Id. at 2494. The framers “settled on a characteristic approach, assigning
6 In this case plaintiffs make very similar claims under parallel provisions of our state
constitution—Article I, Section 19 (equal protection), Article I, Section 12 (freedom of
assembly), Article I, Section 14 (freedom of speech), and Article I, Section 10 (free elections).
Harper I, 380 N.C. at 329−31, 868 S.E.2d at 513–14. Common Cause, for example, asserts
that partisan gerrymandering violates our equal protection clause by “diminish[ing] the
electoral power” of members of the Democratic Party, violates Article I, Sections 12 and 14
by burdening Democratic voters’ rights to freedom of speech and freedom to “associate
effectively” with the Democratic Party, and violates the free elections clause by preventing
elections from reflecting the “will of the people.” See Verified Compl. for Declaratory J. and
Injunctive Relief ¶¶ 189, 200, 180, 184, Harper v. Hall, No. 21 CVS 015426, 2021 WL 6884973(N.C. Super. Ct. Wake County Dec. 16, 2021).
-34-
HARPER V. HALL
Opinion of the Court
the issue to the state legislatures, expressly checked and balanced by the Federal
Congress.” Id. at 2496. Specifically, the framers “addressed the election of
Representatives to Congress in the Elections Clause,” which “assigns to state
legislatures the power to prescribe the ‘Times, Places and Manner of holding
Elections’ for Members of Congress, while giving Congress the power to ‘make or alter’
any such regulations.” Id. at 2495. “At no point was there a suggestion that the
federal courts had a role to play. Nor was there any indication that the Framers had
ever heard of courts doing such a thing.” Id. at 2496. The framers could have limited
partisan gerrymandering in the Constitution or assigned federal courts a role in
policing it, but they did not. As a result, the Supreme Court reasoned that “[t]o hold
that legislators cannot take partisan interests into account when drawing district
lines would essentially countermand the Framers’ decision to entrust districting to
political entities,” that is, to state legislatures and to Congress. Id. at 2497.
The Supreme Court distinguished partisan gerrymandering claims from other
types of redistricting claims that courts have historically adjudicated: “In two areas—
one-person, one-vote and racial gerrymandering—our cases have held that there is a
role for the courts with respect to at least some issues that could arise from a State’s
drawing of congressional districts.” Id. at 2495−96. The Court noted, however, that
“[p]artisan gerrymandering claims have proved far more difficult to adjudicate” than
-35-
HARPER V. HALL
Opinion of the Court
other types of redistricting issues because “while it is illegal for a jurisdiction to
depart from the one-person, one-vote rule, or to engage in racial discrimination in
districting, ‘a jurisdiction may engage in constitutional political gerrymandering.’ ”
Id. at 2497 (quoting Hunt v. Cromartie,7 526 U.S. 541, 551, 119 S. Ct. 1545, 1551
(1999)). Because some level of partisan gerrymandering is constitutional, “[t]he
‘central problem’ ” with such claims is not determining whether a jurisdiction has
engaged in any partisan gerrymandering, which is a simple, yes-or-no delineation.
Id. (quoting Vieth v. Jubelirer, 541 U.S. 267, 296, 124 S. Ct. 1769, 1787 (2004)
(plurality opinion)). Rather, the problem with partisan gerrymandering claims is
“determining when political gerrymandering has gone too far.” Id. (quoting Vieth, 541
U.S. at 296, 124 S. Ct. at 1787). That sort of question requires more than a yes-or-no
answer. Instead, it requires “a standard for deciding how much partisan dominance
is too much.” Id. at 2498 (quoting League of United Latin Am. Citizens v. Perry, 548
U.S. 399, 420, 126 S. Ct. 2594, 2611 (2006) (opinion of Kennedy, J.)).
Because of this inherent difficulty, the Supreme Court stressed that if a
standard for resolving such claims exists, it “must be grounded in a ‘limited and
precise rationale’ and be ‘clear, manageable, and politically neutral.’ ” Id. (quoting
7In Hunt v. Cromartie, the Supreme Court addressed a redistricting challenge arising
from North Carolina. See Cromartie, 526 U.S. at 543, 119 S. Ct. at 1547.
-36-
HARPER V. HALL
Opinion of the Court
Vieth, 541 U.S. at 306–08, 124 S. Ct. at 1793 (Kennedy, J., concurring in the
judgment)). Precise constraints on judicial review of partisan gerrymandering claims
are necessary because
“[t]he opportunity to control the drawing of electoral
boundaries through the legislative process of
apportionment is a critical and traditional part of politics
in the United States.” [Davis v.] Bandemer, 478 U.S. [109,]
145, 106 S.Ct. 2797 [(1986)] (opinion of O’Connor, J.). See
Gaffney [v. Cummings], 412 U.S. [735,] 749, 93 S.Ct. 2321[(1973)] (observing that districting implicates
“fundamental ‘choices about the nature of representation’ ”
(quoting Burns v. Richardson, 384 U.S. 73, 92, 86 S.Ct.
1286, 16 L.Ed.2d 376 (1966))). An expansive standard
requiring “the correction of all election district lines drawn
for partisan reasons would commit federal and state courts
to unprecedented intervention in the American political
process,” Vieth, 541 U.S. at 306, 124 S.Ct. 1769 (opinion of
Kennedy, J.).
Id. (first alteration in original). Accordingly, the Supreme Court concluded that
federal courts could “inject [themselves] into [such] heated partisan issues” only if a
standard existed “that c[ould] reliably differentiate unconstitutional from
‘constitutional political gerrymandering.’ ” Id. at 2499 (first quoting Bandemer, 478
U.S. at 145, 106 S. Ct. at 2817 (O’Connor, J., concurring in the judgment); and then
quoting Cromartie, 526 U.S. at 551, 119 S. Ct. at 1551).
The Supreme Court then examined whether it could locate such a standard in
the Federal Constitution. The Court explained that partisan gerrymandering claims
-37-
HARPER V. HALL
Opinion of the Court
are effectively requests for courts to allocate political power to achieve proportional
representation, something that the Federal Constitution does not require:
Partisan gerrymandering claims invariably sound in a
desire for proportional representation. As Justice O’Connor
put it, such claims are based on “a conviction that the greater
the departure from proportionality, the more suspect an
apportionment plan becomes.” [Bandemer, 478 U.S. at 159,
106 S. Ct. 2797.] “Our cases, however, clearly foreclose any
claim that the Constitution requires proportional
representation or that legislatures in reapportioning must
draw district lines to come as near as possible to allocating
seats to the contending parties in proportion to what their
anticipated statewide vote will be.” Id., at 130, 106 S.Ct. 2797(plurality opinion). See Mobile v. Bolden, 446 U.S. 55, 75−76,
100 S.Ct. 1490, 1504, 64 L.Ed.2d 47 (1980) (plurality opinion)
(“The Equal Protection Clause of the Fourteenth Amendment
does not require proportional representation as an imperative
of political organization.”).
Id. at 2499. Accordingly, partisan gerrymandering claims do not seek to redress a
violation of any particular constitutional provisions; rather, such claims “ask the
courts to make their own political judgment about how much representation
particular political parties deserve—based on the votes of their supporters—and to
rearrange the challenged districts to achieve that end.” Id. (first emphasis added).
Essentially, partisan gerrymandering claims ask courts to “apportion political power
as a matter of fairness.” Id. This judgment call is a policy choice. It is not the kind of
“clear, manageable, and politically neutral” standard required for justiciable issues.
Id. at 2498 (quoting Vieth, 541 U.S. at 306–08, 124 S. Ct. at 1793 (Kennedy, J.,
-38-
HARPER V. HALL
Opinion of the Court
concurring in the judgment)); see also Vieth, 541 U.S. at 291, 124 S. Ct. at 1784(plurality opinion) (“ ‘Fairness’ does not seem to us a judicially manageable
standard. . . . Some criterion more solid and more demonstrably met than that seems
to us necessary to enable the state legislatures to discern the limits of their districting
discretion, to meaningfully constrain the discretion of the courts, and to win public
acceptance for the courts’ intrusion into a process that is the very foundation of
democratic decisionmaking.”).
The Court elaborated that settling on a clear, manageable, and politically
neutral test for “fairness” is extremely difficult because “it is not even clear what
fairness looks like in this context.” Rucho, 139 S. Ct. at 2500. Fairness could mean
increasing the number of competitive districts, in which case the appropriate test
would need to accurately identify and “undo packing and cracking so that supporters
of the disadvantaged party have a better shot at electing their preferred candidates.”
Id. This definition of fairness, however, could backfire because “[i]f all or most of the
districts are competitive . . . even a narrow statewide preference for either party
would produce an overwhelming majority for the winning party in the state
legislature.” Id. (alterations in original) (quoting Bandemer, 478 U.S. at 130, 106 S.
Ct. at 2809).
Alternatively, fairness might be measured by the number of “safe seats” each
-39-
HARPER V. HALL
Opinion of the Court
party receives, in which case the appropriate test would actually require packing and
cracking in the redistricting process to ensure each party wins “its ‘appropriate’ share
of ‘safe’ seats.” Id. (citing Bandemer, 478 U.S. at 130–31, 106 S. Ct. at 2809). This
approach, however, reduces the number of competitive districts and produces what
would seem to be an “unfair” result for “individuals in districts allocated to the
opposing party.” Id.Thus, the Supreme Court concluded that
[d]eciding among just these different visions of
fairness . . . poses basic questions that are political, not
legal. There are no legal standards discernible in the
Constitution for making such judgments, let alone limited
and precise standards that are clear, manageable, and
politically neutral. Any judicial decision on what is “fair” in
this context would be an “unmoored determination” of the
sort characteristic of a political question beyond the
competence of the federal courts.
Id. (quoting Zivotofsky v. Clinton, 566 U.S. 189, 196, 132 S. Ct. 1421, 1427 (2012)).
Next, the Supreme Court concluded that, unlike one-person, one-vote claims,
the Federal Constitution is also devoid of any objective, mathematical metric for
measuring political fairness:
the one-person, one-vote rule is relatively easy to
administer as a matter of math. The same cannot be said
of partisan gerrymandering claims, because the
Constitution supplies no objective measure for assessing
whether a districting map treats a political party fairly. It
hardly follows from the principle that each person must
-40-
HARPER V. HALL
Opinion of the Court
have an equal say in the election of representatives that a
person is entitled to have his political party achieve
representation in some way commensurate to its share of
statewide support.
Id. at 2051.
The Court noted that it is possible for a constitution to provide the explicit
guidance necessary to adjudicate partisan gerrymandering claims and pointed to
several state constitutions and state statutes that expressly do so. Id. at 2507–08. By
contrast, the Federal Constitution contains no such provision.
Finding no manageable standard in the Federal Constitution, the Supreme
Court then turned to the political science-based tests proposed by the Rucho
plaintiffs. Id. at 2503−04. The Supreme Court found these were insufficient as well
because they are not effective at predicting future election results:
The [plaintiff]s assure us that “the persistence of a
party’s advantage may be shown through sensitivity
testing: probing how a plan would perform under other
plausible electoral conditions.” Experience proves that
accurately predicting electoral outcomes is not so simple,
either because the plans are based on flawed assumptions
about voter preferences and behavior or because
demographics and priorities change over time. In our two
leading partisan gerrymandering cases themselves, the
predictions of durability proved to be dramatically wrong.
In 1981, Republicans controlled both houses of the Indiana
Legislature as well as the governorship. Democrats
challenged the state legislature districting map enacted by
the Republicans. This Court in Bandemer rejected that
challenge, and just months later the Democrats increased
-41-
HARPER V. HALL
Opinion of the Court
their share of House seats in the 1986 elections. Two years
later the House was split 50−50 between Democrats and
Republicans, and the Democrats took control of the
chamber in 1990. Democrats also challenged the
Pennsylvania congressional districting plan at issue in
Vieth. Two years after that challenge failed, they gained
four seats in the delegation, going from a 12−7 minority to
an 11−8 majority. At the next election, they flipped another
Republican seat.
Even the most sophisticated districting maps cannot
reliably account for some of the reasons voters prefer one
candidate over another, or why their preferences may
change. Voters elect individual candidates in individual
districts, and their selections depend on the issues that
matter to them, the quality of the candidates, the tone of
the candidates’ campaigns, the performance of an
incumbent, national events or local issues that drive voter
turnout, and other considerations. Many voters split their
tickets. Others never register with a political party, and
vote for candidates from both major parties at different
points during their lifetimes. For all of those reasons,
asking judges to predict how a particular districting map
will perform in future elections risks basing constitutional
holdings on unstable ground outside judicial expertise.
Id. (citations omitted).
In conclusion, the Supreme Court held that partisan gerrymandering claims
are nonjusticiable because there is “no plausible grant of authority in the
Constitution and no legal standards to limit and direct [courts’] decisions.” Id. at
2507. In the final words of the opinion, the Supreme Court warned that adjudication
-42-
HARPER V. HALL
Opinion of the Court
of partisan gerrymandering claims would constitute “an unprecedented expansion of
judicial power,” adding that:
We have never struck down a partisan gerrymander as
unconstitutional—despite various requests over the past
45 years. The expansion of judicial authority would not be
into just any area of controversy, but into one of the most
intensely partisan aspects of American political life. That
intervention would be unlimited in scope and duration—it
would recur over and over again around the country with
each new round of districting, for state as well as federal
representatives. Consideration of the impact of today’s
ruling on democratic principles cannot ignore the effect of
the unelected and politically unaccountable branch of the
Federal Government assuming such an extraordinary and
unprecedented role.
Id.
In Rucho the Supreme Court considered partisan gerrymandering claims
under the Federal Constitution, but the arguments it addressed are similar to those
raised here. While the current claims allege that partisan gerrymandering violates
our state constitution, we find the reasoning of the Supreme Court in Rucho
persuasive because the same arguments, concerns, and predictions have arisen here.
Thus, we now turn our analysis to reviewing the applicable fundamental principles
under our state constitution.
-43-
HARPER V. HALL
Opinion of the Court
III. Fundamental Principles
A. Separation of Powers
The separation-of-powers clause is located within the Declaration of Rights of
Article I of our constitution. The Declaration of Rights is an expressive yet non-
exhaustive list of protections afforded to citizens against government intrusion, along
with “the ideological premises that underlie the structure of government.” John V.
Orth & Paul Martin Newby, The North Carolina State Constitution 46 (2d ed. 2013)
[hereinafter State Constitution]. “The abstractness of the Declaration of Rights has
allowed most of it to survive” in our current constitution. Id. at 6. The placement of
the separation-of-powers clause in the Declaration of Rights suggests that keeping
each branch within its described spheres protects the people by limiting overall
governmental power. The clause does not establish the various powers but simply
states that the powers of the branches are “separate and distinct.” N.C. Const. art. I,
§ 6. The constitutional text develops the nature of those powers. State Constitution
46 (“Basic principles, such as popular sovereignty and separation of powers, are first
set out in general terms, to be given specific application in later articles.”). Thus, the
separation-of-powers clause “is to be considered as a general statement of a broad,
albeit fundamental, constitutional principle,” State v. Furmage, 250 N.C. 616, 627,
109 S.E.2d 563, 571 (1959), and must be considered with the related, more specific
-44-
HARPER V. HALL
Opinion of the Court
provisions of the constitution that outline the practical workings for governance, see
N.C. Const. art. II (providing the framework for legislative power); id. art. III
(providing the framework for executive power); id. art. IV (providing the framework
for judicial power). “Nowhere was it stated that the three powers or branches had to
be equal. In fact, although the balance occasionally shifted, the preponderant power
has always rested with the legislature.” State Constitution 50.
Given that “a constitution cannot violate itself,” Leandro v. State, 346 N.C. 336,
352, 488 S.E.2d 249, 258 (1997), a branch’s exercise of its express authority by
definition comports with separation of powers. A violation of separation of powers
only occurs when one branch of government exercises, or prevents the exercise of, a
power reserved for another branch of government. State ex rel. McCrory v. Berger,
368 N.C. 633, 660, 781 S.E.2d 248, 265 (2016) (Newby, J., concurring in part and
dissenting in part). Understanding the prescribed powers of each branch, as divided
between the branches historically and by the text itself, is the basis for stability,
accountability, and cooperation within state government. See State v. Emery, 224
N.C. 581, 584, 31 S.E.2d 858, 861 (1944) (“[Constitutions] should receive a consistent
and uniform construction . . . even though circumstances may have so changed as to
render a different construction desirable.”).
-45-
HARPER V. HALL
Opinion of the Court
Since 1776, our constitutions have recognized that all political power resides
in the people, N.C. Const. art. I, § 2; N.C. Const. of 1868, art. I, § 2; N.C. Const. of
1776, Declaration of Rights, § I, and is exercised through their elected officials in the
General Assembly, N.C. Const. art. II, § 1; N.C. Const. of 1868, art. II, § 1; N.C. Const.
of 1776, § I; State ex rel. Ewart v. Jones, 116 N.C. 570, 570, 21 S.E. 787, 787 (1895).
“The legislative power is vested in the General Assembly, so called because all the
people are present there in the persons of their representatives.” State Constitution
95. Accordingly, the General Assembly possesses plenary power as well as the
responsibilities explicitly recognized in the text of the state constitution. McIntyre v.
Clarkson, 254 N.C. 510, 515, 119 S.E.2d 888, 891−92 (1961). The structure of the
bicameral legislative branch itself diffuses its power, see Berger, 368 N.C. at 653, 781
S.E.2d at 260–61 (Newby, J., concurring in part and dissenting in part), and the
people themselves limit legislative power by express constitutional
restrictions, see Baker v. Martin, 330 N.C. 331, 338–39, 410 S.E.2d 887, 891–92
(1991).
Most accountable to the people, see N.C. Const. art. II, §§ 3, 5, through the most
frequent elections, id. art. II, §§ 2, 4, “[t]he legislative branch of government is
without question ‘the policy-making agency of our government. . . .’ The General
Assembly is the ‘policy-making agency’ because it is a far more appropriate forum
-46-
HARPER V. HALL
Opinion of the Court
than the courts for implementing policy-based changes to our laws,” Rhyne v. K-Mart
Corp., 358 N.C. 160, 169, 594 S.E.2d 1, 8 (2004) (quoting McMichael v. Proctor, 243
N.C. 479, 483, 91 S.E.2d 231, 234 (1956)); see also Berger, 368 N.C. at 653, 781 S.E.2d
at 261 (Newby, J., concurring in part and dissenting in part) (“The diversity within
the [legislative] branch . . . ensures healthy review and significant debate of each
proposed statute, the enactment of which frequently reaches final form through
compromise.”). The constitutional text provides various express checks on legislative
power. See, e.g., N.C. Const. art. II, § 11 (“Neither house shall proceed upon public
business unless a majority of all of its members are actually present.”); id. art. II, § 22
(providing that, with certain exceptions, all bills shall be subject to the Governor’s
veto); id. art. II, § 24 (prohibiting the General Assembly from enacting various types
of “local, private, or special act[s] or resolution[s]”).
B. Standard of Review
Unlike the United States Constitution, the North Carolina Constitution “is in
no matter a grant of power.” McIntyre, 254 N.C. at 515, 119 S.E.2d at 891 (quoting
Lassiter v. Northampton Cnty. Bd. of Elections, 248 N.C. 102, 112, 102 S.E.2d 853,
861 (1958), aff’d, 360 U.S. 45, 79 S. Ct. 985 (1959)). Rather, “[a]ll power which is not
limited by the Constitution inheres in the people.” Id. at 515, 119 S.E.2d at 891(quoting Lassiter, 248 N.C. at 112, 102 S.E.2d at 861). Because the General Assembly
-47-
HARPER V. HALL
Opinion of the Court
serves as “the agent of the people for enacting laws,” it has the presumptive power to
act, State ex rel. Martin v. Preston, 325 N.C. 438, 448, 385 S.E.2d 473, 478 (1989), and
possesses plenary power along with the responsibilities explicitly recognized in the
constitution, McIntyre, 254 N.C. at 515, 119 S.E.2d at 891−92. The General
Assembly’s textual and plenary power is limited only by the express text of the
constitution. Baker, 330 N.C. at 338–39, 410 S.E.2d at 891–92.
Therefore, the idea of the judiciary “preventing . . . the legislature, through
which the people act, from exercising its power is the most serious of judicial
considerations.” Berger, 368 N.C. at 650, 781 S.E.2d at 259 (Newby, J., concurring in
part and dissenting in part). Accordingly, this Court presumes that legislation is
constitutional. Id. at 639, 781 S.E.2d at 252 (majority opinion). A constitutional
limitation upon the General Assembly must be explicit and a violation of that
limitation must be proved beyond a reasonable doubt. Id. at 639, 781 S.E.2d at 252.
A statute cannot abrogate an express provision of the constitution because the
constitution represents the fundamental law and the express will of the people.
Bayard, 1 N.C. (Mart.) at 7. The judiciary performs this role of judicial review by
determining whether a law conflicts with an express provision of the constitution. See
id. at 6.
-48-
HARPER V. HALL
Opinion of the Court
When this Court looks for constitutional limitations on the General Assembly’s
authority, it looks to the plain text of the constitution just as it would look to the plain
text of a statute. State v. Webb, 358 N.C. 92, 97, 591 S.E.2d 505, 510 (2004). Thus, a
claim that a law is unconstitutional must surmount the high bar imposed by the
presumption of constitutionality and meet the highest quantum of proof, a showing
that the statute is unconstitutional beyond a reasonable doubt.8 Baker, 330 N.C. at334–37, 410 S.E.2d at 889–90.
A proper application of this standard of review is illustrated by the landmark
case of Bayard v. Singleton, the first reported case of judicial review in the nation.
Bayard involved judicial review of a statute that conflicted with an express provision
of the 1776 Declaration of Rights. 1 N.C. (Mart.) at 5. In 1785 the General Assembly
enacted a law that abolished the right to a trial by jury for certain property disputes.
Id. At that time, however, the Declaration of Rights expressly provided for a right to
a trial by jury “in all Controversies at Law respecting property.” N.C. Const. of 1776,
Declaration of Rights, § XIV.
8 The majority in Harper I and Harper II and the dissent here largely ignore the well-
established standard of review that our courts apply when reviewing the constitutionality of
a statute. Notably, courts apply different standards of review when adjudicating other
matters that do not involve the constitutionality of a statute.
-49-
HARPER V. HALL
Opinion of the Court
The Court in Bayard held that the act was unequivocally unconstitutional and
void because it directly conflicted with a clear and express provision of the
constitution. Bayard, 1 N.C. (Mart.) at 7. The Court reasoned that the General
Assembly could not “repeal or alter” an express provision of the constitution by
statute because the constitution represents the fundamental law and the express will
of the people. Id. If the General Assembly could violate the constitution in this
manner, it could defy the express will of the people who are the source of all political
power. Id.; see N.C. Const. art. I, § 2. Thus, this Court declared the statute at issue
unconstitutional. Bayard, 1 N.C. (Mart.) at 7.
This Court, however, did not lightly take on the role of declaring an act of the
General Assembly unconstitutional. The Court noted that it felt “great reluctance” in
involving itself “in a dispute with the Legislature” and took “every reasonable
endeavor” to avoid “a disagreeable difference between” the two branches. Id. at 6. But
in this instance, the Court determined that it had to declare the act void because the
constitution was explicit: “That by the Constitution every citizen had undoubtedly a
right to a decision of his property by a trial by jury.” Id. at 7. Accordingly, the holding
of Bayard is clear: the judiciary performs the role of judicial review, but it only
declares an act of the General Assembly void when it directly conflicts with an express
provision of the constitution.
-50-
HARPER V. HALL
Opinion of the Court
Thus, plainly stated and as applied to this case, the standard of review asks
whether the redistricting plans drawn by the General Assembly, which are presumed
constitutional, violate an express provision of the constitution beyond a reasonable
doubt. When we cannot locate an express, textual limitation on the legislature, the
issue at hand may involve a political question that is better suited for resolution by
the policymaking branch. As “essentially a function of the separation of powers,” the
political question doctrine operates to check the judiciary and prevent its encroaching
on the other branches’ authority. Baker v. Carr, 369 U.S. 186, 217, 82 S. Ct. 691, 710
(1962). Under this doctrine, courts must refuse to review political questions, that is,
issues that are better suited for the political branches. Such issues are considered
nonjusticiable.
Prominent on the surface of any case held to involve a
political question is found a textually demonstrable
constitutional commitment of the issue to a coordinate
political department; or a lack of judicially discoverable
and manageable standards for resolving it; or the
impossibility of deciding without an initial policy
determination of a kind clearly for nonjudicial discretion;
or the impossibility of a court’s undertaking independent
resolution without expressing lack of the respect due
coordinate branches of government; or an unusual need for
unquestioning adherence to a political decision already
made; or the potentiality of embarrassment from
multifarious pronouncements by various departments on
one question.
Id. at 217, 82 S. Ct. at 710; see also Bacon v. Lee, 353 N.C. 696, 716–17, 549 S.E.2d
-51-
HARPER V. HALL
Opinion of the Court
840, 854 (2001). Accordingly, out of respect for separation of powers, a court must
refrain from adjudicating a claim when any one of the following is present: (1) a
textually demonstrable commitment of the matter to another branch; (2) a lack of
judicially discoverable and manageable standards; or (3) the impossibility of deciding
a case without making a policy determination of a kind clearly suited for nonjudicial
discretion. All three of these factors are present here.
IV. Political Question
The claims and arguments at issue in this case are the same as those in Rucho,
only this time they arise under the state constitution instead of the Federal
Constitution. The Declaration of Rights provisions invoked by plaintiffs in this case—
the free elections clause, the equal protection clause, and the freedom of speech and
assembly clauses, N.C. Const. art. I, §§ 10, 12, 14, 19,—are our state constitution’s
counterparts to the Federal Constitutional provisions invoked in Rucho—Article I,
Section 4 (Elections Clause); Article I, Section 2 (composition of the U.S. House of
Representatives); the Equal Protection Clause of the Fourteenth Amendment; and
the First Amendment, which protects the rights to free speech and freedom of
association, see Rucho, 139 S. Ct. at 2491. The dissent in Harper I explained in great
detail that, due to the striking similarities between this case and Rucho, we should
have followed the Supreme Court’s guidance and declared plaintiffs’ claims
-52-
HARPER V. HALL
Opinion of the Court
nonjusticiable. See Harper I, 380 N.C. at 414−24, 868 S.E.2d at 566−72 (Newby, C.J.,
dissenting). The dissent in Harper II reiterated that Rucho was persuasive precedent
from our nation’s highest court and illustrated how all of the justiciability pitfalls
warned of in Rucho permeated the remedial proceedings in this case. See Harper II,
383 N.C. at 166−70, 881 S.E.2d at 206−08 (Newby, C.J., dissenting).
Four justices on this Court “misapprehended” the Rucho analysis in Harper I.
See N.C. R. App. P. 31(a). The remedial proceedings at issue in Harper II confirm
that those four justices were wrong to condemn Rucho as inapplicable to the case at
hand. See Harper II, 383 N.C. at 144−66, 881 S.E.2d at 193−206; Harper I, 380 N.C.
at 356−62, 868 S.E.2d at 529−33 (majority opinion). Today we correct that error.
Under the North Carolina Constitution, redistricting is explicitly and exclusively
committed to the General Assembly by the text of the constitution. The executive
branch has no role in the redistricting process, and the role of the judicial branch is
limited by the principles of judicial review. Moreover, like the Federal Constitution,
our constitution does not provide any judicially discernible or manageable standards
for determining how much partisan gerrymandering is too much. See Rucho, 139 S.
Ct. at 2500. Any attempt to adjudicate such claims forces this Court to make
numerous policy determinations for which there is no constitutional guidance. We
are not authorized or equipped to make these determinations. For all of these
-53-
HARPER V. HALL
Opinion of the Court
reasons, we hold that claims of partisan gerrymandering are nonjusticiable, political
questions under the North Carolina Constitution.
A. Textual Commitment
One prominent characteristic of a political question is “a textually
demonstrable constitutional commitment of the issue to a coordinate political
department.” Bacon, 353 N.C. at 717, 549 S.E.2d at 854 (quoting Baker, 369 U.S. at
217, 82 S. Ct. at 710). The text of our state constitution, as well as that of the Federal
Constitution, expressly assigns the task of redistricting9 to the General Assembly.
Reviewing the historical context of our redistricting and elections process is necessary
to properly understand that our state constitution has committed the issue of
redistricting to the General Assembly for hundreds of years.
North Carolina has had some form of elected, representative body since
1665. As early as 1663, the Lords Proprietors could enact laws in consultation with
the freemen settled in their province. Charter Granted by Charles II, King of England
to the Lords Proprietors of Carolina (Mar. 24, 1663), in 1 Colonial and State Records
of North Carolina 20–23 (William L. Sanders ed., 1886) [hereinafter 1 Colonial and
State Records]. In 1665 certain “concessions” by the Lords Proprietors allowed for the
9 “Districting” and “redistricting” are sometimes referred to as “apportionment” and
“reapportionment.”
-54-
HARPER V. HALL
Opinion of the Court
formation of the predecessor to the General Assembly and the election of freemen
representatives. Concessions and Agreement Between the Lords Proprietors of
Carolina and William Yeamans, et al. (Jan. 7, 1665), in 1 Colonial and State Records
79–81. The 1669 Fundamental Constitutions of Carolina apportioned those
representatives into counties and the counties into precincts. The Fundamental
Constitutions of Carolina (Mar. 1, 1669), in 1 Colonial and State Records 188. The
assembly met and stood for election every two years. Id. at 199–200. Thus, long before
the 1776 constitution, the qualified voters in Carolina were electing their
representatives in districts.
Leading up to the enactment of the 1776 constitution, in 1774 the delegates of
the First Provincial Congress were elected by geographic location, either by town,
which were also known as boroughs, or by county. See Henry G. Connor & Joseph B.
Cheshire, Jr., The Constitution of North Carolina Annotated xii–xiv (1911). The text
of the 1776 constitution established the General Assembly, a gathering of the people
through their elected representatives, as the Senate and the House of Commons. N.C.
Const. of 1776, § I. Senators were elected annually by county without regard to the
population size of that county. Id. § II. Representatives in the House of Commons
were also elected annually, but each county received two representatives and certain
enumerated towns received one as well. Id. § III. Only six towns were initially given
-55-
HARPER V. HALL
Opinion of the Court
separate representation in the House of Commons, id., but other towns were later
added. The 1776 constitution did not contain a specific provision regarding
redistricting. Nonetheless, redistricting occurred through the creation of new
counties—as part of its plenary power, the General Assembly established the
boundaries of the counties from which Senators and Representatives were elected.
See, e.g., Act of Apr. 8, 1777, An Act for dividing Rowan County, and other Purposes
therein mentioned, ch. XIX, 1777 N.C. Sess. Laws 33 (dividing Rowan County to carve
out a new Burke County). Notably, the 1776 Declaration of Rights contained the free
elections and freedom of assembly clauses. N.C. Const. of 1776, Declaration of Rights,
§§ VI, XVIII.
Through the years, the population of the state shifted radically from the east
to the piedmont and west. John V. Orth, North Carolina Constitutional History, 70
N.C. L. Rev. 1759, 1770–71 (1992) [hereinafter Constitutional History]. Nonetheless,
the eastern region received additional representation through the strategic creation
and division of counties. Id. at 1770. The General Assembly created smaller counties
in the east and larger ones in the piedmont and west, keeping the distribution of
representatives in favor of the east despite population growth trends in other
areas. Id. This county-town approach, combined with the power of the General
Assembly to divide existing counties to create new ones, resulted in superior political
-56-
HARPER V. HALL
Opinion of the Court
power in the east. See id. This malapportionment led to civil unrest and a crisis that
culminated with the 1835 constitutional convention. State Constitution 3, 13. During
that time, no one argued that the provisions of the Declaration of Rights or the 1776
constitution made the legislative apportionment acts unconstitutional. Rather, North
Carolinians ultimately recognized the need to amend the text itself to address the
apportionment problem.
In 1835 a constitutional convention met to, among other things, change the
representative system to better address differences in population. See id. That
convention resulted in amendments that provided for a total of fifty senators and
required senatorial districts to be drawn by the General Assembly based on the taxes
paid by each county. N.C. Const. of 1776, amends. of 1835, art. I, § 1. These
amendments also included the predecessor of the WCP, see N.C. Const. art. II, § 3(3),
that prohibited a county from being divided to create the senatorial districts, N.C.
Const. of 1776, amends. of 1835, art. I, § 1.
The 1835 amendments provided for 120 House seats. Id. art. I, § 2. These
amendments eliminated representation for the borough towns, see generally id.,
instead allotting all 120 House seats to counties based roughly on population, id. This
framework allowed the more populated counties to have additional representatives,
but each county was entitled to at least one representative. Id. These amendments
-57-
HARPER V. HALL
Opinion of the Court
alleviated the problem of disproportionate representation in the eastern counties. The
General Assembly was instructed to reconsider the apportionment of the counties
every twenty years and to base reapportionment on population according to the
census taken by order of Congress. Id. art. I, § 3. Likewise, the convention
implemented other changes to representation such as lengthening legislative terms
from one year to two years, id. art. I, §§ 1–2, and allowing the voters to elect the
governor, id. art. II, § 1.
Following the constitutional convention of 1868, the Senate became
apportioned by population. N.C. Const. of 1868, art. II, § 5. Along with the express
limitation imposed by the WCP, the 1868 amendments required senatorial districts
to be contiguous and to be redrawn in connection with the decennial
census. Id. Apportionment of House seats remained the same—allotted to counties
based on population with each county given at least one representative. Id. art. II,
§ 6. The convention lengthened the term of the governor to four years, id. art. III, § 1,
and constitutionally created a separate judicial branch, see id. art. IV, with judges
being elected by the voters for eight-year terms, id. art. IV, § 26. Previously, the
General Assembly elected judges, N.C. Const. of 1776, § XIII, but now judges in North
Carolina became directly accountable to the people through elections, N.C. Const. of
1868, art. IV, § 26.
-58-
HARPER V. HALL
Opinion of the Court
For almost one hundred years, apportionment remained unchanged until the
1960s. At that time, the Speaker of the House received the authority to apportion
House districts. N.C. Const. of 1868, amends. of 1961, art. II, § 5. Then, to comply
with the federal decision in Baker v. Carr, the constitution was amended in 1968 to
reflect the one-person, one-vote requirement. State Constitution 31. This change
affected the structure of the House of Representatives in particular. Id. Significantly,
the number of House members remained at 120, but the representatives were no
longer apportioned by county; instead, the 120 representatives were allotted among
districts now drawn based on equal population. N.C. Const. of 1868, amends. of
1967, art. II, § 5. By the end of the 1960s, the same criteria for proper districts—equal
population, contiguous territory, the WCP, and reapportionment in conjunction with
the decennial census—applied to both Senate and House districts. See id. art. II, §§ 4,
6.
The current version of our constitution, ratified by the people at the ballot box
in 1970, took effect in 1971 and came about as a “good government measure.” State
Constitution 32. This 1971 constitution represented an attempt to modernize the
1868 constitution and its subsequent amendments with editorial and organizational
revisions and amendment proposals. See, e.g., N.C. State Const. Study Comm’n,
Report of the North Carolina State Constitution Study Commission 8–12 (1968).
-59-
HARPER V. HALL
Opinion of the Court
Today our constitution expressly assigns the legislative redistricting authority to the
General Assembly subject to specific enumerated restraints:
The Senators shall be elected from districts. The
General Assembly, at the first regular session convening
after the return of every decennial census of population
taken by order of Congress, shall revise the senate districts
and the apportionment of Senators among those districts,
subject to the following requirements:
(1) Each Senator shall represent, as nearly as may
be, an equal number of inhabitants, the number of
inhabitants that each Senator represents being
determined for this purpose by dividing the population of
the district that he represents by the number of Senators
apportioned to that district;
(2) Each senate district shall at all times consist of
contiguous territory;
(3) No county shall be divided in the formation of a
senate district;
(4) When established, the senate districts and the
apportionment of Senators shall remain unaltered until
the return of another decennial census of population taken
by order of Congress.
N.C. Const. art. II, § 3. Article II, Section 5 establishes the same grant of authority
and limitations for the state House of Representatives. Thus, while the constitution
commits the redistricting responsibility to the General Assembly, it does not leave
the General Assembly completely unrestrained. The constitution expressly requires
that any redistricting plan conform to its explicit criteria.
-60-
HARPER V. HALL
Opinion of the Court
Notably, there is no provision in the state constitution regarding redistricting
of congressional districts. The Federal Constitution, however, commits drawing of
congressional districts to the state legislatures subject to oversight by the Congress
of the United States. “The Times, Places and Manner of holding Elections for
Senators and Representatives, shall be prescribed in each State by the Legislature
thereof; but the Congress may at any time by Law make or alter such Regulations,
except as to the Places of chusing Senators.” U.S. Const. art. I, § 4, cl. 1. This provision
makes clear that the redistricting power is expressly committed to the state
legislative branch.
Additionally, both our constitution and the General Statutes expressly insulate
the redistricting power from intrusion by the executive and judicial branches. The
governor has no role in the redistricting process because the constitution explicitly
exempts redistricting legislation from the governor’s veto power.10 N.C. Const. art. II,
§ 22(5)(b)−(d). Moreover, the General Statutes provide a limited role of judicial review
for courts in reviewing redistricting plans. See N.C.G.S. §§ 120-2.3 to -2.4 (2021). The
10 The North Carolina governor did not gain the veto power until the people approved
an amendment to the North Carolina Constitution in 1996—over two hundred years after
the adoption of our first constitution in 1776. See Act of Mar. 8, 1995, An Act to Provide For
A Referendum to Amend the Constitution to Provide for a Gubernatorial Veto, ch. 5, 1995
N.C. Sess. Laws 6. At that time, the people of North Carolina extended to the governor the
authority to veto many types of legislative enactments but specifically withheld the authority
to veto redistricting legislation. Id. That provision remains unchanged today.
-61-
HARPER V. HALL
Opinion of the Court
General Assembly enacted these statutory provisions in 2003 to clarify and codify the
existing process by which courts already had been reviewing redistricting plans. Act
of Nov. 25, 2003, An Act to Establish House Districts, Establish Senatorial Districts,
and Make Changes to the Election Laws and to Other Laws Related to Redistricting,
S.L. 2003-434, §§ 7−9, 2003 N.C. Sess. Laws (1st Extra Sess. 2003) 1313, 1415−16.
The General Assembly drafted these statutes in response to this Court’s decisions in
Stephenson I, 355 N.C. 354, 562 S.E.2d 377, and Stephenson v. Bartlett (Stephenson
II), 357 N.C. 301, 582 S.E.2d 247 (2003). This Court unanimously upheld these
statutory provisions as proper limitations on the judiciary’s role in the redistricting
process in Stephenson v. Bartlett (Stephenson III), 358 N.C. 219, 230, 595 S.E.2d 112,
119−20 (2004) (“[R]edistricting is a legislative responsibility . . . . Not only do these
statutes allow the General Assembly to exercise its proper responsibilities, they
decrease the risk that the courts will encroach upon the responsibilities of the
legislative branch.”).
Section 1-267.1 requires that a three-judge panel hear challenges to
redistricting plans. N.C.G.S. § 1-267.1 (2021). Specifically, under Section 120-2.3,
courts may review challenges regarding whether a redistricting plan is
“unconstitutional or otherwise invalid.” Id. § 120-2.3. If a court finds a redistricting
plan is unconstitutional, it must specify the precise defects and give the General
-62-
HARPER V. HALL
Opinion of the Court
Assembly an opportunity to remedy any identified defect by enacting a new
redistricting plan. Id. § 120-2.4(a). By statute, a court may not impose a remedial
redistricting plan of its own unless “the General Assembly does not act to remedy”
those defects. Id. § 120-2.4(a1). Even then, a court-imposed redistricting plan may
differ from the General Assembly’s enacted plan “only to the extent necessary to
remedy” the defects identified by the court and will only be used for the next general
election. Id. After the next general election, the General Assembly will replace the
court-imposed map with a new, legislatively enacted map. A court-imposed map is
only used for one election cycle because it is not “established” as that term is used in
Article II, Sections 3(4) and 5(4). See N.C. Const. art. II, §§ 3(4), 5(4) (“When
established, the senate [and representative] districts and the apportionment of
Senators [and Representatives] shall remain unaltered until the return of another
decennial census of population taken by order of Congress.”). This limited role of
judicial review comports with the fact that our constitution expressly assigns the
redistricting authority to the General Assembly. See Stephenson III, 358 N.C. at 230,
595 S.E.2d at 119.
Article II, Sections 3 and 5 commit the redistricting authority to the General
Assembly and set express limitations on that authority. In the landmark case
Stephenson I, this Court considered the express limitations on redistricting in Article
-63-
HARPER V. HALL
Opinion of the Court
II, Sections 3 and 5, and applied them in conformity with federal law. See Stephenson
I, 355 N.C. at 358, 562 S.E.2d at 381. That case dealt with the interplay between the
objective restraints contained in the state constitution and federal redistricting
authorities—namely, Section 2 of the Voting Rights Act (VRA) and the one-person,
one-vote principle.11 See id. at 359, 562 S.E.2d at 382.
The plaintiffs challenged the 2001 state legislative redistricting plans (2001
Plans) as unconstitutional in violation of the WCP of Article II, Sections 3 and 5. Id.
at 358, 562 S.E.2d at 381; N.C. Const. art. II, §§ 3, 5 (“No county shall be divided in
the formation of a senate [or representative] district.”). The defendants argued that
these constitutional provisions were “wholly unenforceable because of the
requirements of the [VRA].” Stephenson I, 355 N.C. at 361, 562 S.E.2d at 383−84.
Thus, before addressing whether the 2001 redistricting plans violated the WCP, this
Court first had to address “whether the WCP is now entirely unenforceable, as [the]
defendants contend, or, alternatively, whether the WCP remains enforceable
11 “Section 2 of the VRA generally provides that states or their political subdivisions
may not impose any voting qualification or prerequisite that impairs or dilutes, on account
of race or color, a citizen’s opportunity to participate in the political process and to elect
representatives of his or her choice.” Id. at 363, 562 S.E.2d at 385. The one-person, one-vote
principle simply requires that districts, to the extent practicable, contain an equal number of
voters. Brown v. Thomson, 462 U.S. 835, 841, 103 S. Ct. 2690, 2695 (1983).
-64-
HARPER V. HALL
Opinion of the Court
throughout the State to the extent not preempted or otherwise superseded by federal
law.” Id. at 369, 562 S.E.2d at 388. In doing so, we explained that
an inflexible application of the WCP is no longer attainable
because of the operation of the provisions of the VRA and
the federal “one-person, one-vote” standard, as
incorporated within the State Constitution. This does not
mean, however, that the WCP is rendered a legal nullity if
its beneficial purposes can be preserved consistent with
federal law and reconciled with other state constitutional
guarantees.
. . . The General Assembly may consider partisan
advantage and incumbency protection in the application of
its discretionary redistricting decisions, see Gaffney v.
Cummings, 412 U.S. 735, [93 S. Ct. 2321,] 37 L. Ed. 2d 298(1973), but it must do so in conformity with the State
Constitution. To hold otherwise would abrogate the
constitutional limitations or “objective constraints” that
the people of North Carolina have imposed on legislative
redistricting and reapportionment in the State
Constitution.
Id. at 371–72, 562 S.E.2d at 389–90. In other words, we recognized that the WCP is
one of the clear and express limitations or “objective constraints” on legislative
redistricting in our constitution. Id. at 371, 562 S.E.2d at 390. We concluded that the
WCP was enforceable to the extent it did not conflict with the one-person, one-vote
principle or the VRA because “the people of North Carolina” expressly chose to limit
the General Assembly in this way. Id. at 371, 374−75, 562 S.E.2d at 390, 391−92; id.at 372−74, 562 S.E.2d at 390−91 (“[T]he WCP remains valid and binding upon the
-65-
HARPER V. HALL
Opinion of the Court
General Assembly during the redistricting and reapportionment process . . . except to
the extent superseded by federal law. . . . Where . . . the primary purpose of the WCP
can be effected to a large degree without conflict with federal law, it should be adhered
to by the General Assembly to the maximum extent possible.”).
Notably, we stated that “[t]he General Assembly may consider partisan
advantage and incumbency protection in the application of its discretionary
redistricting decisions.” Id. at 371, 562 S.E.2d at 390. We supported this statement
with a citation to the Supreme Court’s decision in Gaffney v. Cummings, 412 U.S.
735, 93 S. Ct. 2321 (1973). In that case the Supreme Court observed that
[i]t would be idle, we think, to contend that any political
consideration taken into account in fashioning a
reapportionment plan is sufficient to invalidate it. Our
cases indicate quite the contrary. The very essence of
districting is to produce a different—a more “politically
fair”—result than would be reached with elections at large,
in which the winning party would take 100% of the
legislative seats. Politics and political considerations are
inseparable from districting and apportionment.
Id. at 752−53, 93 S. Ct. at 2331 (emphasis added) (citations omitted). Thus, in
Stephenson I we recognized that partisan considerations are inherently a part of the
redistricting process in our state. We then expressed that the discretionary
consideration of partisan advantage and incumbency protection must be done “in
conformity with the State Constitution.” Stephenson I, 355 N.C. at 371, 562 S.E.2d at
-66-
HARPER V. HALL
Opinion of the Court
390. In other words, the General Assembly’s discretionary considerations are
constrained by the express limitations found in Article II, Sections 3 and 5. “To hold
otherwise,” we explained, “would abrogate the constitutional limitations or ‘objective
constraints’ that the people of North Carolina have imposed on legislative
redistricting and reapportionment in the State Constitution.” Id. at 371−72, 562
S.E.2d at 390. By “constitutional limitations,” we meant the specific constraints in
Article II, Sections 3 and 5.
Having held that the WCP remained enforceable to the extent not preempted
by or otherwise superseded by federal law, we then held that the 2001 Plans violated
the WCP by unduly dividing numerous counties. Id. at 371, 562 S.E.2d at 389–90.
Specifically, the 2001 Plans divided fifty-one of the State’s one hundred counties in
the Senate plan and seventy of the one hundred counties in the House plan. Id. at
360, 562 S.E.2d at 383. We were able to make this determination because the
standard provided by the WCP is express, clear, and easily applied.
Once we found that the 2001 Plans violated the still-valid WCP, we then
crafted detailed criteria harmonizing the WCP and the other express constraints in
Article II, Sections 3 and 5, with the VRA and the federal one-person, one-vote
principle. These standards were clear and manageable because they were based on
-67-
HARPER V. HALL
Opinion of the Court
the express provisions found in our constitution or in federal law. For example, one
of the Stephenson I criteria required that
[i]n counties having a non-VRA population pool
which cannot support at least one legislative district at or
within plus or minus five percent of the ideal population for
a legislative district or, alternatively, counties having a
non-VRA population pool which, if divided into districts,
would not comply with the at or within plus or minus five
percent “one-person, one-vote” standard, the requirements
of the WCP are met by combining or grouping the
minimum number of whole, contiguous counties necessary
to comply with the at or within plus or minus five-percent
“one-person, one vote” standard.
Id. at 383−84, 562 S.E.2d at 397 (emphasis added). The requirement that the General
Assembly group “whole, contiguous” counties together when necessary to create a
district that meets the ideal population requirement is a function of the WCP and the
requirement that “[e]ach [legislative] district shall at all times consist of contiguous
territory.” N.C. Const. art. II, §§ 3(3), 5(3), 3(2), 5(2). Similarly, this Court recognized
that when the General Assembly must group counties together in this way, the
resulting districts in that county grouping might cross over the “interior county
lines”—that is, the county lines that do not create the exterior boundaries of the
county grouping. See Stephenson I, 355 N.C. at 384, 562 S.E.2d at 397. Such
crossovers would violate the WCP but may be necessary to comply with the one-
person, one-vote principle. Thus, in order to enforce “[t]he intent underlying the
-68-
HARPER V. HALL
Opinion of the Court
WCP . . . to the maximum extent possible,” Stephenson I required that districts in
multi-county groupings be “compact” and account for “communities of interest.”12 Id.
at 384, 562 S.E.2d at 397. Compactness and communities of interest are also
important factors under the VRA. See Thornburg v. Gingles, 478 U.S. 30, 50−51, 106
S. Ct. 2752, 2766 (1986).
Stephenson I also required that “[i]n forming new legislative districts, any
deviation from the ideal population for a legislative district shall be at or within plus
or minus five percent for purposes of compliance with federal ‘one-person, one-vote’
requirements.” 355 N.C. at 383, 562 S.E.2d at 397. This requirement is “relatively
easy to administer as a matter of math.”13 Rucho, 139 S. Ct. at 2501. This requirement
12 The Court in Stephenson I recognized that the “impetus” underlying the WCP was
a long-standing respect for counties as “political subdivisions” that “provide essential
services” and “ ‘effectuate the political organization and civil administration of the state’ ” at
the local level. Id. at 365−66, 562 S.E.2d at 385−86 (quoting White v. Comm’rs of Chowan
Cnty., 90 N.C. 437, 438 (1884)). Accordingly, counties were kept whole because they naturally
promote a “clear identity and common interests” among county residents. Id. at 366, 562
S.E.2d at 386. Recognizing that some counties would need to be divided or grouped together
to comply with federal redistricting requirements, and in order to comply with the underlying
intent of the WCP “to the maximum extent possible,” id. at 384, 562 S.E.2d at 397,
Stephenson I required the General Assembly to consider compactness and communities of
interest whenever it had to group multiple counties together.
13 Stephenson I’s plus or minus five percent standard is derived directly from Supreme
Court precedent holding that a population deviation range of ten percent (plus or minus five
percent) generally satisfies the federal one-person, one-vote requirement. See Brown, 462
U.S. at 842, 103 S. Ct. at 2696 (“ ‘[M]inor deviations from mathematical equality among state
legislative districts are insufficient to make out a prima facie case of invidious discrimination
under the Fourteenth Amendment . . . .’ Our decisions have established, as a general matter,
that an apportionment plan with a maximum population deviation under 10% falls within
-69-
HARPER V. HALL
Opinion of the Court
also ensures compliance with Article II, Sections 3(1) and 5(1), which provide that
each senator and representative “shall represent, as nearly as may be, an equal
number of inhabitants.” N.C. Const. art. II, §§ 3(1), 5(1).
Although this Court was very detailed in stating its Stephenson I criteria, each
criterion clearly reflects the fact that the constitution textually commits the
redistricting authority to the General Assembly and only limits that authority in the
ways enumerated in federal law and in Article II, Sections 3 and 5. This Court
harmonized federal redistricting requirements and the directives of our state
constitution, but it did not place any limitations on redistricting that were not derived
from those two sources of law.
In sum, throughout our history our constitutions have invariably committed
redistricting authority to our General Assembly. The General Assembly exercises
that authority subject to the express limitations in our constitution and in federal
this category of minor deviations.” (internal citations omitted) (quoting Gaffney, 412 U.S. at
745, 93 S. Ct. at 2327)); see also Harris v. Ariz. Indep. Redistricting Comm’n, 578 U.S. 253,
259, 136 S. Ct. 1301, 1307 (2016) (“We have further made clear that ‘minor deviations from
mathematical equality’ do not, by themselves, ‘make out a prima facie case of invidious
discrimination under the Fourteenth Amendment . . . .’ We have defined as ‘minor deviations’
those in ‘an apportionment plan with a maximum population deviation under 10%.’ ”
(internal citations omitted) (first quoting Gaffney, 412 U.S. at 745, 93 S. Ct. at 2327; and then
quoting Brown, 462 U.S. at 842, 103 S. Ct. at 2696)); Evenwel v. Abbott, 578 U.S. 54, 59−60,
136 S. Ct. 1120, 1124 (2016) (same); Voinovich v. Quilter, 507 U.S. 146, 160−61, 113 S. Ct.
1149, 1159 (1993) (same); Connor v. Finch, 431 U.S. 407, 418, 97 S. Ct. 1828, 1835 (1977)
(same).
-70-
HARPER V. HALL
Opinion of the Court
law. When the General Assembly acts within the scope of these express limitations,
it is performing its constitutionally assigned role. When the General Assembly
properly performs its constitutionally assigned role, its discretionary decisions
present a political question that is nonjusticiable. Ultimately, the role of our courts
is limited to identifying a redistricting plan that violates those express limitations
and requiring the General Assembly to remedy the specified defects.
B. Judicially Discoverable and Manageable Standards
Another factor that indicates the presence of a political question is the lack of
a judicially discoverable and manageable standard for assessing the matter at hand.
Like the Federal Constitution, our constitution does not provide judicially discernible
or manageable standards for adjudicating partisan gerrymandering claims. The
North Carolina Constitution could contain a provision that expressly prohibits or
limits partisan gerrymandering, and perhaps then our courts could be “armed with a
standard that can reliably differentiate” between constitutional and unconstitutional
partisan gerrymandering. See Rucho, 139 S. Ct. at 2499. Our constitution, however,
contains no such provision.
Almost one hundred years ago, this Court’s opinion in Leonard v. Maxwell
indicated that courts should cautiously consider redistricting claims. 216 N.C. 89, 99,
3 S.E.2d 316, 324 (1939). In that case the plaintiff argued that the General Assembly
-71-
HARPER V. HALL
Opinion of the Court
was malapportioned because it had not reapportioned itself at the first session after
the 1930 census, as required by the constitution. Id. at 98, 3 S.E.2d at 324. As a result,
the plaintiff argued that the 1937 General Assembly was powerless to act including,
“it [wa]s suggested,” to reapportion itself. Id. This Court rejected that argument,
observing that “[t]he question is a political one, and there is nothing the courts can
do about it. [Courts] do not cruise in nonjusticiable waters.” Id. at 99, 3 S.E.2d at 324(internal citation omitted).
Moreover, this Court has previously recognized that the Declaration of Rights
generally does not provide judicially manageable standards for claims related to
gerrymandering. In Dickson I a group of North Carolina voters challenged
redistricting plans passed by the General Assembly in 2011 (2011 Plans) under both
federal and state law. Dickson I, 367 N.C. at 546, 766 S.E.2d at 242, vacated and
remanded on federal grounds, 575 U.S. 959 (2015) (mem.). Among other claims, the
plaintiffs argued that the 2011 Plans violated the “ ‘Good of the Whole’ clause found
in Article I, Section 2” of the North Carolina Constitution’s Declaration of Rights. Id.
at 575, 766 S.E.2d at 260. Article I, Section 2 states:
All political power is vested in and derived from the
people; all government of right originates from the people,
is founded upon their will only, and is instituted solely for
the good of the whole.
-72-
HARPER V. HALL
Opinion of the Court
N.C. Const. art. I, § 2. The plaintiffs argued that the last clause of this provision
constitutes “a specific limitation on the powers of the General Assembly with regard
to redistricting” because the General Assembly “ ‘institutes’ a new form of
government” when it reapportions the legislative districts after every decennial
census. Pl.-Appellants’ Br. at 178−79, Dickson I, No. 201PA12-2, 2013 5669654 (N.C.
Sup. Ct. Oct. 11, 2013).
This Court rejected that claim as nonjusticiable, however, determining that
Article I, Section 2 of the Declaration of Rights did not provide a judicially
manageable standard:
We do not doubt that plaintiffs’ proffered maps represent
their good faith understanding of a plan that they believe
best for our State as a whole. However, the maps enacted
by the duly elected General Assembly also represent an
equally legitimate understanding of legislative districts
that will function for the good of the whole. Because
plaintiffs’ argument is not based upon a justiciable
standard, and because acts of the General Assembly enjoy
“a strong presumption of constitutionality,” Pope v. Easley,
354 N.C. 544, 546, 556 S.E.2d 265, 267 (2001) (per curiam)
(citation omitted), plaintiffs’ claims fail.
Dickson I, 367 N.C. at 575, 766 S.E.2d at 260. We affirmed the trial court’s conclusion
that “the General Assembly applied traditional and permissible redistricting
principles to achieve partisan advantage and that no constitutional violations
-73-
HARPER V. HALL
Opinion of the Court
resulted.” Id. at 546, 766 S.E.2d at 242. Notably, the trial court in that case
specifically stated that partisan gerrymandering is nonjusticiable:
Redistricting in North Carolina is an inherently political
and intensely partisan process that results in political
winners and, of course, political losers. . . .
Political losses and partisan disadvantage are not the
proper subject for judicial review, and those whose power
or influence is stripped away by shifting political winds
cannot seek a remedy from courts of law, but they must
find relief from courts of public opinion in future elections.
Our North Carolina Supreme Court has observed that
“[w]e do not believe the political process is enhanced if the
power of the courts is consistently invoked to second-guess
the General Assembly’s redistricting decisions.”
Dickson v. Rucho, Nos. 11 CVS 16896, 11 CVS 16940, 2013 WL 3376658, at *1−2
(N.C. Super. Ct. Wake County July 8, 2013) (quoting Pender County v. Bartlett, 361
N.C. 491, 506, 649 S.E.2d 364, 373 (2007), aff’d sub nom. Bartlett v. Strickland, 556
U.S. 1, 129 S. Ct. 1231 (2009)). We affirmed the trial court’s analysis. See Dickson I,
367 N.C. at 575, 766 S.E.2d at 260; see also Dickson v. Rucho (Dickson II), 368 N.C.
481, 534, 781 S.E.2d 404, 440–41 (2015) (reiterating our prior holding that Article I,
Section 2 of the North Carolina Declaration of Rights does not provide a justiciable
standard).
The four-justice majority in Harper I should have followed the analysis in
Dickson I. Nevertheless, the Harper I majority departed from this precedent and
-74-
HARPER V. HALL
Opinion of the Court
insisted that our Declaration of Rights plainly provides a standard for identifying
partisan gerrymandering. Even within that opinion, however, the majority could not
consistently enunciate what that standard supposedly is. The Court described a
“constitutional right[ ] of the people to vote on equal terms and to substantially equal
voting power,” as well as an “individual right[ ] of voters to cast votes that matter
equally.” Harper I, 380 N.C. at 323–24, 868 S.E.2d at 510. The Harper I majority also
stated that the constitution protects “the opportunity to aggregate one’s vote with
likeminded citizens to elect a governing majority of elected officials who reflect those
citizens’ views.” Id. at 378, 868 S.E.2d at 544. In another part of the Harper I opinion,
the majority noted a districting plan violates the constitution when it “systematically
makes it harder for one group of voters to elect a governing majority than another
group of voters of equal size.” Id. at 379, 868 S.E.2d at 544. In other parts of Harper
I, however, the majority characterized the standard as a right to aggregate votes “on
the basis of partisan affiliation.” Id. at 390, 392, 868 S.E.2d at 551, 552.
These vague and inconsistent standards are not derived from any express
provision in the constitution. Instead, these standards seem to be grounded in a
desire for some form of proportionality and reflect a judicially created notion of how
much representation is “fair” without explaining what fairness is or how to manage
it. The Supreme Court reached the same conclusion regarding the claims in Rucho:
-75-
HARPER V. HALL
Opinion of the Court
Partisan gerrymandering claims rest on an instinct
that groups with a certain level of political support should
enjoy a commensurate level of political power and
influence. Explicitly or implicitly, a districting map is
alleged to be unconstitutional because it makes it too
difficult for one party to translate statewide support into
seats in the legislature. . . .
Partisan gerrymandering claims invariably sound in
a desire for proportional representation. As Justice
O’Connor put it, such claims are based on “a conviction that
the greater the departure from proportionality, the more
suspect an apportionment plan becomes.”
Rucho, 139 S. Ct. at 2499 (quoting Bandemer, 478 U.S. at 159, 106 S. Ct. at 2824
(O’Connor, J., concurring in the judgment)). These vague notions of fairness do not
answer how to measure whether groups of voters are treated “fairly” or how to predict
the results an election would produce. Moreover, as forewarned by the Supreme Court
in Rucho, these vague notions of fairness did not produce a discernable or workable
standard during the remedial proceedings in this case. See id. at 2499–500
(“ ‘Fairness’ does not seem to us a judicially manageable standard . . . . Some criterion
more solid and more demonstrably met than that seems to us necessary to enable the
state legislatures to discern the limits of their districting discretion [and] to
meaningfully constrain the discretion of the courts . . . .”(first alteration in original)
(quoting Vieth, 541 U.S. at 291, 124 S. Ct. at 1784 (plurality opinion))).
In the remedial phase, the General Assembly attempted to apply the Harper I
-76-
HARPER V. HALL
Opinion of the Court
standard in drawing the Remedial House Plan (RHP), Remedial Senate Plan (RSP),
and Remedial Congressional Plan (RCP). The General Assembly followed the same
process in enacting each plan, yet the Special Masters recommended, and the three-
judge panel concluded, that only the RHP and RSP met the Harper I standard.
Accordingly, the three-judge panel struck the RCP. On appeal, however, the same
four justices from Harper I also struck the RSP as unconstitutional, see Harper II,
383 N.C. at 94, 881 S.E.2d at 162, indicating that neither the General Assembly, the
three-judge panel, the three Special Masters, nor three justices of this Court could
properly understand and apply their standard set forth in Harper I. Constitutional
compliance should not be so difficult. See Rucho, 139 S. Ct. at 2499 (noting that courts
can only adjudicate partisan gerrymandering claims if they are “armed with a
standard that can reliably differentiate unconstitutional from ‘constitutional political
gerrymandering.’ ” (quoting Cromartie, 526 U.S. at 551, 119 S. Ct. 1545)).
The four-justice majority in Harper I did not explain what its standard means
or how it could be reliably met because it could not answer basic questions like how
much partisan gerrymandering is too much and how can courts consistently and
reliably measure partisanship in a redistricting plan. See Harper I, 380 N.C. at 384,
868 S.E.2d at 547 (“We do not believe it prudent or necessary to, at this time, identify
an exhaustive set of metrics or precise mathematical thresholds which conclusively
-77-
HARPER V. HALL
Opinion of the Court
demonstrate or disprove the existence of an unconstitutional partisan
gerrymander.”). Nevertheless, just as the plaintiffs in Rucho argued, see Rucho, 139
S. Ct. at 2503, the Harper I majority indicated that political science metrics could
serve as “possible bright-line standards” for measuring partisan fairness. 380 N.C. at
385–86, 868 S.E.2d at 548 (stating that “a [M]ean-[M]edian [D]ifference of 1% or less
when analyzed using a representative sample of past elections is presumptively
constitutional” and “[i]t is entirely workable to consider the seven percent [E]fficiency
[G]ap threshold as a presumption of constitutionality”).
Although the Harper I majority insisted that “[l]ower courts can and assuredly
will work out more concrete and specific standards,” id. at 384, 868 S.E.2d at 547
(alteration in original) (quoting Reynolds, 377 U.S. at 578, 84 S. Ct. at 1390), on
remand, the selected tests and corresponding scores—as predicted—proved
insufficient as a clear and manageable standard. The General Assembly and the
three-judge panel attempted to use the Mean-Median Difference and Efficiency Gap
metrics to review the General Assembly’s Remedial Plans. But the majority’s
application of these two seemingly straightforward tests led to inconsistent results.
For example, because the Harper I majority indicated that a 1% Mean-Median
Difference and a 7% Efficiency Gap could serve as “possible bright-line standards” for
measuring partisan fairness, id. at 385, 868 S.E.2d at 548, the three-judge panel
-78-
HARPER V. HALL
Opinion of the Court
relied heavily on the advisors’ findings regarding each plan’s Mean-Median
Difference and Efficiency Gap scores in making its findings of fact on remand. Four
out of seven advisors and experts calculated a Mean-Median Difference of less than
1% for both the RHP and the RSP, and all seven advisors and experts calculated an
Efficiency Gap of less than 7% for both plans. Harper II, 383 N.C. at 153, 881 S.E.2d
at 198 (Newby, C.J., dissenting). Accordingly, the three-judge panel held that both
plans were “satisfactorily within the statistical ranges set forth in [Harper I].”
Similarly to the RSP and RHP, five out of eight advisors and experts found
that the RCP had a Mean-Median Difference of less than 1% and an Efficiency Gap
of less than 7%. Id. at 158, 881 S.E.2d at 201. The three-judge panel, however,
concluded without explanation that the RCP was “not satisfactorily within the
statistical ranges set forth in [Harper I].” A majority of advisors and experts found
that all three plans fell within the thresholds set by the Harper I majority, yet for
some reason—a reason that the three-judge panel did not articulate—only the RCP
was unconstitutional. Why was this range of data acceptable for the RSP and RHP,
but not for the RCP? The three-judge panel could not explain its inconsistent results
because these tests do not provide a clear, judicially manageable standard. Instead,
as cautioned by Rucho, these tests “ask[ ] judges to predict how a particular
districting map will perform in future elections [which] risks basing constitutional
-79-
HARPER V. HALL
Opinion of the Court
holdings on unstable grounds outside judicial expertise.” Rucho, 139 S. Ct. at 2503–
04.
Just like the three-judge panel, the same four-justice majority from Harper I
found their own standard unmanageable when they tried to apply it in Harper II. For
example, in declaring the RSP unconstitutional, the Harper II majority believed that
“all but one [a]dvisor” calculated the RSP’s Mean-Median Difference score as greater
than 1%. Harper II, 383 N.C. at 121, 881 S.E.2d at 178.14 According to those four
justices, this evidence supported a conclusion that the RSP did not meet the
statistical thresholds identified in Harper I. Id. The same number of advisors,
however, found that the RHP scored above the 1% Mean-Median Difference threshold
as well. Inexplicably, the four-justice majority in Harper II concluded that this fact
weighed against a finding that the RSP was constitutional but supported a finding
that the RHP was constitutional. Those justices did not say why the same evidence
supported contrary conclusions for two different maps.
Similarly, the Harper II majority believed that the RHP was constitutional
because, collectively, “[t]he [ ] [a]dvisors determined that the RHP yields an average
14 This statement that “all but one [a]dvisor” calculated a Mean-Median Difference
greater than 1% is inaccurate. Half of the advisors, not one, calculated the RSP’s Mean-
Median Difference score as less than 1%.
-80-
HARPER V. HALL
Opinion of the Court
[E]fficiency [G]ap of about 2.88%, [and] an average [M]ean-[M]edian [D]ifference of
about 1.27%.” Id. at 119−20, 881 S.E.2d at 177. The advisors’ average scores for the
RSP were very close to their averages for the RHP. For the RSP, the average of the
advisors’ Efficiency Gap scores was 3.81% and the average of their Mean-Median
Difference scores was 1.29%. Thus, both plans had an average Efficiency Gap score
that was well below the 7% threshold identified in Harper I as presumptively
constitutional. Harper I, 380 N.C. at 386, 868 S.E.2d at 548. Moreover, the average
Mean-Median Difference scores for the RSP and RHP were within two-one-
hundredths of a percentage point of each other. The Harper II majority did not say
why an average Mean-Median Difference of 1.27% weighed in favor of the RHP’s
constitutionality but an average Mean-Median Difference of 1.29% weighed against
the RSP’s constitutionality. If there was something significant about that minute
difference, the Harper II majority did not or could not explain it.15
This standard is not “clear” or “judicially manageable” because, during the
15 Both the RHP and RSP were used during the 2022 election cycle. Significantly,
under the RHP approved by the four-justice majority in Harper II, Republican candidates
won 59% of the house races while receiving about 58% of the aggregate statewide vote. See
North Carolina State Board of Elections,
https://er.ncsbe.gov/?election_dt=11/08/2022&county_id=0&office=NCS&contest=0 (last
visited Apr. 13, 2023) . Under the RSP, which the Harper II majority found unconstitutional,
Republican candidates won 60% of the Senate races while receiving about 59% of the
aggregate statewide vote. Id. It is unclear why this small difference of approximately one
percentage point rendered the RHP constitutional and the RSP unconstitutional.
-81-
HARPER V. HALL
Opinion of the Court
remedial phase of this case, no one—not even the four justices who created it—could
apply it to achieve consistent results. Rucho, 139 S. Ct. at 2500, 2499 (internal
citations and quotations omitted). A constitutional standard must be clear and easily
applied by the branch assigned the duty in question. The approach created by the
four justices in Harper I is neither. See id. at 2498, 2499 (noting that a justiciable
issue has a “clear, manageable, and politically neutral” standard that can “reliably
differentiate” an unconstitutional from a constitutional action (quoting Vieth, 541
U.S. at 306−08, 124 S. Ct. at 1793 (Kennedy, J., concurring in the judgment))). The
remedial proceedings in this case demonstrate that neither the criteria created in
Harper I nor our constitution provide a judicially discoverable or manageable
standard to address claims of partisan gerrymandering.
The dissent argues that a court’s reviewing a legislatively enacted redistricting
statute for claims of partisan gerrymandering is similar to a court’s examining a
speedy trial claim under the constitution or determining a motion to dismiss criminal
charges. This approach, however, contains a fundamental error: it fails to recognize
that the constitution assigns the responsibility of redistricting to the General
Assembly, not to the courts. It forgets this Court’s time-honored standard of review
for legislation. The dissent seems to ignore that the General Assembly fulfills its
redistricting responsibility by enacting laws. Such legislation is entitled to a
-82-
HARPER V. HALL
Opinion of the Court
presumption of constitutionality and requires a showing that the legislation violates
an express provision of the constitution beyond a reasonable doubt. A court’s applying
a constitutional provision to particular facts or evaluating the quality of certain
evidence is fundamentally different than assessing the constitutionality of a statute
through judicial review.
Perhaps the dissent’s analogies reveal a more fundamental misunderstanding
of a court’s role in the redistricting process. The majority in Harper I and the dissent
here seem to imagine a future where redistricting is a court-managed process: a
future where courts endlessly supervise the redistricting process and impose their
own standards in the same way that courts assess which criminal trials are speedy
enough. As previously explained, however, our framers chose a different approach.
They committed redistricting decisions to the wisdom and judgment of the legislative
branch. In short, the dissent’s analogies further reinforce that there is no judicially
discoverable and manageable standard.
A judicially discoverable and manageable standard is necessary for resolving
a redistricting issue because such a standard “meaningfully constrain[s] the
discretion of the courts[ ] and [ ] win[s] public acceptance for the courts’ intrusion into
a process that is the very foundation of democratic decisionmaking.” Rucho, 139 S.
Ct. at 2500 (first quoting Vieth, 541 U.S. at 306−08, 124 S. Ct. at 1793 (Kennedy, J.,
-83-
HARPER V. HALL
Opinion of the Court
concurring in the judgment); and then quoting id. at 291, 124 S. Ct. at 1784 (plurality
opinion)). Here the standard set forth in Harper I does not constrain the discretion of
our courts at all. Instead, it invites limitless judicial involvement because it is so
difficult to apply and leads to inconsistent results. Only the four justices who
enunciated the Harper I standard can say for certain whether their standard has
been met. Accordingly, under the Harper I framework, every redistricting decision
the General Assembly makes would be subject to judicial oversight. This framework
does not constrain judicial discretion; rather, it requires that judicial decisionmaking
dominate the entire redistricting process.
The approach mandated by Harper I would not simply apply to statewide
redistricting decisions. At oral argument, counsel for plaintiffs stated that the Harper
I principles would apply to “all elections” throughout the State because “it stems from
a constitutional principle that speaks to all elections.” See Oral Argument at 49:35,
Harper v. Hall, (413PA21-2) (Mar. 14, 2023), https://www.youtube.com/watch?v=cp-
zlPxuu2I (last visited Apr. 20, 2023). This result would embroil the judiciary in every
local election in every county, city, and district across the state. 16 Municipalities,
16 North Carolina has 100 counties, 552 municipalities, numerous “special districts,”
such as sewer and water districts, and many local boards of education. See How NC Cities
Work, N.C. League of Municipalities, https://www.nclm.org/advocacy/how-nc-cities-work (last
visited Apr. 20, 2023).
-84-
HARPER V. HALL
Opinion of the Court
counties, local boards of education, and special districts frequently hold hundreds, if
not thousands, of local elections. Under the Harper I standard, our courts would need
to ensure that each of these elections provides each member of the relevant local
electorate a sufficient “opportunity to aggregate [his or her] vote with likeminded
citizens to elect a governing majority of elected officials who reflect those citizens’
views.” 380 N.C. at 383, 868 S.E.2d at 546. This process would involve endless
litigation that would task our judges with ensuring that the political makeup of every
city council, county commission, or local board of education adequately reflected the
distribution of Republicans and Democrats in the corresponding locality.
In addition to involving our courts in countless redistricting lawsuits, the
Harper I standard does not provide any guidance for several potential issues that
could arise in these cases. Where the standard does not provide guidance, our courts
would have to utilize their own policy preferences. For example, the Harper I
standard does not tell courts how to account for voters who are affiliated with a
political party other than Republican or Democrat or who are not affiliated with a
party at all. Our judges would have to address these concerns without any “clear,
manageable, [or] politically neutral” guidance. Rucho, 139 S. Ct. at 2498 (quoting
Vieth, 541 U.S. at 306–08, 124 S. Ct. at 1793 (Kennedy, J., concurring in the
judgment)). Harper I provides no guidance to courts on these issues. Instead, it
-85-
HARPER V. HALL
Opinion of the Court
requires courts to use their discretion to “work out” these questions in future
litigation. Harper I, 380 N.C. at 384, 868 S.E.2d at 547. This type of unmoored
discretion is a quintessential characteristic of an unmanageable standard and a
nonjusticiable, political question. As the Supreme Court has noted:
Nor is the goal of fair and effective representation
furthered by making the standards of reapportionment so
difficult to satisfy that the reapportionment task is
recurringly removed from legislative hands and performed
by [ ] courts which themselves must make the political
decisions necessary to formulate a plan or accept those
made by reapportionment plaintiffs who may have wholly
different goals from those embodied in the official plan.
From the very outset, we recognized that the
apportionment task, dealing as it must with fundamental
“choices about the nature of representation,” Burns v.
Richardson, 384 U.S. [87,] 92, [1965], is primarily a
political and legislative process.
Gaffney, 412 U.S. at 749, 93 S. Ct. at 2329.
C. Policy Decisions
Along with failing to provide a discernible and manageable standard, the
approaches created in Harper I and Harper II involve a host of “policy
determination[s] of a kind clearly for nonjudicial discretion.” Baker, 369 U.S. at 217,
82 S. Ct. at 710. Initially, since the state constitution does not mention partisan
gerrymandering, the four justices in Harper I first had to make a policy decision that
the state constitution prohibits a certain level of partisan gerrymandering. Tellingly,
-86-
HARPER V. HALL
Opinion of the Court
the majority was unable to articulate how much partisan gerrymandering is too
much. Essentially, the majority chose to insert into our constitution a requirement
for some type of statewide proportionality based on their view of political “fairness.”
Like the Federal Constitution, however, our constitution does not contain a
proportionality requirement. See Rucho, 139 S. Ct. at 2499. Instead, the creation of
this proportionality requirement was a monumental policy determination made by
the Harper I majority on its own initiative and equated to a judicial amendment to
our constitution.
Then, those four justices determined that our constitution mandates the use of
certain political science tests as a measure of this newly created constitutional
requirement. As the Supreme Court noted in Rucho, however, the definition of
“fairness” and how to measure it “poses basic questions that are political, not legal.”
Id. at 2500. For example, the Harper I majority stated that political science tests
could identify an unconstitutional redistricting plan when “using a representative
sample of past elections.” 380 N.C. at 386, 868 S.E.2d at 548. In doing so, the four-
justice majority in Harper I unilaterally determined that past election results can
accurately predict how individual voters will vote in the future. But there is no reason
to presume this is true because individual voters may vote inconsistently at different
times in their life for a variety of reasons. As the Supreme Court noted in Rucho,
-87-
HARPER V. HALL
Opinion of the Court
voters select candidates based on “the issues that matter to them, the quality of the
candidates, the tone of the candidates’ campaigns, the performance of an incumbent,
national events or local issues that drive voter turnout, and other considerations.”
139 S. Ct. at 2503. Each of these factors is different for each election, and it is not
clear how past election results can possibly predict how each of these factors may
affect individual voters in future elections. The decision to use certain political science
tests, which tests to use, which scores are required, and which past election results
are most predictive of future electoral behavior involve policy choices that are
untethered to the law.
Additionally, in determining that past election results should be used to
calculate political science metrics, the Harper I majority made the policy
determination that past elections are a “better” source of partisan election data than
other potential sources. The Harper I majority even preferred certain past elections
over others. Some might argue, however, that data from past elections does not
measure the distribution of voters among various political groups, but that instead,
it measures the rate of voter turnout. Instead of using past election results, the
Harper I majority might have required partisan data from current voter registration
information. In theory this data set might be a more accurate representation of how
voters might vote in an upcoming election because it reflects current party affiliation
-88-
HARPER V. HALL
Opinion of the Court
statistics instead of past voter turnout. Selecting between past elections, current
voter registration information, or some other data as the “best” source for garnering
partisan election data, however, is exactly the sort of non-judicial policy
determination warned of in Rucho. See Rucho, 139 S. Ct. at 2500 (“Deciding among
just these different visions of fairness (you can imagine many others) poses basic
questions that are political, not legal. There are no legal standards discernible in the
Constitution for making such judgments, let alone limited and precise standards that
are clear, manageable, and politically neutral.”).
Moreover, simply the decision to use these political science metrics at all
requires policy determinations that are not grounded in any constitutional guidance.
Because these tests purport to measure “partisan fairness,” use of these tests
assumes that the chosen past election results are the most relevant factor for
predicting future election results and assumes that voters will continue to vote for
the same party that they have in the past. This is not true since many other
considerations influence a voter’s selection of a candidate. For example,
representative government is grounded in the concept of geographic representation.
Though partisanship may influence a representative’s attention to certain political
issues, the representative is likely to attend to numerous other issues important to
the shared community interests that affect his or her constituents. Indeed there are
-89-
HARPER V. HALL
Opinion of the Court
countless policy issues, and voters and representatives of the same political party
may be likeminded on some issues but not others. The constitution cannot guarantee
that a representative will have identical political objectives as a given constituent
because that is an impossible requirement. Representatives are individuals with
their own beliefs and who pursue their own motivations, often in opposition to other
members of their own party. Partisan fairness metrics do not—and cannot—measure
or quantify these intangible characteristics. The decision to use these “partisan
fairness” tests is a policy determination because it presumes that a voter’s or a
candidate’s partisan affiliation—over all other factors—is the most relevant factor in
predicting future election outcomes.
After making the policy decision that political science tests must be used to
measure partisan fairness, the Harper I majority made yet another policy choice by
selecting two particular political science tests—the Mean-Median Difference and
Efficiency Gap metrics—to serve as its “bright-line standards.” See Harper I, 380 N.C.
at 385, 868 S.E.2d at 548. The Harper I majority was aware of numerous other
potential tests; yet it chose these two as the best measures of its definition of fairness.
See id. at 384, 868 S.E.2d at 547 (recognizing “close-votes, close-seats analysis” and
“partisan asymmetry analysis” as other potential fairness metrics).
-90-
HARPER V. HALL
Opinion of the Court
Furthermore, utilization of these two tests—Mean-Median Difference and
Efficiency Gap—requires a host of policy determinations. During the remedial
process, the General Assembly and each of the four advisors calculated a Mean-
Median Difference and Efficiency Gap score for each of the Remedial Plans. Each
calculated slightly different scores, however, because each utilized different
redistricting software, partisan election data, and calculation methods. The General
Assembly, for example, calculated their scores using Maptitude, a “widely accepted”
redistricting software, and a set of twelve statewide elections selected by plaintiffs’
expert, Dr. Mattingly (Mattingly Election Set). Notably, neither the Special Masters,
the three-judge panel, nor the Harper II majority gave any deference to the General
Assembly’s approach. Each of the advisors selected different redistricting software
and elections sets from those chosen by the General Assembly and by each other. In
turn, the three-judge panel had to weigh each combination of redistricting software,
partisan election data, and calculation methods and determine which was “best.”
Each of these choices constitutes a policy determination that courts are not equipped
to make.
For example, each of the advisors used different redistricting software from the
others, and none chose to use Maptitude, as had the General Assembly. Dr. Grofman
used Dave’s Redistricting App to calculate the Remedial Plans’ Mean-Median
-91-
HARPER V. HALL
Opinion of the Court
Difference and Efficiency Gap scores, and Dr. McGhee used a web-based redistricting
software called PlanScore. It is not clear from Dr. Grofman’s or Dr. McGhee’s reports
how these technologies calculate the relevant metrics or whether they do so
differently from Maptitude.
Likewise, each of the advisors used different sets of elections as their sources
of partisan data to measure the Remedial Plans. Once again, none chose the same set
of elections as each other or as the General Assembly. The General Assembly used
the Mattingly Election Set, which consisted of twelve statewide elections from 2016
and 2020 chosen by one of plaintiffs’ experts. Alternatively, Dr. Jarvis pulled partisan
election data from eleven statewide elections. Nine of these elections matched the
General Assembly’s Mattingly Election Set, but two others did not. Dr. Grofman used
“major statewide races [in] 2016−2020” but did not specify how many elections or
which ones. Dr. Wang, on the other hand, varied the vote totals in each of these
elections “above and below an average [vote total]” in order to “evaluat[e] a range of
future [vote total] scenarios that may arise in the coming decade.” Dr. Wang also
created a composite of vote totals by averaging together three data points: (1) the
average two-party vote share of the 2016 and 2020 presidential elections; (2) the
average two-party vote share of the 2016 and 2020 United States Senate elections;
-92-
HARPER V. HALL
Opinion of the Court
and (3) the average two-party vote share of the 2020 elections for Governor and
Attorney General.
Additionally, Dr. McGhee took a very “different approach” to calculating the
Mean-Median Difference and Efficiency Gap scores. Instead of analyzing which
party’s candidate would win a proposed new district under prior election contests, Dr.
McGhee used PlanScore to “predict” potential partisan outcomes in the future. Dr.
McGhee did not explain which elections PlanScore applied to predict future election
results, nor did he explain the criteria used by PlanScore to make such predictions.
Dr. McGhee also calculated two sets of Mean-Median Difference and Efficiency Gap
scores. He calculated one set from a simulated election that assumed that no
incumbents ran for reelection and another set from a simulated election that assumed
that all incumbents ran for reelection in the proposed district containing their
residence. He did not explain why he made these unrealistic assumptions.
As a result, the General Assembly and each advisor calculated different scores
for the Remedial Plans, even though they all used the same tests. These varying
results prove that the use of two seemingly straightforward fairness metrics actually
involves a multitude of policy choices—the kind of policy choices the Supreme Court
warned of in Rucho. Rucho, 139 S. Ct. at 2503−04 (“For all of those reasons, asking
judges to predict how a particular districting map will perform in future elections
-93-
HARPER V. HALL
Opinion of the Court
risks basing constitutional holdings on unstable ground outside judicial expertise.”).
Because there are “no legal standards discernible in the [c]onstitution” that describe
statewide proportionality or that instruct which tests to use or how to calculate them,
each party and expert simply calculated his scores in whatever way he saw fit. Id. at
2500. Each of these differences illustrates the numerous policy choices that are
inherent in applying the metrics selected in Harper I.
The standard set forth in Harper I is clearly rife with policy determinations
that our courts are not equipped to make. See Baker, 369 U.S. at 217, 82 S. Ct. at 710.
Accordingly, the claims at issue—partisan gerrymandering claims—are
nonjusticiable. Moreover, when a court engages in policy determinations like these,
it ignores our long-standing standard of review that presumes that acts of the General
Assembly are constitutional. See Berger, 368 N.C. at 639, 781 S.E.2d at 252. In part,
the existence of policy choices indicates that a given issue may be nonjusticiable
because the legislative branch—not the judicial branch—is “without question ‘the
policy-making agency of our government.’ ” Rhyne, 358 N.C. at 169, 594 S.E.2d at 8(quoting McMichael, 243 N.C. at 483, 91 S.E.2d at 234). If a court engages in policy
questions that are better suited for the legislative branch, that court usurps the role
of the legislature by deferring to its own preferences instead of the discretion of the
people’s chosen representatives. For this reason, and to protect against this result,
-94-
HARPER V. HALL
Opinion of the Court
the proper starting point in cases challenging an act of the General Assembly is to
assume the General Assembly’s policy choices are constitutional unless proved
otherwise “beyond any reasonable doubt.” Jenkins v. State Bd. of Elections, 180 N.C.
169, 172, 104 S.E. 346, 348 (1920).
Thus, all the policy choices made by the four-justice majority in Harper I and
Harper II demonstrate how that majority utterly ignored the well-established
presumption of constitutionality. By making these policy choices, the majority
replaced the General Assembly’s discretionary policymaking authority with its own.17
This approach flipped the presumption of constitutionality on its head and usurped
the role of the General Assembly—the policymaking branch of government.
In sum, a matter is nonjusticiable if the constitution expressly assigns
responsibility to one branch of government, or there is not a judicially discoverable or
manageable standard by which to decide it, or it requires courts to make policy
determinations that are better suited for the policymaking branch of government. All
three elements are present in the claims at issue in this case. In addition to the
17 As illustrated here, reliance on the tests set forth in Harper I invariably results in
redistricting by a judicial redistricting commission made up of court-appointed special
masters and advisors, which is not authorized anywhere in the constitution. Notably, the
only North Carolina races that did not reflect the statewide voting trends in the 2022 election
cycle were North Carolina’s congressional races held under the Interim Congressional Plan
drawn by the Special Masters and Dr. Grofman.
-95-
HARPER V. HALL
Opinion of the Court
legislature’s plenary power, the constitution expressly assigns the General Assembly
redistricting authority subject only to express limitations. The decision to implement
a proportionality or political fairness requirement in the constitution without explicit
direction from the text inherently requires policy choices and value determinations
and does not result in a neutral, manageable standard. Accordingly, plaintiffs’ claims
of partisan gerrymandering are nonjusticiable, political questions that are “beyond
the reach of” our courts. Rucho, 139 S. Ct. at 2506.
V. Declaration of Rights
Like the plaintiffs in Rucho, plaintiffs here allege that various constitutional
provisions prohibit partisan gerrymandering. In place of the Federal Constitutional
provisions invoked in Rucho, plaintiffs instead argue that comparable state
constitutional provisions expressly limit partisan considerations in redistricting.
Plaintiffs are mistaken; these state constitutional provisions do not expressly limit
the General Assembly’s redistricting authority or address partisan gerrymandering
in any way. Where there is no express limitation on the General Assembly’s authority
in the text of the constitution, this Court presumes an act of the General Assembly is
constitutional. Berger, 368 N.C. at 639, 781 S.E.2d at 252. As previously stated, courts
determine the meaning of a constitutional provision by discerning the intent of its
drafters when they adopted it. Courts look first to the plain language of the text,
-96-
HARPER V. HALL
Opinion of the Court
keeping in mind the historical context of the text’s adoption.
Our Declaration of Rights first appeared in the 1776 constitution and provides
“a statement of general and abstract principles.” State Constitution 6. The
“abstractness” of the Declaration of Rights has “allowed most of it to survive.” Id.
“Because of their abstractness,” many provisions of the Declaration of Rights do not
give rise to “justiciable rights.” Id. at 48; see, e.g., Dickson I, 367 N.C. at 575, 766
S.E.2d at 260 (stating that the “Good of the Whole” clause in Article I, Section 2 of
the constitution does not provide a “justiciable standard”). Rather, the Declaration
of Rights sets out “[b]asic principles” in “general terms,” and these basic terms are
“given specific application in later articles.” State Constitution 46. Here two of the
provisions cited by plaintiffs—the free elections clause and the freedom of assembly
clause—are from our 1776 Declaration of Rights. The other two—the equal protection
clause and free speech clause—first appeared in our 1971 constitution.
A. Free Elections Clause
Article I, Section 10 states that “[a]ll elections shall be free.” N.C. Const. art.
I, § 10. The clause first appeared in the 1776 constitution, which stated that the
“Elections of Members, to serve as Representatives in [the] General Assembly, ought
to be free.” N.C. Const. of 1776, Declaration of Rights, § VI. The 1868 constitution
restated the free elections clause as “[a]ll elections ought to be free.” N.C. Const. of
-97-
HARPER V. HALL
Opinion of the Court
1868, art. I, § 10. In the 1971 constitution, the provision became “[a]ll elections shall
be free,” N.C. Const. of 1971, art. I, § 10, the form that it retains today. See N.C.
Const. art. I, § 10. Even though the word “ought” in both the 1776 and 1868
constitutions was changed to “shall” in the 1971 constitution, this change was not a
substantive revision to the free elections clause. See Report of the North Carolina
State Constitution Study Commission 73−75; see also Smith v. Campbell, 10 N.C. (3
Hawks) 590, 598 (1825) (declaring that “ought” is synonymous with “shall” and noting
that “the word ought, in this and other sections of the [1776 constitution], should be
understood imperatively”).
“Free” means having political and legal rights of a personal nature or enjoying
personal freedom, a “free citizen,” or having “free will” or choice, as opposed to
compulsion, force, constraint, or restraint. See Free, Black’s Law Dictionary (11th ed.
2019). As a verb, “free” means to liberate or remove a constraint or burden. Id.Therefore, giving the provision its plain meaning, “free” means “free from
interference or intimidation.” State Constitution 56.
As with all “[b]asic principles” contained within the Declaration of Rights, we
must consider the free elections clause in the context of later articles that give more
specific application to Article I, Section 10. Id. at 46. The terms “elections” and “free,”
N.C. Const. art. I, § 10, must be read, for example, in the context of Article VI, entitled
-98-
HARPER V. HALL
Opinion of the Court
“Suffrage and Eligibility to Office,” see id. art. VI. The first five sections of Article VI
address the right to vote, and the last five sections concern eligibility to hold office.
See id. Even though “elections shall be free,” they are nonetheless restricted in many
ways by Article VI. See, e.g., N.C. Const. art. VI, § 1 (requiring a North Carolina voter
to be a citizen of the United States and at least 18 years old); id. art. VI, § 2(1)–(2)
(placing residency requirements on voters); id. art. VI, § 2(3) (placing restrictions on
felons’ voting rights); id. art. VI, § 3 (allowing for conditions on voter registration as
prescribed by statute); id. art. VI, § 5 (requiring that votes by the people be by ballot);
id. art. VI, § 7 (requiring public officials to take an oath before assuming office); id.
art. VI, § 8 (outlining certain disqualifications from holding public office); id. art. VI,
§ 9 (prohibiting dual office holding); id. art. VI, § 10 (allowing an incumbent to
continue in office until a successor is chosen and qualified).
Likewise, even though our 1776 constitution stated that elections were “free,”
N.C. Const. of 1776, Declaration of Rights, § VI, other provisions limited the scope of
that phrase. Notably, “free elections” did not mean that everyone could vote, N.C.
Const. of 1776, § VII (limiting the right to vote for senators to “freemen” who were at
least twenty-one years old, lived in their county of residence for at least one year, and
owned at least fifty acres of land in the same county for the preceding six months);
id. § VIII (limiting the right to vote for Representatives in the House of Commons to
-99-
HARPER V. HALL
Opinion of the Court
“freemen” who were twenty-one years old, lived in their county of residence for at
least one year, and paid public taxes); that anyone could run for office, id. § V (only
men who lived in their county of residence for one year and owned at least three
hundred acres of land in fee for one year could serve in the Senate); id. § VI (only men
who lived in their county for at least one year and owned at least one hundred acres
of land in fee or for life for at least six months could serve in the House of
Representatives); that the people were free to vote for all governmental officers, see
id. § XIII (empowering the General Assembly to elect Judges of the Supreme Court
and the Attorney General); see also id. § XV (empowering the General Assembly to
elect the Governor); or that the General Assembly was restricted from apportioning
itself by dividing existing counties, id. §§ II, III (providing each county one senator
and two representatives with no limitations on the General Assembly’s discretion to
create new counties). Clearly, when our framers intended to limit or clarify the scope
of “free elections,” they did so with express provisions in the text. They did not,
however, add anything to our 1776 constitution about partisan gerrymandering.
With respect to the history of the clause, its original intent and inclusion was
to protect against abuses of executive power. Our free elections clause was not
intended to protect the people from their representatives who frequently face election
by the people. Under colonial rule, the English crown appointed the governor for an
-100-
HARPER V. HALL
Opinion of the Court
indefinite period of time. Charles Lee Raper, North Carolina: A Study in English
Colonial Government 27 (1904). As a result, the governor “was very naturally
disposed,” id. at 186, to indulge the interests of the crown as opposed to those of “the
people whose affairs he was to administer,” id. at 27.
Additionally, the governor exercised broad executive, judicial, and legislative
functions. See id. at 28−32. The governor was the “head of the whole administrative
machinery of the province,” id. at 29, and could issue land grants that were legal
“even against the king himself,” id. at 28. He also possessed the authority to create
and establish the colony’s judicial system with any courts of law and equity that he
saw fit and could remove any judge or justice for “sufficient reason.” Id. at 37. In the
legislative realm, the governor possessed a veto power as no law “could be passed
without his assent.” Id. at 35. The governor could call the General Assembly
whenever “occasion demanded it,” id. at 34, and could dissolve it if he saw fit, id. at
35. Additionally, as the three-judge panel found, the Royal Governor “could require
counties and towns to obtain charters of incorporation prior to being able to elect
representatives to the legislature,” a power which inserted the governor squarely into
the issue of apportionment. Moreover, North Carolina colonists were also accustomed
to the English king exercising broad and oppressive executive powers as well. See Our
-101-
HARPER V. HALL
Opinion of the Court
First Revolution: The Remarkable British Upheaval that Inspired America’s
Founding Fathers 167 (2007) [hereinafter Our First Revolution].
For these reasons, there were tensions between North Carolina’s House of
Burgesses and the governor between 1729 and 1776. The two clashed over
representation in the General Assembly, id. at 90−91, the creation of counties, id. at
89−90, 217, the number of members needed to constitute a quorum in the General
Assembly, id. at 216−18, the appointment of agents to England, id. at 206−08, and
the appointment of judges, id. at 207−09, among other issues. Accordingly, by 1776
North Carolinians were inclined to replace “[o]verbearing colonial governors” with a
much weaker executive officer. Constitutional History 1764. As the three-judge panel
found in its 11 January 2022 Judgment,
[i]t was the experience of the people of the State of North
Carolina that was the most important source for the
creation of the 1776 Constitution. By far, the greatest
change in the structure of North Carolina’s government,
other than elimination of the parliament and the Crown,
was the vast reduction in the powers of the Governor and
the substantial increase in the powers of the General
Assembly. These changes were made to make “the governor
that figurehead in law which in fact the colonial legislature
had long sought to make him.”
(Quoting Earle H. Ketcham, The Sources of the North Carolina Constitution of 1776,
6 N.C. Hist. Rev. 215, 230 (1929).) Thus, under the 1776 constitution, the General
Assembly, not the people, chose the governor, N.C. Const. of 1776, § XV, the members
-102-
HARPER V. HALL
Opinion of the Court
of the council of state, id. § XVI, the state treasurer, id. § XXII, the state secretary,
id. § XXIV, the attorney general, id. § XIII, and all judges, id. The governor had no
veto power under the 1776 constitution, see id. §§ XVII−XX, and “he took no formal
role in legislation” because “bills became laws when passed by both houses and signed
by the speakers,” Constitutional History 1764. Additionally, representation in both
the Senate and House of Commons was by county. N.C. Const. of 1776, § II (granting
each county one senator); id. § III (granting each county two representatives and the
borough towns of Edenton, New Bern, Wilmington, Salisbury, Hillsborough, and
Halifax one representative each). Because the General Assembly had the power to
create counties, it also had the power to determine how much representation each
portion of the State received. See, e.g., Act of Apr. 8, 1777, An Act for dividing Rowan
County, and other Purposes therein mentioned, ch. XIX, 1777 N.C. Sess. Laws 33(splitting off part of Rowan County to create Burke County).
Our free elections clause was placed in the 1776 Declaration of Rights at the
same time as other constitutional provisions that both limited executive power and
increased legislative power. Accordingly, any argument that the people added the free
elections clause to the 1776 constitution for the purpose of limiting the General
Assembly’s apportionment authority is inconsistent with this historical context.
Instead, the free elections clause was intended to address abuses of executive power
-103-
HARPER V. HALL
Opinion of the Court
and to protect against interference and intimidation in the voting process. The
historical context occurring in England less than one hundred years earlier confirms
this meaning of the free elections clause.
Our 1776 Declaration of Rights was modeled in part after the English Bill of
Rights, a product of the Glorious Revolution in England in 1688. See Hugh Talmage
Lefler & Albert Ray Newsome, The History of a Southern State: North Carolina 221
(3d ed. 1973). “Today everyone in Britain and the United States is in a sense a
residuary beneficiary of the [Glorious] Revolution, although we can at present take
this for granted since the issues involved now form the accepted bases of our
institutions and societies.” J.R. Jones, The Revolution of 1688 in England 8 (Jack P.
Greene 1972) [hereinafter Revolution of 1688].
In the 1670s and 1680s, numerous European countries, including England,
were moving towards absolutist monarchies. This trend “seemed the way of the
future” throughout the continent. Our First Revolution 7. In England, however,
conflict swelled between King James II and Parliament as the king took various
actions beyond the limits of his authority in order to achieve his legislative agenda.18
18In the modern American context, we might refer to such encroachments as a
separation-of-powers violation. See Berger, 368 N.C. at 660, 781 S.E.2d at 265 (Newby, J.,
concurring in part and dissenting in part) (“A violation of separation of powers occurs when
one branch of government exercises the power reserved for another branch of government.”).
-104-
HARPER V. HALL
Opinion of the Court
King James II also sought to strengthen the crown by increasing the size of the
standing army and continuing regiments that had been raised temporarily for the
purpose of opposing rebellions. Revolution of 1688 61. James hoped to achieve these
goals by creating a compliant Parliament; but by 1685, he realized he could not do so
“without first changing the local officials . . . who conducted and effectively controlled
the elections, and without changing the franchise in many boroughs.”19 Our First
Revolution 109−10. King James shifted local authority by adjusting a county’s or
borough’s charter to embed the king’s agents and ensure a favorable outcome for the
king in the 1685 election. R. H. George, A.M., Ph.D., Fellow of the Royal Hist. Soc’y,
Parliamentary Elections and Electioneering in 1685 176−78 (Oct. 8, 1935). In some
instances, these adjustments altered who could vote in order to limit the franchise to
those most likely to support the king’s preferred candidates. See id. at 176. In other
cases, the adjustments secured for the king’s agents the most powerful political offices
19 The process for selecting members of Parliament varied greatly among counties and
boroughs during this time. Some counties elected two members and others, one. Our First
Revolution 55. Some boroughs elected as many as four members, while others only selected
two or one. Id. There was also disparity between the localities regarding who could vote. “In
some, the right to vote was attached to the ownership of certain pieces of property; in some
it was limited to officers of the borough corporation; in many, all freemen, that is adult males
not bound to service, could vote.” Id. at 56. Moreover, in the boroughs, the size of the
electorate varied widely. Id. at 57. Local officials and large landowners “exerted great
influence over local elections” in both the counties and boroughs. Id. These local differences
“were the result of ancient practice” that had “grown up in response to the demands of
particular communities and private interests” and “reflected a bewildering variety of local
customs.” Id. at 58.
-105-
HARPER V. HALL
Opinion of the Court
and gave them “complete control of the situation.” Id. at 177. Once in power, the
agents fully and immediately exercised their influence on behalf of the king. See id.
at 177, 182, 194−95.
The king’s agents used various tactics to manipulate and intimidate local
officials, would-be parliamentarians, and local business leaders into supporting the
king’s plans. See id. at 168, 188. They intimidated locals through physical scuffles,
threats, demonstrations of force, and beatings, id. at 173−75, and coerced businesses
to support King James’s preferred candidates, some altogether foreign to the locale,
by promising gifts, bribes, or patronage in exchange for compliance or by threatening
to revoke their license to operate, id. at 176−78, 184, 188−90. When the time for
election came, local agents of the king who conducted the polling used devious polling
practices to open, close, and reopen polling to ensure a certain electoral outcome. Id.
at 182, 185, 188.
After the elections of 1685, the resulting Parliament was “agreeable” to King
James at first, id. at 168, but once James presented his legislative agenda, many
parliamentary representatives interpreted his goals “as a danger to constitution and
liberties,” Revolution of 1688 62. Accordingly, King James met with opposition and,
as a result, he discontinued the session of Parliament in November 1685 so he could
unilaterally act to achieve his legislative agenda. See id. at 64−66. Once he
-106-
HARPER V. HALL
Opinion of the Court
discontinued the session of Parliament, he immediately put that agenda into motion.
See id. at 65−74. He repeatedly postponed the next Parliamentary session in an effort
to convince representatives to support his legislative objectives. See id. When those
efforts proved unsuccessful, id., however, he dissolved Parliament in July 1687 and
began a second “campaign to pack” it with members that would support his legislative
agenda, id. at 128, 131, 151; see also Our First Revolution 109. The king’s campaign
“represented a move to make this power complete, total, and permanent,” Revolution
of 1688 151, and was seen as “an attempt to move England toward ‘some form of
absolutism,’ ” Our First Revolution 109 (quoting Revolution of 1688 11−12).
King James once again set about intimidating and manipulating local officials.
Id. at 109−10. He sent agents to canvass justices of the peace and other local officials
to ascertain whether they would support the king’s legislative goals. Id. The king
used their responses to create his short list of “approved parliamentary candidates,”
Revolution of 1688 135, and to purge local officials who did not agree to support his
plans, see id. at 132−33. King James dismissed thousands of county and borough
officials who gave “unsatisfactory” responses. Our First Revolution 110. Additionally,
the king’s agents ensured that local sheriffs attended borough and county elections
to intimidate candidates who were hostile to royal policies. Revolution of 1688 147.
-107-
HARPER V. HALL
Opinion of the Court
King James’s tactics of commandeering his subjects’ support to ensure an
obedient Parliament were entirely unfamiliar to the English people and their
representatives.
Contemporaries were well aware that James was ruling in
a new way, a new way heavily modeled on the methods and
practices of Louis XIV [of France]. Both James’s enemies
and his friends marveled at the rapid increase in royal
power. James II’s “power swelled so fast,” recalled the Whig
critic Lord Delamere, “that he quickly makes all people to
feel the intolerable burden of an unbounded prerogative.”
Barillon agreed that “the royal authority increases
everyday by means of the firm conduct of the King of
England.” James, all concurred, took his measures from
Louis XIV. “The French precedent was too exactly
followed,” lamented one pamphleteer in 1688. “Our King in
imitation of his brother of France,” wrote another
pamphleteer drawing a similar parallel, “strives to bring
all the offices and magistracy of the kingdom, that were
legally of the people’s choice, to be solely and immediately
depending on his absolute will for their being.”
Steve Pincus, 1688: The First Modern Revolution 160−62 (2009). Ultimately,
however, King James II’s absolutism did not prevail in England. Our First Revolution
7. Instead, through the Glorious Revolution and the English Bill of Rights,
Englishmen chose an “alternative . . . constitutional monarchy with limits on
government[ and] guaranteed rights.” Id.
The drafters of the English Bill of Rights very clearly intended to address King
James’s overreaches of executive power and to return authority to Parliament. In the
-108-
HARPER V. HALL
Opinion of the Court
eyes of the drafters, King James had, among other wrongdoings, subverted “the laws
and liberties of th[e] kingdom” by “assuming and exercising [the] power of dispensing
with and suspending of laws, and the execution of laws, without consent of
[P]arliament.” Bill of Rights 1689, 1 W. & M. Sess. 2 c. 2 (Eng.). King James had
exercised “pretended power[s]” beyond the limits of his executive authority by levying
taxes for “the use of the crown” without the permission of Parliament, “raising and
keeping a standing army . . . without the consent of [P]arliament,” “violating the
freedom of election of members to serve in [P]arliament,” prosecuting crimes that
were within Parliament’s jurisdiction in the “court of King’s bench” instead, requiring
“excessive bail” to “elude . . . laws made for the liberty of the subjects,” and imposing
“excessive fines” and “cruel and unusual punishments.” Id. The drafters of the
English Bill of Rights characterized James’s actions as “utterly and directly contrary
to the known laws and statutes, and freedom of this realm.” Id.
Accordingly, after James fled England, the people selected new
representatives, as was their “right,” and the new representatives met “in a full and
free representative of th[e] nation.” Bill of Rights 1689, 1 W. & M. Sess. 2 c. 2 (Eng.).
These new representatives drafted the English Bill of Rights to ensure that their
“religion, laws, and liberties” would no longer “be in danger of being subverted” and
to “vindicat[e] and assert[ ] their ancient rights and liberties.” Id. In many instances,
-109-
HARPER V. HALL
Opinion of the Court
they expressly prohibited the king from acting under “pretended power”—that is,
power he never in fact possessed—without the consent of Parliament.20
The drafters of the English Bill of Rights not only clarified the limits on the
king’s executive power; they also memorialized their “ancient rights and liberties”—
rights that King James had violated and that, the drafters declared, would no longer
be subverted:
[I]t is the right of the subjects to petition the King, and all
commitments and prosecutions for such petitioning are
illegal.
....
[E]lection of members of parliament ought to be free.
[T]he freedom of speech, and debates or proceedings in
parliament, ought not to be impeached or questioned in any
court or place out of parliament.
....
[F]or the redress of all grievances, and for the amending,
strengthening, and preserving of the laws, parliaments
ought to be held frequently.
20Specifically, the English Bill of Rights clarified that the king could not “suspend[ ]”
or “dispens[e] with” laws, levy money for his own use, or raise a standing army in times of
peace without the consent of Parliament. Id. The king also could not require excessive bail,
impose excessive fines, or inflict cruel and unusual punishments. Id.
-110-
HARPER V. HALL
Opinion of the Court
Id. Each of these declarations responded to a specific behavior of King James. The
enumeration of the right to petition the king “direct[ly] rebuke[d]” King James’s
violations of that right. Our First Revolution 192. Under King James, many who
attempted to petition for exemption from certain laws were instead met with
prosecution. Id. The demand for frequent meetings of Parliament responded to
“James’s practice of ruling during most of the 1680s without a Parliament.” Id. The
declaration that elections of Parliamentary members ought to be free had been the
“central tenet” and rallying cry of King James II’s political opponent, William of
Orange, id. at 193:
[A]ccording to the ancient constitution of the English
government and immemorial custom, all elections of
Parliament men ought to be made with an entire liberty,
without any sort of force, or requiring the electors to choose
such persons as shall be named unto them, and the
persons, thus freely elected, ought to give their opinions
freely upon all matters that are brought before them,
having the good of the nation ever before their eyes, and
following in all things, the dictates of their conscience . . . .
William Henry, Prince of Orange, Declaration of the Prince of Orange (Oct. 10, 1688),
reprinted in Our First Revolution 265. By this declaration, the drafters of the English
Bill of Rights sought to secure a “free [P]arliament,” a Parliament where the electors
could vote for candidates of their choice, and the members, once elected, could
legislate according to their own consciences without threat of intimidation or coercion
-111-
HARPER V. HALL
Opinion of the Court
from the monarch.21 Our First Revolution 230. The Glorious Revolution ensured that
Parliamentary elections would be frequent and free from threat and intimidation. For
English citizens, the promises of the English Bill of Rights were fulfilled immediately
and continuously: British Parliament has met every year since 1689. Id. at 231.
In the years leading up to the Glorious Revolution, King James II also sought
to strengthen his control in the American colonies by using tactics similar to those he
used in England, including the elimination of colonial representative assemblies. Id.
The Glorious Revolution set the stage for similar conflicts in Carolina. After the
Glorious Revolution, all colonies reinstated their representative assemblies but still
endured authoritative royal governors. Id. This dynamic catalyzed the American
Revolution because the British colonists saw themselves as Englishmen. They
21 The historical context of the English Bill of Rights indicates that the English free
elections clause was in no way intended to address gerrymandering in apportionment. Rotten
Boroughs—boroughs containing very few residents that elected the same number of
parliamentary members as heavily populated boroughs—existed in England for at least one
hundred years prior to the framing of our constitution. Rotten Boroughs were prevalent in
England before, during, and well after the Glorious Revolution and the signing of the English
Bill of Rights in 1689. At that time, the English people added a free elections clause to their
English Bill of Rights to address threats, coercion, and intimidation in their elections: “Th[e]
election of members of Parliament ought to be free.” Bill of Rights 1689, 1 W. & M. Sess. 2 c.
2 (Eng.). Nevertheless, the Rotten Boroughs continued to exist in England until at least 1832.
As the three-judge panel found, the continued existence of these Rotten Boroughs at the time
of the signing of the English Bill of Rights and their continued use thereafter suggests that
the English people did not intend to address apportionment issues with their free elections
clause.
-112-
HARPER V. HALL
Opinion of the Court
understood that the English Bill of Rights protected them from overreaches of
executive power and secured for them a right to representative government and free
elections. Id. at 231−32.
Accordingly, Carolina colonists saw their Royal Governors’ abuses of executive
power as exercises of the same “pretended power,” Bill of Rights 1689, 1 W. & M.
Sess. 2 c. 2 (Eng.), that “had been stripped from” the king in the English Bill of Rights,
Our First Revolution 232. Thus, when the colonists rebelled and our framers drafted
the 1776 Declaration of Rights, “they were seeking to preserve in their own states
what they believed the [Glorious] Revolution had established.”22 See id. This
historical context produced our free elections clause and freedom of assembly clause.
22 Compare Bill of Rights 1689, 1 W. & M., 2d sess., c. 2 (“That the pretended power of
suspending of [and dispensing with] the laws, or the execution of laws, by regal authority,
without consent of parliament, is illegal.”), and id. (“That levying money for or to the use of
the crown . . . without grant of parliament . . . is illegal.”), and id. (“That the subjects which
are protestants, may have arms for their defence suitable to their conditions, and as allowed
by law.”), and id. (“That election of members of parliament ought to be free.”), and id. (“That
excessive bail ought not to be required, nor excessive fines imposed; nor cruel and unusual
punishments inflicted.”), and id. (“And that for redress of all grievances, and for the
amending, strengthening, and preserving of the laws, parliaments ought to be held
frequently.”), with N.C. Const. of 1776, Declaration of Rights, § V (“That all Powers of
Suspending Laws, or the Execution of Laws, by any Authority, without Consent of the
Representatives of the People, is injurious to their rights, and ought not to be exercised.”),
and id. § XVI (“That the People of this State ought not to be taxed . . . without the Consent
of themselves, or their Representatives in General Assembly, freely given.”), and id. § XVII
(“That the People have a right to bear Arms, for the Defence of the State . . . .”), and id. § VI
(“That Elections of Members, to serve as Representatives in General Assembly, ought to be
free.”), and id. § X (“That excessive Bail should not be required, nor excessive Fines imposed,
-113-
HARPER V. HALL
Opinion of the Court
Given the historical context of the English Bill of Rights, our framers did not
intend the adoption of the free elections clause to limit the General Assembly’s
redistricting authority or to address apportionment at all. As previously noted, North
Carolina experienced issues with apportionment both before and well after the
drafters first placed the free elections clause in the 1776 Declaration of Rights. These
early issues continued until 1835 when the people held a constitutional convention
to, among other things, address the apportionment issues. State Constitution 13. At
that time, they made various changes to their system of representation, see generally
N.C. Const. of 1776, amends. of 1835, but they did not alter the free elections clause,
see id. Thus, the historical context of our free elections clause—both colonial and
English—indicates that “free elections” refers to elections free from interference and
intimidation.
Although the free elections clause has been a part of our constitution since
1776, this Court has rarely been called upon to interpret this provision because its
language is plain: it protects voters from interference and intimidation in the voting
process. We addressed the merits of a free election claim in Clark v. Meyland, 261
nor cruel or unusual punishments inflicted.”), and id. § XX (“That, for redress of Grievances,
and for amending and strengthening the Laws, Elections ought to be often held.”).
-114-
HARPER V. HALL
Opinion of the Court
N.C. 140, 134 S.E.2d 168 (1964). The plaintiff in Clark challenged a statute that
required voters wishing to change their party affiliation to first take an oath that
included the following language: “I will support the nominees of [the] party to which
I am now changing my affiliation in the next election and the said party nominees
thereafter until I shall, in good faith, change my party affiliation in the manner
provided by law . . . .” Id. at 141, 134 S.E.2d at 169. We held that a portion of the
statute requiring certain provisions of the oath was invalid, explaining that:
Any elector who offers sufficient proof of his intent, in good
faith, to change his party affiliation cannot be required to
bind himself by an oath, the violation of which, if not
sufficient to brand him as a felon, would certainly be
sufficient to operate as a deterrant [sic] to his exercising a
free choice among available candidates at the election––
even by casting a write-in ballot. His membership in his
party and his right to participate in its primary may not be
denied because he refuses to take an oath to vote in a
manner which violates the constitutional provision that
elections shall be free. Article I, Sec. 10, Constitution of
North Carolina.
When a member of either party desires to change his
party affiliation, the good faith of the change is a proper
subject of inquiry and challenge. Without the objectionable
part of the oath, ample provision is made by which the
officials may strike from the registration books the names
of those who are not in good faith members of the party.
The oath to support future candidates violates the principle
of freedom of conscience. It denies a free ballot––one that is
cast according to the dictates of the voter’s judgment. We
must hold that the Legislature is without power to shackle
a voter’s conscience by requiring the objectionable part of
-115-
HARPER V. HALL
Opinion of the Court
the oath as a price to pay for his right to participate in his
party’s primary.
Id. at 142–43, 134 S.E.2d at 170 (emphases added) (citing N.C. Const. of 1868, art. I,
§ 10). Thus, we interpreted “free” to mean freedom to vote one’s conscience without
restriction by prior commitment. Nonetheless, an inquiry into the sincerity of one’s
desire to change parties did not violate the clause.
We also considered the free elections clause in State ex rel. Swaringen v. Poplin,
211 N.C. 700, 191 S.E. 746 (1937), in which the plaintiff, a candidate who ostensibly
lost an election for the office of county commissioner of Wilkes County, brought a quo
warranto action, alleging that the Wilkes County Board of Elections fraudulently
deprived him of the office by altering the vote count. Id. at 700–01, 191 S.E. at 746.
In response, the defendant argued the plaintiff’s complaint failed to state facts
sufficient to constitute a cause of action. Id. at 701, 191 S.E. at 746. After the trial
court rejected the defendant’s argument, the defendant appealed, arguing that it was
the sole duty of the County Board of Elections, rather than the judiciary, “to judicially
determine the result of the election from the report and tabulation made by the
precinct officials.” Id. at 701, 191 S.E. at 747. In affirming the trial court’s decision,
we provided the following rationale:
One of the chief purposes of quo warranto or an information
in the nature of quo warranto is to try the title to an office.
This is the method prescribed for settling a controversy
-116-
HARPER V. HALL
Opinion of the Court
between rival claimants when one is in possession of the
office under a claim of right and in the exercise of official
functions or the performance of official duties; and the
jurisdiction of the Superior Court in this behalf has never
been abdicated in favor of the board of county canvassers
or other officers of an election.
In the present case fraud is alleged. The courts are
open to decide this issue in the present action. In Art. I, sec.
10, of the Const. of North Carolina, we find it written: “All
elections ought to be free.” Our government is founded on
the consent of the governed. A free ballot and a fair count
must be held inviolable to preserve our democracy. In some
countries the bullet settles disputes, in our country the
ballot.
Id. at 702, 191 S.E. at 747 (internal citations omitted) (quoting N.C. Const. of 1868,
art. I, § 10). We interpreted “free,” therefore, to mean the right to vote according to
one’s conscience and to have that vote accurately counted.
Based upon its plain meaning as confirmed by its history and by this Court’s
precedent, the free elections clause means a voter is deprived of a “free” election if (1)
a law prevents a voter from voting according to one’s judgment, see Clark, 261 N.C.
at 142, 134 S.E.2d at 170, or (2) the votes are not accurately counted, see Poplin, 211
N.C. at 702, 191 S.E. at 747. Thus, we hold that the meaning of the free elections
clause, based on its plain language, historical context, and this Court’s precedent, is
that voters are free to vote according to their consciences without interference or
-117-
HARPER V. HALL
Opinion of the Court
intimidation. Plaintiffs’ partisan gerrymandering claims do not implicate this
provision.
B. Equal Protection Clause
Article I, Section 19 provides, in relevant part, that “[n]o person shall be denied
the equal protection of the laws; nor shall any person be subjected to discrimination
by the State because of race, color, religion, or national origin.” N.C. Const. art. I,
§ 19. The equal protection clause was added as part of the ratification of the 1971
constitution. State Constitution 68. The addition of the equal protection clause, while
a substantive change, was not meant to “bring about any fundamental change” to the
power of the General Assembly. Report of the North Carolina State Constitution
Study Commission 10.
Our understanding of the equal protection clause has been informed by federal
case law interpreting the Federal Equal Protection Clause. See Texfi Indus., Inc. v.
City of Fayetteville, 301 N.C. 1, 10–11, 269 S.E.2d 142, 149 (1980) (relying almost
entirely on Federal Equal Protection jurisprudence in analyzing a claim under Article
I, Section 19).
Here plaintiffs present the same arguments under our equal protection clause
as were made under the Federal Equal Protection Clause in Rucho. Compare Verified
Compl. for Declaratory J. and Injunctive Relief ¶ 189, Harper I, No. 21 CVS 015426,
-118-
HARPER V. HALL
Opinion of the Court
2021 WL 6884973 (N.C. Super. Ct. Wake County Dec. 16, 2021) (“Partisan
gerrymandering violates the State’s obligation to provide all persons with equal
protection of the law . . . by seeking to diminish the electoral power of supporters of a
disfavored party.”), with Rucho, 139 S. Ct. at 2492 (“[Plaintiffs] alleged that the Plan
violated the Equal Protection Clause of the Fourteenth Amendment by intentionally
diluting the electoral strength of Democratic voters.”). In Rucho the Supreme Court
determined that the plaintiffs’ partisan gerrymandering claims did not implicate the
Federal Equal Protection Clause. See 139 S. Ct. at 2502–04. As the Supreme Court
observed, “judges have no license to reallocate political power between the two major
political parties, with no plausible grant of authority in the Constitution, and no legal
standards to limit and direct their decisions.” Id. at 2507. We find this analysis
persuasive. Under our constitution, a claim of vote dilution allegedly based on one’s
affiliation with a political party does not raise a claim under our equal protection
clause.
This Court has previously explained that “[t]he right to vote on equal terms is
a fundamental right.” Northampton Cnty. Drainage Dist. No. One v. Bailey, 326 N.C.
742, 747, 392 S.E.2d 352, 356 (1990) (emphasis added) (citations omitted). Several of
our cases indicate that the fundamental right to vote on equal terms simply means
that each vote must have the same weight. This historic understanding of equal
-119-
HARPER V. HALL
Opinion of the Court
voting power is stated in Article II, Sections 3(1) and 5(1), requiring that legislators
“represent, as nearly as may be, an equal number of inhabitants.” N.C. Const. art. II,
§§ 3(1), 5(1). This is a simple mathematical calculation. See Rucho, 139 S. Ct. at 2501.
Party affiliation, however, is not mentioned in Article II, Sections 3 or 5.
Early on in its history, North Carolina moved towards representation roughly
based on population, first in the House, see N.C. Const. of 1776, amends. of 1835, art.
I, § 2, and later in the Senate, see N.C. Const. of 1868, art. II, § 5. It was not until
after Baker v. Carr instituted the one-person, one-vote requirement based on the
Federal Equal Protection Clause, see Baker, 369 U.S. at 210, 82 S. Ct. at 706,
however, that apportionment became a strictly population-based system in North
Carolina, see N.C. Const. of 1868, amends. of 1967, art. II, § 5. The 1971 North
Carolina Constitution incorporated these concepts into the text of Article II, see N.C.
Const. of 1971, art. II, §§ 3(1), 5(1), and our courts have applied these concepts in
interpretating our equal protection clause in the context of apportionment, see N.C.
Const. art. II, § 19. Several cases arising after this chronological progression are
helpful when reviewing equal protection claims arising in the context of
apportionment.
This Court’s decision in Northampton County illustrates the concept of
numerically equal voting strength. In that case, a certain drainage district lay partly
-120-
HARPER V. HALL
Opinion of the Court
in Northampton County and partly in Hertford County. 326 N.C. at 744, 392 S.E.2d
at 354. By statute, the Clerk of Superior Court in Northampton County—who was
elected only by Northampton County residents—appointed all the drainage district
commissioners. Id. at 744, 392 S.E.2d at 354. In a suit brought by the drainage
district to recover assessments made against the landowners in Hertford County, this
Court held that the electoral scheme of this drainage district violated the equal
protection clause of Article I, Section 19 because the Hertford County landowners
could not vote for the elected official who appointed all the commissioners, but the
landowners in Northampton County could. Id. at 746, 392 S.E.2d. at 355. This
arrangement infringed on the Hertford County landowners’ fundamental right “to
vote on equal terms” because some members of the district could vote for their elected
official, and some could not. See id. at 746, 392 S.E.2d at 355.
Likewise, in Blankenship v. Bartlett, the plaintiffs demonstrated a “gross
disparity in voting power between similarly situated residents of Wake County” by
making the following showing:
In Superior Court District 10A, the voters elect one judge
for every 32,199 residents, while the voters of the other
districts in Wake County, 10B, 10C, and 10D, elect one
judge per every 140,747 residents, 158,812 residents, and
123,143 residents, respectively. Thus, residents of District
10A have a voting power roughly five times greater than
residents of District 10C, four and a half times greater than
-121-
HARPER V. HALL
Opinion of the Court
residents of District 10B, and four times greater than
residents of District 10D.
363 N.C. 518, 527, 681 S.E.2d 759, 766 (2009). We explained that the above showing
implicated the fundamental “right to vote on equal terms in representative
elections—a one-person, one-vote standard,” id. at 522, 681 S.E.2d at 762–63, and we
thus employed a heightened scrutiny analysis, id. at 523, 681 S.E.2d at 763.
Similarly, in Stephenson I this Court addressed what the fundamental right to
vote on equal terms means when considering the use of multi-member and single-
member districts. See 355 N.C. at 378, 562 S.E.2d at 393. In that case we first found
that the challenged legislative plans—the 2001 Plans—violated the WCP. Id. at 371,
562 S.E.2d at 389–90. Out of respect for the legislative branch, we then sought to give
the General Assembly detailed criteria for fashioning remedial maps. The plaintiffs
“contend[ed] that remedial compliance with the WCP requires the formation of multi-
member legislative districts in which all legislators would be elected ‘at-large.’ ” Id.
at 376, 562 S.E.2d at 392. As such, we “turn[ed] to address the constitutional
propriety of such districts.” Id. at 377, 562 S.E.2d at 393. In doing so, we noted that
“[t]he classification of voters into both single-member and multi-member districts . . .
necessarily implicates the fundamental right to vote on equal terms.” Id. at 378, 562
S.E.2d at 393. We explained that
-122-
HARPER V. HALL
Opinion of the Court
voters in single-member legislative districts, surrounded
by multi-member districts, suffer electoral disadvantage
because, at a minimum, they are not permitted to vote for
the same number of legislators and may not enjoy the same
representational influence or “clout” as voters represented
by a slate of legislators within a multi-member district.
Id. at 377, 562 S.E.2d at 393 (emphasis added). Thus, we concluded that the “use of
both single-member and multi-member districts within the same redistricting plan”
infringes upon “the fundamental right of each North Carolinian to substantially equal
voting power.” Id. at 379, 562 S.E.2d at 394–95. In other words, “substantially equal
voting power” meant that each legislator should represent a similar number of
constituents, which was impossible when using both single-member and multi-
member districts in the same map. This is an application of the one-person, one-vote
concept.
In Harper I, however, four justices expanded the scope of “substantially equal
voting power” from mathematically equal representation under the one-person, one-
vote concept and misconstrued it to create an “opportunity to aggregate one’s vote
with likeminded citizens” based on partisan affiliation. 380 N.C. at 378, 868 S.E.2d
at 544. This idea is not supported by our precedent.
Stephenson I recognized that partisan considerations are permitted in the
redistricting process. 355 N.C. at 371, 562 S.E.2d at 390 (“The General Assembly may
consider partisan advantage and incumbency protection in the application of its
-123-
HARPER V. HALL
Opinion of the Court
discretionary redistricting decisions, but it must do so in conformity with the State
Constitution.” (internal citation omitted)); Rucho, 139 S. Ct. at 2497 (recognizing that
legislators must be permitted to take some “partisan interests into account when
drawing district lines”). The ultimate holding of our Stephenson I decision was that
the WCP of Article II, Sections 3 and 5 must be enforced to the extent compatible
with the VRA and one-person, one-vote principles. Thus, when understanding
Stephenson I in context, it becomes clear that the Court’s statement—that the
General Assembly’s practice of partisan gerrymandering must still conform with the
constitution—refers to the express objective limitations present in Article II, Sections
3 and 5, not to a prohibition or limitation on partisan considerations.
Unlike the classifications in Northampton County, Blankenship, and
Stephenson I, partisan gerrymandering has no impact upon the right to vote on equal
terms under the one-person, one-vote standard. In other words, an effort to
gerrymander districts to favor a political party does not alter individual voting power
so long as each voter is permitted to (1) vote for the same number of representatives
as voters in other districts, and (2) vote as part of a constituency that is similar in
size to that of the other districts. Therefore, following the guidance of the Supreme
Court in Rucho, we hold that a partisan gerrymandering claim does not trigger review
under our equal protection clause. See Gaffney, 412 U.S. at 745 (holding that certain
-124-
HARPER V. HALL
Opinion of the Court
claims were “insufficient to make out a prima facie case of invidious discrimination
under the Fourteenth Amendment so as to require justification by the State”). Claims
that a redistricting plan diminishes the electoral power of members of a particular
political party do not violate Article I, Section 19 of our constitution.
C. Free Speech and Freedom of Assembly Clauses
The freedom of assembly and free speech clauses are found in Article I, Section
12 and Article I, Section 14 respectively. These sections provide as follows:
Sec. 12. Right of assembly and petition.
The people have a right to assemble together to
consult for their common good, to instruct their
representatives, and to apply to the General Assembly for
redress of grievances; but secret political societies are
dangerous to the liberties of a free people and shall not be
tolerated.
Sec. 14. Freedom of speech and press.
Freedom of speech and of the press are two of the
great bulwarks of liberty and therefore shall never be
restrained, but every person shall be held responsible for
their abuse.
N.C. Const. art. I, §§ 12, 14. Like the equal protection clause, the free speech clause
was added to our Declaration of Rights as part of the 1971 constitution. N.C. Const.
of 1971, art. I, § 14. The addition of the free speech clause, while a substantive change,
was not meant to “bring about any fundamental change” to the power of the General
-125-
HARPER V. HALL
Opinion of the Court
Assembly. Report of the North Carolina State Constitution Study Commission 10. Our
understanding of the free speech clause is informed by federal interpretation of the
similar provision in the First Amendment to the Federal Constitution. See State v.
Petersilie, 334 N.C. 169, 183, 432 S.E.2d 832, 840 (1993) (adopting “doctrines
developed by the United States Supreme Court in interpreting the Free Speech
Clause of the United States Constitution . . . for purposes of applying the Free Speech
Clause of the North Carolina Constitution”).
The freedom of assembly clause first appeared in the 1776 Declaration of
Rights and provided “that the People have a right to assemble together, to consult for
their common good, to instruct their Representatives, and to apply to the Legislature,
for Redress of Grievances.” N.C. Const. of 1776, Declaration of Rights, § XVIII. The
freedom of assembly clause was modified by the 1868 constitution by deleting “that,”
the first word of the clause. N.C. Const. of 1868, art. I, § 25. In the 1971 constitution,
the freedom of assembly clause was re-written to the form it has today. N.C. Const.
of 1971, art. I, § 12. As with the 1971 changes to the free speech clause, the most
recent change to the freedom of assembly clause was not meant as a substantive
change, nor was it meant to “bring about any fundamental change” to the power of
the General Assembly. Report of the North Carolina State Constitution Study
Commission 10.
-126-
HARPER V. HALL
Opinion of the Court
The right to free speech is violated when “restrictions are placed on the
espousal of a particular viewpoint,” Petersilie, 334 N.C. at 183, 432 S.E.2d at 840, or
where retaliation motivated by the content of an individual’s speech would deter a
person of reasonable firmness from engaging in speech or association, Toomer v.
Garrett, 155 N.C. App. 462, 478, 574 S.E.2d 76, 89 (2002) (explaining that a viable
retaliation claim requires a showing “that the plaintiff . . . suffer[ed] an injury that
would likely chill a person of ordinary firmness from continuing to engage” in a
“constitutionally protected activity,” including First Amendment activities), appeal
dismissed and disc. rev. denied, 357 N.C. 66, 579 S.E.2d 576 (2003); see Evans v.
Cowan, 132 N.C. App. 1, 11, 510 S.E.2d 170, 177 (1999) (determining “there was no
forecast of evidence” to support a retaliation claim).
It is apparent that a person of ordinary firmness would not refrain from
expressing a political view out of fear that the General Assembly will place his
residence in a district that will likely elect a member of the opposing party. See
Toomer, 155 N.C. App. at 477–78, 574 S.E.2d at 89. It is plausible that an individual
may be less inclined to voice his political opinions if he is unable to find someone who
will listen. Article I, Sections 12 and 14, however, guarantee the rights to speak and
assemble without government intervention, rather than the right to be provided a
receptive audience. See Minn. State Bd. for Cmty. Colls. v. Knight, 465 U.S. 271, 286,
-127-
HARPER V. HALL
Opinion of the Court
104 S. Ct. 1058, 1066 (1984) (stating that individuals “have no constitutional right as
members of the public to a government audience for their policy views”); Johnson v.
Wisc. Elections Comm’n, 967 N.W.2d 469, 487 (Wis. 2021) (“Associational rights
guarantee the freedom to participate in the political process; they do not guarantee a
favorable outcome.” (emphasis added)).
Partisan gerrymandering plainly does not place any restriction upon the
espousal of a particular viewpoint. Rather, redistricting enactments in North
Carolina are subject to the typical policymaking customs of open debate and
compromise. See Berger, 368 N.C. at 653, 781 S.E.2d at 261 (Newby, J., concurring in
part and dissenting in part). As such, opponents of a redistricting plan are free to
voice their opposition.
Article I, Sections 12 and 14 do not limit the General Assembly’s presumptively
constitutional authority to engage in partisan gerrymandering. As with the prior
Declaration of Rights clauses, there is nothing in the history of the clauses or the
applicable case law that supports plaintiffs’ expanded interpretation of them. This
Court and the Court of Appeals have interpreted speech and assembly rights in
alignment with federal case law under the First Amendment. See Petersilie, 334 N.C.
at 184, 432 S.E.2d at 841; Feltman v. City of Wilson, 238 N.C. App. 246, 252–53, 767
S.E.2d 615, 620 (2014); State v. Shackelford, 264 N.C. App. 542, 552, 825 S.E.2d 689,
-128-
HARPER V. HALL
Opinion of the Court
696 (2019). As discussed at length in Rucho, the Supreme Court of the United States
found no manageable standards for assessing partisan considerations in redistricting
despite the existence of similar express protections for speech and assembly rights in
the Federal Constitution. Rucho, 139 S. Ct. at 2505–07.
In summary, none of the constitutional provisions cited by plaintiffs prohibit
the practice of partisan gerrymandering. Each provision must be read in harmony
with the more specific provisions that outline the practical workings for governance.
Notably, Article II, Sections 3 and 5 outline the practical workings of the General
Assembly’s redistricting authority. These provisions contain four express limitations
on the General Assembly’s otherwise explicit redistricting authority, none of which
address partisan gerrymandering.
VI. Stephenson I and the VRA
Because we are overturning Harper I, we must briefly revisit another of
Common Cause’s claims that was based on a holding in that opinion. In its 11
January 2022 Judgment, the three-judge panel concluded that although Stephenson
I requires the General Assembly to draw VRA districts prior to non-VRA districts, it
does not require the General Assembly to conduct an RPV analysis “prior to making
a decision as to whether VRA districts are necessary.” Accordingly, the three-judge
panel dismissed this claim with prejudice. In Harper I the four-justice majority
-129-
HARPER V. HALL
Opinion of the Court
reversed this portion of the 11 January 2022 Judgment and held that our constitution
and Stephenson I require the General Assembly to conduct an RPV analysis before
drawing any legislative districts. See Harper I, 380 N.C. at 401, 868 S.E.2d at 558.
Accordingly, on remand, the General Assembly performed an RPV analysis, and the
three-judge panel found that this analysis satisfied this Court’s directive from Harper
I. Common Cause challenged this finding of fact in its appeal from the three-judge
panel’s remedial order.
The holding from Harper I that required the General Assembly to perform an
RPV analysis before drawing any legislative districts was based on an inaccurate
reading of Stephenson I. In Stephenson I we explained that “Section 2 of the VRA
generally provides that states or their political subdivisions may not impose any
voting qualification or prerequisite that impairs or dilutes, on account of race or color,
a citizen’s opportunity to participate in the political process and to elect
representatives of his or her choice.” 355 N.C. at 363, 562 S.E.2d at 385 (first citing
42 U.S.C. §§ 1973a, 1973b (1994); and then citing Gingles, 478 U.S. at 43, 106 S. Ct.
at 2762). We then stated that “[o]n remand, to ensure full compliance with federal
law, legislative districts required by the VRA shall be formed prior to creation of non-
VRA districts.” Id. at 383, 562 S.E.2d at 396−97. We provided this approach to
alleviate the tension between the WCP and the VRA because the legislative
-130-
HARPER V. HALL
Opinion of the Court
defendants in Stephenson I argued that “the constitutional provisions mandating that
counties not be divided are wholly unenforceable because of the requirements of the
[VRA].” Id. at 361, 562 S.E.2d at 383−84. Thus, the Court in Stephenson I was not
forcing the legislative defendants to conduct an RPV analysis. Rather, the Court was
merely stating that if Section 2 requires VRA districts, those districts must be drawn
first so that the remaining non-VRA districts can be drawn in compliance with the
WCP.
Because the North Carolina Constitution does not require the General
Assembly to conduct an RPV analysis before enacting a redistricting plan, Common
Cause’s arguments regarding the General Assembly’s RPV analysis are inapposite.
Plaintiffs essentially ask this Court to “impose a judicially-mandated preclearance
requirement” where no such requirement exists in our constitution. If Common Cause
believed that the General Assembly was incorrect that no VRA districts were
required, it could have brought a claim under Section 2 of the VRA. Common Cause
did not bring such a claim in this case. Accordingly, the holding in Harper I that
required the General Assembly to undertake an RPV analysis is overruled, and the
-131-
HARPER V. HALL
Opinion of the Court
portion of the 11 January 2022 Judgment dismissing Common Cause’s declaratory
judgment claim with prejudice is affirmed. 23
VII. Petitions for Rehearing Under Rule 31 and Stare Decisis
Rule 31 of the North Carolina Rules of Appellate Procedure states that
[a] petition for rehearing may be filed in a civil action
within fifteen days after the mandate of the court has been
issued. The petition shall state with particularity the
points of fact or law that, in the opinion of the petitioner,
the court has overlooked or misapprehended and shall
contain such argument in support of the petition as
petitioner desires to present.
N.C. R. App. P. 31(a). This rule contemplates that, at times, this Court may need to
revisit a recent decision to correct a mistake. We have never hesitated to rehear a
case when it is clear that the Court “overlooked or misapprehended” the law. See, e.g.,
Jones v. City of Durham, 360 N.C. 367, 367, 629 S.E.2d 611, 611 (2006) (order
granting rehearing); Smith Chapel Baptist Church v. City of Durham (Smith Chapel
I), 349 N.C. 242, 242, 514 S.E.2d 272, 272 (1998) (same); Whitford v. Gaskill, 345 N.C.
762, 762, 489 S.E.2d 177, 178 (1997) (same); Clay v. Emp. Sec. Comm’n, 340 N.C. 83,
87, 458 S.E.2d 198, 199 (1995) (same); Alford v. Shaw, 318 N.C. 703, 703, 351 S.E.2d
23While we do not specifically address the issue of standing here, we note this Court
has addressed the test for standing in Community Success Initiative v. Moore, ___ N.C. ___,
___ S.E.2d. ___ (2023), issued concurrently with this opinion. We overrule the analysis of
standing set forth in Harper I to the extent it conflicts with the decision in Community
Success. See Harper I, 380 N.C. at 353−55, 868 S.E.2d at 528−29.
-132-
HARPER V. HALL
Opinion of the Court
738, 738 (1987) (same); Lowe v. Tarble, 313 N.C. 176, 176, 326 S.E.2d 32, 32 (1985)
(same); Hous., Inc. v. Weaver, 304 N.C. 588, 588, 289 S.E.2d 832, 832 (1981) (same).
Several of these rehearings resulted in new opinions that differed substantially from
the Court’s initial opinion in the case. See, e.g., Jones v. City of Durham, 361 N.C.
144, 146, 638 S.E.2d 202, 202 (2006) (per curiam); Smith Chapel Baptist Church v.
City of Durham (Smith Chapel II), 350 N.C. 805, 806, 517 S.E.2d 874, 876 (1999);
Clay v. Emp. Sec. Comm’n, 340 N.C. 83−84, 457 S.E.2d 725, 726 (1995); Alford v.
Shaw, 320 N.C. 465, 467, 358 S.E.2d 323, 324 (1987) (on rehearing, withdrawing the
Court’s original opinion and reviewing the case “de novo”). It is not uncommon that
rehearing of a case coincides with a change in personnel on the Court who provide a
fresh legal perspective. See, e.g., Smith Chapel II, 350 N.C. at 807, 821, 517 S.E.2d at
876, 883−84. Our decision today simply adheres to these principles. See Sidney
Spitzer & Co. v. Comm’rs of Franklin Cnty., 188 N.C. 30, 32, 123 S.E. 636, 638 (1924)
(“There should be no blind adherence to a precedent which, if it is wrong, should be
corrected at the first practical moment.” (internal citations omitted)). A petition for
rehearing is particularly appropriate here because the four-justice majority in Harper
I expedited the consideration of this matter over the strong dissent of the other three
justices on this Court. See Harper v. Hall, 382 N.C. 314, 316, 874 S.E.2d 902, 904−05
(2022) (order granting motion to expedite hearing and consideration). There was no
-133-
HARPER V. HALL
Opinion of the Court
“jurisprudential reason” to force an expedited consideration of this case. Id. at 317,
874 S.E.2d at 904 (Barringer, J., dissenting) (“Given the absence of any identifiable
jurisprudential reason, the majority’s decision today appears to reflect deeper
partisan biases that have no place in a judiciary dedicated to the impartial
administration of justice and the rule of law.”).
The “doctrine of stare decisis . . . proclaims, in effect, that where a principle of
law has become settled by a series of decisions, it is binding on the courts and should
be followed in similar cases.” State v. Ballance, 229 N.C. 764, 767, 51 S.E.2d 731, 733
(1949) (internal citations omitted). This doctrine reflects the idea that “the law must
be characterized by stability,” and courts should not change the law to reach
particular results. Id. at 767, 51 S.E.2d at 733. When adhering to the doctrine would
“perpetuate error,” however, this Court has never hesitated to refuse to apply it.
Sidney Spitzer & Co., 188 N.C. at 32, 123 S.E. at 638 (“There is no virtue in sinning
against light or in persisting in palpable error, for nothing is settled until it is settled
right.”); see also Mial v. Ellington, 134 N.C. 131, 139, 46 S.E. 961, 964 (1903) (noting
the necessity of overturning a prior decision of this Court where it stood “without
support in reason” and was “opposed to the uniform, unbroken current of authority”
in the state); Ballance, 229 N.C. at 767, 51 S.E.2d at 733 (“[S]tare decisis will not be
applied in any event to preserve and perpetuate error and grievous wrong.”); Rabon
-134-
HARPER V. HALL
Opinion of the Court
v. Rowan Mem’l Hosp., Inc., 269 N.C. 1, 29, 152 S.E.2d 485, 502 (1967) (Lake, J.,
dissenting) (conceding that “a proper exercise of [judicial] power . . . is the result of
its determination that its former decision was an erroneous statement of the law
when the decision was rendered”); Wiles v. Welparnel Constr. Co., 295 N.C. 81, 85,
243 S.E.2d 756, 758 (1978) (“[S]tare decisis will not be applied when it results in
perpetuation of error or grievous wrong, since the compulsion of the doctrine
is . . . moral and intellectual, rather than arbitrary and inflexible.” (internal citation
omitted)).
Sometimes this Court explicitly overrules prior decisions. See, e.g., State v.
Elder, 383 N.C. 578, 603, 881 S.E.2d 227, 245 (2022) (overruling a portion of this
Court’s prior decision in State v. Hall, 305 N.C. 77, 286 S.E.2d 552 (1982));
Cedarbrook Residential Ctr., Inc. v. N.C. DHHS, 383 N.C. 31, 56−57, 881 S.E.2d 558,
576−77 (2022) (overruling Nanny’s Korner Day Care Ctr., Inc. v. N.C. DHHS, 264
N.C. App. 71, 825 S.E.2d 34 (2019)); State v. Kelliher, 381 N.C. 558, 581−83, 873
S.E.2d 366, 383−84 (2022) (abrogating State v. Green, 348 N.C. 588, 502 S.E.2d 819(1998)); Connette ex rel. Gullatte v. Charlotte-Mecklenburg Hosp. Auth., 382 N.C. 57,
71−72, 876 S.E.2d 420, 430−31 (2022) (reversing, with three votes, which is less than
a majority of this Court, the ninety-year-old opinion in Byrd v. Marion Gen. Hosp.,
202 N.C. 337, 162 S.E. 738 (1932)). Other times this Court overrules prior decisions
-135-
HARPER V. HALL
Opinion of the Court
by implication. See, e.g., McAuley v. N.C. A&T State Univ., 383 N.C. 343, 355, 881
S.E.2d 141, 149 (2022) (Barringer, J., dissenting) (noting that the majority opinion
“refuse[d] to follow . . . [ninety] years of this Court’s precedent established in Wray v.
Carolina Cotton & Woolen Mills Co., 205 N.C. 782, 783, 172 S.E. 487, 488 (1934));
State v. Styles, 362 N.C. 412, 415−16, 665 S.E.2d 438, 440−41 (2008) (effectively
abrogating State v. Ivey, 360 N.C. 562, 633 S.E.2d 459 (2006)).
As demonstrated, this Court has not hesitated to revisit and overrule prior
decisions that are erroneous. Regardless, Harper I does not meet any criteria for
adhering to stare decisis—it is neither long-standing nor has it been relied upon in
other cases. See Ballance, 229 N.C. at 767, 51 S.E.2d at 733. Harper I was wrongly
decided and, as a result, Harper II was also wrongly decided. Legislative Defendants
filed a timely petition under Rule 31 of the Rules of Appellate Procedure, and Harper
II was properly reheard. Harper I is overruled, and Harper II is withdrawn and
superseded by this opinion.
VIII. Remedy
In their petition for rehearing, Legislative Defendants asked that if this Court
concludes that plaintiffs’ partisan gerrymandering claims are nonjusticiable, that the
Court also address the appropriate remedy—in other words, what set of maps, if any,
were constitutionally “established” and, therefore, must be used. Article II, Sections
-136-
HARPER V. HALL
Opinion of the Court
3(4) and 5(4) provide that “[w]hen established, the senate [and representative]
districts shall . . . remain unaltered” until the next federal census. N.C. Const. art. II,
§§ 3(4), 5(4) (emphasis added). Because “a constitution cannot violate itself,” Leandro,
346 N.C. at 352, 488 S.E.2d at 258, we must construe the meaning of the phrase
“[w]hen established,” N.C. Const. art. II, §§ 3(4), 5(4), in harmony with the rest of the
constitution.
Looking first to the plain meaning, to “establish” means “[t]o settle, make, or
fix firmly; to enact permanently.” Establish, Black’s Law Dictionary (11th ed. 2019).
This meaning connotes something more than the passage of a redistricting act by the
General Assembly. The General Assembly could certainly amend a redistricting act
up until the time it is used. Once passed and used in the next election, however, the
districts are “established” until the next decennial census unless a court finds them
constitutionally infirm. This understanding of “[w]hen established” is consistent with
our precedent that allows the General Assembly an opportunity to redraw districts
when necessary to remedy court-identified infirmities. See, e.g., Pender County, 361
N.C. at 510, 649 S.E.2d at 376 (“leav[ing] to the General Assembly the decision” of
how to redraw a district that was held to be constitutionally infirm and declining “to
specify the exact configuration” of how the districts should be redrawn). Accordingly,
“[w]hen established” refers to establishment consistent with the constitution. See
-137-
HARPER V. HALL
Opinion of the Court
N.C. Const. art. II, §§ 3, 5 (providing textual limitations); N.C.G.S. §§ 120-2.3 to -2.4
(providing for limited judicial review); see also N.C. Const. art. II, §§ 22(5)(b)-(d)
(exempting restricting legislation from gubernatorial veto).
In our order granting Legislative Defendants’ petition for rehearing, we
specifically asked for briefing on appropriate remedies. See Harper, ___ N.C. at ___,
881 S.E.2d at 550 (order granting Legislative Defendants’ petition for rehearing). As
we did in Stephenson I, “we must now consider the practical consequences of our
holding and address any required remedial measures.” Stephenson I, 355 N.C. at 375,
562 S.E.2d at 392; see also Scott v. Germano, 381 U.S. 407, 409, 85 S. Ct. 1525, 1527
(1965) (“The power of the judiciary of a State to require valid reapportionment or to
formulate a valid redistricting plan has not only been recognized by this Court but
appropriate action by the States in such cases has been specifically encouraged.”).
Legislative Defendants maintain that neither the remedial 2022 Plans nor the
original 2021 Plans were “established” as intended in Article II, Sections 3(4) and
5(4). We agree.
In Harper I four members of this Court wrongly held that partisan
gerrymandering claims are justiciable and violate provisions of the Declaration of
Rights in the North Carolina Constitution. This Court then also erroneously declared
that the 2021 Plans were unconstitutional partisan gerrymanders and “enjoin[ed] the
-138-
HARPER V. HALL
Opinion of the Court
use of [the 2021 Plans] in any future elections.” The 2021 Plans should not have been
enjoined, and this Court should not have ordered the General Assembly to draw
remedial plans using the erroneous standards set forth in Harper I. Nonetheless, this
Court’s Harper I decision forced redistricting criteria upon the General Assembly that
our constitution does not require. Accordingly, the 2022 Plans are a product of a
misapprehension of the law and of Harper I’s violation of separation of powers.
Because Harper I’s misapprehension of our constitutional law generated the
2022 Plans, they were never “established” as that word is used in Article II, Sections
3(4) and 5(4). Additionally, by statute the General Assembly is not required to utilize
the 2022 Plans for future elections. See also N.C.G.S. § 120-2.4(a1) (providing that a
court-imposed remedial map may only be used in the next general election).
Thus, if the 2022 Plans are no longer in force, the question arises whether the
original 2021 Plans are reinstated. In their petition for rehearing and supplemental
brief, Legislative Defendants argued that the 2021 Plans were likewise never
“established” pursuant to Article II, Sections 3(4) and 5(4). Legislative Defendants
point out that the 2021 Plans lasted just over a month before this Court enjoined their
use in the remedial order in Harper I and that the 2021 Plans were never used in an
election. As a direct result of the Harper I decision, the 2022 Plans were drawn,
elections were held based on those remedial districts, and new legislators took their
-139-
HARPER V. HALL
Opinion of the Court
seats in the General Assembly. Legislative Defendants point out that because the
2022 Plans were used in the 2022 election cycle, use of the 2021 Plans for the next
election cycle would “double-bunk” many legislators.24 Legislative Defendants point
to the long history of our cases directing that, when necessary, the General Assembly
must be given the opportunity to redraw constitutionally compliant districts. See, e.g.,
Stephenson II, 357 N.C. at 303, 582 S.E.2d at 248−49; Stephenson I, 355 N.C. at 385,
562 S.E.2d at 398. We agree with Legislative Defendants’ analysis.
Moreover, when reviewing the history behind the General Assembly’s adoption
of the first set of redistricting plans challenged in this case (2021 Plans), it becomes
clear that these plans are also a product of a misapprehension of North Carolina law.
In 2018, just a few years before the enactment of the 2021 Plans, the North Carolina
Democratic Party and a group of North Carolina voters brought a state court action
challenging remedial legislative redistricting plans drawn by the General Assembly
the previous year (2017 Plans).25 See generally Compl., Common Cause v. Lewis, No.
18 CVS 014001, 2019 WL 4569584 (N.C. Super. Ct. Wake County Sept. 3, 2019). The
24The dissent concedes that incumbency protection—that is, avoiding the double-
bunking of incumbent legislators, is a permissible, neutral redistricting criteria.
25 The General Assembly enacted the 2017 Plans after a federal district court found
that several of the legislative districts in the 2011 Plans were racially gerrymandered. See
Covington v. North Carolina, 283 F. Supp. 3d 410, 413 (M.D.N.C. 2018), aff’d in part and
rev’d in part, 138 S. Ct. 2548 (2018) (per curiam).
-140-
HARPER V. HALL
Opinion of the Court
plaintiffs in that case brought the exact claims that are at issue in this case—they
argued that the 2017 Plans were partisan gerrymanders in violation of the free
elections clause, the equal protection clause, and the freedom of speech and assembly
clauses of North Carolina’s Declaration of Rights. Id. at 60−68.
Despite having the benefit of the Supreme Court’s decision in Rucho, the three-
judge panel in Common Cause v. Lewis agreed with the plaintiffs that these
Declaration of Rights provisions prohibit partisan gerrymandering, Lewis, 2019 WL
4569584, at *3, *108−24, *129, and that the General Assembly’s use of partisan
election data to assign voters to districts violated these provisions. See id. at *121.
The panel in Lewis concluded that partisan gerrymandering claims are justiciable
under the North Carolina Declaration of Rights.26 Id. at *126. The Lewis order clearly
represents a mistaken understanding of the North Carolina Constitution—the same
mistaken understanding made by four members of this Court in Harper I and
corrected by this Court today.
26The Lewis panel reached these conclusions even though it had the benefit of the
Supreme Court’s Rucho opinion, which was issued slightly over two months before the Lewis
order. These conclusions also conflicted with this Court’s holdings in Dickson I and Dickson
II that suggested that the Declaration of Rights generally does not provide judicially
manageable standards for claims related to gerrymandering. See Dickson I, 367 N.C. at 575,
766 S.E.2d at 260; Dickson II, 368 N.C. at 534, 781 S.E.2d at 440−41. Of note, the three-judge
panel in Lewis and the three-judge panel in Dickson I consisted of the same three superior
court judges.
-141-
HARPER V. HALL
Opinion of the Court
The panel in Lewis ordered the General Assembly to redraw the 2017 Plans
using specific redistricting criteria and methods enumerated in the Lewis order. Id.
at *136. Many of the required or prohibited criteria in the Lewis order are not derived
from the express language of the constitution. Notably, to prevent partisan
gerrymandering, the Lewis panel explicitly prohibited the General Assembly from
considering any partisan election data in its remedial process.27 Id. As demonstrated
by our opinion today, however, this proscription on the use of partisan data
constituted judicial error because our constitution does not address the use of
partisan data in the redistricting process.
Nevertheless, to comply with the Lewis order the General Assembly proceeded
under the assumption that it could not consider any partisan election data in its
redistricting process without risking a constitutional violation. In 2021, when the
General Assembly first began drawing the 2021 Plans, it convened a Joint Meeting
of the Senate Redistricting and Elections Committee and the House Redistricting
Committee. For purposes of discussing the criteria that would govern the 2021
redistricting process, each Committee member received a copy of the criteria
mandated by the Lewis panel in 2019. One week later, the Joint Redistricting
27 Ironically, the Harper I majority struck the 2021 Plans and then required the
General Assembly to use partisan data in redrawing the plans.
-142-
HARPER V. HALL
Opinion of the Court
Committee adopted finalized criteria for its 2021 map drawing process (Adopted
Criteria). The Adopted Criteria were nearly identical to the criteria mandated by the
Lewis panel. Specifically, the Adopted Criteria, just like the criteria from Lewis,
included a prohibition on consideration of partisan election data. Legislative
Defendants suggest that the Joint Redistricting Committee incorporated this
requirement into its Adopted Criteria because it believed that requirement was
necessary to create constitutionally compliant redistricting plans. See Legislative
Defendants-Appellees’ Brief at 20−21, Harper v. Hall, 380 N.C. 317 (2022) (No.
413PA21-1) (noting that “[t]o avoid violations identified in the 2010 [redistricting]
cycle,” including those identified in the Lewis order, the General Assembly included
prohibition on the consideration of partisan election data in its Adopted Criteria).
As demonstrated by today’s opinion, however, that prohibition does not exist.
Our constitution does not speak to partisan considerations—or any other
considerations not explicitly addressed in the text of our constitution or federal law—
in the redistricting process. Just as this Court’s Harper I decision forced the General
Assembly to draw the 2022 Plans under a mistaken interpretation of our constitution,
the Lewis order forced the General Assembly to draw the 2021 Plans under the same
mistaken interpretation of our constitution. Accordingly, the districts were not
constitutionally “established.” To hold otherwise would perpetuate the same violation
-143-
HARPER V. HALL
Opinion of the Court
of separation of powers that we have attempted to cure today. Thus, the 2021 Plans
are not “established,” as that phrase is used in Article II, Sections 3(4) and 5(4).
The General Assembly shall have the opportunity to enact a new set of
legislative and congressional redistricting plans, guided by federal law, the objective
constraints in Article II, Sections 3 and 5, and this opinion. “When established” in
accordance with a proper understanding of the North Carolina Constitution, the new
legislative plans “shall remain unaltered until the return of” the next decennial
census. N.C. Const. art. II, §§ 3(4), 5(4).
IX. Conclusion
For 200 years our Supreme Court has faithfully sought to implement the intent
of the drafters of our state constitution by interpreting that foundational document
based on its plain language and the historical context in which each provision arose.
Recently, this Court has strayed from this historic method of interpretation to one
where the majority of justices insert their own opinions and effectively rewrite the
constitution. Today we return to the text of the state constitution, correct our course,
and come back to the proper understanding and application of our fundamental
constitutional principles. Apportionment is textually committed to the General
Assembly, and apportionment legislation is entitled to our long-standing standard of
review—a presumption of constitutionality and a required showing that the
-144-
HARPER V. HALL
Opinion of the Court
legislation is unconstitutional beyond a reasonable doubt. There is no judicially
manageable standard by which to adjudicate partisan gerrymandering claims. Courts
are not intended to meddle in policy matters. In its decision today, the Court returns
to its tradition of honoring the constitutional roles assigned to each branch.
This case is not about partisan politics but rather about realigning the proper
roles of the judicial and legislative branches. Today we begin to correct course,
returning the judiciary to its designated lane.
The Court’s authority—possessed of neither the purse nor
the sword—ultimately rests on sustained public confidence
in its moral sanction. Such feeling must be nourished by
the Court’s complete detachment, in fact and in
appearance, from political entanglements and by
abstention from injecting itself into the clash of political
forces in political settlements.
Baker, 369 U.S. at 267, 82 S. Ct. at 737−38 (Frankfurter, J., dissenting).
We have recognized that our constitution allows the General Assembly to enact
laws unless expressly prohibited by the constitutional text. This Court will no longer
change the time-honored meaning of various portions of our constitution by
interpreting the text with the singular aim of reaching a desired outcome. As
explicitly stated in our constitution, the people have the authority to alter their
foundational document, not this Court. The people alone have the final say.
This Court’s opinion in Harper I is overruled. We affirm the three-judge panel’s
-145-
HARPER V. HALL
Opinion of the Court
11 January 2022 Judgment concluding, inter alia, that claims of partisan
gerrymandering present nonjusticiable, political questions and dismissing all of
plaintiffs’ claims with prejudice. This Court’s opinion in Harper II is withdrawn and
superseded by this opinion. The three-judge panel’s 23 February 2022 order
addressing the Remedial Plans is vacated. Plaintiffs’ claims are dismissed with
prejudice.
VACATED.
-146-
HARPER V. HALL
Earls, J., dissenting
Justice EARLS dissenting.
Following the 2010 census and prior to the United States Supreme Court’s
decision in Rucho v. Common Cause, 139 S. Ct. 2484 (2019), one of the Republican co-
chairs of the General Assembly’s redistricting committee, Representative David
Lewis, explained his rationale in presenting redistricting plans that
disproportionately favored Republicans: “I think electing Republicans is better than
electing Democrats. So I drew this map to help foster what I think is better for the
country.” Id. at 2491. Though jarring in its irreverence to democracy, Representative
Lewis simply admitted what all of the evidence subsequently showed about
redistricting maps enacted by the North Carolina General Assembly in recent years:
They stifle the will of North Carolina voters by rigging the system against one party
in favor of another. Representative Lewis’s views carried the day. The General
Assembly adopted a “partisan advantage” redistricting criterion that required the
districts to maintain a ten to three Republican/Democrat congressional delegation.
See Common Cause v. Rucho, 318 F. Supp. 3d 777, 807 (M.D.N.C. 2018), overruled by
Rucho, 139 S. Ct. 2484. Those maps were ultimately held to be unconstitutional under
the North Carolina Constitution in a ruling that was never appealed to this Court.
See Common Cause v. Lewis, No. 18 CVS 014001, 2019 WL 4569584, at *2 (N.C.
Super. Ct. Sept. 3, 2019) (holding that, when these maps were created, “partisan
-147-
HARPER V. HALL
Earls, J., dissenting
intent predominated over all other redistricting criteria resulting in extreme partisan
gerrymander[s]”).
When the General Assembly attempted to enact a new extreme partisan
gerrymander just a few years later following the release of 2020 census data, this
Court rejected the idea that the voters of this state must be hostage to the partisan
objectives of the ruling party in the General Assembly. And for a brief window in
time, the power of deciding who is elected to office was given to the people, as required
by the state constitution. See Harper v. Hall (Harper I), 380 N.C. 317, 339, cert.
granted sub nom. Moore v. Harper, 142 S. Ct. 2901 (2022), vacated, Harper v. Hall,
No. 413PA21-2 (N.C. Apr. 28, 2023); Harper v. Hall (Harper II), 383 N.C. 89 (2022),
vacated, Harper v. Hall, No. 413PA21-2 (N.C. Apr. 28, 2023). In Harper I, this Court
ensured that all North Carolinians, regardless of political party, were not denied their
“fundamental right to vote on equal terms.” Harper I, 380 N.C. at 378 (cleaned up).
Today, the majority strips the people of this right; it tells North Carolinians
that the state constitution and the courts cannot protect their basic human right to
self-governance and self-determination. In so doing, the majority ignores the
uncontested truths about the intentions behind partisan gerrymandering and erects
an unconvincing façade that only parrots democratic values in an attempt to defend
its decision. Despite its lofty prose about the need for principled adherence to the
state constitution, the majority follows none of these principles today. Nor does the
majority even pay passing reference to the anti-democratic nature of extreme
-148-
HARPER V. HALL
Earls, J., dissenting
partisan gerrymandering. These efforts to downplay the practice do not erase its
consequences and the public will not be gaslighted. Our constitution provides that
“[a]ll political power is vested in and derived from the people; all government of right
originates from the people, is founded upon their will only.” N.C. Const. art. I, § 2.
But when Republican lawmakers are free to gerrymander redistricting plans without
constitutional guardrails to ensure their party’s indefinite political domination, this
constitutional requirement is abandoned.
Unchecked partisan gerrymandering allows the controlling party of the
General Assembly to draw legislative redistricting plans in a way that dilutes the
voting power of voters in the disfavored party. In so doing, those who hold political
power can guarantee that they remain in office for decades, making them impervious
to the popular will. Thus, rather than allowing “the people . . . [to] choose whom they
please to govern them,” as Alexander Hamilton once described as “the true principle
of a republic,” 2 Debates on the Constitution 257 (J. Elliot ed. 1891), members of the
General Assembly make this choice for the people, favoring Republicans because they
believe that electing Republicans is better for the country. This is not how democracy
should function.
What is more, the majority abolishes the fundamental right to vote on equal
terms regardless of political party through a process driven by partisan influence and
greed for power. Let there be no illusions about what motivates the majority’s decision
to rewrite this Court’s precedent. Today’s result was preordained on 8 November
-149-
HARPER V. HALL
Earls, J., dissenting
2022, when two new members of this Court were elected to establish this Court’s
conservative majority. To the Court’s new majority, the parties’ briefing after
rehearing was granted did not matter.1 The oral argument held after rehearing was
granted did not matter. The merits of Plaintiffs’ arguments do not matter. For at
stake in this case is the majority’s own political agenda. Today, the Court shows that
its own will is more powerful than the voices of North Carolina’s voters.
To be clear, this is not a situation in which a Democrat-controlled Court
preferred Democrat-leaning districts and a Republican-controlled Court now prefers
Republican-leaning districts. Here, a Democratic-controlled Court carried out its
sworn duty to uphold the state constitution’s guarantee of free elections, fair to all
voters of both parties. This decision is now vacated by a Republican-controlled Court
seeking to ensure that extreme partisan gerrymanders favoring Republicans are
established.2
1 Exhibiting its disregard for the merits of the arguments like those presented by
Plaintiffs, the Court denied two parties’ motions for leave to file amicus curie briefs in support
of Plaintiffs. See Harper v. Hall, 2022-NCSC-121 (March 9, 2023) (order on motion of
Governor Roy Cooper and Attorney General Joshua H. Stein for leave to file amicus brief in
support of plaintiffs-appellants); Harper v. Hall, 2022-NCSC-121 (March 9, 2023) (order on
motion of the Brennan Center for Justice at N.Y.U. School of Law for leave to file amicus
curiae brief in support of plaintiffs-appellants on rehearing). I would have allowed the
motions.
2 For instance, the majority in Harper I recognized that “our responsibility is to
determine whether challenged apportionment maps encumber the constitutional rights of
the people to vote on equal terms and to substantially equal voting power.” Harper I, 380
N.C. at 323. By contrast, today’s majority believes that its responsibility is to protect the
plans that the trial court found to be “egregious and intentional partisan gerrymanders,
designed to enhance Republican performance, and thereby give a greater voice to those voters
than to any others.” Id. at 324.
-150-
HARPER V. HALL
Earls, J., dissenting
In a single blow, the majority strips millions of voters of this state of their
fundamental, constitutional rights and delivers on the threat that “our decisions are
fleeting, and our precedent is only as enduring as the terms of the justices who sit on
the bench.” See Harper v. Hall, No. 413PA21, 2023 WL 1516190 (N.C. Feb. 3, 2022)
(order allowing motion for rehearing) (Earls, J., dissenting) [hereinafter Harper
Order].
I. Background
Though the majority explains the history of this case in depth, it neglects to
make any mention of the practical effect of the maps that sparked and perpetuated
this litigation. In the cases that the majority vacates and overturns today, Harper I
and Harper II, the Court explained at great length the severity of the partisan
gerrymanders that the General Assembly crafted. See Harper I, 380 N.C. at 333–46;
Harper II, 383 N.C. at 100–111, 114–23. I therefore summarize only briefly where
this litigation began.
Following the 2020 Decennial Census, the North Carolina General Assembly
enacted new redistricting plans for the North Carolina House of Representatives, the
North Carolina Senate, and the U.S. House of Representatives (2021 Plans). In
November 2021, North Carolina League of Conservation Voters, Inc. (NCLCV) and
Harper Plaintiffs challenged the plans as unconstitutional partisan gerrymanders in
separate suits that were assigned to the same three-judge panel and consolidated in
-151-
HARPER V. HALL
Earls, J., dissenting
December 2021. That same month, Plaintiff Common Cause moved to intervene in
the litigation, and the three-judge panel granted the motion.
In a 258-page opinion issued in January 2022, the three-judge panel
unanimously found that the 2021 Plans constituted extreme partisan gerrymanders.
Specifically, the trial court found that the 2021 Congressional Plan was an
“intentional, and effective, pro-Republican partisan redistricting” that all but
guaranteed Republicans ten out of fourteen seats in the U.S. House of
Representatives. The trial court further found that “the enacted congressional map
is more carefully crafted to favor Republicans than at least 99.9999% of all possible
maps” using nonpartisan redistricting criteria. Harper I, 380 N.C. at 339. These
results were no accident, the trial court concluded. Instead, “the 2021 Congressional
Plan is a partisan outlier intentionally and carefully designed to maximize
Republican advantage in North Carolina’s Congressional delegation.” The trial court
further explained that “Legislative Defendants offered no defense of the 2021
Congressional Plan. No expert witness opined that it was not the product of an
intentional partisan redistricting.”
The state legislative districts fared no better. For example, the trial court found
that the enacted State Senate Plan
effectuate[d] the same sort of partisan advantage as the
Enacted Congressional Plan. The Enacted Senate Plan
consistently creates Republican majorities and precludes
Democrats from winning a majority in the Senate even
when Democrats win more votes. Even in an essentially
tied election or a close Democratic victory, the Enacted
-152-
HARPER V. HALL
Earls, J., dissenting
Senate Plan gives Republicans a Senate majority, and
sometimes even a veto-proof 30-seat majority. And that
result holds even when Democrats win by larger margins.
Harper I, 380 N.C. at 341.
Similarly, the trial court concluded that “the Enacted House Plan is also
designed to systematically prevent Democrats from gaining a tie or a majority in the
House. In close elections, the Enacted House Plan always gives Republicans a
substantial House majority. That Republican majority . . . persists even when voters
clearly express a preference for Democratic candidates.” Id. The trial court also
concluded that “[t]he 2021 House Plan’s partisan bias creates firewalls protecting the
Republican supermajority and majority in the House.”
So, this is where we started. And when confronted with three different
legislative redistricting plans that were all found to have been intentional attempts
to consolidate Republican power and suppress the will of the voters, this Court chose
to protect the democratic ideals enshrined in our state constitution and the voters
themselves over the political and partisan motivations of a select few in the General
Assembly. Today, the Court reverses course and chooses the latter. Even beyond this
particular decision, the majority has already repeatedly revealed itself to be on a
mission to pursue the agenda of this select few in the legislature. See Holmes v. Moore,
No. 342PA19-3 (N.C. Apr. 28, 2023); Cmty. Success Initiative v. Moore, No. 331PA21
(N.C. Apr. 28, 2023). Its allegiances need no further explanation.
-153-
HARPER V. HALL
Earls, J., dissenting
II. Analysis
A. Remedy
Though it may seem out of order, I begin by addressing the remedy the majority
provides Legislative Defendants today as it is a primer for the lawlessness that recurs
throughout this opinion. The majority makes repeated declarations that “[t]he
constitution is interpreted based on its plain language”—that “[t]he constitution was
written to be understood by everyone, not just a select few.” But the majority also
consistently struggles to apply those principles itself. Nowhere is this more evident
than in the remedy the majority awards Legislative Defendants.
What Legislative Defendants want is a do over—a chance to go back in time
and draw even more egregiously gerrymandered maps than they did before this
litigation began. Because of the majority’s decision today, they now have the
assurance that they will get away with it. And as they correctly predicted, what
Legislative Defendants want, the majority will provide. The majority’s self-
congratulatory exercise of judicial restraint suddenly vanishes when Legislative
Defendants seek a remedy that the state constitution expressly prohibits. Though the
constitutional text may be an inconvenience to the majority’s desire to carry out
Legislative Defendants’ political agenda, it is not something that can be so easily
disregarded at will.
There is a strict constitutional limitation on the General Assembly’s power to
draw state legislative districts. Article II, sections 3 and 5 expressly provide that
-154-
HARPER V. HALL
Earls, J., dissenting
“[t]he General Assembly, at the first regular session convening after the return of every
decennial census of population taken by order of Congress, shall revise [the senate
and the representative] districts and the apportionment of [senators and
representatives] among those districts.” N.C. Const. art. II, §§ 3, 5 (emphasis added).
But these sections further provide that, “[w]hen established,” both the apportionment
of members of the state senate and house of representatives and their districts “shall
remain unaltered until the return of another decennial census of population taken by
order of Congress.” N.C. Const. art. II, §§ 3(4), 5(4). The meaning of this requirement
is simple: Once the districts have been established, or passed, by the General
Assembly, the districts and apportionment of members of the General Assembly are
fixed until the next census.
This Court has applied the provisions strictly. Shortly after the provisions were
ratified in their original form, this Court held that they prohibited the mid-decade
redrawing of the border between Franklin County and Granville County, even though
the border as drawn violated another constitutional provision requiring that “no
county shall be divided in the formation of a Senate district.” N.C. Const. of 1868, art.
II, § 5; Comm’rs of Granville Cnty. v. Ballard, 69 N.C. 18, 20–21 (1873). But the plain
text of article II, sections 3(4) and 5(4) and the history of these provisions simply will
not do for the majority.
Step one in the majority’s scheme is therefore to do away with the remedial
maps (2022 Plans) that Harper I ordered the General Assembly to draw. To that end,
-155-
HARPER V. HALL
Earls, J., dissenting
the majority must first redefine what the word “established” means. The majority
relies on Black’s Law dictionary to define the term “established” as “[t]o settle, make,
or fix firmly; to enact permanently.” Establish, Black’s Law Dictionary (11th ed.
2019). The majority reasons that, using this definition, the 2022 Plans were not
“established” for purposes of article II, sections 3(4) and 5(4) because this definition
“connotes something more than the passage of a redistricting act by the General
Assembly” because the General Assembly was free to amend the maps until they
were used in an election.
But this definition creates a problem for the majority. Not only were the 2022
Plans validly enacted by the General Assembly during its first regular session
following the 2020 Census, they were also used in the 2022 primaries and general
election. That means that the 2022 Plans fall squarely within the majority’s own
definition of the word “established” as used in article II, sections 3(4) and 5(4). Thus,
the majority must create an exception to its definition of the term “established” that
lacks any basis in the constitutional text. Specifically, the majority reasons that,
because the 2022 plans were based on a misapprehension of law, “they were never
‘established’ as that word is used in article II, sections 3(4) and 5(4).”
Interestingly, nowhere in the majority’s definition of the term “established” is
there an exception for such a misapprehension of law—the majority itself holds that
a redistricting plan is established when, as here, it is enacted by the General
Assembly and used in an election. The majority does not provide any legal support
-156-
HARPER V. HALL
Earls, J., dissenting
for the idea that a change in the law justifies the redistricting redo that Legislative
Defendants seek, nor that such a permission is consistent with the text, purpose, or
history of the state constitution’s mid-decade redistricting prohibition. That is
because there is no legal basis for throwing out the 2022 Plans in the middle of the
decade.3
But the majority does not stop there. Cue step two in the majority’s efforts to
carry out Legislative Defendants’ bidding. The majority concludes that, not only must
the 2022 Plans be thrown out, so too must the 2021 Plans that the General Assembly
enacted following the 2020 census before this litigation ever began. Its reasoning is
stunning—the 2021 Plans must be thrown out, it explains, because both because
using the 2021 maps would not sufficiently protect seats for incumbent candidates
and because these plans were allegedly based on a misapprehension of law from a
different case decided years earlier. See Common Cause v. Lewis, No. 18 CVS 014001,
2019 WL 4569584 (N.C. Super. Ct. Sept. 3, 2019). As to the first point, that
incumbents could be better protected through a different map is not a basis for
3 The majority also makes the false assertion that “by statute[,] the General Assembly
is not required to utilize the 2022 Plans for future elections.” See N.C.G.S. § 120-2.4(a1). This
is a blatant mischaracterization of the statute. N.C.G.S. § 120-2.4(a1) provides that, when
the legislature is required to enact a remedial map but fails to “act to remedy any identified
defects” within the timeframe that has been prescribed by a court, the court may impose an
interim plan that will be used in the next election only. N.C.G.S. § 120-2.4(a1). The court-
imposed plan is only “interim” if the General Assembly fails to enact a remedial map on its
own accord. That is not what happened here, as the General Assembly itself passed the
remedial 2022 Plans during its first regular legislative session. Its enacted remedial plans
have the same force and effect as any other redistricting plans that it validly enacts, and they
are treated the same.
-157-
HARPER V. HALL
Earls, J., dissenting
ignoring the constitutional mandate against mid-decade redistricting. The state
constitution does not authorize legislative districts to be redrawn in the middle of a
decade simply to allow the General Assembly to better account for a particular
redistricting criteria and certainly not for the dubious purpose of better protecting
incumbent legislators.
With the respect to the majority’s latter point that the 2021 maps were based
on a misapprehension of law, it relies on a superior court decision that was never
heard by a North Carolina appellate court. Lewis, 2019 WL 4569584 at *2–3. In
Lewis, the plaintiffs brought similar partisan gerrymandering claims against
different legislative maps. Id. The trial court held that the maps were extreme
partisan gerrymanders and violated the state constitution. Id. at 3. But according to
the majority, because of that decision, which is unrelated to this litigation, unrelated
to the 2021 Plans, and was not decided by this Court, when Legislative Defendants
enacted the 2021 Plans over a year later, they were enacted under “a mistaken
understanding of the North Carolina Constitution.” Somehow this mistaken
understanding equates to a failure to establish the legislative plans. In other words,
the majority believes that because it might be possible to enact an even more extreme
partisan gerrymander than was enacted in 2021, the General Assembly should be
allowed to do so, despite the prohibition on mid-decade redistricting of state
legislative districts.
The majority points to the fact that in 2021, when the General Assembly
-158-
HARPER V. HALL
Earls, J., dissenting
started the map drawing process after census data was first released, among the
districting criteria that the General Assembly adopted was the requirement that
partisan election data not be considered in defining legislative districts. The majority
credits Legislative Defendants assertion to this Court in Harper I that the General
Assembly adopted “this requirement . . . because it believed that requirement was
necessary to create constitutionally compliant redistricting plans.” Notably,
Legislative Defendants’ single, vague assertion that the majority hinges its
conclusion on does not argue that the 2021 maps were free of intentional partisan
bias. Such a claim would have been untrue. But the majority refuses to examine any
of the evidence in the record that demonstrates the role partisan considerations
played in the creation of the 2021 Plans and proves that this this criterion was
adopted in name only. This is not surprising—recognizing as much would require the
majority to acknowledge that the General Assembly already took advantage of the
opportunity to enact maps containing extreme partisan gerrymanders.
As has been discussed, almost every shred of evidence in the record shows that
the 2021 maps were extreme partisan gerrymanders, which is why the trial court
specifically found as much. But not only did the 2021 Plans themselves evince that
they were drawn to disproportionately favor Republicans, so too did the events
leading to their enactment. For example, Legislative Defendants claimed that
potential maps must be drawn and submitted in committee hearing rooms using
software that did not account for partisan election data. Defendant Representative
-159-
HARPER V. HALL
Earls, J., dissenting
Destin Hall, the Chair of the House Redistricting Committees assured his colleagues
that the “House as a whole” would “only consider maps that are drawn in this
committee room, on one of the four stations” located in the committee room.
Contrary to these assurances, however, legislators and their staff were able to
use partisan data to draw gerrymandered maps on unofficial devices both inside and
outside of the committee rooms. Evidence at trial revealed that Representative Hall
repeatedly met with members of his staff to review “concept maps” that were created
on unofficial computers using unknown redistricting software and data.
Representative Hall testified that he would then rely on these concept maps when
drawing proposed maps on the committee room computers. In fact, on several
occasions, when drawing maps on the official terminals in the committee rooms,
Representative Hall even brought along a smartphone containing images of the
concept maps so that he could copy the concept map into the public terminal.
Legislative Defendants denied that they used any non-public materials as part
of their map-drawing activities at first, but they were eventually forced to admit that
this was false. The trial court ordered Legislative Defendants to produce the “concept
maps” and related materials. Legislative Defendants failed to do so, and instead
claimed that “the concept maps that were created were not saved, are currently lost
and no longer exist.” Based on this history as well as the extremity of the maps
themselves, the majority’s suggestion that the 2021 Plans were based on the
“incorrect” notion that partisan gerrymandering violates the state constitution is
-160-
HARPER V. HALL
Earls, J., dissenting
plainly false.
Even if it were true that the General Assembly did not consider partisan data
in drawing the 2021 Plans, it would not matter. As already explained, the constitution
proscribes mid-decade redistricting after districts are established. There is no
constitutional caveat providing that a district might become “un-established” if a
change in the law means the districts could have been drawn differently the first time
around. If this were true, legislative redistricting plans would never officially be
established for purposes of article II, sections 3 and 5. The potential for a future
hypothetical change in the law would permanently leave every redistricting plan
enacted by the General Assembly in a state of limbo. The state constitution does not
afford Legislative Defendants a do-over simply because they believe that they can do
a better job of manipulating election outcomes this time around.
Finally, the General Assembly has already expressed its intent that the 2021
Plans should take effect if the 2022 Plans were to be thrown out. Specifically, the
2022 enactments establishing the 2022 Plans (i.e., the remedial plans) for both the
North Carolina Senate and House of Representatives explained that should the
Court’s decision in Harper I be “made inoperable . . . or ineffective,” the 2021 Plans
would, by operation of law, become “again effective.” An Act to Realign the North
Carolina Senate Districts Pursuant to the Order of the North Carolina Supreme
Court in Harper v. Hall, S.L. 2022-2, § 2, 2022 N.C. Sess. Laws 14, 19 (Senate plan);
An Act to Realign North Carolina House of Representatives Districts Pursuant to
-161-
HARPER V. HALL
Earls, J., dissenting
Order of the North Carolina Supreme Court in Harper v. Hall, S.L. 2022-4, § 2, 2022
N.C. Sess. Laws 30, 43 (House plan). Thus, this Court need not speculate about what
the General Assembly intended if, for some reason, the 2022 Plans became
“ineffective.” By ordering that the 2021 Plans be disregarded, this Court violates the
intent of the General Assembly expressed by the body as a whole through formal
legislation, rather than a few of its members involved in this litigation.
None of this matters to the majority. Reason, common sense, and the rule of
law are lost on those who do not care about interpreting the constitution in good faith.
This holding is not a mere error in legal interpretation—I do not think that even the
majority believes itself to be complying with the constitutional text where this remedy
is concerned, as demonstrated by its lack of effort in attempting to support its radical
decision. The remedy afforded here demonstrates how divorced from the law the
majority’s decision is in its entirety. It shatters the notion that the majority is
applying the constitution “based on its plain language” or that “[t]his case is not about
partisan politics.” Put simply, the majority today instructs the General Assembly to
violate the North Carolina constitution. In so doing, it puts on display just how far
this Court has fallen.
B. Partisan Gerrymandering Violates the State Constitution
“No right is more precious in a free country than that of having a voice in the
election of those who make the laws under which, as good citizens, we must live.”
Blankenship v. Bartlett, 363 N.C. 518, 522 (2009) (quoting Wesberry v. Sanders, 376
-162-
HARPER V. HALL
Earls, J., dissenting
U.S. 1, 17 (1964)). As James Madison explained in the Federalist Papers,
“[R]epublican liberty” requires “not only that all power should be derived from the
people; but that those entrusted with it should be kept in dependence on the people.”
Rucho, 139 S. Ct. at 2512 (Kagan, J., dissenting) (quoting The Federalist No. 37, at 4
(James Madison) (J. & A. McLean ed., 1788)). This principle applies not just to the
federal government but to our state as well, for it “is the foundation of democratic
governance.” Id. at 2511–12. Indeed, this very principle is enshrined in our state
constitution, which commands that “[a]ll political power is vested in and derived from
the people; all government of right originates from the people, is founded upon their
will only, and is instituted solely for the good of the whole.” N.C. Const. art. I, § 2.
The extreme partisan gerrymanders that this Court addressed in Harper I and
Harper II made a mockery of those principles and “enabled politicians to entrench
themselves in office as against voters’ preferences. They promoted partisanship above
respect for the popular will. They encouraged a politics of polarization and
dysfunction.” Rucho, 139 S. Ct. at 2509 (Kagan, J., dissenting). In so doing, these
partisan gerrymanders “deprived citizens of the most fundamental of their
constitutional rights: the rights to participate equally in the political process, to join
with others to advance political beliefs, and to choose their political representatives.”
Id. By violating these rights, the plans at issue and the politicians who manipulated
them “debased and dishonored our democracy, turning upside-down the core
American idea that all governmental power derives from the people.” Id. With the
-163-
HARPER V. HALL
Earls, J., dissenting
practice now condoned by this Court’s current majority, the select few in the General
Assembly who crafted the plans, themselves elected under gerrymandered maps, will
make every attempt to entrench their party in the General Assembly indefinitely,
regardless of what North Carolinians have to say about it. See, e.g., Lewis, 2019 WL
4569584, at *8–9, *14–18.
Not only does the majority fail to recognize the anti-democratic nature of these
realities. It goes a step further than any opinion of the full U.S. Supreme Court has
gone before and concludes that, not only is partisan gerrymandering nonjusticiable,
it is actually permitted by the state constitution. As James Madison once cautioned,
the majority misplaces political power “in the Government over the people.” 4 Annals
of Cong. 934 (1794).
Harper I painstakingly laid out the history, requirements, and guarantees of
the constitutional rights that are implicated here—the free elections clause, the equal
protection clause, the free speech clause, and the freedom of assembly clause. I do not
here repeat Harper I’s correct interpretation of these rights, as the principles and
history that Harper I articulated are far more enduring than the majority’s monopoly
on the judicial power. I do, however, address the butchered and curtailed definition
of the free elections clause the majority adopts today and share a few additional
observations about the state’s equal protection clause.
1. The Free Elections Clause
The majority proclaims that “[t]he constitution is interpreted based on its plain
-164-
HARPER V. HALL
Earls, J., dissenting
language” and that “[t]he constitution was written to be understood by everyone, not
just a select few.” It appears that the majority and I agree on at least two points, in
principle at least; we just disagree about what these concepts look like in practice.
The majority’s interpretation of the free elections clause highlights the point. Article
I, section 10 of the North Carolina constitution, known as the free elections clause,
states very simply that “[a]ll elections shall be free.” N.C. Const. art. I, § 10. That is
all. While this clause may seem easy enough for “everyone” to make sense of, not so
in the majority’s view. It takes the Court over twenty pages of convoluted legal
reasoning to explain why the word “free” does not actually mean what one might
think it does. This does not mean that brevity begets accuracy. But neither does the
majority’s odyssey to redefine a simple and explicit requirement in the North
Carolina constitution.
I begin where the majority does: with the dictionary definition of the word
“free.” Moreover, I use the same dictionary definition as does the majority, as the
Court omits a few notable considerations. Black’s Law Dictionary defines the term
“free” as, among other things, “[h]aving legal and political rights; enjoying political
and civil liberty”; “[n]ot subject to the constraint or domination of another; enjoying
personal freedom; emancipated”; “[c]haracterized by choice, rather than by
compulsion or constraint.” Free, Black’s Law Dictionary (11th ed. 2019) (emphases
added). Merriam Webster’s provides additional guidance, encapsulating the
definitions identified above but adding that “free” means “not determined by anything
-165-
HARPER V. HALL
Earls, J., dissenting
beyond its own nature or being: choosing or capable of choosing for itself.” Free,
Merriam-Webster’s Collegiate Dictionary (11th ed. 2022).
With this in mind, we can explore what the free elections clause demands on
its face. In violation of the concept of “free” elections, partisan gerrymandering is a
form of vote dilution—“the devaluation of one citizen’s vote as compared to others,”
Rucho, 139 S. Ct. at 2513 (Kagan, J., dissenting)—that imposes a “constraint” on a
voter’s will. See Free, Black’s Law Dictionary. Justice Kagan explained this process
succinctly in her dissent in Rucho:
A mapmaker draws district lines to “pack” and “crack”
voters likely to support the disfavored party. He packs
supermajorities of those voters into a relatively few
districts, in numbers far greater than needed for their
preferred candidates to prevail. Then he cracks the rest
across many more districts, spreading them so thin that
their candidates will not be able to win. Whether the
person is packed or cracked, his vote carries less weight—
has less consequence—than it would under a neutrally
drawn (non-partisan) map. In short, the mapmaker has
made some votes count for less, because they are likely to
go for the other party.
Id. at 2513–14 (citations omitted). And when done properly, which modern technology
all but assures, it puts representatives, like Legislative Defendants here, in the
business of “rigging elections.” Vieth v. Jubelirer, 541 U.S. 267, 317 (2004) (Kennedy,
J., concurring in the judgment).
A rigged election is not, in any sense of the word, a free election. Nor is an
election in which a voter’s voice is worthless because the election’s results have been
preordained by whoever wields political power in the General Assembly. The majority
-166-
HARPER V. HALL
Earls, J., dissenting
itself acknowledges that the free elections clause was inspired by the English Bill of
Rights, which in turn sought to respond to practices that attempted “to ensure a
certain electoral outcome.” Though the modes of “ensur[ing]” certain electoral
outcomes may have improved with the advent of technology, an election in which the
result is determined by advanced and manipulative map drawing is not,
“[c]haracterized by choice,” as the term “free” requires, but by “constraints” that are
contrived by the legislature alone. See Free, Black’s Law Dictionary.
The majority next turns to the history of the free elections clause. Notably, the
majority does not challenge much of the history surrounding the clause as recounted
in Harper I. In fact, it reiterates much of what Harper I already explained. Instead,
it disagrees with some of the conclusions that Harper I drew from that history.
Because Harper I already successfully completed the task of explaining the historical
underpinnings of the free elections clause, I do not rehash these events here. See
Harper I, 380 N.C. at 373–76. I note only that history cannot be retroactively modified
by the majority.
The majority’s historical analysis warrants a brief comment, however.
Specifically, in analyzing the roots of the free elections clause, the majority examines
a narrow political issue that preceded the clause and the 1776 Declaration of Rights,
namely the tension between North Carolina’s governor and the House of Burgesses
from 1729 until 1776. According to the majority, the free elections clause “was placed
in the 1776 Declaration of Rights at the same time as other constitutional provisions
-167-
HARPER V. HALL
Earls, J., dissenting
that both limited executive power and increased legislative power.” As a result of
these contemporaneous provisions, the majority concludes that “any argument that
the people added the free elections clause to the 1776 constitution for the purpose of
limiting the General Assembly’s apportionment authority is inconsistent with this
historical context.”
This conclusion presents two glaring problems. First, it ignores that the free
elections clause, when first adopted, spoke to the elections of members to the General
Assembly specifically; it did not concern the various disputes that the majority
describes between the governor and the House of Burgesses. Any provisions adopted
to address the balance of power between the governor and the legislative body are
distinct from a provision that demanded the free “election[ ] of members . . . to [the]
General Assembly.” N.C. Const. of 1776, Declaration of Rights, § 6.
Second, and relatedly, was this ongoing feud really the only historically
relevant event that happened in the years leading up to 1776? Can the majority truly
not conceive of anything else that may have driven the people of North Carolina to
embrace the words “election[ ] of members to serve as Representatives in the General
Assembly, ought to be free,” as the clause provided in 1776? N.C. Const. of 1776.
Moreover, might other historical events have inspired an evolved understanding of
the clause as it as well as other constitutional provisions were modified and added
throughout the state’s history, including in 1868? See Harper I, 380 N.C. at 369
(“North Carolina’s Declaration of Rights as it exists today in article I was forged not
-168-
HARPER V. HALL
Earls, J., dissenting
only out of the revolutionary spirit of 1776 but also the reconstruction spirit of 1868.”).
History can, when used properly and appropriately, be useful in giving context
to a constitution. But the majority demonstrates how historical analysis can be
weaponized to paint a distorted picture of a constitution’s historical understanding.
In this way, “it is a magnificent disguise. The judge can do the wildest things, all the
while presenting himself as the passive agent of the sainted Founders—don’t argue
with me, argue with Them.” Richard A. Posner, Bork and Beethoven, 42 Stan. L. Rev.
1365, 1379 (1990). But “bad originalism” has never been a legitimate means of
constitutional interpretation. See id. at 1378.
Finally, the majority attempts to use precedent to support its constrained view
of the free elections clause. As the majority notes, there are few cases that have
interpreted the clause. First, there was Clark v. Meyland, 261 N.C. 140 (1964). There,
the plaintiff sought to change his party affiliation in order to vote in the Republican
primary. Id. at 141. But in order to do so, he was required by statute to take an oath
pledging his allegiance to the new party, including by supporting the nominees from
that party in the subsequent election. Id. Any individual who took the oath falsely
was guilty of a felony. Id. This Court struck down the part of the oath that required
an individual to support the party’s nominees in the future because it “violate[d] the
principle of freedom of conscience. It denies a free ballot––one that is cast according
to the dictates of the voter’s judgment.” Id. at 142. The Court concluded that “the
Legislature is without power to shackle a voter’s conscience by requiring the
-169-
HARPER V. HALL
Earls, J., dissenting
objectionable part of the oath as a price to pay for his right to participate in his party’s
primary.” Id.
Next, the majority cites State ex rel. Swaringen v. Poplin, 211 N.C. 700 (1937),
in which the plaintiff—a candidate for office—claimed that the Wilkes County Board
of Elections fraudulently altered the vote count, leading to the plaintiff’s defeat. Id.at 700–01. Citing the free elections clause and rejecting the Board of Elections’s
argument that it had the sole authority to determine the result of an election, this
Court held that judicial intervention was appropriate and explained that “[a] free
ballot and a fair count must be held inviolable to preserve our democracy.” Id. at 702.
Based on these two cases alone, the majority somehow concludes the free
elections clause encompasses only the right to vote “according to one’s conscience and
to have that vote accurately counted.” This interpretation is confounding. Neither of
these cases in any way limits the free elections clause to the two situations identified
by the majority. The cases that have happened to rule on a specific and limited issue
do not, without more, define the entire scope of a constitutional provision. In
attempting to justify its interpretation of the free elections clause with such an
elementary error in interpreting this Court’s precedent, the majority only emphasizes
how baseless its decision today is. In fact, these errors are so egregious that they
hardly need be explained—they are so glaring that the majority accomplishes the
task on its own.
What is more, if the majority is correct that these cases limit the free elections
-170-
HARPER V. HALL
Earls, J., dissenting
clause to only these two scenarios, then these cases would conflict with the majority’s
own historical analysis of the clause. Again, the majority explains that the
Declaration of Rights was modeled after the English Bill of Rights, which was in turn
an effort to respond to various abuses committed by King James II. But many of the
abuses that the English Bill of Rights sought to address, and therefore the
Declaration of Rights contemplates, do not fit in to the majority’s cabined
interpretation of the free elections clause. For example, the majority explained that,
under King James II, “[w]hen the time for [an] election came, local agents of the king
who conducted the polling used devious polling practices to open, close, and reopen
polling places” to manipulate election outcomes. Under the majority’s newly minted
interpretation of the free elections clause, such a practice would not be proscribed,
and it is certainly not addressed by any other provision in the Declaration of Rights.
2. The Equal Protection Clause
Not only does partisan gerrymandering obstruct the constitution’s promise of
free elections, it also deprives individuals of the “fundamental right to vote on equal
terms,” which is derived from North Carolina’s equal protection clause.4 Stephenson
4 North Carolina’s equal protection clause states that:
[n]o person shall be taken, imprisoned, or disseized of his
freehold, liberties, or privileges, or outlawed, or exiled, or in any
manner deprived of his life, liberty, or property, but by the law
of the land. No person shall be denied the equal protection of
the laws; nor shall any person be subjected to discrimination by
the State because of race, color, religion, or national origin.
N.C. Const. art. I, § 19.
-171-
HARPER V. HALL
Earls, J., dissenting
v. Bartlett (Stephenson I), 355 N.C. 354, 378 (2002). That right “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.” Reynolds v. Sims, 377 U.S. 533, 555
(1964). The majority correctly notes that this Court has stepped in to prevent this
consequence through its one-person, one-vote cases. See Stephenson I, 355 N.C. 354;
Blankenship v. Bartlett, 363 N.C. 518 (2009); Northampton Cnty. Drainage Dist. No.
One v. Bailey, 326 N.C. 742 (1990). These cases recognize that “[e]qual protection
‘requires that all persons similarly situated be treated alike.’ ” Blankenship, 363 N.C.
at 521. Malapportionment—the practice of inequitably apportioning representatives,
allowing certain voters to wield more influence than others—violates this principle
because it deprives individuals of “substantially equal voting power.” Stephenson I,
355 N.C. at 379.
The majority attempts to convince us that this principle of protecting
“substantially equal voting power” is limited to the one-person, one-vote context
because the state constitution specifically contemplates this requirement in article
II, sections 3(1) and 5(1). These sections state that each state senator and each state
representative “shall represent, as nearly as may be, an equal number of inhabitants,
the number of inhabitants that each [senator or representative] represents being
determined for this purpose by dividing the population of the district that he
represents by the number of Senators apportioned to that district.” N.C. Const. art.
II, §§ 3(1), 5(1). The majority asserts that “[p]arty affiliation . . . is not mentioned in
-172-
HARPER V. HALL
Earls, J., dissenting
Article II, Sections 3 or 5.”
Interestingly, however, article II, sections 3(1) and 5(1) apply only to state
senators and members of the North Carolina House of Representatives. Neither of
these provisions nor any other constitutional provision requires that other statewide
offices represent similarly sized constituencies. Even so, in Blankenship, this Court
held that “the right to vote in superior court elections on substantially equal terms”
is protected by North Carolina’s equal protection clause. Blankenship, 363 N.C. at
526. Moreover, this Court reached this interpretation under the state equal
protection clause even though “federal courts [had] articulated that the ‘one-person,
one-vote’ standard [was] inapplicable to judicial elections.” Id. at 522. Thus, this
Court in Blankenship found that North Carolina’s equal protection clause prohibits a
certain practice that was neither mentioned in the state constitution explicitly nor
prohibited by the Federal Constitution.5
Putting the majority’s weak attempt at line drawing aside, partisan
gerrymandering is, in effect, indistinguishable from malapportionment. The only
practical difference is that, rather than diluting votes based on “where [a voter]
5 What is more, article II, sections 3(1) and 5(1)—the provisions on which the majority
relies—also textually contemplate the use of single-member and multi-member districts
within the same redistricting plans. See N.C. Const. art. II, §§ 3(1), 5(1). But as discussed in
depth, see Section II.C.3, in Stephenson I, this Court held that the use of multi-member
districts violates the state constitution’s equal protection clause “unless it is established that
inclusion of multi-member districts advances a compelling state interest.” Stephenson I, 355
N.C. at 381. Thus, Stephenson I further demonstrates that this Court has relied on the state
constitution’s equal protection clause previously in cabining a power that the state
constitution explicitly assigns to the General Assembly.
-173-
HARPER V. HALL
Earls, J., dissenting
happen[s] to reside,” see Reynolds, 377 U.S. at 563, partisan gerrymandering dilutes
votes based on whom an individual happens to vote for. Thus, as with
malapportionment, partisan gerrymandering deprives voters of “substantially equal
voting power” and violates the North Carolina constitution’s equal protection clause.
The majority’s equal protection analysis warrants one final correction. In
particular, the majority implies that the U.S. Supreme Court in Rucho concluded that
partisan gerrymandering does not implicate the federal Equal Protection Clause.
This it did not do, and the majority’s characterization is incorrect. The Supreme
Court’s decision in Rucho was limited to the question of justiciability. Rucho
specifically held that, despite the fact that “such gerrymandering is incompatible with
democratic principles . . . partisan gerrymandering claims present political questions
beyond the reach of the federal courts.” Rucho, 139 S. Ct. at 2506–07 (cleaned up).6
The majority may wish to downplay its legal extremism by analogizing its action
today to that of the nation’s highest court. But it may not accomplish this task by
plainly misstating what the U.S. Supreme Court held.
C. Partisan Gerrymandering is Justiciable
“It has long been understood that it is the duty of the courts to determine the
meaning of the requirements of our Constitution.” Leandro v. State, 346 N.C. 336,
6 In fact, the dissent in Rucho criticized the majority’s refusal to address the claims at
issue in light of the constitutional rights that were implicated by partisan gerrymandering.
See Rucho, 139 S. Ct. at 2509 (“For the first time ever, this Court refuses to remedy a
constitutional violation because it thinks the task beyond judicial capabilities.”).
-174-
HARPER V. HALL
Earls, J., dissenting
345 (1997). This duty holds true where partisan gerrymandering claims are
concerned. The majority, however, invokes the political question doctrine to conclude
that partisan gerrymanders are nonjusticiable political questions. The majority errs
in applying the doctrine to such claims. Indeed, “[t]he doctrine of which we treat is
one of ‘political questions,’ not one of ‘political cases.’ The courts cannot reject as ‘no
law suit’ a bona fide controversy as to whether some action denominated ‘political’
exceeds constitutional authority.” Baker v. Carr, 369 U.S. 186, 217 (1962). The
majority’s conclusion otherwise was wrong when it was first drawn by the dissent in
Harper I, and it is wrong today.
1. A Brief History of Partisan Gerrymandering Jurisprudence
Though the justiciability of partisan gerrymandering claims in the federal
courts has long been debated, a majority of the U.S. Supreme Court only recently
decided that such claims are nonjusticiable. In fact, for several decades, the opposite
view prevailed, and partisan gerrymandering claims were considered justiciable. See,
e.g., Vieth, 541 U.S. at 317 (Stevens, J., dissenting) (“[F]ive Members of the Court . .
. share the view that . . . it would be contrary to precedent and profoundly unwise to
foreclose all judicial review of [partisan gerrymandering] claims that might be
advanced in the future.”); Davis v. Bandemer, 478 U.S. 109, 143 (1986) (plurality
opinion) (holding that “political gerrymandering claims are properly justiciable under
the Equal Protection Clause”), abrogated by Rucho v. Common Cause, 139 S. Ct. 2484
(2019). Then, in 2019, the U.S. Supreme Court changed course. In Rucho, the Court
-175-
HARPER V. HALL
Earls, J., dissenting
held “that partisan gerrymandering claims present political questions beyond the
reach of the federal courts.” 139S. Ct. at 2506–07.
The evolution of the U.S. Supreme Court’s partisan gerrymandering
jurisprudence is not, of course, biding on this Court. Rucho itself was clear that
“[p]rovisions in state statutes and state constitutions can provide standards and
guidance for state courts to apply.” Id. at 2507. But these cases demonstrate that for
decades, U.S. Supreme Court Justices from both sides of the ideological spectrum
agreed that “severe partisan gerrymanders [are incompatible] with democratic
principles,” Vieth, 541 U.S. at 292, and further that their “legislative classifications
‘reflec[t] no policy, but simply arbitrary and capricious action[,]” id. at 316 (Kennedy,
J., concurring in the judgment); see id. at 312 (Kennedy, J., concurring in the
judgment) (recognizing that “the rapid evolution of technologies in the apportionment
field suggests yet unexplored possibilities” with respect to the standards that may
emerge to govern partisan gerrymandering claims).
Times have changed, however, and it is no secret that “ideology in Supreme
Court appointments” has become increasingly important, ushering in a new era of
political polarization on the nation’s highest court. See, e.g., Neal Devins and
Lawrence Baum, Split Definitive: How Party Polarization Turned the Supreme Court
into a Partisan Court, 2016 Sup. Ct. Rev. 301, 319–20 (2017) (explaining that “it
appears that Republican-appointed Justices are more strongly conservative than the
Court’s Democratic-appointed Justices are liberal” and highlighting that, as of 2016,
-176-
HARPER V. HALL
Earls, J., dissenting
legal scholars had “rank[ed] four Roberts Court Republican-appointed Justices as
among the most conservative Justices ever to sit on the Court”). In light of this
increased polarization, it is unsurprising that the previous understanding regarding
partisan gerrymandering’s justiciability became a position of the past by the time
Rucho was decided.
But the U.S. Supreme Court is not the only institution in the country that has
become collateral damage in increasingly partisan battles surrounding voting rights.
Indeed, the decision today demonstrates that this Court has met the same fate. Just
as Rucho followed closely on the heels of a shift in the U.S. Supreme Court’s makeup,
the Court’s decision here follows a midterm election that altered its political
composition. Notably, this Court’s decision to vacate Harper I and Harper II is not
based on a change in or misunderstanding of the controlling law or facts. Instead, the
Court, now armed with the influence of a conservative majority, has an intellectual
disagreement with Harper I’s interpretation of the law. Not only is such a
disagreement not an appropriate basis to vacate a prior decision under these
circumstances, the Court’s decision, which was designed to protect the power of
partisan legislators rather than North Carolina’s voters, stamps a seal of approval on
flagrant violations of the state constitution.
2. Judicially Manageable Standards
The majority reasons that “our constitution does not provide judicially
discernable or manageable standards for adjudicating partisan gerrymandering
-177-
HARPER V. HALL
Earls, J., dissenting
claims” as part of its conclusion that such claims are nonjusticiable political
questions. The majority’s reasoning is largely cribbed from the U.S. Supreme Court
decision in Rucho. Given the majority’s reliance on Rucho, I address the line of
reasoning that was first adopted by the U.S. Supreme Court and is now echoed by
this Court as to why political gerrymandering claims lack judicially manageable
standards. Condensed to its simplest form, the reasoning proceeds as follows.
First, the thinking goes that the Framers of the state and federal constitutions
were aware of the concept of gerrymandering, but neither constitution expressly
prohibited the practice. See Rucho, 139 S. Ct. at 2494–96; Harper I, 380 N.C. at 417
(Newby, J., dissenting). Second, based on this historical practice, some amount of
partisan gerrymandering must be constitutionally permissible, meaning that strict
proportionality is not required by the state or federal constitution. See Rucho, 139 S.
Ct. at 2499; Harper I, 380 N.C. at 417 (Newby, C.J., dissenting). Third, neither
constitution prescribes the exact amount of partisan gerrymandering that is
unconstitutional. See Rucho, 139 S. Ct. at 2501, 2506; Harper I, 380 N.C. at 421
(Newby, C.J., dissenting). This final point coupled with the notion that the “political
science tests” that have been developed to expose partisan gerrymandering are
insufficient yield the conclusion that there is no standard a trial court can reliably
apply to determine whether a partisan gerrymander is unconstitutional. This line of
reasoning can be reduced to a common refrain: “At what point does permissible
partisanship become unconstitutional,” or more simply, “[h]ow much is too much?”
-178-
HARPER V. HALL
Earls, J., dissenting
Rucho, 139 S. Ct. at 2501. This question, the majority thinks, is simply too hard to
answer.
Even if the question is too challenging for this Court’s current majority to fully
grapple with—this particular issue is addressed in more detail below—courts both in
North Carolina and around the country that have successfully confronted this
question as well similar questions in analogous contexts, demonstrating that the
manufactured conundrum is not as mystifying as the majority would have us believe.
The majority attempts to obfuscate the standard laid out in Harper I by
repeatedly asserting that Harper I simply requires a proportionality standard.
Harper I was clear that “the fact that one party commands fifty-nine percent of the
statewide vote share in a given election does not entitle the voters of that party to
have representatives of its party comprise fifty-nine percent of the North Carolina
House, North Carolina Senate, or North Carolina congressional delegation.” Harper
I, 380 N.C. at 387 (majority opinion). To clarify any confusion amongst the members
of the majority, this means that Harper I acknowledged that proportionality is not
the constitutional baseline.
Instead, Harper I explained that the state constitution provides that
voters are entitled to have substantially the same
opportunity to electing a supermajority or majority of
representatives as the voters of the opposing party would
be afforded if they comprised fifty-nine percent of the
statewide vote share in that same election. What matters
here, as in the one-person, one-vote context, is that each
voter's vote carries roughly the same weight when drawing
a redistricting plan that translates votes into seats in a
-179-
HARPER V. HALL
Earls, J., dissenting
legislative body.
Id. To crystalize the point, when the voting strength of a particular group of voters is
artificially diluted based purely on their political preferences, they are deprived of
their “fundamental right to vote on equal terms,” Stephenson I, 355 N.C. at 378,
among other constitutional rights. When such constitutional violations are alleged,
the state constitution requires an inquiry into whether maps enacted by the General
Assembly systematically prevent a political party whose candidates receive a
majority of the statewide votes from having a realistic opportunity to win at least half
of the representative seats that are up for election. That does not mean that the party
must win half of the seats. It simply means the party must not be deprived of the
opportunity to do so though maps that are intended to suppress a particular kind of
voter’s voting power.
There are various empirical and statistical analyses that demonstrate whether
unconstitutional partisan vote dilution has occurred. Relevant here, Harper I clearly
outlined “multiple reliable ways of demonstrating the existence of an
unconstitutional partisan gerrymander,” including the mean-median difference
analysis; the efficiency gap analysis; the close-votes, close seats analysis; and the
partisan symmetry analysis. Harper I, 380 N.C. at 384. Through these analyses, “the
same technologies and data that today facilitate extreme partisan gerrymanders also
enable courts to discover them, by exposing just how much they dilute votes.” Rucho,
-180-
HARPER V. HALL
Earls, J., dissenting
139 S. Ct. at 2517 (Kagan, J., dissenting). 7
“Once a plaintiff shows that a map infringes on their [constitutional rights]”
through impermissible vote dilution, the legislature may still be able to justify the
apparent anomalies by reference to constitutionally acceptable redistricting criteria,
which amount to compelling governmental interests. See Harper I, 380 N.C. at 387.
“[C]ompelling governmental interests in the redistricting context include the
traditional neutral districting criteria expressed in article II, sections 3 and 5 of the
North Carolina Constitution.” Id. at 388. Additionally, incumbency, so long as “it is
applied evenhandedly, is not perpetuating a prior unconstitutional redistricting plan,
and is consistent with the equal voting power requirements of the state constitution,”
as well as other “widely recognized traditional neutral redistricting criteria, such as
compactness of districts and respect for other political subdivisions, may also be
compelling governmental interests.”8 Id.
The majority seems to have two primary objections to the standard laid out in
7 Harper I was careful in declining to “identify an exhaustive set of metrics or precise
mathematical thresholds which conclusively demonstrate or disprove the existence of an
unconstitutional partisan gerrymander.” Harper I, 380 N.C. at 384. As explained later, this
approach exemplifies the understanding that a single case presenting an issue of first
impression for the Court would be insufficient to establish all of the circumstances in which
unconstitutional partisan gerrymandering might occur.
8 “[W]hile adherence to neutral districting criteria primarily goes to whether the map
is justified by a compelling governmental interest, the disregarding of neutral criteria such
as compactness, contiguity, and respect for political subdivisions, particularly when the effect
of the map subordinates those criteria to pursuit of partisan advantage, may also be some
evidence a map burdens the fundamental right to equal voting power.” Harper I, 380 N.C.
384 n.15.
-181-
HARPER V. HALL
Earls, J., dissenting
Harper I. First, the majority is unsatisfied because, while outlining a number of
“political science tests” whose results can evidence an unconstitutional partisan
gerrymander, Harper I and Harper II did not define a single numeric threshold at
which point a metaphoric line can be drawn and a court can conclude that a map
enacted by the General Assembly is unconstitutional because it denies certain voters
of “substantially equal voting power.” This position ignores that “the law is ‘full of
instances’ where a judge’s decision rests on ‘estimating rightly . . . some matter of
degree.’ ” Rucho, 139 S. Ct. at 2522 (Kagan, J., dissenting) (alteration in original)
(quoting Johnson v. United States, 576 U.S. 591, 604 (2015). And in these contexts,
“[t]o the extent additional guidance has developed over the years . . . , courts
themselves have been its author.” Id.Reviewing redistricting plans to determine whether certain voters have been
deprived of “substantially equal voting power” is no different. Indeed, “courts all the
time make judgments about the substantiality of harm without reducing them to
particular percentages. If courts are no longer competent to do so, they will have to
relinquish, well, substantial portions of their docket.” Id. Countless claims require a
court to determine when a harm is sufficiently substantial to constitute a
constitutional violation. We need look no further than the Sixth Amendment of the
U.S. Constitution for an example of this point.
The Sixth Amendment instructs that an “accused shall enjoy the right to a
speedy and public trial,” but what does that mean exactly? U.S. Const. amend. VI.
-182-
HARPER V. HALL
Earls, J., dissenting
The U.S. Constitution certainly does not elaborate, presenting problems that
resemble the majority’s concern about partisan gerrymandering claims. Indeed, as
this Court has explained, “it is impossible to determine precisely when the right [to a
speedy trial] has been denied; it cannot be said precisely how long a delay is too long;
[and] there is no fixed point when the accused is put to a choice of either exercising
or waiving his right to a speedy trial.” State v. McKoy, 294 N.C. 134, 140 (1978). But
the constitutional text’s omission of these details was not cause for the courts to
eventually determine that they were helpless when faced with a claim that an
individual had been denied the right to a speedy trial. I hope the majority would agree
that such a decision would have been a baseless abdication of the judicial function
that would itself defy the judiciary’s role as contemplated by the Constitution.
Instead of abandoning this duty, a “difficult and sensitive balancing” of four
factors has emerged to determine whether a violation has occurred. State v. Farmer,
376 N.C. 407, 414 (2020) (quoting Barker v. Wingo, 407 U.S. 514, 533 (1972)). This
balancing test has developed over time and still provides no precise point at which
the right has been violated. Even so, engaging in this “difficult and highly fact-specific
evaluation” is a mandatory judicial function. Id. at 411. Just as neither the Sixth
Amendment nor its corresponding four-part test define exactly “how long [of] a delay
is too long” for purposes of the right to a speedy trial, McKoy, 294 N.C. at 140, the
North Carolina constitution and the standard that was illuminated by Harper I do
not answer precisely “how much partisan gerrymandering is too much.” This was
-183-
HARPER V. HALL
Earls, J., dissenting
never thought to be a justiciability issue in the Sixth Amendment context, and it is
not a justiciability issue here.
The majority’s only attempt to distinguish this example is based on the notion
that, unlike the Sixth Amendment, “the constitution assigns the responsibility of
redistricting to the General Assembly, not to the courts.” This argument bears on the
separate issue of whether the courts have a constitutionally contemplated role in
presiding over partisan gerrymandering claims. In other words, it is a textual
commitment argument, which is a distinct issue with respect to justiciability. This
argument is not responsive to the point the Sixth Amendment example proves:
judicially manageable standards have been adopted in the face of other constitutional
questions that raise the same “how much is too much” question. The concern that the
majority raises is discussed in full in Section II.C.3. For now, it is enough to respond
that, contrary to the majority’s assertion that “Harper I and the dissent . . . seem to
imagine a future where redistricting is a court-managed process[,]” rather than
exclusively in the hands of the General Assembly, “Harper I and the dissent” imagine
only a future in which the constitutional guarantees of free elections and equal
protection of the laws are enforced—a future in which this Court does not abdicate
the judicial role for its own partisan ends.
With the majority’s irrelevant argument aside, I turn to the capacity of the
courts to interpret the constitutional mandate that voters be afforded “substantially
equal voting power.” Though this mandate is not defined purely in mathematical
-184-
HARPER V. HALL
Earls, J., dissenting
terms, the requirement is grounded in language that courts are accustomed to
interpreting. Most importantly, this Court gave the phrase meaning in the one-
person, one-vote context in Stephenson I. 355 N.C. at 380, 383 (holding that, the right
to “substantially equal voting power” as guaranteed by the state constitution’s equal
protection clause requires that, with respect to legislative apportionment, “any
deviation from the ideal population for a legislative district shall be at or within plus
or minus five percent for purposes of compliance with federal ‘one-person, one-vote’
requirements.”)
The majority attempts to distinguish this example from partisan
gerrymandering claims on the basis that the one-person, one-vote principle is
“relatively easy to administer as a matter of math.” Though lawyers and judges may
not be widely renowned for their mathematical prowess, courts cannot abdicate the
judicial function simply because a legal issue involves a detailed analysis. Both the
state and federal constitutions “forbid[ ] ‘sophisticated as well as simple-minded
modes of discrimination.’ ” Reynolds, 377 U.S. at 563 (quoting Lane v. Wilson, 307
U.S. 268, 275 (1939)). When faced with the one-person, one-vote issue in Reynolds,
the U.S. Supreme Court opined:
We are told that the matter of apportioning representation
in a state legislature is a complex and many-faceted one.
We are advised that States can rationally consider factors
other than population in apportioning legislative
representation. We are admonished not to restrict the
power of the States to impose differing views as to political
philosophy on their citizens. We are cautioned about the
dangers of entering into political thickets and
-185-
HARPER V. HALL
Earls, J., dissenting
mathematical quagmires. Our answer is this: a denial of
constitutionally protected rights demands judicial
protection; our oath and our office require no less of us.
Id. at 566. As Justices on this state’s highest court, our oath, our office, and the
North Carolina electorate demanded the same. Today, a majority of this Court
turns its back on those duties.
Similar language as that found in Harper I’s standard has been given meaning
in other contexts as well. For example, when a criminal defendant seeks to have
charges against him dismissed for insufficient evidence, a trial court ruling on the
motion “need determine only whether there is substantial evidence of each essential
element of the crime and that the defendant is the perpetrator.” State v. Call, 349
N.C. 382, 417 (1998).
In defining this standard, this Court has explained that “[s]ubstantial evidence
is that amount of relevant evidence necessary to persuade a rational juror to accept
a conclusion.” State v. Mann, 355 N.C. 294, 301 (2002). And how much evidence is
that exactly? Over time, the Court has come to recognize that it is something more
than “suspicion or conjecture as to either the commission of the offense or the identity
of the defendant as the perpetrator.” State v. Malloy, 309 N.C. 176, 179 (1983). The
standard is imprecise—reasonable minds regularly disagree about what constitutes
substantial evidence. But one would be hard-pressed to find any member of the legal
community who would insist that the judiciary identify a quantifiable amount of
evidence that meets the standard in all future cases. Such an undertaking would
-186-
HARPER V. HALL
Earls, J., dissenting
likely be impossible—criminal evidence comes in countless forms that serve different
purposes and indicate guilt to varying degrees—and profoundly unwise. Instead of
creating a definition with mathematical precision, over time, both this Court and
lower courts have clarified what constitutes “substantial evidence” in a way that
allows a court to consider the quantity and quality of evidence that might come before
it in a particular case.
That is all that was required here. Unconstitutional partisan gerrymandering
can be demonstrated or disproved through various forms of evidence, including the
tests identified in Harper I, and each allegation involves unique facts that bear on
whether a voter has been deprived of “substantially equal voting power.” That Harper
I allowed future cases to mete out the boundaries of unconstitutional partisan
gerrymandering was not an infirmity indicating that this state’s courts are incapable
of determining what constitutes unconstitutional partisan gerrymandering. Rather,
Harper I described a standard using terminology to which this Court has given
meaning before—even if not with mathematical or scientific exactitude—and
demonstrated the foresight that a single decision could not anticipate every future
scenario in which a constitutional violation has occurred.
The majority takes great issue with Harper I’s promise that “[l]ower courts can
and assuredly will work out more concrete and specific standards in the future.”
Harper I, 380 N.C. at 384 (alteration in original). Despite the majority’s complaints,
this forward-looking approach is not unique to Harper I. Though courts around the
-187-
HARPER V. HALL
Earls, J., dissenting
country regularly decide cases based on standards that lack precise numerical
thresholds, these thresholds may also develop over time. If such flexibility were not
permitted and courts were forced to announce precise constitutional thresholds in the
first instance, many important constitutional claims would have never been resolved.
The one-person, one-vote principle provides an important example.
In Baker v. Carr, 369 U.S. 186, 209 (1962), the U.S. Supreme Court held that
legislative apportionment claims under the Fourteenth Amendment of the U.S.
Constitution were justiciable but did not provide any standard for resolving them.
This decision paved the way for the one-person, one-vote principle itself, which was
developed in broad terms two years later in Reynolds v. Sims. 377 U.S. 533, 578
(1964). Reynolds held that “the Equal Protection Clause requires that a State make
an honest and good faith effort to construct districts, in both houses of its legislature,
as nearly of equal population as is practicable.” Id. at 577. But recognizing that
“[m]athematical exactness or precision is hardly a workable constitutional
requirement,” id., the Court “deem[ed] it expedient not to attempt to spell out any
precise constitutional tests[,]” id. at 578.
Instead, Reynolds allowed lower courts leeway to determine those tests,
explaining that “[l]ower courts can and assuredly will work out more concrete and
specific standards for evaluating state legislative apportionment schemes in the
context of actual litigation.” Id. As the U.S. Supreme Court predicted, the one-person,
one-vote principle took additional form in the years following Reynolds. See, e.g.,
-188-
HARPER V. HALL
Earls, J., dissenting
Brown v. Thomson, 462 U.S. 835, 842 (1983) (holding that “an apportionment plan
with a maximum population deviation under 10% falls within th[e] category” of
“minor deviations . . . from mathematical equality among state legislative districts
[that] are insufficient to make out a prima facie case of invidious discrimination under
the Fourteenth Amendment”); see also Karcher v. Daggett, 462 U.S. 725 (1983); White
v. Weiser, 412 U.S. 783 (1973); Mahan v. Howell, 410 U.S. 315 (1973); Avery v.
Midland Cnty., 390 U.S. 474 (1968). As the majority recognizes, in Stephenson I, this
Court eventually adopted the same threshold that the U.S. Supreme Court developed
over time in its one-person, one-vote cases to analyze whether multi-member districts
are constitutionally compliant. 355 N.C. at 383.
The second issue the majority appears to raise with the standard laid out in
Harper I is that it permits reliance on “political science tests” that are not found
within the text of the constitution itself. But the majority seems to misunderstand
the difference between a constitutional right and the tests that determine whether
such a right has been breached. The former is a cognizable guarantee that must be
contained in the constitution itself whereas the latter is a means by which the courts
assess whether a constitutional violation has occurred. Such tests are almost always
created and adopted by the courts and are rarely found within the constitutional text.
Among the constitutional rights and principles that Harper I determined had
been violated by the 2021 Plans were the free elections clause’s promise that “[a]ll
elections shall be free,” N.C. Const. art I, § 10; see Clark v. Meyland, 261 N.C. 140,
-189-
HARPER V. HALL
Earls, J., dissenting
143 (1964); and the guarantee that North Carolina citizens have “substantially equal
voting power,” “legislative representation,” and “representational influence,”
Stephenson I, 355 N.C. at 377, 379; see also N.C. Const. art. I, § 19. Those principles
are satisfied and the rights of North Carolinians are protected when a plan gives the
party that wins a majority of the statewide vote a substantially equal opportunity as
the opposing party to secure a majority of the open representative seats. The tests
Harper I identified as “reliable ways of demonstrating the existence of an
unconstitutional partisan gerrymander,” namely the mean-median difference
analysis; the efficiency gap analysis; the close-votes, close seats analysis; and the
partisan symmetry analysis, provide credible evidence as to whether legislative
apportionment plans violate those identified constitutional rights. 380 N.C. at 384.
Examples of courts relying on empirical, statistical, and social science analyses
to resolve constitutional issues, despite the absence of these analyses from the text of
the state and federal constitutions, are too numerous to count.9 The majority criticizes
9 See, e.g., Cooper v. Harris, 581 U.S. 285 (2017) (relying on expert statistical analysis
finding that the General Assembly predominately relied on race in drawing 2011 redistricting
plan because the plan disproportionately moved black voters into racially gerrymandered
districts even when controlling for party registration to conclude that the plan constituted an
unconstitutional racial gerrymander); Brown v. Thomson, 462 U.S. 835, 842–43 (1983)
(holding that “an apportionment plan with a maximum population deviation under 10% falls
within th[e] category” of “minor deviations . . . from mathematical equality among state
legislative districts [that] are insufficient to make out a prima facie case of invidious
discrimination under the Fourteenth Amendment,” even though the Constitution does not
reference any such threshold); Gomillion v. Lightfoot, 364 U.S. 339, 341 (1960) (relying on
statistical and social science evidence to conclude that, if the allegations at issue were
uncontradicted at trial, “the conclusion would be irresistible, tantamount for all practical
purposes to a mathematical demonstration, that the [challenged] legislation is solely
concerned with segregating white and colored voters by fencing Negro citizens out of town so
-190-
HARPER V. HALL
Earls, J., dissenting
the analyses adopted in Harper I, however, because they “are not grounded in any
constitutional guidance.” But if this state’s courts were only permitted to act when
the state (or federal) constitution provided a specific and explicit test for determining
when a constitutional violation has occurred, courts would lack the authority to hear
cases involving countless constitutional claims, meaning the courts would be
prohibited from engaging in one of their core constitutional duties.
Finally, the majority attempts to seal the point that Harper I failed to provide
a judicially manageable standard by pointing out that the Court in Harper II was
forced to strike down one of the 2022 Plans that the trial court approved during the
remedial phase because the trial court failed to properly apply Harper I’s standard.
In relying on Harper II as evidence that Harper I failed to define a judicially
manageable standard, the majority does not make the point it believes it does. In fact,
just the opposite.
First, the majority claims that, after Harper I and during the remedial phase,
the General Assembly attempted to apply the Harper I
standard in drawing the Remedial House Plan (RHP),
Remedial Senate Plan (RSP), and Remedial Congressional
Plan (RCP). The General Assembly followed the same
process in enacting each plan, yet the Special Masters
recommended, and the three-judge panel concluded, that
only the RHP and RSP met the Harper I standard.
The majority goes on to complain that, not only did the three-judge panel strike down
as to deprive them of their pre-existing municipal vote”); Brown v. Bd. of Educ., 347 U.S. 483(1954) (relying on academic studies of the psychological impact of segregation on youth as
evidence that racially segregated educational facilities violate the Equal Protection Clause).
-191-
HARPER V. HALL
Earls, J., dissenting
the RCP, the Court in Harper II struck down the RSP as well. What the majority
declines to mention, however, is the blatantly partisan result of the maps that the
General Assembly produced during the remedial phase. Since the majority has
neglected to take on that task, distorting the evidence of partisan gerrymandering
that was before both this Court and the trial court, I do so here.
First, take the RCP. One of the advisors to the Special Masters who were
appointed to assess the constitutional compliance of the remedial 2022 Plans, Dr.
Bernard Grofman, concluded in his report that the Plan “creates a distribution of
voting strength across districts that is very lopsidedly Republican.” Harper II, 383
N.C. at 101. He determined that “[b]ecause they all point in the same direction, the
political effects statistical indicators of partisan gerrymandering strongly suggest the
conclusion that this congressional map should be viewed as a pro-Republican
gerrymander.” Id. (alteration in original). Despite recognizing that “the RCP yielded
an efficiency gap of 6.37%,” he noted that that this was “not . . . proof that there is no
vote dilution” because, applying the other measures identified in Harper I,
“legislative map drawers have apparently sought to draw a congressional map that
just narrowly pass[es] a supposed threshold test for partisan gerrymandering.” Id.
(alterations in original).
Another advisor, Dr. Eric McGhee:
determined that the RCP yielded an efficiency gap of 6.4%,
a mean-median difference of 1.1%, a partisan asymmetry
of 4.9%, and a declination metric of 0.14, all favoring
Republicans. He noted that “[t]he values with incumbency
-192-
HARPER V. HALL
Earls, J., dissenting
factored in all lean more Republican . . . , and this
incumbency effect is greater than it was in the [2021]
enacted plan.” Relatively, he noted that while the RCP
shows improvement from the 2021 enacted plan on several
measures of partisan symmetry, it is “clearly worse” than
the remedial congressional plans proposed by Plaintiffs.
Id. (alterations in original).
Likewise, a third advisor, Dr. Samuel Wang, concluded that the RCP has “an
average efficiency gap of 6.8% and an average mean-median difference of 1.2%, both
favoring Republicans.” Id. In nine out of ten sample elections, he found that the RCP
would allow Republicans to win more seats than Democrats with the same vote share.
Id. “Averaging across all 10 elections, the advantage was 1.7 more seats for
Republicans, or 12% of the 14-seat Congressional delegation.” Id.
Finally, a fourth advisor, Dr. Tyler Jarvis, “determined that the RCP
‘consistently favors Republicans’ across all applicable measures. He determined that
the RCP yields an efficiency gap of 8.8%, a mean-median difference of 0.9%, a partisan
bias of 5.2%, and a declination metric of 11.6%, all favoring Republicans.” Id.
Though a less severe partisan gerrymander than the RCP, the RSP was also
largely inconsistent with Harper I’s mandate. Harper II described these findings in
depth:
Dr. Grofman determined that the RSP “creates a
distribution of voting strength across districts that is very
lopsidedly Republican.” He determined the RSP’s vote bias
indicates “a substantial pro-Republican bias” in which a
statewide majority of Republican voters would be able to
win a majority of the seats while “only a win by
considerably more than 50% of the statewide vote can yield
-193-
HARPER V. HALL
Earls, J., dissenting
the Democrats a majority of the seats.” He determined that
“[b]ecause they all point in the same direction, the political
effects statistical indicators of partisan gerrymandering
argue for the conclusion that th[e] [RSP] should be viewed
as a pro-Republican gerrymander.” He concluded that “the
dilutive effects of th[e] RSP] . . . are still . . . quite
substantial.”
Dr. McGhee determined that the RSP “still favors
Republicans when all seats are open.” He concluded that
the RSP yields an efficiency gap of 4.8%, a mean-median
difference of 2.2%, a partisan asymmetry of 4.8%, and a
declination metric of 0.20, all favoring Republicans. He
observed that “[t]he [efficiency gap] value now clearly falls
below the commonly identified threshold of 7%, though the
[mean-median difference] value falls well above the 1%
number cited by Legislative Defendants.” He determined
that “[a]ll the metric values for both the open seat and
incumbency scenarios are more than 50% likely to favor
Republicans throughout the decade.” He concluded that the
[mean-median difference] and [partisan symmetry]
metrics, which are more relevant for a state legislative plan
because they connect directly to control of the chamber,
suggest that in a tied election Republicans would still hold
27 or 28 [of 50 total] seats, and that Democrats would need
to win as much as 53 percent of the vote to claim 25 seats.
The odds are about three to one that Republicans would
maintain this advantage throughout the decade.
Relatively, Dr. McGhee observed that the Republican
advantage within Plaintiffs’ proposed RSP “is often less
than half the size of the same advantage in the Legislative
Defendants’ [RSP].” “This suggests that there is nothing
foreordained about the advantages in the Legislative
Defendants’ plan.”
Dr. Wang determined that the RSP favors Republicans in
all six metrics evaluated: seat partisan asymmetry, mean-
median difference, partisan bias, lopsided wins, declination
angle, and efficiency gap. Specifically, he determined that
the RSP yields an efficiency gap of 2.2%, a mean-median
difference of 0.8%, and an average partisan asymmetry of
-194-
HARPER V. HALL
Earls, J., dissenting
2.1 seats, all favoring Republicans.
Finally, Dr. Jarvis determined that analysis of the RSP
reveals that it “is often a significant outlier in favor of the
Republicans.” He determined that the RSP yields an
efficiency gap of 4.0%, a mean-median difference of 1.4%,
an average partisan bias of 4.0%, and a declination metric
of 7.0%.
Id. at 103–04 (alterations in original).
By contrast, the advisors to the Special Masters made the following conclusions
about the RHP:
Dr. Grofman determined that although the RHP “creates a
distribution of voting strength across districts that is very
lopsidedly Republican,” it “is genuinely far more
competitive than either of the other two legislatively
proposed maps.” He observed that under the RHP, “unlike
the other maps, the Democrats do not have to win all of the
competitive seats to win a majority in the House. Moreover,
unlike the [RCP and RSP], . . . the competitive seats [in the
RHP] are substantially Democrat in directionality.” He
further noted that: “quit[e] important in judging the
constitutionality of this map in the full context are the facts
that: (a) the Harper plaintiffs have not chosen to offer an
alternative [RHP] but are apparently content to see the
legislative map implemented by the Court, (b) the map was
passed by a clear bipartisan consensus in the legislature,
including members of the legislature who belong to
particular minority communities, and (c) that while it still
is further from being non-dilutive than the NCLCV [RHP]
alternative, it is far closer to Plaintiffs’ map than it is to
the rejected [2021] enacted NC House map.”
He determined that while the RHP's efficiency gap
“remains in a pro-Republican direction,” it is “at the low
level of 2.72[%].” In considering “the totality of the
circumstances . . . and recognizing that this map is still not
ideal (nor need it be),” he concluded that the RHP “simply
lacks the same clear indicia of egregious bias found in the
-195-
HARPER V. HALL
Earls, J., dissenting
previously rejected maps and still found . . . in the [RCP]
and [RSP].”
Dr. McGhee likewise determined that the RHP “still favors
Republicans when all seats are open, but substantially less
[than the 2021 congressional map].” He determined that
the RHP yields an efficiency gap of 3.0%, a mean-median
difference of 1.4%, a partisan asymmetry of 2.9%, and a
declination metric of 0.16, all favoring Republicans. Dr.
McGhee concluded that the RHP “still favors Republicans:
the party would likely hold about 64 of 120 seats with half
the vote, and it would take the Democrats somewhere close
to 52% of the vote to bring that number down to 60.”
Relatively, he determined that the RHP “is very similar to”
NCLCV Plaintiffs’ proposed remedial house map on
metrics of partisan symmetry, that it “do[es] a reasonably
good job of respecting traditional geographic principles,”
and that it reflects “very similar compactness” as Plaintiffs’
proposed remedial House map. He concluded that the
RHP’s partisan symmetry is “closer [to NCLCV's proposed
remedial plan] than was the case for either the [RSP] or the
[RCP],” noting that the NCLCV Plaintiffs’ plan is only “a
little better.” He concluded that this “relatively marginal
improvement hints that it may be difficult to do better
while still abiding by other constraints.”
Dr. Wang determined that the RHP favors Republicans in
all six metrics evaluated: seat partisan asymmetry, mean-
median difference, partisan bias, lopsided wins, declination
angle, and efficiency gap. Specifically, he determined that
the RHP yielded an efficiency gap of 3.1%, a mean-median
difference of 0.9%, a partisan asymmetry of 7.2 seats, and
a declination angle of 4.5 degrees.
Finally, Dr. Jarvis determined that the RHP “appear[s] to
be mostly typical in terms of the number of seats won.” He
determined that the RHP yields an efficiency gap of 2.7%,
a mean-median difference of 1.5%, an average partisan
bias of 2.7%, and a declination metric of 5.7%.
Id. at 102–03 (alterations in original).
-196-
HARPER V. HALL
Earls, J., dissenting
Two observations follow from this evidence. First, contrary to the majority’s
suggestion that Harper I simply required a proportionality standard, the Court in
Harper II approved the RHP, even though three of the four advisors to the Special
Masters determined that the RHP maintained a pro-Republican bias. Though the
majority appears to believe that there is no basis for Harper II’s decision to accept the
RHP but reject the RSP, this conclusion rests solely on the majority’s failure to
consider the totality of the evidence presented for both plans, as discussed below.
Second, as to the RCP, the General Assembly’s refusal to make a legitimate
effort in applying Harper I’s mandate is not evidence that Harper I failed to delineate
a manageable standard. The RCP was rejected by both the three-judge panel and this
Court due to the General Assembly’s own plain and intentional manipulation of the
statistical data. As the Special Masters concluded, “there is substantial evidence from
the findings of the advisors that the proposed congressional plan has an efficiency
gap above 7% and a mean-median difference of greater than 1%.” Id. at 105–106.
More specifically, “none of the Special Masters’ Advisors determined that the RCP
yielded both an efficiency gap below 7% and a mean-median difference below 1%.” Id.
at 117. But this was not all. The evidence demonstrated that the RCP “ ‘consistently
favor[ed] Republicans’ across all applicable measures.” Id. at 117.
Despite the strong evidence across metrics that the RCP represented an
unconstitutional partisan gerrymander, the majority chastises the three-judge panel
for applying this Court’s precedent and concluding that the RCP was “not
-197-
HARPER V. HALL
Earls, J., dissenting
satisfactorily within the statistical ranges set forth in [Harper I].” According to the
majority “[a] majority of advisors and experts found that all three plans fell within
the thresholds set by the Harper I majority, yet for some reason . . . only the RCP was
unconstitutional.” As an initial matter, this statement plainly misstates the advisors’
findings, which are summarized above. Further, it commits the same error that
Harper I and Harper II prohibited by relying exclusively on two of the empirical tests
in isolation, rather than analyzing the evidence in its entirety. See Harper II, 383
N.C. at 93 (explaining that in Harper I, the Court expressly declined to “identify an
exhaustive set of metrics or precise mathematical thresholds which conclusively
demonstrate or disprove the existence of an unconstitutional partisan gerrymander.”)
(quoting Harper I, 380 N.C. at 384).
Harper II was clear that “[c]onstitutional compliance has no magic number.”
383 N.C. at 114. Nor should it for the reasons already explained. Moreover, “[a]n
individual statistical measure standing alone, though helpful, is not dispositive of
constitutional compliance,” id. at 93, because “individual datapoints are vulnerable
to manipulation[,]” id. at 115. The majority proves this point. The majority concludes
that Harper I’s standard must have been applied inconsistently because the
Defendants’ RCP was rejected, even though some of the advisors’ results yielded
either an efficiency gap value or a mean-median difference value within an
acceptable—yet still pro-Republican—range, similarly to the RSP and RHP. In so
concluding, the majority conveniently forgets to acknowledge the substantial amount
-198-
HARPER V. HALL
Earls, J., dissenting
of evidence showing “a very lopsidedly Republican” gerrymander. See id. at 117. The
majority’s analysis shows exactly why Harper II explained that cherry picking
individual tests as proof of constitutional compliance is not sufficient.10
That the trial court was required to evaluate a variety of evidence to determine
whether the RCP as well as the other two maps violated the state constitution does
not demonstrate that Harper I’s standard is judicially unmanageable. The obligation
to weigh the totality of the evidence is a basic evidentiary issue. When overwhelming
and varying evidence in the record points to the same conclusion, a court simply has
a stronger foundation from which to render the correct decision. In fact, that there is
a range of evidence that must be evaluated to reach the correct result does not bear
on the constitutional standard delineated by Harper I in any respect. In the criminal
context, for example, judges and juries must evaluate many different kinds of
evidence, and in assessing guilt or innocence, all of the relevant evidence before the
finder of fact should be considered and afforded the appropriate weight. So too here.
The majority’s refusal to engage in this analysis is not a shortcoming of Harper I—
10 The majority similarly ignores the totality of the evidence demonstrating that the
RSP was an extreme partisan gerrymander. For example, the majority takes umbrage with
the fact that “[t]he Harper II majority did not say why an average Mean-Median Difference
of 1.27% weighed in favor of the RHP’s constitutionality but an average Mean-Median
Difference of 1.29% weighed against the RSP’s constitutionality.” Actually, the majority did
address this issue—several times. To repeat, a single data point such as the average mean-
median calculation among the Advisors to the Special Masters is not dispositive of a plan’s
constitutionality. Harper II, 383 N.C. 89, 123 (2022) (explaining that, with respect to the
RSP, “none of these datapoints are individually dispositive.”). As a result, Harper II’s
rejection of the RSP did not turn on the average of the mean-median values alone.
-199-
HARPER V. HALL
Earls, J., dissenting
the failure belongs to the majority alone. 11
As a final comment, a footnote buried in the majority’s dissent demonstrates
the majority’s continued attempts to mischaracterize what is at stake in this case. In
this footnote, the majority opines:
Both the RHP and RSP were used during the 2022 election
cycle. Significantly, under the RHP approved by the four-
justice majority in Harper II, Republican candidates won
59% of the House races while receiving about 58% of the
aggregate statewide vote. Under the RSP, which the
Harper II majority found unconstitutional, Republican
candidates won 60% of the Senate races while receiving
about 59% of the aggregate statewide vote. It is unclear
why this small difference of approximately one percentage
point rendered the RHP constitutional and the RSP
unconstitutional.
(Citations omitted). As an initial matter, this data appears nowhere in the record,
and it is inappropriate for an appellate court to reach to outside sources for statistical
11 This Court’s decision in Stephenson v. Bartlett, 357 N.C. 301 (2003) (Stephenson II)
further illustrates the point. In Stephenson II, a majority of this Court affirmed a trial court
ruling that districts 6, 10, 11, 14, 16, 21, 26 36 and 44 in the remedial Senate redistricting
plan drawn after the Court invalidated the General Assembly’s first plan in Stephenson I
were unconstitutional under the state constitution as interpreted in Stephenson I because
they were “not compact.” Id. at 314. This Court did not specify what metric determined a
district’s compactness for constitutional purposes even though the software programs used
at the time calculated geographic compactness in nine different ways and did not delineate
how non-compact is too non-compact. There was no objection that the compactness standard
must not be administrable because the General Assembly didn’t comply with it when drawing
remedial districts; no holding that the State Constitution cannot be interpreted to require
geographically compact districts because the word compactness does not appear in the
Constitution; no objection that the court was taking over the function of the legislature by
substituting its own notions of what might be sufficiently geographically compact. It is
impossible to reconcile the Stephenson II opinion with the majority’s decision in this case,
and its failure to apply the same principles here illustrates the majority’s intellectual
dishonesty. The only consistency is that the result of both opinions is to impose on the voters
of this state districting plans that benefit Republican legislators.
-200-
HARPER V. HALL
Earls, J., dissenting
data. More importantly, however, the majority’s representation is highly misleading.
In considering Republican House and Senate candidates’ aggregate share of the
statewide vote, the majority takes advantage of the fact that there are many districts
in which there was no Democratic candidate. Specifically, using the data cited by the
majority, 25% of House districts did not have a Democrat on the ballot, compared to
the 7.5% of districts in which there was no Republican on the ballot. In the Senate,
28% of districts lacked a Democratic candidate, whereas only a single district, which
represents 2% of Senate districts, lacked a Republican candidate. Considering only
the aggregate statewide vote is therefore misleading because it suggests that
Republicans beat more Democrats, entitling them to more seats, than is true in
reality. That the majority has no reservations about engaging in this kind of
statistical manipulation is telling.
When considering races that included only Republican and Democratic
candidates, the results paint a much different story. With respect to the State House
race, though Republicans won 59% of the seats, they only won approximately 53% of
the statewide vote, meaning Democrats won approximately 47% of the statewide vote.
Without the RHP, Republicans likely would have won a supermajority in the House,
despite that, in races in which members of both parties were actually competing, both
parties won a very close share of the statewide vote. As to the State Senate race,
Republicans won 60% of the seats—a supermajority in the Senate—by receiving only
51% of the statewide vote, compared to Democrats’ 49%. Though the RSP was used
-201-
HARPER V. HALL
Earls, J., dissenting
in the 2022 election cycle, allowing Republicans to win a supermajority of seats when
barely able to win a majority of the statewide votes, Harper II eventually struck it
down while retaining the RHP. To clarify any confusion for the majority, the “small
difference” between Republicans winning 59% of the seats with 53% of the vote in the
House versus 60% of seats in the Senate with only 51% of the statewide vote is the
Senate’s veto-proof supermajority.
3. Textual Commitment
Almost sixteen years before the U.S. Supreme Court decided Marbury v.
Madison, 5 U.S. (1 Cranch) 137 (1803), this Court explained that:
the obligation of [judges’] oaths and the duty of their office
require[s] them . . . to give their opinion on that important
and momentous subject; and . . . notwithstanding the great
reluctance they might feel against involving themselves in
a dispute with the Legislature of the State, yet no object of
concern or respect could come in competition or authorize
them to dispense with the duty they owe[ ] the public, in
consequence of the trust they were invested with under the
solemnity of their oaths.
Bayard v. Singleton, 1 N.C. (Mart.) 5–6 (1787). Since then, “[i]t has long been
understood that it is the duty of the courts to determine the meaning of the
requirements of our Constitution.” Leandro v. State, 346 N.C. 336, 345 (1997).
Though the majority is correct that the state constitution assigns the
redistricting authority to the legislature, it does not give the General Assembly
license to “dictate electoral outcomes.” Cook v. Gralike, 531 U.S. 510, 523 (2001).
Recognizing this limitation on the General Assembly’s redistricting authority, this
-202-
HARPER V. HALL
Earls, J., dissenting
Court long ago established that “within the context of state redistricting and
reapportionment disputes, it is well within the ‘power of the judiciary of a State to
require valid reapportionment or to formulate a valid redistricting plan.’ ” Stephenson
I, 355 N.C. at 362 (quoting Scott v. Germano, 381 U.S. 407, 409 (1965) (per curiam));
see also Blankenship, 363 N.C. at 522–28; State ex rel. Martin v. Preston, 325 N.C.
438 (1989).
There is no exception to this principle for redistricting cases, and for good
reason. “Indeed, the need for judicial review is at its most urgent in these cases. For
here, politicians’ incentives conflict with voters’ interests, leaving citizens without
any political remedy for their constitutional harms.” Gill v. Whitford, 138 S. Ct. 1916,
1941 (2018) (Kagan, J., concurring). But the majority lets none of this stand in its
way in carving out its own partisan gerrymandering exception. In so holding, the
majority violates the established principle that “the ‘judicial power’ under the North
Carolina Constitution is plenary, and ‘[e]xcept as expressly limited by
the constitution, the inherent power of the judicial branch of government
continues.’ ” Comm. to Elect Dan Forest v. Emps. Pol. Action Comm., 376 N.C. 558,
607 (2021) (quoting Beard v. N.C. State Bar, 320 N.C. 126, 129 (1987)). No express
limitation on the judicial power exists with respect to the General Assembly’s
redistricting authority, and judicial oversight in such cases, including partisan
gerrymandering cases, is mandatory.
The majority’s conclusion that partisan gerrymandering claims are not
-203-
HARPER V. HALL
Earls, J., dissenting
reviewable by this state’s courts largely turns on the existence of two specific
provisions in the state constitution that restrict the legislature’s redistricting
authority. In particular, the majority points to article II, sections 3 and 5 of the North
Carolina constitution. Article II, section 3 provides:
The Senators shall be elected from districts. The General
Assembly, at the first regular session convening after the
return of every decennial census of population taken by
order of Congress, shall revise the senate districts and the
apportionment of Senators among those districts, subject
to the following requirements:
(1) Each Senator shall represent, as nearly as may be, an
equal number of inhabitants, the number of inhabitants
that each Senator represents being determined for this
purpose by dividing the population of the district that
he represents by the number of Senators apportioned to
that district;
(2) Each senate district shall at all times consist of
contiguous territory;
(3) No county shall be divided in the formation of a senate
district;
(4) When established, the senate districts and the
apportionment of Senators shall remain unaltered until
the return of another decennial census of population
taken by order of Congress.
N.C. Const. art. II, § 3. Article 2, section 5 prescribes the same guidelines and
restrictions for the North Carolina House of Representatives. N.C. Const. art. II, § 5.
Together, the third limitations in both sections are known as the Whole County
Provisions (WCP). In the majority’s view, article II, sections 3 and 5 are effectively
the only limitations in the state constitution that restrict the General Assembly’s
-204-
HARPER V. HALL
Earls, J., dissenting
redistricting powers. Accordingly, the majority believes that “the role of our courts is
limited to identifying a redistricting plan that violates those express limitations.”
This reasoning, of course, ignores that Harper I identified multiple
constitutional protections that prohibit partisan gerrymandering, rendering such an
express provision redundant. That the rights and principles upon which Harper I’s
holding is based are more encompassing than those found in article II, sections 3 and
5 is of no moment. As the majority itself explains, the North Carolina Declaration of
Rights, which contains all of the rights protected by Harper I, speaks in “abstract”
terms. The majority admits that this quality is what has allowed the Declaration of
Rights to survive. To maintain this “abstractness,” the Declaration of Rights
necessarily does not explicitly define every type of conduct or act that constitutes a
constitutional violation.
Whether through narrow and explicit provisions, like article II, sections 3 and
5, or those that are broad and less indefinite, like the free elections clause, the state
constitution protects the rights that are fundamental to our state and upon which our
democracy was founded. It is the duty of the courts to interpret precisely what
conduct these provisions proscribe. This duty is not to be abandoned simply because
a constitutional provision is not sufficiently “explicit.”12 All of this aside, the
12 For this reason, the majority’s reliance on Stephenson I as an appropriate example
of judicial oversight with respect to a redistricting dispute as compared to Harper I is
unavailing. Just as the Court in Stephenson I properly reviewed and ruled unconstitutional
malapportioned maps that violated article II, sections 3 and 5, 355 N.C. at 371, Harper I
-205-
HARPER V. HALL
Earls, J., dissenting
majority’s reasoning also fails to acknowledge that the restrictions articulated in
article II, sections 3 and 5 of the North Carolina constitution were first recognized in
principle by this Court before they were ever added to the state constitution.
In People ex rel. Van Bokkelen v. Canaday, 73 N.C. 198, 220 (1875), this Court
struck down an act of the General Assembly that divided Wilmington, North Carolina
into three wards from which nine members—three members from each ward—of the
Board of Alderman would be elected. The first and second ward consisted of
approximately 400 voters, whereas the third ward had approximately 2,800 voters.
Id. at 225. The Court struck down the malapportioned map as a “plain violation of
fundamental principles, the apportionment of representation.” Id. The Court further
explained that “[o]ur government is founded on the will of the people. Their will is
expressed by the ballot.” Id. at 220.
The principle Van Bokkelen recognized, however, was not expressly contained
in the text of the North Carolina constitution—article II, sections 3 and 5 were not
added until much later—and the U.S. Supreme Court’s one-person, one-vote principle
was not recognized for almost another ninety years. Thus, Van Bokkelen recognized
that, with respect to city representatives, “representation shall be apportioned to the
popular vote as near as may be” nearly one hundred years before express
constitutional provisions requiring the same were adopted. 73 N.C. at 224. This point
properly reviewed and ruled unconstitutional maps that violated the free elections clause,
the equal protection clause, the free speech clause, and the freedom of assembly clause.
-206-
HARPER V. HALL
Earls, J., dissenting
is absent from the majority’s extensive musings about the requirement that there be
an “express” limitation on the General Assembly’s reapportionment power in order
for courts to exercise judicial review.
Finally, the majority exalts this Court’s decision in Stephenson I as an example
of the proper exercise of judicial review over a dispute arising from legislative
redistricting maps. But its reliance on Stephenson I is misplaced.
Stephenson I concerned state House of Representative and Senate maps that
divided counties throughout the state into multiple districts in violation of the WCP,
which “prohibit[ ] the General Assembly from dividing counties into separate Senate
and House districts.” 355 N.C. at 359. The defendants “contend[ed] that the
constitutional provisions mandating that counties not be divided are wholly
unenforceable because of the requirements of the Voting Rights Act.” Id. at 361. The
Court rejected this argument, holding that “the WCP remain[ ] valid and binding
upon the General Assembly during the redistricting and reapportionment process . . .
except to the extent superseded by federal law.” Id. at 372.
The majority recognizes that “[o]nce [the Court] found that the 2001 Plans
violated the still-valid WCP, [it] then crafted detailed criteria harmonizing the WCP
. . . with the [Voting Rights Act] and the federal one-person, one-vote principle.” But
the Stephenson I Court did not only “harmonize” the WCP with federal law. It also
went on to ensure that the legislative maps complied with the state constitution’s
equal protection clause. The Court specifically explained, “the WCP cannot be applied
-207-
HARPER V. HALL
Earls, J., dissenting
in isolation or in a manner that fails to comport with other requirements of the State
Constitution.” Id.at 376. The particular issue the Court was tasked with resolving at
this stage was “[p]laintiffs[‘] conten[tion] that remedial compliance with the WCP
require[d] the formation of multi-member legislative districts” in addition to single-
member districts within the same plan. Id. And so, the Court went on to evaluate
whether such a plan would comply with the requirements of North Carolina’s equal
protection clause in addition to other constraints imposed by federal law.
As part of its state equal protection analysis, the Court explained that “[i]t is
well settled in this State that ‘the right to vote on equal terms is a fundamental right.”
Id. at 378 (quoting Northampton Cnty Drainage Dist No. One v. Bailey, 326 N.C. 742,
746 (1990)). With this in mind, “[t]he classification of voters into both single-member
and multi-member district within plaintiffs’ proposed remedial plans necessarily
implicates the fundamental right to vote on equal terms,” making strict scrutiny the
appropriate standard of review. Id.The Court was faced with a problem, however, in that article II, sections 3(1)
and 5(1)
arguably contemplate multi-member districts by stating
that, for apportionment purposes, each member of the
General Assembly from such a district represents a
fraction of the voters in that district. The principle of ‘one-
person, one-vote’ is preserved because the number of voters
in each member’s fraction of the multi-member district is
the same as the number of voters in a single-member
district.
-208-
HARPER V. HALL
Earls, J., dissenting
Id. at 379. This point is worth emphasizing. Though the state constitution does not
expressly permit partisan gerrymandering, there is an express provision that permits
use of single-member and multi-member districts together.
Were we to accept the Court’s rationale today, this fact would have been the
end of the Court’s inquiry in Stephenson I: enacting maps that use single and multi-
member districts in tandem is a power that is expressly granted to the General
Assembly, and there is no express limitation on this power (as it involves a clause
other than the WCP), so the courts are unable to oversee the General Assembly’s
exercise of this authority. This, of course, is not what Stephenson I did.
Instead, Stephenson I analyzed the practical effects of the combined use of
single and multi-member districts in light “of the fundamental right of each North
Carolinian to substantially equal voting power” under the state equal protection
clause. Id. at 379. The Court concluded that such maps violate this fundamental
right. Id. at 384. As such, based on the principle that “a constitution cannot be in
violation of itself,” the Court determined that article II, sections 3(1) and 5(1) cannot,
as their text suggests, be construed as “affirmative constitutional mandates and do
not authorize use of both single-member and multi-member districts in a manner”
that violates the fundamental right to substantially equal voting power. Id. at 378–
79.
This is all that Harper I did. Where Stephenson I analyzed the General
Assembly’s apportionment powers under article II, sections 3(1) and 5(1) in light of
-209-
HARPER V. HALL
Earls, J., dissenting
the equal protection clause, Harper I analyzed the General Assembly’s redistricting
powers under article II, sections 3 and 5 and the federal Constitution in light of the
state equal protection clause, the free elections clause, the free speech clause, and the
freedom of assembly clause. The majority might disagree about whether partisan
gerrymandering actually violates any of these constitutional provisions. But as
Stephenson I demonstrates, it is simply inaccurate to characterize this issue as
committed solely to the province of the General Assembly.
In sum, the majority’s textual commitment analysis does not establish that
this state’s courts lack a constitutionally contemplated role in ensuring that the
General Assembly respects the will of the voters through constitutionally complaint
maps.
4. Policy Decisions
The majority’s final effort to establish that partisan gerrymandering claims
are nonjusticiable is based on its conclusion that such claims involve “a host of ‘policy
determination[s] of a kind clearly for nonjudicial discretion[,]’ ” quoting Baker, 369
U.S. at 217 (alteration in original). I have already addressed many of the arguments
the majority raises here, and I will not repeat why those arguments fail. A few
additional points are warranted, however.
First, the majority argues that the “political science tests”—or the empirical
analyses—that Harper I identified as means of determining whether a legislative
redistricting plan constitutes an unconstitutional partisan gerrymander are
-210-
HARPER V. HALL
Earls, J., dissenting
insufficient because they use data from past elections to predict how “voters will vote
in the future.” Such data will not provide accurate results, the majority posits,
because “individual voters may vote inconsistently at different times in their life for
a variety of reasons.”
This argument is smoke in mirrors. These tests do not simply permit courts to
“gaze into crystal balls, as the majority tries to suggest.” Rucho, 139 S. Ct. at 2519(Kagan, J., dissenting). Using these reliable analyses that courts around the country
have successfully employed, courts can make “findings about . . . gerrymanders’
effects on voters—both in the past and predictably in the future—[that are] evidence-
based, data-based, statistics-based.” Id. In other words, these tests use the same data
and analyses that the General Assembly uses in attempting to create egregious
partisan gerrymanders in the first place.13 When the General Assembly uses
13 The dissent in Rucho explained clearly why the argument raised by the majority is
not a legitimate concern, particularly in light of the constitutional rights that are at stake:
Mapmakers now have access to more granular data about party
preference and voting behavior than ever before. County-level
voting data has given way to precinct-level or city-block-level
data; and increasingly, mapmakers avail themselves of data sets
providing wide-ranging information about even individual
voters. . . . Just as important, advancements in computing
technology have enabled mapmakers to put that information to
use with unprecedented efficiency and precision. . . . While
bygone mapmakers may have drafted three or four alternative
districting plans, today’s mapmakers can generate thousands of
possibilities at the touch of a key—and then choose the one
giving their party maximum advantage (usually while still
meeting traditional districting requirements). The effect is to
make gerrymanders far more effective and durable than before,
insulating politicians against all but the most titanic shifts in
-211-
HARPER V. HALL
Earls, J., dissenting
advanced technological tools and similar analyses in drawing legislative plans, it does
not simply cross its fingers and hope that it is making a close guess about election
outcomes. It knows with near certainty what the outcomes are going to be. The same
is true when trial courts use this data to determine whether the maps as drawn by
the General Assembly have been gerrymandered on a partisan basis. In
acknowledging the purpose and capabilities of such analyses, the Court in Harper I
“refused to content [itself] with unsupported and out-of-date musings about the
unpredictability of the American voter. . . . They did not bet [North Carolina’s]
future—as today the majority does—on the idea that maps constructed with so much
expertise and care to make electoral outcomes impervious to voting would somehow
or other come apart.” Id.The majority goes on to criticize Harper I for making policy judgments about a
number of issues that, as explained previously, are nothing more than evidentiary
questions. Though I will not repeat this explanation in depth, it is necessary to clarify
what the majority is doing here. As an initial matter, determining how to
discriminate against a certain kind of voter most effectively “reflects no policy, but
simply arbitrary and capricious action.” Baker, 369 U.S. at 226. This issue aside,
rather than pointing out genuine policy disputes, the majority uses the term as a
the political tides. These are not your grandfather’s—let alone
the Framers’—gerrymanders.
Rucho, 139 S. Ct. at 2513 (Kagan, J., dissenting).
-212-
HARPER V. HALL
Earls, J., dissenting
misnomer for what are really just evidentiary judgments. A quick exercise
illuminates the point. Every time the majority uses the term “policy question” or
“policy determination,” replace it with the term “evidentiary judgments.”14 The latter
term is the accurate way to describe the different decisions that the majority explores
and that come before a court analyzing partisan gerrymandering issues. Repeatedly
declaring that these considerations are policy judgments does not make them so.
For example, contrary to the majority’s conclusion “[s]electing between past
elections, current voter registration information, or some other data as the ‘best’
source for garnering partisan election data” is not a “non-judicial policy
determination,” but an evidentiary judgment that a court must resolve in
determining which data yields the most accurate results. This is the kind of judgment
that courts must frequently make in other contexts, and the use of experts in the
particular field can help provide guidance on making the right decision. How this is
a policy question in any respect is unclear.
The majority also takes aim at the fact that a single test, such as the mean-
median difference analysis or the efficiency gap analysis, can yield different results.
14 Note that there is one particular claim in the majority’s analysis where this
comparison will not work. Specifically, the majority states that using “these political science
metrics at all requires policy determinations that are not grounded in any constitutional
guidance.” As explained in depth, this argument simply advances the incorrect notion that
the tests for proving a constitutional violation must be found within the state constitution
itself. Apparently, if a court itself prescribes a test that is sufficient to prove a constitutional
violation, this is a “policy decision.” Many members of the legal community will be surprised
to learn this.
-213-
HARPER V. HALL
Earls, J., dissenting
This is simply another way of expressing the concern addressed above because it
takes issue with the variety of data, as well as “software” and “calculation methods”
that a single analysis can utilize. But when these analyses, despite their different
methods and data, yield results that point in substantially the same direction as
consistently happened in both Harper I and Harper II, there is only greater confidence
that the results are accurate. For example, as the three-judge panel found in Harper
I with respect to the Congressional Plan, “[e]ven though [Plaintiffs’] experts employed
different methodologies, each expert found that the enacted plan is an outlier that
could only have resulted from an intentional effort to secure Republican advantage.”
Further, the trial court explained that “Legislative Defendants offered no defense of
the 2021 Congressional Plan. No expert witness opined that it was not the product of
an intentional partisan redistricting.” In this way, a variety of analyses that employ
different methods only support that the trial court’s conclusion was correct.
D. The Issues Presented Here Have Already Been Decided by this Court
Finally, the majority attempts to convince us that today’s decision—a decision
that used raw partisan power to overturn two of this Court’s precedents—is nothing
out of the ordinary. “We have never hesitated,” the majority explains “to rehear a case
when it is clear that the Court ‘overlooked or misapprehended’ the law.” What the
majority has done today is anything but ordinary. It is an extreme departure from
205 years of practice. “Indeed, data from the Supreme Court’s electronic filing system
indicate that, since January 1993, a total of 214 petitions for rehearing have been
-214-
HARPER V. HALL
Earls, J., dissenting
filed, but rehearing has been allowed in only two cases.” Harper Order at 550 (Earls,
J., dissenting).
Nothing has changed since Harper I and Harper II were decided. “The legal
issues are the same; the evidence is the same; and the controlling law is the same.
The only thing that has changed is the political composition of the Court.” Id. at 550–
51. Now emboldened by its sheer political might, it takes the extraordinary step of
overturning not just the two cases at issue here, but also a third voting rights case
that this Court decided just months ago. See Holmes v. Moore, 383 N.C. 171 (2022),
rev’d, No. 342PA19-3 (N.C. Apr. 28, 2023)
Rehearing in this case never should have been granted. The cases that the
majority cites to justify its conduct confirm this. For example, the majority cites only
two cases in which rehearing was granted in this millennium. The scarcity of such
instances speaks for itself.15
These cases need not be distinguished: That they were the only two cases that were
15
granted rehearing in the last twenty-three years proves that rehearing is granted in
exceedingly rare instances. Even so, as explained in my dissent to the Court’s order granted
rehearing:
The Court most recently granted rehearing in Jones v. City of
Durham, 361 N.C. 144 (2006). There, the Court granted
rehearing for the limited purpose of reconsidering specific
evidence in a negligence action that involved a single plaintiff,
rather than to consider abolishing a constitutional right that
belongs to millions of voters. There was no dissent to the per
curiam final opinion of the Court, indicating the absence of any
partisan divide over the issue. The other case in which the Court
permitted rehearing was Smith Chapel Baptist Church v. City of
Durham, 350 N.C. 805 (1999). That case similarly did not
involve a fundamental issue central to the structure of our
-215-
HARPER V. HALL
Earls, J., dissenting
The cases that the Court cites in which rehearing was granted over twenty
years ago offer no more support for its mischaracterization of the remedy. For
example, in Whitford v. Gaskill, 345 N.C. 762 (1997), rehearing was granted for the
sole and limited purpose of modifying the final clause of the last paragraph on the
last page of an opinion. Specifically, a party sought to have this clause changed from
stating “for entry of judgment consistent with this opinion,” to “for further
proceedings not inconsistent with this opinion.” Id. at 762. Thus, rehearing was not
granted to overturn the result of a previous case, but rather to provide more accurate
instructions to the trial court regarding the proper way to proceed in the litigation.
In Alford v. Shaw, 320 N.C. 465 (1987), the Court granted rehearing because it
originally misunderstood the pertinent legal issue. In other words, it did not
originally address the question the case presented. In Lowe v. Tarble, rehearing was
granted without explanation, but the Court did not overturn its previous decision on
rehearing, explaining that “the question [at issue] is no longer debatable; it has been
resolved against defendants.” 313 N.C. 460, 462 (1985). And in Housing, Inc. v.
Weaver, 304 N.C. 588 (1981), the Court granted rehearing for the limited purpose of
rescinding a previous order that denied a party’s petition for a writ of certiorari and
allowed the petition instead. The case had not even been argued, let alone decided
democracy and had no impact whatsoever on elections.
Harper Order at 550 n.1 (Earls, J., dissenting).
-216-
HARPER V. HALL
Earls, J., dissenting
(and affirmed by a separate case). Id.These cases “demonstrate that rehearing in this Court is used cautiously; it is
rarely permitted, and when allowed, it is limited in scope.” Harper Order at 552
(Earls, J., dissenting). By contrast, the majority has used rehearing in this case to
“upend the constitutional guarantee that voters in the State will enjoy ‘substantially
equal voting power,’ regardless of their political affiliations.” Id. “Such a change . . .
fundamentally alter[s] the political rights of every voter in North Carolina.” Id.(quoting Harper I, 380 N.C. at 376).
The Court cites only one case in which the outcome changed on rehearing after
an adjustment in the Court’s composition. See Smith Chapel Baptist Church v. City
of Durham, 350 N.C. 805 (1999). This case did not involve voting rights or
redistricting. Nevertheless, even if it were analogous, a politically motivated decision
in a single case over twenty years ago does not excuse or justify such conduct going
forward. Instead, it highlights the fact that, despite ideological differences, this Court
has historically abided by its own precedent out of “[r]espect for the institution and
the integrity of its processes.” Harper Order at 550 (Earls, J., dissenting).
III. Conclusion
Following decisions such as this, we must remember that, though the path
forward might seem long and unyielding, an injustice that is so glaring, so lawless,
and such a betrayal to the democratic values upon which our constitution is based
will not stand forever. As Harper II explained, the rights that prohibit partisan
-217-
HARPER V. HALL
Earls, J., dissenting
gerrymandering in this state “are . . . the enduring bedrock of our sacred system of
democratic governance, and may be neither subordinated nor subverted for the sake
of passing political expediency.” Harper II, 383 N.C. at 95.
I dissent from this Court’s majority opinion and its shameful manipulation of
fundamental principles of our democracy and the rule of law. I look forward to the
day when commitment to the constitutional principles of free elections and equal
protection of the laws are upheld and the abuses committed by the majority are
recognized for what they are, permanently relegating them to the annals of this
Court’s darkest moments. I have no doubt that day will come.
Justice MORGAN joins in this dissenting opinion.
-218-