Billie Johnson v. Wisconsin Elections Commission ( 2022 )

2022 WI 14SUPREME COURT OF WISCONSIN
CASE NO.: 2021AP1450-OA
COMPLETE TITLE: Billie Johnson, Eric O'Keefe, Ed Perkins and
Ronald Zahn,
Petitioners,
Black Leaders Organizing for Communities, Voces
de la Frontera, League of Women Voters of
Wisconsin, Cindy Fallona, Lauren Stephenson,
Rebecca Alwin, Congressman Glenn Grothman,
Congressman Mike Gallagher, Congressman Bryan
Steil, Congressman Tom Tiffany, Congressman
Scott Fitzgerald, Lisa Hunter, Jacob Zabel,
Jennifer Oh, John Persa, Geraldine Schertz,
Kathleen Qualheim, Gary Krenz, Sarah J.
Hamilton, Stephen Joseph Wright, Jean-Luc
Thiffeault, and Somesh Jha,
Intervenors-Petitioners,
v.
Wisconsin Elections Commission, Marge Bostelmann
in her official capacity as a member of the
Wisconsin Elections Commission, Julie Glancey in
her official capacity as a member of the
Wisconsin Elections Commission, Ann Jacobs
in her official capacity as a member of the
Wisconsin Elections Commission, Dean Knudson in
his official capacity as a member of the
Wisconsin Elections Commission, Robert Spindell,
Jr. in his official capacity as a member of the
Wisconsin Elections Commission and Mark Thomsen
in his official capacity as a member of the
Wisconsin Elections Commission,
Respondents,
The Wisconsin Legislature, Governor Tony Evers,
in his official capacity, and Janet Bewley
Senate Democratic Minority Leader, on behalf of
the Senate Democratic Caucus,
Intervenors-Respondents.
ORIGINAL ACTION
OPINION FILED: March 1, 2022
SUBMITTED ON BRIEFS:
ORAL ARGUMENT: January 19, 2022
SOURCE OF APPEAL:
COURT:
COUNTY:
JUDGE:
JUSTICES:
NOT PARTICIPATING:
ATTORNEYS:
For the petitioners, there were briefs filed by Richard M.
Esenberg, Anthony F. LoCoco, Lucas T. Vebber and Wisconsin
Institute for Law & Liberty, Milwaukee. There was oral argument
by Richard M. Esenberg.
For the intervenors-petitioners Black Leaders Organizing
for Communities, Voces de la Frontera, League of Women Voters of
Wisconsin, Cindy Fallona, Lauren Stephenson and Rebecca Alwin,
briefs, including amicus briefs, were filed by Douglas M.
Poland, Jeffrey A. Mandell, Rachel E. Snyder, Richard A. Manthe,
Carly Gerads and Stafford Rosenbaum LLP, Madison; Mel Barnes and
Law Forward, Inc., Madison; Mark P. Gaber (pro hac vice),
Christopher Lamar (pro hac vice)and Campaign Legal Center,
Washington, D.C.; Annabelle Harless (pro hac vice) and Campaign
Legal Center, Chicago. There was oral argument by Douglas M.
Poland.
For the intervenors-petitioners Congressmen Glenn Grothman,
Mike Gallagher, Bryan Steil, Tom Tiffany and Scott Fitzgerald
there were briefs, including amicus briefs, filed by Misha
Tseytlin, Kevin M. LeRoy, and Troutman Pepper Hamilton Sanders
LLP, Chicago. There was oral argument by Misha Tseytlin.
For the intervenors-petitioners Lisa Hunter, Jacob Zabel,
Jennifer Oh, John Persa, Geraldine Schertz and Kathleen
Qualheim, there were briefs, including amicus briefs filed by
Charles G. Curtis, Jr. and Perkins Coie LLP, Madison; Marc Erik
Elias (pro hac vice), Aria C. Branch (pro hac vice), Daniel C.
2
Osher (pro hac vice), Jacob D. Shelly (pro hac vice), Christina
A. Ford (pro hac vice), William K. Hancock (pro hac vice) and
Elias Law Group LLP, Washington, D.C. There was oral argument
by John Devaney (pro hac vice), Perkins Coie LLP, Washington,
D.C.
For the intervenors-petitioners Citizens Mathematicians and
Scientists Gary Krenz, Sarah J. Hamilton, Stephen Joseph Wright,
Jean-Luc Thiffeault and Somesh Jha, briefs were filed by Michael
P. May, Sarah A. Zylstra, Tanner G. Jean-Louis and Boardman &
Clark LLP, Madison, and David J. Bradford (pro hac vice) and
Jenner & Block LLP, Chicago. There was oral argument by Sam
Hirsch (pro hac vice), Jenner & Block LLP, Washington, D.C.
For the respondents Wisconsin Elections Commission, Marge
Bostelmann, Julie Glancey, Ann Jacobs, Dean Knudson, Robert
Spindell, Jr. and Mark Thomsen there were letter-briefs filed by
Steven C. Kilpatrick, assistant attorney general, Karla Z.
Keckhaver, assistant attorney general, Thomas C. Bellavia,
assistant attorney general.
For the intervenors-respondents the Wisconsin Legislature
there were briefs, including amicus briefs, filed by Kevin M.
St. John and Bell Giftos St. John LLC, Madison; Jeffrey M.
Harris (pro hac vice), Taylor A.R. Meehan (pro hac vice), James
P. McGlone and Consovoy McCarthy PLLC, Arlington, Virginia and
Adam K. Mortara and Lawfair LLC, Chicago. There was oral
argument by Taylor A.R. Meehan.
For the intervenor-respondent Governor Tony Evers there
were briefs filed by Joshua L. Kaul, attorney general, Anthony
D. Russomanno, assistant attorney general and Brian P. Keenan,
assistant attorney general. There was oral argument by Anthony
D. Russomanno.
3
For the intervenor-respondent Janet Bewley, State Senate
Democratic Minority Leader on behalf of the State Senate
Democratic Caucus there were briefs filed by Tamara B. Packard,
Aaron G. Dumas and Pines Bach LLP, Madison. There was oral
argument by Tamara B. Packard.
There was an amicus brief filed on behalf of William
Whitford, Hans Breitenmoser, Mary Lynne Donohue, Wendy Sue
Johnson and Deborah Patel by Ruth M. Greenwood (pro hac vice),
The Election Law Clinic, Harvard Law School, Cambridge, MA; with
whom on the brief were law student-practitioners Mary F. Brown,
Mark R. Haidar, Meredith A. Manda, Sarah A. Sadlier, Corey M.
Stewart, Harvard Law School and Jakob Feltham and Hawks Quindel,
S.C., Madison.
There was an amicus brief filed on behalf of Concerned
Voters of Wisconsin by Joseph S. Goode, Mark M. Leitner, John W.
Halpin and Laffey, Leitner & Goode, L.L.C., Milwaukee.
There was an amicus brief filed on behalf of Non-Party
Legal Scholars by Allison Boldt, Robert Yablon and the
University of Wisconsin Law School, Madison.
There was an amicus brief filed by Daniel R. Suhr,
Thiensville.
4
2022 WI 14NOTICE
This opinion is subject to further
editing and modification. The final
version will appear in the bound
volume of the official reports.
No. 2021AP1450-OA
STATE OF WISCONSIN : IN SUPREME COURT
Billie Johnson, Eric O'Keefe, Ed Perkins and
Ronald Zahn,
Petitioners,
Black Leaders Organizing for Communities, Voces
de la Frontera, League of Women Voters of
Wisconsin, Cindy Fallona, Lauren Stephenson,
Rebecca Alwin, Congressman Glenn Grothman,
Congressman Mike Gallagher, Congressman Bryan
Steil, Congressman Tom Tiffany, Congressman
Scott Fitzgerald, Lisa Hunter, Jacob Zabel,
Jennifer Oh, John Persa, Geraldine Schertz,
Kathleen Qualheim, Gary Krenz, Sarah J.
Hamilton, Stephen Joseph Wright, Jean-Luc
Thiffeault, and Somesh Jha, FILED
Intervenors-Petitioners, MAR 3, 2022
v. Sheila T. Reiff
Clerk of Supreme Court
Wisconsin Elections Commission, Marge
Bostelmann in her official capacity as a member
of the Wisconsin Elections Commission, Julie
Glancey in her official capacity as a member of
the Wisconsin Elections Commission, Ann Jacobs
in her official capacity as a member of the
Wisconsin Elections Commission, Dean Knudson in
his official capacity as a member of the
Wisconsin Elections Commission, Robert
Spindell, Jr. in his official capacity as a
member of the Wisconsin Elections Commission
and Mark Thomsen in his official capacity as a
member of the Wisconsin Elections Commission,
Respondents,
The Wisconsin Legislature, Governor Tony Evers,
in his official capacity, and Janet Bewley
Senate Democratic Minority Leader, on behalf of
the Senate Democratic Caucus,
Intervenors-Respondents.
HAGEDORN, J., delivered the majority opinion of the Court, in
which ANN WALSH BRADLEY, DALLET, and KAROFSKY, JJ., joined. ANN
WALSH BRADLEY, J., filed a concurring opinion, in which DALLET
and KAROFSKY, JJ., joined. ZIEGLER, C.J., filed a dissenting
opinion, in which ROGGENSACK and REBECCA GRASSL BRADLEY, JJ.,
joined. ROGGENSACK, J., filed a dissenting opinion, in which
ZIEGLER, C.J., and REBECCA GRASSL BRADLEY, J., joined. REBECCA
GRASSL BRADLEY, J., filed a dissenting opinion, in which
ZIEGLER, C.J., and ROGGENSACK, J., joined.
ORIGINAL ACTION. Relief granted.
¶1 BRIAN HAGEDORN, J. Every ten years, states are
required to redraw the boundaries for congressional and
legislative districts to account for population changes. This
means the maps enacted into law in 2011 cannot constitutionally
serve as the basis for future elections. The responsibility to
adopt new district boundaries is not ours in the first instance,
but that of the legislature and governor via the legislative
process.
¶2 Shortly after the completion of the 2020 decennial
census, a group of voters petitioned this court to declare the
2011 maps unconstitutional and remedy the malapportionment. We
granted the petition, and subsequently granted intervention to
all parties that sought it, mindful that relief from this court
2
No. 2021AP1450-OA
would be necessary only if the legislative process failed.1 We
have given the political branches a fair opportunity to carry
out their constitutional responsibilities. They have not done
so. Both this court and the United States Supreme Court have
held that this failure implicates the constitutional rights of
voters. State ex rel. Reynolds v. Zimmermann, 22 Wis. 2d 544,
562, 126 N.W.2d 551 (1964); Reynolds v. Sims, 377 U.S. 533, 566
(1964). We are therefore left with the unwelcome task of
filling the gap.
¶3 The members of this court were not of one mind
regarding how——or even whether——to approach this undertaking.
But having taken this case, we sought input from the parties on
the law that governs, as well as the process by which we should
fashion a remedy.
¶4 In an order issued on November 17, 2021, and an
opinion issued on November 30, 2021, we set out the basic
process and criteria we would use to guide our decision.
Johnson v. Wis. Elections Comm'n, 2021 WI 87, 399 Wis. 2d 623,
967 N.W.2d 469. Rather than craft our own map, we invited all
parties to this litigation to submit one proposed map for each
set of districts where new district boundaries are required:
congress, state senate, and state assembly. We said we would
choose maps that minimize changes from current law and evaluate
maps for compliance with state and federal law. Id., ¶¶38, 72.
1For a summary of this case's prior procedural history, see
Johnson v. Wis. Elections Comm'n, 2021 WI 87, ¶¶5-6, 399
Wis. 2d 623, 967 N.W.2d 469.
3
No. 2021AP1450-OA
In so concluding, we rejected an approach that involved this
court making significant policy decisions or weighing competing
policy criteria. We also rejected invitations to consider the
partisan makeup of proposed districts. By focusing on legal
requirements and using the maps currently reflected in Wisconsin
law as a reference point, we sought to minimize our involvement
in the numerous policy and political decisions inherent in map-
drawing.
¶5 Following our November 30 opinion, parties submitted
proposed maps, briefs, and expert reports. And we heard over
five hours of argument regarding which proposed maps best comply
with the parameters we established.
¶6 Although not bound by any map proposal, we approached
this task hoping to select submissions from the parties that
best satisfied the criteria we set forth. We did so both at the
suggestion of the parties and in recognition of our limitations.
While we appreciate the hard work of the parties, we must
acknowledge that each proposal makes changes that appear
unnecessary to account for population changes or to otherwise
comply with the law. But rather than modify submissions we
received, we determine that the best approach is to choose the
maps that best conform with our directives, imperfect though
they may be.
¶7 Congressional maps. We received proposed
congressional maps from four parties: the Citizen
4
No. 2021AP1450-OA
Mathematicians and Scientists,2 the Congressmen,3 Governor Tony
Evers, and the Hunter intervenors-petitioners.4 The first
question is which map most complies with our least-change
directive. With only eight districts, core retention——a measure
of voters who remain in their prior districts——is the best
metric of least change, and the map submitted by Governor Evers
easily scores highest. His map moves 5.5% of the population to
new districts, leaving 94.5% in their current districts. In raw
numbers, the Governor's proposal to move 324,415 people to new
districts is 60,041 fewer people than the next best proposal.
In addition, Governor Evers' submission complies with the
federal Constitution and all other applicable laws. We
therefore adopt Governor Evers' proposed congressional map.
¶8 State legislative maps. We received six state
legislative map proposals from: the BLOC intervenors-
2The Citizen Mathematicians and Scientists include Gary
Krenz, Sarah J. Hamilton, Stephen Joseph Wright, Jean-Luc
Thiffeault, and Somesh Jha.
3The Congressmen include Congressmen Glenn Grothman, Mike
Gallagher, Bryan Steil, Tom Tiffany, and Scott Fitzgerald.
The Wisconsin Legislature endorsed the Congressmen's
proposed congressional map, but did not advance any arguments on
the merits of this proposed map.
4The Hunter intervenors-petitioners include Lisa Hunter,
Jacob Zabel, Jennifer Oh, John Persa, Geraldine Schertz, and
Kathleen Qualheim.
5
No. 2021AP1450-OA
petitioners,5 the Citizen Mathematicians and Scientists, Governor
Evers, the Hunter intervenors-petitioners, Senator Janet Bewley,6
and the Wisconsin Legislature. The proposed senate and assembly
maps making the least changes from current law are once again
those of Governor Evers. In their senate proposals, both
Governor Evers and the Legislature move a nearly identical 7.8%
of voters to different districts (92.2% core retention), with a
slight edge to the Legislature for moving 1,958 fewer people.
However, in their assembly map proposals, Governor Evers moves
14.2% of voters to new districts, while the Legislature moves
15.8% (85.8% vs. 84.2% core retention), a difference that
affects 96,178 people. No other proposal comes close. And
beyond core retention, no other measure of least change alters
the picture. The Governor's proposed senate and assembly maps
produce less overall change than other submissions.
¶9 We also conclude that Governor Evers' proposals
satisfy the requirements of the state and federal constitutions.
Under the Wisconsin Constitution, all districts are contiguous,
sufficiently equal in population, sufficiently compact,
appropriately nested, and pay due respect to local boundaries.
Governor Evers' proposed maps also comply with the federal
constitution's population equality requirement.
5 The BLOC interventors-petitioners included the
organizations Black Leaders Organizing for Communities, Voces de
la Frontera, and League of Women Voters of Wisconsin, in
addition to Cindy Fallona, Lauren Stephenson, and Rebecca Alwin.
6 Senate Minority Leader Janet Bewley intervened as a
respondent on behalf of the Senate Democratic Caucus.
6
No. 2021AP1450-OA
¶10 Regarding the Voting Rights Act (VRA), the 2011 maps
enacted into law include six majority-Black assembly voting
districts in the Milwaukee area. Governor Evers, along with
several other parties, argues the VRA now requires a seventh
majority-Black assembly district in the Milwaukee area. As a
map-drawer, we understand that our duty is to determine whether
there are "good reasons" to believe the VRA requires a seven-
district configuration. In assessing the information presented
by the parties, we conclude there are good reasons to believe a
seventh majority-Black district is needed to satisfy the VRA.
Governor Evers' assembly map accomplishes this. For these
reasons, we adopt Governor Evers' proposed remedial state senate
and state assembly maps.
I. FRAMEWORK FOR OUR DECISION
¶11 In our prior opinion in this case, we laid out more
fully the analytical framework for our final decision. For
completeness, we briefly summarize our approach here. Before
our November 30 opinion, the parties offered a variety of
arguments regarding which factors we could or should consider in
providing remedial maps. See Johnson, 399 Wis. 2d 623, ¶7. We
concluded we would minimize judicial policymaking by starting
with the 2011 maps previously enacted into law, and change only
what is "necessary to resolve constitutional or statutory
7
No. 2021AP1450-OA
deficiencies."7 Id., ¶72. We further concluded that the
partisan makeup of districts would not play a role in our
decision. Id., ¶39. We were not unanimous in these
conclusions, but it is how we as a court decided to proceed. 8 So
we invited parties to submit maps that minimize deviations from
existing district boundaries and abide by all relevant laws.
¶12 With this framework in mind, we begin our analysis by
probing which maps make the least change from current district
boundaries. From there, we examine the relevant law to ensure
that the map producing the least change also comports with all
state and federal legal requirements.
7 The concurrence agreed with this approach and added that
if there were equally compelling arguments on least change, we
could look to traditional redistricting criteria to assist our
decision-making. Johnson, 399 Wis. 2d 623, ¶83 (Hagedorn, J.,
concurring). Our selection of remedial maps in this case is
driven solely by the relevant legal requirements and the least
change directive the majority adopted in the November 30 order——
not a balancing of traditional redistricting criteria.
8 The dissent argued that "[t]rue neutrality could be
achieved by instead adhering to the neutral factors supplied by
the state and federal constitutions, the Voting Rights Act, and
traditional redistricting criteria." Johnson, 399 Wis. 2d 623,
¶94 (Dallet, J., dissenting). Thus, the dissent proposed
conducting a more open balancing of various policy interests,
including population equality, compactness, and respect for
political subdivision boundaries. Id. It also viewed
partisanship as "one of the many factors a court must balance
when enacting remedial maps." Id., ¶110.
8
No. 2021AP1450-OA
II. CONGRESSIONAL MAP
A. Least Change
¶13 Wisconsin has eight congressional districts, so
evaluating which maps changed the least is far simpler than for
legislative maps, where modifications are necessarily more
numerous and granular. The core retention figures are therefore
especially helpful. Core retention represents the percentage of
people on average that remain in the same district they were in
previously. It is thus a spot-on indicator of least change
statewide, aggregating the many district-by-district choices a
mapmaker has to make. Core retention is, as multiple parties
contended from the beginning of this litigation, central to a
least change review.9
¶14 The parties' submissions rate as follows on core
retention:
9 Three parties asked us to adopt a least change approach,
and each made it abundantly clear that core retention is central
to that inquiry. In briefing advocating a least change approach
(before our November 30 opinion), the Legislature explained that
a least change approach is one that "maximizes core retention."
The Congressmen agreed, arguing that a "'least-change' approach
would simultaneously 'minimize voter confusion,' and maximize
'core retention' by limiting the number of people placed in
different congressional districts." The Johnson petitioners
were in full accord: "Preserving the cores of prior districts
is the foundation of 'least change' review." While core
retention is not the only relevant metric, every party
understood that our adoption of a least change approach would
place core retention at the center of the analysis.
9
No. 2021AP1450-OA
Total People Moved Average Core Retention
Governor Evers 324,415 94.5%
Congressmen 384,456 93.5%
Hunter 411,777 93.0%
MathSci10 500,785 91.5%
¶15 As these numbers reveal, the Governor's map moves the
fewest number of people into new districts. It is not a close
call. The Governor's proposal moves 60,041 fewer people than
the next closest submission, that of the Congressmen.11 The
parties do not offer any other measures of least change that
counterbalance the Governor's superior core retention.
¶16 The most significant counterargument on least change
comes from the Congressmen. They argue that the Governor's
proposal makes what they call "gratuitous changes" that are
unexplained. For example, they point to the swapping of
communities between congressional districts 4 and 1. These
changes are unnecessary, the Congressmen maintain, because
district 4 is already substantially underpopulated. In other
10 In briefing, the Citizen Mathematicians and Scientists
helpfully employed the "MathSci" moniker to refer to their maps.
11 Before oral argument, the Congressmen sought leave to
submit a second map for consideration in addition to their
initial proposal. We granted motions by two other parties to
modify their proposals, but we denied the Congressmen's motion
because our November 17 order limited parties to a single
congressional map. Granting the Congressmen's motion would have
allowed them to present two congressional maps, while everyone
else was permitted only one.
10
No. 2021AP1450-OA
words, they argue that the unstated and unexplained motives
behind these changes should doom the Governor's proposal. We
see two problems with this argument.
¶17 First, nothing in our prior orders or opinion required
an explanation of changes at any level of granularity. In fact,
the November 30 opinion did not give the parties any specific
instructions beyond our rubric for deciding the case generally.
The concurrence encouraged parties to explain "why their maps
comply with the law, and how their maps are the most consistent
with existing boundaries." Johnson, 399 Wis. 2d 623, ¶87
(Hagedorn, J., concurring). But neither that concurrence nor
any order of the court asked for an explanation for every change
or provided guidance regarding what level of specificity would
satisfy the court.12
¶18 Second, the Congressmen's argument elevates form over
substance. In their submission, the Congressmen propose
significant changes to congressional districts 3 and 7. They
explain these changes by referencing population changes in
district 2. But the districts most in need of change are
district 2 in and around Dane County (which needs to shrink),
and district 4 in Milwaukee County (which needs to grow).
Applying a least change approach, the more logical place to
adjust district boundaries to account for these population
changes would be the districts both adjacent to and in between
Moreover, rejecting every map with unexplained changes
12
would require us to exclude every proposed state legislative
map. All of them contain numerous unexplained changes.
11
No. 2021AP1450-OA
congressional districts 2 and 4——not district 3 on Wisconsin's
western border and district 7 in the north and northwest. So
while the Congressmen offer an explanation for the change, it
does not appear to be a particularly good one. Perhaps, as the
Congressmen posited, the Governor has other motives; perhaps so
do the Congressmen. But rather than weigh motives and pick and
choose which changes we approve of and which we don't, we look
to which maps actually produce the least change, not which
explained their changes the most comprehensively.
¶19 The most principled way to address least change for
congressional maps is to choose the map that, in the aggregate,
moves the fewest number of people into new districts. In this
regard, the Governor's proposed map is superior to every other
proposal. It is the map with the least change.
B. Compliance with the Law
¶20 Having concluded the Governor's proposal best complies
with our directive to minimize deviations from current district
boundaries, we next consider whether it complies with all
relevant laws. The Wisconsin Constitution contains no explicit
requirements related to congressional redistricting. And no
party develops an argument that the Wisconsin Constitution
requires something for congressional districts not already
12
No. 2021AP1450-OA
necessary under the United States Constitution.13 Further, no
one argues that any congressional submission we received runs
afoul of the VRA. The only legal question that remains concerns
population equality under the United States Constitution.
¶21 The Governor's map comes close to perfect equality.
The mathematically ideal district contains 736,714.75 persons,
and the Governor's districts have either 736,714 people, 736,715
people, or 736,716 people. Thus, the total deviation between
the most and least populated districts is two persons. Several
parties argue——mostly at oral argument——that the Governor's two-
person deviation violates the United States Constitution. This
is, at best, a strained reading of the law.
¶22 To be sure, the Supreme Court has explained that there
is "no excuse for the failure to meet the objective of equal
representation for equal numbers of people in congressional
districting other than the practical impossibility of drawing
equal districts with mathematical precision." Mahan v. Howell,
410 U.S. 315, 322 (1973). On the other hand, the Supreme Court
has been willing to accept "small differences in the population
of congressional districts" "so long as they are consistent with
constitutional norms." Karcher v. Daggett, 462 U.S. 725, 740
(1983). As the Court explained, "Any number of consistently
13 As we noted in our prior opinion, the parties previously
disputed whether the Wisconsin Constitution imposes requirements
consistent with the Equal Protection Clause of the federal
Constitution. But that issue would not have any substantive
impact on our decision, so we did not (and here do not) address
it. See Johnson, 399 Wis. 2d 623, ¶13 n.4.
13
No. 2021AP1450-OA
applied legislative policies might justify some variance,
including, for instance, making districts compact, respecting
municipal boundaries, preserving the cores of prior districts,
and avoiding contests between incumbent Representatives." Id.In Tennant v. Jefferson County Commission, the Supreme Court
upheld a 4,871-person deviation in West Virginia's congressional
districts, noting the deviation advanced the state's interests
in maximizing core retention and maintaining whole counties.
567 U.S. 758, 762, 764-65 (2012) (per curium).
¶23 Moreover, many states have adopted districts with
minor variations. According to one source cited in briefing,
following the 2010 census, 14 states implemented maps with
greater than single-person deviations: Arkansas (428), Georgia
(2), Hawaii (691), Idaho (682), Iowa (76), Kansas (15), Kentucky
(334), Louisiana (249), Mississippi (134), New Hampshire (4),
Oregon (2), Texas (32), Washington (19), and West Virginia
(4,871).14 If the law is clear that a two-person deviation (or
more) is unacceptable, then nearly a third of states with more
than one congressional district have apparently not gotten the
message. We know of no case in which a court has struck down a
map based on a two-person deviation.
¶24 In addition, this minor population deviation is
justified under Supreme Court precedent by our least change
objective. In this very proceeding, we have determined that the
https://www.ncsl.org/research/redistricting/2010-ncsl-
14
redistricting-deviation-table.aspx
14
No. 2021AP1450-OA
least change approach should guide our decision. Core retention
is central to this analysis, and as our prior discussion
reveals, the Governor's map does far better on this metric than
any other map. Selecting a map from among those submitted to us
with a maximum deviation of one person would require us to adopt
a map that does substantially worse on core retention. The
United States Supreme Court held that maximizing core retention
was an acceptable justification for a far greater deviation in
Tennant. We see no reason why that rationale would not apply
with equal force here. We conclude the two-person deviation
between the most- and least-populated districts in the
Governor's proposed map does not violate the United States
Constitution.
¶25 In sum, we adopt Governor Evers' proposed
congressional map because it best follows our directive to make
the least changes from existing congressional district
boundaries while complying with all relevant state and federal
laws.
III. STATE LEGISLATIVE MAPS
A. Least Change
¶26 Our least change inquiry for state legislative maps is
a bit more complicated. This is due in part to the sheer number
of districts involved. In addition, the Wisconsin Constitution
requires that three assembly districts be nested within each
senate district, meaning we need to analyze assembly and senate
maps jointly. Wis. Const. art. IV, § 5. Nevertheless, we again
15
No. 2021AP1450-OA
begin our least change inquiry by comparing core retention
scores for each senate and assembly map we received.
¶27 The parties' senate map submissions rate as follows on
core retention, in order from least to most change:
Total People Moved Average Core Retention
Legislature 459,061 92.2%
Governor Evers 461,019 92.2%
Senator Bewley 576,321 90.2%
BLOC 610,568 89.6%
Hunter 1,128,878 80.8%
MathSci 1,513,824 74.3%
¶28 The parties' assembly map submissions rate as follows
on core retention, again in order from least to most change:
Total People Moved Average Core Retention
Governor Evers 837,426 85.8%
Legislature 933,604 84.2%
BLOC 939,513 84.1%
Senator Bewley 984,336 83.3%
Hunter 1,586,059 73.1%
MathSci 2,299,629 61.0%
¶29 Taken together, the Governor's maps score best on core
retention. Although the Legislature's senate map moves 1,958
16
No. 2021AP1450-OA
fewer people than the Governor's senate map, that slightly
better performance is outstripped by the Governor's vastly
superior core retention in the assembly, where the Governor
moves 96,178 fewer people than the Legislature. No maps from
any other party perform nearly as well as the Governor's on core
retention.
¶30 Other metrics of least change are helpful, but only
minimally so in this case. Both the Legislature and the
Governor do comparably well minimizing the number of voters who
would have to wait six years between senate elections.15 The
Legislature's senate map has this effect on 138,753 people,
whereas the Governor's does so for 139,606 people. On
geographic core retention, the Governor's senate map moves 5.0%
of the state's geography from one district to another, versus
the Legislature's 7.1%. And the Governor's assembly map moves
11.3% of the state's geography from district to district,
against the Legislature's 16.5%. Finally, both the Governor and
the Legislature pair three incumbents——one pair of senators and
two pairs of representatives for the Governor, and three pairs
of representatives for the Legislature.16 Ultimately, none of
these considerations outweigh the Governor's superior
performance on core retention.
15 See Johnson, 399 Wis. 2d 623, ¶83 n.9 (Hagedorn, J.,
concurring); id., ¶94 n.5 (Dallet, J., dissenting).
16 Some parties argue that considering incumbency is
improper. As a standalone value, that may be true. But as an
indicator of least change from existing districts, it could
constitute a helpful data point.
17
No. 2021AP1450-OA
¶31 Two other least-change approaches offered by the
parties are worth further discussion. First, the Legislature
argues that the Governor's maps are not acceptable because they
change Milwaukee-area districts more than other submissions.
Looking to the degree of change region-by-region has merit, but
we see little benefit to its application here. Some of the
changes to the Governor's maps in the Milwaukee area are driven
by modifications arguably required by the VRA (more on this
below). This necessarily creates a cascading effect on nearby
districts. But even if the Legislature's Milwaukee-specific
complaints have merit, its conclusion does not. Although the
Legislature's proposed maps may move fewer voters in some
Milwaukee-area districts, the Governor's proposed maps move
fewer voters throughout the rest of the state, leaving 13
assembly districts outside Milwaukee entirely unchanged from
their prior configurations. The Legislature does not explain
why we should reject the Governor's map for its changes to
Milwaukee, while accepting the Legislature's proposal to change
districts even more elsewhere.
¶32 Second, the Legislature argues that we should weigh as
a measure of least change the total number of counties and
municipalities split under each proposal. We fail to see why
this is a relevant least-change metric, however. If a
municipality was split under the maps adopted in 2011, reuniting
that municipality now——laudable though it may be——would produce
more change, not less. Particularized data about how many
counties or municipalities remain unified or split may be a
18
No. 2021AP1450-OA
useful indicator of least change. But no party saw fit to
provide that data.17 What we did receive was raw counts of the
total county and municipal splits under each proposal, and that
information provides no insight into which map makes the least
change to existing district boundaries.18
¶33 Viewing various least change metrics as a whole, and
relying most heavily on the preeminent core retention metric, we
conclude the Governor's legislative maps produce the least
change from current law.
B. Compliance with the Law
¶34 Next we consider whether the Governor's legislative
maps adhere to all relevant laws, starting with the Wisconsin
Constitution. As we explained in our prior opinion, the
Wisconsin Constitution requires that districts be compact,
17The Legislature provided an accounting of county and
municipal splits in the proposed legislative maps, but no one
submitted data documenting how many of those splits were present
in the 2011 maps, or how many previously split municipalities
were unified. The Legislature highlighted a handful of new
municipal splits in the Governor's map, but those examples were
limited to Waukesha County and Dane County. Without statewide
data, these geographically-limited data points do not allow for
a meaningful comparison of each proposal's overall performance
on this metric.
18Similarly, population deviation is not an indicator of
least change. Quite the opposite. Given the malapportionment
here, maximizing population equality requires more change to
current districts, not less. That is why, recognizing the
tension between these two goals, our instructions to the parties
were to redistrict according to population while minimizing
change to existing districts.
19
No. 2021AP1450-OA
contiguous, and proportionally populated; they must respect
certain local political boundaries; and the districts must
"nest" three assembly districts within each senate district.
Johnson, 399 Wis. 2d 623, ¶¶28-38; Wis. Const. art. IV, §§ 3-5.
Our cases have long recognized these requirements operate as a
floor with space for mapmaker discretion. Zimmerman, 22
Wis. 2d at 566 ("[T]here are choices which can validly be made
within constitutional limits.").
¶35 Therefore, in analyzing compliance with the Wisconsin
Constitution, we look to whether the maps meet constitutional
standards, not whether they perform comparatively better or
worse on these metrics than other maps we received. We do not,
for example, scrutinize proposed maps to determine which are
more compact or which contain the smallest population
deviations. Our concern is simply whether districts are
sufficiently compact and sufficiently equal in population to
comply with the constitution. Proposed maps are either lawful
or they are not; no constitutional map is more constitutional
than another. For our purposes, so long as a map complies with
constitutional requirements, better performance on these metrics
becomes commendable, but not constitutionally required. In
other words, they become policy choices——maybe good ones, but
policy choices nonetheless. And we have already stated our aim
to avoid deciding between competing policies. Johnson, 399
Wis. 2d 623, ¶3.
¶36 The Governor's proposed maps fall comfortably within
the relevant constitutional requirements as laid out in our
20
No. 2021AP1450-OA
cases. The districts are contiguous and properly nested. See
Wis. Const. art. IV, §§ 4-5. And with respect to the other
requirements, the Governor's maps are consistent with historical
practice and court-sanctioned requirements for compactness,
respect for local boundaries,19 and population equality.
Regarding population equality in particular, the Governor's
population deviations——1.20% for the senate and 1.88% for the
assembly——are well under the deviations previously adopted by
the legislature and those prescribed by this court.20 See Wis.
Stat. § 4.001(1) (1971-72) (noting that under the 1972 maps "no
district deviates from the state-wide average for districts of
its type by more than one per cent" (for an absolute population
19 As explained in our prior opinion, the geographic
limitations in the Wisconsin Constitution can no longer be fully
enforced given the United States Supreme Court's directives on
population equality. Johnson, 399 Wis. 2d 623, ¶35.
20 The Legislature's expert in this case agreed, explaining
that the "conventional maximum[]" for population deviation is
"+/- 5.0%," for an absolute deviation of 10%. The Governor's
maps are far below this.
If the Wisconsin Constitution requires better performance
than this on population deviation, we have never said so. Nor
have we understood State ex rel. Attorney General v. Cunningham,
81 Wis. 440, 51 N.W. 724 (1892), and State ex rel. Lamb v.
Cunningham, 83 Wis. 90, 53 N.W. 35 (1892), to afford mapmakers
no leeway on population deviation. To the contrary, in State ex
rel. Bowman v. Dammann, we declined to strike down maps despite
our conclusion that "fairer results with respect to equality of
representation" could have been accomplished. 209 Wis. 21, 30,
243 N.W. 481 (1932). We explained that only a "wide and bold
departure" from population equality was beyond the mapmaker's
discretion. Id. Were it otherwise, every map submitted would
violate the constitution, since better performance on population
deviation is certainly possible.
21
No. 2021AP1450-OA
deviation of 2%)); State ex rel. Reynolds v. Zimmerman, 23
Wis. 2d 606, 618-25, 128 N.W.2d 16 (1964) (adopting legislative
districts after legislative impasse with substantially larger
population deviations than those proposed here). They are also
well within the population equality requirements of the Equal
Protection Clause, which are more relaxed for state legislative
districts than for congressional districts.21 Harris v. Az.
Indep. Redistricting Comm'n, 578 U.S. 253, 259 (2016) ("[W]e
have refused to require States to justify deviations of 9.9% and
8%." (citations omitted)); Wis. St. AFL-CIO v. Elections Bd.,
543 F. Supp. 630, 634 (E.D. Wis. 1982) ("We believe that a
constitutionally acceptable plan . . . should, if possible, be
kept below 2%.").
¶37 We next examine whether the Governor's proposed maps
comply with the Equal Protection Clause's limits on race-based
districting and the VRA.
¶38 Under the Equal Protection Clause, "strict scrutiny
applies when race is the predominate consideration in drawing
the district lines such that the legislature subordinates
traditional race-neutral districting principles to racial
considerations." Shaw v. Hunt, 517 U.S. 899, 907 (1996)
(cleaned up). If racial considerations predominate in a map's
configuration, the state must "prove that its race-based sorting
In the last decennial redistricting cycle, dozens of
21
states enacted legislative maps with population deviations
exceeding those in the Governor's maps——most by a wide margin.
https://www.ncsl.org/research/redistricting/2010-ncsl-
redistricting-deviation-table.aspx
22
No. 2021AP1450-OA
of voters serves a 'compelling interest' and is 'narrowly
tailored' to that end." Cooper v. Harris, 137 S. Ct. 1455, 1464
(2017) (quoting another source). The Supreme Court "has long
assumed that one compelling interest is complying with operative
provisions of the Voting Rights Act." Id.¶39 "Section 2 [of the VRA] prohibits any 'standard,
practice, or procedure' that 'results in a denial or abridgement
of the right . . . to vote on account of race.'" Id. (quoting
52 U.S.C. § 10301(a)). The Supreme Court has "construed that
ban to extend to vote dilution——brought about, most relevantly
here, by the dispersal of a group's members into districts in
which they constitute an ineffective minority of voters." Id.(cleaned up). This means the VRA, when triggered, may require
the race-conscious drawing of majority-minority districts. Id.
at 1470.
¶40 Our VRA inquiry comes in an unusual procedural
posture. Often cases under the VRA present as a challenge to
particular districts in legislatively drawn maps. But our task
is to produce districts in the first instance without the
benefit of a trial and a fully-developed factual record
regarding the performance of specific districts. Sitting in
this posture, we follow the instructions provided by the Supreme
Court in Cooper:
When a State invokes the VRA to justify race-based
districting, it must show (to meet the "narrow
tailoring" requirement) that it had "a strong basis in
evidence" for concluding that the statute required its
action. Or said otherwise, the State must establish
that it had "good reasons" to think that it would
23
No. 2021AP1450-OA
transgress the Act if it did not draw race-based
district lines. That "strong basis" (or "good
reasons") standard gives States "breathing room" to
adopt reasonable compliance measures that may prove,
in perfect hindsight, not to have been needed.
Id. at 1464 (citations omitted). Under this precedent, a
mapmaker may draw districts with racial considerations in mind
provided "a strong basis in evidence," or "good reasons,"
suggest the VRA requires the mapmaker to do so.
¶41 A typical § 2 challenge is analyzed under a two-step
framework, beginning first with the so-called Gingles22
preconditions, then proceeding to whether minority voting power
is diluted under the totality of the circumstances. See
Rodriguez v. Bexar County, 385 F.3d 853, 859 (5th Cir. 2004).
Here, the Governor argues——as do several other parties——that
seven majority-Black assembly districts are required by the
VRA.23 Applying Cooper, we analyze whether a strong basis in
evidence suggests the Gingles preconditions are satisfied, and
if so, whether there are good reasons to think minority voting
power would be diluted under the totality of the circumstances
with fewer majority-Black districts. We see our inquiry as
limited to determining whether the Governor's proposal is within
the "leeway" states have "to take race-based actions reasonably
22 Thornburg v. Gingles, 478 U.S. 30, 50-51 (1986).
No one suggests the Governor's senate map violates either
23
the Equal Protection Clause or the VRA.
24
No. 2021AP1450-OA
judged necessary under a proper interpretation of the VRA."24
Cooper, 137 S. Ct. at 1472.
¶42 Beginning with step one, we first determine whether
there are "good reasons" to think the three Gingles
preconditions are met for the Black voting age population in the
Milwaukee area. In Cooper, the Court explained the
preconditions as follows:
First, a minority group must be sufficiently large and
geographically compact to constitute a majority in
some reasonably configured legislative district.
Second, the minority group must be politically
cohesive. And third, a district's white majority must
vote sufficiently as a bloc to usually defeat the
minority's preferred candidate. . . . If a State has
good reason to think that all the Gingles
preconditions are met, then so too it has good reason
to believe that § 2 requires drawing a majority-
minority district. But if not, then not.
Id. at 1470 (cleaned up).
¶43 First, it is undisputed that the Black voting age
population in the Milwaukee area is "sufficiently large and
geographically compact" to form a majority in seven "reasonably
configured legislative district[s]."25 Id. (quoting another
24 To be clear, this case does not involve a claim under the
Equal Protection Clause or VRA. Rather, as remedial map-
drawers, we strive to act in compliance with the Constitution
and applicable federal laws necessarily relying on the more
limited record before us. A standard VRA claim is brought after
the adoption of new districts. Such a claim would proceed much
differently, requiring a fully developed factual record and
detailed findings regarding the performance of specific
districts.
25 Several parties, including the Governor, calculate Black
voting age population by including "multi-race subcategories" in
addition to "non-Hispanic Black" and "non-Hispanic (Black +
25
No. 2021AP1450-OA
source). Six such districts were created by the 2011 maps, and
the parties' submissions demonstrate that it is now possible to
draw a seventh sufficiently large and compact majority-Black
district.
¶44 Second, it is also undisputed that Black voters in the
Milwaukee area are politically cohesive. Experts from multiple
parties analyzed voting trends and concluded political cohesion
existed; no party disagreed.
¶45 Finally, turning to the third Gingles precondition,
the parties offered a strong evidentiary basis to believe white
voters in the Milwaukee area vote "sufficiently as a bloc to
usually defeat the minority's preferred candidate." Id.
(quotation marks omitted). Experts from multiple parties argued
this requirement was satisfied by looking at various election
contests, with the most comprehensive expert analysis
calculating that white voters in the Milwaukee area defeat the
preferred candidate of Black voters 57.14% of the time when
relevant elections are analyzed.26 We received little in the way
White)" categories. The Legislature excludes "multi-race
subcategories" from its calculations but raises no objection to
the inclusion of those categories. See Georgia v. Ashcroft, 539
U.S. 461, 473 n.1 (2003) ("[W]e believe it is proper to look at
all individuals who identify themselves as black."), superseded
by statute on other grounds, Ala. Legis. Black Caucus v.
Alabama, 575 U.S. 254, 276-77 (2015).
BLOC's expert "analyzed eight elections between Black and
26
white candidates in nonpartisan or Democratic primaries and
Spring generals in jurisdictions that cover either Milwaukee
County, Milwaukee City, or both." In a subsequent report, the
expert explained that he omitted the 2018 lieutenant governor
primary from his analysis because "it [did] not simulate an
26
No. 2021AP1450-OA
of alternative data or analysis to counter this. To the
contrary, throughout briefing, all parties appeared to assume
the VRA requires at least some majority-Black districts in the
Milwaukee area. This can only be true if racially polarized
voting that usually defeats the minority's preferred candidate
exists. It was not until oral argument that anyone meaningfully
contended the third Gingles precondition was not met. To the
extent it was suggested in the substantial briefing we received,
it was virtually unsupported by expert analysis or argument.27
It is telling that no party saw fit to develop an argument
supported with data suggesting the VRA preconditions are not
satisfied with respect to the Black voting age population in and
around Milwaukee. We further observe that the federal court
drawing maps in 1992 assumed racially polarized voting in
Milwaukee and drew majority-Black districts to comply with the
VRA. Prosser v. Elections Bd., 793 F. Supp. 859, 868-71 (W.D.
Wis. 1992). No court has concluded otherwise since then. Based
on the data we were provided, historical practice, and the
election in which white bloc voting might defeat the choice of
Black voters." The Legislature's expert critiqued the omission,
and noted that supplementing BLOC's election data with it could
alter the analysis. The Legislature's expert did not argue that
any other additional elections besides the 2018 lieutenant
governor primary should have been included in BLOC's analysis.
Before oral argument, the strongest suggestion that the
27
Gingles preconditions might not be satisfied was a comment in
one of the Legislature's expert reports suggesting "serious
doubts about whether the Gingles threshold standard is currently
met in Milwaukee County." But an alternative analysis was not
conducted, nor did the Legislature's briefing advance or develop
this in any meaningful way.
27
No. 2021AP1450-OA
absence of any sufficiently developed counterargument, we
conclude there are good reasons to think all three Gingles
preconditions are satisfied.
¶46 Moving to the second step, § 2 of the VRA requires
consideration of the totality of the circumstances to determine
whether members of a racial group "have less opportunity than
other members of the electorate to participate in the political
process and to elect representatives of their choice." 52
U.S.C. § 10301(b). The Supreme Court has pointed to various
factors that might be relevant to this determination, including
those listed in a Senate Report from the 1982 amendments to the
VRA, and most pertinently here, "whether the number of districts
in which the minority group forms an effective majority is
roughly proportional to its share of the population in the
relevant area."28 League of United Latin Am. Citizens v. Perry,
28 The Senate Report factors include:
the history of voting-related discrimination in the
State or political subdivision; the extent to which
voting in the elections of the State or political
subdivision is racially polarized; the extent to which
the State or political subdivision has used voting
practices or procedures that tend to enhance the
opportunity for discrimination against the minority
group . . .; the extent to which minority group
members bear the effects of past discrimination in
areas such as education, employment, and health, which
hinder their ability to participate effectively in the
political process; the use of overt or subtle racial
appeals in political campaigns; and the extent to
which members of the minority group have been elected
to public office in the jurisdiction. The Report
notes also that evidence demonstrating that elected
officials are unresponsive to the particularized needs
of the members of the minority group and that the
28
No. 2021AP1450-OA
548 U.S. 399, 426 (2006). In Johnson v. De Grandy, the Court
explained that proportionality is highly relevant, but not the
exclusive measure of minority voting strength. 512 U.S. 997,
1020-21 (1994). The Court added that § 2 does not require a
mapmaker to maximize minority representation. Id. at 1017. In
all of this, we keep in mind that "States retain broad
discretion in drawing districts to comply with the mandate of
§ 2." Shaw, 517 U.S. at 917 n.9.
¶47 Here, we cannot say for certain on this record that
seven majority-Black assembly districts are required by the VRA.
But based on our assessment of the totality of the circumstances
and given the discretion afforded states implementing the Act,
we conclude the Governor's configuration is permissible.
¶48 The 2011 maps enacted into law created six majority-
Black districts in the Milwaukee area. Over the last decade,
policy underlying the State's or the political
subdivision's use of the contested practice or
structure is tenuous may have probative value.
League of United Latin Am. Citizens v. Perry, 548 U.S. 399, 426
(2006) (quoting Gingles, 478 U.S. at 44-45).
Like other courts in this posture, we find these factors
less helpful in the context of this case. In Prosser, for
example, the federal court that provided new maps for Wisconsin
in 1992 did not even mention the Senate Report factors, focusing
instead other relevant considerations. See Prosser, 793
F. Supp. at 869-71. Similarly, when the U.S. Supreme Court has
faced VRA challenges regarding the number of majority-minority
districts drawn, it has focused much of its attention on
considerations not mentioned in the Senate Report, such as
proportionality. See Johnson v. De Grandy, 512 U.S. 997, 1017-
21 (1994); Perry, 548 U.S. at 436-42.
29
No. 2021AP1450-OA
the Black population in Wisconsin grew by 4.8% statewide, while
the white population fell by 3.4%. Based on the current census,
the Black voting age population statewide is between 6.1% and
6.5%, although the precise number is subject to some dispute.
Proportionality would therefore suggest somewhere between six
and seven majority-Black assembly districts are appropriate.
Looking a bit deeper, a significant proportion of Wisconsin's
Black population lives in Milwaukee County where the subject
districts are principally located. And there, the Black voting
age population increased 5.5%, while the white voting age
population decreased 9.5%. The baseline of six districts ten
years ago, combined with population trends since then and
statewide population numbers now, suggest a seventh majority-
Black district may be required.
¶49 In addition, we have some concern that a six-district
configuration could prove problematic under the VRA. The
Legislature, for example, submitted a configuration with five
majority-Black districts, and a sixth just under a majority.
One of its proposed districts has a Black voting age population
of 73.28%, a level some courts have found to be unlawful
"packing" under the VRA. Ketchum v. Byrne, 740 F.2d 1398, 1418
(7th Cir. 1984). Packing occurs when a mapmaker draws district
lines that pack minority voters "into one or a small number of
districts to minimize their influence in the districts next
door." De Grandy, 512 U.S. at 1007. The risk of packing Black
voters under a six-district configuration further suggests
30
No. 2021AP1450-OA
drawing seven majority-Black districts is appropriate to avoid
minority vote dilution.
¶50 Viewing the totality of the circumstances, we see good
reasons to conclude a seventh majority-Black assembly district
may be required. To be clear, the VRA does not require drawing
maps to maximize the number of majority-minority districts, and
we do not seek to do so here. See De Grandy, 512 U.S. at 1016-
17. Rather, on this record, we conclude selecting a map with
seven districts is within the leeway states have to take
"actions reasonably judged necessary" to prevent vote dilution
under the VRA. Cooper, 137 S. Ct. at 1472.
¶51 Based on the foregoing, we conclude the Governor's
legislative maps comply with all relevant legal requirements.
Because they are also the maps that produce the least change
from the previously enacted maps, we adopt them.
IV. CONCLUSION
¶52 To remedy the unconstitutional malapportionment of the
2011 congressional and state legislative maps, we adopt the
Governor's proposed congressional and state legislative maps.
Beginning with the August 2022 primary elections, the Wisconsin
Elections Commission is enjoined from conducting elections under
the 2011 maps and is ordered to implement the congressional and
legislative maps submitted by Governor Evers for all upcoming
elections. This order shall remain in effect until new maps are
enacted into law or a court otherwise directs.
By the Court.——Relief granted.
31
No. 2021AP1450-OA.awb
¶53 ANN WALSH BRADLEY, J. (concurring). I join the
majority opinion, which selects the Governor's congressional and
state legislative maps, not because I approve of the "least
change" approach. I do not.
¶54 Having previously voiced my dissent to the adoption of
that approach, a majority of the court in a prior order
nevertheless embraced "least change" as the framework that would
govern the proceedings in this case. Circumscribed by that
decision and the parties' reliance upon it when crafting their
submissions, I join today's majority opinion because the
Governor's maps adhere most closely to the court's earlier
directive. Accordingly, I respectfully concur.
I
¶55 This case came to us as an original action petition
filed before the legislature and Governor had even acted on any
redistricting legislation. I joined the dissent from the order
granting the petition due to the myriad "reasons for preferring
a federal forum" and because this court had "no experience in
drawing district maps." Johnson v. Wis. Elections Comm'n, No.
2021AP1450-OA, unpublished order, at 16, 18 (Wis. Sept. 22,
2021, amended Sept. 24) (Dallet, J., dissenting).
¶56 The court then solicited briefing from the parties on
several topics, ranging from procedure to substance to timing.
Specifically, the court sought the parties' input on how it
should conduct these proceedings, what criteria it should
consider, and when final maps should be in place.
1
No. 2021AP1450-OA.awb
¶57 After redistricting legislation was passed by the
legislature and vetoed by the Governor, thus failing the
political process, a majority of the court advised that it would
apply the "least change" approach to reapportion Wisconsin's
congressional and state legislative districts in light of the
2020 census. That is, the existing maps would serve as a
template and this court would implement "only those remedies
necessary to resolve constitutional or statutory deficiencies."
Johnson v. Wis. Elections Comm'n, 2021 WI 87, ¶72, 399 Wis. 2d
623, 967 N.W.2d 469; see also id., ¶85 (Hagedorn, J.,
concurring). I again joined the dissent from this decision
because it had "potentially devastating consequences for
representative government in Wisconsin." Id., ¶88 (Dallet, J.,
dissenting). We then received initial map submissions followed
by additional rounds of briefing, culminating in over five hours
of oral argument.
II
¶58 The shortcomings of "least change" were on display
throughout these proceedings. For example, "least change," as
set forth in the court's prior order, is unmoored from any legal
requirement for redistricting. The parties struggled with
reconciling it with the United States Constitution, Wisconsin
Constitution, and Voting Rights Act.
¶59 Further, beyond core retention, it was unclear if some
metrics would carry more weight than others. Throughout
briefing and oral argument, the "least change" approach did not
and could not offer an explanation for the tradeoffs and
2
No. 2021AP1450-OA.awb
discretionary decisions that are intrinsic to map-drawing. If
this process has shown us anything, it is that the court should
depart from the "least change" approach if and when
redistricting arrives before it in the decades to come.
¶60 Although some advance that "least change" is an
apolitical approach, this court recognized that redistricting is
"inherently political" when it previously (and wisely) refrained
from jumping into the fray. Jensen v. Wis. Elections Bd., 2002
WI 13, ¶10, 249 Wis. 2d 706, 639 N.W.2d 537. It dictates where
candidates can run for office and for whom voters can cast their
vote. The process affords the chance to "restore the core
principle of republican government, namely, that voters should
choose their representatives, not the other way around." Ariz.
State Legislature v. Ariz. Indep. Redistricting Comm'n, 576 U.S.
787, 824 (2015) (internal citation omitted).
¶61 The people of Wisconsin deserve both a fair process
and fair maps. We have cautioned that "[j]udges should not
select a plan that seeks partisan advantage." Jensen, 249 Wis.
2d 706, ¶12 (quoting Prosser v. Elections Bd., 793 F. Supp. 859,
867 (W.D. Wis. 1992)). Here, the "least change" approach
necessarily enshrines the partisan advantage adopted by the
political branches ten years ago. Its application undermines,
rather than fulfills, the promise of a truly representative
government.
¶62 That being said, I am bound by the court's earlier
determination in this case. Although I disapprove of the "least
3
No. 2021AP1450-OA.awb
change" approach, I am limited by that prior determination and
obligated to apply it here.
¶63 Indeed, a majority of the court previously placed
limitations on the parties' submissions by setting forth general
criteria to be employed. The parties relied on those
limitations when preparing their maps and arguments. Because
they were directed to use a "least change" approach, the parties
did not sufficiently argue any other standard for distinguishing
between the submitted maps. Furthermore, the submitted maps may
have been far different had the parties known this court would
entertain criteria other than "least change" as preeminent.
Thus, as the majority opinion well explains, the Governor's maps
adhere most closely to the court's prior order.
¶64 I therefore join the majority opinion in its entirety
and respectfully concur.
¶65 I am authorized to state that Justices REBECCA FRANK
DALLET and JILL J. KAROFSKY join this concurrence.
4
No. 2021AP1450-OA.akz
¶66 ANNETTE KINGSLAND ZIEGLER, C.J. (dissenting). The
majority opinion demonstrates a complete lack of regard for the
Wisconsin Constitution and the Equal Protection Clause. Short
on legal analysis and long on ipse dixit, the majority opinion
amounts to nothing more than an imposition of judicial will.
The majority deems the language of the Wisconsin and United
States Constitutions to be mere policy. I dissent because here,
the majority's decision to select Governor Tony Evers' maps is
an exercise of judicial activism, untethered to evidence,
precedent, the Wisconsin Constitution, and basic principles of
equal protection. Even those in the majority recognize that
that there exists a "struggle[]" to reconcile the least change
approach they adopt with the United States Constitution,
Wisconsin Constitution, and the Voting Rights Act ("VRA").1
Concurrence, ¶58.
1 Three of the four justices in the majority would have
preferred the federal courts to have drawn the maps for
Wisconsin. See Johnson v. Wis. Elections Comm'n, No.
2021AP1450-OA, unpublished order (granting petition for leave to
commence original action), at 15-18 (Wis. Sep. 22, 2021)
(Dallet, J., dissenting) (explaining the advantages of federal
court litigation and concluding that the court should not have
accepted this original action). They clearly disagree with the
least change approach, and the concurrence is far from a
wholesale endorsement of the analysis in the majority opinion,
which adopts its own version of least change. See concurrence,
¶¶53-64. Those three justices assert there was a "struggle[]"
the parties were forced to confront when attempting to reconcile
least change with the United States Constitution, the Wisconsin
Constitution, and the VRA. Id., ¶58. Yet the majority opinion
neither recognizes nor resolves any "struggle[]" that exists
between its version of least change and the law. This calls
into question whether the majority opinion is really a lead
opinion with only Justice Hagedorn fully adopting the reasoning
therein. Id.
1
No. 2021AP1450-OA.akz
¶67 Lacking in substantive legal analysis, the majority is
imbued with personal preference. The majority disrespects the
VRA and instead cabins voters for purportedly "good reasons" in
districts based solely on race, which is nothing short of a
violation of the Equal Protection Clause. But to the majority,
the Equal Protection Clause is a mere box to check, a speedbump
on the path to dividing Wisconsin into racial categories. Not
one case cited by the majority supports its race-based
determination.2 Moreover, the majority implements a previously
unknown, judicial test: "core retention." Because the
majority's adoption of the Governor's maps is unconstitutional,
and conflicts with the record and well-established
jurisprudence, I must dissent.
¶68 For the reasons explained below, I conclude that the
court should have adopted the maps submitted by the Wisconsin
Legislature ("the Legislature") and Congressmen Glenn Grothman,
Mike Gallagher, Bryan Steil, Tom Tiffany, and Scott Fitzgerald
("the Congressmen"), or in the alternative, the maps submitted
by the Citizen Mathematicians and Scientists ("CMS"). The court
could have also drawn its own maps or directed the parties to
submit new maps that had record support and complied with the
law. The maps submitted by the Governor are unconstitutional
and fatally flawed.
See Cooper v. Harris, 581 U.S. ___, 137 S. Ct. 14552
(2017); Shaw v. Hunt, 517 U.S. 899 (1996); League of United
Latin Am. Citizens v. Perry, 548 U.S. 399 (2006) ("LULAC");
Johnson v. De Grandy, 512 U.S. 997 (1994). VRA caselaw,
including these precedents, are discussed in greater detail in
Section II.A, infra.
2
No. 2021AP1450-OA.akz
I. SUMMARY
A. No Support For Drawing Districts On The Basis Of Race.
¶69 Because the Governor has not demonstrated a VRA
violation, there can be no race-based remedy, let alone one
constructing a new district and changing six others in Milwaukee
to include exactly 51% black populations. It is undisputed that
the Legislature's maps and the maps submitted by CMS are the
only race-neutral maps submitted. Either performs better than
the Governor's maps under the constitution and the law.
Alternatively, we could design or draw our own maps, or combine
positive characteristics of several maps. Further, we could
have requested additional briefing to direct the parties, or the
Legislature or Governor specifically, to improve their maps and
provide greater record justification for their decisions. We
now are the map drawers, we are the government actors, and we
are the ones that must satisfy strict scrutiny by using racial
classifications. It is our duty to be responsible to the law.
¶70 The majority adopts the Governor's maps, which
unambiguously divided districts in the Milwaukee area on the
basis of race alone. The only valid justification for doing
this is if a VRA violation were shown, requiring a race-based
remedy. Completely absent, however, is any demonstration of a
VRA violation. Without a violation, there can be no remedy
because to take race-based action would violate the Equal
Protection Clause. In other words, a VRA remedy is
constitutionally permissible only as required to remedy a VRA
violation. Stated even differently, specific evidence must
3
No. 2021AP1450-OA.akz
demonstrate that white voters block a minority group's vote, and
due to a variety of local conditions the minority group does not
have the opportunity to effectively participate in democratic
elections, inside a district or area where a minority could be
made into an effective electoral majority. District-specific
evidence must demonstrate that the majority-minority group is
unable to elect the candidate of its choice in a specific
district. We have exactly zero evidence of any such thing
happening in these districts in Milwaukee. There is zero
evidence on the conditions and environment of local communities
warranting a race-based remedy. Yet, the majority incorrectly
surmises that there is "good reason" to nonetheless invent this
remedy.
¶71 The parties were free to engage in discovery, depose
experts, and gather the requisite information to advocate for
their positions. The Governor completely failed to evidence any
factual support for his race-based designs. The only party that
even attempted to provide the evidence sufficient to justify a
race-based remedy, the Black Leaders Organizing for Communities
("BLOC"), agrees that when examining the existing record, the
Governor's maps do not comply with the VRA, and are thus
unconstitutional.
¶72 Nonetheless, the majority places its imprimatur on the
Governor's maps, which carve seven Assembly districts with
populations that are curiously at almost exactly 51% African-
American populations. His maps reduce, not increase, the
minority percentage in most majority-minority districts. His
4
No. 2021AP1450-OA.akz
maps add what was referred to in VRA parlance as "white filler,"3
to these districts. The majority cites no support for its VRA
remedy that adds white voters and reduces black voter
percentage.
¶73 The majority fails to follow VRA jurisprudence and
instead the majority invents a new, heretofore unknown standard,
evolved from its own creation of the law and relying heavily on
alleged party concessions, not evidence. So says the majority,
if there are "good reasons" to create race-based districts, the
court is endowed with the authority to do as it wishes,
regardless of the complete lack of evidence to support any VRA
violation. Tellingly, the majority engages in no substantive
strict scrutiny analysis of the racial assignment of Milwaukee
voters, even though such scrutiny is required as a part of the
legal analysis.
B. Least Change Is Not Core Retention.
¶74 In our November 30, 2021 opinion in this case, we
concluded that our "judicial remedy should reflect the least
change necessary for the maps to comport with relevant legal
requirements." Johnson v. Wis. Elections Comm'n, 2021 WI 87,
¶¶24-63, 72, 399 Wis. 2d 623, 967 N.W.2d 469. Nowhere in that
opinion did we use the phrase "core retention". Not only were
the parties not advised that core retention would be the
decisive factor in the court's decision, but the parties were
explicitly "invited" by the concurrence to consider factors
3Counsel from CMS at oral argument explained how map
drawers construct majority-minority districts when considering
race.
5
No. 2021AP1450-OA.akz
wholly unrelated to least change.4 Johnson, 399 Wis. 2d 623,
¶¶83, 87 (Hagedorn, J., concurring) (noting that "traditional
redistricting criteria" would assist in the selection of maps).
The concurrence, which received no votes in support, was
perfectly free to include core retention in its analysis. It
did not, and for a very simple reason: no one, neither among
the parties nor the court, understood core retention was the
sole factor for determining least change and further, for
selecting maps. The core retention analysis in the majority is
an invention, made after-the-fact to justify a policy
preference.
¶75 The law instructs us to consider more than one number:
population deviation and local government divisions,
fundamentally underlie the validity of any core retention
number. Even so, the Governor's core retention numbers are
worse than the Legislature's in the Wisconsin Senate. While the
Governor's maps move fewer individuals overall, those same maps
have inordinately high population deviations among districts,
far greater than the deviations in the Legislature's maps. The
Governor's maps also divide an extraordinary number of local
communities, orders of magnitude more than the Legislature's
maps. We are constitutionally required to minimize population
deviations and local government splits. Given this significant
constitutional interest, we should adopt either the
Sitting as a court of seven, the concurrence had no
4
authority to alone direct the court's business. For further
explanation on the November 30 concurrence, see footnote 19,
infra.
6
No. 2021AP1450-OA.akz
Legislature's or CMS's maps, which score the best out of all the
submitted maps, or the court should create a map out of the best
of each.
¶76 We were tasked with selecting legislative and
congressional maps that best conform with the law while also
making as little change as possible to existing district lines.
We accepted another round of briefing and expert reports, and we
held over five hours of oral argument. Despite this extensive
opportunity to prepare, Governor Tony Evers presented maps that
had marked population deviation and divided dozens and dozens of
local municipalities.
C. The Governor's Congressional Maps Are Unconstitutional.
¶77 Knowing that the Legislature and the Congressmen
intended to submit legislative and congressional maps that were
already passed by the Wisconsin Legislature in 2021, the
Governor simply designed maps that met his own partisan ends,
which appear to be based solely on core retention. In so doing,
the Governor substantially increased population deviation and
local government splits and engaged in an unsubstantiated racial
gerrymander. In other words, the Governor inflated the core
retention number at the expense of the Wisconsin public.
Inexplicably, the majority now adopts the Governor's maps in
full, resting entirely on "core retention" as determinative.
¶78 The court refused to allow the Congressmen to submit
amended maps, conflicting with our duty to consider all
available information and the fact that other parties, including
the Governor, were permitted to amend their maps. Nonetheless,
7
No. 2021AP1450-OA.akz
the Governor has a greater population deviation, and under well-
established constitutional law, there is no de minimus deviation
for congressional districts. The Governor explained that his
deviation was caused by his lack of understanding that a lower
deviation was required. But carelessness is not a valid
justification for excessive deviation. The Governor's (and now
Wisconsin's) congressional maps are unconstitutional. The court
should have adopted the Congressmen's map, or in the alternative
CMS's map, which includes the lowest deviation available, and
are both least change.
II. STATE LEGISLATIVE MAPS
¶79 In our November 30 opinion, we indicated that any map
would need to comply with federal and state legal requirements
and be the least change possible to existing legislative
districts. Six parties submitted maps for the Wisconsin Senate
and Assembly: the Legislature, CMS, the Hunter Intervenor-
Petitioners ("Hunter"), Senator Janet Bewley, the Governor, and
BLOC. The maps submitted by the Legislature and CMS achieve
minimal changes to existing district lines while best complying
with the demands of the Wisconsin Constitution and federal law.
For the most part, the parties argued for the adoption of either
the Legislature's or the Governor's maps.
A. The Equal Protection Clause And The VRA
¶80 The maps adopted by the majority are nothing short of
a racial gerrymander, and the Governor failed to present any
material evidence warranting this substantial departure from the
principles of equal protection.
8
No. 2021AP1450-OA.akz
¶81 Fatally, the majority provides at most a cursory
analysis on the VRA and the Equal Protection Clause, mustering a
mere five pages to apply an incredibly important and complex
area of law. See Ipse Dixit, Oxford English Dictionary (2022)
("An unproved assertion resting on the bare authority of some
speaker."). Just as BLOC warned, the majority's VRA analysis is
woefully inadequate at best. Its use of an aggressive race-
based remedy for no showing of a VRA violation, simply because
it can, is untenable and legal error.
¶82 The majority's use of race to draw seven bare-
majority-minority districts undermines that which the VRA was
properly meant to correct. It utilizes racial categories to
move minority voters into newly created districts, with newly
defined constituencies, which could not have been reasonably
created using traditional race-neutral redistricting methods.
Notably, the majority cites broad quotes taken from United
States Supreme Court precedent, but it conspicuously omits any
detailed description of the facts and outcomes of those cases,
i.e., what those cases actually stand for.5 No real attempt at
grappling with the vast nuances of VRA caselaw, from lower
courts to the United States Supreme Court, was given. By
5 For instance, the majority cites Cooper, 137 S. Ct. 1455,
Shaw v. Hunt, 517 U.S. 899, LULAC, 548 U.S. 399, De Grandy, 512
U.S. 997. In Cooper and Shaw, the Court struck down race-based
district maps under the Equal Protection Clause due to the lack
of support for VRA compliance. In LULAC, the Court found that
maps drawn in Texas lacked support under the VRA, and in
De Grandy, the Court held that the VRA did not apply at all,
where a plaintiff sought maximization of majority-minority
districts. A more complete analysis on the VRA is provided
below.
9
No. 2021AP1450-OA.akz
adopting the Governor's maps, the majority is now bringing to
the fore the incendiary and constitutionally suspect category of
race. The majority has a legal responsibility to more fully and
thoroughly explain itself. Below, I attempt to fill the void in
substance the majority leaves for future courts and the public.
¶83 What's next? Perhaps a federal court challenge before
the United States Supreme Court.6 Although braving a face of
finality, the majority opinion practically begs that the adopted
maps be subject to further litigation.
¶84 I first discuss the legal background of the Equal
Protection Clause, and then turn to a discussion on the VRA and
its application to this case.
1. The Equal Protection Clause
6 The parties to this lawsuit were given the opportunity to
present evidence, advance support for their favored maps, and
critique and oppose the maps ultimately adopted. The next step
for the case is appeal to the United States Supreme Court. See
Coleman v. Thompson, 501 U.S. 722, 730 (1991) (explaining that
the Supreme Court "reviews a state court decision on direct
review pursuant to 28 U.S.C. § 1257"). The parties are
precluded from relitigating this case in a separate federal
lawsuit. Migra v. Warren City Sch. Dist. Bd. of Educ., 465 U.S.
75, 81 (1984) (explaining that "a federal court must give to a
state-court judgment the same preclusive effect as would be
given that judgment under the law of the State in which the
judgment was rendered"); Wickenhauser v. Lehtinen, 2007 WI 82,
¶22, 302 Wis. 2d 41, 734 N.W.2d 855 (stating the elements of
claim preclusion). "Congress had empowered only [the United
States Supreme] Court to exercise appellate authority to reverse
or modify a state-court judgment." Exxon Mobil Corp. v. Saudi
Basic Indus. Corp., 544 U.S. 280, 284 (2005). Further, under
the "Rooker-Feldman" doctrine, "cases brought by state-court
losers complaining of injuries caused by state-court judgments
rendered before the district court proceedings commenced and
inviting district court review and rejection of those judgments"
fall outside federal district courts' subject matter
jurisdiction. Lance v. Dennis, 546 U.S. 459, 464 (2006).
10
No. 2021AP1450-OA.akz
¶85 Recognizing the deeply American value that individuals
should be equally protected under the law, the United States
Supreme Court has repeatedly held that government cannot sort or
distinguish individuals on the basis of race without
extraordinary justifications. "Distinctions between citizens
solely because of their ancestry are by their very nature odious
to a free people, and therefore are contrary to our traditions
and hence constitutionally suspect." Fisher v. Univ. of Texas,
Austin, 570 U.S. 297, 309 (2013) (citations and quotations
omitted). The Court has recognized that government-sanctioned
distinctions "threaten to stigmatize individuals by reason of
their membership in a racial group and to incite racial
hostility." Shaw v. Reno, 509 U.S. 630, 643 (1993). "Because
racial characteristics so seldom provide a relevant basis for
disparate treatment, the Equal Protection Clause demands that
racial classifications be subjected to the most rigid scrutiny."
Fisher, 570 U.S. at 309-10 (cleaned up). Classifications based
on race "are constitutional only if they are narrowly tailored
to further compelling governmental interests." Grutter v.
Bollinger, 539 U.S. 306, 326 (2003). This is a "searching
judicial inquiry," id., that rejects "any but the most exact
connection between justification and classification." Parents
Involved in Community Schools v. Seattle Sch. Dist. No. 1, 551
U.S. 701, 720 (2007) (quotations removed).
¶86 The Supreme Court has understood the pernicious nature
of dividing up individuals into legislative districts based on
11
No. 2021AP1450-OA.akz
race, and has applied the Equal Protection Clause to
redistricting. The Court is exacting in its scrutiny:
The idea is a simple one: At the heart of the
Constitution's guarantee of equal protection lies the
simple command that the Government must treat citizens
as individuals, not as simply components of a racial,
religious, sexual or national class. When the State
assigns voters on the basis of race, it engages in the
offensive and demeaning assumption that voters of a
particular race, because of their race, think alike,
share the same political interests, and will prefer
the same candidates at the polls. Race-based
assignments embody stereotypes that treat individuals
as the product of their race, evaluating their
thoughts and efforts——their very worth as citizens——
according to a criterion barred to the Government by
history and the Constitution. They also cause society
serious harm. . . .
Racial classifications with respect to voting carry
particular dangers. Racial gerrymandering, even for
remedial purposes, may balkanize us into competing
racial factions; it threatens to carry us further from
the goal of a political system in which race no longer
matters——a goal that the Fourteenth and Fifteenth
Amendments embody, and to which the Nation continues
to aspire.
Miller v. Johnson, 515 U.S. 900, 911-12 (1995) (cleaned up).
¶87 With this is mind, it is striking how explicitly the
Governor——and the majority——divide up Wisconsin districts solely
by race. While in 2011 the Legislature drew six assembly
districts that have a majority of black voting-age populations
("BVAP"), ranging from 51% to 62%, the Governor carves seven
districts by race with the exactness of only the most gifted
social scientists. According to the Governor himself, he drew
seven districts with BVAP ranging from 50.1% to 51.4%. At oral
argument and in briefing, it was clear that race imbued the
decisions of the Governor in drawing districts. Explaining his
12
No. 2021AP1450-OA.akz
district boundaries, he stated the intent was "to produce seven
majority Black districts in the Assembly." There is simply no
way to deny that the Governor created "[d]istinctions between
citizens solely because of their ancestry," and if his maps are
adopted, they must overcome strict scrutiny. Fisher, 570 U.S.
at 309; Grutter, 539 U.S. at 326.
¶88 On the other hand, it is undisputed that the
Legislature drew race-neutral maps. The Legislature sought to
retain districts that have high percentages of black individuals
to as close to the same as they were drawn in 2011, i.e., "least
change." See Johnson, 399 Wis. 2d 623, ¶72. The core retention
statistics from high BVAP districts differ dramatically between
the Legislature and the Governor. For the Legislature, the core
retention numbers for those districts were 87.7%, 85.4%, 88.1%,
100.0%, 94.3%, and 86.4%. By contrast, high BVAP districts for
the Governor had core retention percentages of 85.8%, 56.1%,
58.7%, 91.3%, 58.5%, 75.9%, and 12.7%. It is clear from the
data that the Legislature emphasized as little disruption as
possible for districts representing high percentages of African-
American citizens, as it did for all citizens, regardless of
race. By contrast, the Governor's driving motivation was race.
The Legislature confirmed at oral argument that the drawing of
its districts was driven by race-neutral constitutional criteria
and least change, not race.
¶89 Core retention numbers for high BVAP districts were
not available for CMS. However, the varying percentages of BVAP
in the maps presented help satisfy any concern that their
13
No. 2021AP1450-OA.akz
district choices were "motivated by a racial purpose or object."
Miller, 515 U.S. at 913. CMS has seven districts varying from
35.2% to 83.2% BVAP.7 The Legislature similarly has six
districts ranging from 45.8% to 71.5%. By comparison, the
Governor has seven districts with pinpoint accuracy of 50% to
51% BVAP. While the Governor has the hallmarks of an
unconstitutional racial gerrymander in violation of the Equal
Protection Clause, the Legislature and CMS do not.
2. The VRA
¶90 The Governor contends that his maps would survive
strict scrutiny because his seven districts are required under
§ 2 of the VRA. Through argument, it was made clear that the
Governor believed seven majority-minority districts with exactly
51% BVAP must be drawn because it is mathematically possible to
do so. That has never been the law. Fundamentally, drawing a
map based on race, to create another district because it can be
created, is a clear violation of equal protection. No VRA
violation has been demonstrated by district-specific evidence.
Despite the opportunity to engage in discovery, the Governor
presents no evidence on Wisconsin election history at all, no
evidence on the unique and specific history and socio-economic
experiences of minorities in the districts they seek to
manufacture. At most, BLOC (not the Governor) submitted
7 At oral argument, CMS also noted the striking degree to
which race infused the court's consideration and discussions,
along with the Governor's and others' race-based proposals.
Unlike the Governor, CMS affirmed that race should not and
cannot be the motivating factor behind drawing districts.
14
No. 2021AP1450-OA.akz
argument (not evidence) about Milwaukee as a county. Absent the
requisite showing, no district can be reconfigured based upon
race without violating the constitutional prohibition against
race-based action. Because there is no such evidence, the
Governor's maps fail and do not withstand constitutional
scrutiny.
¶91 The only support presented in an attempt to justify
race-based districts was submitted by a party who contends the
Governor's maps violate the VRA: BLOC. The majority does not
explain this but cites to BLOC's VRA record evidence to support
its choice of the Governor's map. See majority op., ¶45
(restating BLOC's number that African-American preferred
candidates are blocked "57.14%" of the time). Even BLOC offers
only broad assertions that are county specific, and a dearth of
district-specific race vote blocking. No party except BLOC
presented any details on the state and condition of minority
communities in the districts at issue, and even that evidence is
deeply flawed.
¶92 The United States Supreme Court has "assumed
that . . . complying with operative provisions of the Voting
Rights Act of 1965" can serve as a compelling interest.
However, the government must still satisfy the narrow tailoring
and "searching judicial inquiry" that strict scrutiny requires.
Parents Involved in Community Schools, 551 U.S. at 720; Bush v.
Vera, 517 U.S. 952, 978 (1996) ("Strict scrutiny remains,
nonetheless, strict."). There must be a "strong basis in
evidence" that the VRA requires the drawing of districts on race
15
No. 2021AP1450-OA.akz
to ameliorate harm and lack of access experienced by a minority
community. Miller, 515 U.S. at 922; accord Shaw v. Reno, 509
U.S. at 653 ("[R]acial bloc voting and minority-group political
cohesion [the requirements of a VRA redistricting violation]
never can be assumed, but specifically must be proved in each
case in order to establish that a redistricting plan dilutes
minority voting strength in violation of § 2."). "Strong" in
the context of evidence is defined as "convincing; hard to
refute, ignore, or deny." Strong, Oxford English Dictionary
(2022). This is not, as the majority appears to take it, a
minor procedural speedbump on the way toward racialized district
lines. See, e.g., Cooper v. Harris, 581 U.S. ___, 137
S. Ct. 1455, 1464 (2017) (holding that the State of North
Carolina lacked evidence to support race-based district
boundaries after examining in detail electoral history in the
districts at issue); Vera, 517 U.S. at 965-83 (examining in
detail the record justifying the district lines in Texas,
concluding that race motivated the district boundaries, and
reasoning that the districts at issue were insufficiently
compact to justify application of the VRA); Miller, 515 U.S. at
920-27 (reviewing in the context of § 5 of the VRA that the
record of the case, the justifications underlying district lines
in Georgia, and communications between the state and federal
government, and concluding that race-based district lines were
not justified under the VRA); Shaw v. Hunt, 517 U.S. at 916(concluding, even assuming the existence of "strong evidence" to
support the use of race under the VRA, simply creating majority-
16
No. 2021AP1450-OA.akz
minority districts where racially polarized voting occurs absent
a targeted remedy for the geographically compact voters harmed
fails to satisfy strict scrutiny).8
¶93 The operative language in § 2 of the VRA is that
election procedures and practices cannot, in the "totality of
the circumstances," create
political processes leading to nomination or election
in the State or political subdivision are not equally
open to participation by members of a [protected]
class of citizens . . . in that its members have less
opportunity than other members of the electorate to
8 The majority contends that a complete record to support
racially motivated district lines can be produced in a lawsuit
after the maps are enacted. Majority op., ¶41 n.24
(distinguishing a "VRA claim brought [] after the adoption of
new districts" from the review provided by the majority, reliant
upon a "limited record"). Under the majority's theory, VRA
requirements apply only when a government is brought to court.
However, state actors must consider whether there is a "strong
basis" to support race-based distinctions prior to engaging in
remedial action. See Shaw v. Hunt, 517 U.S. at 910 ("[T]he
institution that makes the racial distinction must have had a
strong basis in evidence to conclude that remedial action was
necessary, before it embarks on an affirmative-action
program."); see, e.g., Cooper, 137 S. Ct. at 1469-72 (examining
the motivation and support for applying a race-based remedy
under the VRA at the time of redistricting); Miller v. Johnson,
515 U.S. at 920-27 (reviewing the justifications for a state's
use of race in redistricting at the time of adoption of the
maps); Bethune-Hill v. Vir. State Bd. of Elections, 580 U.S.
___, 137 S. Ct. 788, 801-02 (2017) (examining the evidence and
justifications for a race-based distinctions at the time
legislative districts were drawn). As a court, the majority
should be considering the law when it selects its maps; the VRA
is the law.
17
No. 2021AP1450-OA.akz
participate in the political process and to elect
representatives of their choice.[9]
52 U.S.C. § 10301(b). The United States Supreme Court has
recognized that a violation of the statute is not dependent on
an "intent to discriminate against minority voters." Thornburg
v. Gingles, 478 U.S. 30, 44 (1986). Instead, courts must look
at effects to determine if the votes of a minority group have
been "diluted" to impair the ability of those minorities "to
elect representatives of their choice." 52 U.S.C. § 10301(b).
"[T]he 'essence' of a [VRA] § 2 vote dilution claim is that a
certain electoral law, practice, or structure causes an
inequality in the opportunities enjoyed by black and white
voters to elect their preferred representatives." Georgia v.
Ashcroft, 539 U.S. 461, 478 (2003).
¶94 Recognizing the broad remedial goals of § 2 of the VRA
and its more generalized application, untied to discriminatory
intent, the Supreme Court has held that the drawing of districts
could constitute an illegal impairment of minority voting rights
by permitting a white majority to override the minority's choice
in candidate. "[I]nteracting with social and historical
conditions," district lines that prevent a cohesive minority
from electing their preferred candidate "impairs the ability of
9The statute also states that "nothing in this section
establishes a right to have members of a protected class elected
in numbers equal to their proportion in the population." 52
U.S.C. § 10301(b). The United States Supreme Court has made
clear that there is a difference between minority-preferred
candidates and minority candidates. "[T]he ultimate right of
§ 2 is equality of opportunity, not a guarantee of electoral
success for minority-preferred candidates of whatever race."
De Grandy, 512 U.S. at 1014 n.11.
18
No. 2021AP1450-OA.akz
a protected class to [exercise voting rights] on an equal basis
with other voters." Johnson v. De Grandy, 512 U.S. 997, 1007
(1994). If certain conditions are met, a map may require the
"drawing of majority-minority district[s]." Cooper, 137 S. Ct.
at 1470.
¶95 The Supreme Court has demanded that three specific
elements be met before it finds that the creation of additional
majority-minority districts are necessary: "(1) the racial
group is sufficiently large and geographically compact to
constitute a majority in a single-member district; (2) the
racial group is politically cohesive; and (3) the majority votes
sufficiently as a bloc to enable it usually to defeat the
minority's preferred candidate." League of United Latin
American Citizens v. Perry, 548 U.S. 399, 425 (2006) (cleaned
up) ("LULAC").
¶96 These three elements of the so-called "Gingles test"
are necessary prerequisites for the creation of majority-
minority districts. They do not necessarily prove that an
election scheme fits the standard of "imped[ing] the ability of
minority voters to elect representatives of their choice" under
§ 2 of the VRA. Gingles, 478 U.S. at 48. To meet the standard,
there must be a proven record of discriminatory effects. Taken
from a 1982 report from the United States Senate, courts have
recognized as potentially significant:
the history of voting-related discrimination in the
State or political subdivision; the extent to which
voting in the elections of the State or political
subdivision is racially polarized; the extent to which
the State or political subdivision has used voting
19
No. 2021AP1450-OA.akz
practices or procedures that tend to enhance the
opportunity for discrimination against the minority
group . . . ; the extent to which minority group
members bear the effects of past discrimination in
areas such as education, employment, and health, which
hinder their ability to participate effectively in the
political process; the use of overt or subtle racial
appeals in political campaigns; and the extent to
which members of the minority group have been elected
to public office in the jurisdiction. The Report
notes also that evidence demonstrating that elected
officials are unresponsive to the particularized needs
of the members of the minority group and that the
policy underlying the State's or the political
subdivision's use of the contested practice or
structure is tenuous may have probative value.
LULAC, 548 U.S. at 426 (citing Gingles, 478 U.S. at 44-45).
¶97 None of the factors above are dispositive; however,
the three Gingles factors must be met before a court considers
whether the totality of the circumstances justifies a race-based
remedy. Courts consider the "totality of the circumstances" as
a second step to determine if the minority opportunities to
participate in the electoral process have been impeded. This is
an intensively fact-based analysis; it requires submission of
testimony and detailed expert reports on the state and
conditions of a localities' minority community, the extent they
face discrimination, the extent past discrimination still
impairs their ability to participate, current election rules,
and how those rules impact minorities. De Grandy, 512 U.S. at
1011 ("[E]quality or inequality of opportunity were intended by
Congress to be judgments resting on comprehensive, not limited,
canvassing of relevant facts"); Gingles, 478 U.S. at 45 ("[T]he
question whether the political processes are 'equally open'
depends upon a searching practical evaluation of the 'past and
20
No. 2021AP1450-OA.akz
present reality,' and on a 'functional' view of the political
process.").
¶98 To show that a district map is in violation of the VRA
and requires the creation of additional majority-minority
districts, there must be thorough factual findings. The Supreme
Court has repeatedly refused to apply a VRA remedy without
detailed factual evidence demonstrating the existence of the
Gingles factors, even prior to engaging in the more fact-
intensive "totality of the circumstances," i.e., the
characteristics of the minority community and their voter
behavior. See, e.g., Cooper, 137 S. Ct. at 1471-72 (concluding
that a majority-minority district created for VRA compliance was
unconstitutional because past election data showed super-
majority vote percentages by the candidate preferred by African-
Americans and effective white-bloc voting, the third Gingles
factor, was not proven, despite the possibility that new white
voters were added who could change the voting results); Bartlett
v. Strickland, 556 U.S. 1, 19-20 (2009) (plurality) (concluding
that § 2 of the VRA does not apply where the parties did not
prove a change in district lines would create a majority
African-American district, reasoning that the first Gingles
factor was not met); LULAC, 548 U.S. at 432 (holding that a
majority-Hispanic district was required but an existing map
creating a majority-Hispanic district failed to satisfy the VRA
because different Hispanics in different areas had "differences
in socio-economic status, education, employment, health, and
other characteristics," and there was insufficient evidence of
21
No. 2021AP1450-OA.akz
"compactness" under the first Gingles factor); Gonzalez v. City
of Aurora, 535 F.3d 594, 600 (7th Cir. 2008) (concluding that no
evidence was provided that voting opportunities for Hispanics in
a municipality were impaired, the plaintiff did not
"build . . . a factual record," and no VRA claim lay despite
Hispanics being dramatically less represented as a portion of
their population); Clarke v. City of Cincinnati, 40 F.3d 807,
812-13 (6th Cir. 1994) (noting that the electoral history for
the public offices at issue demonstrated that "47 percent of
blacks' preferred black candidates were elected" and thus there
was "no reason to find that blacks' preferred black candidates
have 'usually' been defeated" under Gingles).
¶99 Furthermore, well-established Supreme Court precedent
states that § 2 violations are determined by examining
individual districts and specific voting groups. Cooper, 137
S. Ct. at 1471-72, 1471 n.5 ("[G]eneralized conclusion[s]" of
state-wide racial polarization in voting "fails to meaningfully
(or indeed, at all) address the relevant local question:
whether, in a new version of District 1 created without a focus
on race, black voters would encounter sufficient white bloc-
voting to cancel their ability to elect representatives of their
choice." (cleaned up)); LULAC, 548 U.S. at 432, 437 (explaining
that VRA analysis requires "an intensely local appraisal" of the
relevant district); Shaw v. Hunt, 517 U.S. at 917 ("For example,
if a geographically compact, cohesive minority population lives
in south-central to southeastern North Carolina, as the Justice
Department's objection letter suggested, District 12 that spans
22
No. 2021AP1450-OA.akz
the Piedmont Crescent would not address that § 2 violation.");
Abbott v. Perez, 585 U.S. ___, 138 S. Ct. 2305, 2333-34 (2018)
(noting, despite evidence of a "long history of discrimination"
in Texas, a "pattern of disadvantage" for minorities, and
racially polarized voting in the region, there was insufficient
evidence of "present local conditions" to support a VRA remedy);
United States v. City of Euclid, 580 F. Supp. 2d 584, 604-12
(N.D. Ohio 2008) (examining in detail the need for a race-based
VRA remedy by considering the conditions and experiences of
specific African-American communities in a town of 50,000);
Comm. for a Fair & Balanced Map v. Ill. State of Bd. of
Elections, 835 F. Supp. 2d 563, 583 (N.D. Ill. 2011) (noting
that "northern and southern enclaves" of a Hispanic district had
"a common heritage and share[d] common core value[s]").
¶100 The inquiry is emphatically not to create "the maximum
number of majority-minority districts," regardless of the on-
the-ground characteristics of the minority neighborhoods and
communities at issue. De Grandy, 512 U.S. at 1016 (reversing a
district court's finding of § 2 violation because more Hispanic
majority-minority districts could have been created); Gonzalez,
535 F.3d at 598 ("But neither § 2 nor Gingles nor any later
decision of the Supreme Court speaks of maximizing the influence
of any racial or ethnic group."); Bartlett, 556 U.S. at 15("Nothing in § 2 grants special protection to a minority group's
right to form political coalitions.").
¶101 Thus, from these legal principles a picture of narrow
VRA compliance for this court emerges. Legislative boundaries
23
No. 2021AP1450-OA.akz
must be drawn to create effective majority-minority districts
only where proof is offered, and accepted by a court, that
existing districts or districts drawn using race-neutral
criteria would result in white voters, as a bloc, preventing
minorities from electing candidates that they support and that
represent them. In addition, evidence must be offered and
accepted that the minority needs representation from their
choice candidate due to depressed socio-economic statistics as a
result of current and historical discrimination, election
practices and procedures that encourage or facilitate racial
discrimination, and the lack of non-choice candidates to respond
to the "particularized needs of the members of the minority
group," among other factors. LULAC, 548 U.S. at 426, 440.
¶102 Further, there must be available the creation of
districts with majority-minority composition. Id. (stating the
first Gingles factor of "the racial group is sufficiently large
and geographically compact to constitute a majority in a single-
member district" (emphasis added)); Bartlett, 556 U.S. at 19(holding that § 2 does not require the creation of below-
majority "opportunity districts," reasoning that "a party
asserting § 2 liability must show by a preponderance of the
evidence that the minority population in the potential election
district is greater than 50 percent."). As the United States
Supreme Court explained in Cooper, when voters outside the
minority group act as sufficient "crossover" to "help [the]
minority to elect its candidate of choice," "it is difficult to
see how the majority-bloc-voting requirement could be met" under
24
No. 2021AP1450-OA.akz
Gingles. Cooper, 137 S. Ct. at 1471. If there is not
substantial proof that a majority-minority district can be
created, that minority voters are barred from effective
participation, or that minorities are blocked by white voters
from having representation, any consideration of race during
redistricting would violate the constitution. Id. at 1464-65.
Without the need to draw districts under the VRA, race-neutral
"traditional districting principles such as compactness,
contiguity, and respect for political subdivisions" must control
this court's decision. Shaw v. Reno, 509 U.S. at 647.
i. Gingles Factors and Bloc Voting
¶103 Despite the high demands of the VRA, coupled with the
need to meet VRA standards to justify the use of race to create
government policy under the Equal Protection Clause, it is
striking how insubstantial a record the Governor has provided to
support his racially driven maps. Courts have made it very
clear that substantial evidence must be produced of all three
Gingles factors to permit racial motivations in district
boundaries. Cooper, 137 S. Ct. at 1471-72; Bartlett, 556 U.S.
at 19-20; LULAC, 548 U.S. at 425; Gonzalez, 535 F.3d at 600;
Clarke, 40 F.3d at 812-13. However, unlike the leading cases on
the VRA, only BLOC engages in any detailed analysis on electoral
history. See LULAC, 548 U.S. at 423-29 (describing in detail
the electoral history, by race, of an at issue congressional
district to find a VRA violation); Cooper, 137 S. Ct. at 1470-72(explaining the electoral history of an area to determine that a
25
No. 2021AP1450-OA.akz
majority-minority district fell outside the VRA and was thus
unconstitutional).
¶104 The Governor presents, and the majority opinion
accepts, zero evidence of election history to support the
application of the Gingles factors to the current maps, the
Legislature's maps, or other race-neutral alternatives to
support his division of districts by race. Further, the
Governor presents no electoral history evidence to prove the
existence of the Gingles factors in any of the specific
districts he drew. Such evidence is also lacking to show the
Governor's maps comply with the VRA, as compared to BLOC's maps,
which also include seven black-majority districts. In a twist
of fate, this leaves open the possibility that VRA compliance is
not met for the Governor's maps, even if the VRA is triggered
and requires raced-based districts.
¶105 The only thing the Governor does do that approaches
objective or scientific argument is cite population percentages
of African-Americans in Wisconsin. The Governor thereby
concludes that seven districts of a bare 51% BVAP can be drawn,
and must be drawn. This notwithstanding that the United States
Supreme Court has explicitly rejected the same logic on numerous
occasions. De Grandy, 512 U.S. at 1016 (rejecting a claim that
§ 2 requires states to create "the maximum number of majority-
minority districts"); Bartlett, 556 U.S. at 15 ("Nothing in § 2
grants special protection to a minority group's right to form
political coalitions."); Gonzalez, 535 F.3d at 598 ("But neither
§ 2 nor Gingles nor any later decision of the Supreme Court
26
No. 2021AP1450-OA.akz
speaks of maximizing the influence of any racial or ethnic
group."). Stopping here, the Governor has failed to provide any
evidence specific to his proposed districts warranting a finding
of white bloc voting that can effectively overcome a
politically-cohesive black voting bloc, let alone strong and
convincing evidence sufficient to overcome strict scrutiny. See
Miller, 515 U.S. at 922. This alone should counsel the court to
reject the Governor's map and adopt the race-neutral maps
presented by either the Legislature or the CMS.
¶106 This is exactly the form of analysis that the Michigan
Supreme Court recently applied. Detroit Caucus v. Indep.
Citizens Redistricting Comm'n, ___ N.W.2d ___, 2022 WL 329915(Mem) (Mich. Feb. 3, 2022). The court found that "a conclusory
expert affidavit with no accompanying bloc-voting analysis" was
insufficient to support the use of race to create additional
majority-minority districts which the state could have drawn,
but did not. Id. at *2. The Governor in this case has
presented little more evidence than the inadequate VRA showing
made in Detroit Caucus. Notably, when a full and complete
election history analysis was performed in Michigan,
"significant white crossover voting for Black-preferred
candidates" was found. Id.
¶107 Furthermore, the Governor's maps actually reduce the
percentage of African-American voters in the relevant districts
from their existing levels. The VRA is invoked only when
minorities, due to a mobilized and oppositional majority, cannot
effectively participate and elect preferred candidates.
27
No. 2021AP1450-OA.akz
Gingles, 478 U.S. at 48; De Grandy, 512 U.S. at 1007. The maps
adopted by the majority reduce this population allegedly
overpowered by a white majority, instead of giving it a greater
voice within the aggrieved districts. Of course then, the
districts cannot be so aggrieved, and no evidence exists so to
invoke the VRA. In other words, before a change is to be made
under the VRA, there must be a violation of the VRA so to invoke
its remedy. The remedy is to cure the suppressed voter effect
by giving minority voters greater voice, not reducing their
voice. Alone, this statistic puts a dagger in the Governor's
map.
¶108 Lacking any support in the record, one might turn to
the presentations made by BLOC, the only other party that
supported racially-motivated district lines but also provided
electoral evidence. In fact, the majority's sole citation to
electoral history evidence relied on BLOC's expert report. See
majority op., ¶45 (restating BLOC's statistics on the rate in
which African-American preferred candidates are blocked). Yet
even that evidence is flawed. BLOC selects eight oddly
identified races from Milwaukee County (two comptroller races,
and one race each for sheriff, democratic gubernatorial primary,
state assembly, mayor, Milwaukee county executive, and state
superintendent) to evidence the region's electoral history.
Only one election was examined that involved the public offices
at issue in this case: assembly, senate, and congressional
elections. This is markedly at odds with traditional VRA
analysis. See, e.g., Cooper, 137 S. Ct. at 1471-72 (examining
28
No. 2021AP1450-OA.akz
the electoral history of a congressional district at issue in
the challenge); LULAC, 548 U.S. at 427-28 (explaining electoral
history in the congressional district at issue); City of Euclid,
580 F. Supp. 2d at 598-600 (describing non-applicable elections
in the context of a detailed review of city council elections at
issue in the lawsuit); Harper v. City of Chicago Heights, 824
F. Supp. 786, 790, 799-800 (N.D. Ill. 1993) (examining the
electoral history of specific city commissioner offices at
issue).
¶109 While some elections may be of more probative value
than others, the provision of only eight elections, and only one
of which involving the elected offices at issue, can hardly
demonstrate the extent to which black people, under existing and
race-neutral maps, lack the same "opportunity . . . to
participate in the political process and to elect
representatives of their choice" as do white people. 52 U.S.C.
§ 10301(b); see Bone Shirt v. Hazeltine, 336 F. Supp. 2d 976,
996 (D.S.D. 2004) (explaining a common hierarchy of election
history value, when such history is available, noting that
"[e]ndogenous elections, contests within the jurisdiction and
for the particular office that is at issue, are more probative
than exogenous elections").10 Undoubtedly, dozens of elections
have occurred in the Milwaukee-area state assembly, senate, and
10 If this were otherwise, it is highly likely that
governments would simply cite state-wide general election
results (white versus minority percentages) to justify racially
motivated district lines, in almost every state in almost every
region of the country. This would be a dramatic expansion of
the permissible use of race in American election practices.
29
No. 2021AP1450-OA.akz
congressional districts at issue in the past 10 years alone.11
The court's focus is on the "totality of the circumstances" and
whether as a whole African Americans are denied the opportunity
to effectively participate in electoral democracy. 52 U.S.C.
§ 10301(b). The consistent election of candidates of choice for
the African-American community into public office in the
districts at issue would be highly probative. Yet the record is
completely devoid of any evidence that the voters in these
districts were blocked from voting in the candidates of their
choice in a way that would invoke the VRA.
¶110 Even under BLOC's selective analysis, white voters
engaged in bloc voting to prevent the candidate of choice for
African-Americans four times. That is around a 50% rate——hardly
the kind of strong evidence needed to overcome strict scrutiny.
Compare Clarke, 40 F.3d at 812-13 (even when considering
applicable electoral history, concluding that minority-preferred
candidates were not "usually" defeated when the minority-
preferred candidate was selected in 47% of elections). BLOC
disaggregated allegedly polarized election results for each
individual district it drew for only three races (a Democratic
gubernatorial primary, a Milwaukee county executive race, and a
state superintendent race). But how can the court effectively
perform an "intensely local appraisal" of district-specific
evidence when election results for these districts are provided
for a mere three races, none of which were for the elected
The dissent of Justice Roggensack, which follows this
11
dissent, identifies many such elections of black-preferred
candidates in districts that are predominantly white.
30
No. 2021AP1450-OA.akz
offices at issue? LULAC, 548 U.S. at 437. Of the three races
selected for district-specific treatment, only one of them had a
head-to-head race where voters did not split votes between
several candidates (thus preventing a more complete picture of
voter preferences).
¶111 The district-specific evidence of two races BLOC
provided was limited only to BLOC's proposed assembly districts.
BLOC did not provide detailed district analyses of the current
maps, an alternative race-neutral map, nor any other party's
maps outside one Democratic gubernatorial primary in 2018. In
the process of this litigation, the court has not been made
aware of a single case that found the existence of a strong
evidentiary record, applied the VRA, and satisfied strict
scrutiny through use of one election result, let alone a result
from an exogenous election (from a partisan primary between
candidates with strong support from the African-American
community).12 Compare LULAC, 548 U.S. at 427-28 (examining
partisan general election results); Cooper, 137 S. Ct. at 1470-
71 (reviewing partisan general election results); City of
Euclid, 580 F. Supp. 2d at 598-99 (explaining electoral history
for non-partisan general election results); Harper, 824 F. Supp.
See
12 Wisconsin Governor Exit Polls, CNN,
https://www.cnn.com/election/2018/results/wisconsin/governor
(last visited Feb. 10, 2022) (explaining how the Governor was
elected statewide on the support of 85% of the African-American
population).
31
No. 2021AP1450-OA.akz
at 790 (reviewing non-partisan general election results).13 To
understate the point, this substantially limits the ability of
the court to effectively judge if African-American voters are
having their candidates blocked and their voices unlawfully
stifled, therefore justifying race-based redistricting. See,
e.g., Comm. for Fair & Balanced Map, 835 F. Supp. 2d at 587
13 Of course, considering the wide-sweeping scope of VRA
review, primary elections may be valid considerations when
determining if a racial group has equal opportunity to
participate in elections. See Thornburg v. Gingles, 478 U.S.
30, 59 (1986) (reviewing both general and primary election
results). However, party makeups can change dramatically over
time. At some points in history, a party may contain voters
with markedly different views on the treatment of minorities.
See, e.g., Glenn T. Eskew, George C. Wallace, Encyclopedia of
Alabama, (Jun. 10, 2021) (describing the political history of
George Wallace, an outspoken supporter of racial segregation and
a lifelong Democrat). BLOC's analysis presents serious
questions of whether current Democratic primary elections in
Wisconsin, standing alone, are substantially probative on the
ability of African-Americans to have effective opportunities,
voices, and representation in democratic government.
32
No. 2021AP1450-OA.akz
(concluding that white bloc voting was not met where an expert
failed to provide evidence on specific districts at issue).14
14Furthermore, race-based redistricting under § 2 of the
VRA applies only where voting is polarized to such an extent
that a white majority blocks African-American-supported
candidates so that the only way African-American individuals can
effectively participate in democracy is to create majority-
minority districts. See Bartlett v. Strickland, 556 U.S. 1, 19
(2009) (plurality) (holding that § 2 does not require the
creation of below-majority "opportunity districts"); Cooper, 137
S. Ct. at 1464-65. A bare majority of African-American voters
is unlikely, absent extraordinary polarization, to prevent white
bloc-voting (if it exists) from stopping effective African-
American representation. Along these lines, courts attempting
to ensure VRA compliance have accepted the need to create VRA
districts with BVAP percentages materially greater than a bare
51% majority. See, e.g., Comm. for a Fair & Balanced Map v.
Ill. State Bd. of Elections, 835 F. Supp. 2d 563, 582 (N.D. Ill.
2011) ("60 percent of voting-age population is reasonably
required to ensure minorities a fair opportunity to elect a
candidate of their choice."); Hastert v. State Bd. of Elections,
777 F. Supp. 634, at 647 (N.D. Ill. 1991) (noting that a "65%
minority population [or 60% minority voting-age population]
concentration [is] generally regarded as necessary to ensure
minorities a reasonable opportunity to control a district");
Baumgart v. Wendelberger, No. 01-C-0121, 2002 WL 34127471, at *5
(E.D. Wis. May 30, 2002) (recognizing expert testimony that "a
minority district requires an African–American voting age
population of at least 60% to guarantee the election of
candidates of choice"); United States v. City of Euclid, 580
F. Supp. 2d 584, 594 n.11 (N.D. Ohio 2008) (explaining that the
efficacy of a "narrow" majority-minority district is subject to
question and this is remedied by majority-minority districts in
excess of "60%"); Baldus v. Members of Wis. Gov't Accountability
Bd., 849 F. Supp. 2d 840, 851 (E.D. Wis. 2012) (creating a
majority-minority Hispanic district, effective at 67.7% voting-
age population); African American Voting Rights Legal Defense
Fund, Inc. v. Villa, 54 F.3d 1345, 1348 n.4 (8th Cir. 1995)
("[A] guideline of 65% of total population (or its equivalent)
has achieved general acceptance in redistricting
jurisprudence."); Ketchum v. Byrne, 740 F.2d 1398, 1403 (7th
Cir. 1984) ("A guideline of 65% of total population has been
adopted and maintained for years by the Department of Justice
and by reapportionment experts and has been specifically
approved by the Supreme Court."). When commenting on total
voter population percentage, the court in Prosser explained that
33
No. 2021AP1450-OA.akz
¶112 Strikingly, under BLOC's analysis, the Governor's maps
do not satisfy the VRA, and are thus unconstitutional. The
majority not only lacks evidence to support the maps it adopts,
but the only party who even attempted to prove a VRA need
determined those maps were illegal.15
ii. Totality of the Circumstances
¶113 The Gingles factors are only "necessary
prerequisites," they are not "sufficient" to justify a race-
effective majority-minority districts require 65% minority
populations "(50 percent plus 5 percent to reflect the lower
average age of blacks and hence lower voting population, 5
percent to reflect a lower fraction of registered voters, and 5
percent to reflect a lower turnout)." Prosser v. Elections Bd.,
793 F. Supp. 859, 869 (W.D. Wis. 1992). Even if evidence
supported the race-based remedy offered by the Governor, his
bare-majority districts fall outside the mainstream of accepted
VRA redistricting measures.
Even if, due to specific electoral statistics and
15
community-based evidence in Milwaukee, a seventh high-BVAP
district were required, that in no way explains why the
remaining six high-BVAP districts must be drawn with a scalpel
to reach exactly 51% BVAP. Racially motivated government action
must be "narrowly tailored" to satisfy strict scrutiny. Grutter
v. Bollinger, 539 U.S. 306, 326 (2003); see, e.g., Shaw v. Hunt,
517 U.S. at 916-18 (concluding that districts drawn on the basis
of race were not "narrowly tailored" because the government drew
district lines from scattered minority communities which may
have different VRA needs and were thus not sufficiently
compact). The VRA must be tied to individuals and their
specific communities, not general categories of race. Shaw v.
Hunt, 517 U.S. at 917 (affirming that the VRA protects
"individual[s]" not "the minority as a group"); LULAC, 548 U.S.
at 437 ("A local appraisal is necessary because the right to an
undiluted vote does not belong to the minority as a group, but
rather to its individual members."); De Grandy, 512 U.S. at 1016(explaining that, even when the Gingles factors and the totality
of the circumstances require race-based redistricting, the VRA
does not support creating "the maximum number of majority-
minority districts").
34
No. 2021AP1450-OA.akz
based remedy under the VRA. Gingles, 478 U.S. at 50; De Grandy,
512 U.S. at 1011. In addition to the Gingles factors, the VRA
requires proof that the "totality of the circumstances" supports
the drawing of districts on the basis of race. Gingles, 478
U.S. at 50; De Grandy, 512 U.S. at 1011; LULAC, 548 U.S. at 436;
Bartlett, 556 U.S. at 24. Totality of the circumstances is an
independent, separate requirement; to apply a race-based remedy
a totality of the circumstances analysis must be provided. The
majority's description of the totality of the circumstances is
shockingly insubstantial.
¶114 Proportionality of majority-minority districts to the
"citizen voting-age population" can be relevant to the totality
of the circumstances analysis. LULAC, 548 U.S. at 436. The
Legislature's expert notes that various data files show an
African-American citizen voting-age population ("CVAP") of
either 6.1% of 6.4% (taken from two different U.S. Census data
files). The Governor fails to present evidence on the issue.
While BLOC strenuously opposes the Legislature's numbers, their
expert suggests an African-American CVAP of 6.5%. Even if
BLOC's number were accepted, a proportionality analysis would
not support seven assembly districts. There are 99 assembly
districts, 6.5% of 99 is 6.4, which rounding to the nearest
whole number would be 6. At the very least, a proportionality
analysis does not provide strong support for a seventh district.
¶115 The majority notes that the African-American CVAP in
Wisconsin falls between 6.1% and 6.5%, but it fails to complete
the final step of a proportionality inquiry: multiplying the
35
No. 2021AP1450-OA.akz
CVAP by the relevant number of seats, here 99. Majority op.,
¶48. It thus states a misleading statistic of 6.5% and hopes
the reader confuses it for a complete proportionality analysis.
Further, the majority relies heavily on population trends among
black and white individuals, as well as demographic statistics
in Milwaukee County. See majority op., ¶48 ("[A] significant
proportion of Wisconsin's Black population lives in Milwaukee
County where the subject districts are principally located.").
Yet the United States Supreme Court in League of United Latin
American Citizens v. Perry explicitly rejected the use of
"regional" as opposed to "statewide" proportionality analysis
for statewide districting plans. 548 U.S. at 436-38. And
proportionality refers to the percentage of a given race in a
state. Id. at 436 (explaining that the proportionality of a
race is determined by comparing the number of minority districts
to "the [minority] share of the citizen voting-age population").
Proportionality does not encompass an increase or decrease of
anything, i.e., population trends amongst the African-American
population. The majority both twists the natural meaning of
English and refuses to comply with explicit Supreme Court
directives.
¶116 Beyond proportionality, the majority fails to discuss
any of the 1982 Senate Report factors relied upon by courts to
determine if the VRA applies. Gingles, 478 U.S. at 43-45;
LULAC, 548 U.S. at 426; see, e.g., City of Euclid, 580 F. Supp.
2d at 604-12 (providing a totality of the circumstances
analysis). Those factors lay at the heart of a totality of the
36
No. 2021AP1450-OA.akz
circumstances analysis; they are the reason why racially
motivated maps may satisfy strict scrutiny. Gingles, 478 U.S.
at 50; De Grandy, 512 U.S. at 1011; LULAC, 548 U.S. at 426;
Bartlett, 556 U.S. at 24. Nonetheless, the factors are
completely ignored.
¶117 The majority shortcuts the required analysis and
instead relies on the flawed belief that proportionality is the
preeminent consideration for totality of the circumstances.
Majority op., ¶46 n.28, ¶¶47-50 (stating that courts
"focus[] . . . [their] attention on considerations not mentioned
in the Senate Report, such as proportionality," and examining
only proportionality in a totality of the circumstances
analysis). That is flatly contradicted by established United
States Supreme Court precedent. De Grandy, 512 U.S. at 1011-12(rejecting the argument that proportionality is determinative of
VRA compliance and noting that "[n]o single statistic provides
courts with a shortcut"); Gingles, 478 U.S. at 47 ("The essence
of a § 2 claim is that a certain electoral law, practice, or
structure interacts with social and historical conditions to
cause an inequality in the opportunities enjoyed by black and
white voters to elect their preferred representatives."); LULAC,
548 U.S. at 426, 436-42 (laying out the Senate Factors as
considerations for totality of the circumstances analyses and
examining both proportionality and several Senate Factors when
determining the VRA required redrawing of certain districts in
Texas). By statute, the VRA requires examination of the
"totality of the circumstances," 52 U.S.C. § 10301; nowhere in
37
No. 2021AP1450-OA.akz
the statute does it state or imply that proportionality should
be the primary "focus[] . . . of [the court's] attention."
Majority op., ¶46 n.28.
¶118 There is a simple reason no real support is provided
by the majority for the totality of the circumstances: there is
none. The only party who even attempted to argue for VRA
application under the totality of the circumstances was BLOC.
The Governor presented no totality of the circumstance support
for his districts. Either the majority does not rely on BLOC,
and thus zero evidence is available to support the application
of the VRA, or, in the alternative, the majority must rely
solely on BLOC's analysis. In either case, there is no
justification for use of race in drawing the Governor's maps.
¶119 BLOC's totality of the circumstances analysis is
deeply flawed and is in the form of an expert opinion alone.
This lone source of evidence is highly debatable, and strikes an
unmistakable tone of partisanship, attacking political opponents
and disfavored policies. Such conclusory opinion evidence does
not amount to the kind of factual district-specific evidence
that could support a conclusion that a VRA violation has
occurred and the remedy must be creation of seven bare-majority
districts. Cooper, 137 S. Ct. at 1471-72; LULAC, 548 U.S. at
432; Shaw v. Hunt, 517 U.S. at 917; City of Euclid, 580
F. Supp. 2d at 604-12; Comm. for Fair & Balanced Map, 835
F. Supp. 2d at 583.
¶120 For instance, BLOC claims Milwaukee's choice to close
polling locations during the COVID-19 Pandemic and voter ID laws
38
No. 2021AP1450-OA.akz
demonstrate the existence of racially discriminatory election
practices. No evidence or explanation is provided as to how
these basic administrative and perfectly legitimate election
practices "tend to enhance the opportunity for discrimination
against the minority group." Gingles, 478 U.S. at 44-45. This
is far cry from the "poll tax, an all-white primary system, and
restrictive voter registration time periods," used in the past
in parts of the country to mask disenfranchisement of African-
American voters. LULAC, 548 U.S. at 439-40; see also De Grandy,
512 U.S. at 1018 ("In a substantial number of voting
jurisdictions, that past reality has included such reprehensible
practices as ballot box stuffing, outright violence,
discretionary registration, property requirements, the poll tax,
and the white primary; and other practices censurable when the
object of their use is discriminatory, such as at-large
elections, runoff requirements, anti-single-shot devices,
gerrymandering, the impeachment of office-holders, the
annexation or deannexation of territory, and the creation or
elimination of elective offices.").
¶121 The State of Wisconsin must strive to eliminate any
voting practice that facilitates unjust discrimination.
According to BLOC, must the state control election
administration in Milwaukee to prevent consolidation of polling
locations and covert discriminatory practices? Must the state
revoke its Voter-ID laws? See Frank v. Walker, 768 F.3d 744,
753-54 (7th Cir. 2014) (upholding a direct § 2 VRA challenge
against Wisconsin's Voter-ID law, noting "[s]ection 2(b) tells
39
No. 2021AP1450-OA.akz
us that § 2(a) does not condemn a voting practice just because
it has a disparate effect on minorities," there was no finding
"blacks . . . have less 'opportunity' than whites to get photo
IDs," and black individuals had equal if not higher voter
registration and turnout in the 2012 election as compared to
white individuals); Brnovich, 141 S. Ct. at 2345 (noting that "a
distorted picture can be created" by the manipulative use of
statistics, such as "[i]f 99.9% of whites had photo IDs, and
99.7% of blacks did, it could be said that blacks are three
times as likely as whites to lack qualifying ID (0.3 ÷ 0.1 = 3)"
(quotations omitted)); Crawford v. Marion Cnty. Elections Bd.,
553 U.S. 181, 204 (2008) ("The application of [Indiana's Voter-
ID law] to the vast majority of Indiana voters is amply
justified by the valid interest in protecting the integrity and
reliability of the electoral process.").
¶122 BLOC also looks at general socio-economic correlations
between white and African-American individuals in Wisconsin,
including the lower rates of African-American homeownership and
lower average incomes, and concludes, without any substantial
analysis on the extraordinary complexities of causation, that
this is the result of current and past discrimination. The
accepted fact that African-American individuals experienced
despicable forms of discrimination, specifically racial housing
covenants in the Milwaukee-area, is certainly a factor impacting
VRA analyses, but mere conclusions of discriminatory effects for
all African-American individuals in Milwaukee from race-based
correlations is not substantial evidence of discriminatory
40
No. 2021AP1450-OA.akz
hindrances on the ability of African-American individuals "to
participate effectively in the political process." Gingles, 478
U.S. at 44-45. It is the burden of those seeking to use race in
district boundaries to prove the need for such practices. Mere
inferences and assumptions cannot be sufficient.
¶123 Further, BLOC asserts proof of race baiting and
racially motivated campaigning by pointing to statements from
Republicans and conservatives critiquing the Black Lives Matter
organization, taking knees during national anthems, and
defunding the police. Notably, despite the fact that BLOC
relies heavily on Democratic primary data to demonstrate bloc-
voting and the need for race-drawn districts, the racial animus
directed toward African-American individuals in campaigns and
public messages all allegedly come from conservative
Republicans. There is no evidence offered by BLOC that the
Democratic public officials who at times defeat African-American
preferred candidates, such as the Governor in his Democratic
primary, are "unresponsive to the particularized needs of the
members of" the African-American community. Gingles, 478 U.S.
at 44-45; see LULAC, 548 U.S. at 426, 440 (explaining in detail
that a current representative for a district subject to VRA
scrutiny was "unresponsive" to the needs of the minority
community). Shockingly, BLOC contends that African-American
candidates have only had "mixed success" in the districts at
issue. Relying on exogenous and state-wide elections, BLOC
ignores the fact that the current assembly, senate, and
41
No. 2021AP1450-OA.akz
congressional districts have elected African-American office
holders in the vast majority of elections.
¶124 The evidence offered by BLOC of the totality of the
circumstances is hardly localized to the historical, societal,
and economic experiences of specific neighborhoods in Milwaukee.
Underlying BLOC's analysis is the assumption that all African-
American individuals in Wisconsin have the same history,
experiences, and effects of discrimination, and there is no need
to go further than broad strokes of correlations, debatable
assumptions, and talking-points. See LULAC, 548 U.S. at 432(examining in a VRA analysis that different Hispanics in
different parts of Texas had "differences in socio-economic
status, education, employment, health, and other
characteristics"); Comm. for Fair & Balanced Map, 835
F. Supp. 2d at 583 (noting that "northern and southern enclaves"
of a Hispanic district had "a common heritage and share[d]
common core value[s]"); City of Euclid, 580 F. Supp. 2d at 605-
07 (explaining in detail, with numerous experts reports, record
evidence, and testimony, forms of official discrimination
against a discrete African-American community in Euclid, Ohio).
Individuals, communities, and societal groups differ, even if
they are the same race. In fact, the maps offered by the
Legislature and CMS recognize that many of the African-Americans
moved under the Governor's maps are located in discrete and
compact neighborhoods. Following traditional redistricting
criteria, and putting together those with shared communities,
interests, and experience, the Legislature's and CMS's districts
42
No. 2021AP1450-OA.akz
fluctuate in BVAP to recognize this geographical reality.16 By
comparison, for their purported benefit, the majority chooses to
displace many African Americans and move them into districts
with little societal, cultural, and economic similarities.17
iii. The Majority Opinion and Party Concessions
¶125 Despite all its faults, BLOC at least provided some
evidence supporting their VRA claims. The Governor presented
nothing, let alone district-specific evidence. This flies in
the face of well-accepted precedent on overcoming strict
scrutiny and proving VRA needs. See Vera, 517 U.S. at 965-83;
Miller, 515 U.S. at 920-27; Shaw, 517 U.S. at 916; Cooper, 137
S. Ct. at 1471-72; Bartlett, 556 U.S. at 19-20; Perry, 548 U.S.
at 432; Gonzalez, 535 F.3d at 600; Clarke, 40 F.3d at 812-13;
City of Euclid, 580 F. Supp. 2d at 604-12; Committee for a Fair
and Balanced Map, 835 F. Supp. 2d at 583; Harper, 824 F. Supp.
See John Johnson, Neighborhoods Where Milwaukee Isn't
16
Segregated, Marquette University Law School (Feb. 9, 2022),
https://law.marquette.edu/facultyblog/2022/02/neighborhoods-
where-milwaukee-isnt-segregated/ (describing the demographic
makeup of the many unique neighborhoods in Milwaukee).
In the process, to obtain his 51% BVAP districts, the
17
Governor shifted white voters (referred to as "filler" voters at
oral arguments) into new districts to achieve targeted racial
proportions. The VRA by its text does not apply solely to any
one race, and both the Equal Protection Clause and the Fifteenth
Amendment's prohibition on racial discrimination in voting
practices apply to all races. See 52 U.S.C. § 10301; U.S.
Const. amend. XIV; U.S. Const. amend. XV; Shaw v. Reno, 509 U.S.
630, 657 (1993) ("Racial gerrymandering, even for remedial
purposes, may balkanize us into competing racial factions; it
threatens to carry us further from the goal of a political
system in which race no longer matters——a goal that the
Fourteenth and Fifteenth Amendments embody, and to which the
Nation continues to aspire.").
43
No. 2021AP1450-OA.akz
At 790, 799-800. Yet that does not seem to bother the majority
as they walk blindfolded into a buzz saw of Equal Protection
law. Given that, under BLOC's analysis, the Governor's maps
violate the VRA, the majority's maps may bear the usual stigma
of violating the Equal Protection Clause and the VRA at the same
time.
¶126 Counterintuitively, a linchpin of the majority's VRA
analysis is an alleged lack of evidence and argument. The
majority opinion may leave the reader with the misperception
that all litigants at this court agreed that a racial
gerrymander under the VRA was necessary. See majority op., ¶45
(noting "little . . . alternative data or analysis" to counter
BLOC's election history and indicating that the "parties
appeared to assume the VRA requires" race-based district lines).
That is patently inaccurate. In briefing, the Legislature was
clear that its maps both provided African-Americans equal
opportunity "to participate in the political process and to
elect representatives of their choice" (thus satisfying the
VRA), 52 U.S.C. § 10301(b), and was not motivated by race (thus
satisfying the Equal Protection Clause), Miller, 515 U.S. at
911-12. The Legislature asserted that the Governor's maps
"reveal a policy of prioritizing mechanical racial targets above
all other districting criteria (save one-person, one-vote),
meaning there is ample evidence that race motivated the drawing
of particular lines." Further, the Legislature claimed,
correctly, that the Governor "offered zero evidence that the
existing districts do not give all voters equal opportunity to
44
No. 2021AP1450-OA.akz
elect their candidate of choice." In the Legislature's reply
brief, it argued the Governor presented "novel and likely
unconstitutional" arguments in support of seven bare majority-
minority districts, labeled by the Legislature as an
"unconstitutional racial gerrymander." The Legislature
reaffirmed in the same brief that its "redistricting plan was
drawn without regard to race." Further, the Legislature's
expert, John Alford, described in many pages of detail the
computational and data concerns with the evidence submitted by
BLOC to support application of the VRA. He stated explicitly,
"[T]he election patterns detailed by [BLOC] raise serious doubts
about whether the Gingles threshold standard is currently met in
Milwaukee County." Finally, Mr. Alford observed that, even using
BLOC's election data, the black-preferred candidate was blocked
in less than 50% of elections.
¶127 The central goal of the Legislature's proposed maps
was to conserve existing boundaries for districts with high
BVAP, not draw districts to maximize majority-minority
45
No. 2021AP1450-OA.akz
districts. The Legislature's race-neutral intentions were
confirmed at oral argument.18
¶128 But, even so, why is the majority attached to party
briefing? They have a responsibility to read the law,
understand available evidence, and come to the correct
18 The majority adds in an argument that the Legislature's
districts in some way "pack" African-American voters into a
district with above 70% BVAP. Majority op., ¶49. The
Legislature has one district at 71.5% BVAP. As the majority
notes, it is well established that the VRA requires the creation
of race-based districts where minorities are
"fragment[ed] . . . among several districts where a bloc-voting
majority can routinely outvote them," or where minorities are
"pack[ed] . . . into one or a small number of districts to
minimize their influence in the districts next door."
De Grandy, 512 U.S. at 1007; see majority op., ¶49. But the
United States Supreme Court has clarified that the VRA applies
only to the creation of majority-minority districts; it does not
require splitting up high minority-percentage districts to more
effectively spread the minority's political influence.
Bartlett, 556 U.S. at 19; Cooper 137 S. Ct. at 1471 (explaining
that without the need for a majority-minority district
sufficient white crossover would undermine the satisfaction of
the Gingles factors). Thus, the inquiry is whether there has
been presented evidence of effective white bloc voting to
prevent minorities in a specific area and district from
successfully electing candidates they support. Even if the
Legislature drew a higher BVAP district following race-neutral
redistricting criteria such as preserving continuity of
interests, geographic compactness, and local government lines,
without the requisite evidence of a VRA violation in a separate,
neighboring district where a majority-minority district could be
created, no race-based remedy under the VRA can be used. Here,
there is no such district-specific evidence. The majority does
not cite a single case holding that merely having a high BVAP
district, without the need to prove the Gingles factors or the
need for a race-based remedy under the totality of the
circumstances, violates the VRA. See Ketchum, 740 F.2d at 1403-
06, 1418 (case cited by the majority, noting the commonly
accepted target of 65-70% minority population percentages in
applying a VRA remedy, after a VRA violation in relevant
districts has been established).
46
No. 2021AP1450-OA.akz
conclusion. See State v. Hunt, 2014 WI 102, ¶42 n.11, 360
Wis. 2d 576, 851 N.W.2d 434 ("Because it is our constitutional
duty to say what the law is, we are not bound by a party's
concessions of law."). They, not the litigants, are the
government actors. U.S. Const. amend. XIV sec. 1 ("No State
shall . . . deny to any person within its jurisdiction the equal
protection of the laws." (Emphasis added.)); Brentwood Academy
v. Tenn. Secondary Sch. Athletic Ass'n, 531 U.S. 288, 295 (2001)
(noting that only those "outside formally governmental
organizations" fall outside the coverage of the Fourteenth
Amendment); Johnson v. California, 543 U.S. 499, 505 (2005)
("Under strict scrutiny, the government has the burden of
proving that racial classifications are narrowly tailored
measures that further compelling governmental interests."
(Emphasis added.)). They are the ones choosing a map for the
State of Wisconsin, endorsing district boundaries unambiguously
motivated by race. See, e.g., De Grandy, 512 U.S. 997(reviewing under traditional Equal Protection and VRA standards
maps approved by the Florida Supreme Court). The court, acting
on behalf of the State of Wisconsin, not the parties, must
overcome strict scrutiny. See Grutter, 539 U.S. at 326(describing strict scrutiny demands when the government treats
individuals differently on the basis of race); Vera, 517 U.S. at
978 ("Strict scrutiny remains, nonetheless, strict."); see,
e.g., Cooper, 137 S. Ct. at 1464; Miller, 515 U.S. at 920-27;
Shaw v. Hunt, 517 U.S. at 916.
47
No. 2021AP1450-OA.akz
¶129 Ultimately, the majority's focus on the parties'
positions is a tactic of distraction. The majority may
understand that it lacks sufficient evidence to support race-
driven maps proposed by the Governor, so to compensate, it turns
around and reasons that the Governor's maps cannot be rejected
with what it views as inadequate argument on the part of the
Legislature and other parties. But this merely begs the
question: why is the court adopting a racially motivated map
without support in the record? The majority does not cite a
single case standing for the proposition that a state action can
survive strict scrutiny by pointing to the fact that other
private, non-state actors did not present evidence or arguments
in favor of a constitutional course of action. Under the
majority's logic, could the Legislature, when it passes maps at
the next redistricting cycle, draw districts on the basis of
race, without evidence supporting the application of the VRA, by
simply allowing third-party stakeholders an opportunity to
object? The majority's reasoning is foreign to constitutional
jurisprudence.
¶130 The majority also cites a prior Wisconsin federal
court decision that adopted districts in the 1990s with majority
BVAP. Prosser v. Elections Bd., 793 F. Supp. 859 (W.D. Wis.
1992); see majority op., ¶45. That decision did not analyze the
Gingles factors, the history of electoral success for African-
American preferred candidates, or the totality of the
circumstances, as is required to prove the need for a VRA
remedy. Cooper, 137 S. Ct. at 1471-72; Bartlett, 556 U.S. at
48
No. 2021AP1450-OA.akz
19-20; LULAC, 548 U.S. at 432; Gonzalez, 535 F.3d at 600;
Clarke, 40 F.3d at 812-13. It was also issued prior to almost
every major United States Supreme Court precedent on the VRA,
for example: Shaw v. Reno, Shaw v. Hunt, Johnson v. De Grandy,
Miller v. Johnson, Bush v. Vera, League of United Latin American
Citizens v. Perry, Bartlett v. Strickland, and Cooper v. Harris.
Nonetheless, the contention that a decision from the 1990s on
conditions warranting a race-based remedy supports the same
remedy today is similar to asserting that a race-based remedy in
Michigan warrants the same in Wisconsin. Both theories are
antithetical to a proper VRA analysis. The circumstances of the
actual individuals on the ground today, in their specific
communities, is what drives a VRA review, not assumptions
derived from how other individuals of the same race were treated
at different times, in different places, and under different
circumstances. Cooper, 137 S. Ct. at 1471-72; LULAC, 548 U.S.
at 432; Shaw v. Hunt, 517 U.S. at 917; City of Euclid, 580 F.
Supp. 2d at 604-12; Comm. for Fair & Balanced Map, 835 F. Supp.
2d at 583. No caselaw is cited for the proposition that
"historical practice," relied upon by the majority, can either
support race-based district lines or satisfy strict scrutiny.
Majority op., ¶45. Surely, many governments in the past would
have relied on such an argument to support racially motivated
policies and practices.
¶131 History is littered with racial animus, hostility,
discrimination, and desperate treatment. The Equal Protection
Clause demands that governments in the United States rise above
49
No. 2021AP1450-OA.akz
the human temptation of dividing by race and treat individuals
how basic dignity demands they be treated: as individuals.
Only in specific cases, with exacting and quantifiable
information, and with narrowly targeted remedies, may government
discard equal protection guarantees. Fisher, 570 U.S. at 309-
10; Miller, 515 U.S. at 911-12, 922; Shaw v. Reno, 509 U.S. at
653. Lowering the bar for equal protection and allowing it to
be ignored without extraordinary evidence, and relying primarily
on conclusory analysis and a court's subjective observations,
would mark a material turn for equal protection jurisprudence
and an unwelcome departure from foundational American
principles. See majority op., ¶¶43-49 (relying heavily on party
concessions, incomplete evidence, and an out of context standard
of "good reasons" to justify unambiguous racial
classifications). If that path is followed, a Pandora's box of
racial grouping, jealousy, division, and animosity may open more
fully. And we all may look back in regret at the day equal
protection was made into an insubstantial and secondary
interest.
¶132 Given the serious lack of evidence supporting the need
to draw districts as explicitly based on race as is done by the
Governor, this court should abide by its constitutional duty to
treat all Wisconsinites the same regardless of race. Vera, 517
U.S. at 965-83; Miller, 515 U.S. at 922; Shaw v. Reno, 509 U.S.
at 653; Cooper, 137 S. Ct. at 1464; Fisher, 570 U.S. at 309-10.
The court has no lawful, constitutional basis to adopt any other
50
No. 2021AP1450-OA.akz
maps than the race-neutral, constitutional, least change maps
submitted by the Legislature or, in the alternative, CMS.
B. Least Change Is More Than One Core Retention Number.
¶133 Core retention is the percentage of individuals that
are retained in the same legislative districts as the maps in
existence prior to this lawsuit. Never before oral argument did
we conclude that the core retention number alone was the sole
factor to be considered. In our November 30 opinion, we stated
that "our judicial remedy should reflect the least change
necessary for the maps to comport with relevant legal
requirements." Johnson, 399 Wis. 2d 623, ¶72. We did not limit
the factors and considerations that can be taken into account
when determining whether a map made as little changes as
possible while complying with the law. Certainly, we did not
hold that the map that moves the lowest number of people will be
selected, regardless of any other change or constitutional
consideration. Our majority opinion on November 30 simply never
mentioned that phrase, "core retention." A majority of this
court nonetheless takes a myopic approach and refuses to look
beyond core retention or even evaluate the underpinnings of how
those numbers were achieved. See majority op., ¶24 ("[L]east
change approach should guide our decision" and "[c]ore retention
is central to analysis.").
¶134 Fundamental jurisprudence instructs that the data that
underlies the core retention numbers may be considered, but in
conjunction with other valid considerations such as county and
municipality division and population deviation. Such routine
51
No. 2021AP1450-OA.akz
considerations are valid, as is discussed in caselaw, and more
importantly, they are constitutionally required. The author of
the majority opinion now distances himself from these basic
principles and even his own writing, which explicitly indicated
"traditional redistricting criteria" would be considered.
Johnson, 399 Wis. 2d 623, ¶83 (Hagedorn, J., concurring).19
The majority opinion's author refused to sign onto small
19
parts of the November 30 opinion and wrote a separate
concurrence because, in that Justice's view, the November 30
opinion unduly limited the court's discretion in selecting a new
map. "Legal standards establish the need for a remedy and
constrain the remedies we may impose, but they are not the only
permissible judicial considerations when constructing a proper
remedy," the November 30 concurrence declared triumphantly.
Johnson v. Wis. Elections Comm'n, 2021 WI 87, ¶83, 399
Wis. 2d 623, 967 N.W.2d 469 (Hagedorn, J., concurring). In
fact, there was a specific factor the concurrence gave special
favor to: "one universally recognized redistricting criterion
is communities of interest," i.e., local communities and
governments. Id. (Hagedorn, J., concurring). The concurrence
contemplated reliance on this factor when multiple maps were
comparable on the issue of least change:
Suppose we receive multiple proposed maps that comply
with all relevant legal requirements, and that have
equally compelling arguments for why the proposed map
most aligns with current district boundaries. In that
circumstance, we still must exercise judgment to
choose the best alternative. Considering communities
of interest (or other traditional redistricting
criteria) may assist us in doing so.
Id. (Hagedorn, J., concurring).
Despite the urge to make this apparently principled opinion
known in a concurrence, the same logic is absent in the majority
opinion. Not only does the opinion cast as insignificant basic
constitutional interests in maintaining local government
boundaries, but it also adopts maps with substantially greater
divisions of communities of interests, all the while having
immaterial differences on the (now controlling) least-change
metric of core retention. Time changes all things, but
presumably not that quickly.
52
No. 2021AP1450-OA.akz
¶135 Now, four of my colleagues inexplicably adopt core
retention as the sole factor even though the phrase cannot be
found in the November 30 majority or concurring opinions. This
comes out of thin air and much to the surprise of three members
of the court. While the Governor retains 85.8% of individuals
in their existing districts, the Legislature retains 84.2%, a
1.6% difference. However, the Legislature scores better than
the Governor in the senate, moving several thousand less
individuals.20 The Governor moves around 95,000 less people in
the assembly. Thus, overall, combining the figures for the
senate and assembly, the Governor moves less people than the
Legislature, although they are fairly close in measure. By
comparison, CMS has a 61% core retention in the assembly and a
74.3% core retention in the senate.
¶136 One is left to wonder: If the Legislature knew that
core retention was the only criteria to be used, might it have
submitted different maps if given the chance? Recall, all
parties had the benefit of knowing the Legislature's maps before
submitting their own. The Legislature advanced support for maps
The parties in this lawsuit submitted maps under guidance
on what they viewed as the deciding factors for the author of
the November 30 concurrence. It was not an unreasonable
inference that that Justice's vote may decide the outcome of
this case. Yet now that Justice, writing the majority opinion,
claims soft, non-legal factors such as communities of interest
are not of material importance when the court can identify a map
with the lowest core retention. This is a classic example of
shifted standards.
I recognize that the percentages in the senate are very
20
close; with rounding the Governor and the Legislature have a
92.2% core retention in the senate.
53
No. 2021AP1450-OA.akz
passed by the Assembly and Senate in 2021, which all parties
could examine in advance. No such privilege was afforded to the
Legislature vis-a-vis the Governor's maps.
¶137 To be clear, core retention is a useful statistic for
evaluating the amount of changes in a given map, but it cannot
be the only consideration for the court. Our November 30
opinion made clear that any map must not only consider
statistics reflecting the amount of change, but it must do so
while comporting best with other legal interests such as per
capita representation and retaining local communities. Johnson,
399 Wis. 2d 623, ¶¶24-38, 72 (describing legal considerations in
detail). The November 30 opinion made clear that the
constitutional requirements must be met. Id., ¶38 ("In
determining a judicial remedy for malapportionment, we will
ensure preservation of these justiciable and cognizable rights
explicitly protected under the United States Constitution, the
VRA, or Article IV, Sections 3, 4, or 5 of the Wisconsin
Constitution."). We made clear that in remedying any
malapportionment in the existing maps we must not "inadvertently
choose a remedy that solves one constitutional harm while
creating another." Id., ¶34. As explained below, while the
Governor has higher core retention numbers than the Legislature
and CMS, he did so by sacrificing other constitutional
considerations. As we stated in our November 30 opinion, the
law does not countenance such a result.
C. One-Person-One-Vote
54
No. 2021AP1450-OA.akz
¶138 The United States Supreme Court has continuously and
unambiguously reminded us that, in apportioning state
legislative districts, "the overriding objective must be
substantial equality of population among the various districts,
so that the vote of any citizen is approximately equal in weight
to that of any other citizen in the State." Reynolds v. Sims,
377 U.S. 533, 579 (1964); see also Baumgart, 2002 WL 34127471,
at *2 (quoting Connor v. Finch, 431 U.S. 407, 409 (1977)) ("With
respect to reapportionment, population equality is the 'most
elemental requirement of the Equal Protection Clause.'"). The
Constitution "does not permit a State to relegate considerations
of equality to secondary status and reserve as the primary goal
of apportionment the service of some other state interest."
Mahan v. Howell, 410 U.S. 315, 340, modified, 411 U.S. 922(1973) (Brennan, J., concurring in part).
¶139 The United States Supreme Court, recognizing the
interests of federalism and respect for state sovereignty, has
acknowledged that "some leeway in the equal-population
requirement should be afforded States in devising their
legislative reapportionment plans . . . [and that] when state
legislative districts are at issue we have held that minor
population deviations do not establish a prima facie
constitutional violation." Chapman v. Meier, 420 U.S. 1, 23
(1975). Likewise, the Court has explained that "the
Constitution permits 'such minor deviations only as may occur in
recognizing certain factors that are free from any taint of
arbitrariness or discrimination.'" Swann v. Adams, 385 U.S.
55No. 2021AP1450-OA.akz
440, 444 (1967) (quoting Roman v. Sincock, 377 U.S. 695, 710
(1964)). The State of Wisconsin has an independent requirement
of population equality. Article IV, Section 3 of the Wisconsin
Constitution states that new maps must be
"apportion[ed] . . . according to the number of inhabitants."
¶140 In analyzing the deviation and the extent to which
minor deviations are acceptable under the United States
Constitution, courts follow a two-step process. The first step
is to calculate the ideal population. 81A C.J.S. States § 140.
This is done through simple math: population of the state
divided by the number of applicable districts. Once the ideal
population is calculated, it is then possible to determine the
extent to which a given district population deviates from the
ideal. Id. There is not a mathematical formula extracted from
the Equal Protection Clause establishing "what range of
percentage deviations is permissible, and what is not." Mahan,
410 U.S. at 329.
¶141 While we do know that "[c]ourt-enacted maps are held
to a higher standard . . . the Supreme Court has not explained
how much higher." Essex v. Kobach, 874 F. Supp. 2d 1069, 1082
(D. Kan. 2012) (citing Connor, 431 U.S. at 414). District
courts around the country have generally sought to adopt maps
that, at most, include a 2% deviation. See, e.g., Colleton
Cnty. Council v. McConnell, 201 F. Supp. 2d 618, 655 (D.S.C.
2002).
¶142 However, while courts have attempted to reach at most
2% population deviation when drawing maps, this does not mean
56
No. 2021AP1450-OA.akz
that courts reach it and then quit. The continual goal of courts
when drawing maps is minimizing population disparities. In
Smith v. Cobb Cnty. Bd. of Elections & Registrations, the United
States District Court for the Northern District of Georgia was
tasked with drawing the maps for Cobb County, Georgia. 314
F. Supp. 2d 1274 (N.D. Ga. 2002). Like other courts, it
declared that the "most important goal in fashioning this
remedial plan was to minimize the population deviations among
the four districts . . . ." Id. at 1300. Among the plans
presented to it by the parties was a plan that kept population
deviation at 1.77%. Id. However, in following its declared
goal, the court still redrew the maps itself and ended with a
population deviation of 1.51%. Id. at 1302.
¶143 Further, the State of Wisconsin has an independent
requirement of population equality. Article IV, Section 3 of
the Wisconsin Constitution states that new maps must be
"apportion[ed] . . . according to the number of inhabitants."
Federal courts, respecting the independent sovereignty of
states, have permitted greater deviations than what would be
permitted for congressional districts. But that does not imply
that the Wisconsin Constitution does not place independent
demands on Wisconsin's own legislative districts. Chapman, 420
U.S. at 23. Notably, while the demands of population equality
under the United States Constitution are based on the Equal
Protection Clause, the demands under the Wisconsin Constitution
are derived from Article IV, Section 3 on the apportionment of
districts, not equal protection. See Evenwel v. Abbott, 578
57
No. 2021AP1450-OA.akz
U.S. 54, 58-61 (2016) (describing the different legal standards
for state and federal districts under the United States
Constitution). When the federal government interprets and
applies its own apportionment clause in Article I, Section 2 of
the United States Constitution, it demands "as close to perfect
equality as possible," with little leniency for excess
deviation. Id.¶144 In line with these principles, the November 30 opinion
stated that the population deviation should be "as close an
approximation to exactness as possible" under the Wisconsin
Constitution. Johnson, 399 Wis. 2d 623, ¶28 (quotations
omitted). Minimizing population deviation as much as
practicable has been established for over a century in Wisconsin
and at least since State ex rel. Attorney General v. Cunningham,
81 Wis. 440, 484, 51 N.W. 724 (1892).
¶145 In Wisconsin, federal courts have played a role in
drawing the legislative maps for the past three redistricting
cycles. The federal courts' determinations came only after the
Wisconsin Supreme Court chose not to take up the issue. The
federal courts recognize redistricting is our responsibility, if
the legislative and executive branches fail. Nonetheless, each
time, the federal panel has stated that population equality
remained its chief goal and adopted plans as consummate with
that goal as practicable. See Wis. State AFL-CIO v. Elections
Bd., 543 F. Supp. 630, 637 (E.D. Wis. 1982) (describing that
their plan with a population deviation of 1.74% exemplifies the
"condition that, in a representative form of government, the
58
No. 2021AP1450-OA.akz
vote of each person be, to the extent reasonably possible, equal
in weight to the vote of another"); Prosser, 793 F. Supp. at 866(stating that "[b]elow 1 percent, there are no legally or
politically relevant degrees of perfection," and adopting a map
with deviation of 0.52 percent); Baumgart, 2002 WL 34127471, at
*7 (detailing that the court's "attempt to keep population
deviation between districts as low as possible" yielded a
deviation of 1.48%). Last cycle, in 2011, the Legislature
enacted a map with a "maximum deviation for assembly districts
[of] 0.76% and 0.62% for senate districts." Baldus v. Members
of Wis. Gov't Accountability Bd., 849 F. Supp. 2d 840, 851 (E.D.
Wis. 2012). The existing levels of deviation, by surviving the
constitutional and political processes, are a useful basis for
comparison when evaluating the deviations proposed in the
respective maps. Our November 30 opinion stated that the
population deviation should be "as close an approximation to
exactness as possible" under the Wisconsin Constitution.
Johnson, 399 Wis. 2d 623, ¶28 (quotations omitted); see also
Cunningham, 81 Wis. at 484.
¶146 With this law in hand, the Governor's maps that have
been adopted by a majority of this court are highly concerning.
They contain some of the largest deviations from one-person-one-
vote that were presented to us: 1.883% for the assembly
districts and 1.179% for the senate districts, over double the
deviations adopted in the prior maps. Apparently, to the
majority, this dramatic departure from the existing maps is not
relevant to the least change inquiry. Meanwhile, the
59
No. 2021AP1450-OA.akz
Legislature (0.759% for the assembly districts and 0.574 for the
senate districts) and CMS (0.736% for the assembly districts and
0.501% for the senate districts) have substantially lower
population deviations.
¶147 It is clear from the comparisons between the 2011
maps, historically adopted maps, and the maps proposed by the
parties, the Governor failed to heed the instructions this court
gave in Cunningham and repeated in its November 30 opinion.
While the Governor keeps population deviations below a largely
arbitrary line of 2 percent, this is by no means the end of the
analysis. See Cunningham, 81 Wis. at 484; Cobb Cnty., 314
F. Supp. 2d at 1300-02. The Governor fails to provide any
explanation for why his maps have over double the magnitude in
population distortions compared to the 2011 maps other than
vaguely asserting compliance with "least change." Notably, the
Legislature was able to design maps with almost the same core
retention, while also keeping deviation orders of magnitude
lower. The Legislature's effort is proof positive that the
Governor's population deviations among districts were entirely
unnecessary. Given advanced software, there is little doubt
that if the Governor were not striving for other goals, based at
least in part on race and likely in large part on politics, his
core retention could have remained the same while lowering
population deviations. But while political considerations are
not included in the constitution, population equality is. See
Johnson, 399 Wis. 2d 623, ¶53 (explaining that partisanship is
60
No. 2021AP1450-OA.akz
not a legally recognized interest found in the Wisconsin or
United States Constitutions).
¶148 The court's interest is in making populations "as
nearly as [equal] as possible," and thus, the court should adopt
either the Legislature's map or CMS's map. Abrams v. Johnson,
521 U.S. 74, 98-99 (1997); Johnson, 399 Wis. 2d 623, ¶28. The
population deviations included in the Governor's maps allow him
to inflate his core retention numbers, undercut the
Legislature's numbers, and assert he has provided the least
change maps. In the process, however, he ignored interests
recognized in both the United States and Wisconsin Constitutions
that individuals should have as close to equal influence in
elections as possible. We should embrace this foundational
democratic value, not just explain it away.21
D. Dividing Local Communities
¶149 Under Article IV, Section 4 of the Wisconsin
Constitution, assembly districts must be drawn "to be bounded by
county, precinct, town or ward lines." As we explained in our
November 30 opinion:
Applying the one person, one vote principle may make
bounding districts by county lines nearly impossible.
See Wis. State AFL-CIO v. Elec. Bd., 543 F. Supp. 630,
635 (E.D. Wis. 1982) (stating the maintenance of
county lines is "incompatib[le] with population
equality"); see also 58 Wis. Att'y Gen. Op. 88, 91
(1969) ("[T]he Wisconsin Constitution no longer may be
21Particularly if we adopted the approach endorsed by the
November 30 concurrence, whereby the court may consider
"traditional redistricting criteria" when selecting between two
least-change maps. Johnson, 399 Wis. 2d 623, ¶83 (Hagedorn, J.,
concurring).
61
No. 2021AP1450-OA.akz
considered as prohibiting assembly districts from
crossing county lines, in view of the emphasis the
United States Supreme Court has placed upon population
equality in electoral districts."). Nonetheless, the
smaller the political subdivision, the easier it may
be to preserve its boundaries. See Baumgart v.
Wendelberger, No. 01-C-0121, 2002 WL 34127471, at *3
(E.D. Wis. May 30, 2002) ("Although avoiding the
division of counties is no longer an inviolable
principle, respect for the prerogatives of the
Wisconsin Constitution dictate that wards and
municipalities be kept whole where possible.").
Johnson, 399 Wis. 2d 623, ¶35.
¶150 Courts have recognized for many years that this
provision serves to protect local communities, which are central
features of individual identity for voters and are the building
blocks of Wisconsin's democracy. State ex rel. Reynolds v.
Zimmerman, 22 Wis. 2d 544, 555, 126 N.W.2d 551 (1964)
(explaining that the primary goal of "per capita equality of
representation" must still comply with the Wisconsin
Constitution's "geographical limitations" under Article IV,
Section 4); Jensen v. Wis. Elections Bd., 2002 WI 13, ¶6 n.3,
249 Wis. 2d 706, 639 N.W.2d 537 (explaining that the Wisconsin
Constitution demands "respect for municipal boundaries");
Baumgart, 2002 WL 34127471, at *3 (stating that in redistricting
after the 1980 and 1990 censuses, conducted in federal court,
the courts "did not divide any wards in their respective
reapportionment plans, and the 1992 panel rejected a proposed
plan that achieved 0% population deviation by splitting wards");
Prosser, 793 F. Supp. at 863 ("To be an effective
representative, a legislator must represent a district that has
a reasonable homogeneity of needs and interests; otherwise the
62
No. 2021AP1450-OA.akz
policies he supports will not represent the preferences of most
of his constituents.").
¶151 Given this constitutional interest in preserving
communities of interest and local governments, it is not
surprising that the Legislature, when it drew the existing maps
in 2011, sought to limit the amount of county and municipal
splits. The Legislature in 2011 permitted 46 county splits in
its senate map and 58 county splits in its assembly map. It
created 48 municipal splits in the senate and 78 municipal
splits in the assembly. Although the number of municipal splits
increased over time as local governments changed size and
annexed new areas, it is clear from past practice that the state
has strived to minimize divisions of local communities.
¶152 The Governor, and the majority who adopted his maps,
do not seem to care. Without detailed explanation, they divide
an inordinate number of local communities. In the adopted map,
they included 42 county splits in the senate and 53 in the
assembly. There were 117 municipal splits in the senate and 175
in the assembly, and they split 179 wards in the senate and 258
in the assembly. See Prosser, 793 F. Supp. at 866 (explaining
that wards are "the basic unit of Wisconsin state government for
voting purposes . . . [y]ou vote by ward"). On January 10,
2022, we permitted the Governor to amend his map, and he used
the opportunity to reduce local government divisions. However,
according to the Governor's own numbers, he still retained 76
municipal splits in the senate and 115 in the assembly. Like
population deviation, the Governor's stark departure from
63
No. 2021AP1450-OA.akz
standards for local government divisions used to draw the
existing maps is of little concern to the majority's least
change analysis. Only core retention is considered by the
majority.
¶153 My colleagues on the other side devalue these
extraordinary divisions concluding that they are of no
consequence. I disagree because local changes at polling places
are of great significance to those affected and are deserving of
consideration. For people living in Brookfield, Glendale, and
De Pere, their communities are now divided. Their neighbors
sharing common interests, government, and organizations must
seek representation from different officials representing
different constituencies across unique geographies. Many
Wisconsinites may no longer engage in the most fundamental form
of democratic engagement: discussing and deliberating shared
election choices with those having similar interests and
identities. Although division of local governments may appear
to be simply a number, it most assuredly is not. It is a
constitutional requirement, not some policy choice. Wis. Const.
art. IV, § 4.
¶154 With the adoption of the Governor's maps, local
communities are the losers. The majority finds this of no
consequence, yet Wisconsin is made up of few large cities and
many local municipalities. Dividing a town or a county in
localities of hundreds of thousands of inhabitants may not be
noticeable by all those residents; however, that is not true for
the many, many small communities around the state. In accepting
64
No. 2021AP1450-OA.akz
the Governor's maps, the majority opinion chooses to favor the
big city interests over more rural identities.
¶155 By contrast, the maps offered by the Legislature and
CMS keep divisions of local communities to a minimum. The
Legislature has comparable county splits to the Governor, with
42 county splits in the senate and 53 splits in the assembly.
CMS outperforms all parties in this metric by including 28
county splits in the senate and 40 in the assembly. Where the
parties diverge substantially is in municipalities. The
Legislature includes a striking low number of municipal splits
with 28 in the senate and 48 in the assembly.22 CMS, by
comparison, has 31 municipal splits in the senate and 70 in the
assembly. Finally, while the Governor demonstrated little to no
concern for ward lines, both the Legislature and CMS divided
zero current ward boundaries. Given the minimal difference in
core retention between the Legislature and the Governor, and the
obvious technical ability to limit local government divisions,
the Legislature's and CMS's maps provides powerful evidence that
the drastic number of local government splits made by the
Governor's maps were entirely unnecessary and represented
significant change. If those drawing the Governor's maps were
not so motivated by race and politics, perhaps they may have
considered the Wisconsin Constitution.
¶156 Further, if my colleagues would consider
constitutional mandates as more than a policy choice, they would
Among municipalities, the Governor split 50 towns.
22 The
Legislature, by contrast, split only 16. At the time the 2011
maps were passed, they contained 30 town splits.
65
No. 2021AP1450-OA.akz
be required to conclude that the Governor's maps are not
constitutionally compliant. In addition, they would be forced
to recognize that the core retention figures of their preferred
maps are artificially inflated at the expense of the people and
their local communities. Nonetheless, the majority proceeds to
adopt the Governor's maps, carving up Wisconsin communities for
the stated and unstated interests of the Governor.
¶157 Both the Legislature and CMS demonstrated that
mapmakers could have minimized the changes to existing maps
while still respecting in large respect the boundaries by which
Wisconsinites organize themselves at the local level. While,
under existing one-person-one-vote jurisprudence from the United
States Supreme Court, local government boundaries cannot be
retained in full, that in no way implies that local government
divisions are of no concern to this court, as the majority
appears to believe. Johnson, 399 Wis. 2d 623, ¶35. In our
November 30 opinion, we reaffirmed decades of caselaw that the
citizens of Wisconsin have a constitutionally protected interest
in "preserv[ing] [local government] boundaries." Johnson, 399
Wis. 2d 623, ¶35 (noting "respect for the prerogatives of the
Wisconsin Constitution dictate that wards and municipalities be
kept whole where possible"); Reynolds, 22 Wis. 2d at 555;
Jensen, 249 Wis. 2d 706, ¶6 n.3; Baumgart, 2002 WL 34127471, at
*3; Prosser, 793 F. Supp. at 863.
¶158 The Legislature and CMS took our directives and
constitutional demands seriously. The Governor did not. In
adopting the Governor's maps through its fixation on core
66
No. 2021AP1450-OA.akz
retention, the majority turns a blind eye to the constitution's
clear call to consider these boundary line changes.
III. CONGRESSIONAL MAPS
¶159 Only four parties submitted congressional maps: the
Congressmen; the Governor; Hunter; and CMS. The Governor's map
is unconstitutional under the Equal Protection Clause, and the
court should adopt the Congressmen's map, or in the alternative,
CMS's map.
A. Least Change
¶160 As explained in the analysis on state maps, least
change is not defined by a single statistic. Johnson, 399
Wis. 2d 623, ¶72. Nowhere in the November 30, 2021 opinion did
we hold that core retention is the sole determinant of a least
change inquiry. Id.¶161 Among other factors and considerations, core retention
can be a useful statistic to consider. Here, the Governor has
the highest core retention with 94.5%. The Congressmen come in
second with 93.5%, followed by Hunter at 93% and CMS at 91.5%.
Thus, the Governor moves around 50,000 fewer people than the
Congressmen.
¶162 Of note, however, the Congressmen attempted to
introduce an amended map, which would have had the lowest core
retention of any maps. Given the extraordinary importance of
this case, and the need to fairly consider all positions and
evidence presented by the parties, the court should have no
issue accepting such requests. Our duty is to consider how best
to redistrict, and more information is better than less.
67
No. 2021AP1450-OA.akz
¶163 The Congressmen's amended map moved almost 100,000
fewer people than the Governor's map. Furthermore, both the
Governor and BLOC were permitted to amend their maps, mostly to
reduce their local government splits and make their maps more
attractive for the court to adopt. Nonetheless, the court, in a
January 10, 2022 order, chose not to consider the second map
submitted by the Congressmen. Johnson v. Wis. Elections Comm'n,
No. 2021AP1450-OA, unpublished order (Wis. Jan. 10, 2022). Due
to this ruling, only the first map submitted by the Congressmen
is reviewed. However, the majority is not relegated to adopting
only one party's map. It is endowed with the authority to draw
the best map, yet it failed to do so.
¶164 Even though the majority is purportedly driven by the
single statistic of core retention, it apparently is not
concerned enough to seek out or adopt the map that scored best
on that metric. The court, post argument, regularly allows
supplemental submissions. We did in this case. If there ever
was a case to ensure that we have the best possible information
at our disposal, this is it. Curiously, a majority of the court
does not want it.
B. One-Person-One-Vote
¶165 The Governor's map cannot be accepted because he has
an unnecessary and unexplained deviation from perfect population
equality. Population equality for congressional districts is
governed by Article I, Section 2 of the United States
Constitution, not the Equal Protection Clause. Evenwel, 578
U.S. at 58-61.
68
No. 2021AP1450-OA.akz
¶166 In our November 30 opinion, we quoted the United
States Supreme Court in declaring that, "[There is] no excuse
for the failure to meet the objective of equal representation
for equal numbers of people in congressional districting other
than the practical impossibility of drawing equal districts with
mathematical precision." Johnson, 399 Wis. 2d 623, ¶25 (quoting
Mahan, 410 U.S. at 322). "[P]opulation alone" is the "sole
criterion of constitutionality in congressional redistricting
under Art. I, § 2[.]" Id. CMS aptly argues that the Governor's
congressional map should not pass scrutiny because it "fail[s]
to satisfy even this fundamental requirement [by exhibiting]
more than the mathematical minimum population deviation between
districts."
¶167 The Supreme Court, in recognizing that a zero
deviation will not always be possible, gave the following
instructions for evaluating a plan that varies from the
precision of mathematical equality:
First, the court must consider whether the population
differences among districts could have been reduced or
eliminated altogether by a good-faith effort to draw
districts of equal population. Parties challenging
apportionment legislation must bear the burden of
proof on this issue, and if they fail to show that the
differences could have been avoided the apportionment
scheme must be upheld. If, however, the plaintiffs can
establish that the population differences were not the
result of a good-faith effort to achieve equality, the
State must bear the burden of proving that each
significant variance between districts was necessary
to achieve some legitimate goal.
Karcher v. Daggett, 462 U.S. 725, 730–31 (1983). The court
further reaffirmed that "there are no de minimis population
69
No. 2021AP1450-OA.akz
variations," so long as those variations can "practicably be
avoided." Id. at 734.
¶168 A useful example of this burden shifting mechanism can
be found in Larios v. Cox, 300 F. Supp. 2d 1320 (N.D. Ga.),
aff'd, 542 U.S. 947 (2004). In Larios, a three-judge panel
heard several challenges to the congressional and state
legislative reapportionment plans enacted by the Georgia General
Assembly in 2001 and 2002. Id. at 1321. In the relevant
portion of the opinion, the panel examined whether the
plaintiff's challenge to the congressional maps enacted by the
state legislature complied with the United States Constitution's
one-person-one-vote requirement. "[T]he total population
deviation for the [legislature's] final Congressional Plan was
only seventy-two people." Larios, 300 F. Supp. 2d at 1354. At
the trial, expert testimony concluded that:
[I]t would be possible to draw a congressional map for
the State of Georgia with a population deviation of
plus or minus one person that (1) complied with the
Voting Rights Act; (2) split fewer counties than the
present plan; (3) is more compact than the present
plan; and (4) divides fewer voting precincts than the
present plan.
Id. at 1354.
¶169 Under the Karcher framework, the panel reasoned that
"[t]he fact that such a plan could have been produced all but
invalidates any argument that the [legislature] made a good
faith effort to achieve a zero deviation." Larios, 300
F. Supp. 2d at 1354 (citing Karcher, 462 U.S. at 736). On this
basis, the panel determined that the plaintiffs had met their
burden and that the burden was now put on the Legislature to
70
No. 2021AP1450-OA.akz
show that a "consistently applied legislative policy" justified
the deviation. Id. The State of Georgia contended that
it did not further reduce the population deviation
because to do so would have required either splitting
more precincts [which Georgia has a history of not
doing] or further splitting existing split precincts
along something other than an easily recognizable
boundary [as doing so would make it hard for voters
and election officials to accurately ascertain which
voting district they reside].
Id. Additionally, the court found that, although the plaintiffs
showed that the population deviation could be remedied, they did
not prove that it could be done without splitting precincts
along something other than recognizable boundary lines. Id. at
1355. Therefore, the panel found that "[g]iven the relatively
small total deviation of only seventy-two people and the
importance of the state's interest in avoiding voter confusion,
we find that the congressional districts do not violate
plaintiffs' rights under the one-person, one-vote principles of
Art. I, § 2." Id.
¶170 In this case, the Legislature and CMS can point to the
fact that their maps have a mathematically precise population
deviation as a means of invalidating any argument that the
Governor made a good-faith effort to achieve zero deviation.
Therefore, the burden of explaining what "consistently applied"
state policy justifies the larger than minimum population
deviation falls on the Governor.
¶171 Rather than address this deviation, the Governor
denies that it exists. The Governor's population deviation is
two. Population deviation (taken as a range of deviation) is
71
No. 2021AP1450-OA.akz
determined by taking the Governor's maximum deviation above the
ideal (one person) and adding it to the Governor's minimum
deviation below the ideal (one person). See Evenwel, 578 U.S.
at 59 (explaining that population deviation, when conducting a
population equality analysis, is calculated by a comparison
"between the largest and smallest district"). 1 + 1 = 2.
However, the Governor, in his briefs, asserts that his deviation
is the same as the Congressmen's: one person. This assertion
stems from the incorrect, semantic wordplay of his expert who,
in her initial report, calculated that "[t]he largest deviation
is 1 person, with all districts ranging from 1 person below to 1
person above the ideal population." The "largest" difference
between the average population may be one person, but that is
not the relevant statistic. Population deviation is the
difference between the smallest and largest district.
Importantly, this range of deviation is later acknowledged in
the Governor's expert report.
¶172 Despite this burden and the need to explain why his
districts have greater than necessary population inequality, the
Governor at oral argument stated a population deviation of two
was included because the Governor did not believe a lower
population deviation was required under law. No explanation or
details were provided as to why the deviation was necessary,
applying reasonable priorities such as "making districts
compact, respecting municipal boundaries, preserving the cores
of prior districts, and avoiding contests between incumbent
Representatives." Karcher, 462 U.S. at 740. As explained
72
No. 2021AP1450-OA.akz
above, the United States Constitution requires exactness of
population absent the "practical impossibility of drawing equal
districts with mathematical precision." Johnson, 399
Wis. 2d 623, ¶25. Both CMS and the Congressmen showed a lower
population deviation could be done, and they too achieved high
core retention.
¶173 Given advanced software technology and the immense
financial resources put to use in this litigation, it was
abundantly possible for the Governor to achieve a deviation of
one while retaining the same least change characteristics, such
as core retention. Due to a misunderstanding of law, and
misstatement of the definition of population deviation, the
Governor overlooked the driving consideration of drawing
congressional districts "with populations as close to perfect
equality as possible." Evenwel, 578 U.S. at 59. But
carelessness cannot satisfy the Governor's burden of proving
"with some specificity that the population differences were
necessary to achieve some legitimate state objective." Tennant
v. Jefferson Cnty. Comm'n, 567 U.S. 758, 760, 763-65 (2012) (per
curium) (quotations omitted) (holding that a congressional map
in West Virginia was legal where the state justified its
deviations by pointing to protection of local communities,
limiting incumbent pairings, and reducing change in district
lines). By contrast, the Governor's deviation was not the
result of "a good-faith effort to achieve absolute equality" and
is thus insufficient. Id. (quoting Karcher, 462 U.S. at 730).
73
No. 2021AP1450-OA.akz
¶174 The majority picks sides and litigates for the
Governor, claiming that the two person deviation was necessary
for least change. See majority op., ¶24 ("[The Governor's]
minor population deviation is justified under Supreme Court
precedent by our least change objective.") This is a whitewash:
the Governor admitted that a lower deviation could be done
without issue, but permitted a deviation of two because he did
not believe a lower deviation was necessary. Neither the
Governor nor any other party argued that a deviation of two
individuals was required to ensure a least change map.
Furthermore, it is facially preposterous: with advanced
computer technology, the Governor could have readily reduced his
population deviation while maintaining his core retention.
Simply put, the Governor failed to present a "legitimate state
objective" for his unnecessary deviation. Tennant, 567 U.S. at
760; see also Karcher, 462 U.S. at 730–31 ("[T]here are no de
minimis population variations.")
¶175 Only the Congressmen's map and CMS's map should be
considered by this court. The Congressmen have higher core
retention than CMS and should be adopted. Nonetheless, CMS
offers a reasonable alternative. The Governor's maps are
fatally and constitutionally flawed. The majority errs in
adopting them.
IV. CONCLUSION
¶176 For the foregoing reasons, I respectfully dissent.
¶177 I am authorized to state that Justices PATIENCE DRAKE
ROGGENSACK and REBECCA GRASSL BRADLEY join this dissent.
74
No. 2021AP1450-OA.pdr
¶178 PATIENCE DRAKE ROGGENSACK, J. (dissenting). The
2020 census shows that Wisconsin's growth in population requires
reapportionment of its congressional and state legislative
districts. Reapportionment presents a three dimensional puzzle,
each piece of which has statutory and constitutional
requirements. I write to address one error of Governor Evers's
map reapportioning Wisconsin's Assembly Districts, which four
members of this court have adopted. In Wisconsin's single
member districts, the Assembly map conflicts with the Voting
Rights Act of 1965, formerly set out in 42 U.S.C. § 1973, now
within 52 U.S.C. § 10301. In adopting the Governor's map, a
majority of this court engages in racial gerrymandering contrary
to the Equal Protection Clause of the Fourteenth Amendment of
the United States Constitution, which prohibits separating
voters into different voting districts based on the race of the
voter. Bethune-Hill v. Virginia State Bd. of Elections, 580
U.S. __, __, 137 S. Ct. 788, 797 (2017). It is my hope that the
United States Supreme Court will be asked to review Wisconsin's
unwarranted racial gerrymander, which clearly does not survive
strict scrutiny.
¶179 The United States Constitution requires that
apportionment be as equal as practicable because population
disparity in voting districts for the same legislative body
dilutes the power of some voters. Concerns about voter
inequality have been the foundation of the Supreme Court's one-
person-one-vote decisions. Reynolds v. Sims, 377 U.S. 533, 558
1
No. 2021AP1450-OA.pdr
(1964) (explaining that the concept of voter equality "can mean
only one thing——one person, one vote").
¶180 The Supreme Court has required near mathematical
equality for congressional maps. Abrams v. Johnson, 521 U.S.
74, 98 (1997). Somewhat more leeway is given when drawing
boundaries for state legislative districts. Evenwel v. Abbott,
578 U.S. 54, 59 (2016). However, court-drawn maps are held to a
more exacting standard of population equality than are
legislatively drawn maps. Abrams, 521 U.S. at 98.
¶181 The Voting Rights Act prohibits any standard, practice
or procedure that results in denial or abridgement of the right
to vote on account of race. 52 U.S.C. § 10301(a); Cooper v.
Harris, 137 S. Ct. 1455, 1464 (2017).1 Subsection (b) provides
the required examination for assessing whether race is
precluding equal opportunity for a protected class:
A violation of subsection (a) is established if, based
on the totality of circumstances, it is shown that the
political processes leading to nomination or election
in the State or political subdivision are not equally
open to participation by members of a class of
citizens protected by subsection (a) in that its
members have less opportunity than other members of
the electorate to participate in the political process
and to elect representatives of their choice. The
extent to which members of a protected class have been
elected to office in the State or political
subdivision is one circumstance which may be
considered: Provided, [t]hat nothing in this section
establishes a right to have members of a protected
The provisions of 52 U.S.C. § 10301 have been referred to
1
as Section 2 of the Voting Rights Act of 1965 subsequent to the
1982 amendment. See Johnson v. De Grandy, 512 U.S. 997, 1009-10
(1994).
2
No. 2021AP1450-OA.pdr
class elected in numbers equal to their proportion in
the population.
§ 10301(b).
¶182 Over the years, the Supreme Court has addressed the
Voting Rights Act in decisions that explain how it is to be
applied in various contexts. Thornburg v. Gingles, 478 U.S. 30(1986), is the seminal Supreme Court case that sets the
analytical framework that is required when the Voting Rights Act
is addressed.2 Gingles establishes all three threshold
"prerequisites" that must be affirmatively proved before further
consideration of a claim of, or potential remedy for, a
violation of § 2 of the Voting Rights Act can be addressed in
reapportionment. First, there must be proof that a minority
group is "sufficiently large and geographically compact to
constitute a majority [in a single-member district];" second,
the minority group must be "politically cohesive"; and third,
the "white majority [] vote[ed] sufficiently as a bloc to
[enable it] usually [to] defeat the minority's preferred
candidate." Cooper, 137 S. Ct. at 1470 (citing Gingles, 478
U.S. at 51); Johnson v. De Grandy, 512 U.S. 997, 1009-10 (1994)
(also citing Gingles, 478 U.S. at 51).
¶183 Cooper is particularly helpful in its instructions
about how to employ the Gingles "prerequisites." Cooper sets
out the "three threshold conditions" for proving voter dilution3
2Thornburg v. Gingles, 478 U.S. 30 (1986) arose in a
challenge to multi-member districts. Its analysis has been
applied to single-member district challenges as well.
De Grandy, 512 U.S. at 1000.
Voter dilution, a violation of § 2 of the Voting Rights
3
Act, may occur when a cohesive minority group is fragmented
3
No. 2021AP1450-OA.pdr
and then explains that these showings are needed to establish
that racially polarized voting prevents the minority group's
choice in the district as actually drawn because the minority
group is submerged in a larger white voting population. Cooper,
137 S. Ct. at 1470.
¶184 In determining whether the third Gingles
"prerequisite" was met, the Court reviewed the success of black
candidates in past elections. Id. It noted that in North
Carolina, where Cooper arose, "electoral history provided no
evidence that a § 2 plaintiff could demonstrate the third
Gingles prerequisite——effective white bloc-voting." Id. The
Supreme Court in Cooper concluded that when an elective district
"functioned, election year in and election year out, as a
'crossover' district, in which members of the majority help a
'large enough' minority to elect its candidate of
choice . . . it is difficult to see how the majority-bloc-voting
requirement could be met——and hence how § 2 liability could be
established." Id. (citing Bartlett v. Strickland, 556 U.S. 1,
13, 16 (2009)).
¶185 The three Gingles prerequisites are factual conditions
that must be proved in order to establish the first step of a
claim under § 2 of the Voting Rights Act. All three
preconditions must be met before considerations of race could
lawfully affect drawing district boundaries. As the Supreme
Court has explained, "In a § 2 case, only when a party has
among several districts or packed into too few districts. Id.
at 1002.
4
No. 2021AP1450-OA.pdr
established the Gingles requirements does a court proceed to
analyze whether a violation has occurred based on the totality
of the circumstances." Bartlett, 556 U.S. at 11-12. However,
to escape the parties' failure to establish the Gingles
requirements, the majority resorts to protesting that "no party
saw fit to develop an argument" that the Gingles requirements
were not satisfied.4 Nevertheless, if we permit this abdication
to form the basis of the law of the State of Wisconsin, the
results in this case will effect an unconstitutional, racially
gerrymandered map. Our judgments are precedents, and the proper
interpretation of the law as it relates to these judgments
cannot simply be left to the parties. Young v. United States,
315 U.S. 257, 259 (1942). Instead, as this state's highest
court, it is our duty to ensure the proper interpretation of the
law.
¶186 Milwaukee is Wisconsin's only county that has a
sufficiently large and geographically compact black population
of voters that could meet the Gingles preconditions. The black
voters of Milwaukee do vote cohesively for candidates of their
choice. However, Milwaukee's history for at least the last ten
years is that of crossover voting where white voters help black
voters elect candidates of their choice.
¶187 Notwithstanding the Supreme Court's clear
instructions, the majority opinion ignores the historical record
of black voters choosing candidates of their choice and assigns
voters based solely on their race to create seven majority-
4 Majority op., ¶45.
5
No. 2021AP1450-OA.pdr
minority voting assembly districts in Milwaukee County. The
Supreme Court "has made clear that unless each of the three
Gingles prerequisites is established, 'there neither has been a
wrong nor can be a remedy.'" Cooper, 137 S. Ct. at 1472
(quoting Growe v. Emison, 507 U.S. 25, 41 (1993) (emphasis in
Cooper)). The Supreme Court in Cooper struck down North
Carolina's racial gerrymander "whose necessity is supported by
no evidence and whose raison d'etre is a legal mistake."
Cooper, 137 S. Ct. at 1472.
¶188 The map adopted by the majority opinion violates the
Voting Rights Act for the same reason as North Carolina's choice
did in Cooper. Factually, Wisconsin has had significant
experience with electing black candidates through white
crossover voting.
¶189 For example, in 2016, Gwen Moore, a black
congresswoman, was elected to Congressional District 4, which
has only 33.3% black residents. However, she received 76.74% of
the vote.5 She was reelected in 2018 with 75.61% of the vote;
and reelected in 2020 for a third time with 74.65%. That her
vote totals exceed the percentage of black residents in her
district evidences that white voters have crossed over to
support her elections.
The record of votes achieved by black candidates comes
5
from state public records of election outcomes and are therefore
"capable of accurate and ready determination by resort to
sources whose accuracy cannot reasonably be questioned." Wis.
Stat. § 902.01.
6
No. 2021AP1450-OA.pdr
¶190 Mandela Barnes, a black state-wide candidate, is
another example of white crossover voting. In 2018, Mandela
Barnes was elected over a white primary opponent for Lieutenant
Governor with 67.86% of the vote.6
¶191 David Clarke, a black county-wide candidate, provides
repetitive examples of white crossover voting. Clarke was
elected Milwaukee County Sheriff in 2006 with 77.85% of the
vote; reelected in 2010 with 80.42% and reelected again in 2014
with 79.12% of the vote. Each time he was elected with the
assistance of white crossover voting, as shown by his percentage
victories that are well above the black resident percentage of
Milwaukee County.7 White crossover voting also helped elect
David Crowley, a black candidate, as the Milwaukee County
Executive in 2020. He formerly held a position in Wisconsin's
Assembly.
¶192 Since 2012, Lena Taylor, a black state senator, has
been elected repeatedly to Senate District 4 with vote totals
showing white voter support. For example, in 2012, Lena Taylor
obtained 86.6% of the vote; in 2016 she obtained 98.33% of the
vote; and in 2020, she obtained 98.34% of the vote. 61.7% of
the residents of Senate District 4 are black.
¶193 La Tonya Johnson, a black state senator, has been
elected repeatedly to public office with vote totals showing
6 Wisconsin's black population of voting age is
approximately 6.4%.
7 Approximately 26% of Milwaukee County's residents are
black.
7
No. 2021AP1450-OA.pdr
support from white voters. For example, in 2014, she was
elected to Assembly District 17 with 87.25% of the vote, and in
2016 she was elected to Senate District 6 with 98.89% of the
vote. 65.4% of the residents of Assembly district 17 are black
and 62.1% of Senate District 6 are black residents. Leon Young,
a black assemblyman was elected to Assembly District 16 in 2014,
unopposed.8 In 2014, Jason Fields, a black assemblyman, was
elected to Assembly District 11, unopposed.9
¶194 The majority opinion ignores Milwaukee County's
historical record of white crossover voting that has provided
repeated support for black candidates during at least the last
ten years. The majority opinion does so in order to create
seven majority-minority districts in Milwaukee County. In so
doing, the majority opinion comes squarely within the
prohibition that assigning voters to voting districts by race
violates the Equal Protection Clause of the Fourteenth
Amendment.
¶195 The majority opinion says that it relies on Cooper for
the racial gerrymander that it creates in Milwaukee County. The
majority opinion clearly misunderstands Cooper, which overturned
racial gerrymandering that occurred in North Carolina. Let's
look at Cooper and why the majority opinion fails to follow it.
¶196 Justice Kagan begins her discussion in Cooper with the
Equal Protection Clause of the Fourteenth Amendment, which she
explains, "limits racial gerrymanders in legislative districting
8 61.5% of the residents of Assembly District 16 are black.
9 63.7% of the residents of Assembly District 11 are black.
8
No. 2021AP1450-OA.pdr
plans." Id. at 1463. As Justice Kagan explained, the Equal
Protection Clause "prevents a State, in the absence of
'sufficient justification,' from 'separating its citizens into
different voting districts on the basis of race.'" Id. (quoting
Bethune–Hill, 137 S. Ct. at 797). When allocation of voters by
race has occurred, that allocation must withstand strict
scrutiny such that the State must prove "its race-based sorting
of voters serves a 'compelling interest' and is 'narrowly
tailored' to that end." Id. at 1464.
¶197 In order to meet the narrow tailoring for the racial
assignment of voters, the State must establish by factual proofs
that it had "good reasons" to believe that the Voting Rights Act
would be violated if voters were not assigned based on their
race. Id. Cooper explained what it means by "good reasons"
sufficient to satisfy strict scrutiny. First, Cooper emphasized
that the "good reason" to which it referred was factual proof of
"good reason to think that all the 'Gingles preconditions' are
met, then so too it has good reason to believe that § 2 requires
drawing a majority-minority district. . . . But if not, then
not." Id. at 1470. Second, as the Supreme Court said as it
examined factual evidence, "[h]ere, electoral history provided
no evidence that a § 2 plaintiff could demonstrate the third
Gingles prerequisite——effective white bloc-voting." Id.
¶198 It is Cooper's "good reason" phrase that the majority
opinion picked up as its foundation for assigning voters to
districts based on race. The majority said, "we conclude there
are good reasons to believe a seventh majority-Black district is
9
No. 2021AP1450-OA.pdr
needed to satisfy the VRA."10 It did so without understanding
that factual proofs of the Gingles preconditions are necessary
before it could satisfy "good reason" for assigning voters by
race in districting.
¶199 The majority showed how limited its understanding of
Cooper is by its dismissive treatment of Cooper's requirement to
factually prove the three Gingles preconditions.11 Factual proof
is exactly what "good reasons" requires and what the majority
lacks as it contravenes the Equal Protection Clause by assigning
voters to districts based on their race. As Cooper carefully
explained, there must be proof of effective white bloc-voting
that prevents the minority's ability to elect the candidate of
its choice before a § 2 violation can arise. Id.
¶200 As the factual evidence above showed, black voters in
Milwaukee are able to elect candidates of their choice, election
year in and election year out, for congresswoman, state
senators, state assembly persons, sheriff and Milwaukee County
Executive to name only a few. Just as in North Carolina in
Cooper, proof of the third Gingles precondition to § 2 liability
is absent from the majority opinion. The Voting Rights Act is
violated by the majority opinion just as it was by the State of
North Carolina in Cooper.
¶201 It is beyond dispute that the Governor's districting
plan adopted by a majority of this court assigns voters to
districts based on race. "Racial classifications are
10 Majority op., ¶10.
11 Id., ¶45.
10
No. 2021AP1450-OA.pdr
antithetical to the Fourteenth Amendment, whose 'central
purpose' was 'to eliminate racial discrimination emanating from
official sources in the States.'" Shaw v. Hunt, 517 U.S. 899,
907 (1996). Such an assignment violates the Equal Protection
Clause of the Fourteenth Amendment unless the racial assignment
serves a compelling state interest and is narrowly tailored to
meet that interest. Cooper, 137 S. Ct. at 1464.
¶202 Just as it ignores the lack of factual proof for the
three Gingles preconditions, the majority opinion identifies no
compelling state interest to which its racial gerrymander is
narrowly tailored. Instead, it asserts that if a seventh black
majority district were not drawn, a § 2 violation may occur, but
it "cannot say for certain on this record."12
¶203 To justify its weak position, the majority cites to
the black population of Wisconsin increasing and the white
state-wide population decreasing in the last ten years, both by
less than five percent.13 However, the majority does not
identify whether any of that population change occurred in
Milwaukee County; or whether if it occurred in Milwaukee County,
it occurred in the area of Milwaukee County where the majority
opinion creates a seventh black majority district.
¶204 This is not a small error because the means chosen to
accomplish a race-based purpose "must be specifically and
narrowly framed to accomplish that purpose." Shaw, 517 U.S. at
908. To meet that standard, the racial assignment of voters
12 Id., ¶47.
13 Id., ¶48.
11
No. 2021AP1450-OA.pdr
must be remedial to the specific location of the compelling
state interest identified. Id. at 915.
¶205 However, just as in Shaw, the seventh district that
the majority creates is not remedial to correcting an identified
compelling state interest. Stated otherwise, creation of a
seventh district in one area of Milwaukee County is not a
narrowly tailored remedy for a population change for the entire
State of Wisconsin, which the majority asserts as justification
for creating the seventh district. The creation of the seventh
black majority district in Milwaukee County cannot survive
strict scrutiny.
¶206 Accordingly, because proof of meeting the third
Gingles precondition has not been provided, as is required
before voters may be assigned to voting district by race, and
because the seventh black majority district does not survive a
strict scrutiny inquiry, the majority errs, and I respectfully
dissent.
¶207 I am authorized to state that Chief Justice ANNETTE
KINGSLAND ZIEGLER and Justice REBECCA GRASSL BRADLEY join this
dissent.
12
No. 2021AP1450-OA.rgb
¶208 REBECCA GRASSL BRADLEY, J. (dissenting).
[H]e who would place the supreme power in mind, would
place it in God and the laws; but he who entrusts man
with it, gives it to a wild beast, for such his
appetites sometimes make him; for passion influences
those who are in power, even the very best of men:
for which reason law is reason without desire.
Aristotle, A Treatise on Government Bk. III, ch. XVI (William
Ellis trans., 1912) (circa 384–22 B.C.),
https://www.gutenberg.org/files/6762/6762-h/6762-
h.htm#link2H_INTR.
¶209 Just three months ago, we said this court "will
confine any judicial remedy to making the minimum changes
necessary in order to conform the existing congressional and
state legislative redistricting plans to constitutional and
1
No. 2021AP1450-OA.rgb
statutory requirements."1 Johnson v. Wis. Elections Comm'n, 2021
WI 87, ¶8, 399 Wis. 2d 623, 967 N.W.2d 469. Now, the majority
overrides the United States Constitution, the Wisconsin
Constitution, and federal statutory law in favor of a policy
1 In a deceptive caricature of our November 30, 2021
opinion, Justice Ann Walsh Bradley (joined by two other
justices) claims "'least change,' as set forth in the court's
prior order, is unmoored from any legal requirement for
redistricting. The parties struggled with reconciling it with
the United States Constitution, Wisconsin Constitution, and
Voting Rights Act." Concurrence, ¶58. Although in this opinion
the new majority indeed untethers the least-change approach from
the law, in this court's November 30 opinion (not an "order"),
we consistently defined "least change" to mean "making only
those changes necessary for the maps to comport with the one
person, one vote principle while satisfying other constitutional
and statutory mandates." Johnson v. Wis. Elections Comm'n, 2021
WI 87, ¶8, 399 Wis. 2d 623, 967 N.W.2d 469; see also id., ¶¶4,
8, 64, 72, 81. Although the majority corrupts the least-change
approach by "unmoor[ing]" it from the law——treating a single
measure of least change, core retention, as an extra-legal
criterion taking precedence over the law——that is not the way we
described it three months ago. Any "struggle[]" to
"reconcil[e]" the least-change approach with the law stems not
from our "prior order" but from a misapplication of the least-
change approach that allows core retention (an extra-legal
criterion) to override the United States Constitution, the
Wisconsin Constitution, and the VRA. Contrary to the
concurrence's disingenuous description, we never said core
retention was a "metric" that would carry any weight, let alone
"more weight than others." Concurrence, ¶59. We never told the
parties that core retention was "preeminent," id., ¶63; we told
them to submit maps that made only those changes necessary to
comply with the law. Although three justices in the majority
believe core retention plays far too great a role in the
majority's analysis, they join it anyway, then lament about it
in a separate writing. Despite six justices agreeing core
retention should not be the sole governing criterion in this
case, a majority nevertheless selects the Governor's maps
ostensibly on this basis. Contrary to the concurrence, nothing
in our November 30 opinion compels this; properly applied, our
November 30 opinion stands in opposition to the majority's
decision.
2
No. 2021AP1450-OA.rgb
goal it deems "commendable"2——"core retention"3——a phrase
appearing nowhere in either our November 30, 2021 opinion nor
even in Justice Hagedorn's concurrence to that opinion (which no
one joined). Elevating their subjective policy preferences over
the law, members of the majority abandon a remedy for
malapportionment grounded in the law and instead entangle
themselves in legislative (and therefore blatantly political)
policymaking by choosing maps based upon what the majority deems
"best,"4 justified by what the majority determines are "good
reasons,"5 and using criteria the majority deems "helpful."6
¶210 In doing so, the majority flouts not only this court's
precedent but the constitutional separation of powers. "Because
the judiciary lacks the lawmaking power constitutionally
conferred on the legislature" we promised to "limit our remedy
to achieving compliance with the law rather than imposing policy
choices." Id. The majority now reneges on that promise,
relegating constitutional mandates to "policy choices" that may
be protected or disregarded at the whim of the majority of this
court.7 The majority's decision represents a startling departure
2 Majority op., ¶35.
3 Id., ¶¶7–8, 13 & n.9, 14–15, 22, 24, 26–30, 33.
4 Id., ¶6.
5 Id., ¶45.
Id., ¶13.
6 The majority is most transparent about its
"involvement" in making "numerous policy and political
decisions," see id., ¶4, thereby abandoning its neutral role.
7 Id., ¶35.
3
No. 2021AP1450-OA.rgb
from the rule of law and an alarming affront to the people of
Wisconsin who elected us to uphold the constitutions.
¶211 The majority's dispositive guidepost——core retention——
exists nowhere in the United States Constitution, the Wisconsin
Constitution or any statutory law. Absent from the law, it does
not appear in our November 30 opinion among the purely legal
criteria we directed the parties to employ in proposing maps.
Nevertheless, the majority belatedly invokes core retention as
justification for its preferred maps, allowing an extra-legal
criterion to take precedence over the Equal Protection Clause,
the Voting Rights Act (VRA), and Article IV——the "exclusive
repository" of "the standards under the Wisconsin Constitution
that govern redistricting." Id., ¶63. "It is 'the province and
duty of the judicial department to say what the law is[,]' and
not what we think it should be." Town of Wilson v. City of
Sheboygan, 2020 WI 16, ¶51, 390 Wis. 2d 266, 938 N.W.2d 493(Rebecca Grassl Bradley, J., concurring) (quoting Marbury v.
Madison, 5 U.S. (Cranch) 137, 177 (1803)) (modification in the
original). Instead of following the law this court declared
just three months ago, the majority instead adopts maps based on
its subjective policy preferences, fulfilling the fears of many
citizens concerned about a judicially-partisan outcome.
¶212 Remedying unconstitutional malapportionment——
inequality in the number of citizens in each legislative or
4
No. 2021AP1450-OA.rgb
congressional district——was this court's sole task in this case,8
and would not have been a particularly challenging one, if the
majority had confined itself to applying the law. The majority
flunks every constitutional test by adopting maps that are not
even remedial, exhibiting avoidable population inequality (in
violation of Article IV, Section 3 of the Wisconsin
Constitution, Article 1, Section 2 of the United States
Constitution, and the Equal Protection Clause) and excessive
county, town, and ward splits (in violation of Article IV,
Section 4 of the Wisconsin Constitution).
¶213 For over a century, this court has required "as close
an approximation to exactness as possible" in apportioning
population by legislative districts under the Wisconsin
Constitution. State ex rel. Attorney General v. Cunningham, 81
Wis. 440, 484, 51 N.W. 724 (1892). The only justification for
deviating from exactness is compliance with other constitutional
requirements (mainly, Section 4). State ex rel. Lamb v.
Cunningham, 83 Wis. 90, 150, 53 N.W. 35 (1892). Similarly,
nearly fifty years ago the United States Supreme Court declared
there is "no excuse for the failure to meet the objective of
equal representation for equal numbers of people in
congressional districting other than the practical impossibility
8 The entire point of this proceeding was to
"remedy . . . malapportionment, while ensuring the maps satisfy
all other constitutional and statutory requirements." Johnson,
399 Wis. 2d 623, ¶4. Instead, the majority overrides the
constitutional command of one person, one vote because
"population deviation is not an indicator of least change."
Majority op., ¶32 n.18. The constitution is not expendable at
the majority's caprice.
5
No. 2021AP1450-OA.rgb
of drawing equal districts with mathematical precision." Mahan
v. Howell, 410 U.S. 315, 322 (1973) (emphasis added). The
majority conveniently does not address these precedents other
than to pay lip service to them.
¶214 Irrefutably, the majority could have adopted maps with
practically perfect population equality; the Citizen
Mathematicians and Scientists drew such maps. Not only does the
majority adopt an assembly map and a congressional map with
unconstitutional population deviations, it also inflicts a
constitutional harm not present in the 2011 maps by severing the
boundaries of numerous local communities with no lawful
justification for doing so. The Governor did not sacrifice
population equality to preserve local communities, so his
population deviation is unjustifiable and therefore
unconstitutional.
¶215 If all of these constitutional failings weren't enough
to disqualify the Governor's maps, their constitutionally
impermissible dilution of the Black vote in Milwaukee County
should be. In Johnson v. De Grandy, the United States Supreme
Court rejected the "rule of thumb apparently adopted by the
District Court" in that case (and by the majority in this case)
"that anything short of the maximum number of majority-minority
districts consistent with the Gingles conditions would violate
§ 2 [of the VRA]" as "caus[ing] its own dangers, and they are
not to be courted." 512 U.S. 997, 1016 (1994). Expanding the
number of Black opportunity districts to seven may on the
surface appear to augment Black voting strength, but in reality
6
No. 2021AP1450-OA.rgb
it jeopardizes the effectiveness of each district by spreading
the population too thin,9 with each of the Governor's opportunity
districts hovering just above or just below 50%.10
¶216 I also write to address an issue with recurring
significance beyond redistricting. Justice Hagedorn's November
30 concurring opinion——which no one joined——is not the
"controlling" opinion of this court.11 Setting aside Justice
Hagedorn's departure from his November 30 position in announcing
new views as the majority author at this late stage of the case,
his November 30 concurrence was simply that and the majority
opinion controls the issues presented. The apparent confusion
9 Some elected officials characterized plans to reduce the
Black voting-age population percentages in Milwaukee as part of
"a national effort to dilute minority communities to create more
Democratic seats." See, e.g., Assembly Floor Session, at
2:18:05 (Nov. 11, 2021) (statement of Rep. Sylvia Ortiz-Velez
(AD8)), https://wiseye.org/2021/11/11/wisconsin-state-assembly-
floor-session-42.
10 The parties present slightly different ways of measuring
Black voting-age population. According to the Legislature, this
population includes "non-Hispanic Black" and "non-Hispanic
(Black + White)." Legislature's Resp. Br., at 22. The
Legislature omits other "multi-race subcategories[.]" Id. In
contrast, other parties, including BLOC, ask that these
subcategories be included. BLOC's Reply Br., at 8 n.1. If the
goal is to draw seven majority-minority districts (which the
majority suggests is the case), this definitional dispute is
critical. In fact, according to the Legislature's definition,
none of the Governor's seven supposedly VRA-mandated Black
opportunity districts are above 50.0% (although one is exactly
50.0%). Legislature's Resp. Br., at 22.
11 The Hunter Intervenor-Petitioners expressly labelled
Justice Hagedorn's concurrence "controlling[.]" Hunter
Intervenor-Petitioners' Resp. Br., at 6. A number of other
parties treated it as controlling without giving it that label.
7
No. 2021AP1450-OA.rgb
caused by his concurrence derailed the case presentations of
several parties.
¶217 To prevent the court's policy-driven mapmaking in the
future, the next time this court resolves a redistricting
dispute it should consider withdrawing language from State ex
rel. Reynolds v. Zimmerman, which prohibited the Legislature
from implementing state legislative redistricting plans by joint
resolution. 22 Wis. 2d 544, 569–70, 126 N.W.2d 551 (1964).
That precedent should be revisited because it does not comport
with the constitutional text, which assigns the Legislature
alone the responsibility of redistricting. The Legislature
suggested this court may need to revisit Zimmerman, depending on
how it decided to proceed in this case.12 This issue is worthy
of the court's attention.
¶218 As a final matter, in the interest of ensuring
procedural due process, this court should have allowed all
parties to submit substantive modifications to their proposed
remedial maps. The majority disingenuously states, "we invited
all parties to this litigation to submit one proposed map for
each set of districts[.]"13 True, we asked each party to submit
only "one" set of proposed remedial maps; however, we permitted
the Governor and BLOC to make critical changes that went well
beyond correcting drafting errors. For example, the Governor
Legislature's 10/26/21 Br., at 20-22 ("Zimmerman is on
12
shaky ground in light of the language of . . . Article IV, § 3
and historical context.").
13 Majority op., ¶4 (emphasis added).
8
No. 2021AP1450-OA.rgb
originally proposed a remedial assembly map that split 80 towns,
but his modified map splits 50, a reduction of nearly 40%.14 The
Congressmen asked to submit a modified map, but the same
majority that now adopts the Governor's modified maps denied the
Congressmen this opportunity.15 Instead, the majority
inexplicably rushes to select the Governor's unlawful maps,
eschewing reasoned law for its own desires. I dissent.
I. THE MAJORITY'S REMEDY VIOLATES THE CONSTITUTIONS
¶219 The majority guts state constitutional mandates. In
our November 30th opinion, we outlined the "discrete
requirements" of Article IV, Sections 3 and 4. Johnson, 399
Wis. 2d 623, ¶63. Section 3 requires state legislative
districts to be drawn "according to the number of inhabitants."
Section 4 requires assembly districts "to be bounded by county,
precinct, town, or ward lines[.]"16 We declared these sections
"explicitly protect[] . . . justiciable and cognizable rights,"17
14Johnson v. WEC, No. 2021AP1450-OA, unpublished order, at
3 (Wis. Jan. 10, 2022) (Roggensack, J., dissenting).
15 Id.16 In one of this court's seminal cases on redistricting,
Chief Justice Lyon explained a precinct was a form of local
government that ceased to exist when a part of Article IV of the
Wisconsin Constitution became fully operative. State ex rel.
Attorney General v. Cunningham, 81 Wis. 440, 520, 51 N.W. 724(1892) (Lyon, C.J., concurring) ("[T]he precinct of the
constitution disappeared when the uniform system of town and
county government prescribed, by the constitution (art. 4, sec.
23) became fully operative. We have now no civil subdivisions,
other than towns and wards, which are the equivalent of the
precinct of territorial times."). Under Article IV, "precinct"
does not mean election precinct.
17 Johnson, 399 Wis. 2d 623, ¶38.
9
No. 2021AP1450-OA.rgb
dedicated eleven paragraphs to expounding how these sections are
satisfied,18 and repeatedly promised Wisconsinites we would
uphold these sections when selecting remedial state legislative
maps.19 The majority in this opinion reverses course, treating
Sections 3 and 4 as mere hortative statements with no operative
effect. The majority goes so far as to suggest Section 4 may
not even be a commendable policy goal——at least, not as
commendable as core retention.20 Despite the constitutional
command, the majority actually frowns upon minimizing the number
of county, town, and ward splits to the extent such an effort
produces more change from prior maps than the majority deems
acceptable.21 Least change is an approach designed to minimize
changes to predecessor maps, but it should go without saying
that the court must in all respects comply with the law. The
Wisconsin Constitution is the supreme law of this state, which
all members of this court swore an oath to uphold. The people
of Wisconsin should be alarmed at the majority's dismissiveness
toward the constitution.
18 Id., ¶¶28–38.
19 Id., ¶¶8, 34, 38, 81. Justice Hagedorn agreed without
reservation, writing in his solo concurrence, "remedial maps
must comply with . . . Article IV, Sections 3, 4, and 5 of the
Wisconsin Constitution[.]" Id., ¶82 n.4 (Hagedorn, J.,
concurring).
Majority op., ¶32 ("[T]he Legislature argues that we
20
should weigh as a measure of least change the number of counties
and municipalities split under each proposal. We fail to see
why this is a relevant least-change metric, however.").
Id. ("If a municipality was split under the maps adopted
21
in 2011, reuniting that municipality now——laudable though it may
be——would produce more change, not less.").
10
No. 2021AP1450-OA.rgb
¶220 In 1892, this court rejected the majority's current
construction of Article IV, Sections 3 and 4 as mere
recommendations for being a "dangerous doctrine," which "should
not be encouraged even to the extent of discussing the question"
because "[t]he convention, in making the constitution, had a
higher duty to perform than to give . . . advice." Cunningham,
81 Wis. at 485. It expressly held, "the restrictions on the
power . . . to make an apportionment, found in sections 3[]
[and] 4 . . . are mandatory and imperative, and are not subject
to . . . discretion[.]" Id. at 486. Later that same year, this
court declared the requirements of these sections are
"absolutely binding" and even the Legislature has "no
power . . . to dispense with any one of them." Lamb, 83 Wis. at
148. The majority now endorses this "dangerous doctrine,"
effectively overruling the Wisconsin Constitution. The majority
barely mentions Cunningham or Lamb, despite implicitly
withdrawing language from both seminal decisions.
¶221 The majority's departure from precedent is, indeed,
dangerous. Wisconsin's founders knew political actors would act
politically.22 They did not impose a partisan fairness
requirement on the redistricting process, Johnson, 399
Wis. 2d 623, ¶¶53–63, because telling partisans in the
Legislature not to act for partisan advantage would have been
like ordering water to be dry. Cf. The Law and Policy of
Redistricting Reform, Fed. Soc'y, at 1:06:20 (Apr. 26, 2019),
Gerrymandering was a common practice by 1840.
22 Rucho v.
Common Cause, 588 U.S. __, 139 S. Ct. 2484, 2495 (2019)
(citation omitted).
11
No. 2021AP1450-OA.rgb
https://www.youtube.com/watch?v=nOi-BEo8ZFc&t=1618s (statement
of Larry Obhof). The founders did, however, impose the
requirements of Article IV, Sections 3 and 4 to limit the extent
to which one party could take control of the state by
gerrymandering.23 Cunningham, 81 Wis. at 486.
¶222 The majority assures future political actors they can
adopt state legislative redistricting plans with population
deviation nearing 2% that cannot be justified by a good-faith
attempt to preserve political boundaries. For comparison, the
assembly map passed by the Legislature and signed by the
Governor in 2011 had a population deviation of 0.76%. Baldus v.
Members of Wis. Government Accountability Bd., 849 F. Supp.
2d 840, 851 (E.D. Wis. 2012). Instead of mentioning this
feature of the 2011 map, the majority resorts to a
legislatively-drawn map from the 1970s that purportedly had a 2%
population deviation.24 Every assembly map drawn by a federal
court in the history of Wisconsin has had a lower population
deviation than the map the majority adopts. Baumgart v.
Wendelberger, No. 01-C-0121, 2002 WL 34127471, at *7 (E.D. Wis.
May 30, 2002) (1.48%); Prosser v. Elections Bd., 793 F. Supp.
859, 866 (W.D. Wis. 1992) (0.52%); Wis. State AFL-CIO v.
They also adopted Article IV, Section 5, which states, in
23
relevant part, "no assembly district shall be divided in the
formation of a senate district." No one has ever treated
Section 5 as anything less than an absolute constitutional
requirement. Not a single assembly district is divided in the
formation of any senate district in any proposed remedial plan
submitted to this court.
24 Majority op., ¶36.
12
No. 2021AP1450-OA.rgb
Elections Bd., 543 F. Supp. 630, 637 (E.D. Wis. 1982) (1.74%).
The majority's assurances that "the Governor's maps are
consistent with . . . court-sanctioned requirements
for . . . population equality"25 is simply false. This court has
never recognized a safe harbor for population deviation——until
now. H. Rupert Theobald, Equal Representation: A Study of
Legislative and Congressional Apportionment in Wisconsin, in
Wisconsin Blue Book 71, 72 (1970) ("The Wisconsin
Constitution has, since 1848, required districts 'according to
the number of inhabitants', and it does not recognize a 'minimal
deviation' which could be disregarded.").26
¶223 A 2% automatic safe harbor is quite the gift to
political actors, affording them unprecedented map-drawing
discretion. Although all but one member of the current majority
25 Id.According to the majority, this court has never required
26
less population deviation than is present in the maps it adopts.
Id., ¶36 n.20. However, this court has not decided a
redistricting case since the rise of the one person, one vote
principle. Even before the United States Supreme Court
established the primacy of this principle in the 1960s, this
court never recognized any sort of safe harbor, below which maps
are per se constitutional. Instead, it has always examined
whether other constitutional criteria (not extra-legal criteria
such as core retention) justify the population deviation.
Neither the Governor nor the majority has pointed to any such
criteria as justification. The question is not whether "better
performance on population deviation is . . . possible." Id. As
the majority acknowledges, it "certainly" is. Id. The question
is whether any legal rationale supports the deviation the
majority asserts is permissible——not just for the Governor's
maps but for any map. There isn't any. Under controlling
precedent, population deviation cannot be judged in isolation,
without consideration of all other constitutional criteria.
13
No. 2021AP1450-OA.rgb
decried the 2011 maps as "sharply partisan,"27 they now embrace a
tool for promoting partisan gerrymanders.28 When a partisan
gerrymander coexists with population inequality, a subset of the
people become more politically powerful than the rest of the
population, raising serious concerns that the people, as a
whole, have lost control over their own government. Minimizing
population deviation is the key limitation on partisan
gerrymandering, as evidenced by England's "infamous rotten
boroughs." Johnson, 399 Wis. 2d 623, ¶30 (citing The Federalist
No. 56, at 349 (James Madison) (Clinton Rossiter ed., 1961)).
¶224 The constraints on the Legislature's redistricting
power are "very simple and brief;" undermining any one of them
grants the body significantly more leeway than the constitution
permits. Id., ¶58 (quoting Cunningham, 81 Wis. at 511 (Pinney,
J., concurring)). While this court is bound by the least-change
approach, the Legislature is not. At any time, the Legislature
and the Governor may implement redistricting plans through the
political process, which would supplant this court's remedy.29
Id., ¶19 (majority opinion) (quoting State ex rel. Reynolds v.
Zimmerman, 23 Wis. 2d 606, 606, 128 N.W.2d 16 (1964) (per
27Johnson, 399 Wis. 2d 623, ¶¶88, 106 (Dallet, J.,
dissenting) (citations omitted).
Of course, notwithstanding a partisan gerrymander, when
28
map drawers comply with the constitutional command to achieve
population equality, "[v]oters retain their freedom to choose
among candidates irrespective of how district lines are drawn."
Id., ¶55 (majority opinion) (citation omitted).
Majority op., ¶52 ("This order shall remain in effect
29
until new maps are enacted into law or a court otherwise
directs.").
14
No. 2021AP1450-OA.rgb
curiam)). Under the majority's new redistricting paradigm, one
side of the political aisle may be politically obliterated, much
like the words "according to the number of inhabitants" under
the majority's atextual interpretation. The majority's opinion
is a wolf that does not even try to masquerade as a sheep. See
Morrison v. Olson, 487 U.S. 654, 699 (1988) (Scalia, J.,
dissenting).
¶225 The majority rationalizes constitutionally
impermissible population inequality by declaring "the Governor's
maps are consistent with historical practice and court-
sanctioned requirements for compactness, respect for local
boundaries, and population equality."30 So much for the
constitution. The majority points to maps this court approved
long ago, with substantial population inequality, which the
majority proclaims constitutes a baseline by which to measure
proposed remedial maps in this case. The majority's reliance on
cases predating the primacy placed by the United States Supreme
Court on population equality undermines its analysis entirely.
¶226 In Cunningham and Lamb, this court explained that
Article IV, Sections 3 and 4 exist in tension. While Section 3
requires population equality, Section 4 renders political
boundary lines inviolable——specifically, the lines dividing
counties, towns, and wards. Grouping people into perfectly
equal districts while respecting political boundaries, in which
unequal populations live, is challenging. In Cunningham and
Lamb, this court gave Sections 3 and 4 near equal weight: "[I]t
30 Id., ¶36.
15
No. 2021AP1450-OA.rgb
is impossible to secure exact and equal representation, by
reason of the constitutional hindrances mentioned [mainly,
Section 4]; and it is because of such hindrances, and only
because of such hindrances, that the legislature, under the
constitution, are at liberty to depart from equality of
representation." Lamb, 83 Wis. at 150 (emphasis added); see
also id. at 155 ("It follows that the constitution requires the
legislature to apportion the state into senate and assembly
districts 'according to the number of inhabitants,' as nearly as
can be done consistently with other provisions of the
constitution mentioned."). In particular, this court prohibited
county splits, at the expense of population equality. Id. at
148 ("It was determined in the former case [Cunningham], and is
now conceded, that no county line is to be broken in the
formation of any assembly district.").
¶227 This court twice reaffirmed Cunningham and Lamb. In
1932, this court declared the Legislature "bound by
constitutional mandate to avoid unnecessary inequalities in
representation;" however, it also noted "it was recognized in
[Cunningham and Lamb] that the Constitution contains other
provisions which militate against absolute equality . . . . For
example, the requirement that the districts be bounded by
county, . . . town, or ward lines[.]" State ex rel. Bownman v.
Dammann, 209 Wis. 21, 27, 243 N.W. 481 (1932).
¶228 A few decades later, this court reiterated that "the
constitution itself commits the state to the principle of per
capita equality of representation subject only to some
16
No. 2021AP1450-OA.rgb
geographic limitations in the execution and administration of
this principle." Zimmerman, 22 Wis. 2d at 556 (emphasis added).
That statement was not a passing remark. This court emphasized
the importance of population equality multiple times:
It is assumed by all parties and understood by this
court that a mathematical equality of population in
each senate and assembly district is impossible to
achieve, given the requirement that the boundaries of
local political units must be considered in the
execution of the standard of per capita equality of
representation.
It is equally clear, however, that a valid
reapportionment 'should be as close an approximation
to exactness as possible, and [that] this is the
utmost limit for the exercise of legislative
discretion.'
. . . .
[T]he legislature must apportion in direct ratio to
population, subject only to (1) practical limitations
in execution of this principle, and (2) precise
constitutional restrictions about observance of
governmental boundaries in drawing district lines.
Id. at 563–66. Until the United States Supreme Court ruled
otherwise, substantial population inequality was permissible,
but it had to be justified almost entirely by the preservation
of political boundaries. Our November 30 opinion stressed the
importance of the principle articulated in Zimmerman, although
we also recognized federal constitutional law uprooted the
balance this court had struck between Article IV, Sections 3 and
4, rendering population equality of paramount importance in
redistricting. Johnson, 399 Wis. 2d 623, ¶¶35, 38 (citations
omitted).
17
No. 2021AP1450-OA.rgb
¶229 Post-Zimmerman, federal constitutional law changed.
No longer may Article IV, Sections 3 and 4 be given
approximately equal weight. In 1964, the United States Supreme
Court held, "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
practicable." Reynolds v. Sims, 377 U.S. 533, 577 (1964). That
same year, the Court confirmed even state senate districts had
to comply with the one person, one vote principle. Maryland
Comm. for Fair Representation v. Tawes, 377 U.S. 656, 674–75
(1964). On the eve of Wisconsin's next redistricting cycle, the
assembly requested an opinion from the attorney general
regarding the application of Sections 3 and 4 in light of these
binding precedents. 58 Wis. Att'y Gen. Op. 88 (1969). The
attorney general responded, "[i]n my opinion, the Wisconsin
Constitution no longer may be considered as prohibiting assembly
districts from crossing county lines, in view of the emphasis
the United States Supreme Court has placed upon population
equality among electoral districts." Id. at 91. In another
opinion two years later, the attorney general explained town and
ward lines still needed to be followed but only "insofar as may
be consistent with population equality[.]" 60 Wis. Att'y Gen.
Op. 101, 106 (1971); see also Michael Gallagher, Joseph Kreye &
Staci Duros, Redistricting in Wisconsin 2020 17 (2020),
https://docs.legis.wisconsin.gov/misc/lrb/wisconsin_elections_pr
oject/redistricting_wisconsin_2020_1_2.pdf (explaining respect
for the unity of political subdivisions is "by no means
18
No. 2021AP1450-OA.rgb
obsolete" but that these boundaries were followed "much more
meticulously in Wisconsin, and elsewhere, before the advent of
one person, one vote"); Theobald, A Study of Legislative and
Congressional Apportionment in Wisconsin, at 72 ("As long as
they do not conflict with the equal population requirements, all
other apportionment provisions of the Wisconsin Constitution
must be given full effect." (emphasis added)). Accordingly,
every proposed remedial map in this case splits substantially
more counties, towns, and wards than would have been permissible
under Cunningham and Lamb.
¶230 Under the original understanding of Article IV,
Section 3, population inequality was permissible only if a
"constitutional hindrance[]," i.e., compliance with another
constitutional requirement, compelled it. Lamb, 83 Wis. at 150.
In Reynolds, the United States Supreme Court changed the
calculation, but the majority nevertheless chooses maps in
accordance with a bad interpretation of bad law, embracing both
population inequality and fractured political boundaries.
¶231 While the truth may be inconvenient for the majority,
pretending Zimmerman sanctions the Governor's maps because the
maps approved in Zimmerman had "substantially larger population
deviations"31 ignores binding precedent of the United States
Supreme Court. The majority relegates the United States Supreme
Court's directive on population equality to a single footnote,
acknowledging "the geographic limitations in the Wisconsin
31 Id.
19
No. 2021AP1450-OA.rgb
Constitution can no longer be fully enforced"32 as a result. The
majority neglects to acknowledge that those "geographic
limitations" in Article IV, Section 4 can no longer justify the
extent of population inequality approved in Zimmerman.
¶232 While federal constitutional law precludes us from
giving perfect effect to Article IV's original meaning, we could
nonetheless achieve population equality while preserving
political boundaries, something the majority makes no attempt to
do. The remedial maps proposed by the Governor, which the
majority adopts as its own, have both greater population
deviation and more splits than the Legislature's proposed
remedial maps. The Governor offers no explanation for his
population deviation other than a passing reference to least
change, despite this court's direction to the parties to be
mindful of both Sections 3 and 4. Specifically, the Governor's
assembly map has more than twice the population deviation of the
Legislature's map (1.88% compared to the Legislature's 0.76%),33
and double the municipal splits (115 compared to the
Legislature's 52),34 and hundreds more ward splits (the
Legislature split zero wards).35 The ward splits are
particularly difficult to justify because "the smaller the
32 Id., n.19 (citing Johnson, 399 Wis. 2d 623, ¶35).
33 Resp. Expert R. Thomas M. Bryan, at 3.
Suppl. R. Supp. Governor Evers's Proposed Corrected State
34
Legislative District Plans, at 5; Expert R. Thomas M. Bryan, at
18.
The Governor and the Legislature split the same number of
35
counties.
20
No. 2021AP1450-OA.rgb
political subdivision, the easier it may be to preserve its
boundaries." Johnson, 399 Wis. 2d 623, ¶35 (citing Baumgart,
2002 WL 34127471, at *3). While one person, one vote
necessitates breaking up counties (large units of people), it
does not necessitate dividing the smallest political units
recognized in the state.
¶233 The Governor argues town splits are relevant but not
village and city splits based on the language of Article IV,
Section 4. His interpretation is consistent with Lamb, 83
Wis. at 148. Even so, he asks this court to split 50 towns by
adopting his proposed remedial assembly map——and the majority
obliges.36 In comparison, the Legislature's map has 52 total
municipal splits, of which only 16 are town splits (the rest are
village and city splits).37 At the time of adoption, the 2011
assembly map split 30 towns.38 A 67% increase in town splits
hardly reflects "least change."
¶234 The majority mischaracterizes the record to justify
the high number of splits. It states:
Particularized data about how many counties or
municipalities remain unified or split may be a useful
indicator of least change. But no party saw fit to
provide that data. What we did receive was raw counts
of total county and municipal slits under each
Suppl. R. Supp. Governor Evers's Proposed Corrected State
36
Legislative District Plans, at 5.
37 Expert R. Thomas M. Bryan, at 18.
See Legislature's Reply Br., at 13 ("How many towns were
38
split by Act 43 is ascertainable by reading the statute,
identifying in text every town split. There were 30[.]"
(citing Wis. Stat. § 4.001, et seq.)).
21
No. 2021AP1450-OA.rgb
proposal, and that information provides no insight
into which map makes the least change to existing
district boundaries.[39]
Problematically, the majority seems to sanction an illegal map——
containing an unlawful number of splits——because the map
performs well on a single extra-legal criterion, core retention.
The majority's approach violates its duty to uphold the
Wisconsin Constitution.40
¶235 Contrary to the majority's assertion, the Legislature
did provide detailed split analyses,41 which it discussed at
length in its response brief. Its expert provided a breakdown
of every county and municipal split in every proposed remedial
map (except for the Governor's modified maps).42 To determine
whether a proposed map retained an existing split or added one
may be tedious, but it is not particularly difficult to
ascertain. The current statutes explicitly state when a split
39 Majority op., ¶32 (second emphasis added).
40Adding together the number of county, town, and ward
splits, the assembly map the majority adopts likely has more
splits than any map ever implemented in this state. While the
majority compares population deviation in its maps with past
maps, it does not endeavor to make analogous comparisons for
splits.
41 Resp. Expert R. Thomas M. Bryan, at App. 2.
This expert report was submitted before the Governor was
42
allowed to modify his maps to reduce the number of splits. The
fact that this court allowed the Governor to modify his maps
while denying other parties the opportunity illustrates the
serious due process problems triggered by the majority's
acceptance of the Governor's modified maps. They have not been
subjected to the same level of adversarial scrutiny as other
maps. The Governor's motion to file modified maps was filed on
January 6, 2022——conveniently, two days after the deadline for
submitting reply briefs and reply expert reports.
22
No. 2021AP1450-OA.rgb
occurs. For example, Wis. Stat. § 4.44(1) declares the 44th
Assembly District includes "[t]hat part of the town of Harmony
comprising U.S. census tract 1202, blocks 3004 and 3095," while
Wis. Stat. § 4.45(1)(a) declares the 45th Assembly District
includes "[t]he towns of Albany, Decatur, Jefferson, Spring
Grove, and Sylvester." By comparing the split analyses to the
existing statutes, the Legislature explained in its response
brief "[t]he Governor would split 7 new municipalities in
Waukesha County's Assembly District 99, including Oconomow[o]c
and Pewaukee. Similarly, the Governor would add 8 municipal
splits in Dane County, including Stoughton and Sue Prairie, even
though not previously split[.]"43
¶236 Adding to its infirmities under the law, the
majority's map effectuates a racial gerrymander. The Governor
admits he drew his proposed remedial assembly map with the
express purpose of creating seven Black majority-minority
assembly districts. Such race-driven redistricting must survive
strict scrutiny. The United States Supreme Court has assumed
compliance with the VRA can be a compelling state interest.
Abbott v. Perez, 138 S. Ct. 2305, 2315 (2018). However, VRA
violations "never can be assumed, but specifically must be
proved in each case in order to establish a redistricting plan
dilutes minority voting strength in violation of § 2 [of the
VRA]." Shaw v. Reno, 509 U.S. 630, 653 (1993). A state must
have "a strong basis in evidence" demonstrating that without
explicit consideration of race, a redistricting plan would
43 Legislature's Resp. Br., at 16.
23
No. 2021AP1450-OA.rgb
transgress the VRA. Cooper v. Harris, 137 S. Ct. 1455, 1464
(2017) (quoting Alabama Legislative Black Caucus v. Alabama, 135
S. Ct. 1257, 1274 (2015)).
¶237 The majority assumes a remedial assembly map with
fewer than seven Black majority-minority districts would violate
the VRA. This assumption is inappropriate, and the Governor has
failed to establish "a strong basis in evidence" for a seventh
district. The majority suggests the VRA requires the drawing of
a seventh Black majority-minority district because Wisconsin's
Black voting-age population approaches seven percent. However,
Section 2 of the VRA declares "That nothing in this section
establishes a right to have members of a protected class elected
in numbers equal to their proportion in the population." 52
U.S.C. § 10301(b). In De Grandy, the United States Supreme
Court held the failure to maximize the number of opportunity
districts is not a VRA violation.44 512 U.S. at 1017.
Opportunity is generally measured, the Court said, against
44 Maximization has been rejected because it carries a
heavy price: "if the number of minority-majority districts is
maximized, then it necessarily follows that black influence is
elsewhere minimized, which reduces the number of districts in
which blacks, fully participating in an integrated process, can
hold the balance of power." In re Apportionment of the State
Legislature—1992, 486 N.W.2d 639, 654 n.66 (1992) (citation
omitted)). In turn, even if Black voters collectively perform
better, a portion of the Black voting population is
"relegate[d]" to the status of "second class . . . wards of the
political/electoral system." Id. Many Black voters object to
their votes being diluted "within . . . their district merely to
secure the chance that . . . their allies in other
districts . . . [are] able to vote more like-minded partisans to
the legislature." Cf. Larry Alexander & Saikrishna B. Prakash,
Tempest in an Empty Teapot: Why the Constitution Does Not
Regulate Gerrymandering, 50 Wm. & Mary L. Rev. 1, 27 (2008).
24
No. 2021AP1450-OA.rgb
"rough" proportionality. See id. at 1000, 1023. The author of
the majority opinion in De Grandy, writing in dissent in another
VRA case, explained:
Several baselines can be imagined; one could, for
example, compare a minority's voting strength under a
particular districting plan with the maximum strength
possible under any alternative. Not surprisingly, we
have conclusively rejected this approach; the VRA was
passed to guarantee minority voters a fair game, not a
killing. See Johnson v. De Grandy, 512 U.S. 997,
1016–1017, 114 S.Ct. 2647, 129 L.Ed.2d 775 (1994). We
have held that the better baseline for measuring
opportunity to elect under § 2, although not
dispositive, is the minority's rough proportion of the
relevant population. Id., at 1013–1023, 114 S.Ct.
2647.
Bartlett v. Strictland, 556 U.S. 1, 29 (2009) (Scouter, J.,
dissenting) (citation omitted). The majority skims over De
Grandy.45
¶238 The Black voting-age population is between 6.1% and
6.5%, as Chief Justice Ziegler explains in her dissent.46
Wisconsin has 99 assembly seats——not 100——so, even taking the
high estimate of 6.5%, the proportional share of Black assembly
districts, rounded to the nearest whole number, would be six,
not seven (99 × 0.065 = 6.4). Accordingly, even if the Gingles
45Some United States Supreme Court justices have been quite
critical of the emphasis placed on proportionality; nonetheless,
it is the law we are bound to follow. Holder v. Hall, 512
U.S. 874, 943–44 (1994) (Thomas, J., concurring) ("Few words
would be too strong to describe the dissembling that pervades
the application of the 'totality of the circumstances' test
under our interpretation of § 2. It is an empty incantation——a
mere conjurer's trick that serves to hide the drive for
proportionality that animates our decisions.").
46 Chief Justice Ziegler's dissent, ¶114.
25
No. 2021AP1450-OA.rgb
preconditions were satisfied, six districts is sufficient to
constitute rough proportionality. See, e.g., Bodker v. Taylor,
No. Civ.A.1:02-CV-999ODE, 2002 WL 32587312, at *8–9 (N.D. Ga.
June 5, 2002) (noting Black people constituted 45.2% of the
population and had only 42.35% of the seats but nonetheless
finding "the court's map conforms with Section 2 of the Voting
Rights Act" because rough "proportional representation" was
achieved and while not "dispositive," proportionality is "strong
evidence" that "minorities have an equal opportunity to
participate" particularly "where there is simply no evidence
before the court about social, historical or other circumstances
that might impact whether minorities in Fulton County are denied
equal opportunity for political participation").47 Justice
Roggensack provides many "good reasons" to believe the
majority's conclusory analysis of the third Gingles precondition
is wanting.
¶239 Rough proportionality is not a safe harbor, but it is
"obviously an indication that minority voters have an equal
opportunity, in spite of racial polarization, 'to participate in
the political process and to elect representatives of their
choice,' 42 U.S.C. § 1973(b)[.]" De Grandy, 512 U.S. at 1020.
47BLOC referred to Bodker in its brief and included a copy
of the opinion in its appendix. It also referred to and
provided a copy of Stenger v. Kellett, No. 4:11CV2230, 2012 WL
601017, at *12 (E.D. Mo. Feb. 23, 2012) ("[B]ecause the African
American 'effective minority' districts are in approximate
proportion to their population of St. Louis County, the plan
would likely not violate the Voting Rights Act even if the
Gingles factors were met, given the totality of the
circumstances in this case.").
26
No. 2021AP1450-OA.rgb
Just like least change is not reflected by a single number, a
proper VRA analysis is not governed by a "single statistic[.]"
Id. Nevertheless, the "central teaching" of De Grandy is clear:
"[P]roportionality . . . is always relevant evidence in
determining vote dilution . . . . Thus, in evaluating . . . the
totality of the circumstances a court must always consider the
relationship between the number of majority-minority voting
districts and the minority group's share of the population."
Id. at 1025 (O'Connor, J., concurring) (citing Thornburg v.
Gingles, 478 U.S. 30, 99 (1986) (O'Connor, J., concurring in
judgment)). The requisite proportionality analysis is missing
from the majority opinion.
¶240 "[E]xplicit race-based districting embarks us on a
most dangerous course." Id. at 1031 (Kennedy, J., concurring in
part and concurring in the judgment). "[R]acial classifications
violate the very essence of the lofty ideals of individual
equality for which this country strives. The concept of racial
classification ought to be repugnant to all Americans." Robert
Redwine, Comment, Constitutional Law: Racial and Political
Gerrymandering——Different Problems Require Different Solutions,
51 Okla. L. Rev. 373, 399 (1996). In the absence of strong
evidence demonstrating a VRA violation will result from the lack
of a seventh district, this court should "unerringly and
unapologetically . . . exalt[] the ideal of individual equality
without regard to race." Id. Exhibiting highly suspect racial
classifications, the majority's remedy violates the Equal
Protection Clause.
27
No. 2021AP1450-OA.rgb
II. JUSTICE HAGEDORN'S SOLO CONCURRENCE
¶241 Justice Hagedorn wrote a solo concurrence to our
November 30 opinion, which many parties treated as the
controlling opinion. No justice joined it, and it does not
constitute binding precedent. In Wisconsin, a solo concurrence
can never be controlling. A point of law is the opinion of this
court only if a majority of justices both agree on the point and
join the mandate. State v. Dowe, 120 Wis. 2d 192, 194 352
N.W.2d 660 (1984) (per curiam) (citations omitted); Piper v.
Jones Dairy Farm, 2020 WI 28, ¶22, 390 Wis. 2d 762, 940
N.W.2d 701 (citations omitted). Justice Hagedorn joined all but
six of the 81 paragraphs comprising our November 30 opinion.
The 75 paragraphs joined by four justices in the majority
constitute the majority opinion of the court.
¶242 Perhaps the parties mistakenly assumed the position of
the United States Supreme Court on this issue applies to
Wisconsin Supreme Court cases. The United States Supreme Court
will consider and count concurring opinions in cases lacking an
opinion joined by a majority. In Marks v. United States, the
Court held, "[w]hen a fragmented Court decides a case and no
single rationale explaining the result enjoys the assent of five
justices, 'the holding of the Court may be viewed as that
position taken by those Members who concurred in the judgment on
the narrowest grounds[.]'" 430 U.S. 188, 193 (1977) (quoting
Gregg v. Georgia, 428 U.S. 153, 169 n.15 (1976) (plurality)).
Federal courts understand the so-called Marks Rule differently.
Some give precedential effect to the narrowest opinion that
joined the mandate; others search for a "common denominator"
28
No. 2021AP1450-OA.rgb
that "must embody a position implicitly approved by at least [a
majority] of Justices who support the judgment." See United
States v. Epps, 707 F.3d 337, 348 (D.C. 2013) (quoting King v.
Palmer, 950 F.2d 771, 781 (D.C. Cir. 1991) (en banc)).
¶243 The Marks Rule does not apply to this case, but even
if it did, Justice Hagedorn's solo concurrence would not be
controlling. This court has never applied the Marks Rule to
interpret its own precedent, but only to interpret federal
precedent. See State v. Griep, 2015 WI 40, ¶36, 361
Wis. 2d 657, 863 N.W.2d 567. Even if this court had adopted the
Marks Rule (which has been the subject of substantial scholarly
criticism),48 it would not apply. On many points, Justice
Hagedorn's concurrence is broader than the majority opinion, and
some of its conclusions lack any common rationale with the
majority. For example, Justice Hagedorn said extra-legal
criteria could be considered in selecting a map——but only those
extra-legal criteria he deemed important in his subjective
judgment.49 Three justices in the majority would have stuck to
The
48 parties' reliance on Justice Hagedorn's solo
concurrence illustrates one problem with the Marks Rule.
Justice Hagedorn represents one-seventh of this court, yet his
opinion has nonetheless been treated as controlling by most of
the parties in this case. The "least popular view[s]" of a
single justice do not reflect the law. See Richard M. Re,
Beyond the Marks Rule, 132 Harv. L. Rev. 1943, 1944 (2019).
Although Justice Hagedorn believes this court can define
49
what constitutes a community of interest and then protect that
community in selecting a map, he acknowledges, "[i]t is not a
legal requirement[.]" Johnson, 399 Wis. 2d 623, ¶83 (Hagedorn,
J., concurring). In contrast, Justice Hagedorn was unwilling to
consider another extra-legal criterion: partisan fairness.
Id., ¶87. This inconsistency has never been explained. Justice
Hagedorn agrees this court lacks the institutional competency to
29
No. 2021AP1450-OA.rgb
the law alone, showing an unbridgeable philosophical divide
regarding the propriety of extra-legal criteria advanced by the
concurrence.
¶244 In fairness to the parties who mistook Justice
Hagedorn's solo concurrence for the opinion of this court,
perhaps their confusion stemmed from Justice Hagedorn's own
words. In his concurrence, he "invited" the parties to submit
proposed remedial maps and briefing in conformity with his
idiosyncratic views50——never mind that only this court, acting
through a majority of participating justices, can "invite"
parties to do anything. Justice Hagedorn may have cast the
deciding vote in this case, but he does not have the power to
act as a supreme court of one.
¶245 Justice Hagedorn's solo concurrence is also
inconsistent with the views he now expresses as the majority
author. Never once did he mention "core retention" in his
concurrence——nor did the majority, and the dissent used the
phrase only once, in passing.51 In contrast, today's rather
define what constitutes partisan fairness and which political
communities deserve special consideration. For the same
reasons, "it is not for the Court to define what a community of
interest is and where its boundaries are, and it is not for the
Court to determine which regions deserve special consideration."
Id., ¶71 n.7 (majority opinion) (quoting In re Legislative
Districting of the State, 805 A.2d 292, 298 (Md. 2002)).
50 Id., ¶63 (Hagedorn, J., concurring).
51 Id., ¶97 (Dallet, J., dissenting) (citation omitted).
30
No. 2021AP1450-OA.rgb
short majority opinion52 uses the phrase a striking 27 times.53
Justice Hagedorn now says, as the majority author, core
retention is the "preeminent . . . metric"54 and "especially
helpful."55 We never determined "core retention is . . . central
to least change review,"56 despite some parties stating in
briefing before our November 30 opinion that it might be
important to consider, nor did we determine that it is a
"preeminent . . . metric" or "especially helpful." We never
mentioned it at all, until now.
¶246 While we determined that the least-change approach
should guide this court's decision, no one thought that meant
maximizing core retention——not even Justice Hagedorn. There is
a reason the majority does not direct the reader to any portion
of our November 30 opinion to support the proposition that core
retention is dispositive: this majority made it up.
¶247 Justice Hagedorn's concurrence contemplates a
situation that should (as a statistical matter) never occur if
The majority opinion addresses several issues but spans a
52
mere 32 pages. In contrast, the three-judge federal district
court opinion in Singleton v. Merrill, one of the most recent
successful VRA challenges in the context of redistricting, is
225 pages. __ F. Supp. 3d __, 2022 WL 265001 (N.D. Ala. Jan.
24) (per curiam), stayed sub nom. pending cert. review, Merrill
v. Milligan, 142 S. Ct. 879 (Mem). In this case, the only full-
fledged VRA analyses come from the three dissents.
53 Majority op., ¶¶7–8, 13 & n.9, 14–15, 22, 24, 26–30, 33.
54 Id., ¶33.
55 Id., ¶13.
56 Id.
31
No. 2021AP1450-OA.rgb
core retention is the "preeminent . . . metric" in selecting
maps——a tie:
Suppose we receive multiple proposed maps that comply
with all relevant legal requirements, and that have
equally compelling arguments for why their proposed
map most aligns with current district boundaries. In
that circumstance, we still must exercise judgment to
choose the best alternative. Considering communities
of interest (or other traditional redistricting
criteria) may assist us in doing so.
Johnson, 399 Wis. 2d 623, ¶83 (Hagedorn, J., concurring).
Justice Hagedorn envisioned parties presenting "equally
compelling arguments" regarding least change, which is an odd
turn of phrase if he really meant, "I will vote for whichever
maps have the best core retention." The chance of two proposed
remedial maps having the same core retention probably approaches
the chance of winning the lottery. No reasonable person would
read Justice Hagedorn's concurrence and think a slight
difference in core retention would be dispositive, yet that is
exactly what the majority now holds.
¶248 Justice Hagedorn's misunderstanding of the least-
change approach, first displayed in his concurrence, infects the
majority opinion in a more fundamentally erroneous way than
equating least change with core retention. The majority spends
substantial time discussing Tennant v. Jefferson County
Commission, 567 U.S. 758, 764–65 (2012) (per curiam).
Specifically, the majority states:
In Tennant[,] . . . the Supreme Court upheld a 4,871-
person deviation in West Virginia's congressional
districts, noting the deviation advanced the state's
interests in maximizing core retention and maintaining
whole counties. . . .
32
No. 2021AP1450-OA.rgb
The United States Supreme Court held that maximizing
core retention was an acceptable justification for far
greater deviation in Tennant.[57]
There are multiple problems with the majority's reliance on
Tennant.
¶249 First, our November 30 opinion did not recognize least
change, let alone core retention, as a "state interest." The
least-change approach reflects this court's limited power to
remedy violations of law, which does not include the power to
write statutes out of whole cloth. Johnson, 399 Wis. 2d 623, ¶8
(majority opinion) ("Because the judiciary lacks the lawmaking
power constitutionally conferred on the legislature, we will
limit our remedy to achieving compliance with the law rather
than imposing policy choices."). "A least-change approach is
nothing more than a convenient way to describe the judiciary's
properly limited role in redistricting." Id., ¶72.
¶250 The majority errs by treating core retention as a
state interest of critical importance, at the expense of
applying the text of the Wisconsin Constitution. At most, core
retention may indicate whether this court has exceeded its
jurisdiction by delving into political decision-making. In
choosing the Governor's maps, the majority does not limit itself
to "making only those changes necessary for the maps to comport
with the one person, one vote principle while satisfying other
constitutional and statutory mandates (a 'least-change'
approach)," id., ¶5, but instead implements Justice Hagedorn's
57 Id., ¶¶22, 24.
33
No. 2021AP1450-OA.rgb
previously articulated view, which permits tipping the scales
with concededly extra-legal criteria. Id., ¶83 (Hagedorn, J.,
concurring).58
¶251 Second, the West Virginia State Legislature drew the
map under review in Tennant. 567 U.S. at 760–61. Courts have
long been held to higher standards than legislative bodies when
drawing maps precisely because courts do not get to determine,
in the first instance, what constitutes a state interest (at
least not normally).59 The majority's reliance on Tennant is
misplaced.
¶252 That Justice Hagedorn's majority opinion is a
perversion of least change is self-evident from the opinion's
very structure. The majority "begin[s] [its] analysis by
probing which map makes the least change from current district
boundaries. From there, [it] examine[s] the relevant law[.]"60
As in any case, the court is supposed to begin with the law.
Without first knowing what the law requires, there is no way for
the court to "mak[e] only those changes necessary for the maps
to comport with the [law]." Johnson, 399 Wis. 2d 623, ¶5
(majority opinion). The majority's fundamentally flawed
analysis produces an illegitimate remedy.
58Justice Ann Walsh Bradley confirms the majority
privileged policy over the law in her concurrence, which is
joined by all members of the majority except Justice Hagedorn.
59 Chief Justice Ziegler's Dissent, ¶141.
60 Majority op., ¶12.
34
No. 2021AP1450-OA.rgb
III. ZIMMERMAN
¶253 Nearly sixty years have passed since this court last
resolved redistricting litigation. In that case, this court
declared a redistricting plan cannot be implemented by joint
resolution. Zimmerman, 22 Wis. 2d at 559. While Zimmerman has
been precedent for many years, it is the only case to address
that issue, and this court has never had the opportunity to
revisit it because every redistricting case that followed was
heard exclusively in federal court. Unlike a fine wine,
precedent does not necessarily get better with age.61
¶254 With respect to state legislative redistricting
plans,62 the foundation for Zimmerman is weak. The text of
Article IV, Section 3 does not contemplate a role for the
Governor in the drawing of assembly and senate maps. Compare
Wis. Const. art. IV, § 3 ("[T]he legislature shall apportion and
district anew the members of the senate and assembly[.]"), with
e.g., id. art. I, § 21(1) ("Writs of error . . . shall be issued
by such courts as the legislature designates by law." (emphasis
61 See Montejo v. Louisiana, 129 S. Ct. 2079, 2093 (2009)
(Alito, J., concurring) ("The dissent, finally, invokes
Jackson's antiquity, stating that 'the 23–year existence of a
simple bright-line rule' should weigh in favor of its retention.
Post, at 2098. But in Gant, the Court had no compunction about
casting aside a 28–year–old bright-line rule. I can only assume
that the dissent thinks that our constitutional precedents are
like certain wines, which are most treasured when they are
neither too young nor too old, and that Jackson, supra at 23, is
in its prime, whereas Belton, supra at 28, had turned brownish
and vinegary.").
62Article IV, Section 3 governs assembly and senate
districts, not congressional districts.
35
No. 2021AP1450-OA.rgb
added)). While the Legislature's prerogative to enact laws is
subject to a gubernatorial veto, the constitution does not
describe the Legislature's duty to redistrict as lawmaking,
suggesting the constitution denies the Governor a role in the
process.63
¶255 In contrast, at the time the Wisconsin Constitution
was adopted, Article XIV, Section 11 expressly provided
congressional redistricting would involve both the Legislature
and the Governor. Wis. Const. Art. XIV, § 11 (1848), repealed
1982 (declaring the state's two congressional districts, and
saying they shall be in force "until otherwise provided by law"
(emphasis added)). Differences in language typically signal
differences in meaning, particularly when two provisions of the
same document use different language to describe analogous
concepts. See Parsons v. Associated Banc-Corp., 2017 WI 37,
¶26, 374 Wis. 2d 513, 893 N.W.2d 212 (quoting Antonin Scalia &
Bryan A. Garner, Reading Law 170 (2012)) ("'A word or phrase is
presumed to bear the same meaning throughout a text; a material
variation in terms suggests a variation in meaning.' . . . The
fact that the same section of the state constitution refers
generally to a matter being 'prescribed by law' and specifically
to the legislature 'provid[ing]' something 'by statute' strongly
Legislature's 10/26/21 Br., at 21 ("The Legislature's
63
power to reapportion its districts is specifically enumerated in
the state constitution, distinct from its lawmaking
power. . . . [The text of Article IV, Section 3] does not
provide that 'the legislature should enact legislation to
apportion anew' or 'the legislature shall by law apportion
anew.'").
36
No. 2021AP1450-OA.rgb
suggests that 'law' in that section has a broader meaning than
simply 'statutory law.'" (modification in the original)).
¶256 The difference between the text of Article IV, Section
3 and the now repealed Article XIV, Section 11 is particularly
telling in light of early Wisconsin history. Under territorial
law, the Governor had an explicit role in reapportionment.
Although he did not draw districts, the Governor was responsible
for assigning a number of representatives to each district. The
law provided, in relevant part:
As soon as practicable after having been furnished
with the enumeration of the inhabitants of the
Territory, . . . the Governor of the Territory shall
apportion the thirteen members of the Council, and
twenty-six members of the House of Representatives,
among the several electoral districts as organized by
law, according to their population, as near as may be,
as shown by the census taken by virtue of this act.
1842 Laws Wis. Terr. 50. Wisconsin's founders did not preserve
this particular gubernatorial role, and we should be skeptical
of the idea they gave him an entirely different role——the power
of vetoing redistricting plans——without using language even
nearly as explicit.64 See generally James T. Austin, The Life of
The Legislature did not try to enact redistricting plans
64
by joint resolution until the 1960s, despite gubernatorial
vetoes of redistricting legislation. State ex rel. Reynolds v.
Zimmerman, 22 Wis. 2d 544, 553, 126 N.W.2d 551 (1964). To some
extent, this customary practice may inform original meaning, but
it is evidence of lesser value and of course secondary to the
plain meaning of the words, as illuminated by historical context
surrounding their adoption. See, e.g., SEIU v. Vos, 2020 WI 67,
¶28, 393 Wis. 2d 38, 946 N.W.2d 35 (Hagedorn, J., majority op.)
("The text of the constitution reflects the policy choices of
the people, and therefore constitutional interpretation
similarly focuses primarily on the language of the
constitution." (citation omitted)); Coulee Catholic Schs. v.
LIRC, 2009 WI 88, ¶57, 320 Wis. 2d 275, 768 N.W.2d 868 ("The
37
No. 2021AP1450-OA.rgb
Elbridge Gerry 347 (1829) (explaining Governor Elbridge Gerry
signed the first so-called "gerrymander" into law because, in
light of "precedents," he doubted whether he could veto the
legislation).
¶257 The Legislature alone has the constitutionally-
prescribed duty to enact a state legislative redistricting plan
each decade. Johnson, 399 Wis. 2d 623, ¶13. While a veto may
frustrate the Legislature's policy agenda, it does not normally
hinder the Legislature from fulfilling an obligation assigned to
it by the supreme law. Whether the Governor actually has the
power to inhibit a co-equal branch's ability to perform its
duty, absent express constitutional authorization, is
questionable.
¶258 The Legislature's duty was critical to an argument
advanced by several "legal scholars"65 in an amicus brief. They
claimed, "the whole reason for this litigation is that the
authoritative, and usually final, indicator of the meaning of a
provision [of the Wisconsin Constitution] is the text——actual
words used." (citation omitted)); Jacobs v. Major, 139
Wis. 2d 492, 504, 407 N.W.2d 832 (1987) ("We need go no further
than holding that Art. I, sec. 3 has [a] plain, unambiguous
meaning[.]"); Black v. City of Milwaukee, 2016 WI 47, ¶54, 369
Wis. 2d 272, 882 N.W.2d 333 (Rebecca Grassl Bradley, J.,
concurring) ("I give priority to the plain meaning of the
words[.]" (citation omitted)). The Legislative and Executive
branches cannot, through tacit understanding, change the
constitutional allocation of powers. Bartlett v. Evers, 2020 WI
68, ¶210, 393 Wis. 2d 172, 945 N.W.2d 685 (Kelly, J.,
concurring/dissenting).
65The legal scholars include (in the order listed in the
brief's appendix) Richard Briffault, Joseph Fishkin, James A.
Gardner, Michael S. Kang, D. Theodore Rave, David Schultz, Kate
Shaw, and Robert Yablon.
38
No. 2021AP1450-OA.rgb
legislature breached its constitutional duty to redistrict by
failing to pass a bill with gubernatorial support or a veto-
proof majority."66 This viewpoint is peculiar, but it highlights
a problem with Zimmerman. The Legal Scholars blame this
litigation solely on the Legislature, but an analogous charge
could be levied against the Governor if in fact the executive
has any constitutional role to play in redistricting despite the
absence of a provision granting him one. As long as this
court's precedent permits the Governor to veto redistricting
plans, redistricting is as much his duty as it is the
Legislature's——but that is inconsistent with the way we have
described the duty. E.g., Johnson, 399 Wis. 2d 623, ¶79 ("[T]he
legislature must implement a redistricting plan each cycle.").
¶259 This court's precedent significantly increases the
likelihood of judicial involvement in what should be a purely
political process.67 If the political process fails to produce
Amicus Br. Legal Scholars, at 5. The majority similarly
66
misstates the Legislature's duty, saying "[w]e have given the
political branches a fair opportunity to carry out their
constitutional responsibilities. They have not done so."
Majority op., ¶2. Actually, the Legislature has. The
Legislature fulfilled its constitutional duty to "apportion and
district anew the members of the senate and assembly, according
to the number of inhabitants," but the Governor vetoed the
Legislature's plans. See Wis. Const. Art. IV, § 3. The
majority describes our responsibilities as an "unwelcome task,"
majority op. ¶2, which is a strange way of describing the job we
were elected to perform.
67Johnson v. WEC, No. 2021AP1450-OA, unpublished order, at
11 (Wis. Sept. 22, 2021, amended Sept. 24) (Rebecca Grassl
Bradley, J., concurring) (explaining Zimmerman creates "a
constitutional conundrum").
39
No. 2021AP1450-OA.rgb
redistricting plans, this court has a duty to remedy
constitutional and other legal defects in the existing maps;
however, if this court's precedent defines the process
differently than the Wisconsin Constitution, this court has a
duty to align its precedent with the text of the constitution.
We cannot mistake "the law" for "the opinion of the judge"
because "the judge may mistake the law."68 Introduction, William
Blackstone, Commentaries *71; see also Bryan A. Garner et al.,
The Law of Judicial Precedent 397 (2016) ("The primary and most
important factor to weigh in considering whether to overrule an
earlier decision is its correctness.").
IV. CONCLUSION
¶260 Our November 30 opinion in this case cabined the
court's redistricting decision-making to the confines of the
law. Unfortunately prophetic, it also cautioned that if four
The 2011 assembly and senate maps were adopted by law and
68
are codified as statutes (except for a minor change to the
assembly map made by a federal court). Johnson, 399
Wis. 2d 623, ¶14 (majority opinion). A joint resolution cannot
replace duly enacted law——even when that law has been declared
unconstitutional. Id., ¶72 n.8. Contra id., ¶93 n.3 (Dallet,
J., dissenting) ("[B]oth the Wisconsin and U.S. Constitutions
require that all maps be redrawn every ten years to account for
population shifts since the prior census. These are the sunset
provisions. In this respect, the 2011 maps are unlike an
ordinary unconstitutional statute, since they were enacted
without any expectation of longevity." (citations omitted)).
Perhaps this court should consider, as a remedy, allowing
the Legislature to redistrict by joint resolution. Unless a
court adopts the Governor's maps as it did in this case, a
court-ordered remedy ultimately denies the Governor control
anyway. Zimmerman does not prohibit the Legislature from
implementing redistricting plans by joint resolution in the
event of an impasse.
40
No. 2021AP1450-OA.rgb
members of this court cast aside those confines, "judges would
refashion this court as a committee of oligarchs with political
power superior to both the legislature and the governor."
Johnson, 399 Wis. 2d 623, ¶80 (citation omitted). In this
opinion, the majority abandons the law, perverts the least-
change approach into a license for policymaking, and
subordinates constitutional commands, statutory restrictions,
and precedent to the majority's preferences. I dissent.
¶261 I am authorized to state that Chief Justice ANNETTE
KINGSLAND ZIEGLER and Justice PATIENCE DRAKE ROGGENSACK join
this dissent.
41
No. 2021AP1450-OA.rgb
1