Baldus et al v. Brennan et al

Filing 210

MEMORANDUM OPINION AND ORDER by Circuit Judge Diane P. Wood, District Judge J.P. Stadtmueller, and District Judge Robert M. Dow, Jr., on 3/22/12 as follows: GRANTING the plaintiffs' and intervenor-plaintiffs' Sixth Claim for relief - the c ourt having found that New Assembly Districts 8 and 9 violate the Voting Rights Act and accordingly the Government Accountability Board is ENJOINED from implementing Act 43 in its current form; DISMISSING with prejudice plaintiffs' and interven or-plaintiffs' remaining claims; DENYING as moot 75 the intervenor-defendants' motion for judgment on the pleadings; DENYING as moot 128 defendants' motion for summary judgment; DENYING as moot 117 plaintiffs' motion to d efer a judicial decision; DENYING as moot 119 intervenor-plaintiffs' motion to defer a judicial decision; having been granted at trial, TERMINATING administratively 179 James Troupis' motion for clarification; TERMINATING administrat ively 160 defendants' motion in limine, as related to the presentation of evidence related to redistricting anomalies; DENYING 198 intervenor-defendants' motion to dismiss for lack of standing; and, DENYING 126 Citizens for Fair and Public Redistricting's motion to appear as amicus curiae; each party shall bear its own costs. (cc: all counsel) (nm)

UNITED STATES DISTRICT COURT
EASTERN DISTRICT OF WISCONSIN
ALVIN BALDUS, CARLENE BECHEN, ELVIRA
BUMPUS, RONALD BIENDSEIL, LESLIE W DAVIS,
III, BRETT ECKSTEIN, GLORIA ROGERS, RICHARD
KRESBACH, ROCHELLE MOORE, AMY RISSEEUW,
JUDY ROBSON, JEANNE SANCHEZ-BELL,
CECELIA SCHLIEPP, TRAVIS THYSSEN, CINDY
BARBERA, RON BOONE, VERA BOONE,
EVANJELINA CLEERMAN, SHEILA COCHRAN,
MAXINE HOUGH, CLARENCE JOHNSON,
RICHARD LANGE, and GLADYS MANZANET

Case No. 11-CV-562
JPS-DPW-RMD

Plaintiffs,
TAMMY BALDWIN, GWENDOLYNNE MOORE and
RONALD KIND,
Intervenor-Plaintiffs,
v.
Members of the Wisconsin Government
Accountability Board, each only in his official
capacity: MICHAEL BRENNAN, DAVID
DEININGER, GERALD NICHOL, THOMAS CANE,
THOMAS BARLAND, and TIMOTHY VOCKE, and
KEVIN KENNEDY, Director and General Counsel for
the Wisconsin Government Accountability Board,
Defendants,
F. JAMES SENSENBRENNER, JR., THOMAS E.
PETRI, PAUL D. RYAN, JR., REID J. RIBBLE, and
SEAN P. DUFFY,
Intervenor-Defendants.
VOCES DE LA FRONTERA, INC., RAMIRO VARA,
OLGA VARA, JOSE PEREZ, and ERICA RAMIREZ,

v.

Case No. 11-CV-1011
JPS-DPW-RMD

Members of the Wisconsin Government
Accountability Board, each only in his official
capacity: MICHAEL BRENNAN, DAVID
DEININGER, GERALD NICHOL, THOMAS CANE,
THOMAS BARLAND, and TIMOTHY VOCKE, and
KEVIN KENNEDY, Director and General Counsel for
the Wisconsin Government Accountability Board,

MEMORANDUM
OPINION and
ORDER

Plaintiffs,

Defendants.

Tried to the Court February 21–24, 2012
Decided March 22, 2012
Before WOOD, Circuit Judge, STADTMUELLER District Judge, and DOW,
District Judge.
PER CURIAM . There was once a time when Wisconsin was famous for
its courtesy and its tradition of good government. In 2006, James J. Conant
was able to write that:
The most important feature of Wisconsin’s society,
government, and politics during the twentieth century
was its progressive nature. Wisconsin had a highly
developed civil society, its elected and administrative
officials continuously attempted to improve the state’s
political institutions, and they attempted to enhance the
economic and social circumstances of the state’s citizens.
Throughout the century Wisconsin’s politics were issueoriented, state government institutions operated free of
scandal, and the administration of state policies and
programs was conducted efficiently and effectively.1
Students of American history still read about Robert M. La Follette,
Sr., an independent thinker who came to prominence at the end of the 19th
century and whose views defied the partisan pigeonholes of his day. More
recently, Wisconsin has been called a “purple” state–that is, a state whose
people regularly elect comparable numbers of Democrats and Republicans.
Over roughly the last half-century, six Republicans and six Democrats have
served as governor. Over the same time, one of its two seats in the U.S.
Senate has been held continuously by a Democrat, while the other one has
been occupied by three Republicans and two Democrats.
1

James J. Conant, WISCONSIN POLITICS
LABORATORY OF DEMOCRACY at xv (2006).

Page 2 of 38

AND

GOVERNMENT : AM ERICA ’S

This bipartisan tradition has not, unfortunately, exempted Wisconsin
from the contentious side of the redistricting process that takes place every
ten years in the wake of the United States Census. Before the events leading
to this lawsuit, the last time the Wisconsin legislature successfully passed a
redistricting plan was in 1972, following the 1970 census. See Wis. Stat.
§ 4.001(1); Wisconsin State AFL-CIO v. Elections Board, 543 F. Supp. 630, 631
(E.D. Wis. 1982). After the 1980 census, however, the state was not so
fortunate. Beginning with that round, decennial litigation was just as much
a feature of the political scene as was decennial redistricting. See Wisconsin
State AFL-CIO, 543 F. Supp. 630 (1980 census); Prosser v. Elections Board, 793
F. Supp. 859 (W.D. Wis. 1992) (1990 census); Arrington v. Elections Board, 173
F. Supp. 2d 856 (E.D. Wis. 2001) (2000 census) and then Baumgart v.
Wendelberger, Nos. 01 121 and 02 366, 2002 WL 34127471 (E.D. Wis. May 30,
2002) (per curiam), amended by 2002 WL 34127473 (E.D. Wis. July 11, 2002 )
(also 2000 census). In 1982, 1992, and 2002, Wisconsin’s legislative districts
were drawn by a three-judge court. It is notable that in each of these earlier
cases, the only contested matter related to the state’s legislative districts; until
now, no one has called on the federal court to intervene with respect to the
state’s congressional districts.
Now it is our turn. The U.S. Constitution, see Art. I, sec. 2, cl. 3,2
requires the federal government to conduct an actual enumeration of the U.S.

2

The Enumeration Clause is actually the fourth to appear in the original
Constitution, but the original Clause 3, which established the infamous three-fifths
rule for counting population, was abrogated by section 2 of the Fourteenth
Amendment. We therefore skip over the now-repudiated clause and count the
Enumeration Clause as the third. See Utah v. Evans, 536 U.S. 452, 457 (2002); Dep’t
of Commerce v. U.S. House of Representatives, 525 U.S. 316, 362 (1999) (Stevens, J.,
dissenting).

Page 3 of 38

population once every ten years; that enumeration provides the basis for
representation in the House of Representatives. Article IV, section 3, of the
Wisconsin Constitution independently requires the state legislature to update
its senate and assembly districts following each federal census. In 2010, the
Bureau of the Census complied with its constitutional duty, and on
December 21, 2010, it announced and certified that Wisconsin’s population
was 5,686,986 as of April 1, 2010. This represented a slight increase over the
2000 population of 5,363,675. These new numbers required Wisconsin to take
a fresh look at both its state assembly and senate districts, and its eight
congressional districts (the overall number of districts remained the same) to
ensure compliance with the one-person, one-vote principle announced by the
Supreme Court in Reynolds v. Sims, 377 U.S. 533 (1964). Wisconsin has
attempted to do so. Regrettably, like many other states, Wisconsin chose a
sharply partisan methodology that has cost the state in dollars, time, and
civility. Nevertheless, our task is to assess the legality of the outcome, not
whether it lived up to any particular ideal.
1.

The Redistricting Process
The mid-term election in November 2010 resulted in a shift in political

control in the State of Wisconsin. Scott Walker, the Republican candidate for
governor, defeated Tom Barrett, the Democratic candidate, in the race to
replace Governor Jim Doyle, a Democrat. Control of both Wisconsin’s State
Assembly (its lower house) and State Senate shifted from the Democratic to
the Republican party. Thus, as of the time the Census results were certified
and the state was ready to begin drawing whatever new legislative and
congressional district lines were necessary, all three critical players were in
the hands of a single party for the first time in many years. (Throughout this

Page 4 of 38

opinion, when we refer to “legislative” redistricting, we mean the two state
houses; we use the term “congressional” redistricting for the lines drawn for
seats in the U.S. House of Representatives.) The new governor and legislators
were sworn in on January 3, 2011, and the very next day the Republican
legislative leadership announced to members of the Democratic minority that
the Republicans would be provided unlimited funds to hire counsel and
consultants for the purposes of legislative redistricting. They informed the
Democrats that they would not receive any funding for this process.
True to their word, the Republicans immediately began work in
earnest, retaining the law firm of Michael Best & Friedrich LLP (“Michael
Best”) to advise their caucus. Every effort was made to keep this work out of
the public eye and, most particularly, out of the eye of the Democrats.
Indeed, it was widely assumed that redistricting work would not begin until
Wisconsin’s units of local government had a chance to review their ward
lines. Wards in Wisconsin are the smallest unit of government. In the past,
redistricting has always proceeded on a “bottom up” basis: ward lines would
be redrawn based on the new census figures, villages and towns would recompute their populations, and the counties would build on those figures.
The Census does not use these units of government; instead, it proceeds on
the basis of “census blocks” that do not always correspond to local
government boundaries. Some care must be taken, therefore, to translate the
census data into information that is compatible with actual governing units.
As we noted, the venue of the redistricting work was the offices of
Michael Best. The actual drafters included: Adam Foltz, a staff member to
Assembly Speaker Jeff Fitzgerald; Tad Ottman, a staff member to Senate
Majority Leader Scott Fitzgerald; and Joseph Handrick, a consultant with the

Page 5 of 38

law firm of Reinhart Boerner Van Duren s.c. Others involved in the process
were James Troupis, Eric McLeod, Ray Taffora, Speaker Fitzgerald, Majority
Leader Fitzgerald, Sarah Troupis, Robin Vos, Senator Rich Zipperer, and Dr.
Keith Gaddie. The drafters relied on a computer program called autoBound
to work with various district lines. They testified that the partisan makeup
of the potential new districts played no part at all in their decisions.
Handrick, for instance, testified that he did not know if partisan makeup was
considered, that he had no access to voting data from past elections, and that
only “population equality, municipal splits, compactness, contiguity, [and]
communities of interest” were considered. Foltz testified that he worked with
legal counsel and experts, and that Speaker Fitzgerald, Senator Fitzgerald,
Robin Vos, and Senator Zipperer advised him where to draw the boundaries.
In June and July 2011, Foltz had meetings about redistricting with
every single Republican member of the State Assembly. He did not meet
with any Democrats. Nevertheless, he testified that it was not “a part of the
goal to increase the Republican membership in the legislature.” Before his
meetings with the Republicans, each person was required to sign a
confidentiality agreement promising not to discuss anything that was said.
Ottman had similar meetings, conducted under the same cloak of secrecy.
The drafters did not limit their outreach to public officials; they also held
meetings behind closed doors with selected outsiders. In January 2011, they
met with certain private business interests, including representatives from
realtor and banking associations, and a hybrid state chamber of commerce
called Wisconsin Manufacturers & Commerce.
In addition, the drafters reached out to certain members of the Latino
community. They contacted Jesus Rodriguez, a co-founder and member of

Page 6 of 38

Hispanics for Leadership, a political organization comprised of local business
people, educators, and community advocates who work toward “getting the
most representation possible for the Latino community on all levels.”
Rodriguez is also the President of Hispanics for School Choice, a nonprofit
organization dedicated to advancing school choice for Hispanic children,
notably through school vouchers. Hispanics for School Choice, available
online at http://www.hispanicsforschoolchoice.com/ (last visited March 14,
2012). Through Hispanics for School Choice, Rodriguez developed a
professional and personal relationship with former Assembly Speaker Scott
Jensen (a Republican), who presently serves as a senior advisor for another
school

choice

Children.

advocacy

American

organization,

Federal

for

American

Children,

Federation

available

online

for
at:

http://www.federationforchildren.org/ (last visited March 15, 2012).
Troupis also contacted the Mexican American Legal Defense Education Fund
(MALDEF), a national Latino civil rights organization, in an attempt to secure
its support for the Republicans’ plan. He hoped to “take the largest legal
fund for the Latino community off the table in any later court battle,” by
courting their approval.
The process followed for the congressional districts was somewhat
different. Like the state legislature, the Wisconsin congressional delegation
ended up with a majority of Republicans after the 2010 mid-term elections
(specifically, five Republicans and three Democrats, as we can see from the
intervening parties to this case). In keeping with long-standing practice, the
legislature in 2011 permitted the incumbent Wisconsin members of the House
of Representatives to draft a map delineating the new congressional districts.
Andrew D. Speth, chief of staff to Republican Congressman Paul D. Ryan, Jr.,

Page 7 of 38

took primary responsibility for that task. Speth had some communications
with Erik Olson, chief of staff to Democratic Congressman Ronald Kind, and
later on, Congressman Ryan consulted with the three Democratic members
of Wisconsin’s delegation. In meetings that Speth held with the Republican
members, they expressed their desire to draw districts that would maximize
the chances for Republicans to be elected. (Note the contrast with the
disclaimers of partisanship offered by those who were working on the
legislative redistricting process.) Speth’s first complete draft was ready by
May 13, 2011. That draft was shared exclusively with the Republicans. A
second draft of June 1, 2011, was circulated to all members of Wisconsin’s
House delegation. The Democrats offered an alternative map two days later,
but it was quickly rejected for failing to reflect minimal deviation from the
ideal population for each district. Speth finalized a draft on June 8, 2011.
On July 8, 2011, the bills that would become Act 43 (legislative
redistricting) and Act 44 (congressional redistricting) were introduced by the
Republican leadership in the Wisconsin legislature. Simultaneously, the bill
that became Act 39 was introduced. This was crucial, because it was Act 39
that permitted the legislature to draw new districts before Wisconsin’s
municipalities drew or re-drew their ward lines based on the new Census.
Instead, upending more than a century of practice in Wisconsin, Act 39
required the municipalities to adjust their ward lines to the new state
legislative districts. The legislature held a single public hearing on Acts 43
and 44 on July 13, 2011. On July 19, 2011, the legislature passed Act 43, and
on July 20, 2011, it passed Act 44; both bills were then transmitted to the
Governor. Act 39 was passed on July 25, 2011.

Page 8 of 38

2.

Procedural History
On August 9, 2011, Governor Scott Walker signed into law each of the

three critical bills discussed here: Act 39, which enabled redistricting based
only on census blocks; Act 43, establishing the new legislative districts for
both the State Assembly and the State Senate; and Act 44, establishing the
new lines for Wisconsin’s eight congressional districts. In the meantime,
correctly suspecting that something like the process we have described was
afoot, on June 10, 2011, a group of voters filed suit against the Wisconsin
Government Accountability Board (GAB) and its members in their official
capacity, alleging that Acts 43 and 44 both violate federal and state law. The
GAB is the state body charged with administering and enforcing all of
Wisconsin’s laws related to campaign finance, elections, ethics, lobbying, and
contract disclosures. We refer to these voters as the Baldus plaintiffs, using
the name of the lead party. On October 31, 2011, Voces de la Frontera, Inc.
(“Voces”), an organization that describes itself as a grassroots group
organized under the laws of Wisconsin, filed its own complaint against GAB
and its members. Voces charged that Act 43 violates Section 2 of the Voting
Rights Act; it did not challenge Act 44.
Because these two lawsuits qualified as actions “challenging the
constitutionality of the apportionment of congressional districts or the
apportionment of any statewide legislative body,” 28 U.S.C. § 2284(a), the
Chief Judge of the United States Court of Appeals for the Seventh Circuit
entered an order assigning this litigation to a three-judge court and
appointing this panel to serve as the members of the court. In an order
entered on November 21, 2011, the court permitted the three Democratic
members of Wisconsin’s delegation to the United States House of

Page 9 of 38

Representatives to intervene as plaintiffs, and it permitted the five
Republican members of that delegation to intervene as defendants. The next
day, the court consolidated the Baldus and the Voces cases. Pretrial discovery
took place on an expedited basis, with the expectation that trial would begin
on February 21, 2012. That morning, however, the court urged the parties to
make one last good-faith effort to settle, in the interest of all citizens of the
State of Wisconsin and out of respect for the role of the state legislature in
redistricting matters. See Perry v. Perez, 132 S.Ct. 934, 940 (2012). Those
efforts, unfortunately, were unsuccessful, and so the trial continued on
February 23 and 24, 2012.
With the benefit of the full record, the panel now makes its findings
of fact and conclusions of law for the case. For ease of reading, these are not
presented separately. Our analysis of each of the plaintiffs’ claims leads us
to the conclusion that Act 43 violates Section 2 of the Voting Rights Act of
1965, 42 U.S.C. § 1973(a), by improperly diluting the citizen voting age
population of Latinos across Assembly Districts 8 and 9. Otherwise, we find
no judicially redressable injury in any of the plaintiffs’ and intervenorplaintiffs’ remaining claims.
3.

Analysis of Claims
3.1

Overview

Before turning to our detailed analysis of each claim, we must review
what is still properly before us. For ease of reference, we begin by
summarizing the claims of the Second Amended Complaint, noting the
statute to which each one pertains, the current status of the claim, and, where
pertinent, which parties are pursuing it:

Page 10 of 38

!

Claim One (Act 43): Legislative boundaries unconstitutionally
sacrifice redistricting principles required by the U.S.
Constitution. This claim went to trial on behalf of the Baldus
plaintiffs.

!

Claim Two (Act 43): The new legislative districts violate
federal standards because they impermissibly disrupt local
governmental boundaries. The Baldus plaintiffs were the only
ones raising this, and they abandoned it at trial. Trial Trans.
Vol. VI, 398-99.

!

Claim Three (Act 43): The statute violates federal law because
it disenfranchises 299,704 voters whose state senate districts
have been moved. It does so by shifting them from an oddnumbered district to an even-numbered district; this shift
means that voters in the affected districts will have to wait six
years, not just four, until they have an opportunity to vote
again for their state senator. This claim went to trial for the
Baldus plaintiffs only.

!

Claim Four (Act 44): Congressional districts are not compact
and fail to preserve communities of interest, in violation of the
Equal Protection Clause. The Baldus plaintiffs abandoned this
claim at trial, but the congressional plaintiff-intervenors
continued to maintain it.

!

Claim Five (Acts 43 and 44): The legislative (Act 43) and
congressional (Act 44) districts reflect partisan gerrymandering
forbidden by both the Equal Protection Clause and the First
Amendment. The Baldus plaintiffs abandoned this claim at
trial; the congressional plaintiff-intervenors continued to
maintain it.

!

Claim Six: (Act 43): The new legislative districts violate the
Voting Rights Act of 1965, 42 U.S.C. §§ 1973-1973aa-6, in two
ways: first, because Act 43 “packs” African-American voters in
Milwaukee into six districts and thus foregoes the opportunity
to create a seventh “influence” district; and, second, because
the statute “cracks” the Latino community into two districts,
Page 11 of 38

neither one of which is a majority-minority district of citizen
voting age Latinos. The Baldus plaintiffs abandoned at trial
their challenge to the African-American districts, but they,
along with the Voces de la Frontera plaintiffs in the
consolidated case, pursued the second claim at trial. (Indeed,
this claim consumed nearly all of the trial time.)
!

Claim Seven (Act 43): Act 43 is unconstitutional because the
legislative drafters used race or ethnicity as the predominant
reason for drawing certain districts, in violation of the Equal
Protection Clause as explained in cases such as Shaw v. Reno,
509 U.S. 630 (1993). The Baldus plaintiffs abandoned this claim
at trial.

!

Claim Eight (Act 43): The legislative redistricting accomplished
in Act 43 violates the Equal Protection Clause because the new
districts break up communities of interest. This claim went to
trial.

!

Claim Nine (Act 43): This claim seeks a declaratory judgment
and an injunction requiring the GAB not to use the new 2012
district lines for any recall elections that may take place in
Wisconsin between the present time and the date of the general
election in November. This claim, too, went to trial.

In summary, therefore, most of this case remains before us in one way
or another. Only Claim Two, part of Claim Six (dealing with the AfricanAmerican districts), and Claim Seven are entirely out of the case.
As we noted earlier, the total “official” population of Wisconsin for
purposes of redistricting is 5,686,986. Using that number and applying the
“one person, one vote” command, the ideal population for each of
Wisconsin’s 33 senate districts is 172,333, and for its 99 assembly districts
57,444. As for the congressional districts, the ideal population is either
710,873 or 710,874. (Two of the eight congressional districts must have an
additional person because the total population does not divide evenly by
Page 12 of 38

eight.) As we discuss below, there is some deviation from these ideals for the
legislative districts, but, reflecting the capabilities of modern computer
programs, the congressional districts from a headcount standpoint could not
be improved.
The Baldus plaintiffs are seeking declaratory and injunctive relief
against Wisconsin’s Government Accountability Board and its members.
They ask the court to bar the implementation of both redistricting plans for
the reasons we have just outlined and discuss in more detail below. The
intervenor-plaintiffs are still pursuing a claim against the new congressional
districts drawn in Act 44. The Voces de la Frontera plaintiffs, joined by the
Baldus plaintiffs, charge that Act 43 violates Section 2 of the Voting Rights
Act because it dilutes the Latino community’s voting strength in the 8th
Assembly District. Both the GAB defendants and the congressional
intervenor-defendants filed motions to dismiss both complaints on the
pleadings; they have also denied that there is any cognizable federal
violation in either Act.
We now turn to the merits of the case. In doing so, we say nothing
about any arguments that could be understood to be based on allegations
that the state officials have failed to follow state law. As defendants rightly
point out, such claims are beyond our authority under the principles
announced in Pennhurst State School & Hospital v. Halderman, 465 U.S. 89, 106
(1984). We note, however, that the facts underlying some points that touch
on state law may still be relevant for the federal issues that are properly
before us. Thus, for example, we may examine plaintiffs’ allegation that Act
43 fails to honor traditional redistricting criteria or to maintain local
government boundaries, even if we refrain from expressing any opinion on

Page 13 of 38

the question whether this may also state a claim for violating the Wisconsin
Constitution.
3.2

Claims One and Eight: The New Legislative Districts Fail To
Comply with Constitutional Standards and Are Not Justified
by any Legitimate State Interest.

Claim One addresses redistricting principles in general, while Claim
Eight focuses on the specific principle against breaking up communities of
interest. Since the latter is subsumed within the former, we have grouped
these two claims together for purposes of discussion.
Only 323,026 people needed to be moved from one assembly district
to another in order to equalize the populations numerically, but instead Act
43 moves more than seven times that number–2,357,592 people–for a net
change that results in districts that are roughly equal in size. Similarly, only
231,341 people needed to move in order to create equal senate districts, but
Act 43 moves 1,205,216–more than five times as many. Even accepting the
argument urged by the GAB that one cannot change one district without
affecting another, these are striking numbers. (Physicists would remind us
that the amplitude of waves, whether in water or in air, diminishes unless
one is in a vacuum because energy is absorbed; so too, a “wave” of
population shifts in one corner of Wisconsin is likely to dissipate long before
the other corners are reached.)
When Act 43 is compared to the 2010 census precise ideal population,
the total population deviation (from the most populous to the least populous
district) is 438 persons for the newly adopted assembly districts, and 1,076
persons for the newly adopted senate districts. Plaintiffs, therefore, allege
that the Act violates the “one-person, one-vote” principle. Reynolds v. Sims,
377 U.S. 533, 568 (1964). And indeed, it is an interesting question whether
Page 14 of 38

deviations that might have been acceptable in an earlier time ought to be
tolerated now that–as Wisconsin proved in Act 44–it is possible for a
computer to draw not one, but an unlimited number of districts with the
perfect number of voting inhabitants. But putting that thought to one side,
it was the plaintiffs who had the initial burden to show (1) the existence of a
population disparity that (2) could have been reduced or eliminated by (3)
a good-faith effort to draw districts of equal proportion. Karcher v. Daggett,
462 U.S. 725, 730 (1983). If the plaintiffs accomplish this, the burden shifts
to the GAB to show that “each significant variance between districts
was necessary to achieve some legitimate goal.” Karcher, 462 U.S. at 731.
Accepted justifications include: core retention; avoidance of split
municipalities; contiguity; compactness; and maintenance of communities of
interest. Id. at 740; Wisconsin State AFL-CIO, 543 F. Supp. at 636.
The defendants do not defend the state’s new legislative districts on
the ground that they are the best that could be managed. What the threejudge court said in 1992 remains just as true today: “representative
democracy cannot be achieved merely by assuring population equality across
districts,” Prosser, 793 F. Supp. at 863; factors like homogeneity of needs and
interests, compactness, contiguity, and avoidance of breaking up counties,
towns, villages, wards, and neighborhoods are all necessary to achieve this
end. Id. Nor do the defendants appear to argue that it is impossible to draw
a plan that serves these democratic and neutral purposes. That is plainly not
the case, since the court-drawn plans have consistently and scrupulously
striven to be politically neutral. Id. at 867; see generally Abrams v. Johnson, 521
U.S. 74, 98 (1997). Instead, defendants begin by observing that the Supreme
Court has said that “state reapportionment statutes are not subject to the
same strict standards applicable to reapportionment of congressional seats.”
Page 15 of 38

White v. Regester, 412 U.S. 755, 763 (1973). As Abrams pointed out, the Court
has also held state legislatures to a lower standard of population equality
than it imposes on courts. 521 U.S. at 98. Although times may be changing,
in the past the Court has opined that “an apportionment plan with a
maximum population deviation under 10% falls within this category of
minor deviations” that are insufficient to make out a prima facie case. Brown
v. Thomson, 462 U.S. 835, 842 (1983).
This does not mean, of course, that deviations under the 10% point are
beyond challenge. It does, however, indicate that plaintiffs bear a greater
burden to show a violation of their voting rights for deviations of 10% or
lower. Several courts have expressed this thought by concluding that
legislative population disparities under 10% are subject to a rebuttable
presumption of validity, but that they may nevertheless be unconstitutional
if the drafting process was arbitrary, discriminatory, or otherwise
unsupported by traditional redistricting criteria. Daly v. Hunt, 93 F.3d 1212,
1220 (4th Cir. 1996); Cecere v. County of Nassau, 274 F. Supp. 2d 308, 311-12
(E.D.N.Y. 2003); Montiel v. Davis, 215 F. Supp. 2d 1279, 1285-86 (S.D. Ala.
2002) (three-judge court); Hulme v. Madison County, 188 F. Supp. 2d 1041,
1047 (S.D. Ill. 2001); Abate v. Rockland County Legislature, 964 F. Supp. 817, 819
(S.D.N.Y. 1997); Marylanders for Fair Representation v. Schaefer, 849 F. Supp.
1022, 1032 (D. Md. 1994) (three-judge court). Notably, the Northern District
of Georgia concluded in a case quite similar to ours that it “need not decide
[ ] whether the mere use of a 10% population window renders Georgia’s state
legislative plans unconstitutional, because the policies the population
window was used to promote in this case were not free from any taint
of arbitrariness or discrimination.” Larios v. Cox, 300 F. Supp. 2d 1320,
1341-42 (N.D. Ga. 2004). The Georgia court was appalled by the “baldly
Page 16 of 38

unconstitutional scheme” to protect the legislative influence of traditional
communities at the expense of growing populations elsewhere and to protect
incumbents in a discriminatory and arbitrary fashion. Id. With this case law
in mind, we address the merits of the plaintiffs’ claim.
The plaintiffs’ redistricting expert, Dr. Kenneth Mayer, testified that
Act 43 fails to comply with traditional redistricting principles. He was
particularly concerned with the excessive shifts in population, disregard for
core district populations, arbitrary and partisan motivations related to
compactness, and unnecessary disenfranchisement. The defendants’ experts,
in particular Professor Bernard Grofman, had little to say about these points
beyond the generic comment that when an underpopulated district must
seize population from a neighboring district in order to reach its optimal size,
the neighboring district may need to do the same, until such time
as an overpopulated district is encountered and matters balance out.
Conspicuously missing from Professor Grofman’s testimony was anything
precise about the magnitude of the population shifts here; Dr. Mayer, in
contrast, offered testimony about many districts that could have been
balanced out by moving vastly fewer numbers of people.
While we share Dr. Mayer’s concerns in many respects and find
ourselves largely unpersuaded by Professor Grofman’s incomplete testimony
to the contrary, we return to the degree of the deviations, which were
nowhere close to the 10% number that the Supreme Court mentioned in 1983.
The maximum deviation for assembly districts is 0.76% and 0.62% for senate
districts. Numbers like these place a very heavy burden on the plaintiffs to
show a constitutional violation. In the final analysis, they have failed to
surmount that burden. We come to that conclusion not because we credit the
testimony of Foltz, Ottman, and the other drafters to the effect that they were
Page 17 of 38

not influenced by partisan factors; indeed, we find those statements to be
almost laughable. But the partisan motivation that, in our view, clearly lay
behind Act 43 is not enough to overcome the de minimis population
deviations that the drafters achieved, at least under this theory. We therefore
find no merit in Claims One or Eight and conclude that they must be
dismissed.
Before leaving this point, we add a few words about communities of
interest. It is important not to assume that the mere ability to elect a
representative of one’s preferred political party is a perfect substitute for the
ability to elect a representative who will more broadly identify with and
serve his or her constituents’ needs. The two major political parties are both
big tents that contain within them people of significantly different
viewpoints. That is precisely why, especially when the court must also take
into account the rights of minority groups as we must with Assembly
Districts 8 and 9, careful attention is necessary. As we discuss in greater
detail in section 3.5 below, the concept of community of interest is one that
sweeps in much more than party label. Thus, for example, even if the reconfigured Assembly District 8 were seen as a reliably Democratic district,
as Professor Grofman testified, that does not necessarily mean that the
successful candidate would be the candidate of choice for the Latino
community there. The whole point of the analysis under section 2 of the
Voting Rights Act is to ensure that qualifying minorities have an opportunity
to elect representatives who will have strong voices on the topics that matter
to them. Thus, the concept of community of interest will have an important
role to play when we come to Claim Six. Untethered from section 2 of the
Voting Rights Act, however, we do not have enough evidence before us to

Page 18 of 38

conclude that the remaining new districts created by Act 43 can be disturbed
on that basis alone.
3.3

Claim Three: Disenfranchisement of Voters For State
Senators

In Claim Three, the Baldus plaintiffs assert that the movement of
299,704 voters (5.26% of all persons in Wisconsin, according to the 2010
Census) from certain even-numbered senate districts to odd-numbered
senate districts deprives those voters of their constitutional right to vote for
a state senator in a regular election for two years. (Obviously, as the
defendants point out, those voters have the right to vote for any other office
on the ballot, but we do not understand defendants to be arguing that a voter
can constitutionally be deprived of the right to vote in a particular
race–maybe for the House of Representatives–as long as he/she may vote for
dog-catcher or the library board. The right to vote is a fundamental right for
every elective office in a democracy.) Pursuant to Wisconsin Constitution
Article IV, section 5, state senators serve four-year, staggered terms with half
of the senators elected in presidential years and the other half during midterm years. The redistricting plan shifts voters among senate districts in a
manner that causes certain voters who previously resided in an even-number
district (which votes in presidential years) to be moved to an odd-numbered
district (which votes in mid-term years); this shift means that instead of
voting for a state senator in 2012, as they would have done, they must wait
until 2014 to have a voice in the composition of the State Senate. The number
of persons experiencing this type of disenfranchisement per district ranges
from 133 to 72,431, with an average of 17,630 for the 17 districts involved.
The Baldus plaintiffs argue that this disenfranchisement violates the
Equal Protection Clause’s requirement that “a State make an honest and
Page 19 of 38

good faith effort” to avoid vote dilution. Reynolds, 377 U.S. at 577. Some
degree of temporary disenfranchisement in the wake of redistricting is seen
as inevitable, and thus as presumptively constitutional, so long as no
particular group is uniquely burdened. Donatelli v. Mitchell, 2 F.3d 508, 515-16
(3d Cir. 1993); Republican Party of Oregon v. Keisling, 959 F.2d 144, 145-46 (9th
Cir. 1992). The Supreme Court has never articulated a hard-and-fast standard
for how much of this type of disenfranchisement is too much, nor did the
Baldus plaintiffs offer any concrete standard to which we might turn.
Although the GAB suggested that earlier maps drawn by courts for
Wisconsin have established a floor of 500,000, or even 750,000, for
permissible moves, we reject that proposition. These numbers cannot be
assessed in a vacuum, and we have no indication of any other factors that
might have compelled these significant numbers. Each case, and each decade,
should be assessed on its own record, and factors like the number of people
moved, the overall population shifts in the state (both internally and from
out-of-state), the impact on particular demographic groups, and comparable
points, will all enter into the assessment. It is important to us here that the
evidence presented at trial did not indicate that any particular group will
suffer more disenfranchisement than the remainder of the population. While
we are sympathetic to the nearly 300,000 voters who have lost their
opportunity to vote for a state senator for two years, we find that Act 43 does
not violate the Equal Protection Clause on this basis.
3.4

Claims Four and Five: Congressional Districts Fail
Constitutional Standards for Compactness, Communities of
Interest, and Partisan Gerrymandering

The intervenor-plaintiffs (the three Democratic members of Congress
from Wisconsin) assert that Act 44 violates Reynolds by focusing on
Page 20 of 38

population equality to the detriment of other principles, especially that of
effective representation. They had no other choice, given the fact (as the
parties stipulated) that Act 44 apportions the 2010 census population of the
state of Wisconsin perfectly. Lacking any evidence of population deviation
whatsoever, the intervenor-plaintiffs have no traction on this aspect of their
Equal Protection Clause claim. Whatever else may have happened in
Wisconsin, it has without a doubt preserved the one person, one vote
principle for its citizens.
Second, the intervenor-plaintiffs argue that the Republican majority’s
legislative leadership in the Wisconsin legislature systematically created
congressional districts to give their party an unfair electoral advantage in an
attempt to preserve political majorities. The intervenor-defendants demur to
that point, asserting frankly that there is nothing wrong with political
considerations motivating redistricting. They further argue that the
intervenor-plaintiffs did not offer a workable standard for the court to use in
evaluating the political gerrymandering claim.
Justice Kennedy made a similar comment in his opinion concurring in
the judgment in Vieth v. Jubelirer, 541 U.S. 267 (2004). Writing for a plurality,
Justice Scalia had argued that Davis v. Bandemer, 478 U.S. 109 (1986), which
had recognized the possibility of a constitutional claim based exclusively on
the existence of partisan political gerrymandering, should be overruled. But
five members of the Court did not agree with him, even though it was also
the case that they could not agree on exactly what legal standard ought to
apply in these cases. Interestingly, however, Justice Kennedy’s pivotal
opinion on this point appeared to throw the ball to the litigating parties to
come up with a manageable legal standard.

Page 21 of 38

Whether the parties bear full responsibility for the development of the
law, or if the Court shares that duty, is a topic beyond the scope of this case.
We do note that the right to vote is an individual right, not a group right, see
Burdick v. Takushi, 504 U.S. 428 (1992) (strict scrutiny is required for election
laws that impose a severe restriction on an individual’s right to vote). And
few acts in a democracy are more expressive than the individual’s marking
a ballot (in whatever way it is done these days) to indicate which candidates
he or she would like to see win the race. If, as the Supreme Court has held,
the First Amendment protects persons from politically-based hiring
decisions, see Elrod v. Burns, 427 U.S. 347 (1976), O’Hare Truck Service, Inc. v.
City of Northlake, 518 U.S. 712, 717 (1996), Board of County Com’rs, Wabaunsee
County v. Umbehr, 518 U.S. 668 (1996), Rutan v. Republican Party of Illinois, 497
U .S. 62 (1990), then perhaps the Court will find some day that the First
Amendment also protects persons against state action that intentionally uses
their partisan affiliation to affect the weight of their vote. Legislative districts
drawn behind a Rawlsian veil of ignorance would arguably give each voter
the best chance to express his or her views without anyone putting a thumb
on the scale in advance. But those developments, we concede, lie in the
future, and so we return to the case at hand.
For now, we find that the intervenor-defendants have the better of the
argument, because we are unable to discern what standard the intervenorplaintiffs propose. Their failure to offer a workable standard means that no
one has had a chance to test a suggested rule through the adversarial process.
Without a specific proposal on the table, we are unable to evaluate the merits
of this partisan gerrymandering claim. To the extent that the point is about
process rather than results, we add that our review of the drafting of Act 44

Page 22 of 38

leads us to believe that it was a significantly more bipartisan process than
that associated with the drafting of Act 43. As discussed above, Speth did
begin by meeting with the Republican members of Congress to discuss their
priorities and concerns about redistricting. But well before the process was
over, Congressman Ryan consulted his three Democratic colleagues to
discuss their preferences. Speth testified that he attempted to incorporate all
of the feedback (not just the Republican comments) into the draft. He
avoided putting incumbents together in the same district, and he did not flip
districts from majority-Democrat to majority-Republican or vice versa.
Accordingly, we hold that the intervenor-plaintiffs cannot succeed on their
partisan gerrymandering claim.
3.5.

Claim Six: Voting Rights Act Claim of Latinos

This claim, which concerns only Act 43, is the most troubling. As
matters now stand, both the Baldus and the Voces plaintiffs charge only that
the legislative redistricting plan, as it applies to one particular area of
Milwaukee County, violates the rights of Latino voters under Section 2 of the
Voting Rights Act. To succeed, plaintiffs are required to meet the threshold
requirements for such a case spelled out in Thornburg v. Gingles, 478 U.S. 30,
48-51 (1986): (1) the minority groups are sufficiently large and geographically
compact to create a majority-minority district; (2) the minority groups are
politically cohesive in terms of voting patterns; and (3) voting is racially
polarized, such that the majority group can block a minority’s candidate
from winning. If plaintiffs can meet this threshold, the court must evaluate
the totality of the circumstances to determine whether the minority groups
have been denied an equal opportunity to participate in the political process

Page 23 of 38

and elect candidates of their choice. 42 U.S.C. § 1973(b); Gingles, 478 U.S. at
44-45.
The defendants argue that the districts drawn in Act 43 give Latinos
60.52% of the voting age population in New Assembly District 8 and 54.03%
of the voting age population in New Assembly District 9. As the trial
unfolded, however, it appeared that they conceded that the relevant measure
is citizen voting age population, at least for an ethnic group with as high a
proportion of lawful non-citizen residents as the Latinos. This is correct. For
the obvious reason that non-citizens are not entitled to vote, we cannot
ignore citizenship status, particularly given the Supreme Court’s express
endorsement of the centrality of this point. League of United Latin American
Citizens v. Perry, 548 U.S. 399, 429 (2006) (concluding that citizen voting age
population “fits the language of § 2 because only eligible voters affect a
group’s opportunity to elect candidates”).
The defendants’ expert, Dr. Gaddie, chose not to testify on this claim.
He was, however, involved in the drafting process for Assembly Districts 8
and 9. Foltz testified that Dr. Gaddie instructed the drafters on how to draw
the Latino districts in the way he believed was appropriate, but that the
drafters did not follow his instructions for the final version. Rather, Foltz
said, changes from Dr. Gaddie’s recommendations were made in response
to MALDEF’s input. Ottman testified that Dr. Gaddie “looked at some of the
minority district configurations that we had prepared and kind of evaluated
them.”
Turning to the Gingles factors, the parties have stipulated that the
Latino community in the area of Milwaukee covered by both former and
New Assembly Districts 8 and 9 satisfies the first criterion (i.e., the Latino

Page 24 of 38

group is sufficiently large and geographically compact to create a majorityminority district). Professor Grofman, the defendants’ lead Section 2 expert,
testified at trial that he agrees with the plaintiffs that they have also satisfied
the second requirement (i.e., that the Latinos in Milwaukee are politically
cohesive in their voting behavior). Finally, Dr. Grofman agreed that we may
accept Dr. Mayer’s racial polarization analysis for the third inquiry (i.e., that
voting is racially polarized, such that the majority group can block the Latino
candidate from winning). In fact, when asked whether “the issue in this case
is more about the totality of the circumstances” than the Gingles factors, Dr.
Grofman agreed. We see no reason to disagree with this assessment, which
as far as it goes is shared by Dr. Mayer.
Inquiry into the totality of the circumstances inevitably requires us to
get into the weeds and decide, based on all of the facts in the record, whether
Latinos in the vicinity of New Assembly Districts 8 and 9 have been denied
an equal opportunity to participate in the political process and elect
candidates of their choice. 42 U.S.C. § 1973(b). The parties do not dispute that
Milwaukee’s Latino community bears the socioeconomic effects of historic
discrimination in employment, education, health, and other areas, and that
its depressed socioeconomic status hinders its ability to participate in the
electoral process on an equal basis with other members of the electorate.
The dispute surprisingly centers on whether two Latino influence
districts are superior to one majority-minority district.3 The defendants assert
3

“In majority-minority districts, a minority group composes a numerical,
working majority of the voting-age population. …At the other end of the spectrum
are influence districts, in which a minority group can influence the outcome of an
election even if its preferred candidate cannot be elected.” Bartlett v. Strickland, 556
U.S. 1, 13 (2009). In light of LULAC, supra, we understand the Court to be referring
to citizen voters, where that qualification is pertinent.

Page 25 of 38

that “[t]he Latino community itself is divided on this point.…[S]ome
members want the chance to have a second seat, they want 8 and 9 as they
were prepared.” There is a preliminary legal question to be answered,
however, which will dictate whether this argument can prevail. It is whether,
in a Section 2 claim, a state is entitled to deprive a minority group of one
majority-minority district and substitute for that two influence districts. We
have searched both Supreme Court decisions and those of other courts
around the country, and we cannot find anything holding that this is an
acceptable trade-off. In fact, the Supreme Court specified in Bartlett v.
Strickland that “[u]nder present doctrine, § 2 can require the creation of
[majority-minority] districts” but that “[t]his Court has held that § 2 does not
require the creation of influence districts.” Bartlett, 556 U.S. at 13 (2009). We
interpret the Court’s language to mean that if we find a Section 2 violation,
the creation of influence districts in lieu of a majority-minority district is not
on the menu of options for relief.
The defendants argue that “[o]thers, such as Voces, appear to want to
make a 100-percent guarantee in 8 sacrificing the influence that was given to
them in 9.” But this argument flies in the face of Section 2’s protection against
cracking minority populations–in a sense, its assurance that a bird in the
hand really is better than two in the bush, even though everyone realizes that
a good hunter might actually snare both of the latter. The fundamental
question for a Section 2 claim is whether the redistricting plan in Act 43
provides Latino voters with an opportunity to elect a candidate of choice.
LULAC, 548 U.S. at 430-31.
Dr. Mayer estimates that eligible Latinos constitute 47.07% of New
Assembly District 8’s and 40.52% of New Assembly District 9’s citizen voting

Page 26 of 38

age population. Taking into account not only that number, but also actual
voting experience in the races for the 2011 primary for Milwaukee County
Circuit Judge, the 2008 State Superintendent of Public Instruction general
election, the 2008 12th Aldermanic race, the 2008 Milwaukee City Attorney
race, and the 2004 State Senate election, he concludes that this number is not
enough to create the opportunity that Section 2 mandates. He also identified
36 elections since 1989 in which one or more Latino candidates ran against
one or more Caucasian, non-Latino candidates, and showed that only four
Latino candidates won over this time period, which represents only 11.1%
success by Latino candidates. He proposes an alternative Assembly District
8 with a Latino voting age population of 70.07%, which he estimates amounts
to 60.06% citizen voting age population.
One of the defendants’ other experts, Peter Morrison, largely agreed
with Dr. Mayer’s conclusions. Mr. Morrison, a demographer, estimates that
by November 2012 Latinos will constitute at least 44.9% of the citizen voting
age population in New Assembly District 8; that number coincides with Dr.
Mayer’s estimate. Morrison estimates that the Latino share of the citizen
voting age population is increasing at a rate of at least 1.1% annually, which
means Latinos will continue to lack an effective majority in New Assembly
District 8 through 2018. The parties’ experts also agree that Latinos do not
have an effective majority in New Assembly Districts 8 and 9, but that if the
lines were drawn differently, they could immediately achieve a majorityminority district in a reconfigured Assembly District 8.
In light of this evidence, coupled with the fact that voting is racially
polarized and cohesive in this area, it is apparent that Latino voters have a
distinctly better prospect of electing a candidate of choice with one majority-

Page 27 of 38

minority district than with two influence districts. Notably, the 25,590
individuals who were added to New Assembly District 8 include a high
percentage of Caucasian voters who come from neighborhoods where the
effects of past discrimination are less burdensome than those experienced by
the Latinos from the predecessor Assembly District 8. The newly
added voters–who represent 45% of the New Assembly District 8–are
approximately 41% non-Latino and are expected to continue to engage in
racially polarized voting. Moreover, given the lower degree of historic
discrimination, they are more likely to register and cast a ballot on election
day. Dr. Mayer testified that the voter turnout rates among the newly-added
Caucasian voters in New Assembly District 8 are higher by a factor of 10
when compared to Latino voters in that new district.
This is where our earlier observations about community of interest
come back into play. The evidence shows that the new lines for Districts 8
and 9 will be disruptive to the Latino community of interest. This is so
despite Professor Grofman’s prediction that the voters of New District 8 are
likely to support candidates from the Democratic party. But the Democratic
candidate favored by the Latino community will not necessarily be the same
as the Democratic candidate favored by the new non-Latino voters in the
district. The latter are people who, as the record shows, have a vastly higher
turn-out rate than do the Latinos. In other words, we may simply have a
situation in which the real race is at the primary level, not during the general
election, but all of the same problems will simply be pushed back one stage.
As the Supreme Court pointed out in O’Brien v. Brown, 409 U.S. 1, 15-16
(1972), quoting from United States v. Classic, 313 U.S. 299, 308 (1941), “where
the state law has made the primary an integral part of the procedure of

Page 28 of 38

choice, or where in fact the primary effectively controls the choice, the right
of the elector to have his ballot counted at the primary, is likewise included
in the right protected by Article I, § 2.”
Dr. Grofman argues that Latinos in New Assembly District 8 can elect
their candidate of choice because the district is more properly understood as
a coalition district. In a coalition district, “two minority groups form a
coalition to elect the candidate of the coalition’s choice.” Bartlett, 556 U.S. at
13. Dr. Grofman believes that non-Latino minorities in New Assembly
District 8–specifically African-American voters–will support the Latinos’
candidate of choice. This argument is more commonly presented by plaintiffs
seeking to protect the minority coalition’s Section 2 rights. Nixon v. Kent
County, 76 F.3d 1381 (6th Cir. 1996) (en banc). That said, the Supreme Court
has suggested that a proven coalition district may dodge Section 2
intervention. Johnson v. De Grandy, 512 U.S. 997, 1020 (1994) (“If the lesson of
Gingles is that society’s racial and ethnic cleavages sometimes necessitate
majority-minority districts to ensure equal political and electoral
opportunity, that should not obscure the fact that there are communities in
which minority citizens are able to form coalitions with voters from other
racial and ethnic groups, having no need to be a majority within a single
district in order to elect candidates of their choice.”).
Certainly, if the GAB had offered concrete evidence demonstrating
that New Assembly District 8 is a coalition district, such a showing would
have supported a finding of no Section 2 violation. But Dr. Grofman did not
conduct a racial polarization analysis, for Latinos or any other minority
community. The only racial polarization analysis in the record is Dr. Mayer’s,
where he independently examined the Latino and African-American

Page 29 of 38

communities. While the plaintiffs have abandoned their Section 2 claim on
behalf of African-American voters, we note that Dr. Mayer showed that
African-Americans tend to vote nearly unanimously for African-American
candidates, whereas Caucasian voters were “uniformly less likely to support
the African-American candidate, often by huge margins.” Dr. Mayer also
testified at trial that there is no evidence of coalition building in New
Assembly District 8 among Latino, African American, or Asian voters. He
concluded, in fact, that “there’s quite a bit of tension” among the distinct
racial groups. Testimony by Christine Neumann-Ortiz, the founder of Voces
de la Frontera, and Pedro Colon, Milwaukee County Circuit Court Judge,
support Dr. Mayer’s conclusion. Presented with this record, we cannot make
the leap that African-American voters would prefer a Latino candidate to a
Caucasian candidate merely because they tend to prefer African-American
candidates to a Caucasian candidate. Dr. Grofman would have us rely on his
hunch that African-Americans would vote for Latino candidates, but that is
plainly inadequate. Section 2 rights are too valuable to be evaluated on an
expert’s unsubstantiated prediction.
Dr. Grofman also argues that there is no Section 2 violation because
the current Assemblywoman, Jocasta Zamarripa, from the former Assembly
District 8 is Latina. He did not estimate the extent of such an incumbency
advantage and whether it sufficiently counteracts Dr. Mayer’s concerns with
low registration and low voter turnout among Latino voters. It is no matter,
however, because Dr. Grofman’s supposition utterly ignores the radical
reconfiguration that the New Assembly District 8 imposes. Assemblywoman
Zamarripa is not an incumbent with respect to fully 45% of the population
of New Assembly District 8. It seems to this court that the alleged

Page 30 of 38

incumbency advantage tracks the racial divide, thus rendering its
significance minimal.
Finally, the defendants assert that New Assembly District 8 should
pass muster under the Voting Rights Act because Act 43 is, in certain ways,
consistent with this court’s 2002 map. The former Assembly District 8 had a
Latino voting age population in 2002 of 58.3%, which is less than the New
Assembly District 8’s 60.52% Latino voting age population. We first point out
that the Supreme Court did not highlight the importance of focusing on
citizenship status until 2006. LULAC, 548 U.S. at 429. Since 2006, we have
been required to take real voting majorities into consideration; just as we do
not include children in those numbers, we cannot include non-citizens who
do not enjoy the franchise. Second, the record shows that the Latino
community’s success under the 2002 map is mixed at best. We are not tied
down by history when better evidence of the present and likely future is
before us, and when the last decade has produced demographic changes that
now make it possible to draw an effective majority-minority district. We find
persuasive, in this context, the experience of the 2008 Milwaukee City
Attorney race between Grant Langley (Caucasian) and Pedro Colon, in which
Langley won the position. When Colon ran, he had been the Assemblyman
from the former Assembly District 8 for 10 years. He was on the joint finance
committee during his tenure, and had previously run for mayor, and thus
was hardly an unknown to Milwaukee’s voters. Colon won nearly every
ward in the former Assembly District 8. He lost every ward in those areas
that represent the former Assembly District 9–areas that would be added to
New Assembly District 8 by Act 43. Whether or not this election result was,
as Dr. Mayer put it, “a dry run of what the future holds under Act 43,” we

Page 31 of 38

cannot turn a blind eye to this evidence, which supports the need for a
functioning majority-minority district for Milwaukee’s Latino community,
not just one or two influence districts.
3.6

Claim Nine: Use of the New Districts in Any Future Recall
Election Before November 6, 2012

Plaintiffs ask the court to declare as unconstitutional and enjoin the
conduct of any special or recall elections under Act 43 prior to November 6,
2012. The defendants counter that no case or controversy exists because the
GAB Board does not intend to conduct recall elections in accord with the
legislative districts created by Act 43. At first we had trouble understanding
why this claim reflected any kind of controversy between the parties, because
the GAB has insisted that it intends to conduct the recall elections under the
2002 district lines, just as plaintiffs want. But the plot thickens when we
realize that there is pending litigation in the state courts of Wisconsin in
which some Republican plaintiffs have sued the GAB to compel it to conduct
the recall elections using the 2012 districts.
This puts us in a difficult position. On the one hand, there is no
authoritative statement from the state (either its legislature, or a court
proceeding, or an administrative order from the GAB) with which any decree
of this court would conflict. But there’s the rub: there is also nothing concrete
on which any such decree could operate. We have concluded, based on the
GAB’s formal representations to us in the present proceeding, that there is
no question ripe for determination before us at this time. We take the GAB
at its word that it will use the 2002 districts. This is sensible, especially in light
of the command in the Wisconsin Constitution not to re-district more than
once each 10 years. State ex rel. Smith v. Zimmerman, 266 Wis. 307, 63 N.W.2d

Page 32 of 38

52 (1954).4 District lines may be close to perfect from a population standpoint
when they are initially drawn, but they slip away from perfection as time
goes on, people are born, die, move, and become naturalized citizens. Both
the state and the federal Constitutions recognize that line-drawing is
essential, and they have both chosen a 10-year period for that line. Taking
that into account, it becomes clear that there is nothing unconstitutional at all
about the 2002 districts for the period of time between the adoption of the
map based on the 2000 census and the adoption of the map based on the 2010
census. We note as well that we have no authority to enjoin on-going state
court proceedings, see the Anti-Injunction Act, 28 U.S.C. § 2283, and so our
hands are tied with respect to the state court case. If, however, a time comes
when the GAB proposes to take a different action, either on its own or by
virtue of a state court ruling, and there is a live controversy, plaintiffs may
return to this court and present whatever arguments they may have on this
question.
4.

Conclusion
In conclusion, we find that the Baldus and Voces plaintiffs are entitled

to relief on their Section 2 claim concerning New Assembly Districts 8 and 9,
because Act 43 fails to create a majority-minority district for Milwaukee’s
Latino community. Two influence districts have never been held to be an

4

Indeed, the GAB claimed before trial that it is barred by the Wisconsin
Constitution from making any amendments to the redistricting plan for the next ten
years. We saw nothing in the Wisconsin Constitution or in Zimmerman that stood
in the way of further revision by the General Assembly in the context of reaching
a settlement with the plaintiffs, but for present purposes we will take the GAB at
its word that it finds its hands tied to make any changes to the plan whatsoever
until 10 years has elapsed, and assume that this position will also require it to argue
to a competent court that any effort on the part of the legislature to advance the
effective date of Act 43 is blocked by the state constitution.

Page 33 of 38

adequate substitute for such a district under the factual circumstances that
we have before us. This holding is not intended to affect any other district
drawn by Act 43. Indeed, to avoid disrupting other lines, the court
emphasizes that the re-drawing of the lines for Districts 8 and 9 must occur
within the combined outer boundaries of those two districts. Recognizing as
we have throughout this litigation the primary role that the state has in this
area, we are giving the legislature the first opportunity to address this point,
but it must act quickly given the impending elections. This should not be an
impossible task, given that Dr. Mayer has prepared at least one alternative
configuration that should be a useful starting point.
As for the other claims, we find that although the drafting of Act 43
was needlessly secret, regrettably excluding input from the overwhelming
majority of Wisconsin citizens, and although the final product needlessly
moved more than a million Wisconsinites and disrupted their long-standing
political relationships, the resulting population deviations are not large
enough to permit judicial intervention under the Supreme Court’s
precedents. Act 44 has zero population deviation, which is why we find that
the intervenor-plaintiffs have no meritorious “one person, one vote” claim.
The intervenor-plaintiffs’ partisan gerrymandering claim never made it out
of the gate because no workable standard was offered to the court.
Tempers can flare when people are excluded from the political
process, whether they are shut out because of their party affiliation, because
of their race, because of their economic status, or because of any other trait.
Such a contentious atmosphere is neither necessary nor desirable. We know
that it is not necessary, because courts hold themselves to a higher standard
and have succeeded in drawing successful maps time and again. We should

Page 34 of 38

have learned that it is not desirable because of the rancor that it fosters. Some
states, like Iowa and California, have adopted nonpartisan systems that seem
successfully to have overcome this. New York is seriously thinking right now
of taking a similar step, and there has been some talk of it in Wisconsin in the
wake of this litigation. But we must deal with the here-and-now, and we
therefore must acquiesce in the approach that Wisconsin (not alone among
the states in this circuit, we hasten to add–see Committee for a Fair and Balanced
Map v. Illinois State Bd. of Elections, — F. Supp. 2d —, 2011 WL 6318960, (N.D.
Ill. Dec 15, 2011); Radogno v. Illinois State Bd. of Elections, — F. Supp. 2d —,
2011 WL 5025251 (N.D. Ill. Oct. 21, 2011))–has chosen.
Before concluding, the court must finally address a number of motions
that the parties have submitted and that remain outstanding, all of which
may now be dispatched.
The first, the intervenor-defendants’ Motion for Judgment on the
Pleadings (Docket #75), has effectively been granted by the Court’s
determination that all of the plaintiffs’ and consolidated plaintiffs’ Act 44
claims fail. Thus, the motion requires no further ruling beyond dismissing it
as moot. Similarly, the Court is also obliged to deny the defendants’ Motion
for Summary Judgment (Docket #128) as moot, given that, through this
order, the Court has addressed the substance of all of the outstanding claims
in this matter.
Further, the plaintiffs’ and intervenor-plaintiffs’ motions to defer a
judicial decision (Docket #117, #119), in which they invited the Court to delay
ruling on the intervenor-defendants’ Motion for Judgment on the Pleadings
(Docket #75), have also become moot. Through the passage of time, allowing
trial to proceed before rendering a decision on the intervenor-defendants’

Page 35 of 38

motion, the Court effectively deferred its decision. Therefore, the Court will
also deny the motions to defer as moot.
A number of other motions may be taken care of administratively. At
trial, the Court clarified the scope of its prior order relating to the subpoena
issued to James Troupis, essentially granting Mr. Troupis’s motion for
clarification. (Tr. 58:14–60:4 (clarifying scope of Court’s prior order, as
requested by Mr. Troupis in Docket #179)). Thus, that motion (Docket #179)
can be administratively terminated as having been granted at trial.
Additionally, the defendants’ motion in limine (Docket #160) is hereby
administratively terminated. The Court never specifically addressed the
motion at trial, and the parties did not go to great lengths to elicit testimony
regarding anomalies in redistricting boundaries (the subject of the motion in
limine); further, in this order, the Court does not discuss those anomalies.
Therefore, the Court has no reason to grant or deny that motion (Docket
#160), and will thus terminate it without making a decision on its merits.
Finally, the Court must also deny a request from members of the
group Citizens for Fair and Competitive Redistricting to appear as amicus
curiae. (Docket #126). Through counsel, that group submitted a proposed
brief and several maps which, taken together, urge the Court to adopt an
entirely different redistricting plan than the plan adopted by the legislature.
(Docket #126). Heeding the instruction of the United States Supreme Court
that “[r]edistricting is ‘primarily the duty and responsibility of the State,’” the
Court will not tread into the black water of re-drawing the redistricting
boundaries itself. Perry, 132 S. Ct. at 940 (citing Chapman v. Meier, 420 U.S. 1,
27, 95 S. Ct. 751, 42 L. Ed. 2d 766 (1975)). Instead, as discussed above, the
Court will allow the Legislature to sort out the redistricting maps’ infirmities

Page 36 of 38

on its own. The Court will thus deny the amicus’ request to appear without
consideration of the group’s submissions.
Accordingly,
IT IS ORDERED that the plaintiffs’ and intervenor-plaintiffs’ Sixth
Claim for relief be and the same is hereby GRANTED, the Court having
found that New Assembly Districts 8 and 9 violate the Voting Rights Act,
and, accordingly, the Government Accountability Board is hereby
ENJOINED from implementing Act 43 in its current form;
IT IS FURTHER ORDERED that plaintiffs’ and intervenor-plaintiffs’
remaining claims be and the same are hereby DISMISSED with prejudice;
IT IS FURTHER ORDERED that the intervenor-defendants’ motion
for judgment on the pleadings (Docket #75) be and the same is hereby
DENIED as moot;
IT IS FURTHER ORDERED that the defendants’ motion for
summary judgment (Docket #128) be and the same is hereby DENIED as
moot;
IT IS FURTHER ORDERED that the plaintiffs’ motion to defer a
judicial decision (Docket #117) be and the same is hereby DENIED as moot;
IT IS FURTHER ORDERED that the intervenor-plaintiffs’ motion to
defer a judicial decision (Docket #119) be and the same is hereby DENIED as
moot;
IT IS FURTHER ORDERED that James Troupis’ motion for
clarification (Docket #179), having been granted at trial, be and the same is
hereby TERMINATED administratively;
IT IS FURTHER ORDERED that the defendants’ motion in limine, as
related to the presentation of evidence related to redistricting anomalies

Page 37 of 38

(Docket #160), having not been addressed at trial or in this Order, be and the
same is hereby TERMINATED administratively;
IT IS FURTHER ORDERED that the intervenor-defendants’ motion
to dismiss for lack of standing (Docket #198) be and the same is hereby
DENIED;
IT IS FURTHER ORDERED that the motion of Citizens for Fair and
Public Redistricting to appear as amicus curiae, (Docket #126) be and the same
is hereby DENIED; and
IT IS FURTHER ORDERED that each party is to bear its own costs.
The Clerk is directed to enter judgment accordingly.

Page 38 of 38