The League of Women Voters of Florida, etc. v. Ken Detzner Opinion

Supreme Court of Florida

____________

No. SC14-1905

____________

THE LEAGUE OF WOMEN VOTERS OF FLORIDA, etc., et al.,

Appellants/Cross–Appellees,

vs.

KEN DETZNER, et al.,

Appellees/Cross–Appellants.

[July 9, 2015]

PARIENTE, J.

In this appeal involving legal issues of first impression, we review a trial

court’s finding that the 2012 “redistricting process” and the “resulting map”

apportioning Florida’s twenty-seven congressional districts were “taint[ed]” by

unconstitutional intent to favor the Republican Party and incumbent lawmakers.1

1. This Court previously considered two issues arising out of the pre-trial

discovery process—one concerning the legislative privilege and the other

concerning the discovery of documents in the possession of non-party political

consultants—and released three opinions while the litigation was pending. See

League of Women Voters of Fla. v. Fla. House of Representatives (Apportionment

IV), 132 So. 3d 135, 138 (Fla. 2013) (addressing and largely rejecting claims of

legislative privilege); League of Women Voters of Fla. v. Data Targeting, Inc.

(Apportionment V), 140 So. 3d 510, 514 (Fla. 2014) (permitting the use during

trial of evidence obtained from non-party political consultants, pending further

Cognizant that this Court’s role is not to select a redistricting map that performs

better for one political party or another, but is instead to uphold the purposes of the

constitutional provision approved by Florida voters to outlaw partisan intent in

redistricting, the crux of what we must decide is whether the trial court gave the

appropriate legal effect to its finding that the Florida Legislature drew the state’s

congressional districts in violation of the Florida Constitution.

Added to the Florida Constitution in 2010, the Fair Districts Amendment

sought to eliminate the age-old practice of partisan political gerrymandering—

where the political party and representatives in power manipulate the district

boundaries to their advantage—by forbidding the Florida Legislature from drawing

a redistricting plan or an individual district with the “intent to favor or disfavor a

political party or an incumbent.” Art. III, § 20(a), Fla. Const. “The desire of a

political party to provide its representatives with an advantage in reapportionment

is not a Republican or Democratic tenet, but applies equally to both parties.” In re

Senate Joint Resolution of Legislative Apportionment 1176 (Apportionment I), 83

So. 3d 597, 615 (Fla. 2012). As observed when a three-judge panel of a federal

district court examined Florida’s last decennial congressional redistricting plan in

appellate review); Bainter v. League of Women Voters of Fla. (Apportionment

VI), 150 So. 3d 1115, 1117 (Fla. 2014) (upholding trial court ruling ordering

production of documents in the possession of non-party political consultants).

-2-

2002, the “raw exercise of majority legislative power does not seem to be the best

way of conducting a critical task like redistricting, but it does seem to be an

unfortunate fact of political life around the country.” Martinez v. Bush, 234 F.

Supp. 2d 1275, 1297 (S.D. Fla. 2002).

With the voters’ approval of the Fair Districts Amendment, that unfortunate

fact of political life was banned in Florida. Our citizens declared that the

Legislature must “redistrict in a manner that prohibits favoritism or

discrimination.” Apportionment I, 83 So. 3d at 632. And the Eleventh Circuit

Court of Appeals similarly declared that “[f]ar from dictat[ing] electoral outcomes,

the provision seeks to maximize electoral possibilities by leveling the playing

field.” Brown v. Sec’y of State of Fla., 668 F.3d 1271, 1285 (11th Cir. 2012)

(internal quotation marks omitted).

Like the voters of Arizona, who adopted an independent redistricting

commission recently upheld by the United States Supreme Court as consistent with

the “fundamental premise that all political power flows from the people,” the

Florida voters endeavored “to address the problem of partisan gerrymandering—

the drawing of legislative district lines to subordinate adherents of one political

party and entrench a rival party in power.” Ariz. State Legislature v. Ariz. Indep.

Redistricting Comm’n, No. 13-1314, 2015 WL 2473452, at *4, 21 (U.S. June 29,

2015). As the United States Supreme Court has recognized, “partisan

-3-

gerrymanders . . . [are incompatible] with democratic principles.” Id. at *4

(alteration in original) (quoting Vieth v. Jubelirer, 541 U.S. 267, 292 (2004)

(plurality opinion)). In short, the Fair Districts Amendment was designed “to

restore ‘the core principle of republican government,’ namely, ‘that the voters

should choose their representatives, not the other way around.’ ” Id. at *21

(quoting Mitchell N. Berman, Managing Gerrymandering, 83 Texas L. Rev. 781,

781 (2005)).2

Presented in this case with a first-of-its-kind challenge under the Fair

Districts Amendment, the trial court found that the Legislature’s 2012

congressional redistricting plan was drawn in violation of the Florida

Constitution’s prohibition on partisan intent. We affirm that finding. We

2. We reject the Legislature’s federal constitutional challenge to the Fair

Districts Amendment. The Supreme Court’s recent opinion in the Arizona case

confirms that neither the “Elections Clause” of the United States Constitution, U.S.

Const. art. I, § 4, cl. 1, nor federal law, 2 U.S.C. § 2a(c), prohibits the people of a

state, through the citizen initiative process, from directing the way in which its

congressional district boundaries are drawn. As the Supreme Court explained,

“[b]anning lawmaking by initiative to direct a State’s method of apportioning

congressional districts” would “stymie attempts to curb partisan gerrymandering,

by which the majority in the legislature draws district lines to their party’s

advantage.” Ariz. State Legislature, 2015 WL 2473452, at *20; see also Brown,

668 F.3d at 1280 (rejecting a federal constitutional challenge to the Fair Districts

Amendment based on reasoning wholly consistent with the Supreme Court’s

reasoning in Arizona State Legislature).

-4-

conclude, however, that the trial court failed to give proper legal effect to its

determination that the Fair Districts Amendment was violated.

In reaching this conclusion, we recognize that the trial court had scant

precedent to guide it in approaching the legal issues presented. And, we commend

the trial court for the tremendous effort that was expended in deciding this novel

challenge under the Fair Districts Amendment.

Nevertheless, we conclude that two legal errors significantly affected the

trial court’s determination of the appropriate legal effect of its finding of

unconstitutional intent. First, the trial court erred in determining that there was no

distinction between a challenge to the “plan as a whole”—a challenge, in effect, to

the map produced from the unconstitutional “process”—and a challenge to

individual districts. Second, the trial court erred in the standard of review it

applied, which was improperly deferential to the Legislature’s decisions after

finding a violation of the Fair Districts Amendment’s prohibition on partisan

intent. Although it found the existence of unconstitutional intent, the trial court

relied solely on objective “tier-two” constitutional indicators, such as compactness

and the use of political or geographical boundaries, rather than on the direct and

circumstantial evidence of “tier-one” unconstitutional intent presented at trial.

In other words, the trial court analyzed the Legislature’s map as if it had not

found the existence of unconstitutional intent, affording deference to the

-5-

Legislature where no deference was due. Once a direct violation of the Florida

Constitution’s prohibition on partisan intent in redistricting was found, the burden

should have shifted to the Legislature to justify its decisions in drawing the

congressional district lines.

Relying on the finding of unconstitutional intent, the challengers have urged

that the entire plan should be redrawn. Certain factors support this approach,

which would require the Legislature to begin the redistricting process anew on a

blank slate. For example, we are aware that the starting point for drawing the 2012

congressional redistricting map was the 2002 map, which was drawn prior to the

Fair Districts Amendment with, at that time, legally permissible partisan intent. In

fact, the Legislature itself had, in defending against a racial gerrymandering claim

directed at the 2002 map, “stipulated” that its intent “was to draw the congressional

districts in a way that advantages Republican incumbents and potential

candidates.” Martinez, 234 F. Supp. 2d at 1340. We also acknowledge that a

three-judge federal district court panel concluded that the Florida Legislature’s

“overriding goal with respect to congressional reapportionment” in 2002 was to

“maximize the number of districts likely to perform for Republicans.” Id. at 1300-

01. These are considerations now explicitly outlawed by the Florida Constitution’s

prohibition on partisan political gerrymandering.

-6-

Based on the findings and evidence in this case, however, we ultimately

reject the challengers’ request that the entire plan must be redrawn or that this

Court should, at this time, perform the task of redrawing the districts. Although we

conclude that the trial court’s finding of unconstitutional intent required the burden

to shift to the Legislature to justify its decisions regarding where to draw the lines,

we also conclude that the challengers still must identify some problem with the

Legislature’s chosen configuration. They did so in this case with respect to

Districts 5, 13, 14, 26, and 27—showing a nexus between the unconstitutional

intent and the district—as well as for Districts 21, 22, and 25, which they

contended were problematic either for “tier-two” reasons or because the

Legislature unjustifiably rejected a less favorable configuration.

Accordingly, while we affirm the trial court’s finding that the Legislature’s

enacted map was “taint[ed]” by unconstitutional intent, we reverse the trial court’s

order upholding the Legislature’s remedial redistricting plan. We relinquish this

case to the trial court for a period of 100 days from the date of this opinion, with

directions that it require the Legislature to redraw, on an expedited basis,

Congressional Districts 5, 13, 14, 21, 22, 25, 26, 27, and all other districts affected

by the redrawing, pursuant to the guidelines set forth in this opinion. We

emphasize the time-sensitive nature of these proceedings, with candidate

qualifying for the 2016 congressional elections now less than a year away, and

-7-

make clear that we take seriously our obligation to provide certainty to candidates

and voters regarding the legality of the state’s congressional districts. Upon the

completion of the redrawing of the map, the trial court shall hold a hearing where

both sides shall have an opportunity to present their arguments and any evidence

for or against the redrawn map, and the trial court shall then enter an order either

recommending approval or disapproval of the redrawn map.

We commend both parties for their professionalism in presenting the case to

this Court and now proceed to discuss in detail the legal issues that have been

raised on appeal, the background of this case, the evidence presented, and our legal

reasoning.

I. CHALLENGE TO TRIAL COURT’S FINDINGS

On appeal in this Court, the challengers seek affirmance of the trial court’s

finding of unconstitutional partisan intent in drawing the state’s congressional

districts—a finding that was based on both direct and circumstantial evidence.

Their primary contention of error, however, is that the trial court applied an unduly

deferential standard of review, thereby precluding it from imposing a more

meaningful remedy for its finding of unconstitutional intent to favor the

Republican Party and incumbents.3

3. The issues raised on appeal by the challengers are: (1) the trial court erred

in requiring only two districts to be redrawn after finding constitutionally improper

intent in the enacted congressional redistricting plan; (2) Congressional Districts 5,

-8-

The Legislature, while seeking affirmance of the trial court’s approval of the

remedial redistricting plan, nevertheless takes issue with the trial court’s finding of

unconstitutional intent.4 In particular, the Legislature contests, first, the trial

court’s finding of a connection between the evidence and the Legislature itself,

13, 14, 21, 22, 25, 26, and 27 are independently unconstitutional; (3) this Court

should craft a meaningful remedy, either by adopting a constitutionally valid plan

or assisting the Legislature so that it can adopt a plan that complies with the

Florida Constitution; and (4) the trial court erred in rejecting the challengers’

attempt to re-open the evidence to introduce additional allegations of improper

partisan intent.

We summarily reject the challengers’ claim regarding the trial court’s denial

of their motion to re-open the evidence. Although the e-mail the challengers

sought to introduce after the close of evidence did provide some additional

circumstantial support for their claim of improper intent, the challengers

themselves have conceded that it was cumulative to other evidence. Thus, while it

may have been relevant evidence and properly introduced during the trial if the

challengers had been able to obtain it sooner, the trial court did not abuse its

discretion in denying the motion to re-open the case, and the challengers were not,

in any event, prejudiced since the trial court found the existence of unconstitutional

intent.

4. The Legislature also raises the following three issues on cross-appeal: (1)

the trial court’s order improperly discourages public participation in the

redistricting process; (2) under the Florida Constitution, the controlling intent is the

intent of the Legislature as a collective body; and (3) article III, section 20, of the

Florida Constitution is invalid because it violates the United States Constitution.

As to the claim regarding public participation, we clarify that we do not read

the trial court’s order as discouraging public input in redistricting. There is

nothing inherently in violation of the law or the Florida Constitution for an

individual to anonymously submit a map to the Legislature for consideration or to

submit a map through a third party. We conclude that any comments by the trial

court to the contrary were made in the specific context of the facts and

circumstances of this case and do not amount to error.

-9-

including the trial court’s decision to ascribe the intent of a few individuals to the

Legislature as a collective body. Second, the Legislature asserts that, even

assuming the existence of unconstitutional intent, the trial court’s finding pertains

solely to the two invalidated districts and not to the broader process or map as a

whole. Accordingly, the Legislature argues that any remedy that may have been

necessary has already been provided through the enactment of the remedial

redistricting plan.

We address these issues in the following way. After setting forth a

comprehensive overview of the factual and legal background of the case, including

a review of the evidence relied on by the trial court in finding unconstitutional

intent, our analysis begins by considering the “intent” standard and the trial court’s

application of that standard in this case. Upon determining that the trial court

appropriately framed the “intent” inquiry, we turn to the legal sufficiency of the

trial court’s finding of unconstitutional intent. We conclude that competent,

substantial evidence supports the trial court’s finding and that this finding pertains

to the plan as a whole and not solely to the two invalidated districts. We then

proceed to consider the proper legal effect of this finding as we review each

challenged district. Finally, we address the remedy.5

5. We conclude—as agreed by both parties—that amici curiae LatinoJustice

PRLDEF, Florida New Majority, and Mi Familia Vota lack standing to challenge

the validity of Congressional District 9. Amici curiae did not appear in the trial

- 10 -

II. THE FLORIDA CONSTITUTION’S PROHIBITION ON PARTISAN

POLITICAL GERRYMANDERING

In February 2012, “the Florida Legislature approved the decennial plan

apportioning Florida’s twenty-seven congressional districts, based on population

data derived from the 2010 United States Census.” League of Women Voters of

Fla. v. Fla. House of Representatives (Apportionment IV), 132 So. 3d 135, 139

(Fla. 2013). After the adoption of the Legislature’s 2012 congressional

redistricting plan, two separate groups of plaintiffs (“the challengers”)6 filed civil

complaints in the Second Judicial Circuit Court in and for Leon County,

challenging the validity of the plan under new state constitutional redistricting

court to raise this claim, and it is well-settled that amici are not permitted to raise

new issues. See Riechmann v. State, 966 So. 2d 298, 304 n.8 (Fla. 2007).

6. We use the term “challengers,” which has been used by this Court in

prior opinions during the course of this litigation, to refer collectively to the

plaintiffs in the trial court, who are the Appellants/Cross–Appellees in this Court.

These litigants that challenged the constitutionality of the congressional

redistricting plan enacted in 2012 include two separate groups, which have

described themselves as the “Coalition plaintiffs” and the “Romo plaintiffs.” The

“Coalition plaintiffs” consist of the League of Women Voters of Florida, Common

Cause, and four individually named parties. The National Council of La Raza was

formerly a member of the “Coalition plaintiffs” but later voluntarily dismissed all

claims and withdrew as a party in the case prior to the trial. The “Romo plaintiffs”

consist of lead plaintiff Rene Romo and six other individually named parties.

There has rarely been a need to distinguish between the two groups for purposes of

the issues to come before this Court, and the circuit court consolidated the two

lawsuits filed by these groups that challenged the Legislature’s 2012 congressional

redistricting plan.

- 11 -

standards approved by the Florida voters in 2010 and now enumerated in article

III, section 20, of the Florida Constitution. “Those standards, governing the

congressional reapportionment process, appeared on the 2010 general election

ballot as ‘Amendment 6’ and, together with their identical counterparts that apply

to legislative reapportionment (‘Amendment 5’), were generally referred to as the

‘Fair Districts’ amendments.” Id.7 As this Court has previously noted, “[t]here is

no question that the goal of minimizing opportunities for political favoritism was

the driving force behind the passage of the Fair Districts Amendment.”

Apportionment I, 83 So. 3d at 639.

In Apportionment I, during this Court’s first review involving the new

constitutional standards, we commended the Legislature for what it claimed at that

time to be an unprecedented transparent redistricting process, in which the

Legislature engaged in twenty-six public hearings around the state and obtained

public input as it went about its task of redistricting. See 83 So. 3d at 637 n.35,

664. In truth, public input in redistricting was not unique to the 2012 process. The

Legislature held thirty-three public hearings during the 1992 redistricting and

7. “Amendment 5 is now codified in article III, section 21, of the Florida

Constitution. The standards in article III, section 20—governing congressional

reapportionment—and those in article III, section 21—governing legislative

reapportionment—are identical.” Id. at 139 n.1.

- 12 -

twenty-four public hearings prior to the enactment of the 2002 map. See Martinez,

234 F. Supp. 2d at 1288.

Based on the new constitutional standards that applied for the first time to

the 2012 process, transparency became legally significant under the Florida

Constitution. This Court explained that “if evidence exists to demonstrate that

there was an entirely different, separate process that was undertaken contrary to the

transparent effort in an attempt to favor a political party or an incumbent in

violation of the Florida Constitution, clearly that would be important evidence in

support of the claim that the Legislature thwarted the constitutional mandate.”

Apportionment IV, 132 So. 3d at 149. Indeed, the challengers’ principal claim in

this litigation challenging the constitutional validity of the Legislature’s 2012

congressional redistricting plan involved evidence of the type of “entirely different,

separate process” this Court warned would be “important evidence” of a

constitutional violation.

Specifically, the challengers argued that the Legislature cooperated and

collaborated with partisan political operatives aligned with the Republican Party to

produce a redistricting plan that was drawn in contravention of article III, section

20, with the intent to favor incumbents and the Republican Party, which was the

controlling political party in the Legislature at the time of the 2012 redistricting.

Before the approval of the Fair Districts Amendment, this Court had previously

- 13 -

acknowledged, in 1992, that there was “little doubt that politics played a large

part” in the adoption of prior redistricting plans in this state, explaining that the

protection of incumbents and favoritism of one party over another was inevitable—

and certainly “not illegal.” In re Senate Joint Resolution 2G, Special

Apportionment Session 1992 (In re Apportionment Law—1992), 597 So. 2d 276,

285 (Fla. 1992). But at that time, such partisan intent was not legally prohibited.

The acceptability of partisan political gerrymandering in this state

dramatically changed in 2010. With “fairness” as its “focus,” the Fair Districts

Amendment now “expressly prohibits” redistricting “practices that have been

acceptable in the past, such as crafting a plan or district with the intent to favor a

political party or an incumbent.” Apportionment I, 83 So. 3d at 605, 607, 616.

These “express new standards” thus afford Florida citizens “explicit constitutional

protection” under article III, section 20, of the Florida Constitution, “against

partisan political gerrymandering.” Apportionment IV, 132 So. 3d at 138-39.

Specifically, article III, section 20, of the Florida Constitution, provides in

its entirety as follows:

In establishing congressional district boundaries:

(a) No apportionment plan or individual district shall be drawn

with the intent to favor or disfavor a political party or an incumbent;

and districts shall not be drawn with the intent or result of denying or

abridging the equal opportunity of racial or language minorities to

participate in the political process or to diminish their ability to elect

representatives of their choice; and districts shall consist of

contiguous territory.

- 14 -

(b) Unless compliance with the standards in this subsection

conflicts with the standards in subsection (a) or with federal law,

districts shall be as nearly equal in population as is practicable;

districts shall be compact; and districts shall, where feasible, utilize

existing political and geographical boundaries.

(c) The order in which the standards within subsections (a) and

(b) of this section are set forth shall not be read to establish any

priority of one standard over the other within that subsection.

Art. III, § 20, Fla. Const.

Under article III, section 20, “there is no acceptable level of improper

intent.” Apportionment I, 83 So. 3d at 617. The prohibition on improper partisan

intent in redistricting applies, “by its express terms,” to “both the apportionment

plan as a whole and to each district individually” and does not “require a showing

of malevolent or evil purpose.” Id. A finding of partisan intent therefore renders

the Legislature’s redistricting plan constitutionally invalid, as the Florida

Constitution expressly “outlaw[s] partisan political gerrymandering.”

Apportionment IV, 132 So. 3d at 137. As we explained in Apportionment I:

The Florida Constitution now expressly prohibits what the

United States Supreme Court has in the past termed a proper, and

inevitable, consideration in the apportionment process.

Florida’s express constitutional standard, however, differs from

equal protection political gerrymandering claims under either the

United States or Florida Constitutions. Political gerrymandering

claims under the Equal Protection Clause of the United States

Constitution focus on determining when partisan districting as a

permissible exercise “has gone too far,” so as to “degrade a voter’s or

a group of voters’ influence on the political process as a whole.”

In contrast to the federal equal protection standard applied to

political gerrymandering, the Florida Constitution prohibits drawing a

plan or district with the intent to favor or disfavor a political party or

- 15 -

incumbent; there is no acceptable level of improper intent. It does not

reference the word “invidious” as the term has been used by the

United States Supreme Court in equal protection discrimination cases,

and Florida’s provision should not be read to require a showing of

malevolent or evil purpose.

83 So. 3d at 616-17 (citations omitted).

“Florida’s constitutional provision prohibits intent, not effect,” which is to

say that a map that has the effect or result of favoring one political party over

another is not per se unconstitutional in the absence of improper intent. Id. at 617.

“Thus, the focus of the analysis must be on both direct and circumstantial evidence

of intent.” Id. “One piece of evidence in isolation may not indicate intent, but a

review of all of the evidence together may lead this Court to the conclusion that the

plan was drawn for a prohibited purpose.” Id. at 618. The relevant inquiry for

discerning improper partisan intent “focuses on whether the plan or district was

drawn with this purpose in mind.” Id.

A. TRIAL COURT’S FINDING OF UNCONSTITUTIONAL INTENT

The challengers’ claim of unconstitutional intent in the enacted

congressional redistricting plan was that the Legislature communicated and

collaborated with partisan political operatives, in the shadow of the Legislature’s

purportedly open and transparent redistricting process, to produce a map favoring

Republicans and incumbents. After hearing all the evidence presented during a

twelve-day bench trial held from late May to early June 2014, and evaluating the

- 16 -

credibility of all the witnesses, the trial court found that the challengers had proven

their case and concluded that the Florida Legislature’s enacted 2012 congressional

redistricting plan was drawn in violation of article III, section 20.

The introductory paragraph of the trial court’s judgment stated that “districts

5 and 10 were drawn in contravention of the constitutional mandates of Article III,

Section 20,” but, in its discussion throughout the course of its forty-one-page order,

the trial court more generally referred to and found that a group of partisan

political operatives “conspire[d] to manipulate and influence the redistricting

process” and succeeded in “infiltrat[ing] and influenc[ing] the Legislature, to

obtain the necessary cooperation and collaboration” to “taint the redistricting

process and the resulting map with improper partisan intent.” (Emphasis supplied.)

Specifically, the trial court stated, in pertinent part, as follows:

[The challengers’] theory of the case regarding improper intent

is that Republican leadership in the House and the Senate, their key

staff members, and a small group of Republican political consultants

conspired to avoid the effective application of the Fair District

Amendments to the redistricting process and thereby successfully

fashioned a congressional map that favors the Republican Party and

its incumbents. The strategy they came up with, according to the

[challengers], was to present to the public a redistricting process that

was transparent and open to the public, and free from partisan

influences, but to hide from the public another secretive process. In

this secretive process, the political consultants would make

suggestions and submit their own partisan maps to the Legislature

through that public process, but conceal their actions by using proxies,

third persons who would be viewed as “concerned citizens,” to speak

at public forums from scripts written by the consultants and to submit

- 17 -

proposed maps in their names to the Legislature, which were drawn

by the consultants.

What is clear to me from the evidence, as described in more

detail below, is that this group of Republican political consultants or

operatives did in fact conspire to manipulate and influence the

redistricting process. They accomplished this by writing scripts for

and organizing groups of people to attend the public hearings to

advocate for adoption of certain components or characteristics in the

maps, and by submitting maps and partial maps through the public

process, all with the intention of obtaining enacted maps for the State

House and Senate and for Congress that would favor the Republican

Party.

They made a mockery of the Legislature’s proclaimed

transparent and open process of redistricting by doing all of this in the

shadow of that process, utilizing the access it gave them to the

decision makers, but going to great lengths to conceal from the public

their plan and their participation in it. They were successful in their

efforts to influence the redistricting process and the congressional

plan under review here. And they might have successfully concealed

their scheme and their actions from the public had it not been for the

[challengers’] determined efforts to uncover it in this case.

The closer question is whether the Legislature in general, or the

leadership and staff principally involved in drawing the maps,

knowingly joined in this plan, or were duped by the operatives in the

same way as the general public. The Defendants argue that if such a

conspiracy existed, there is no proof that anyone in the Legislature

was a part of it. If portions of the operatives’ maps found their way

into the enacted maps, they say, it was not because leadership or staff

were told or knew they came from this group, but rather because the

staff, unaware of their origins, saw the proposals as improving the

draft maps they were working on.

The most compelling evidence in support of this contention of

the Defendants is the testimony of the staff members who did the bulk

of the actual map drawing for the Legislature. I had the ability to

judge the demeanor of Alex Kelly, John Guthrie and Jason Poreda at

trial and found each to be frank, straightforward and credible. I

conclude that they were not a part of the conspiracy, nor directly

aware of it, and that significant efforts were made by them and their

bosses to insulate them from direct partisan influence. I accept that

their motivation in drawing draft maps for consideration of the

- 18 -

Legislature was to produce a final map which would comply with all

the requirements of the Fair District Amendments, as their superiors

had directed them.

That being said, the circumstantial evidence introduced at trial

convinces me that the political operatives managed to find other

avenues, other ways to infiltrate and influence the Legislature, to

obtain the necessary cooperation and collaboration to ensure that their

plan was realized, at least in part. They managed to taint the

redistricting process and the resulting map with improper partisan

intent. There is just too much circumstantial evidence of it, too many

coincidences, for me to conclude otherwise.

(Emphasis supplied.)

Having reviewed the trial court’s factual findings and the record, and

viewing the evidence in the light most favorable to the trial court’s finding of

unconstitutional intent, we set forth the following relevant factual background of

the case. See Berges v. Infinity Ins. Co., 896 So. 2d 665, 676 (Fla. 2004)

(explaining that it “is not the function of this Court to substitute its judgment for

that of the trier of fact”); Markham v. Fogg, 458 So. 2d 1122, 1126 (Fla. 1984)

(stating that an appellate court “should not substitute its judgment for that of the

trier of fact” as long as there is competent, substantial evidence to support the

findings, and concluding upon review of conflicting evidence that there was

“ample credible evidence adduced at the trial to sustain the trial judge’s findings”);

see also Hausdorff v. Hausdorff, 913 So. 2d 1267, 1268 (Fla. 4th DCA 2005)

(viewing the evidence in the light most favorable to the challenged judgment in

evaluating whether competent, substantial evidence supported the trial court’s

- 19 -

rulings); Mesick v. Loeser, 311 So. 2d 132, 136 (Fla. 2d DCA 1975) (findings by

the lower court as a trier of fact come to the appellate court “clothed with a heavy

presumption of correctness and where there is substantial competent evidence to

sustain the actions of the trial court,” the appellate court cannot substitute its own

opinion on the evidence but “must indulge every fact and inference in support of

that judgment,” which is the equivalent of a jury verdict). We note, given the

nature of the challengers’ claim, that circumstantial evidence is often essential in

proving a conspiracy—and indeed may be the only type of evidence available. See

Anheuser-Busch, Inc. v. Campbell, 306 So. 2d 198, 199 (Fla. 1st DCA 1975) (“It is

a well settled rule that circumstantial evidence is admissible in civil conspiracy

cases.”); see also Resnick v. State, 287 So. 2d 24, 26 (Fla. 1973) (holding that a

criminal conspiracy need not be proved by only direct evidence).

As we recount the facts, we emphasize that not every meeting held or every

communication made was improper, illegal, or even violative of the letter of the

Fair Districts Amendment. We set forth the pertinent facts in the record because,

collectively, the evidence that the challengers were able to uncover after a

protracted discovery process demonstrates a different scenario than the entirely

open and transparent process touted by the Legislature when this Court considered

the original apportionment challenges to the state Senate and House maps in

- 20 -

Apportionment I. This is, indeed, what the trial court—which heard and

considered all this evidence—found.

We also emphasize that since many of the e-mails were deleted or destroyed,

we still may have only a partial picture of the behind-the-scenes political tactics.

As the trial court found, “the Legislators and the political operatives systematically

deleted almost all of their e-mails and other documentation relating to

redistricting.” The Legislature did so even though it had acknowledged that

litigation over the redistricting plan was “a moral certainty.” Indeed, if not for the

production of some documents from the political consultants, including Marc

Reichelderfer and Pat Bainter, there would be no record of the separate process

undertaken by the consultants and no way to establish whether or not this process

involved the collusion of the Legislature and ultimately affected the enacted map,

as the trial court concluded.

We further understand that “taking the politics out of politics” is itself a

difficult challenge, considering that partisan political gerrymandering was the

norm for both political parties during prior redistricting processes in this state.

Nevertheless, the facts that we recount provide the backdrop as to why we reject

the Legislature’s defense—which focuses on the political consultants’ efforts to

“influence the redistricting process” and “make themselves relevant” despite their

“exclusion from the decision-making process”—that depicts the political

- 21 -

consultants and a few errant staffers as independent, self-motivated culprits,

individuals who did not have the ability to and did not, in fact, influence the

Legislature’s decisions regarding where to draw the lines. And, finally, we

emphasize that a finding of unconstitutional intent to favor a political party or

incumbent does not necessarily mean that those who made the decisions acted with

“malevolent or evil purpose,” which is not required for a finding of

unconstitutional intent under the Fair Districts Amendment. Apportionment I, 83

So. 3d at 617.

B. EVIDENCE OF UNCONSTITUTIONAL INTENT

A month after the Florida voters approved the Fair Districts Amendment

during the November 2010 general election, then-Speaker of the House Dean

Cannon authorized a meeting in December 2010 at the headquarters of the

Republican Party of Florida, involving Republican political consultants and

legislative staffers, to discuss the upcoming redistricting process. This gathering

was described by one of the consultants at trial as a meeting of “people that, prior

to passage of the [new constitutional standards], would have generally been

involved in the redistricting process.”

The four key political consultants in attendance, who became major figures

in the redistricting trial, were (1) Rich Heffley, (2) Frank Terraferma, (3) Marc

Reichelderfer, and (4) Pat Bainter. Heffley is a consultant who has worked with

- 22 -

many Republican legislators and candidates for public office, including Senator

Don Gaetz, the Chairman of the 2012 Senate Committee on Reapportionment.

Heffley had been involved in prior redistricting processes in Florida in 1992 and

2002 and, by the summer of 2011, was being paid $10,000 per month by the

Republican Party of Florida for unspecified redistricting services. Terraferma is

also a consultant who has worked for a number of Republican legislators and

candidates, including Representative Will Weatherford, the Chairman of the 2012

House Redistricting Committee. Terraferma had previously been hired by Heffley

to work for the Republican Party of Florida and went back to work for the party as

Director of House campaigns in 2011. He was described by employees of a

national Republican organization, in an invitation for a meeting held in

Washington, D.C., in June 2011 with key individuals involved in the redistricting

process, as a “genius map drawer.” Reichelderfer is another consultant who has

worked with several Republican legislators and candidates, including former

Speaker Dean Cannon. Reichelderfer is also one of Cannon’s longtime personal

friends, dating back over twenty years to their days together as Young

Republicans. He was, at the time of the 2012 redistricting, considered part of

Cannon’s “inner circle,” and he had a good working relationship with Heffley.

Bainter is the owner of a Gainesville, Florida, based political consulting firm

known as Data Targeting, Inc., which has as one of its largest clients the

- 23 -

Republican Party of Florida. Between January of 2011 and November of 2012, the

Republican Party of Florida paid Data Targeting, Inc., almost $3 million for

consulting, polling, and direct mail services.

These four consultants, along with employees of the Republican Party of

Florida, met in the initial December 2010 meeting with Alex Kelly, the staff

director for the House Redistricting Committee; Chris Clark, the chief legislative

aide for Senator Gaetz; and attorneys for the Legislature. At a second meeting the

following month, in January 2011, the consultants met with Senator Gaetz,

Representative Weatherford, Alex Kelly, and Kelly’s Senate counterpart, John

Guthrie.

These meetings were not open to the public and there is no record of what

was discussed. As the trial court stated, “[n]o one who testified at trial about [the

meetings] seemed to be able to remember much about what was discussed, though

all seemed to agree that the political consultants were told that they would not have

a ‘seat at the table’ in the redistricting process,” as they had during redistricting in

years past. According to the trial court, “[n]o one clearly articulated what that

meant exactly, but there was testimony that they were told that they could still

participate in redistricting through the public process ‘just like any other citizen.’ ”

Reichelderfer, the consultant who has worked with then-Speaker Cannon,

testified that one topic of discussion at the meetings, as the trial court noted, was

- 24 -

“whether a privilege could be identified to prevent disclosure of redistricting-

related communications among political consultants, legislators, and legislative

staff members.” The conclusion reached at the meetings, according to the trial

court, was “that no privilege would apply.” After the first meeting, in December

2010, Reichelderfer prepared a memorandum that included the following question:

“Communication with outside non-lawyers—how can we make that work?”

Another question included in the Reichelderfer memorandum was,

“Evolution of maps—Should they start less compliant and evolve through the

process—or—should the first map be as near as compliant as possible and change

very little?” Reichelderfer acknowledged at trial that it was “possible” he

discussed with Speaker Cannon the issues identified in this initial memo he

prepared. The trial court would later reference Reichelderfer’s memo in rejecting

part of the Legislature’s argument that there could be “no improper partisan intent

in the drafting of the maps” because, the Legislature asserted, “as things

progressed, each succeeding map that was drawn was an improvement over the one

before it in terms of compactness, leaving cities and counties intact and following

geographical boundaries.” “Coincidentally,” the trial court stated, “that

corresponds with a strategy suggested from Reichelderfer’s notes, i.e., start with

less compliant maps and work toward a more compliant map.”

- 25 -

The trial court found that there was “no reason to convene two meetings just

to tell active political partisans of the Republican Party that they would not ‘have a

seat at the table.’ ” The trial court also noted “a few curious things about these

meetings and their connection to subsequent events that are troubling.”

Specifically, even though the consultants supposedly had no “seat at the

table,” the trial court found that they continued to be involved in the process. In

June 2011, an e-mail was sent from Senator Gaetz’s e-mail address to legislators to

provide information about upcoming public hearings regarding redistricting. A

“blind copy” of this e-mail was sent to Heffley, the consultant under contract with

the Republican Party of Florida, and to Terraferma, the “genius map drawer.” The

trial court found that this was evidence that either Senator Gaetz or “someone in

his office” was “keeping these operatives in the loop.”

Another e-mail, sent in October 2011 from Terraferma to Representative

Weatherford, reported that Kirk Pepper, the Deputy Chief of Staff for then-Speaker

Cannon, was “huddled on a computer” at the Republican Party of Florida’s

headquarters, working with consultant Heffley on “[c]ongressional redistricting if I

had to guess?” Pepper acknowledged at trial that he must have been speaking with

Heffley at the Republican Party of Florida’s headquarters at the time, but stated

that he “never met with Rich Heffley about redistricting.” He had no explanation

as to why Terraferma, whom Pepper had previously worked with at the Republican

- 26 -

Party of Florida, would have thought otherwise. The trial court found that it was

“possible that Terraferma was mistaken or simply speculating without any basis,”

but this communication caused the trial court to “wonder why [Terraferma] would

make this assumption if Pepper really had nothing to do with the redistricting

process.”

As it turned out, Pepper acted as a conduit between the consultants and the

Legislature. According to testimony relied on by the trial court, Cannon staffer

Pepper “regularly” provided advance, non-public copies of draft redistricting maps

to consultant Reichelderfer. The evidence, which came from document production

by Reichelderfer since, as the trial court noted, neither Pepper nor Speaker Cannon

preserved any records, demonstrated that between November 2011 and January

2012, Pepper transmitted to Reichelderfer—through his personal e-mail account, a

“Dropbox” account he later deleted, and a thumb drive—at least twenty-four draft

congressional redistricting maps prepared by the Legislature, mostly before they

were released to the public. In some instances, Pepper sent Reichelderfer maps the

Legislature prepared but never released to the public.

Although Pepper testified at trial that he acted “without Speaker Cannon’s

approval” and, in retrospect, considered his decision to provide Reichelderfer with

maps to have been “a mistake,” Pepper was later hired by Cannon’s private firm

after Cannon left office. Cannon described Pepper as “a loyal employee,” but

- 27 -

testified that he did not know about Pepper’s transmission of maps to Reichelderfer

until it was reported in the media during the litigation in this case.

While they denied doing so, the trial court found that Pepper and

Reichelderfer “communicate[d] about the political performance of the maps.” In

one instance, after Reichelderfer expressed concerns that the draft of a Central

Florida district occupied by incumbent Republican Representative Daniel Webster

was “a bit messed up,” Pepper asked Reichelderfer, “[p]erformance or

geography?” Reichelderfer acknowledged during testimony at trial that

“performance” in that context would “[g]enerally” refer to the political

performance of the district, although there is no record of his response to Pepper.

Reichelderfer testified that he could not recall whether or how he answered that

question. He spoke on the phone “regularly” with Pepper but denied having

“specific conversations about political performance.”

Despite asking, “[p]erformance or geography?” Pepper testified at trial that

he did not want to know from Reichelderfer if there was a problem with the

political performance of that particular district. Instead, he provided a lengthy

explanation that his question was a “sarcastic” response to remind Reichelderfer

“to be quiet,” because they were not supposed to talk about redistricting or the

political implications of certain maps. Pepper stated of his question, “[i]t’s like if

you were talking to someone that you knew very well and had known for a long

- 28 -

period of time, you could say something in writing that other people might take

differently than you meant it.” The trial court discredited Pepper’s explanation as

“very unusual and illogical.”

After receiving maps from Cannon staffer Pepper, Reichelderfer modified

the maps to increase the Republican performance of the districts, and he and the

other consultants traded numerous maps back and forth with each other. Of

significance, the trial court found that some of Reichelderfer’s modifications

corresponded to the actual decisions the Legislature ultimately made.

In one graphic example, cited by the trial court, Reichelderfer’s revisions

changed the performance of Districts 5, 7, 9, and 10 from four Democratic

performing or leaning seats to two Democratic and two Republican performing

seats, as eventually reflected in the actual map enacted by the Legislature. Another

map, which was known to have been drawn by Terraferma, shared eleven identical

districts with a map submitted through the public process by an individual named

Alex Posada, who denied ever creating or submitting the map and stated that he

had not authorized anyone to submit a map using his name.

For his part, Reichelderfer described his interest in the Legislature’s maps as

important to him “professionally” to “know the lay of the land,” similar to

Bainter’s explanation that his interest was an “after-the-fact” one merely for the

sake of his own “[k]nowledge”—even though the evidence presented at trial

- 29 -

demonstrated that the consultants spent considerable time, including weekends,

early mornings, and late nights, making revisions to draft maps, and even though

communications between these consultants regarding the maps referred to having

“a job to do,” wanting to “spread” the maps “around,” and “[h]ead[ing] up” to

Tallahassee to “[t]ell[] folks to look at” certain maps.

The trial court found that the consultants “did their best to evade answering

direct questions” at trial, “often using semantic distinctions to avoid admitting

what they had done.” As this Court previously noted with respect to documents

produced by Bainter that included communications among the consultants

regarding maps, “the documents support[ed] the challengers’ claim that Bainter

was not just drawing maps out of casual ‘after-the-fact interest,’ but was actively

engaged in an extensive process to draw maps favorable to a particular political

party or incumbent and facilitate the submission of those maps to the Legislature

through ‘shell people’ without any indication that the maps were drawn by the

political consultants.” Apportionment VI, 150 So. 3d at 1129. For instance, one e-

mail produced by Bainter stated that a Republican activist in Gainesville was

“getting” him “10 more people at least,” while another e-mail indicated that if one

of the consultants could “think of a more secure and failsafe way to engage our

people, please do it.”

- 30 -

The trial court found that the Bainter documents “evidenced a conspiracy to

influence and manipulate the Legislature into a violation of its constitutional duty”

to redistrict in a neutral, non-partisan fashion, and explained that those documents

were “very helpful” in demonstrating not only that the consultants “were

submitting maps to the legislature” through third parties, but “how extensive and

organized that effort was, and what lengths they went to in order to conceal what

they were doing.” The trial court also found it “hard to imagine” that the

legislative leaders and staffers who allegedly told these consultants that they could

not be involved, other than through the public process, “would not have expected

active participation in the public redistricting process by those political consultants

at the meetings” and would not have questioned both why the consultants were not

in attendance at the public hearings and why none of the maps coming from the

public had any of the consultants’ names on them. “I would think,” the trial court

opined, “that the staff and legislative leaders would find [this lack of public

participation by the consultants] extremely strange, that they might even ask why

not. But they didn’t.”

According to the trial court, however, the consultants had no need to

publicly participate in order to influence the Legislature’s redistricting plan.

Throughout the process, Reichelderfer was in direct contact with Speaker Cannon.

In one late November 2011 e-mail from Cannon to Reichelderfer, which copied

- 31 -

Pepper, Cannon commented that “we are in fine shape” as long as “the Senate

accommodates the concerns that you [Reichelderfer] and Rich [Heffley] identified

in the map that they put out tomorrow.”

Cannon testified at trial that these “concerns” he was referring to were that

the House and Senate “not roll out maps that were either completely inconsistent

with one another or designed to show some inadequacy in terms of either minority

representation or defect in [the House’s] maps,” so that reconciliation between the

two chambers would be difficult. The trial court found Cannon’s explanation to be

“a stretch given the language used.”

The evidence also revealed that Cannon asked Reichelderfer and Heffley,

who was described as being “close” to Senator Gaetz, to serve, as the trial court put

it, “as go betweens for leadership of the two chambers regarding the redistricting

process.” According to testimony relied on by the trial court, the asserted reason

for Reichelderfer’s and Heffley’s involvement was “purportedly because of a lack

of a good working relationship between the Speaker of the House and the President

of the Senate.”

The trial court was skeptical of that explanation, however, stating that “by

all accounts, the actual staff members of each chamber who were working on the

maps got along well with each other, as did the chairmen of the redistricting

committees.” The trial court actually found the staff members who testified at trial

- 32 -

to be “straightforward and credible” and “not a part of the conspiracy.” In any

event, the trial court specifically found that “in their insider roles, Heffley and

Reichelderfer did not have to speak directly to staff map drawers, or even

leadership, to infect and manipulate the map drawing and adoption process.”

At trial, Reichelderfer admitted to discussing “global” redistricting concerns

with Speaker Cannon, but denied talking to Cannon “specifically about individual

maps.” Reichelderfer lived near Cannon, their families spent time together,

Reichelderfer saw Cannon on the weekends, and Reichelderfer met with Cannon to

discuss issues he was dealing with as Speaker.

Reichelderfer also correctly informed other consultants about which of the

Legislature’s draft maps was most “relevant,” meaning which was most likely to

advance in the process. Among the seven congressional maps released to the

public by the House on December 6, 2011, the map identified by Reichelderfer as

the map most likely to advance was the map that was revised to become the

House’s final proposed congressional map. At trial, Reichelderfer could not

“recall specifically” how he knew that map to be the most likely to advance in the

process, simply stating that if he “had that information for sure,” he wouldn’t have

used the qualifier “I think” in his response. He testified that he “could have” just

thought it “was the easiest to pair up with the Senate version of the map.”

- 33 -

Communications among the consultants revealed particular emphasis on

certain areas of the map. For instance, in one e-mail referencing a configuration in

a draft map that kept District 14 contained entirely within Hillsborough County—a

configuration less favorable to Republicans than the configuration ultimately

enacted, which crossed Tampa Bay to pick up voters from Pinellas County in

District 14—Terraferma noted to Heffley that “Tampa is far from perfect.” The

enacted configuration of Districts 13 and 14—where District 14 includes a portion

of Pinellas County, rather than being strictly within Hillsborough—produced one

safe Democratic seat and one seat that either party could win, rather than two

naturally-occurring seats favorable to Democrats. This was the configuration

preferred by the consultants.

In another e-mail between Terraferma, Heffley, and Reichelderfer sent on

the same day the Senate released a public map that did not divide the City of

Homestead—a division considered by the consultants to be important to favor

Republicans—Terraferma noted that District 26 was “pretty weak.” Heffley

responded, “The [H]ouse needs to fix a few of these,” and Terraferma, copying

Reichelderfer, responded, “yes.” The enacted configuration did, indeed, split the

City of Homestead between Districts 26 and 27, which turned one Republican

district and one Democratic district into two Republican-leaning districts.

- 34 -

The decision to split Homestead was one of several key decisions made in a

non-public meeting between Senator Gaetz, Representative Weatherford, and the

two staff directors of the respective redistricting committees. While the meeting of

two legislators in private does not result in a violation of article III, section 4(e), of

the Florida Constitution—which requires all meetings between “more than two

members” of the Legislature to be open to the public—the lengths to which the

legislators went to avoid triggering the requirements for a public meeting in the

final stages of negotiating and making changes to the districts raises questions as to

the motivation of the Republican leadership. It also stands in stark contrast to

statements from that leadership proclaiming that the 2012 redistricting process

would be the most open and transparent in Florida’s history. And, it can be readily

distinguished from other legislative decisions where private negotiations are

undertaken, since redistricting involves “a constitutional restraint on the

Legislature’s actions.” Apportionment IV, 132 So. 3d at 147.

Indeed, many final revisions that affected numerous districts in some way—

such as the decision to push the Black Voting Age Population (BVAP) of District 5

over 50%, add an appendage to District 10, split Homestead, and increase the

Hispanic Voting Age Population (HVAP) of Districts 9 and 14—were made in this

non-public meeting that occurred after the House and Senate had each passed their

versions of the congressional map. The decisions regarding District 5 and District

- 35 -

10 specifically contributed to the trial court’s decision to invalidate those two

districts.

There was, in general, either conflicting or vague testimony as to why

certain decisions were made in this meeting, including that the decisions were

necessary to comply with the federal Voting Rights Act or some other policy

concern. Because the meeting was not public, however, there is no official record

of the reason for these decisions, which ultimately benefitted the Republican Party.

One example of a key decision made during this non-public meeting was the

decision to push the BVAP of District 5 over 50%. Although he could not recall

specifics, Representative Weatherford testified that making District 5 a majority-

minority district was “important to the Senate” and that the Senate made a

“compelling case” for raising the BVAP of the district over 50%. The highest

BVAP for District 5 in any of the House’s draft maps was slightly over 48%.

Senator Gaetz testified that the Senate believed it was important to increase the

BVAP to over 50% to protect against a federal Voting Rights Act challenge, and

that he also favored keeping the City of Sanford in the district, which the House’s

version of the map did not do.

Before Representative Weatherford met with Senator Gaetz, Speaker

Cannon met separately with Representative Weatherford and staff in another non-

public meeting. Speaker Cannon anticipated that the Senate would ask to make

- 36 -

District 5 a majority-minority district and apparently instructed the House during

this non-public meeting to agree to the Senate’s request. Ensuring that the BVAP

of District 5 ended up over 50% was of particular concern to Reichelderfer, the

consultant who was part of Speaker Cannon’s “inner circle.”

At trial, Reichelderfer testified, without specificity, that he believed pushing

the BVAP of District 5 over 50% was important “to comply with the Federal

Voters Rights Act,” based on a general recollection of discussions with lawyers

whose names he could not recall. He thought it would be “politically damaging” if

the map was invalidated because of a successful Voting Rights Act challenge, even

though the 2002 version of District 5 did not have a BVAP of over 50% and was

not invalidated during Voting Rights Act litigation. See Martinez, 234 F. Supp. 2d

at 1307 (noting that the BVAP of the 2002 version of District 5 was “only” 46.9%,

but that the district “will afford black voters a reasonable opportunity to elect

candidates of choice and probably will in fact perform for black candidates of

choice”). At the same time, increasing the BVAP of District 5—as occurred from

early versions of the Legislature’s draft maps to the enacted version—decreased

the Democratic performance of surrounding districts.

The trial court found the Legislature’s justification for making District 5 a

majority-minority district to be “not compelling” and invalidated the enacted

version of District 5. The Legislature’s decision—made in a non-public meeting,

- 37 -

after Cannon’s instruction in a separate non-public meeting, consistent with a

concern Reichelderfer had long expressed—is therefore circumstantial evidence of

collusion between the Legislature and the consultants, particularly where the trial

court found there to have been no showing that it was legally necessary to create a

majority-minority district.

There is no record from the time many of these key decisions were made to

explain the Legislature’s reasoning. This is, of course, partly because the final

decisions were made in a non-public meeting. But it is also because the

Legislature, as the trial court found, deleted almost all e-mails and documentation

related to redistricting.

Former Speaker Cannon testified that his e-mails were automatically deleted

after six months unless specifically saved as having “significant archival or legal

significance.” If that were the case, then exchanges between Speaker Cannon and

consultant Reichelderfer that occurred in late November 2011—discovered from

document production by Reichelderfer—would not have been deleted until May

2012 unless they were intentionally deleted before that time. But May 2012 was

several months after the lawsuit was filed in this case, naming Cannon as a party

and making a reality what the Legislature itself had previously acknowledged, as

far back as December 2012, to have been “a moral certainty” from “start to finish”

during the redistricting process—that records related to redistricting would be

- 38 -

sought by the challengers and relevant to adjudicating the constitutionality of the

Legislature’s redistricting plan.

Ultimately, based on the evidence the challengers uncovered and presented

at trial, the trial court found that there was “just too much circumstantial evidence”

and “too many coincidences” to reach any conclusion other than that the political

operatives had “infiltrate[d] and influence[d] the Legislature” in order to “obtain

the necessary cooperation and collaboration” to “taint the redistricting process and

the resulting map with improper partisan intent.” While it is sometimes said that it

is “hard to believe in coincidence,” the trial court determined in this case that, as

the saying goes, it was “even harder to believe in anything else.” After reviewing

all the evidence, both direct and circumstantial, the trial court thus concluded that

the plan was drawn with improper partisan intent.

C. STEPS AFTER FINDING UNCONSTITUTIONAL INTENT

Despite its finding of unconstitutional partisan intent, however, the trial

court invalidated only Districts 5 and 10, rejecting challenges to seven other

individual districts. The trial court determined that there was no “distinction”

between a challenge to the plan as a whole and a challenge to specific districts, and

therefore “focused on those portions of the map” that it found to be “in need of

corrective action in order to bring the entire plan into compliance with the

constitution.”

- 39 -

Its finding of unconstitutional intent notwithstanding, the trial court applied

a deferential standard of review in analyzing each challenged district, “deferring to

the Legislature’s decision to draw a district in a certain way, so long as that

decision does not violate the constitutional requirements.” Believing that the

“more reliable” indicators of whether the plan was drawn with the intent to favor a

political party or incumbent were the tier-two constitutional measures, the trial

court “first examine[d] the map for apparent failure to comply with tier-two

requirements of compactness and utilization of political and geographical

boundaries where feasible, then consider[ed] any additional evidence that supports

the inference that such districts are also in violation of tier-one requirements.”

Applying this analysis as to District 5, the trial court noted that the decision

to increase the BVAP of District 5 over 50% was made at a non-public meeting at

the end of the redistricting process and ultimately found that there was no showing

“that it was legally necessary to create a majority-minority district.” The trial court

therefore concluded that the challengers had proved “that District 5 unnecessarily

subjugates tier-two principles of compactness” and that “portions of District 5 were

drawn to benefit the Republican Party, in violation of tier-one.”

As to District 10, the trial court noted an “odd-shaped appendage” and found

that the challengers had “shown that the district could be drawn in a more compact

fashion, avoiding this appendage.” The trial court therefore concluded, based in

- 40 -

part on an inference it drew from the existence of the odd-shaped appendage that

had no legal justification, that District 10 was drawn to benefit the Republican

Party and the incumbent.

Accordingly, the trial court required Districts 5, 10, and “any other districts

affected thereby” to be redrawn. But the trial court rejected the challenges to

Districts 13, 14, 21, 22, 25, 26, and 27, concluding that the challengers had not met

their burden to demonstrate unconstitutionality and had not shown more than “de

minimis” tier-two violations.

As a remedy, the challengers urged the trial court to adopt one of their

remedial plans, draw its own remedial plan, or hire an independent expert to draw

a remedial plan. After a hearing, the trial court declined the challengers’

suggestions and determined that the Legislature should redraw the plan.

The Legislature held a special session in August 2014 to enact a remedial

redistricting plan. During this session, the chairs of the respective redistricting

committees again conducted non-public meetings with staff and counsel to

negotiate the features of the revised plan. The Legislature made modest changes to

correct the specific tier-two deficiencies identified in Districts 5 and 10,8 and, after

8. In redrawing Districts 5 and 10, the Legislature’s remedial redistricting

plan also slightly altered the boundaries of five other congressional districts—

Districts 6, 7, 9, 11, and 17. All of the remaining districts were unchanged from

the configuration enacted in the Legislature’s 2012 redistricting plan.

- 41 -

the plan was signed into law, the trial court held another hearing to consider the

validity of the revised plan and whether it could be implemented in time for the

2014 elections.

Concluding that the challengers’ objections to the validity of the remedial

plan were without merit, the trial court approved the Legislature’s remedial

redistricting plan and ordered the then-impending 2014 elections to proceed under

the unconstitutional 2012 plan due to time constraints, with the remedial plan to

take effect for the 2016 elections. The 2016 effective date for the remedial plan

has not been challenged.

The challengers appealed the trial court’s initial order containing its factual

findings and legal conclusions, as well as its subsequent order approving the

remedial redistricting plan, and the Legislature cross-appealed, attacking certain

aspects of the trial court’s judgment but ultimately seeking affirmance of the order

approving the remedial plan. The First District Court of Appeal then certified the

trial court’s judgment for direct review by this Court. See League of Women

Voters of Fla. v. Detzner, No. 1D14-3953, 2014 WL 4851707, at *2 (Fla. 1st DCA

Oct. 1, 2014). We accepted jurisdiction under article V, section 3(b)(5), of the

Florida Constitution, and heard oral argument. See League of Women Voters of

Fla. v. Detzner, No. SC14-1905, 2014 WL 5502409, at *1 (Fla. Sup. Ct. order filed

Oct. 23, 2014).

- 42 -

III. ISSUES OF “INTENT”

Having set forth this comprehensive background, we now turn to the legal

issues pertaining to the trial court’s finding of unconstitutional intent. First, we

consider the “intent” standard itself and whether the trial court correctly applied

the standard in this case. Then, we review the legal sufficiency of the trial court’s

finding.

A. THE “INTENT” STANDARD

Article III, section 20, of the Florida Constitution, prohibits an

apportionment plan or individual district from being “drawn” with the “intent to

favor or disfavor a political party or an incumbent.” Art. III, § 20(a), Fla. Const.

All parties in the litigation, the trial court stated, “agreed that it is the Legislature’s

intent”—not the intent of, for instance, one rogue “staff member charged with

actually drawing the map,” or of political consultants with no influence on the

Legislature—“that is at issue.” But how to determine the Legislature’s intent in

this unique context, where the Florida Constitution contains an explicit prohibition

on certain improper legislative intent in “draw[ing]” the redistricting plan, is a

much more difficult proposition.

In Apportionment I, 83 So. 3d at 617, this Court explained that “the Florida

Constitution prohibits drawing a plan or district with the intent to favor or disfavor

a political party or incumbent.” (Emphasis supplied.) There is, this Court held,

- 43 -

“no acceptable level of improper intent.” Id. The “intent” standard “applies to

both the apportionment plan as a whole and to each district individually.” Id.

(emphasis supplied). This Court’s precedent discussing the “intent” standard in the

course of prior cases during this litigation has demonstrated this principle—that

improper intent, particularly if “part of a broader process to develop portions of the

map,” may “directly relate to whether the plan as a whole or any specific districts

were drawn with unconstitutional intent.” Apportionment IV, 132 So. 3d at 150

(emphasis supplied).

In a traditional lawsuit involving a challenge to a statutory enactment, courts

determine legislative intent through statutory construction, looking to the actual

language used and any other tools—such as the history of legislative changes and

any appropriate interpretive canons—to assist in discerning the Legislature’s intent

in enacting the law. See, e.g., Heart of Adoptions, Inc. v. J.A., 963 So. 2d 189,

198-99 (Fla. 2007) (setting forth the general principle of statutory interpretation

that “legislative intent is determined primarily from the statute’s text” and applying

rules of statutory construction “to determine the legislative intent behind the

provision,” including reading related statutory provisions together to achieve a

consistent whole and avoiding readings that would render part of a statute

meaningless). As this Court has previously explained, however, determining

whether the Legislature acted with the type of improper “intent” that is prohibited

- 44 -

by the “specific constitutional mandate of article III, section 20(a), is entirely

different than a traditional lawsuit that seeks to determine legislative intent through

statutory construction.” Apportionment IV, 132 So. 3d at 150.

Specifically, this Court held in largely rejecting claims of legislative

privilege in Apportionment IV that, because the decision-making process itself is

the case, “the communications of individual legislators or legislative staff

members, if part of a broader process to develop portions of the map, could

directly relate to whether the plan as a whole or any specific districts were drawn

with unconstitutional intent.” Id. This Court further stated that the “existence of a

separate process to draw the maps with the intent to favor or disfavor a political

party or an incumbent is precisely what the Florida Constitution now prohibits,”

and that evidence of this separate process would “clearly” be “important” to help

support a “claim that the Legislature thwarted the constitutional mandate.” Id. at

149.

Following this Court’s precedent, which “emphasize[s] that this case is

wholly unlike the traditional lawsuit challenging a statutory enactment,” id. at 151,

the trial court framed the “intent” inquiry as determining “the motive in drawing”

the districts. We agree that this was the correct approach. Under this framework,

the trial court appropriately concluded that “the actions and statements of

- 45 -

legislators and staff, especially those directly involved in the map drawing

process[,] would be relevant on the issue of intent.”

Case law supports the trial court’s conclusion, which is consistent with our

decision in Apportionment IV, that the intent of individual legislators and

legislative staff members involved in the drawing of the redistricting plan is

relevant in evaluating legislative intent. The United States Supreme Court, for

example, has recognized that the actions of individual legislators and staff

members may be relevant in discerning legislative intent in the context of

redistricting.

In Easley v. Cromartie, 532 U.S. 234, 254 (2001), the Supreme Court

reviewed “direct” evidence, relied on by a federal district court evaluating a claim

of racial predominance in North Carolina’s congressional redistricting plan,

involving an e-mail sent from “a legislative staff member responsible for drafting

districting plans” to two state senators. The Supreme Court noted that the e-mail’s

“reference to race” offered “some support” for the district court’s conclusion that

the North Carolina Legislature used race as the “predominant factor” in drawing

the boundaries of a particular district. Id.; see also Vill. of Arlington Heights v.

Metro. Hous. Dev. Corp., 429 U.S. 252, 267 (1977) (stating that the “specific

sequence of events leading up to the challenged decision also may shed some light

on the decisionmaker’s purposes” and that “[d]epartures from the normal

- 46 -

procedural sequence also might afford evidence that improper purposes are playing

a role”).

Other redistricting cases have confirmed this principle. In Texas v. United

States, 887 F. Supp. 2d 133, 165 (D.D.C. 2012),9 cited by the trial court, a three-

judge federal district court panel stated that its “skepticism about the legislative

process that created [a challenged district] [wa]s further fueled by an email sent

between staff members on the eve of the Senate Redistricting Committee’s markup

of the proposed map.” See also Smith v. Beasley, 946 F. Supp. 1174, 1210 (D.S.C.

1996) (stating that “the evidence [wa]s clear that the Reapportionment

Subcommittee delegated to its staff . . . the responsibility of drawing the district

lines,” and subsequently evaluating the actions of those staff members). In other

words, the federal district court looked to the actions of legislative staff members

directly involved in the redistricting process to assist in evaluating whether the

Legislature was acting with improper intent. Whether the actions of individual

legislators or staffers ultimately signify constitutionally improper intent—as the

trial court concluded in this case, despite finding the professional staff to be

9. The federal district court’s opinion in Texas was subsequently vacated on

other grounds by the United States Supreme Court after that Court issued its recent

decision in Shelby County, Alabama v. Holder, 133 S. Ct. 2612 (2013), holding a

portion of the Voting Rights Act unconstitutional. See Texas v. United States, 133

S. Ct. 2885 (2013).

- 47 -

credible—is a separate question from whether their intent is relevant, in the first

place, to evaluating the intent of the Legislature in drawing the redistricting plan.

In support of its contrary argument that “[c]ourts across the country . . .

refuse to impute the personal motivations of individual legislators to the legislative

body as a collective whole,” the Legislature offers a catalogue of citations to cases

from other jurisdictions. But, as the challengers have pointed out, these cases and

the arguments made by the Legislature in this case closely mirror the exact cases

and arguments this Court distinguished and rejected for the same basic principle in

Apportionment IV.

In that case, this Court specifically stated that “this case is completely

distinguishable from the various circuit court orders and cases outside the

reapportionment context from other jurisdictions cited by the Legislature that have

quashed subpoenas of legislators or legislative staff members where the testimony

of an individual member of the Legislature was not directly relevant to any issue in

the case.” Apportionment IV, 132 So. 3d at 150. Indeed, in Apportionment IV, we

determined that the actions of the individual legislators and legislative staff

members involved in the drawing of the redistricting plan were directly relevant to

assessing whether the plan itself was drawn with improper intent. See id. at 137

(“[T]he issue presented to the Court is whether Florida state legislators and

legislative staff members have an absolute privilege against testifying as to issues

- 48 -

directly relevant to whether the Legislature drew the 2012 congressional

apportionment plan with unconstitutional partisan or discriminatory ‘intent.’ ”).

Accordingly, we hold that the trial court correctly framed the “intent”

inquiry and reject the Legislature’s assertion that the finding of unconstitutional

intent could not be ascribed to the Legislature as a whole. Having reached the

conclusion that the trial court did not err in evaluating the actions of legislators and

legislative staff members in finding unconstitutional “intent,” as prohibited by

article III, section 20, we turn next to the legal sufficiency of the trial court’s

finding.

B. LEGAL SUFFICIENCY OF UNCONSTITUTIONAL INTENT

Our review of the trial court’s finding of unconstitutional intent in the

congressional redistricting plan takes place against the backdrop of the trial court’s

specific finding that the Legislature “systematically deleted almost all of their e-

mails and other documentation relating to redistricting.” The Legislature did so

despite knowledge that litigation over the constitutionality of its redistricting plan

was inevitable.

In fact, as far back as 2008, the Legislature argued to this Court that

“litigation challenging reapportionment under the new standards” would increase

as a result of the Fair Districts Amendment. See Advisory Op. to Att’y Gen. re

Standards for Establishing Legislative Dist. Boundaries, 2 So. 3d 161, 165 (Fla.

- 49 -

2009). And, the Legislature informed the trial court in this case that litigation “was

‘imminent’ long before the days preceding the filing of” the challengers’ lawsuit.

From “start to finish,” the Legislature asserted, the 2012 redistricting process,

“more than any other, was conducted in an atmosphere charged with litigation.”

To be sure, the Legislature did preserve some records related to

redistricting—documents showing, for instance, the time and location of public

meetings or other generally benign details of the process. But the Legislature

saved virtually no communications among legislators and staff and none of the

communications—which, as a result of this case, we now know to have occurred—

involving the outside political consultants.

The Legislature had no specific policy requiring it to preserve

communications regarding redistricting, even though it knew litigation was certain

to occur, and admits that its record-retention policies applied in the same manner to

redistricting as they applied to all types of legislative business. The House’s

policy, for example, specified that “records that are no longer needed for any

purpose and that do not have sufficient administrative, legal, or fiscal significance

to warrant their retention shall be disposed of systematically.” Fla. H.R. Rule

14.2(b) (2010-2012).

To the extent the Legislature argues that it had no reason to know it needed

to preserve these records because it could not have anticipated this Court’s

- 50 -

decision in Apportionment IV rejecting its broad claim of legislative privilege over

communications related to redistricting, the Legislature had, according to

testimony at trial, determined as early as January 2011 that no privilege would

apply to any of its communications with outside political consultants. In other

words, the Legislature clearly knew that communications between, for instance,

Speaker Cannon and consultant Reichelderfer would not be privileged, that they

would be sought in litigation, and that litigation was certain to occur. Yet, Speaker

Cannon did not preserve these records—and the only reason we now know these

communications occurred is because records were produced during the litigation

by Reichelderfer. The same is true of non-public draft redistricting maps sent to

Reichelderfer by legislative staffer Kirk Pepper, using a personal e-mail account

and a since-deleted “Dropbox” account.

The trial court stated that there was “no legal duty on the part of the

Legislature to preserve these records, but you have to wonder why they didn’t,”

given that litigation was certain to occur. Although the Legislature’s failure to

preserve records apparently did not violate a specific rule of legislative procedure

regarding records retention—even though at least some of these records likely did

have sufficient legal significance to have warranted their retention—Florida courts

have, in any event, found a duty to preserve evidence in other circumstances when

a party should reasonably foresee litigation. See Am. Hospitality Mgmt. Co. of

- 51 -

Minn. v. Hettiger, 904 So. 2d 547, 549 (Fla. 4th DCA 2005) (noting holdings that

“a defendant could be charged with a duty to preserve evidence where it could

reasonably have foreseen the claim”). And this Court, in rejecting the

Legislature’s broad claim of legislative privilege in Apportionment IV, clearly held

that the “purpose behind the voters’ enactment of the article III, section 20(a),

standards will be undermined” if “the Legislature alone is responsible for

determining what aspects of the reapportionment process are shielded from

discovery.” 132 So. 3d at 149.

Even in the absence of a legal duty, though, the spoliation of evidence

results in an adverse inference against the party that discarded or destroyed the

evidence. As this Court explained in Martino v. Wal-Mart Stores, Inc., 908 So. 2d

342, 346 (Fla. 2005), Florida courts may impose sanctions, including striking

pleadings, against a party that intentionally lost, misplaced, or destroyed evidence,

and a jury could infer under such circumstances that the evidence would have

contained indications of liability. If the evidence was negligently destroyed, a

rebuttable presumption of liability may arise. Id. at 347. In other words, as

recognized by the Fourth District Court of Appeal, “an adverse inference may arise

in any situation where potentially self-damaging evidence is in the possession of a

party and that party either loses or destroys the evidence.” Golden Yachts, Inc. v.

Hall, 920 So. 2d 777, 781 (Fla. 4th DCA 2006) (quoting Martino v. Wal-Mart

- 52 -

Stores, Inc., 835 So. 2d 1251, 1257 (Fla. 4th DCA 2003), approved, 908 So. 2d

342); see also Nationwide Lift Trucks, Inc. v. Smith, 832 So. 2d 824, 826 (Fla. 4th

DCA 2002) (stating that “[c]ases in which evidence has been destroyed, either

inadvertently or intentionally, are discovery violations” that may be subject to

sanctions).

The trial court was, therefore, justified in drawing an adverse inference

against the Legislature in adjudicating the challengers’ claim of unconstitutional

partisan intent. And we too must consider the Legislature’s “systematic[]

delet[ion]” of redistricting records in evaluating whether the trial court’s finding is

supported by competent, substantial evidence.

Turning to the merits of the trial court’s finding, we have little trouble

concluding that competent, substantial evidence of unconstitutional intent exists in

the record. The Legislature asserts that the trial court did not find improper intent

in the plan as a whole and, in particular, contends that there was no collaboration

between partisan operatives and the Legislature in drawing the congressional

redistricting plan. While acknowledging that partisan operatives “sought to

influence the redistricting process,” the Legislature states that “at no time did the

Legislature participate in their efforts.” If features from the operative-created maps

made it into the enacted map, the Legislature says, it is simply because those

features were obvious or the similarities “superficial,” and not because the

- 53 -

operatives’ “frenetic efforts to make themselves relevant” were successful. In

other words, the Legislature argues that it “did not conspire with the operatives,

despite the operatives’ efforts.”

We reject the Legislature’s attempt to water down the trial court’s findings

and the inferences the trial court drew from the circumstantial evidence presented

by pointing to an alleged lack of connection between the “parallel” process and the

Legislature. The trial court found that it was “convince[d]” by the “circumstantial

evidence introduced at trial” that the political operatives “obtain[ed] the necessary

cooperation and collaboration” from the Legislature to ensure that the

“redistricting process and the resulting map” were “taint[ed]” with “improper

partisan intent.” Indeed, the trial court specifically found that the operatives “were

successful in their efforts to influence the redistricting process and the

congressional plan under review.”

Nevertheless, the Legislature asserts that any conclusion that the whole plan

was motivated by partisan intent “assumes the complicity of professional staff,”

which is an “assumption” it claims the trial court rejected. While the trial court did

find the professional staff to be “credible” and not to have been “part of the

conspiracy,” the trial court immediately dismissed the Legislature’s argument

about the effect of the staff having been insulated from the improper intent—which

it called the “most compelling evidence in support” of the Legislature’s defense—

- 54 -

by stating that the “political operatives managed to find other avenues, other ways

to infiltrate and influence the Legislature, to obtain the necessary cooperation and

collaboration” to “taint the redistricting process and the resulting map with

improper partisan intent.” And while the trial court made no explicit credibility

determinations regarding any of the legislators who testified, the trial court did

specifically reject the innocuous explanations provided by former Speaker Cannon

and his staffer, Pepper, for their communications with the political consultants.

There is also no doubt that the trial court’s finding of unconstitutional intent

pertained to the “process” of redistricting and the “enacted map” as a whole—to

use the trial court’s own words—rather than solely to the two specifically

invalidated districts as the Legislature contends. In finding “too much

circumstantial evidence” to reach any conclusion other than that the “redistricting

process” and the “resulting map” were “taint[ed]” by “improper partisan intent,”

the trial court pointed specifically to the following evidence: the Legislature’s

destruction of “almost all” e-mails and “other documentation relating to

redistricting”; early meetings between legislative leaders and staff with political

consultants regarding the “redistricting process”; and the “continued involvement”

of political consultants in the “redistricting process.” None of this evidence relied

on by the trial court was district-specific. The dissent’s contrary interpretation of

- 55 -

the trial court’s finding of unconstitutional intent renders meaningless the trial

court’s extensive discussion of—and critical findings related to—this evidence.

We also reject the Legislature’s suggestion that the trial court’s

determination, in its order approving the remedial redistricting plan, that the

Legislature had corrected the identified deficiencies in the map is dispositive in

evaluating the scope of its finding of unconstitutional intent. Instead, as detailed in

the next sections, the trial court’s decision to approve the Legislature’s remedial

redistricting plan flowed from the legal errors made in its original judgment.

Accordingly, for all these reasons, we affirm the trial court’s finding of

unconstitutional intent. We turn next to the trial court’s two legal errors, which

significantly affected its determination of the proper effect of its finding that the

Legislature violated the Florida Constitution.

IV. TRIAL COURT’S FIRST LEGAL ERROR: FAILING TO PROPERLY

ANALYZE THE CHALLENGE TO THE PLAN “AS A WHOLE”

The first legal error committed by the trial court was its determination that

there was no distinction between a challenge to the redistricting plan “as a whole”

and a challenge to individual districts. This error led to the trial court’s failure to

give any independent legal significance to its finding of unconstitutional intent

when examining the challenges to individual districts.

Specifically, the evidence presented and considered by the trial court—

evidence that actually led the trial court to find the existence of constitutionally

- 56 -

improper partisan intent—included evidence pertaining both to the plan “as a

whole” and to “specific districts.” Indeed, the trial court explicitly noted this,

stating that “[o]ne of [the challengers’] claims is that the entire redistricting

process was infected by improper intent.” (Emphasis supplied.)

Yet, despite its findings that partisan political consultants had “made a

mockery” of the process and “managed to taint the redistricting process and the

resulting map with improper partisan intent,” the trial court rejected the

challengers’ distinction between their challenge to improper intent in the

redistricting plan “as a whole”—a challenge, in effect, to the map that was

produced from the process—and their challenge to “individual districts,” stating as

follows:

[The challengers] distinguish between their challenge to the

redistricting plan as a whole, as being drawn with the intent generally

to favor the Republican Party, and their challenge to several

individual districts, as being specifically drawn with such intent. I

find this to be a false dichotomy, a distinction without difference. The

redistricting plan is the result of a single act of legislation. If one or

more districts do not meet constitutional muster, then the entire act is

unconstitutional. The districts are part of an integrated indivisible

whole. So in that sense, if there is a problem with a part of the map,

there is a problem with the entire plan. [FN 5]

[FN 5] This is consistent with the approach taken by

[this] Court in Apportionment I. The Court invalidated

the entire Senate plan but gave specific instructions as to

which districts required corrective action. Id. at 684-686.

That does not mean, however, that portions of the map not

affected by those individual districts found to be improperly drawn

would need to be changed in a redrawn map, even if a general intent

to favor or disfavor a political party or incumbents was proven. What

- 57 -

would be the point if the other districts are otherwise in compliance?

Such a remedy would go far beyond correcting the effect of such

noncompliance, but rather would require a useless act that would

encourage continued litigation. Therefore, I have focused on those

portions of the map that I find are in need of corrective action in order

to bring the entire plan into compliance with the constitution.

(Emphasis supplied.)

The dissent asserts that “[a]t no point does the trial court indicate that it

would permit some level of unconstitutional intent in the drawing of any district.”

Dissenting op. at 114. But the trial court specifically concluded that districts could

be “in compliance” with the constitutional standards “even if a general intent to

favor or disfavor a political party or incumbents was proven.” This statement

clearly indicates that the trial court considered a general improper intent to lack

any independent legal significance unless it was accompanied by another

constitutional violation, which is an interpretation that simply does not square with

the Florida Constitution or this Court’s precedent.

This Court has held that “the Florida Constitution prohibits drawing a plan”

with improper intent. Apportionment I, 83 So. 3d at 617. This Court has also held

that “there is no acceptable level of improper intent.” Id. And, this Court has held

that the “intent” standard “applies to . . . the apportionment plan as a whole.” Id.

Accordingly, under these holdings, the trial court’s “general” finding of improper

intent in the “process” must have some independent legal significance.

- 58 -

The trial court, however, failed to give effect to that finding of improper

intent, in part because it never separately considered the challenge to the plan as a

whole and, critically, never gave any weight to the general improper intent in

analyzing the individual district challenges. The challengers correctly note that the

trial court’s finding of improper intent was based extensively on the existence of a

“different, separate process that was undertaken contrary to the [Legislature’s

public] transparent [redistricting] effort in an attempt to favor a political party or an

incumbent.” Apportionment IV, 132 So. 3d at 149. And, as this Court stated in

Apportionment IV, the existence of such a “parallel” process is “important

evidence in support of the claim that the Legislature thwarted the constitutional

mandate.” Id.

In error, the trial court gave no legal weight to the existence of this separate

process. The trial court’s decision to invalidate District 5 was supported by

numerous factors distinct from the “parallel” process, including that the district as

enacted was “not compact,” was “bizarrely shaped,” and did not “follow traditional

political boundaries as it winds from Jacksonville to Orlando,” narrowing at one

point to the width of a highway. The trial court found improper intent to benefit

the Republican Party as to District 5 based on “the decision to increase the district

to majority BVAP, which was accomplished in large part by creating [a] finger-

like appendage jutting into District 7.” Then, the trial court simply “buttressed”

- 59 -

this “inference” of improper intent, based on the existence of the “oddly shaped

appendage[],” through “the evidence of improper intent in the redistricting process

generally, and as specifically related to the drawing of District 5,” but did not

independently rely on the “general” improper intent in any legally significant way.

In other words, aside from referencing the increase in the BVAP of District

5 over 50% during a non-public meeting at the end of the redistricting process, the

trial court’s decision to invalidate District 5 was based solely on blatant tier-two

violations. While this Court had to resort to evaluating tier-two violations as a

means to infer improper intent when considering the challenges to the Senate and

House maps in Apportionment I, as we emphasized at that time, we were

constrained because we had no factual record and no direct evidence of improper

intent. Exactly the opposite was true in this case.

The trial court’s decision to invalidate District 10 is analogous. Noting an

“odd-shaped appendage which wraps under and around District 5, running between

District 5 and 9,” the trial court stated that the challengers had “shown that the

district could be drawn in a more compact fashion, avoiding this appendage.” The

trial court’s conclusion that District 10 “was drawn to benefit the Republican Party

and the incumbent” was “based in part on the inference that the Florida Supreme

Court suggested [in Apportionment I] could be drawn from oddly shaped

appendages that had no legal justification”—an “inference” that, as with District 5,

- 60 -

was simply “buttressed by the general evidence of improper intent” in the process

and by objective indicators relied on by this Court in Apportionment I.

In rejecting challenges to seven other individual districts, the trial court

never referred to the “general evidence of improper intent” that it found to exist in

the “process.” Rejecting the challenge to Districts 13 and 14, in the Tampa Bay

area, the trial court stated that, “[u]nlike Districts 5 and 10, there are no flagrant

tier-two deviations” from which the trial court could “infer” improper intent.

“[U]nlike changes made to District 5 by the [legislative] leaders during conference

committee”—the “evidence of partisan intent specifically related to District 5,”

where the House agreed with the Senate’s request to push the BVAP over 50%—

the trial court determined that it could not conclude, “on partisan effect alone,” that

certain decisions were made in drawing Districts 13 and 14 “with the intent to

benefit the Republican Party or the incumbent member of Congress.”

Likewise rejecting the challenge to Districts 21 and 22, the trial court

concluded that the challengers had “not met their burden of showing unnecessary

deviation from tier-two requirements,” nor had they “shown that improper intent

led to the adoption of Districts 21 and 22.” Similarly, with respect to Districts 25,

26, and 27, the trial court determined that the challengers had “not proved

invalidity” because the “totality of the evidence” did not establish that the

“configuration” of these districts “was based on unlawful partisan intent.” At no

- 61 -

point in addressing the validity of any of these districts—in which the trial court

rejected the challengers’ contention that the districts were drawn with improper

partisan intent—did the trial court address any effect of its findings regarding how

the “process” had been “taint[ed]” with “improper partisan intent.”

In determining that there was no distinction between a challenge to the

“whole map” and a challenge to individual districts, the trial court relied on this

Court’s prior decision to invalidate the entire state Senate plan in Apportionment I.

Citing this Court’s decision as support, the trial court stated that this Court

“invalidated the entire Senate plan but gave specific instructions as to which

districts required corrective action.”

The trial court was correct that this Court invalidated the whole Senate plan,

to the extent that it determined the plan did “not pass constitutional muster” for the

purposes of this Court’s article III, section 16, declaratory judgment review.

Apportionment I, 83 So. 3d at 683. But, unlike here, this Court in Apportionment I

did not find a general improper intent in the state Senate plan, aside from the

district numbering system that was manipulated to favor incumbents. Nor could

we have, based on the nature of the limited record before us.

In Apportionment I, we expressed our conclusion regarding the Senate plan

as follows:

We have held that Senate Districts 1, 3, 6, 9, 10, 29, 30, and 34

are constitutionally invalid. The Legislature should remedy the

- 62 -

constitutional problems with respect to these districts, redrawing these

districts and any affected districts in accordance with the standards as

defined by this Court, and should conduct the appropriate functional

analysis to ensure compliance with the Florida minority voting

protection provision as well as the tier-two standards of equal

population, compactness, and utilization of existing political and

geographical boundaries. As to the City of Lakeland, the Legislature

should determine whether it is feasible to utilize the municipal

boundaries of Lakeland after applying the standards as defined by this

Court. In redrawing the apportionment plan, the Legislature is by no

means required to adopt the Coalition’s alternative Senate plan.

Finally, we have held that the numbering scheme of the Senate plan is

invalid. Accordingly, the Legislature should renumber the districts in

an incumbent-neutral manner.

Id. at 686.

In other words, this Court identified very specific deficiencies in the Senate

plan—eight individual districts that were invalid, the failure to conduct a

functional analysis, and the district numbering scheme. This Court did not

conclude that the whole plan was unconstitutional because of improper intent in

the whole plan, and this Court did not analyze—and could not have analyzed—the

plan in that manner. Therefore, in relying on Apportionment I in this way, the trial

court failed to give any actual effect to its finding in this case that the “whole plan”

challenge had been proven through the direct and circumstantial evidence of

improper partisan intent presented at trial.

Accordingly, for all these reasons, we conclude that the trial court erred in

failing to recognize any distinction between a challenge to the redistricting plan “as

a whole” and a challenge to individual districts. This error significantly affected

- 63 -

the trial court’s determination of the proper scope and legal effect of its finding of

unconstitutional intent, particularly with regard to its analysis of the challenges to

individual districts, and ultimately contributed to its decision to approve a remedy

that was effectively no different than the remedy if there had been no finding of

unconstitutional intent.

V. TRIAL COURT’S SECOND LEGAL ERROR: APPLYING A

DEFERENTIAL STANDARD OF REVIEW

The trial court’s error in failing to properly analyze the challenge to the plan

“as a whole” was compounded by its error in the deferential standard of review it

applied after finding the existence of unconstitutional intent. Certainly, we

recognize the difficult task the trial court faced, considering numerous issues of

first impression and attempting to be faithful to this Court’s redistricting decisions.

And we commend the trial court for the superb and professional manner in which it

handled this difficult litigation.

But, we conclude nevertheless that the trial court failed to recognize the

critical differences between this Court’s “facial” review of the state legislative

redistricting plans in Apportionment I and the nature of the fact-based claims

presented in this case. This legal error in the standard of review, as with the legal

error in not recognizing the independent significance of the challenge to the plan

“as a whole,” led to the trial court’s failure to give any independent legal

significance to its finding of unconstitutional intent when examining the challenges

- 64 -

to individual districts. Once the trial court found unconstitutional intent, there was

no longer any basis to apply a deferential standard of review; instead, the trial court

should have shifted the burden to the Legislature to justify its decisions in drawing

the congressional district lines.

The trial court’s error as to the standard of review can be traced to its

analysis in evaluating the challengers’ claims, which it set forth as follows:

It seems that the more reliable focus in such an inquiry would be on

what was actually produced by the Legislature, the enacted map.

Specifically, an analysis of the extent to which the plan does or does

not comply with tier two requirements is a good place to start. Can

one draw a map that meets tier-two requirements but nonetheless

favors a political party or an incumbent? Sure, but it is more difficult.

Furthermore, a failure to comply with tier-two requirements not

only supports an inference of improper intent, it is an independent

ground for finding a map unconstitutional. See Apportionment I, 83

So. 3d [at] 640-641. Additional direct and circumstantial evidence of

intent may serve to strengthen or weaken this inference of improper

intent. Therefore, I first examine the map for apparent failure to

comply with tier-two requirements of compactness and utilization of

political and geographical boundaries where feasible, then consider

any additional evidence that supports the inference that such districts

are also in violation of tier-one requirements.

(Emphasis supplied.) In other words, the trial court began by asking whether there

was any tier-two violation—whether the district was compact, and whether it

followed existing political and geographical boundaries where feasible. Then, the

trial court considered the direct and circumstantial evidence of tier-one improper

intent only as “additional evidence” to “strengthen or weaken” an “inference of

improper intent” that was identifiable from tier-two deficiencies. The trial court

- 65 -

did so despite finding that the direct and circumstantial evidence itself had

established a violation of the tier-one constitutional standards.

Although the trial court relied on Apportionment I as support for the

standard of review it applied, the standard from that case—a facial review based on

purely objective, undisputed evidence in the limited record before the Court—does

not directly translate to this one—a fact-intensive challenge based on direct and

circumstantial evidence developed during an adversarial trial. Discerning which

aspects of the standard set forth in Apportionment I apply and which do not is thus

of critical importance.

In Apportionment I, this Court rejected the arguments of the Attorney

General and the House of Representatives “that a challenger must prove facial

invalidity beyond a reasonable doubt,” as is generally considered to be the standard

applied to a typical lawsuit challenging the constitutionality of a legislative

enactment outside the context of redistricting. 83 So. 3d at 607. This Court

considered the “beyond a reasonable doubt” standard to be both “a departure from

[its] precedent in legislative apportionment jurisprudence” and “ill-suited” to the

nature of its review. Id. “Unlike a legislative act promulgated separate and apart

from an express constitutional mandate,” this Court stated, “the Legislature adopts

a joint resolution of legislative apportionment solely pursuant to the ‘instructions’

- 66 -

of the citizens as expressed in specific requirements of the Florida Constitution

governing this process.” Id. at 607-08.

Although the legislative redistricting plan comes before this Court “with an

initial presumption of validity,” this Court explained that “the operation of this

Court’s process in apportionment cases is far different than the Court’s review of

ordinary legislative acts,” including “a commensurate difference in [its]

obligations.” Id. at 606. Noting that the “new requirements” of the Fair Districts

Amendment “dramatically alter[ed] the landscape with respect to redistricting,”

this Court held that its scope of review had “plainly increased, requiring a

commensurately more expanded judicial analysis of legislative compliance.” Id. at

607. As this Court would later reason, “the framers and voters” of the Fair

Districts Amendment “clearly desired more judicial scrutiny” of the Legislature’s

decisions in redistricting. Fla. House of Representatives v. League of Women

Voters of Fla. (Apportionment III), 118 So. 3d 198, 205 (Fla. 2013).

“It is this Court’s duty, given to it by the citizens of Florida, to enforce

adherence to the constitutional requirements and to declare a redistricting plan that

does not comply with those standards constitutionally invalid.” Apportionment I,

83 So. 3d at 607. However, this Court acknowledged in the context of its review in

Apportionment I that it would “defer to the Legislature’s decision to draw a district

in a certain way, so long as that decision does not violate the constitutional

- 67 -

requirements.” Id. at 608. This Court emphasized that its “responsibility [wa]s

limited to ensuring compliance with constitutional requirements.” Id.

“[E]ndeavoring to be respectful to the critically important role of the Legislature,”

this Court stated that its duty was “not to select the best plan, but rather to decide

whether the one adopted by the legislature is valid.” Id. (quoting In re

Apportionment Law—1992, 597 So. 2d at 285).

Echoing this Court’s language in Apportionment I, the trial court

determined—based on “the nature of the legislation and the nature of what is

reviewed”—that it should apply the same standard to the challenge presented in

this case. Therefore, reciting the principles from Apportionment I, the trial court

set forth the standard for its review as follows:

I will therefore, in this case, apply the standard of review

articulated in Apportionment I, deferring to the Legislature’s decision

to draw a district in a certain way, so long as that decision does not

violate the constitutional requirements, with an understanding of my

limited role in this process and the important role of the Legislature.

My duty “is not to select the best plan” but to determine whether [the

challengers] have proved the plan invalid. Apportionment I, 83 So. 3d

597 at 608.

The trial court then cited this standard, and its deferential review, in rejecting

challenges to certain individual districts.

We conclude that the trial court was correct, initially, in rejecting the

“beyond a reasonable doubt” standard, as this Court did in Apportionment I. As

this Court stated, “[u]nlike a legislative act promulgated separate and apart from an

- 68 -

express constitutional mandate, the Legislature adopts a joint resolution of

legislative apportionment solely pursuant to the ‘instructions’ of the citizens as

expressed in specific requirements of the Florida Constitution governing this

process.” Apportionment I, 83 So. 3d at 607-08. Just as there is a difference in

evaluating legislative intent with respect to the specific constitutional mandate

outlawing improper partisan intent in redistricting, so too is there “a difference

between the Court’s role in reviewing a legislative apportionment plan to

determine compliance with constitutionally mandated criteria and the Court’s role

in interpreting statutes.” Id. at 607 n.5. The “reason for the different standard,”

the trial court correctly noted, is that “the inquiry is into the process, the end result,

and the motive behind the legislation”—not “a question of searching for a

reasonable interpretation of a statute which would make it constitutional.”

In this respect, the trial court was right to rely on Apportionment I in

concluding that the nature of the legislation and the specific constitutional mandate

outlawing partisan political gerrymandering require a different standard of review

than applied in traditional cases challenging legislative enactments. Where the

trial court erred, however, was in discounting the differences between

Apportionment I and this case to conclude that the same standard must apply, even

though this case involved direct and circumstantial evidence of tier-one

- 69 -

constitutional violations that this Court had no ability to review in Apportionment

I.

As this Court has explained, its review in Apportionment I was quite

different than the challenge presented in this case. Unlike the fact-intensive

challenge here, in which the parties had an opportunity to present extensive

evidence during an adversarial trial pertaining to whether the plan and individual

districts were drawn with improper intent, this Court’s review in Apportionment I

was “a facial review based on objective, undisputed evidence in the limited record

before the Court.” Apportionment III, 118 So. 3d at 200.

In Apportionment I, 83 So. 3d at 634, this Court looked to objective

measures and tier-two requirements—such as the existence of “bizarre shape[s]”

and “appendages”—in an effort to discern whether the map was drawn with

improper intent. As this Court stated, “in the context of Florida’s constitutional

provision, a disregard for the constitutional requirements set forth in tier two is

indicative of improper intent, which Florida prohibits by absolute terms.” Id. at

640.

The evidence of improper intent in this case, to the contrary, involved direct

and circumstantial evidence of tier-one violations of the constitutional intent

standard. Yet, despite the existence of testimony and fact-based claims regarding

improper intent from a voluminous record that extended far beyond the legislative

- 70 -

record to which this Court was constrained in Apportionment I, the trial court still

determined that tier-two requirements—compactness and the use of political and

geographical boundaries where feasible—were the “more reliable” indicators of

improper intent, explaining that “a failure to comply with tier-two requirements”

would “support[] an inference of improper intent,” and that “[a]dditional direct and

circumstantial evidence of intent may serve to strengthen or weaken this inference

of improper intent.” Based on this assumption, the trial court proceeded to “first

examine the map for apparent failure to comply with tier-two requirements of

compactness and utilization of political and geographical boundaries where

feasible, [and] then consider[ed] any additional evidence that supports the

inference that such districts are also in violation of tier-one requirements.”

Not surprisingly under this framework of analysis, only where the trial court

found a tier-two violation—the appendages in Districts 5 and 10—did the trial

court conclude that a district had been drawn with improper intent to favor a

political party or incumbent. The independent finding that the “redistricting

process” and the “resulting map” were “taint[ed]” with “improper partisan intent”

was relegated to “buttress[ing]” the “inference” of improper intent based on the

tier-two violation.

We conclude that the trial court erred in focusing first on tier-two violations

at the expense of the evidence of tier-one violations—violations it specifically

- 71 -

found based on the evidence presented. The trial court’s error was then

exacerbated by its decision to apply an unduly deferential standard to its review of

the map, even after finding the existence of unconstitutional partisan intent.

Certainly, this Court explained in Apportionment I that the judiciary’s role

in reviewing an apportionment plan enacted by the Legislature is “not to select the

best plan, but rather to decide whether the one adopted by the legislature is valid.”

Apportionment I, 83 So. 3d at 608 (quoting In re Apportionment Law—1992, 597

So. 2d at 285). At that time, this Court stated the general principle that it would

“defer to the Legislature’s decision to draw a district in a certain way, so long as

that decision does not violate the constitutional requirements.” Id.

But, in Apportionment I, this Court was conducting a “facial” review of the

legislative apportionment plan, without fact-finding, to determine whether any

improper intent existed in the plan. Unlike that context, here, the trial court found

the existence of improper intent, based on evidence presented during an adversarial

trial, yet still applied a deferential standard of review. That was error.

The trial court conducted its review as if it were premature to directly

address the impact of the tier-one violations the trial court itself specifically found.

In particular, the trial court found that the Legislature had “cooperat[ed]” and

“collaborat[ed]” with partisan political operatives to draft an apportionment plan

favoring the Republican Party and incumbents—in other words, a finding of a tier-

- 72 -

one constitutional violation. While the Legislature is generally entitled to

deference as a result of its role in the redistricting process, that deference applies

only “so long as [its redistricting] decision[s] do[] not violate the constitutional

requirements.” Apportionment I, 83 So. 3d at 608.

Once a tier-one violation of the constitutional intent standard is found, there

is no basis to continue to afford deference to the Legislature. To do so is to offer a

presumption of constitutionality to decisions that have been found to have been

influenced by unconstitutional considerations. The existence of unconstitutional

partisan intent is contrary to the very purpose of the Fair Districts Amendment and

to this Court’s pronouncements regarding the state constitutional prohibition on

partisan political gerrymandering.

Accordingly, after reaching the conclusion that the “redistricting process”

and the “resulting map” had been “taint[ed]” by unconstitutional intent, the burden

should have shifted to the Legislature to justify its decisions, and no deference

should have been afforded to the Legislature’s decisions regarding the drawing of

the districts. In other contexts, states have placed the burden on their legislatures

to justify the validity of a redistricting plan when the plan has “raised sufficient

issues” with respect to state constitutional requirements. In re Legislative

Districting of State, 805 A.2d 292, 325 (Md. 2002).

- 73 -

Because there are many ways in which to draw a district that complies with,

for example, the constitutional requirement of compactness, which party bears the

burden of establishing why a decision was made to accept or reject a particular

configuration can ultimately be determinative. This can be seen in reviewing the

seven maps initially released to the public by the House.

All of these maps were considered by the Legislature to be maps that

complied with the tier-two constitutional standards. But, in one of the maps,

designated as H000C9001, there were as few as 14 Republican districts based on

2008 presidential election data and 15 Republican districts based on 2012

presidential data. In the map chosen by the House to move forward in the process,

designated as H000C9011, there were 16 Republican districts under both the 2012

and 2008 presidential results. And, after additional revisions, the Legislature’s

enacted map performed with 17 Republican districts under the 2008 data and 16

using the 2012 data—actually more favorable to Republicans than the performance

of the admittedly gerrymandered 2002 districts under the same data.10 This

consistent improvement in the Republican performance of the map—even when

10. The 2002 benchmark plan performed with 15 Republican districts under

the 2008 presidential data and 14 Republican districts under the 2012 data, though

there were only twenty-five total districts in that map, as compared to twenty-seven

total districts in the 2012 map after Florida gained two districts based on the results

of the 2010 Census.

- 74 -

comparing maps the Legislature itself produced and considered tier-two

compliant—reveals that there are many ways to draw constitutionally compliant

districts that may have different political implications.

Since the trial court found that the Legislature’s intent was to draw a plan

that benefitted the Republican Party, the burden should have been placed on the

Legislature to demonstrate that its decision to choose one compact district over

another compact district, or one tier-two compliant map over another tier-two

compliant map, was not motivated by this improper intent. This is particularly true

where the challengers presented evidence that the Legislature’s choices ultimately

benefitted the Republican Party and also showed alternative maps that performed

more fairly.11 Unlike in Apportionment I, where this Court remained deferential to

the Legislature’s decisions in the absence of a finding of improper intent, there is

11. The Legislature has strongly disputed the relevance of these alternative

maps, going so far as to assert that this Court should not consider the alternative

maps at all because they were either drawn by partisan operatives aligned with the

Democratic Party or of unknown origin. But alternative maps are not on trial

themselves, as is the Legislature’s map, and they can provide “relevant proof that

the Legislature’s apportionment plans consist of district configurations that are not

explained other than by the Legislature considering impermissible factors, such as

intentionally favoring a political party or an incumbent”—as the trial court found

the Legislature to have done in this case. Apportionment I, 83 So. 3d at 611.

Nevertheless, we have reviewed only the alternative maps actually introduced into

evidence during the trial and remedial proceedings, rather than any of the

summary-judgment maps, and have relied on those maps only insomuch as they

show alternate ways—not necessarily the best or legally required way—to

configure the districts.

- 75 -

no longer any basis for this Court to be deferential to the Legislature in fulfilling its

own “solemn obligation to ensure that the constitutional rights of its citizens are

not violated and that the explicit constitutional mandate to outlaw partisan political

gerrymandering . . . in redistricting is effectively enforced.” Apportionment IV,

132 So. 3d at 137.

VI. LEGAL EFFECT OF FINDING UNCONSTITUTIONAL INTENT

Having now concluded that the trial court erred in the standard of review it

applied, we proceed to consider the legal effect of the trial court’s finding of

unconstitutional intent under the appropriate standard. In so doing, we reject the

dissent’s view that we have “transgressed the boundaries of proper appellate

review” and “abandon[ed]” the “restraints of the appellate process.” Dissenting

op. at 122, 124. The Legislature vigorously defended the challenged districts.

Rather than foster additional delay and risk another election under unconstitutional

districts, we have all the record evidence necessary to evaluate now whether the

Legislature’s justifications can withstand legal scrutiny.

A. DISTRICT 5

We begin with District 5, which has been a focal point of the challenge to

the Legislature’s redistricting plan. Initially, the trial court invalidated District 5 as

“visually not compact, bizarrely shaped,” and in contravention of “traditional

political boundaries as it winds from Jacksonville to Orlando,” narrowing at one

- 76 -

point to the width of a highway. After the Legislature removed an appendage from

Seminole County and widened the district, however, the trial court upheld the

remedial version of District 5, concluding that while still “not a model of

compactness,” the revised district is “much improved.” Deferring to the

Legislature, the trial court summarily rejected the challengers’ proposed East-West

configuration of the district, determining that although this configuration was

“somewhat more compliant” with the constitutional standards, there were

“legitimate non-partisan policy reasons for preferring a North-South configuration

for this district over an East-West configuration.”

The trial court did not elaborate as to what any of these “non-partisan policy

reasons” were. The only legal justification offered by the Legislature, with the

support of the Florida State Conference of NAACP Branches, for preferring a

North-South configuration to an East-West orientation is to comply with the

minority voting protection requirements of the Florida Constitution and the federal

Voting Rights Act—specifically, that a North-South configuration is necessary to

avoid diminishing the ability of black voters to elect a candidate of their choice.

The challengers contend, though, that the North-South configuration of this

district is a linchpin to the Legislature’s efforts to draw a map that favors the

Republican Party. They allege that the North-South configuration overpacks

Democratic-leaning black voters into the district—that is, places more black voters

- 77 -

in the district than is necessary to ensure that they can elect a candidate of choice—

thereby diluting the influence of Democratic minorities in surrounding districts.

The challengers rely in part on a trilogy of cases in federal court that trace the

unique history of this district, culminating in a 2002 decision from a three-judge

panel finding that the Legislature’s “overriding goal with respect to congressional

reapportionment” was to adopt a plan that “would maximize the number of

districts likely to perform for Republicans.” Martinez, 234 F. Supp. 2d at 1300-01;

see also DeGrandy v. Wetherell, 794 F. Supp. 1076, 1087-88, 1090 (N.D. Fla.

1992) (adopting a redistricting plan drawn by an independent expert, which created

the predecessor to District 5 as a black majority-minority district); Johnson v.

Mortham, 926 F. Supp. 1460, 1466-67, 1472, 1495 (N.D. Fla. 1996) (noting that

the prior version of District 5 split every one of the fourteen counties that made up

the district, and even split individual precincts, and declaring that the district was

“racially gerrymandered” in violation of the Equal Protection Clause).

Even as redrawn by the Legislature and approved by the trial court, District

5 clearly does not strictly adhere to the Florida Constitution’s tier-two

requirements of compactness and the utilization of political or geographical

boundaries where feasible. It splits seven counties and has numerical compactness

scores of .127 on the Reock measure and .417 on the Convex Hull measure, where

1 is the best score. The critical determination, then, is whether the North-South

- 78 -

configuration of this district, which extends from Jacksonville to Orlando, is

necessary to comply with either the federal Voting Rights Act or the tier-one state

constitutional requirement that no district shall be drawn in such a way as to

diminish the ability of black voters to elect a representative of their choice—the

justifications offered by the Legislature.

Having reviewed the arguments of the parties in detail and studied the

unique nature of this district, we conclude that the Legislature has failed to meet its

burden to demonstrate that District 5, even as revised, passes constitutional muster.

We further conclude that, because the trial court found that District 5 was a key

component of the Legislature’s unconstitutional intent in the drawing of the

congressional redistricting plan, the trial court erred in conducting only a cursory

review of the remedial district and deferring to the Legislature’s North-South

configuration on the basis of unstated “non-partisan policy reasons.”

Since the Legislature cannot prove that the North-South configuration is

necessary to avoid diminishing the ability of black voters to elect a candidate of

their choice, we hold that District 5 must be redrawn in an East-West manner.

While the dissent suggests that this holding displaces the Legislature’s chosen

configuration with a configuration drawn by operatives aligned with the

Democratic Party, see dissenting op. at 121, the argument for an East-West

orientation of this district is not exclusive to the Democratic Party. In fact, an

- 79 -

East-West orientation is the only alternative option, and one that the Legislature’s

own map drawers—insulated, the dissent itself states, from partisan influence—

considered during the redistricting process.

We reach our conclusion as to the continued unconstitutionality of District 5

for several reasons. First, the Legislature’s configuration was entitled to no

deference in light of the trial court’s finding of unconstitutional intent. The trial

court clearly found that the Legislature’s intent in drawing the congressional

redistricting plan generally, and District 5 specifically, was to benefit the

Republican Party. The Legislature’s configuration also had the effect of

benefitting the long-time incumbent of the district, Congresswoman Corrine

Brown, who previously joined with leading Republicans in actively opposing the

Fair Districts Amendment and redistricting reform. See Brown, 668 F.3d 1271.

Indeed, the remedial version of District 5 still retains approximately 80% of its

2002 benchmark—a redistricting map that was admittedly gerrymandered to favor

the Republican Party and incumbents. See Martinez, 234 F. Supp. 2d at 1340.

Retaining the same basic shape, while merely tweaking a few aspects of the

district, does not erase its history or undo the improper intent that the trial court

found. The trial court’s decision to defer to the Legislature’s configuration is

contrary to the proper standard that should have applied—shifting the burden to the

- 80 -

Legislature to justify its enacted configuration—particularly where the trial court

itself continued to acknowledge that the district is “not a model of compactness.”

We conclude that the Legislature cannot justify its enacted configuration.

Despite the Legislature’s repeated contentions that a North-South orientation of the

district is the only option and is essential to avoid diminishing the ability of black

voters to elect a candidate of their choice, there is simply insufficient evidence to

support that assertion. Indeed, legislative staffer Alex Kelly initially drew an East-

West version of the district, with a BVAP of 44.96%, and concluded that such a

configuration would be constitutionally compliant. The Legislature relies on the

trial court’s finding that Kelly was straightforward and credible elsewhere, but

offers no persuasive explanation as to why this version of District 5 was rejected or

why Kelly’s assessment in this circumstance was incorrect.

During the trial, the Legislature argued that it had increased the BVAP of

District 5 over 50%—a decision made during a non-public meeting at the end of

the redistricting process—in order to prevent vote dilution and avoid retrogression.

The trial court specifically found that argument to be “not supported by the

evidence” and there to have been no showing that a majority-minority district was

“legally necessary.”

After redrawing the district, the BVAP of remedial District 5 is 48.11%.

The Legislature continues to argue that any additional diminishment in the BVAP

- 81 -

would prevent black voters from electing a candidate of their choice. But neither

the evidence, nor the case law, bears this out.

As of 1996, following the decision in Johnson that required the predecessor

district to be redrawn, the predecessor to District 5 had a total black population of

47.0% and a total BVAP of 42.7%. Martinez, 234 F. Supp. 2d at 1308. By 2000,

the benchmark district had a total black population of 50.8% and a total BVAP of

46.7%. Id. The district performed for the black candidate of choice in every

election from 1992 through 2000. Id.

In 2002, the total black population of the district was 51.4%. Id. at 1307.

The total BVAP was 46.9%. Id. The federal court in Martinez determined that the

BVAP of 46.9% “will afford black voters a reasonable opportunity to elect

candidates of choice and probably will in fact perform for black candidates of

choice.” Id. The actual election results show this to be true—the district has

continued to perform for the black candidate of choice in every election from 2000

through the present.

The challengers’ proposed East-West configuration of the district has a

BVAP of 45.12%—higher than the BVAP in the initial draft district drawn by

Alex Kelly. This is well within the range of the 42.7%, 46.7%, and 46.9% BVAP

percentages that were addressed by the federal court in Martinez and considered to

be sufficient to “afford black voters a reasonable opportunity to elect candidates of

- 82 -

choice” and to “in fact perform for black candidates of choice.”12 Id. “This is so

in part because,” the federal court in Martinez stated,

blacks constitute 61.3% of registered Democrats in [the predecessor to

District 5], and Democrats constitute 63.8% of registered voters.

Republicans constitute only 22.7% of registered voters. Actual voting

also is strongly Democratic; in the 2000 presidential election, voters

in [the predecessor to District 5] voted 63.7% for Mr. Gore and 34.2%

for Mr. Bush. The black candidate of choice is likely to win a

contested Democratic primary, and the Democratic nominee is likely

to win the general election.

Id. at 1308.

The same logic applies to an East-West configuration of the district. Black

voters constitute 66.1% of registered Democrats under this configuration, and

Democrats constitute 61.1% of registered voters. Republicans, by contrast,

constitute only 23.0% of registered voters. This compares very favorably to the

same respective numbers in the 2002 district upheld by the federal court in

Martinez.

12. Although the dissent states that our review of minority voting strength

as to the East-West configuration of District 5 ultimately amounts to “we know

retrogression when we see it,” dissenting op. at 121, we clearly rely on long-

standing precedent applied by the three-judge federal district court panel in

Martinez—the last time this exact district was challenged. Our conclusion that a

BVAP of 45.12% does not diminish the ability of black voters to elect a candidate

of choice—a BVAP percentage squarely within the range of prior BVAP

percentages that precedent has established not to diminish the ability of black

voters to elect a candidate of choice—is hardly subjective or arbitrary.

- 83 -

Thus, in an East-West orientation of the district, the black candidate of

choice is still likely to win a contested Democratic primary, since black voters

constitute 66.1% of registered Democrats. And the Democratic candidate is still

likely to win the general election, since Democratic voters outnumber Republicans

61.1% to 23.0%. In other words, just as noted in Martinez as a basis for

concluding that the prior version of District 5 afforded black voters a reasonable

opportunity to elect a candidate of choice, “[t]he black candidate of choice is likely

to win a contested Democratic primary, and the Democratic nominee is likely to

win the general election.”13 Martinez, 234 F. Supp. 2d at 1308.

The Legislature’s contrary argument rests entirely on the premise that the

BVAP of the district cannot be decreased from 48.11% to 45.12%. Of course, the

trial court already rejected the Legislature’s argument, based on the same asserted

interest in protecting black voters, that the BVAP needed to be over 50%, and in

urging this Court to uphold the revised district, the Legislature has now tacitly

conceded that 48.11% is sufficient.

13. Contrary to the dissent’s accusation that we fail to apply any objective

standard to our retrogression review of the minority voting strength of an East-

West district, see dissenting op. at 120-21, our analysis is consistent with the

standard set forth by this Court in Apportionment I: “To undertake a retrogression

evaluation requires an inquiry into whether a district is likely to perform for

minority candidates of choice.” 83 So. 3d at 625. This is precisely what we have

done with respect to a proposed East-West orientation of District 5.

- 84 -

But, beyond that, the United States Supreme Court has recently articulated—

in a case with a similar claim of overpacking black voters to maintain the

continued political dominance of the Republican Party in surrounding districts—

that the BVAP itself cannot be viewed in a vacuum. In Alabama Legislative Black

Caucus v. Alabama, 135 S. Ct. 1257, 1272 (2015), the Supreme Court emphasized

that it is the “ability to elect a preferred candidate of choice,” not “a particular

numerical minority percentage,” that is the pertinent point of reference.

The language of the Voting Rights Act that protects against adopting a

redistricting plan that “has the purpose of or will have the effect of diminishing the

ability of [the minority group] to elect their preferred candidates of choice”—

language incorporated into our tier-one state constitutional standards—“does not

require maintaining the same population percentages.” Id. at 1272-73. Instead, the

Supreme Court has told us, this requirement “is satisfied if minority voters retain

the ability to elect their preferred candidates.” Id. at 1273. Providing an example,

the Supreme Court stated that “it would seem highly unlikely that a redistricting

plan that, while increasing the numerical size of the district, reduced the percentage

of the black population from, say, 70% to 65% would have a significant impact on

the black voters’ ability to elect their preferred candidate.” Id.

Accordingly, we reject the Legislature’s argument that an East-West version

of the district would diminish the ability of black voters to elect a candidate of their

- 85 -

choice. We also reject the dissent’s contention that an East-West district causes

the redistricting map to become significantly less compact. See dissenting op. at

118-20. There is no doubt, as noted by the dissent, that the East-West orientation

is longer, with a correspondingly greater perimeter and area. But length is just one

factor to consider in evaluating compactness.

As this Court stated in Apportionment I, “the object of the compactness

criterion is that a district should not yield ‘bizarre designs.’ ” 83 So. 3d at 634.

And as the Supreme Court of Washington has recognized, in a decision cited

favorably by this Court in Apportionment I, “the phrase ‘as compact as possible’

does not mean ‘as small in size as possible,’ but rather ‘as regular in shape as

possible.’ ” Kilbury v. Franklin Cnty. ex rel. Bd. of Cnty. Comm’rs, 90 P.3d 1071,

1077 (Wash. 2004).

There is no doubt that an East-West version of District 5 is visually less

“unusual” and “bizarre” than the meandering North-South version enacted by the

Legislature. Apportionment I, 83 So. 3d at 634. There is also no doubt that the

numerical compactness scores actually favor the East-West orientation: the

different configurations have essentially the same Reock score (.12 for a proposed

East-West version of the district, and .13 for the Legislature’s North-South, where

1 is the most compact), while an East-West district fares significantly better on the

Convex Hull measure (.71 for the East-West as compared to .42 for the North-

- 86 -

South, where 1 is again the most compact). Further, an East-West orientation

allows for fewer incorporated city and county splits than the Legislature’s North-

South district—another consideration in determining tier-two compliance.

The reality is that neither the North-South nor the East-West version of the

district is a “model of compactness,” as the trial court stated. Other factors account

for this phenomenon, “including geography and abiding by other constitutional

requirements such as ensuring that the apportionment plan does not deny the equal

opportunity of racial or language minorities to participate in the political process or

diminish their ability to elect representatives of their choice.” Id. at 635. And

while the dissent cherry-picks a favorable statistic to highlight the supposed

decrease in the compactness of District 2 under an East-West version of District 5,

see dissenting op. at 120, the challengers have demonstrated that the decrease in

the compactness of District 2 is an outlier; in fact, as few as four and as many as

seven other districts can be drawn in a more compact manner by drawing District 5

from East to West.

The bottom line is that none of the Legislature’s justifications for its

gerrymandered version of District 5, and none of its complaints about an

alternative East-West configuration, can withstand legal scrutiny. Because the trial

court erred in deferring to the Legislature’s enacted North-South configuration, and

- 87 -

because the Legislature cannot justify this configuration, District 5 must be

redrawn in an East-West orientation.

B. DISTRICTS 13 & 14

We turn next to Districts 13 and 14, in the Tampa Bay area, which the

challengers contended were drawn in violation of the constitutional requirements.

In addition to relying on the trial court’s finding that the entire map was tainted by

unconstitutional intent, the challengers asserted specifically that the Legislature’s

configuration of Districts 13 and 14 mirrored the configuration known to have

been favored by political operatives, in which District 14 was drawn to cross

Tampa Bay from Hillsborough County, splitting Pinellas County and the City of

St. Petersburg to move a portion of the black population from District 13 into

District 14. In support, the challengers pointed to an e-mail communication from

consultant Frank Terraferma to consultant Rich Heffley and an employee of the

Republican Party of Florida, which described this region as “far from perfect” in a

draft map where District 14 did not cross Tampa Bay.

The enacted configuration of these two districts, which crossed Tampa Bay,

added more Democratic voters to an already safely Democratic District 14, while

- 88 -

ensuring that District 13 was more favorable to the Republican Party.14 The

challengers thus contended that the Legislature’s configuration of these districts

was directly connected to the trial court’s finding that the enacted map was

unconstitutionally drawn to favor the Republican Party.

The trial court denied the challenge to these districts, reasoning that there

were “no flagrant tier-two deviations from which” to “infer unlawful intent.” The

trial court stated that it “simply” could not “conclude, on partisan effect alone, that

the decision to incorporate portions of South St. Petersburg into District 14 was

done with the intent to benefit the Republican Party or the incumbent member of

Congress.”

We conclude that the trial court erred in rejecting the challenge to these

districts. The trial court erroneously required a “flagrant tier-two deviation” in

order to “infer unlawful intent,” rather than viewing the configuration of these

districts through the lens of the direct and circumstantial evidence of improper

intent presented at trial. Once the trial court found unconstitutional intent, the

Legislature’s enacted configuration was no longer entitled to deference, and it

14. Indeed, although District 13 still leans Democratic under the elections

data relied on by the parties, it actually elected Republican Representative David

Jolly over Democrat Alex Sink in a close election in 2014.

- 89 -

becomes the Legislature’s burden to justify its decision to draw the districts in a

certain way.

The Legislature’s asserted justification for picking up voters from Pinellas

County in District 14 was to increase minority voting strength in that district,

which the Legislature considered to be preferable—though not required—from a

state constitutional tier-one and federal Voting Rights Act perspective. The trial

court did not, however, make any findings that it was necessary to add black voters

from Pinellas County to District 14 in order to avoid diminishing the ability of

black voters to elect a representative of their choice.15

During trial, the challengers showed that it is possible not to cross Tampa

Bay and still maintain tier-two compliance. In fact, as the charts below indicate,16

following the county boundary significantly increases the numerical compactness

scores of District 13, although it does cause a decrease in the scores of surrounding

districts.

15. District 14 was, prior to 2012, and still is, under the 2012 map,

represented by Kathy Castor, a white Democratic congresswoman.

16. We use graphical depictions of maps that were included in the

challengers’ brief because those maps show the particular areas of concern. The

Legislature did not contest the accuracy of these graphics. In any event, we

include them only as visual aids and have, in our analysis, relied solely on the data

and maps introduced into evidence during the trial.

- 90 -

Reock Score17 Convex Hull Score18

Enacted Plan Romo Alternative Enacted Plan Romo Alternative

(H000C9047) (Romo A) (H000C9047) (Romo A)

CD12 0.40 0.38 0.81 0.79

CD13 0.46 0.57 0.82 0.91

CD14 0.36 0.28 0.69 0.60

CD15 0.44 0.33 0.75 0.67

CD16 0.42 0.32 0.81 0.80

CD17 0.64 0.39 0.83 0.68

AVG. 0.45 0.38 0.79 0.74

17. The Reock, or circle-dispersion, method of quantifying compactness

“measures the ratio between the area of the district and the area of the smallest

circle that can fit around the district.” Apportionment I, 83 So. 3d at 635. “This

measure ranges from 0 to 1, with a score of 1 representing the highest level of

compactness as to its scale.” Id.

18. The Area/Convex Hull method, which “measures the ratio between the

area of the district and the area of the minimum convex bounding polygon that can

enclose the district,” also ranges from 0 to 1, “with a score of 1 representing the

highest level of compactness.” Apportionment I, 83 So. 3d at 635. “A circle,

square, or any other shape with only convex angles has a score of 1” under this

measure. Id.

- 91 -

In rejecting the challenge to these districts, the trial court emphasized the

decrease in the overall compactness scores in the region, ultimately determining

that the Legislature “was not required to make this tradeoff in compactness to

avoid splitting Pinellas County.” However, as this Court has recognized, following

county lines may result in a reduction in compactness scores. See Apportionment

I, 83 So. 3d at 635 (explaining that the compactness of the districts “cannot be

considered in isolation” because other factors influence a district’s compactness,

including the “Legislature’s desire to follow political or geographical boundaries

or to keep municipalities wholly intact”).

The trial court’s decision to defer to the Legislature’s configuration was

contrary to the proper standard that should have applied once the trial court found

that the Legislature’s intent in drawing the congressional redistricting plan was to

benefit the Republican Party. Because the Legislature cannot justify its enacted

configuration of these districts based on race—the only justification that was

offered—the trial court should have invalidated these districts. Accordingly,

Districts 13 and 14 must be redrawn to avoid crossing Tampa Bay.

C. DISTRICTS 26 & 27

The challengers also mounted an individual attack against the validity of

Districts 26 and 27, claiming that the enacted configuration of these two districts

needlessly divided the City of Homestead to Republican gain—turning one

- 92 -

Republican district and one Democratic district into two Republican-leaning

districts.19 In support, the challengers relied on the general evidence of improper

intent in the plan as a whole, as well as specifically on an e-mail chain between

consultants Heffley, Terraferma, and Reichelderfer that took place after the Senate

released a draft map that did not split Homestead. In this e-mail chain, the

operatives stated that the configuration of these districts was “pretty weak” and

that the House “need[ed] to fix” it. The Senate’s draft version, not splitting

Homestead, is shown on the left below, with the enacted map on the right.

The trial court denied the challenge to these two districts, stating that any

tier-two differences between the enacted map and an alternative map introduced

19. The two districts are actually represented by members of the Republican

Party. The performance data relied on by the parties shows that these two districts

are Republican under the 2008 presidential and 2010 gubernatorial elections but

Democratic under the 2012 presidential election.

- 93 -

into evidence during trial by the challengers were de minimis since the enacted and

alternative plans split about the same number of counties and cities in the region.

The trial court stated that it would have been “selecting a plan that [it] found

subjectively better rather than determining if [the challengers] have proved the

enacted plan invalid,” if it were to invalidate the enacted configuration of these

districts “based on the objective tier-two evidence” presented. The trial court also

summarily concluded that it did not find the enacted configuration to have been

“based on unlawful partisan intent.”

We conclude that the trial court erred in rejecting the challenge to these

districts. Based on the trial court’s finding of unconstitutional intent to benefit the

Republican Party, the burden should have shifted to the Legislature to justify its

configuration of these districts. Thus, instead of deferring to the Legislature’s

configuration and refraining from “selecting a plan [it] found subjectively better,”

the trial court should have required the Legislature to demonstrate that the decision

to split Homestead between Districts 25 and 26 was not done to benefit the

Republican Party. Because the Legislature’s asserted justification for its

configuration of these districts—to protect minority voting rights—simply cannot

be justified, these districts must be redrawn to avoid splitting Homestead.

D. DISTRICT 25

- 94 -

Along with the individual challenge to Districts 26 and 27 based on the split

of Homestead, the challengers also argued that nearby District 25 needlessly

divided Hendry County, in violation of the constitutional requirements. The trial

court summarily rejected this challenge, considering it in conjunction with the

challenge to Districts 26 and 27 and concluding simply that the challengers had not

proved invalidity because they had not demonstrated more than “de minimis” tier-

two deficiencies.

This was error. Having found improper intent in the adoption of the

redistricting plan, the trial court should not have deferred to the Legislature’s

configuration but should have, instead, shifted the burden to the Legislature to

justify its decision to divide Hendry County.

The decision to adopt a configuration of District 25 that split Hendry

County—as the Senate’s map had done but the House’s had not—was made in a

non-public meeting at the end of the redistricting process. There is thus no record

- 95 -

from the time this decision was made to explain why the Legislature chose the

Senate’s configuration of this district over the House’s, even though the Senate’s

configuration rendered the district less numerically compact while splitting a

county boundary and without improving the compactness of the adjacent district,

District 20.

The Legislature’s asserted justification at trial and on appeal in this Court for

splitting Hendry County is chiefly based on concerns related to preclearance

requirements under Section 5 of the Voting Rights Act, since Hendry County was a

“covered” jurisdiction to which Section 5 applied—that is, a county for which the

state had to obtain federal permission prior to enacting any law related to voting in

that county. The Legislature argues that, if it had placed Hendry County entirely

within District 25, the Department of Justice would have denied preclearance.

We reject the Legislature’s justification for its decision to split Hendry

County for at least two reasons. First, the House itself had drawn District 25 with

Hendry County almost entirely included in the district, and the House considered

its map to be constitutionally compliant. The Legislature’s concerns about

preclearance thus appear to be post-hoc rationalizations for the enacted

configuration.

Second, to the extent preclearance is offered as a justification, preclearance

concerns are no longer applicable after the United States Supreme Court’s decision

- 96 -

in Shelby County v. Holder, 133 S. Ct. 2612 (2013). In that case, the Supreme

Court invalidated the “coverage formula” in Section 4 of the Voting Rights Act,

thereby effectively invalidating the preclearance process established by Section 5

of the Act unless and until Congress creates another formul

This text is long and has been trimmed here. Open the source document for the complete record.