PRESS ROBINSON, еt al v. KYLE ARDOIN, in his official capacity as Secretary of State for Louisiana
CIVIL ACTION 22-211-SDD-SDJ
UNITED STATES DISTRICT COURT MIDDLE
June 6, 2022
RULING AND ORDER
Before the Court are the Motion for Preliminary Injunction1 filed by the Robinson Plaintiffs and the Motion for Preliminary Injunction2 by the Galmon Plaintiffs. Defendant Secretary Ardoin and the Intervenor Defendants filed Oppositions,3 to which Plaintiffs filed Replies.4 The Court also received a Brief Amicus Curiae in Support of Neither Party5 from a group of mathematics and computer science professors at Louisiana State and Tulane Universities. A five-day hearing on the Motions was held, beginning May 9, 2022 and ending May 13, 2022. After the hearing, Plaintiffs and Defendants (along with the Intervenor Defendants) both filed Proposed Findings of Fact,6 as well as post-hearing briefs.7
For the reasons set forth herein, the Court concludes that Plaintiffs are substantially likely to prevail on the merits of their claims brought under Section 2 of the
The appropriate remedy in this context is a remedial congressional redistricting plan that includes an additional majority-Black congressional district. The United States Supreme Court instructs that the Legislature should have the first opportunity to draw that plan.9 Therefore, the Court ORDERS the Louisiana Legislature to enact a remedial plan on or before June 20, 2022. If the Legislature is unable to pass a remedial plan by that date, the Court will issue additional orders to enact a remedial plan compliant with the laws
BACKGROUND
I. Procedural Posture
In April 2021, the United States Census Bureau delivered the 2020 Census data that would drive the state of Louisiana‘s redistricting process. Under the new numbers, Louisiana‘s congressional apportionment was unchanged from 2010, holding steady at six seats in the U.S. House of Representatives.11 The task of redrawing those six districts fell upon the Louisiana Legislature, where the drawing of new maps was guided in part by Joint Rule No. 21, passed by the Louisiana Legislature in 2021 to establish criteria that would “promote the development of constitutionally and legally acceptable redistricting plans.”12 Joint Rule 21 provided as follows:
Joint Rule No. 21. Redistricting criteria
A. To promote the development of constitutionally and legally acceptable redistricting plans, the Legislature of Louisiana adopts the criteria contained in this Joint Rule, declaring the same to constitute minimally acceptable criteria for consideration of redistricting plans in the manner specified in this Joint Rule.
B. Each redistricting plan submitted for consideration shall comply with the Equal Protection Clause of the Fourteenth Amendment and the Fifteenth Amendment to the U.S. Constitution; Section 2 of the Voting Rights Act of 1965, as amended; and all other applicable federal and state laws.
C. Each redistricting plan submitted for consideration shall provide that each district within the plan is composed of contiguous geography.
D. In addition to the criteria specified in Paragraphs B, C, G, H, I, and J of this Joint Rule, the minimally acceptable criteria for consideration of a redistricting plan for the House of Representatives, Senate, Public Service Commission, and Board of Elementary and Secondary Education shall be as follows:
(1) The plan shall provide for single-member districts.
(2) The plan shall provide for districts that are substantially equal in population. Therefore, under no circumstances shall any plan be considered if the plan has an absolute deviation of population which exceeds plus or minus five percent of the ideal district population.
(3) The plan shall be a whole plan which assigns all of the geography of the state.
(4) Due consideration shall be given to traditional district alignments to the extent practicable.
E. In addition to the criteria specified in Paragraphs B, C, G, H, I, and J of this Joint Rule, the minimally acceptable criteria for consideration of a redistricting plan for Congress shall be as follows:
(1) The plan shall provide for single-member districts.
(2) The plan shall provide that each congressional district shall have a population as nearly equal to the ideal district population as practicable.
(3) The plan shall be a whole plan which assigns all of the geography of the state.
Leading up to their redistricting session, legislators held a series of “roadshow” meetings across the state, designed to share information about redistricting and solicit public comment and testimony, which lawmakers described as “absolutely vital to this process.”13 Citizens who engaged in the process at the roadshows were assured that “your ideas and recommendations matter to me and they matter to us.”14 The Legislature convened on February
Having long telegraphed that he would,16 Louisiana Governor John Bel Edwards vetoed H.B. 1 and S.B. 5 on March 9, 2022.17 The Legislature voted to override the Governor‘s veto on March 30, 2022.18 That same day, the Robinson and Galmon Plaintiffs filed their Complaints in this Court, alleging that the 2022 congressional map dilutes Black voting strength in violation of the
After the Complaints were filed, Patrick Page Cortez, the President of the Louisiana State Senate, and Clay Schexnayder, the Speaker of the Louisiana House of Representatives (collectively, “the Legislative Intervenors“), moved to intervene as Defendants in the suit, as did Louisiana Attorney General Jeff Landry (“Attorney General Landry” or “the Attorney General“).20
The motions now before the Court -- the Motion for Preliminary Injunction24 by the Robinson Plaintiffs and the Motion for Preliminary Injunction25 by the Galmon Plaintiffs –
were filed on April 15, 2022. Therein, Plaintiffs urge the Court to enjoin Secretary Ardoin from conducting the 2022 congressional elections under the enacted district maps, to set a deadline for the Legislature to enact a compliant map and, if the Legislature fails to do so, to order that the November 2022 election be conducted under one of the illustrative plans proposed by Plaintiffs.26
After a more condensed schedule proposed by the Court drew objections from Defendants, the Court set the Motions for a five-day evidentiary hearing to begin May 9, 2022.27 On the eve of the preliminary injunction hearing, Attorney General Landry filed a Motion to Stay, arguing that the Supreme Court‘s forthcoming merits decision in Merrill v. Milligan, 142 S. Ct. 879 (2022),28 “could be dispositive of this litigation” and will, “[a]t the very least. . .be informative to the Parties’ claims and defenses in the instant case.”29 The Court denied that motion, reasoning that “[t]he blow to judicial economy and prejudice to Plaintiffs that would result from granting the moved-for stay cannot be justified by speculation over future Supreme Court deliberations. . .”30
II. Factual and Legal Background
constitutional and statutory requirements. Perhaps most fundamentally, the “one person, one vote” rule requires that districts be drawn such that one person‘s “vote in a congressional election” is “nearly as is practicable ... worth as much as another‘s.”32 The United States Supreme Court has observed that “to say that a vote is worth more in one district than in another would not only run counter to our fundamental ideas of democratic government, it would cast aside the principle of a House of Representatives elected ‘by the People,’ a principle tenaciously fought for and established at the Constitutional Convention.”33 To that end, districts must be drawn as close to equal in population as possible, and states must “justify population differences
More nuanced are the requirements regarding the consideration of race in redistricting. As many courts have observed, mapdrawers are pulled in one direction by the Equal Protection Clause, which “forbids ‘racial gerrymandering,’ that is, intentionally assigning citizens to a district on the basis of race without sufficient justification.”35 The
Section 2 of the
(a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color, or in contravention of the guarantees set forth in section 10303(f)(2) of this title, as provided in subsection (b).
(b) A violation of subsection (a) is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population.38
A state violates Section 2 “when a state districting plan provides ‘less opportunity’ for racial minorities ‘to elect representatives of their choice.‘”39 “A plaintiff may allege a Section 2 violation in a single-member district if the manipulation of districting lines fragments politically cohesive minority voters among several districts or packs them into one district or a small number of districts, and thereby dilutes the voting strength of members of the minority population.”40 Thornburg v. Gingles41 sets forth three threshold conditions for a claim of vote dilution under Section 2:
district“; second, “that it is politically cohesive“; and third, “that the white majority votes sufficiently as a bloc to enable it ... usually to defeat the minority‘s preferred candidate.”42
If a party establishes the threshold Gingles requirements, the Court will “proceed to analyze whether a violation has occurred based on the totality of the circumstances.”43 The totality of the circumstances determination is made by reference to the “Senate Factors,” which are derived from a report of the Senate Judiciary Committee accompanying the 1982 amendments to the
At the totality of the circumstances stage, courts also consider “whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area.”46 When a statewide districting plan is the subject of a vote dilution claim, “the proportionality analysis ordinarily is statewide.”47
The redistricting process is emphatically within the province of the state legislatures.48 Federal court review, then, represents “a serious intrusion on the most vital of local functions”49 and calls for sensitivity to “the complex interplay of forces that enter
a legislature‘s redistricting calculus.”50 Further, a “presumption of good faith . . . must be accorded legislative enactments.”51
III. Statement of Claims and Defenses
A. Plaintiffs’ Claims
Plaintiffs argue that the enacted map “artificially limits Black voters’ influence” by packing them into CD 2 and cracking them throughout the other five districts. Plaintiffs contend that the maps, “coupled with high levels of racially polarized voting. . .greatly dilute the ability of the State‘s Black voters to elect their candidates of choice.”52 Relying on the illustrative plans prepared by their experts, Anthony Fairfax and William Cooper, Plaintiffs assert that “Louisiana‘s Black community is sufficiently large and geographically compact to comprise more than 50% of the voting-age population in a second congressional
Plaintiffs argue that “[i]t is beyond dispute that Black voters in Louisiana have voted as a cohesive bloc,”54 and that “white voters voted in bloc against the candidate supported by Black voters”55 Thus, Plaintiffs aver that all of the threshold conditions of a vote dilution claim under Gingles are met here. Further, they contend that the vestiges of Louisiana‘s long and irrefutable history of discrimination have resulted in modern day disparate socioeconomic conditions, segregated communities, and unequal educational outcomes,
all of which contribute to a totality of circumstances that denies a meaningful opportunity for Black voters to elect their preferred candidates.
Absent an injunction preventing the enacted maps from controlling the 2022 congressional election, Plaintiffs allege, they will suffer irreparable harm. Moreover, they argue that “preserving the rights of Louisianans is strongly in the public interest and the threat of disenfranchising Black Louisianans vastly outweighs the minimal potential administrative burden that an injunction might impose on Defendant.”56 Though Plaintiffs acknowledge that the Purcell doctrine proscribes judicial intervention on the eve of an election, they distinguish Purcell and progeny factually and point out that here, “the election is over six months away” and that counsel for Louisiana‘s Speaker of the House and Senate President are on the record in companion state court redistricting lawsuits as representing that “[t]he election deadlines that actually impact voters do not occur until October 2022. . .Therefore, there remains several months on Louisiana‘s election calendar to complete the process.”57 Indeed, they explained, Louisiana‘s “election calendar is one of the latest in the nation.”58 Since Louisiana has only six congressional districts, and alternative maps with two majority-minority districts were introduced and debated during the legislative redistricting process, Plaintiffs submit that “only a brief period” should be necessary to craft a VRA-compliant map.59 Plaintiffs argue that these challenges pale in comparison to the harm from proceeding with the 2022 elections under maps that violate Section 2 of the VRA.60
B. Secretary of State Ardoin
Secretary Ardoin begins by questioning Plaintiffs’ standing to maintain this action, arguing that although the Galmon Plaintiffs challenge the entire congressional plan, they “only have Plaintiffs living in Congressional Districts 2, 5, and 6.”61 Second, Secretary Ardoin contends that Plaintiffs are unlikely to succeed on the merits of their Voting Rights Act claim, arguing that their claim fails because the second majority-majority district they propose is not geographically compact. Specifically, the Secretary objects to the manner in which Plaintiffs’ illustrative plans “combine[] portions of EBR [East Baton
Even if Plaintiffs’ proposed districts were sufficiently compact, Secretary Ardoin disputes that the districts would “perform” – that is, that they would actually provide Black voters with an opportunity to elect the candidate of their choice – because Plaintiffs’ illustrative plans create only a “bare majority”66 of Black voting-age population (“BVAP“) in the proposed second majority-minority districts. Further, Secretary Ardoin argues that
Plaintiffs cannot make the requisite showing of racially polarized voting and White bloc voting.
Secretary Ardoin next avers that the totality of the circumstances analysis “show[s] that minority voters possess the same opportunities to participate in the political process and elect their candidate of choice.”67 The Secretary urges that a second majority-minority district is untenable because Louisiana has a “substantial interest in maintaining the continuity of representation in its districting plans.”68 Per the Secretary, the Black population in Louisiana is “remaining flat or even declining”69 such that drawing a second Black congressional district is not justified.
Lastly, Secretary Ardoin argues that the Purcell doctrine forecloses the possibility of judicial intervention in the form of an injunction for the 2022 election cycle. He cites a handful of recent cases where courts have applied Purcell, and the declaration of Louisiana election official Sherri Hadskey, who attests that the process of assigning Louisiana voters to their new districts in the state election database system is complicated and time-consuming, and that doing so before the 2022 cycle would cause “significant cost, confusion, and hardship.”70
C. Intervenor Defendant - Attorney General Landry
The Attorney General argues that Plaintiffs are not substantially likely to succeed on their merits of their claims “as to the first and third Gingles preconditions.”71 Plaintiffs only appear to have proposed a sufficiently numerous and geographically compact
second majority-minority district, he explains, by using “statistical manipulation.”72 Specifically, Attorney General Landry argues, Plaintiffs’ use of the “Any Part Black” metric, which is a census category including anyone who identifies as
The Attorney General challenges the compactness of the illustrative plans as “combin[ing] Black communities from far-flung parts of Louisiana in the same district.”76 The proposed maps are an example of racial gerrymandering, he argues, which is impermissible “even when the purported purpose of the racial gerrymander is in seeking to comply with the dictates of the Voting Rights Act.”77 As for racially polarized voting, the Attorney General argues that partisan affiliation, not race, best explains the tendency of Black Louisianans to vote similarly.
The Attorney General also advances an argument recently credited by the District Court for the Eastern District of Arkansas in Arkansas State Conference of the NAACP v. Arkansas Board of Apportionment, No. 4:21-CV-01239-LPR, 2022 WL 496908 (E.D. Ark. Feb. 17, 2022),78 finding that there is no private right of action under Section 2 of the
that the Court should dismiss Plaintiffs’ claims on that basis. Finally, the Attorney General echoes Secretary Ardoin‘s argument that the Purcell doctrine dictates that it is too late for relief to be granted as to the 2022 congressional election cycle.79
D. The Legislative Intervenors
The Legislators begin by noting that, in the push-pull of the VRA and the Equal Protection Clause, neither proportional representation nor a desire to maximize minority representation are sufficient reasons to create a new majority-minority district. After reviewing the history of the 1990s Hays litigation and the 2020 redistricting process, the Legislators argue that the enacted congressional map should not be invalidated because the “Legislature had before it no evidence justifying race-based redistricting.”80 The Legislators argue that race must have predominated in the drawing of Plaintiffs’ illustrative plans because “[a] set of 10,000 computer-simulated redistricting plans generated without racial criteria and according to neutral principles produces zero majority minority congressional districts . . .let alone two as Plaintiffs demand.”81
Further, the Legislators argue that the illustrative plans offered by Plaintiffs disregard communities of interest because they “combine[] urban Baton Rouge and its suburbs in some way with the distant rural communities of Louisiana‘s delta parishes. . . who share race in common
Court‘s or Plaintiffs,‘” they contend, and Plaintiffs’ plans “dismantle the Legislature‘s legitimate and race-neutral goals.”84 Seconding Attorney General Landry, the Legislators also argue that the “DOJ Black” definition should be used to assess whether Plaintiffs’ proposed maps feature two districts with greater than 50% BVAP.
The Legislators argue that “white bloc voting. . .is low enough (and crossover voting is high enough) to permit Black voters to elect their preferred candidates without 50% BVAP districts.”85 Thus, the cohesion of Black voters or the polarization of the electorate “carries no legal significance.”86 What may appear as cohesive Black voting is equally likely to be the product of partisan politics, the Legislators assert, noting that “[i]t is difficult for any Democratic candidate, white or Black, to win in Louisiana, except under special circumstances.”87
The Legislators argue that Plaintiffs’ case also fails under the totality of the circumstances because they “do not focus on alleged discrimination against a discrete group in a discrete locality, relying instead on statewide elections and statewide ideals of proportionality.”88 Overall, the Legislators assert, it is not clear whether Black Louisianans would be better off with the status quo of one majority-minority district with a strong BVAP of roughly 58%, or two majority-minority districts that only slightly exceed 50% BVAP. In any event, they argue, Purcell demands that the Court abstain from tinkering with the November election.
LAW AND EVIDENCE
I. STANDARD OF REVIEW
“[A] preliminary injunction is ‘an extraordinary remedy never awarded as of right.‘”89 “To obtain a preliminary injunction, the movant must establish four elements: (1) a substantial likelihood of success on the merits, (2) a substantial threat of irreparable injury if the injunction is not issued, (3) that the threatened injury if the injunction is denied outweighs any harm that will result if the injunction is granted, and (4) that the grant of an injunction will not disserve the public interest.”90
II. APPLICABLE LAW
“The essence of a § 2 claim is that a certain electoral law, practice, or structure interacts with social and historical conditions to cause an inequality in the opportunities enjoyed by black and white voters to elect their preferred representatives.”91 Thornburg v. Gingles sets forth three threshold conditions for a claim of
“The ‘geographically compact majority’ and ‘minority political cohesion’ showings are needed to establish that the minority has the potential to elect a representative of its own choice in some single-member district. And the ‘minority political cohesion’ and
‘majority bloc voting’ showings are needed to establish that the challenged districting thwarts a distinctive minority vote by submerging it in a larger white voting population.”93 “Unless these points are established, there neither has been a wrong nor can [there] be a remedy.”94 Consequently, if Plaintiffs fail to establish any one of these three conditions, the Court need not consider the other two.95
Under the first prong of Gingles, “a party asserting § 2 liability must show by a preponderance of the evidence that the minority population in the potential election district is greater than 50 percent.”96 Because “only eligible voters affect a group‘s opportunity to elect candidates,”97 this requirement is analyzed in terms of Black voting-age population (or “BVAP“). Proving the existence of a sufficiently large minority population does not end the inquiry; compactness is also required. If the minority population is dispersed such that a reasonably compact majority-minority district cannot be drawn, “Section 2 does not require a majority-minority district....”98
“While no precise rule has emerged governing § 2 compactness, the inquiry should take into account traditional districting principles such as maintaining communities of interest and traditional boundaries.”99 “Community of interest” is a term of art that has no universal definition in the redistricting context. Visual assessments are appropriate when assessing compactness. “[B]izarre shaping of” a district that, for example, “cut[s] across
pre-existing precinct lines and other natural or traditional divisions,” suggests “a level of racial manipulation that exceeds what § 2 could justify.”100
To determine whether Plaintiffs satisfy the first Gingles requirement, the Court compares the enacted plan with Plaintiffs’ illustrative plans.101 The Court‘s comparison is for the limited purpose of evaluating Gingles I, which requires a district that is “reasonably compact and regular“;102 compactness is not a “beauty contest[]”103
The second and third requirements of Gingles require Plaintiffs to establish that voting in the challenged districts is racially polarized.104 As the Supreme Court has explained, “in the absence of significant white bloc voting it cannot be said that the ability of minority voters to elect their chosen representatives is inferior to that of white voters.”105
If Plaintiffs establish all three Gingles requirements, the Court then analyzes whether a Section 2 violation has occurred based on the “totality of the circumstances.” At this step, the Court considers the Senate Factors, which include:
the history of voting-related discrimination in the State or political subdivision; the extent to which voting in the elections of the State or political subdivision is racially polarized; the extent to which the State or political subdivision has used voting practices or procedures that tend to enhance the opportunity for discrimination against the minority group, such as unusually large election districts, majority vote requirements, and prohibitions against bullet voting; the exclusion of members of the minority group from candidate slating processes; the extent to which minority group members bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process; the use of overt or subtle racial appeals in political
campaigns; and the extent to which members of the minority group have been elected to public office in the jurisdiction.106
Supreme Court precedent also dictates that the Court must consider whether the number of majority-Black districts in the enacted plan is roughly proportional to the Black share of the population in Louisiana.107
Not relevant to the Court‘s inquiry is whether the Louisiana Legislature intended to dilute the votes of Black Louisianans. The Court‘s Section 2 analysis “assess[es] the impact of the contested structure or practice on minority electoral opportunities on the basis of objective factors.”108 The Legislature‘s intent is therefore “the wrong question.”109 “The ‘right question . . . is whether ‘as a result of the challenged practice or structure plaintiffs do not have an equal opportunity to participate in the political processes and to elect candidates of their choice.‘”110
A. Gingles I – Numerosity and Reasonable Compactness
To satisfy the first Gingles requirement, Plaintiffs must establish that Black voters as a group are “sufficiently large and geographically compact to constitute a majority in some reasonably configured legislative district.”112 To establish that, Plaintiffs rely upon the testimony of expert witness William Cooper (“Cooper“).
Defendants stipulated to Plaintiffs’ tender of Cooper as an expert in redistricting, demographics, and census data. Drawing on his 30 years of experience as a demographer, including testifying in more than 50 voting-related federal cases, Cooper opines that “African Americans in Louisiana are sufficiently numerous and geographically compact to allow for two majority-Black U.S. House districts in a six-district plan.”113 Cooper offers four illustrative maps — all of which contain two majority-Black congressional districts based on his analysis.114
Cooper testified that between 2010 and 2020, Louisiana gained approximately 125,000 residents, with all of the gain attributable to minority populations, and about half attributable to gains in the Black population. Conversely, Cooper documented an overall decline in the White population in Louisiana since 1990. Cooper concludes that the Black population, counted using the Any Part Black metric, increased from 32.80% in the 2010 census to 33.13% in the 2020 census — an increase of 56,234 people. According to Cooper, Any Part Black “is the appropriate Census classification to use in
Population by Race and Ethnicity
| All Ages | 1990 | Percent of Total Populatio | 2000 | Percent of Total Populatio | 2010 | Percent of Total Population | 2020 | Percent of Total Population |
|---|---|---|---|---|---|---|---|---|
| Total Population | 4,219,973 | 100.00% | 4,468,976 | 100% | 4,533,372 | 100% | 4,657,757 | 100.00% |
| NH White* | 2,776,022 | 65.78% | 2,794,391 | 62.53% | 2,734,884 | 60.33% | 2,596,702 | 55.75% |
| Total Minority Pop. | 1,443,951 | 34.22% | 1,674,585 | 37.47% | 1,798,488 | 39.67% | 2,061,055 | 44.25% |
| Latino | 93,044 | 2.20% | 107,738 | 2.41% | 192,560 | 4.25% | 322,549 | 6.92% |
| NH Black* | 1,291,470 | 30.60% | 1,443,390 | 32.30% | 1,442,420 | 31.82% | 1,452,420 | 31.18% |
| NH Asian* | 39,302 | 0.93% | 54,256 | 1.21% | 69,327 | 1.53% | 85,336 | 1.83% |
| NH Hawaiian and PI*# | NA | NA | 24,129 | 0.54% | 28,092 | 0.62% | 1,706 | 0.04% |
| NH American Indian and Alaska Native | 17,539 | 0.42% | 1,076 | 0.02% | 1,544 | 0.03% | 25,994 | 0.56% |
| NH Other*~ | 2,596 | 0.06% | 4,736 | 0.11% | 6,779 | 0.15% | 16,954 | 0.36% |
| NH Two or More Races# | NA | NA | 39,260 | 0.88% | 57,766 | 1.27% | 156,096 | 3.35% |
| SR Black (Single-race Black) | 1,299,281 | 30.79% | 1,451,944 | 32.49% | 1,452,396 | 32.04% | 1,464,023 | 31.43% |
| AP Black (Any Part Black) | NA | NA | 1,468,317 | 32.86% | 1,486,885 | 32.80% | 1,543,119 | 33.13% |
117
Cooper opines that, based on the new Census data, “[t]here are a variety of ways to draw two majority-Black congressional districts in Louisiana while adhering to traditional redistricting principles.”118 Cooper‘s four illustrative maps are reproduced below:119
Cooper‘s illustrative plans include two majority-Black congressional districts. A majority-Black district is one in which the Any Part Black Voting Age Population (APBVAP) exceeds 50% in the district. Cooper testified that APB is the obviously appropriate metric, since it has been accepted in many cases throughout the country since the 2003 Supreme Court case Georgia v. Ashcroft.120 Cooper stated that he has relied upon APB since just before the 2010 census, and has applied it in several cases this year, as well as in the 2017 Louisiana case Terrebonne NAACP v. Jindal.121 Cooper
cross-referenced his BVAP data with a registered voter file provided by the Secretary of State in summer 2021, which verified that his CD 2 and CD 5 have over 50% Black registered voters, as well. Further, Cooper testified that even using the most conservative definition of BVAP possible, single race non-Hispanic citizen voting age population, CD 2 and CD 5 in his illustrative plans still exceeded 50% BVAP:
2016-2020 Citizen Voting Age Population by Plan
| 2022 Plan | % NH SR Black CVAP | % NH White CVAP | NH Black CVAP to NH White CVAP Margin | July 2021 Black Registered Voters |
|---|---|---|---|---|
| District 2 | 61.89% | 31.34% | 30.55% | 61.52% |
| Illustrative Plan 1 | ||||
| District 2 | 53.35% | 39.31% | 14.04% | 52.33% |
| District 5 | 50.94% | 46.19% | 4.75% | 51.84% |
| Illustrative Plan 2 | ||||
| District 2 | 53.66% | 39.53% | 14.13% | 52.72% |
| District 5 | 51.26% | 45.92% | 5.34% | 51.53% |
| Illustrative Plan 3 | ||||
| District 2 | 53.40% | 39.31% | 14.09% | 52.33% |
| District 5 | 52.78% | 44.86% | 7.92% | 53.35% |
122
Cooper testified that he could have maximized the Black population in his proposed
Analyzing
that the BVAP in the enacted CD 2 is 58.65%, while the other five districts have under 34% BVAP. Cooper concluded that the enacted plan “packs and cracks Black voters.”123
Cooper explained that the purpose of the illustrative plans he drew, and of illustrative plans generally, is to demonstrate to the court that it is possible to draw a map with two majority-minority congressional districts that satisfies the first prong of Gingles I — numerosity and compactness — while adhering to traditional redistricting principles. Cooper noted that Plaintiffs did not engage him to produce such a plan no matter what; only to do so if it was feasible. He testified that in other cases, he has declined to draw illustrative maps where it was not possible to add majority-minority districts without violating traditional redistricting principles. This was not such a case. In his view, configuring Louisiana congressional maps with two majority-minority districts with fidelity to traditional redistricting principles was easy and obvious.
Cooper‘s illustrative maps all take roughly the same shape, reaching from East Baton Rouge and St. Landry Parishes in the south to the Delta Parishes along the Louisiana-Mississippi border. Cooper explained that one difference between his illustrative maps and the enacted map is that he made CD 2 and CD 6, which he considered to be irregularly shaped, more regular. On direct examination, he described how his maps perform with respect to traditional redistricting principles. First, he stated, all of his maps comply with one person-one vote; in fact, three of his four plans are zero-deviation plans, meaning that all of the districts have perfectly equal population sizes. Illustrative Plan 4 has an overall deviation of plus or minus 150 persons, which resulted from reconfiguring the maps to split zero precincts.
Cooper opined that his plans “are across-the-board superior to the 2022 plan in terms of parish splits, municipal splits, and CBSA splits,”124 as documented in the following charts:
| Parish Splits | Populated 2020 VTD Splits | Populated Municipal Splits | Single-Parish Populated Municipal Splits* | CBSA splits | |
|---|---|---|---|---|---|
| 2022 Plan | 15 | 0 | 30 | 25 | 18 |
| Illustrative Plan 1 | 10 | 13 | 24 | 18 | 14 |
| Illustrative Plan 2 | 11 | 7 | 30 | 22 | 16 |
| Illustrative Plan 3 | 10 | 12 | 29 | 23 | 17 |
125
Figure 3
Political Subdivision Splits
| Parish Splits | Populated 2020 VTD Splits | Populated Municipal Splits | Single-Parish Populated Municipal Splits* | CBSA splits | |
|---|---|---|---|---|---|
| 2022 Plan (HB 1) | 15 | 1 | 36 | 25 | 18 |
| Illustrative Plan 4 | 10 | 0 | 30 | 21 | 14 |
126
To evaluate splits in political subdivisions, Cooper used Core Based Statistical Areas (“CBSAs“), which are regions defined by the Office of Management and Budget comprised of urban centers and the surrounding areas. The composition of CBSAs is influenced by commuting patterns, commercial activity, and communities of interest. Cooper testified that his illustrative plans split fewer CBSAs than the enacted map and that, overall, his maps better preserved CBSAs and other political subdivisions.
| Plan | Reock | Polsby-Popper | ||||
|---|---|---|---|---|---|---|
| Low | High | Low | High | |||
| HB 1 | ||||||
| Mean of All Districts | .37 | .18 | .50 | .16 | .06 | .34 |
| CD 2 | .18 | .06 | ||||
| Illustrative Plan 1 | ||||||
| Mean of All Districts | .36 | .23 | .53 | .19 | .09 | .27 |
| CD 2 | .23 | .15 | ||||
| CD 5 | .33 | .09 | ||||
| Illustrative Plan 2 | ||||||
| Mean of All Districts | .41 | .23 | .53 | .19 | .09 | .27 |
| CD 2 | .23 | .12 | ||||
| CD 5 | .33 | .09 | ||||
| Illustrative Plan 3 | ||||||
| Mean of All Districts | .38 | .23 | .52 | .18 | .08 | .31 |
| CD 2 | .23 | .15 | ||||
| CD 5 | .30 | .08 | ||||
| Illustrative Plan 4 | ||||||
| Avg. of All Districts | .37 | .23 | .56 | .18 | .08 | .29 |
| CD 2 | .23 | .15 | ||||
| CD 5 | .35 | .09 | ||||
128
Cooper testified that the superior compactness achieved by his plans is easily verifiable by making a simple visual comparison of his plans to the enacted plan, with its “very oddly shaped” CD 2 and wraparound CD 6:
129
Cooper reported that all of his illustrative maps have contiguous districts. Although the enacted plan can technically say the same, Cooper was critical of how that result was achieved. For example, he pointed out that CD 2 flanks both sides of Interstate 10 around the Mississippi River Bridge in Baton Rouge; although being bisected by a body of water
does not technically vitiate the district‘s contiguity, it is not an ideal configuration. Likewise, looking at the enacted plan‘s District 6, Cooper commented that to get from parts of St. John Parish from the rest of the district, one would have to either swim across Lake Maurepas into Livingston Parish or take Interstate 55 and drive through another district entirely.
Based on Cooper‘s analysis, under the enacted plan, only 31% of Louisiana‘s Black population lives in a majority-minority district, while 91.5% of the White population lives in a majority-White district.130 Cooper testified that in his illustrative plan, approximately 50% of Black people in Louisiana would live in a majority-Black district, while 75% of White people would live in a majority-White district.
On cross-examination, Cooper again emphasized that although Plaintiffs did ask him to draw maps with two majority-minority districts, he did not have a goal of creating create two such districts no matter what. He asserted that he followed traditional principles and considered all of the relevant factors. Defendants’ cross of Cooper focused heavily on the inclusion of East Baton Rouge and the Delta Parishes in one congressional district. Cooper testified that, based on the data, he found very clearly defined neighborhoods that were overwhelmingly Black in some areas. This compactness in the Black population made it easy to join Black areas to other Black areas to draw a majority-Black district. Socioeconomic factors also made the combination of East Baton Rouge and East Carroll Parish a natural one; Cooper testified that he found nothing unusual at all about including them in the same district, though he agreed that poverty is much higher
in East Carroll Parish, with much lower median income for the Black population, and that educational attainment was likewise much lower in East Carroll Parish.
As for respecting communities of interest, Cooper agreed that there is no universal definition of a community of interest, but noted that he tried to keep the Acadiana region relatively intact in his maps, and believed that he did so successfully. Conversely, Cooper stated that nothing in his analysis indicated that the areas around Fort Polk should necessarily be joined as a community of interest, so he did not prioritize that.
Plaintiffs’ expert Anthony Fairfax also gave opinion testimony related to Gingles I. Defendants stipulated to Fairfax‘s expertise in demography, redistricting, and census data. Fairfax prepared a report and two supplemental reports in this case,131
132
Fairfax opines that his illustrative maps prove that Louisiana‘s Black population is sufficiently large and geographically compact to “easily meet[] the first preconditions of [Gingles].”133 Specifically, Fairfax opines that:
It is possible to draw an Illustrative Plan that adheres to federal and state redistricting criteria and contains two majority-Black congressional districts. The Illustrative Plan was drawn with race not predominating and continues to perform as well or better than the enacted plan HB1 on eight out of eight redistricting criteria including: 1) population deviation (equal population or “one person, one vote“); 2) contiguity; 3) compactness; 4) political subdivision splits for parishes; 5) political subdivision splits for Voting Tabulation Districts (“VTDs“); 6) preserving communities of interest for census places; 7) preserving communities of interest for landmark areas; and 8) fracking.134
Fairfax testified that he started with the enacted plan as a baseline. Thus, his illustrative plans retain the current majority-Black district in CD 2, but with adjustments designed to “lessen the presence of District 2 in Baton Rouge and create a more single metro district”135 anchored in New Orleans. Like Cooper, Fairfax drew various versions of CD 5 that connect the Bаton Rouge area to the Delta Parishes along the Louisiana-Mississippi border. Also like Cooper, Fairfax used the Any Part Black definition to conclude that the majority-minority districts in his illustrative plans topped the 50% BVAP mark. Illustrative Plan 1 features a CD 2 with a BVAP of 50.96% and a CD 5 with 52.05% BVAP.136 His Plans 2 and 2A also feature two majority-minority districts (with APBVAP ranging from 51.55% to 51.98%).137
Fairfax stated that APB is a commonly used and accepted definition in this type of analysis, but in response to Defendants’ criticism of that measure, he also assessed his proposed districts’ population using the
Fairfax testified that he looked at equal population, contiguity, compactness, splits, communities of interest, and fracking when drawing his maps. Consideration of Legislature‘s
Table 1 - Illustrative Plan and HB 1 Mean Compactness Measurements
| District | Reock | Polsby-Popper | Convex Hull | Performed Best |
|---|---|---|---|---|
| Illustrative Plan Mean | .42 | .18 | .69 | 3 of 3 |
| Illustrative Plan 2 Mean | .39 | .20 | .71 | 3 of 3 |
| Illustrative Plan 2A Mean | .39 | .20 | .71 | 3 of 3 |
| HB1 Plan Mean | .37 | .14 | .62 | 0 of 3 |
139
As for parish splits, Fairfax‘s plans split either 12 or 14 parishes, while the enacted plan splits 15. Moreover, Fairfax explained that it was his belief, based on studying the
enacted plan, that the Legislature prioritized the elimination of VTD splits — the enacted plan splits none of them. Fairfax‘s plans likewise split none. With respect to communities of interest, Fairfax testified that he analyzed them by considering the number of times his illustrative plans split census places and landmark areas. The category of “census places” includes government entities such as cities and towns, as well as Census Designated Places (CDPs).140 While the enacted plan splits 32 census places, Fairfax‘s plans are superior on this index, splitting either 26 or 31. Fairfax defended his use of census places as communities of interest, explaining that even more than a city or town boundary, a census place is a locally well-known unit and a good indicator of the existence of a community of interest, even though it may not have a governmental body.
In his supplemental report, Fairfax addressed the performance of his Illustrative Plans 1 and 2 on state and traditional redistricting criteria, as compared to the enacted map:
| Criteria | Illustrative Plan 1 | Illustrative Plan 2 | HB1 Plan |
|---|---|---|---|
| Equal Population | 51 | 58 | 65 |
| Contiguity | Y | Y | Y |
| Parish Splits | 14 | 12 | 15 |
| VTD Splits | 0 | 0 | 0 |
| COI Census Places Splits | 31 | 26 | 32 |
| COI Landmark Splits | 58 | 58 | 58 |
| Compactness (mean) | .42, .18, .69 | .39, .20, .71 | .37, .14, .62 |
| Fracking | 5 | 5 | 8 |
Source: Illustrative and HB1 Plans extracted from Maptitude for Redistricting reports 141
Fairfax concludes that his “Illustrative Plan 2 in addition to performing better than the HB1 Plan, exceeds many of Illustrative Plan 1‘s performance on state and traditional
redistricting criteria.”142 Fairfax explains that he considered the testimony of Louisiana residents from the roadshows held by the legislature during the redistricting process to validate his impressions of communities of interest. His supplemental report cites particular testimony and explains how the testimony specifically influenced his consideration of communities of interest.
Fairfax testified that socioeconomic data was another important guiding factor for assessing communities of interest and to ensure respect for commonalities in mapping the districts. He explained how he used the mapping software‘s capabilities to overlay data onto his proposed districts related to, for example, median household income, educational attainment, food stamp percentage, poverty level, percentage of renter households, and community resilience estimates. This information led him to conclude that areas in Ouachita Parish, Rapides Parish, Evangeline Parish, Baton Rouge, and Lafayette could be appropriately grouped together. For example, by overlaying data related to the percentage of the population with no high school education in a given area, it was easy to see that the areas shaded red and orange in the map below, indicative of more people with no high school education, followed a pattern that “clearly define[d] the boundaries of District 5
144
Likewise, by overlaying data from the U.S. Census Bureau‘s Community Resilience Estimate (CRE), Fairfax observed a commonality in the areas that he drew into CD 5:
145
income, communication barriers, number of persons per room in the house, lack of health insurance, and others. So, Fairfax explains, this indicator also suggested to him that the areas in his CD 5 were appropriately grouped together.
On cross-examination, Defendants asked Fairfax to elaborate on how race factored into his drawing process. Fairfax confirmed that he had the ability to display BVAP in his mapping software, but that he only used that feature at the beginning of his process, to get a sense of possible areas where a majority-minority district could be drawn. After he got that initial sense of where BVAP levels were strong, Fairfax testified that he “turned off” the BVAP overlay and focused instead on socioeconomic data, which he used to make more granular mapping decisions. Fairfax did not deny that he considered race to some extent in
Fairfax also elaborated on how his proposed CD 5 was developed. He testified that he started with the existing congressional map and began trimming the area to the west to make the Delta region in Northeast Louisiana a more substantial presence in the district. He then expanded down further to add different areas, starting in the north and working his way south. Fairfax clarified that he does not simply add territory to the district until he reaches the 50% threshold; the BVAP in the district goes up and down along with way based on the addition and subtraction of different
redistricting criteria. In his words, he was asked to focus on the demographics. Bryan explains that he set out to “explore the demographic definition of minorities and show how different definitions can generate different conclusions about whether a district is ‘majority’ or not.”146 On this topic, Bryan opines that “only by the most generous definition of Black, the any part black (APB) measure, do any of the Illustrative Plans meet the traditional majority minority criteria of over 50% +1.”147 Bryan’s demographic analysis is reflected in the following tables setting forth his findings regarding the BVAP of districts in the illustrative plans:
Table III.A.4 Robinson Illustrative Plan Black Share of Voting Age Population
| Illustrative Plan | Black Alone | Black DOJ | Any Part Black |
|---|---|---|---|
| 1 | 16.84% | 17.24% | 18.29% |
| 2 | 48.73% | 49.39% | 50.96% |
| 3 | 16.77% | 17.29% | 17.91% |
| 4 | 30.76% | 31.25% | 31.90% |
| 5 | 50.63% | 51.25% | 52.05% |
| 6 | 15.31% | 15.68% | 16.19% |
Table III.A.5 Galmon Illustrative 1 Plan Black Share of Voting Age Population
| Illustratuve 1 Plan | Black Alone | Black DOJ | Any Part Black |
|---|---|---|---|
| 1 | 16.95% | 17.35% | 18.18% |
| 2 | 47.77% | 48.41% | 50.16% |
| 3 | 18.55% | 19.10% | 19.75% |
| 4 | 30.68% | 31.17% | 31.82% |
| 5 | 48.62% | 49.22% | 50.04% |
| 6 | 16.36% | 16.74% | 17.24% |
Table III.A.6 Galmon Illustrative 2 Plan Black Share of Voting Age Population
| Illustrative 2 Plan | Black Alone | Black DOJ | Any Part Black |
|---|---|---|---|
| 1 | 15.29% | 15.67% | 16.51% |
| 2 | 48.27% | 48.92% | 50.65% |
| 3 | 20.39% | 20.93% | 21.59% |
| 4 | 27.52% | 28.00% | 28.65% |
| 5 | 48.65% | 49.25% | 50.04% |
| 6 | 18.74% | 19.14% | 19.67% |
Table III.A.7 Galmon Illustrative 3 Plan Black Share of Voting Age Population
| Illustrative 3 Plan | Black Alone | Black DOJ | Any Part Black |
|---|---|---|---|
| 1 | 17.35% | 17.74% | 18.52% |
| 2 | 47.77% | 48.41% | 50.16% |
| 3 | 16.82% | 17.35% | 17.98% |
| 4 | 31.79% | 32.29% | 32.96% |
| 5 | 50.23% | 50.81% | 51.63% |
| 6 | 15.14% | 15.53% | 16.09% |
Bryan opined that all of the plans (though he did admittedly not examine Robinson Illustrative Plans 2 and 2A) only achieve two majority-Black districts by using the most expansive metric, Any Part Black.
Bryan testified regarding what he called his “splits analysis,” which he used to conclude that “[b]ased on the surgical, divisive nature of the splits in each of the Plaintiffs’ Illustrative Plans across Louisiana’s places . . . race was the prevailing factor in their design.”149 He explained that his goal was to count how many pieces of geography are split by a given plan, then assess the impact of those splits by examining how many and what races are impacted by
For example, Bryan testified, under the illustrative plans, the city of Baton Rouge is split between CD 2 and CD 6. Roughly 65% of the total Baton Rouge population lives in CD 6, he said, but 94.59% of the White people in Baton Rouge live in CD 6. Conversely, although only about 34% of Baton Rouge residents live in CD 2, 57.21% of Black Baton Rougeans live in CD 2. Bryan testified that these numbers provide some evidence of “misallocation,” stating that, if all else was equal, you would expect White residents to be allocated into the districts in the same way the total population is distributed. Instead, he claims, White voters are “over-indexed” in CD 6. Bryan’s report includes a number оf charts setting forth this “misallocation” analysis under different plans and in different localities.152
Bryan further testified regarding his impressions of “misallocation” in Lafayette and Monroe. In the illustrative plans, Lafayette is split between CD 3 and CD 5. According to Bryan, if the districts were drawn race-blind, there should be equal amounts of White and Black voters in each district. Instead, he claimed, they were drawn very precisely, such that blocks with 50% or more Black population ended up on the same side of the line. He testified that the same pattern is observable in Baton Rouge, where the district line, in his opinion, was drawn block by block to precisely divide the Black and White populations. Overall, Bryan testified, he found that all the splits he analyzed occurred in such a way that a disproportionate share of the Black population was drawn into a majority-minority district. Therefore, he concluded that race was a prevailing factor in the designs of the illustrative plans. Responding to a criticism of his report by Anthony Fairfax, Bryan admitted that he did not consider socioeconomic factors in drawing his conclusions.
On cross-examination, Bryan agreed that the district court in Caster v Merrill, where he offered testimony, found that Any Part Black was the proper metric to apply in
Bryan confirmed that his conclusion about racial predomination was based on data, a visual examination of the maps, and the index of misallocation. He testified that he has never produced an index of misallocation before, and that he is not aware of it ever being credited by a court. In response to questioning about the assumptions built into the “misallocation” inquiry, Bryan testified that the notion of populations being “misallocated” is based on two assumptions: first, that the Black population is evenly distributed in an area and that a split is created randomly. Plaintiffs’ counsel asked Bryan to examine this map from his report:
Plaintiffs’ counsel further asked Bryan to explain how it can be gleaned whether a “misallocation” is due to Black population being concentrated in a certain area versus racial intent in the drawing of lines. Bryan stated that he infers misallocation, because although the map drawer could theoretically divide areas in any number of ways, based on his observation, the lines are drawn to maximize division by putting the line right between Black and White populations. Ultimately, however, he conceded that he cannot say how much of a given “misallocation” is attributable to race-based line drawing or highly segregated populations. Nor, he reiterated, did he examine socioeconomic data or other traditional redistricting principles in determining that race prevailed.
Defendants’ next witness speaking to Gingles I was Dr. Christopher Blunt. Dr. Blunt, a PhD in political science with a public opinion consulting practice, was tendered and accepted as an expert in political science with an emphasis in quantitative political science and data analysis based on Plaintiffs’ stipulation. Dr. Blunt was asked “to analyze and determine whether a race-blind redistricting process, following traditional districting criteria, would or would not be likely to produce a plan with two majority-minority districts.”153 He did so by simulating a set of 10,000 possible congressional maps using commercially available software called REDIST, and computing the BVAP percentage for each of the six districts in the computer-simulated plans, using the APB metric.
Dr. Blunt concluded that “[n]one of the simulated plans produces even one majority-minority
Dr. Blunt testified that his simulations did not consider or account for race, partisanship, or prior district boundaries. Further, he explained that due to limitations of the simulation software he used, he was only able to preserve communities of interest to the extent that a community of interest was located entirely within the bounds of one parish. He stated that he was hesitant to include communities of interest in his simulation in any event, because the term has no firm legal definition, and because a community of interest could serve as a proxy for race, and his intent was to simulate a race-blind redistricting process.
Dr. Blunt opined that his simulated plans were more compact than the illustrative plans and split fewer parishes. He stated that, even when he set the constraints to allow unlimited splits, he found that the results did not change significantly. Simulations without split constraints yielded a highest BVAP district of 46.06% BVAP, but there was still not a single majority-minority district in any plan.
On cross-examination, Dr. Blunt testified that he has never run a simulation of electoral districts before, or, in fact, conducted any simulation analysis previously in his career. He explained that did not write the code for his simulations, but he did write the instructions to execute the underlying algorithm.
Dr. Blunt agreed that the simulations do not provide a valid comparison if traditional redistricting principles are not part of the constraints. So, he stated, he looked to the
Defendants also called Dr. M.V. Hood III, a political science professor at the University of Georgia, to testify as an expert in the fields of political science, quantitative
Table 1. District Core Retention Comparisons
| District | Enacted | Robinson | Galmon-1 | Galmon-2 | Galmon-3 |
|---|---|---|---|---|---|
| 1 | 97.9% | 68.5% | 71.4% | 80.6% | 63.6% |
| 2 | 98.8% | 81.3% | 85.2% | 80.6% | 85.2% |
| 3 | 100% | 76.0% | 80.9% | 88.6% | 72.3% |
| 4 | 93.8% | 72.3% | 69.3% | 70.8% | 70.3% |
| 5 | 89.1% | 49.1% | 52.3% | 53.5% | 47.0% |
| 6 | 98.5% | 55.4% | 58.6% | 64.7% | 61.1% |
| Mean | 96.4 | 67.1 | 69.6 | 73.1 | 66.6 |
| S.D. | 4.1 | 12.4 | 12.6 | 12.7 | 12.8 |
| Range | 10.9 | 32.2 | 32.9 | 35.1 | 38.2 |
Based on this data, Dr. Hood concluded that for the enacted plan, “most districts appear to be a close approximation of their corresponding configurations from the benchmark plan.”158 Dr. Hood opines that Plaintiffs’ illustrative plans, by contrast, have lower overall core retention.
Dr. Hood also analyzed the same districts using a geographic similarity index, which is used to determine the degree to which districts share a common geography. Like core retention, the similarity index is expressed from 0 to 100%, with a higher score indicating more geographic overlap. Dr. Hood calculated that, comparing the 2022 enacted plan to the benchmark 2011 map, the districts have a mean similarity value of 88.3, which, he opines, is an indication that “the congressional districts in the 2022 enacted plan strongly resemble the previous districts from the benchmark plan.”159 The illustrative plans that he examined had lower mean similarity scores, ranging from 41.0 to 46.4%.160 Thus, he concludes, “the plaintiff-proposed [districts] deviate to a greater degree from the benchmark plan.”161
Dr. Hood also performed a racial composition analysis to determine the percentage of Black population contained within each congressional district for the 2011 map, the 2022 enacted mаp, and Plaintiffs’ maps. He used the DOJ Black definition, which, he explains, “combines all single-race Black identifiers who are also non-Hispanic with everyone who is non-Hispanic and identifies as white and Black.”162 Applying his DOJ Black metric to the illustrative maps,
Table 4. District Percentage Black Comparisons, 2020 Voting Age Population
| District | Benchmark | Enacted | Robinson | Galmon-1 | Galmon-2 | Galmon-3 |
|---|---|---|---|---|---|---|
| 1 | 13.7% | 12.5% | 17.2% | 17.4% | 15.7% | 17.7% |
| 2 | 57.0% | 57.0% | 49.4% | 48.4% | 48.9% | 48.4% |
| 3 | 23.8% | 23.9% | 17.3% | 19.1% | 20.9% | 17.4% |
| 4 | 32.6% | 33.1% | 31.3% | 31.2% | 28.0% | 32.3% |
| 5 | 32.4% | 32.3% | 51.2% | 49.2% | 49.3% | 50.8% |
| 6 | 24.1% | 23.3% | 15.7% | 16.7% | 19.1% | 15.5% |
Dr. Hood agreed that a desire to maximize core retention is not a consideration that trumps compliance with the
Defendants tendered Dr. Alan Murray as an expert in the field of demographic analysis, spatial analytics as it relates to race, and statistics. Dr. Murray was asked to evaluate the spatial distribution of BVAP and WVAP164 in Louisiana. Based on his spatial statistical analysis, he concluded that Black and White voters “are not at all similarly geospatially distributed, with significant clusters of concentrated groupings.”165 Rural areas, he noted, are “dominated by high percentages of white population, but urban areas have clusters of high percent white population as well.”166 Meanwhile, the Black population is clustered particularly in urban areas, “although these urban areas are separated from each other.”167 Additionally, in his report, Dr. Murray calculated the distance between the centers of various Louisiana cities and found, inter alia, that the cities of Monroe and Baton Rouge are 152 miles apart.168
On cross-examination, Dr. Murray stated that he had no basis to disagree with the opinions offered by any of Plaintiffs’ experts. He testified that he has no opinion on whether two majority-minority districts can be drawn consistent with traditional redistricting principles. He further stated that he expresses no opinion on numerosity or compactness. He did not review any of Plaintiffs’ illustrative plans as part of his analysis. Dr. Murray testified that he has never seen a population where the Black population is not heterogeneously distributed. Therefore, he stated, Louisiana’s distribution of Black and White residents is not unusual.
B. Gingles II and III
To satisfy the second and third Gingles requirements, namely that Black voters are “politically cohesive” and “that the white majority votes sufficiently as a bloc to enable it . . . usually to defeat the minority‘s
Defendants stipulated to Plaintiffs’ tender of Dr. Palmer as an expert in redistricting with an emphasis in racially polarized voting and data analysis. An Associate Professor of Political Science at Boston University, Dr. Palmer has previously testified as an expert in eight federal court voting cases; he prepared a report and a rebuttal report in this case.170 Dr. Palmer found “strong evidence of racially polarized voting across Louisiana,” and “in each of the six individual congressional districts.”171 His analysis revealed that “Black-preferred candidates are largely unable to win elections in Louisiana,” and that on a district level, “Black-preferred candidates are only regularly successful in the 2nd Congressional District, which is a majority-Black district.”172 Lastly, Dr. Palmer concluded that Black-preferred candidates would be “generally able to win elections in the Second and Fifth Congressional Districts”173 under Plaintiffs’ illustrative maps.
At the hearing, Dr. Palmer explained that racially polarized voting occurs when voters of different races prefer different candidates. He testified that racially polarized voting is not always present; for example, in the Bethune-Hill case, he found it in some districts but not others. Dr. Palmer explained that his analysis is based on a statistical process known as ecological inference (EI), which estimates group-level preferences based on aggregate data. According to Dr. Palmer, EI is the best available method and has been widely recognized by courts. He emphasized that it is not his project, nor is it within the capabilities of EI, to investigate the reasons behind racially polarized voting. In other words, his analysis sets out to determine how different racial groups vote, not why they vote that way.
Dr. Palmer’s analysis relied on precinct-level election results and voter turnout by race, as compiled by the Louisiana Secretary of State. That data was then paired with precinct-level shape files of the congressional districts. Dr. Palmer examined 22 statewide elections in Louisiana from 2012 to 2020, looking at the final round of voting fоr each race and the runoff rounds, when runoffs occurred. His analysis began by examining the support for each candidate in a given race by each demographic group, in order to determine if members of the group cohesively support a single candidate. Then, Dr. Palmer compared the preferences of White voters to the preferences of Black voters to see if there was evidence that they supported different candidates.
In 18 of the 22 elections he examined, Dr. Palmer found that there was a Black candidate of choice. In 21 of the 22, there was a White candidate of choice. Of the 18 elections with a Black candidate of choice, Dr. Palmer found that White voters had a different candidate of choice and strongly opposed the Black candidate of choice 17 times. Relatedly, Dr. Palmer found that the 18 Black candidates of choice were supported by 91.4% of Black voters and 20.8% of White voters. Among the 21 White candidates of choice, the average
Dr. Palmer also testified about the performance analysis he conducted of Plaintiffs’ illustrative districts, opining that under Cooper’s three illustrative maps, the Black candidate of choice would statistically, more often than not, be able to win. He reached this conclusion by calculating the percentage of the vote won by the Black-preferred
Asked during his testimony about the possibility that polarized voting patterns may be attributable to partisan polarization, not race, Dr. Palmer stated again that his purview is to identify voting patterns that emerge, not to explain the reasons behind them. In other words, his inquiry is statistical, not social. Dr. Palmer offered a rebuttal of Defendants’ expert Dr. Alford’s argument that because President Obama, who is Black, received a smaller share of the Black vote than did Hillary Clinton, who is White, the relevant pattern is partisan in nature, not racial. Dr. Palmer disputed the assumption that the Black-preferred candidate is necessarily Black; according to his findings, in the 18 elections with a Black-preferred candidate, that candidate was Black only 9 times.
On cross-examination, Dr. Palmer clarified that his analysis was based on statewide elections and not congressional elections because there have not been any elections conducted under the current map, and it would be impossible to combine data from different districts to comport with the new boundaries, because there are different candidates on the ballot in each district. Statewide elections are the most germane to analysis because the same candidate is up for election in every precinct.
Defendants also queried Dr. Palmer about the impact of White crossover voting on his analysis. White crossover voting occurs when White voters vote for the Black-preferred candidate. Dr. Palmer agreed that, on account of White crossover voting, it could be possible for CD 2 and CD 5 to be drawn at below 50% BVAP and still elect Black-preferred candidates. Dr. Palmer testified that the existence of White crossover voting does not negate the existence of racially polarized voting, however.
Plaintiffs also presented opinion testimony on Gingles II and III requirements from Dr. Lisa Handley. Defendants stipulated to Dr. Handley’s expertise in the area of redistricting with a focus on racially polarized voting. Relying on her 35 years as a redistricting expert and her previous testimony in dozens of voting rights cases, Dr. Handley opined that “[v]oting in the State of Louisiana is racially polarized.”177 This polarization, she explained, “impedes the ability of Black voters to elect candidates of their choice unless congressional districts are drawn that provide Black voters with an opportunity to elect their preferred candidates to the U.S. House of Representatives.”178 Based on her review of the illustrative maps prepared by Anthony Fairfax, Dr. Handley opined that “it is possible to create an additional congressional district that would provide Black voters with an
Dr. Handley testified that there is a “quite stark” pattern of racially polarized voting in Louisiana. She stated that voting is polarized if White and black voters vote differently. Or, to put it another way, she explained that if Black voters voting alone would elect a different candidate than White voters voting alone, an election is racially polarized. Dr. Handley testified that she uses three statistical techniques to assess racially polarized voting: homogenous precinct analysis (HP), ecological regression (ER), and ecological inference (EI).180 HP and ER were used and accepted by the Supreme Court as far back as Gingles, she stated. EI, which was developed later, has since become a widely accepted technique, as well. Dr. Handley testified that if estimates of racially polarized voting are similar across statistical measures, the conclusions are more probative.
Dr. Handley analyzed recent statewide election contests that included Black candidates. In her report, she notes that courts consider election contests that include minority candidates to be more probative than contests with only White candidates, because this approach recognizes that it is not sufficient for minority voters to be able to elect their preferred candidate only when that candidate is White. Additionally, Dr. Handley explains that she conducted a racial bloc voting analysis for congressional elections with Black candidates from 2016 – 2020, because endogenous elections – elections for the office at issue in the suit – are considered “particularly probative in a vote dilution claim.”181
Dr. Handley’s analysis demonstrated that the 15 statewide contests that included Black candidates were racially polarized, with Black voters “very cohesive in support of their preferred candidates,” and “white voters consistently bloc voted against these candidates.”182 Across the 15 contests she studied, the average percentage of Black voter support for the Black-preferred candidate was 83.8%, or an even stronger 93.5% in contests with only two candidates.183 As for the nine congressional elections she analyzed, Dr. Handley testified that six of them – the ones that did not occur in majority-Black CD 2 – were quite racially polarized. Dr Handley’s analysis is reproduced below.
Dr. Handley stated that polarization was less evident in CD 2; in her report, she finds that the CD 2 contests were “probably not racially polarized.”185 At the hearing, she explained that CD 2 is distinguishable for its relatively high amount of White crossover voting.
Dr. Handley also analyzed whether the Legislature’s enacted map provides opportunities for Black voters to elect the candidate of their choice. She analyzed this by recompiling election results from previous elections into the district boundaries in CD 2, 3, 4, 5, and 6, to see how those elections would have played out.186 In the enacted plan, Dr. Handley found, only CD 2 is an opportunity district:187188
| Enacted Plan District | Effectiveness Score #1: Percent of Contests Black-Preferred Candidate Wins or Advances to Runoff From All 15 Elections | Effectiveness Score #2: Percent of Two-Candidate Contests Black-Preferred Candidate Wins |
|---|---|---|
| 1 | 0.0% | 0.0% |
| 100.0% | 100.0% | |
| 3 | 6.7% | 0.0% |
| 4 | 26.7% | 0.0% |
| 5 | 26.7% | 0.0% |
| 6 | 6.7% | 0.0% |
Dr. Handley opines that, by contrast, Plaintiffs’ illustrative plans feature two districts that perform for Black voters. She explained that a “district-specific, functional analysis of this plan reveals that it offers two districts that are likely to provide Black voters with an opportunity to elect the candidates of their choice to Congress: Districts 2 and 5.”189
Specifically, Dr. Handley concluded that in the CD 5 proposed in Cooper’s Illustrative Plan 1, the Black-preferred candidate is likely to win or advance to a runoff in 80% of all election contests, and likely to win 77.8% of two-candidate contests.190 And, in the CD 5 proposed in Cooper’s Illustrative Plans 2 and 2A, Dr. Handley concluded that the Black-preferred candidate is likely to win or advance to a runoff in 86.7% of contests, and to win 77.8% of all two-candidate contests.191 Her results appear in the following table from her report:
| Illustrative Plan District | Effectiveness Score #1: Percent of Contests Black-Preferred Candidate Wins or Advances to Runoff From All 15 Elections | Effectiveness Score #2: Percent of Two-Candidate Contests Black-Preferred Candidate Wins |
|---|---|---|
| 1 | 13.3% | 0.0% |
| 2 | 100.0% | 100.0% |
| 3 | 0.0% | 0.0% |
| 4 | 26.7% | 0.0% |
| 5 | 80.0% | 77.8% |
| 6 | 0.0% | 0.0% |
Based on her analysis of the data, Dr. Handley concluded that because of the clearly racially polarized voting in Louisiana, Black voters can only elect their candidate of choice if a district is drawn that gives them that opportunity.
On cross-examination, Dr. Handley agreed that it was theoretically possible for an “effective” district – that is, a district where Black voters have the opportunity to elect the candidate of their choice – to have less than 50% BVAP. On redirect, Dr. Handley discussed the BVAP percentages in Cooper’s illustrative plans and asserted that her conclusions on the effectiveness of the proposed districts did not change depending on which definition of Black was used. Regardless, she stated, none of the enacted plan districts perform to allow Black voters an opportunity to elect their preferred candidate except for CD 2.
Defendants offered various expert witnesses on the inquiry into racially polarized voting. Dr. John Alford was tendered and accepted, pursuant to Plaintiffs’ stipulation, as an expert on Gingles II and racially polarized voting. Dr. Alford was retained to provide analysis related to evidence of racially polarized voting in this case; his first step, he explained, was to attempt to replicate the ecological inference (EI) analysis performed by Plaintiffs’ expert witnesses, Dr. Handley and Dr. Palmer. Dr. Alford opined at the hearing that EI is the most reliable and widely used of the available techniques in this area, and that it is the “gold standard” widely relied on by experts.
Dr. Alford reported that he very closely replicated Dr. Handley and Dr. Palmer‘s results, “with only the slight variation that one would expect given the inherent variation associated with [EI] estimations.”193 Because he found their data reliable, Dr. Alford relied for his report “entirely” on Dr. Handley and Dr. Palmer‘s EI conclusions
Defendants also offered the report and testimony of Dr. Jeffrey Lewis, tendered as an expert in the fields of political science, census data analysis, and statistics, specifically racially polarized voting. Plaintiffs stipulated to the tender of Dr. Lewis. In his report, Dr. Lewis describes the scope of his inquiry as calculating “the fraction of voters in the November 3, 2020 Presidential General election who identified as Black in the second and fifth districts of the illustrative Louisiana congressional district plans proposed by the plaintiffs.”194 Next, he attempted to “estimate the support of Black and of white (non-Black) voters for Biden/Harris in the same election among voters residing in each of those illustrative districts.”195 Lastly, Dr. Lewis set out to “calculate the support for Biden/Harris among all voters residing in each illustrative district and the support that Biden/Harris would have received in those same districts in the absence of any white ‘crossover’ voting.”196
Dr. Lewis concludes that the illustrative districts proposed by Plaintiffs could be “effective” — that is, they could still provide Black voters an opportunity to elect the candidate of their choice — if they were drawn to have less than 50% BVAP. The reason, he asserts, is White crossover voting. Dr. Lewis testified that his analysis suggests that Biden/Harris would have received over 50% of the vote in all of the illustrative districts, “even if the BVAP in those districts was reduced to as low as 30 percent in the second district or as low as 48 percent in the fifth district.”197 His findings are set forth in the following chart from his report:
District 2
| Percent support for Biden/Harris: | |||||
|---|---|---|---|---|---|
| Plan | Percent Black voters | All voters | Black voters (EI Estimate) | White voters (EI Estimate) | Without white cross-over votes |
| C130: Robinson/Fairfax | 51.18 | 69.15 | 99.13 | 38.35 | 50.73 |
| C131: Galmon-1/Cooper-1 | 49.94 | | 98.57 | 39.01 | 49.23 |
| C132: Galmon-2/Cooper-2 | 50.33 | 69.11 | 97.69 | 40.39 | 49.16 |
| C133: Galmon-3/Cooper-3 | 49.79 | 68.76 | 98.52 | 38.85 | 49.06 |
District 5
| Percent support for Biden/Harris: | |||||
|---|---|---|---|---|---|
| Plan | Percent Black voters | All voters | Black voters (EI Estimate) | White voters (EI Estimate) | Without white cross-over votes |
| C130: Robinson/Fairfax | 49.66 | 53.34 | 96.91 | 12.10 | 48.13 |
| C131: Galmon-1/Cooper-1 | 47.53 | 54.37 | 96.58 | 13.16 | 45.90 |
| C132: Galmon-2/Cooper-2 | 46.88 | 53.24 | 96.28 | 13.18 | 45.14 |
| C133: Galmon-3/Cooper-3 | 49.00 | 55.94 | 96.75 | 13.93 | 47.41 |
Dr. Lewis conceded that his conclusion was based on a single exogenous election and that a “complete analysis. . .would require consideration of additional elections and more extensive consideration of whether EI estimates of support for each candidate were reliable in this context among other things.”199
To address Gingles III, Defendants offered the opinion testimony of Dr. Tumulesh Solanky, tendered as an expert in the fields of mathematics and statistical analysis. Dr. Solanky was asked to examine voting patterns in the state of Louisiana, focusing in particular on East Baton Rouge Parish. Dr. Solanky opined that East Baton Rouge Parish votes “very differently” compared to the other parishes that are part of Plaintiffs’ illustrative CD 5. Specifically, he stated, East Baton Rouge votes more strongly in favor of the minority-preferred candidate than other parishes. Dr. Solanky relies on the 2020 presidential election as the basis for his conclusion. He found that although there were 13.0% more White voters than Black voters who participated in the election, the minority preferred candidate (Biden) won by 13.0% in East Baton Rouge Parish. Therefore, Solanky opines, “it is apparent that. . .White voters did not vote as a bloc to defeat the black (minority) preferred candidate.”200
On cross-examination, Dr. Solanky conceded that he is not a voting expert and that his only familiarity with the concept of racially polarized voting is derived from reading the other expert reports in this case. He emphasized that he did not conduct a racially polarized voting analysis in his report; instead, he investigated the assumption that Black and White voters vote similarly regardless of which parish they live in. He found that assumption not to be true.
C. The Senate Factors and Proportionality
Plaintiffs offered several expert witnesses who presented testimony relevant to the Court‘s consideration of the Section
- Senate Factor 5: There are large gaps in educational attainment, unemployment, and other socioeconomic indicators between Black and White Louisianans. Research shows that these disparities are the result of contemporary and historical discrimination by government and market institutions and actors. Educational attainment and other socioeconomic indicators are important predictors of voting behavior.
- Senate Factor 5: Several cities in Louisiana are marked by racial residential segregation, which has been shown to affect voting. These patterns of residential segregation are the result of contemporary and historical racial discrimination by government and market actors.
- Senate Factor 5: Health outcomes vary by race in Louisiana; health is also an important predictor of voter turnout. Health disparities in Louisiana are shaped by government and market policies that affect the sites of environmental hazards as well as access to health care.
- Senate Factor 5: Criminal justice involvement also affects voting, and criminal justice outcomes vary by race in Louisiana. Black people are overrepresented in Louisiana‘s correctional populations. Research has shown that racial discrimination played a role in racial disparities in criminal justice in Louisiana in the past and continues to do so today. Patterns of criminal justice outcomes cannot be explained fully by the differential commission of crimes by race.
- Senate Factor 6: Political campaigns in Louisiana have historically been and remain marked by implicit and explicit racial appeals.
- Senate Factor 7: Black people are one third of Louisiana‘s overall population, yet are underrepresented among elected officials at all levels of government, including among executives (such as Governor, Lieutenant Governor, and Mayors), federal and state legislators, and judges.
- Senate Factor 8: Policy outcomes, such as with respect to infrastructure, do not track the specific needs of the minority community in several ways. Moreover, Black Louisianans often express the belief that they are not valued equally by elеcted representatives in both public comments and surveys.
-
Senate Factor 9: Although supporters of SB5 and HB1 offered several justifications for passing SB5 and HB1, including respect for traditional redistricting principles such as minimizing deviations from the ideal district population, compactness, keeping precincts and parishes whole, keeping traditional district boundaries, and maintaining communities of interest, the Legislature ultimately elected not to pass legislation proposing maps with two majority-minority districts that more closely conformed to these traditional redistricting principles than SB5 and HB1. - Senate Factor 9: Sponsors of SB5 and HB1 provided no evidence that they tried to draw an additional majority-minority district, nor did they provide evidence that adding a second majority-minority district would fail to allow Black Louisianans an opportunity to elect a candidate of their choice.
(PR-14, p. 6-7).
Dr. Burch also offered opinions on partisanship and race. She opines that, based on her examination of relevant political science literature, “racial identity and racial attitudes shape partisanship and party cohesion, and have become increasingly linked since 2008.”201 For example, Dr. Burch asserts, Black voters consistently identify strongly with the Democratic Party, a fact which scholars have concluded is not explained by socioeconomic status, policy preferences, or ideology.202 Instead, Dr. Burch explains, this unified Democratic support is caused by the “sense of racial linked fate, or the degree to which a Black person believes that their fate is tied to the fate of the race, and in the social pressure to conform to group ideas of Black uplift.”203 Overall, in Dr. Burch‘s opinion, attributing the polarization of voting in Louisiana to party cohesion instead of race is flawed because it “ignores the rather strong evidence in the literature that race and racial attitudes increasingly drive partisanship and vote choice.”204
Plaintiffs offered the expert testimony of Dr. Allan Lichtman, a professor of American Politics at American University who has testified as an expert in roughly 100 cases, including voting rights cases considered by the Supreme Court, and in this Court in Terrebonne Parish Branch NAACP v. Jindal.205 Dr. Lichtman opined that all nine of the Senate Factors are present in Louisiana contemporarily and operate to impede the ability of Black voters to participate in politics and elect candidates of their choice. The factors do not exist in isolation, he stated; instead, they synergistically work together to produce vote dilution.
As to Senate Factor 1, Dr. Lichtman opined that Louisiana has not only a history of voting discrimination, but an “ongoing history.” He cited issues like at-large elections, the closure of polling places, and felon disenfranchisement as operating to affect voter access. Louisiana‘s poor educational attainment and outcomes also impairs Black Louisianans’ ability to engage, he stated, since research indicates that education is a prime determinant of political participation.
Dr. Lichtman next addressed Senate Factor 2, which asks whether voting is racially polarized. Like Dr. Palmer and Dr. Handley, he opined that Louisiana has “extreme” racially polarized voting. According to Dr. Lichtman, Black voters vote almost
As to Senate Factor 5, which inquires as to “the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process,”206 Dr. Lichtman opines that Louisiana has “major” socioeconomic disparities, extending to almost every area of significance in people‘s lives. He cited income, unemployment, poverty, dependance upon welfare, homeownership rates, vehicle ownership, internet access, and educational attainment as areas in which Black Louisianans are significantly less well-off than White ones. The record evidence summarizes the socioeconomic disparities in the following charts:
Senate Factor 5
| MEASURE | AFRICAN AMERICANS | NON-HISPANIC WHITES |
|---|---|---|
| MEDIAN HOUSEHOLD INCOME | $32,631 | $61,967 |
| MEDIAN FAMILY INCOME | $42,430 | $79,574 |
| PER-CAPITA INCOME | $19,464 | $34,690 |
| ALL PERSONS IN POVERTY | 29.4% | 12.7% |
| CHILDREN IN POVERTY | 43.2% | 15.0% |
| UNEMPLOYED | 8.0% | 4.2% |
| EMPLOYED MANAGEMENT, PROFESSIONAL | 26.3% | 40.4% |
| HOUSEHOLD RECEIVED FOOD STAMPS LAST 12 MONTHS | 27.0% | 8.6% |
| HOMEOWNERS | 48.8% | 76.6% |
| MEDIAN HOME VALUE | $132,400 | $186,700 |
| NO VEHICLE AVAILABLE IN HOUSEHOLD | 16.5% | 4.7% |
| NO BROADBAND INTERNET | 15.7% | 27.8% |
| Source: U. S. Census, American Community Survey, 2019, 1-Year Estimates. | ||
| MEASURE | AFRICAN AMERICANS | NON-HISPANIC WHITES |
|---|---|---|
| INFANT MORTALITY RATE PER 1,000 LIVE BIRTHS | 12.4 | 5.3 |
| LOW BIRTH WEIGHT BABIES | 15.5% | 7.8% |
| LIFE EXPECTANCY | 72.4 | 76.7 |
| ADULT MEN POOR OR FAIR HEALTH | 25% | 18% |
| ADULT WOMEN POOR OR FAIR HEALTH | 28% | 21% |
| COMPOSITION OF MEDICAL SCHOOL GRADS | 8.3% | 72% |
| NON-ELDERLY MEDICAID | 47% | 20% |
| WITH PRIVATE INSURANCE | 43.3% | 67.5% |
| WITH NO INSURANCE | 8.9% | 6.9% |
| MEASURE | AFRICAN AMERICANS | NON-HISPANIC WHITES |
|---|---|---|
| HIGH SCHOOL GRADUATES OR MORE AGE 25+ | 82.1% | 88.9% |
| BACHELOR‘S DEGREE+ AGE 25+ | 17.0% | 28.9% |
| % AT OR ABOVE BASIC 8TH GRADE MATH | 41% | 77% |
| % AT OR ABOVE BASIC 8TH GRADE READING | 52% | 79% |
| HIGH SCHOOL GRADUATION RATE | 73% | 83% |
Dr. Lichtman also testified about Senate Factor 6, related to the use of racial appeals in campaigns, concluding that Louisiana campaigns feature both subtle and overt racial appeals, and stated that such appeals are used by winning campaigns in Louisiana. Dr. Lichtman cited advertisements and campaign materials promoted by David Vitter, Mike Foster, Steve Scalise, Mike Johnson, John Kennedy, and various Republican-affiliated organization that, in his view, constituted racial appeals. As for Senate Factor 7, which calls for an analysis of “the extent to which members of the minority group have been elected to public office in the jurisdiction,”210 Dr. Lichtman points out that no Black person has held statewide office in Louisiana since Reconstruction.
Dr. Lichtman further concluded that with respect to Senate Factor 8, which looks to whether elected officials are responsive to the particularized needs of the minority group, the state
Senate Factor 9 asks “whether the policy underlying the state or political subdivision‘s use of such voting qualification, prerequisite to voting, or standard, practice or procedure is tenuous.” Dr. Lichtman opines in his report that “Louisiana has no significant justification for its failure to create a second majority-Black district in the post-2020 redistricting process.”211 At the hearing, he disputed Defendants’ assertion that the concept of “core retention” is a valid reason not to create another majority-minority district. First, core retention, while it may be a preference, is not a legal requirement like one person, one vote. Second, Dr. Lichtman testified that prioritizing core retention risks freezing in the inequities of the previous map. If core retention was the key factor in redistricting as Defendants assert, Dr. Lichtman stated, there would never be a remedy for
Nor, Dr. Lichtman testified, does the fact that the 2011 districting map was precleared by the Justice Department provide justification for enacting a carbon copy during this round of redistricting. After all, he explained, preclearance does not mean that a map is not violative of the
Dr. Lichtman opined that essentially all of the Senate Factors support a finding of vote dilution with respect to the Louisiana congressional maps. On cross-examination, he acknowledged that the Black candidate of choice prevailed in the last gubernatorial races in Louisiana, but cautioned that “one swallow does not make a spring.” Asked whether the mayor of Baton Rouge is Black, Dr. Lichtman stated that she is, adding that the fact that the majority-Black city of Baton Rouge has a Black mayor only proves the point that Black-preferred candidates can win in Black jurisdictions, but they are being shut out in White jurisdictions and White districts.
Plaintiffs further offered the reports212 and testimony of R. Blakeslee Gilpin, a
Dr. Gilpin cites property requirements, poll taxes, literacy tests, and the grandfather clause as historical examples of denying Black Louisianans the ability to vote. Dr. Gilpin notes that there is no record of “a single Black Louisianan elected to office until the 1940s,” and from 1910 until 1949, “less than 1% of Louisiana‘s voting age African-American population was able to register to vote.”213 The passage of the VRA in 1965 was not a magic bullet, he asserts. In fact, he explains, the
Dr. Gilpin cites very recent examples that, in his view, demonstrate ongoing discrimination against Black voters in Louisiana. In April 2021, the City of West Monroe entered into a consent decree after the Justice Department asserted that the at-large system used for the Board of Aldermen was proven to disenfranchise Black voters; despite Black residents comprising 30% of the electorate, no Black candidate had ever been elected to the Board. West Monroe agreed to end the practice. Gilpin further cites Louisiana‘s resistance to expand absentee voting during the COVID-19 pandemic, which this Court in Harding v. Edwards214 found placed undue burdens on Black voters.
In his supplemental report, Dr. Gilpin responds to the dispute in this case about the appropriate metric for counting Black voters, be it Any Part Black, “DOJ Black,” or some other measure. Ironically, he explains, the state has long attempted to “designate anyone who could possibly be counted as Black to prevent them from voting.”215 Although Defendants’ resistance to the use of Any Part Black cuts in the opposite direction, toward restricting who can be counted as Black, in Dr. Gilpin‘s opinion, the attempt “is disturbingly reminiscent of this long history of imposing racial categories to disenfranchise its Black citizens.”216
Overall, Dr. Gilpin concludes, the “state of Louisiana‘s long history of racial discrimination is without dispute.”217 The powers that be in Louisiana, he opines, subscribe to the notion that there is an appropriate level of “white political control,”218
Several of Plaintiffs’ lay witnesses offered testimony relevant to various Senate Factors, as well. Mike McClanahan, the state president of the Louisiana NAACP, testified that in his view, anyone with “one drop of Black blood” is Black, no matter what they look like on the outside. According to McClanahan, the Louisiana NAACP engages in voter registration, voter engagement, and voter training efforts involving Black Louisianans. His organization was acutely aware of the importance of the current redistricting cycle, he testified, and undertook efforts related to the Census because they knew the data collected would feed into the redistricting process. When the legislature was holding “roadshow” meetings to solicit public input on new maps, McClanahan participated in a weekly call to coordinate with members across the state to ensure attendance and participation at the roadshows. He himself testified at a roadshow, as well. Based on the maps that the Legislature enacted, McClanahan said, the legislators at the roadshows must have been asleep, or listening “with deaf ears.” In his view, the enacted map was not responsive to the pleas of Black Louisianans — it did not reflect the data, the testimony of the public, or the issues raised in legislative hearings.
Once the map passed the Legislature, McClanahan testified that the Louisiana NAACP‘s strategy was to persuade the Governor to veto it. He explained that his membership did all in their power to get to the Governor, including calling him, holding rallies, engaging on social media, and having legislators contact him on their behalf. When the Governor did, in fact, veto the map, McClanahan felt optimistic but skeptical because of the possibility of a veto override. When the Legislature convened to vote on the veto override, McClanahan and some of his members went to the Capitol, attending session in both houses and, he said, walking the building to ensure their voices were heard. When the vote came in to override Governor Edwards’ veto, McClanahan testified that he saw legislators high-fiving one another, cheering, and jumping in the air. To him, this was a slap in the face of everyone who had participated in the process.
Asked about the effects of racism in Louisiana, McClanahan testified that he lives it, every day, all of his life. He testified that Blaсk quality of life is reduced in so-called “Cancer Alley,” a strip of parishes from Baton Rouge to New Orleans where, he said, chemical plants set up shop in Black neighborhoods and harm residents with their pollution. He cited officer-involved shootings and the fact that several police departments in Louisiana have operated under consent decrees as evidence that, for Black Louisianans, law enforcement does not serve and protect them equally. McClanahan testified that access to quality health care is limited for many Black Louisianans, and noted that during the COVID-19 pandemic Black people had a significantly higher death rate than the rest of the population. All of this, he believes, reflects a lack of elected official responsiveness to issues which disproportionately affect Black Louisianans.
On cross-examination, McClanahan agreed that he has been involved on various state committees related to police
Plaintiff Dr. Dorothy Nairne offered fact testimony related to a number of the Senate Factors and to the real-world consequences of splitting parishes. Dr. Nairne is Black and resides in Napoleonville, in Assumption Parish. Dr. Nairne testified that she is a registered voter and a regular voter in CD 6, where she is represented by Rep. Garrett Graves. She stated that although she has contacted his office on several occasions, she does not see Rep. Graves at events in her community and he does not campaign in her community. Dr. Nairne explained that she lives “right on the cusp” of the split between CD 6 and CD 2. Her neighbors across the street are part of CD 2, while she is in CD 6. She testified that this disconnection makes it very difficult to organize and speak with one voice about issues affecting Assumption Parish. Dr. Nairne does community work in the river parishes area related to environmental justice and racial justice. She testified that the way the river parishes are split under the congressional map means that although they work together, but they don‘t vote together. Of this situation, Dr. Nairne said, “I do not believe my interests are represented. I am alienated.”
Plaintiffs’ counsel showed Dr. Nairne one of their illustrative maps — Fairfax‘s Illustrative Plan 1 — and zoomed in on her part of the state:
In this map, Dr. Nairne testified, she would be in the same district, CD 2, with the people she organizes with, the river parishes into Orleans and Jefferson. “This map makes sense to me,” she stated, adding that if this map was implemented, she knew exactly which households she would go visit to engage them in the political process.
Ashley Shelton is the President and CEO of Power Coalition, a Louisiana civil engagement organization and one of the Plaintiffs in this case. Shelton testified that she is a lifelong Baton Rouge resident and that her work with communities of color
Shelton described that while citizens were speaking to legislators at the redistricting roadshows, the legislators were doodling, not looking up, and not paying attention while people told their story of generations of their families working to vote. Shelton testified that roadshow testimony consistently offered two messages to legislators: first, that the voters wanted a fair and equitable process, and second, that there should be a second majority-minority district to honor the increase in the Black population.
Shelton organized a rally of 250 people of color and allies at the state capitol to, in her words, say “hey, we‘re watching you.” On the day of the veto override, Shelton testified, the vote was along racial lines. Conservative politicians cheered and celebrated, which Shelton said was deflating and felt like “a true sign of disenfranchisеment.” Now, Shelton explains, her organization is engaging with voters who feel disengaged because their efforts around redistricting were unsuccessful.
Shelton testified that Black voters in Louisiana face discrimination when it comes to voting. She stated that Black voters experience polling place changes or closures more frequently; a recent consolidation of a black polling location in New Orleans East, for example, made it a lot harder for chronic voters in that area to access the polls. Though the move was only a few miles, the new site required crossing Interstate 10. Often, Shelton explained, Black voters lack access to transportation. Many Black residents face housing insecurity and may not always be able to afford a cell phone or broadband internet.
In Shelton‘s view, no one makes an effort to talk to Black voters. In her experience with Power Coalition, when she creates a “universe” of voters to target for outreach, she can get 60-65% of them to turn out to vote. This proves to her that it is possible to engage Black voters but that no one is addressing Black concerns or including them in the process. Shelton testified that Black people do not vote for Democrats simply because they are Democrats — they vote for the candidate who they believe will vote with their interests. In her experience, she testified, White and conservative candidates have not centered the issues that she cares about and therefore, would not be her candidate of choice. Overall, she said, she feels that neither party has been particularly responsive to the Black community.
As exhibits to their memoranda in opposition to the Motion for Preliminary Injunction, Defendants offered the reports of Dr. Jeff Sadow219 and Mike Hefner.220 Dr. Sadow was called to testify at the hearing via videoconference, but due to difficulties with his internet connection, was not able to testify. The Court permitted Defendants to call him at a later time, but they did not. Nor did Defendants call Hefner as a witness at the hearing. Their reports were not offered as substantive evidence at the hearing. Additionally, the reports are hearsay, and there was no opportunity for cross-examination. Accordingly, the Court did not consider these reports.
To address the Purcell issue in this case, namely, the parties’ dispute over whether the November election cycle is too close to allow time for a remedy to be implemented, Plaintiffs called Matthew Block, executive counsel to Louisiana Governor John Bel Edwards. Block testified that he has experience working with the Secretary of State to develop special election plans that become necessary due to emergencies such as hurricanes and other natural disasters. He testified that during Governor Edwards’ term, there have been nine instances where election dates, qualifying dates, polling locations, or other aspects of election administration had to be altered, most recently last year after Hurricane Ida. Block noted that after Ida, election dates were pushed back a month, from October/November to November/December, and that in 2020, the April/May elections were moved twice as a result of the COVID-19 pandemic.
Block testified that the state was able to successfully administer these elections, despite the need for last-minute change. He stated that he was unaware of any electoral chaos that ensued, and that he has heard nothing to dispute that the Secretary of State was able to successfully administer these elections. Overall, Block asserted, the Governor, the Secretary of State, and local officials have a lot of experience with adjusting elections. Turning to the facts of this case, Block testified that, based on his experience working with the Legislature, it would be possible for the body to draw a new map, especially because there were bills previously filed that offer alternative maps for consideration, so the process would not be starting from scratch. The Governor has the power to call an extraordinary session of the Legislature, he stated, and the Legislature can also initiate one itself.
On cross-examination, Defendants engaged Block on the topic of Governor Edwards’ efforts on behalf of the Black community in Louisiana. Block agreed that the Governor has Black support and tries to be responsive to the needs of the Black cоmmunity. Block confirmed that Governor Edwards expanded Medicaid, is a proponent of criminal justice reform, helped pass a bill restoring voting rights for many felons, and supported a constitutional amendment requiring unanimous jury verdicts. Defendants listed several other examples that, in their view, represented Governor Edwards’ responsiveness to the Black community — hiring Black officials in his administration, making Juneteenth a state holiday, convening a task force to track inequities in health care, and offering free COVID vaccines and testing — and Block confirmed that, indeed, Governor Edwards did all of the above. Block explained that the Governor vetoed the Legislature‘s enacted map because he believed that a second majority-minority district was necessary to comply with Section 2 of the VRA, and because he believed that a fair map would have a second majority-minority district.
Defendants called Sherri Hadskey, the Commissioner of Elections for the Louisiana Secretary of State. Hadskey oversees election operations and election administration, including implementation of new districting plans. Hadskey testified that
her office has already undertaken significant administrative work related to the Legislature’s enacted map by reassigning and notifying voters who find themselves in a new congressional district under that plan. According to Hadskey, this effort involved voters in fifteen parishes, and 250,000 voting cards have been sent to voters who changed districts under the new map.
Hadskey further testified that Louisiana’s election administration resources have been strained by COVID-19 and the delay in receiving census data. If forced to implement one of Plaintiffs’ illustrative plans, she testified that her office would have to undo the coding for the fifteen parishes that saw changes under the enacted plan; code the new changes under an illustrative plan; and timely notify voters and potential candidates of these changes. Hadskey expressed concern that the process would be rushed, potentially causing errors that would give rise to confusion. The process of updating records and notifying voters impacted by districting changes under the enacted map took about three weeks, she testified.
Citing a recent issue in Calcasieu Parish, where voter information was entered incorrectly, leading to the issuance of incorrect ballots, Hadskey testified that she fears these issues could occur on a larger scale if a new map is handed down in June or July. Moreover, Hadskey testified that a national paper shortage could interfere with reprinting ballot envelopes, voter notification cards, and other items required under the law. Hadskey testified that she is “extremely concerned” about the prospect of administering the congressional election under a new map, noting that in her thirty-year career at the Secretary of State’s office, she has never moved a federal election.
FINDINGS OF FACT AND CONCLUSIONS OF LAW
I. PRELIMINARY MATTERS
A. Standing
On the issue of standing, Secretary Ardoin is laboring under misapprehensions of both the facts and the law. As an initial matter, the jointly stipulated facts in this case establish that Plaintiff Edwin René Soulé resides in Congressional District 1,221 and that the Louisiana NAACP, Plaintiff herein, has members “who live in every parish and in each of the six congressional districts in the enacted congressional plan.”222 Setting aside those factual corrections, the Court finds that, in the context of a vote dilution claim under
arguably more expansive view of standing in the vote dilution context, finding that the plaintiffs had standing where “[i]t is conceded that each voter resides in a district where their vote has been cracked or packed. That is enough.”225
In the instant case, Plaintiffs allege that they have suffered the injury of vote dilution because Black voters in Louisiana are packed into one majority-Black district that hoards Black population to prevent another majority-minority district from being drawn, and because Black voters outside of that majority-Black district have been fractured across the congressional map to prevent them from concentrating their voting strength in a district where they would have an opportunity to elect the candidate of their choice. The Fifth Circuit in Harding explained that, “[i]n vote dilution cases, the ‘harm arises from the particular composition of the voter‘s own district, which causes his vote—having been packed or cracked—to carry less weight than it would carry in another, hypothetical district.‘”226 Plaintiffs herein have pled such harm and, as the Fifth Circuit counsels, “[t]hat is enough.”227 Accordingly, the Court rejects Defendants’ argument that Plaintiffs lack standing.
B. Challenges to Gingles
Intervenor Defendant Attorney General Landry invites the Court to toss Gingles onto the trash heap, repeatedly arguing that the well-worn Gingles test is endangered and, possibly, bound for extinction. The Attorney General candidly acknowledges that Thornburg v. Gingles, 478 U.S. 30 (1986) and its progeny are controlling,228 but warns that the Supreme Court “has signaled. . .that it will be reviewing vote dilution claims under
C. Private Right of Action Under Section 2 of the VRA
Defendants advance another argument premised on dicta: that
Inviting this Court to disregard Morse and scores of
While this issue has been flagged,239 it is undisputed that the Supreme Court and federal district courts have repeatedly heard cases brought by private plaintiffs under
D. How to Count Black Voters
Because the numerosity of Black voters is central to the Gingles I inquiry, deciding who counts as Black is a threshold issue. Three definitions have been advanced by the parties’ experts: Any Part Black, “DOJ Black,” and Single-Race Black. For several reasons, the Court concludes that the Any Part Black metric is appropriate when considering the Gingles I precondition of numerosity. This conclusion is supported foremost by the United States Supreme Court’s discussion of the issue in Georgia v. Ashcroft, 539 U.S. 461 (2003), a 2003 Voting Rights Act case. There, the Court wrote:
Georgia and the United States have submitted slightly different figures regarding the black voting age population of each district. The differing figures depend upon whether the total number of blacks includes those people who self-identify as both black and a member of another minority group, such as Hispanic. Georgia counts this group of people, while the United States does not do so. . . Moreover, the United States does not count all persons who identify themselves as black. It counts those who say they are black and those who say that they are both black and white, but it does not count those who say they are both black and a member of another minority group. Using the United States’ numbers may have more relevance if the case involves a comparison of different minority groups. Here, however, the case involves an examination of only one minority group‘s effective exercise of the electoral franchise. In such circumstances, we believe it is proper to look at all individuals who identify themselves as black.241
There is no question that the instant case is a case involving “an examination of only one minority group’s effective exercise of the electoral franchise.”242 Thus, this Court will follow the Supreme Court and “look at all individuals who identify themselves as black.”243 This conclusion is further supported by the dissenting comments of Chief Justice Roberts in the Supreme Court’s grant of an emergency application for stay in Merrill v. Milligan. Therein, Justice Roberts stated that the District Court for the Northern District of Alabama, which applied the Any Part Black metric in its analysis, had “properly applied existing law in an extensive opinion with no apparent errors for our correction.”244 If, in the eyes of the
Chief Justice, a court using Any Part Black “cannot be faulted for its application of Gingles,”245 this Court would be remiss to apply another standard.
The Any Part Black definition is deeply rooted in Louisiana history; testimony established that the state employed a rigid system of categorizing its citizens as Black if they had any “traceable amount”246 of Black blood. It would be paradoxical, to say the least, to turn a blind eye to Louisiana’s long and well-documented expansive view of “Blackness” in favor of a definition on the opposite end of the spectrum. The Court declines to define Black Voting Age Population (BVAP) in a way that gatekeeps Blackness in the context of this Voting Rights case. Finally, the weight of the evidence presented shows that two majority-minority congressional districts that satisfy Gingles and respect traditional redistricting principles can be drawn in Louisiana even if more restrictive definitions of Black are applied.247
II. LIKELIHOOD OF SUCCESS ON THE MERITS
A. Gingles I
1. Numerosity
The Court finds that Plaintiffs have established that the Black Voting Age Population (BVAP) is “sufficiently
Plaintiffs have put forth several illustrative maps which show that two congressional districts with a BVAP of greater than 50% are easily achieved. Defendants’ expert witness Dr. Bryan concluded the same.252 Moreover, Plaintiffs have established that they can meet the 50% plus threshold required by Gingles I even if a more restrictive metric is used. Cooper calculated the BVAP for his illustrative majority-minority districts using the Non-Hispanic Single-Race Black Citizen Voting Age Population definition and found that even using this most restrictive definition of Black, the Gingles I numerosity requirement was achieved. The statistical results of the impact of this narrow definition of “Black” are reproduced from the record evidence below:
Figure 5
2016-2020 Citizen Voting Age Population by Plan
| % NH SR Black CVAP | % NH White CVAP | NH SR Black CVAP to NH White CVAP Margin | July 2021 Black Registered Voters | |
|---|---|---|---|---|
| 2022 Plan | ||||
| District 2 | 61.31% | 31.45% | 29.86% | 61.46% |
| Illustrative Plan 1 | ||||
| District 2 | 52.82% | 39.31% | 13.51% | 52.33% |
| District 5 | 50.37% | 46.19% | 4.18% | 51.84% |
| Illustrative Plan 2 | ||||
| District 2 | 53.07% | 39.53% | 13.54% | 52.72% |
| District 5 | 50.71% | 45.92% | 4.79% | 51.53% |
| Illustrative Plan 3 | ||||
| District 2 | 52.82% | 39.31% | 13.51% | 52.33% |
| District 5 | 51.72% | 44.86% | 6.86% | 53.35% |
| Illustrative Plan 4 | ||||
| District 2 | 52.63% | 39.53% | 13.10% | 52.23% |
| District 5 | 50.78% | 45.75% | 5.03% | 52.17% |
Table 2 — Illustrative Plan 2’s Black Voting Age Population
| District | DOJ BVAP | DOJ BVAP% | AP BVAP | AP BVAP% |
|---|---|---|---|---|
| 1 | 97,079 | 16.07% | 103,416 | 17.12% |
| 2 | 299,351 | 50.02% | 308,535 | 51.55% |
| 3 | 103,263 | 17.60% | 106,965 | 18.23% |
| 4 | 186,380 | 31.25% | 190,267 | 31.90% |
| 5 | 300,776 | 50.96% | 305,661 | 51.79% |
| 6 | 97,834 | 16.46% | 100,925 | 16.98% |
Note: DOJ BVAP includes Not-Hispanic Black Alone plus Not-Hispanic Black and White combined race; APBVAP includes “Any Part” Black (which contains Hispanic Black VAP)
(PR-86, p. 7)
Dr. Hood also concluded that two of Plaintiffs’ plans demonstrate that two majority-Black districts can be achieved using the ‘DOJ Black’ definition:
Table 3. District Percentage Black Comparisons, 2020 Total Population
| District | Benchmark | Enacted | Robinson-2A | Galmon-4 | LSU/Tulane |
|---|---|---|---|---|---|
| 1 | 15.0% | 13.7% | 17.5% | 18.9% | 15.8% |
| 2 | 59.1% | 59.1% | 51.9% | 50.3% | 41.7% |
| 3 | 25.5% | 25.7% | 19.0% | 20.6% | 23.5% |
| 4 | 34.4% | 34.9% | 32.9% | 32.4% | 32.9% |
| 5 | 34.4% | 34.4% | 53.7% | 52.1% | 34.3% |
| 6 | 25.6% | 24.7% | 17.5% | 18.2% | 44.4% |
Although Defendants argue that Plaintiffs can only succeed at Gingles I using the Any Part Black definition, they fail to refute the record evidence to the contrary. Accordingly, the Court concludes that Plaintiffs are substantially likely to prove Gingles I numerosity should this matter proceed to the merits.
2. Compactness
For the reasons which follow, the Court finds that Plaintiffs have demonstrated that they are substantially likely to prove that Black voters are sufficiently “geographically compact”255 to constitute a majority in a second congressional district. The Court heard opinion testimony on this topic from William Cooper and Anthony Fairfax, who were accepted as expert witnesses by the Court upon Defendants’ stipulation to their expertise in the fields of redistricting, demographics, and census data.
Cooper and Fairfax offered several illustrative plans which included two majority BVAP congressional districts, CD 2 and CD 5. Both Cooper and Fairfax testified that the illustrative plans they drew performed better than the enacted plan on well recognized and widely-used statistical measures of compactness. Specifically,
The record evidence and testimony established following mean compactness scores for the enacted plan as compared to the illustrative plans:
| Plan | Reock | Polsby-Popper | Convex Hull |
|---|---|---|---|
| Enacted Plan | 0.37 | 0.14 | 0.62 |
| Fairfax Illustrative Plan 1 | 0.42 | 0.18 | 0.69 |
| Fairfax Illustrative Plan 2 | 0.39 | 0.20 | 0.71 |
| Fairfax Illustrative Plan 2A | 0.39 | 0.20 | 0.71 |
| Cooper Illustrative Plan 1 | 0.36 | 0.19 | X256 |
| Cooper Illustrative Plan 2 | 0.41 | 0.19 | X |
| Cooper Illustrative Plan 3 | 0.38 | 0.18 | X |
| Cooper Illustrative Plan 4 | 0.37 | 0.18 | X |
Cooper and Fairfax demonstrated, without dispute, that in terms of the objective measures of compactness, the congressional districts in the illustrative plans are demonstrably superior to the enacted plan.
Like the question of numerosity, Defеndants did not meaningfully refute or challenge Plaintiffs’ evidence on compactness. Rather, Defendants challenged the Cooper and Fairfax illustrative maps as improperly, and Defendants submit unlawfully, motivated by considerations of race. Defendants offered opinion testimony from Drs. Bryan, Blunt, Hood and Murray to show that race was the predominant factor in configuring a second majority-BVAP congressional district in the illustrative plans.
On stipulation of the parties, the Court heard opinion testimony from Thomas Bryan, offered by the Defendants as an expert in the field of demographics. Bryan quite candidly acknowledged that he testified as an expert in a redistricting case for the first time earlier this year in Caster v. Merrill, and that the Alabama District
Bryan opined that race was a prevailing factor in the design of Plaintiffs’ illustrative plans based on his “index of misallocation,” which purports to flag areas where a disproportionate share of the Black population was grouped into a majority-minority district. Bryan testified that he does not know if this “misallocation” analysis has ever been credited by a court in a voting rights case – he did not offer it in the Alabama case – and that he was unaware of any case in which the “index of misallocation” was accepted as probative or persuasive by a court in the voting rights context.
Even if this “misallocation” method is accepted, the factual assumptions upon which his conclusions rest are absent in this case. Hence, Bryan’s conclusions are unsupported by the facts and data in this case and thus wholly unreliable. Bryan testified that his analysis is based on two assumptions – that the Black population is evenly distributed and that district splits are created randomly – both of which, he admitted, are not supported by the evidence in this case. Bryan testified that it is still possible to perform the misallocation analysis when those assumptions are not borne out, but he did not explain why, if the underlying assumptions are false, his resulting opinion is reliable. Ultimately, Bryan conceded that that he could not say how much of the “misallocation” he observed was attributable to a racially-motivated mapdrawing process, as opposed to being reflective of the reality that the Black population in Louisiana is highly segregated. This admission seriously undermines the reliability of his opinion that Plaintiffs’ maps are the product of racial predominance. Furthermore, the Court accords Bryan’s racial predominance opinion little weight because he testified that he did not account for compactness, communities of interest, or incumbent protection in concluding that race predominated in Plaintiffs’ maps.
Finally, the Court finds that Bryan’s analysis lacked rigor and thoroughness, which further undermines the reliability of his opinions. On cross-examination, Bryan was asked about Cooper’s findings that his illustrative districts had greater than 50% BVAP even using the single-race Black definition and several other methods for measuring BVAP. He testified that he looked at it but had no opinion to offer about it. For the foregoing reasons, the Court gives very little weight to Bryan’s analysis and conclusions.
Defendants offered opinion testimony from Dr. Christopher Blunt, stipulated by the parties as an expert in the field of political science with an emphasis in quantitative political science and data analysis. Dr. Blunt opined that a computer simulation he used to generate congressional districts did not generate even a single majority-BVAP congressional district. Defendants argue that Dr. Blunt’s simulations prove that the majority-BVAP districts produced in the illustrative plans are the product of racial predominance in the
The Court considers Dr. Blunt to be well-qualified by education and experience in the tendered field of expertise. However, Dr. Blunt has no experience, skill, training or specialized knowledge in the simulation analysis methodology that he employed to reach his conclusions. He testified that had never attempted a simulations analysis before this case and has never published on the topic, taught, or even taken a course on it. Dr. Blunt’s simulation analysis experience is best described as novice.
Dr. Blunt testified that he downloaded publicly available code and wrote the instructions to execute the underlying algorithm. Several times, in response to questions about his analysis, Dr. Blunt admitted that he was limited in his ability to go “under the hood” of the code he was using to program in parameters that would account for certain redistricting criteria. Dr. Blunt conceded the importance of including all the relevant redistricting criteria variables into his simulations. However, he testified that he was only able to account for population equality, contiguity, compactness, and minimization of parish splits. Admittedly, his simulations were performed without regard to minimizing precinct splits, respecting communities of interest, incumbency protection, or even the criterion considered paramount by Defendants, core retention. In short, the simulations he ran did not incorporate the traditional principles of redistricting required by law. Accordingly, his opinions merit little weight.
The Court heard opinion testimony from Dr. M.V. Hood, offered by the Defendants and stipulated to be an expert in the fields of political science, quantitative political analysis, and election administration. Dr. Hood offered opinions on the performance of the illustrative plans by reference to the criteria of core retention, and opinions on compactness of BVAP statewide. The Court finds that he was generally credible but that his conclusions are not particularly helpful to the Court. The Court detects no error in Dr. Hood’s core retention analysis and gives it some weight, though the conclusion that the illustrative plans have lower core retention than the enacted map, which was drawn using a target of “least-change,” is hardly a blockbuster. Further, the Court notes that the importance to be assigned to core retention as a traditional redistricting principle is hotly disputed in this case (see infra), and Dr. Hood willingly admitted and agreed that a desire to preserve core retention does not trump the
The Supreme Court directs that Gingles I compactness “refers to the compactness
The Court accepted Defendants’ witness Dr. Alan Murray as an expert in the fields of demographic analysis, spatial analytics as it relates to race, and statistics. The Court finds Dr. Murray’s opinions unhelpful and unilluminating for several reasons. Dr. Murray employed “spatial analysis” to reach the conclusion that the Black and White populations in Louisiana are heterogeneously distributed. This is nothing more than a commonsense observation which is not a whit probative of the compactness of the districts in the Plaintiffs’ illustrative plans. In fact, Dr. Murray never looked at the illustrative plans. The time-tested, generally accepted statistical measures of compactness used by other experts in this case are qualitatively superior evidence and far more probativе of compactness.
Dr. Murray has no background or experience in redistricting; he did not review any of Plaintiffs’ illustrative plans, and, most notably, he testified that he has no basis to disagree with any of the opinions offered by Plaintiffs’ experts in this case. Lastly, Dr. Murray testified that he is not aware of any court considering the type of “spatial analysis” that he performed in the context of a
In weighing the opinions of the competing expert witnesses the Court finds the Plaintiffs’ Gingles I experts Cooper and Fairfax qualitatively superior and more persuasive on the requirements of numerosity and compactness.
Cooper has extensive experience drawing maps for redistricting and has been repeatedly recognized and accepted as an expert in federal voting rights cases. Cooper has familiarity with the unique voting laws and processes in Louisiana, having worked on redistricting projects in in Shreveport and in Terrebonne, Point Coupee, Madison, and East Carroll Parishes. The Court finds that Cooper’s reports261 in this case were clear, substantiated by unrefuted
Anthony Fairfax’s thirty years of experience in preparing redistricting plans make him well-qualified, in the Court’s view, and his report and supplemental reports are extremely thorough and methodologically sound. Like Cooper, Fairfax remained steady under cross-examination and candidly described his process in detail. The Court did not observe inconsistencies in his testimony, nor any reason to question the veracity of Fairfax’s testimony. The Court credits in particular Fairfax’s testimony where he discussed how race contributed to the illustrative plans that he drew. Fairfax did not deny that he used his mapping software to assess the location of BVAP in Louisiana initially, but he was adamant and credible in his testimony that race did not predominate in his mapping process. Rather, he testified that he only considered race to the extent necessary to test for numerosity and compactness as required by Gingles I.
The weight afforded to Plaintiffs’ experts, Cooper and Fairfax, is appropriate considering not a single defense expert disputed that Plaintiffs’ illustrative plans are generally more compact than the enacted plan based on statistical measures.
The Court’s assessment of reasonable compactness is also informed by a visual inspection of the shapes of the districts in Plaintiffs’ illustrative plans. Overall, the Court observes that the districts proposed in the illustrative maps are regularly shaped, without “tentacles, appendages, bizarre shapes, or any other obvious irregularities,”263 save a few narrow finger-shaped boundaries. Compared to the shape of CD 2 and the wraparound shape of CD 6 in the enacted plan, the illustrative plans are visually more compact.
Next, the Court turns to the question of whether Plaintiffs’ illustrative plans demonstrate reasonable compactness when viewed through the lens of “traditional districting principles such as maintaining communities of interest and traditional boundaries.”264 As an initial matter, the Court will not extensively analyze the traditional criteria of equal population and contiguity, because the evidence makes
The first factor to consider is whether Plaintiffs’ illustrative plans respect existing political subdivisions, such as parishes, cities, and towns. The evidence presented by the parties largely related to parishes and to VTDs, also referred to as precincts. As for parish splits, the Court finds that Plaintiffs’ illustrative plans split fewer parishes than the enacted plan. The enacted plan splits 15. Fairfax’s Illustrative Plan 1 splits 14, while his Plans 2 and 2A split 12 parishes. Cooper’s Illustrative Plans 1 through 4 split 10, 11, 10, and 10 parishes, respectively. Accordingly, the Court finds that Plaintiffs’ illustrative maps respect political subdivision boundaries as much or more so than the enacted plan with regard to parish splits.
As for precinct splits, the
The Court next considers whether Plaintiffs’ illustrative plans respect “communities of interest.” The term “communities of interest” has no universally agreed-upon definition. The
All redistricting plans shall respect the established boundaries of parishes, municipalities, and other political subdivisions and natural geography of the state to the extent practicable. However, this criterion is subordinate to and shall not be used to undermine the maintenance of communities of interest within the same district to the extent practicable.266
By its
Defendants did not call any witnesses to testify about communities of interest. This strikes the Court as a glaring omission, given that
In their post-hearing briefs, Defendants criticize the fact that “Mr. Cooper drew Vernon Parish, home of Fort Polk, and Shreveport, home of Barksdale Air Force Base, into different districts in his illustrative plans even though they were joined in the enacted plan.” 270 Defendants offer no assessment of how Plaintiffs’ maps treat their other two stated communities of interest, preserving the Acadiana region and joining cities with their suburbs. The Court does not find that splitting one argued community of interest is fatal to a finding that Cooper‘s districts are geographically compact without sacrificing communities of interest. Cooper analyzed the enacted plan and identified splits of 18 CBSAs and 30 populated municipalities. Defendants offered no evidence of why the splits in their plan are less offensive to traditional redistricting principles than the ones in Cooper‘s.
Regardless, the inquiry under Gingles I is not whether Plaintiff‘s illustrative maps represent the most perfect or preferable way to draw a majority-Black district; there is no need to show that the illustrative maps would “defeat [a] rival compact district[ ]” in a “beauty contest[ ].” 271 The relevant question is whether, taking into account traditional redistricting principles including communities of interest, a reasonably compact and regular majority-Black district can be drawn.
Courts struggle with analyzing and giving meaning to the subjective redistricting criteria that counsels respect for “communities of interest.” This Court offers no recipe for the definition of “community of interest,” but based on the testimony and evidence, the Court finds that Plaintiffs’ experts demonstrated that they gave careful thought to selecting objectively verifiable indicators to identify for assessing communities of interest and calculating how often their maps split them. By those metrics, Plaintiffs’ maps split locally relevant areas less often than the enacted map. To the extent that “communities of interest” is a term susceptible to clear definition, the Court finds that Plaintiffs made a strong showing that their maps respect them and even unite communities of interest that are not drawn together in the enacted map (St. Landry Parish and
Next, the Court turns to the final two traditional redistricting criteria: incumbency protection and core retention. Avoiding incumbent pairing was not one of the criteria that the Legislature included in its
Lastly, the Court considers core retention. Defendants’ expert Dr. Hood testified that the core retention scores for the illustrative plans are lower than those for the enacted plan, reflecting that the enacted plan retains more of the benchmark district cores than the illustrative plan. The Court does not question this conclusion - in fact, it finds that nothing could be more obvious. Plaintiffs’ illustrative maps were intended to demonstrate that it is possible to draw, minding the other necessary criteria, two majority-minority districts in Louisiana instead of one. Naturally, their maps are less similar to the benchmark.
Moreover, the Court struggles to grasp why Defendants elevate the importance of core retention. They cite no case which treats core retention as dispositive of, or even central to, the Gingles I inquiry. Furthermore, the Legislature‘s own redistricting rule is silent on core retention. As Plaintiffs highlight,
Thus, the Court concludes that, although Plaintiffs’ illustrative maps have lower core retention than the enacted plan, that fact is entitled to essentially no weight under
Ultimately, the Court finds that the illustrative plans developed by Plaintiffs’ experts satisfy the reasonable compactness requirement of Gingles I. In Defendants’ post-hearing brief, they assert that “it is only through a very specific set of contortions that a second majority-minority district can be extracted from Louisiana‘s demographics.” 277 If Plaintiffs’ maps are the result of improper “contortions,” those contortions somehow went undetected by the numerous statistical measures employed to demonstrate their adherence to traditional districting principles. Plaintiffs’ maps have roughly zero population deviation, contiguous districts, districts that are at least as geographically compact as the districts in the enacted plan - in fact, they are almost always more geographically compact. Plaintiffs’ maps protect incumbents, reflect communities of interest, and respect political subdivisions, splitting fewer parishes than the enacted map. Cooper and Fairfax both offered persuasive testimony regarding how they balanced all of the relevant principles, including the Legislature‘s
3. Equal Protection: Hays and Racial Gerrymandering
Defendants insist that the illustrative maps are racial gerrymanders as a matter of law. They cite the Hays series of cases from the 1990s, wherein Louisiana congressional maps with two majority-minority districts were invalidated as racial gerrymanders. 278 In 1993, the District Court for the Western District of Louisiana took up Hays v. State of Louisiana (Hays I), a private action which challenged a legislatively-enacted congressional map on
The Western District found that the Hays I map (depicted below) was the result of racial gerrymandering. The Court colorfully described the map as follows:
Like the fictional swordsman Zorro, when making his signature mark, District 4 slashes a giant but somewhat shaky “Z” across the state, as it cuts a swath through much of Louisiana. It begins north of Shreveport-in the northwestern corner of Louisiana, just east of the Texas border and flush against the Arkansas border-and sweeps east along that border, periodically extending pseudopods southward to engulf small pockets of black voters, all the way to the Mississippi River. The district then turns south and meanders down the west bank of the Mississippi
River in a narrow band, gobbling up more and more black voters as it goes. As it nears Baton Rouge, the district juts abruptly east to swallow predominantly black portions of several more parishes. Simultaneously, it hooks in a northwesterly arc, appropriating still more black voters on its way to Alexandria, where it selectively includes only predominantly black residential neighborhoods. Finally, at its southern extremity, the district extends yet another projection-this one westward towards Lafayette-adding still more concentrations of black residents. On the basis of District 4‘s physiognomy alone, the Plan is thus highly irregular, suggesting strongly that the Legislature engaged in racial gerrymandering.279
After the Western District‘s ruling, the Legislature adopted a new redistricting plan, and the finding of racial gerrymandering was vacated and remanded for further consideration in light of the new plan.
Defendants have repeatedly invoked Hays as a cautionary tale in this litigation, suggesting that because a map with two majority-Black districts was previously invalidated by a court, there can never be an acceptable map with two Black districts. In fact, the Legislative Intervenor Defendants use the word “insanity” to describe efforts to draw two, quipping in their Opрosition to the Motion for Preliminary Injunction that “[i]nsanity is doing the same thing over and over and expecting different results.” 281 For the 2020 redistricting cycle, they assert, the Louisiana Legislature kept Hays in mind and “did not succumb to this malady.” 282
Defendants’ assertion that Hays automatically vitiates the validity of Plaintiffs’ illustrative plans is refutable by a cursory visual inspection of the Hays maps. In the Hays I map, District 2 appears on the map of Louisiana with the coherence of a sneeze. It is not disputed that Plaintiffs’ illustrative plans draw a second majority-Black district by connecting parts of East Baton Rouge Parish with the Delta Parishes in their proposed CD 5. But apart from that commonality, the layout of their CD 5 is scarcely similar to Hays I‘s CD 4. Take, for example, Cooper‘s Illustrative Plan 1:
Instead of a narrow and jagged band reaching from the far northwest of the state all the way south toward the Gulf Coast, Cooper‘s map appears as a relatively compact, reasonable shape. Cooper‘s Illustrative Plan 2 reaches further to the northwest, but still avoids the plunge to the coast:
Likewise, Anthony Fairfax‘s illustrative maps connect East Baton Rouge to the Delta Parishes in compact form and have none of the deranged twists and turns of the map at issue in Hays I.
The district shown in black, CD4, was a majority-minority district that the Western District described as “an inkblot which has spread indiscriminately across the Louisiana map.” 285 Notably, this CD 4 does not commit what Defendants make out to be the cardinal sin of including East Baton Rouge Parish and the Delta Parishes in the same district. Clearly, then, it was not the combination of those areas that the Western District rejected - it was the diffuse and nonsensical configuration of the majority-minority districts. Plaintiffs’ expert Anthony Fairfax testified that the majority-minority districts in Hays were extremely non-compact, to the point that he would never draw them.
The invocation of Hays is a red herring. By every measure, the Black population in Louisiana has increased significantly since the 1990 census that informed the Hays map. According to the Census Bureau, the Black population of Louisiana in 1990 was 1,299,281. 286 At the time, the Census Bureau did not provide an option to identify as more than one race. The 2020 Census results indicate a current Black population in Louisiana of 1,464,023 using the single-race Black metric, and 1,542,119 using the Any Part Black metric. 287 So, by the Court‘s calculations, the Black population in Louisiana has increased by at least 164,742 and as many as 242,838 since the Hays litigation. Hays, decided on census data and demographics 30 years ago, is not a magical incantation with the power to freeze Louisiana‘s congressional maps in perpetuity. Hays is distinguishable and inapplicable. Defendants argue vociferously that race was the predominant factor in the creation of CD 2 and CD 5 in Plaintiffs’ illustrative maps. A plan that links locations solely on the basis of race is suspect race-based redistricting, they argue, and cannot satisfy Gingles I. Defendants assert that Cooper and Fairfax had “racial target[s]” 288 and that drawing two majority-minority districts was “non-negotiable” 289 for them. Because race was “the overriding reason for choosing one map
The Court rejects this argument, for both legal and factual reasons. As discussed supra, there is an inherent tension between the
The Supreme Court explicitly acknowledges that some consideration of race is permissible in the context of the
redistricting differs from other kinds of state decision making in that the legislature always is aware of race when it draws district lines, just as it is aware of age, economic status, religious and political persuasion, and a variety of other demographic factors. That sort of race consciousness does not lead inevitably to impermissible race discrimination.296
Plaintiffs argue that Defendants’ effort to import a requirement that map-making be demonstrably race neutral into Gingles I was explicitly rejected by the Fifth Circuit in the 1996 case Clark v. Calhoun County. 297 In Clark, the Fifth Circuit considered whether racial predominance is a factor in the Gingles I inquiry and concluded, quite clearly, that it is not. In Clark, the Fifth Circuit considered Calhoun County‘s argument that, because the plaintiffs’ predominant concern in drawing their proposed districts was race, those
The Clark court “agree[d] with the County‘s reading of Miller but disagree[d] that Miller is relevant to the first Gingles factor.” 300 “In contrast to Miller‘s focus on motivation,” the Fifth Circuit wrote, “the first Gingles factor requires that the plaintiff demonstrate that the minority group is ‘sufficiently large and geographically compact to constitute a majority in a single-member district.‘” 301 This demonstration is typically made, the court observed, by drawing hypothetical majority-minority districts. Based on Supreme Court precedent, 302 the Clark court held:
Miller‘s emphasis on purpose does not apply to the first Gingles precondition. In neither case did the Court suggest that a district drawn for predominantly racial reasons would necessarily fail the Gingles test. To the contrary, the first Gingles factor is an inquiry into causation that necessarily classifies voters by their race.303
The court went on:
[W]e do not understand Miller and its progeny to work a change in the first Gingles inquiry into whether a sufficiently large and compact district can be drawn in which the powerful minority would constitute a majority. To be sure, this test of causation insists upon a compact district, and a remedial response narrowly tailored to remedying a found violation must also be compact. As we will explain, however, that tailored response must use race at the expense of traditional political concerns no more than is reasonably necessary to remedy the found wrong.304
Further, the Clark court drew a distinction between the districts proposed by the plaintiffs, which “were ‘simply presented to demonstrate that a majority-black district is feasible in Calhoun County‘” 305 under Gingles I, and the remedial map that would ultimately be developed by the County in response to the court‘s ruling. A remedial map, the court explained, “must use race at the expense of traditional political concerns no more than is reasonably necessary to remedy the found wrong.” 306 This makes sense, since illustrative maps drawn by demographers for litigation are not state action and thus the
In a strained attempt to get around the well-reasoned holding of Clark and piggyback an
In the first, Abbott v. Perez, 309 the Supreme Court invalidated a lower court‘s decision to “defer[] a final decision on the § 2 issue and advise[] the plaintiffs to consider [it] at the remedial phase of the case.” 310 This is no more than a recognition of the hornbook legal principle that liability must be decided before a remedy can be ordered. Abbott does not hold that the liability inquiry and the remedial inquiry are the same. The Abbott court pointed out the lower Court‘s error in deferring part of the
The other case advanced by the Defendants, Anne Harding v. County of Dallas, Texas, 312 is likewise unavailing. 313 Finally, the Court finds that Wright v. Sumter Cnty. Bd. of Elections & Registration, 314 an Eleventh Circuit case, also poses no obstacle here. In Wright, the court instructed that “a district court‘s remedial proceedings bear directly on and are inextricably bound up in its liability findings.” 315 With that in mind, the Eleventh Circuit affirmed the district court‘s decision, finding that the challenged “district map impermissibly diluted black voting strength in violation of section 2 of the Voting Rights of 1965.” 316 The district court then, “with the help of a well-qualified special master, drew new district boundaries that plainly remedied the violation.” 317 Wright is not authority for the proposition that the legal analysis applicable to liability and remedy are “one and the same.” Wright states the obvious, that the liability and remedial phases are highly interrelated; it does not state that all legal theories applicable to the remedy apply with equal force during liability. As keen as Defendants are to bake the racial predominance inquiry into Gingles I, the Court finds no legal basis for doing so. Clark clearly sets forth the Fifth Circuit‘s rejection of the conflation of the racial gerrymandering doctrine with the vote dilution claims raised by Plaintiffs here.
Defendants also argue that, in Bethune-Hill, the Supreme Court clarified
In any event, the ”
Plaintiffs’ expert witnesses William Cooper and Anthony Fairfax explicitly and credibly testified that they did not allow race to predominate over traditional districting principles as they developed their illustrative plans. Defendants dismiss this testimony as “semantic,” 322 and they cite both Cooper and Fairfax‘s statements that they used 50% BVAP as a threshold as evidence that they employed unconstitutional racial targets. They further cite Cooper‘s statement that he “was specifically asked to draw two [majority-minority districts] by the plaintiffs.” 323 This is not the “gotcha” moment that Defendants make it out to be. It is well-established that in a vote dilution case, the method by which a plaintiff can prove numerosity to satisfy Gingles I is the production of illustrative maps demonstrating that it is possible to draw an additional 50% + majority-minority district. So, the fact that Plaintiffs asked Cooper to draw such a map is no surprise. And, while Cooper did testify that Plaintiffs asked him to draw two majority-Black districts, he also testified that he “did not have a goal to under all circumstances create two majority-Black districts” because “when developing a plan you have to follow traditional redistricting principles.” 324 And Fairfax‘s testimony established how he considered socioeconomic data extensively in deciding where to draw his lines. Overall, the Court found Cooper and Fairfax to be highly credible witnesses, and it credits their testimony that race did not predominate in their drawing as sincere.
In any event, if Plaintiffs’ experts engaged in race-predominant map drawing, their illustrative plans would surely betray this imbalanced approach by being significantly less compact, by disregarding communities of interest, or some other flaw. But the Court found that Plaintiffs’ plans outperformed the enacted plan on every relevant criteria. Moreover, the accusations that Defendants level at Plaintiffs’ illustrative plans - that they pick up areas of BVAP with “surgical precision” and unite far-flung areas with little in common - apply equally to the enacted plan‘s CD 2. Testimony at the hearing established that the enacted CD 2 is very non-compact and includes Baton Rouge and New Orleans, two major cities with significantly different economies and representation needs, in the same district.
Race-blind map drawing is not required by precedent - in fact, racially conscious map drawing has been recognized as necessary to comply with the
In Alabama‘s view . . the advent of computerized districting should change the way the first Gingles condition operates. Plaintiffs can now use technology to generate millions of possible plans, without any аttention to race. Alabama claims that some number of those plans (what number is unclear) must contain an additional majority-Black district for Section 2 plaintiffs to satisfy the first Gingles condition. But whatever the pros and cons of that method, this Court has never demanded its use; we have not so much as floated the idea, let alone considered how it would work.325
This Court declines to supplant thirty years of guiding precedent in vote dilution cases in favor of simulation maps created by someone who was performing such a simulation for the first time and whose maps bear absolutely no resemblance to the enacted plan or the previous plan.
B. Gingles II and III – Racially Polarized Voting
Gingles II asks whether Black voters are “politically cohesive,” 326 and Gingles III whether White voters vote “sufficiently as a bloc to usually defeat [Black voters‘] preferred candidate.” 327 Based on the testimony and reports of expert witnesses at the preliminary injunction hearing, the Court finds that the Plaintiffs are substantially likely to prove both prongs.
Gingles II asks whether Black voters are “politically cohesive” - in other words, whether Black voters usually support the same candidate in elections. On this factor, Plaintiffs offered opinions from Dr. Maxwell Palmer and Dr. Lisa Handley and the Defendants offered opinions from Dr. John Alford.
Dr. Palmer was offered by Plaintiffs as an expert in the field of redistricting with an emphasis in racially polarized voting and data analysis. Defendants stipulated to Dr. Palmer‘s expertise in the tendered field. Dr. Palmer opines that Louisiana has “a clear pattern of racially polarized voting,” where “Black voters have a clear candidate of choice in most statewide elections,” 328 and “Black and White voters consistently support different candidates.” 329
Dr. Handley was tendered and accepted, based on Defendants’ stipulation, as an expert in redistricting with an emphasis in racially polarized voting and data analysis. She reached the same conclusion as Dr. Palmer, opining that “[v]oting in recent elections in Louisiana is starkly racially polarized.” 330 The opinions of Drs. Handley and Palmer were based on a significant amount of historical voting data that they gathered and analyzed. Their conclusions were not seriously disputed at the hearing. Defendants’ expert Dr. Alford testified that he found no errors in Dr. Palmer‘s and Dr. Handley‘s work. Defendants’ expert witness Dr. Solanky testified that he does not dispute Palmer‘s and Handley‘s conclusions with respect to Gingles II.
Dr. John Alford testified as an expert for the Defendants on racially polarized voting. He does not dispute that voting in Louisiana is polarized as between Black and White voters; rather, it is his opinion that polarized voting in Louisiana is attributable to partisanship, not race. The Court does not credit this opinion as helpful, as it appears to answer a question that Gingles II does not ask and in fact squarely rejects, 331 namely, why Black voters in Louisiana are politically cohesive. Further, the Court finds that Dr. Alford‘s conclusions conflict with the opinions of other experts in this case who employed more robust methodology. Dr. Alford merely looked at the results reported by Dr. Palmer and Dr. Handley and opined that polarized voting “may be correlated with race, but whatever accounts for the correlation, the differential response of voters of different races
to the race of the candidate is not the cause.”332 Not only does this statement appear to concede that Dr. Alford does not know exactly why voting is polarized (“whatever accounts for the correlation“), Dr. Palmer‘s well-accepted ecological inference analysis contradicts it. Dr. Palmer demonstrated that the race of the candidate does have an effect; he found that Black voters support Black candidates more often in a statistically observable way. The Court finds that Dr. Alford‘s opinions border on ipse dixit. His opinions are unsupported by meaningful substantive analysis and are not the result of commonly accepted methodology in the
The Court rejects Defendants’ attempt to append an additional requirement to Gingles II, namely, that Black voters’ cohesion must be shown to be caused by or attributable to race instead of something else, like partisanship. The Court finds no basis for this requirement in the law.334
The Court credits Dr. Palmer‘s opinions and conclusions, finding that his methods were sound and reliable. His testimony was clear and straightforward, raising no issues that would cause the Court to question his credibility. Likewise, the Court credits the testimony and conclusions of Dr. Lisa Handley, who was accepted as an expert in redistricting with a focus on racially polarized voting. Dr. Handley‘s extensive expertise in the area of redistricting and voting rights is reflected in her CV and was аpparent from her testimony, which was thorough, careful, well-supported by data, facts and soundly reasoned. The Court finds the opinion testimony of Drs. Palmer and Handley to be both probative and reliable.
Defendants offered Dr. Tumulesh Solanky as an expert in the fields of mathematics and statistical analysis, to which Plaintiffs stipulated. While the Court does not question Dr. Solanky‘s credentials in the fields of mathematics and statistical analysis, the Court finds there is little, if any, connection between his expertise and his opinions. Solanky opined that “there is no evidence of legally significant racially polarized voting in [East Baton Rouge Parish],”335 and that the second minority-majority district proposed by Plaintiffs is created by “pull[ing] out Black voters primarily from [East Baton Rouge Parish].”336 According to his testimony, he has no experience in analyzing racially polarized voting patterns. Solanky used an admittedly narrow data set as the basis for his conclusions. He analyzed only East Baton Rouge Parish, which he conceded is not populous enough to form its own congressional district and would need to be analyzed with as many as 18 other parishes to form an opinion regarding the degree of polarization in a district. Dr. Solanky does not offer any opinion about majority bloc voting in any congressional district under the enacted or illustrative plans. The Court finds that Dr. Solanky‘s analysis is of limited utility, since at most it speaks to White voting behavior in one parish out of 64. Moreover, Dr. Solanky himself observed that East Baton Rouge is an outlier in terms of White crossover voting compared to surrounding parishes. Voting behavior in a small area that is concededly an outlier is not probative of voting patterns districtwide. Dr. Solanky‘s opinions are unhelpful and do not inform the Court‘s analysis under Gingles II.
Based on the evidence and the opinions of experts, the Court concludes that Plaintiffs have demonstrated that Black voters in Louisiana are politically cohesive.
Gingles III requires an inquiry into whether White voters in Louisiana vote “sufficiently as a bloc to usually defeat [Black voters‘] preferred candidate.”337
The Fifth Circuit and the Supreme Court have held that “the question here is not whether white residents tend to vote as a bloc, but whether such bloc voting is ‘legally significant.‘”341 Defendants posit that White bloc voting is not legally significant if “white crossover voting is sufficient to enable the Black community to elect its preferred candidates of choice in districts below 50 percent BVAP.”342 In Covington v. North Carolina, which was affirmed by the Supreme Court in 2017, the District Court for the Middle District of North Carolina held that “a general finding regarding the existence of any racially polarized voting, no matter the level, is not enough.”343 Because a statistically significant level of racially polarized voting could be found at, say, 51%-49%, the court explained, merely statistically significant levels “cannot be construed as conclusive evidence of the third Gingles factor.”344
The Covington court criticized the plaintiff‘s experts because they “never conducted an inquiry to determine whether racially polarized voting sufficient to enable the majority usually to defeat the candidate of choice of African-American voters was present in the challenged districts.”345 The same cannot be said of Plaintiffs’ experts in this case. Both Dr. Palmer and Dr. Handley examined this issue, amassed detailed data, and arrived at the same conclusion: that White voters consistently bloc vote to defeat the candidates of choice of Black voters.
The Court also finds, based on the work of Dr. Palmer and Dr. Handley, that Plaintiffs’ illustrative districts would not be opportunity districts in name only but would actually perform to allow Black voters a genuine opportunity to elect the candidate of their choice. Defendants seize on the fact that Plaintiffs’ experts all agreed, during their testimony, that it is possible for districts drawn below 50% BVAP to still “perform” because there may be enough White crossover voting to allow Black voters an opportunity to elect their preferred candidate. It is true that the Covington court called for an analysis of crossover voting under Gingles III, noting that high levels of crossover voting undermine a finding of legally significant polarized voting.346
The Court accepted Defendants’ witness Dr. Jeffrey Lewis as an expert in the fields of political science, census data analysis, and statistics, specifically racially polarized voting. Dr. Lewis advances the opinion that majority minority districts are unnecessary because in his view, Black voters have a meaningful opportunity to elect candidates of their choice owing to White crossover voting. Dr. Lewis‘s analysis is informed by a single election, the 2020 Presidential general election. Using data from that single election, he constructs a hypothetical in illustrative CD 2 and CD 5 where there are no White crossover votes for the Black-preferred candidate, from which he concludes that, without White crossover voting, the Black-preferred candidates, Biden/Harris, would not have been elected except in one illustrative district.348 This hypothetical based on limited data is not helpful to the Court‘s assessment of whether Plaintiff‘s illustrative maps “perform” for Black-preferred candidates. Likewise, Dr. Lewis‘s conclusion that districts with as little as 30% BVAP could perform for Black-preferred candidates due to White crossover voting was based on his analysis of one exogenous election. His opinion is simply unsupported by sufficient data and is accordingly unreliable. Dr. Lewis states that further analysis was not possible due to “time limitations.”349 The Court finds this excuse less than persuasive, especially since Dr. Lewis performed his analysis using the data gathered by Dr. Palmer.350
White crossover voting was inherently included in the analysis performed by Dr. Palmer and Dr. Handley, and the levels they found were insufficient to swing the election for the Black-preferred candidate in any of the contests they examined. The fact that Plaintiffs’ experts agreed, hypothetically, that a sub-50% BVAP district could perform under unspecified circumstances, is not sufficient to overcome the conclusions reached by their robust statistical analysis. Although Defendants insist
C. Senate Factors and Proportionality
Gingles counsels that a Section 2 violation is established:
if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of [a racial minority group] ... in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice.356
Courts have concluded that “it will be only the very unusual case in which the plаintiffs can establish the existence of the three Gingles factors but still have failed to establish a violation of § 2 under the totality of circumstances.”357 Indeed, here the Court finds that Plaintiffs have established that they are substantially likely to prevail in showing that the totality of the circumstances weighs in their favor. The Court first analyzes the Senate Factors (beginning with Factors 2 and 7, which Gingles marks as the “most important”358) and then turns to the proportionality issue.
1. Senate Factor 2
Senate Factor 2 examines “the extent to which voting in the elections of the state or political subdivision is racially polarized.”359 The Court already found, supra, that voting in Louisiana is racially polarized based on the substantial and mostly unrebutted evidence brought forward by Plaintiffs’ expert witnesses, who opined that the polarization is “stark.” Further,
Dr. Lichtman summarized that the Democratic and Republican parties have undergone a role reversal since the 1860s, and that Black voters were loyal not to the “label” of Republican but to their racial identity. Dr. Lichtman testified at the hearing that party labels have no meaning; what matters to voters is what candidates represent. “[P]arty identification is conjoined with race, although party labels ha[ve] come to mean the opposite of what they once were,”360 he writes. Dr. Handley‘s report also cites peer-reviewed scholarly studies which show that the racial attitudes of the parties, and their positions on race-related issues, are what drives support for a particular party.
The analytical evidence of voter polarization which forms the bases for Drs. Palmer and Handley‘s opinions is bolstered and substantiated by the historical voting patterns described by Drs. Lichtman and Gilpin. The polarization evidence is further bolstered by fact testimony, such as Ashley Shelton who testified that in her lived experience, Black voters in Louisiana prefer Democratic candidates, not because of the party label, but because Democrats are more likely to discuss the issues that matter to Black voters. The Court finds that Senate Factor 2 weighs heavily in favor of Plaintiffs.
2. Senate Factor 7
This factor assesses “the extent to which members of the minority group have been elected to public office in the jurisdiction.” It is undisputed that there has not been a Black candidate elected to statewide office in Louisiana since Reconstruction.361 Since 1991, only four Black Louisianans have been elected to Congress.362 Before 1991, there was one Black Congressperson elected, and it was during Reconstruction.363 Louisiana has never had a Black Congressperson elected from a non-majority-Black district.364 This underrepresentation persists at other levels of government, as well.
3. Senate Factor 1
This inquiry considers “[t]he extent of any history of official discrimination in the state ... that touched the right of the members of minority group to register, to vote, or otherwise to participate in the democratic process.”366
The Legislative Intervenors candidly concede that Louisiana has a “sordid history of discrimination.”367 In another recent voting rights case in this District, the Court found it “indisputable that Louisiana has a long history of discriminating against black citizens.”368 The expert historians who testified also recounted well-documented, undisputed historical facts of discriminatory voting laws and practices. Dr. Gilpin reported about voting restrictions like poll taxes, property ownership requirements, and literacy tests, which were first implemented before Black Louisianans were granted the right to vote. These mechanisms were first enforced against immigrants, and later against Black Louisianans after their right to vote was recognized. Dr. Gilpin recounted that Black voting in Louisiana reached its peak in 1896, when Black voters made up almost 45% of registered voters.369 This ushered in a period of onerous restrictions, which rendered Black voting all but futile. For example, the Grandfather Clause, enacted in 1898, prohibited a Black citizen from voting unless they could establish that either their father or grandfather had voted before January 1, 1867.370 As a result, Black voting plummeted dramatically. Registration purges, the Understanding Clause, and other restrictions disenfranchised Black voters to the point that, between 1910 and 1948, fewer than 1% of Black Louisianans of voting age were able to register to vote.371 By the passage of the 1965 Voting Rights Act, only one third of the Black population was registered.372
Nor did the Voting Right Act foretell an era free from racially motivated voting discrimination in Louisiana. Instead, the Act‘s provision for supervision of state practices meant that Louisianans were more aware of attempts to disenfranchise Black voters. From 1965 to 1999, the U.S. Attorney General issued 66 objection letters to more than 200 voting changes, and from 1990 until the end of preclearance in 2013, an additional 79 objection letters were issued.373 Recently, in 2021, the City of West Monroe entered into a consent decree with the Department of Justice related to its use of all at-large districts for elections to the Board of Aldermen; Dr. Gilpin explains that at-large elections commonly result in disenfranchisement of Black voters.
Dr. Gilpin opines that Black voter suppression results from modern day practices such as restricting access to polling
Defendants argue that “Plaintiffs did not present any meaningful recent evidence of official discrimination.”377 While “tragic,” they say, “it is in the distant past and is not especially probative of this Section 2 case in 2022.”378 Defendants tout the fact that Louisiana was recently ranked 7th in the nation for election integrity379 and the fact that there is White crossover voting380 as evidence that discrimination is a thing of the past. They assert that the State‘s abysmal preclearance history is merely indicative of “backsliding”381 and not discrimination.
Senate Factor 1 explicitly calls for an inquiry into any history of voting-related discrimination, but as discussed above, practices which result in barriers to Black voters continue. In this context, taking a broader look is well justified. A Black Louisianan born in 1965, the year the Voting Rights Act was passed, is only 57 years old today. This is not ancient history. Defendants’ contention that there is no evidence of Black voters being denied the right to vote is irrelevant. This case presents claims of vote dilution.
The Court is not persuaded by the facts adduced by Defendants in support of their argument that the “State of Louisiana has made significant strides in addressing its inequitable past as part of its recent history.”382 It is laudable that, for example, the LSU African-American Diversity Organization hosts programs including Umoja, a welcome event for freshman and transfer students;383 that the Legislature recently created a task force to study the lack of minority candidates for athletic director and head coach positions in the state;384 or that Juneteenth has been recognized as a holiday and that “many” members of the Louisiana State Bar Assoсiation have agreed to by abide by a Statement of Diversity Principles.385 But to the extent these facts are offered as mitigation of the repugnant history of discrimination in Louisiana, they fall completely flat.
Lastly, the Court finds that Louisiana‘s history of discrimination has been recognized by other federal courts. In the 2017 case Terrebonne Par. Branch NAACP v. Jindal, Judge James Brady analyzed
There is no sincere dispute regarding Senate Factor 1. The evidence of Louisiana‘s long and ongoing history of voting-related discrimination weighs heavily in favor of Plaintiffs.
4. Senate Factor 3
The Court next examines “[t]he extent to which the state ... has used ... voting practices or procedures that may enhance the opportunity for discrimination against the minority group.”390 Plaintiffs cite Louisiana‘s open primary system and majority-vote requirement as an example of discriminatory voting procedures, explaining that if a Black candidate wins a plurality of the vote in a White jurisdiction, they will have to face the White-preferred candidate in a runoff, where Black candidates rarely win. According to Dr. Lichtman, this system was enacted in 1975 to protect White incumbents from electoral challenges.391 Defendants dispute that the system was motivated by racial bias, arguing that Louisiana configured its primaries in this manner after a previous system was struck down by the Supreme Court in Foster v. Love.392
Dr. Lichtman points to the 2015 Lieutenant Governor race, the 2017 Treasurer race and the 2018 Secretary of State race as evidence that Louisiana‘s open primary system hinders the ability of Black-preferred candidates to win. The Court credits Dr. Lichtman‘s conclusion that the Black-preferred candidate lost in the runoff in each of these three elections, but the Court is unpersuaded that the results of three exogenous elections prove the point. The Court finds Senate Factor 3 neutral.
5. Senate Factor 4: Candidate Slating
There is no slating process for Louisiana‘s congressional elections, so this factor is not relevant and the Court makes no finding.393
6. Senate Factor 5
Factor 5 concerns “[t]he extent to which members of the minority group in the
Common sense suggests that a Louisianan who is barely getting by, who has limited access to transportation, who is in poor health, or who is functionally illiterate, is ill-equipped to exercise the franchise. However, a plain reading of Senate Factor 5 requires a showing of hindrance. The Court further agrees the Court‘s observation in Caster requiring explicit proof of impaired participation overlooks “the question whether the lasting effects of discrimination make it harder for Black [voters] to participate at the levels that they do.”396 Nonetheless, the Fifth Circuit counsels Courts to require “evidence of reduced levels of black voter registration, lower turnout among black voters, or any other factor tending to show that past discrimination has affected their ability to participate in the political process.”397 Without specific evidence that these disparities manifest themselves in political participation outcomes, the Court finds that Senate Factor 5 is neutral.
7. Senate Factor 6
Plaintiffs’ experts Dr. Burch and Dr. Lichtman spoke to “[w]hether political campaigns have been characterized by overt or subtle racial appeals.”398 They cite the following examples of racial appeals in Louisiana politics, among others:
- David Duke, a former Grand Wizard of the Ku Klux Klan, won three statewide elections (1990 U.S. Senate, 1991 gubernatorial open primary, and a 1991 gubernatorial runoff) by appealing “to white racial fears“;399
- Former Governor Mike Foster, during the 1995 gubernatorial runoff election, stated that predominantly White Jefferson Parish “is right next to the jungle in New Orleans and it has a very low crime rate“;400
- Incumbent Republican Senator David Vitter in 2010 released a campaign ad depicting Hispanic immigrants sneaking through a hole in a fence and being welcomed by people holding the banner of his opponent and a giant check made out to “all illegal aliens“;401
- Another ad by Vitter that photoshopped Governor Edwards next to President Barack Obama and stated that Edwards and Obama were scheming to release 5,500 dangerous
“thugs” and “drug dealers” back onto the streets;402 - Eddie Rispone, a 2019 gubernatorial candidate, produced an ad blaming Governor John Bel Edwards for crimes committed by people released from prison, featuring mugshots of Black men, then accused Governor Edwards and his family of “taking advantage of black people in Louisiana. . .since Louisiana was born.”403
This is some evidence that racial appeals are used in Louisiana politics, but the persuasive weight of the evidence is minimal. The Court finds that this factor weighs neither for nor against Plaintiffs.
8. Senate Factor 8
Senate Factor 8 invites inquiry related to “[w]hether there is a significant lack of responsiveness on the part of elected officials to the particularized needs of the members of the minority group.”404
Plaintiffs rely on the reports of Dr. Burch and Dr. Lichtman for this factor, arguing that the disparities in health, housing, employment, education, and criminal justice faced by Black Louisianans are “indicative of a failure on the part of elected officials to address the needs of Black residents.”405 Plaintiffs highlight the roadshow testimony of various Louisianans cited by Dr. Burch as evidence that Black Louisianans have a sense of being overlooked by politicians.406 On the other hand, Defendants argue persuasively that evidence of disparities is not, in and of itself, evidence of non-responsiveness.
The Defendants elicited testimony from Governor Edwards’ executive counsel Matthew Block that the Governor supported Medicaid expansion, criminal justice reform, and has appointed black officials to positions of authority. Though not overwhelming, this is evidence of at least one elected official‘s responsiveness to “particularized needs.”
Plaintiffs’ showing on this factor was somewhat anecdotal and indirect, and the Court does not have sufficient evidence before it to conclude that there is a significant lack of responsiveness by elected officials in Louisiana. This factor favors Defendants.
9. Senate Factor 9
Lastly, in considering the totality of the circumstances, the Court assesses “[w]hether the policy underlying the Plan is ‘tenuous.’ While the Legislative Intervenors joined this suit on the premise that they could represent “the policy considerations underpinning”407 the enacted plan, Defendants offered no direct evidence on that point. Not a single elected official testified about the policies or considerations underpinning the enacted plan. By contrast, Dr. Burch examined the legislative record and points to specific evidence that the reasons in favor of enacted plan offered by legislators “lacked empirical support, were vague or contradictory, or were based on misunderstandings.”408
For example, although several members cited population equality as one of their foremost priorities for the new map, when Senator Fields presented an amendment with lower population deviation and a second majority-minority district, Senator Hewitt retreated from the previously expressed
Defendants’ advancement of the importance of core retention is, in this Court‘s view, evidence of the tenuousness of the justifications for the enacted plan. As ably stated by Dr. Lichtman, core retention is nothing more than a guarantee that inequities in the map will be frozen in place despite changes in population. Defendants’ argument that they prioritized avoiding racial gerrymanders, anemically pointing to the Hays maps, is likewise unconvincing. The Defendants offered no persuasive or legally relevant illumination of the policies served by H.B. 1. The Court finds this his factor weighs in favor of Plaintiffs.
10. Proportionality
Section 2 expressly provides that “nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population;”410 however, the Supreme Court has held that “whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area” is a “relevant consideration” in the totality-of-the-circumstances analysis.411 “[P]roportionality ... is obviously an indication that minority voters have an equal opportunity, in spite of racial polarization to participate in the political process and to elect representatives of their choice....”412
The Court finds that Black representation under the enacted plan is not proportional to the Black share of population in Louisiana. This fact is not disputed. Although Black Louisianans make up 33.13% of the total population and 31.25% of the voting age population, they comprise a majority in only 17% of Louisiana‘s congressional districts.413 Accordingly, the Court finds that the proportionality consideration weighs in favor of Plaintiffs.
The Court concludes that on the record before it, the totality of the circumstances weighs in favor of Plaintiffs’ request for relief and finds that the Plaintiffs are substantially likely to prevail on the merits of their vote dilution claim.
III. IRREPARABLE HARM
The Court concludes that Plaintiffs have demonstrated that they will suffer an irreparable harm if voting takes place in the 2022 Louisiana congressional elections based on a redistricting plan that violates federal law. Voting is a “fundamental political right, because it is preservative of all rights.”414 “[O]nce the election occurs, there can be no do-over and no
Defendants do not dispute or directly address Plaintiffs’ arguments about irreparable harm. Instead, they posit that the real threat of irreparable harm is an injunction from this Court, which they claim “would pose an unacceptable risk of constitutional injury to hundreds of thousands of Louisiana residents”416 because this Court would be re-imposing a map with two majority-minority districts, which was struck down as an unconstitutional racial gerrymander in Hays almost thirty years ago. The Court considered and rejected Defendants’ contention that Hays maps are useful comparators here or that Hays is instructive, applicable, or otherwise persuasive. It is not. Accordingly, the Court finds that Plaintiffs will suffer an irreparable harm without a preliminary injunction. If the 2022 election is conducted under a map which has been shown to dilute Plaintiffs’ votes, Plaintiffs’ injury will persist unless the map is changed for 2024.
IV. BALANCE OF EQUITIES
Elements three and four of the preliminary injunction test, “[t]he balance of the equities and the public interest[,] “merge when the Government is the opposing party.”417 The Court concludes that protecting voting rights is quite clearly in the public interest, while allowing elections to proceed under a map that violates federal law most certainly is not. The irreparable harm to Plaintiffs’ voting rights outweighs the administrative burden articulated by Defendants.
The Court further concludes that the implementation of a remedial congressional map is realistically attainable well before the 2022 November elections in Louisiana. In the 2006 case Purcell v. Gonzalez, the Supreme Court vacated a lower court injunction enjoining Arizona‘s voter identification procedures because the injunction came “just weeks before an election.”418 The High Court reasoned that “Court orders affecting elections . . .can themselves result in voter confusion and consequent incentive to remain away from the polls. As an election draws closer, that risk will increase.”419 The Supreme Court vacated the lower court‘s injunction for the reasons that “[g]iven the imminence of the election and thе inadequate time to resolve the factual disputes, our action today shall of necessity allow the election to proceed without an injunction suspending the voter identification rules.”420 Defendants call Purcell a “seminal opinion”421 which requires that this Court do nothing.
As the Caster court points out, Purcell is not the only opinion ever advanced by the Supreme Court on the subject of timing. In Reynolds v. Sims, for example, the Court explained that “once a State‘s legislative apportionment scheme has been found to be unconstitutional, it would be the unusual case in which a court would be justified in not taking appropriate action to insure that no further elections are conducted
A necessity standard was endorsed in Upham v. Seamon, where the Supreme Court stated that “we have authorized District Courts to order or to permit elections to be held pursuant to apportionment plans that do not in all respects measure up to the legal requirements, even constitutional requirements. Necessity has been the motivating factor in these situations.”425 Defendants urge consideration Justice Kavanaugh‘s concurrence in Merrill v. Milligan, where he opined that:
Running elections state-wide is extraordinarily complicated and difficult. Those elections require enormous advance preparations by state and local officials, and pose significant logistical challenges. The District Court‘s order would require heroic efforts by those state and local authorities in the next few weeks—and even heroic efforts likely would not be enough to avoid chaos and confusion.426
The Court concludes that neither Purcell nor any other case compels this Court to withhold immediate relief. The evidence presented at the hearing demonstrates that, although the administrative tasks that would be necessitated by a new congressional map would challenge the Secretary of State‘s office, the effort required would not be a heroic undertaking. Sherri Hadskey, the Commissioner of Elections, testified that after the Governor‘s veto was overridden and the enacted map became law, her office was able to update their records and send out mailings to all impacted voters in less than three weeks. The number of voters affected by a remedial map would likely be greater than the number affected by the change from the 2011 map to the 2022 map. Nevertheless, even if it took twice as long to accomplish, six weeks would be enough. Hadskey further testified that she was concerned that, if the process was rushed, her office may code voter information incorrectly, leading to incorrect information on ballots.
The Court finds that, although Hadskey‘s testimony demonstrated general concern about the prospect of having to issue a new round of notices to voters, she did not provide any specific reasons why this task cannot be completed in sufficient time for November elections, nearly 6 months from now. Likewise, the Court is not persuaded that the national paper shortage referenced by Hadskey in her testimony presents a significant burden. There was no evidence that State‘s paper demands would be impacted by map changes. Plaintiffs question, as does the Court, how paper usage is affected by the shape of Louisiana‘s congressional districts. Further,
Defendants repeatedly claim that June 22, 2022 is a critically impending deadline. June 22 is the deadline for potential congressional candidates to qualify by nominating petition. Notably, Hadskey testified that it is “extremely rare” for candidates to qualify by nominating petition. The evidence was that most pay the filing fee аnd qualify during the July 20-22 qualifying period. July 20 is more than six weeks from now. And Louisiana‘s unique election schedule, with the open primary not occurring until November, allows the State 6 months to accomplish what needs to be done.
Defendants warn the Court against issuing an injunction that would “act like [a] hurricane[],”427 but the Court finds the metaphor shallow. Placing a bureaucratic strain on a state agency in order to rectify a violation of federal law is not analogous to a natural disaster.
Most importantly, the Court finds that the credibility of Defendants’ assertions regarding the imminence of deadlines lacks credence. The Legislative Intervenors in this case, President Cortez and Speaker Schexnayder, made judicial admissions to the contrary in March 2022 in a currently pending state court case regarding the very congressional redistricting at issue here.428 President Cortez and Speaker Schexnayder asserted that: “the candidate qualification period could be moved back, if necessary, as other states have done this cycle, without impacting voters.”429 They further represented that: “[t]he election deadlines that actually impact voters do not occur until October 2022. . .Therefore, there remains several months on Louisiana‘s election calendar to complete the process.”430 There was no rush, they assured the court, because Louisiana‘s “election calendar is one of the latest in the nation.”431
In the context of the state court suit, the Legislative Intervenors were attempting to demonstrate that judicial intervention to resolve the impasse on redistricting was not necessary, and in that context, they painted a very different picture than the one they paint for this Court. The Court sees no reason why the statement that qualifying deadlines could be moved without ill effect does not apply with equal force in this context.
Likewise, the Court relies upon the statement of Secretary Ardoin‘s counsel in the state court proceeding that Louisiana does not have “a hard deadline for redistricting,”432 and “the Legislature. . .can
Defendants have not pointed to a single piece of evidence that an order from this Court would require the type of “heroic efforts” that Justice Kavanaugh warns about. The Court credits the testimony of Matthew Block that state election officials and officeholders have significant experience with adjusting the time, place, and manner of elections, thanks to natural disasters like Hurricane Ida and the COVID-19 pandemic. Major deadlines are still months away; overseas absentee ballots are not due to be mailed until September 24, 2022, and limited early voting begins October 18, 2022.434 And the qualifying deadline is adjustable, according to Defendants themselves.
There are a number of recent Supreme Court actions that merit consideration on this topic. In Moore v. Harper, the Supreme Court denied an application for stay out of North Carolina, with Justice Kavanaugh explaining his vote to deny stay as follows:
In light of the Purcell principle and the particular circumstances and timing of the impending primary elections in North Carolina, it is too late for the federal courts to order that the district lines be changed for the 2022 primary and general elections, just as it was too late for the federal courts to do so in the Alabama redistricting case last month.435
From the dissent, penned by Justice Alito and joined by Justices Thomas and Gorsuch, this Court is given to understand that the application for stay was received by the Supreme Court “only seven days before the deadline for candidates to file on March 4.”436 This stands in stark contrast to the timing in the instant case, where the most commonly-utilized method of candidate qualifying does not begin for more than six weeks. Remarkably, the dissent stated that even seven days before qualifying, “promptly granting a stay would have been only minimally disruptive.”437
In a per curiam opinion in Wisconsin Legislature v. Wisconsin Elections Commission, the Supreme Court reversed the imposition of redistricting maps and remanded to the Wisconsin Supreme Court, expressing confidence that the court would have “sufficient time to adopt maps consistent with the timetable for Wisconsin‘s August 9th primary election.”438 The Court‘s statement came on March 23, 2022, when the August 9th primary date was 139 days away. Louisiana‘s open congressional primary will occur on November 8, 2022, more than 150 days from now. Under the Supreme Court‘s guiding precedent, this provides “sufficient time” for the State to adjust its procedures and accomplish the tasks necessary to administer the election.
In Merrill v. Milligan, Justice Kavanaugh explained that in his view, а stay of the Caster court‘s order was necessary because “the primary elections begin (via absentee voting) just seven weeks from now, on March 30.”439 This timing is distinguishable from the instant case, where the candidates for office will not even be
The Court finds that a remedial congressional plan can be implemented in advance of the 2022 elections without excessive difficulty or risk of voter confusion. Defendants’ assertion that “an entirely new congressional plan”440 will be required at significant cost and hardship rings hollow. The Legislature would not be starting from scratch; bills were introduced during the redistricting process441 that could provide a starting point, as could the illustrative maps in this case, or the maps submitted by the amici.442
Other courts that have invalidated redistricting plans have imposed deadlines in the range of ten days to two-and-a-half weeks for the legislature to craft a new plan.443 With a new map in place by mid-June, the Secretary of State would have roughly five weeks before the July 20 qualifying period begins to update records and notify voters. This effort “does not rise to the level of a significant sovereign intrusion.”444 Given the timing of Louisiana‘s election and election deadlines, the representations made by Defendants in related litigation, and the lack of evidence demonstrating that it would be administratively impossible to do so, the Court finds that the State has sufficient time to implement a new congressional map without risk of chaos.
V. REMEDY
Defendants argue that a preliminary injunction is improper because Plaintiffs seek relief that “would establish a state of affairs that never before existed and does not preserve the status quo pending trial.”445 The Court finds useful instruction in the Fifth Circuit‘s analysis of the “status quo” issue in Canal Auth. of State of Fla. v. Callaway:
It must not be thought, however, that there is any particular magic in the phrase ‘status quo.’ The purpose of a preliminary injunction is always to prevent irreparable injury so as to preserve the court‘s ability to render a meaningful decision on the merits. It often happens that this purpose is furthered by preservation of the status quo, but not always. If the currently existing status quo itself is causing one of the parties irreparable injury, it is necessary to alter the situation so as to prevent the injury, either by returning to the last uncontested status quo between the parties, by the issuance of a mandatory injunction, or by allowing the parties to take proposed action that the court finds will minimize the irreparable injury. The focus always must be on prevention of injury by a proper order, not merely on preservation of the status quo.446
The Court pays heed to the principle that “[f]ederal-court review of districting legislation represents a serious intrusion on the most vital of local functions. It is well settled that ‘reapportionment is primarily the duty and responsibility of the State.‘”450 Thus, “[f]ederal courts are barred from intervening in state apportionment in the absence of a violation of federal law precisely because it is the domain of the States, and not the federal courts, to conduct apportionment in the first place.”451 The State has a “sovereign interest in implementing its redistricting plan.”452
The Supreme Court “has repeatedly held that redistricting and reapportioning legislative bodies is a legislative task which the federal courts should make every effort not to pre-empt.”453 Upon a federal court‘s finding that a redistricting plan violates the federal law, “it is therefore, appropriate, whenever practicable, to afford a reasonable opportunity for the legislature to meet [applicable federal legal] requirements by adopting a substitute measure rather than for the federal court to devise and order into effect its own plan. The new legislative plan, if forthcoming, will then be the governing law unless it, too, is challenged and found to violate”454 federal law.
After a determination that a redistricting plan violates Section 2, “[s]tates retain broad discrеtion in drawing districts to comply with the mandate of § 2.”455 The State may elect to use one of Plaintiffs’ illustrative plans, but is not required to do so, nor must it “draw the precise compact district that a court would impose in a successful § 2 challenge.”456 Overall, “the States retain a flexibility that federal courts enforcing § 2 lack, both insofar as they may avoid strict scrutiny
The Court‘s imposition of a particular map becomes necessary only if the Legislature fails to adopt its own remedial map according to the Court‘s deadline. “Legislative bodies should not leave their reapportionment tasks to the federal courts; but when those with legislative responsibilities do not respond, or the imminence of a state election makes it impractical for them to do so, it becomes the unwelcome obligation of the federal court to devise and impose a reapportionment plan pending later legislative action.”458
Accordingly, this Court provides the Legislature an opportunity to enact a new map that is compliant with
IT IS ORDERED.
Baton Rouge, Louisiana, this 6th day of June, 2022.
SHELLY D. DICK
CHIEF DISTRICT JUDGE
MIDDLE DISTRICT OF LOUISIANA
Notes
(AG-2, p. 20-22).