Opinion
3:22-cv-22
11-17-2023
Turtle Mountain Band of Chippewa Indians, et al., Plaintiffs, v. Michael Howe, in his Official Capacity as Secretary of State of North Dakota, Defendant.
FINDINGS OF FACT AND CONCLUSIONS OF LAW
Peter D. Welte, Chief Judge United States District Court
Plaintiffs Turtle Mountain Band of Chippewa Indians (“Turtle Mountain Tribe”), Spirit Lake Tribe (“Spirit Lake Tribe”), Wesley Davis, Zachery S. King, and Collette Brown assert the State of North Dakota's 2021 legislative redistricting plan dilutes Native American voting strength by unlawfully packing subdistrict 9A of district 9 with a supermajority of Native Americans and cracking the remaining Native American voters in the region into other districts, including district 15-in violation of Section 2 of the Voting Rights Act of 1965. Defendant Michael Howe, the Secretary of State of North Dakota, denies the Section 2 claim, arguing the 2021 redistricting plan is lawful.
Section 2 of the VRA prohibits any “standard, practice, or procedure” that “results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color[.]” 52 U.S.C. § 10301(a). It prohibits what the Tribes claim happened here-“the distribution of minority voters into districts in a way that dilutes their voting power.” Wis. Legislature v. Wis. Elections Comm'n, 142 S.Ct. 1245, 1248 (2022) (citing Thornburg v. Gingles, 478 U.S. 30, 46 (1986)). In Gingles, the United States Supreme Court identified three preconditions that must be initially satisfied to proceed with a Section 2 voter dilution claim:
1. The minority group . . . is sufficiently large and geographically compact to constitute a majority in a single-member district;
2. The minority group . . . is politically cohesive; and,
3. The white majority votes sufficiently as a bloc to enable it-in the absence of special circumstances . . . usually to defeat the minority's preferred candidate.478 U.S. at 50-51. Failure to prove any of the three preconditions defeats a Section 2 claim. Clay v. Bd. of Educ., 90 F.3d 1357, 1362 (8th Cir. 1996). If all preconditions are met, then there is a viable voter dilution claim, and the analysis shifts to determining whether, under the totality of the circumstances, members of the racial minority group have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. 52 U.S.C. § 10301(b).
A four-day bench trial was held on June 12, 2023. After consideration of the testimony at trial, the exhibits introduced into evidence, the briefs of the parties, and the applicable law, what follows are my findings of fact and conclusions of law. And as explained below, the Tribes have established a Section 2 violation of the VRA.
I. FINDINGS OF FACT
A. The Parties
Two Tribes and three individual voters make up the Plaintiffs. For the Tribes, the Turtle Mountain Tribe is a federally recognized Tribe under 88 Fed. Reg. 2112 (2023), possessing “the immunities and privileges available to federally recognized Indian Tribes[.]” Jamie Azure is its Chairman. Doc. 117 at 10:25-11:4. The Turtle Mountain Reservation is located entirely within Rolette County in northeastern North Dakota and covers 72 square miles. A large portion of Turtle Mountain's trust land is also located in Rolette County. Id. at 13:12-14:23; Id. at 15:11-16:4. The Turtle Mountain Tribe has over 34,000 enrolled members, and approximately 19,000 members live on and around the Turtle Mountain Reservation, including on Turtle Mountain trust lands in Rolette County. Id. at 13:12-14:23.
The second Tribe is the Spirit Lake Tribe, which is also a federally recognized Tribe. Douglas Yankton, Sr. is its former Chairman. He served as Chairman during the 2021 redistricting process. Doc. 115 at 45:12-22. The Spirit Lake Tribe is located on the Spirt Lake Reservation. The Spirit Lake Reservation covers approximately 405 square miles, primarily in Benson County in northeastern North Dakota. Id. at 47:10-48:2, 55:13-23. The Spirit Lake Tribe has approximately 7,559 enrolled members, with approximately 4,500 members living on or near the Spirit Lake Reservation. Id. at 47:10-48:2.
Three individual voters join the Tribes as Plaintiffs: Wesley Davis, Zachary King, and Collette Brown. Davis and King are enrolled members of the Turtle Mountain Tribe. They live on the Turtle Mountain Reservation, are eligible to vote, and plan to continue voting in elections. They currently reside in what is now Senate district 9 and House subdistrict 9A. Doc. 108 at 6. Brown is an enrolled member of the Spirit Lake Tribe. She lives on the Spirit Lake Reservation, is eligible to vote, and plans to continue voting in elections. She resides in district 15. Doc. 116 at 7:8-9:11.
The Secretary is sued in his official capacity as Secretary of State of North Dakota. Doc. 108 at 7. The Secretary is responsible for “supervis[ing] the conduct of elections,” and “publish[ing] . . . a map of all legislative districts.” N.D. Cent. Code §§ 16.1-01-01(1) & (2)(a).
B. North Dakota's 2021 Redistricting Plan
Article IV, Section 2 of the North Dakota Constitution requires the state legislature to redraw the district boundaries of each legislative district following the Census that happens every 10 years. The North Dakota Legislative Assembly (“Legislative Assembly”) is required to “guarantee, as nearly as is practicable, that every elector is equal to every other elector in the state in the power to cast ballots for legislative candidates.” N.D. Const., Art. IV, Sec. 2. It is also required to “fix the number of senators and representatives and divide the state into as many senatorial districts of compact and contiguous territory as there are senators” and requires that the “senate must be composed of not less than forty nor more than fifty-four members, and the house of representatives must be composed of not less than eighty nor more than one hundred eight members. These houses are jointly designated as the legislative assembly of the state of North Dakota.” Id., Sec. 1. So, one Senator and at least two House members are allocated to each district. Section 2 of Article IV allows the House members to be either elected at-large from the district or elected from subdistricts created within the district. Id., Sec. 2.
1. North Dakota's Legislative Districts Before the 2021 Redistricting
Recall that the Tribes challenge changes made to districts 9 and 15. For the decade prior to the 2021 redistricting, district 9 was entirely within Rolette County. Doc. 108 at 3. It had a Native American voting age population (“NVAP”) of 74.4%, did not contain any subdistricts, and contained the entirety of the Turtle Mountain Reservation, and its trust land located within Rolette County. Id. This map shows the pre-2021 legislative districts in the region:
(Image Omitted)
Pl. Ex. 103.
2. 2021 Redistricting Process and Plan
As a result of the COVID-19 pandemic, the 2020 Census data was delayed. Doc. 116 at 149:18-150:2. While waiting for the new data, on April 21, 2021, Governor Burgum signed House Bill 1397. It established a legislative management redistricting committee (“Redistricting Committee”) that was required to develop and submit a redistricting plan by November 30, 2021, along with implementation legislation. Doc. 108 at 1.
On May 20, 2021, then-Chairman Yankton sent a letter to the Redistricting Committee, requesting they schedule public hearings on each of the reservations located within North Dakota. Pl. Ex. 155. In response, the North Dakota Tribal and State Relations Committee held a joint meeting with the Tribal Council of the Turtle Mountain Tribe at the Turtle Mountain Community College on the Turtle Mountain Reservation. Def. Ex. 305; Doc. 108 at 2.
Redistricting was discussed at the joint meeting for roughly 30 minutes. Def. Ex. 418 at 17:18-21; Def. Ex. 305. Chairman Azure testified he became aware that redistricting had been added to the meeting agenda shortly before the meeting began. Doc. 117 at 29:21-31:24. He testified the Tribe had limited information about the 2020 Census population data and the discussion focused primarily on a population undercount. Id. at 29:21-31:24. One individual spoke in favor of subdistricts generally during the 30-minute discussion. Id. at 70:4-73:19.
Eventually, on August 12, 2021, the Census Bureau released redistricting data in legacy format (meaning the format used in specific redistricting software). Doc. 108 at 2. The Census data was released in a user-friendly format to the public on September 16, 2021. Id. at 2. The Redistricting Committee held public meetings in Bismarck on August 26, 2021, in Fargo on September 8, 2021, and again in Bismarck on September 15 and 16. Additional public meetings of the Redistricting Committee were held in Bismarck on September 22 and 23, and September 28 and 29. Id. at 3.
Brown testified on behalf of the Spirit Lake Tribe at the August 26 Redistricting Committee meeting. She advocated for the Redistricting Committee to consider tribal input and for the use of single member districts to elect representatives to the House. Def. Ex. 327. Brown also encouraged the Redistricting Committee to comply with the requirements of the VRA. Id.
On September 1, 2021, the Tribal and State Relations Committee held a public meeting at the Spirit Lake Casino and Resort on the Spirit Lake Reservation and discussed redistricting. Doc. 108 at 2. Chairman Yankton testified that Spirit Lake may be interested in a legislative subdistrict to elect its House member. Def. Ex. 334. At subsequent meetings, representatives of Spirit Lake requested a subdistrict. Def. Ex. 351; Def. Ex. 398.
At its September 28 and 29 meetings, the Redistricting Committee released several proposals for creating two subdistricts in district 9. Def. Ex. 405. One proposal extended district 9 to the east to incorporate population from Towner and Cavalier Counties, created a subdistrict in district 9 that generally encompassed the Turtle Mountain Reservation, and placed Spirit Lake in an at-large district with no subdistrict. Def. Ex. 408.
About a month after that proposed plan was introduced, the Tribes each consulted their leadership, obtained an analysis of racially polarized voting, created a new proposal for district 9, and sent a letter to the Governor and legislative leaders with their proposal. Pl. Ex. 156 at 19-24; Doc. 115 at 77:5-79:18; Doc. 117 at 34:14-36:11. The letter stated that the Redistricting Committee's proposal as to district 9, which placed the Turtle Mountain Reservation in a subdistrict, was a VRA violation. It also stated that the Turtle Mountain Tribe did not request to be placed in a subdistrict. Pl. Ex. 156 at 19-24. Included in the letter was an illustration of an alternative district map, where the Turtle Mountain and Spirit Lake Reservations were placed into a single legislative district with no subdistricts. Pl. Ex. 156 at 19-24; Doc. 108 at 4. Effectively, this alternative district combined Rolette County with portions of Pierce and Benson Counties, instead of combining Rolette County with portions of Towner and Cavalier Counties. Compare Pl. Ex. 156 at 19-24 with Def. Ex. 408. The letter stated that voting in the region is racially polarized, with Native American voters preferring different candidates than white voters. Id. at 19-24.
Then, at the November 8, 2021, Redistricting Committee meeting, Senator Richard Marcellais, who represented district 9 since his election in 2006, spoke in favor of the Tribes' proposed district. Def. Ex. 429 at 21-23. Representative Marvin Nelson from district 9 also spoke in favor of the proposal. Id. at 33-35. Representative Joshua Boschee moved for the adoption of an amendment to include the Tribes' proposal, but the amendment did not pass. Doc. 108 at 4. The Redistricting Committee passed and approved its final redistricting plan and report, which recommended passing the original proposal involving districts 9 and 15 (extending district 9 to the east to incorporate population from Towner and Cavalier Counties, creating a subdistrict in district 9 encompassing the Turtle Mountain Reservation, and placing Spirit Lake in an at-large district with no subdistrict).
The next day, the House of Representatives debated and passed House Bill 1504, the redistricting legislation accompanying the Redistricting Committee's final plan and report. Id. at 5. Then the Senate debated House Bill 1504. Senator Marcellais moved for an amendment (similar to the one he proposed to the Redistricting Committee), but it did not pass. Id. The Senate passed House Bill 1504, which was signed by Governor Burgum on November 11, 2021. Id.
3. 2021 Redistricting Plan As Enacted
As enacted, the 2021 redistricting plan created 47 legislative districts and subdivided district 9 into single-member House subdistricts 9A and 9B. Id. The plan extended district 9 eastward to include portions of Towner and Cavalier Counties, with the Towner County and Cavalier County portions included with parts of Rolette County in subdistrict 9B. Pl. Ex. 100. It also placed the Turtle Mountain Reservation into Senate district 9 and House subdistrict 9A and placed portions of Turtle Mountain trust lands located within Rolette County into House subdistrict 9B. Doc. 108 at 5. The plan placed the Spirit Lake Reservation in district 15. Doc. 108 at 5.
According to the 2020 Census, the NVAP of Rolette County is 74.4%. The NVAP of the portion of Towner County in district 9 is 2.7%. There is an NVAP of 1.8% in the portion of Cavalier County in district 9. Pl. Ex. 1 at 16. Subdistrict 9A has a NVAP of 79.8% and subdistrict 9B has a NVAP of 32.2%. Pl. Ex. 42 at 7; Doc. 115 at 134:13-19, 136:7-137:25. District 15 has a NVAP of 23.1%. Doc. 115 at 135:3-13; Doc. 108 at 4.
Voters in Senate district 9 and Senate district 15 each elect one Senator. Doc. 108 at 5. Voters in House subdistricts 9A and 9B each elect one representative to the House of Representatives. Id. Voters in district 15 elect two representatives at-large to the House of Representatives. Id. This is the 2021 plan's map of the legislative districts in northeastern North Dakota:
(Image Omitted)
Pl. Ex. 101.
C. The Tribes' Proposed Plans
In support of their Section 2 claim, the Tribes produced two proposed plans containing alternative district configurations that demonstrate the Native American population in northeast North Dakota is sufficiently large and geographically compact to constitute an effective majority in a single multimember district. This is the first proposed plan:
(Image Omitted)
Pl. Ex. 105. And this is the second proposed plan:
(Image Omitted)
Pl. Ex. 106. Both feature a district 9 that has a majority NVAP. The first proposed plan has a NVAP of 66.1%, and the second has a NVAP of 69.1%. Doc. 115 at 134:22-135:2, 135:14-17, 166:1-3.
D. Trial Testimony and Evidence on Section 2 Claim
At trial, former Chairman Yankton (Doc. 115 at 41-120), Collette Brown (Doc. 116 at 6-44), former Senator Richard Marcellais (Doc. 116 at 44-71), former House of Representatives member Marvin Nelson (Doc. 116 at 170-189), and Chairman Jamie Azure (Doc. 117 at 10-66) testified as fact witnesses for the Tribes. Erika White (Doc. 117 at 186-203) and Bryan Nybakken (Doc. 118 at 6-38), two representatives of the Secretary of State's office, testified as fact witnesses for the Secretary.
Four expert witnesses testified. Dr. Loren Collingwood (Doc. 115 at 120-201), Dr. Daniel McCool (Doc. 116 at 72-143), and Dr. Weston McCool (Doc. 116 at 144-170) testified as expert witnesses for the Tribes. Dr. M.V. Hood III (Doc. 117 at 72-182) testified as an expert witness for the Secretary.
Former Chairman Yankton testified to the shared representational interests, socioeconomic status, and cultural and political values of Turtle Mountain Tribal members and Spirit Lake Tribal members. Doc. 115 at 50:24-52:11, 52:24-73:9; Doc. 117 at 22:4-16-27:15, 28:18-25; 50:3-7; 52:23-53:1, 55:9-12. 115. He also testified as to the political cohesiveness of the Tribes, explaining that the voters who live on the Turtle Mountain Reservation and the voters who live on the Spirit Lake Reservation vote similarly. Doc. 115 at 52:12-53:25.
He also testified specifically as to the 2018 election (which is a key point of contention in this case), where Native American voter turnout was particularly high. He stated that there were unique circumstances that led to increased Native American voter turnout in 2018. Those circumstances included the election being a high-profile race, a backlash by Native American voters (who perceived North Dakota as trying to block them from voting by imposing a residential address requirement to vote), and the significant national attention and resources that flowed into the Tribes following the decision allowing the address requirement to go into effect just before the election. He testified that those resources-and resulting high voter turnout among Native American voters-was unlike anything he had seen, before or since. Doc. 115 at 80:18-86:17.
Dr. Loren Collingwood testified next. Doc. 115 at 119. Dr. Collingwood is an Associate Professor of Political Science at the University of New Mexico. Id. at 120. He teaches statistical programming, along with American politics, among other things. He has published several papers on the VRA and is qualified as an expert on voting behavior, race and ethnicity, racially polarized voting, map drawing, electoral performance, and redistricting analysis. Id. at 128:7-17.
Dr. Collingwood's expert testimony was extensive. He opined on each of the three Gingles preconditions. He reviewed the statistical data and analysis he used in reaching his expert conclusions as to racially polarized voting, white bloc voting, and the NVAP in the as-enacted districts compared to the Tribes' proposed districts. His expert reports were also admitted and received as exhibits. Pl. Ex. 1, 42.
Dr. Collingwood concluded that all three Gingles preconditions were met in districts 9 and 15. He found that racially polarized voting is present in North Dakota statewide and specifically in districts 9 and 15. He also found that, in statewide elections featuring Native American candidates, white voters vote as a bloc to Native American voters in all of the elections analyzed. He opined on the NVAP percentages. He further opined that there is racially polarized voting in district 9, subdistricts 9A and 9B, and district 15. Doc. 115 at 144-45.
Dr. Collingwood also opined on white bloc voting. Id. at 153-66. After wide review of his statistical analysis, he concluded that the white voting bloc usually defeats the Native American-preferred candidate of choice in districts 9, 9B, and 15. Id.
As to the 2018 election, Dr. Collingwood testified that the election was “an anomalous election.” Id. at 156. He noted that he had “never seen any turnout number like this, ever.” Id. As a result, he gave the 2018 election results less probative value and less weight, though the results were still included in his analysis. Id. at 158.
Collette Brown testified next for the Tribes. Doc. 116 at 6. Brown is the Gaming Commission Executive Director for the Spirit Lake Gaming Commission. Id. at 8. She ran for the Senate seat in district 15 in the 2022 election. Id. at 9. She spoke about the need for Native American representation and some of the difficulties she faced in her election campaign. Id. at 14. Brown also testified about her involvement in the 2021 redistricting process. Id. at 23. She stated that the Tribes did not request the subdistricts in district 9A and 9B. Id. at 23.
Former Senator Richard Marcellais testified next. Marcellais is an enrolled member of the Turtle Mountain Tribe and was the elected state Senator for district 9 from 2006-2022. Id. at 45, 48. He testified that he lost the 2022 election, and that after his loss, there are no Native Americans serving in the North Dakota Senate. Id. at 53.
Dr. Daniel McCool then testified as the second expert witness for the Tribes. Dr. Daniel McCool is a political science professor at the University of Utah. He specializes in Native American voting rights and Native American water rights. Id. at 73. He opined on the presence of the Senate Factors in North Dakota and the impact of the 2021 redistricting plan on Native Americans. Id. at 81. He reviewed in detail his expert report and concluded that there was substantial evidence of all of the Senate Factors, except factors four and six. Id. at 89-126. He concluded that, under the totality of the circumstances, Native Americans in North Dakota have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. Id.
Dr. Weston McCool testified as the third expert witness for the Tribes. He is a National Science Foundation post-doctoral research fellow with the Anthropology Department at the University of Utah. Id. at 144. His expertise is in quantitative data analysis and analytical methods. Id. He opined specifically as to Senate Factor 5. He reviewed his statistical analysis of seven socioeconomic variables, including education, employment, and health. Id. at 161. He concluded that Native Americans in the counties at issue bear the effects of discrimination along the socioeconomic factors articulated by Senate Factor 5, and the disparities serve as obstacles to hinder Native Americans' ability to effectively participate in the political process. Id.
Next former Representative Marvin Nelson testified. Doc. 116 at 170. He testified as to his experience representing Rolette County from 2010 to 2022. Id. at 172.
The final witness for the Tribes was Turtle Mountain Tribal Chairman Jamie Azure. Doc. 117 at 11. He testified about the Turtle Mountain Tribe and its membership. Id. at 14. He also spoke about the legislative district make-up before the 2021 redistricting plan, relative to the Tribes' Reservations and trust lands. Id. at 17. And as to the 2021 redistricting plan, he testified about the Tribes sharing community interests and that the Tribes did not request the subdistricts as enacted in district 9. Id. at 19.
Chairman Azure also spoke at length about the 2018 election. Id. at 20. He discussed the record voter turnout that year because of concerns over a voter identification law. He noted there was “a lot of attention” and many national resources were directed at the Tribes. Id. He also said he had never seen that level of Native American voter engagement in his life and has not seen it since. Id. at 21.
The first witness for the Secretary was expert witness Dr. M.V. Hood, III. He is a political science professor at the University of Georgia and director of the School of Public and International Affairs Survey Research Center. Doc. 117 at 72. Dr. Hood is an expert on American politics, election administration, southern politics, racial politics, and Senate electoral politics. Id. at 75:12-76:7.
Dr. Hood's expert testimony was extensive. He reviewed his expert report (Pl. Ex. 81) and opined on each of the three Gingles preconditions. Doc. 117 at 72:2-182:20. Notably, he testified that he agreed that the first precondition was met but questioned whether there was enough data to prove the second precondition. Id. at 89.
On the third precondition (white bloc voting), he reached a different result than Dr. Collingwood. Id. He analyzed the same elections as Dr. Collingwood (Doc. 117 at 83:14-18), though he statistically weighed the elections differently, and concluded that white bloc voting was not present in district 9 at-large and as-enacted. Id. at 86. He stated that “Gingles 3 is not met because the Native American candidate of choice is not typically being defeated by the majority white voting bloc.” Id. at 89. Dr. Hood also testified that he did not review the 2022 election results. Id. at 162.
As to the 2018 election, Dr. Hood testified that the Native American turnout in 2018 was historically high and that the results should not necessarily be excluded from a performance analysis. Dr. Hood testified that those 2018 results “prove[] that Native American turnout can be that high” and that if “[i]t was that high in 2018,” it could be that high again. Id. at 86:7-15.
Erika White, the North Dakota Election Director, testified next. She spoke about the role of the Secretary in North Dakota elections and the processes and deadlines that are imposed on state elections by statute. Doc. 117 at 192. She testified too about the redistricting process.
The Secretary's final witness was Brian Nybakken, the Elections Systems Administration Manager in the Secretary's Elections Office. Doc. 118 at 6-33. He testified about the elections systems in place in North Dakota, auditor training, voter identification requirements, and certain election issues pertaining to Native Americans in North Dakota. Id.
II. CONCLUSIONS OF LAW
Section 2 of the VRA prohibits any “standard, practice, or procedure” that “results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color[.]” 52 U.S.C. § 10301(a). A violation of Section 2 is established if it is shown that “the political processes leading to [a] nomination or election” in the jurisdiction “are not equally open to participation by [minority voters] 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.” Id. § 10301(b). “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 minority and white voters to elect their preferred candidates.” Bone Shirt v. Hazeltine, 461 F.3d 1011, 1017-18 (8th Cir. 2006) (cleaned up).
Section 2 prohibits “the distribution of minority voters into districts in a way that dilutes their voting power.” Wis. Legislature v. Wis. Elections Comm'n, 142 S.Ct. 1245, 1248 (2022) (citing Gingles, 478 U.S. at 46). Recall that, under Gingles, three preconditions must be initially satisfied to proceed with a Section 2 voter dilution claim:
1. The minority group . . . is sufficiently large and geographically compact to constitute a majority in a single-member district;
2. The minority group . . . is politically cohesive; and,
3. The white majority votes sufficiently as a bloc to enable it-in the absence of special circumstances . . . usually to defeat the minority's preferred candidate.478 U.S. at 50-51. Failure to prove any of the three preconditions defeats a Section 2 claim. Clay v. Bd. of Educ., 90 F.3d 1357, 1362 (8th Cir. 1996).
If all preconditions are met, then there is a viable voter dilution claim, and the analysis shifts to determining whether, under the totality of the circumstances, members of the racial minority group have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. 52 U.S.C. § 10301(b); see also Johnson v. De Grandy, 512 U.S. 997, 1011-12 (1994) (once the three preconditions are met, the totality of the circumstances is addressed). To assess the totality of the circumstances, the Court considers the factors identified in the Senate Judiciary Committee Majority Report accompanying the bill that amended Section 2 (also known as the “Senate Factors”). S. Rep., at 28-29, U.S. Code Cong. & Admin. News 1982, pp. 206-207; Gingles, 478 U.S. at 36. Two other factors are also relevant: (1) was there a significant lack of response from elected officials to the needs of the minority group, and (2) was the policy underlying the jurisdiction's use of the current boundaries tenuous. Gingles, 478 U.S. at 44; Bone Shirt, 461 F.3d at 1022.
The Senate Report stresses that these factors are “neither comprehensive nor exclusive.” Gingles, 478 U.S. at 45. The extent to which voting is racially polarized (Senate Factor 2) and the extent to which minorities have been elected under the challenged scheme (Senate Factor 7) predominate the analysis. Missouri State Conf. of the Nat'l Ass'n for the Advancement of Colored People v. Ferguson-Florissant Sch. Dist., 894 F.3d 924, 938 (8th Cir. 2018); Bone Shirt, 461 F.3d at 1022; Cottier v. City of Martin, 551 F.3d 733, 740 (8th Cir. 2008); Harvell v. Blytheville Sch. Dist. No. 5, 71 F.3d 1382, 1390 (8th Cir. 1995).
A. The Gingles Preconditions
1. Gingles 1: Sufficiently Large and Geographically Compact
The first Gingles precondition requires a Section 2 plaintiff to demonstrate that the minority group (here, Native Americans) is sufficiently large and geographically compact to constitute a majority in a potential district. Gingles, 478 U.S. at 50. This is also known as the “majority-minority standard.” Jeffers v. Beebe, 895 F.Supp.2d 920, 931 (E.D. Ark. 2012). As explained in Gingles, “unless minority voters possess the potential to elect representatives in the absence of the challenged structure or practice, they cannot claim to have been injured by that structure or practice.” Gingles, 478 U.S. at 50. So, this precondition focuses on electoral potential-and specifically here, whether Native American voters have the potential to constitute the majority of voters “in some reasonably configured legislative district.” See Cooper v. Harris, 581 U.S. 285, 301 (2017); see also Houston v. Lafayette Cnty., Miss., 56 F.3d 606, 611 (5th Cir. 1995). Hence the analysis for the first precondition considers the proposed district(s) and not the existing district. See, e.g., Bone Shirt, 461 F.3d at 1018.
While the first precondition refers to a minority constituting a majority in a “single-member district,” the analysis is done on a case-by-case basis, and the Gingles factors “cannot be applied mechanically and without regard to the nature of the claim.” See Voinovich v. Quilter, 507 U.S. 146, 158 (1993).
As an initial matter, the Secretary argues the first precondition is not met because district 9, as-enacted, better reflects traditional redistricting criteria than the Tribes' proposed districts. He also asserts that the first precondition is not met as to district 15. But a Section 2 claim is not a competition between which version of district 9 better respects traditional redistricting criteria. See Allen v. Milligan, 143 S.Ct. 1487, 1505 (2023) (noting Gingles 1 is not a “beauty contest” between plaintiffs' maps and the state's districts). The claim is not defeated simply because the challenged plan performs better on certain traditional redistricting criteria than the proposed plan. Id. (finding that plaintiffs' demonstrative plans were reasonably configured, even where the enacted plan arguably performed better on certain traditional redistricting criteria than the demonstrative plans).
With that issue resolved, the question is whether Native American voters have the potential to constitute the majority of voters in some reasonably configured legislative district. The parties agree that Native American voters have the potential to constitute the majority of voters in both proposed versions of district 9. The NVAP in the Tribes' first proposed plan is 66.1%. Doc. 15 at 134:22-135:2, 135:14-17, 166:1-3. The NVAP in the Tribes' second proposed plan is 69.1%. Id. So, the remaining issue is whether these proposed districts are “reasonably configured.” See Johnson v. De Grandy, 512 U.S. 997, 1008 (1994).
De Grandy articulated this standard in the context of single-member districts. Here, given the comparison of subdistricts to multimember districts, it is more useful to consider the number of representatives that Native American voters have an opportunity to elect.
A district is reasonably configured “if it comports with traditional districting criteria, such as being contiguous and reasonably compact.” Milligan, 143 S.Ct. at 1503. Courts may also consider other traditional redistricting criteria, including respect for political boundaries and keeping together communities of interest. Id. at 1505 (considering respect for political subdivisions and communities of interest as traditional redistricting criteria); Alabama Legislative Black Caucus v. Alabama, 575 U.S. 254, 259 (2015) (citing compactness and not splitting counties or precincts as examples of traditional redistricting criteria, amongst others).
The evidence at trial shows that the Tribes' proposed plans comport with traditional redistricting principles, including compactness, contiguity, respect for political boundaries, and keeping together communities of interest. First, as to contiguity and compactness, the proposed districts are made up of a contiguous land base (Pl. Exs. 105, 106) and contain no obvious irregularities as to compactness. Indeed, the evidence at trial demonstrated that the proposed districts did not appear more oddly shaped than other districts, and both proposed districts are reasonably compact. See Doc. 115 at 139:17-23, 141:4-8; Pl. Ex. 1 at 32, 39. The proposed plans are also comparatively compact when viewed against other districts in the 2021 redistricting plan. Pl. Ex. 1 at 32, 39. Statistically too, Dr. Collingwood testified the compactness scores of the proposed districts are within the range of compactness scores for other districts in the 2021 redistricting plan. See Doc. 115 at 139:17-140:5, 141:24-143:20; Pl. Ex. 1 at 32, 39; Pl. Ex. 42 at 9-11; Pl. Ex. 126, 128, and 129.
The Tribes' proposed plans also respect existing political boundaries, including Reservation boundaries, and keep together communities of interest. As to political boundaries, the proposed plans keep together the Turtle Mountain Reservation and its trust lands. Pl. Exs. 105, 106. The plans similarly preserve and keep together two communities of interest. Several witnesses testified that the Tribes represent a community of interest because of their geographic proximity and their members shared representational interests, socioeconomic statuses, and cultural values. Doc. 115 at 50:24-52:11, 52:24-73:9; Doc. 117 at 22:4-16-27:15, 28:18-25; 50:3-7; 52:23-53:1, 55:9-12. Chairman Azure and former Chairman Yankton persuasively testified to all those shared interests. Id. As to representational interests, the Tribes often collaborate to lobby the Legislative Assembly on their shared issues, including gaming, law enforcement, child welfare, taxation, and road maintenance, among others. See Doc. 115 at 56:12-61:18, 64:1-70:6; Doc. 116 at 21:11-21; Doc. 117 at 25:23-28:8. The residents on the Tribes' Reservations also have similar socioeconomic and education levels-levels that differ from the white residents in neighboring counties. Pl. Ex.73 at 513; Doc. 116 at 156:17-159:8; 161:13-161:24. Residents of the Tribes also participate in similar cultural practices and events and share cultural values. See Doc. 117 at 18:14-19:13.
All this evidence shows that the Tribes' proposed plans comport with traditional redistricting principles, including compactness, contiguity, respect for political boundaries, and keeping together communities of interest. The proposed plans demonstrate that Native American voters have the potential to constitute the majority of voters in some reasonably configured legislative district. And as a result, the Tribes have proven by a preponderance of the evidence that the first Gingles precondition is satisfied.
The Secretary expresses concern that the districts under the Tribes' proposed plans would be illegal racial gerrymanders. But even assuming race was the predominate motivating factor in drawing the districts, establishing (and then remedying) a Section 2 violation provides a compelling justification for adopting one of the proposed plans. See Cooper, 581 U.S. at 292.
2. Gingles 2: Racially Polarized Voting and Political Cohesion
“The second Gingles precondition requires a showing that the Native American minority is politically cohesive.” Bone Shirt, 461 F.3d at 1020. “Proving this factor typically requires a statistical and non-statistical evaluation of the relevant elections.” Id. (citing Cottier, 445 F.3d at 1118). “Evidence of political cohesiveness is shown by minority voting preferences, distinct from the majority, demonstrated in actual elections, and can be established with the same evidence plaintiffs must offer to establish racially polarized voting, because political cohesiveness is implicit in racially polarized voting.” Id.
The parties and their experts agree that voting in districts 9 and 15 (when voting at large) is racially polarized, with Native American voters cohesively supporting the same candidates. Doc. 108 at 6. Based on the evidence at trial, voting in subdistricts 9A and 9B is also racially polarized, with Native American voters cohesively supporting the same candidates. Pl. Ex. 13, 14; Doc. 115 at 145:23-146:2. Although subdistricts 9A and 9B do not contain enough precincts for a full statistical analysis, subdistrict 9A has an NVAP of 68.5%. Pl. Ex. 1 at 15. That, combined with the undisputed political cohesiveness of district 9 at-large, demonstrates that voters in subdistrict 9A are politically cohesive. Pl. Ex. 1 at 15; Doc. 115 at 149:7-150:25.
Dr. Hood agreed that Native American voters are politically cohesive in subdistricts 9A and 9B. Pl. Ex. 80 at 4-6; Doc. 117 at 139:19-140:16. He testified that his conclusion assumed that the vote distribution within in each subdistrict “mirrors the overall district.” Doc. 117 at 140:1-16. Testimony from Chairman Azure and former Chairman Yankton confirms the statistical data. Both testified that voters living on the Turtle Mountain Reservation and Spirit Lake Reservation vote similarly. Doc. 116 at 16:5-19:19, 28:14-25; Doc. 115 at 52:12-53:25.
The statistical evidence, combined with the lay witness testimony, shows that the Native American minority is politically cohesive. The Tribes have proven by a preponderance of the evidence that the second Gingles precondition is met.
3. Gingles 3: White Bloc Voting
With the first and second preconditions met, the analysis turns to the third precondition, which is the chief point of disagreement between the Tribes and the Secretary. The third Gingles precondition “asks whether the white majority typically votes in a bloc to defeat the minority candidate.” Bone Shirt, 461 F.3d at 1020. “This is determined through three inquiries: (1) identifying the minority-preferred candidates; (2) determining whether the white majority votes as a bloc to defeat the minority preferred candidate, and (3) determining whether there were special circumstances . . . present when minority-preferred candidates won.” Id. (cleaned up).
Not all elections are equally relevant in assessing white bloc voting. “Endogenous and interracial elections are the best indicators of whether the white majority usually defeats the minority candidate.” Id. “Although they are not as probative as endogenous elections, exogenous elections hold some probative value.” Id. In addition, “[t]he more recent an election, the higher its probative value.” Id.There is no requirement that a particular number of elections be analyzed in determining whether white bloc voting usually defeats minority-preferred candidates. Gingles, 478 U.S. at 57 n.25. “The number of elections that must be studied in order to determine whether voting is polarized will vary according to pertinent circumstances.” Id.
An endogenous election is an election where a district (or subdistrict) is electing a direct representative for that district (or subdistrict), as opposed to an exogenous election, which in this case, are statewide elections.
In assessing the third precondition, courts look to the districts in which it is alleged that Native American preferred candidates are prevented from winning, not on neighboring “packed” districts. Bone Shirt, 461 F.3d at 1027 (Gruender, J., concurring) (“If the State's approach were correct, packing would be both the problem and the solution-i.e., having illegally packed Indians into one district, the State could then point out that Indians are sometimes able to elect their preferred candidate in the packed district”); De Grandy, 512 U.S. at 1003-04 (focusing on whether white voters vote as a bloc “to bar minority groups from electing their chosen candidates except in a district where a given minority makes up the voting majority”). Finally, courts must also consider whether “special circumstances . . . may explain minority electoral success in a polarized contest.” Gingles, 478 U.S. at 57 & n.26. Special circumstances must be considered if “the election was not representative of the typical way in which the electoral process functions.” Ruiz v. City of Santa Maria, 160 F.3d 543, 557 (9th Cir. 1998).
i. Subdistrict 9B
Starting with subdistrict 9B, the parties agree that a white bloc voting usually defeats Native American preferred candidates in subdistrict 9B when the three most probative election types are considered. And the evidence at trial supports that conclusion.
Because the challenged plan that created the subdistrict was enacted in 2021, the only endogenous election data available is from the 2022 election. Nonetheless, the data is highly probative. One of two state legislative elections in subdistrict 9B's boundaries was the district 9 at-large Senate election, which featured a Native American candidate,[ who lost:
In all tables below, the Native American preferred candidates are marked with an asterisk.
| Election | Result | Native American Candidate Win or Lose |
| 2022 State Senate District 9 | Weston: 63.0% Marcellais*: 36.8% | Lose |
Pl. Ex. 1 at 21. The other endogenous election in subdistrict 9B featured two white candidates. The Native American preferred candidate, incumbent Marvin Nelson, also lost:
| Election | Result | Native American Candidate Win or Lose |
| 2022 State House District 9B | Henderson: 56.5% Nelson*: 37.6% | Lose |
Id. Beyond the 2022 endogenous election data, there are four exogenous (or statewide) elections since 2016 that featured Native American candidates that voters in precincts within the boundaries of now-subdistrict 9B voted in.[ In each of those contests, the Native American candidate lost:
To account for the lack of subdistrict specific election data, this data is generated from collecting precinct data from those precincts now in subdistrict 9B.
| Election | Result | Native American Candidate Win or Lose |
| 2022 Public Service Commissioner | Fedorchak: 64.4% Moniz*: 35.3% | Lose |
| 2016 Insurance Commissioner | Godfread: 58.4% Buffalo*: 41.6% | Lose |
| 2016 Public Service Commissioner | Fedorchak: 60.2% Hunte-Beaubrun*: 32.4% | Lose |
| 2016 U.S. House | Cramer: 62.2% Iron Eyes*: 32.9% | Lose |
Id. at 17-20.
The next set of data focuses on the most recent three election cycles, where special circumstances were not present-here, the 2022, 2020, and 2016 elections.[ Per the table below, the defeat rate of the Native American preferred candidates was 100% for every election cycle:
As discussed in detail below, the 2018 election involved special circumstances that made it atypical.
| Election | Result | Native American Preferred Candidate Win or Lose | Defeat Rate for Native American Preferred Candidates |
| 2022 Agricultural Commissioner | Goehring: 70.9% Dooley*: 28.9% | Lose | 2022 Defeat Rate: 100% |
| 2022 Attorney General | Wrigley: 65.6% Lamb*: 34.3% | Lose | |
| 2022 Public Service Commissioner (4 Year) | Haugen Hoffart: 65.4% Hammer*: 34.3% | Lose | |
| 2022 Secretary of State | Howe: 57.1% Powell*: 33.7% | Lose | |
| 2022 U.S. House | Armstrong: 61.4% Mund*: 38.4% | Lose | |
| 2022 U.S. Senate | Hoeven: 60.6% Christiansen*: 27.5% | Lose | |
| 2020 Auditor | Gallion: 59.8% Hart*: 40.1% | Lose | 2022 + 2020 Defeat Rate: 100% |
| 2020 Governor | Burgum: 65.3% Lenz*: 29.8% | Lose | |
| 2020 President | Trump: 60.8% Biden*: 37.0% | Lose |
| 2020 Public Service Commissioner | Kroshus: 60.4% Buchmann*: 39.8% | Lose | 2022 + 2020 Defeat Rate: 100% |
| 2020 Treasurer | Beadle: 58.6% Haugen*: 41.2% | Lose | |
| 2020 U.S. House | Armstrong: 64.4% Raknerud*: 33.4% | Lose | |
| 2016 Governor | Burgum: 61.7% Nelson*: 35.8% | Lose | 2022 + 2020 + 2016 Defeat Rate: 100% |
| 2016 President | Trump: 56.6% Clinton*: 33.8% | Lose | |
| 2016 Treasurer | Schmidt: 53.6% Mathern*: 39.8% | Lose | |
| 2016 U.S. Senate | Hoeven: 72.9% Glassheim*: 22.1% | Lose |
Pl. Ex. 1 at 17-20. This evidence establishes that white bloc voting usually-and always in the most probative elections-defeats the Native American preferred candidates in subdistrict 9B. As a result, the third precondition is met as to subdistrict 9B.
ii. District 15
The parties also agree that the same conclusion follows as to district 15. Again, the only endogenous election is the 2022 state legislative election, where two Native-American preferred candidates appeared on the ballot. Both were defeated:
| Election | Result | Native American Candidate Win or Lose |
| 2022 State Senate District 15 | Estenson: 65.5% Brown*: 33.8% | Lose |
| 2022 State House District 15 | Frelich: 41.6% Johnson: 38.6% Lawrence-Skadsem*: 19.7% | Lose |
Pl. Ex. 1 at 27. There have been no endogenous all-white elections in district 15. Four exogenous elections since 2016 have featured Native American candidates within the boundaries of district 15. In each of those contests-100% of the time-the Native American candidate lost:
| Election | Result | Native American Candidate Win or Lose | |
| 2022 Public Service Commissioner | Fedorchak: 69.3% Moniz*: 30.6% | Lose |
| Election | Result | Native American Candidate Win or Lose |
| 2016 Insurance Commissioner | Godfread: 64.6% Buffalo*: 35.4% | Lose |
| 2016 Public Service Commissioner | Fedorchak: 63.8% Hunte-Beaubrun*: 27.6% | Lose |
| 2016 U.S. House | Cramer: 65.5% Iron Eyes*: 27.9% | Lose |
Pl. Ex. 1 at 17-20. As shown below, Native American preferred candidates have lost every exogenous all-white election in the record:
| Election | Result | Native American Preferred Candidate Win or Lose | Defeat Rate for Native American Preferred Candidates |
| 2022 Agricultural Commissioner | Goehring: 75.0% Dooley*: 24.9% | Lose | 2022 Defeat Rate: 100% |
| 2022 Attorney General | Wrigley: 70.9% Lamb*: 29.0% | Lose | |
| 2022 Public Service Commissioner | Fedorchak: 69.3% Moniz*: 30.6% | Lose | |
| 2022 Public Service Commissioner (4 Year) | Haugen Hoffart: 70.4% Hammer*: 29.4% | Lose | |
| 2022 Secretary of State | Howe: 61.2% Powell*: 27.8% | Lose | |
| 2022 U.S. House | Armstrong: 62.8% Mund*: 37.1% | Lose | |
| 2022 U.S. Senate | Hoeven: 58.5% Christiansen*: 24.8% | Lose | |
| 2020 Auditor | Gallion: 65.4% Hart*: 34.5% | Lose | 2022 + 2020 Defeat Rate: 100% |
| 2020 Governor | Burgum: 67.6% Lenz*: 25.8% | Lose | |
| 2020 President | Trump: 64.3% Biden*: 33.0% | Lose |
| 2020 Public Service Commissioner | Kroshus: 64.1% Buchmann*: 35.7% | Lose | 2022 + 2020 Defeat Rate: 100% |
| 2020 Treasurer | Beadle: 63.2% Haugen*: 36.3% | Lose | |
| 2020 U.S. House | Armstrong: 68.7% Raknerud*: 28.1% | Lose | |
| 2016 Governor | Burgum: 71.1% Nelson*: 24.8% | Lose | 2022 + 2020 + 2016 Defeat Rate: 100% |
| 2016 President | Trump: 57.6% Clinton*: 31.2% | Lose | |
| 2016 Treasurer | Schmidt: 59.5% Mathern*: 31.8% | Lose | |
| 2016 U.S. Senate | Hoeven: 75.7% Glassheim*: 18.5% | Lose |
Pl. Ex. 1 at 27-30.
Again, like subdistrict 9B, all this evidence establishes that white bloc voting usually-and always in the most probative elections-defeats the Native American preferred candidates in district 15. As a result, the third precondition is met as to district 15.
iii. District 9
District 9 at-large presents a much closer call and is the central point of disagreement between the parties. The Secretary disputes whether the white vote bloc usually defeats the Native American preferred candidate in (as-enacted and at-large) district 9. But based on the evidence at trial, the Tribes proved by a preponderance of the evidence that a white bloc voting does usually defeat Native American preferred candidates in the as-enacted and at-large district 9.
Without question, and consistent with case law, the most probative election in district 9 at-large is the 2022 Senate election. The election featured each of the three factors that makes an election more probative-(1) it is an endogenous election, (2) it featured a Native American candidate, and (3) it is part of the most recent election cycle. Native American incumbent Senator Marcellais lost his bid for reelection despite Native American voters casting roughly 80% of their ballots for him. Pl. Ex. 1 at 15; see Bone Shirt, 461 F.3d at 1021 (affirming finding that Gingles 3 was satisfied where “[i]n the only mixed-race endogenous election . . . the Indian-preferred candidate for state senate lost even though he received 70 percent of the Native-American vote”). As the 2022 election data shows, Senator Marcellais, the Native American candidate, was defeated by his opponent, the candidate of choice of white voters in the district:
| Election | Result | Native American Candidate Win or Lose |
| 2022 State Senate District 9 | Weston: 53.7% Marcellais*: 46.1% | Lose |
Pl. Ex. 1 at 17. Moving to the statewide exogenous elections since 2016, four have featured Native American candidates within the current boundaries of district 9. In those elections, the Native American candidate lost half of the elections:
| Election | Result | Native American Candidate Win or Lose |
| 2022 Public Service Commissioner | Fedorchak: 54.1% Moniz*: 45.7% | Lose |
| 2016 Public Service Commissioner | Fedorchak: 46.5% Hunte-Beaubrun*: 46.1% | Lose |
| 2016 Insurance Commissioner | Godfread: 43.2% Buffalo*: 56.8% | Win |
| 2016 U.S. House | Cramer: 46.9% Iron Eyes*: 49.3% | Win |
Pl. Ex. 1 at 17-20. When all contests featuring Native American candidates (whether endogenous or exogenous) are taken together, the defeat rate for Native American candidates is 60%.
Among exogenous all-white elections, Native American preferred candidates lost 100% of the 2022 elections, 67% of the 2022 and 2020 elections combined, and 56% of the 2022, 2020, and 2016 elections combined:
| Election | Result | Native American Preferred Candidate Win or Lose | Defeat Rate for Native American Preferred Candidates |
| 2022 Agricultural Commissioner | Goehring: 60.2% Dooley*: 39.6% | Lose | 2022 Defeat Rate: 100% |
| 2022 Attorney General | Wrigley: 55.3% Lamb*: 44.6% | Lose | |
| 2022 Public Service Commissioner (4 Year) | Haugen Hoffart: 55.2% Hammer*: 44.6% | Lose | |
| 2022 Secretary of State | Howe: 47.5% Powell*: 42.3% | Lose | |
| 2022 U.S. House | Armstrong: 52.8% Mund*: 47.0% | Lose | |
| 2022 U.S. Senate | Hoeven: 51.3% Christiansen*: 36.4% | Lose | |
| 2020 Auditor | Gallion: 46.5% Hart*: 53.4% | Win | 2020 Defeat Rate: 33% |
| 2020 Governor | Burgum: 52.8% Lenz*: 43.1% | Lose | |
| 2020 President | Trump: 47.2% Biden*: 50.8% | Win | |
| 2020 Public Service Commissioner | Kroshus: 46.4% Buchmann*: 53.4% | Win | |
| 2020 Treasurer | Beadle: 45.6% Haugen*: 54.2% | Win | |
| 2020 U.S. House | Armstrong: 50.6% Raknerud*: 47.0% | Lose | |
| 2016 Governor | Burgum: 48.3% Nelson*: 48.7% | Win | 2016 Defeat Rate: 25% |
| 2016 President | Trump: 44.2% Clinton*: 45.1% | Win | |
| 2016 Treasurer | Schmidt: 41.6% Mathern*: 50.0% | Win | |
| 2016 U.S. Senate | Hoeven: 59.7% Glassheim*: 33.9% | Lose |
Pl. Ex. 1 at 17-20. From this data, a pattern emerges: the more recent the election, the more likely the Native American preferred candidate is to lose. When averaged together, the total defeat rate is 56%. Beyond that, even when the 2018 election results (which, as explained below, was an atypical election) are factored in, the 100% defeat rate for Native American candidates of choice in the most recent election is highly probative and compelling evidence of white bloc voting. Said another way, giving each election the appropriate weight per Eighth Circuit and Supreme Court case law, the evidence proves by a preponderance that Native American candidates of choice will not be successful over 50% of the time in as-enacted and at-large district 9.
iv. 2018 Election and Special Circumstances
One of the key differences of opinion between Dr. Collingwood and Dr. Hood concerns the probative value and weight of the 2018 election. “Only minority electoral success in typical elections is relevant to whether a Section 2 majority voting bloc usually defeats the minority's preferred candidate.” Ruiz, 160 F.3d at 557. So, a central issue is whether 2018 was a typical election, deserving equal weight as other elections, or whether it was an atypical election, deserving less weight than other elections. The Secretary argues that 2018 is a typical election deserving equal weight; the Tribes assert that the 2018 election was atypical and deserves less weight.
In 2018, a North Dakota voter identification law was upheld that required a residential address to vote. The voter identification requirement affected the number of Native Americans eligible to vote and resulted in significant national and regional attention to Native American voters and increasing voter turnout. Voter turnout did increase dramatically, as compared to years prior and since:
ElectionWhite Electorate ShareNative American Electorate Share
| Election | White Electorate Share | Native American Electorate Share |
| 2014 | 67% | 33% |
| 2016 | 63% | 37% |
| 2018 | 50% | 50% |
| 2020 | 63% | 37% |
| 2022 | 60% | 40% |
Pl. Ex. 42 at 4-5. Because of the increase in Native American voter turnout, Native American preferred candidates also performed much better than in any other years, prior or since. Pl. Ex. 1 at 18.
Chairman Azure and former Chairman Yankton persuasively testified about the extraordinary resources that poured into North Dakota's Native American reservations in the lead up to the 2018 election. Doc. 115 at 80:18-86:17; Doc. 117 at 21:8-12. The voter identification law caused a backlash among Native American voters, which was aided by substantial financial resources promoting get-out-the-vote efforts on the reservations. Id. National celebrities gave concerts and performances on the reservations to promote turnout. Id. Both testified that the resources-and resulting turnout among Native American voters-was unlike anything they have seen before or since. Id.
That testimony is supported by the data. Native American turnout in 2018 was unusually high. Not only did it exceed statewide turnout and approach white turnout in district 9, but it inverted the normal pattern of lower turnout in midterm versus presidential elections:
(Image Omitted)
Pl. Ex. 43.
With those facts in mind, the experts offer competing opinions on the probative value of the 2018 election. Dr. Hood concluded that the third precondition was not met in as-enacted and at-large district 9 because Native American preferred candidates were successful in over 50% of the elections he reviewed. To reach that conclusion and opinion, Dr. Hood reviewed the election data from Dr. Collinwood's report and added together the elections in at-large district 9 and subdistrict 9A and 9B. Pl. Ex. 81 at 4. He also included the election data from the 2018 election. Doc. 117 at 143. In other words, Dr. Hood considered all election data equally and gave no probative weight or value to any one election. Doc. 117 at 85:19-86:6. Also, and importantly, Dr. Hood did not consider the 2022 election results. Id. at 150.
Dr. Collingwood reached a different conclusion. He concluded the 2018 election presented special circumstances, including unprecedented voter turnout, that “warrant and counsel against mechanically interpreting” the results. Pl. Ex. 1 at 18. As a result, he gave the 2018 election less weight when calculating white bloc voting in district 9. He also did consider the 2022 election, weighed that election more heavily, and concluded that the Native American preferred candidate “lost every single contest.” Pl. Ex. 1 at 21. Dr. Collingwood opined that the third precondition is met because “white voters are voting as a bloc to prevent Native Americans from electing candidates of choice in recent elections, in endogenous elections . ., and in 60% of contests across all tested years in which the Native American preferred candidate was a Native American.” Pl. Ex. 1 at 43.
Having heard the testimony by both experts at trial, along with having reviewed their respective reports, Dr. Collingwood's conclusions and analysis are more credible because they follow the general directives of the Eighth Circuit in weighing elections in VRA cases. Indeed, the Eighth Circuit has recognized that endogenous elections should be considered more probative than exogenous elections; elections with a Native American candidate are more probative than elections that do not feature a Native American candidate; and that more recent elections have more probative value than less recent elections. Bone Shirt, 461 F.3d at 1020-21. Dr. Hood gave all elections equal probative value and generally weighed all elections the same. But Dr. Collingwood's report and methodology more closely tracks the instruction from the Eighth Circuit in weighing election data in VRA cases, making it more credible and reliable. In addition, Dr. Hood's testimony at trial acknowledged that endogenous elections, elections featuring Native American candidates, and more recent elections are more probative. Doc. 117 at 142:9-143:7. He also testified that the 2022 endogenous election for the district 9 Senate seat was the “single most probative” election because it featured all three probative characteristics (id. at 143:12-17), but he did not consider the 2022 endogenous election in reaching his conclusions (id. at 150).
Substantively and statistically, Dr. Hood's conclusion on the third precondition rests on adding together all data from district 9 and subdistricts 9A and 9B. But recall that subdistrict 9A has a near 80% NVAP, and Native American preferred candidates win 100% of the time. A district with a packed minority population is not one where the defeat of minority preferred candidates is to be expected, and it should not be considered as part of the third Gingles precondition. See Bone Shirt, 461 F.3d at 1027. And importantly, as Dr. Hood testified and acknowledged at trial, if subdistrict 9A was removed from his analysis, the Native American preferred candidates defeat rate is 59.5%. Doc. 117 at 148:16-24. That alone also satisfies the third Gingles precondition.
Having reviewed the testimony and evidence, giving the elections the appropriate weight consistent with Eighth Circuit case law, the Tribes have proven by a preponderance of the evidence that the white majority typically votes in a bloc to defeat the minority candidate in as-enacted and at-large district 9. As such, the third Gingles precondition is also established as to as-enacted and at-large district 9.
B. Totality of the Circumstances and the Senate Factors
With the Gingles preconditions met, the Section 2 analysis turns to the totality of the circumstances and analysis of the Senate Factors. The Senate Factors come from the Senate Committee report to the 1982 amendment to the VRA and directs courts to consider the following factors in determining whether the totality of the circumstances indicate a Section 2 violation:
(1) the extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process;
(2) the extent to which voting in the elections of the state or political subdivision is racially polarized;
(3) the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;
(4) if there is a candidate slating process, whether the members of the minority group have been denied access to that process;
(5) 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;
(6) whether political campaigns have been characterized by overt or subtle racial appeals;
(7) the extent to which members of the minority group have been elected to public office in the jurisdiction.S.R. No. 97-417 at 28-29 (1982); Gingles, 478 U.S. at 44-45. Two additional factors are also probative in determining a Section 2 violation: (1) was there a significant lack of response from elected officials to the needs of the minority group; and (2) was the policy underlying the jurisdiction's use of the current boundaries tenuous. Gingles, 478 U.S. at 44. “[T]his list of typical factors is neither comprehensive nor exclusive. While the enumerated factors will often be pertinent to certain types of § 2 violations, particularly to vote dilution claims, other factors may also be relevant and may be considered. Furthermore, . . . there is no requirement that any particular number of factors be proved, or that a majority of them point one way or the other.” Id.at 45 (internal citations omitted).
1. Senate Factors 2 and 7
“Two factors predominate the totality-of-circumstances analysis: the extent to which voting is racially polarized and the extent to which minorities have been elected under the challenged scheme.” Bone Shirt, 461 F.3d at 1022. As to Senate Factor 2, the extent of racially polarized voting, the record reflects a high level of racially polarized voting in districts 9 and 15 and subdistricts 9A and 9B. That evidence is largely undisputed and was discussed at length above. As to Senate Factor 7-the extent to which Native Americans have been elected-the only election under the 2021 redistricting plan in 2022 resulted in the loss of a Native American Senator (who had held the seat since 2006). Brown, a Native American, also lost the district 15 race. In effect, as a result of the 2021 redistricting plan, Native Americans experienced a net-loss of representation. Both factors weigh the totality of the circumstances towards a Section 2 violation.
2. Remaining Senate Factors
This leaves factors one, three,[ and five,[ along with tenuousness, lack of response, and proportionality. As to the first Senate Factor, which considers historical discrimination practices, the Tribes offered expert testimony from Dr. Daniel McCool. He testified as to the long history of mistreatment of Native Americans in North Dakota and discussed evidence of contemporary discrimination against Native Americans, including many successful voting discrimination claims affecting Native Americans. Doc. 116 at 90-126. The evidence of discrimination in the democratic and political process against Native Americans in North Dakota is well-documented and undisputed by the Secretary. So, the first Senate Factor 1 weighs toward a Section 2 violation.
Senate Factor 4, which addresses candidate slating processes, is not applicable on these facts.
The parties agree that Senate Factor 6 is not at issue.
Next, as to the third Senate Factor, which considers discrimination through voting practices and procedures, the Tribes suggest that the 2021 redistricting plan itself is the best evidence of voting practices or procedures that enhance the opportunity for discrimination. But beyond that blanket assertion, there is no evidence that the Secretary used the 2021 redistricting plan to enhance the opportunity of discrimination against Native Americans. As a result, the third Senate Factor does not weigh toward finding Section 2 violation.
Senate Factor 5 considers the effects of discrimination against Native Americans more broadly, in such areas as education, employment, and health care. Dr. Weston McCool offered undisputed evidence as to the lower socio-economic status of Native Americans in North Dakota and that Native Americans continue to experience the effects of discrimination across a host of socioeconomic measures, which results in inequal access to the political process. Doc. 116 at 148. And the Secretary did not challenge that evidence. Senate Factor 5 weighs toward a Section 2 violation.
The three remaining factors in the totality of the circumstances analysis are tenuousness, lack of response, and proportionality. Tenuousness looks at the justification and explanation for the policy or law at issue. “The tenuousness of the justification for the state policy may indicate that the policy is unfair.” Cottier v. City of Martin, 466 F.Supp.2d 1175, 1197 (D.S.D. 2006).
While the actions of the Legislative Assembly may not have ultimately went far enough to comply with Section 2 of the VRA, the record establishes that the Secretary and the Legislative Assembly were intensely concerned with complying with the VRA in passing the 2021 redistricting plan and creating the districts and subdistricts at issue. The justification by the Secretary for the 2021 redistricting plan is not tenuous, and this factor does not weigh in favor of a Section 2 violation.
The next factor is lack of response. The Tribes generally assert the Legislative Assembly was unresponsive to the needs of the Native American community. But the Secretary presented ample evidence of Tribal representatives and members generally advocating for subdistricts. Doc. 116 at 28, 32-33, 33-34, 134, 141. Again, the record is clear that the Legislative Assembly sought input from the Tribes and their members and attempted to work with the Tribes to comply with the VRA, even though the VRA compliance measures fell short. Also recall that the redistricting plan was developed under a truncated timeline because of the COVID-19 pandemic. On these facts, one cannot find a lack of response by the Secretary and the Legislative Assembly, and as a result, this factor does not weigh in favor of a Section 2 violation.
The final factor is proportionality. Based on their share of statewide VAP, Native Americans should hold three Senate seats and six House seats. However, under the 2021 redistricting plan, Native Americans hold zero seats in the Senate and two House seats. Either of the proposed plans would yield one Senate seat and three House seats. While certainly not dispositive, this obvious disparity as to proportionality is further evidence of vote dilution under the totality of circumstances.
All told, while a closer decision than suggested by the Tribes, the two most critical Senate Factors (2 and 7) weigh heavily towards finding a Section 2 violation. Those factors, together with the evidence on Senate Factors 1, 5, and proportionality, demonstrates that the totality of the circumstances deprive Native American voters of an equal opportunity to participate in the political process and to elect representatives of their choice, in violation of Section 2 of the VRA.
III. CONCLUSION AND ORDER
“Determining whether a Section 2 violation exists is a complex, fact-intensive task that requires inquiry into sensitive and often difficult subjects.” Missouri State Conf. of the Nat'l Ass'n for the Advancement of Colored People v. Ferguson-Florissant Sch. Dist., 201 F.Supp.3d 1006, 1082 (E.D. Missouri 2016). This case is no exception. It is evident that, during the redistricting process, the Secretary and the Legislative Assembly sought input from the Tribes and other Native American representatives. It is also evident that the Secretary and the Legislative Assembly did carefully examine the VRA and believed that creating the subdistricts in district 9 and changing the boundaries of districts 9 and 15 would comply with the VRA. But unfortunately, as to districts 9 and 15, those efforts did not go far enough to comply with Section 2.
“The question of whether political processes are equally open depends upon a searching practical evaluation of the past and present reality, and on a functional view of the political process.” Id. (citing Gingles, 478 U.S. at 45). Having conducted that evaluation and review, the 2021 redistricting plan, as to districts 9 and 15 and subdistricts 9A and 9B, prevents Native American voters from having an equal opportunity to elect candidates of their choice in violation of Section 2 of the VRA. The Secretary is permanently enjoined from administering, enforcing, preparing for, or in any way permitting the nomination or election of members of the North Dakota Legislative Assembly from districts 9 and 15 and subdistrict 9A and 9B. The Secretary and Legislative Assembly shall have until December 22, 2023, to adopt a plan to remedy the violation of Section 2. The Tribes shall file any objections to such a plan by January 5, 2024, along with any supporting expert analysis and potential remedial plan proposals. The Defendant shall have until January 19, 2024, to file any response. The first election for the state legislative positions in the remedial district shall occur in the November 2024 election.
IT IS SO ORDERED.
LET JUDGMENT BE ENTERED ACCORDINGLY.