Baber v. Dunlap
| Decision Date | 13 December 2018 |
| Docket Number | NO. 1:18-CV-465-LEW,1:18-CV-465-LEW |
| Citation | 376 F.Supp.3d 125 |
| Parties | Brett BABER, et al., Plaintiffs v. Matthew DUNLAP, et al., Defendants |
| Court | U.S. District Court — District of Maine |
Joshua A. Tardy, Irwin Tardy & Morris, Newport, ME, Andrew G. Woodson, Eric Wang, Pro Hac Vice, Lee E. Goodman, Wiley Rein LLP, Washington, DC, Joshua A. Randlett, Rudman & Winchell, Bangor, ME, for Plaintiffs.
Phyllis Gardiner, Thomas A. Knowlton, Office of the Attorney General Six State House Station, Augusta, ME, Patrick N. Strawbridge, Consovoy McCarthy Park PLLC, Boston, MA, for Defendants.
DECISION AND ORDER ON MOTION FOR PRELIMINARY INJUNCTION
Although the Court scheduled the hearing on December 5, 2018 to address Plaintiffs' Motion for Preliminary Injunction, the parties agreed that the question of injunctive relief should be consolidated with a final ruling on the merits of the action. Therefore, pursuant to Rule 65(a)(2), this Decision and Order will be accompanied by a final judgment in favor of Defendants.
On November 6, 2018, the State of Maine held a general election at which races for federal office were governed by Maine's "Act to Establish Ranked-Choice Voting" ("RCV Act").1 This new manner of holding federal elections2 is the product of a popular initiative, the history of which has been set forth previously and is not repeated here.3 See, e.g. , Maine Republican Party v. Dunlap , 324 F.Supp.3d 202, 204–06 (D. Me. 2018) ; Maine Senate v. Sec'y of State , 183 A.3d 749 (Me. 2018) ; Opinion of the Justices , 162 A.3d 188 (Me. 2017). Under the RCV system employed in Maine, when there are three or more candidates on the ballot, a candidate cannot be declared the winner of the election following tabulation of the votes without securing a majority of the ballots validly cast (i.e., excluding ballots invalidated due to overvotes (marking more than one candidate at the same level of ranking) or undervotes (failing to rank a candidate) ). 21-A M.R.S. § 723-A.
Plaintiffs, Brett Baber, Terry Hamm-Morris, Mary Hartt, and Bruce Poliquin, are residents of Maine's Second Congressional District. Plaintiffs participated in Maine's November 6, 2018, general election, at which each cast a vote for Bruce Poliquin to continue serving as Representative of the Second Congressional District in the United States House of Representatives. They maintain that the RCV Act is unconstitutional, both facially and as applied, and that it violates the Voting Rights Act. They maintain that the ballot form and instructions were too confusing and that the manner by which Defendant Dunlap tabulated the votes diluted the votes cast by Poliquin supporters and otherwise disenfranchised too many Maine voters to withstand scrutiny.
The ballot for the Second District house race provided a choice among four candidates, a space to enter a write-in candidate, and a manner by which to rank the candidates, in the following form:
Rep. to Congress 1st Choice 2nd Choice 3rd Choice 4th Choice 5th Choice
District 2
Bond, Trffany L. 0 0 0 0 0
Portland
Independant
Golden, Jared F. 0 0 0 0 0
Lewiston
Democratic
Hoar, William R.S. 0 0 0 0 0
Southwest Harbor
Independent
Poliquin, Bruce 0 0 0 0 0
Oakland
Republican
Write-in 0 0 0 0 0
The ballot included the following instructions:
Plaintiffs each filled in the circle for Bruce Poliquin shown in the first-choice column of the ballot. They did not fill in any other circles. Many other voters took the same approach. Some voters expressed equivalent support for Mr. Poliquin, but filled in the Poliquin circle in every column of the ballot. These voters, in other words, elected not to rank any candidate other than their preferred candidate. Other voters expressed their support for the other candidates in the same fashion. Many other voters chose to rank every candidate. In all, given five potential candidates and five columns, there were 120 different orders in which to rank the candidates, assuming one nominated a write-in and then went on to rank every candidate. There were several other ways in which one might respond to the ballot. For example, 5,582 voters submitted their ballots without filling in any circles.
Following the election, Defendant Dunlap oversaw a process in which his office gathered the ballots and tabulated the election results. On November 7, Defendant Dunlap announced that, based on the tabulation of all "first choice" votes, no contestant in the race achieved victory by a majority. The results of the initial tabulation were as follows:
Candidate Votes
Bruce Poliquin 134,184
Jared Golden 132,013
Tiffany Bond 16,552
William Hoar 6,875
TOTALS 289,6244
[Editor's Note: The preceding image contains the reference for footnote4 ].
Pursuant to the RCV Act, because no candidate achieved a majority, Defendant Dunlap was required to conduct a further tabulation of the votes. Because it was mathematically impossible for Ms. Bond or Mr. Hoar to be elected, Defendant Dunlap performed a "batch elimination" of those candidates. Id. § 723-A(1)(A). He then reviewed the ballots in which the eliminated candidates were named as first choice, to determine if those ballots indicated a preference between the remaining two candidates, Mr. Poliquin and Mr. Golden. If so, then those ballots were redistributed accordingly.5 On November 26, 2018, Defendant Dunlap published and certified a final tabulation of the votes. The results were as follows:
Candidate Votes Percentage
Jared Golden 142,440 50.62%
Bruce Poliquin 138,931 49.38%
TOTALS 281,3716 100%
[Editor's Note: The preceding image contains the reference for footnote6 ].
Because the Secretary of State certified Jared Golden as the winner of the RCV election, Plaintiff Bruce Poliquin requested a recount pursuant to 21-A M.R.S. § 737-A. The recount is under way at this time.
The matter came on for hearing on December 5, 2018. Before hearing oral argument, the Court permitted Plaintiffs to call to the witness stand Dr. James G. Gimpel, Ph.D., a professor at the University of Maryland – College Park. Among other areas of expertise, Dr. Gimpel is well studied in the area of voter behavior. Dr. Gimpel testified that alternative systems for conducting elections, such as RCV, are generally considered by their proponents to be "systems to enhance participation." According to Dr. Gimpel, interest in these systems is growing and, undoubtedly, will lead to more litigation like the litigation now before this Court. Dr. Gimpel has formed the opinion that RCV (or "instant run-off") systems do not offer advantages over a plurality system, or over a majority system that resolves close elections by means of an actual run-off. The primary flaw he sees in RCV is that, unlike ordinary elections and ordinary run-offs, voters are required to make predictions about who will be left standing following an initial tabulation of the votes.
While Dr. Gimpel concedes that many voters have sufficient information to make reliable predictions, he believes that a portion of the voting public has insufficient interest and information to make a meaningful assessment about likely outcomes. In his view, RCV is "flat out unfair to the uninformed voter." He also maintains that the instructions Defendant Dunlap provided with the ballot leave such voters "clueless."
Dr. Gimpel contends that the data of voting behavior for this election (i.e., the ballots in this election) reinforce his opinion. He observes that thousands of voters cast ballots that were invalid, and that the most logical inference is that those voters guessed wrong due to an information deficit. Dr. Gimpel presumes that the voters in this category are predominantly independent voters, meaning they have no party affiliation. By his reasoning, independent voters such as Ms. Bond and Mr. Hoar's supporters are, on average, less informed on the issues.7 According to Dr. Gimpel, this information deficit is demonstrated by the fact that many of the voters who identified Ms. Bond or Mr. Hoar as their first choice neglected to rank another candidate. In his view, this is proof that they believed Ms. Bond or Mr. Hoar would be victorious. He finds it hard to believe that voters would "drop out like this" if they were presented with a simple choice between a Republican candidate and a Democratic candidate.8 On the other hand, he also testified that independent voters, on average, are not as likely to turn out for elections in the first place.9
On cross examination, Dr. Gimpel testified that he did not interview or consider any interviews or studies of actual Maine voters, but that, remarkably, he would like to develop a survey to evaluate what voters were thinking. When asked to articulate why the ballot and the voting instructions were confusing, Dr. Gimpel testified that the worst thing about the ballot and instructions is what was omitted from the instructions. In particular, Dr. Gimpel suggested the instructions should have explained the various ways in which a vote could be invalidated. Dr. Gimpel was asked to evaluate Plaintiffs' contention that RCV is flawed because it fails to produce "monotonic" results.10 Dr. Gimpel testified that he did not detect a monotonicity problem in this particular election. Finally, Dr. Gimpel opined that...
To continue reading
Request your trial
-
Kohlhaas v. State, Supreme Court No. S-18210
...Voting , 109 CAL. L. REV. 1773, 1788-95 (2021) (detailing the history of majority provisions).166 See Baber v. Dunlap , 376 F. Supp. 3d 125, 130-31 (D. Me. 2018).167 See AS 15.15.350(c) -(e) ; AS 15.15.360(a)(1).168 AS 15.15.350(d).169 Id.170 AS 15.15.350(d)(2), (g)(2).171 AS 15.15.350(d).1......
-
Hagopian v. Dunlap, 1:20-cv-00257-LEW
...identifying a burden,8 Defendants are likely to establish that the burden is a rational and fair means of conducting an election. Baber , 376 F. Supp. 3d at 144.2. Second Claim – Compelled Speech Plaintiffs very much do not want to cast a ballot for anyone other than Senator Collins, and th......
- Duncan v. O'Shea