Symposium: How to win the partisan-gerrymandering cases

Daniel Tokaji is Associate Dean for Faculty and Charles W. Ebersold and Florence Whitcomb Ebersold Professor of Constitutional Law at The Ohio State University Michael E. Moritz College of Law.

No one said this would be easy. For decades, critics of partisan gerrymandering have been knocking on the U.S. Supreme Court’s door, seeking a ruling that extreme gerrymanders violate the U.S. Constitution. Even as the problem has worsened, the court has refused to open that door – though it hasn’t locked it shut either. Most recently, the Supreme Court denied relief last term in cases out of Wisconsin and Maryland. In the Wisconsin case (Gill v. Whitford), the court unanimously concluded that the plaintiffs had failed to demonstrate their standing to assert that a Republican-drawn state legislative redistricting plan violated their right to vote. In the Maryland case (Benisek v. Lamone), the court affirmed the denial of a preliminary injunction against a congressional redistricting plan drawn by Democrats.

Two partisan-gerrymandering cases are before the Supreme Court again this term. On March 26, the justices will hear arguments in challenges to North Carolina’s congressional redistricting plan (Rucho v. Common Cause) and Maryland’s congressional redistricting plan, which is back before the court (Lamone v. Benisek). Three-judge district courts granted relief in both cases. So there is hope. But after last term’s rulings and the retirement of Justice Anthony Kennedy – long seen as a potential fifth vote to strike down partisan gerrymanders – there are reasons to doubt the plaintiffs’ chances in the Supreme Court.

If either of these challenges is to succeed, the plaintiffs will have to present the best argument that can be marshalled. That means pressing the claim that extreme partisan gerrymandering violates the First Amendment right of expressive association. Both the North Carolina and Maryland district courts accepted this idea, though a winning argument before the Supreme Court will look significantly different than either of the lower court opinions. In the remainder of this post, I offer three suggestions for the plaintiffs and their amici on the argument that is most likely to succeed.

1. Follow Justice Kagan’s lead

Although last term’s decision in Gill was unanimous in holding that the Wisconsin plaintiffs had failed to establish standing on their right-to-vote claim under the equal protection clause, four justices suggested an alternative legal theory. Justice Elena Kagan’s concurring opinion (joined by Justices Ruth Bader Ginsburg, Stephen Breyer and Sonia Sotomayor) offered a theory grounded in the First Amendment right of expressive association. Though the plaintiffs had made such an argument below, Kagan wrote that they “did not advance it with sufficient clarity or concreteness to make it a real part of their case.”

The concurring justices sketched out what a First Amendment association claim might look like – and how it would differ from the equal protection clause arguments that have been the focal point of partisan-gerrymandering litigation until now. As Kagan’s concurrence explained, representative democracy depends on citizens being able to “band together” with others to advance their shared political beliefs. The Supreme Court has long looked with disfavor on laws and practices that weaken a disfavored group’s ability to join together for this purpose. That includes political parties, ranging from major parties (i.e., Republicans and Democrats) to third parties and supporters of independent candidates.

Among the laws that the Supreme Court has struck down as violative of the First Amendment are restrictions on people joining with like-minded others through their votes. An example is Anderson v. Celebrezze, cited in Kagan’s concurrence. In Anderson, the court struck down Ohio’s restrictions on ballot access for third-party presidential candidates. Following previous cases like Williams v. Rhodes, another Ohio ballot-access case, the court held that the First Amendment right of association extends to state laws that limit access to the ballot, because elections are a critical forum for “expression of views on issues of the day, and a candidate serves as a rallying point for like-minded citizens.” On the other hand, the court recognized that some regulation of the electoral process is necessary, so not all restrictions imposed by the state are constitutionally suspect. Anderson prescribed a balancing test, under which the burdens on voting and association should be weighed against the state’s interest.

I was among a group of law professors who submitted an amicus brief in the Wisconsin case, explaining why extreme partisan gerrymanders violate the right of expressive association set forth in Anderson and other cases. Kagan’s concurring opinion develops this idea. She urges consideration of evidence outside the voting process itself – that is, evidence of effects on the disfavored party other than the defeat of their candidates in legislative elections. Citing Anderson, Kagan suggests that partisan gerrymandering may cause “difficulties fundraising, registering voters, attracting volunteers, generating support from independents, and recruiting candidates to run for office (not to mention eventually accomplishing their policy objectives).” These extra-electoral effects are among the burdens on expressive association that partisan gerrymandering may impose, and should be considered as a part of the balancing test.

The First Amendment association theory sketched out in Kagan’s concurrence presents the most promising basis for invalidating extreme partisan gerrymanders. First, the association theory is supported by over 50 years of Supreme Court precedent, which has understood voting and attendant political activities to be a form of association protected by the First Amendment. Holding that extreme redistricting also violates the right of association would be a logical and reasonable extension of this precedent. Second, it best captures the injury inflicted by partisan gerrymandering. That includes systemic injury inflicted on a group of people through the dilution of their votes, as well as effects on the disfavored party and its supporters outside the electoral process. Third, the right of association provides an appropriately nuanced legal standard. Anderson and its progeny require that the burdens imposed by partisan gerrymandering be weighed against the state’s legitimate interest. This avoids the trap of arguing that a redistricting plan is constitutionally suspect if there is any evidence of partisan intent. That argument is a nonstarter. There is probably some partisan motivation underlying virtually every redistricting plan. That’s a bridge too far for this court.

For those who remain unpersuaded by these arguments for pressing the First Amendment association theory, here is one more: Kagan and the three other justices who joined her concurrence suggested it. They and only they have the opportunity to talk with their colleagues at conferences. We should take these justices at their word about what arguments are most likely to persuade one or more of their colleagues.

2. Don’t reinvent the wheel

Over the years, lawyers and legal scholars have engaged in lots of hand-wringing over the legal standard that should govern partisan-gerrymandering claims. That’s understandable, given the significant risk that the Supreme Court will declare partisan gerrymandering to be a nonjusticiable political question due to the absence of judicially discoverable and manageable standards. Four justices would have taken partisan gerrymandering claims off the table back in 2004, in Vieth v. Jubelirer, an equal protection challenge to Pennsylvania’s congressional redistricting plan.

As Kennedy suggested in his Vieth concurrence, the First Amendment right of association provides a constitutional standard that is both discoverable and manageable. Anderson and its progeny prescribe a balancing test, under which courts should first consider the “character and magnitude” of the burden on associational and voting rights. Next, it should consider the “precise interests” advanced by the state for the burden imposed, including not only their “legitimacy and strength” but the extent to which it is “necessary” to burden the rights of voters, candidates and parties. While emphasizing that there is no “litmus-paper test” and that “hard judgments” are necessary, the court said that “the State’s important regulatory interests are generally sufficient to justify reasonable, nondiscriminatory restrictions.” Under Anderson, courts should consider not only the magnitude of the injury, but also its character – and in particular, whether it discriminates against a particular group.

Courts have applied the Anderson balancing standard to a wide variety of election law cases over the years, including ones involving ballot access, write-in voting and primary systems. More recently, a majority of justices applied this standard in Crawford v. Marion County Election Board, an unsuccessful 2008 challenge to Indiana’s voter ID law. Since then, lower courts have applied this balancing standard to various burdens on participation, including limits on early voting and the counting of provisional ballots. Courts have upheld some of these practices and invalidated others. But the standard has proven manageable, allowing courts to carefully review the evidence regarding both the burdens imposed by the law and the state’s legitimate justifications.

Such a standard is no less manageable in the context of partisan gerrymandering. To be sure, it is no bright-line rule. The reality is that partisan gerrymandering necessarily involves complicated questions about the effects of a redistricting plan on individual voters, parties and the system as a whole. Courts must also carefully consider the state’s legitimate reasons for drawing a plan that may have some discriminatory effects on the nondominant party – such as geographic features, preservation of communities and adherence to local government boundaries. To quote Anderson, there is no “litmus-paper test” that will distinguish extreme partisan gerrymanders from run-of-the-mill plans. Courts will have to consider a multitude of factors and exercise judgment, as they so often do in constitutional litigation.

That’s what the district court opinions in both the Wisconsin and Maryland cases try to do, although there are problems in how both courts characterize the First Amendment standard. The North Carolina opinion throws several First Amendment arguments up against the wall – viewpoint discrimination, speaker discrimination, retaliation and political association – evidently hoping that one of them will stick. It winds up adopting a three-part standard, cobbled together from these various lines of precedent, requiring plaintiffs to demonstrate (1) intent to burden supporters of the disfavored party, (2) an actual burden on their speech and association, and (3) a causal relation between the two. Although this captures the factors that courts should generally consider, the court would have been better off sticking with the balancing test prescribed by Anderson and its progeny.

The Maryland district court opinion has similar problems. In their previous trip to the Supreme Court, the Benisek plaintiffs argued that partisan gerrymandering should be considered a form of retaliation forbidden under the First Amendment. But retaliation is a poor fit for the injury effected by gerrymandering. Retaliation cases are backward-looking, tending to focus on the individualized harm done to someone because of perceived political beliefs or affiliations. Gerrymandering, by contrast, inflicts an injury that is systemic and forward-looking, denying a political group and its supporters a fair opportunity to compete for public support. Thankfully, the most recent district court opinion in the Maryland case subtly backs away from (without completely abandoning) the retaliation comparison. But it too endorses a three-part test, requiring (1) specific intent, (2) injury and (3) causation.

The plaintiffs and their amici are better off sticking with the tried-and-true First Amendment association standard set forth in Anderson and its progeny, with the refinement that Kagan suggested in Gill. They should focus on how partisan gerrymandering diminishes representation in legislative bodies and harms people who wish to associate with the disfavored party outside of elections.

3. Tell a story

This brings me to my third and most important point. If either the North Carolina or Maryland plaintiffs are to succeed, they will have to tell a compelling story about how the disfavored party and its members have actually been harmed by partisan gerrymandering. This requires a more granular presentation of the facts and law than we’ve seen before, what my colleague Ned Foley calls a “particularistic” approach.

Plaintiffs have done a great job of presenting empirical research on how gerrymandering systematically disadvantages the disfavored party. That’s especially true in the Wisconsin case last term and the North Carolina case this term. The record shows how partisan gerrymandering locks the dominant party in power while excluding the other major party, using multiple statistical measures that all point in the same direction.

I love a good scatterplot or S-curve as much as anyone. And this type of evidence will be necessary for plaintiffs to prevail. But it won’t be sufficient. This reality is best captured in Chief Justice John Roberts’ reference to empirical research as “sociological gobbledygook” at oral argument last term. This remark is unfair. Even so, it captures the reality that judges aren’t social scientists and can’t be expected to rely exclusively on statistical evidence of partisan disadvantage as a basis for striking down a district or plan.

What every great lawyer knows how to tell – and every judge loves to hear – is a good story. Missing from prior cases is a compelling narrative showing how real people are hurt by gerrymandered districts. This is one of the lessons from the unanimous opinion in Gill. The Supreme Court’s opinion emphasized the “individualized and personal” nature of the right to vote, insisting that plaintiffs show a “district specific” injury for standing. The evidence of statewide injury to the nondominant party and its voters wasn’t enough. I disagree with this ruling, because I think partisan gerrymandering by its nature inflicts a systemic rather than atomistic harm. And it’s still possible that the court could accept a statewide association claim, without proof of individualized harm to identified people. But such a ruling is unlikely.

If the plaintiffs in either the North Carolina or Maryland case hope to win, they will have to tell a more compelling story than they have so far. That requires evidence of real people who have been hurt – for example, by their communities being cracked or packed by redistricting. An example is the story that in 2006 moved the court to action – albeit on a different kind of claim – in League of United Latin American Citizens v. Perry. The court heard the story of Latino voters in Laredo, Texas, who were becoming more politically active and on the cusp of finally electing their candidate of choice before the mid-decade gerrymandering engineered by Tom DeLay stole away that opportunity.

The best stories have good guys as well as bad guys. So in addition to showing how the dominant party drew the plans to help themselves, the plaintiffs must show how real-life voters of the other major party were hurt. An example is how students at North Carolina A&T State University, a historically black institution, were split into two districts by that state’s congressional districting plan. One student explains here how “my vote and those of my peers were diluted and our ability to elect the representatives we chose was undermined” by the way these districts were drawn. The plaintiffs should also show how gerrymandered districts have stymied political organizing by the disfavored party and would-be members. The lower-court opinions in both the North Carolina and Maryland cases describe this evidence in general terms, but a more particularistic – and more humanized – account will be essential if plaintiffs are to prevail in either case.

Winning a partisan-gerrymandering claim has never been easy. But to have any realistic hope of prevailing, the plaintiffs and their amici will have to present the facts and law in a way that’s different from the lower-court opinions. That means following Kagan’s lead on seeing partisan gerrymandering as a violation of associational rights, applying the time-tested balancing test of Anderson and its progeny, and telling a story of how real human beings are harmed by gerrymandered districts.

* * *

Past cases linked to in this post:

Anderson v. Celebrezze, 460 U.S. 780 (1983)
Benisek v. Lamone, No. 17-333 (U.S. Jun. 18, 2018)
Crawford v. Marion County Election Board, 553 U.S. 181 (2008)
Gill v. Whitford, No. 16-1161 (U.S. Jun. 18, 2018)
League of United Latin American Citizens v. Perry, 457 F. Supp. 2d 716 (E.D. Tex. 2006)
Vieth v. Jubelirer, 541 U.S. 267 (2004)
Williams v. Rhodes, 393 U.S. 23 (1968)