Recent commentary on the Supreme Court’s stay order in Merrill v. Milligan has focused on what the decision portends for the future of Section 2 doctrine. But the Court’s radical expansion of Purcell threatens to sow more immediate electoral chaos as well, drastically undercutting Black representation in Congress in 2022 and playing a potentially decisive role in which party takes control of the House of Representatives. Below I explore what might be done.
The (Latest) Purcell Problem
As Merrill brings the problems with Purcell back to the fore—problems illuminated by Prof. Hasen and Prof. Codrington in their scholarship—I want to briefly highlight one more. If the conservative majority is going to apply Purcell this far out from an election, an enormous amount of pending litigation will be caught in its sweep. For redistricting litigation in particular, the consequences are easy to track and hard to deny.
If the Voting Rights Act had been enforced in Alabama, Black voters would’ve had the opportunity to elect two Members of Congress in 2022. Now they will only elect one. A similar case is pending in Georgia, where a stay order from the Supreme Court could prevent an additional Black opportunity district from being created before 2022. Another case may well arise in Louisiana. And DeSantis seems bent on dismantling a Black opportunity district in Florida too.
In short, if the Supreme Court continues to apply Purcell as broadly as it did in Merrill, there may be several maps that have been declared unlawful before the 2022 elections that are nonetheless in place for those elections. This is a recipe for disaster. The impact on Black representation in Congress is an outrage all on its own. But it’s also conceivable that control of the House in 2022 could be determined by a handful of seats—seats that might only swing to Republicans because the Supreme Court effectively suspended enforcement of the VRA for most of the year leading up to the general election.
The last thing this country needs is a “Purcell Congress,” controlled by a bare majority elected from unlawful districts.
So, how to avoid (yet another) emerging threat to public trust and electoral legitimacy? There are no easy fixes, but below are three places for advocates to apply pressure if we want to avoid this outcome.
The Supreme Court
Unfortunately, the most straight forward fix also seems the least likely: The Supreme Court should step back from the brink and rein in its dangerous extension of Purcell.
First, extending Purcell this far out from an election deeply undermines the rule of law. Purcell makes sense when an election is close and the risk of confusion is demonstrably high: in such a scenario, judicial intervention could do more harm than good. Applying Purcell outside of these narrow circumstances encourages lawlessness and undermines electoral legitimacy. In Merrill, the Court issued a stay nine months before the general election and four months before the primary. That kind of “equitable balancing” threatens to let a judge-made remedial guideline eclipse the full force of constitutional and statutory law. The Constitution and the Voting Rights Act don’t only apply in odd-numbered years.
Nor is it difficult to draw a straight line from Purcell to democratic instability. The more Purcell applies, the more officials will be elected under unlawful conditions, and the more the public will understandably question the legitimacy of the results. And the longer Purcell applies, the longer legislators will know they can act free from judicial review, and the greater the incentive will be to rig election rules in their favor. (That incentive is especially acute right now, given that congressional redistricting is still underway. Merrill sends a strong signal to would-be lawbreakers that the coast is clear.)
Second, invoking Purcell with respect to brand new, recently enacted maps is nonsensical. As Deuel Ross and Prof. Shaw note in a recent (and unnerving) episode of Strict Scrutiny, the whole purpose of Purcell is to protect the status quo and avoid the confusion that can come from disturbing expectations about the rules governing an election. At the beginning of a redistricting cycle, there are no settled expectations about the configuration of districts because the maps are new either way. Relying upon Purcell in such circumstances has the perverse effect of requiring administrators, candidates, and voters to learn to navigate one map in one election year and another map the next election year—cementing a new unlawful status quo only to upend it. This creates confusion where there would otherwise be none.
All of which sets up the final point: There was a time when the Supreme Court rightly understood that district courts are better positioned to navigate the delicate, intensely local, and fact-laden judgments necessary to craft an equitable remedy in redistricting cases. Before issuing another stay under Purcell, the conservative majority should revisit Upham v. Seamon, 456 U.S. 44 (1982).
Today, Upham is best known for the principle that a court’s redistricting remedies must be tailored to the specific violations it identifies. But the case is also a perfect example of how appellate courts should approach Purcell. In Upham, the Supreme Court held that the district court had erred by imposing its own policy preferences in its interim remedial map. Nonetheless, the Court concluded, “it does not necessarily follow that [the district court’s] plan should not serve as an interim plan governing the forthcoming congressional elections.” The Court left the ultimate question up to the district court:
The filing date for candidates, which was initially postponed by the District Court, has now come and gone. The District Court has also adjusted other dates so that the primary elections. . . may be held. . . Because we are not now as familiar as the District Court with the Texas election laws and the legal and practical factors that may bear on whether the primary elections should be rescheduled, we . . . remand the case to that court for further proceedings. . . . [W]e leave it to that court in the first instance to determine whether to modify its judgment and reschedule the primary elections for Dallas County or, in spite of its erroneous [interim map], to allow the election to go forward in accordance with the present schedule.
Upham reflects the long-standing (but perhaps no-longer-standing) norm that the Supreme Court should only disregard a district court’s factual findings or decide fact-intensive questions itself in extreme circumstances.
To be clear: Upham is entirely consistent with the idea that, sometimes, it’s too late to make a change. On remand, the district court in Upham let the election go forward without any further changes to the schedule or the map. But the point is that the district court made that decision based on its own findings about the consequences of further judicial intervention.
Upham is also far more consistent with the spirit of Purcell than the Supreme Court’s approach in Merrill. In Merrill, Justice Kavanaugh asserts in a footnote that “[c]orrecting an erroneous lower court injunction . . . does not itself constitute a Purcell problem.” But, of course, Upham directly contradicts that position. In Upham, the Court engaged in error correction and then deferred to the judgment of the district court regarding next steps precisely because it recognized that disruption of the status quo by a reviewing court poses the same risk of unsettling expectations and sowing confusion as intervention by a district court.
For all these reasons, one would hope that the Supreme Court might reconsider its rash extension of Purcell. Unfortunately, there seems little realistic chance that the Supreme Court will suddenly resume the level of historical deference it once exercised.
Another option? Congress could and should pass a “Purcell fix.” The Freedom to Vote: John R. Lewis Act contained language that would’ve addressed this problem, see § 5006(c), and the need for that solution has only grown more urgent.
Republican Senators regularly point to Section 2 as a reason why additional voting rights protections are unnecessary. Even if this were true (and it most certainly isn’t), the Supreme Court’s aggressive use of Purcell effectively nullifies those protections for 2022. As Congress debates Electoral Count Act reform and considers other ways to shore up public confidence in our elections, putting statutory limits on Purcell is becoming an ever more critical piece of the puzzle.
Litigators and Lower Courts
The last line of defense are the judges currently overseeing pending redistricting cases. Of course, their work product is ultimately subject to the whims of the justices, but there are steps judges can take (and litigators can suggest) to protect the institutional role and prerogatives of the lower courts along the way. I offer a deeper dive regarding these points in Justice Denied: Equity, Elections, and Remedial Redistricting Rules, 19 J.L. Soc’y 229 (2019), but I will summarize some key takeaways here.
First, a district court can issue the merits of its ruling in a declaratory judgment and withhold injunctive action until it is ready to impose a remedial map.
By issuing a declaratory judgment alone, the district court advises the parties as early as possible about the lawfulness of the map and gives the legislature an opportunity to adopt its own replacement map or adjust the primary calendar as it sees fit. By declining to enter an injunction at this time, however, the district court can proceed to develop an appropriate remedy without the case being immediately appealable to the Supreme Court under 28 U.S.C. § 1253. Instead, any appeal would go to the relevant U.S. Court of Appeals in the first instance—a court that will (hopefully) be more likely to follow existing precedent.
Next, the district court can finalize a remedial map as well as developing a robust factual record that explicitly compares the costs, burdens, and risk of confusion associated with each potential course of action, including the costs, burdens, and risks associated with any stay that would leave an unlawful map in place for one election only to have that map replaced ahead of the next election.
To be sure, many of these facts will be developed before and during trial. But to the extent the record needs further supplementing to address the factors relevant to Purcell, including the risks associated with a stay, those findings can be made before the court enjoins the violative map and imposes a remedial map.
None of this is meant to be intra-branch nose-thumbing. Injunctions and stays are supposed to be well-grounded in factual findings, and any finding of likely success of the merits necessarily implies that the unlawful map must be modified at some point—whether before or after the election. The only way to truly weigh the equities is to develop the facts and examine the burdens associated with both courses of action. For this reason, district courts should make explicit findings about the consequences of creating an unlawful status quo, electing representatives from a discriminatory map, and then resetting that status quo after a single election. In the absence of these facts, the Supreme Court may be more likely—as in Merrill—to fill in the gaps with its own unsupported and generic assumptions.
Finally, the district court can implement its remedial map along with a finding about the latest date the court believes any changes could take place before the costs would outweigh the benefits. The court can further indicate that if its decision is reversed on the merits prior to that date, the old map will be automatically restored, but if the court’s decision is reversed on the merits after that date, the remedial map will remain in place until after the election, at which point the old map will be automatically restored.
This approach accomplishes a few things. It ensures that the “status quo” upon appeal to the Supreme Court is, in fact, a lawful map. It provides the Supreme Court more context for how its own intervention could impact the equitable calculus, based on the timeline and associated findings set out by the district court. And, it is consistent with Upham, recognizing that even a technically unnecessary judicial intervention will—at some point—become the status quo worth honoring to avoid any further confusion. As a practical matter, establishing presumptive conditional triggers also helps minimize the risk that time-sensitive windows are missed during the “handoff” between courts during the appeals process.
* * *
None of the approaches above are ideal. Both the Supreme Court and Congress seem unlikely to act, and the recommendations for districts courts are a “last best” option. As Merrill shows, even the most thorough factual findings and most rigorous analysis by a district court can be wiped out in an instant on the shadow docket—a dynamic so disrespectful and disheartening to litigants and judges alike that, as Prof. Tolson recently noted, it may well be the point.
Yet, the steps above could at least help make the stakes and consequences clear. There is a genuine risk that the Supreme Court’s conservative majority swings control of a coequal branch of government in 2022 by clearing the path for enough candidates to be elected from unlawful congressional districts. That may be unavoidable. But it should not be acceptable. It should not be easy. And it should not be normalized.