106. The Supreme Court and the 2024 Election, Part II

Wednesday's grant of emergency relief in the Virginia case epitomizes what's wrong with the "shadow docket"; the question now is whether we're in for a similarly unexplained ruling in Pennsylvania.

Welcome back to “One First,” a weekly newsletter that aims to make the U.S. Supreme Court more accessible to all of us. In addition to Monday’s regular installment and Thursday’s bonus issue, it seemed worth providing additional coverage of the Court this week to look carefully at its role in the election—especially its unexplained decision Wednesday morning allowing Virginia to remove more than 1600 registered voters from the rolls (based on the claim that they are non-citizens); and its impending ruling on a dispute over “naked” ballots in Pennsylvania, which may come as soon as today. If you’re finding this coverage useful, and aren’t already a regular subscriber, I hope you’ll consider becoming one:

I previewed both of these disputes in Monday’s regular newsletter issue. But to make a long story short, Wednesday’s ruling in the Virginia case is the epitome of what’s wrong with the “shadow docket.” The Court (1) granted emergency relief; (2) in a case with a sharp partisan valence; (3) along its usual ideological lines; and (4) with zero explanation as to why the applicants were entitled to it (or why the Fourth Circuit erred in its written opinion holding that they were not). Indeed, whether you like the bottom line the justices reached or not, these kinds of cases are the precise disputes in which all of us—state election officials; voters; and everyone in between, including the Court itself—would be better off with more clarity as to why the Supreme Court did what it did. For reasons sketched out below, I’m skeptical that such an opinion would’ve been persuasive; but it sure would’ve been better than what we got.

To briefly recap how we got here: Exactly 90 days before next Tuesday’s election, Virginia Governor Glenn Youngkin issued an executive order directing state election officials to switch from monthly to daily culling of the state’s voter rolls for anyone suspected of being a non-citizen—including, perhaps most importantly, individuals who, when they completed their driver’s license application, neglected to check the box identifying as a U.S. citizen, or individuals who became naturalized citizens after moving to Virginia and first getting on the grid. (There’s a lot of misinformation out there about these cases, but it is clear that at least some U.S. citizens—and lawfully registered voters—were removed from the rolls as part of this process; the Prince William County Registrar noted at a public hearing that he had to remove 43 voters who had previously affirmed their citizenship and had voting histories.)

The federal government and private plaintiffs brought two separate suits against the state, claiming that these actions violated the so-called “quiet period” mandate of the National Voter Registration Act, which bars states from implementing any systematic program aimed at removing the names of ineligible voters from voter registration lists in the 90 days before an election—entirely to avoid what appears to have happened in Virginia, i.e., the risk of disenfranchising eligible voters by accident and leaving them with insufficient notice and/or time to correct the error. Indeed, the NVRA specifically authorizes suits like these, within 30 days of an election, to ensure that its protections can be judicially enforced.

Last Friday, a federal judge in Virginia granted a preliminary injunction—ordering the state to put the roughly 1600 registered voters it had disenfranchised under Youngkin’s executive order back on the rolls. On Sunday afternoon, a unanimous Fourth Circuit panel refused to stay almost all of that injunction (including the parts that matter)—and released a six-page opinion explaining why.

In a nutshell, the court of appeals offered three reasons for denying the stay. First, on the merits, it concluded that Virginia’s argument that it hadn’t violated the NVRA’s quiet period was unlikely to succeed—because it is based on an interpretation of the statute that “violates basic principles of statutory construction.” Second, it explained why the so-called “Purcell principle” also did not justify putting the district court’s ruling on hold—because, among other things, Purcell would render the quiet period unenforceable (and because Purcell is, at most, a judge-made constraint on the equitable powers of federal courts, it can be overridden by a statute that authorizes, if not demands, such intervention). Finally, it also explained why Virginia’s argument that it would suffer “irreparable harm” without a stay was “weak,” especially when balanced against the disenfranchisement risk that led Congress to create the quiet period in the first place.

Analytically, the holding that Purcell didn’t apply explained why Virginia had to satisfy the traditional standard for a stay pending appeal; and the merits and irreparable harm holdings provided independent reasons why it failed to meet that standard. Thus, to disagree with the Fourth Circuit, the Supreme Court would have to conclude either that (1) Purcell does apply—thus justifying a freeze of the district court’s election-eve intervention without regard to the merits or the equities; or (2) Virginia meets the traditional standard for a stay both because it has a likelihood of success on the merits and because it will suffer irreparable harm without a stay. (These hurdles—and the difficulties in explaining why the Fourth Circuit was wrong—are why I wrote on Monday that I thought it unlikely that the Court would intervene.)

We’ll never know exactly what the Court concluded. All we know is that, a few minutes before 10 ET on Wednesday morning, the Court issued an unsigned and entirely unexplained order staying the district court’s injunction in full—with Justices Sotomayor, Kagan, and Jackson publicly noting, but not explaining, their dissents.¹ This is the sum total of what the Court published:

Did the majority find Virginia’s statutory arguments more persuasive than the Fourth Circuit (and the district court) had? Did it disagree with the Fourth Circuit’s refusal to apply Purcell? Did it believe that Virginia really was suffering “irreparable harm” justifying emergency relief because of unsubstantiated claims of non-citizens voting (which is, in any event, a felony)? We’ll never know—and that’s the problem.

I’ve written before (in Chapter 7 of my book) about why it is so important for the Court to explain itself—especially when it grants emergency relief in a manner that divides the justices along ideological lines. The virtue of an explanation is that it’s the best (indeed, the only) defense the justices have against claims that they’re simply voting their partisan political preferences. Indeed, the Court itself has long viewed its ability to provide principled justifications as essential to its public credibility (and legitimacy)—not because we’ll necessarily agree with the principles the justices espouse, but because we’ll hopefully at least agree that they are principles.

And when Justice Kavanaugh this April attempted in a concurring opinion to explain why the Court shouldn’t always have to write in such cases, I wrote a post explaining why, with all due respect, his arguments were both self-contradictory and patently unpersuasive. If anything, the importance of providing a written rationale goes up dramatically in election-related disputes, because that’s where there is the most immediate and direct risk of the justices at least appearing to play partisan favorites—all the more so when, at least publicly, the decision appears to have divided the justices straight down partisan lines. If anything, the Virginia case is an especially extreme example in this regard because, unlike what’s true in most election cases, here, the issue turned on a federal statute that expressly contemplates (and, indeed, authorizes) litigation on the eve of an election. To not explain why that statute nonetheless couldn’t be enforced in these circumstances raises separation-of-powers questions that the typical election dispute doesn’t.

I’m not indifferent to the difficulty of providing a rationale in a case in which the timing was so compressed. But even a sentence or two would be better than nothing—to indicate, at the very least, what the principal ground for granting relief (traditional equities or Purcell) actually was. Indeed, when the Court intervened earlier this year in a Louisiana redistricting case, it at least did that much. Ditto in the Texas abortion case in September 2021—where a majority was able to cobble together a 401-word paragraph in a little more than 24 hours. Simply put, if it’s important enough for the Court to intervene, it should be important enough for the justices to drop everything and tell us why they’re doing so—even if they don’t have time to write a lengthy opinion.

Fortunately, the immediate stakes of the Virginia ruling are likely to be relatively modest, at least nationally. Virginia has same-day voter registration for anyone who was wrongfully removed from the rolls; and even if that number ends up being in the hundreds, and not everyone is able to correct the error, it’s hard to imagine any federal races turning on such a slim margin (the Virginia state legislature is, alas, another matter). And the one (and only) upshot of the Court not explaining itself is that its ruling can hardly be held out as a precedent for the conclusion that Virginia’s behavior did not violate the NVRA; obviously, the Court said nothing of the kind.

But the real concern Wednesday’s ruling raises is that the justices may be willing to intervene in a similar fashion in election-related disputes with higher stakes—such as the pending application seeking a stay of the Pennsylvania Supreme Court’s ruling that would allow for the counting of in-person provisional ballots cast by voters whose mail-in ballots are rejected for failing to include the required secrecy sleeve. Like the Virginia case, that case also raises an applicability-of-Purcell question (does Purcell even apply to rulings by state courts?) and a messy merits question (about whether the Pennsylvania Supreme Court decision reflects the kind of egregious behavior that triggers the so-called “Independent State Legislature” theory after Moore v. Harper). Given that there’s a non-frivolous possibility that the presidential election may very well come down to Pennsylvania, and that intervening in Pennsylvania might lead to requests to intervene in other swing states with Democratic majorities on the state supreme court, it would be the apex of judicial irresponsibility for the Court to grant emergency relief in that case without at least some (persuasive) explanation for why it’s doing so.

My book on the shadow docket has a rather provocative subtitle: “How the Supreme Court Uses Stealth Rulings to Amass Power and Undermine the Republic.” I’ve gotten a fair amount of criticism for leveling such a charge at the Court. But as I explained in the book, it’s in election cases that the charge has the most purchase, for it’s in those cases that, if the Court is going to intervene when it comes to how we exercise the most basic right of our democratic republic—choosing who will represent us—it has the highest obligation to do so based upon neutral legal principles rather than partisan politics, and to tell us what those principles are.

The Court shirked that obligation in the run-up to the 2020 election—in a way that, fortunately, ended up not making a difference. It flubbed it again on Wednesday. We can only hope that it breaks that streak in the Pennsylvania case—and going forward—not only for the sake of preserving at least some public confidence in the Court today and going forward, but for the sake of the Republic itself.

1. I’ll note, because some media outlets reported the ruling as “6-3,” that we don’t know for a fact that that was the vote count. It is possible, as I’ve explained before, that there was a “stealth” dissent—and that one of the Republican appointees voted against the stay, but chose not to publicly indicate as much. I have no specific reason to believe that that’s what happened here, but I also can’t prove that it didn’t.