The Supreme Court Needs to Show Its Work

illustration
Getty / The Atlantic

Updated at 9:35 a.m. ET on March, 10, 2021.

A little after 8 p.m. eastern time on Friday, February 26—a time when people were logging off for the weekend, parents were putting kids to bed, and the last thing on anyone’s mind was the Supreme Court—the justices issued their latest in a series of controversial orders blocking local and state restrictions on indoor religious services. Like most of the Court’s prior actions in COVID-19 cases, the order in Gateway City Church v. Newsom was unsigned, unscheduled, and (mostly) unexplained. In freezing regulations in Santa Clara County, California (a step that two lower courts had refused to take), the order cryptically explained that such a result was “clearly dictated” by a similar order that the Court had handed down in a challenge to California’s statewide rules three weeks earlier—an order that was itself given without any majority rationale. Even though Santa Clara County had explained at some length why its restrictions differed from the state’s, an unnumbered majority of justices tersely responded that the lower courts’ decisions were “clearly erroneous” because they had failed to properly decipher the meaning of the Court’s earlier unexplained action in a different case about a different set of restrictions.

This confusion is the product of the so-called shadow docket, a term coined by the University of Chicago law professor William Baude in 2015 to capture the work that the justices do in cases that have not yet been (and may never be) argued on the merits. Although the phrase is relatively new, justices or the full Court have, since the Court’s beginning, issued orders to control their caseloads—and for most of the Court’s history, doing so has been entirely uncontroversial. That’s because an overwhelming majority of these rulings are anodyne—giving parties more time to file a brief, allowing interested nonparties to file briefs as “friends of the Court,” lengthening oral arguments, and so on. As Congress, from 1925 to 1988, gradually but consistently gave the Court more control over its docket, these case-management orders were joined by more significant decisions regarding whether to grant certiorari—the means by which the justices decide whether to take up a case for plenary review. Even then, though, the norm is to preserve the status quo—to deny certiorari. (The Court grants roughly 1 percent of the 6,500 petitions it receives each year.)* And in the few cases in which the Court agrees to hear the appeal, those orders don’t themselves do much; rather, they are a promise of further review—an additional round of briefing, an oral argument, and, eventually, a signed opinion from the Court.

More in this series

Imperfect though it may have been, the result of the Court’s traditional approach was to maximize (at least from the Court’s perspective) the transparency of its legal reasoning, to provide the clearest possible guidance to lower courts in pending and future cases, and to limit the precedential effect of the Court’s decision making to those rulings that received plenary consideration. Some cases, of course, presented exigencies requiring the Court to move faster than it might have wanted to, but its response was usually to confine any rulings under those circumstances to the facts of the particular dispute. The less the Court was able to follow regular order, the less it sought to accomplish or resolve. But the rise of the shadow docket in recent years has upset that understanding—with troubling implications for the Court’s legitimacy.

Most worryingly, the shadow docket has seen a significant uptick in recent years in a type of procedural ruling that used to be exceedingly rare—a grant of “emergency relief” while an appeal of a lower-court decision is pending. The first category of these rulings is stays—orders that freeze the effect of lower-court injunctions—which Americans saw over and over again, for example, with regard to Trump-administration immigration policies such as the second iteration of the travel ban, the border wall, the “Remain in Mexico” asylum policy, and so on. In each of these cases, a lower court held the policy to be unconstitutional or otherwise unlawful. By issuing stays of those rulings, the justices allowed the policies to go into effect for the duration of the appellate process—even though no court ultimately held that these policies were actually lawful, and even though thousands (if not millions) of individuals were adversely affected in the interim.

A second category of these rulings has come in the form of injunctions—as in Gateway City Church—where the Court has itself been the institution to block a local or state regulation after multiple lower courts would not. As with the stays in immigration cases, these rulings have enormous real-world effects—for example, preventing states from enforcing public-health measures designed to mitigate the spread of COVID-19.

Third, and perhaps most aggressively, the Court has also lifted stays in a number of death-penalty cases, including a series of rulings that cleared the way for the federal government’s resumption of executions last summer. Indeed, of the 13 executions that the federal government carried out in the last six months of the Trump administration, five were possible only because the Supreme Court stayed lower-court injunctions—without ever resolving the legal arguments that had led the lower courts to pause the executions. And those five rulings led the U.S. Court of Appeals for the D.C. Circuit to overturn trial-court injunctions in at least three of the other eight cases.

It’s not just that these rulings have profound real-world consequences; it’s that they are anathema to principles of judicial decision making. Most of the orders in these cases come with little to no analysis, or without any separate writings from the justices in the majority trying to explain the basis for their decision, as in the Gateway City Church case. Sometimes, the public is even left to guess as to who cast the crucial votes—as happened on February 11, when the Court turned down Alabama’s request to allow an execution to go forward without permitting the inmate’s spiritual adviser in the execution chamber. Seven of the nine justices publicly revealed their vote, but two did not. Although at least one of the remaining justices had to have joined the concurring justices to produce a majority, we have no idea how either Samuel Alito or Neil Gorsuch voted, or their reasoning.

The lack of clarity as to the vote count would be problematic enough if the substance of these rulings were easily digestible to observers. But it hasn’t been. Instead, confusion over the rulings has led one district judge to attempt to create a taxonomy for understanding how to give precedential effect to these orders, and provoked a sharp division in a Fourth Circuit case in August, when the majority and the dissent vigorously disagreed as to how much weight one of these shadow-docket rulings ought to receive. It’s often hard to understand how these rulings meet the Court’s own criteria for emergency action. For instance, the justices agreed on January 12 to stay a lower-court opinion that had cleared the way for pregnant women to obtain Mifeprex—an abortion-inducing medication that usually must be obtained in person—remotely during the pandemic. The “emergency” application that the Trump administration had filed seeking a stay in that case was docketed on August 26. In contrast, the Trump administration’s expedited appeal for a full hearing of a lower-court decision barring the exclusion of undocumented immigrants from the calculations that go into mapping congressional districts was docketed on September 22, was argued before the full Court on November 30, and was resolved in a signed opinion handed down on December 18.

These shortcomings have been intensified by how many more of these rulings we’ve seen in recent years. During the George W. Bush and Obama administrations combined, the federal government sought emergency relief from the justices a total of eight times—once every other Supreme Court term. (The Court granted four of these requests and denied four.) During the Trump administration, that number surged to 41 requests in four years. And the justices largely acquiesced—granting 28 of those requests in full or in part. Expanding the data set to encompass applications brought by other parties, dozens of rulings in the past few years have used the shadow docket to alter the status quo—compared with only a handful over the previous few decades.

Perhaps unsurprisingly, these rulings are also becoming far more divisive: Of the eight requests for emergency relief from the Bush and Obama administrations, only one provoked any public dissent. In contrast, a majority of the Trump administration’s requests provoked at least three justices to publicly register a dissent, and 10 of them led the justices to publicly split 5–4. In each of the first two terms since Justice Anthony Kennedy left the Court in 2018, there were on average five times as many 5–4 rulings on the shadow docket as there were during his last term. And the public divisions among the justices haven’t been limited to cases in which the federal government has sought relief.

Finally, because there are more of them and they’re more divisive, shadow-docket disputes appear to be taking up far more of the Court’s attention. During the October 2019–October 2020 term, the Court handed down signed opinions in only 53 cases argued on the merits before the Court—the lowest total since 1862. Although some of that can be blamed on COVID-19 (which led the Court to postpone 10 cases to the current term), the justices are currently on track to hand down signed opinions in only 56 cases during the term that began in October 2020—which would be the second-lowest total since the Civil War. Correlation is not causation, of course. But it seems to be more than a coincidence that, as the Court has spent more of its time on emergency applications, it has had less bandwidth for its bread and butter.

All of this has happened without any changes to the Court’s procedures, rules, or governing statutes. Indeed, since 1988, Congress has virtually abandoned any role in regulating the Court’s docket—something that it did frequently (and at times aggressively) throughout the country’s first 200 years. That does not mean, however, that Congress is powerless to respond to this troubling phenomenon. If the problem is that disputes are taking too long to reach the Court on the merits, Congress can speed up the appellate process. If the problem is that the federal government is too often subject to “nationwide” injunctions from remote district judges with partisan agendas, Congress can allow the government to transfer such cases to the federal courts in Washington, D.C. If the problem is that courts are being too tolerant of local and state COVID-19 restrictions, Congress could at least attempt to clarify when injunctive relief is and is not appropriate—as it has done in statutes such as the Prison Litigation Reform Act.

But real change has to start with the justices, who—other than Sonia Sotomayor—have been reluctant to acknowledge even the quantitative rise in their shadow-docket decisions, let alone the possibility that such an uptick might be normatively problematic. Perhaps that’s because the justices have an easier time signing on to an order that’s accompanied by no rationale, or voting to keep a stay of an execution in place, when no one outside the Supreme Court will know that they did so. But to whatever extent such motivations help explain why we’ve seen such a dramatic uptick in these obscure rulings, they are not justifications. For a Court whose legitimacy depends largely on the public’s perception of its integrity, the growth of unseen, unsigned, and unexplained decisions that disrupt life for millions of Americans can only be a bad thing—and is reason enough for the Court to bring more of these rulings out of the proverbial and literal shadows.


* This article originally misstated the percentage of petitions the Court grants each year.