The Supreme Court term that ended July 1 marked the first for Justice Amy Coney Barrett, and as a result was the first in which conservatives had a decisive 6–3 majority. Many liberals had braced themselves for the worst, expecting that with the addition of a third Trump-appointed justice, Chief Justice John Roberts would no longer be able to exercise his sometimes moderating influence and the Court would veer sharply to the right. But with the notable exception of a disturbingly partisan voting rights decision, the doomsayers were wrong. In many of its most controversial cases, the Court came to surprisingly bipartisan agreement. Its three liberals were in the majority in nearly half of its nonunanimous decisions. And the Court ruled more often in favor of defendants than the government in criminal cases.
Before Barrett’s appointment, Roberts was seen as the Court’s swing vote. In 2019 and 2020 he voted with the liberal justices to invalidate a Louisiana abortion restriction, to overturn Trump’s attempt to eliminate protection from deportation for undocumented immigrants who had come to the US as minors, to extend federal employment discrimination protections to LGBT workers, and to strike from the census a question about citizenship that was designed to suppress responses from immigrants. With the support of Justices Brett Kavanaugh and Neil Gorsuch, Roberts also wrote the Court’s two opinions rejecting Trump’s efforts to block subpoenas for his personal financial records. And perhaps most significantly, in 2012 Roberts broke rank with his four conservative colleagues and voted with the liberals to uphold the Affordable Care Act (ACA) against a claim that it was beyond Congress’s power to enact.
It has generally been thought that the ability to be a mediating justice depends on being the median justice; there must be four justices to one’s right and four to one’s left. That’s why Justices Sandra Day O’Connor and Anthony Kennedy exercised that power for so many years. For a short period, after Kavanaugh took Kennedy’s seat, Roberts had such a swing vote. But now that Barrett has replaced Ruth Bader Ginsburg, he no longer does. If Roberts alone votes with the liberals today, he merely makes a three-justice dissent into a four-justice dissent.
It turns out, however, that there is more than one way to exercise a moderating influence. The Court this term decided few of its important cases by a 6–3 or even a 5–4 margin. More often, it found ways to reach a broader consensus, leading to a large number of “strange bedfellow” opinions. And it did that primarily by deciding cases on the narrowest possible grounds. Roberts has long championed such “minimalism.” As he puts it, “If it is not necessary to decide more to a case, then in my view it is necessary not to decide more to a case.” Minimalism avoids overly broad decisions that the Court may later regret. And as this term demonstrates, it also has the virtue of allowing justices with very different legal philosophies and worldviews to agree.
Fulton v. City of Philadelphia, for example, could have been one of the term’s most controversial decisions. It pitted the free exercise of religion against guarantees of equality for same-sex couples, one of the most hotly contested issues in today’s culture wars.1 Catholic Social Services sued Philadelphia and argued that because it opposes same-sex marriage, the First Amendment gives it the right to receive millions of dollars from the city to certify families as suitable for foster children—even as it refuses to certify any same-sex couples. Philadelphia responded that both its law and its contract required city contractors not to discriminate on the basis of sexual orientation and that there is no free exercise license to discriminate, especially when performing a government service using public funds.
Catholic Social Services invited the Court to overrule Employment Division v. Smith (1990), which held that there is no religious free exercise right to be exempt from neutral and generally applicable legal obligations—that is, laws that apply across the board to conduct, regardless of its secular or religious character. In Smith itself, the Court rejected a claim that Native Americans had a free exercise right to use peyote, even though it was a central part of their religious rituals. It was sufficient, the Court ruled, that California law prohibited any use of peyote, whether for religious or nonreligious purposes. Catholic Social Services argued that religious adherents should be exempt from even such neutral laws unless the state can satisfy “strict scrutiny” by showing that denying a religious exemption is necessary to further a compelling state interest.
To almost everyone’s surprise, the Court ruled unanimously for Catholic Social Services. It did so by issuing an extremely narrow ruling, relying on factors so peculiar to the case that the decision is unlikely to apply to any other city’s foster care program or nondiscrimination law. It cited an unusual provision in the city contract that it interpreted as allowing the city unfettered discretion to give individual exemptions from the nondiscrimination requirement. The Court reasoned that as a result of this provision, the law was not “generally applicable,” and the city had to show why it could not grant a religious exemption to Catholic Social Services. But as both Justices Gorsuch and Samuel Alito pointed out in separate opinions, Philadelphia need only remove the contractual provision permitting exemptions (which it has never used) in order to compel Catholic Social Services to abide by the nondiscrimination requirement in the future. As Alito put it, “This decision might as well be written on the dissolving paper sold in magic shops.”
As a result, the Court did not have to decide whether to overrule Smith. Justices Alito, Gorsuch, and Clarence Thomas proclaimed themselves ready to do so, but significantly, Barrett and Kavanaugh wrote separately to note that even if Smith were to be reconsidered in a future case, there are difficult questions to be addressed about what to replace it with.
In Fulton, as in Masterpiece Cakeshop v. Colorado Civil Rights Commission (2018), which involved a bakery that refused to bake a wedding cake for a gay couple, the conservative justices appear to have blinked. In both cases, they declined to endorse the expansive First Amendment license to discriminate that the religious right had sought. (The Court resolved the cake case on similarly narrow grounds, finding that the civil rights commission that adjudicated it was itself religiously biased.) On July 2 the Court announced that it would not review another such case, Arlene’s Flowers v. Washington, in which the lower courts had ruled that a religiously motivated florist had no right to refuse to sell flowers for a same-sex wedding. The Court is evidently not ready to take on the broad question the religious right has been pressing.
In California v. Texas, the Court again reached broad consensus by ruling narrowly on a potentially divisive case. Seven justices rejected a constitutional challenge to the ACA, which has been the subject of hyperpartisan battle ever since Barack Obama proposed it. The case focused on the same “individual mandate” that the Court upheld, 5–4, against a constitutional challenge in 2012. In that earlier decision, the court ruled that the mandate, which required people to obtain health insurance or pay a tax penalty, was a permissible exercise of Congress’s power to tax. But Trump’s tax cuts eliminated the penalty for failing to purchase insurance. Because the provision no longer imposed a tax, the lawsuit maintained, it was no longer a permissible exercise of Congress’s taxing power. And because the individual mandate was integral to the ACA, the lawsuit contended that the Court should invalidate the entire statute.
The Court, however, declined to decide any of these questions. It turned the challengers away at the threshold, finding that because a failure to buy insurance carried no consequences, the plaintiffs could not show that the mandate caused them any injury, and therefore there was no concrete dispute to decide. Only Alito and Gorsuch dissented.
In the Court’s most important free speech case, Mahanoy Area School District v. B.L., it again reached wide consensus through a narrow decision.2 The case arose when Brandi Levy, a high school freshman frustrated that she had not made the varsity cheerleading team, sent out a Snapchat one weekend while hanging out at a local Cocoa Hut. The message, which went only to her “friends” and was designed to disappear within twenty-four hours, said, “Fuck school fuck softball fuck cheer fuck everything.” When another student took a photograph of the message and showed it to a cheerleading coach, the school suspended Brandi from the junior varsity cheerleading team for a year. It argued that it could have punished such speech had it been uttered on campus and should have the same authority to punish student speech outside school.
The case implicated the speech rights of 50 million public school students as well as the ability of school authorities to regulate out-of-school speech that has in-school effects. The Court could have announced a standard for the regulation of speech outside school, but it did not. It focused instead on the particular facts of Brandi’s speech, which plainly did not disrupt the school in any meaningful sense. Rather than announce a formal rule, the Court identified “three features of off-campus speech that often, even if not always, distinguish schools’ efforts to regulate that speech from their efforts to regulate on-campus speech.” Off campus, young people are under the control of their parents, not of the schools. If the school’s argument were correct, there would be no place where students could speak freely. And as “nurseries for democracy,” schools have a responsibility to instill in students the value of free speech and should teach tolerance by modeling tolerance.
In Nestle USA, Inc. v. Doe, a human rights lawsuit alleging that Nestlé and Cargill aided and abetted child slave labor on cocoa plantations in Ivory Coast, the Court yet again came together on a narrow proposition and avoided more controversial issues. The plaintiffs, former child slaves on cocoa plantations, sued under the Alien Tort Statute, which authorizes foreign nationals to sue in US federal court for injuries sustained in violation of fundamental international legal norms, such as those banning torture, summary execution, and slavery. The Court ruled that because the plaintiffs did not claim that any of Nestlé and Cargill’s aiding and abetting took place in the United States, they could not pursue their lawsuit here. The corporate defendants had urged the Court to go further, by ruling that corporations could never be sued under the statute, and that the statute permits suits only for international law violations recognized at the time it was first enacted in 1789, such as piracy, which would render the law a dead letter. But the Court declined to address either of these arguments, and as a result the decision garnered eight votes.
In a majority of its criminal cases, the Court also reached liberal results despite its conservative makeup. In two cases involving home searches, the Court reaffirmed that the police must obtain a warrant to search a home, overturning lower court rulings that they could dispense with a warrant if they were pursuing a fleeing suspect or trying to respond to a potentially suicidal resident. The votes were 7–2 and 9–0. In two other cases, the Court rejected broad interpretations of federal criminal laws, reading them narrowly to favor criminal defendants. In one of those cases, Van Buren v. United States, the vote was 6–3, but the six were the three liberal justices joined by the three Trump appointees, and Justice Barrett wrote the opinion.
In Torres v. Madrid, involving a police shooting, Roberts and Kavanaugh joined the three liberal justices to rule that the use of physical force to attempt to stop a suspect is a “seizure” requiring probable cause under the Fourth Amendment even if the suspect does not in fact stop. And in a case with echoes of the killing of George Floyd, Lombardo v. City of St. Louis, Roberts, Kavanaugh, and Barrett joined the liberal justices to vacate a lower court’s dismissal of a lawsuit by the survivors of a man killed by being pinned to the floor for fifteen minutes while shackled in a jail cell after he was arrested for trespass and failure to appear in court for a traffic ticket.
The term saw many unusual alliances. And while Roberts has often been assumed to be the most likely swing vote, this past term the conservative justice who most often sided with the three liberals in nonunanimous cases was Kavanaugh, followed by Barrett. So much for writing off the Trump justices.
The principal exception to this trend was Brnovich v. Democratic National Committee, handed down on the last day of the term. Here, the Court’s partisan colors shone through, as the six Republican-appointed justices erected significant barriers to lawsuits under Section 2 of the Voting Rights Act, which forbids any voting practice that denies equal opportunity to voters on account of race.
In Shelby County v. Holder (2013), the Court had invalidated another provision of the Voting Rights Act, Section 5, which required states with a history of discrimination to show that any changes they proposed in their voting rules would not have a disparate impact on minority voters. In erasing the states’ obligation to obtain such “preclearance,” the Court at the time pointed to Section 2, which allows plaintiffs to sue states if they adopt rules or practices that have such an impact. But the Court in Brnovich has now made that alternative much more difficult, potentially insulating many voter suppression measures, such as those recently adopted in Florida and Georgia, from meaningful challenge.
The Court did so by offering a set of “guideposts,” nowhere found in the statute’s language, each one skewed to sustain voting practices even when they disproportionately burden minority voters. The Court effectively said that in order to prevail, challengers would have to show that the practices were unusual, imposed a greater than usual burden, affected a large number of people, and did not have a reasonable justification.
Justice Elena Kagan wrote a brilliant and impassioned dissent, accusing her conservative colleagues of abandoning their professed commitment to interpreting statutes according to their literal terms by superimposing nontextual obstacles. The Voting Rights Act, she underscored, was passed to prohibit practices that have a disparate impact on minority voters. But the majority has now ensured that many such practices will be sustained:
The majority fears that the statute Congress wrote is too “radical”—that it will invalidate too many state voting laws. So the majority writes its own set of rules, limiting Section 2 from multiple directions. Wherever it can, the majority gives a cramped reading to broad language…. What is tragic here is that the Court has (yet again) rewritten—in order to weaken—a statute that stands as a monument to America’s greatness, and protects against its basest impulses.
In addition to Brnovich, the Court decided only a handful of cases by the 6–3 conservative–liberal margin that some feared would predominate. In Cedar Point Nursery v. Hassid, its six conservatives concluded that a California law that allowed union organizers temporary access to farms before and after work and during the lunch hour was a “taking without just compensation,” in violation of the Fifth Amendment. In Jones v. Mississippi, the conservatives ruled that children convicted of murder could be sentenced to life in prison without parole as long as the judge had discretion to consider the defendant’s youth, and did not require the judge to make a specific finding that the child was “permanently incorrigible,” as an earlier decision had suggested.
In Edwards v. Vannoy, Justice Kavanaugh wrote for his conservative colleagues in declining to make retroactive last year’s decision (which he also wrote) that the Constitution requires unanimous jury verdicts in criminal cases. And in Americans for Prosperity Foundation v. Bonta, the Court struck down as violating the right to association a California law that required every charity to provide the state attorney general with a list of its top donors, concluding that the state had more narrowly tailored ways to further its interest in investigating fraud (a result the ACLU supported).
Other than the Voting Rights Act case, perhaps the Court’s most troubling decisions came in cases in which it never heard arguments and only intermittently issued opinions. In a series of emergency appeals challenging pandemic restrictions on social gatherings, the Court repeatedly ruled that religious institutions and groups had to be exempted. Roberts generally voted with the liberal justices, but the other five Republican-appointed justices consistently voted to exempt religious activities.
In Tandon v. Newsom, for example, the Court struck down the application to religious gatherings of a California rule that limited in-home gatherings to no more than three households. The law applied to all in-home gatherings, whether religious or secular; it did not single out religion. But the Court reasoned that because the state allowed larger groups to gather in hair salons, retail stores, and restaurants, it had to let larger groups gather in homes for religious purposes. Justice Kagan dissented, noting that retail stores and homes are not remotely comparable settings, and arguing that “the law does not require that the State equally treat apples and watermelons.” But apparently, if the apple or the watermelon is religious, it now does.
These cases were all decided on an emergency basis without full briefing or oral argument. The more nuanced approach the Court took in Fulton suggests that it may be more successful at reaching consensus when it has time to consider its options and the implications of its decisions more fully. In its pandemic cases, it never did.
Still, the main takeaway from this term is that the Court defied expectations. The addition of Justice Barrett did not lead, as so many had predicted, to a sharp tilt to the right, just as Kavanaugh’s replacing Kennedy failed to do the year before. In both terms, the Court appears to have sought common ground across partisan divides in some of its mostly hotly contested cases. This is in everyone’s interest. It is in liberals’ interest because if the Court voted along party lines, liberals would lose every time. It’s in moderates’ interest because it produces more moderate results. And it is in conservatives’ interest, and indeed, the interest of the institution itself, because the Court’s legitimacy rests on its acting pursuant to law rather than politics. If the Court were to vote, like virtually all legislative bodies these days, along strict party lines, it would undermine its own authority.
The Voting Rights Act decision demonstrates, however, that defying expectations and rising above party lines becomes more difficult as the stakes increase. The Court has already announced that next year it will be deciding major cases on abortion, the right to carry concealed guns, and public aid to religious schools. It may also take up a challenge to Harvard’s affirmative action program. Next term, then, will be a better test of whether the Roberts Court can continue to avoid partisan politics in the name of minimalism and the rule of law.
—July 7, 2021
The ACLU, of which I am the national legal director, represented two organizations that intervened in the case to support the City of Philadelphia. For more on these questions, see my “A New Assault on Marriage Equality,” The New York Review, December 3, 2020. ↩
The ACLU represented Brandi Levy, the student in the case, and I argued it before the Supreme Court. ↩