Aimee Stephens outside the Supreme Court after oral arguments by David Cole in her case challenging the legality of discrimination against transgender employees

Michael Reynolds/EPA-EFE/Shutterstock

Aimee Stephens outside the Supreme Court after oral arguments by David Cole in her case challenging the legality of discrimination against transgender employees, Washington, D.C., October 2019. Stephens died in May.

When the Supreme Court’s 2019–2020 term opened last October, with LGBTQ discrimination, gun rights, abortion, immigration, and state funding of religious schools on the docket, it promised to be both controversial and deeply partisan. The Court then added cases on whether President Trump could block Congress and a grand jury from obtaining his personal financial records, and whether employers who objected to contraception on religious grounds could refuse to provide their employees with insurance that paid for it. This was Justice Brett Kavanaugh’s second term on the Court, and it afforded a real test of how his replacement of the more moderate Anthony Kennedy would alter the Court’s character. Republicans, who had long made picking “reliable” conservative justices a priority, were eager to reap their rewards.

But in the end, the Court’s term was much less conservative than anyone had expected. In many of its most prominent cases, one or more conservative justices joined their liberal colleagues, declaring that federal civil rights law bars LGBTQ discrimination, striking down a Louisiana restriction on abortion, and invalidating President Trump’s effort to rescind deportation protection and work authorization from “Dreamers,” the 700,000 young undocumented migrants who came to the United States as children, and whom President Obama had given protection under the Deferred Action for Childhood Arrivals (DACA) program. Seven justices joined forces to reject President Trump’s contention that he could ignore subpoenas of his financial records, with even his own appointees, Kavanaugh and Neil Gorsuch, ruling against him. And Justice Gorsuch joined the four liberal justices—Ruth Bader Ginsburg, Stephen Breyer, Elena Kagan, and Sonia Sotomayor—to declare that about half of Oklahoma belonged to the Muscogee (Creek) Nation for purposes of authority to prosecute Native American defendants.

The conservatives won their share of decisions, to be sure. The Court restricted immigrants’ rights to judicial review of deportation orders, and required states to fund scholarships for religious schools if they funded secular schools. It excused Catholic schools from abiding by civil rights laws in hiring and firing teachers who provide religious instruction, and upheld President Trump’s order permitting employers to deny their employees contraceptive coverage. And in a series of unexplained or barely explained decisions reviewing requests for emergency stays of lower court orders, the Court four times ruled against efforts to protect voting rights, and twice cleared the way for federal executions by summarily overruling lower courts that had identified serious legal questions that still needed to be resolved.

Still, the primary question the term raised was why liberals prevailed in so many significant decisions. For many commentators, the answer was simple: Chief Justice John Roberts. In all but one of the liberal outcomes in high-profile cases, he was in the majority. But Roberts is famously enigmatic, so identifying him as the crucial swing vote does not provide much of an explanation. A review of the Court’s most important decisions provides further clues.

The most consequential decision of the term was Bostock v. Clayton County.1 The question in Bostock was whether discrimination on the basis of sexual orientation or transgender status is prohibited by Title VII, a 1964 civil rights law that bars businesses from discriminating against employees “because of sex.” There is little doubt that when Congress enacted that law, no member contemplated that the law would protect LGBTQ individuals. Sex between same-sex partners was still a crime in many states. Most federal appellate courts ruled over the years since 1964 that discrimination on the basis of sexual orientation is distinct from discrimination “because of sex” and therefore was not covered by Title VII. And Congress repeatedly failed to adopt proposed amendments to the statute that would have explicitly barred employment decisions based on sexual orientation and gender identity.

Yet in an opinion written by Gorsuch and joined by Roberts and the four liberal justices, the Court held that discrimination on the basis of sexual orientation or transgender status is, by definition, a form of sex discrimination. The logic is simple: when an employer fires a woman for being attracted to women but does not fire a man for being attracted to women, he is treating the employees differently “because of sex.” Similarly, the funeral home that fired Aimee Stephens when she came out as transgender objected to her living and identifying as a woman because she was assigned a male sex at birth. Had she been assigned a female sex at birth, her employer would not have objected. She was treated differently, the Court reasoned, because of her sex, even if “sex” as used in Title VII includes only one’s sex assigned at birth.


The decision interprets a single law, but its reasoning has far-reaching consequences. It should extend to all federal and state laws that prohibit sex discrimination, thus affording LGBTQ individuals sweeping civil rights protections in virtually every area of life, including housing, education, credit, and public accommodations. The Court left open the possibility that religious employers might assert a “free exercise” right to discriminate, but did not address this question because no one had asked it to do so.

Gorsuch justified this liberal result through a method of statutory interpretation, “textualism,” that Justice Antonin Scalia long championed, and that the conservative justices generally endorse. It maintains that the Court is bound by the literal words of a statute, not Congress’s intent or purpose, much less the statute’s “legislative history.” The textualist method has produced liberal results from conservative justices before. As Scalia wrote, in a 1998 case extending Title VII protection to a man harassed by other men on an all-male oil rig, “male-on-male sexual harassment in the workplace was assuredly not the principal evil Congress was concerned with when it enacted Title VII,” but “it is ultimately the provisions of our laws rather than the principal concerns of our legislators by which we are governed.”

In a second major win for liberals, June Medical Services v. Russo, Roberts again joined the four liberal justices to strike down a Louisiana law that required doctors performing abortions to have admitting privileges at a nearby hospital. The law was virtually identical to a Texas law the Court had struck down four years earlier in Whole Woman’s Health v. Hellerstedt. In that 5–3 decision, Kennedy had provided the swing vote, over Roberts’s dissent. Yet in June Medical, Roberts chose adherence to precedent, or stare decisis, over his own view of the law’s constitutionality. He made clear that he continues to believe that Whole Woman’s Health was “wrongly decided.” But despite the fact that his four conservative colleagues would have gladly joined him to overturn it, Roberts blinked.

Stare decisis is a fundamentally conservative idea. It reflects a commitment to follow the collective wisdom of the past. As Roberts wrote, quoting the English jurist William Blackstone, stare decisis is designed “to keep the scale of justice even and steady, and not liable to waver with every new judge’s opinion.” Roberts added that the doctrine expresses humility, furthers predictability, and guards against arbitrary judicial discretion. Stare decisis is not an “inexorable command,” he cautioned, and some of the Court’s most celebrated decisions, including Brown v. Board of Education, have overturned precedents. But in June Medical, Roberts’s commitment to precedent prevailed over his own deeply held views on the specific constitutional question.

Roberts took pains to note that Louisiana had not asked the Court to reconsider Planned Parenthood v. Casey, the 1992 decision that had rejected an effort to overturn Roe v. Wade, which had first established women’s constitutional right to abortion in 1973. That request will soon come, as anti-abortion forces, emboldened by Trump’s appointments to the Court, have increasingly enacted laws that ban abortion in most circumstances and can be upheld only if Roe is overturned. But if Roberts follows his reasoning in June Medical, the right to reproductive choice should survive. The stare decisis argument for upholding Roe is much stronger than that for upholding Whole Woman’s Health, as Roe has been law for nearly fifty years and has already been reaffirmed once. Most importantly, women have long relied on Roe’s protections to help plan their childbearing and pursue their lives; as Justice Ginsburg has argued, it is an essential part of women’s equality. If Roberts was willing to adhere to a recent precedent from which he strongly dissented, he should reaffirm Roe and Casey as well.

But while June Medical was a win for stare decisis and reproductive rights, it was a qualified win. Even as Roberts claimed to be bound by Whole Woman’s Health, he wrote separately to repudiate part of that decision. In Whole Woman’s Health the Court held that if the burdens a law imposes outweigh its putative benefits, it is unconstitutional. Roberts rejected such a “balancing” test and maintained that laws violate the right to abortion only if they have “the purpose or effect of placing a ‘substantial obstacle’ in the path of a woman seeking abortion…or are not reasonably related to a legitimate state interest,” a standard announced in Casey. Roberts contended that the balancing test is unprincipled, because the values on either side are often incommensurable. How does one balance a woman’s interest in bodily autonomy and equality against a state interest in potential life? Quoting Scalia, he argued that the inquiry would be like “judging whether a particular line is longer than a particular rock is heavy.”


So while Roberts agreed to be bound by the result in June Medical, he did not deem himself constrained by its reasoning. Stare decisis, however, is a commitment to apply the same principles in the next case, not just to reach the same result. We’ve seen this sort of move before. In Casey, the Court rewrote Roe v. Wade in the act of affirming it. The resulting “undue burden” test allowed states to take further steps to restrict abortion in the name of protecting both women’s health and potential life. And given Roberts’s willingness to redefine what he’s “bound” by even as he claims his hands are tied, the biggest risk is not that Roe will be overturned entirely but that the Court will “redefine” it in ways that leave it a mere shadow of what it once promised. Still, in this term, the right to abortion survived.

In the biggest surprise of the term, Department of Homeland Security v. Regents of the University of California, the Court invalidated President Trump’s attempt to end the DACA program. Again Roberts provided the fifth vote, siding with the liberals over his conservative colleagues. The case seemed a long shot in the Supreme Court. President Obama had defended the legality of the program on the ground that the president has discretion in how to enforce immigration laws, and that the decision to “defer action” on these individuals’ immigration status was encompassed within that discretion. But if Obama had the discretion not to enforce immigration law against the Dreamers, surely Trump had the discretion to resume enforcement.

Roberts did not dispute that Trump could have ended DACA. Instead, he found that the way he did so was “arbitrary and capricious,” largely because the Department of Homeland Security failed even to consider the substantial “reliance interests” of the DACA recipients, who had enrolled in school, launched careers, started businesses, and even married and had children in the expectation that they could rely on the program’s continuation. These expectations, Roberts made clear, would not necessarily preclude the program’s revocation, but at a minimum they had to be considered.

The decision also repudiated Trump’s attempt to deflect responsibility for ending DACA to the courts. His administration claimed that it had to end the program because a federal appeals court had ruled a similar immigration program illegal. Roberts rejected that conclusion, noting that the appeals court decision, whether or not it was correct, left plenty of room for the administration to make decisions about how swiftly to end DACA and what parts it might keep. In doing so, Roberts put responsibility for ending the program squarely on the administration, where it belongs.

President Trump has already made noises about seeking to rescind DACA again, this time with more fulsome reasons. But that effort is likely to be tied up in the courts at least until the next administration, so the fate of the Dreamers is now, for all practical purposes, in the hands of the voters.

On the last day of the term, the Court issued two decisions rejecting Trump’s contention that he is immune from subpoenas that a New York grand jury and three congressional committees had issued for his financial records as a private citizen, including his long-elusive tax records. Roberts wrote both decisions, and this time was joined not just by the four liberals but also by Gorsuch and Kavanaugh. Indeed, in rejecting Trump’s argument for absolute immunity from state legal process, the Court was unanimous. (Justices Clarence Thomas and Samuel Alito dissented only because they felt that state prosecutors should have to meet a heightened standard of justification to subpoena the president.) In this, the most politicized dispute it faced, the Court succeeded in rising above the partisan politics that shape so many responses to this president.2

Chief Justice John Roberts and Associate Justices Elena Kagan, Neil Gorsuch, and Brett Kavanaugh

Leah Millis/Getty Images

Chief Justice John Roberts and Associate Justices Elena Kagan, Neil Gorsuch, and Brett Kavanaugh at President Donald Trump’s third State of the Union address, Washington, D.C., February 4, 2020

In the New York grand jury case, Trump v. Vance, Roberts opened his decision with classic understatement and clarity: “In our judicial system, ‘the public has a right to every man’s evidence.’ Since the earliest days of the Republic, ‘every man’ has included the President of the United States.” Recounting with a historian’s relish and flair the treason trial of Aaron Burr, Roberts noted that Burr subpoenaed President Thomas Jefferson in his defense, that Jefferson asserted immunity, and that Chief Justice John Marshall, presiding over the trial, overruled the objection. Marshall acknowledged that under English common law, the king could not be subpoenaed. But, Roberts explained, quoting Marshall, “a king is born to power and can ‘do no wrong.’ The President, by contrast, is ‘of the people’ and subject to the law.”

In the congressional subpoena case, Trump v. Mazars, also decided by a vote of 7–2, with only Alito and Thomas dissenting, the Court rejected the president’s claim that Congress should have to make a heightened showing of justification, akin to what is required to overcome the executive privilege that protects confidential presidential deliberations, even though these subpoenas sought only his private papers. But the Court also dismissed Congress’s contention that its power to demand a president’s private papers is limitless. It reaffirmed that Congress has broad authority to issue subpoenas, but ruled that concerns about separation of powers warranted more careful judicial scrutiny to ensure that Congress has a legitimate purpose, that it can’t get the information elsewhere, and that the request is not overly burdensome. If you are skeptical about these limits, ask yourself whether you’d want Senate Majority Leader Mitch McConnell to have unchecked authority to demand private records from President Joe Biden.

Some have maintained that as a practical matter, Trump actually won, because, pending further legal proceedings in the lower courts, his tax records will not be made public before the November election. But that was never a legitimate purpose of the subpoenas, and those who expected to see the records as soon as the Supreme Court ruled failed to take into account the snail’s pace at which courts operate. In any event, the far more important issue was whether a president could put himself above the law and beyond oversight. The answer to that question was a resounding, and glorious, no.

Why did such a conservative Court reach such liberal results? Four factors seem to have played a part. First, the liberal causes that prevailed largely did so by appealing to conservative values. Textualism, stare decisis, the need to provide reasons when upsetting settled expectations, and the notion that we are all equal before the law are fundamentally conservative ideals. Many are also liberal ideals, of course, but to prevail before this Supreme Court, liberals have to be able to speak across the aisle. That’s a good thing, and this term, they did it.

Second, many of the liberal victories came on issues that had generated intense public support, buoyed by decades of political organizing and advocacy. The Dreamers have organized as a group and gained widespread support from universities, businesses, state governments, and colleagues. The public is divided on abortion, but thanks to long-standing efforts of women’s rights and reproductive freedom groups, the majority oppose overturning Roe v. Wade.

Nowhere was the relevance of political organizing more evident than in Bostock. What made the argument that LGBTQ discrimination is inherently sex discrimination more “logical” in 2020 than it had been in the preceding fifty years has much more to do with changes in American society than with analogical reasoning. In that time, gay, lesbian, bisexual, and transgender people have increasingly come out, formed political organizations, and asserted their rights to be treated with equal dignity and respect. At the oral argument in Bostock, Gorsuch asked me what the Court should do if it concluded that we had a good textualist argument but that ruling for us would cause “massive social upheaval.” My response was that the decision would not in fact cause social upheaval; LGBTQ individuals lived among us, were present in the courtroom, and the sky had not fallen. But that response was credible only because of the political progress LGBTQ organizations had achieved over the last half-century of hard-fought struggle.

Third, President Trump’s attacks on judges, and on the values that conservative and liberal judges alike prize, may have contributed to the Court’s decisions. In 2018 Roberts, who is ordinarily very reserved, took the highly unusual step of responding to a Trump attack on judges. Roberts insisted that

we do not have Obama judges or Trump judges, Bush judges or Clinton judges. What we have is an extraordinary group of dedicated judges doing their level best to do equal right to those appearing before them.

In each of the high-profile cases in which the Court ruled for liberals, it pointedly rejected the Trump administration’s often radical arguments. It stood for conservatism with a small “c” against Trump’s extremism—especially in the subpoena decisions, in which the Court stood up for the rule of law against the president’s assertion of personal privilege.

Finally, and most important, the hyperpartisan character of the country surely prompted the Court to distinguish itself—to liberals’ advantage. While nothing dissuades legislators and executive branch officials from acting in blatantly partisan ways—indeed, we generally expect them to do so—the justices would risk losing their legitimacy if they did. The notions that a judge should be open to argument and should decide cases based on law and reason rather than partisan or political considerations may seem quaint and naive in a highly skeptical age. But they are central to the very idea of judging. At Roberts’s confirmation hearings, he argued that, as an advocate before the Court, he would be “very frustrated” if “it all came down to politics.” It would not just be frustrating for the lawyers but deeply corrosive of the institution itself, a danger that Roberts understands perhaps better than any of the other justices. Largely as a result, the Supreme Court is the only federal institution that looks good in the summer of 2020.