Among the many talents of John Roberts was an ability to mask his ambitions. Though conversational in chambers, about the day’s headlines or what he did over the weekend, he was by nature reticent. Law clerks around the building, both his own and others, didn’t feel they got to know him as well as, say, Clarence Thomas or Steve Breyer. So when the chief justice a few years ago let his guard down—unprompted—it astonished his listeners. And it revealed a hunger for greatness that went along with an abiding affection for the institution he led.
Several clerks were taking Roberts to lunch a few blocks from the Court. On the walk, one innocently asked him, “How do you like the job?” Instead of something like “It’s the privilege of a lifetime,” he offered a telling response. Roberts began by reminding the clerks that there had been only 16 before him who had occupied the center seat. In that sense, he was delighted to be No. 17. But Roberts understood the history of the Court and of the nation. Even among the chiefs, he observed, there had been only one John Marshall, who served for 34 years at the beginning of the 19th century and wrote the seminal Marbury v. Madison, which solidified the Court’s place in the constitutional blueprint. Marshall “had the opportunity to decide the great questions because the Constitution was undeveloped,” Roberts told the clerks.
“It’s not like that anymore,” he said. “I was born in the wrong era.”
Perhaps not quite so. The chief might yet have the chance to be great. In his early 60s, he was still the third-youngest member of the Court. If he served until he was 80, he wouldn’t even reach the halfway point of his tenure until 2020.
Consider the near-term future of the Court. Trump’s newest nominee would veer it further to the right, joining conservative stalwarts Gorsuch, Thomas and Alito. With Ginsburg, Breyer, Sotomayor and Kagan on the left, that means the new swing justice would be the chief. The newly constituted Court is truly the Roberts Court.
Roberts seemed to appreciate what greatness in his own time looked like. His votes on gun control, campaign finance and voting rights were indeed injudicious intrusions on democracy and ran counter to his exhortations about how same-sex marriage ought to be left to the political process. But in 2012, in the signal case of Roberts’s time on the Court and the most important appeal since Bush v. Gore—on the fate of Obamacare—the chief justice might have proven himself a worthy descendant of Marshall.
As the decisive vote in NFIB v. Sebelius, the 5-to-4 ruling that upheld the Affordable Care Act (ACA), Roberts kept the Court out of the political maelstrom. The law in question was the signature policy achievement of Barack Obama’s presidency. Lower courts had disagreed on the constitutionality of the law’s so-called individual mandate, which required most Americans to have health insurance or to pay a penalty based on income and what basic insurance would have cost them. (Congress ultimately repealed the mandate in late 2017, with Trump’s rabid support.) The justices agreed to resolve the conflict. Had the justices invalidated the law, the Court could well have become the overriding issue in the November 2012 presidential election between Obama and Mitt Romney. Roberts understood the institutional toll that might take on the Court.
The individual mandate aimed to address a dilemma: Millions of people participated in the health care market, yet failed to have health insurance. They were able to obtain health care, for example, because hospitals, legally and ethically, couldn’t turn away emergencies. Some of those uninsured people paid their bills. Most did not—those consumers received services for nothing. The costs wound up being absorbed by others, which wound up raising the price for those who did have insurance. The mandate made almost everybody pay, one way or another, for their participation in a market that involved trillions of dollars and more than 15 percent of the U.S. gross domestic product. Opponents of the ACA—those who had lost the legislative debate, in the arena of democracy—went to court to undo the law. The constitutionality of the mandate turned on two questions: Did Congress have power under the Commerce Clause to coerce individuals to engage in a specific activity? And could the ACA’s penalty qualify as a “tax,” which therefore could be levied under the Constitution’s Taxing Clause (which stated “Congress shall have power to lay and collect taxes”)?
Most constitutional experts thought the first answer was easy: yes. That rendered a second answer unnecessary. But the oral arguments at the Court in March 2012, over the course of three days, showed otherwise on both questions. If people chose not to buy insurance, different justices asked, how could those people be participating in the health care marketplace? By that logic—inaction amounts to action—there were few bounds to government regulation. Scalia pushed the point with an analogy that seemed specious, but then was invoked three more times during the arguments. “Everybody has to buy food sooner or later,” he said. “Therefore, everybody is in the market. Therefore, you can make people buy broccoli.” The bit wasn’t Scalia’s spontaneous invention. He was echoing political conservatives who had been making the broccoli argument for three years. Justices borrowed ideas from the academy and beyond all the time. But to the extent it looked like a member of the Court was doing the bidding of Rush Limbaugh and the Wall Street Journal editorial page, it was unseemly.
The broccoli analogy was far-fetched, even if Kennedy wound up steaming about it in chambers in the weeks after the arguments. The failure of some people to buy a vegetable hardly had an impact on the overall food market; and for broccoli itself, reduced demand might drive down the price. But health care spending was a very different market. As the government lawyer defending the ACA, Solicitor General Donald Verrilli, explained: “Virtually everybody” is “either in that market or will be in that market,” and its “distinguishing feature” is that “people cannot generally control when they enter that market or what they need when they enter.”
But the simpler response might have been, sure, it would be theoretically fine for Congress to require people to buy broccoli (or be penalized for not doing so)—if legislators rationally concluded that diet affected health, health affected interstate commerce, and trying to foster intake of green veggies would improve health. A National Eat-Your-Broccoli Act would be silly, but not unconstitutional.
The penalty-qua-tax question was even less complicated, despite the penalty being a tool to affect conduct rather than to raise revenue. Obama and various lawmakers strained to say the penalty wasn’t a tax, but that scarcely meant it couldn’t be construed as an exercise of Congress’s taxing power. (Opponents of the ACA did call the mandate a tax.) Nor did it matter that the ACA itself failed to designate the penalty a tax. Judges routinely cut through labels to get at what a statute or executive order really meant. Word games by politicians didn’t dictate constitutional law. A rose was still a rose.
The biggest problem for the mandate was that the Obama administration had to contradict itself legally. Before insisting on Day 2 of the arguments that the penalty for refusing to buy health insurance was a tax, the administration on Day 1 claimed the penalty was not a tax. That’s because the obscure 19th-century federal Tax Anti-Injunction Act prohibited litigants from challenging a tax before they actually paid it. Since the ACA penalty hadn’t yet taken effect, the Tax Anti-Injunction Act barred a current lawsuit if the penalty was a tax. The administration wanted the uncertainty resolved, so for purposes of the Tax Anti-Injunction Act, it had to argue the penalty wasn’t a tax.
Justice Alito was quick to point out the incongruity. “Today, you are arguing that the penalty is not a tax,” he told the SG. “Tomorrow, you are going to be back and you will be arguing that the penalty is a tax?”
Alito knew better. Legal interpretation, both constitutional and statutory, often required semantics—otherwise called judging. It was the only sensible way to deal with text and subtext. The legal battleground in NFIB v. Sebelius was just another chapter in the debate between the literalism championed by Scalia and the pragmatism advocated by Breyer. The debate was always a bit bogus. All judges engaged in interpretation—the only issue was what they called it and the degree to which they conceded it.
Roberts himself, no fan of either constricted literalism or unbounded pragmatism, seemed to admit as much when he pressed one of the lawyers challenging the ACA. “The idea that the mandate is something separate from whether you want to call it a penalty or tax just doesn’t seem to make much sense,” he said. The chief was foreshadowing how he would come out in the case—and with much accompanying drama.
At the justices’ conference on March 30, two days after arguments, Roberts took his seat at the head of the oblong walnut table, under three crystal chandeliers. In this private setting, within the suite of rooms that made up the chief’s chambers, neither he nor his colleagues wore their robes. At the far end of the baize-topped table, opposite the chief, was the most senior associate justice, Scalia. Seated to the chief’s right were the next three most senior: Kennedy, Thomas and Ginsburg. To the chief’s left were the four most junior: Breyer, Alito, Sotomayor and Kagan.
Right behind Scalia was the room’s black marble fireplace, with a matching clock on the mantel. Above the mantel, directly facing Roberts, hung a circa-1834 portrait of Chief Justice Marshall. Roberts himself had selected the various portraits in the room, which also included those of Justices Robert Jackson, Benjamin Cardozo and the first John Harlan. (Cardozo, for example, got wall space because he was a New Yorker, just like three current justices: Ginsburg, Sotomayor and Kagan. Roberts chose Jackson because he so revered his writing.) The chief particularly liked the idea of the venerable Marshall gazing down at the conference, perhaps a muse for the current members of the Court. As always, the conference after NFIB v. Sebelius began with handshakes all around—a tradition that dates to the late 19th century. The custom was a reminder that the justices all had the same goals in mind, though Justice James F. Byrnes described it as more like nine boxers about to go to their corners before coming out punching.
Even though the chief owned but one vote, he had an important role at conference. He summarized each case, getting a chance to frame the issues. He went first in the discussion and had the opportunity to cast the opening vote. And when he was in the majority, he determined who wrote the opinion, even though longer-tenured associate justices might be part of the majority.
Roberts outlined the difficult questions raised in the arguments. He carefully acknowledged competing positions on the tax question. But he was certain the mandate was impermissible under the Commerce Clause, so he voted to strike down at least that part of Obamacare, concluding as well that the Taxing Clause couldn’t save the mandate. (Whether the rest of the law could survive in the absence of the mandate was left unresolved for now.) Scalia went next—another vote against the mandate (and with it the entire law). Kennedy and Thomas agreed with Scalia’s position—two more against. One more vote was needed. While Ginsburg and Breyer then voted to uphold the mandate, Alito sided with Scalia, and the conservatives had their majority, even with Sotomayor and Kagan joining Ginsburg and Breyer. The chief assigned himself the opinion.
However, that’s where the storyline got fuzzier. Based on Roberts’s vote, the other justices, immediately after the conference, walked back to their respective chambers and reported to their clerks that the Court was throwing out the most important piece of social legislation in two generations. Clerks reacted predictably—with elation or despair. But Roberts had doubts on the Taxing Clause question, even if he didn’t identify them that way in the conference. Votes there are tentative. Justices aren’t locked in. They may re-evaluate based on what someone else’s opinion looks like, or if they’re writing for the Court themselves, they may discover “the opinion doesn’t write.”
That’s what happened to Roberts during the next two months. He instructed different clerks in his chambers to write alternative drafts of an opinion, the different outcomes turning on the tax question. He finally decided that the mandate could be considered a tax and therefore represented a legitimate congressional exercise of the taxing power. The Commerce Clause aspect became irrelevant. He revealed his new position to the other justices during a subsequent conference. Much as Roberts didn’t like how the ACA became law, he thought it was constitutional. On June 28, 2012, the Court announced its decision.
It was the first time the chief had joined the four liberals in a 5-to-4 vote, let alone a ruling that big. (One scholar said it was more like a 4–1–4 decision.) “The text of a statute can sometimes have more than one possible meaning,” Roberts wrote. “And it is well established that if a statute has two possible meanings, one of which violates the Constitution, courts should adopt the meaning that does not do so.” Bearing in mind that principle of interpretation, he said the mandate “may reasonably be characterized as a tax. Because the Constitution permits such a tax, it is not our role to forbid it, or to pass upon its wisdom or fairness.” Even as he granted Congress its due, his opinion scorned congressional ineptitude, suggesting, as one scholar observed, “the fools couldn’t even figure out how to structure [the ACA] to render it constitutional.”
Roberts viewed what he had done as diligent judging. The other four conservatives thought otherwise. Had Roberts voted initially with the liberals, the conservatives obviously would have been disappointed. But to them, Roberts had caved. Scalia vented about it in a way that surprised even those at the Court who were used to his eruptions. In private, some of Roberts’s defenders suggested he didn’t actually reverse his position because he hadn’t committed in the first place. But internal Court correspondence and draft opinions themselves make clear he indeed had reconsidered. It wasn’t that Roberts had decided his initial view was wrong—only that an alternate view was permissible. Because that alternate view deferred to Congress—and in favor of constitutionality—the chief concluded he was required to adopt the alternative view.
Conservatives were so livid that they wrote Roberts off as another David Souter—disloyal to the president and the party who gave him a lifetime seat. According to that narrative, in 1992, only two years after President George H.W. Bush put Souter on the Court, Souter had voted the “wrong” way in Casey, the abortion ruling that mostly upheld Roe v. Wade; Justices Kennedy and O’Connor also stood convicted of a double-cross on Roe. And it was Kennedy who Scalia, Thomas and Alito thought was most likely to stray in NFIB v. Sebelius. In fact, it was Kennedy who devoted the most energy to trying to win Roberts back, even as Roberts tried to convince Kennedy to join him.
Doctrinally speaking, Roberts’s reasoning wasn’t the most persuasive—any more, for example, than Earl Warren’s had been in Brown v. Board of Education. If one tried to diagram Roberts’s Obamacare opinion, it might resemble a pretzel. But his logic was good enough, and it kept the Court out of a political quagmire that could do it grave injury. What the Court did, and what it did not do, weren’t robotic commands. Decisions required judgment, prudence, timing. In that sense, writing an opinion was akin to doing what the best umpires did. Stephen Jay Gould, the evolutionary biologist, once famously explained why it had been right for the umpire Babe Pinelli to call a pitch—quite outside—a strike to finish off Don Larsen’s perfect game in the 1956 World Series. “Context matters,” Gould wrote. “Truth is a circumstance, not a spot.”
Roberts’s point that the Court wasn’t opining on the ACA’s “wisdom or fairness” was a sop to the right, which he repeated when he announced the decision in the courtroom. He emphasized the ruling “is not in any way based on our judgment about whether the Affordable Care Act is good policy.” Instead, as he wrote in his opinion, that judgment, “under the Constitution,” was “reserved to the people.”
To his conservative critics, that naturally wasn’t enough. The chief justice hadn’t only voted wrong. He’d committed the sin of changing his mind—the second “switch in time that saved nine” at the Court. (The first had been in 1937 when another Roberts on the Court, Owen Roberts, reversed sides on the constitutionality of New Deal legislation.) Ever since, conservatives have smoldered over what they saw as John Roberts’s betrayal. These resentments were what led to Trump’s fulminations against Roberts during the 2016 campaign.
The day after the Obamacare ruling, Roberts exhibited his usual sense of humor. Attending a conference of lower-court judges, arriving early, he found himself alone at the head table. “It’s lonely here, all by myself,” he sardonically remarked to a friend. Later, playing to a big audience, Roberts mentioned he was headed off to Malta to teach a two-week course on American law. “Malta, as you know, is an impregnable island fortress,” he explained. “It seemed like a good idea.”
The conservative justices and their clerks knew Roberts had switched and someone let word slip out. Such finger-pointing leaks—reported a month before the ruling, by the conservative columnist George Will and others—were unprecedented. The Court simply didn’t let anything about pending decisions get out. But the fact of the leaks reflected how much the conservative justices felt they had been duped. It may also have been a clumsy attempt to intimidate Roberts back into the fold.
Conservatives spun such efforts as merely a response to shameful political pressure from lefty law professors, senators and Obama himself. “Taunting” and “get[ting] under Roberts’s skin”—“politics at its filthiest”—was “beneath the dignity of the Court,” a conservative Washington Post columnist called it weeks before the ruling. Never mind that the widely read piece by Kathleen Parker itself could fairly be called an attempt at working the ref. All of it, on both sides of the aisle, was just more evidence that smart people now saw the Supreme Court as a political body susceptible to lobbying that extended beyond the legal briefs.
Four days after the Obamacare ruling, the most dramatic leak of all happened. Citing “two sources with specific knowledge of the deliberations,” Jan Crawford of CBS News reported that Roberts had changed his mind. If those sources were the conservative justices or clerks, the leak no longer could serve to win Roberts back. Now, it was just about trying to embarrass him.
At any rate, Roberts hardly had cast his lot fully with the Court’s liberal bloc. Much of his opinion focused on the Commerce Clause, which he found proscribed the mandate. That finding didn’t alter the outcome of the appeal because Roberts held that Congress still had authority under the Taxing Clause to enact the mandate. But his forceful views on the limits of the Commerce Clause had the faraway potential to radically shrink the power of the federal government to regulate the workplace, promote social welfare, set education policy, and more. Since the New Deal, the Court had almost always upheld legislation against Commerce Clause challenges. Roberts’s opinion questioned nearly eight decades of constitutional law. The mandate “does not regulate existing commercial activity,” he wrote. “It instead compels individuals to become active in commerce by purchasing a product, on the ground that their failure to do so affects interstate commerce. Construing the Commerce Clause to permit Congress to regulate individuals precisely because they are doing nothing would open a new and potentially vast domain to congressional authority.” That, he said, was not what the Framers had in mind.
Although the four other conservatives declined to join that part of his opinion—or even acknowledge Roberts had written on the Commerce Clause—their views on the subject coincided with his. He was with them more than he was against them. Theirs was a counterrevolutionary view of the Commerce Clause. Many scholars pointed out there weren’t a lot of federal laws attempting to compel activity. The “broccoli horrible,” as Ginsburg mocked it, was less an actual possibility than a caricature of Big Government—Alice Waters meets George Orwell.
Even so, Roberts’s opinion on the Commerce Clause was significant for two reasons. First, he chose to write it at all. There was no need to do so, since the appeal was resolved on the ground of the taxing power. After all, this was the chief who had often said the Court should be in the business of deciding no more than it had to. “Our usual practice,” he had counseled in Northwest Austin, the 2009 case on the Voting Rights Act, “is to avoid the unnecessary resolution of constitutional questions.” Three years before, in a commencement speech at Georgetown, he stressed, “If it’s not necessary to decide more to dispose of a case, in my view it is necessary not to decide more.” Whatever ruminations he’d had about the Commerce Clause were superfluous. Second, he was providing a dog whistle to both libertarians and conservatives that a majority of the justices might be receptive to fresh attempts to curtail federal regulatory authority. By giving Congress and the president what they sought at that instant—but laying down a long-term marker on federal power—Roberts was imitating what John Marshall had deftly executed in Marbury v. Madison 209 years earlier.
Marshall’s middle-ground ruling found the new administration of President Jefferson had wrongly acted under the federal law at issue. But a constitutional provision on jurisdiction deprived the Court of the power to do anything about it. Jefferson, a political adversary, thus had no incentive to object, even as Marshall, unconstrained, established the authority of the Court to strike down acts of Congress that the justices said violated the Constitution. Marshall put himself “in the delightful position…of rejecting and assuming power in a single breath,” as a leading historian of the Court extolled it. Marbury was “a masterwork of misdirection, a brilliant example of Marshall’s capacity to sidestep danger while seeming to court it.”
Similarly, in NFIB v. Sebelius, Roberts, avoiding a confrontation with a sitting president, was able to rein in Commerce Clause doctrine on the sly—“to advance in one direction while his opponents are looking in another,” as Bickel had praised Marshall. A few conservative politicians appreciated what Roberts had pulled off. Most did not. And whatever points he had registered on the constitutional scoreboard would appear only in the distant future.
At the White House, Obama was ecstatic the ACA had survived, even if part of the ruling limited the law’s expansion of Medicaid. (Seven justices agreed that Congress had overstepped its authority by trying to coerce states into joining the expansion; if states had refused, they risked losing even their existing funding.) The president was relieved as well. But privately, he professed intellectual admiration for the chief’s masterstroke, as much as he disagreed with half of it. “He’s clearly playing the long game,” Obama told Vice President Biden as they and several aides in the Oval Office pored over the 187 pages of opinions.
Obama had always thought the institutionally minded Roberts was more likely than Kennedy to provide a fifth vote to uphold the ACA. And Obama himself, early on in the litigation in various federal courts, had signed off on the Justice Department’s request to argue the mandate was a tax, even though he’d gone on Good Morning America to assert it was not. Obama understood that Roberts was more likely than Kennedy to buy into the tax argument. But Obama had not anticipated Roberts’s next beat, on the Commerce Clause. Of the various presidential speeches that had been prepared in advance of the rulings, none contemplated exactly what the chief ended up doing.
It may well be that Roberts voted to uphold Obamacare chiefly as a way to disguise remaking the Commerce Clause. But such a ploy has yet to pay dividends in any new rulings, and it won him few new admirers on the right. On the Court itself, his conservative colleagues gave him no credit for threading a needle. Moreover, if Roberts really believed the ACA was suspect, he had the chance in another appeal three years later, King v. Burwell, to kill it. But with Kennedy also joining the majority this time, Roberts declined to do so.
There’s a more basic, less Machiavellian explanation for what motivated him. He thought NFIB v. Sebelius was a close case. On the one hand, he believed that Commerce Clause doctrine was not so boundless as to justify regulation based on inactivity—and given the attention that issue had received in the lower courts, he had to address it. But on the other hand, the Taxing Clause was probably broad enough to allow the mandate. In a close case, Congress deserved the benefit of the doubt. Seeing himself as the steward of the Court’s legitimacy, Roberts chose to keep the Court out of harm’s way. He had failed to do so in Shelby County, in Citizens United and in Heller, but now perhaps, at least in this one case, four months before a presidential election, he had rethought his role and become a judicial statesman—an advocate not for one or another issue, but for the Court itself.
What Roberts actually wrote offered no hint of his motivation, and he of course didn’t confess he had switched sides. During oral arguments, he had already indicated that mere labels didn’t end the inquiry. Nor was it conclusive that the Obama administration claimed the mandate wasn’t a tax under the Tax Anti-Injunction Act. After the conference, he simply rethought the legal issues, deciding the appeal on the merits.
But yes, in a close case, more than the particular issues, he made the Court’s institutional needs paramount. Roberts hadn’t been on the Court for Bush v. Gore, but he had seen the price the Court paid for the ruling. Maybe the chief was acting with the judicial humility he had counseled for others—judicial “politics” in the best sense. If so, in an age when the Court has become ascendant, Roberts’s destiny might be, as he told clerks, to do great things.
Maybe.
With a Republican in the White House through at least January 2021, conservatives saw the Court as their best way to accomplish policy ends. (Liberals no doubt would’ve done the same thing if Hillary Clinton had won and Democrats had taken back the Senate.) Even as Donald Trump floundered, members of his party repeatedly touted Justice Neil Gorsuch as the greatest achievement of the Trump administration. Now, with another appointment to the Court, following Kennedy’s retirement, Republicans saw the opportunity, finally, to assume conservative control of the third branch.
As that possibility increased, much of the legal right discarded talk of merely returning the Court to its more minimalist roots or merely undoing such rulings as Roe. With five votes, such changes were too modest. Now the goal would be to get the justices to recognize new constitutional rights—to do precisely what conservatives had denounced liberals for doing for decades. Conservatives of course didn’t admit it would be just more judicial activism. Instead, many rebranded their aim as “judicial engagement.”
Judicial engagement, a phrase that went back only to 2011, dismissed the idea that the democratic branches deserved any overweening deference if they infringed on freedoms. “The rightful liberties of the people are so capacious they cannot all be enumerated or listed,” explained Randy Barnett, a leading libertarian at Georgetown law school. Among other things, judicial engagement adored the open-ended Ninth Amendment, which stated that the existence of certain rights spelled out in the Constitution (like freedom of expression or freedom of religion) didn’t preclude other, unlisted ones that were “retained by the people.” Judges should be the ones to protect those natural rights, according to the theory.
The chief proponent of judicial engagement was the Institute for Justice, the libertarian public interest shop and “merry band of litigators” that had been behind Heller. In establishing its Center for Judicial Engagement, the institute said its mission was to “reinvigorate the courts’ role as bulwarks of liberty.” It urged the appointment of avenging judges who would “fully enforce the limits our Constitution places on the government’s exercise of power over our lives,” rather than “abdicate” their duties by showing “misguided deference to other branches.” Deference, the institute explained, was “fake judging.” In the age of Trump, it has been especially active in trying to seed lower courts with test cases that someday the Supreme Court might use to undo accepted doctrine on federal authority.
The problem with judicial engagement was determining how a conservative justice’s view of “rightful liberties” could properly be distinguished from a liberal justice’s view of “rightful liberties.” As Justice William Brennan had taught, with five votes, you could do anything. To its credit, the institute was consistent. It supported the gay couples in Obergefell and took on free speech causes dear to liberals. But in the hands of judicial engagers, the relevant liberties usually centered on the chains of economic regulation—like minimum-wage and maximum-hour laws, rules on health and safety in the workplace, environmental statutes, a government directive to buy health insurance, zoning restrictions on land, and mandatory participation in Social Security. Judicial engagement sounded a lot like restoration of the Lochner era, when the Court, in the name of capitalism, struck down New Deal legislation. In justices like Gorsuch, Thomas and Alito, judicial engagement had friends.
Roberts had his complaints about federal regulatory power. He made that much clear when he said the ACA violated the Commerce Clause. He, too, might yet side with radicals on the Court who want to scupper the Voting Rights Act entirely, to topple what was left of campaign finance regulation, and to protect carrying guns in public. But at the end of the day, the chief’s instinct for preserving the Court’s prestige had prevailed in the Obamacare ruling. And it might also occur to him that repositioning judicial aggrandizement as judicial engagement ought to be resisted, in defense of the Court.
If so, then Roberts might well be hesitant to overturn Roe—despite his disapproval of the ruling. Whatever grief he suffered by alienating himself further from conservatives wasn’t worth the cost to the Court of looking like the handmaid of radicals. In short, the conventional wisdom about the pending demise of Roe in a post-Kennedy Court was hardly obvious.
In the 2017–18 term of the Court, Roberts, as well as other justices, sent mixed signals on any embrace of humility. We’re modest, except when we’re not. On the one hand, several rulings seemed to indicate the Court was willing to defer to other branches of government: to Trump on his travel ban and congressional indifference about it; to states that engaged in bald-faced partisan gerrymandering, as well as gerrymandering based on race; to states that wanted out-of-state Internet merchants to collect sales taxes; and to states that wanted to aggressively purge their voting rolls of nonvoters. But on the other hand, the Court was perfectly happy to ignore choices by other governmental actors—most notably at the state level, about discrimination based on sexual orientation and about what medical information had to be provided to pregnant women—as well as, on one major occasion, its own 41-year-old precedent concerning public unions.
Pollyanna would say the rulings merely reflected nine conscientious jurists grappling with the imprecision of constitutional law. The more accurate assessment, though, would acknowledge that the 2017–18 term was a pageant of hypocrisy, driven less by an honest reckoning of hard cases than by conforming to a conservative political agenda. So, the Court gave the thumbs-up to:
the exercise of vast power over immigration by a Republican in the White House;
established political parties, which typically meant keeping in office more Republicans than Democrats;
claims of religious freedom prevailing over discrimination laws protecting gays (in the closely watched case involving a Colorado baker of wedding cakes); and
claims of pro-life free speech prevailing over the medical interests of pregnant women.
And it was thumbs-down to:
the idea that President Trump might have been motivated by anti-Muslim bias in enacting his travel ban;
Democratic votes that had been diluted by reapportionment maneuvering;
minorities inhibited from voting or whose voting power was diluted by reapportionment; and
organized labor (which leans toward Democrats and generously funds them).
That last decision—the precedent-busting Janus v. AFSCME—typified the conservative majority’s goals. The issue was whether government employees who opted not to join unions could still be compelled to help pay for collective bargaining and grievance procedures that benefited them. More than 20 states have laws requiring such “agency fees,” which restrict freeloaders. In 1977, the Court unanimously blessed such laws. The Roberts Court reached out to say otherwise. Writing for a 5–4 majority, Justice Alito said agency fees violated the First Amendment: The nonmembers of a union were being forced to subsidize the speech of others.
But that wasn’t exactly true. The 1977 ruling already barred agency fees going to overt political activities. Negotiating a collective bargaining agreement hardly sounded like the same thing. It’s a question of degree. If government mandates that all lunch counters and hotels must be desegregated, are the First Amendment freedoms of association and speech of business owners violated purely because those individuals don’t like the idea of integration? Are the rights of a candy bar manufacturer infringed when it has to include a calorie count on the label? Few would see constitutional violations in either situation. Besides, government employees already give up some First Amendment rights because there are some things they’re not allowed to say on the job. Nonetheless, the Court in Janus found a First Amendment violation. No longer was the First Amendment a shield for the powerless—it was now a sword to be wielded by big business, just as it had been used in Citizens United by big money.
Why was it that, according to Alito, the 1977 decision had to be abandoned? It “was poorly reasoned,” he alleged, as if poor reasoning didn’t live in a whole lot of precedents. No, the transparent explanation for Janus was that the conservatives didn’t like unions and had the votes to weaken them.
“Don’t like a decision?” Justice Kagan taunted in dissent. “Just throw some gratuitous criticisms into a couple of opinions and a few years later point to them as ‘special justifications’ ” for dispensing with stare decisis.
The two rulings, in late June, that best exposed the Court’s hypocrisy involved partisan gerrymandering and Trump’s travel ban. In each unusual case, in which the justices most assuredly should have intervened to rein in others in government who had overstepped constitutional bounds, they shrunk from their duty—the equivalent of a judicial shrug. Neither ruling would likely have gone the same way if President Obama’s nominee—Merrick Garland—rather than Gorsuch, had been on the Court.
The reapportionment cases, Gill v. Whitford (about Republican-controlled Wisconsin) and Benisek v. Lamone (about Democratic-controlled Maryland), constituted the Court’s best chance in a generation to finally fix the worst structural defect in our politics. Indeed, the two cases could have been the Court’s most important rulings in years and showed the justices at their best—not picking ideological winners and losers, but enforcing rules of the game.
In the age of supercomputers and hyperpartisanship, legislators in some states had become especially adept at gerrymandering. The party in power drew up congressional districts and state legislative districts that entrenched its power. The mapmakers’ technique was to “pack” and “crack.” Packing meant concentrating supermajorities of one party in districts so it won overwhelmingly, but in only a few districts. Cracking entailed spreading the disfavored party among districts so it won nothing in the cracked districts. Packing and cracking worked exquisitely well to “waste” votes of the party out of power. What benefit the practices had for democracy was far less apparent. In Wisconsin, for example, in 2012 Republicans won 48.6 percent of the statewide vote for state assembly, but 60 of the 99 seats. Two years later, Republicans got 63 of the seats with 52 percent of the statewide vote.
Such numbers illustrated what the scholar John Hart Ely was talking about when he explained that elected legislatures alone, Republican or Democratic, couldn’t be trusted with reapportionment. They just had too much interest vested in maintaining the status quo—in keeping themselves in office. In this instance, democracy didn’t foster democracy. In order to “clear the channels of political change,” it was obvious that “representative government cannot be trusted.” Under those circumstances, as Kagan wrote in Gill, it was up to the Court to play its hand. “Here,” she said, “politicians’ incentives conflict with voters’ interests, leaving citizens without any political remedy for their constitutional harms.”
Kagan, along with the three other liberals, joined a unanimous Court in Gill to reject the challenge on technical grounds—that the particular plaintiffs challenging Wisconsin’s districts lacked “standing” to bring the case. That was because those voters didn’t actually live in a gerrymandered district, which meant they didn’t have sufficient skin in the game. (The Court sent Gill back to a lower court to reconsider; the two dissenters wanted the case dismissed entirely. In Benisek, the Maryland case, the Court also ducked the main issue, finding that the challengers had waited too long to file a lawsuit.)
The Court’s objection in Gill wasn’t ridiculous, but the justices could easily have concluded that the plaintiffs’ injuries were statewide in nature. Instead, they preferred to skirt the issues, in large part because Kennedy was on the fence—and certainly not prepared to join Kagan’s four-justice concurrence. Kennedy had been straddling the problem of partisan gerrymandering for more than a decade, but still wasn’t ready—even as he was retiring—to take a side. A flimsy declaration by the Court didn’t do anybody much good, and with Kennedy departing, there remained the possibility of an eventual resolution on the constitutional merits. Kagan’s opinion—by laying out how challengers next time could solve the standing problem by presenting a First Amendment right-of-association claim—offered a road map on how to win a Court majority. Whether the newest justice, taking Kennedy’s seat, might provide the decisive fifth vote was another story.
Gill was all the more galling given that the Court, on the same day, did intervene in another dispute—supposedly governed by the First Amendment—that involved the political process. In a case from Minnesota, the justices struck down a state law that prohibited individuals in a polling place from wearing hats, T-shirts, badges, and other gear that expressed a political message. So, evidently, the Court knew that some constitutional rights in the context of elections needed vindication.
While the decision on Trump’s travel ban may not have long-term significance, it demonstrated that the Court seemed content to look the other way when doing so dovetailed with a partisan agenda. The five justices who upheld Trump’s executive order in Trump v. Hawaii were associated with the GOP (just as the four dissenters were connected to Democrats). This wasn’t a reprise of Bush v. Gore, but one would be forgiven for believing political preferences might have had something to do with the outcome. If, say, a President Hillary Clinton had issued an executive order that appeared to be animated by bias, would the same five justices have approved? As in the reapportionment cases, the Court could have distinguished itself by checking another branch of the government that had exceeded constitutional limits.
The travel ban had its origins in Trump’s presidential campaign, when he unrelentingly advocated a “Muslim ban.” He read on TV a “Statement on Preventing Muslim Immigration” that called for a “total and complete shutdown of Muslims entering the United States until our country’s representatives can figure out what’s going on.” As a candidate, he also declared that “Islam hates us” and the United States was “having problems with Muslims coming into the country.” Shortly after Election Day, when Trump was asked whether violence in Europe had affected his plans to “ban Muslim immigration,” he replied, “You know my plans.”
A week after Trump was sworn in, he issued his first executive order banning foreign travelers from several mainly Muslim countries; according to an adviser, the president called it a “Muslim ban” since “we can’t allow people coming into the country who have this hatred of the United States.” After lower courts blocked the first order, Trump issued another version, insisting that it was “much tougher.” It was a third order, in effect since December 2017 and imposing indefinite restrictions on travel, that the Supreme Court considered.
Trump’s orders on foreign travel might have been unremarkable under other circumstances. The president has extraordinary discretion over border policy, because Congress had delegated that authority. But Trump’s anti-Muslim enmity, though expressed chiefly during his campaign, raised the question of whether he was discriminating on the basis of religion. According to Roberts, in a 5-to-4 decision, the travel ban was legal because it was neutral on its face and justified by national security—Trump’s other, noxious statements were beside the point. To blunt Trump’s words, Roberts explained that “we must consider not only the statements of a particular president, but also the authority of the presidency itself.”
Sotomayor would have none of it. In a vehement dissent from the bench, she recounted a litany of Trump’s words. “Let the gravity of those statements sink in,” she said. They “were spoken or written by the current president of the United States.” In her written opinion, she concluded with a defense of judicial action: “Our Constitution demands, and our country deserves, a judiciary willing to hold the coordinate branches to account when they defy our most sacred legal commitments.” In this case, she was right.
Trump v. Hawaii was especially infuriating, Sotomayor said, because the Court used it to officially repudiate Korematsu, the 1944 ruling that upheld the wartime detention of Japanese Americans. Trump’s travel ban, according to Sotomayor, was little different. The parallels, she said, were manifest: “an ill-defined national-security threat to justify an exclusionary policy of sweeping proportion”; “an exclusion rooted in dangerous stereotypes about…a particular group’s supposed inability to assimilate and desire to harm the United States”; and “strong evidence that impermissible hostility and animus motivated the government’s policy.” In redeploying “the same dangerous logic underlying Korematsu, the Court “replaces one gravely wrong decision with another.”
Anthony Kennedy remained an essential fifth vote in close cases like Gill and Trump v. Hawaii, as well as Janus. But the 2017–18 term was the first time in more than a decade in which he didn’t vote with the liberals in at least one 5-to-4 ruling. (The year before, he was with the liberals in those cases half the time.) There were 18 decisions in 2017–18 that split 5–4 on ideological issues—and Kennedy sided with the conservative justices every time. It was the most riven Court in years: Only 34 percent of its rulings were unanimous, down 23 percentage points from the prior term. When Kennedy had the opportunity to act as a brake on the conservative onslaught, he was in the back seat. He had relished his time in the center of the storm. But he decided he was done.
The liberal justices couldn’t miss Kennedy’s vanishing act. Combined with Gorsuch’s arrival the year before, it conspired to produce a term of abject misery (even if Gorsuch in his first full year didn’t prove to be quite as conservative as Thomas or Alito). And with Kennedy’s replacement—the 53-year-old Brett Kavanaugh—looming, the liberals knew the future looked grim. What they feared most was that an energetically retrograde Court would suffer in the eyes of a public that no longer respected it. The justices always faced that risk if their anti-democratic ways diverged too far from the body politic. But it was hardly clear that the emboldened conservative majority, unshackled by Donald Trump, appreciated the peril.
Various structural reforms have been suggested to restrain an arrogant Court, to depoliticize the appointments process, and to lower the stakes for filling its vacancies. The best of these ideas was to eliminate life tenure and to phase in staggered, non-renewable terms—say, for 18 years, which eventually would result in a vacancy every two years. In such a system, presidents would no longer have disproportionate incentive to name youthful justices—Trump’s advisers call age 55 the “sell-by date” for Court candidates—and no single justice could have protracted influence. The last six justices to leave the Court served, on average, for 29 years. One well-known judge called the federal judiciary “the nation’s premier geriatric occupation.” When the Framers created the Court, Americans who survived childhood had a life expectancy of around 60 years. When he was a lawyer in the Reagan administration, Roberts himself endorsed a 15-year term limit for federal judges. But that kind of change would require a constitutional amendment, and the political party in power at the Court would never support it.
Likewise, a statute requiring justices to move to lower federal courts after a period of years—they’d retain their life tenure—would be politically unrealistic. So, too, would be the idea of staffing the Supreme Court sequentially with nine experienced judges chosen at random from the federal appellate courts, who would serve for a limited time before returning to their original posts. Nor was it likely, back in the real world, that any president would unilaterally begin nominating moderates, the better to create a nonideological Court whose rulings wouldn’t be seen as extensions of partisan agendas. Obama might have come closest to that ideal in naming Merrick Garland, but Obama did so only because of political calculation.
No, the best chance for the Court to reclaim its stature is from within, and it is John Roberts who provides the Court’s best hope. If the Court is to become a less dangerous branch, he has the opportunity, the temperament and maybe the skill to help to make it so.
In recent decades, most conservatives deplored the Court for Roe and Obergefell, while their liberal counterparts saluted those decisions. Just as predictably, liberals denounced Bush v. Gore, Heller, Citizens United and Shelby County, while conservatives applauded. If mere outcomes are the lodestar for evaluating the work of the justices, that altogether makes sense. But the modern history of the Court shows that the real institutional problem isn’t this or that unwise ruling. It’s that the justices simply are involving themselves too often and with too much certitude.
Other branches surely can do grievous harm, and have: tolerance of slavery for decades; failing to head off the Great Depression; 58,000 American deaths in the Vietnam War; and countless missteps on taxation, education, debt, infrastructure and other vast swaths of policy. But the Supreme Court’s accretion of power—steady, subtle, unstated—produces its own danger. We have come to expect from the Court what it should not deliver. And in turn we demand even less from our democratically chosen representatives.
Once upon a time, the exemplar of judicial deference—Oliver Wendell Holmes Jr., who served on the Court from 1902 to 1932—wrote to his British friend, Harold Laski, and quipped: “If my fellow citizens want to go to hell, I will help them. It’s my job.” Holmes was right, even if hell wasn’t the optimal destination. Roberts said much the same thing, more forgivingly, in the first Obamacare ruling: “It is not our job to protect the people from the consequences of their political choices.”
Impatiently, myopically, with deep distrust in our elected representatives, we have come to believe that democracy is broken. And we have come to see the justices as our saviors. With so much dysfunction elsewhere in government, the justices see themselves that way, too. But we need more politics, not less politics. It is a sign of weakness that we countenance an almighty Court to resolve so many of our hardest choices. We do not need, nor should we want, the Court to save us from ourselves.