In the past two months, Ruth Bader Ginsburg has fractured three ribs and had two cancerous nodules removed from her left lung. She was absent from oral argument last week and will miss this week’s arguments as well. Doctors say they expect her to be back on the bench in February, and until then she will review transcripts from her home and participate in the court’s decision-making remotely. But her return to work has not quieted critics who say that Ginsburg should have retired long ago and that her health issues are the latest evidence that justices should not be allowed to serve for life.
Ginsburg, who is 85, suggested this summer that she intends to serve “at least five more years” on the court. She is far from the first justice to linger on the bench into advanced age. John Paul Stevens retired at 90 in 2010, making him the oldest serving justice since Oliver Wendell Holmes stepped down from the bench two months shy of his 91st birthday in 1932. Stevens’ extended tenure produced significantly less hand-wringing than Ginsburg’s—a contrast partly attributable to Stevens’ hale health but also possibly driven by the gender bias that Ginsburg has battled throughout her career. Yet while the focus on Ginsburg may be out of proportion, the concerns generated by a graying judiciary cannot be blithely dismissed. Fears of judicial gerontocracy have flared at several earlier points in American history, including long before the court had any female members.
The late Antonin Scalia waved off the idea of limiting the terms of justices as “a solution in search of a problem,” but the problem is not an imaginary one: Some justices really have clung to their positions long after their mental faculties have left them. Justice Henry Baldwin remained on the court for nearly a dozen years after his 1832 hospitalization for “incurable lunacy.” One of Justice Nathan Clifford’s colleagues described him as a “babbling idiot” in the final years before his death in 1881. Justice Stephen Field in the mid-1890s and Justice Joseph McKenna in the early to mid-1920s each reportedly spent the end of their tenures in a haze.
“Mental decrepitude” on the Supreme Court has continued into the modern era, as historian David Garrow has documented. Frank Murphy, who served in the 1940s, was likely addicted to illegal drugs by the end of his tenure, and his biographer wrote that “on at least one occasion,” with Murphy in absentia, his law clerk and two fellow justices “jointly decided what Murphy’s votes should be.” Justice Charles Whittaker teetered on the brink of nervous breakdown for much of his five-year stint on the court in the late 1950s and early 1960s. Hugo Black stayed on for more than two years after his wife concluded in 1969 that “his mentality has been impaired.”
Nor was Black the last justice whose mind slipped while he was still on the bench. In 1975, his last year on the court, William O. Douglas was so severely disabled by a stroke that his fellow justices agreed to delay any decision in which Douglas’ vote could swing the outcome. Justice William Rehnquist developed a dependence on a sedative that caused him to experience hallucinations during withdrawal; at one point in late 1981, he tried to escape from George Washington University Hospital in his pajamas. Rehnquist recovered, but two of his colleagues—Lewis Powell and Thurgood Marshall—faced doubts about their mental capacities at the tail end of their careers.
The history of cognitive decline on the high court teaches two lessons. First, there is a real risk of a substantial time lag between the onset of mental deterioration and a justice’s retirement. But second, and as important, this is a risk that can be contained. No justice—no matter how deranged—can do serious doctrinal damage without the acquiescence of at least half his colleagues. And when a justice is so utterly incapacitated that he is unable to break 4-4 ties, the court can continue to function with an even number of active members. Originally, the court had only six justices; during the Civil War, it had 10; and it has functioned fine with eight members during prolonged vacancies. Indeed, there are notable virtues to having an even number of justices—one of them being that it then takes more than a knife’s-edge majority to overturn a lower court decision or strike down a law nationwide.
The proposed solutions to “mental decrepitude” on the Supreme Court each come with flaws of their own. A common proposal is to fix the lengths of justices’ terms, with 18 years being the number most often suggested. Yet 18-year terms would not lay to rest the problem of mental decline. Murphy had been on the court for only eight years when his apparent drug dependence reached its height. Whittaker finally suffered a nervous breakdown less than five years into his term. And the Rehnquist pajama incident occurred just nine years into his 33-year tenure. Granted, the risk of mental disability increases with advanced age, and 18-year term limits might on balance lead to a younger bench. Or they might not. Presidents might be inclined to select older nominees if justices could serve for only 18 years rather than for life.
Other ostensible benefits of 18-year terms are also likely to prove illusory. Advocates argue that a fixed term length will lower the stakes of confirmation battles. Perhaps, but fights over open court seats will be fierce whether the appointee wields influence over abortion and the death penalty for 18 years or for longer. What fixed term lengths will do, without a doubt, is to ensure that these fights occur more frequently. If the goal is to defuse some of the tension surrounding Supreme Court confirmations, then creating more vacancies is a curious choice.
Assuming that terms are staggered, then the 18-year proposal would also ensure that a seat on the court opens at least every two years. This is sometimes cited as an advantage, as it would narrow the inequity across presidents who have disparate opportunities to influence the court based on the number of vacancies that arise during their terms. For example, William Howard Taft, a one-term president, appointed six justices, while fellow one-termer Jimmy Carter named none. But it would also mean that every two-term president would choose four—or in the event of early retirements or deaths, even more—members of the court. That possibility is disconcerting given that justices are, empirically, much more likely to vote with the administration when the president who appointed them is still in office. This “loyalty effect,” which my colleagues Lee Epstein and Eric Posner have documented, limits the court’s efficacy as a check on presidential overreach. Staggered 18-year terms would likely lead to a larger number of “loyal”—pliant—justices on the court at any given moment.
Fixed term lengths would also raise the question of what term-limited justices will do after their 18 years expire. Some might try to monetize their experience by going into private practice. Others might seek elected office. Consciously or unconsciously, a justice might adjust her decisions with a view toward pleasing potential employers or future voters. While today nothing stops a justice from leaving the bench for practice or politics, very few do—at least in the modern era—and the Supreme Court remains one of the few governmental institutions that is immune from the revolving door. Term limits could change that for the worse.
Finally, term limits would lead to what in game theory is known as the “last period” problem. Justices who anticipate that they will interact with each other year after year can expect a concession in one case to be reciprocated later on. But as a term-limited justice approaches the 18-year mark, not only would her incentive to cooperate diminish, but her colleagues’ incentives to cooperate with her would too. Moreover, this dynamic potentially affects not only the last period of play, but also the period before the last period, and the period before that, and so on, leading to an unraveling of cooperation on the court. One advantage of the status quo is that justices rarely announce—and sometimes do not decide on—their retirements until shortly before they leave the bench. Term limits, by making end dates more predictable, would undermine the incentives for soon-departing justices to behave cooperatively and for their colleagues to cooperate with them.
Instead of fixed term lengths, some have suggested a mandatory retirement age for justices—either 70 or 75. These proposals have many of the same flaws as term limits, though a richer pedigree. Several Democratic lawmakers introduced constitutional amendments to set a mandatory retirement age for justices of 70 or 75 as an alternative to President Franklin D. Roosevelt’s ill-fated 1937 court-packing plan. In 1954, the Senate voted 58-19 to approve an amendment requiring all justices and federal judges to retire at 75, but the House never took up the proposal. Meanwhile, more than 30 states have adopted mandatory retirement ages for the judges on their highest courts—with most setting the cap at 70 or 75. (Vermont, an outlier, requires retirement at 90.)
But like fixed terms, pushing justices off the bench as a birthday present for hitting 70 or 75 would not eliminate the risk of mental deterioration. Frank Murphy’s disability struck in his late 50s; Charles Whittaker’s nervous breakdown came in his early 60s; Rehnquist’s hospitalization for sedative dependence occurred when he was 57. Others will reach the age of 70 or 75 with still many years of work ahead, thus raising the risk that post-judicial career prospects may taint their decisions.
An age cutoff at 70 or 75 would not appreciably lower the confirmation stakes either. Neil Gorsuch turns 70 in 2037, Brett Kavanaugh in 2035. Even with an age cap of 70, either justice could one day decide whether a female born in 2019 can get an abortion. Mandatory retirement would, however, raise the same last-period problem as fixed term lengths. It might not produce a more cognitively capable court, but it would likely lead to a less cooperative one.
A third proposal targets the problem of disability more directly. In the 1970s, Senator Sam Nunn of Georgia repeatedly introduced legislation that would have allowed a panel of 12 federal judges to force the retirement of a Supreme Court justice or lower court judge if a majority of the panel concluded he was mentally or physically incapacitated. Senator Howell Heflin of Alabama introduced a constitutional amendment with a similar goal in 1989. These proposals sidestep some of the pitfalls of term limits and age caps, such as the last-period problem and the potential that justices would be swayed by post-court career prospects. But they also raise the risk of justices being ousted not because they are incompetent but because they are ideological outliers.
For all three proposals—term limits, age caps and the removal of judges determined to be disabled—there are serious questions as to whether reform requires a constitutional amendment. Article III of the Constitution states that justices and lower court judges “shall hold their offices during good behaviour,” a phrase whose meaning the Supreme Court has never fully explicated. According to one view, “good behaviour” means that impeachment is the only way to cut a justice’s term short. In another view, the constitutional requirement is satisfied if Supreme Court justices are demoted to the lower courts or to auxiliary status once they serve for 18 years or reach age 70 or 75, as long as their salaries are unaffected.
But whether reform would require a constitutional change or simply a statutory enactment, the calls to end life tenure for justices should be batted away this time as they have been before. Term limits and age caps would lead to more frequent (but not less bruising) confirmation battles, weaker incentives for the court’s members to cooperate, and stronger motivations for political posturing as justices consider the prospect of post-judicial careers. Judicial disability panels, while not raising all the same problems as term limits and age caps, would open up new opportunities for gamesmanship if members sought to force retirements to gain political advantage. The Supreme Court, while not immune from ideological strife, is one of the few remaining institutions in American life in which liberals and conservatives interact collegially and find common ground on a wide range of issues. The proposals to end life tenure would put that at risk.
The severity of Ginsburg’s current health condition pales in comparison with the ailments that have afflicted many of her predecessors on the bench—and unlike them, there is no sign that she has lost any of her intellectual edge. The fact that the court has faced, and survived, the much more serious impairment of several of its members suggests that the problem of judicial disability, while undeniable, is also manageable. In comparison with presidential incapacity, the threat of which prompted the 25th Amendment, the incapacity of Supreme Court justices is both more common and less dangerous. Fixed terms, age caps, and forced retirement are all strong medicine for the problem of judicial disability. In light of the flaws inherent in each, the better course of treatment is none at all.