Judicial recusal is when a judge voluntarily disqualifies herself from hearing a case. The reasons for this usually include conflicts of interest, such as financial investments with one of the parties or personal or professional relationships with the parties or their attorneys. Federal law requires judges to recuse themselves under certain circumstances; for example, when a district court judge gets promoted to the appellate level or a trial judge sits by designation on an appeal, she cannot hear a case that she decided previously. For the most part, at the trial and appellate levels, there has been little controversy over the recusal of federal judges because when they disqualify themselves—whether voluntarily or on the basis of one of the parties requesting it—another federal judge is capable of serving as a substitute.^[1]

Such is not the case for the justices of the US Supreme Court because the nation’s Constitution requires that there be “one Supreme Court.” No one, not even retired justices, can substitute for one of the Court’s nine members. If any one of them disqualifies themselves, then the Court would be left shorthanded. In most cases, this would not pose a problem, but in close decisions, one justice’s recusal could leave the Court evenly divided, and the lower court’s decision would stand. As a result, Supreme Court justices feel a “duty to sit,” an expression that usually refers to not leaving the Court evenly divided but just as likely reflects the singularly unique position of the Supreme Court within the federal hierarchy. There is, after all, only one Supreme Court.

Like other federal judges, today’s Supreme Court justices are legally bound to disqualify themselves. However, for much of the High Court’s history, there was no recusal requirement. Instead, justices exercised their discretion and sense of propriety when deciding whether to sit on a case, and they adopted common-law principles, or the prevailing practices of other federal judges (Frank 1947). Whenever questions arose over their disqualification, controversy more often followed from their participation in a case rather than from their recusal. According to Stempel, Supreme Court history has been “rife with questionable recusal practices and conduct widely condemned by today’s standards of disqualification” (1987, 608). Frost similarly observed that “on many occasions during the past 200 years the public has focused on a judge’s questionable decision not to recuse and has found the laws governing that decision to be wanting” (2005, 532–33). All of that changed in 1948, when Congress, for the first time, mandated that justices must disqualify themselves under certain circumstances.

For example, the 1948 legislation required that justices disqualify themselves when they had been “of counsel” or when they had been “a material witness” in a case before the Court.^[2] Of course, justices continued to exercise considerable discretion over their recusal decisions, as the new legislation left it to their opinion as to whether they were “so related to or connected with” a party in a case or to their lawyer. The issue that prompted the 1948 legislation involved, in part, Justice Hugo Black’s participation in a case argued by his former law partner, Crampton Harris, in Jewell Ridge Coal v. United Mine Workers, 325 U.S. 161 (1945) and Justice Robert Jackson’s concurring opinion to a denial for rehearing where Jackson intimated the impropriety of Black’s having participated, as the case raised “a question as to the qualification of one of the Justices to take part in the decision.”^[3]

For the next quarter of a century, by and large Supreme Court disqualification remained, as Virelli put it, a “personal, independent, unreviewable decision by an individual justice whether to participate in an individual case” (2012, 1547). Therefore, in 1974 Congress amended the recusal law to create a “stricter, more comprehensive, and objective” standard for Supreme Court disqualification (Stempel 1987, 594).

This study focuses on that period from 1948 to 1974, when Congress first required justices to disqualify themselves but the justices had few precedents to guide them other than past practices. Specifically, this study will examine the recusal decisions of Justice Tom C. Clark, who served the longest (1949–67) of any justice whose service commenced post-1948 but concluded pre-1974.^[4] As a case study, Justice Clark offers several compelling reasons to review recusal behavior. First, Clark had served as the US attorney general for four years before President Harry S. Truman elevated him to the High Court. As a result, he recused himself from more cases than most of his Court colleagues, especially during his first four terms. Hume found that Clark had a recusal rate of 4 percent over eighteen terms; the other justices with whom he served with higher recusal rates were short-term Justices Goldberg (4.5 percent) and Fortas (6.25 percent) and former attorney general Robert Jackson (4.5 percent; 2014, figure 1). Interestingly, more former attorneys general have gone to the Supreme Court than former solicitors general, although more attention has focused on the recusal patterns of former solicitors general (Beamer 2012; Hume 2014; Stempel 1987; Virelli 2011). In fact, when Clark retired from the Court, he acknowledged that “for the first time in 69 years the membership did not include a former Attorney General. No other public office has furnished so many members of the Court” (1969b).

Another reason for examining Clark’s recusal decisions—or more specifically, his decisions to participate in cases when he might have disqualified himself—relates to Congress’s 1974 amendments. The impetus for the 1974 amendments was, in part, Justice William Rehnquist’s decision to participate in Laird v. Tatum, 408 U.S. 1 (1972) after the respondents requested his disqualification. Laird challenged the government’s domestic surveillance program. Before joining the Court, Rehnquist had served as head of the Office of Legal Counsel at the Justice Department, and because he had testified before Congress about the program, respondents argued that he had advised the White House about the program’s constitutionality. Rehnquist denied the respondents’ request and instead joined a five-person majority to support the government’s contention that the respondents lacked standing. In a surprising move, which was unnecessary and admittedly unprecedented, Rehnquist defended his participation in a lengthy memorandum^[5] that became the object of widespread commentary and criticism (“Justice Rehnquist’s Decision to Participate” 1973; MacKenzie 1974; Stempel 1987). Consequently, Congress amended the recusal statute, most likely in response to Rehnquist’s memorandum. Stempel characterized Rehnquist’s defense as “pernicious,” calling it “an important catalyst” in congressional reform efforts (2009, 81). The amended statute required justices to disqualify themselves when their “impartiality might reasonably be questioned.” In other words, the appearance of partiality triggered mandatory recusal. Actual bias did not need to be proved; justices had to avoid even the appearance of bias.^[6]

For our purposes, Justice Clark warrants scrutiny because Rehnquist’s memorandum drew on past cases when there was perceived bias, but it neglected to consider any instances when Clark chose to disqualify himself. For example, Rehnquist mentioned Justices Murphy’s and Jackson’s behaviors in Schneiderman v. United States, 320 U.S. 118 (1943); Chief Justice Stone’s congressional testimony in 1943; Justice Black’s participation in United States v. Darby, 312 U.S. 100 (1941) and Jewell Ridge Coal v. United Mine Workers, 325 U.S. 161 (1945); Justice Frankfurter’s participation in United States v. Hutcheson, 312 U.S. 219 (1941); and Justice Hughes’s participation in Adkins v. Children’s Hospital, 261 U.S. 525 (1922) and West Coast Hotel v. Parrish, 300 U.S. 379 (1937). However, Rehnquist disregarded those instances when Clark participated in questionable cases—cases that presented the appearance of partiality, which could have substantially aided in his Laird defense. Instead of citing post-1948 judicial behavior, Rehnquist relied on pre-1948 instances to contend that “none of the former Justices of this Court since 1911 have followed a practice of disqualifying themselves in cases involving points of law with respect to which they had expressed an opinion or formulated policy prior to ascending to the bench”.^[7]

Such a claim appeared anachronistic, as there was no pre-1948 congressional mandate for Supreme Court recusal. MacKenzie considered Rehnquist’s defense “irrelevant” (1974, 209); others have called it “misplaced” (“Justice Rehnquist’s Decision to Participate,” 1973, 123). Flamm observed that Rehnquist’s examples amounted to no more than “historical relics from a time when society’s attitudes toward judicial bias were more relaxed” (2007, 916).

Therefore, this study will review Justice Clark’s recusal decisions, paying particular attention to those instances when he participated in questionable decisions. Although Beamer examined “controversial and overlooked recusal affairs to prove that a consistent flaw exists,” she did not consider Clark or specifically reference former attorneys general (2012, 1). As the first justice appointed following the enactment of the 1948 recusal statute, Clark’s entire Court service was encompassed by the law until the 1974 revisions. His recusal decisions may shed light on the efficacy of the 1974 amendments and the current debate over further reforms.

Supreme Court recusal, although understudied and often misunderstood, has become, as Virelli put it, one of “the most high profile and controversial issues involving the Court” (2012, 137). In another context, Virelli observed, “Supreme Court recusal . . . has been a regular topic of passionate debate since the Founding” and today has become one of the Court’s “most hotly debated issues” (2011, 1182 and 1185). Over the past four decades, concern about Supreme Court disqualification has become more urgent and widespread, leading to a number of news stories and scholarly articles on the topic, and two recent books have closely examined it to evaluate the merits of proposed reforms (Hume 2017; Virelli 2016). In a review of Hume’s book, Black observed that Supreme Court recusal “lands in that scholarly sweet spot of being really important but woefully understudied” (2018). Justice Clark’s recusal behavior deserves a closer look. The one time that I found Clark’s behavior questioned was when he participated in a relatively obscure case, Madsen v. Kinsella, 343 U.S. 341 (1952), involving the trial of an American citizen by military commission (Kastenberg and Merriam 2016, 166–67). In addition, Hume mentioned only that Clark had to frequently disqualify himself because of his former position as the attorney general (2017).

An Attorney General Becomes Associate Justice

Tom Clark served as the US attorney general during the postwar period of the Truman administration, from 1945 to 1949. During that time, he dealt with a number of issues that eventually made their way to the Supreme Court. As a result, in his first term he sat out of nearly one hundred cases, of which the Court decided fifteen on the merits. He found this aggravating, as litigants were entitled to a full Court, he believed, and “there were quite a few affirmances by an evenly divided Court” (T. Clark 1969b). How did Clark make his decision to disqualify himself from these cases? Principally, he relied on the advice of Court colleagues, as the new federal statute was still untested. Shortly after starting his judicial service, he requested guidance from Solicitor General Philip Perlman, who advised Clark to consult with Justice Jackson and the law clerk to the recently deceased Justice Murphy, whom Clark replaced (Perlman 1949). Section 455 of the Judicial Code read, in its entirety,

Any justice or judge of the United States shall disqualify himself in any case in which he has a substantial interest, has been of counsel, is or has been a material witness, or is so related to or connected with any party or his attorney as to render it improper, in his opinion, for him to sit on the trial, appeal, or other proceeding therein.^[8]

Clark felt ethically bound to disqualify himself from cases connected to his service as attorney general, though strictly speaking, the law did not require recusal for past government service. Typically, an attorney general’s connection to a case before the Court was nominal, since as Frank observed, the attorney general had “only theoretical responsibility for minor cases in their departments” (1947, 624). Chief Justice Harlan Stone similarly suggested that the standard for disqualification should have “something more to do with the case than the merely pro forma relations which an Attorney General often has with cases in the Department of Justice” (Frank 1970, 48). Whatever his reasons for disqualification, there was no obligation for Clark to explain his decision, just as there were no means of enforcing his recusal. According to Lubet, “By tradition, most Supreme Court Justices do not announce their reasons for recusal. It is therefore impossible to know with any precision what the bases were for the great majority of these disqualifications” (1996, 659; see also Bassett 2005, 50; and Flamm 2007, 915). During his entire Court service, Clark decided for himself whether his “substantial interest” in a case warranted his recusal. As Frank summarized, for the next quarter of a century, justices generally did not disqualify themselves unless (a) they had a dollar interest, (b) they were related to a party or to counsel, (c) their former law office was involved, or (d) when serving in government, “they dealt with the precise matter” before the Court (1970, 50).

Clark’s service as the attorney general offered substantial reasons for disqualification, and two-thirds of his recusals occurred during his first four terms. In total, he sat out of seventy-three cases decided on the merits. So far as the record shows, none of those were due to illness or injury, but during the first week of January 1953, he was in Texas attending his mother’s funeral and missed oral arguments in five cases decided on the merit.^[9] Most often, he disqualified himself from antitrust decisions^[10] and cases involving suspected Communists.^[11] He also sat out of several alien property claims. In 1942, President Roosevelt had established the alien property custodian by Executive Order 9095, but in 1946, President Truman terminated the office by Executive Order 9788 and transferred its authority to the attorney general.^[12] In addition, Clark missed deportation or denaturalization cases because, as part of a 1940 reorganization plan,^[13] the powers and functions of the Immigration and Naturalization Service were transferred from the Labor Department to the Justice Department.^[14] In some instances, Clark disqualified himself because he had argued before the Court as the attorney general on the very issue before the Court. In other instances, he had publicly supported Truman administration policies. Several of these issues will be examined more closely, paying particular attention to cases in which Clark participated when there was ample reason for him to recuse himself.

His inconsistency in this regard should not impugn his character, however. According to Frank, “In repeated instances, Attorneys General rise to the Supreme Court, and they must then decide whether they can sit on cases which were in the Department when they themselves were there,” but Clark was “immensely sensitive to the value of high ethical appearance as well as substance” (1970, 47 and 48). That he disqualified himself from seventy-three decisions on the merits before Congress imposed more objective recusal standards speaks of his integrity. According to Hume, Clark’s recusal rate was twice the average of all justices from the modern era and well above the vast majority (2014, figure 1; 2017, figure 2.1). In fact, Clark was so committed to the appearance of justice that he retired from the Court principally because his son, Ramsey, became the US attorney general. There was no legal requirement for him to do so, just as there was no legal requirement for him to disqualify himself from cases handled by Ramsey at the Justice Department. Nevertheless, Clark retired to make room for his son’s advancement, writing to one well-wisher, “I believe it would be best for me to retire. Litigants have enough problems without having a father-son psychology to face. And while there is no actual conflict the potential is there, and the appearance of justice is as important and effective as the real thing” (1967).^[15]

Ramsey Clark at the Justice Department

Clark’s retirement notwithstanding, his recusal record with respect to his son, Ramsey, was haphazard at best. Eight years before Ramsey became the attorney general, he had argued a case at the Supreme Court involving Safeway Stores, Safeway Store v. Oklahoma Retail Grocers, 360 U.S. 334 (1959), and his father disqualified himself. Then after Ramsey was confirmed as the attorney general, Clark disqualified himself from Honda v. Clark, 386 U.S. 484 (1967), an appeal where Ramsey was named as a party in an alien property case. However, these were the only two instances of Clark recusing himself due to his son’s direct involvement in a case before the Court.

During the four years that Ramsey served as the assistant attorney general of the Justice Department’s Lands Division and then two years as the deputy attorney general and the acting attorney general, his father did not disqualify himself from cases brought by the Justice Department. “The cases came to the Supreme Court that I had participated in in the lower courts, and had actually been involved in formulating the position taken by the government,” Ramsey told an interviewer. “Dad always sat on the cases; he never, as far as I know, disqualified himself in any cases, because I happen to have been in it” (1968). Ironically, Justice Clark chose to retire when his son became the attorney general to avoid the appearance of a conflict of interest, yet when his son headed the Lands Division, the question of a conflict of interest hardly arose. “I don’t know why it didn’t,” Ramsey recalled. He continued,

In fact, your proximity to actual cases in the Supreme Court is much greater as Assistant Attorney General than as Attorney General, because they are much closer to specifics and to individual cases. . . . Actually, as Deputy I was much further away from the work of the Supreme Court than I had been as Assistant Attorney General in the Lands Division where we would have six or eight cases before the Court in a single term (1968).

Frank acknowledged that “there was in fact no functional difference between the son’s relationship to the Supreme Court as Deputy Attorney General or, certainly, as Acting Attorney General, a post he held for a long time, and his final position as Attorney General” (1970, 47). Of course, Tom and Ramsey Clark’s situation was without precedence. The only other time that a similar father-son relationship created a potential conflict of interest was in 1930, when Charles Evans Hughes Jr. resigned his position as the solicitor general because his father became the chief justice of the United States.

Former Law Clerks

Other than his son appearing before the Court, the next closest “family” to Clark who posed the potential for a conflict of interest were his former law clerks. Given that his Court service extended for eighteen terms, it was not unusual that some of his former clerks worked on cases where he had to decide whether to participate. Of course, the prevailing recusal statute said nothing with respect to former law clerks, so Clark’s decisions as to whether he was “so related to or connected with any party or his attorney as to render it improper” for him to participate were left to his discretion. As a result, his decisions appeared indiscriminate, especially since some of them changed case outcomes.

Clark twice recused himself when his fourth-term clerk Frederick Rowe presented cases to the Court. He sat out of a case involving illegal oil pipeline rebates where Rowe prepared the briefs for one of the pipeline companies, United States v. Atlantic Refining, 360 U.S. 19 (1959), and he sat out of a case involving public utilities regulation that Rowe argued on appeal, Commercial Communications v. Public Utilities Commission, 359 U.S. 341 (1959). In his final term, Clark also disqualified himself from three antitrust cases when his second-term clerk Donald Turner appeared before the Court as the assistant attorney general of the Antitrust Division. Turner had worked on the briefs in two of the cases, United States v. Sealy, 388 U.S. 350 (1967) and United States v. Arnold, Schwinn, 388 U.S. 365 (1967) and presented the government’s arguments in the third, United States v. 1st City Nat Bank, 386 U.S. 361 (1967). One of Justice William Brennan’s former law clerks, Richard Posner, also assisted on the briefs and argued before the Court; however, Brennan did not recuse himself.

Clark also participated in cases when his former law clerks appeared before the Court. For example, he participated in Bushnell v. Ellis, 366 U.S. 418 (1961), argued by his first-term clerk Percy Don Williams, who represented an indigent defendant; Escobedo v. Illinois, 378 U.S. 478 (1964), where his fourth-term clerk Bernard Weisberg argued as an amicus curiae for the American Civil Liberties Union; and United States v. John Hancock Mutual Insurance, 364 U.S. 301 (1960), argued by his eighth-term clerk Harry Hobson. Although he had disqualified himself from two of Frederick Rowe’s cases, he also participated in cases presented by Rowe, and sometimes his participation determined the outcome. For instance, Clark provided the fifth vote in Federal Trade Commission v. Henry Broch, 363 U.S. 166 (1960), an antitrust decision adverse to Rowe’s corporate client. Two years later, Rowe was back at the Court after the US Court of Appeals modified a cease and desist order on remand, and Clark participated in Federal Trade Commission v. Henry Broch, 368 U.S. 360 (1962), but Rowe lost a second time. In a habeas corpus petition where Rowe represented an indigent defendant, Payne v. Madigan, 366 U.S. 761 (1961), Clark participated, but instead of providing the fifth vote, his participation created an evenly divided Court due to Justice Felix Frankfurter’s nonparticipation, and the Court affirmed the lower court’s ruling. Considering the nature of the appeal, that may have been a strategic move, as Clark was regarded generally as tough on criminal defendants, and affirmance meant that the defendant remained incarcerated for the prescribed sentence. A similar instance occurred when Clark participated in United States v. Huck Mfg., 382 U.S. 197 (1965), argued by his second-term clerk Donald Turner, and Clark’s participation led to affirmance by an evenly divided Court, as Justice Abe Fortas did not participate.

When his former law clerk Assistant Attorney General Donald Turner assisted in preparing the brief in Baltimore & Ohio v. United States, 386 U.S. 372 (1967), Clark’s participation determined the outcome. The case involved an order of the Interstate Commerce Commission, which had permitted a proposed railway merger. In his majority opinion, Clark reversed the lower court, thereby precluding the proposed merger. Had he disqualified himself, then an evenly divided Court would have affirmed the lower court’s decision. Of course, Clark may have participated only because Justice Brennan did, even though Brennan’s former clerk Richard Posner had also assisted on the brief.

By participating in a case involving an Interstate Commerce Commission (ICC ) order and a proposed railway merger, particularly one involving his former law clerk, Clark appeared inconsistent. Years earlier, he had disqualified himself from St. Joe Paper v. Atlantic Coast Line Railroad, 347 U.S. 298 (1954), which involved an ICC plan to reorganize a bankrupt railway by merging it with another railway, and in a split decision, the Court forestalled the proposed merger because Justice Hugo Black also sat out of the case. Clark’s decision to participate in Baltimore & Ohio may have been strategic, or it may have been simply waiting long enough. After all, Turner had been his law clerk more than fifteen years beforehand, and Clark’s recusal in St. Joe Paper was more than a dozen years previously. As he once told an interviewer, “Indeed, I sat in cases involving Safeway stores even though I represented them in 1937. I think you can go to ridiculous lengths in abstention” (1973b). However, in one antitrust case, Safeway Stores v. Vance, 355 U.S. 389 (1958) and its companion, Nashville Milk v. Carnation, 355 U.S. 373 (1958), decided twenty years after his personal association with Safeway had ended, Clark’s participation determined the outcome.

Antitrust Cases

Antitrust cases represented the greatest number of cases where Clark disqualified himself, which was unsurprising given that he had served for several years as a lawyer in the Justice Department’s Antitrust Division and for a short time had headed the division. As the attorney general, he reportedly had “started more antitrust suits than any previous Attorney General,” and he was proud of his record for filing “almost as many [antitrust] suits as we formerly had in the history of the United States” (“Enforcing the Antitrust Laws”; see also “Clark, Cautious Trust Buster”). At the Court, he disqualified himself from cases involving the Du Pont Corporation because as the attorney general, he had made high-profile attacks against the industrial giant. For example, he sat out of United States v. Du Pont, 351 U.S. 377 (1956); United States v. Du Pont, 353 U.S. 586 (1957); and United States v. Du Pont, 366 U.S. 316 (1961). He also recused himself from a number of movie industry cases because he had argued before the Court in a major film distribution monopoly case, United States v. Paramount Pictures, 334 U.S. 131 (1948), which led to similar recusals in Sutphen Estates v. United States, 342 U.S. 19 (1951); Hughes v. United States, 342 U.S. 353 (1952); and Partmar v. Paramount, 347 U.S. 89 (1954). As a result, he necessarily sat out of other antitrust cases that relied on the principles announced in Paramount Pictures, such as United States v. Gypsum, 340 U.S. 76 (1950) and United States v. New Wrinkle, 342 U.S. 371 (1952).

That Clark disqualified himself from so many antitrust cases, particularly ones involving the movie industry, did not necessarily mean that he had to recuse himself from all of them. For example, when the Court considered the “scope and effect” of Paramount Pictures in Theatre Enterprises v. Paramount, 346 U.S. 537 (1954), Clark wrote the majority opinion and found the petitioners’ allegations “untenable” and any relevance to Paramount Pictures “slight.” Having had experience with—even having expressed an opinion on—an issue before the Court did not automatically disqualify a justice. As Justice Rehnquist argued in his defense for participating in Laird,

It would be not merely unusual, but extraordinary, if they had not at least given opinions as to constitutional issues in their previous legal careers. Proof that a Justice’s mind at the time he joined the Court was a complete tabula rasa in the area of constitutional adjudication would be evidence of lack of qualification, not lack of bias.^[16]

However, Clark participated in two antitrust decisions so closely related to Paramount Pictures that his impartiality appeared questionable. The first involved an alleged antitrust violation in the promotion of professional boxing. In International Boxing Club v. United States, 358 U.S. 242 (1959), Clark wrote the five-person majority opinion. He had argued in Paramount Pictures to end a movie industry monopoly, and in International Boxing Club he effectively ended a professional boxing monopoly. Most significantly, in his International Boxing Club opinion he made frequent references to significant holdings from Paramount Pictures, at one point announcing, “The case which most squarely governs this case is United States v. Paramount Pictures.”^[17]

Clark’s participation in International Boxing Club had two plausible explanations, which may temper the impression of partiality. When the case first appeared at the Court, United States v. International Boxing Club, 348 U.S. 236 (1955), it did not present the antitrust questions controlled by Paramount Pictures but instead questioned whether professional boxing could receive the same antitrust exemption as professional baseball. Chief Justice Earl Warren ruled that boxing was different from baseball for the purposes of interstate commerce and sent the case back to the district court, which subsequently found an antitrust violation. Clark had participated in the first International Boxing Club decision, which was unquestionably appropriate, so when the case returned to the Court as an antitrust violation, it may have seemed reasonable to participate a second time. There was also the length of time involved, as Clark’s arguments in Paramount Pictures were a decade in the past.

The second antitrust case where Clark’s participation should have raised questions about his impartiality involved a named defendant that he had prosecuted as the attorney general. One of the original defendants in Paramount Pictures was Loew’s, and Clark recused himself from United States v. Loew’s, 339 U.S. 974 (1950), an appeal of the Paramount Pictures decision. A dozen years later, Loew’s was back before the Court in a case questioning whether block booking of copyrighted feature motion pictures for television violated antitrust law. Although Loew’s new appeal involved television broadcasting instead of movie theater film distribution, the connection to Paramount Pictures was unmistakable.

Relying on Paramount Pictures, Justice Arthur Goldberg ruled that “tying agreements,” no matter the medium, were illegal. “Appellants attempt to distinguish the Paramount decision in its relation to the present facts,” Goldberg announced in United States v. Loew’s, 371 U.S. 38 (1962), but “appellants cannot escape the applicability of Paramount Pictures.” The degree to which Paramount Pictures controlled the new television medium was indubitable:

The principles underlying our Paramount Pictures decision have general application to tying arrangements involving copyrighted products, and govern here. . . . Enforced block booking of films is a vice in both the motion picture and television industries, and that the sin is more serious (in dollar amount) in one than the other does not expiate the guilt for either. Appellants’ block booked contracts are covered by the flat holding in Paramount Pictures.^[18]

On previous occasions, Clark had disqualified himself from cases that relied on principles announced in Paramount Pictures, such as No Pac Ry v. United States, 356 U.S. 1 (1958), when the Court considered whether tying agreements were unreasonable restraints on trade. However, this time he participated, and the only reasonable explanation for his participation was that enough time had passed.

Communist-Inspired Decisions

Clark served as the attorney general during the immediate postwar period, when Communist or subversive infiltration was keenly feared. As part of President Truman’s federal loyalty program, Clark had made public the infamous “Attorney General’s List of Subversive Organizations,” and because he was the original named respondent, he recused himself when the attorney general’s list was challenged in Joint Anti-Fascist Refugee Committee v. McGrath, 341 U.S. 123 (1951), which was the appeal of Joint Anti-Fascist Refugee Committee v. Clark, 177 F. 2d 79 (D.C. Cir. 1949). He also disqualified himself from Smith Act prosecutions, such as Dennis v. United States, 341 U.S. 494 (1951), which led to the convictions of American Communist Party leaders. The Smith Act punished advocating the violent overthrow of government. Clark also sat out of Dennis v. United States, 339 U.S. 162 (1950), which challenged the trial jury’s impartiality when the defendant refused to appear before the House Un-American Activities Committee; Sacher v. United States, 343 U.S. 1 (1952), which involved contempt charges against Dennis lawyers; and In re Isserman, 345 U.S. 286 (1953), which considered disbarment of Dennis lawyers.

One of the more significant Communist-inspired cases where Clark recused himself was American Communications Association v. Douds, 339 U.S. 382 (1950), which also became the basis for one of his more questionable participation decisions in a case that appeared to overturn Douds. During his first term, Clark sat out of Douds, which challenged the non-Communist oath for union officers in the Taft-Hartley Act, a law that Congress had passed over President Truman’s veto and that Attorney General Clark had sought to repeal. In the Court’s majority opinion, Chief Justice Fred Vinson upheld Taft-Hartley’s non-Communist oath, concluding that it was not a bill of attainder. Had Clark participated in Douds, undoubtedly he would have supported Vinson’s majority position, as Clark’s anti-Communist views were well established—his opposition to Taft-Hartley had more to do with his views on labor—and during his first term, Clark reflectively voted alongside Vinson in 98 percent of cases decided on the merits.

Fifteen years later, the Court considered United States v. Brown, 381 U.S. 437 (1965), a case that challenged a law making it a crime for Communist Party members to serve as union officers or employees. The new law, which had replaced that portion of Taft-Hartley requiring a non-Communist oath, was enacted a decade after Clark joined the Court, and Brown became the first time that the Court reviewed a conviction under it. Speaking for a five-person majority, Chief Justice Earl Warren invalidated the law principally because it was a bill of attainder, and Clark joined three other dissenters. Although Brown did not consider the First Amendment freedom of association as had Douds, the similarities between the two cases were unmistakable.

The purpose of the two laws was the same. According to Chief Justice Vinson, the non-Communist oath of Taft-Hartley was intended to “prevent political strikes,” and Chief Justice Warren held that its replacement was to minimize “the danger of political strikes.” In fact, Warren conceded that the new law “was designed to accomplish the same purpose” as Taft-Hartley “but in a more direct and effective way”^[19], while the four Brown dissenters considered Douds “obviously overruled.”^[20] Dissenting along with Clark were Justices Byron White, John Marshall Harlan, and Potter Stewart. Not incidentally, Victor Rabinowitz and Leonard Boudin appeared as counsel in both cases. They represented the appellants in Douds, where they lost the decision, and they returned in Brown with amicus briefs, where their side prevailed.

Clark’s participation in Brown seemed suspicious because Brown appeared to overrule a case where he had recused himself. As with most of his questionable decisions, enough time had passed to give the appearance of impartiality, and this time Clark’s participation did not change the outcome, so it made little difference. Furthermore, Clark had changed. He was no longer a freshman justice who uncritically followed the chief justice; he was now fourth in seniority on the historically activist Warren Court. In fact, only two other justices—Black and Douglas—who had served with him in Douds remained on the Court for Brown. Black, who had dissented in Douds, joined the majority in Brown. Douglas, on the other hand, had missed Douds and half of Clark’s first term due to serious injuries he suffered from falling off a horse that then fell on him, but he also joined the majority in Brown. The country was also different, as concern with Communist infiltration had waned considerably since the demise of McCarthyism, and the Court reflected this shift in perspective.

By the 1960s, public criticism of the Court was directed at its rulings, not at recusal decisions. For example, the Court faced harsh public criticism following its school prayer decisions, Engel v. Vitale, 370 U.S. 421 (1962) and Abington v, Schempp, 374 U.S. 203 (1963), and its criminal prosecution decisions, such as Mallory v. United States, 354 U.S. 449 (1957) and Escobedo v. Illinois, 378 U.S. 478 (1964). Just as important, Justice Arthur Goldberg had as much reason to disqualify himself from Brown as did Clark. Fifteen years earlier, Goldberg had worked on the briefs for the Congress of Industrial Organizations (CIO) in a case consolidated with the Douds ruling, and he had filed amicus curiae briefs for the CIO in the final Douds decision. Yet Goldberg participated in Brown, which may have given Clark reason for doing so. While Goldberg provided the crucial fifth vote to overturn a law that the Brown majority argued was “not necessarily controlled by Douds,”^[21] Clark and the dissenting justices asserted that Congress was now forbidden “to do precisely what was validated in Douds.”^[22]

The Submerged Lands Act

Commonly referred to as the “tidelands” controversy, the states and federal government had a long-standing dispute over who controlled offshore oil, gas, and mineral rights. As the attorney general, Clark had argued in United States v. California, 332 U.S. 19 (1947) that the federal government should have “paramount rights in and power over” California’s three-mile belt along the coast. After his successful argument defending national sovereignty, Clark tried to forestall congressional efforts to restore state authority, writing, “The United States should not by legislative action transfer to the respective coastal states dominion and control over the lands and resources underlying the ocean areas adjacent to our shores but should retain and administer these valuable national assets for the benefit of all” (1948). Nonetheless, Congress tried to restore state sovereignty over the tidelands, once while Clark was still the attorney general and again after he joined the Court, and both times President Truman vetoed the measures.

It was unsurprising, then, when Clark disqualified himself during his first term from a pair of cases raising the issue of state sovereignty. Other states, such as Louisiana and Texas, had joined California in urging state sovereignty, but the Court continued to rule against them in United States v. Louisiana, 339 U.S. 699 (1950) and United States v. Texas, 339 U.S. 707 (1950). The Court found no reason to treat Louisiana any differently than California, even though Texas’s unique status as an independent republic caused some disagreement among the justices. Texans were especially furious with the Court’s decisions—in part because Attorney General Clark had previously indicated in a press statement that Texas was different: “As a Republic it owned all of the lands within its boundaries, including the marginal sea commonly called tidelands. This area was under the sovereignty of Texas during the Republic and was retained by it under the provisions of the Act of Admission” (Daniel 2010).

Following these Supreme Court losses, several additional states not party to the original disputes renewed court challenges after President Dwight D. Eisenhower signed the 1953 Submerged Lands Act, which granted to coastal states title to submerged lands and natural resources within their traditional boundaries, limited to three miles.^[23] When the Court considered these new state challenges, Clark participated, even though as the attorney general he had argued for and publicly defended national sovereignty. Instead of recusing himself, he joined the Court’s per curiam opinion in Alabama v. Texas, 347 U.S. 272 (1954) to uphold the Submerged Lands Act on the constitutional ground that Congress could dispose of territory under its control any way it wanted. Because California was involved, the chief justice and former California governor Earl Warren did not participate.

The most striking aspect of Clark’s participation in the Submerged Lands Act decision was not the timeframe involved—only four years had elapsed from his recusal in tidelands cases to his participation in one—but his reversal of position. Given that he had successfully argued in favor of and had publicly supported national sovereignty, it was surprising that he voted to uphold a law granting state sovereignty, effectively reversing the Court’s prior decisions. By reversing his own position as the attorney general, Clark gave the impression of being unbiased and probably deflected attention away from whether he should have disqualified himself. However, in three subsequent tidelands cases, United States v. Louisiana, 363 U.S. 1 (1960); United States v. Florida, 363 U.S. 121 (1960); and United States v. California, 381 U.S. 139 (1965), Clark continued to recuse himself.

The Steel Seizure Case

The one time that Clark’s participation in a case publicly raised concern was during the high-profile incident when President Truman ordered his secretary of commerce, Charles Sawyer, to seize the nation’s steel mills. By that time, the Korean War was entering its third year, and a threatened strike of steelworkers potentially endangered America’s forces overseas. Contemporary commentators, such as syndicated columnist Drew Pearson, expected Clark to support the president’s action because he had made statements to that effect as the attorney general. Pearson predicted that Clark would vote in favor of the president (1952), and Truman probably counted on Clark’s assent. According to Marcus, “Everyone anticipated that Clark would vote for the President” (1994, 190).

As the attorney general, Clark had advised Truman regarding labor relations and industrial work stoppages that threatened the nation’s well-being. In fact, when coal miners threatened to go on strike in defiance of a court injunction and the government took control of the mines, Clark successfully defended the administration in United States v. Mine Workers, 330 U.S. 258 (1947), calling the miners’ defiance an “insult to the United States itself.” In addition, Attorney General Clark had made public statements to the Senate Committee on Labor, which were widely reproduced in the press, defending the president’s authority to interfere in proposed strikes. “With regard to the question of the power of the Government,” Clark contended, “I might point out that the inherent power of the President to deal with emergencies that affect the health, safety and welfare of the entire Nation is exceedingly great” (1949).

Therefore, in the steel seizure case, Clark was on record as having advised the president on the very issue before the Court, and while the case was under review, initially it did appear as though he would support Truman’s action. “There was quite a clamor for me to disqualify myself in that case,” Clark remembered twenty years afterward. “They all said I’d prejudged it—that I was going to decide in favor of the President” (1973a). Nevertheless, rather than supporting the president, Clark joined a Court majority to deny Truman the authority to seize private property. However, he would not join Justice Hugo Black’s opinion in Youngstown Sheet & Tube v. Sawyer, 343 U.S. 579 (1952), which ruled that Truman lacked constitutional authority or congressional authorization for his action.

Instead, Clark concurred in the judgment insofar as Congress had “laid down specific procedures to deal with the type of crisis confronting the President.” Without such statutory guidance, Clark argued, “the Constitution does grant to the President extensive authority in times of grave and imperative national emergency,” but “the President’s independent power to act depends upon the gravity of the situation confronting the nation.”^[24] In other words, Justice Clark maintained the position he had taken as the attorney general—that the president possessed “exceedingly great” inherent power—but only so long as Congress did not prescribe a specific alternate procedure to follow under the circumstances. Unfortunately, as Dickson observed, Truman was “infuriated by Tom Clark’s ‘betrayal’ and never forgave his old friend” (2001, 182n32).

This middle ground—acknowledging the president’s authority to act yet denying its exercise—may have been in response to the dissenting justices, who used Clark’s views as the attorney general to defend Truman’s action. Chief Justice Fred Vinson, writing in dissent, also drew attention to two former attorneys general, Frank Murphy and Robert Jackson, whose views were “in line” with Clark’s. After the Court announced the steel seizure decision, public attention shifted from Clark’s recusal to his changed position, what he called a self-reversal:

I found myself in that spot in the Steel case where the press pointed out that as Attorney General I had advised Senator Thomas that the inherent powers of the President were very great, but as a Justice had stricken down the President’s seizure of the steel mills. My reply was that when writing the Senator, I was Attorney General; now I am a Justice! (1969b)

Once again, by altering the position he had taken as the attorney general, Clark diverted attention away from his participation in the steel seizure decision.

Contemporary Issues

As previously mentioned, in 1974 Congress amended the federal recusal statute to require justices to disqualify themselves when their “impartiality might reasonably be questioned.” Congress did this in large part to address concerns over Justice William Rehnquist’s defense of his participation in Laird v. Tatum using pre-1948 examples of questionable judicial behavior. Thirty years later, Justice Antonin Scalia sparked renewed interest in Supreme Court recusal when he similarly denied a motion to disqualify himself from a case involving Vice President Richard Cheney as a named party, Cheney v. U.S. District Court, 542 U.S. 367 (2004). The controversy began when Scalia, among others, went on a duck hunting trip with Cheney, and afterward news coverage and critical commentary became so intense that Scalia issued a memorandum defending his participation in the case.^[25] A spate of duck-themed articles followed Scalia’s participation in Cheney (Freedman 2004; Goodson 2005; Riffle 2005). Like Rehnquist’s defense, Scalia’s memorandum was remarkable given the well-entrenched custom on the Court to avoid explaining recusal or participation decisions, what Weaver called a “conspiracy of silence” (1975).

Justice Scalia’s defense relied, in part, on a doctrine first propounded by Rehnquist, what came to be known as the “duty to sit.” This duty to sit has become such a compelling reason to avoid recusal that Hume concluded it oftentimes outweighs other ethical considerations favoring disqualification (2017, 4). However, the duty to sit has come increasingly under fire, as other commentators have argued that Congress repudiated the doctrine with the 1974 amendments requiring disqualification regardless of the potential for leaving the Court shorthanded (Bassett 2005, 21; Beamer 2012, 3; Stempel 1987, 607; 2009, 814 and 818; Virelli 2016, 15–16).

Nevertheless, justices continued to participate in questionable decisions, more often than not without explanation. For example, when the Court considered the constitutionality of the Affordable Care Act (Obamacare) in National Federation of Independent Business v. Sebelius, 132 S. Ct. 2566 (2012), critics sought Justice Elena Kagan’s recusal because she had served as the solicitor general when, presumably, the government developed its defense of the program (Hume 2017, 1, 4, and 41). Calls were similarly heard for Justice Neil Gorsuch to disqualify himself from Janus v. AFSCME, 138 S. Ct. 2448 (2018) because of his close ties to political interests who funded one side of the controversy (Warren 2017; “Ginsburg and Gorsuch Should Explain” 2017). In neither instance did Kagan or Gorsuch explain their participation despite publicized misgivings about their impartiality. Undoubtedly, they subscribed to the duty to sit requirement, as their disqualifications would have left the Court evenly split, which may have been the sole objective of those advocating for their recusals.

More recently, disqualification became an issue during the confirmation hearings of Judge Brett Kavanaugh. In addition to some arguably partisan attacks on Kavanaugh’s qualifications—not to mention the scandalous accusations of prior sexual misconduct—there was also concern about his impartiality, particularly when deciding cases involving a sitting president who faced the potential for indictment. Because he had previously written articles suggesting that the president could not face criminal charges while in office, some commentators believed that Kavanaugh should disqualify himself from any case involving President Donald Trump, who faced the possibility of a criminal investigation (Hurley 2018; Nichols 2018; Rizzo 2018). Thus Kavanaugh’s nomination raised the specter of Supreme Court disqualification even before his confirmation occurred.

Calls for Reform

The most salient issue regarding Supreme Court recusal has centered on who has the authority to decide when a justice must disqualify themselves. For most of the Court’s history, that decision belonged to the justices themselves; however, for the past seven decades, Congress has required justices to disqualify themselves when their participation raised ethical questions, but that decision has always been unreviewable and unenforceable. Even after Congress amended the law to provide for more objective recusal standards, for the next forty-five years, as Hume observes, “the justices have at times conducted their business as though the 1974 revisions never happened” (2017, 117). As a result, a number of reform measures have been proposed to correct what Stempel calls the Court’s “flawed practice of permitting each justice to make a final and unreviewable decision” (1987, 590).

Calls for reform most often followed from a justice’s participation in a particular case when there were arguably reasons for their recusal. More sustained criticism of a justice’s behavior inevitably followed from his participation in close decisions, when his disqualification could have potentially changed the outcome. Reform measures have ranged from leaving the law intact but changing the Court’s practice to repealing the law entirely and relying on “more indirect constitutional tools” (Virelli 2011, 1185).

One proposal, for example, would retain the law in terms of the criteria used for deciding when to recuse, but justices with the potential for bias in borderline cases would be required to explain their participation in a “statement of interest.” These statements would then become a part of the public record, which would draw the other justices’—and the public’s—attention toward the ethical implications of their participation. In addition, other proposals would require similar statements whenever a justice disqualified herself because justices have never explained their recusal when they did sit out of cases (Bassett 2005, 50; Hume 2017, 130–31).

Beyond these reporting requirements, another area of potential reform involves some type of review, usually by the other justices, of a particular justice’s decision to participate. For example, one proposal would require a “statement of interest” whenever a party requested a justice’s recusal but was denied, and then the full Court would review that decision (Goodson 2005, 186 and 213–20). Another proposal would amend the current recusal statute to require justices to inform parties of factors that might prompt their disqualification, which could then lead to requests for their recusal. Again, if the parties’ request were denied, then the Court would review and possibly overturn that decision (Stempel 1987, 590 and 643; see also Hume 2017, 127–28). Such a method of reporting and oversight may still be possible without amending the law, as traditional forms of adjudication already permit it (Frost 2005, 535 and 592).

Yet a third proposal would address the most compelling reason justices sometimes ignore the ethical implications of their participation—the “duty to sit.” Unlike other federal courts where visiting judges can sit by designation, Supreme Court justices are irreplaceable if they choose to disqualify themselves. Therefore, to avoid the possibility of an evenly divided Court, another proposal would allow replacements for recused justices, such as retired justices or chief judges on the US Court of Appeals (Hume 2017, 128–29).

Of course, all of these proposals, from reporting requirements, to judicial oversight, to potential substitutions, presuppose that Congress can legislatively compel justices to disqualify themselves. However, legislative control over Supreme Court behavior may be an “unconstitutional intrusion” on the separation of powers principle. Therefore, before the Court rules on its constitutionality, Congress should repeal the current recusal statute affecting justices. As an alternative to legislation, Virelli has argued, Congress could rely on constitutionally sounder methods, such as impeachment and removal, which may be the only effective means of enforcing the current statute regardless (2011, 1185; 2012, 1535–36; 2016, chap. 5). Bassett, on the other hand, thought that it was “unclear whether the statute potentially infringes on separation of powers” (2005, 50).

Conclusion

When President Truman nominated his attorney general to a seat on the Supreme Court, there was little consternation over Clark’s past government service, which included eight years in the Justice Department before serving four years as the attorney general. Whether Clark would have to disqualify himself because of his close connection to issues that he had considered as the attorney general never came up during his confirmation. There was some controversy over Truman nominating a non-Catholic to fill the seat of the recently deceased Justice Frank Murphy, and a few Senators objected to Clark’s nomination for his failure to appear before the judiciary committee for questioning, a familiar practice today that was then still developing. When Clark did recuse himself, usually there was a recognizable connection to his service as the attorney general, but at times it was difficult to discern. For example, he sat out of several cases where the most plausible explanation for his recusal was the case’s Texas-based origins: Dalehite v. United States, 346 U.S. 15 (1953) involved the notorious Texas City disaster, where over five hundred people perished in one of the deadliest industrial accidents in US history; United States v. Harriss, 347 U.S. 612 (1954) involved federal lobbying law violations and a Texas corporation’s failure to report its lobbying efforts affecting agricultural prices; United Gas v. Memphis Gas, 358 U.S. 103 (1958) involved a Federal Power Commission decision affecting gas prices where a Texas corporation served as one of the parties; and Mercantile National Bank v. Langdeau, 371 U.S. 555 (1963) was an appeal from the Supreme Court of Texas involving Dallas-based banks and insurance that turned on the question of venue.

As the first justice appointed to the Court following enactment of the 1948 recusal statute, Clark used his best judgment to decide whether he would participate in cases. By and large, he was free to exercise his discretion, and for the most part, he behaved appropriately. Rarely did his decisions affect the outcome of cases, and only once was his decision to participate publicly questioned.

Following the 1974 amendments, concern about Supreme Court disqualification persisted, leading to calls for further reform. However, it is unclear whether additional statutory amendments will make a difference in how justices behave. Since recusal decisions seldom change case outcomes or public perceptions of the Court, as Hume has argued, justices routinely disqualify themselves when the stakes are low. When the stakes are high and their participation will determine the outcome, then policy preferences overcome ethical concerns (2017, 15–16 and 117–18). In other words, even when there are good reasons for a justice’s recusal, she will still participate in cases when the outcome matters to her.

Clark understood this as well as the separation of powers problem behind legislative efforts. Writing in another context, he observed,

To many minds the judge is the law. To bend the judge to anyone’s will, therefore, raises grave questions. This makes it the more necessary to give pause to any temporary popular demand to straightjacket the judges. . . . It cannot be gainsaid that the founders intended for each branch of the federal government to be master of its own house. This is especially true in the judiciary where history teaches that the bulwark of a free society is its courts. (1970)

Recent attention to Supreme Court recusal, with its attendant calls for reform, should heed Clark’s example. There may be no constitutionally effective way—save impeachment—to enforce Supreme Court disqualification; the law will have to serve as well as the justices who abide by it. Although he departed from the federal bench in disgrace following allegations of sexual misconduct, the Ninth Circuit Court of Appeals judge Alex Kozinski put it best when reflecting on rules and restrictions to promote ethical judicial behavior: “The hard truth is that none of these things really matters. Judicial ethics, where it counts, is hidden from view, and no rule can possibly ensure ethical judicial conduct. Ultimately, there is no choice but to trust the judges” (2004, 1106).