Partial Recusal and the Supreme Court
Louis J. Virelli III*
INTRODUCTION
At her historic confirmation hearing, Justice Ketanji Brown Jackson was asked about judicial recusal—the process by which a judge is removed from a case, typically for ethical reasons. More specifically, she was asked whether, if confirmed, she would recuse herself from two consolidated affirmative action cases scheduled for the Supreme Court’s upcoming term. The cases involved constitutional and statutory challenges to admissions policies at Harvard and the University of North Carolina. Justice Jackson served on Harvard’s Board of Overseers for six years prior to her confirmation. Whether Justice Jackson’s service on the Board constituted grounds for recusal became a topic of heated public debate almost immediately upon her nomination. Concerns were raised about how her relationship with Harvard, a named party in the case, would impact her ability to remain impartial, or even to appear so. By contrast, others argued that because the Board was not involved in university admissions policy, and because there was a companion case against a different school, Justice Jackson should not be precluded from participating.[1] Justice Jackson resolved the dispute at her hearing by taking the unusual step of giving a definitive answer to a question about recusal. She stated that she planned to recuse herself from the case if confirmed.[2]
The fact that a Supreme Court nominee was asked about her views on recusal is not surprising. The last several Supreme Court confirmation processes have involved questions about the nominees’ views on judicial recusal, and the nominees have answered with varying degrees of transparency and specificity.[3] What was unusual about Justice Jackson’s situation was the nature of the recusal controversy and what it can teach us about recusal at the Court. This essay uses the framework of the affirmative action litigation to introduce and preliminarily explore the concept of “partial recusal” at the Court. Part I briefly introduces judicial recusal, including the unique features of recusal at the Supreme Court. Part II addresses the idea of partial recusal, which has been generally undertheorized and effectively overlooked with respect to the justices’ recusal practices. It then distinguishes between two types of partial recusal—horizontal and vertical—and makes a case for why each could be useful for the Court.
I.
Judicial recusal is as old as courts themselves. Since Justinian’s time, judges have removed themselves from cases for ethical reasons.[4] While the accepted grounds for recusal have evolved over the centuries, the underlying purposes of judicial recusal have not. Recusal serves two fundamental goals—to protect the fairness of the proceedings for litigants and to promote public confidence in the courts. It serves both of these ends by ensuring that judges are not only impartial arbiters of the cases before them, but also that they appear so to the observing public. Courts owe their legitimacy to public sentiment; when the public stops believing in the integrity of its judges, the judicial system is no longer able to perform its primary, dispute resolution function.[5]
Recusal in federal courts is governed by statute. The statute requires federal judges and Supreme Court justices to recuse when they experience certain conflicts of interest (generally financial or familial) and when a reasonable person would conclude that a jurist’s “impartiality might reasonably be questioned.”[6] Although the statute ostensibly applies equally to all federal judges, including members of the Supreme Court, it does not in practice. This is primarily because lower court judges are replaceable; when they recuse from a case, another federal judge is authorized to sit in their stead.[7] The recused judge is no longer part of the suit, but the number and status of the judges deciding the case remains the same. The same is not true for Supreme Court justices. Due to Article III’s mandate that there be “one supreme Court,” justices are not replaceable. Recusal of one or more justices therefore changes the makeup of the Court for that case. The result could be a tie vote in the case of a single recusal, or in the case of multiple recusals, a loss of quorum or even exclusion of the entire Court.[8] I have argued elsewhere that the distinctive institutional consequences of Supreme Court recusal make statutory recusal requirements for the justices unconstitutional because they threaten the Court’s ability to exercise the “judicial power” explicitly assigned to it in Article III.[9] More importantly for present purposes, I have also argued that the institutional considerations inherent in Supreme Court recusal decisions require a different approach to recusal at the Court.[10] Whereas a lower court judge need only consider how her participation affects the litigants and the public’s perception of the proceeding, a Supreme Court justice must balance those ethical concerns against impediments to the Court’s ability to decide cases properly before it.[11]
Imagine a situation where the Supreme Court was called on to decide the outcome of a presidential election and a recusal statute required (quite sensibly as an ethical matter) that no justice could participate in a case involving a party that the justice had disparaged in public. If one of the justices had revealed that they considered a candidate unfit for public office, should they recuse? If they were a replaceable lower court judge, almost certainly. Having revealed their preference in the election, a judge’s comments could easily cause a reasonable observer to question their impartiality in the case. But what if they are a Supreme Court justice? Recusal would address ethical concerns about deciding a case when an adjudicator has formulated a personal opinion about one of the parties, but it would also mean that a presidential election would be decided by either a diminished Court or a regional court of appeals in the case of a tie vote.[12] At minimum, consideration by less than nine justices could create doubt about the authority and credibility of the Court as an institution.
II.
Given the realities of recusal at the Court, Justice Jackson’s circumstances in the affirmative action cases raise some novel and interesting questions about what is collectively referred to here as “partial recusal.” How, for example, should we treat the companion case? The concern about Justice Jackson’s participation was based on her prior relationship with Harvard. Does the simple fact that the Court chose to consolidate two similar cases against different parties mean she is precluded from deciding a case against North Carolina? If Justice Jackson were a lower court judge, this would not be a significant issue—she could simply recuse from both cases out of an abundance of caution and allow a replacement judge to resolve them both. But what about as an unreplaceable Supreme Court justice? Must North Carolina accept a decision from an incomplete Court because of a potential conflict with a different party in another, albeit related, case?
And what about the two questions presented in each case? The first question presented was whether the Court should overrule its prior decision in Grutter v. Bollinger,[13] in which it held that universities could consider race in their admissions policies without violating the Equal Protection Clause. This issue arises in both of the consolidated cases. The second issue presented is slightly different for each of the consolidated cases. In the case against Harvard, the second question is whether Harvard’s treatment of Asian American applicants violates Title VI of the Civil Rights Act. The case against North Carolina involves similar issues of race neutral alternatives in admissions but does not mention Asian American applicants specifically.
If Justice Jackson were a member of a circuit court panel, it would make sense for her to recuse herself from the entire case. A replacement judge without any potential conflicts could assume her place and decide the case in its entirety without having to manage the logistics of having different judges participating in different parts of the same or similar cases. But since Justice Jackson cannot be replaced once recused, it may be preferable for the litigants and the Court if she only refrained from deciding issues for which the grounds for recusal were strongest. Board membership may preclude her from addressing whether Harvard’s policy violates Title VI, but does it necessarily require recusal on the question of whether to overrule Grutter? Surely if that were the only issue in a case against North Carolina, there would be no basis for Justice Jackson to recuse. Adding Harvard as a defendant complicates matters, but a purely legal analysis involving stare decisis is not as closely related to membership on Harvard’s Board of Overseers as a review of university policy under Title VI.
In practice, partial recusal rarely comes up. When it does, circuit courts are split as to whether the federal recusal statute permits partial recusal.[14] The issue has never been litigated at the Court, nor have the justices made any public statements regarding whether, as a policy matter, they could or would consider partial recusal. This indicates at least two things about partial recusal that make it ripe for additional investigation, especially with regard to its application to Supreme Court justices. First, uncertainty among the circuit courts indicates both the practical significance of, and lack of clarity about, the issue in general. Second, the fact that the issue has only been addressed by circuit courts applying the federal recusal statute to lower court judges means the non-statutory considerations that are unique to the justices’ recusal decisions have so far been entirely overlooked.
The circumstances in the affirmative action cases are merely a starting point for a much broader, and novel, inquiry into whether partial recusal is a viable and useful option for the Court. That inquiry considers partial recusal in two dimensions: “horizontal” and “vertical”. “Horizontal recusal” is defined here as the decision to recuse from a subset (one or more, but not all) of cases within a consolidated case. “Vertical recusal” is the decision to recuse from a subset (one or more, but not all) of the issues within a single case.[15]
A. Horizontal Recusal
How should a Supreme Court justice manage recusal decisions involving multiple, consolidated cases with distinct ethical ramifications? One obvious approach would be for a justice to simply treat the consolidated cases collectively, as a lower court judge might. This has the advantage of administrative simplicity and clarity—if the Court is treating several cases as one for purposes of oral argument, then any decision by a justice to recuse should apply equally to all cases. It also serves the ethical purpose of protecting against actual or perceived cross-contamination from one case to another. If a justice participates in an oral argument addressing issues from multiple cases, even if that justice were to refrain from participating in the case in which they have a potential conflict, the effect of that conflict may be perceived as spilling over to the other consolidated case.
On the other hand, if the grounds for recusal from one case are clearly contained within the details of that case, for example in cases where a justice exhibits apparent party bias, has a personal relationship with a lawyer or witness for a party, or has a financial interest with respect to one of multiple defendants, institutional considerations may counsel in favor of participating in the remaining case(s). Imagine consolidated redistricting cases in which a justice lived in one of the states where a map was being challenged. The justice may have enough of a direct interest in their own state’s map that real or apparent conflicts would support recusal, but that interest would not exist with regard to the other state’s map. Recusing from both cases would result in an incomplete Court addressing issues that may be of national importance and increase the possibility of a tie vote, which could lead to a non-precedential decision affirming the court below. That may be a necessary cost of avoiding the ethical conflict in the first case, but there may be good reasons not to change the number and makeup of the Court for the other.
Without horizontal recusal as an available solution, parties could use consolidation strategically to require recusal of seemingly unsympathetic justices even where recusal would not be appropriated if the case were litigated separately. For example, if counsel for the state in the second redistricting case above—the one that does not involve the justice’s home state—sought recusal of the justice on the grounds that their conflict of interest contaminates the entire litigation, it could afford a dangerous opportunity (and precedent) for justice-shopping that could cause significant delays and confusion at the Court. One response to this point may be that it is really an argument against consolidation, and that any problem horizontal recusal may seek to address is better resolved by litigating cases separately. This may be true, but that raises the question of the utility of consolidating cases in the first place. Assuming the Court consolidates cases for valid procedural or other reasons, it is not necessarily true that the benefits of consolidation are less than the costs of horizontal recusal.
In addition to the Court’s own institutional concerns, there is a similar argument in support of horizontal recusal from the litigants’ perspective. The possibility of a tie vote in a consolidated case that does not raise clear ethical concerns for the parties could leave them without a definitive resolution to their legal questions despite significant investment in time and money to bring that case before the Court. To the extent horizontal recusal can protect litigants’ investment in their own cases with little or no ethical cost, it should at least be considered a viable option.
Finally, horizontal recusal may result in more recusals—and, in turn, the perception of higher ethical standards—at the Court. It is not uncommon for justices’ decisions not to recuse to be controversial, especially in high profile cases.[16]This contributes to the perception that justices are participating in cases in which they may not be impartial. However, if a more surgical recusal mechanism were available, such that a justice could recuse from one of multiple consolidated cases, they may be more willing to trade the institutional costs of withdrawing from a single case for the ethical benefits of recusing from that case. This greater incentive to recuse is particularly valuable for the Supreme Court because the justices’ recusal decisions are unreviewable and thus effectively unconstrained by any statutory standard.[17] Therefore, at least one overarching effect of horizontal recusal may be increased public confidence in the judiciary. The justices may be more likely to recuse, and the idiosyncrasies of Supreme Court recusal may be more readily apparent to the observing public in a justice’s decision to employ horizontal recusal.[18]
B. Vertical Recusal
Vertical recusal—recusing from one of multiple issues in a single case—offers similar, but not identical, costs and benefits to horizontal recusal. The administrative costs of vertical recusal may be slightly less than for horizontal recusal. Vertical recusal can be achieved by a justice simply leaving the bench during one portion of the argument. Horizontal recusal requires a justice to sit through the full argument and then limit their participation in the ultimate resolution of the litigation to only the case or cases from which they are not recused. Although the ethical consequences of the two approaches may be about the same, the act of recusal is simpler and clearer when the justice can be physically removed from the portion of the case that raises ethical concerns about their participation.
Vertical recusal is—like horizontal recusal—also vulnerable to cross-contamination, especially if the concern is over a justice’s party bias. If the multiple issues in a case are related such that either could swing the outcome in a preferred party’s favor, a biased (or seemingly biased) justice who recuses from one issue may still be able to benefit their preferred party via the remaining issue. This is one way to understand the choice confronting Justice Jackson in the affirmative action case. If a reasonable person would question Justice Jackson’s impartiality as to all things relating to Harvard based on her service on its Board of Overseers, then her participation in the affirmative action case with respect to the issue of whether to overrule Grutter, even though not specific to Harvard, could still be seen as problematic and thus merit recusal from the entire case.[19]
Conversely, vertical recusal allows for more targeted balancing of institutional concerns. Singling out specific issues within a case that could raise ethical issues for a justice allows the Court to maximize ethical benefits without unduly impinging on the Court’s ability to fully and diligently decide other important issues in that case. Returning to the example of Justice Jackson, the case for her recusal from the first issue presented in the affirmative action cases (whether to overrule Grutter) is far weaker than from the second (whether Harvard’s policy violates Title VI). That is not to say that her recusal from the first issue would necessarily be inappropriate, but only that the two analyses are substantively different.
The first issue requires application of the law of stare decisis to a generalized question of constitutional law. If Grutter is in fact overruled, it will undoubtedly have some impact on whether Harvard’s policies survive legal scrutiny. That could be a viable basis for questioning Justice Jackson’s impartiality but is far less compelling than recusal from issue two, which challenges Harvard’s admissions policy for “engaging in racial balancing, overemphasizing race, and rejecting workable race-neutral alternatives.”[20] Issue two directly implicates Justice Jackson’s relationship with Harvard while still being potentially outcome determinative.
The institutional consequences of participating in the two issues are also not the same. Overruling Grutter will have permanent consequences for not only all of American higher education but also for the doctrine of equal protection more generally. Allowing less than the full complement of justices to decide such a monumental legal issue is potentially problematic, especially in a tied or close case, and must be treated as such when weighed against the ethical benefits of recusal. By contrast, although the Court’s resolution of issue two will have some precedential value, it is arguably less important that a full court consider whether a specific school’s policy satisfies Title VI because that decision is narrower in application (Harvard and schools with similar admissions policies) and scope (Title VI).
As with horizontal recusal, vertical recusal’s more tailored analysis may have the added benefit of encouraging more recusals. By targeting only those issues that contain, or appear to contain, a direct personal conflict for the justice, the justices may be more willing to recuse from one issue with the knowledge that they can still fulfill the Court’s institutional obligation to decide the remaining questions before them. Their decisions may also be more palatable to the observing public. Much of the controversy over Supreme Court recusal revolves around decisions not to recuse and often neglects to consider the institutional issues inherent in those decisions. Vertical recusal offers justices an opportunity to show the public a more nuanced and willing approach to recusal that may make the justices’ decisions not to recuse from specific issues easier for the public to accept and understand.
Another related benefit to more targeted recusal decisions is their educational function. Much of the controversy around recusal at the Court focuses on whether the justices have predetermined notions of what the law is in a given area. This is most evident at the justices’ confirmation hearings, where Senators often grill nominees on their judicial philosophy.[21]While perhaps an appropriate qualification for becoming a justice, legal philosophy is not a proper grounds for recusal in a given case. Put another way, whether the public may believe that a particular justice favors a given legal position in case is not, nor should be it be, a reasonable grounds on which to question that justice’s impartiality. As Justice Scalia explained for the Court in Republican Party of Minnesota v. White, legal predispositions are evidence of a qualified, not a biased, judge.[22]
Vertical recusal creates an opportunity for justices to publicly distinguish between legal issues that could trigger a justice’s personal interest and those that merely raise a legal question about which the justice may have established opinions. This is a feature, rather than a flaw, of vertical recusal. Allowing justices to focus on recusal from issues that appear to directly benefit either the justice or a party to whom the justice has a personal connection not only does a better job of balancing institutional and ethical benefits in a given case, but also signals that recusal is not designed to create an ideological blank slate among the justices in every case. This lesson is valuable to the extent it can shift public expectations toward a more realistic—and useful—application of recusal standards to the Court.
III.
This essay advocates for greater attention to partial recusal at the Supreme Court. Supreme Court justices make recusal decisions in a far more complex environment than their colleagues on the lower courts. Public discourse over recusal at the Court is often focused—rightly—on the justices’ ethical obligation to not only remain impartial in the case before them but to appear so as well. What is most frequently overlooked, however, is that unlike recusal in the lower courts, Supreme Court recusal necessarily changes the size and makeup of the reviewing Court. As a result, whereas partial recusal likely creates more problems than it solves in the lower courts, the arguments for and against partial recusal are far different for the justices. This essay presents partial recusal as a potentially viable option for Supreme Court justices in light of their need to balance the ethical demands of their position with the Court’s ability to fulfill its constitutional duty to resolve cases properly before it. It is not meant as an endorsement of partial recusal in any specific circumstance, but rather, an introductory discussion of the promise of such an approach. To the extent an unreviewable Court has additional tools for effectuating sound recusal decisions, it should enhance recusal’s ability to achieve its twin goals of protecting litigants from biased decisionmakers and promoting public confidence in the integrity and legitimacy of the Court.
* Louis J. Virelli III is a Professor of Law at Stetson University College of Law, where he teaches and writes in the areas of constitutional and administrative law. He is the author of several publications on Supreme Court recusal, including Disqualifying the High Court: Supreme Court Recusal and the Constitution (2016).
[1] Compare Jonathan Turley, Judge Jackson Should Recuse Herself From Major Discrimination Case Before the Court, The Hill (Mar. 1, 2022), https://thehill.com/opinion/judiciary/596249-judge-jackson-should-recuse-herself-from-major-discrimination-case-before, with Jacob Gershman, Supreme Court Nominee Ketanji Brown Jackson’s Harvard Service Raises Questions for Admissions Cases, Wall St. J. (Mar. 4, 2022), https://www.wsj.com/articles/judge-jackson-to-face-more-questions-in-supreme-court-hearings-11648035000?mod=RSSMSN.
[2] See Felicia Sonmez et al., Supreme Court Nominee Jackson Says She Would Recuse Herself From Harvard Affirmative Action Case, Wash. Post(Mar. 23, 2022), https://www.washingtonpost.com/politics/2022/03/23/ketanji-brown-jackson-supreme-court-hearing-live-updates/.
[3] See, e.g., Louis J. Virelli III, 98 Denv. L. Rev. Forum 1, 9–11 (2021) (discussing recusal questions in Justice Barrett’s confirmation),
https://static1.squarespace.com/static/5cb79f7efd6793296c0eb738/t/6063ccd45487fa23c0ba986e/1617153237215/Supreme+Court+Recusal_Virelli_DLR+Final.pdf;Laurence H. Tribe et al., Unresolved Recusal Issues Require a Pause in the Kavanaugh Hearings, Brookings (Sept. 4, 2018), (discussing recusal and Justice Kavanaugh), https://www.brookings.edu/research/unresolved-recusal-issues-require-a-pause-in-the-kavanaugh-hearings/.
[4] See Louis J. Virelli III, Disqualifying the High Court: Supreme Court Recusal and the Constitution 1–2 (2016) (outlining judicial recusal’s ancient roots).
[5] See, e.g., Planned Parenthood v. Casey, 505 U.S. 833, 865–66 (1992) (“[T]he Court's legitimacy depends on making legally principled decisions under circumstances in which their principled character is sufficiently plausible to be accepted by the Nation.”).
[6] 28 U.S.C. § 455.
[7] See, e.g., 28 U.S.C. §§ 291–92, 294 (allowing for designation of circuit, district, and senior judges, respectively).
[8] See Virelli, supra note 4, at 78–85 (offering examples over the last century of how recusal can and has threatened the Court’s ability to decide cases).
[9] See Louis J. Virelli III, The (Un)Constitutionality of Supreme Court Recusal Standards, 2011 Wisc. L. Rev. 1183.
[10] See Virelli, supra note 4, at 160–64.
[11] See id.
[12] Durant v. Essex Co., 74 U.S. (7 Wall.) 107, 111 (1868) ("[N]o affirmative action can be had in a cause where the judges are equally divided in opinion as to the judgment to be rendered or order to be made."); Edward A. Hartnett, Ties in the Supreme Court of the United States, 44 Wm. & Mary L. Rev. 643, 651–52 (2002) (quoting 28 U.S.C. § 2109 and noting that while the statute is not “an explicit congressional requirement” of the rule that a tie vote constitutes an affirmance of the court below, “it clearly represents Congress' awareness and endorsement of that rule”).
[13] 539 U.S. 306 (2003).
[14] The majority of circuits approve of partial recusals under 28 U.S.C. § 455 on efficiency and case-management grounds, but the Ninth and Eleventh Circuits do not. See, e.g., In re Moore, 955 F.3d 384, 389–90 (4th Cir. 2020) (concluding that partial recusal is allowed under 28 U.S.C.A. § 455(a)); Decker v. GE Healthcare Inc., 770 F.3d 378, 389 (6th Cir. 2014) (same); Ellis v. United States, 313 F.3d 636, 642 (1st Cir. 2002) (“Today, we make that approval explicit: we hold that a judge may, in an appropriate case, decide certain issues and recuse himself or herself as to others.”); Pashaian v. Eccelston Props., Ltd., 88 F.3d 77, 84–85 (2d Cir. 1996) (holding that district court did not err in deciding a motion before effecting recusal and that such was “a practical and appropriate resolution”); United States v. Kimberlin, 781 F.2d 1247, 1258–59 (7th Cir. 1985) (finding no abuse of discretion where the trial judge issued a limited recusal order). But see Murray v. Scott, 253 F.3d 1308, 1311 (11th Cir. 2001) (concluding that judge's recusal must be from whole proceeding); United States v. Feldman, 983 F.2d 144, 145 (9th Cir. 1992) (“[W]hen a judge determines that recusal is appropriate it is not within his discretion to recuse by subject matter or only as to certain issues and not others.”).
[15] As the affirmative action cases demonstrate, horizontal and vertical recusal are not mutually exclusive. They are dealt with separately here in order to highlight their conceptual differences and better emphasize the features and limitations of each.
[16] See, e.g., Huma Khan, Should Supreme Court Justices Clarence Thomas, Elena Kagan Sit Out Health Care Case?, ABCNEWS, (Feb. 10, 2011), http://abcnews.go.com/Politics/supreme-court-justice-clarence-thomas-sit-health- care/story?id=12878346.
[17] I have argued elsewhere that Supreme Court recusal is governed as a matter of law by the Due Process Clause, but readily admit that there is no existing—or theoretically realistic—way to enforce that standard against the justices. See Virelli, supra note 4, at 120–46 (articulating a due process-based recusal standard for the justices). The relevant ethical codes do not include the justices, and even if they did, the Court has no reviewing body, and lower court judges do not have authority to pass on the legality of the justices’ conduct. See, e.g., Code of Conduct for United States Judges, https://www.uscourts.gov/judges-judgeships/code-conduct-united-states-judges; Russell Wheeler, What’s So Hard About Regulating Supreme Court Justices’ Ethics?—A Lot, Brookings (Nov. 30, 2011), https://www.brookings.edu/research/whats-so-hard-about-regulating-supreme-court-justices-ethics-a-lot/ (contending that a panel of lower-court judges sitting in review of Supreme Court recusal decisions “would most likely violate the Constitution’s ‘one Supreme court’ mandate”).
[18] This is true regardless of whether a justice explains their horizontal recusal decision in writing. Returning to the example of Justice Jackson’s potential involvement in the affirmative action case, her choice to recuse from only the case in which Harvard is a party would send the message to interested observers that she was carefully considering the ethical impacts of her involvement as well as the institutional consequences of removing herself from both suits.
[19] Note that this argument would not affect whether horizontal recusal is merited. In fact, if the focus is on Justice Jackson’s relationship with Harvard, participation in the consolidated case against North Carolina is far less problematic, even though the same issue regarding Grutter is also part of that case.
[20] Petition for Writ of Certiorari, Students for Fair Admissions, Inc., v. President & Fellows of Harvard Coll., No. 20-1199 (Feb. 25, 2021),
https://www.supremecourt.gov/DocketPDF/20/201199/169941/20210225095525027_Harvard%20Cert%20Petn%20Feb%2025.pdf.
[21] See, e.g., Jess Bravin, First Question for Jackson: What’s Your Legal Philosophy?, WSJ (Mar. 22, 2022), https://www.wsj.com/livecoverage/supreme-court-confirmation-hearings-ketanji-brown-jackson-2022-03-22/card/draft-judicial-philosophy-nQ5I2WE7XaXbK3AbGYNH.
[22] 536 U.S. 765, 778 (2002) (“Indeed, even if it were possible to select judges who did not have preconceived views on legal issues, it would hardly be desirable to do so.”).
Any reproduction of the Article, including, but not limited to its publication, posting, or excerption in print, or on the internet, shall give attribution to the Article’s original publication on the online MSLR Forum, using the following method of citation:
“Originally published on May 11, 2022 Mich. St. L. Rev.: MSLR Forum.”