During Brett Kavanaugh’s 2018 US Senate confirmation hearing, several women accused the US Supreme Court nominee of sexual misconduct. These allegations prompted investigations that raised questions about Kavanaugh’s personal fitness to serve on the Court. Upon Kavanaugh’s confirmation, elected officials and legal scholars expressed lingering concerns that his alleged misconduct would not only taint his reputation but also imperil public attitudes toward the Court. Senator Dianne Feinstein, the Senate Judiciary Committee’s ranking member, lamented that “[c]onfirming Brett Kavanaugh in the face of credible allegations of sexual assault…undermines the legitimacy of the Supreme Court.”Footnote 1 Similarly, New York University law professor Wendy Weiser opined that Kavanaugh would “harm the actual credibility, legitimacy, and authority of the US Supreme Court…[i]f half the country believes one of [its members] committed sexual assault and lacks integrity….”Footnote 2
As the sole branch of the federal government that lacks a public mandate via elections, the federal courts rely on a strong public perception of legitimacy to ensure acceptance of and compliance with rulings (Caldeira and Gibson Reference Caldeira and Gibson1992). While scholars have long considered legitimacy stable and enduring, recent studies identify threats to the judiciary’s diffuse support, including political polarization (Armaly and Enders Reference Armaly and Enders2021; Bartels and Johnston Reference Bartels and Johnston2013), attacks from extrajudicial actors (Collins and Eshbaugh-Soha Reference Collins and Eshbaugh-Soha2019; Rogowski and Stone Reference Rogowski and Stone2021), and unpopular court decisions (Christenson and Glick Reference Christenson and Glick2015). As threats mount, pundits share unease that courts may lose legitimacy and be perceived as another partisan institution.
Justice Kavanaugh’s confirmation proceedings highlight yet another peril for judicial legitimacy: scandal, or allegations of personal misbehavior. Unfortunately for the courts, the allegations of Kavanaugh’s misconduct were not an isolated instance of scandal because numerous judges have been accused of improprieties in the modern era.Footnote 3 Whereas existing studies show that scandal damages executives’ and legislators’ reputations and electoral success (Basinger Reference Basinger2019; Doherty, Dowling, and Miller Reference Doherty, Dowling and Miller2011), they seldom consider how scandal impacts institutional support, which for judges—unlike elected officials—is the fundamental source of political authority (Caldeira and Gibson Reference Caldeira and Gibson1992). Therefore, we must confront whether scandals pose risks to judicial legitimacy.
We examine whether scandals erode public support for the federal judiciary using survey experiments that leverage stylized and real-world instances of scandal. Across three experiments, we find that scandals diminish public support for individual judges. However, we find no evidence that scandals undermine support for judicial decisions or diffuse support for the courts. These results are consistent across respondents irrespective of their partisan alignment with judges. Unlike other threats to legitimacy, this suggests that members of the public may not react to judicial scandal through a partisan lens. Our findings demonstrate that the public sanctions individual judicial actors when they engage in unbecoming conduct. The public tends to hold errant judges accountable through their disapprobation but does not appear to sanction the courts writ large. These results may allay normative concerns about scandals posing a serious threat to the legitimacy of the courts.
We examine whether scandals erode public support for the federal judiciary using survey experiments that leverage stylized and real-world instances of scandal.
SCANDAL AS A THREAT TO PUBLIC SUPPORT
Legitimacy is essential to all well-functioning political institutions, including the judiciary, which lacks a democratic mandate and is reliant on other entities for compliance and enforcement (Caldeira and Gibson Reference Caldeira and Gibson1992). While public approval of the judiciary’s performance may oscillate, diffuse support historically has been stable, drawing on a “reservoir of goodwill” generated by public fidelity to and socialization into democratic norms and values (Gibson and Nelson Reference Gibson and Nelson2014). Yet, diffuse support is not entirely impervious. For instance, polarization makes the public less acquiescent to courts when rulings deviate from individuals’ preferences (Bartels and Johnston Reference Bartels and Johnston2013). When political elites attack or politicize courts (Collins and Eshbaugh-Soha Reference Collins and Eshbaugh-Soha2019), the public alters its perceptions of legitimacy given their partisan alignment with relevant elites (Armaly and Lane Reference Armaly and Lane2022; Rogowski and Stone Reference Rogowski and Stone2021). Court decisions can harm diffuse support when judicial policies diverge from public preferences (Christenson and Glick Reference Christenson and Glick2015).
Scandals may pose similar perils to public support for the courts. Established threats to legitimacy focus on environmental factors and the actions of political elites. However, scandals concern personal behavior. Public perceptions of legitimacy stem, in part, from the notion of judges being “appropriate, proper, and just” (Tyler Reference Tyler2006, 376). Legal professionals have emphasized the importance of judges’ personal integrity to sustain “public confidence” in courts; the American Bar Association exhorts judges in its guidelines to avoid even “the appearance of impropriety” (Geyh Reference Geyh and Bybee2007). Moreover, the Code of Conduct for United States Judges, which binds lower-court federal judges, urges them to “avoid impropriety and the appearance of impropriety in all activities” because “public confidence in the judiciary is eroded by irresponsible or improper conduct by judges.”Footnote 4 A recent nationally representative survey substantiates the emphasis on personal integrity, with respondents indicating that the most important characteristic for judicial nominees is their “moral character” (Krewson and Schroedel Reference Krewson and Schroedel2020, 1437).
Research outside of the judicial context indicates that scandals harm political actors’ public standing and electoral success by providing negative signals about their character (Basinger Reference Basinger2019; Doherty, Dowling, and Miller Reference Doherty, Dowling and Miller2011). Furthermore, in a case study of the 1990s US House of Representatives banking scandal, Bowler and Karp (Reference Bowler and Karp2004) establish a connection between scandals and institutional support. They show that respondents gave more negative evaluations of Congress if their member was implicated in the scandal and they reported hearing about it.
Regrettably, we know little about the repercussions of scandal within judicial contexts. Research on judicial nominations suggests that scandalized nominees receive less support from the President’s copartisans in the US Senate and also attract stronger opposition from out-partisan senators (Cameron and Segal Reference Cameron and Segal1998). Research on state courts sheds additional light on the negative ramifications of scandal on judicial institutions writ large. First, work by Casey (Reference Casey1988) reveals a bifurcation in public support for a state Supreme Court marred by scandal. On the one hand, “cynical bystanders” remain indifferent to scandal. On the other hand, “disaffected” members of the public see the court as illegitimate and desire reforms. Second, Gibson (Reference Gibson2009) highlights how state court legitimacy is subverted through campaign contributions. Donations to judicial candidates prompt the public to perceive judges—and, consequently, the judiciary as a whole—to have conflicts of interest, which is precisely the type of ethical concern that some scandals implicate. Whereas the public may remove state court judges for perceptions of impropriety, federal judges—with lifetime appointments—are immune from electoral repercussions. Consequently, when they are not held to account, scandalized judges’ continued presence on the bench could erode diffuse support for the federal judiciary over time.
THEORETICAL EXPECTATIONS
We posit that judicial scandals have negative consequences for public support for individual judges, their decisions, and the judiciary as a whole. Just as scandals sully the reputations of elected officials, allegations of personal misconduct may lead the public to doubt whether a judge has sufficient integrity. The public may express less faith in rulings associated with scandalized judges, whose dubious character can undermine otherwise principled legal decisions. Together, these phenomena illustrate how scandal could diminish specific support for judicial actors or their opinions. Furthermore, because scandals are negatively valenced events that can bring unwanted scrutiny to judicial institutions that are not consistently salient to the public, they may prompt the public to question whether (1) judges are as “appropriate, proper, and just” as they previously believed; and (2) the institutions through which they exercise power deserve broad deference (Tyler Reference Tyler2006, 376). Thus, we might expect scandal to permeate from implicated jurists, thereby harming the judiciary’s diffuse support.
Finally, we expect that partisanship may condition how individuals perceive scandals. Existing research establishes that the public processes a Supreme Court nominee’s approval through shared partisanship with the President (Bartels and Kramon Reference Bartels and Kramon2022). With regard to scandal, out-partisans may be eager to punish scandalized judicial actors, whereas copartisans may rally behind judges who they believe are falsely accused. Indeed, this phenomenon was on ready display during and after Brett Kavanaugh’s confirmation, which Senator Lindsey Graham (R-SC) called “the most unethical sham since I’ve been in politics.”Footnote 5
RESEARCH DESIGN
Exploring the effects of institutional actors’ behavior on public opinion poses inferential challenges. Ideally, we could observe public attitudes toward the judiciary in the presence and absence of scandal, ceteris paribus, to determine whether scandalous accusations induce disparate levels of support. These conditions elude us because we cannot observe the counterfactual in which a scandalized jurist does not face allegations or vice versa. Whereas panel data could enable researchers to explore the effect of a scandal on perceptions of the courts, repeated measures of support among the same respondents are “woefully scarce” (Gibson and Caldeira Reference Gibson and Caldeira2009, 5). Moreover, panels seldom are serendipitously timed to capture responses before and after unforeseen events that could affect public opinion toward the courts.
Acknowledging these challenges, we deployed three survey experiments to examine the effects of scandal on support. Our designs include realistic scenarios involving judges who attract public attention and vary whether a judge faces scandalous allegations. Through random assignment, we can isolate the causal effects of scandal on public attitudes toward the courts. To enhance generalizability, each experiment focuses on a distinct context in which the public might encounter news of judicial scandal. As long as our experiments yield substantively similar results, we can discount that certain situational details varying across scenarios (e.g., the elapsed time since the scandal occurred or a judge’s position in the judicial hierarchy) underlie our findings. Our experimental protocols, including vignette and question wordings, are in online appendix B.
Our designs include realistic scenarios involving judges who attract public attention and vary whether a judge faces scandalous allegations.
Furthermore, our experimental design promotes external validity (Gaines, Kuklinski, and Quirk Reference Gaines, Kuklinski and Quirk2007). First, by basing our treatments on real-world scandals, we ensure that our stimuli mirror the range of misconduct to which the public is exposed. Second, whereas our first two experiments are stylized and measure only the immediate effects of scandal, our third experiment leverages Justice Brett Kavanaugh’s alleged misconduct to assess whether those accusations affect present-day perceptions of the Supreme Court. This experiment incorporates the real-world political milieu to examine whether a justice associated with a well-known scandal exerts enduring effects on support.
Studies 1 and 2 are similar in design: each provides a stylized account of a federal judge. Study 1—fielded in January 2020 on Amazon’s Mechanical Turk with approximately 1,400 respondents—examines how scandals that emerge during the Supreme Court confirmation process affect public support. Nominations present a prime opportunity to uncover past improprieties that would threaten a nominee’s confirmation, making them a salient context when the public encounters information that might erode Court support. Study 1 presents respondents with common information about a Supreme Court nominee who ultimately was confirmed several years ago. Respondents are randomly assigned details about Senate vetting and any allegations of scandal that arose. The control condition merely apprises respondents that senators questioned the judge during a hearing. Our three treatment conditions inform respondents that the vetting process uncovered evidence of past improprieties.
Study 2—fielded in August 2020 on Lucid with approximately 1,650 respondents—explores how scandalous allegations concerning sitting federal judges affect public perceptions of the judiciary. Although incumbent judges rarely face the same scrutiny as during their confirmation, allegations of misbehavior sporadically arise and capture national attention, casting negative light on the judiciary and sometimes prompting resignations or impeachment. Study 2 presents a stylized news article wherein a DC Circuit Court of Appeals judge pledges to remain on the bench despite personal news that might have prompted resignation or retirement. The control condition presents a judge who intends to remain despite a cancer diagnosis; the three treatment conditions present various misconduct allegations.
The treatment conditions in Studies 1 and 2 draw on three of four scandal types in Basinger et al.’s (Reference Basinger, Brown, Harris, Gulati, Dagnes and Sachleben2013) typology—ethical, sexual, and financial—and each treatment is based on real-world scandals associated with nominees or sitting judges.Footnote 6 Additionally, Studies 1 and 2 manipulate whether Democratic President Barack Obama or Republican President George W. Bush nominated the judges. Varying the nominating president allows us to assess whether copartisanship conditions the effects of scandal on attitudes toward the courts (Bartels and Johnston Reference Bartels and Johnston2013).
Study 3—fielded in August 2020 on Lucid with approximately 850 respondents—leverages Brett Kavanaugh’s role on the Supreme Court to investigate enduring effects of scandal on public support. Our experiment provides respondents with a stylized article describing the Court’s 2020 ruling in Ramos v. Louisiana. To determine if making salient Justice Kavanaugh’s presence influences public support, we randomized the opinion author to be either Justice Kavanaugh or Justice Neil Gorsuch—another conservative justice appointed by President Donald Trump who did not face any allegations of scandal.Footnote 7 , Footnote 8
Following the vignettes in all three studies, we queried respondents about their attitudes. For specific support, Studies 1 and 2 ask respondents for their approval for the judicial actor, whereas Study 3 asks whether respondents approve of the Court’s Ramos decision. To measure diffuse support, we asked respondents to express their level of agreement with the six-statement battery on legitimacy from Gibson, Caldeira, and Spence (Reference Gibson, Caldeira and Spence2003).Footnote 9
RESULTS
We present the results in figure 1. Point estimates in the left panel indicate the effect of the treatments relative to the control condition on specific support for the judge or opinion featured in each experiment using a dichotomized version of our outcome measures. Point estimates in the right panel indicate the effect of those treatments on diffuse support for the judiciary. Treatment effects in the left panel can be interpreted as percentage-point differences in respondents’ approval of the featured judge or opinion; treatment effects in the right panel indicate differences in diffuse support on a 0–1 scale.
Effect of Scandal on Specific and Diffuse Support
Linear regression coefficients for treatments effects. Positive values along the x-axis reflect higher levels of support. Bars around point estimates represent 95% confidence intervals. Specific support is measured with a binary indicator for whether the respondents approved of the judge (Supreme Court nomination and lower court) or opinion (Supreme Court opinion) featured in the vignette. Diffuse support is measured on a 0–1 scale using respondents’ answers to the six questions used by Gibson, Caldeira, and Spence (Reference Gibson, Caldeira and Spence2003). Additional details on the vignettes and question wordings are in online appendix B.
Turning to specific support, we observe consistently negative effects of scandal on respondents’ evaluations of the judges in the Supreme Court nomination and lower-court experiments. The approval rating of the judge in our Supreme Court nomination experiment ranges from 24 to 46 percentage points lower when the judge faces scandalous allegations relative to the control condition. Surprisingly, Justice Kavanaugh’s authorship of the majority opinion in our opinion experiment led to a small increase—although not statistically distinguishable—in approval for the opinion compared to when authorship was attributed to Justice Gorsuch. Our results indicate that scandal exerts negative effects on the public’s support for individual judges, but they do not provide evidence that support for a ruling is affected by a scandalized jurist’s association with it.
We observe markedly different results in the right panel of figure 1 concerning the effect of scandal on diffuse support. Across the seven treatments in our three experiments, six manifest null effects; the seventh—Study 2’s sexual scandal—exerts a statistically distinguishable but substantively small decrease in diffuse support of 0.05. Thus, although the public sanctions scandal-ridden jurists for misdeeds, we find scant evidence that such misdeeds damage judicial institutions. Taken together, these findings demonstrate that the sporadic presence of scandal-ridden individuals does not upend diffuse support for courts.
CONCLUSION
Modern politics pose wide-ranging hazards for public support of the courts—including scandals. Our experimental findings involving individual judges demonstrate that the public negatively evaluates scandalous personal behavior. However, we find no evidence that scandal has broader institutional repercussions. That the public appears to differentiate between the personal behavior of individual judges and the legitimacy of judicial institutions provides some normative reassurance concerning the public’s ability to exercise accountability over judges and the judiciary (Przeworski, Stokes, and Manin Reference Przeworski, Stokes and Manin1999). However, the normative desirability of the public drawing this distinction depends on the prevalence of judicial impropriety. When scandalous behavior is rare, individuals should hold accountable the accused judge by expressing displeasure—which may generate pressure for individual sanctions—but should not punish the broader judiciary—which otherwise is populated with upstanding jurists and ostensibly not responsible for a single judge’s misdeeds. However, this dynamic would lose its normative luster if a new equilibrium emerged in which scandalous behavior among judges became more pervasive. The public should hold accountable institutions in which unbecoming personal conduct that may impair their ability to make impartial, principled rulings is ubiquitous.
Our experimental findings involving individual judges demonstrate that the public negatively evaluates scandalous personal behavior.
This article focuses attention on how personal attributes of judges and other elites, including their prior history of scandal, may influence public evaluations of their respective institutions. Although we find no evidence that scandal damages the judiciary’s diffuse support, myriad opportunities remain to probe the relationship between elites and public support for institutions. Future research should consider whether the effect of scandal on institutional support hinges on the public’s prior level of support. While the federal judiciary’s reservoir of goodwill may shield it from sporadic scandals, scandals may damage legitimacy of institutions with weaker baseline support, such as Congress (Bowler and Karp Reference Bowler and Karp2004). Future inquiries should probe the public’s tolerance for judicial scandal because the public may (and should) diminish its reservoir of goodwill for the courts if scandal becomes pervasive. Exploring the conditions under which a critical mass of scandal-ridden judges or a spate of controversial decisions by the Supreme Court may undermine the foundations of judicial legitimacy would garner interest from a broad array of audiences. Additionally, contextual details—such as (1) a scandal’s recency or severity, (2) the notoriety of the scandalized elites, or (3) how elites respond to allegations of misbehavior—may condition public evaluations of elites and the institutions they inhabit (e.g., Miller and Reeves Reference Miller and Reeves2022; Pereira and Waterbury Reference Pereira and Waterbury2019). Finally, researchers should consider how elites’ characteristics—such as their expertise and ability to descriptively represent the polity—inform the public’s support for institutions more broadly. When attributes of key elites diverge from those of the public, institutional support may suffer.
ACKNOWLEDGMENTS
Authors are listed in alphabetical order per disciplinary convention; however, this ordering does not reflect the authors’ relative contribution to this research project. We thank Eileen Braman, Kayla Canelo, Nathan Carrington, Nicole Kalaf-Hughes, Jonathan King, Christopher Krewson, Elizabeth Lane, Alice Sanford, Jessica Schoenherr, Marcy Shieh, James Spriggs, Andrew Stone, Logan Strother, Guillermo Toral, and Nicholas Waterbury for helpful comments. We also sincerely thank the anonymous reviewers, whose comments significantly improved this article. We presented earlier iterations of this article at the Midwest Political Science Association and the American Political Science Association annual meetings in 2021. This project adheres to the ethical and transparency obligations outlined in APSA’s “Principles and Guidance for Human Subjects Research.” Details on adherence to these principles are in online appendix B. This research was approved by the Institutional Review Boards at Bowling Green State University (#1521845, approved January 17, 2020; #1619591, approved July 21, 2020; and #1638063, approved August 4, 2020) and the University of North Dakota (#201912-138, approved December 18, 2019). This research was supported by a grant from the Bowling Green State University Office of Sponsored Programs and Research (Grant #33000321) and funding provided by the Bowling Green State University political science department.
DATA AVAILABILITY STATEMENT
Research documentation and data that support the findings of this study are openly available at the PS: Political Science & Politics Harvard Dataverse at https://doi.org/10.7910/DVN/OM9HES.
Supplementary Materials
To view supplementary material for this article, please visit http://doi.org/10.1017/S104909652200138X.
CONFLICTS OF INTEREST
The authors declare that there are no ethical issues or conflicts of interest in this research.
Open access
