Why the Supreme Court’s Leak Investigation Failed

The inquiry revealed once again the need for real oversight of the judicial branch.

Illustration showing a hand holding a magnifying glass pointed at a banana peel on the floor (the kind that could cause someone to slip)
Getty; The Atlantic

Immediately following the leak of its draft opinion in Dobbs v. Jackson Women’s Health last spring, the Supreme Court in a press release described the incident as an “egregious” breach of trust. Chief Justice John Roberts directed the marshal of the Supreme Court to investigate the leak. That investigation resulted in a report, issued earlier this year, that did not uncover the source of the leak.

This result did not surprise me. As the inspector general of the Department of Justice and acting inspector general of the Department of Defense for many years, I was often asked to conduct leak investigations. They are notoriously difficult to resolve.

Not all leaks are as egregious as this one. Many do not warrant a full investigation or the upheaval such an investigation can cause to an organization, particularly given that they are rarely successful. But this one clearly did, because the leak damaged the integrity of the Court and its decision-making process, which the Court recognized.

Unfortunately, the leak investigation it conducted demonstrates how not to conduct a leak investigation. It also illustrates, again, the need for internal oversight at the Court, which should have a better permanent capacity to police itself by more credibly investigating alleged misconduct and by identifying faulty procedures proactively.

The first problem in the Court’s leak investigation was whom the Court asked to conduct it. The marshal of the Court, Gail A. Curley, is responsible for overseeing the Supreme Court building’s operations, providing security for the justices and the building, disbursing payrolls, and managing the courtroom, including calling it to order. Curley is a former Army lawyer who does not have experience or expertise in conducting this type of complex investigation.

Even more problematic, the marshal did not have the necessary independence to conduct the investigation. In essence, she was asked to investigate her bosses, the justices, who are in the universe of potential leakers. They supervise her and can fire her. She was conflicted from the start. That is no reflection on Curley or her integrity. Any marshal would have been placed in the same position.

The second problem was how the investigation was conducted. According to the report itself, the probe focused on Court personnel—law clerks and permanent employees, who were intensively interrogated. They were required to sign notarized affidavits. Their personal cellphones were scrutinized. They were questioned as to whether they had talked about Court decisions with anyone, including their spouses, in the Dobbs case or other cases, prior to public release. According to the report, follow-up was pursued on leads relating to the clerks and employees.

The investigation did not treat the justices in the same way. When the report was issued, whether they had even been interviewed was not clear. It makes no mention of investigating the justices at all.

That omission prompted an outpouring of questions and criticism. In response, a day after the report was issued, Curley released a short statement declaring that she had spoken “with each of the Justices, some on multiple occasions.” She said that the justices had “actively cooperated in this iterative process, asking questions and answering mine.” She stated that she had followed up on all credible leads and that none had implicated the justices, so she did not think it necessary to ask them to sign sworn affidavits.

This double standard of investigation was baffling. What was this “iterative” process, with the justices “asking questions and answering mine”? That’s not how investigators normally conduct interviews. Why weren’t the justices required to sign affidavits like the clerks and Court employees were? Were the justices’ phones scrutinized like the clerks’ phones? Were the justices also questioned about whether they had discussed cases with their spouses or whether their spouses had access to the Dobbs draft opinion? Did the justices take the draft opinion home to work on, which may have given family members access to it? What were the credible leads that Curley alluded to in her cryptic statement?

Neither the report nor Curley’s statement gave any reason for this double standard. It was an Alice in Wonderland investigation: The conclusion seemed to come first—let’s focus on the clerks and employees, not the justices, as the likely source of the leak—rather than result from a consistent investigation focusing on all potential leakers.

The third problem was that the investigation’s report was incomplete. A report is credible when it comprehensively explains the investigative process, the facts found, and the investigator’s analysis based on those facts. Others may disagree with the analysis, but the investigation and the evidence should at least be fully described. This report did not do that. It left many questions unanswered, such as the ones above about how the justices were treated, who the “seasoned attorneys and trained federal investigators with substantial experience” assisting Curley in the probe were, and whether the justices had input into how the investigation was conducted.

In addition, the Court’s statement accompanying the report contains a troubling assertion: “The leak was no mere misguided attempt at protest. It was a grave assault on the judicial process.” This implies that the leaker was believed to be someone “protesting” the substance of the draft Dobbs opinion, presumably because the leaker opposed a decision to overturn Roe v. Wade.

That is one plausible theory about who leaked the opinion. Another plausible one is quite different: that the draft opinion was leaked to bolster support for it, and to make any wavering justices in the majority feel pressure to stick with their tentative decision to overturn Roe.

The report itself does not provide sufficient evidence to conclude which theory is more likely. But the comment suggesting that the opinion had been leaked in “protest” undermines confidence that the investigation was unbiased.

Finally, in an attempt to buttress the report’s credibility, Chief Justice Roberts asked Michael Chertoff, a former U.S. attorney, judge on the Third Circuit Court of Appeals, assistant attorney general, and secretary of Homeland Security, to opine on the thoroughness of the investigation. Chertoff wrote in a statement accompanying the report that Curley and her team had conducted a thorough inquiry and that he could not identify any additional useful investigative measures. Yet this testimonial makes the report less rather than more persuasive, as it ignores the report’s weaknesses and the double standard in how the investigation was conducted.

Subsequently, CNN reported that Chertoff and the Court have long-standing financial ties; Chertoff received nearly $1 million in contracts to provide security assessments for the justices. Someone with this sort of connection to the Court may have a financial incentive to maintain good relations with it, and is thus not in the best position to provide an unbiased opinion about the thoroughness of its internal investigation. This creates a conflict, or at the very least the appearance of one.

To its credit, the report identified many flawed Court procedures that had contributed both to the leak and to the difficulty in investigating it. These include the Court’s inability to track who had access to Court documents, its lack of written policies for safeguarding sensitive documents, and its lax and outdated information-security policies.

However, the existence of these widespread deficiencies again demonstrates the need for dedicated, experienced internal oversight at the Court, such as an inspector general for the judiciary, whose job is to help identify problems before they contribute to egregious breaches.

No one likes oversight, but every institution needs it, especially insular institutions such as the Court, which operates without much transparency. The Department of Justice, the FBI, the Department of Defense, and other executive-branch agencies initially resisted oversight by an inspector general. They argued that it could undermine their independence, second-guess their decisions, or harm their operations. None of that proved true. Inspectors general were established in each agency, and although not perfect or a solution to every problem, their oversight helped make the agencies more accountable, transparent, and able to address allegations of misconduct.

The value of such oversight should not be a partisan issue. I have worked with strong leaders in the administrations of each party who recognized its benefits. For example, Attorney General John Ashcroft supported independent oversight, even when it resulted in criticism of his actions, because he said it helped in the effort to continuously improve agency operations. Similarly, Secretary of Defense James Mattis encouraged the inspector general’s office to bring problems to his attention, which he said was crucial to correcting deficiencies in agency operations. Such oversight might be painful in the short run, but it makes the agency healthier in the long run.

The same is true at the Supreme Court. Dedicated, permanent oversight could examine Court processes to help prevent problems before they occur, and more credibly investigate misconduct when it happens. Unfortunately, the flawed and incomplete leak investigation by the Court undermined, rather than strengthened, trust in the Court and in its ability to police itself.