Mandamus and the Proper Role of Amicus Curiae in the Michael Flynn Case Commentary
(c) Wikimedia Commons (Gage Skidmore)
Mandamus and the Proper Role of Amicus Curiae in the Michael Flynn Case

On Monday, Judge Emmet G. Sullivan of the U.S. District Court for the District of Columbia filed his response to the petition for mandamus brought by General Michael Flynn in the D.C. Circuit regarding Sullivan’s handling of the Justice Department (DOJ) Motion to Dismiss the false statement charges against Flynn. Flynn had pleaded guilty to these charges in 2017 and again in 2018. The mandamus petition asked the D.C. Circuit to, 1) grant the government’s motion to dismiss, 2) block Sullivan’s appointment of a retired federal judge, John Gleeson, as an amicus curiae (friend of the court) opposing the motion, and, 3) assign the Flynn case to a different district judge.

Sullivan’s Monday response—prepared by Washington, D.C. attorney Beth Wilkinson, who also represented Justice Brett Kavanaugh in his confirmation hearings—argued against each of these three steps, contending that Sullivan could decide the government’s motion, appoint an amicus, and maintain impartiality toward Flynn. The D.C. Circuit should grant the motion in limited part by constraining the amicus’s role. That would involve holding Judge Gleeson to analysis of the legal and factual issues raised by the government’s motion to dismiss on its face and barring Judge Gleeson from either additional inquiry into DOJ’s motives or the second facet of his charge from Judge Sullivan: determining whether contempt of court charges should be filed against Flynn. To preserve the ordinary course of adjudication of the government’s motion and avoid a judicial merry-go-round of judges in the long-running Flynn case, the D.C. Circuit should deny the mandamus petition’s first and third requests and allow Judge Sullivan to decide the motion to dismiss, subject to appeal.

As Sullivan’s response notes, a writ of mandamus is an “extraordinary” remedy—one well outside the norm—because it makes a sitting judge into an adverse party. A petition for mandamus argues that the judge has so exceeded their power that immediate appellate review is necessary, replacing the typical course in which a trial judge makes a decision and the appellate court reviews that ruling. To win a mandamus petition, a party must meet a high threshold, showing that, 1) there are no other adequate ways to safeguard the party’s rights; 2) the arguments for mandamus are beyond dispute; and, 3) granting the writ is “appropriate under the circumstances.” To explore how this demanding standard works in Flynn’s case, we need to briefly provide some background on the twists and turns of that prosecution.

Lawfare readers doubtless know that former Lieutenant General Michael Flynn, who had served as head of the Defense Intelligence Agency under President Obama and was eventually removed from that position, became a national security advisor to then-candidate Donald Trump during the 2016 presidential campaign. Once Trump won, he designated Flynn for the vital post of National Security Advisor. At that point, Flynn reached out to Sergey Kislyak, the Russian ambassador to the United States. That outreach was particularly fraught, given the backdrop of Russian interference in the 2016 election.

The Mueller Report (Vol. 1, p. 9) described “numerous links” between persons close to the Russian government and persons with ties to the Trump campaign. Special Counsel Robert Mueller noted that he did not find sufficient evidence of criminal behavior on the part of Trump campaign figures to support prosecutions on this score. Moreover, the Mueller Report did not identify any contacts during the campaign between Flynn and Russian figures regarding election interference. Nevertheless, against the backdrop of the Russia probe that culminated in the Mueller Report, Flynn’s stated belief in the need for a closer relationship with Russia and his efforts to cultivate that relationship with Kislyak in late December, 2016 triggered the interest of the Federal Bureau of Investigation (FBI).

Flynn had conversations with Ambassador Kislyak about two matters: 1) Russian reactions to sanctions just issued by President Obama to respond to Russian election interference, and, 2) Russian assistance in delaying or defeating a U.N. Security Council resolution sponsored by Egypt asking Israel to halt settlements in the West Bank. After the conversations took place, Flynn told Vice President-Elect Pence and other incoming officials that he had not engaged in such talks. (Mueller Rpt., Vol. II, pp. 29-30.).

The FBI had conducted lawful surveillance of Kislyak, and thus knew the contents of Flynn’s communications with the Russian ambassador. On January 24, 2017, FBI agents interviewed Flynn about those contacts. Against the urging of Acting Attorney General Sally Yates, FBI Director James Comey made the decision to proceed with the interview without first informing the White House of Flynn’s lie to Pence. The FBI also did not notify DOJ, including Acting Assistant Attorney General Mary McCord, of the timing of the interview.

Positions differ about whether Flynn’s statements to the agents at his interview crossed the line from hedging and half-truths into outright false statements. Both Flynn today and the DOJ’s motion to dismiss assert that Flynn didn’t cross that sometimes blurry line. In contrast, Flynn’s guilty plea and critics of the government’s motion to dismiss assert that the line was crossed (see this post by Robert Litt, former General Counsel for the Office of the Director of National Intelligence during the Obama administration). FBI Director James Comey told Congress that the question of whether Flynn lied is a “close one.” (Motion to Dismiss, Exh. 5, p. 10).

The tension between Flynn’s guilty plea and the contentions in the government’s motion to dismiss is an important basis for Judge Sullivan’s appointment of former Judge Gleeson as an amicus (Sullivan Response, pp. 5-8). Flynn is currently seeking to withdraw his plea, citing new government disclosures discussed in the motion to dismiss that according to Flynn show that the FBI staged the interview to get Flynn to lie and then either prosecute him or “get him fired.” In his plea and related documents in the voluminous case file, Flynn acknowledged under oath that he had made false statements to his FBI interviewers. In materials supporting his plea to one count of making false statements, Flynn also acknowledged under oath that he had made false statements and material omissions in another matter, involving filings about his ties to Turkey under the Foreign Agents Registration Act (FARA).

A prosecution for false statements hinges on the falsity and materiality of the defendant’s words. The government’s motion to dismiss falls short on the falsity element, particularly given its failure to address why Flynn would have pleaded guilty twice to making false statements if his statements did not rise to that level. The government’s discussion of materiality fails on a far more epic scale. Its restrictive reading of materiality badly distorts the law and offers dangerously inadequate guidance to both DOJ lawyers and the FBI. That said, the government’s motion to dismiss offers the germ of an argument that continuing the Flynn false statement charges would not promote either counterintelligence goals or the productive collaboration of DOJ lawyers and the FBI. However, the government motion’s distorted conception of materiality obscures exactly the arguments that DOJ should now stress.

The concept of materiality in the D.C. Circuit and most other federal appellate courts is broad, to ensure that persons answering questions posed by an agency such as the FBI do not deceive the questioners and thereby impair the agency’s mission. In United States v. Moore, decided by a panel that included Judges Douglas Ginsburg, David Tatel, and then-Judge Brett Kavanaugh, the court—in an opinion by Judge Ginsburg—held that a statement is material “if it has a natural tendency to influence, or is capable of influencing, either a discrete decision or any other function of the agency to which it was addressed” (emphasis added).

The government’s motion to dismiss the charges against Flynn relies on a narrow definition of materiality that clashes with the Moore court’s holding. The motion notes what all parties accept: that the FBI agents questioning Flynn already knew the contents of Flynn’s communications with Kislyak. According to DOJ, since the agents knew what was said, actually interviewing Flynn about the communications did not contribute to the FBI’s function and was hence immaterial. (Motion to Dismiss, p. 14.). Moreover, the motion notes correctly that, prior to learning of Flynn’s contacts with Kislyak, the FBI had drafted a recommendation to close the Flynn investigation, since at the time of that recommendation it had not acquired information to suggest that Flynn was a Russian agent. (Motion to Dismiss, p. 13). According to DOJ, the new information about the Kislyak contacts did not justify continuing the investigation. Indeed, DOJ argues that the calls were “entirely appropriate on their face.” (Motion to Dismiss, p. 14.).

To reach this conclusion, the motion to dismiss ignores the entire context and premise of the Russia probe. That probe started because the combination of clear Russian campaign interference and “numerous” contacts between Russia and Trump campaign figures created a reasonable basis for investigating whether the contacts related to the election interference in a manner that violated U.S. law. Ignoring this predicate, DOJ’s motion to dismiss instead analyzes the FBI’s investigation by assuming facts that Special Counsel Mueller only stated after lengthy and careful investigation: that insufficient evidence existed to warrant criminal prosecution of Flynn or any other Trump campaign figure for collaboration in Russian election interference. But that analysis adopts the wrong perspective for assessing the FBI agents’ belief in January, 2017, which must address what the agents knew (and didn’t know) at the time, not some later date once their investigation was complete. In this sense, DOJ’s conception of materiality puts the cart before the horse.

Viewed from the standpoint of law enforcement in January, 2017, continuing the Flynn investigation was material to the FBI’s vital counterintelligence function. The FBI and other agencies have a counterintelligence mission that augments the FBI’s law enforcement task. As part of that mission, the FBI is always on the lookout for unauthorized foreign interventions on U.S. sovereign prerogatives, including the activities of U.S. officials and the conduct of the U.S. political system. Continuing the Flynn investigation in light of his contacts with Kislyak and learning whether Flynn would level with the FBI about those contacts was entirely consistent with this counterintelligence mission.

The weakness of DOJ’s legal arguments against Flynn’s guilt actually masks credible policy arguments that DOJ could have stressed. For example, suppose that DOJ chose to abandon its narrow definition of materiality. Instead, DOJ might argue straightforwardly that, with the benefit of Mueller’s full investigation, it had become clear that holding Flynn to his plea would no longer serve counterintelligence goals. For example, DOJ could have addressed the concern raised in Marty Lederman’s excellent post that Flynn’s lie to Pence made him subject to Russian blackmail. Whatever the merits of this concern at the time of Flynn’s interview with the FBI, it ceased to be a concern once the White House fired Flynn after learning of his lying to Pence. DOJ gestures at this argument by invoking the “equities” of the case (Motion to Dismiss, p. 20), but fails to make it squarely.

That brings us to the standard of review under Rule 48(a) of the Federal Rules of Criminal Procedure, which requires “leave of court” to dismiss pending charges. As Thomas Ward Frampton wrote in a perceptive new piece, Rule 48(a)’s conditioning of dismissal of charges on the “leave of court” does not merely protect defendants against prosecutorial harassment through repeated cycles of charges and dismissals—a rationale the Supreme Court cautioned against in Rinaldi v. United States. Rule 48(a) also requires leave of court to ensure some check on prosecutorial discretion (see also this piece by Andrew Crespo, Laura London᷈o Pardo, Kristy Parker & Nathaniel Sobel). The “leave of court” requirement does not license courts to substitute their predilections for prosecutors’ discretionary weighing of priorities and policies. That kind of freewheeling judicial review would impinge on the separation of powers. However, as the D.C. Circuit noted in United States v. Ammidown (1973), a court reviewing a Rule 48(a) motion can inquire whether the motion displays “considered judgment” and constitutes an “application made in good faith.”

While the brief of the Solicitor General (SG) in the D.C. Circuit supporting the mandamus petition suggests that Rule of 48(a)’s “leave of court” language permits only the most perfunctory judicial review, the case law does not relegate the court to this nugatory role. According to the SG’s brief, the Constitution mandates that the court merely rubber-stamp the prosecution’s motion in each case, since by definition no party before the court opposes it. Hence, there is no legal dispute or “case or controversy” between the parties to bring to court. Since Article III of the Constitution only gives federal courts jurisdiction over “cases and controversies,” any inquiry into the basis for DOJ’s motion to dismiss exceeds the district court’s power. (SG Brief, pp. 12-16.) This constitutional argument seems unduly stark. Federal courts on occasion make inquiries on their own, including those for decisions on whether or not to accept a guilty plea or appoint a prosecutor to try a lawyer or party for contempt of court. Indeed, a case that the SG cites, United States v. HSBC Bank USA, N.A., suggests that a requirement that a court automatically grant a prosecutor’s motion to dismiss would fail to adequately test whether a Rule 48(a) motion was made in bad faith and was hence “clearly contrary to manifest public interest.” The HSBC court saw no constitutional problem with the court undertaking this inquiry, at least in a bounded manner.

Judge Sullivan’s response rejects the rubber-stamp role envisioned by both the DOJ motion to dismiss and the SG’s brief in the D.C. Circuit. The response cites Ammidown and a more recent D.C. Circuit decision, United States v. Fokker Servs. BV (2016), for the proposition that a court need not automatically grant a Rule 48(a) motion. Instead, as Ammidown explained, the court may at least inquire whether the dismissal serves “due and legitimate prosecutorial interests.” However, Judge Sullivan’s response understates the deferential and cabined nature of the court’s inquiry and how that limits the role of an amicus curiae.

Returning to the standard governing writs of mandamus and the relief sought by Flynn from the D.C. Circuit, it is hardly “beyond dispute” that Judge Sullivan exceeded his power in appointing Judge Gleeson to serve as amicus, as long as Judge Gleeson focuses on the merits of the government’s legal and factual arguments. District courts frequently permit the submission of briefs by amici curiae to educate the judge about the merits of the case. That role is even more important here, given the government’s failure to fully assess the falsity of each of Flynn’s statements and the unduly parsimonious and unworkable definition of materiality that the government proffered.

That said, it is indeed “beyond dispute” that the broader role outlined for Judge Gleeson in Judge Sullivan’s original order would be counterproductive. The second component of the order would allow Judge Gleeson to make recommendations regarding whether Flynn should be prosecuted for contempt of court. In addition, Judge Sullivan’s order might be read to authorize Judge Gleeson to conduct a roving inquiry into prosecutorial motives for the motion to dismiss. Judge Sullivan’s response alludes to this scenario in suggesting that the usual “presumption of regularity” accompanying prosecutors’ decisions is inappropriate where the government’s motion was made by an Acting U.S. Attorney and not joined by any of the line attorneys who had worked on the case for a substantial period of time. (Sullivan Response, pp. 27-28.) Moreover, critics of the government’s motion have suggested that it is one aspect of a concerted plan by Attorney General William Barr to discredit the entire Russia probe.

Expanding the amicus’s role in the way that Judge Sullivan’s order provides would have substantial costs for both legal institutions and individual rights. The expanded amicus mandate would chill a defendant’s right to seek to withdraw a plea prior to sentencing and impair the independence of prosecutorial decisionmaking. Any defendant would hesitate to move to amend or withdraw a plea if the price was a contempt finding and a possible fine or imprisonment. As the Supreme Court has held, a contempt prosecution must meet a high standard: the party to be held in contempt must have “obstructed” the judicial process. In re McConnell (1962). That high standard usually requires a showing that a party has defied a court’s order or disrupted actual judicial proceedings. A defendant’s effort to withdraw a plea does not rise to that level. Moreover, prosecutors would be reluctant to move for Rule 48(a) relief in appropriate cases if that triggered a protracted inquest into their motivations. Even the prospect of a contempt citation would discourage defendants and prosecutors alike, even in cases where the relief they sought was necessary and appropriate.

Considering the first prong of the mandamus test, the injury from such effects would be irreparable and beyond an appellate court’s ability to cure with a reversal of a denial of the Rule 48(a) motion. These harms would inure to the defendant and DOJ even if Judge Sullivan ultimately granted the government’s motion. Granting a writ of mandamus to head off these harms is clearly “appropriate under the circumstances.”

However, that is the extent of the relief that petitioners should obtain from the D.C. Circuit on their mandamus petition. Given the persistent twists and turns of the Flynn prosecution, the district court should take the first crack at adjudicating the government’s motion. Preempting the ordinary course of business through a ruling in the D.C. Circuit would short-circuit that important phase in the process. Moreover, reassigning the case to another judge would also be inappropriate, given the case’s tangled history and voluminous filings. Judge Sullivan is the second judge to preside over the Flynn matter, and he has done so for over 2 1/2 years. Appointing a third judge in the Flynn matter would not be wise or efficient, absent a showing of clear prejudice against Flynn on Judge Sullivan’s part.

Judge Sullivan has acted reasonably throughout a challenging case. Acting on the government’s recommendation when it still wished to prosecute Flynn, Sullivan repeatedly put off sentencing to allow the government to extract the maximum cooperation from the defendant. While at one point Judge Sullivan may have engaged in hyperbole by referring to Flynn’s contacts with Kislyak as arguably “treasonous,” those remarks seemed designed to persuade Flynn of the seriousness with which the judge took the cooperation that Flynn had pledged to provide. (Sullivan Response at 10.). Moreover, Judge Sullivan had entered a Brady v. Maryland disclosure order early in his time on the case that resulted in Flynn getting a raft of helpful materials, including the FBI agents’ post-interview “impressions” that Flynn was not seeking to deceive them. Viewed as a whole, these facts show a judge committed to his duty in a difficult case, not a judge prejudiced against the defendant.

In sum, Judge Sullivan’s response is a persuasive answer to Flynn’s mandamus petition, except on the question of the amicus’s role. There, the D.C. Circuit should narrow the scope of Judge Gleeson’s purview. Granting the mandamus petition to that extent will enhance the prospects for a fair, principled, and prudent adjudication of the government’s motion to dismiss the false statements charge against Flynn.


Professor Peter Margulies teaches National Security Law at Roger Williams University School of Law in Rhode Island. He has spoken at events sponsored by the American Bar Association, Yale Law School, and Columbia Law School on the intersection of surveillance, cybersecurity, and privacy in U.S. and EU law. In addition, Professor Margulies has served as co-counsel for amici curiae in prominent cases, including Humanitarian Law Project v. Holder, 561 U.S. 1 (2010) (holding that statute prohibiting material support to foreign terrorist organizations did not violate the First Amendment).

With Geoff Corn, Jimmy Gurule, and Eric Jensen, Professor Margulies is co-author of NATIONAL SECURITY LAW: PRINCIPLES AND POLICY (Wolters Kluwer 2d ed. 2019). Professor Margulies’ articles include, Sovereignty and Cyber Attacks: Technology’s Challenge to the Law of State Responsibility, 14 Melb. J. Int’l L. 496 (2013), Global Cybersecurity, Surveillance, and Privacy: The Obama Administration’s Conflicted Legacy, 24 Ind. J. Global Legal Stud. 459 (2017), and Autonomous Weapons in the Cyber Domain: Balancing Proportionality and the Need for Speed, 96 U.S. Naval War Coll. Int’l L. Stud. __ (forthcoming 2020),


Suggested citation: Peter Margulies, Mandamus and the Proper Role of Amicus Curiae in the Michael Flynn Case, JURIST – Academic Commentary, June 5, 2020,

This article was prepared for publication by Tim Zubizarreta, JURIST’s Managing Editor. Please direct any questions or comments to him at

Opinions expressed in JURIST Commentary are the sole responsibility of the author and do not necessarily reflect the views of JURIST's editors, staff, donors or the University of Pittsburgh.