Courts & Litigation Criminal Justice & the Rule of Law Democracy & Elections

In Fulton County, Fear Not Removal

Anna Bower, Alan Z. Rozenshtein, Benjamin Wittes
Thursday, September 7, 2023, 9:04 AM
The issue is complicated, but removal of the Fulton County case to federal court would not be a disaster—and is probably the right answer.
Former White House Chief of Staff Mark Meadows in Nashville in 2023. (Gage Skidmore,; CC BY-SA 2.0,

Published by The Lawfare Institute
in Cooperation With

To hear many pundits discuss the question of Mark Meadows’s motion to “remove” the Fulton County election interference case to federal court, you’d think it posed a straightforward question.

“Even though the legal hurdle is low and the law is favorable to federal officers, Meadows faces a seemingly insurmountable barrier,” write a trio of analysts in Just Security.

“It’s a legal, historical and moral imperative ... to keep the case in Fulton County,” writes Jennifer Rubin in the Washington Post.

The Post editorializes:

There are many valid reasons for protecting federal officers from harassment by state and local prosecutors, a reality underscored by the South’s dark history of resisting federal power. But Mr. Meadows’s effort to keep Mr. Trump in the White House isn’t comparable to U.S. marshals trying to enforce federal rights that Jim Crow states were violating. Mr. Meadows’s legally deficient motion for removal threatens states’ ability to enforce their laws and uphold the integrity of their elections.

Slow down, folks. 

The question of whether the Fulton County case should be removed to federal court—in whole or in part, temporarily pending resolution of certain immunity questions or all the way through trial—is actually a gnarly one with a lot of ins and outs and twists and turns. It’s a hard question, on which case law provides inadequate guidance and about which the answer is not obvious. (Writing in Lawfare, Lee Kovarsky argues that Meadows’s removal motion is “weak” but, to his credit, recognizes that Meadows has a “nontrivial chance of succeeding.”) Federal removal is also not the hill on which the rule of law is going to make its last stand. 

Meadows—and Donald Trump, for that matter—can get a fair trial in Fulton County court. They can also get a fair trial in federal court. The fate of justice and accountability for Jan. 6 does not turn on the success of the removal litigation.

In our view, removal—at least for purposes of litigating the forthcoming issue of federal Supremacy Clause immunity—is probably the best approach for U.S. District Judge Steve Jones, who last week held a day-long evidentiary hearing at which Meadows testified. This is a provocative proposition, in part because there is so much anxiety about the perceived disadvantages to the prosecution if the case is removed. But these perceived disadvantages are largely a function of myths that require a bit of busting before one can examine the question of removal dispassionately. 

One misconception is that removal to federal court would render the charges eligible for presidential pardons should Trump or a sympathetic Republican prevail in the next election. But this is just wrong. Removal does not change the nature or substance of the state charges in the indictment. The charges would remain Georgia state charges under Georgia law. And presidential pardons can be granted only for “offenses against the United States,” meaning federal crimes. So even if the case were removed to federal court, neither Trump nor any other president could issue pardons related to the Fulton County case.  

Then there’s the matter of the jury pool—assuming the matter ever reaches a jury. While some reports have suggested that the jury pool in federal court would be significantly more conservative than the jury pool in Fulton County, that’s not accurate. Standard practice under the jury plan in the Northern District of Georgia involves drawing jurors from the division in which the case is assigned—here, the Atlanta division. That division comprises 10 counties in the metro Atlanta area: Cherokee, Clayton, Cobb, DeKalb, Douglas, Fulton, Gwinnett, Henry, Newton, and Rockdale. While some of those counties are more politically conservative than Fulton County, the overall difference is not dramatic. According to the Wall Street Journal, 73 percent of voters in Fulton County voted for Joe Biden during the 2020 election, while only 26 percent voted for Trump. Across the 10 counties in the Atlanta division, by comparison, Biden garnered 65 percent of the vote to Trump’s 33 percent.

Others have worried that Trump or others could move for a district-wide venire, which would include the much more conservative counties in the rural parts of North Georgia. But we doubt that any such effort would succeed. Requests for a district-wide venire have, to our knowledge, only twice been granted in Georgia federal district courts. And those requests were granted in the Southern District of Georgia, which follows a different jury plan that specifically authorizes discretionary use of district-wide juries. The jury plan in the Northern District of Georgia contains no similar provision.

Another source of anxiety around removal is rooted in the idea that District Attorney Fani Willis and her team would be litigating on unfamiliar turf and according to federal procedural rules with which they may have little experience. Here, too, we think the concern is overblown. For one thing, Georgia’s evidence code is not all that different from the Federal Rules of Evidence. 

And Willis’s office is hardly new to litigating in federal court in any event: Last year, after Fulton County prosecutors brought state law charges against two federal marshals who killed a man during a 2016 shootout, a judge in the Northern District of Georgia permitted removal to federal court. Since then, Fulton County prosecutors have continued to litigate the case in federal court. If the Meadows (or Trump) case were to be removed, Willis and her team will be able to draw on its experience with the Heinze case. And, if necessary, Willis can also hire outside counsel with experience in federal courts to assist with the prosecution.

The biggest realistic impact of removal on the trial likely involves the issue of cameras in the courtroom—which are the norm in Georgia state court but strictly verboten in federal trial courts. On this issue, removal would make a huge difference, but it’s not obvious that the difference favors the defense. Attention, after all, is Trump’s lifeblood. And it could be that having the case litigated in public throughout the election cycle is just what he would need to keep people talking about him while he’s stuck in court instead of campaigning. 

It is also important to separate the question of removal from the question of whether Supremacy Clause immunity may protect Meadows, Trump, and others. These two questions are related, but they are emphatically not the same. And some of the anxiety about removal has its roots in the concern that the same logic that leads to removal will ultimately lead to the case’s dismissal. 

This conflation is a mistake. The test for removal invokes the question of immunity, but it is substantially less rigorous than the test for Supremacy Clause immunity. As the Eleventh Circuit described the test for removal back in 2017, to win removal, a defendant:

must show that [he] is a person within the meaning of the statute who acted under a federal officer. Second, [he] must show that [he] performed the actions for which [he] is being sued under color of federal office. Stated another way, [he] must show “a causal connection between what the officer has done under asserted official authority and the action against him.” Third, [he] must raise a colorable federal defense. 

In other words, to win on his removal motion, Meadows does not need to show that he has a winning argument for federal Supremacy Clause immunity. He needs to show only that he has a colorable argument for Supremacy Clause immunity. Think of this gap, the space between a colorable and a winning argument in support of immunity, as the space in which a federal—not a state—court needs to decide that the defendant must stand trial. 

Color us skeptical that Meadows has a winning argument for federal Supremacy Clause immunity. The grand jury has accused him of taking a string of actions that the district attorney argues are wholly outside of his role as the president’s chief of staff:

Count 1 of the Indictment alleges that the defendant, while associated with a criminal enterprise, unlawfully conspired and endeavored to conduct and participate in the enterprise through a pattern of racketeering activity, both directly and indirectly. That conspiracy contained a common plan and purpose to unlawfully change the outcome of Georgia’s presidential election in Mr. Trump’s favor. 

Count 28 charges the defendant with soliciting Georgia Secretary of State Brad Raffensperger to violate his oath of office by unlawfully “altering, adjusting, or otherwise influencing” the certified returns of the presidential election in Georgia. Neither of these alleged acts fall within the scope of a Chief of Staff’s official duties, and the defendant’s proffered federal defense of Supremacy Clause immunity cannot apply to them. 

We believe that when the record is full, this argument will ultimately prevail—both against Meadows and against Trump. Based on the allegations in the indictment along with what we know from the Jan. 6 committee report, the Jan. 6 federal indictment, and years of dogged journalism, it’s increasingly clear that Trump knew that he lost and intentionally listened only to sycophants who told him what he wanted to hear. Trump’s bad faith, and that of those around him, is so monumentally egregious that denying him, Meadows, and the others in his coterie immunity would not meaningfully chill future good-faith presidents in the exercise of their legal authorities.

But that is not the question before the court just yet. The question right now is whether Meadows was operating under the color of his office, and whether in doing so, he has a colorable claim to a federal defense of this type. And on this matter, Meadows’s attorneys have a point

Nothing Mr. Meadows is alleged in the indictment to have done is criminal per se: arranging Oval Office meetings, contacting state officials on the President’s behalf, visiting a state government building, and setting up a phone call for the President. One would expect a Chief of Staff to the President of the United States to do these sorts of things. 

Before the court has amassed the whole record and heard full briefing and evidence on the question of immunity, on what basis is it to conclude that the meetings and calls Meadows arranged were in support of a coup—rather than the sort of things “one would expect a Chief of Staff to the President of the United States to do” as part of his federal duties? Surely the judge is not merely to accept the prosecutor’s say-so as to Meadows’s mens rea. Just as surely, he is not to glean it from extra-record sources like the assumptions of cable news commentators or the Jan. 6 committee report. The judgment has to be based on record evidence.

Yet the evidentiary hearing held by Judge Jones last week only goes so far in answering this question. In it, over the course of a number of hours, Meadows testified that he took each of the specific acts of which the indictment accuses him in his capacity as chief of staff to President Trump as part of his federal function. In a couple of cases, he contested the facts at issue, but broadly, he contended that he did the things the indictment alleges but claimed he did them as part of his performance of his federal duties. 

In turn, the prosecution sought to trip Meadows up and get him to admit that he was acting for political purposes outside of his federal role but on behalf of Trump the candidate, or Trump the person. And it did so with some success at times—getting him, for example, to admit that the meetings he arranged were with campaign lawyers, not government lawyers. The prosecution also put on testimony from Raffensperger and another lawyer that cast Meadows’s activity wholly in a campaign, not a governmental, role.

But that, in turn, set up a bit of a conundrum in Jones’s mind: What if some, but not all, of the specific acts of which the defendant is accused were within the scope of his federal employment? The judge sought supplemental briefing on this question, and the parties took predictable positions on the subject.

In our view, removal is appropriate because the decision that a federal immunity defense is not available to Meadows (and others) should itself come from a federal court. 

Imagine, for a moment, that a presidential election truly was stolen. Under such circumstances, wouldn’t we want the president to aggressively investigate that possibility and be minimally deterred by the possibility of later liability in protecting the integrity of the vote? If so, then we cannot categorically state that actions like those Trump and Meadows took are outside the powers of the presidency without a close examination of the facts. This does not mean that immunity should attach to any such action; it might or might not attach, depending on the attendant facts and the incentive effects of not granting immunity to that action, given all of its underlying context. But it does mean that a judge needs a fully developed record to make the decision. Do we really want the final word on that question to be a state court system answerable only to the people of a single state?

What’s more, much of the evidence that we use for our judgment about Trump’s state of mind and bad faith is either in prosecutorial allegations or otherwise in venues—the Jan. 6 committee report or news reports—that have not gone through the adversarial legal process. It is true that D.C. District Judge Amit Mehta in the Blassingame litigation held that Trump’s actions did not deserve immunity. To the extent that he based that on a categorical argument regarding the president’s role in the Electoral College certification, we disagree with his reasoning. And while the decision can also be justified on the grounds that Trump’s conduct on Jan. 6 was so egregious as not to create worries about the chilling effect on legitimate presidential conduct—this is what we expect the D.C. Circuit to rule—the procedural posture of Blassingame was also fundamentally different from the Georgia case. Blassingame was a civil case decided on a motion to dismiss, in which all reasonable inferences were to be taken against Trump. Here, while Meadows has the burden of establishing the case for removal, he’s not required to admit to everything in the indictment.

There are two more reasons why removal is desirable here. The first is that the law of presidential official acts and immunity is, even by the standards of Article II doctrine, unclear and unsettled. We know that this case is going to present the question of the scope of the president’s official duties, both through litigation by Meadows and others and likely through Trump himself. The law of presidential immunity should be decided in federal, not state, court.

And indeed the question will be decided in federal court no matter what, because the federal questions will make their way up through the Georgia state system and ultimately be appealable to the Supreme Court on direct review and on any subsequent collateral attack. But that’s a particularly messy way to present questions we all know are coming and that present the most important presidential power questions to come before the Supreme Court since Watergate. Having a federal district court review the immunity claim at the front end and having that matter appealed cleanly would be a much more orderly mode of adjudication.

There are broader policy considerations, especially around the dangers of allowing state-level prosecutors to go after presidents, that point in the same direction. There’s a case for not letting any county-level prosecutor indict a former president without some minimal degree of federal oversight—especially when that oversight doesn’t meaningfully impede the state’s legitimate interest in prosecuting the case. That oversight is what removal provides, a gut check by the federal court system whether there is some federal doctrine or policy that would protect the officeholder against the state criminal action. 

In brief, the right procedure is for the case to be removed, for the decision on Supremacy Clause immunity to be rendered in federal court and—if Meadows’s motion to dismiss is denied—for the case to be remanded to Fulton County court for trial. 

Indeed, removal need not be permanent. If Meadows, Trump, or one of their co-defendants were to succeed in removing the case to federal court, Judge Jones could later remand the case back to state court once the “federal defense” that supported removal is dismissed.  

The district court’s authority for remanding in such circumstances derives from the mandatory remand provision in 28 U.S.C. § 1447, which provides that, “[i]f at any time before final judgment it appears that the district court lacks subject matter jurisdiction, the case shall be remanded.” At the removal stage, subject matter jurisdiction in § 1442 removal actions is predicated on the existence of a “colorable” federal defense. As such, once a district court determines that there is no valid federal defense or immunity, it can—and may indeed be required to—remand the case to state court.

Unfortunately, we are not aware of any criminal cases that were properly removed to federal court under § 1442 and later remanded back to state court following denial of a federal defense or immunity claim. And in the context of civil cases removed under § 1442, there is some unfortunate authority in the opposite direction. At least two circuit courts, though not the Eleventh Circuit, have held that the district court does not have discretion to remand any portion of the case once it acquires jurisdiction by removal under § 1442.

In Jamison v. Wiley, a federal employee was sued in state court for civil claims alleging sexual assault and intentional infliction of emotional distress. The case was removed to federal court under § 1442, but the district court subsequently remanded the case back to state court after it found that the defendant had not acted in the scope of his employment as a federal officer. On appeal, the Fourth Circuit held that the lower court erred in remanding the case back to state court. The court concluded that “[n]othing in the federal removal statutes authorizes the remand of a case that has been properly removed under § 1442(a)(1) on the ground that the federal employee’s immunity defense is later rejected.” Citing an influential litigation treatise, the court reasoned:

By raising a colorable federal defense in his removal petition, the defendant-official transforms the otherwise nonremovable state-law action into one that falls within the federal court’s “arising under” jurisdiction. … That the federal court ultimately rejects the federal defense that supported removal under § 1442(a)(1) does not mean that it thereby loses subject matter jurisdiction over the removed action; “the jurisdiction of the federal courts over a properly removed action will not be defeated by later developments in the suit.”

The Fourth Circuit later reaffirmed the principle it set out in Jamison in Mangold v. Analytic Services, Inc. While Mangold dealt with a different federal removal provision under the Westfall Act, 28 U.S.C. § 2679, the court similarly held that the district court erred in remanding the case to state court after it rejected the federal immunity defense. The court wrote that “the effect recognized in Jamison surely must be the same: that the jurisdiction properly acquired by the removal was effectively mandatory and did not permit a discretionary remand following denial of the federal immunity defense.”

The Sixth Circuit—albeit in a footnote—has also cited the language in Jamison to stand for the proposition that a federal court does not lose subject matter jurisdiction if it ultimately rejects the federal defense raised in support of removal.

We think these cases are, at the very least, contestable. Where federal jurisdiction is predicated on the presence of a federal defense, rejection of the federal defense should at minimum give the court the discretion to remand to state court. After all, the Supreme Court has observed that  “one of the most important reasons for removal is to have the validity of the defense of official immunity tried in a federal court.” Once that question has been resolved, the federal interest in preventing the state of Georgia from handling its own cases is gone. 

But the stakes in the question are, in any event, relatively low. If Judge Jones determines that he must keep the case for trial himself, so be it. 

There may even be certain advantages to trying the case in federal court. With little more than a year before the 2024 presidential election, time is of the essence—yet past examples of multi-defendant RICO trials in Fulton County courts have been anything but efficient. In 2014, jury selection in the 12-defendant RICO case related to the Atlanta Public Schools test cheating scandal lasted more than six weeks, with a verdict being handed down more than six months later. Presently, a sprawling RICO case against more than a dozen individuals involved in an alleged criminal gang, dubbed “Young Slime Life,” is on month eight of jury selection.  In the Trump case, jury selection alone could take a year or more. 

Compounding the general glacial pace of RICO trials in Fulton County is the fact that cases in Fulton County courts are already backlogged. Last year, an order issued by Chief Judge Ural Glanville estimated that the total pending felony criminal matters in Fulton County amounted to 18,014 cases. 

Against this backdrop, a Trump RICO case in federal court may move at a more efficient pace. Jury selection in federal courts rarely lasts more than several weeks. And federal judges tend to be more exacting when it comes to case management—a product, in part, of their relatively light caseloads as compared to state superior court judges, as well as the relatively greater resources available to them, particularly with respect to court administrators and clerks. 

The broad point here is that those lining up to plant the flag on opposing removal in the Fulton County case are oversimplifying. They are oversimplifying some very murky law. They are also oversimplifying the equities, which don’t obviously favor trial in state court. And they are being too dismissive of the federal interest in ensuring that federal courts resolve questions of federal defenses.  

Anna Bower is Lawfare’s Legal Fellow and Courts Correspondent. Anna holds a Bachelor of Laws from the University of Cambridge and a Juris Doctorate from Harvard Law School. She joined Lawfare as a recipient of Harvard’s Sumner M. Redstone Fellowship in Public Service. Prior to law school, Anna worked as a judicial assistant for a Superior Court judge in the Northeastern Judicial Circuit of Georgia. She also previously worked as a Fulbright Fellow at Anadolu University in Eskişehir, Turkey. A native of Georgia, Anna is based in Atlanta and Washington, D.C.
Alan Z. Rozenshtein is an Associate Professor of Law at the University of Minnesota Law School, a senior editor at Lawfare, and a term member of the Council on Foreign Relations. Previously, he served as an Attorney Advisor with the Office of Law and Policy in the National Security Division of the U.S. Department of Justice and a Special Assistant United States Attorney in the U.S. Attorney's Office for the District of Maryland.
Benjamin Wittes is editor in chief of Lawfare and a Senior Fellow in Governance Studies at the Brookings Institution. He is the author of several books.

Subscribe to Lawfare