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

Removal in the Georgia Prosecution: A Low Bar but Weak Arguments

Lee Kovarsky
Friday, August 25, 2023, 9:47 AM
Trump and Meadows have a shot at removing their case to federal court—not because their arguments are good, but because the bar is low.
Mark Meadows in 2018. (Gage Skidmore,; CC BY-SA 2.0,

Published by The Lawfare Institute
in Cooperation With

On Aug. 14, District Attorney Fani Willis indicted Donald Trump in Fulton County, Georgia. The indictment against Trump and 18 others alleges 41 counts of racketeering and unlawful interference with Georgia’s tabulation of votes in the 2020 presidential election, as well as with its certification of presidential electors. Trump’s former chief of staff, Mark Meadows, is one of the 19 defendants, and he has a very strong legal team. Meadows has already attempted to “remove” his prosecution to state court and (late Friday night) also moved to dismiss the whole prosecution because of Supremacy Clause immunity. Two more defendants have filed removal notices since then, and the rest will surely follow suit.

 Observers confidently estimate Trump’s legal jeopardy at their own peril, but one can already make some educated guesses about how the removal attempts might play out. Meadows has a nontrivial chance of succeeding because of the overlap between his charged behavior and the unique responsibilities of his former office. Trump’s chances of removal are something less than Meadows’s for reasons that I’ll explain below. But removal is probably less of a defendant windfall than people think.

What Is “Removal?”

In the parlance of federal procedure, a defendant is sometimes permitted to “remove” a case to federal court even though the plaintiff or prosecution initiated the case in state court. Ordinarily, removal happens in the context of civil litigation, but it can occasionally take place during criminal prosecutions. The provision for “federal officer removal” is 28 U.S.C. § 1442(a)

A civil action or criminal prosecution that is commenced in a State court and that is against  … any of the following may be removed[;]: (1) … [A]ny officer (or any person acting under that officer) of the United States … , in an official or individual capacity, for or relating to any act under color of such office[.]

There are some special rules that apply when the matter to be removed is a criminal prosecution—I’ll reference them as needed. 

The statute reflects the idea that federal officers shouldn’t find themselves in state court for acts undertaken as part of their office. So, if a state prosecutes a federal official “for or relating to” behavior under color of their federal office, then the official can remove when they also show that they have a colorable federal defense. Meadows is seeking to remove under this provision, asserting Supremacy Clause immunity—which protects federal officers from lawsuits arising out of conduct that they subjectively and reasonably believed to be necessary and proper to their official duties. Trump and the other co-defendants will almost certainly do the same. Removal will ultimately be resolved when, after the defendants file removal notices, the court decides whether to “remand” some or all of the prosecutions to state court. (The remand decision will happen, at least for Meadows, sometime after an Aug. 28 hearing.) 

Before I discuss why the defendants are trying to remove and whether they’re likely to succeed, a couple of misconceptions about removal are worth dispelling now. First, removal changes the court system in which a case is tried, but it does not change the substantive law used to decide who wins. Whether the trial is in a Georgia court or in a federal one, the prosecution will seek to prove—and the jury will decide liability for—offenses under Georgia law. For example, the broader Georgia Racketeer Influenced and Corrupt Organizations (RICO) statute applies, not its narrower federal counterpart. And because any conviction will be for a Georgia crime, the president does not have power to pardon, commute, or reprieve the sentence. Therefore, in the case of a second presidential victory, Trump will not be able to pardon himself.

Why Do Trump and His Co-defendants Want to Remove?

There are (at least) three reasons why Trump and the other defendants might want to move the litigation to federal court. The first, and the most frequently discussed, is the jury pool. Were Trump to be tried before Fulton County jurors, the jury would be pulled from a county that voted 73-26 percent for President Joe Biden. Fulton County is also majority non-white and 45 percent Black. The demography of the jury pool is particularly salient because race would be front and center at any trial; Trump’s election fraud claims traffic heavily in ugly ideas about the illegitimacy of Black votes and the unreliability of Black election officials. (Think Ruby Freeman, the Black poll worker baselessly accused of committing election fraud, often in incendiary terms.) 

At the same time, the difference between Fulton County and federal jury pools is less dramatic than some have suggested. There’s a broad misconception that the federal jury pool would be drawn from all of Georgia’s northern federal district, but that’s almost certainly untrue. The Northern District subdivides into federal “divisions,” and it’s almost certain that all jurors would come from the Atlanta division. That division consists of 10 counties that form or neighbor the city: Cherokee, Clayton, Cobb, DeKalb, Douglas, Fulton, Gwinnett, Henry, Newton, and Rockdale. With the exception of Cherokee, which is the third smallest of the group and about 11 percent of Fulton’s size, all of these counties voted against Trump by significant margins. DeKalb, which is the second most populous county in the group, voted 83-17 percent in President Biden’s favor. Nor are these monolithically white exurbs. DeKalb county is 64 percent non-white and 55 percent Black. Gwinnett, the third largest county in the division, voted against Trump by “only” a 58-42 percent margin, and it is majority white—at a tick less than 51 percent.    

The upshot is that the media might be somewhat overstating the jury-pool benefit that comes with federal removal. The defendants hope (among other things) to seat a single juror willing to “nullify,” by which I mean a juror willing to acquit the former president and his co-defendants notwithstanding facts or law. The odds of seating a nullifier may increase slightly with a federal jury pool, but—for the reasons discussed above—not by much. There may also be some slight defendant advantage to a whiter federal pool, which could be important given the racially inflammatory nature of Trump’s election fraud claims.           

The second reason Trump and his co-defendants prefer a federal forum is the possibility of delay that goes with a switch from state to federal court. As most experienced lawyers know, any change of venue entails procedural skirmishes that precede the difficult work of conducting the trial and fixing guilt. These procedural fights would be especially prominent in a criminal prosecution—when a federal court must decide, on virtually every procedural issue, whether to apply federal or Georgia law. Every law student will recognize these issues as a dreaded Erie (v. Tompkins) problem, which requires federal courts to choose between state and federal procedure as they adjudicate the substance of state law.

There are some other procedural fights to expect, too. For example, there will probably be requests to “transfer venue.” Whereas removal moves a case from a state court to a federal court in that same state, a transfer moves a case from one federal court to another—and may move it from a federal court in State A to a federal court in State B. There may also be a dispute over whether the removal of one defendant’s case requires that all defendants “ride along” and receive a federal trial. In that scenario, only one defendant out of the 19 would need to meet the criteria for officer removal. (Meadows has already telegraphed that argument in footnote 4 of his motion on Supremacy Clause immunity.)

Nevertheless, and as was the case with the jury pool, there are also reasons to believe that the opportunity for procedural delay is capped. Unlike removal in civil cases, the statute applicable here provides that “a notice of removal … shall not prevent the State court in which [a] prosecution is pending from proceeding further[.]” Moreover, the possibilities for procedural delay shrink if a federal judge wants to run a tight ship—and the defendants got a bad draw in federal court. Meadows’s removal notice was assigned (randomly) to U.S. District Court Judge Steve Jones, an Obama appointee who spent 18 years as a state-court judge and who was confirmed by a unanimous Senate vote. Internal operating procedures probably mean that Jones will get all filings associated with the Willis prosecution. Jones moved very quickly in response to Meadows’s removal notice, scheduling a hearing on the matter for Monday, Aug. 28. He seems exceedingly unlikely to dither. When Meadows and Jeffrey Clark filed long-shot emergency motions to stop DA Willis from arresting them, Jones denied both motions extremely quickly, in separate six-page orders. His approach to pretrial management seems to be closer to that of D.C. federal Judge Tanya Chutkan, who presides over the Jan. 6 prosecution, than that of Aileen Cannon, the federal judge in Florida who presides over the federal documents case.           

The third reason that Trump and his co-defendants might prefer a federal forum is probably the most obvious: Up the federal pyramid are a circuit court and a Supreme Court with many judges likely to be receptive to Trump’s arguments. If the prosecution remains in state court, then the opportunities to involve those justices might seem smaller. To reach the Supreme Court by way of a state-court prosecution, any conviction would have to involve a federal issue decided in a way that triggers Supreme Court jurisdiction. Federal issues creating the possibility of U.S. Supreme Court review are just a lot more likely to appear in the course of a federal trial.

Still, even this third perceived advantage is limited—Georgia appellate courts might be more Trump protective than meets the eye, and federal appellate courts less so. The Georgia Supreme Court is no hotbed of Democratic activism; it consists of eight judges appointed by Republican governors and one who prevailed in a nonpartisan election. And the U.S. Court of Appeals for the 11th Circuit, which has appellate jurisdiction over the federal district court in Atlanta, decisively shut down Trump’s special-master gambit in the Mar-a-Lago investigation. For its part, the U.S. Supreme Court refused to block a documents subpoena in New York’s criminal investigation and a grand jury subpoena for Sen. Lindsey Graham (R-S.C.) in the very Georgia investigation at issue.

Will Trump and His Co-defendants Succeed?

Meadows has a small-but-nontrivial shot at removing, and Trump’s chances are less than Meadows’s—but still not zero. Removal requires roughly that (a) the remover be an “officer (or any person acting under that officer) of the United States,” (b) the conduct at issue be or “relate to” acts taken “under color of such office,” and (c) there be some colorable federal defense. The primary federal defense at issue at prong c is called Supremacy Clause immunity, which bars prosecution of federal officials for acts authorized by federal law that are “necessary and proper” to their federal duties. (Yes, the underlying criteria for Supremacy Clause immunity look a lot like prong b of the removal test.)     

Trump and Meadows won’t have problems on what I am calling prong a because they clearly are—and in Meadows’s case, also work for—federal officers. Trump was the president of the United States and, as Trump’s chief of staff, Meadows was the head of the Executive Office of the President. The tough questions start on prong b, where Meadows has a stronger argument than Trump that his conduct relates to acts taken under color of his office. Meadows’s job description was, speaking colloquially, to advise and assist the president as the president pursues public interests. In the removal notice, Meadows has argued that he did no more than that: “arrange[] meetings for the President,” “communicate[] with state officials,” and visit the Georgia election audit site in order to secure the country’s election integrity interest (not Trump’s). His dismissal motion repeats, in sum and substance, that characterization of the indicted conduct.      

The overlap between the charged conduct and the formal requirements of Meadows’s office is much greater than the comparable area of overlap for Trump. DA Willis responds, persuasively, that Meadows could not have been acting under color of his office because the indicted conduct was political activity that violated the Hatch Act. But whereas Meadows can at least say with a straight(ish) face that the indictment alleges no more than what his job requires, Trump can’t. The indictment alleges far more substantial Trump conduct that is also much more culpable—conduct that Trump cannot explain away by reference to his office. Federal law assigns the sitting president no role in the counting and tabulation of presidential votes, and it does so for conflict-of-interest reasons that require little discussion. And even if there were some such role, the idea that the indicted conduct represents something other than Trump’s pursuit of his private interest doesn’t pass a laugh test.

Because of the unique federal defense involved, the analysis of removal prong c resembles that of removal prong b. And for many of the same reasons, Meadows has a better argument than Trump. Meadows has moved to dismiss on the grounds of Supremacy Clause immunity, which requires him to show that the charged conduct was authorized by federal law and that he did no more than what was necessary and proper in the discharge of his federal duties. (He also moved to dismiss on First and 14th Amendment grounds that won’t dictate the outcome of the removal fight.) Although Meadows’s prong c arguments will be better than Trump’s, I break with some other commentators in that I still consider them weak. In his motion to dismiss, Meadows harps on the fact that he has not been charged with federal crimes, but that’s a nonsequitur. That there was no federal charge does not mean that federal law affirmatively authorized the conduct, and it certainly doesn’t mean that the conduct was necessary and proper to official duties.         

Nor do I think that, simply because Meadows (or Trump) asserts the existence of a fact issue as to their intent, there is a colorable federal defense requiring removal. For starters, even if intent went to the colorability of the federal defense on prong c, it doesn’t resolve whether the conduct was sufficiently related to the federal office at prong b. And there are real problems even with the federal defenses on prong c, especially for Trump. I doubt seriously that Trump can make a colorable claim of presidential immunity. The leading case recognizing the immunity is Nixon v. Fitzgerald, which is expressly limited to civil damages. And in that case, Chief Justice Warren Burger concurred expressly to underscore the importance of that limitation. Moreover, Trump is unlikely to meet the presidential immunity standard articulated in the civil cases because the indicted conduct doesn’t fall within the “outer perimeter” of presidential authority. Finally, for both Meadows and Trump, Supremacy Clause immunity requires a showing that the conduct in question was “necessary and proper” to the office—but the necessary-and-proper showing requires that the defendants demonstrate both a subjective belief that the conduct was necessary and proper to their office and that the belief was reasonable. Absent such reasonable belief, the defense is not colorable and there can be no removal. If I had to bet on the outcome of the removal skirmish, then I would bet on the defendants losing on some combination of the Hatch Act and this reasonable-belief requirement.

Notwithstanding my skepticism about Meadows’s arguments on both prongs b and c, and even though I think Trump’s arguments are weaker than Meadows’s, the defendants have a nontrivial shot at removal, for two major reasons. First, in order to remove, the defendants won’t need to prove that their conduct related to their official duties or that they will prevail on federal defenses—they need only make showings that are “plausible” or “colorable.” (Judge Jones twice mentioned the colorability requirement in his order calendaring the Aug. 28 removal hearing.) With respect to the colorability of the federal defense on prong c, Jefferson County v. Acker emphasized that a defendant need not “win his case before he can have it removed.” Acker also makes clear that a colorable defense is something less than a “clearly sustainable” one. Finally, Acker in some ways aligns the showings necessary to meet prongs b and c, remarking that “[j]ust as requiring a clearly sustainable defense rather than a colorable defense [on prong c] would defeat the purpose of the removal statute, so would demanding an airtight case on the merits in order to show the required causal connection [on prong b].” Acker’s language is particularly important because Acker was cited heavily in Georgia v. Heinze, a prominent and potentially precedential Northern District of Georgia case from 2022 on the question of federal officer removal.

Second, some of the influential lower-court opinions on prongs b and c, which I cannot canvass exhaustively here, are quite favorable to the defendants.  On prong b, the 11th Circuit has adopted a very defendant-favorable interpretation of a 2011 statutory amendment, requiring—in contrast to some other appellate courts—only a “connection” or “association” between the acts in question and the federal office. The 11th Circuit emphasized that this standard is “quite low.” And with respect to prong c, Heinze includes the suggestion that “all that is required” is that the defendants make “alleg[ations] that they were acting as federal officers in accordance with federal law and therefore entitled to immunity.”

In sum, the officer-removal standard likely to control here is just quite low. And there are other defendant-favorable arguments in play, too. For example, Trump might also be able to argue that a successful Meadows removal requires that the whole case be removed to federal court without respect to the individual removability of other defendants. Finally, there is some abstract, intuitive force behind the idea that a prosecution for overturning a federal election, based on the conduct of the then-sitting president, should be in federal court.

* * *

In the end, there might be less at stake to removal than many believe. The Fulton County and Atlanta Division jury pools aren’t that different, plus Trump and his co-defendants drew a federal judge who is likely to press the case forward. Still, the defendants’ removal arguments are not especially good, relying in at least some measure on a court’s willingness to indulge the fiction that Trump and his co-defendants were engaged in a coordinated election interference to protect the public interests of the United States—rather than to promote the private interests of Donald Trump. If the case does find its way into federal court, it won’t be because the defendants’ arguments on their underlying defenses are strong; it will be because courts are exercising the utmost caution.

Lee Kovarsky is the Bryant Smith Chair in Law at the University of Texas, where he also co-directs the school's Capital Punishment Center. He has litigated Texas death penalty cases for almost twenty years.

Subscribe to Lawfare