The Constitutionality of Trying a Former President for Impeachment–A Reply to Frank Bowman

By Philip Bobbitt
Friday, February 12, 2021, 2:53 PM

Frank Bowman recently published an essay in Lawfare that criticized arguments I made in an essay on the site. Bowman, whose impeachment book, “High Crimes and Misdemeanors: A History of Impeachment for the Age of Trump,” is one of the standard recent works on the subject, is a thoughtful, serious and knowledgeable scholar. Yet I believe his criticism mistakes the principal arguments I advanced in my piece. In such a situation, the presumption has to be that I made my arguments in a confusing or misleading way, and so I am happy that Lawfare has given me the opportunity to set the record straight.

Bowman has organized his critique along the lines of the modalities of constitutional argument, and I will do the same. In each case, I will state my position briefly; repeat the critique; and where I can, rebut it.

Constitutional Text

Very briefly, the center of my textual argument lies in this language from Article II, Section 4:

The President, Vice President and all civil Officers of the United States, shall be removed from Office on Impeachment for, and Conviction of, Treason, Bribery, or other high Crimes and Misdemeanors.

My point is quite straightforward: This language is the only text in the Constitution that sets forth who may be impeached and convicted, and on what grounds. It and only it inscribes the limits of those persons subject to the various impeachment clauses.

There is no doubt in my mind that the language “The President, Vice President and all civil Officers” does not include former officials. They no more remain “civil officers” than I remain a Princeton undergraduate. Perhaps because I state this conclusion so definitely, Bowman charges me with claiming infallibility in my assessment of constitutional meaning. But the meaning of the constitutional text, as a matter of textual argument, isn’t up to me. It’s up to the ordinary English speaker. That is the nature of this modality of argument in U.S. constitutional construction. If anyone reading this text thinks, “[t]he President, Vice President and all civil officers of the United States” means “all persons who once were or who now are the President, Vice President and all other current or former civil officers,” then there is not much I can say to resolve our disagreement.

I put to one side the patently frivolous argument that the category of officials who can be tried in the Senate is broader than the category of “civil officers” because language in Article I, Section 3, Clause 6 refers to “persons” and not “civil officers.” (“[N]o Person[s] shall be convicted without the Concurrence of two thirds of the Members present.”) Surely, no one is contending that all persons, whether citizens or not, whether natural persons or non-natural persons like corporations, whether minors or most pertinently whether they hold or have ever held or are even eligible to hold office, are subject to impeachment and conviction for the offenses that might lead to the removal of a civil officer.

Now how does Bowman reconstruct my argument?

First, he says the text “does not directly address the question whether a president, vice president or other civil officer can be impeached by the House while in office, but tried, convicted and sanctioned by the Senate after leaving office.” I confess I simply don’t see this. Whatever may be the case with Trump’s impeachment by the House when he was a sitting president, how can he be convicted if he is no longer a “civil officer”? Who is to be convicted according to the text of Article II, Section 4, if not the “President, Vice President and all civil Officers of the United States”? How much more directly could the text specify who is subject to conviction?

Second, he takes my argument to be that because the text of Article II requires the removal of the convicted official, it cannot apply to a person who cannot be removed. It’s not possible to impeach a dead person, a person who had resigned or a person previously convicted and removed, a person who had never been elected or appointed in the first place, and so on. This was Justice Joseph Story’s position in “The Commentaries”:

If then there must be a judgment of removal from office, it would seem to follow that the Constitution contemplated that the party was still in office at the time of the impeachment. If he was not, his offense was still liable to be tried and punished in the ordinary tribunals of justice. And although a judgment of disqualification might still be pronounced, the language of the constitution may create some doubt, whether it can be pronounced without being coupled with a removal from office.

These are perfectly acceptable arguments, but they are not my principal arguments. If the word “removed” were to be replaced with “punished,” my point would be exactly the same. My point is not a matter solely of removal. The Congress has no power to impeach and convict any person not a “civil officer.” Full stop.

That is not to say that removal is irrelevant to the issue. There is a rather fey textual argument I have heard that runs like this: Of course you can disqualify someone who is not a civil officer of the United States. After all, once an officer has been impeached, convicted and removed, everyone agrees he can be disqualified from future office even though he is now not a “civil officer.” And if he’s not a civil officer in the first place, well then you don’t need to bother removing him. Therefore, removal is not in fact a predicate for disqualification. This Escher-like reasoning affirms the importance of removal as a predicate and captures a bit of the lunacy loose in the debate at present.

Bowman also taxes me with the “subtle” stratagem of putting my discussion of the text of Article II first in my essay, before the texts of Article I, Section 2 and Article I, Section 3. The former provides, “The House of Representatives shall chuse their Speaker and other Officers; and shall have the sole Power of Impeachment,” while the latter provides, “The Senate shall have the sole Power to try all Impeachments.” Article I, Section 3 also provides that

Judgment in Cases of Impeachment shall not extend further than to removal from Office, and disqualification to hold and enjoy any Office of honor, Trust or Profit under the United States: but the Party convicted shall nevertheless be liable and subject to Indictment, Trial, Judgment and Punishment, according to Law

Since they come first, they should be read first, he suggests. In fact, the placement of the grounds for impeachment and the persons to whom it applies in Article II is important, but not because it is subordinated by ordinal numbering to Article I. The provisions for the basis for impeachment are in Article II because that is the article that governs the executive and the vast majority of serving civil officers are members of the executive branch; the senators and representatives of Article I are not civil officers. But let us play out Bowman’s suggestion.

Bowman argues that reading the provisions in the order he suggests leads to a substantively different interpretation. “Read as part of this more natural sequence,” he writes, “the removal language of Article II places no necessary restriction on the class of officials who may be impeached, convicted and penalized” (emphasis added ). I will reiterate here that my argument is not about removal, but there is also a more important consideration: If the “class of persons subject to impeachment” requirement of Article II can simply be erased because a general power to impeach, try and convict is given the House and Senate in Article I, what would become of the “grounds for impeachment” requirement of Article II? Is it also superfluous? Does it also place “no necessary restriction” on the grounds for impeachment? Would “high Crimes and Misdemeanors” mean, as is sometimes claimed, whatever the House thinks it is at any given time, the House being given “the sole Power of impeachment” in Article I?

Why would so scrupulous a scholar as Frank Bowman propose a step that might open the door to such possibilities?

Let us track his rationale as follows: The removal penalty in Article II, though mandatory, is merely appurtenant to the powers of impeachment, trial and conviction provided the houses of Congress in Article I. Additionally, disqualification from future service is a discretionary penalty. If the House has impeached an officer for committing high crimes and misdemeanors while in office, and the officer has since left office before the Senate has determined “the proper penalty,” thus making removal impossible, disqualification is still available. Thus, for an officer who has vacated a post, removal may be moot, but disqualification remains available.

But the problem raised by the inability to remove a private person from an office no longer held wasn’t a problem of mootness; that’s the least of it. The problem was that the inability to remove a convicted person was irrefutable evidence that the person couldn’t be convicted in the first place. As Article II clearly requires, removal occurs “on Impeachment for, and Conviction of, Treason, Bribery or other high Crimes and Misdemeanors.” It is not a discretionary penalty.

Moreover, the Article I language on which this argument relies is not so much a grant of authority as it is a limitation. Article I provides, “Judgment in Cases of Impeachment shall not extend further than to removal from Office and disqualification to hold and enjoy any Office of honor, Trust or Profit under the United States” (emphasis added). As such, Article I relies on and supplements authority granted in Article II: “The President, Vice President and all civil Officers of the United States shall be removed from Office on Impeachment for, and Conviction of, Treason, Bribery and other High Crimes and Misdemeanors.”

Bowman elevates the disqualification language, providing him with the basis for the claim that to vindicate this language we are compelled to imply any necessary additional authority to bring it into action. In other words, we must not allow a president to avoid the penalty of disqualification by resigning.

To my great surprise Bowman asserts that “a penalty is neither an element of an offense nor a ‘ground’ for asserting jurisdiction; it is a consequence of conviction after trial by the body possessing jurisdiction.”

How true. The existence of a possible penalty cannot be used to bootstrap jurisdiction on the grounds that absent a finding of jurisdiction, the penalty is not going to be assessed. But immediately after noting this, Bowman writes, “Bobbitt argues that a person whose circumstances are such that he cannot be made to suffer a mandatory penalty therefore escapes jurisdiction altogether.”

Frankly, I don’t know how the penalty of disqualification is mandatory. Of the 20 persons impeached by the House, eight have been convicted and only three have been disqualified from future office. But Bowman offers to help by giving two hypotheticals:

Consider a con man convicted under a fraud statute which provided, upon conviction, for mandatory forfeiture of all the proceeds of the crime plus a term of imprisonment imposed at the discretion of the judge. Would anyone think it sensible to conclude that the con man would be immune from prosecution and the possibility of prison if, before his trial, he blew all the loot in Vegas

This assumes the very position that Bowman has warned us about: It infers jurisdiction from the penalty. Suppose the fraud statute requires the use of the U.S. mails. Would the con man be subject to prosecution if he stole his gains—regardless of what he did with them—but it could not be proved he used the instrumentalities of interstate commerce?

The second hypothetical is also helpful. Bowman proposes:

Or imagine a law passed in a pastoral society that said: “A goatherd who steals a goat from the flock he is tending shall, upon conviction, be removed from the owner’s employment. Judgment in cases of goat theft shall extend no further than removal from current employment, and disqualification from ever again working in the livestock industry.” Suppose that, after apprehension, but before trial, the goat thief was fired from his job. Would it make sense to argue that the thief therefore could no longer be tried or, if convicted, banned from the herding trade for the protection of a society for whom honest stewardship of livestock is central to survival?

The problem with this analogy is that goatherds, like cowboys and masons, remain goatherds even when they’re between jobs. Being a “goatherd” doesn’t entail holding a particular status under the law, it just entails making that one’s profession. If Article II had said “politicians” rather than “civil officers,” there would be some force to the analogy. A more apt hypothetical statute would read, “A licensed goatherd who steals ….” Now why would someone not see this evident distinction? I think because it’s not text at all, but prudence that is driving the argument: “Would it make sense,” Bowman asks, to let the thieving goatherd go free to continue his ways “in a society for whom honest stewardship of livestock is central to survival?” And I guess the answer is, from a prudential point of view, that it depends on how much value the society places on law. Down that path lies extrajudicial killing, and I suspect that over the millennia, many a horse thief and cattle rustler has been hanged on the basis of just such a rationale. Perhaps we’re beyond those days, but lying behind Bowman’s hypothetical is the argument that the plain restraints of the law that govern the processes of legal punishments can be disregarded when they are thought to be inconvenient.

Bowman has a second textual argument that I will call, in friendly chaffing, the “McConnell move,” partly because Bowman cites the distinguished judge and legal scholar Michael McConnell. This argument goes like this: Article I, Section 3, Clause 6 gives the Senate the power to try “all” impeachments. President Trump was in fact a “civil officer” when he was impeached by the House while still in office. Ergo, the Senate can try his impeachment. They are not required to try only civil officers because they are empowered to try all impeachments; they have the constitutional authority to try anyone the House has validly impeached.

The problem with using Article I’s grant of power to the Senate to erase the grounds for and limits on this power granted by Article II is that such a move ignores the basis for conviction provided in Article II. Only Article II states the grounds for, and the persons subject to, impeachment and conviction. Nothing in Article I supplements or purports to supplement those grounds and those persons.

A better formulation of the McConnell move would be this: The former official could be impeached while in office (Article I provides that the House “shall have the sole Power of Impeachment”); perhaps, since the original impeachment was valid under the terms of Article II, a former official can even be tried (Article II provides that the Senate “shall have the sole Power to try all Impeachments”), but there is no authority to convict anyone who is not a civil officer as provided by Article II. This is not absurd: The events of this week—in which the House impeachment managers made dramatic and devastating presentations before the Senate, and before the American people—will make a historic contribution to the understanding of just what happened on Jan. 6 and how it came about.

This is the true precedent—if there is any—set by the Belknap case that is so often mis-cited as providing “solid” authority for the impeachment of a former civil officer. In fact, the former secretary of war was acquitted by the Senate on the grounds that he was no longer subject to conviction owing to his resignation. It may be that the exposure of the underlying offense can justify such a maneuver. By contrast, there are three examples of the Senate declining to try impeachments following resignations: judges Delahay, English and Kent in 1873, 1929 and 2009, respectively, in much more quotidian and less historic contexts.

This is a matter of judgment. Even though the text of Article I, Section 3 does indeed grant the Senate the sole power to try all impeachments, this is not the same as saying it grants the Senate the ability to try all impeachments. It’s not hard to think of many hypotheticals in which some event occurs—the death of the impeached party—that would deprive the Senate of the ability to try the case. The first impeachment, in 1797-1799, was dismissed for want of jurisdiction (either because the defendant was no longer a civil officer, having been expelled from the Senate, or because he had never been a civil officer since senators and members of Congress are not “officers of the United States”) and thus obviously was not “tried.” Trump’s impeachment resolution, which provides the basis on which the Senate is to conduct a trial, names the defendant as “President Donald J. Trump,” but there is in fact no longer such a person.


Next, Bowman takes on a structural argument I make to advance my position:

After treating the text, Bobbitt moves on to what he calls a “structural” claim. … The more fundamental problem with Bobbitt’s “structural” argument is that it isn’t really a structural argument at all. A structural argument seeks the meaning of a contested passage of the Constitution by considering its place and function within the overall structure of the Constitution, with particular attention to other passages that address the same objective.

I don’t hold a copyright on the term “structural argument,” but my own view—and that of Charles L. Black Jr., whose celebrated lectures, “Structure and Relationship in Constitutional Law,” give the modern basis for explicating this modality of argument—is a bit different from Bowman’s. I would say that “structural arguments are inferences from the existence of constitutional structures and the relationships which the Constitution ordains among these structures.” Similarly, “To say that the institutional relationships promulgated by the Constitution require or are incompatible with or tolerate a particular answer … is to use a structural mode of arguments.” Among these structures are the three branches of the federal government, federalism, a national economic union, and the relation of voters to their representatives in a democratic republic. This is emphatically not the same as deducing some general mission of the Constitution, calling that a “structure” and seeing whether a particular rationale supports that mission.

Well, so what? What difference does it make what labels one uses? I think there are two unfortunate consequences for Bowman’s essay. First, he ignores “what [Bobbitt] calls a structural claim.” To put it very simply, it is that the structures of elected government, and the relationship of the electorate to those structures, are jeopardized by adding to the power of elected officials the ability to disqualify potential competitors. Perhaps because he thinks this is the wrong label for my arguments, he helpfully recasts them in prudential terms and attempts to refute them on that basis.

Second, with a similar insouciance toward the requirements of the various modal forms, he substitutes a mélange of prudential, textual and historical arguments without the sort of supporting arguments that these modalities require. But I think it can be shown that, whatever label one affixes to these arguments, there are good reasons—good prudential and originalist reasons—to reject them.

My argument is brief: I say,

Next consider structure. The interpretation that persons are subject to impeachment and conviction even if they are not civil officers would greatly expand the Senate’s ability to prevent future office-holding. … [U]nder the “alternative penalty” rationale [which does not require removal of the convicted person because disqualification is only an alternative to removal] any person who was once a civil officer might be impeached and convicted and by this means disqualified from any future office. Is it really compatible with the system of democratic representation to provide for an “alternative” penalty by so greatly expanding the number of persons who can be banned by a majority of senators from holding elective office—including by running against them?

This is an argument about the structure of our democracy and the relationship of the electorate to that structure. To which Bowman replies,

In the 230 years since the impeachment clauses were written, there has never been a case remotely like either of these imagined horrors. Given the universally recognized difficulty of bringing any impeachment to a successful conclusion, the idea that Congress would run amok as postulated is fantasy

Maybe. I am not arguing that this will happen—although I am far less sanguine than Bowman—I am simply asking whether permitting the members of Congress such power to determine the composition of their electoral opponents is compatible with the system the country has adopted.

Bowman’s version of structural argument is this:

To say that one is interpreting a constitutional passage “structurally” is to say that, if presented with two textually plausible readings, one should prefer the reading that best promotes the evident aims of the constitutional structure taken as a whole. Any proper “structural” reading of the impeachment clauses should give maximum effect to their overall purpose of protecting constitutional order against dangerous presidents, while avoiding artificially narrow constructions that defeat that end.

Thus Bowman concludes that the Framers’ concerns (“their overall purpose”) should drive our analysis:

As I have described elsewhere, by virtue of their classical educations and their pre-Convention study of prior republics, the Framers were especially leery of a particular kind of would-be autocrat—the “demagogue”—who might gain power and overthrow republican government by appeals to the passions of the masses. The danger posed by a person of that kind, even if once expelled from office, persists so long as he retains the potential to overwhelm rational opposition by inflaming popular sentiment and so too returning to power. Thus, elections, removal and even prosecution do not wholly address the problem of the aspiring populist autocrat. The Framers filled the resultant gap in their constitutional structure by also including the discretionary remedy of permanent disqualification from national office

I’d call this a historical or an originalist argument. The problem with it is that there is not one shred of historical evidence that the Framers and ratifiers in fact intended the Senate’s power to disqualify persons from holding federal office to apply to those persons who do not in fact hold federal office. To say that that conclusion is “illogical” or that the Framers can’t have meant that is simply to assume that the objective one seeks—the disqualification of Donald Trump—validates one’s reasoning.

What we do know is that several state constitutions provided for impeachment trials that would take place after the relevant government official was no longer in office. We know that the Framers rejected this language and also the parliamentary practice in Britain of trying private persons for treason and other crimes by means of impeachments. We know that the Constitution was criticized for insulating presidents from impeachment while the states provided for more expansive liability. It was in answer to this criticism that James Madison wrote in Federalist #39 that “in Delaware and Virginia [the chief magistrate] is not impeachable till out of office. The President of the United States is impeachable at any time during his continuance in office” (emphasis added).

The Framers prohibited bills of attainder as part of their efforts to reject the British law of impeachment. Only the extraordinary measure of impeachment can permit legislative action that is generally conceded would otherwise be deemed a bill of attainder. Moreover, we know that the Framers regarded the impeachment of private citizens in Britain as a bill of attainder. So what are we to conclude if we wish to give effect to their intentions? That the legislative punishment of private citizens is permissible so long as the victim was formerly a civil officer, a position rejected by the Framers when it was suggested they follow the example of the 1776 Virginia convention?

Finally, there’s the decided verdict of Joseph Story—professor of law at Harvard, associate justice of the Supreme Court, confidante and protégé of John Marshall and author of the most respected of the first generation of constitutional treatises—on the precise point of whether a party must be in office to be impeached, as quoted in my original essay and above.

Call it what you will, but it simply won’t do to broadly state the ambition of the Framers to protect the country from demagogues and then conclude from that that the clearest of textual barriers, and a good deal of other historical evidence, can be disregarded.


I, like Bowman, also explore the question of precedent for the current moment.

As everyone who is following this debate will know, a Tennessee senator, William Blount, was impeached by the House in the summer of 1797. Before a trial could be held in the Senate, it voted to expel Blount. Subsequently that winter, his trial began and in a narrow vote, the Senate defeated a resolution that asserted that Blount was an impeachable officer. The Senate failed to make clear whether this was because no senator was a civil officer or because, owing to his resignation, Blount was no longer a civil officer. Whichever one concludes, the Blount precedent provides no support for the assertion that a non-civil officer can be impeached.

More often cited is the impeachment of Grant’s secretary of war, William Belknap, in 1876, as noted above. Enmeshed in a scandal, Belknap gave the president his resignation just minutes before the House was scheduled to vote on his impeachment. Nevertheless, the House went ahead and sent to the Senate five articles of impeachment. The House managers argued that Belknap should not be allowed to escape justice by resigning, and the Senate initially agreed with this assessment. The 22 senators who opposed this resolution, however, were sufficient to thwart the two-thirds majority required for conviction and Belknap was acquitted on the grounds, one may infer, that an impeached official who has resigned may not be convicted. This case is frequently cited as affirming the proposition that a resigned civil officer may be impeached because the Senate initially voted to take jurisdiction. That it does not support this reliance may be concluded not only from the vote for acquittal—which we know was done by a sufficient number of senators who thought Belknap, having resigned, could not be convicted—but also from this week’s events.

The Senate voted this week, by a majority vote, to reject the constitutional claim that Donald Trump was no longer subject to trial and conviction. If in the ensuing weeks, 34 or more senators vote to acquit on the grounds that Trump is indeed not subject to conviction on constitutional rather than evidentiary grounds, can it be claimed that the initial vote determines the precedent for future impeachments when that decision has been repudiated?

Of these two precedents, Bowman concludes:

Although the Blount and Belknap cases are not definitive proof of senatorial impeachment jurisdiction over former officials, they are better precedent for such jurisdiction than against it.

Perhaps for that reason, Bobbitt proffers the puzzling contention that the fact that Congress chose not to impeach Richard Nixon after his resignation is evidence for the proposition that it could not have had it wished.

I introduced the Nixon example not because, as Bowman puts it, I contend that “Congress chose not to impeach Richard Nixon after his resignation[.]” On the contrary, there is no evidence that the leadership of Congress even considered this a constitutional option.

Rather, I think the Nixon case is instructive, and far more relevant as a precedent, because it is more recent than Blount and Belknap. The case of a president, elected by the nation, is more relevant to the case of Donald Trump than that of a senator or secretary of war; there is ample evidence in contemporary records on which we can draw. These are among the relevant standards for the application of doctrinal argument—precedent.

Those records emphatically do not demonstrate that “Congress chose not to impeach Richard Nixon” (emphasis added). What they disclose is that, as I wrote, “no one—not Rep. Peter Rodino, the chair of the House Judiciary Committee; not Sen. Sam Ervin, who chaired the Senate Watergate Committee; and not even the lawyers for the various congressional committees—ever seems to have suggested that the Congress should go ahead with impeachment despite the fact that Rodino met with the House leadership before the resignation and discussed the disqualification penalty.” Remember that even after Nixon was pardoned, he was still amenable to impeachment on Bowman’s view. Ford’s pardon put Nixon beyond any federal judicial penalties, so impeachment was likely even more compelling an option if anyone had thought it available and would have been the only way to achieve disqualification.

Is this historical incident a governing precedent? No, certainly not. But it is instructive because it indicates that people in Congress—and in the media and the law journals—who thought about a similar problem didn’t even regard the course of impeaching a former president as a constitutional possibility. That may not have been for legal reasons; legislative precedent is hard to evaluate, not because scholars don’t know the real reasons for a constitutional decision by legislators, as is sometimes asserted, but, rather, because no rationale is usually proffered and adopted by the deciding body.


Bowman and I also differ on certain prudential questions. What I believe is driving the dubious claim that, as a New Yorker columnist put it, “[i]n fact, there is strong support for post-Presidential proceedings in constitutional history and in precedent,” is the concern that without the availability of impeachments against former presidents, there “would be no real penalty for a late-in-the-term coup attempt.” Bowman seems in sympathy with this argument and characterizes my piece as arguing that “the Constitution prohibits the completion of this exercise in republican self-protection because the president sought to overthrow democracy late in his term and the Senate didn’t start his trial fast enough to allow both removal and disqualification.” A similar argument is that a president might commit an egregious impeachable offense—like a sustained campaign to de-legitimate the electoral process, culminating in a violent attempt to thwart the certification of election results—and then resign in order to escape punishment. In the absence of impeachment, what would prevent such a person from running for office again?

The short answer is supplied by 18 U.S.C. § 2383 and 18 U.S.C. § 2384. The former provides that “[w]hoever incites, sets on foot, assists, … insurrection against the authority of the United States or the laws thereof, or gives aid or comfort thereto, shall be fined under this title or imprisoned not more than ten years, or both; and shall be incapable of holding any office under the United States,” while the latter criminalizes “two or more persons … prevent[ing], hinder[ing], or delay[ing] the execution of any law of the United States, [and that] … they shall each be fined under this title or imprisoned not more than twenty years, or both.”

Taken together, these laws give ample scope to punishing the kind of campaign that was waged against the certification of the electoral slates and culminated in the Jan. 6 attack on Congress. Moreover, disqualification from future office is one of the penalties. Of course, a criminal trial is a wholly different proposition; the burden of proof is higher than in an impeachment proceeding, and the trial is before an impartial jury rather than the Senate, which has divided along party lines in recent impeachments. But isn’t that in fact preferable to stretching the Constitution to get at a president who has so manifestly committed a constitutional and a statutory crime?


Sometimes a constitutional argument is hard to classify. In my essay, I argued that one reason in favor of my interpretation of the language of Article II is that if private persons—including former officers—could be convicted and punished by means of impeachment, the prohibition against bills of attainder in Article I, Section 9, Clause 3 would be violated. In other words, when you strip the disqualification penalty from removal, you are left with a naked legislative trial and punishment of private persons, precisely of the kind practiced by the British. One might argue that this is a historical argument. British impeachment practice permitted the prosecution of former officials and bills of attainder, and the Framers and ratifiers clearly intended to reject the latter. Or is this a textual argument, construing the language of Article I strictly (“a bill of attainder is an act of a legislature declaring a person guilty of some crime and punishing that person without the normal judicial process”)? Is it a matter of constitutional ethos, the notion that the process that is “due” citizens who are not public officials is different? I doubt, for example, whether we could court-martial private persons, including those who formerly served in armed forces when their crimes were committed, see U.S. ex rel. Toth v. Quarles, unless they remained in some connection with the services (see U.S. v. Begani, pending before the U.S. Court of Appeals for the Armed Forces).

In another forum, Bowman characterized my original Lawfare piece (“really a dreadful exercise in setting up & knocking down straw men that almost completely ducks the serious arguments on the other side“), and I fear my reply may not have altogether quieted his outrage. The hanging of horse thieves, the harassing impeachments of former presidents and other high officials after they have left office, courts-martials of former military personnel—or impeachments of them when we fear “a man on horseback” and want to bar him from running for office—these must seem pretty far-fetched. Nevertheless, it is a useful reminder to recall that the Trump campaign proposed impeaching Hillary Clinton before the 2016 election for misdeeds done when she had been—though she no longer was—secretary of state. Their goal? To disqualify her from being eligible for the presidency.

If it is true that the country has just passed four years of increasingly aggressive disregard, indeed derision of, historic constitutional norms, one might conclude that it is time for accountability beyond what the election provided. If Trump’s behavior is without parallel, perhaps he merits some unparalleled action in response; while it is true that the country has never impeached and convicted a former civil officer—no president, no judge, no cabinet member, in our history—we have never before experienced such a sustained assault on the Constitution by an officeholder.

I am also sorely conscious of the irony that Trump’s behavior might take us another step down the path of “constitutional hardball,” where decent, thoughtful and law-abiding citizens have been goaded into behavior that is uncivil at the least and unconstitutional at the worst.

I have the greatest respect for Frank Bowman. I would never say that I was infallible—if I did, my family would roll their eyes—only that I, like him, am trying to do my best. In this dispute, I am not saying that Bowman and “the vast majority of other scholars with expertise in the field” should simply adopt my views. My message is, regardless of the near-unanimity of opinion: Don’t simply rely on how many persons are willing to endorse a constitutional position (“Around here,” Dean Acheson once memorably said, “we don’t count heads, we weight them”); carefully consider what may appear to be a contrarian argument; make the effort to deal with “the serious arguments on the other side”; and question your judgment where Donald Trump’s behavior is concerned, so that he doesn’t tempt you into supporting a ruling that, however it can be plausibly supported by talented and zealous advocates in this egregious case, ultimately undermines the Constitution and the people’s faith in it.

Moreover, although this may not disturb many of my readers, maintaining that the Senate obviously has the power to try a private person, that there are no serious arguments otherwise, will de-legitimate the Republican acquittal. Doubtless that’s just fine for many people. It may even be that pressing for a Senate trial, which has little chance of a conviction, is driven largely by the effort to expose Republicans in the Senate to public criticism. But this is just the sort of thing for which I have criticized jurists such as Judge Robert Bork and Justice Antonin Scalia: the claim that those who disagree with us cannot possibly have a legitimate reason for doing so. It is just as destructive in our hands as in theirs.

I may be wrong in my constitutional analysis; many people in my dolorous field for whom I have the highest regard and the tenderest affections disagree with me. But my arguments are not “flimsy” and, even when made by those I happen to oppose politically, they are not unconventional or without foundation.

Trying and convicting a private person—whenever he may have been impeached and whatever office he may once have held—directly contradicts the text of the Constitution. Despite exaggerated claims to the contrary in the media, there is no “solid” precedential authority for such a step and what history does exist that has focused on the question—the Federalist Papers, “The Commentaries”—goes the other way. A trial in the Senate of a private person is not “obviously constitutional.” What is driving excellent scholars and journalists into contortions is really one issue only: What should the country do if a truly egregious, unconstitutional, impeachable act is done by a president whose term ends before he can be tried or who resigns—as Nixon did—to avoid impeachment?

On this question, Richard Danzig and I depart from a great many of the distinguished commentators on the subject. Because criminal prosecution—which includes the remedy of barring the convicted person from future office—as well as censure by concurrent resolutions in both houses are available, I think it is a mistake to twist the Constitution in order to fashion a weapon to punish Donald Trump.

Although I mention the statutory option of barring Trump from future office, I do so mainly to remind readers that such disqualification would be the only consequence of an impeachment conviction now, if it withstood judicial scrutiny. In fact, I am not afraid of the most effective method of rebuke in a democracy: an election. Trump lost the Republican majority in the House in 2018, almost single-handedly lost the Republican majority in the Senate in 2020 and decisively lost the presidency. Preventing him from running again may well be in the interest of the Republican Party, but I see no need to protect the country from another four years of Trump: The voters will do that. In a democracy, that—and the most punctilious adherence to the Constitution—is where the country should place its chief reliance.