Published by The Lawfare Institute
in Cooperation With
Weeks after the Jan. 6 riot at the Capitol Building, the fallout continues as the Senate convenes for the second impeachment trial of former President Trump over Trump’s behavior before and on the day of the riot. Meanwhile, some members of the House have accused other members of inviting future rioters into the Capitol before Jan. 6. And questions of how Congress disciplines its own members are relevant once again after a majority of the House voted to remove Rep. Marjorie Taylor Greene from her committee assignments for anti-Semitic and violent statements that she made before entering Congress. Also pending in the House is a motion to expel her for those statements.
The clauses in the Constitution governing expulsion and exclusion from Congress, impeachment, disqualification from public office, and pardons are obscure and have rarely been interpreted by Congress or the courts. Yet these clauses are interrelated and can lead to striking and provocative conclusions. Any member of Congress who assisted the Jan. 6 assault on the Capitol can be not only expelled from Congress but also excluded upon reelection. Short of conviction by the Senate in the impeachment trial, former President Trump can be disqualified from holding any federal or state office if the Senate makes a finding of fact that he engaged in the Jan. 6 insurrection. And, perhaps most strikingly, pardons issued and the bills signed by Trump after Jan. 6 are constitutionally invalid.
Expulsion and Exclusion Under Article I
Article I, Section 5, Clause 2 of the Constitution authorizes each House to “punish its members for disorderly behavior and, with a two-thirds vote, expel a member.” To best explain how expulsion interacts with Section 3 of the 14th Amendment, I’ll first review the expulsion process separate from Section 3, and then consider them alongside one another.
Greene was sworn into Congress on Jan. 3, the first day of the present session. She can be expelled for statements—speeches and tweets—that were made before that date. Article I, Section 5, Clause 2 is an enumerated power that is written in unqualified language and gives the House plenary authority to expel a member. The House alone can determine the cause for expulsion—any conduct that reflects negatively on the House. The House has consistently taken that position and the Supreme Court endorsed it in In re Chapman, one of its two cases that discuss expulsion.
An expulsion decision by the House is almost certainly not subject to judicial review. United States v. Nixon held that decisions by the House and Senate on impeachment are not judicially reviewable. The case for exclusive jurisdiction in each House to discipline its members, including with the ultimate punishment of expulsion, is at least as strong.
This does not mean that there are no constraints against the House abusing its expulsion power. The constitutional constraint is the requirement of a two-thirds vote. There are also serious policy constraints: political partisanship, fear of constituent backlash and a natural reluctance to expel a member for misbehavior because that could be turned onto any other member. In particular, one would expect members of Congress to be reluctant to expel a member for statements outside of the House. Expulsions have in fact been quite rare. Only 15 senators have been expelled from the Senate, 14 of whom were senators representing Confederate states; in the House, only five members have been expelled, three of whom were Confederates. Lesser punishments, which both chambers can impose by majority vote, have effectively replaced expulsion. Censure has been used 23 times in the House, and reprimand has been used 11 times in that chamber. The chambers can also use withdrawal of privileges like committee assignments.
The House has sometimes taken the position that a member should not be expelled for conduct that occurred prior to that session of Congress, at least when that conduct was known at the time of the election—which would put a constraint on expelling Greene for her statements before Jan. 3. The reasoning is that the voters knew about the misconduct and elected the member anyway. Expelling the member would amount to the House overriding the will of the people. But this is an issue of policy and not of constitutional law. In the Chapman case, the Supreme Court endorsed the 1797 Senate expulsion of William Blount for actions taken before he took office. Whether the House adheres to its precedents on pre-session conduct is a matter of discretion.
Imagine that Greene is expelled for her pre-Jan. 3 statements but runs again and is reelected. In this scenario, the House cannot refuse to seat her. Exclusion is very different from expulsion: Exclusion requires only a majority vote, and the House can refuse to seat a member only for not satisfying the qualifications listed in the Constitution for age, citizenship and residency under Article I, Section 2, Clause 2. Madison was emphatic in The Federalist Papers that neither House could impose additional qualifications, and the Supreme Court so held unanimously in Powell v. McCormack. Powell was accused of embezzling House funds during earlier sessions, and the House voted to exclude him. The Supreme Court held that he could not be constitutionally excluded, while acknowledging that he could be expelled on a two-thirds vote. So if Greene was expelled for her vile statements, she would have to be seated if reelected. Theoretically, she could be expelled again, but not as a practical matter.
Section 3, Expulsion, Exclusion and Disqualification
Now consider the effect of Section 3 of the 14th Amendment. That provision states that no person can be a member of Congress or hold any federal or state office if the individual previously took an oath to support the Constitution as a member of Congress, as an officer of the United States, or as a state legislator, executive, or judge, and then engaged in insurrection or rebellion against the United States or gave aid and comfort to the enemies of the United States (the last term being a constitutional definition of treason).
Suppose that the FBI uncovers evidence that before Jan. 3, a member of Congress conspired with extremist groups to storm the Capitol for the purpose of overturning the presidential election. Of course, that member could be expelled for this behavior independently of Section 3, as outlined above. But any pre-Jan. 3 conduct might not fall within Section 3 because that provision applies only to those who had taken the constitutional oath—that is, prior officeholders or members of the military. Greene, for example, was neither before entering Congress, so Section 3 would not apply even if she were found to have taken part in the insurrection prior to becoming a member of Congress.
This limitation of Section 3 may not seem to make sense, but the oath requirement was added to target the white Southern aristocracy that had governed the South before the Civil War—a group that Republicans blamed for the Civil War. If Section 3 had been written to apply to all people who engaged in the rebellion, it would have disqualified the vast majority of all white Southerners. Gerard Magliocca describes this history in his excellent recent article discussing the development and post-1868 enforcement of Section 3.
If the FBI uncovers evidence that a member of Congress assisted the insurrection between Jan. 3 and Jan. 6, by contrast, the member would be subject to Section 3 in an expulsion proceeding. What if the member then runs for and wins reelection? As described, expulsion alone does not disqualify a one-time member of Congress from being sworn in again—so the question is whether adding Section 3 into the mix would change the outcome. To put it another way, does Section 3 only state grounds for expulsion, or does it also impose a disqualification requirement?
Facially, Section 3 appears to impose a new qualification for being a member of Congress: “No person shall be a Senator or Representative in Congress” who engaged in insurrection, rebellion or treason against the United States after having taken the constitutional oath. Compare this to Article I, Section 2, Clause 2, which set the pre-amendment age, citizenship and residency qualifications for representatives: “No person shall be a Representative who”—exactly the same categorical language as used in Section 3. And the same categorical “No person shall be” language is found in Article I, Section 3, Clause 3, which set the age, citizenship and residence qualifications for being a senator.
Age, citizenship and residence were the initial three qualifications for being a member of Congress. The 14th Amendment added a fourth qualification—those who violated their constitutional oath by engaging in insurrection, rebellion or treason cannot be representatives or senators. Thus, Section 3 would prevent a member from being seated after winning reelection if that member engaged in the insurrection after taking the constitutional oath as a member of Congress on Jan. 3.
If this is correct, there is another important implication. Normally, expulsion is discretionary—and, as a result, has been imposed very infrequently. But if Section 3 sets a qualification for being a member of Congress, it would follow that a violation of Section 3 would make expulsion mandatory. Suppose representatives or senators were seated with the understanding that they were residents of the states they represent when they were elected, but this understanding later turns out to be incorrect. Or suppose it turned out, after being seated, that they were younger than the prescribed constitutional minima or not citizens for the prescribed period of time. The qualifications clauses state that under these circumstances they can’t be members of Congress (“No person shall be a Representative [or] Senator ...”). Under these circumstances, expulsion would appear to be constitutionally mandated. The same should be true for violating Section 3, which sets a new qualification.
Section 3 and Impeachment
This analysis of Section 3 and expulsion and exclusion applies to the president and impeachment as well. And it leads to some unexpected consequences.
In analyzing Section 3’s applicability to an impeachment trial of the president, the first step is to determine the breadth of that constitutional provision. Section 3 does not explicitly mention the president—it refers to any person who took the constitutional oath as an “officer of the United States”—and there is case law that statutes using this general term presumptively do not apply to the president when he isn’t mentioned explicitly. This is a rule of construction for statutes, based on the notion that Congress would normally hesitate to apply regulatory legislation to the president personally. But this notion does not apply to construing a provision of the Constitution, a document that itself regulates the president and his eligibility to hold that office.
Section 3 applies literally to any “officer of the United States.” According to the dictionary, the first definition of “officer” is “one who holds an office of trust, authority, or command.” That certainly describes the president as chief executive. The second definition of “officer” is “one who holds a position of authority or command in the armed forces,” which describes the president as commander in chief. And consider the breadth of Section 3: Its jurisdiction is greater even than impeachment because it applies to every person who took an oath as a member of Congress, or as a military officer of the United States, or as a state legislator, judge or executive—three categories of individuals who are not subject to impeachment. Similarly, the disqualification requirements of Section 3 are also broader than impeachment because they include all state as well as federal offices. And it is difficult to imagine that the Republicans who were responsible for enacting Section 3 intended to exempt President Andrew Johnson, whose loyalty they questioned (to put it mildly). Given the breadth and history of Section 3, holding that the president is not an “officer of the United States” within the meaning of the Constitution would be fairly startling.
Just as a violation of Section 3 creates a mandatory ground for expelling and excluding a member of Congress, the same mandatory requirement should apply to an impeachment trial of the president. Ordinarily, the commission of an impeachable offense doesn’t mean that the Senate has to convict the official or disqualify him or her from holding federal office; President Clinton’s case is a good example. Similarly, even for an official convicted by the Senate in an impeachment trial, disqualification from holding office is ordinarily a discretionary penalty.
But Section 3 is written with the same categorical language as the presidential qualifications clause in Article II, Section 1, Clause 5: “No person” shall be “eligible to the Office of the President” who does not meet the citizenship, age and residence requirements. A person who does not meet those three qualifications cannot be president.
Section 3 added a fourth qualification for presidential eligibility: “No person” who was a public official and engaged in insurrection against the United States can hold the office of president. Because Section 3 creates a new qualification for holding the office of president, a violation of Section 3 should require the Senate to convict the individual and disqualify them from being eligible to hold federal or state office. Another way of stating this is that Section 3 describes a particularly serious impeachable offense—so serious that a violator cannot hold any other federal or state office.
Even if two-thirds of senators do not vote to convict Trump, there are still several possibilities for enforcing Section 3 against him. Section 3 was enforced outside of Congress post-1868 under a federal statute authorizing the attorney general or the U.S. attorneys to bring quo warranto actions—that is, writs challenging eligibility of office—to bar violators from holding public office. The most dramatic was a lawsuit to remove a majority of the Tennessee Supreme Court. That quo warranto statute was repealed in 1948. It could be reenacted and applied to Trump and others who took the constitutional oath and engaged in insurrection.
A censure resolution by the Senate or a concurrent censure resolution by the House and Senate could represent a form of public accountability for official conduct that threatened constitutional government. However, it’s doubtful whether such formal expressions of opinion could have any legal effect outside of Congress. The Constitution prohibits bills of attainder, which are legislative declarations of guilt and punishment, and giving legal effect to a censure resolution to disqualify a person from holding any state or federal office appears to be such a declaration. Section 5 of the 14th Amendment gives Congress the power to enforce Section 3 by “appropriate legislation”—but a censure resolution is not legislation. A joint resolution signed by the president is legislation but could also be an inappropriate bill of attainder, especially if it targeted one individual. And if such a legal effect of censure resolutions doesn’t seem troubling as applied to a discredited former president, think of such disabilities being imposed through legislative resolutions during a new McCarthy era.
Of course, there is no bill of attainder problem in a criminal prosecution. As a private person, Trump could potentially face prosecution for inciting insurrection—as unlikely as that might seem. It’s an interesting question whether the government could satisfy the standard for incitement under Brandenburg v. Ohio to hold Trump criminally liable for his speech—a matter that doesn’t arise in the impeachment context for a public official’s speech, as was the case in Trump’s first impeachment for inviting a foreign government to investigate a political rival. If Trump were convicted in a criminal trial, that would constitute a finding that he engaged in insurrection, which would then warrant judicial enforcement of Section 3 in sentencing.
Section 3 can also be enforced in civil litigation. In the language of constitutional law, Section 3 appears to be self-executing, reading, “No person shall ….” That is, Section 3 applies without enabling congressional legislation—just like Section 1 of the 14th Amendment, which operates under a similar command and reads, “No state shall ….” This means that if Trump ran for the presidency in 2024, he could be barred from appearing on ballots by state election officials—and the decision to bar him would be judicially reviewable. If election officials did not bar Trump, an opposing candidate or political party could sue in state court to remove him under Section 3. However, this solution would appear to require a factual hearing on whether Trump engaged in insurrection and was therefore disqualified, which would leave Trump’s eligibility to hold the office of president unsettled for four years and then subject to potentially lengthy and contentious litigation. The intersection of impeachment and civil litigation may offer a solution to this problem.
Impeachment is a constitutional exception to the prohibition of bills of attainder, and the Senate is the trier of fact and judge of the law in an impeachment trial. The Senate can make findings of fact in the trial previous to the vote on conviction. Two findings of fact could be made: First, Trump took the constitutional oath, and second, he engaged in insurrection against the United States. These findings can be made by a majority vote, just like rulings on witnesses and evidence. Factual findings made in a trial in which Trump is a party could be admissible and presumably binding on him in future litigation. So if Trump again ran for president—or, for that matter, mayor of Palm Beach—the findings of fact made by the Senate could be dispositive in ballot eligibility litigation and could trigger the disqualification requirements of Section 3 without the necessity of new findings of fact.
One objection to the use of such findings would be a challenge to the Senate’s jurisdiction over the impeachment trial. But congressional jurisdiction clearly attached, because Trump was impeached by the House when in office for conduct committed in office. I am not aware of any criminal or civil case in which jurisdiction properly attached through indictment or the filing of a complaint and service of process, respectively, and was then lost through extraneous circumstances—such as the defendant resigning or leaving office, or even leaving the country.
Another objection would be that findings of fact in an impeachment trial should not be treated the same as findings in a judicial proceeding. But Congress was chosen by the Framers to be the court that adjudicates alleged “high crimes and misdemeanors” by federal executive officers and judges.
Section 3 and Pardons
If Trump engaged in an insurrection on Jan. 6 and was therefore disqualified from public office under Section 3 as of that date, this has some startling ramifications. It raises the issue of whether the actions that he took between Jan. 7 and Jan. 20—including the pardons he issued and the bills he signed into law—are constitutionally valid. If, as I have argued, Section 3 mandates a qualification requirement for eligibility to hold the office of the presidency, those pardons and bills would be invalid. The ordinary rule is that the decisions of a federal official who holds office in violation of the Constitution are not law.
Trump’s post-Jan. 6 pardons could be challenged by the Justice Department in criminal cases and the bills that he signed could be challenged in civil cases by injured parties. Admittedly, the former is less likely than the latter.
It is of course not obvious that the courts would entertain those challenges on the merits. Using the “political question” doctrine to avoid the merits seems initially attractive. A pardon, for example, is textually committed to the president. But the issue is whether he was legally president in light of Section 3’s disqualifications, and there is no “textually demonstrable commitment” to Congress (or the president) to make that determination in all cases. This provision of the Constitution appears to be self-executing and applies to numerous federal and state officials apart from the president. And raising “political question” avoidance could present the government with a Catch-22: The government cannot easily prevent enforcement of Section 3 in private litigation without waiving its authority to challenge unauthorized pardons in criminal cases. Such challenges to presidential capacity could raise these and probably more intricate and unprecedented issues. The only obvious conclusion at this point is that the enforcement of Section 3 can lead into uncertain terrain with potentially remarkable consequences.
It may seem strange to place contemporary reliance on Section 3 of the 14th Amendment—a provision of the Constitution that was enacted in the wake of the Civil War, that has not been utilized since 1872, and that few people even realized existed before Jan. 6, 2021. But Section 3 has not been repealed and is as much the supreme law of the land as any other provision in the Constitution. Americans should be pleased that Section 3 has laid dormant for so long. That is a testament to a norm of constitutional government that was obeyed for more than a century until it was recently shattered—and which desperately needs to be restored.