Democracy’s Other Boundary Problem: The Law of Disqualification

Almost all national constitutions contain one or more ways to disqualify specific individuals from political office. Indeed, the U.S. Constitution incorporates at least four overlapping pathways toward disqualification. This power of disqualifying specific individuals or groups stands at the heart of the complex project of maintaining democratic rule. In practice, disqualification can work both as an instrument for preserving democratic rule and also as a knife against it. This Article is the first to systematically analyze the complex positive and normative questions raised by disqualification. We offer both a positive account of the function that disqualification plays in constitutional ordering and a normative account of the role that it should play. Drawing on domestic and comparative evidence, we develop the blueprint for an “optimal” disqualification regime. This regime would disqualify officials who pose a clear threat to a relatively minimalist, electorally focused conception of democracy, while avoiding overuse for less pressing ends or, worse, abuse for anti-democratic purposes. It would contain multiple pathways that would be calibrated to avoid the possibility of partisan arbitrage. These pathways would generally regulate individuals as such rather than groups and parties. Usually, they would not run directly through elected bodies. The substantive prerequisite for disqualification would more often be stated as a rule than as a standard, and the ensuing prohibitions would more often be temporary than permanent. This optimal approach leads to specific reform recommendations for the United States. First, we demonstrate that Section 3 of the Fourteenth Amendment should be given greater specificity and shape via statute, just as Congress did after the Civil War and as it is empowered to do now via its authority to “enforce” the terms of the Reconstruction Amendments. Second, we develop a case for a framework statute setting forth a judicial mechanism for enforcing the two-term limit on chief executives under the Twenty-Second Amendment. Finally, we propose a decoupling of impeachment from disqualification, creating two distinct institutional pathways for disqualification.

Table of Contents Show

Introduction

On January 6, 2021, then-President Donald Trump addressed a rally on the Capitol Mall and called again on his supporters to challenge Congress’s certification of Joe Biden as President. Some of those supporters broke away and attacked the Capitol. Five people died in the resulting violence. On January 13, 2021, as a result of President Trump’s actions on and surrounding January 6, he faced a second impeachment proceeding.[1] As the Senate trial proceeded against him, a central disagreement arose concerning whether he should be able to stand for public office in the future.[2] The House impeachment brief put the case for disqualification in no uncertain terms: “To protect our democracy and national security—and to deter any future President who would consider provoking violence in pursuit of power—the Senate should convict President Trump and disqualify him from future federal officeholding.”[3] Law professors also called for the invocation of a “little-known” provision in Section 3 of the Fourteenth Amendment permitting the disqualification of those who had “engaged in insurrection or rebellion against” the U.S. Constitution.[4] Disqualification, some noted, would simply relegate Trump to the same status as noncitizen residents and those who became citizens via naturalization: ineligible to run for the presidency.[5] Others, while sympathetic to the aim of excluding President Trump from public office, raised concerns about how this bar would be created in practice.[6] And some raised a worry that, whatever its practicalities, an effort to disqualify former President Trump via the Fourteenth Amendment or otherwise would be unavailing: he could simply find a proxy to run for him, wielding formidable influence through his role as de facto head of the Republican Party rather than via a formal office.[7]

As the drive to disqualify President Trump shows, there are multiple paths, including those found in the U.S. Constitution, to cast out specific individuals from a formally elected or appointed role in political life. The effort to disqualify President Trump is but one specific instance in a long and checkered history in which disqualification mechanisms have been used for both good and ill. Among the many examples of disqualification that can be drawn from both U.S. and international history include:

• In 1870, a U.S. Attorney in Tennessee brought actions against three members of the state supreme court, alleging that they had given “aid and comfort” to the “traitorous organization known as the Confederate States of America” and urging that the state justices be removed from office.[8] These actions were brought pursuant to the Enforcement Act of 1870.[9] In the same period, the Senate also refused to seat a former Confederate on Section 3 grounds.[10] In 1872, however, Congress enacted an amnesty law removing most Section 3 disabilities.[11]

• In January 1966, the Georgia House of Representatives declined to seat civil rights leader Julian Bond for giving “aid and comfort to the enemies of the United States and Georgia” by criticizing the Vietnam War.[12] The U.S. Supreme Court held that the legislature’s action violated the First Amendment.[13] Three years later, the Court held, on different constitutional grounds, that the U.S. House of Representatives lacked power to exclude another African-American representative, Adam Clayton Powell.[14]

• In 2004, Lithuania’s parliament impeached President Rolando Paksas for granting citizenship to a campaign donor. Lithuania’s constitution contained no clause whereby removed presidents could be permanently barred from ever running again. But when Paksas sought another term in presidential office, the legislature passed an amendment to the electoral law prohibiting any impeached leader from competing again for office.[15]

• In 2014, Ukrainian President Viktor Yanukovych was driven from office due to mass protests and took refuge in Russia. The Ukrainian parliament passed a “lustration” law, banning those who had worked for Yanukovych from holding office for the next several years.[16] This barred 2,686 people, including the former president, from seeking future office.[17]

• Also in 2014, the Constitutional Court of South Korea banned the Unified Progressive Party, which espoused pro-North Korean views, on the grounds that its activities violated the basic democratic order.[18] As a consequence of that party ban, five members of the National Assembly lost their seats immediately.

• Finally in 2018, former Brazilian president Luiz Inácio Lula da Silva was ruled ineligible to run in an impending presidential election due to a corruption conviction—throwing that race into disarray.[19] Three years later, another court annulled Lula’s conviction, once again scrambling the electoral calculus.[20] And in 2023, Brazil’s Electoral Court held another recent former president, Jair Bolsonaro, ineligible to run for office until 2030 because of false claims the Court found that he had made about electoral fraud.[21]

One could say that a democratic choice was taken away from the people in each of these cases. In all of these incidents, however, disqualification was justified by reference to the past acts or affiliations—not the status, origin, or identity—of a political actor. In each instance, no matter how popular or unpopular the individuals, they could no longer play a role in domestic politics, either temporarily or permanently, because of what they had (allegedly) done.

These cases show how democracies that pride themselves on broad participation in selecting leaders and broad opportunities to run for office can use disqualification as a constraint on democratic possibility. They demonstrate that the power to disqualify in practice stands at the heart of the complex project of democratic rule. Janus-faced, disqualification can be both an instrument for preserving democratic rule and a knife for its murder.

This Article is the first to isolate and examine democratic disqualification as a general problem in constitutional law. We draw on U.S. and comparative examples to offer both a positive account of the function that disqualification does in practice play in many constitutional orders and a normative account of the role that it should play.[22] The focus of our inquiry is a class of mechanisms for identifying and excluding specific individuals or groups, whether by discrete adjudication or general legislative rule, from public office, whether temporarily or permanently. This distinguishes “disqualification” mechanisms from the ex ante categorical exclusion of certain classes of persons—such as noncitizens, minors, or, even more dubiously, women or racial and ethnic minorities—from public office. Disqualification is also distinct from criminal prosecution or conviction. The latter may (but need not) be an instrument for disqualification. Furthermore, political exclusion can be and often is implemented through mechanisms that go well beyond the criminal justice process. This definition enables us to develop both the positive and normative implications of allowing the polity the power to police itself through targeted, or group-focused, exclusions from political power.

Our initial aim is to surface the concept of democratic disqualification as a distinct object of constitutional choice. It presents a specific iteration of a more pervasive problem of democratic design: the tension between democratic self-realization and democratic self-destruction. Democratic institutions have a reasonable claim to set the terms of political participation. The forms of elections, rules for candidate and voter qualification, and ballot access rules are all commonly matters for democratic decision. Yet at the same time, there is a risk that the power to set rules for the democratic game will be used to fence out disfavored groups, to entrench incumbents beyond electoral challenge, and to create an image of democratic competition without its substance. The resulting tension is apparent in debates over the drawing of electoral district lines, the choice of electoral systems, and the question of who becomes a citizen in the first place. There is, indeed, a voluminous literature on some of these issues—in particular, on the question of whether and how a democratic polity can set entrance conditions on political involvement, including the right to vote and run for office. The focus of this literature is typically on where the “boundaries” of the people lie.[23] More generally, it is common ground that a democracy can benefit from many nondemocratic institutions, such as election commissions, independent prosecutors, and independent constitutional courts, which all can be turned against the larger cause of democracy. We argue by analogy to these problems that disqualification is properly understood as a power of democratic prophylaxis but also democratic destruction. Unbounded, it imperils democracy as a going concern. Absent, it means lost opportunities to deepen and defend democracy.

The category of disqualification unifies several otherwise disparate elements of the U.S. Constitution and lines of case law that until now have been analyzed separately. And the U.S. Constitution is not alone in folding in disqualification mechanisms. Around the world, constitutions employ a range of mechanisms to allow the targeted removal of individuals or groups from political life. Drawing on the Comparative Constitutions Project database of organic documents enacted since the late 1780s, we develop a more complete typology of disqualification mechanisms than the U.S. Constitution currently allows.

How should these disqualification regimes be judged? The evaluation of a constitution’s disqualification rules, we argue, centers on a balancing of the risk to democracy on the one hand and the gains to be had from democratic control of individual exit on the other. It raises the question of whether a constitution’s design—say, linking disqualification to impeachment and permitting group lustration of seditionists in the U.S. case—minimizes costs while maximizing benefits to democracy. Further, the justification for a specific disqualification regime is necessarily relative. It turns not just on its absolute efficacy in advancing democracy, but also on its relative efficacy in comparison to other feasible legal devices.

With these considerations in mind, we conclude that well-designed disqualification rules have a place in democracy’s arsenal. But in practice many are too imprecise in operation to be effective or insufficiently hedged with safeguards to avoid the substantial risks of abuse. We also offer a necessarily rough estimate of what an “optimal” disqualification regime might look like. In brief, it would contain multiple pathways, calibrated to avoid the possibility of partisan manipulation. These different mechanisms would lean toward the regulation of individuals and not groups. They would usually not run directly through elected bodies. Instead, they would leverage the distinct institutional strengths of administrative agencies and courts. The substantive threshold for disqualification would more often be stated as a rule than as a standard.[24] And finally, the ensuing prohibitions would more often be temporary than permanent.

Finally, we apply our “optimal” framework as an analytic lens to improve design and practice in the United States. The United States already contains a surprisingly robust set of disqualification mechanisms, but the Trump example suggests that the disqualification mechanisms in the U.S. Constitution are too fragmented and cumbersome to respond to contemporary threats to democracy. Impeachment is too difficult a tool to wield, especially in light of the modern American party system. The example involving President Trump suggests that it may be all but dead as an effective disqualification tool. Section 3 of the Fourteenth Amendment, which has received a sudden infusion of scholarly and journalistic attention, shows more promise.[25] We argue that it could be revitalized and improved via a carefully crafted statute, one which would expand on and offer precision to the substantive standard, rely on courts rather than political actors for enforcement, and so limit opportunities for partisan manipulation. We further highlight the underappreciated role of presidential term limits, entrenched in the Twenty-Second Amendment, as safeguards of democracy. These term limits have costs, but their rule-like character is an important advantage, especially considering the standard-like quality of other disqualification mechanisms. They have never truly been challenged in the United States, unlike in many other countries. As we explain below in greater detail, we think that it is possible that under stress, term limits could prove unexpectedly fragile despite their textual incorporation in the Twenty-Second Amendment.[26]

Most ambitiously, we suggest the attractiveness of creating a new, non-legislative pathway for disqualification, which would work alongside the existing impeachment mechanism. The aim would be to create a channel that would be less likely to fall prey to either partisan paralysis or partisan capture and that could be keyed to a broader range of modern anti-democratic threats than Section 3’s narrow, historically bound focus on “insurrection or rebellion.”

Our argument proceeds in six parts. Part I situates the problem of democratic disqualification in the context of democratic theory. Part II maps the existing disqualification mechanisms found in the U.S. Constitution. This reveals a surprising breadth of tools but also a high degree of fragmentation across a governance system riddled with ambiguities and inertia. Part III locates the U.S. design in comparative context. It provides a global, comparative map of disqualification tools. This is organized around two axes: individual versus group forms of disqualification, and forward-looking versus backward-looking restrictions. This parsimonious framework allows us to explore and classify a surprisingly varied set of tools that govern individual exit from democratic competition. These include “militant democracy” bans of anti-democratic parties, lustration or purges of “tainted” officials during democratic transitions, disqualification of officials following conviction by impeachment or via analogous judicial or administrative process, and term limits. Although we show that the landscape here is broad, we also suggest that there are significant problems with the way many of these tools function in practice. Using the insights of the first three parts, Part IV elucidates a better path forward. We outline what an “optimal” disqualification regime would look like, illuminating key choices. Part V explores how the existing system of disqualification tools could be improved, and more wisely deployed, in the United States, notwithstanding the limits imposed by the constitutional text. A short Conclusion follows.

I. The Problem of Disqualification in Democratic Theory

Democratic theory offers a lens through which to analyze the appropriate scope and operation of disqualification measures. To begin our inquiry, we draw on this body of theory to show why disqualification is a double-edged sword: both a warranted protection for and a risk to democratic government. The resulting tension is the central normative problem in democratic disqualification rules. This problem has been grasped, at least in part, in two other discussions. These are known as the boundary problem of democracy and the problem of ensuring the integrity of guardianship institutions. By considering these normative dilemmas and their practical implications (which have been extensively discussed in the literature before), we can better craft our vision of the ideal disqualification regime.

A good place to start is the claim advanced in defense of President Trump to the effect that his second impeachment attempt was “about as undemocratic as you can get.”[27] To understand disqualification from the perspective of democratic theory is to understand why that statement has a grain of truth, even as it also obscures a more powerful counterargument that builds on democracy itself.

A.    Democracy as a Problem of Ruling and Being Ruled

At its linguistic root, democracy is defined by the idea of the people as their own rulers.[28] Rule by the people requires that “all the members [of the polity] are to be treated (under the constitution) as if they were equally qualified to participate in the process of decisions about the policies the association will pursue.”[29] Democracy hence may not require elections: a system of lotteries for filling office, as the kind used in ancient Greece, achieves political equality and popular control without any voting mechanism.[30] However put into practice, democracy requires that members of the polity be able to both make and then revise decisions over time. When a democratic system ceases to be responsive to members of the polity at all—for example, when it becomes beholden to the wishes of a small clique holding power or exclusively to the top decile of the electorate as defined by wealth—that failure of ongoing responsiveness is also a failure of democracy.[31] A status quo “protected from the majority by all kinds of institutional trenches” is not meaningfully democratic even if elections occur because it is not dynamically responsive to the polity over time.[32] It is thus no accident that one of the oldest definitions of democracy, offered by Aristotle, centers on its durability over time by insisting on the idea of “ruling and being ruled in turn.”[33]

This ambition of democratic durability creates a foundational set of challenges for a democracy. Many decisions that a democracy necessarily makes implicate a risk of entrenchment. That is, a transiently ascendant group might seize a decision by the majority as an opportunity to entrench itself into absolute power positions, thereby ending the process of “ruling and being ruled in turn.”[34] Thus, a democratic system often requires (or at least benefits from) institutions designed to protect against entrenchment. Such institutions will often be situated outside popular control to ensure their efficacy in moments of need.

Yet the very institutions that are intended to guard democracy can also be turned to democracy’s undoing. This problem is implicated in the classic question of “who guards the guardians?”[35] The central challenge in the design of democracy-guardian institutions is how to maximize their efficacy against democratic threats while at the same time minimizing the risk that they pose in potentially hindering democracy.

B.     The Problem of Ruling and Being Ruled in Practice

Democratic theorists have worked through these tensions through the lens of two analytic problems in the design of democratic institutions. The first is a threshold question called the “boundary problem,” which concerns who is in the self-governing demos and who is outside of it.[36] How, that is, should the line between citizens and those without political rights be drawn? Second, theorists ask whether and how nondemocratic (or counter-majoritarian) institutions, known as “guardian institutions,” can be justified as instruments for preserving democracy. When, that is, should nondemocratic institutions be woven into the fabric of democracy? And when are those nondemocratic elements a problem for democracy? In resolving both these problems, the key question is whether a given institutional choice extends or undermines democracy’s durability. The same factor is key to thinking about disqualification.

To understand the boundary problem, theorists consider first how democracies draw limits on the geographic and demographic scope of the polity. Citizenship rules, franchise eligibility limits, and other laws all serve this end.[37] All implicate the normative question of how democratic power can be exercised without first identifying the self-governing entity that holds such power.[38] To avoid a fatal circularity, it is necessary to “include all interests that are actually affected by the actual decision.”[39] When people are affected in “non-trivial” ways by a polity’s decision, there is a moral argument for including them in democratic deliberations.[40] But doing so in a polity such as the United States opens a “Pandora’s box” since it means some large, perhaps unknowable, number of persons beyond the polity’s physical boundaries must be intermittently consulted on political decisions.[41] Any solution to the boundary problem, therefore, cannot rest on an appeal to a simple intuition about the right to be consulted when a state decision affects you. But neither is the exercise of democratic choice over boundaries by a given body of people obviously legitimate. After all, at issue is the very question of which people get to exercise such choice. Solving the boundary question rather requires a set of normative judgments about the justifications for, and the limits to, democratic rule.[42]

In practice, the boundary problem and its attendant normative questions are resolved by the transient occupants of political office. Transient political advantage, not high theory, is brought to bear. In determining the scope of the polity, those who happen to control the state at a given moment can act in ways that redound their partisan interests.

Rules for apportioning power within a polity can involve similar questions because they can also work as means of “zeroing out” political power for some groups. In the United States, the long-standing categorical exclusions of women, racial minorities, and other groups have directly entrenched the power of those who were able to exercise the franchise. As in many other countries, the vote was thus restricted to property-owning White males until the early twentieth century. Elsewhere, exclusions along these lines have been so severe as to undermine the claim of many countries to even be considered democracies.[43] Another prominent example is the assignment of power to expand or contract the geographical bounds of political constituencies. This obviously shapes the electorate’s ability to express its collective preferences, but in many cases such assignments are mechanisms for self-dealing.[44] In the United States, much of the constitutional power to manage elections is given to partisan state legislatures, many of which continually shape electoral rules to favor majority parties.[45]

Because there is no unique answer to these boundary-like problems, it seems inevitable that democratic decision-makers will often have the discretion to make judgments about democratic bounds with the aim of entrenching themselves. With this risk, democratic decisions to rule out at the threshold certain groups and individuals are often properly viewed with great skepticism.[46] Wise constitutional designers will hedge the risk of abuse with institutional checks or substantive limits.

The second problem that raises a parallel set of normative considerations is the decision whether to use nondemocratic guardian institutions to safeguard democracy.[47] The theory behind such bodies is that institutional mechanisms are necessary to respond to the risk that elected actors will abuse their authority and entrench themselves in power, and these institutions must be insulated from direct democratic control.[48] But what is to prevent the occupants of these offices in turn from abusing their powers?

A historic example of guardianship in ancient Athens took the form of a cultural norm. There, leaders cultivated a civil virtue of self-restraint by which citizens “restrain[ed themselves] from self-aggrandizing actions that compromise[d] another’s dignity.”[49] The norm itself was not democratic. It was, instead, elitist. But it served democratic ends.

Perhaps the most well-known example of a guardian institution is the judiciary. As John Hart Ely’s famous justification of judicial review notes, courts are in theory a means of responding to, and fixing, failures in the political process.[50] Often, the decision to channel the responsibility for protecting democratic principles (say, by banning parties or prohibiting certain election-related speech) into the courts reflects a belief that judges stand outside politics and can therefore be trusted to defend the political process. But a guardian institution standing apart from democratic currents is also an institution that may derail that well-functioning democratic process. In the United States, perhaps the best-known instance in which this is alleged to have happened was the 2000 presidential election. In Bush v. Gore, the Court intervened in the process of counting votes in the pivotal state of Florida in a way that seemed to many to hand the election to George W. Bush.[51] On one view of that case, the Court was asked to act as a neutral institution working to keep democracy on the tracks. It instead struck out on its own and determined the outcome of the election based on the conservative majority’s raw partisan self-interest.[52] In the eyes of those who took this view, the high Court shifted gears at that moment from being an instrument meant to facilitate democracy through the neutral enforcement of legal rules to instead an instrument of democratic degradation.

The large academic literature devoted to the “counter-majoritarian” difficulty presented by the federal courts can also be understood as an attempt to grapple with this double-edged quality of judicial action in relation to democracy:[53] the potential for it to be both a boon to democracy and an abuse of public trust.[54]

C.    Democratic Disqualification as a Problem of Ruling and Being Ruled

A parallel set of normative issues arises when a polity exercises a formal legal power to exclude a person or group from political power. Like the exercise of the power to define boundaries or to safeguard democracy via guardianship bodies, the use of a disqualification power implicates a balance between the individual entitlement to meaningful participation and the communal power of self-determination on the one hand, and the risk of entrenchment against “being ruled” on the other. Indeed, decisions about who, how, and when to exclude by disqualification are as much matters of collective self-determination as setting qualifications for inclusion and drawing boundaries on the polity at the outset. They require the exercise of a like discretion. Such discretion, however, can be used wisely or abused for self-entrenchment by incumbents. So disqualification entails the same normative trade-off as the boundary problem and the design of nondemocratic guardian institutions.

There are differences between the boundary problem and guardianship design on the one hand and disqualification on the other.[55] As an initial matter, ruling parties may view due process or related protections differently when faced with different sorts of exclusion. When facing disqualification, an existing officeholder might plausibly demand greater procedural protection than a person who has yet to assume an office or yet to participate politically in the polity. Second, regulating exit rules for sitting public officials is pragmatically challenging in a way that drawing a democratic boundary is not. Incumbents’ power can make them a substantial threat to the democratic order. They have the practical tools to resist a legal instruction to vacate their offices. Alternative tools of accountability, such as the use of criminal charges, may be impossible, or at least difficult to use, when it comes to incumbent officeholders. This is in part because those officeholders may enjoy legal protections from prosecution and in part because some officeholders may exercise control over the criminal process.[56] Alternatively, it may be very difficult politically for even the most independent of prosecutors to bring charges against a sitting chief executive.

Despite these practical differences, the study of boundary problems and guardian institutions as problematic in the design of democratic institutions yields several lessons for the design of disqualification rules given the normative problems.

First, there is no close correlation between the nature of the accountability tool used and the goal of positively preserving democracy. For example, democratic and nondemocratic guardian institutions can both vindicate and threaten the durability of self-government. By extension, it should not be assumed that a democracy-oriented disqualification will itself be democratically controlled (although it could be). Second, a single institution can at different points in time be a defender and a threat to democratic rule. Its orientation toward the democratic project can fluctuate over time. Consider the U.S. Supreme Court. This body defends the right to vote in some cases. In other instances, it halts presidential vote counting in a potentially outcome-determinative manner or else throws up needless obstacles to democratic participation. Accordingly, in designing democratic disqualification mechanisms, it is not enough to ask whether an institution will perform well in the abstract. It is necessary rather to look closely at its present composition and its wider setting to make a more context-specific judgment. It is also necessary to assess the risk over time of a previously beneficial institution being turned into a tool for entrenchment. Third, exercises of boundary-defining and guardianship powers demand an inevitable measure of discretion, but they are also amenable to demands for justification and account. It is possible, for example, to demand that constraints on citizenship be based on widely shared normative premises (and not, say, antipathy to a racial group or nationality). With respect to any exercise of power, such a justification can clarify whether the use of the power inflicts harm to the quality of democracy in a way that outweighs any offsetting benefits. For example, the Supreme Court’s jurisprudence on the individual right to vote can be understood as an effort to test whether exclusions from the right to vote are truly justified or merely self-dealing.[57] At the same time, the justifications embraced by the Court are often (plausibly) condemned as inadequate (as in the case of Bush v. Gore). At a minimum, this suggests that optimal disqualification design will elicit a specificity in proffered justifications to prevent abuses of power that undermine democracy.

II. Democratic Disqualification in the U.S. Constitution

Many believe that the U.S. Constitution has “undemocratic” elements to it.[58] Academics have highlighted the structural choices within the U.S. Constitution that constrain majoritarian rule or that facilitate minority control of the national government.[59] Yet missing from this important debate to date are the individualized mechanisms through which, in defense of democracy, the Constitution can be used to expel particular persons or groups or exclude them from political life. We call these individually targeted tools “retail” mechanisms as a way to distinguish them from “wholesale” mechanisms that sweep categorically across whole populations by geography, race, or gender.

This Part documents those retail mechanisms. Federal chief executives and legislators face different institutional risks when it comes to the prospect of disqualification.[60] We begin by focusing on mechanisms of political exclusion targeting the executive, primarily the U.S. presidency through impeachment with its sequel, disqualification; the anti-insurrection provision of the Fourteenth Amendment; and the two-term limit for presidents. We then turn to legislative exclusion mechanisms of the kind Julian Bond experienced in Georgia. All purport to preserve democratic rule by excluding persons from office.

To be clear, our account of disqualification mechanisms does not exhaust the tools of democratic exclusion in the United States, especially with respect to voting. Pursuant to the Twenty-Sixth Amendment, persons under the age of eighteen are excluded from the franchise.[61] More controversially, and closer to our topic, many states exclude people convicted of felonies from voting and officeholding. We leave felony disenfranchisement to one side because the U.S. Supreme Court has construed it as “an affirmative sanction” for crime that is permitted under other general protections for voting in the Constitution.[62] It is not a democratic prophylaxis in the sense that concerns us here. To be sure, felony disenfranchisement can be adopted because it advantages one party over another.[63] But it is not well understood as a form of disqualification on the basis of the threat to democracy posed by a specific individual or small group. No one, that is, thinks that an individual person convicted of a felony poses a direct threat to democracy; instead, disenfranchisement is justified on the basis that people convicted of felonies deserve to be stripped of the right to vote.

A.    Disqualification Following Impeachment

Perhaps the best-known vehicle for disqualification today is the impeachment process whereby the House and Senate can act against the President, other executive officials, or judges.[64] The best-known effect of impeachment is immediate removal from office. But the Constitution states that conviction on an impeachment charge may have the additional consequence of “disqualification to hold and enjoy any Office of honor, Trust or Profit under the United States.”[65] Thus, Congress has the ability not only to remove officials from office but also to prevent them from holding certain offices again.[66] In this regard, impeachment is meant to offer a durable response to the threat of “tyranny, arising out of the subversion of the constitution and the accompanying destruction of republican liberty.”[67]

Several aspects of post-impeachment exclusion are left unclear by text and history. The Senate is plainly the relevant decision-making body. Less obvious, however, is the voting rule for disqualification—a simple majority of the Senate, or the two-thirds supermajority required for conviction on the impeachment charge and removal from office.[68] The scope of the Disqualification Clause has also been a subject of debate.[69] The phrasing indicates that it applies to federal offices (i.e., those that are “under the United States”) and not state offices. But which federal offices? One scholar argues that disqualification applies to all executive and judicial offices (including the presidency and vice presidency) but not to legislative ones.[70] Another scholar disagrees, finding that disqualification applies only to appointed offices but exempts all elected positions, such as the presidency.[71] Although we concede the force of the textual argument, we find it implausible to imagine that disqualification was originally intended or understood to prevent someone from being appointed as Postmaster General (Second Class) out of concern for corruption but not to stop them from rising to the White House. We prefer a purposive reading that makes practical sense of the text.

The text is also unclear as to whether conviction in an impeachment trial is a necessary prerequisite for disqualification or whether disqualification can be achieved via other mechanisms.[72] To illustrate, it is not clear during an impeachment trial whether two distinct votes are necessary (one on removal and a second on disqualification) or whether both questions could be bundled into the same package. Nor is there any text in the Constitution that rules out the possibility of removal without the predicate of impeachment.

With respect to procedure, the historical practice in the Senate has been to hold two separate votes and to disqualify officials based on a simple majority once they have been convicted and removed via the two-thirds supermajority specified in the Constitution.[73] But there is at least a plausible argument that this is unconstitutional. The Constitution’s text does not specify a different voting threshold for the different consequences of an impeachment conviction.[74] Further, the Supreme Court has been reluctant to countenance shifts from a supermajority to a majority rule for legislative disqualification.[75] It is unclear why a different rule would apply to Article II.

Not only is the constitutional text unclear in theory, but many of these issues remain unresolved in practice as well. This is not simply because of the brevity of the relevant constitutional text. Many other terse elements of the Constitution have generated detailed regulatory frameworks (think of the reticulated jurisprudence that has developed under the Free Speech Clause of the First Amendment). Rather, legal uncertainty flows from the fact that disqualification has been rarely used. The House has impeached federal officials only twenty-one times in U.S. history.[76] Eight of the impeached—all federal judges—were convicted by the Senate, but the others either had their proceedings discontinued following resignation or were acquitted.[77] Only three convicted judges (West H. Humphreys in 1862, Robert W. Archbald in 1913, and G. Thomas Porteous, Jr., in 2010) were also disqualified from future office by a separate vote.[78] In all other cases, the judge was removed, and no vote was held on whether to disqualify. In 1989, for instance, Alcee Hastings was impeached and removed from the federal bench for allegedly accepting bribes.[79] He then went on to win election to the U.S. House of Representatives in 1992, where he remained until his death in office in April 2021.[80] Conviction on an impeachment charge has, in short, rarely served as a basis for disqualification from future political life.

B.    Insurrection and Disqualification

The second constitutional pathway to disqualification, little known until January 6, 2021,[81] emerged from the post-Civil War cauldron of racial conflict. In 1868, the Republican Congress proposed, and the states ratified, the Fourteenth Amendment in response to the continued failure of the southern states to respect Black civil rights.[82] In April 1868, Representative George Henry Williams (R-OR) proposed language for a constitutional amendment that would have excluded “all persons who voluntarily adhered to the late insurrection, giving it aid and comfort . . . from the right to vote for Representatives in Congress and for electors for President and Vice-President of the United States.”[83] In the Senate, this bar on voting was amended to include a bar on officeholding for those who had taken an oath in support of the Confederacy.[84] Section 3 of the Fourteenth Amendment ultimately wrought a disqualification rule along the lines approved by the Senate:

No Person shall be a Senator or Representative in Congress, or elector of President and Vice-President, or hold any office, civil or military, under the United States, or under any State, who, having previously taken an oath, as a member of Congress, or as an officer of the United States, or as a member of any State legislature, or as an executive or judicial officer of any State, to support the Constitution of the United States, shall have engaged in insurrection or rebellion against the same, or given aid or comfort to the enemies thereof. But Congress may by a vote of two-thirds of each House, remove such disability.[85]

As Professor Ruti Teitel observes, Section 3 is akin to the lustration provisions found in other constitutions and laws around the world.[86] These constitutional clauses bar people from certain offices based on their past activities or affiliations with some past, tainted regime. In this case, Congress aimed to bar officials who had served in the Confederacy and made war on the United States. As such, Section 3 is not just an instrument of Reconstruction but also an instrument of what later came to be known as “transitional justice.”[87]

Because Section 3 is relatively unfamiliar, we begin by summarizing its historical use and then discuss its parallels and divergences from impeachment.

1.     Historical Uses

The active history of Section 3 is significant, albeit brief. Even before the Fourteenth Amendment was ratified in 1868, Congress relied on so-called “ironclad oaths” to bar Confederates from office.[88] These oaths were sometimes narrowed by the judiciary.[89] In 1869, Justice Salmon Chase, riding circuit, answered an important question about the scope of Section 3 by holding that it was not “self-executing”: it required a congressional law to give it effect, and the courts would not act on their own to remove tainted officials.[90] Congress passed such a law, the Enforcement Act of 1870, allowing federal prosecutors to bring quo warranto actions against state officials covered by Section 3. The law stated that those actions would take priority on federal dockets and fined ineligible officials who took office knowing they were barred from doing so.[91] Subsequently, the Grant administration used the Enforcement Act of 1870 and related statutes against a number of officials, including three members of the Tennessee Supreme Court and the Tennessee Attorney General.[92]

Yet the half-life of disqualifications under this act was fleeting. Almost immediately, Congress began passing many bills to exempt thousands of former Confederates from being disqualified.[93] In 1872, as the Ku Klux Klan was conducting a campaign of racial terror across the South, Congress passed a sweeping amnesty law that lifted most war-related disabilities, except for those who had held a few very high offices before the Civil War.[94] The proponents of the 1872 amnesty law argued that disqualification had not fulfilled its goals and stymied attempts to tie the amnesty law to new civil rights legislation.[95] The ironclad oath laws were also repealed.[96] This broad amnesty legislation was in sync with Congress’s broader retreat from Reconstruction from 1877 onward.[97]

The largely dormant and “vestigial” Section 3, as we have discussed, gained new life after the January 6, 2021, storming of the U.S. Capitol and the ensuing impeachment of former President Trump. Many commentators called for the use of Section 3 as an alternative to disqualification by impeachment, particularly after it became clear that there were insufficient votes for a conviction in the Senate impeachment trial.[98] What made the Section 3 mechanism appealing was its ability to achieve disqualification via radically different channels from those of impeachment.

2.     Comparison to Impeachment

Section 3, like impeachment, aims to defend and reinforce a larger democratic order. But whereas impeachment is a prophylaxis against threats internal to the governing apparatus, Section 3 supplies a response to the problem of external threats to the federal government. Hence, Section 3’s disqualification rule differs in important ways from disqualification in the wake of impeachment. Consider four key differences between these two paths to disqualification.

First, Section 3 likely applies to a narrower range of conduct than impeachment but to a broader range of officials. The terms “insurrection or rebellion” and “aid and comfort” are likely more targeted than the “Treason, Bribery, or other high Crimes and Misdemeanors” triggers for impeachment.[99] Even on a minimalist reading, the substantive range of offenses in the Impeachment Clause meriting conviction would extend to a larger set of crimes than Section 3.[100] Historical practice supports this intuition. The first insurrection statute was enacted in 1790.[101] In it, the language of “aid and comfort” tracks just one kind of treason identified in Article III.[102] The 1790 statutory language may in particular be limited to instances of disloyalty in favor of an enemy state.[103] Given the context of its adoption, Section 3 is best read to apply whenever an enemy of “the Constitution” is given aid and comfort, whether or not an enemy state is implicated. Unlike impeachment, however, Section 3 cannot plausibly be limited to a subset of federal officials. It covers a large swath of federal and state officers alike.

Second, Section 3 is categorical, whereas impeachment is individualized. It contemplates disqualifying all officials who fall within the prohibited category of having engaged in insurrection or rebellion against the United States. As such, it applies to any person who has previously taken an oath to support the United States, whether as a federal or state official, with the necessary relationship to the insurrection. There is also no plausible argument that, like the impeachment power, it is triggered solely by an act covered by a criminal statute. Notwithstanding the narrow concern of its drafters with Reconstruction,[104] Section 3 is also framed in general and prospective terms. It applies not solely to the Confederacy but to any subsequent insurrection. It contains no sunset clause and remains part of the constitutional text. Depending on how it is used, however, it is either an “enduring expression of the historical politics that shaped the identity of the American Union”[105] or a “vestigial portion . . . of the constitution”[106] otherwise in desuetude. Further, unlike impeachment, it sweeps beyond incumbent officials. Rather, it suffices that a person has (a) at one point in time taken a covered oath to the Constitution and (b) engaged in a prohibited act, whether or not they were in office at the time.

Third, Section 3 lacks even a scintilla of detail concerning the process of disqualification, particularly with respect to executive branch officials.[107] Even though the impeachment provisions leave many questions unresolved, they at least identify the relevant decision-maker, as well as a substantive standard.

One possibility is that Section 3’s application is automatic: it requires no prior criminal process or conviction because it operates directly on persons. This was the theory deployed in September 2022 when a New Mexico state judge removed county commissioner Couy Griffin from office for his participation in the January 6 insurrection.[108] The judge disqualified Griffin from holding future office. While this interpretation is in harmony with the original operation of Section 3, it may raise due process and perhaps bill of attainder concerns (at least where the disqualifying body is legislative in nature).

An alternative construction gives Congress authority to determine how Section 3 is enforced. Section 5 of the Fourteenth Amendment vests Congress with authority to enforce “by appropriate legislation, the provisions of this article.”[109] On its face, this gives Congress discretion over the enforcement of Section 3. Although the Supreme Court has narrowed Congress’s enforcement discretion with respect to the identification of rights under Section 1 of the Fourteenth Amendment,[110] Congress still seems to have sufficient leeway to opt for either some regularized mechanism for Section 3 disqualification or, alternatively, a more ad hoc approach by which Congress itself votes on specific disqualifications. But, as we have observed, the latter reading is in tension with Article I’s prohibition on bills of attainder.[111]

Assuming that Section 3 disqualification does require a congressional enactment laying out the overarching process, it is nevertheless procedurally distinct from impeachment. There is, for example, no requirement that individual decisions about disqualification must run through the national Congress. The Griffin case suggests that state courts can operate as an alternative venue. And there is no good reason for thinking that the voting rule must be super-majoritarian. Congress could plausibly create a completely judicial procedure initiated by a U.S. attorney or other official. Alternatively, it could create an administrative structure akin to denaturalization, in which an agency subject to post hoc judicial review makes the decision.[112]

Fourth, the consequences of Section 3 are more extensive than the consequence of disqualification by impeachment. Under Section 3, disqualification attaches to both federal and state offices. Federal impeachment can yield a bar to holding any “office of honor, trust or profit under the United States,” but it cannot preclude future state office holding. Section 3 has a broader set of consequences than impeachment.

Nevertheless, the invocation of Section 3 in anticipation of and response to a failed presidential impeachment underscores the possibility that Section 3 disqualification might be invoked in some instances as a substitute for disqualification via impeachment. This constitutional redundancy is salient because of the supermajority voting rule for conviction after impeachments. In periods of strong partisan polarization, such as our own, this voting threshold will often be difficult for a legislative chamber to reach. Where it applies, Section 3 does not require the identification of a criminal prohibition. In addition, Section 3 disqualification makes a larger swath of the nation’s political offices unavailable to the targeted individual or group. To understand the scope of constitutional disqualification, therefore, it is at a minimum necessary to hold both mechanisms in view simultaneously.

C.    Term Limits

In October 2017, as President Barack Obama was launching into a speech on behalf of a gubernatorial candidate in New Jersey, the audience began chanting “four more years.” President Obama replied, “I will refer you both to the Constitution as well as to Michelle Obama to explain why that will not happen.”[113] The Twenty-Second Amendment to the U.S. Constitution, indeed, states that “[n]o person shall be elected to the office of the President more than twice.”[114] As conventionally glossed, the Twenty-Second Amendment means that any president who has served for two full terms is thereafter subject to a permanent ban on again holding the presidency.

The U.S Constitution’s presidential term limit is singular. Other federal officials, including members of Congress and federal judges, cannot constitutionally be subject to a term limit. The Court invalidated a state law imposing term limits on members of Congress on the ground that it impermissibly added qualifications beyond those found in the text of Article I.[115] Judicial term limits have also been proposed but are considered constitutionally questionable because of their conflict with textual protections from removal.[116] In contrast, many states impose term limits by legislation or constitutional provision on their state legislatures (fifteen) and governors (thirty-seven).[117]

For most of U.S. history, there were no explicit term limits even on presidents. In Federalist No. 72, Alexander Hamilton argued against presidential term limits. He contended that indefinite reelection would be advantageous since it allowed an experienced president to deploy his or her expertise, and that desire for reelection would help to ensure good behavior lacking in a lame duck.[118] Notwithstanding Hamilton’s views, George Washington initiated a norm that presidents would serve no more than two terms in office.[119] After Franklin D. Roosevelt swept past that informal norm in the 1940 (and 1944) election, however, Congress and the states acted to formalize a two-term limit through passage of the Twenty-Second Amendment.[120] Hamilton’s intuition was thus decisively rejected.

In resorting to a two-term limit on presidents, the United States is in good company. An overwhelming majority of presidential and semi-presidential systems around the world now have presidential term limits.[121] The U.S. approach of two terms followed by an absolute bar on reelection is the modal choice worldwide.[122] In contrast, the United States’s history of compliance with the presidential term limit is somewhat unusual. Globally, attempts at evading term limits are quite common and often succeed, even though such limits are entrenched into constitutional text.[123] This comparative experience raises the question of whether the Twenty-Second Amendment would be an effective constraint on a White House incumbent determined to serve more than eight years by either legalistic sleight of hand or brute force. Perhaps the Amendment’s clear text should not be taken as foreclosing all efforts to obtain a third term in the White House.

Obviously, term limits produce a different kind of disqualification from either impeachment or Section 3. To begin with, both impeachment and Section 3 have a negative connotation. But term limits, unlike impeachments, constrain politicians who are popular enough that they might otherwise continue in office. They punish the successful, not the culpable. Unlike impeachment and Section 3, which are backward looking, term limits are a forward-looking prophylaxis. Moreover, the Twenty-Second Amendment draws the class of affected persons far more narrowly than the other disqualification provisions: a presidential term limit will reach no more than a handful of people at any one time, and there is little uncertainty about their identities.

Yet like Section 3 and impeachment, the Twenty-Second Amendment safeguards the democratic character of the polity. It is a shield against an incumbent president entrenching themselves beyond democratic correction. A second similarity is the absence of a clear remedial pathway. Like Section 3, the Twenty-Second Amendment sets out no process for enforcement. It says nothing about what follows from a flagrant violation. At least until now, the Amendment has been self-enforcing. Presidents who won two terms, like Reagan, Clinton, and Obama, have not tried to find workarounds. But what if a two-term president simply refused to leave office? Would it make a difference if that person had won in the Electoral College? Could a federal court enjoin them from taking the oath of office? Neither text nor history provides answers.

One plausible view is that “the Supreme Court would [not] invalidate the election of a candidate to a third term even now, with the Amendment on the books. The Court might regard the validity of the election as a political question to be determined through the electoral process.”[124] But if it is not the Court’s role to enforce the Twenty-Second Amendment, then whose responsibility is it? The absence of past circumvention efforts means that it is hard to know now which institution—the Electoral College, state legislatures, or the certifying Congress—would have the job of resisting an attempted violation of the Twenty-Second Amendment. The mere fact that there are multiple institutions that could play this role, apart from the courts, creates confusion and uncertainty. It is possible to imagine coordination problems derailing enforcement efforts—even when the political will exists to act.

A further difficulty arises because the circumstances of such an attempt might vary quite dramatically from one case to another. In particular, it is possible to imagine such attempts having the imprimatur of an election victory or being in derogation of the apparent result.[125] Thus, it is hard to generalize about the political circumstances in which presidential term limits violations could play out—and hence the best institutional response.

D.    Legislative Exclusion and Expulsion

The impeachment and term limit provisions apply to the federal executive branch alone. Section 3 applies to both the President and legislators. There is also a set of exclusionary possibilities that pertains solely to legislators. Again, these operate alongside—perhaps as substitutes, perhaps as complements to—the powers created by impeachment, Section 3, and term limits.

Two relevant powers merit attention here: the power to exclude elected legislators and the power to expel legislators from the congressional body once seated. First, each House of Congress has power to act as “judge” of its members’ “qualifications” as a means to exclude candidate members at the threshold.[126] Legislative qualifications are “defined and fixed in the Constitution, and are unalterable by the legislature.”[127] In 1965, the Supreme Court enjoined an effort by the House to prevent Representative Adam Clayton Powell from being seated on the basis of alleged criminal conduct.[128] Powell v. McCormack holds that Congress cannot supplement the Constitution’s list of prerequisites for legislators as a means of disqualifying those selected by the voters.[129] The matter of Julian Bond, in contrast, hedges the power of state legislators to decline to seat members by invoking the First Amendment.[130] Both cases concerned majority-White chambers’ efforts to keep African American representatives from being seated in a moment of civil rights activism and public turmoil. Both, therefore, are properly read as reflecting the Warren Court’s awareness of the way in which racial dynamics warped facially neutral governmental processes. Powell and Bond are relevant here because they establish limits to federal and state legislatures’ powers to refuse to seat members. Powell is based on legal ground internal to the constitutional design of federal legislative powers. Bond, in contrast, concerns constraints imposed from outside the state constitutional text by the First Amendment. Both also attest to the willingness of the courts to step in and police the wielding of legislative powers to exclude an elected representative.

Second, both the House and Senate have a separate power to “expel” their own members by a two-thirds vote.[131] The Constitution, however, does not textually link this expulsion power to a member’s qualifications. Nor does it explicate the grounds for expulsion. Many successful expulsions occurred during the Civil War to remove pro-Confederate members.[132] But these exercises of the expulsion power have not generated a clear precedential legacy. Unlike the exclusion power, the expulsion power does not have a clear scope.

Despite these differences, the exclusion and the expulsion powers are linked. After Powell, a federal legislative chamber must specify up front whether it is voting for exclusion or expulsion: having teed up an exclusion vote (which requires the identification of a constitutional ground upon which to penalize an elected member), the chamber cannot switch gears and claim it was expelling.[133] As a result, exclusion by supermajority vote is not a freely available substitute for expulsion. In effect, a chamber must turn square corners if it wants to defy the voters’ will in selecting a particular person.

The powers of congressional exclusion and expulsion interact in rich and ambiguous ways with the impeachment and Section 3 powers. Expulsion is an obvious means to apply Section 3’s prohibition to insurrectionary legislators. Under Powell, a legislator who had participated in an uprising against the Constitution could not be denied their seat based on their past involvement in an insurrection as a threshold matter. But they could be expelled later by a supermajority vote, once having been seated. This raises an intriguing, and hardly implausible, possibility that a legislator might be seated by necessary force of law, with a vote to expel failing when it achieves more than a majority but less than a supermajority. The higher vote threshold for legislative exclusion on grounds of insurrection implicit in this scheme also has implications for the process that might be employed for executive officials: if a supermajority rule is constitutionally mandated for insurrectionary legislators, why should a lower voting threshold be used for those in the executive branch? That is, notwithstanding the breadth of Congress’s Section 5 power to implement the Fourteenth Amendment’s terms, could the details of Article I offer an implicit procedural floor for Article II in terms of the voting rule?

E.     The Constitutional Matrix of Democratic Disqualification

This Part has identified and explored five intersecting pathways of democratic disqualification created in the U.S. Constitution: impeachment, Section 3 of the Fourteenth Amendment, presidential term limits, legislative exclusion, and legislative expulsion. These mechanisms generate a range of pathways by which individuals can be excluded from democratic office, and Section 3 even adds a possible group-based disqualification.

Despite this rich constitutional framework, the law of democratic disqualification is full of gaps. Impeachments, exclusions, and expulsions are all rare events in American history. Whatever the law may be, the moral and rhetorical force of an individual’s claim to remain in office based on an electoral victory seems to have had a powerful buffering effect. The net result is a paucity of “historical gloss” that could clarify ambiguity.[134] In particular, the procedural channels by which disqualification occurs remain highly uncertain. In the context of the second Trump impeachment and other responses to the January 6, 2021, Capitol riot, this uncertainty may have had meaningful consequences. The lack of clarity may have thwarted potential responses to those events. Inaction begets uncertainty, and this in turn begets inaction.

The U.S. law of democratic disqualification, in short, remains incomplete. It cries out for supplement and context through broader study. With that in mind, the next Part widens the analytic lens to develop a more general theory of disqualification in other democracies.

III. The Comparative Law of Democratic Disqualification

Disqualification is as old as democracy itself. The Cleisthenic democracy of fifth-century Athens had a procedure called ostraka.[135] Once a year, the Assembly was asked whether it wished to conduct an ostracism. If the Assembly assented, votes would be cast by placing an ostraka, or potsherd, scratched with the name of one citizen who would be exiled from the polis for ten years (but not deprived of property or name). The results were not always encouraging. One day, a man called Aristides the Just, having commanded navies against the Persian host, was stopped by an illiterate fellow citizen and asked to help spell the name “Aristides” for an ostraka. When the puzzled general Aristides asked why, the aggrieved citizen explained that “I’m sick of hearing him called ‘the just.’”[136]

Today, disqualification is employed with (we hope) a bit more wisdom by a number of democratic polities to safeguard democracy. This Part analyzes this comparative experience to generate a more comprehensive account of disqualification’s possibilities and to situate the American experience.

We begin by offering a simple typology of disqualification strategies. We organize this typology along two main design choices—a group versus an individual approach, and an approach focused on past actions versus future threats. With this simple framework in hand, we identify and discuss the four permutations (in effect mapping out a two-by-two matrix of design options). This exercise yields a simple mapping of the various costs and benefits associated with disqualification.

A.     Mapping Disqualification Rules

Disqualification rules are ubiquitous in modern democracies. But they come in different forms. This Part maps out the major kinds of disqualification rules found in modern constitutions. Our aim is to highlight the richness of the field and the different ways constitutional democracies have grappled with similar problems.

Table 1 provides a basic typology of disqualification-related rules found in diverse constitutional settings. It is organized along two design margins. First, disqualification rules can operate either on the group level or on the individual level. That is, they can either disqualify actors en masse because of membership in a certain party or affiliation with a discredited regime, or they can work in a granular way, focusing on the conduct and characteristics of the individual actor at issue. The U.S. Constitution mainly operates at the individual level with the exception of Section 3. But the U.S. example, as we show below, also usefully demonstrates the possibility of ambiguity across the group/individual boundary.

Second, disqualification rules can be backward looking, focusing on the prior acts of an individual or group, or forward looking, seeking to identify organizations or actors that pose ongoing and serious threats to constitutional stability. This distinction between punitive and preventive functions, of course, is imperfect: even where a measure is forward looking, its application will nevertheless turn on past actions in most cases. But the latter are important because they predict future anti-democratic acts. The distinction, familiar from the criminal law, is a useful experiment.

We consider each of these categories in turn. By drawing on the deeper well of comparative experience, we can identify empirical regularities in the way each disqualification design operates. We can hence compare the normative implications of foreign designs with the U.S. experience. We further recognize other important design decisions not captured in this matrix. To supplement our analysis, we flag three other design margins worth considering: substantive scope (the basis for disqualification), temporal reach (how long disqualification lasts), and mechanism (who decides on disqualification and under what decision rule). The distinctive costs and benefits of democratic disqualification in each case shape the way in which the choices along these various design margins play out in practice.

B.    Lustration: Group-Based, Backward-Looking Disqualification

Many transitional democracies have some kind of lustration rule “screening,” “barring,” or even removing candidates from “public office” based on their association with a prior regime,[137] characterized as “tainted” in some fashion.[138] By nature, lustration is often an instrument of political transitional justice. Appropriate in extraordinary circumstances such as the wake of civil wars or authoritarian regimes, it is harder to justify during times of “normality” given rule-of-law and due process objections.[139]

Lustration is closely associated with the transition from communism after the Iron Curtain fell. In the post-Cold War Czech Republic, for example, some fifteen thousand individuals were removed or barred from public office.[140] In the eastern portion of reunified Germany, lustration under reunification treaty provisions resulted in some 54,926 people being removed or barred from office.[141] Lustration has also been used in post-invasion Iraq and elsewhere.[142] The obvious domestic U.S. analogue to these measures is Section 3, targeting Confederate supporters.[143] Unlike the American example, however, lustration programs elsewhere have been implemented in concerted and sustained ways. That experience provides a clearer perspective on the practical and normative questions that ensue.

The most studied and thus best-understood lustration policies are those that were deployed in post-Communist Eastern Europe in the 1990s.[144] Despite the many similarities among countries in the former eastern bloc, these nations adopted diverse approaches to Communist collaborators and officials. In some, such as Czechoslovakia (and later in the Czech Republic) and the Baltic states, robust lustration measures disqualified those who had carried out certain functions within the old regime, such as informing for the secret police, from some public positions within the new order.[145] Other countries adopted less draconian measures. The newly unified Germany, for example, did not automatically disqualify those who had served as secret informants for the former East German regime. Instead, it subjected those informants to a process that evaluated their fitness to serve on a case-by-case basis.[146] Similarly, Hungary and Romania did not disqualify officials en masse. Officials instead had a choice of resigning and keeping records secret or staying in their positions and having them released.[147] In Poland, officials were asked to confess whether they cooperated with the security apparatus of the old regime. Only those who made false denials were disqualified.[148] Bulgaria and Albania opted for limited or no lustration. Bulgaria mandated disclosure of the names of those who worked for the previous regime, only to see this measure invalidated in court.[149] Considerable debate rages as to which of these approaches was most effective at achieving goals such as building trust in the government, promoting social reconciliation, or preventing the “demoral[ization]” that ensues when those in power under an old regime remain in office despite complicity with past wrongs.[150]

Regardless of these differences in approach, the design of lustration mechanisms implicates common challenges. The first design challenge encompasses deciding the range of activities under the old regime that lead to disqualification and the range of positions from which a person is disqualified. In practice, legislatures tend to draw such boundaries narrowly, sometimes increasingly so over time. At other times, courts have further narrowed the scope of disqualification. Respecting those who have been listed as collaborators in the files of secret police forces, for example, courts and administrative actors have often found that merely being named on a list is not enough for disqualification. There also needs to be evidence of a person’s activities on behalf of the regime.[151] Further, regimes often limit the range of posts from which a lustrated official is disqualified to include only certain high positions. As a result, even under more aggressive regimes such as that of the Czech Republic, the actual number of barred officials appears to be (relatively) small in relation to the potential set of former officials.[152]

A major driver of the relative narrowness of lustration measures is practical. Successor regimes often face shortages of qualified personnel.[153] So while lustration may send an important signal about the moral commitments of a new regime,[154] overly broad lustration risks seeming vindictive and may even hinder reconciliation. The U.S. “de-Baathification” policy in Iraq after the 2003 war has become a symbol of the dangers of casting too broad a net.[155] Initially, the United States disqualified all members of the previous ruling Baath party without inquiring into individuals’ motives or activity. Later, the occupying coalition switched to a more discretionary model, but damage had already been done.[156] Iraq lost much of its capacity across the security services and other areas of the state. De-Baathification inevitably widened key cleavages in Iraqi society.[157] This helped to fuel an insurgency in the years after the initial U.S. invasion.[158]

A second, related design challenge is temporal: how long should lustration endure? In some cases, the temporal scope is explicit. Bulgaria, for example, passed a lustration law imposing a five-year ban on certain kinds of Communist collaborators from holding core offices. In Lithuania, the period covered by the lustration law was ten years.[159] But despite the use of explicit temporal endpoints, regimes seem to struggle to contain the reach of lustration. In the Czech Republic, lustration rules were initially put in place with a time limit, but they have been continuously extended.[160] According to one perceptive observer, the country’s exclusionary approach to lustration created a culture where social actors became accustomed to such exclusions as an ordinary part of public administration.[161] In other words, the program became quasi-permanent well after the bureaucracy itself had been effectively cleansed of the worst holdovers from the ancien régime. Across Eastern Europe, lustration and the communist legacy shape political debate to this day. This is in striking contrast to the United States, where federal authorities quickly ended disqualification of most former Confederates, closing further debate.

A different dynamic unfolded in Poland. Lustration was applied later and with less determination. Used to disqualify only officials who falsely stated that they had not been collaborators, the lustration rule was further narrowed by the Polish Constitutional Court and the ordinary judiciary to ensure compliance with due process norms.[162] These judicial interventions in turn fed a political reaction, associated with the Law and Justice Party, for more aggressive action against former Communist officials. The current Law and Justice Party-led administration, which is viewed as driving democratic erosion in Poland, has made anti-communism into a rallying cry and a justification to “purge” the judiciary and other institutions alleged to still be “tainted.” The government defended recent moves weakening judicial independence, for example, in part as necessary “anti-communist” measures. The government justified its actions by saying that Poland previously “lack[ed] a practical method to hold to account judges who were directly and shamefully involved with the communist system” and that this failure “negatively affects the public trust in the judiciary—and thus the rule of law itself.”[163]

A third design challenge concerns the institutions charged with enforcing lustration. Jurisdictions work through both administrative officials and judges to apply either unified or variegated standards. In Germany and the Czech Republic, officials conducted initial screenings after consulting state security service files, with their decisions subject to ex post judicial review.[164] In Germany, courts hence often reversed dismissals, narrowing the scope of acts for which someone could be disqualified.[165] Different German states, however, applied widely different standards, with some (such as Saxony) taking an aggressive approach while others were far more restrained.[166] In Poland, by contrast, a special lustration court, modeled on criminal adjudication in many respects, had initial responsibility for disqualification.[167] After criticism that this tribunal was unduly slow and too favorable to the accused, the Polish Parliament in 2007 moved the primary responsibility for lustration to an administrative body, the Institute of National Remembrance, with more limited possibilities of judicial review and a broader remit.[168] This new body’s proponents argued that the change would accelerate lustration. Meanwhile, critics complained of its politicization and a consequent loss of due process.[169]

While the sheer diversity of lustration designs belies any easy inference about optimal design, certain conclusions do emerge from this survey. First, like disqualification mechanisms in the United States, lustration often proves narrower than its formal scope might suggest. Practical and political concerns limit its operation, even after a fairly broad disqualification measure has been enacted. Second, where lustration has been used, as in Iraq, it has interacted with ongoing political fissures in socially and politically damaging ways. Lustration seems to advance its goal of facilitating transitions without having unintended spillover effects only when it is narrowly focused. Finally, the tendency of lustration regimes to linger past their intended termination points to a risk that what is initially intended to be tightly circumscribed to a transitional period may end up leaching out into “ordinary” politics. Transitional mechanisms can help ensure that legacy officials are not too “tainted” by association with the old regime. Tightly limited substantive standards and sunset mechanisms will work best to achieve this end while minimizing spillover into ordinary politics after the transition.

C.     Militant Democracy/Party Bans: Group-Based, Forward-Looking Disqualification

German jurist and refugee from Nazi rule Karl Loewenstein coined the term “militant democracy” shortly before World War II. He aimed to capture the idea that a democratic system might take proactive steps to thwart anti-democratic actors from seizing power from within.[170] The term has come to encompass “the use of legal restrictions on political expression and participation to curb extremist actors in democratic regimes.”[171] Nazi use of constitutional tools such as legislative prorogation (i.e., shuttering a parliament before its expected end) and rule by emergency decree loomed large for scholars such as Loewenstein. His ideas went on to influence constitutional designers in Germany and beyond.[172] Today, militant democracy’s most important institutional form is the ban on anti-democratic parties. This has been deployed at various times in Germany, Finland, Czechoslovakia, South Korea, France, Spain, and the United Kingdom.[173] Although not formally a bar against specific persons’ participation in politics, a party ban is often a de facto disqualification of known individuals closely identified with the party.

We begin by outlining core historical experiences with this instrument, then drawing out the ways in which its design illuminates both the costs and benefits of disqualification. The primus inter pares of militant democratic constitutional provisions is Article 21 of the 1945 Basic Law promulgated in West Germany after World War II. Pursuant to this provision, political “[p]arties that, by reason of their aims or the behavior of their adherents, seek to undermine or abolish the free democratic basic order or to endanger the existence of the Federal Republic of Germany shall be unconstitutional.”[174] Further, all parties’ “internal organization must conform to democratic principles” and use of their funds must also be transparent.[175]

The German Constitutional Court has power to enforce Article 21.[176] In 1951, the German government asked the Constitutional Court to ban both the Socialist Reich Party and the Communist Party. In the Socialist Reich Party case of 1952, the Court acted quickly and with relative ease.[177] It found that the party’s platform leaned heavily on former Nazi ideas and imagery, that the party recruited unrepentant former Nazis to fill its ranks, and that it was organized in a top-down, undemocratic manner. In essence, the Court found the party was an attempt to revive a totalitarian, Nazi-like regime in form and substance. The Court had more difficulty in the Communist Party case, which was handed down four years later. It ultimately upheld the ban in a detailed opinion.[178] The Court focused on the ideological goals of the party. It held that the Communist Party’s ultimate ambition for a dictatorship of the proletariat was inconsistent with the fundamental values of the German Basic Law, including human dignity and multi-party democracy, even when pursued via constitutional means. Lurking in the background of the decision, of course, was the security threat posed by then-Communist East Germany.

After these decisions, Article 21’s party-ban mechanism fell into disuse. The only cases filed in recent years have been two attempts to ban a new far-right party, the National Democratic Party (NDP). The first effort to do so was overturned in 2003 on procedural grounds.[179] In the second in 2017, the Court reached the merits of the disqualification and denied the government’s petition.[180] It held that the NDP indeed had an anti-constitutional platform, which aimed to create an exclusive, ethnic-based nation-state that rejected liberal democratic values. Even though the platform was anti-constitutional, the Court concluded that there was no need for a ban because the prospects of the party achieving its aims were slight. In so doing, it rejected its earlier view in the Communist Party case that a party need not have a realistic near-term prospect of achieving its aims to be banned.

The NDP holding has potential ramifications for any extension of Article 21’s prohibitory scope. In March 2020, the Federal Office for the Protection of the Constitution (BfV) classified the Alternative for Germany (AfD) party as “not compatible with the Constitution.”[181] A year later, AfD was placed under electronic surveillance by the BfV.[182] The agency’s underlying report explained that a “substantial portion of the party . . . seeks to awaken or strengthen a fundamental rejection of the German government and all other parties and their representatives.”[183] Many used “Nazi-era insults like ‘Volksverräter’ when discussing their political competitors, a racially laden term which essentially translates to ‘traitor to your own people.’”[184] In light of this evidence, and given the AfD’s consistent resistance to moderation, an exercise of Article 21 power was seen as a feasible subsequent step.[185] The NDP case’s predicate of “potentiality,” moreover, would be far easier to satisfy with respect to the AfD, which became the largest opposition party in the Bundestag in 2017.

From its German roots, the militant-democratic idea of party bans has now diffused around the world.[186] Some 29 percent of constitutional courts have the ability to adjudicate the legality or constitutionality of political parties.[187] Party bans have recently been imposed across various regions and contexts, ranging from Spain and Turkey to Israel and South Korea.[188] For example, a South Korean court in 2014 disqualified the small left-wing United Progressive Party. The court cited the party’s alleged links with North Korea, at the behest of former President Park Geun-Hye, after affiliates were arrested for an alleged plot with North Korea.[189] The idea of party bans has also gained recognition under international law.[190] The modern U.S. approach, where party bans stand in clear tension with the First Amendment, is an exception.[191]

Creating a militant democracy, like a lustration regime, entails a set of design decisions about its scope and manner of operation. The choices for calibrating these design margins will rest on the costs and benefits of party-focused disqualification. Article 21 of the German Basic Law and many other militant democratic provisions are durable rather than transitional. On the one hand, such provisions should be rationed like lustration rules and cabined to transitional periods in which democracy is “fragile.”[192]

Party bans present many of the same design choices as lustration. A first question, as with lustration, is scope: what should the substantive trigger be for a prohibition, and who should it affect? This first question is relatively straightforward in a democracy’s early years when the “anti-canonical” party is likely easily identified based on the nature of the preceding regime. This explains the ease of the 1952 Socialist Reich Party decision. Over time, the question of how to identify a threat to democracy—as opposed to the manifestation of a substantial portion of public opinion with a legitimate claim to voice—becomes much more difficult. Exacerbating the problem, modern authoritarian parties tend to have largely ambiguous ideologies.[193] Their aims are not flagrantly anti-democratic. Instead, they seek to win power through democratic means but then use their power to consolidate their position, effectively tilting the electoral playing field in their favor. The relatively “mainstream” nature of modern authoritarian threats makes them challenging to ban and raises the prospect of errors.

Reflecting this difficulty, the 2014 South Korean court opinion on the United Progressive Party was criticized for incorrectly applying the proportionality analysis by questioning whether the party’s alleged aims were realistically achievable or sufficiently “imminent.”[194] There are two overlapping rationales for such a risk requirement.[195] First, banning a party unlikely to inflict damage is not worth it, and second, a ban might make the situation worse, for example, by driving the party’s supporters and financing networks underground.[196] It is possible that the emergence of terrorist groups in Germany like the Red Army Faction might be explained in terms of bans on socialist politics, although the connection is hard to prove with certainty.

Any future adjudication in Germany about the disqualification of the AfD party will also be fraught with difficulty. It may well be both that AfD reflects the preferences of a substantial minority of the German electorate and that it disdains and will predictably act against the systemic institutional foundations of electoral democracy. Either banning or permitting AfD’s operation would impose substantial strains on democratic institutions. The first would burden the “legitimate interests in participation” that even anti-democratically inclined citizens possess.[197] The second might lead to the AfD capturing political power and then shutting down democratic competition. Deciding whether banning or permitting AfD is more costly places immense epistemic burdens on the BfV and the German Constitutional Court. The uncertainty and difficulty attendant to the decision make it much easier for officials acting in bad faith to misuse the disqualification authority in favor of a party ban.

Next, there is the question of temporal scope. Militant democracy provisions are generally part of the permanent constitutional order and are not transitional. But the risk of their abuse might increase as time elapses because the merits of its application to a given set of facts will become harder to settle. Yet, the threat to democratic order may not necessarily decline over time. A convergence of political dynamics and structural constitutional features might provoke unexpected recursions of the threat to democracy after a constitutional order has matured.[198] An important question is how courts and other actors should treat these provisions as time goes on. Should party bans be read more restrictively to account for increasing regime stability (as the German Constitutional Court has in effect done), or should they be given a static meaning that makes bad faith manipulation more difficult?

Further, just as with respect to lustration, there are questions of what the downstream consequences of party bans ought to be and how long they should last. In some countries, a ban on a party only covers the party itself and not the individuals comprising it. This is true in Germany, where the Constitutional Court has held that mere membership in a proscribed political party or organization is insufficient to ban someone from politics or the civil service altogether.[199] In South Korea, however, the Constitutional Court held that the effect of its party-ban decision was also to strip five legislators from the banned party of their seats.[200] It noted that while the Constitution itself was silent on this issue, failing to remove the legislators would mean that the Court had not protected the constitutional order.[201] In Turkey, when the Constitutional Court banned the Welfare Party in 1998, it similarly removed a party’s parliamentary members from their seats and banned them from founding or joining any other political party for five years.[202]

Finally, and again in parallel to the lustration context, there is a question of mechanism design: who exactly is to decide when a ban goes into effect? Perhaps because party banning is such a sensitive function, it is often left to constitutional courts instead of administrative actors or ordinary courts to make the final determination.[203] But there are contexts where critics have raised concerns that the state machinery is unfairly targeting minority groups, reflecting concerns of majoritarian tyranny. The most well-known examples of this dynamic stem from Turkey, where Kurdish parties have often been banned by the Constitutional Court.[204] Reviewing these bans, the European Court of Human Rights (ECHR) has reversed them on several occasions, finding them either unjustified or disproportionate. In 2010, for example, the ECHR ruled that a ban on the Kurdish People’s Democracy Party was inconsistent with the European Convention because the party was merely critical of the regime without actually advocating armed resistance or violence against it.[205]

Even in cases where party bans do not target discrete ethnic minorities or similar groups, there are still risks of misuse. In the extreme, a ban can be used to remove a pro-democratic party from the electoral arena, rather than inhibit an anti-democratic one. For example, in 2020, Thailand’s Constitutional Court disbanded, on relatively thin evidence, the broadly popular Future Forward Party, which largely represented urban and young voters.[206] An even more extreme case occurred in Cambodia in 2013, when the Constitutional Court, acting at the instigation of the ruling party, banned the upstart opposition Rescue Party on the fabricated ground that it was linked to foreign actors in the United States and somehow posed a threat to the “liberal multi-party democracy.”[207] Ironically, the decision’s effect was to sink the only opposition party, stripping it of all fifty-five of its seats and banning its leaders from politics for five years. In the subsequent election of 2018, the ruling party won all 125 seats in the National Assembly.[208]

The difficulty posed by each of these design decisions is compounded by the fact that militant democracy simply might not work in the long term. In Turkey, Islamist parties were banned numerous times in the 1970s, 1980s, and 1990s. They nonetheless managed to reconstitute after each ban. By the time the Welfare Party was banned in 1998, it was the single largest party in Parliament and its leader had already served as prime minister.[209] The Turkish court held that the party violated constitutional provisions on secularism because of its Islamic bent, and the ECHR subsequently upheld the prohibition.[210] Although much of its leadership was banned, affiliates founded the Justice and Development Party in 2001. After nearly being banned itself in 2008, the Justice and Development Party became Turkey’s dominant party. It has taken the country in an increasingly authoritarian direction.[211] The Turkish experience may suggest a limit on the ability of mass disqualification and militant democracy to repress anti-democratic movements with genuine popular support within a democracy. Designers can perhaps slow the emergence of such a movement—acting as a “speed bump”[212]—but will be hard-pressed in delaying it indefinitely given the durability of public support.

Concern about the efficacy of party bans has led constitutional designers to develop a set of “softer” mechanisms to address anti-democratic parties. In Germany, following the failure to ban the NDP, parliamentary leaders pushed through an amendment that added new sections to Article 21. In part, these new provisions allow the State to deny government funding to any parties that have an anti-democratic platform, regardless of the probability or imminence requirement undergirding the 2017 Constitutional Court decision.[213] Denial of public funding, which is especially crucial in Germany, may be a lesser but still potent alternative to a ban.[214] Israel illustrates another alternative approach. Its statutory framework for elections creates a gap between the legal dissolution of a party and a prohibition on its running for seats in the Knesset, the Israeli legislature. The first of these measures is subject to a more demanding standard than the second.[215] As a result, a political system may allow space for some parties organized around even “reprehensible” ideas to exist and recruit while still denying them the right to seek certain nationally elected offices.[216] This might allow for ideas to be ventilated in the public sphere without the risk of anti-democratic actors gaining a foothold in key democratic institutions as a way of building up their power.

D.    Impeachment and Other Forms of Individual, Backward-Looking Disqualification

Whereas lustration and militant-democracy party bans operate against groups, impeachment-linked disqualification rules are often applied in individual cases. Recent comparative studies of impeachment address its substantive scope and mechanism choices and provide baseline empirics of its substantive scope and procedural vehicles.[217] After briefly summarizing what is known about impeachment, we focus on a question previous work has not addressed: how is the choice to disqualify either linked to, or decoupled from, the mechanism for removal from office?

1.     Impeachment Regimes

Almost all (90 percent) of constitutions with a presidency speak to impeachment.[218] The substantive scope of impeachment varies widely. Crimes and constitutional violations supply the most common bases for removal of a president.[219] Although the procedural parameters of impeachment vary greatly among different national constitutions, it is typically a lower legislative chamber that begins an impeachment process by a supermajority vote.[220] Ex post judicial review is often, but not always, available.[221] As in the United States, successful impeachment globally is quite rare. One recent study found “between 1990 and 2018 . . . at least 210 proposals [to impeach] in 61 countries, against 128 different heads of state,” but only 10 successful removals.[222] The evidentiary basis for analyzing disqualification by impeachment is correspondingly thin.

Within this array of choices about the substantive, temporal, and procedural parameters of impeachment, we focus on the decision whether to link impeachment and disqualification and, if so, in what fashion. Unlike the relative convergence in comparative constitutional design in respect to impeachment, there is sharp disagreement over the connection between removal and disqualification of individual political malefactors. Disqualification is not itself always even a potential consequence of impeachment. Some constitutions limit the effects of impeachment to removal alone, leaving open the possibility of a subsequent run for office.[223] On the other extreme, other organic laws make disqualification an automatic consequence of impeachment alongside removal from office.[224] Still others follow the U.S. approach by allowing (but not requiring) the relevant institution to impose disqualification as a supplemental penalty for impeachment.[225] Finally, some constitutions are simply unclear about the relationship between impeachment, removal, and disqualification.

The Brazilian Constitution, for instance, permits the Senate to remove a convicted official from office and to deprive them of political rights, including the right to hold office for eight years. Yet the Brazilian Constitution is ambiguous on whether or not both consequences automatically apply upon conviction.[226] During the 2016 impeachment trial of President Dilma Rousseff, the Senate decided to vote separately on these two questions. The choice to hold two separate votes was endorsed by the President of the Supreme Federal Tribunal, who presided over the trial.[227] The Senate then voted to remove Rousseff by a sixty-one to twenty vote, comfortably exceeding the necessary two-thirds margin. But the subsequent disqualification vote was closer—forty-two to thirty-six—falling short of the needed two-thirds margin. In consequence, Rousseff remained free to run for future office, and indeed she ran unsuccessfully for a Senate seat in 2018.[228] The Rousseff example is a reminder of how the voting rule can be a dispositive factor.[229] It also suggests that the decision to split voting into separate removal and disqualification decisions can have outcome-determinative effects. That is, had legislators been forced to decide simultaneously on whether to remove and disqualify Rousseff, it is possible they would have reached a different outcome on either question.[230]

2.     Individual Disqualification Outside Impeachment

Where impeachment and disqualification are decoupled rather than bundled together in a single process, the question of disqualification may be channeled through administrative or judicial channels. In Brazil in June 2023, as we noted in the Introduction, former President Jair Bolsonaro was barred from seeking office until 2030 after that country’s Electoral Court held that he had made false claims about rampant electoral fraud to a group of foreign diplomats in the runup to his 2022 reelection loss.[231] Here, we offer detailed accounts of the specific non-legislative vehicles used in three regimes—Pakistan, Israel, and Colombia. In each of these cases, substitutes for impeachment have been much more often utilized than the relatively rare mechanism of impeachment. It is possible that this selection effect can be explained in terms of these countries’ pervasive distrust of political actors, as well as a highly active judiciary. However, it is also possible that the choice of institutional channel influences the rate of disqualification. We then introduce the more general possibility of hitching disqualification to a criminal conviction.

Under Section 7A of Basic Law: the Knesset, the Israeli legislature can prevent candidates from running for office if they engage in speech denying the Jewish and democratic nature of the State, incite racism, or support the armed struggle of a state or terrorist organization against Israel.[232] The same procedure is used for party and individual bans. Decisions are made by a central election committee comprising current legislators and a Supreme Court judge. In application, it has sometimes been deployed against far-right Jewish candidates who incite hatred against Arabs.[233]

Supplementing this channel, Israeli courts have adopted an aggressive program of removing officials and blocking appointments based on a judge-crafted concept of “good character.”[234] Israeli law on the books disallows officials from serving in office upon conviction, but the Supreme Court has gone further to deem officials ineligible from remaining in or holding office if they are under indictment.[235] More broadly, the Court has prohibited appointments within the government or military based on behavior showing poor character (such as sexual harassment) even when no criminal charges were formally filed.[236] In such decisions, the Court has relied on administrative and disciplinary records and media reports. Under its doctrine, disqualification need only meet an administrative law standard of “substantial evidence.”[237] Several recent petitions targeted media statements made by prospective appointees as potential grounds for disqualification.[238] Disqualification is becoming more frequent. There were eighteen cases in the 1990s and twenty-five in the 2000s, but already twenty-one in the first half of the 2010s alone.[239]

Similarly, the Pakistani Supreme Court has relied on Article 62(1)(f) of that country’s 1973 Constitution, which bars persons who are not “honest” and “righteous” from Parliament and Provincial Assemblies.[240] In 2017, the Court invoked this article to remove and disqualify then-Prime Minister Nawaz Sharif after his name was linked to overseas property holdings in the so-called Panama Papers.[241] The Court found Sharif to be unfit for office because he had failed to disclose an employment relationship and assets from a Dubai-based company on his nomination documents.[242] The Court subsequently held that Sharif could not lead the ruling Pakistan Muslim League party because a disqualified official cannot head a political party.[243] Other officials have been disqualified on similar grounds. In a subsequent decision reviewing petitions filed by seventeen former officials in 2018, the Court held that disqualification under Article 62(1)(f) constituted a permanent, lifetime bar from seeking office.[244]

Colombia uses an independent administrative actor, called the General Procurator, to make decisions on democratic suspensions, removals, and disqualifications.[245] The General Procurator can act following a wide range of allegations for “disciplinary faults,” including conflicts of interest, exceeding designated powers, and non-compliance with duties.[246] For example, in 2013, the General Procurator removed Bogotá Mayor Gustavo Petro from office for allegedly botching the reverse privatization of garbage collection in the city and banned him from seeking office for fifteen years.[247] The charges hinged on the allegation that Petro had used operators with insufficient experience to do the job and violated the legal rights of free enterprise.[248] While Petro was removed from office, the decision was appealed and eventually reversed by the Council of State in 2017 on the grounds that the General Procurator had shown insufficient evidence of bad intent.[249] In 2020, the Inter-American Court of Human Rights held that the General Procurator’s conduct and process impinged on Petro’s rights under a regional human rights accord.[250] This left Petro free to run for president in 2018, when he finished as the runner-up, and 2022, when he won the presidency.[251]

The General Procurator remains an enormously powerful figure, with removals and disqualifications continuing apace. Between July 1, 2019, and June 30, 2020, the General Procurator’s office took disciplinary action against 152 mayors, 113 councilors, and 1 governor, with about one-third of those being disqualified from seeking future office for some period of time.[252] Accounting also for unelected officials, the General Procurator’s disciplinary chamber issued 703 first-instance decisions imposing sanctions in that period.[253] In a 2023 decision, the Constitutional Court curtailed the Procurator’s powers over elected officials, holding that it could not remove or disqualify those officials without first receiving a favorable decision from an administrative court.[254]

Finally, consider the possibility of linking disqualification to any criminal conviction. As noted above, Israel compels convicted officials to resign by law, and in Pakistan, certain criminal convictions lead to a five-year ban from office.[255] More generally, some thirty-five countries (18 percent) today do not allow any person convicted of a felony to become head of state.[256] Fifteen of 134 (11 percent) that have a separate head of government have the same restriction. Today, fifty-one countries (28 percent) restrict people convicted of felonies from serving in the lower house of the legislature, while twenty-seven of eighty-five (32 percent) of those with an upper house have the same restriction.[257] In many U.S. states, people convicted of felonies cannot run for or hold elected office.[258] But the U.S. Constitution appears to prohibit such a restriction for federal office. The Supreme Court has held that Article I set forth the exclusive qualifications for becoming a member of Congress (and, presumably, this applies to Article II regarding presidential power).[259]

Of course, the criminal process channel can be abused for incumbency-protecting ends, which might potentially lead to the political capture of prosecutorial agencies.[260] Vladimir Putin’s Russia has become notorious for sidelining and discrediting political opponents by selective use of tax laws and other financial crimes.[261] Prominent opposition figure Alexei Nalvany, for instance, was disqualified from challenging Putin for the presidency based on an allegedly sham 2014 fraud conviction.[262]

3.     Implications

Individualized, retroactive disqualification occurs via an impressive array of institutional channels and is based on a varied set of substantive bases. Yet despite the heterogeneity of these individualized, backward-looking regimes, several conclusions concerning design choices and their implications can be drawn from this survey. In some contexts, they suggest the underappreciated value of using independent, non-legislative bodies, albeit designed to avoid political capture.

First, the most important design parameter is the mechanism for applying disqualification. Perhaps surprisingly, moving disqualification decisions outside elected bodies is correlated with increases in the rate of disqualification. Contrary to the concern that elected actors would use disqualification as an instrument for entrenchment, it seems that reposing a discrete and individualized disqualification power in the legislature generates a détente, or a fear-based equilibrium: the focused threat of retaliation against whomever opens the possibility of individualized removal induces all to avoid escalating the conflict. In contrast, non-elected actors labor under no parallel disincentive, dampening the active deployment of disqualification powers. Where disqualification is understood to be a tool for promoting “public trust in government,”[263] reliance upon judicial or administrative action rather than parliamentary actors may reduce the perception, and even the risk, of insider self-dealing. Since legislative disqualification is not a credible signal of commitment to the rule of law, it generates less value to the democratic system and, as such, is observed less often. It may well be a self-defeating constitutional design choice.

Second, the question of whether removal and disqualification should be subject to different voting rules remains undertheorized. Other constitutional systems resolve the issue in various ways, but, as the Rousseff impeachment demonstrates, the voting threshold and the decision to hold two votes instead of one can be highly consequential.[264] There are normative arguments cutting in both directions. On the one hand, if an expansive, political account of impeachment and removal as an appropriate response to political seizures is adopted, then the voting rule for disqualification should be more demanding than the one used for removal. After all, once an impeachment has resolved an immediate crisis, the need for enduring exclusion becomes less clear. Many of those who pose a serious threat while in office are later unelectable, so a lifetime exclusion may be unnecessary. While impeachment creates something like a general public good, disqualification imposes a concentrated cost on a single member of the polity. On the other hand, a high-threshold supermajority rule may in practice be impossible to meet given predictable partisan divisions over a disqualification decision. One solution to this dilemma may be to adopt a legislative impeachment rule and then—like Israel, Pakistan, and Colombia—a judicial or administrative disqualification protocol.

Third, notwithstanding this argument in their favor, non-legislative disqualification procedures remain controversial. The Israeli, Pakistani, and Colombian disqualification regimes have all been critiqued as nondemocratic interventions in the political process that thwart popular will.[265] In addition, the vagueness of the standards that are applied—“good character” in Israel and “honesty” and “righteousness” in Pakistan—has prompted concerns about lawlessness and instability. They are vulnerable to the argument that only democratic institutions should impinge on democratic choices. Yet provided that a disqualifying institution was initially established by democratic processes, there is reason to doubt the force of this objection. Concern about the ambiguity of substantive standards is also easily remedied. While textual vagueness may be an appropriate choice for the substantive standard for impeachment since it effectively delegates normative discretion to future legislatures, the case for vesting unelected bodies with any parallel discretion is weak.

E.     Term Limits: Individual, Forward-Looking Disqualification

A final option for disqualification is an ex ante and determinate criterion for ineligibility, which eliminates uncertainty over substance and procedure. The archetypal rule of this kind is the term limit.

Term limits prevent officials from entrenching themselves in office by categorically barring terms of more than a certain number of prescribed years. While there is considerable debate on their merits, all sides agree that such limits restrict the electoral choice of the polity, which would otherwise be free to reject or accept a candidate who previously served in office.[266] The justification for a categorical bar on specific individuals (as opposed to parties, which are free to continue winning elections with other candidates) turns on the large risk of personal entrenchment and corrupt enrichment via public office.[267] Incumbency advantage, earned or undeserved, may be so great that the most important predicates of democracy, such as ex post accountability and uncertainty of tenure in office, may be compromised in the absence of term limits.[268] The argument from entrenchment has the most force in relation to chief executives. As presidents serve for more terms, they become increasingly able to use formal and informal tools to gain control over additional parts of the state, including institutions like courts.[269] In other words, experience has shown that presidents in office for long periods of time are able to systematically tilt the electoral playing field increasingly in their favor using their institutional powers.

On the other hand, a term limit provision impedes prima facie legitimate democratic choice. More pragmatically, the possibility of reelection might also engender electoral accountability that produces better governance. Additional time in office allows officials to gain expertise and opportunities to carry out their policy programs.[270] Where a policy can only be executed in the long term, such as climate change mitigation, this may well be to a democracy’s advantage.

Legislative and presidential term limits present distinct issues. The case for legislative term limits rests on less certain ground. Legislative term limits may weaken the legislature as an institution, allowing presidents to tilt the separation of powers even more heavily in their favor.[271] Alternatively, they might solve a collective action problem among voters who seek to limit the scale of pork barreling while preserving their representative’s ideological convergence with the median voter.[272] These effects will net off in different ways under different empirical conditions. Perhaps as a result, term limits for legislators are uncommon but far from unheard of.[273] For example, fifteen U.S. states include term limits for members of their state legislatures.[274]

Presidential terms limits, in contrast, have now become “defining features of democracy.”[275] The vast majority of presidential or semi-presidential systems include a term limit for their presidents. The small proportion (12 percent) that do not carry term limits tend to be non-democracies, often because the term limit was removed at the behest of an autocratic chief executive.[276] Even for presidents, only a small percentage of systems (8 percent) prohibit all possibility of reelection.[277] The most common design provides for an absolute bar on any presidential reelection after two terms have been served. Some countries allow nonconsecutive terms, while others restrict officeholders to two terms in a lifetime.[278] A sizable number of systems include an alternative form of disqualification where presidents must leave office after serving either one or two terms, but only temporarily. They can return after sitting out for a set period of time, usually one term.[279] Chile offers an interesting recent example. Its last four presidencies have been held by two presidents from different sides of the political spectrum (Michelle Bachelet and Sebastián Piñera), each alternating service for one term. As with other forms of disqualification, term limits sometimes require only a temporary exit.

Unlike legislators, presidents seek to evade term limits and remain in office with considerable frequency. Two recent studies found that about one quarter of presidents attempted to change or circumvent term limits and remain in office through different methods such as formal constitutional amendment, judicial reinterpretation or removal, or reliance on controllable “shadow presidents.”[280] Most evasion attempts succeeded. Here, as elsewhere, making the disqualification effective may require some creative constitutional design. For example, use of temporary rather than permanent disqualification may make sense in this context, if only because presidents may be more willing to leave power if they think they can return down the line.[281] A temporary exclusion mitigates entrenchment risk without creating a perverse incentive to cling to power.

Term limits are not the only species of forward-looking individual disqualification, although they are the instrument most obviously addressed to concerns about democracy maintenance. Other constitutions have their own rules. Article 44 of the Australian Constitution, for example, addresses qualifications to run for parliament, including not holding foreign citizenship.[282] In 2017, seven sitting parliamentarians were found to be ineligible to run for the offices they held, in some cases because they held foreign citizenship through grandparents. The High Court confirmed their disqualification and ineligibility from future office.[283] Other constitutions have maximum age limits or mandatory retirement ages, after which officials must leave power, although these are less common than minimum ages (which serve as entry rather than exit conditions).[284] Retirement ages are common for judges both comparatively and within U.S. states, although less common for other officials.[285] Maximum age limits work in a similar way to term limits. They tend to ensure that officials will not remain in power for life and may create room for younger politicians. They also, presumably, respond to a set of biases that officials in certain positions over the specified age will be less likely to be able to carry out the job effectively or that their levels of performance will be too risky or unpredictable.

Finally, Iran’s unelected Council of Guardians vets all candidates for the presidency, Parliament, and certain other positions.[286] Relevant criteria include not only age, experience, and education, but also commitment to religious values and the principles of the Iranian Constitution. The latter criteria are non-transparent, and the decisions cannot be appealed. The vast majority of potential candidates are typically excluded. In the 2021 presidential election, for instance, over 600 candidates applied to run, but only seven were approved. Among those disqualified was a former two-term president, Mahmoud Ahmadinejad, and all female candidates.[287] In the 2020 parliamentary elections, about half of applicants were disqualified, including about one-third (or eighty) of incumbents.[288] The demerits of such a blatant system of political control need no elaboration.

In summary, there are substantial grounds for installing presidential term limits in a constitutional design. Indeed, it is almost de rigor in democratic constitutions. But there are more fragile reasons for using them in respect to legislators. The modal form of presidential term limit means that after a first term, presidents are lame ducks. While this has costs, it is plausible to think that the mechanism plays an important function in deterring a specific form of anti-democratic behavior.

F.     The Spectrum of Disqualification Rules

The constitutional design of democratic disqualification rules defies easy summary. To begin with, it divides into no less than four paths—call them lustration, party bans, impeachment, and term limits—that swell, narrow, and involute in unexpected ways. Sometimes they cut across each other; at other times, they run widely apart. All serve the same goal of securing democratic rule not just for the day, but for the longer term. Despite their importance, their heft, and the gravitational tug they exert on political life, they tend to be treated distinctly—as if they were all doing different things.

While all four vehicles for disqualification can work as bulwarks of democratic rule, some do so better than others. Three tentative conclusions, therefore, can be drawn from the available stock of comparative evidence without taking the evidence for more than it is worth.

First, comparative experience suggests that term limits, while imperiled by evasion efforts, provide a rough (if imperfect) prophylaxis against presidential entrenchment. Second, regarding individual misconduct, an institutional separation between removal and disqualification seemingly yields a more robust mechanism than legislative oversight. Situating a disqualification power outside the legislature implies risks of overbreadth and capture, but, at least in the small number of cases at hand, these concerns appear not to dominate. Third, group-based measures have a mixed track record: whereas lustration and party bans can be effective during transitions into democracy, they both raise difficult questions as they endure into a maturing democracy. Where a potentially prescribed party has considerable public support, such as in the cases of the AKP party in Turkey and the AfD in Germany, both the normative and the practical problems of applying a party ban (or indeed, a lustration rule) peak. Group-based measures, in short, may have efficiencies of scale, but these seem to diminish over time.

IV. Optimal Disqualification Rules

After analyzing the U.S. experience, comparative experience, and the relevant set of theoretical considerations at hand, we propose an “optimal” disqualification mechanism. We use the term optimal as a shorthand to capture the sense of design choices that are, all else being equal, more likely to promote rather than undermine democracy. At the same time, it would be a mistake to assume that disqualification regimes should look the same everywhere or be insensitive to local conditions. Empirical evidence of the sort marshalled in Parts II and III, as well as the theoretical work accomplished in Part I, provide a starting point, rather than a fixed star, in the constellation of constitutional possibilities. Our proposals should be read in that tentative spirit.

A.     Substantive Thresholds for Disqualification

In the abstract, the correct substantive standard for disqualification is easily stated: disqualification is justified when specific actors or groups pose a clear anti-democratic threat. Disqualification’s costs to democracy should be countenanced if and only if they are outweighed by the danger posed by an actor who threatens to erode democracy upon winning office. The case for permitting other considerations to factor in is weak, since disqualification is targeted at the specific concern that an actor who threatens to erode democracy will gain state power.[289]

Democratic preservation, admittedly, can be an ambiguous standard. It is possible to gain clarity by positing a relatively minimalist definition of democracy, which foregrounds free and fair elections, basic political rights of speech and association, and rule-of-law constraints on state power.[290] These elements tend not to dissipate immediately. In place of military coups and sudden democratic implosions, parties and actors tend to attack democracy gradually, using legal tools and constitutional changes to consolidate power and to repress the opposition.[291] The result may be a legal regime where elections continue to be held, but incumbents tilt the electoral playing field heavily in their favor— so-called hybrid or competitive authoritarian regimes.[292] Given the incremental way in which modern democracies tend to wither, democracies must be able to identify and prevent political actors or groups from undertaking incremental authoritarian projects, if they are able to gain power. The theory and practice of militant democracy, in contrast, assumes a thoroughly authoritarian or totalitarian party—a model that is increasingly inapt.[293]

The evidentiary difficulties in discerning whether a party or official presents a future risk to democracy can be fruitfully addressed by operationalizing the norm of democratic commitment by a mix of rules and standards, but with a preference for the former. At present, some disqualification norms are clear and rule-like. Term limits, mandatory retirement ages, and felony disqualification have this character. Many others are standards with a more discretionary, open-textured character.

In our view, an optimal disqualification system should lean more on rules than standards. The latter are likely to be difficult to apply because specific judgments will rest on speculative evaluation of what might come to pass. Rules, such as term limits, can be written so that officials can be disqualified without requiring open-ended assessment of the power they have accrued. Rules are also likely to have lower compliance costs than standards because it is harder to interfere with their application.[294]

Standards are warranted to the limited extent it is unfeasible to spell out all the forms that democratic threats may take ex ante.[295] Thus, a broad power to remove groups who seek to “undermine or abolish the free democratic basic order” may make some sense,[296] as does a discretionary power to proscribe individuals who commit certain forms of impeachable offenses such as “high Crimes and Misdemeanors.”[297] Standards should focus on the kinds of acts that are likely, in light of recent experience, to pose a threat to the stability of the democratic order. A broader focus on the character or morality of public officials, while appropriate in other contexts, is likely to lead to an unduly activist regime.

Standards are also warranted when rules would present a risk of overreach. This depends on context. Felony disenfranchisement and disqualification rules, for example, are sometimes used as a substitute for discretionary forms of disqualification like impeachment. In the United States, they have an ugly, racist history.[298] But in a different context, this history may be absent, and a focused felony disqualification rule might have a different effect.

The choice between rules and standards also turns on venue. Presently, broad standards appear in legislative processes such as impeachment; think here again of the Constitution’s “high Crimes and Misdemeanors” language. In legislative contexts, very broad standards may make sense because the democratic legitimacy of the legislature lends weight to the disqualification decision, even when it is made on an ambiguous substantive basis. Political forces provide a constraint on abuse. The same kinds of ambiguous standards, such as “good character,” may be more dangerous when applied by administrative or judicial bodies. Such broad standards invite adjudicators to reach well beyond the goal of protecting electoral democracy itself, which should be the core concern of a disqualification regime.

B.     Individuals, Not Groups

One basic decision in disqualification regimes is the choice between groups and individuals as targets of exclusion. Lustration rules and militant democracy focus on groups. Impeachment and term limits focus on the individual. Group disqualifications are surprisingly common in constitutional democracies, which is odd considering well-founded modern norms against group punishment or accountability.[299] The rationale seems to be their utility as instruments of transitional justice, not as permanent bars for “tainted” officials from the polity.[300] In this capacity, they serve mainly to stabilize and legitimize emerging democracies. As political transitions end, lustration rules generally ease, either because laws sunset or are repealed, or officials are reinstated by the requisite actors. Of course, there is some risk that lustration provisions will become entrenched, with society becoming more accustomed to large-scale disqualifications over time.[301]

The more sweeping protection that militant democracy and lustration norms provide may be important to prevent democracy from being eroded from within. But group-based measures should be considered only where individual forms of accountability, such as impeachment, are insufficient, and where the failure to act will imperil democracy itself.[302] Beyond the distinctive context of new democracies seeking to overcome previous autocratic regimes, group-based mechanisms of disqualification are likely so prone to abuse and overuse that they must be either watered down or completely abandoned. Hence, as we saw in Part III, many Eastern European countries winnowed down lustration rules by using alternatives to flat disqualification: individual assessments of suitability in Germany, revelation rather than disqualification in Hungary, and disqualification only for the dishonest in Poland.[303] Part III observed that temporary provisions often calcify into permanence over time. But this is undesirable. Where group-based disqualification persists out of political inertia, it should be saddled with such onerous procedural constraints, deployable only as a last resort, and, as the NDP case showed, utilized only where the threat is substantial rather than merely hypothetical.[304]

The task of banning a person from political life raises a sufficiently weighty risk of entrenchment that it must be hedged around with procedural constraints that promote individualized, and individually justified, sanctions. Some actor (either a court or administrative agency, or both) must decide whether a specific individual’s record is sufficient for disqualification.[305] Across nearly all lustration systems, the extent of desirable due process is contested.[306] Often, the process of assigning individual guilt based on long-ago actions has narrowed the effective scope of lustration considerably. Likewise, militant democracy norms wrestle with the question of how to treat individuals based on their membership in a proscribed party. German law draws a sharp distinction between the organization and the individual: a party ban only has direct effect at the level of the party itself, and mere membership in a proscribed party is insufficient for sanction.[307] Elsewhere, this line is blurred. For example, the South Korean Constitutional Court has punished both individuals and parties, reasoning that focusing on individuals alone allows members of the sanctioned party to remain in office, which would be insufficient to defend democracy.[308]

In contrast, there is no real case for minimizing or abolishing mechanisms that carry out individualized bans from political life. The very commonality of such rules suggests a felt need for an individualized safety valve for democracy.[309] One cannot rule out, of course, the possibility that some individual forms of disqualification might also be able to “sunset” over time. But the possibility has not to our knowledge been explored in existing scholarship or design. As a first cut, it may be more reasonable to phase out term limits than other exit rules. The balance between the protection provided by presidential term limits and the benefits provided by reelection, such as democratic accountability and continuity, may shift depending on the “fragility” of the context. But the fact that virtually all presidential systems include some form of term limit (and those lacking such limits are usually autocratic) suggests a continuing vulnerability ameliorated by presidential term limits.

Indeed, recent events have suggested that the gap between a “fragile” and a “consolidated” democracy may be far thinner than political scientists and comparative constitutional lawyers have long assumed.[310] Arguably, democracies require some degree of “militancy” throughout their lives as a stabilization mechanism.[311] There is a much stronger case for doing so via robust enforcement of individual disqualification mechanisms rather than through group mechanisms like lustration or party bans.

C.     Process and Decision-Makers

A second set of questions is procedural: who decides whether disqualification proceeds, and through what steps? As we have seen in Part III, the options for a decision-maker include a court, an administrative agency, or a legislature. Section 3 of the Fourteenth Amendment simply fails to address the question, making it a suboptimal approach conducive to uncertainty and disuse. The theoretical material we canvassed in Part I also does not support a strong conclusion about the merits of any one or another decision-maker.[312]

Disqualification by the legislature may yield greater public legitimacy for decisions but comes with a greater risk of repression. A legislative process may also prove too difficult to deploy against even clear threats to democracy. This risk may be especially acute where the legislative processes turn on a supermajority voting rule, as is the case with impeachment virtually everywhere. Such processes are likely to be unresponsive even when an actor or group poses a clear threat to democracy, as many commentators argued was the case during President Trump’s second impeachment trial in early 2021.[313]

These shortcomings make impartial, expert-driven venues more appealing. Constitutional designers sometimes turn to judges because of their perceived impartiality. Courts are usually seen as the appropriate institution to handle the sensitive decision of whether to ban anti-constitutional parties. They are also often charged with reviewing lustration decisions. In some contexts, courts are criticized for allowing too much process or moving too slowly. In Poland, lawmakers have looked increasingly toward administrative venues over time because they felt courts were relying on a criminal law model too cumbersome for lustration.[314] Similarly, in the United States, post-Civil War courts were criticized for resisting the disqualification of former Confederate leaders and more broadly for limiting and obstructing Reconstruction itself, even though it was Congress that took the lead in removing disabilities and retreating from Reconstruction in other areas.[315] But courts are not necessarily restrained in disqualification decisions. Activist courts in Israel and Pakistan have constructed and applied loose standards to disqualify large numbers of officials.[316] This suggests that there is a risk in certain contexts that courts will adopt too aggressive a standard for disqualification precisely because they do not trust the political process to police itself.

Similarly, administrative disqualifications pose a risk of excessive zeal, depending on the selection and culture of the agency. In lustration debates, opposition to administrative venues often turns on the risk that agencies will “de-civilize” lustration by weakening due-process protections.[317] Colombia, which relies heavily on administrative institutions to disqualify political officials for perceived malfeasance or corruption on relatively vague grounds, is a potential illustration of this concern.[318] Perhaps for this reason, designers rarely rely on administrative actors alone. Instead, they allow for some judicial review over administrative decisions to disqualify.

On balance, while there is no perfect procedure, it would be a mistake to rely solely on legislators and elected actors. Especially where parties are very strong, disciplined, and polarized, the risk that disqualification will be instrumentalized negatively is too great. At the same time, a purely technocratic process will suffer a potentially disabling legitimacy deficit. A better system would include parallel paths through both administrative and judicial forums. In comparison to the current scheme, making alternative venues available is likely to increase the rate of disqualification. Clear constraining criteria for disqualification (which we refer to in more detail below), effective procedural protections, and meaningful judicial review of administrative decisions to exclude can mitigate a risk of excessive exclusion. This arrangement has the further advantage of directing attention to the key question of instrumental rationality—the link between the means of disqualification in a specific case and the end of preserving the process of “ruling and being ruled in turn.”

D.    Consequences: Temporary, Not Permanent, Bans

Disqualification can be temporary or permanent. Attention to the trade-off to democracy-related costs and benefits suggests a reason to err on side of temporary measures. By forcing dangerous or unfit officials out of office for some period of time, time-limited measures can protect the sound operation of democracy during periods of stress. Permitting an eventual return to power limits the burden placed by exclusion on democratic norms. It likely also boosts the probability of compliance, limits social unrest by supporters, and avoids legitimacy costs.[319] Despite these factors, temporary bans are presently unevenly distributed. Party bans, for example, generally seem to be permanent rather than temporary. Their durability is tempered somewhat by the possibility that adherents often reform under a new, slightly altered party banner, as occurred repeatedly in Turkey.[320]

A temporary bar requires ex ante specification of its duration. This is likely to have an arbitrary flavor. Because of its rule-like quality, it will be either under- or over-inclusive with respect to particular democratic threats. To mitigate this concern, a disqualification regime might allow a court, agency, or legislature to make case-by-case determinations when actors are suitable for reentering the political sphere. As we discuss below, the Fourteenth Amendment arguably permits a supermajority of Congress to make such a determination. Other lustration systems also include an implicit sunset provision.[321] For instance, a lustration statute itself might be set to expire after a fixed period of time—although even then, it might be repeatedly extended, as in the Czech Republic.[322]

E.     One Path or Many? The Advantages of Pluralism

A common feature of constitutional design in this area is the inclusion of more than one pathway to disqualification. The U.S. Constitution’s four different mechanisms are exemplary rather than unique. At first blush, this seems unwise: why risk possible “system effects” from the interactions between different mechanisms?[323] Why not consolidate disqualification in one venue?

But plural disqualification channels are well justified, even if their interaction effects need attention. There are compelling reasons why constitutions contain different disqualification processes. They tend to serve distinct, overlapping functions. Depending on whether an official hopes to gain an elected or non-elected office, moreover, bars to future officeholding raise different normative concerns. Varying channels may be established at different historical moments to achieve different ends. Lustration measures, as we have argued already, are best thought of as transitional measures, phased out as a polity gains distance from its autocratic past. To force all these temporally and normatively distinct tasks into a single functional form would be undesirable.

Even as they create plural disqualification mechanisms, a constitution’s designers, as well as officials executing these provisions, should attend to interactions between different disqualification mechanisms. In particular, plural pathways create a risk of inconsistent, and hence manipulable, regimes. Splitting disqualification into different processes should not provide a pathway for different regimes to be applied to mainstream and peripheral parties. South Korea’s Constitution, for example, contains both an impeachment procedure and a procedure for prohibiting anti-constitutional parties.[324] Both give the Constitutional Court the final word about removing an official (after a motion has been passed by two-thirds of the legislature) or banning a party. In carrying out the former inquiry, the Court asks not only whether the official has committed an impeachable offense, but also whether the offense warrants removal given the substantial effect removal will have on the constitutional order.[325]

In contrast, during proceedings to ban the allegedly pro-North Korean United Progressive Party, the Court offered only a brief discussion of proportionality. Its decision, commentators argued, diverged sharply from relevant international standards.[326] In effect, the Court focused almost entirely on the party’s platform and activities without asking whether such a small movement could realistically achieve its goals. The careful way in which the Court evaluated proportionality in the impeachment context is at odds with this light touch for a wholesale disqualification. Perhaps the unique and imminent threat of North Korea to South Korean interests garnered more weight from the Court. But there is no reason, in South Korean law or as a matter of first principle, to think the robustness of the inquiry should vary across both contexts. Indeed, if anything, the risks of abuse are greater with respect to party bans, which are more readily used as a tool for majoritarian abuse.

Making it too easy for unelected officials to disqualify elected ones can itself create tension. In Israel and Pakistan, judicial standards conflict with legislative ones, which increases the opportunity for abuse. For example, Israeli courts have disqualified officials on the basis of relatively vague standards of good conduct. These judge-made doctrines overlap with other textual disqualification provisions that use rather different standards.[327] The Israeli Supreme Court’s use of a “good conduct” standard, for instance, sits uneasily alongside statutory requirements holding that officials must be removed and disqualified from office upon conviction of serious crimes.[328] In effect, the Court allows officials to be disqualified upon indictment, even as the laws themselves require resignation only upon conviction.[329] Similarly, in Pakistan, the Supreme Court’s aggressive use of Article 62 to permanently disqualify officials is hard to square with the more carefully drawn restrictions found in Article 63. The latter states that those convicted of certain categories of serious crimes are barred from office for five years.[330] Yet, the judicial interpretation of Article 62, which has been used to cover similar conduct, has been used to bar officials for life.[331] Such disparities invite partisan manipulation to protect incumbents—defeating the primary object of disqualification mechanisms.

* * *

To summarize, an “optimal” disqualification regime aims at disqualifying officials who pose a clear threat to a relatively minimalist, electorally focused conception of democracy. This ensures an opportunity to “rule and be ruled in turn,” while avoiding overuse for less pressing ends, or even worse, abuse for anti-democratic purposes. It would contain plural pathways, calibrated to avoid the possibility of partisan arbitrage. These would lean toward the regulation of individuals rather than groups. They would not usually run directly through elected bodies. The prerequisite for disqualification would more often be stated as a rule than as a standard. And the ensuing prohibitions would more often be temporary rather than permanent to calibrate and minimize the damage done to democracy from the disqualification itself.

V. Disqualification Under the U.S. Constitution: A Reconsideration

How might one apply these design principles to the undertheorized disqualification regime found in the United States? The challenges faced by American democracy, as many have noted, are not unique. There is no reason why comparative and theoretical experience cannot inform reform efforts aimed at improving the present disqualification systems. Of course, the difficulty of constitutional amendment under Article V curtails the set of feasible interventions. But the ensuing space for design changes is still significant.

As a threshold matter, Part II demonstrated that the U.S. Constitution already contains a variety of different disqualification mechanisms. These include a presidential term limit designed to prevent strong chief executives from distorting the democratic political order; a legislative power to disqualify presidents, judges, and officials upon Senate conviction for impeachable offenses; a lustration-like provision in Section 3 of the Fourteenth Amendment; and the legislative powers of exclusion and expulsion. The U.S. Constitution thus already contains a set of different mechanisms, as we have argued the optimal regime would. But many of those mechanisms are not working well, either when interacting together or separately.

The disqualification regime faced its most significant test in modern history after the insurrection of January 6, 2021. The impeachment pathway gained more support for presidential removal than it had at any point since the impeachment of President Andrew Johnson in 1868. But it still failed to disqualify President Trump from office in a Senate vote that was not close to achieving the two-thirds threshold. And the removal process failed despite a set of facts that hewed quite close to the core purpose of a disqualification regime in a democracy. The articles of impeachment accused President Trump of inciting an insurrection by setting a mob on Congress on the day that it was meeting to certify the results of the presidential election, after spending weeks attempting to delegitimize and reverse the electoral results via a number of different routes.[332] To many, this suggests a need for renewed development of alternative pathways.

A.     Against “Militant Democracy”

The U.S. Constitution’s paths to disqualification largely focus on individuals rather than groups. Other nations, as we have noted, take different paths. In Canada, for example, the far-right Proud Boys were designated a terrorist group in February 2021.[333] In the United States, this seems less likely to happen. Historically, Congress and state legislatures acted to outlaw some political parties (such as Communist movements) and to suppress their members.[334] First Amendment concerns have not prevented the application of group-based sanctions to foreign organizations.[335] But domestic organizations are likely to have more robust constitutional claims.[336] Militant democracy of the sort used in other systems around the world, therefore, is not readily available in the United States, given the constraints posed by the Constitution.[337]

We think that this equilibrium is unlikely to change because of the entrenched quality of First Amendment rules. But setting aside feasibility constraints, is this a loss? Miguel Schor has recently promoted the addition of an explicit militant democracy dimension to U.S. constitutionalism.[338] He contends that the Framers envisioned a kind of militant democracy to protect against factions and demagogues but chose the wrong tools and aimed against the wrong threats for the modern age.[339]

Even setting the constraints of the First Amendment aside, experiences with militant democracy elsewhere around the world are mixed enough to warrant more caution than Schor allows. In the previous Section, we note a clear preference for individual rather than group forms of disqualification. This is especially true in established rather than fragile democracies. And indeed, both lustration rules and party-banning powers often seem paired with explicit or implicit sunsets, and so are transitional in nature.

Militant democracy clauses are surprisingly common but in practical experience run into a predictable set of problems. These include the risk of overuse in situations where actors pose no clear threat to the democratic order (as arguably happened in South Korea).[340] More dramatically, they include the risk of outright abuse, where party bans become an anti-democratic tool to repress the opposition, as in Turkey and Cambodia.[341] And finally, they include the very common, salient risks that anti-democratic parties may not be easily identifiable in modern politics and that bans may prove ineffective even if they are attempted.[342] While we share Schor’s sense that U.S. tools of disqualification must be made more robust, we are skeptical that the addition of explicit militant democracy tools is the way to do it.

B.     Revising and Reviving Section 3 of the Fourteenth Amendment

Section 3 of the Fourteenth Amendment raises similar concerns because it is in some ways a group form of disqualification, a variant of the lustration clauses found around the world. These clauses, as we have explained at length, play an important role in transitions to democracy or in post-conflict scenarios. But they are paradigmatically transitional, with their anti-democratic risks rising (and pro-democratic benefits falling) if they become permanently entrenched in the constitutional order.

Section 3, though, is written in general terms: it is not textually limited to its origins after the Civil War. Comparative experience suggests it could be given an “individualistic” valence more appropriate for a mature democracy. In particular, Section 3 could be given greater specificity and shape via statute, as Congress indeed did after the Civil War and as it is empowered to do now via its authority to “enforce” the terms of the Reconstruction Amendments.[343]

Such a statute would be useful to clarify both the substantive standard for application and the procedure for disqualification. It might also address other issues, such as incorporating our argument in favor of temporary rather than permanent bans in most cases. Via a carefully crafted disqualification statute, Section 3 could thus be moved some distance towards the direction of our “optimal” regime, outlined in Part IV. Since Justice Chase held after the Civil War that Section 3 is not self-executing, there is a good chance that such a statute would be a prerequisite for any modern use.[344]

As it is, Section 3’s threshold of “insurrection or rebellion” invites careless application and fatal underreach. It is probably too narrow to deal with most modern threats to democracy. Incumbents now use a variety of quite legal means to entrench themselves in office. These may be challenging to fit within the terms “insurrection or rebellion.” Different, but equally vexing, problems attend the substantive standard for impeachment. The standard for impeachment—“Treason, Bribery, or other high Crimes and Misdemeanors”—is potentially more elastic and open to a range of modern democratic threats.[345] However, that standard has been read in a “legalistic” fashion throughout much of U.S. history, for example as requiring or being anchored to the commission of crimes.[346] President Trump’s defense team argued during his first impeachment that the acts alleged against him were not impeachable because they did not constitute express violations of criminal law.[347] The fact that he may have grossly abused state power in an effort to discredit a political opponent and remain in office—precisely the kind of threat that a reasonable impeachment and disqualification regime would guard against—was irrelevant under that argument.

The threshold for impeachment is best understood, in historical, functionalist, and comparative terms, to allow a broad and relatively political, rather than narrowly legalistic, use of impeachment.[348] A similar clarification is warranted for Section 3. One could, indeed, imagine a statutory framework fleshing out the meaning of “insurrection and rebellion” and elaborating in more detail a substantive threshold keyed to the need to preserve democracy as an ongoing concern. Such a standard should be written broadly to catch future threats rather than being confined to a particular historical incident. Attempts to subvert the electoral process should be at the core of such a “modernized” statutory definition. The standard would thus aim at specific, individualized acts undertaken to attack democracy rather than, as with classic lustration mechanisms, membership in a tainted regime or group. Even with such a clarification, of course, Section 3 may remain underinclusive and simply unusable against some kinds of democratic threats.

Further, the statute should address the process through which disqualification would proceed, as indeed the post-Civil War legislation did. Under the 1870 Enforcement Act, disqualification for most officials proceeded via the initiative of federal prosecutors in suits brought against allegedly ineligible state officials, with the federal courts acting as arbiters. [349] A statute laying out a similar procedure may have some merit in the contemporary United States. A turn to courts would be a shift from the dominant constitutional paradigm for disqualification. Impeachment and legislative exclusion both operate through the political process and not through an administrative agency or a court. Indeed, the United States is striking in its virtually exclusive reliance on political rather than judicial or administrative routes for disqualification.

As suggested by our analysis in Part IV, there is good reason to think that the Constitution has not struck the right balance on this question. The Constitution’s heavy reliance on elected actors, especially members of Congress, creates a risk that disqualification decisions will be based on partisan rather than system-level grounds. This can lead to either under- or over-enforcement, and the prevailing result has been paralysis. As has occurred elsewhere, judicial involvement may help to unblock the channels of disqualification where true anti-democratic threats have been identified. Courts are not a cure-all solution. Comparative and historical experience demonstrates that they may become either too passive or too active in playing their assigned role, or merely too entangled in politics, a particularly salient risk for the already highly politicized federal judiciary.[350] But they could be tapped as part of a revitalized Section 3 that holds some promise of sanctioning anti-democratic attacks on democracy.

Finally, the interaction between Section 3 and other disqualification tools in the Constitution merits attention. Unsurprisingly, given that Section 3 and these other mechanisms were added to the Constitution at different times and for different purposes, they do not always harmonize well. The events of January 2021 showed that the relationship between impeachment, legislative expulsion, and Section 3 demands closer attention. The other routes to disqualifying individuals under the Constitution based on past conduct, impeachment and legislative expulsion, require a two-thirds supermajority. These supermajority requirements help ensure that removal and disqualification are not purely majoritarian acts.

Section 3 contains no such supermajority requirement for disqualification; indeed, it flips the script by requiring a two-thirds vote in each chamber of Congress to lift a disqualification, rather than impose one.[351] Its aggressive use may lead to majoritarian arbitrage from the more stringent channels. Indeed, once it became clear that impeachment would fail to disqualify former President Trump, many commentators argued in favor of Section 3 as an alternative route to disqualify him and some of his collaborators. Thus, one risk is that the lustration provision in Section 3 could allow majority actors a means to repress political minorities. There is no way to close this gap entirely. But the kind of statute we envision in this Section may ameliorate this risk. Again, providing for a primarily judicial rather than political process would lessen the concern somewhat. So too would a standard keyed to categories of disqualifying threats to democracy, as opposed to a standard written exclusively for a particular historical event.

C.    The Importance of the Twenty-Second Amendment

The near-moribund status of impeachment and (at least in its current form) Section 3 as instruments of disqualification means that the Twenty-Second Amendment’s presidential term limit is a singularly important protection for the U.S. democratic order. Placing a lifetime two-term limit on presidents, regardless of individual competence or attitude, is a crude way to protect democracy and carries real costs. Popular and effective presidents with no nefarious agenda are arbitrarily forced to leave office, despite popular opinion. At the same time, the clear, rule-like quality of the term limit can be a major advantage, avoiding the difficult and politically fraught judgments attending the nebulous standards for disqualification found in impeachment and Section 3.

Particularly under circumstances where other prohibitory mechanisms have fallen into desuetude, the Twenty-Second Amendment embodies the correct calculus that, on balance, a hard limit wards against dangers that are greater than the costs it imposes. Hamilton in Federalist No. 72 was correct to note that presidential term limits impose costs, but he underestimated their benefits as a bulwark of democracy—perhaps because he and other Founders assumed impeachment and other disqualification mechanisms would be more protective than they have turned out to be.[352] A raft of recent comparative work shows that presidents who do not surrender power often pose a threat to basic electoral democracy because they leverage their power to tilt the playing field, making future elections increasingly unfair.[353] Furthermore, as we have already noted, the democratic records of presidential systems without term limits on their chief executives are invariably grim.[354] One benefit of considering term limits as part of the U.S. Constitution’s broader disqualification regime is to highlight their underappreciated role in limiting anti-democratic threats.

Yet the United States’s presidential term limit regime may be more vulnerable to evasion than commonly appreciated. It has been followed relatively uneventfully since adoption in the mid-twentieth century. This period of tranquility may be deceptive.[355] Unlike many democracies around the world, the United States has never experienced a serious evasion attempt. But past may not be prologue. It would be dangerous to assume that no such attempt will happen in the future.[356] As we showed above in Part II, the current regime is riddled with ambiguity about enforcement. Should an evasion attempt be made, whether brazenly or with subtlety, it is unclear which institution would be responsible for stopping it. The courts might not act, deferring to political institutions that may also be paralyzed.

Thus, there is a powerful case for a framework statute setting forth a judicial mechanism for enforcing the two-term limit on chief executives.[357] Ideally, enforcement would precede a presidential election and perhaps focus on the presence on the ballot of a candidate who is barred by law. At present, courts might be likely to steer clear of such a dispute, invoking the political question doctrine.[358] A new law could force their hand. Such a statute would have to identify appropriate plaintiffs (for example, the attorney general of a state) and elaborate a clear norm detailing the Twenty-Second Amendment’s application to different scenarios. It would also have to specify a remedy. For example, a district court could be authorized to issue an injunction against including an illegitimate candidate on state ballots. In effect, this is the mirror image of orders now issued mandating a candidate’s inclusion.[359] It is also akin to orders the Supreme Court has recently issued mandating that certain votes not be counted in an ongoing election.[360]

D.    New Institutional Pathways

Until now, we have assumed that constitutional amendment was off the table and focused on those improvements that could be made by statute or even informal shift. What, however, if current partisan formations and amendment difficulties were bracketed? What reforms might then be desirable? The Twenty-Second Amendment, after all, was ratified a mere seventy years ago. The assumption that the constitutional order is frozen in time may well be an artifact of the present political moment—and that itself may be transient. Under those assumptions, more ambitious reform might be conceived.

If constitutional amendment were on the table, one could re-envision disqualification from the ground up, constructing a system very close to our theoretical optimum (although tailored to the U.S. context). One could construct a new pathway keyed towards the protection of democracy, focused on identifying threats posed via individual acts (rather than group membership), and placed largely in non-legislative hands. As we have noted above, a revitalization project around Section 3, which would focus on the drafting of a careful statutory framework, might be able to play that role to some degree. But Section 3 is inevitably limited, especially because of its fairly narrow substantive focus on “insurrection or rebellion,” which hardly exhausts the universe of modern anti-democratic threats.

Wholesale reform must start from the observation that no system primarily or wholly reliant on super-majoritarian decision-making by Congress is likely to work well in the current American context, given the mix of a dominant two-party system and an increasingly polarized polity. This suggests the need to rely on other institutions, such as judicial or administrative agencies. The latter approaches, we have shown, are common overseas. One way to structure such a system would be to retain the existing impeachment procedure as is while also creating a new pathway for disqualification more reliant on administrative or judicial actors. In other words, we would propose decoupling impeachment and disqualification, creating two distinct institutional pathways. Section 3 already suggests this differentiation.

We would suggest broadening the grounds for disqualification beyond “insurrection or rebellion,” a standard designed primarily to deal with the particular problems posed by the Civil War. This new institutional pathway would almost certainly become more aggressive in making disqualification decisions. That would be its aim. There is, of course, a corresponding risk of excessive use. This risk could be controlled, however, with a clearer substantive threshold for political expulsion and more detailed ex ante guidance as to the actions sufficient to warrant disqualification. The language should focus on the kinds of actions that pose a threat to democratic stability, not broader issues of the character or morality of public officials. The focus in Israel, Pakistan, and Colombia on character, and perhaps even corruption, sweep too broadly into the democratic sphere.[361]

Finally, in designing a new pathway for disqualification, the United States would be better served with temporary exclusions of the sort found elsewhere rather than the more permanent bars contained in the current text of the federal constitution. Many systems around the world, as we have shown, use temporary exclusions from power to defend the democratic order while softening the tension with democracy that is implicit in any disqualification regime. Disqualifications of five to eight years may help preserve democracy against immediate threats, while also increasing both compliance and incentives for actors to deploy disqualification as a sanction.[362] Temporary bans also allow the length of disqualification to be calibrated to the degree of the offense and nature of the threat posed to the democratic order.

Conclusion

In this Article, we have highlighted the diverse ways in which democratic constitutions, including our own, disqualify individuals or organizations from seeking future office. We suggest that the American way of handling disqualification, while sprawling, creaky, and fragmented, has both costs and benefits. On the one hand, it has been proved resistant to capture and redeployment as an instrument of partisan entrenchment. On the other hand, it is far too slow to confront modern democratic threats. There are no perfect fixes. But reflection on comparative experience and democratic theories can help flesh out opportunities for improvement. Given the prevalence of anti-democratic threats in recent years, both within the United States and globally, the need to develop more effective boundary conditions for democracy will remain an urgent and unceasing task that, unfortunately, seems unlikely in the current climate of political gridlock.

Copyright © 2023 Tom Ginsburg, Aziz Z. Huq, and David Landau.

Tom Ginsburg, Leo Spitz Professor of Law, University of Chicago Law School. Aziz Z. Huq, Frank and Bernice J. Greenberg Professor of Law, University of Chicago Law School. David Landau, Mason Ladd Professor, Florida State University College of Law.

Thanks to Danielle Wenger for research assistance. For helpful comments, thanks to Eric Orts and audiences at the University of Chicago, the University of Pennsylvania, Harvard Law School, and the Comparative Law Work in Progress Workshop at the University of Illinois. DOI available at https://doi.org/10.15779/Z388P5VB2H.

           [1].     The first impeachment of Donald Trump in 2019 focused on the abuse of power and the obstruction of Congress in his efforts to corruptly use his powers to influence Ukraine’s actions. Nicholas Fandos & Michael D. Shear, Trump Impeached for Abuse of Power and Obstruction of Congress, N.Y. Times (Dec. 18, 2019), https://www.nytimes.com/2019/12/18/us/politics/trump-impeached.html [https://perma.cc/NFE5-F4RL].

           [2].     See, e.g., Dennis Aftergut, Opinion, Impeaching Lays Groundwork for Disqualification— Even Without a Conviction, Hill (Jan. 12, 2021), https://thehill.com/opinion/white-house/533745-impeaching-lays-groundwork-for-disqualification-even-without-a-conviction [https://perma.cc/CFU8-BJ8F]. The Constitution identifies “disqualification to hold and enjoy any Office of honor, Trust or Profit under the United States” as a potential downstream consequence of impeachment. U.S. Const. art. I, § 3, cl. 7.

           [3].     Trial Memorandum of the United States House of Representatives in the Impeachment Trial of President Donald J. Trump, S. Doc. No. 117-3, vol. 1, at 71 (2021).

           [4].     Bruce Ackerman & Gerard Magliocca, Opinion, Impeachment Won’t Keep Trump from Running Again. Here’s a Better Way, Wash. Post (Jan. 11, 2021), https://www.washingtonpost.com/opinions/2021/01/11/impeachment-wont-keep-trump-running-again-heres-better-way/ [https://perma.cc/7G4Q-4UMJ].

           [5].     Richard D. Bernstein, Lots of People Are Disqualified from Becoming President, Atlantic (Feb. 4, 2021), https://www.theatlantic.com/ideas/archive/2021/02/trump-disqualification-president/617908/ [https://perma.cc/2VR6-EPAL].

           [6].     Daniel Hemel, Using the 14th Amendment to Bar Trump from Office Could Take Years, Wash. Post (Jan. 12, 2021), https://www.washingtonpost.com/outlook/2021/01/12/14th-amendment-insurrection-trump-removal-problems/ [https://perma.cc/2XXE-FAB4] (also raising concerns about the Bill of Attainder Clause as an impediment to legislative action under Section 3 and noting historical support for a judicial channel). For an extended analysis, see Daniel Hemel, Disqualifying Insurrectionists and Rebels: A How-To Guide, Lawfare (Jan. 19, 2021) [hereinafter Hemel, Disqualifying Insurrectionists], https://www.lawfareblog.com/disqualifying-insurrectionists-and-rebels-how-guide [https://perma.cc/V6XW-2NFD].

           [7].     Tom Ginsburg, Disqualification Won’t Keep Trump Out of the White House, Balkanization (Feb. 11, 2021), https://balkin.blogspot.com/2021/02/disqualification-wont-keep-trump-out-of.html [https://perma.cc/L6VL-FZVX].

           [8].     Sam D. Elliott, When the United States Attorney Sued to Remove Half the Tennessee Supreme Court: The Quo Warranto Cases of 1870, 49 Tenn. Bar J. 20, 26 (2013).

           [9].     Gerard N. Magliocca, Amnesty and Section Three of the Fourteenth Amendment, 36 Const. Comment. 87, 109–10 (2021).

         [10].     Id. at 110–11.

         [11].     Id. at 111.

         [12].     Bond v. Floyd, 385 U.S. 116, 123 (1966).

         [13].     Id. at 136–37.

         [14].     Powell v. McCormack, 395 U.S. 486, 508–10, 550 (1969) (holding that the House could only exclude Powell based on grounds enumerated in the Constitution’s text).

         [15].     Paksas v. Lithuania, App. No. 34932/04, Eur. Ct. H.R. ¶ 31 (2011), https://hudoc.echr.coe.int/eng?i=001-102617 [https://perma.cc/Z6N7-FXMK].

         [16].     Gabriella Gricius, Transitional Justice: Lustration and Vetting in Ukraine and Georgia, 5 J. Liberty & Int’l Affs. 26, 35–36 (2019).

         [17].     Yuliya Zabyelina, Lustration Beyond Decommunization: Responding to the Crimes of the Powerful in Post-Euromaidan Ukraine, 6 State Crime J. 55, 71 (2017).

         [18].     Hunbeobjaeopanso [Const. Ct.], Dec. 19, 2014, 26-2 Hun-Da 1 (S. Kor.).

         [19].     Shasta Darlington & Manuela Andreoni, Brazilian Court Rules that “Lula” Cannot Run for President, N.Y. Times (Aug. 31, 2018), https://www.nytimes.com/2018/08/31/world/americas/lula-president-brazil.html [https://perma.cc/ASM8-E5ZF].

         [20].     Ernesto Londoño & Letícia Casado, Brazil’s Ex-President “Lula” May Run for Office Again as Court Cases Are Tossed, N.Y. Times (Mar. 8, 2021), https://www.nytimes.com/2021/03/08/world/americas/brazil-lula-supreme-court.html [https://perma.cc/7YUE-4NKJ].

         [21].     Jack Nicas, Brazil Bars Bolsonaro from Office for Election-Fraud Claims, N.Y. Times (June 30, 2023), https://www.nytimes.com/2023/06/30/world/americas/bolsonaro-brazil-banned-office.html [https://perma.cc/XDF7-5HLU].

         [22].     In previous work, we have similarly distinguished between the normative and the descriptive questions that can be ventilated using a comparative constitutional law lens. See Tom Ginsburg, Aziz Huq & David Landau, The Comparative Constitutional Law of Presidential Impeachment, 88 U. Chi. L. Rev. 81, 86–87 (2021) [hereinafter Ginsburg et al., The Comparative Constitutional Law] (enumerating both descriptive and normative questions).

         [23].     Frederick G. Whelan, Prologue: Democratic Theory and the Boundary Problem, 25 Nomos 13, 13 (1983).

         [24].     We use Louis Kaplow’s now-canonical definition of rules and standards: a rule is a legal norm that is given content before regulated subjects act, whereas a standard is a legal norm that is given content after regulated subjects act. See Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L.J. 557, 559–63 (1992).

         [25].     Ackerman & Magliocca, supra note 4; Hemel, Disqualifying Insurrectionists, supra note 6; Magliocca, supra note 9, at 93 n.29; Gerard Magliocca, The 14th Amendment’s Disqualification Provision and the Events of Jan. 6, Lawfare (Jan. 19, 2021), https://www.lawfareblog.com/14th-amendments-disqualification-provision-and-events-jan-6 [https://perma.cc/99EY-WN7Z]; James Wagstaffe, Time to Reconsider the 14th Amendment for Trump’s Role in the Insurrection, Just Sec. (Feb. 11, 2021), https://www.justsecurity.org/74657/time-to-reconsider-the-14th-amendment-for-trumps-role-in-the-insurrection/ [https://perma.cc/4V2J-42WV]; Philip Zelikow, A Practical Path to Condemn and Disqualify Donald Trump, Lawfare (Jan. 22, 2021), https://www.lawfareblog.com/practical-path-condemn-and-disqualify-donald-trump [https://perma.cc/5S25-VTMG]; John Nichols, Calls to Disqualify Trump Using the 14th Amendment Grow Louder, Nation (Feb. 19, 2021), https://www.thenation.com/article/politics/14th-amendment-trump-foner/ [https://perma.cc/3ZQ7-C7HT].

         [26].     See infra text accompanying Part II.C.

         [27].     Eric Tucker & Mary Clare Jalonick, Trump Lawyer: Impeachment Case “Undemocratic,” Ill-Advised, AP News (Feb. 2, 2021), https://apnews.com/article/donald-trump-impeachment-updates-5adebdc5d8c30622f5e116e40c9396e6 [https://perma.cc/TTD9-W6C5].

         [28].     Hélène Landemore, Open Democracy: Reinventing Popular Rule for the Twenty-First Century 1 (2020).

         [29].     Robert A. Dahl, On Democracy 35 (1998); see also Landemore, supra note 28, at 7 (describing democracy as “a regime of political equals”).

         [30].     See, e.g., David Van Reybrouck, Against Elections: The Case for Democracy 138–49 (2016) (arguing for a sortition-based democracy); see also Josiah Ober, Demopolis: Democracy Before Liberalism in Theory and Practice 28 (2017) (noting that Athenian democracy was “never centered on the use of a majority-voting rule to elect officeholders”).

         [31].     We recognize that some theorists of democracy go a step further and contend that electoral mechanisms are doomed to fail. See, e.g., Alexander A. Guerrero, Against Elections: The Lottocratic Alternative, 42 Phil. & Pub. Affs. 135, 141 (2014). For application to the United States, see Larry M. Bartels, Unequal Democracy: The Political Economy of the New Gilded Age 253 (2008) (finding that “senators in [1989-1994] were vastly more responsive to affluent constituents than to constituents of modest means”). We do not go that far here.

         [32].     Adam Przeworski, Why Bother with Elections? 21 (2018).

         [33].     Landemore, supra note 28, at 10.

         [34].     Cf. Daryl Levinson & Benjamin I. Sachs, Political Entrenchment and Public Law, 125 Yale L.J. 400, 409 (2015) (“Political entrenchment implies not just the absence of political change but some kind of special constraint on the usual processes of political change.”).

         [35].     The phrase “quis custodiet ipsos custodes” (“who guards the guards themselves”) is generally associated with Juvenal. Susanna Braund, Juvenal, in Oxford Research Encyclopedia of Classics (2015). For applications to political theory, see generally Martin Shapiro, Who Guards the Guardians? Judicial Control of Administration (1988).

         [36].     For an overview of the extensive literature on the boundary problem, see generally Sarah Song, The Boundary Problem in Democratic Theory: Why the Demos Should Be Bounded by the State, 4 Int’l Theory 39 (2012).

         [37].     On the use of citizenship and immigration law for political entrenchment, see Adam B. Cox & Eric A. Posner, The Rights of Migrants: An Optimal Contract Framework, 84 N.Y.U. L. Rev. 1403, 1447 (2009).

         [38].     Whelan, supra note 23, at 15–16; see also Robert A. Dahl, Democracy and Its Critics 119–31 (1989).

         [39].     Robert E. Goodin, Enfranchising All Affected Interests, and Its Alternatives, 35 Phil. & Pub. Affs. 40, 52 (2007).

         [40].     James Lindley Wilson, Making the All‐Affected Principle Safe for Democracy, 50 Phil. & Pub. Affs. 169, 171 (2022).

         [41].     Goodin, supra note 39, at 64 (citation omitted).

         [42].     See, e.g., David Miller, Democracy’s Domain, 37 Phil. & Pub. Affs. 201, 225 (2009) (offering one such account).

         [43].     See, e.g., Tom Ginsburg & Aziz Z. Huq, How to Save a Constitutional Democracy 18 (2018) (arguing that the United States was not a democracy until 1919 because of its failure to accord women the vote).

         [44].     See Samuel Issacharoff & Richard H. Pildes, Politics as Markets: Partisan Lockups of the Democratic Process, 50 Stan. L. Rev. 643, 644 (1998) (exploring “ways in which dominant parties manage to lock up political institutions to forestall competition”).

         [45].     Ginsburg & Huq, supra note 43, at 161–62 (discussing this risk).

         [46].     Cf. Harper v. Va. Bd. of Elections, 383 U.S. 663, 666 (1966) (holding that “a State violates the Equal Protection Clause of the Fourteenth Amendment whenever it makes the affluence of the voter or payment of any fee an electoral standard”).

         [47].     Levinson & Sachs, supra note 34, at 405 (pointing out that formal and functional entrenchment can be substitutes).

         [48].     Ginsburg & Huq, supra note 43, at 192–97.

         [49].     Ober, supra note 30, at 120.

         [50].     See generally John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (1980).

         [51].     Bush v. Gore, 531 U.S. 98 (2000).

         [52].     For an unusually comprehensive version of this criticism, see Michael J. Klarman, Bush v. Gore Through the Lens of Constitutional History, 89 Calif. L. Rev. 1721, 1724–27 (2001).

         [53].     Michael J. Klarman, Majoritarian Judicial Review: The Entrenchment Problem, 85 Geo. L.J. 491, 550 (1997). For a critique and reconstruction of this claim focused on the Roberts Court’s attacks on democratic choice, see Aziz Z. Huq, The Counterdemocratic Difficulty, 117 Nw. U. L. Rev. 1099, 1109 (2023).

         [54].     See David Landau & Rosalind Dixon, Abusive Judicial Review: Courts Against Democracy, 53 U.C. Davis L. Rev. 1313, 1358–60 (2020).

         [55].     The line between entry and exit conditions is, at any rate, sometimes blurred. Some kinds of rules may function as both restrictions on entry and exit to the electoral sphere. For example, some constitutions prohibit bankrupts from running for office (an entry condition) but also require officials to leave office if they declare bankruptcy while holding it (an exit condition). See, e.g., Constitution art. 99(2)(f) (2010) (Kenya) (“A person is disqualified from being elected a member of Parliament if the person . . . is an undischarged bankrupt.”); id. art. 103(1)(g) (providing for removal if the member becomes disqualified for election to Parliament under art. 99(2)(d) to (h)).

         [56].     A long-standing (although controversial) view holds that sitting presidents may not be criminally indicted while in office, for example. See Akhil Reed Amar & Brian C. Kalt, The Presidential Privilege Against Prosecution, 2 NEXUS 11, 11 (1997).

         [57].     See Edward B. Foley, Voting Rules and Constitutional Law, 81 Geo. Wash. L. Rev. 1836, 1847 (2013) (noting the Supreme Court’s Anderson-Burdick balancing test for determining when a state’s rules and procedures for administering the electoral process violate the 14th Amendment).

         [58].     Sanford Levinson, Our Undemocratic Constitution: Where the Constitution Goes Wrong (And How We the People Can Correct It) 9 (2006).

         [59].     Id. at 25–79 (developing criticisms of the federal legislative process as undemocratic); Jacob S. Hacker & Paul Pierson, Let Them Eat Tweets: How the Right Rules in an Age of Extreme Inequality 171–96 (2020) (summarizing evidence of the lack of democratic responsiveness of the U.S. government for all but the very wealthy). For a deployment of the same trope aimed at the Supreme Court itself, see Joshua P. Zoffer & David Singh Grewal, The Counter-Majoritarian Difficulty of a Minoritarian Judiciary, 11 Calif. L. Rev. Online 437, 438–39 (2020).

         [60].     Because our subject is democratic disqualification, we do not address federal judges. They are, however, subject to a similar impeachment regime to executive branch officials.

         [61].     U.S. Const. amend. XXVI.

         [62].     Richardson v. Ramirez, 418 U.S. 24, 54 (1974). That said, these laws, which very commonly have racist origins, impact not only the right to vote, but also the right to run for state office in some states. Jeff Manza, Christopher Uggen & Angela Behrens, The Racial Origins of Felon Disenfranchisement, in Locked Out: Felon Disenfranchisement and American Democracy 41, 41–42 (Jeff Manza & Christopher Uggen eds., 2006). The same is true for state laws that prohibit voting unless all fees and fines are paid off. See Beth A. Colgan, Wealth-Based Penal Disenfranchisement, 72 Vand. L. Rev. 55, 66–67 (2019).

         [63].     Cf. Edward M. Burmila, Voter Turnout, Felon Disenfranchisement and Partisan Outcomes in Presidential Elections, 1988–2012, 30 Soc. Just. Rsch. 72, 73 (2017) (noting different effects before and after 2000).

         [64].     See Ginsburg et al., The Comparative Constitutional Law, supra note 22, at 83–84.

         [65].     U.S. Const. art. I, § 3. To be precise, Article I states that impeachment’s consequences “shall not extend” beyond the enumerated ones.

         [66].     For a cogent discussion of the connection between impeachment and removal, see Michael J. Gerhardt, The Constitutional Limits to Impeachment and Its Alternatives, 68 Tex. L. Rev. 1, 46–47 (1989); see also Raoul Berger, Impeachment: The Constitutional Problems 194–95 (enlarged ed. 1974) (discussing how the power of removal was discussed at the Philadelphia Convention).

         [67].     Josh Chafetz, Impeachment and Assassination, 95 Minn. L. Rev. 347, 422 (2010).

         [68].     See Robert J. Reinstein, Expulsion, Exclusion, Disqualification, Impeachment, Pardons: How They Fit Together, Lawfare (Feb. 11, 2021), https://www.lawfareblog.com/expulsion-exclusion-disqualification-impeachment-pardons-how-they-fit-together [https://perma.cc/XY44-T9DN] (exploring the interplay between the voting requirements for expulsion and exclusion under Article I and those for expulsion, exclusion, and disqualification under Section 3 of the Fourteenth Amendment).

         [69].     For representative entries in this debate, see, for example, Frank O. Bowman, III, The Constitutionality of Trying a Former President Impeached While in Office, Lawfare (Feb. 3, 2021), https://www.lawfareblog.com/constitutionality-trying-former-president-impeached-while-office [https://perma.cc/MHT8-VANH]; Keith E. Whittington, Can a Former President Be Impeached and Convicted?, Lawfare (Jan. 15, 2021), https://www.lawfareblog.com/can-former-president-be-impeached-and-convicted [https://perma.cc/8EFK-GW5E]; Jed Handelsman Shugarman, Impeach an Ex-President? The Founders Were Clear: That’s How They Wanted It, Politico Mag. (Feb. 11, 2021), https://www.politico.com/news/magazine/2021/02/11/donald-trump-impeachment-ex-president-founders-468769 [https://perma.cc/6P4U-WQFJ]; cf. Harold J. Krent, Can President Trump Be Impeached as Mr. Trump? Exploring the Temporal Dimension of Impeachments, 95 Chi.-Kent L. Rev. 537, 538 (2020).

         [70].     See Benjamin Cassady, “You’ve Got Your Crook, I’ve Got Mine”: Why the Disqualification Clause Doesn’t (Always) Disqualify, 32 Quinnipiac L. Rev. 209, 218 (2014).

         [71].     See Seth Barrett Tillman, Originalism and the Scope of the Constitution’s Disqualification Clause, 33 Quinnipiac L. Rev. 59, 63 (2014); see also Seth Barrett Tillman, Who Can Be President of the United States? Candidate Hillary Clinton and the Problem of Statutory Qualifications, 5 Brit. J. Am. Legal Stud. 95, 99–112 (2016) (interpreting the scope of the term “office under the United States”).

         [72].     See Gerhardt, supra note 66, at 46–47 (discussing possible interpretations of the removal power).

         [73].     See, e.g., Michael J. Gerhardt, The Federal Impeachment Process: A Constitutional and Historical Analysis 80–81 (3d ed. 2019) (discussing this procedure).

         [74].     See U.S. Const. art. I, § 3, cls. 6–7 (“And no Person shall be convicted without the Concurrence of two thirds of the Members present. Judgment in Cases of Impeachment shall not extend further than to removal from Office, and disqualification to hold and enjoy any Office of honor.”).

         [75].     Cf. Powell v. McCormack, 395 U.S. 486, 507–09 (1969) (noting limits on Congress’s power to expel members and suggesting that they cannot be circumvented by refusing to seat a member).

         [76].     Ginsburg et al., The Comparative Constitutional Law, supra note 22, at 114.

         [77].     See Lawrence J. Trautman, Presidential Impeachment: A Contemporary Analysis, 44 U. Dayton L. Rev. 529, 536, 543 (2019).

         [78].     Jonathan Turley, The Executive Function Theory, the Hamilton Affair, and Other Constitutional Mythologies, 77 N.C. L. Rev. 1791, 1826, 1830–31 (1999) (discussion of Humphreys and Archbald); J. Richard Broughton, Conviction, Nullification, and the Limits of Impeachment as Politics, 68 Case W. Rsrv. L. Rev. 275, 278 (2017) (referring to the author’s discussion of Thomas Porteus).

         [79].     David E. Kyvig, The Age of Impeachment: American Constitutional Culture Since 1960, at 308–09 (2008).

         [80].     Id.; Cassady, supra note 70, at 211–14.

         [81].     Ackerman & Magliocca, supra note 4.

         [82].     The historical background is summarized in Eric Foner, The Second Founding: How the Civil War and Reconstruction Remade the Constitution (2019).

         [83].     Benjamin B. Kendrick, The Journal of the Joint Committee of Fifteen on Reconstruction 116 (1914).

         [84].     Gregory E. Maggs, A Critical Guide to Using the Legislative History of the Fourteenth Amendment to Determine the Amendment’s Original Meaning, 49 Conn. L. Rev. 1069, 1106 (2017).

         [85].     U.S. Const. amend. XIV, § 3.

         [86].     Ruti G. Teitel, Transitional Justice 164 (2000).

         [87].     See id. at 161 (defining transitional justice as “pragmatic resolutions intended as transitory for a particular political period of reconstruction”).

         [88].     Id. at 154.

         [89].     See Ex parte Garland, 71 U.S. 333, 337 (1867). Garland held that the use of the ironclad oath as a qualification rather than a penalty violated the Ex Post Facto and Bill of Attainder Clauses. For historical context, see generally Barry Friedman, The History of the Countermajoritarian Difficulty, Part II: Reconstruction’s Political Court (N.Y.U. Sch. of L. Pub. L. & Legal Theory, Working Paper No. 49, 2002).

         [90].     See In re Griffin, 11 F. Cas. 7, 22–27 (C.C.D. Va. 1869) (No. 5,815) (declining to invalidate a conviction issued by a judge who “was a member of the legislature of Virginia in 1862, during the late Rebellion, and as such voted for measures to sustain the so-called Confederate States in their war against the United States” on the ground that such officials “are not removed therefrom by the direct and immediate effect of the prohibition to hold office” under Section 3).

         [91].     See Magliocca, supra note 9, at 112–13.

         [92].     See id. at 35. Subsequent debates about Section 5, however, did not touch on Section 3. Christopher W. Schmidt, Section 5’s Forgotten Years: Congressional Power to Enforce the Fourteenth Amendment Before Katzenbach v. Morgan, 113 Nw. U. L. Rev. 47, 58–82 (2018) (discussing congressional debates in the early twentieth century, including on anti-lynching and anti-poll tax measures).

         [93].     See Magliocca, supra note 9, at 112–13.

         [94].     Amnesty Act, Pub. L. No. 42-193, 17 Stat. 142 (1872).

         [95].     See Magliocca, supra note 9, at 114–15.

         [96].     See Teitel, supra note 86, at 155–56 (noting the Supreme Court’s rejection of these oaths in two cases: Ex parte Garland and Cummings v. Missouri).

         [97].     See Magliocca, supra note 9, at 124–27.

         [98].     See Magliocca, supra note 25 (“A review of the basic parameters of Section 3 suggests it is the best legal framework available for addressing the extraordinary events at the Capitol with respect to the eligibility of participants to hold public office.”); see also Wagstaffe, supra note 25; Zelikow, supra note 25.

         [99].     U.S. Const. art. II, § 4.

      [100].     A source of continuing disagreement about impeachment concerns its application to conduct for which there is no applicable criminal statute. See, e.g., Raoul Berger, The President, Congress, and the Courts, 83 Yale L.J. 1111, 1139 (1974) (discussing and rejecting President Nixon’s lawyers’ argument that the Impeachment Clause extends only to statutory offenses).

      [101].     Neutrality Act of 1794, Pub. L. No. 3-50, § 1, 1 Stat. 381–82 (1794) (deeming it a “high misdemeanor” to “accept and exercise a commission to serve a foreign prince or state in war by land or sea”). Insurrection is also a state law offense, used in the early twentieth century against political radicals. See, e.g., Herndon v. State, 174 S.E. 597, 612, 615–16 (Ga. 1934).

      [102].     U.S. Const. art. III, § 3 (“Treason against the United States, shall consist only in levying War against them, or in adhering to their Enemies, giving them Aid and Comfort.”).

      [103].     See Charles Warren, What Is Giving Aid and Comfort to the Enemy:, 27 Yale L.J. 331, 333 (1917–1918) (arguing that “giving aid and comfort is generally committed in connection with a war waged against the United States by a foreign power”).

      [104].     See Maggs, supra note 84, at 1105–07.

      [105].     See Teitel, supra note 86, at 155.

      [106].     See Magliocca, supra note 9, at 87.

      [107].     As we explain below, there is an express constitutional power of expulsion from one of the two legislative houses that provides an obvious procedural channel for insurrectionary legislators. See infra Part II.D.

      [108].     See generally State ex rel. White v. Griffin, No. D-101-CV-2022-00473, slip op. (N.M. Santa Fe Jud. Dist. Sept. 6, 2022). The judge in that case made findings to the effect that (1) Griffin took an oath to support the Constitution of the United States, (2) the January 6 attack was an “insurrection” against the Constitution of the United States, and (3) Griffin “engaged in” that insurrection. Id. at 27.

      [109].     U.S. Const. amend. XIV, § 5.

      [110].     See Bd. of Trs. of the Univ. of Ala. v. Garrett, 531 U.S. 356, 365 (2001) (“Accordingly, § 5 legislation reaching beyond the scope of § l’s actual guarantees must exhibit ‘congruence and proportionality between the injury to be prevented or remedied and the means adopted to that end.’” (citation omitted)).

      [111].     See U.S. Const. art I, § 9, cl. 3.

      [112].     See 8 U.S.C. § 1451(e) (2018) (providing for denaturalization when a citizen has been convicted of knowing procurement of naturalization by fraud). The government has the burden of proving fraud in the acquisition of citizenship, and “the facts and the law should be construed as far as is reasonably possible in favor of the citizen.” Schneiderman v. United States, 320 U.S. 118, 122 (1943).

      [113].     CBC News, Barack Obama Greeted with Chants of “Four More Years” at Campaign Rally, YouTube (Oct. 19, 2017), https://www.youtube.com/watch?v=_wIgoZXNjOQ [https://perma.cc/G32X-27YU]. In fact, on the first day of President Obama’s term in office, a resolution was introduced in Congress to amend the Constitution to do away with term limits. See H.R.J. Res. 5, 111th Cong. (2009).

      [114].     U.S. Const. amend. XXII, § 1. The Amendment goes on to say that “no person who has held the office of President, or acted as President, for more than two years of a term to which some other person was elected President shall be elected to the office of the President more than once.” Id.

      [115].     U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779, 783 (1995).

      [116].     Cf. Daniel Hemel, Can Structural Changes Fix the Supreme Court?, 35 J. Econ. Persps. 119, 138 (2021) (discussing and rejecting the arguments in favor of imposing term limits and noting that proposals like term limits “would likely require constitutional amendments”).

      [117].     See The Term-Limited States, Nat’l Conf. of State Legislatures (Nov. 12, 2020), https://www.ncsl.org/research/about-state-legislatures/chart-of-term-limits-states.aspx [https://perma.cc/92EF-VATV]; Which States Have Term Limits on Governor?, U.S. Term Limits, https://www.termlimits.com/which-states-have-term-limits-on-governor/ [https://perma.cc/3XJX-BK58].

      [118].     The Federalist No. 72 (Alexander Hamilton).

      [119].     Kyvig, supra note 79, at 325. Congress did consider imposing presidential term limits many times before they were enacted. See Stephen W. Stathis, The Twenty-Second Amendment: A Practical Remedy or Partisan Maneuver?, 7 Const. Comment. 61, 63–64 (1990) (“By the time the Eightieth Congress convened in 1946, more than two hundred attempts had been made to amend the Constitution and fix the tenure of the president.”).

      [120].     See Stathis, supra note 119, at 72. Stathis argues that party politics played a role and notes that Republicans spearheaded the initiative, with virtually all the opposition coming from parts of the Democratic Party. See id. at 68–69.

      [121].     See Rosalind Dixon & David Landau, Constitutional End Games: Making Presidential Term Limits Stick, 71 Hastings L.J. 359, 372 tbl.1 (2020); Tom Ginsburg & Zachary Elkins, One Size Does Not Fit All: The Provision and Interpretation of Presidential Term Limits, in The Politics of Presidential Term Limits 37, 39 (Alexander Baturo & Robert Elgie eds., 2019) (arguing that “term limits have become more common in presidential systems”).

      [122].     Dixon & Landau, supra note 121, at 372; see also Tom Ginsburg, James Melton & Zachary Elkins, On the Evasion of Executive Term Limits, 52 Wm. & Mary L. Rev. 1807, 1810–15 (2011) [hereinafter Ginsburg et al., On the Evasion of Executive Term Limits] (evaluating the debate over term limits and the success of historical attempts by executives to evade them).

      [123].     See Mila Versteeg, Timothy Horley, Anne Meng, Mauricio Guim & Marilyn Guirguis, The Law and Politics of Presidential Term Limit Evasion, 120 Colum. L. Rev. 173, 176 (2020).

      [124].     David A. Strauss, The Irrelevance of Constitutional Amendments, 114 Harv. L. Rev. 1457, 1494–95 (2001).

      [125].     Or a victory in the popular vote, but not the Electoral College, or vice versa. The range of possible outcomes is daunting to encompass.

      [126].     U.S. Const. art. I, § 5.

      [127].     The Federalist No. 60 (Alexander Hamilton).

      [128].     Powell v. McCormack, 395 U.S. 486, 489 (1969) (holding that age, residency, and citizenship are the only qualifications by which the houses may judge members under their Article I, Section 5 judging powers).

      [129].     Id.

      [130].     Bond v. Floyd, 385 U.S. 116, 137 (1966).

      [131].     U.S. Const. art. I, § 5. Many state legislatures have a similar power to “expel” members on vague grounds with a supermajority vote. In April 2023, for example, the Republican-dominated Tennessee state legislature expelled two Democratic representatives by a two-thirds vote for “disorderly conduct” after they joined a protest on the floor of the state House of Representatives that called for stricter gun control laws. Both were subsequently reappointed to the House to fill the resulting vacancies after a vote by their respective county commissions. See Eliza Fawcett & Emily Cochrane, Tennessee House Expulsions: What You Need to Know, N.Y. Times (May 20, 2023), https://www.nytimes.com/article/tennessee-house-democrats-expulsion-shooting-gun-control.html [https://perma.cc/ZZ68-N7UA]. That same month, the Republican-controlled Arizona House of Representatives voted to expel a Republican representative on a bipartisan vote, also requiring a two-thirds supermajority, after she invited a conspiracy theorist to give testimony before that body. See Neil Vigdor, Arizona House Republicans Expel One of Their Own, N.Y. Times (Apr. 12, 2023), https://www.nytimes.com/2023/04/12/us/politics/liz-harris-expelled-arizona-house-republicans.html [https://perma.cc/46C6-ETAE].

      [132].     1 Asher C. Hinds, Hinds’ Precedents of the House of Representatives §§ 448–452 (1907).

      [133].     For an argument that an expulsion cannot be a basis of a subsequent refusal to seat a member, see Dorian Bowman & Judith Farris Bowman, Article I, Section 5: Congress’ Power to Expel—An Exercise in Self-Restraint, 29 Syracuse L. Rev. 1071, 1090–92 (1978).

      [134].     Indeed, “historical practice is most commonly invoked in connection with debates over the scope of presidential power.” Curtis A. Bradley & Trevor W. Morrison, Historical Gloss and the Separation of Powers, 126 Harv. L. Rev. 411, 417 (2012).

      [135].     Paul Cartledge, Democracy: A Life 70–72 (2016).

      [136].     Id. at 71.

      [137].     Jens Meierhenrich, The Ethics of Lustration, 20 Ethics & Int’l Affs. 99, 99 (2006) (“The practice of lustration ordinarily revolves around, first, the screening of candidates from public office; second, the barring of candidates from public office; and third, the removal of holders of public office.”).

      [138].     See Adam Czarnota, Lustration, Decommunisation and the Rule of Law, 1 Hague J. Rule L. 307, 326–327 (2009).

      [139].     For a summary, see Cynthia M. Horne, International Legal Rulings on Lustration Policies in Central and Eastern Europe: Rule of Law in Historical Context, 34 Law & Soc. Inquiry 713, 716–17 (2009).

      [140].     Natalia Letki, Lustration and Democratisation in East-Central Europe, 54 Eur.-Asia Stud. 529, 539 (2002).

      [141].     Katy Crossley-Frolick, Sifting Through the Past: Lustration in Reunified Germany, in Lustration and Consolidation of Democracy and the Rule of Law in Central and Eastern Europe 197, 208–09 (2007).

      [142].     Meierhenrich, supra note 137, at 101–02.

      [143].     See supra Part I.C.

      [144].     See, e.g., Monika Nalepa, After Authoritarianism: Transitional Justice and Democratic Stability 91–97 (2022) (offering examples from the Eastern European context).

      [145].     See Czarnota, supra note 138, at 319–22, 326–28.

      [146].     See A. James McAdams, Judging the Past in Unified Germany 64–65 (2001).

      [147].     See Czarnota, supra note 138, at 325–26; Roman David, Lustration and Transitional Justice: Personnel Systems in the Czech Republic, Hungary, and Poland 65–91 (2011) (offering a wider account of how both the Hungarian and Romanian systems of lustration worked).

      [148].     See Aleks Szczerbiak, Politicising the Communist Past: The Politics of Truth Revelation in Post-Communist Poland 21, 24 (2018).

      [149].     Horne, supra note 139, at 718.

      [150].     See Eric A. Posner & Adrian Vermeule, Transitional Justice as Ordinary Justice, 117 Harv. L. Rev. 762, 808 (2004); Nalepa, supra note 144, at 253–58 (summarizing findings of empirical analyses of lustration’s effects on democratic quality).

      [151].     See McAdams, supra note 146, at 79.

      [152].     See David, supra note 147, at 74.

      [153].     See Posner & Vermeule, supra note 150, at 777–78.

      [154].     Cf. Aysegul Keskin Zeren, From De-Nazification of Germany to De-Baathification of Iraq, 132 Pol. Sci. Q. 259, 261 (2017) (explaining how de-Baathification in Iraq was justified by comparisons to the Nazi regime).

      [155].     See, e.g., Cherish M. Zinn, Consequences of Iraqi De-Baathification, 9 Cornell Int’l Affs. Rev. 80, 80–81 (2016); James P. Pfiffner, US Blunders in Iraq: De-Baathification and Disbanding the Army, 25 Intel. & Nat’l Sec. 76, 76 (2010).

      [156].     See David, supra note 147, at 232–33; David Pimentel & Brian D. Anderson, Judicial Independence in Postconflict Iraq: Establishing the Rule of Law in an Islamic Constitutional Democracy, 46 Geo. Wash. Int’l L. Rev. 29, 33 (2013) (noting that “lustration can run the risk of interfering with the independence of the judiciary by allowing politically-motivated lustration proceedings to threaten and coerce judges”).

      [157].     See Zeren, supra note 154, at 276–77.

      [158].     Pfiffner, supra note 155, at 76. Much the same dynamic has been observed in Afghanistan. Tricia Bacon & Daniel Byman, De-Talibanization and the Onset of Insurgency in Afghanistan, Stud. Conflict & Terrorism 1, 1–2 (2021).

      [159].     See Czarnota, supra note 138, at 326–27. Note that the Bulgarian law was struck down on other grounds. See id.

      [160].     See David, supra note 147, at 71–72.

      [161].     See id.

      [162].     See, e.g., Trybunal Konstytucyjny [Pol. Const. Tribunal], May 11, 2007, K 2/07 (narrowing a major amendment to the law); Trybunal Konstytucyjny [Pol. Const. Tribunal], May 28, 2003, K 44/02 (striking down an amendment to the law); Trybunal Konstytucyjny [Pol. Const. Tribunal], Oct. 26, 2005, K 31/04 (holding that former security service personnel had a right to see their own files).

      [163].     Chancellery of the Prime Minister, White Paper on the Reform of the Polish Judiciary 13, 18 (2018).

      [164].     McAdams, supra note 146, at 74.

      [165].     Id. at 77.

      [166].     Id. at 75.

      [167].     Szczerbiak, supra note 148, at 21; see Mark S. Ellis, Purging the Past: The Current State of Lustration Laws in the Former Communist Bloc, 59 Law & Contemp. Probs. 181, 192–94 (1996).

      [168].     See Ellis, supra note 167, at 192–94.

      [169].     See id.

      [170].     The two key articles are Karl Loewenstein, Militant Democracy and Fundamental Rights, I, 31 Am. Pol. Sci. Rev. 417 (1937), and Karl Loewenstein, Militant Democracy and Fundamental Rights, II, 31 Am. Pol. Sci. Rev. 638 (1937).

      [171].     Giovanni Capoccia, Militant Democracy: The Institutional Bases of Democratic Self-Preservation, 9 Ann. Rev. L. & Soc. Sci. 207, 207 (2013).

      [172].     See Jan-Werner Muller, Militant Democracy, in The Oxford Handbook of Comparative Constitutional Law 1253, 1258 (Michel Rosenfeld & Andras Sajo eds., 2012) (calling Loewenstein’s militant democracy “highly influential in the Federal Republic of Germany”).

      [173].     Capoccia, supra note 171, at 209 (discussing several examples); Carole J. Petersen, Prohibiting the Hong Kong National Party: Has Hong Kong Violated the International Covenant on Civil and Political Rights?, 48 H.K.L.J. 789, 796 (2018) (discussing South Korean examples).

      [174].     Grundgesetz [GG] [Basic Law] art. 21(2), translation at https://www.gesetze-im-internet.de/englisch_gg/englisch_gg.html [https://perma.cc/9QFP-6BAV].

      [175].     Id. art. 21(1).

      [176].     See id. art. 21(2).

      [177].     BVerfGE, 1 BvB 1/51, Oct. 23, 1952. For discussion, see Samuel Issacharoff, Fragile Democracies, 120 Harv. L. Rev. 1405, 1431–32 (2007).

      [178].     See BVerfGE 5, 85, Aug. 17, 1956.

      [179].     See Thilo Rensmann, Procedural Fairness in a Militant Democracy: The “Uprising of the Decent” Fails Before the Federal Constitutional Court, 4 German L.J. 1117, 1122 (2003).

      [180].     BVerfG, 2 BvB 1/13, Jan. 17, 2017, https://www.bundesverfassungsgericht.de/SharedDocs/Entscheidungen/DE/2017/01/bs20170117_2bvb000113.html [https://perma.cc/US2X-DGAG].

      [181].     Katrin Bennhold, Germany Places Part of Far-Right Party Under Surveillance, N.Y. Times (Mar. 12, 2020), https://www.nytimes.com/2020/03/12/world/europe/germany-afd.html [https://perma.cc/MJ3L-SBFN].

      [182].     Id.

      [183].     Jörg Diehl, Ann-Katrin Müller, Ansgar Siemens & Wolf Wiedmann-Schmidt, German Officials Seek to Turn up the Heat on the AfD, Spiegel Int’l (Mar. 5, 2021), https://www.spiegel.de/international/germany/monitoring-the-right-wing-german-officials-seek-to-turn-up-the-heat-on-the-afd-a-c44c0a38-ee6b-4dc1-ac94-9fbee59b706f [https://perma.cc/BX3K-ST9D].

      [184].     Id.

      [185].     Article 21 also forbids officials from “interven[ing] in the electoral campaign in favor or to the detriment of a political party.” Thomas Kliegel, Freedom of Speech for Public Officials vs. the Political Parties’ Right to Equal Opportunity: The German Constitutional Court’s Recent Rulings Involving the NPD and the AfD, 18 German L.J. 189, 193 (2017). The federal minister of education’s criticism of AfD has been found to violate that rule. See id. at 203–04. The creates an odd dynamic in which ministers cannot criticize a political party, but a domestic intelligence agency can investigate it and then move to have it proscribed.

      [186].     See Gregory H. Fox & Georg Nolte, Intolerant Democracies, 36 Harv. Int’l L.J. 1, 37 (1995) (finding that “some form of party prohibition procedure is common to most democratic systems”).

      [187].     See Tom Ginsburg & Zachary Elkins, Ancillary Powers of Constitutional Courts, 87 Tex. L. Rev. 1431, 1443 tbl.1 (2009).

      [188].     See, e.g., Basic Law: the Knesset § 7A (1958) (excluding from the electoral arena any party that rejects the democratic and Jewish character of the state, as well as any party whose platform is deemed to incite to racism); Víctor Ferreres Comella, The New Regulation of Political Parties in Spain, and the Decision to Outlaw Batasuna, in Militant Democracy 133, 133-34 (András Sajó ed., 2004).

      [189].     Hunbeobjaeopanso [Const. Ct.], Dec. 19, 2014, 26-2 Hun-Da 1 (S. Kor.).

      [190].     Fox & Nolte, supra note 186, at 69 (finding that “the international community is not hopelessly divided on the problem of anti-democratic actors”).

      [191].     See Issacharoff, supra note 177, at 1415; Miguel Schor, Militant Democracy in America, Int’l J. Const. L. Blog (Feb. 16, 2021) [hereinafter Schor, Militant Democracy in America], http://www.iconnectblog.com/militant-democracy-in-america/ [https://perma.cc/T95F-D5UZ] (arguing that “militant democracy runs counter to contemporary American views of free speech”). Even in U.S. law, there is an exception for speech coordinated within foreign terrorist organizations. See Holder v. Humanitarian L. Project, 561 U.S. 1, 40 (2010).

      [192].     Issacharoff, supra note 177, at 1421.

      [193].     See Tom Gerald Daly & Brian Christopher Jones, Parties Versus Democracy: Addressing Today’s Political Party Threats to Democratic Rule, 18 Int’l J. Const. L. 509, 511 (2020); David Landau, Abusive Constitutionalism, 47 U.C. Davis L. Rev. 189, 219–20 (2013).

      [194].     Jongcheol Kim, Dissolution of the Unified Progressive Party Case in Korea: A Critical Review with Reference to the European Court of Human Rights Case Law, 10 J.E. Asia & Int’l L. 139, 149–50 (2017).

      [195].     See Gelijn Molier & Bastiaan Rijpkema, Germany’s New Militant Democracy Regime: National Democratic Party II and the German Federal Constitutional Court’s “Potentiality” Criterion for Party Bans, 14 Eur. Const. L. Rev. 394, 394–95 (2018).

      [196].     See id. at 399.

      [197].     Alexander S. Kirshner, A Theory of Militant Democracy: The Ethics of Combatting Political Extremism 4 (2014).

      [198].     See Paul Pierson & Eric Schickler, Madison’s Constitution Under Stress: A Developmental Analysis of Political Polarization, 23 Ann. Rev. Pol. Sci. 37, 38–39 (2019) (developing that argument for the U.S. context).

      [199].     See Muller, supra note 172, at 1256–62.

      [200].     Hunbeobjaeopanso [Const. Ct.], Dec. 19, 2014, 26-2 Hun-Da 1 (S. Kor.).

      [201].     Id.

      [202].     Refah Partisi v. Turkey, 2003-II Eur. Ct. H.R. 20.

      [203].     See Ginsburg & Elkins, supra note 187, at 1446–49.

      [204].     See Issacharoff, supra note 177, at 1440–42.

      [205].     See Hadep & Demir v. Turkey, App. No. 28003/03, Eur. Ct. H.R. (2010), https://hudoc.echr.coe.int/?i=001-102256 [https://perma.cc/K5DN-YAFQ].

      [206].     See Rebecca Ratcliffe, Thai Court Dissolves Opposition Party Future Forward, Guardian (Feb. 21, 2020), https://amp.theguardian.com/world/2020/feb/21/thai-court-dissolves-opposition-party-future-forward [https://perma.cc/6ML8-AEK7].

      [207].     Ministry of Interior v. Nat’l Rescue Party, Verdict No. 340 (S. Ct. of Cambodia 2017).

      [208].     See Cambodia Top Court Dissolves Main Opposition CNRP Party, BBC News (Nov. 16, 2017), https://www.bbc.com/news/world-asia-42006828 [https://perma.cc/AV3N-T893]; see also Hannah Beech, Cambodia Re-Elects Its Leader, a Result Predetermined by One, N.Y. Times (July 29, 2018), https://www.nytimes.com/2018/07/29/world/asia/cambodia-election-hun-sen.html [https://perma.cc/7LPC-MFFK].

      [209].     See Refah Partisi v. Turkey, 2003-II Eur. Ct. H.R. 20.

      [210].     See id. at 30.

      [211].     See Meltem Müftüler-Baç & E. Fuat Keyman, Turkey Under the AKP: The Era of Dominant-Party Politics, 23 J. Democracy 85, 92 (2012).

      [212].     See Rosalind Dixon & David Landau, Transnational Constitutionalism and a Limited Doctrine of Unconstitutional Constitutional Amendment, 13 Int’l J. Const. L. 606, 615 (2015) (developing the idea of law as a speed bump or deterrent, rather than an absolute prohibition).

      [213].     See Molier & Rijpkema, supra note 195, at 406–08.

      [214].     See id. at 408 (noting that parties in Germany “are highly dependent on state funding” and questioning whether denial of public funding would in fact be a “de facto party ban”).

      [215].     See Issacharoff, supra note 177, at 1449–50 (documenting these differences and observing that “[a]t least in theory, a ban on running for the Knesset is less draconian than outlawing an entire party”).

      [216].     See id.

      [217].     See Ginsburg et al., The Comparative Constitutional Law, supra note 22, at 127–35.

      [218].     Id. at 117.

      [219].     Id. at 127.

      [220].     See id. at 129 (asserting that the “continued tenure” of the head of government is dependent on the lower legislative chambers).

      [221].     See id. at 131–32.

      [222].     Id. at 118.

      [223].     See, e.g., Daehanminkuk Hunbeob [Hunbeob] [Constitution] art. 65(4) (S. Kor.) (“A decision on impeachment shall not extend further than removal from public office.”); Constitución de la República del Paraguay art. 225 (stating that an impeachment trial has “the sole purpose of removing [officials] from office”).

      [224].     See, e.g., Constituição da República de Angola art. 127 (“Conviction shall lead to removal from office and disqualification from standing for another term of office.”); Constitución Política de la República de Chile art. 49(1) (“On the declaration of culpability, the accused [person] is removed from his position and he cannot hold public function, whether of public election or not, for a period of five years.”).

      [225].     See, e.g., Constitución Nacional [Const. Nac.] art. 60 (Arg.).

      [226].     Constituição Federal [C.F.] [Constitution] art. 52 (Braz.). For discussion, see Alexandra Rattinger, The Impeachment Process of Brazil: A Comparative Look at Impeachment in Brazil and the United States, 49 U. Mia. Inter-Am. L. Rev. 129, 155 (2018).

      [227].     See Rattinger, supra note 226, at 155.

      [228].     See id. at 145; Brazil’s Former Leader Rousseff Fails in Bid for Senate Seat, France 24 (Aug. 10, 2018), https://www.france24.com/en/20181008-brazils-former-leader-rousseff-fails-bid-senate-seat [https://perma.cc/R58X-CM8X].

      [229].     See also Powell v. McCormack, 395 U.S. 486, 548 (1969) (illustrating a holding where voting rules proved dispositive).

      [230].     See Aziz Z. Huq, The Constitutional Law of Agenda Control, 104 Calif. L. Rev. 1401, 1413–14 (2016) (analyzing this phenomenon in terms of Arrovian cycling).

      [231].     See Nicas, supra note 21 (noting that Bolsonaro faces fifteen other pending complaints in front of the Electoral Court that could also result in a ban).

      [232].     Basic Law: the Knesset § 7(A) (1958) (amend. 1985). The Knesset also has the power under § 42A (3) of Basic Law, by a three-quarters supermajority vote, to expel a member if they incite racism or support an armed struggle against the State.

      [233].     For discussions, see Gur Bligh, Defending Democracy: A New Understanding of the Party-Banning Phenomenon, 46 Vand. J. Transnat’l L. 1321, 1338 (2013); Rivka Weill, On the Nexus of Eternity Clauses, Proportional Representation, and Banned Political Parties, 16 Election L.J. 237, 244–45 (2017); Suzie Navot, Fighting Terrorism in the Political Arena: The Banning of Political Parties, 14 Party Pol. 745, 756–59 (2008); Raphael Cohen-Almagor, Disqualification of Lists in Israel (1948–1984): Retrospect and Appraisal, 13 Law & Phil. 43, 43–45 (1994).

      [234].     See Yoav Dotan, Impeachment by Judicial Review: Israel’s Odd System of Checks and Balances, 19 Theoretical Inquiries 705, 720 (2018).

      [235].     Id. at 727.

      [236].     Id. at 723.

      [237].     Id. at 733–34.

      [238].     Id. In March 2023, however, the Knesset passed a measure making it more difficult to remove a prime minister. Hadas Gold & Amir Tal, Israel Passes Law Shielding Netanyahu from Being Removed amid Protests over Judicial Changes, CNN (Mar. 24, 2023), https://www.cnn.com/2023/03/23/middleeast/israel-judicial-reforms-prime-minister-law-intl/index.html [https://perma.cc/E9CW-TP52].

      [239].     Data on file with authors.

      [240].     Pakistan Const. art. 62(1)(f) (“A person shall not be qualified to be elected or chosen as a member of Majlis-e-Shoora (Parliament) unless . . . he is sagacious, righteous and non-profligate, honest . . . there being no declaration to the contrary by a court of law.”).

      [241].     Imran Ahmed Khan Niazi v. Muhammad Nawaz Sharif, (2017) PLD (SC) 265 (Pak.).

      [242].     Id.

      [243].     See Kay Johnson & Asif Shahzad, Pakistan Supreme Court Rules Ousted PM Sharif Cannot Lead His Party, Reuters (Feb. 21, 2018), https://www.reuters.com/article/us-pakistan-politics/pakistan-supreme-court-rules-ousted-pm-sharif-cannot-lead-his-party-idUSKCN1G51NX [https://perma.cc/2NCF-94CX].

      [244].     See Sami Ullah Baloch v. Abdul Karim Nousherwani, (2018) PLD (SC) 405 (Pak.).

      [245].     See L. 1952/19, art. 2, enero 28, 2019, Diario Oficial [D.O.] (Colom.).

      [246].     See id. art. 26.

      [247].     See Bogota Mayor Gustavo Petro Sacked and Banned from Office, BBC News (Dec. 10, 2013), https://www.bbc.com/news/world-latin-america-25310963 [https://perma.cc/ML8S-MX6F].

      [248].     See, e.g., Sanción anulada: Petro arrolló a Ordóñez en el Consejo de Estado, Semana (Nov. 15, 2017), https://www.semana.com/nacion/articulo/anulan-sancion-a-petro-en-el-consejo-de-estado/547323/ [https://perma.cc/NV56-R9TU].

      [249].     See Procuraduría General destituyó e inhabilitó por 15 años a Gustavo Petro, El Espectador (Dec. 9, 2013), https://www.elespectador.com/judicial/procuraduria-general-destituyo-e-inhabilito-por-15-anos-a-gustavo-petro-article-463082/ [https://perma.cc/7QGU-E25D].

      [250].     See Petro Urrego v. Colombia, Inter-Am. Ct. H.R. (ser. C) No. 406 (July 8, 2020).

      [251].     See Benjamin Russell, The Evolution of Colombia’s Gustavo Petro, Americas Q. (Apr. 25, 2022), https://www.americasquarterly.org/article/the-evolution-of-colombias-gustavo-petro/ [https://perma.cc/3FVD-DMWM]; Ivan Briscoe, Gustavo Petro’s Big Win: Colombia’s First Leftist President Could Transform the Region, Foreign Affs. (June 19, 2022), https://www.foreignaffairs.com/articles/colombia/2022-06-19/gustavo-petros-big-win [https://perma.cc/NLA9-YDL3].

      [252].     Procuraduría General de la Nación, Informe de Gestión 2019/2020 25 tbls.2-1, 2-2 (2020).

      [253].     Id. at 14 tbl.1-1.

      [254].     See Procuraduría no podrá tumbar judicialmente a elegidos por voto: Corte Constitucional, El Espectador (Feb. 16, 2023), https://www.elespectador.com/judicial/procuraduria-ya-no-podra-destituir-alcaldes-ni-gobernadores-corte-constitucional/ [https://perma.cc/A9SQ-325V].

      [255].     See Pakistan Const. art. 63.

      [256].     Data from Compar. Consts. Project, https://comparativeconstitutionsproject.org/ [https://perma.cc/L2Y4-ANPB].

      [257].     See id. The Dominican Republic and Mexico have the same restriction for the Chief Justice, while eight countries (Afghanistan, Bolivia, Colombia, the Dominican Republic, Honduras, Myanmar, and Panama) do so for the Supreme Court. The Maldives, the Dominican Republic, and Myanmar do the same for all judges. Finally, seven countries (Bolivia, Colombia, Congo, the Dominican Republic, Haiti, Myanmar, and Thailand) foreclose people convicted of felonies from serving on the Constitutional Court. See id.

      [258].     See Mich. Const. art. XI, § 7; Va. Const. art. II, §§ 1, 5 (establishing that people convicted of felonies may not vote or stand for elected state or local office unless their civil rights have been restored).

      [259].     See Powell v. McCormack, 395 U.S. 486, 550 (1969) (holding that the House of Representatives can only exclude an elected member for a reason stated in the Constitution: having at least twenty-five years of age, having been a citizen for seven years, and being an inhabitant of the state they represent).

      [260].     Aziz Z. Huq, Legal or Political Checks on Apex Criminality: An Essay on Constitutional Design, 65 UCLA L. Rev. 1506, 1525 (2018) (describing dynamics that might lead to capture).

      [261].     See, e.g., Ozan O. Varol, Stealth Authoritarianism, 100 Iowa L. Rev. 1673, 1708–09 (2015).

      [262].     See Alexei Nalvany Has Been Disqualified from the Presidential Race, Free Russia (Dec. 25, 2017), https://www.4freerussia.org/alexei-navalny-disqualified-presidential-race/ [https://perma.cc/S8S3-ANLZ].

      [263].     See, e.g., Dotan, supra note 234, at 729 (noting that the Israeli Court has justified its doctrines by a need for “public trust in governmental institutions”).

      [264].     See supra notes 226–230.

      [265].     See Dotan, supra note 234, at 729 (opining that Israeli practice “does not easily resonate with principles of self-government and democratic accountability”); Menaka Guruswamy, Adjudicating “Honesty”: Prime Minister(s) and the Supreme Court of Pakistan, Int’l J. Const. L. Blog (Dec. 5, 2017), iconnectblog.com/adjudicating-honesty-prime-ministers-and-the-supreme-court-of-pakistan-i-connect-column/ [https://perma.cc/HW2N-ZBA7] (arguing that the Pakistani Court’s review came “at the expense of constitutional democracy”).

      [266].     Paul Jacob, From the Voters with Care, in The Politics and Law of Term Limits 27, 38–39 (Edward H. Crane & Roger Pilon eds., 1994); Mark P. Petracca, Restoring “The University in Rotation”: An Essay in Defense of Term Limitation, in The Politics and Law of Term Limits, supra, at 57, 65; Charles W. Stein, The Third Term Tradition: Its Rise and Collapse in American Politics 14–15 (1943).

      [267].     Alexander Baturo, The Stakes of Losing Office, Term Limits and Democracy, 40 Brit. J. Pol. Sci. 635, 635–36 (2010).

      [268].     See Javier Corrales & Michael Penfold, Manipulating Term Limits in Latin America, 25 J. Democracy 157, 165 (2014).

      [269].     See id.; Gideon Maltz, The Case for Presidential Term Limits, 18 J. Democracy 128, 131 (2007).

      [270].     See Bruce E. Cain & Marc A. Levin, Term Limits, 2 Ann. Rev. Pol. Sci. 163, 176–77 (1999); Ginsburg et al., On the Evasion of Executive Term Limits, supra note 122, at 1824.

      [271].     See Thad Kousser, Term Limits and the Dismantling of State Legislative Professionalism 33–38 (2005) (offering acute evidence of this from California).

      [272].     Einer Elhauge, Are Term Limits Undemocratic?, 64 U. Chi. L. Rev. 83, 86 (1997).

      [273].     See Eur. Comm’n for Democracy Through L. (Venice Comm’n), Report on Term Limits 3–4 (2019).

      [274].     The Term-Limited States, Nat’l Conf. of State Legislatures (Nov. 12, 2020), https://www.ncsl.org/research/about-state-legislatures/chart-of-term-limits-states.aspx [https://perma.cc/92EF-VATV].

      [275].     Maltz, supra note 269, at 129–30.

      [276].     See Dixon & Landau, supra note 121, at 372 tbl.1; Ginsburg & Elkins, supra note 121, at 41, 44.

      [277].     Dixon & Landau, supra note 121, at 372 tbl.1.

      [278].     See id.; Ginsburg et al., On the Evasion of Executive Term Limits, supra note 122, at 1836.

      [279].     See Dixon & Landau, supra note 121, at 372 tbl.1 (finding that about 15 percent of systems take this route).

      [280].     See Versteeg et al., supra note 123; Ginsburg et al., On the Evasion of Executive Term Limits, supra note 122, at 1847.

      [281].     See Dixon & Landau, supra note 121, at 363.

      [282].     See Australian Constitution s 44 (“Any person who is under any acknowledgment of allegiance, obedience, or adherence to a foreign power, or is a subject or a citizen or entitled to the rights or privileges of a subject or a citizen of a foreign power . . . shall be incapable of being chosen or of sitting as a senator or a member of the House of Representatives.”).

      [283].     Re Canavan (2017) 263 CLR 284 (Austl.).

      [284].     See, e.g., Const. of Uganda, art. 102 (“A person is not qualified for election as President unless that person is– . . . (b) not less than thirty-five years and not more than seventy-five years of age.”).

      [285].     See, e.g., Alysia Blackham, Judges and Retirement Ages, 39 Melb. U. L. Rev. 738, 740 (2016).

      [286].     See, e.g., Mandana Naini, Iran’s Second Chamber? The Guardian Council, 12 J. Leg. Stud. 198, 203 (2006).

      [287].     See Iran Election 2021: Seven Candidates Qualified to Run for President, IFP News (May 25, 2021), https://ifpnews.com/iran-elections-2021-seven-candidates-qualified-to-run-for-president/ [https://perma.cc/9QXB-FGKU].

      [288].     Iran Hardliners Bar Dozens of Current Lawmakers from Running Again, Radio Farda (Jan. 12, 2020), https://en.radiofarda.com/a/iran-hardliners-bar-dozens-of-current-lawmakers-from-running-again/30372980.html [https://perma.cc/D2VF-YUE2].

      [289].     See, e.g., Issacharoff, supra note 177, at 1410–11 (referring to the “intuition that democratic elections require, as a precondition to the right of participation, a commitment to the preservation of the democratic process”); Fox & Nolte, supra note 186, at 51; Refah Partisi v. Turkey, 2003-II Eur. Ct. H.R. 20, 31 (“It necessarily follows that a political party whose leaders incite to violence or put forward a policy which fails to respect democracy or which is aimed at the destruction of democracy and the flouting of the rights and freedoms recognized in a democracy cannot lay claim to the Convention’s protection.”).

      [290].     See Ginsburg & Huq, supra note 43, at 19; Rosalind Dixon & David Landau, Competitive Democracy and the Constitutional Minimum Core, in Assessing Constitutional Performance 268, 277 (Tom Ginsburg & Aziz Huq eds., 2016).

      [291].     See Aziz Huq & Tom Ginsburg, How to Lose a Constitutional Democracy, 65 UCLA L. Rev. 78, 93–94 (2018).

      [292].     See Steven Levitsky & Lucan A. Way, Competitive Authoritarianism: Hybrid Regimes After the Cold War (2010).

      [293].     See Daly & Jones, supra note 193, at 520; Landau, supra note 193, at 219–20.

      [294].     See Versteeg et al., supra note 123, at 187–88; Ginsburg & Elkins, supra note 121, at 39. Term limit provisions can and sometimes are reinterpreted by courts, for example, to deny them retroactive effect so that presidents can stay in office longer. See, e.g., Versteeg et al., supra note 123, at 221.

      [295].     See Dixon & Landau, supra note 212, at 614; Samuel Issacharoff, Constitutional Courts and Democratic Hedging, 99 Geo. L.J. 961, 1002 (2011) (citing this as a justification for the judge-made unconstitutional constitutional amendment doctrine).

      [296].     GG [Basic Law] art. 21.

      [297].     U.S. Const. art. II, § 4.

      [298].     See generally Manza et al., supra note 62, at 41–68 (outlining the discriminatory history of felony disenfranchisement and disqualification rules).

      [299].     See generally Joel Feinberg, Collective Responsibility, 65 J. Phil. 674 (1968) (arguing against group accountability).

      [300].     See Teitel, supra note 86, at 163–64.

      [301].     See David, supra note 147, at 228.

      [302].     See, e.g., Fox & Nolte, supra note 186, at 51–52.

      [303].     See supra Part II.C.

      [304].     See supra note 180.

      [305].     See supra Part II.C.

      [306].     See, e.g., McAdams, supra note 146, at 77–83; Szczerbiak, supra note 148, at 45–46.

      [307].     See Muller, supra note 172.

      [308].     See Hunbeobjaeopanso [Const. Ct.], Dec. 19, 2014, 26-2 Hun-Da 1 (S. Kor.).

      [309].     We do not mean to say, of course, that the application of these rules will be the same across all contexts.

      [310].     See, e.g., Mark A. Graber, Sanford Levinson & Mark Tushnet, Constitutional Democracy in Crisis? (2018) (presenting essays on constitutional crisis in established democracies); Tarunabh Khaitan, Killing a Constitution with a Thousand Cuts: Executive Aggrandizement and Party-State Fusion in India, 14 Law & Ethics Hum. Rts. 49, 52 (2020); Jacques Rupnik, Explaining Eastern Europe: The Crisis of Liberalism, 29 J. Democracy 24, 25 (2018); Madhav Khosla & Milan Vaishnav, The Three Faces of the Indian State, 32 J. Democracy 111, 111 (2021).

      [311].     See Schor, Militant Democracy in America, supra note 191.

      [312].     There are parallels to the choice between courts and agencies, but disqualification raises distinctive concerns. See generally Matthew C. Stephenson, Legislative Allocation of Delegated Power: Uncertainty, Risk, and the Choice Between Agencies and Courts, 119 Harv. L. Rev. 1035 (2006) (describing the conditions under which rational legislators delegate interpretive authority to agencies rather than courts).

      [313].     See, e.g., Paul Blumenthal, A Broken Party Acquitted Donald Trump in His Second Impeachment, Huffington Post (Feb. 13, 2021), https://www.huffpost.com/entry/donald-trump-impeachment-acquit_n_6027e114c5b6f88289fbfc0f [https://perma.cc/9BCE-7DVG].

      [314].     See Szczerbiak, supra note 148, at 30–31.

      [315].     See Teitel, supra note 86, at 154–57; Magliocca, supra note 9, at 112–17.

      [316].     See supra Part II.D.

      [317].     See Szczerbiak, supra note 148, at 35.

      [318].     See supra Part IV.D.

      [319].     See id.

      [320].     Toygar Sinan Baykan, The Justice and Development Party in Turkey: Populism, Personalism, Organization 1–8, 32–38 (2018) (offering a summary of historical changes in Turkey’s governing party).

      [321].     See, e.g., McAdams, supra note 146, at 85.

      [322].     See David, supra note 147, at 71–72.

      [323].     See generally Adrian Vermeule, System Effects and the Constitution, 123 Harv. L. Rev. 4 (2009), for a discussion about the dangers of failing to account for system effects.

      [324].     Hunbeob [Constitution] arts. 65, 111 (S. Kor.).

      [325].     See Youngjae Lee, Law, Politics, and Impeachment: The Impeachment of Roh Moo-hyun from a Comparative Constitutional Perspective, 53 Am. J. Compar. L. 403, 414 (2005).

      [326].     See Kim, supra note 194, at 148–50.

      [327].     See supra Part II.D.

      [328].     See supra text accompanying notes 234–239.

      [329].     See Dotan, supra note 234, at 733–34.

      [330].     See Pakistan Const. art. 63(g)–(h).

      [331].     See Sami Ullah Baloch v. Abdul Karim Nousherwani, (2018) PLD (SC) 405 (Pak.).

      [332].     H.R. Res. 24, 177th Cong. (2021).

      [333].     Bill Chappell, Canada Lists Proud Boys as a Terrorist Group, Alongside ISIS and Al-Qaida, NPR (Feb. 3, 2021), https://www.npr.org/2021/02/03/963682181/canada-lists-proud-boys-as-a-terrorist-group-alongside-isis-and-al-qaida [https://perma.cc/3Y2Y-BMHR].

      [334].     See, e.g., Communist Control Act of 1954, Pub. L. No. 83-637, 68 Stat. 775 (outlawing the Communist Party of the United States and criminalizing membership in it); Whitney v. California, 274 U.S. 357, 372 (1927) (unanimously upholding the conviction of a member of the Communist Labor Party of California under that state’s Criminal Syndicalism Act).

      [335].     See Aziz Z. Huq, Preserving Political Speech from Ourselves and Others, 112 Colum. L. Rev. Sidebar 16, 16–17 (2012) (discussing the legal regime for such designations).

      [336].     Brian Michael Jenkins, Five Reasons to Be Wary of a New Domestic Terrorism Law, RAND Blog (Feb. 24, 2021), https://www.rand.org/blog/2021/02/five-reasons-to-be-wary-of-a-new-domestic-terrorism.html [https://perma.cc/T5UF-BS48]. The most obviously relevant decision is Brandenburg v. Ohio, 395 U.S. 444 (1969).

      [337].     See Schor, Militant Democracy in America, supra note 191; Miguel Schor, Trumpism and the Continuing Challenges to Three Political-Constitutionalist Orthodoxies [hereinafter Schor, Trumpism and the Continuing Challenges] (June 2021) (unpublished article) (on file with authors).

      [338].     See Schor, Trumpism and the Continuing Challenges, supra note 337.

      [339].     See id.

      [340].     See supra text accompanying notes 192–202.

      [341].     Ministry of Interior v. Nat’l Rescue Party, Verdict No. 340 (S. Ct. of Cambodia 2017); see Hadep & Demir v. Turkey, App. No. 28003/03, Eur. Ct. H.R. (2010), https://hudoc.echr.coe.int/?i=001-102256 [https://perma.cc/K5DN-YAFQ].

      [342].     See Daly & Jones, supra note 193, at 532.

      [343].     U.S. Const. amend. XIV, § 5.

      [344].     See In re Griffin, 11 F. Cas. 7 (C.C.D. Va. 1869) (No. 5,815).

      [345].     U.S. Const. art. II, § 4.

      [346].     See Stephen M. Griffin, Presidential Impeachment in Tribal Times: The Historical Logic of Informal Constitutional Change, 51 Conn. L. Rev. 413, 423–25 (2019) (recounting historical developments).

      [347].     See Trial Memorandum of President Donald J. Trump, S. Doc. No. 116-12, pt. III, at 145 (2020).

      [348].     See, e.g., Cass R. Sunstein, Impeachment: A Citizen’s Guide 56 (2017); Charles L. Black, Jr. & Philip Bobbitt, Impeachment: A Handbook 36 (2018); Michael J. Gerhardt, The Federal Impeachment Process: A Constitutional and Historical Analysis 105 (3d ed. 2019); Laurence H. Tribe & Joshua Matz, To End a Presidency: The Power of Impeachment 42 (2018); Ginsburg et al., The Comparative Constitutional Law, supra note 22, at 89–90.

      [349].     See Enforcement Act, ch. 114, § 14, 16 Stat. 140 (1870).

      [350].     Neal Devins & Lawrence Baum, Split Definitive: How Party Polarization Turned the Supreme Court into a Partisan Court, 2016 S. Ct. Rev. 301, 301 (2016).

      [351].     U.S. Const. amend. XIV, § 3 (“Congress may by a vote of two-thirds of each House, remove such disability.”).

      [352].     The Federalist No. 72 (Alexander Hamilton).

      [353].     Versteeg et al., supra note 123, at 184–85; Ginsburg et al., On the Evasion of Executive Term Limits, supra note 122, at 1819–21.

      [354].     See Dixon & Landau, supra note 121, at 372 tbl.1 (2020).

      [355].     See Andrew Solender, Trump Says He Will “Negotiate” Third Term Because He’s “Entitled” to It, Forbes (Sept. 13, 2020), https://www.forbes.com/sites/andrewsolender/2020/09/13/trump-says-he-will-negotiate-third-term-because-hes-entitled-to-it/ [https://perma.cc/CD6J-Y5R5].

      [356].     See id.

      [357].     See Versteeg et al., supra note 123, at 179; Rosalind Dixon & David Landau, Tiered Constitutional Design, 86 Geo. Wash. L. Rev. 438, 507–11 (2018).

      [358].     See Strauss, supra note 124, at 1494–95.

      [359].     See, e.g., Williams v. Rhodes, 393 U.S. 23, 35 (1968) (affirming order to this effect).

      [360].     See, e.g., Republican Nat’l Comm. v. Democratic Nat’l Comm., 140 S. Ct. 1205, 1208 (2020) (ordering Wisconsin not to count ballots received, postmarked, or delivered after Election Day).

      [361].     See supra text accompanying notes 234–251.

      [362].     See Dixon & Landau, supra note 121, at 363 (discussing how non-permanent presidential term limits may increase incentives to comply in some contexts).