Narrow but Deep: The McCulloch Principle, Collective-Action Theory, and Section Three Enforcement

Neil S. Siegel* 

Abstract

In Trump v. Anderson, 144 S. Ct. 662 (2024), the Supreme Court of the United States held that the Colorado Supreme Court erred in excluding former President Donald J. Trump from the Republican Party’s primary ballot in the state. The Court reasoned that the Constitution makes Congress, not the states, solely responsible for enforcing Section 3 of the Fourteenth Amendment. Scholars of Section 3 have persuasively argued that Section 3 is self-executing, so the Court’s rationale lacks a sound basis in the original or contemporary meaning of the text of the Civil War Amendments, the original intent of their drafters, or the Court’s own precedent interpreting them. This Article nonetheless argues that the Court’s judgment is justifiable on structural grounds identified in the author’s recent book, The Collective-Action Constitution (2024). As envisioned in McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316 (1819), the Constitution’s federal structure bars states from enforcing Section 3 against candidates for President or Vice President, at least if they enjoy substantial support within their own political party. More than two centuries ago, McCulloch articulated a structural principle that disables states from causing multistate collective-action problems by interfering with a function of the national governing process. That structural, collective-action principle extends in parallel fashion to actions by states that interfere excessively with a function of the national political process. The Presidency, along with the Vice Presidency, is a uniquely national office because all states, and all voters in states, play a role in determining who will run for that office and ultimately occupy it. Just as “a part” may not tax “the whole” because the whole is not represented in the part, so a part may not make presidential eligibility decisions that significantly undermine the capacity of the whole to determine who will represent it in the White House. Legal scholars can justly criticize much of the Court’s reasoning in Trump v. Anderson, but not the result that it reached.

Introduction

In Trump v. Anderson, some Republican and unaffiliated voters in Colorado argued that former President Donald J. Trump was constitutionally disqualified from becoming President again by Section 3 of the Fourteenth Amendment.[1] Section 3 prohibits, among other individuals, oath-breaking insurrectionists from holding any civil office under the United States:

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.[2]

The Colorado Supreme Court agreed with the plaintiffs that Section 3 barred Trump from re-ascending to the Presidency and ordered the Colorado secretary of state to exclude him from the Republican Party’s primary ballot in the state.[3]

The U.S. Supreme Court reversed. All nine Justices agreed that states were not permitted to enforce Section 3 against presidential candidates. It was not clear in all of the opinions, however, what specific constitutional provision or principle state enforcement of Section 3 would violate.[4] Moreover, five Justices further held that “responsibility for enforcing Section 3 against federal officeholders and candidates rests with Congress and not the States,” and that “Section 5 [of the Fourteenth Amendment] vests in Congress the power to enforce it.”[5] In so holding, the Court in Trump v. Anderson decided more than it had to—or should have. The rationale for precluding states from enforcing Section 3 against presidential and Vice Presidential candidates like Trump—whatever it is, exactly—may or may not apply to candidates for the U.S. Senate or House of Representatives, yet the Court’s opinion speaks of “federal officeholders and candidates” generally, with no apparent limitation.[6]

More importantly, Section 3 experts have persuasively argued—skillfully and at length—that the provision is self-executing.[7] The Court’s reasoning to the contrary finds scant support in the constitutional text of the Civil War Amendments, the original intent of their drafters, or the Court’s precedent interpreting them. As Professors William Baude and Michael Stokes Paulsen have observed, “Section Three’s language is language of automatic legal effect: ‘No person shall be’ directly enacts the officeholding bar it describes where its rule is satisfied.”[8] Textually, Section 3 has immediate, self-executing effect as a constitutional rule.

Professor Mark Graber, whose knowledge of the work of the drafters of the Reconstruction Amendments is stunning, observes that “contemporary questions, such as the precise means for implementing Section 3, did not explicitly arise when the 39th Congress was considering the Fourteenth Amendment,” but that “[w]hat commentary exists suggests that members of Congress assumed all provisions of the Fourteenth Amendment were self-executing,” and “Section 3 was no exception.”[9] Referring to Chief Justice Salmon P. Chase’s decision on circuit in Griffin’s Case,[10] Professor Graber notes that “[o]ne important precedent suggests that Section 3 was not self-executing, but that precedent is inconsistent with the general practice at the time and inconsistent with the past practice of the justice who wrote the opinion.”[11] Professor Gerard Magliocca, who has worked on the history of Section 3 and Griffin’s Case in particular, came to the same conclusion.[12]

The Court itself long ago answered the question of self-execution in line with the text and original intent. Referring to the Thirteenth Amendment in 1883, the Court stated that “[t]his amendment, as well as the Fourteenth, is undoubtedly self-executing without any ancillary legislation, so far as its terms are applicable to any existing state of circumstances.”[13] That is dicta, but it is dicta that has almost universally been viewed as correct before and after the Court uttered it. No one doubts that, to borrow the language used by the Court in the Civil Rights Cases regarding Section 1 of the Thirteenth Amendment, “[b]y its own unaided force,”[14] Section 1 of the Fourteenth Amendment imposes its commands on the states—and on the federal government, which is bound by the Citizenship Clause. Nothing in the language of Section 3—or Section 4—suggests that it should be treated differently.

Finally, the Court correctly pointed out that Section 5 of the Fourteenth Amendment “empowers Congress to prescribe” procedures for enforcing Section 3.[15] For several reasons, however, it is implausible to infer from Section 5’s expansion of congressional authority that Section 3 lacks legal effect until Congress acts. First, the Court’s inference does not make sense of the language of Section 5: the “power to enforce, by appropriate legislation, the provisions of this article” draws no distinction between Section 1—which is undoubtedly self-executing—and Section 3. Further, such an enforcement power presupposes that “the provisions” are already rules of law. Section 5 expanded the scope of the Fourteenth Amendment; it did not render Section 3 inoperative by Section 5’s inclusion.[16]

Second, the Court drew an inference from the existence of an enforcement provision that no one would accept in the parallel instances of Section 1 of each Civil War Amendment. As the Court in the Civil Rights Cases wrote of the Thirteenth Amendment, “legislation may be necessary and proper to meet all the various cases and circumstances to be affected by [a self-executing constitutional rule], and to prescribe proper modes of redress for its violation in letter or spirit.”[17] But it does not follow that the existence of enforcement legislation is a condition precedent to the application of the rule itself.

Third, to concede that the commands set out in Section 1 of the Fourteenth Amendment control government behavior regardless of congressional action, while maintaining that Section 3 is operative only so long as Congress has enacted—and not repealed—enforcement legislation, is to misunderstand the objectives of the Amendment’s drafters. As Professor Graber writes, its Republican supporters in Congress—leadership and “backbenchers” alike—were driven by “an overarching concern with preventing rebel rule,” and they understood that a future Congress dominated by ex-Confederates and their Northern Democratic allies could undo anything a Republican Congress enacted.[18]

In sum, other scholars have demonstrated that textualist, originalist, and doctrinal modalities of constitutional interpretation do not justify the result that the Court reached in Trump v. Anderson. Those are not, however, the only modalities of constitutional interpretation. This Article demonstrates that the result is defensible on a different ground. A structural approach provides an alternative, independently sufficient, and narrow but deep ground of decision.[19] As envisioned in McCulloch v. Maryland,[20] the Constitution’s federal structure bars states from enforcing Section 3 against candidates for President or Vice President, at least if they enjoy substantial support within their own political party. More than two centuries ago, McCulloch articulated a structural principle that disables states from causing multistate collective-action problems by interfering with a function of the national governing process—in McCulloch, the proper functioning of the branches of the Second National Bank within the territory of individual states. That structural, collective-action principle extends in parallel fashion to actions by states that interfere excessively with a function of the national political process. The Presidency, along with the Vice-Presidency, is a uniquely national office because all states, and all voters in states, play a role in determining who will run for that office and ultimately occupy it. Just as “a part” may not tax “the whole” because the whole is not represented in the part,[21] so a part may not make presidential eligibility decisions that significantly undermine the capacity of the whole to determine who will represent it in the White House.

All nine Justices in Trump v. Anderson shared the structural intuition that “federalism principles” prohibited what Colorado did.[22] This Article supplies the structural theory that validates their intuition and so responds to scholarly critics of the decision.[23] Justice Amy Coney Barrett came close to nailing it when she concluded that “States lack the power to enforce Section 3 against Presidential candidates” and “[t]hat principle is sufficient to resolve this case.”[24]

Part I articulates the structural, collective-action principle identified by the Court in McCulloch and reflected in many provisions of the constitutional text. Part II explains why that principle bars states from enforcing Section 3 against presidential and vice presidential candidates who at least enjoy significant support within their political party, but not necessarily against candidates for other federal offices. Part III anticipates legal and prudential objections. The Conclusion identifies an unfortunate consequence of the Court’s failure to grasp the structural argument presented here and discusses the argument’s broader implications for the relationship between the original Constitution and the post-Civil War Constitution.

1. McCulloch’s Structural,
   Collective-Action Principle

This Part begins the work of explicating a structural, collective-action account of Section 3’s enforcement. It does so by identifying, generalizing, and modernizing the structural principle limiting state authority that Chief Justice John Marshall invoked in McCulloch v. Maryland.[25] Part II applies that principle to Section 3 in the case of state efforts to enforce it against presidential candidates.

1. Marshall’s Structural Analysis

McCulloch began as a lawsuit filed by Maryland against James W. McCulloch, the cashier of the Maryland branch of the Second National Bank. Maryland sued in state court to collect a $100 fine ($2,531.22 in 2025 dollars), which was the penalty for circulating a bank note without the required Maryland stamp. The penalty could be avoided if the bank paid a $15,000 annual tax to the state ($379,683.07 today).[26] McCulloch became a test case for the constitutionality of the bank. Moreover, if the Court upheld the constitutionality of the bank, McCulloch would also be a test case for the authority of states to tax branches of the bank within their borders.

Legal historian H. Jefferson Powell has justly called Chief Justice Marshall’s opinion for the Court in McCulloch, which broadly upheld Congress’s power to create the bank and categorically barred states from taxing it, “the greatest opinion written by the single greatest figure in the history of constitutional law.”[27] After interpreting Congress’s power broadly but not limitlessly, thereby enabling the federal government to solve collective-action problems for the states, Marshall interpreted state power narrowly in the setting at issue, thereby preventing states from harming the federal government or causing certain collective-action problems for one another.[28] Regarding the second holding, Marshall concluded for the Court that states may not tax federal institutions within their borders, even though taxation is a concurrent power and nothing in Article I, Section 10 (which contains certain restrictions on state taxation authority)[29] prohibits states from doing so. “There is no express provision for the case,” Marshall acknowledged.[30] But he offered two other arguments. The first was a questionable syllogism, and the second was a promising structural argument.

The three premises of the syllogism were that (1) the power to create was the power to preserve; (2) the power to destroy was incompatible with the power to preserve; and (3) the power to tax was the power to destroy. The conclusion that followed from those premises was that state power to tax the bank was incompatible with federal power to create it, which Marshall established earlier in the opinion. Moreover, given principles of federal supremacy (and the Supremacy Clause),[31] which he also established earlier, his final conclusion was that federal power to create the bank trumped state power to tax its branches.[32]

Marshall’s logic was not vulnerable, but one premise was: the power to tax was not necessarily the power to destroy, particularly given the possibility of judicial review of exercises of the taxing power. Maryland’s tax evidenced hostility toward the bank, but the tax did not destroy it, and more than a century later, Justice Oliver Wendel Holmes, Jr., would memorably write in dissent in Panhandle Oil Company v. Mississippi ex rel. Knox that “[t]he power to tax is not the power to destroy while this Court sits.”[33] Granted, Marshall was writing during a period of state hostility to the bank, and so he may have wanted to nip the problem in the bud out of concern that states could not be trusted to tax nondestructively.[34] But as shown below, that concern was better captured by his structural argument. Marshall also deemed inappropriate judicial efforts to determine when uses of the taxing power were excessive. “We are not driven to the perplexing inquiry, so unfit for the judicial department,” he wrote, “what degree of taxation is the legitimate use, and what degree may amount to the abuse of the power.”[35] As also explained below, however, the modern view is that judicial power can appropriately consider questions of degree and impact. For example, in NFIB v. Sebelius, the modern Court conducted such an inquiry regarding uses of federal taxation authority.[36]

Marshall’s structural argument was more persuasive. “The only security against the abuse of [the taxing] power, is found, in the structure of the Government itself,” he asserted, for “[i]n imposing a tax, the legislature acts upon its constituents,” who can hold their representatives accountable for the taxes they impose.[37] Americans who reside outside a state are not represented in the state’s legislature, however, so “the means employed by the government of the Union have no such security, nor is the right of a state to tax them sustained by the same theory.”[38]

Later in his opinion, Marshall insisted that each state is only “a part” of “the whole,” and a part may not tax the whole, even non-destructively.[39] As a matter of democratic principle, he reasoned, no legislature in America was permitted to target individuals and entities not represented in the political process that elected the legislators who imposed the tax. To restate his structural argument as an ethos argument from the American Revolution, no taxation without representation. His “intelligible standard” was that state taxation authority extends only to “[a]ll subjects over which the sovereign power of a State extends.”[40] His argument anticipated the process theory of judicial review by well more than a century. The Court proposed the basic idea in its famous footnote four in United States v. Carolene Products Company,[41] and Professor John Hart Ely developed it decades later into a theory of judicial review.[42]

1. Generalizing Marshall’s Analysis

Collective-action reasoning supports Marshall’s structural analysis of when state taxation is democratically illegitimate and when it is legitimate. A state internalizes all the benefits of taxing federal institutions because the taxing state obtains all the tax revenue from doing so and reduces potential federal competition with its institutions only, not with those in other states. By contrast, the taxing state externalizes most of the costs—the impaired functioning of the federal institutions—onto sister states because those institutions serve the whole nation. Because the taxing state internalizes all the benefits of taxing federal institutions and externalizes most of the costs, it is incentivized to overtax federal institutions. Further, all states are similarly situated. But if they all externalize costs in that way, all of them will likely end up worse off from their own perspectives than if none of them do so. Even if only some states interfere with a function of the national governing process by taxing a federal institution, it still contravenes structural, collective-action principles of federalism protecting that process for states to externalize costs onto sister states that are greater than the benefits they are internalizing.

More precisely, multistate collective-action problems take three basic forms.[43] First, cooperation problems arise when all members of a group of states would be better off by their own estimations if every member cooperated than if no member cooperated, but some or all group members prefer not to cooperate while others do. The Prisoners’ Dilemma, a famous example of that kind of collective-action problem, captures situations in which one group of states “free rides” off the contributions of another group of states to collective action or the two groups “race to the bottom” after some states disadvantage themselves relative to others by regulating businesses or individuals in ways that other states do not. Second, coordination problems arise when some or all states would need to coordinate their behavior to solve a problem but there are multiple ways of doing so and there may be disagreements about how to do so. For example, creating national networks of transportation and communication would require the states to coordinate their regulatory behavior.

Those classic collective-action problems of game theory can be called “Pareto collective-action problems” because all states would be better off, or at least not worse off, by their own estimations, if collective action succeeded. Given the number of states and the extraordinarily demanding theoretical requirement (called Pareto optimality or efficiency) that all states be no worse off, Congress would almost never be able to act if it were authorized to solve only Pareto collective-action problems.[44]

By contrast, a third category of collective-action problems, which the author has called “cost-benefit collective-action problems,” refers to situations in which some states would regard themselves as far better off if collective action succeeded but other states would deem themselves somewhat worse off. Thus, solving such a collective-action problem would help the first group of states more than it would harm the second group. That requirement is called cost-benefit efficiency (or Kaldor-Hicks efficiency). An outcome is cost-benefit efficient if it produces more benefits than costs.[45] Almost all multistate collective-action problems in the history of U.S. federalism are of that variety. For one early, dramatic example, Rhode Island regarded itself as sufficiently worse off by constitutional reform in the 1780s that it boycotted the Constitutional Convention and (along with North Carolina) still had not ratified the Constitution when George Washington was inaugurated as President. Rhode Island’s opposition did not prevent the rest of the states from moving forward with reform and making themselves far better off than Rhode Island was worse off.[46]

In sum, collective-action problems that harm all states are Pareto inefficiencies, whereas collective-action problems that harm some states more than they benefit other states are cost-benefit inefficiencies. When this Article refers to multistate collective-action problems, it includes both Pareto and cost-benefit collective-action problems. States are not permitted to undermine a function of the national governing process by causing either kind of multistate collective-action problem.

Similar collective-action logics help explain other constitutional provisions and principles that constrain state power. For example, even when a state taxes imports and exports because doing so is “absolutely necessary” to executing its inspection laws, “the net Produce” of such taxes “shall be for the Use of the Treasury of the United States,”[47] thereby preventing the taxing state from making a profit. Moreover, all such state laws “shall be subject to the Revision and Control of the Congress,”[48] thereby empowering Congress—where all states are represented—to further protect against cost externalization by states onto sister states. Similarly, the dormant commerce doctrine usually prevents states from causing collective-action problems by taxing or regulating in ways that discriminate against interstate commerce or that unduly burden it.[49]

Likewise, most provisions in Article IV, including the Full Faith and Credit Clause,[50] the Privileges and Immunities Clause,[51] the Extradition Clause,[52] the Admission Clause,[53] the Territory Clause,[54] and the Republican Form of Government Clause[55] help prevent states from “racing to the bottom” by harming one another or the federal government, among other potential collective-action problems.[56] For example, states that routinely disrespected the judicial judgments of sister states, or that treated visitors from sister states like unwelcome foreigners, or that refused extradition requests from sister states, would almost certainly provoke retaliation, potentially making all or most of the states involved worse off from their own perspectives. And the Republican Form Clause prohibits state governments from being organized as monarchies in part because of the concern, prominent at the Founding, that monarchies were more likely than democracies to externalize massive costs by going to war against sister states or the federal government. An ancient example invoked during the ratification debates was the despotic king of Macedon, who first used political cunning to get admitted to the joint government of the Greek city-states and then gained control of each of them.[57] Whatever benefits the king secured for Macedon (or, more likely, for himself) were swamped by the costs that he imposed on other city-states by destroying their ability to govern themselves. The no-monarchies rule imposed by the Republican Form Clause prohibits such cost-benefit inefficiency and so reflects the McCulloch principle’s condemnation of state behavior that externalizes costs onto sister states that exceed the benefits it secures for itself.

1. Modernizing Marshall’s Analysis

A modern court applying the structural, McCulloch principle to an issue not firmly settled by precedent (as the bar on state taxation of the federal government basically is) would need to translate the principle to the modern context of American law. In particular, a contemporary court should follow Holmes, not Marshall, in deciding whether the state interference or impact on the federal matter at issue was severe enough to warrant invalidation on McCulloch grounds. That is, the court would rightly look to the degree of interference, as Holmes wanted to do in Panhandle Oil Company, rather than simply treat the issue as a binary, yes/no matter. As Holmes noted in that case, Marshall’s premise rested on a Founding-era assumption that questions of degree are ordinarily non-justiciable:

It seems to me that the State Court was right. I should say plainly right, but for the effect of certain dicta of Chief Justice Marshall which culminated in or rather were founded upon his often quoted proposition that the power to tax is the power to destroy. In those days it was not recognized as it is today that most of the distinctions of the law are distinctions of degree. If the States had any power it was assumed that they had all power, and that the necessary alternative was to deny it altogether. But this Court which so often has defeated the attempt to tax in certain ways can defeat an attempt to discriminate or otherwise go too far without wholly abolishing the power to tax. The power to tax is not the power to destroy while this Court sits. The power to fix rates is the power to destroy if unlimited, but this Court while it endeavors to prevent confiscation does not prevent the fixing of rates. A tax is not an unconstitutional regulation in every case where an absolute prohibition of sales would be one.[58]

Later American jurists and lawyers would come to reject the assumption that “[i]f the States had any power . . . they had all power.”[59]

Although the degree of interference is relevant to the constitutional inquiry, such questions can be difficult and involve judgment calls. Because Congress possesses greater democratic legitimacy and institutional competence than the federal courts, Congress has greater leeway to prohibit states from interfering with federal programs. When states disagree, collective-action problems do not simply exist or not in a technical, scientific way. Cost-benefit collective-action problems have a certain objective structure, but their existence and significance require assessing the extent to which states are externalizing costs that are greater than the benefits they are internalizing. The assessor that matters most for constitutional purposes is either the Constitution itself or the government institution with the most democratic legitimacy to make such judgment calls. That institution is Congress—the first branch of government—where all states and all individuals are represented, in contrast to state governments, where only one state and some Americans are represented. In McCulloch, Chief Justice Marshall explained that key difference between the democratic legitimacy of the states and the people collectively in Congress and the democratic legitimacy of the states individually outside it.[60] Congress is also more broadly representative of all the states and all the people than is the presidency, which does not balance interests and include both political parties at a given time to anywhere near the same extent that Congress does.[61] And federal judges, who are appointed with life tenure rather than elected and re-elected, possess substantially less democratic legitimacy than the presidency. It follows that Congress enjoys greater discretion than the federal courts to prohibit states from interfering with federal institutions and programs.

Similarly, Congress possesses greater power under the Interstate Commerce Clause to regulate activities that affect interstate commerce than the Court enjoys under the dormant commerce doctrine. The Court will uphold non-discriminatory state regulations protecting public health, traffic safety, the local environment, or state roads if it concludes, after relatively deferential judicial review, that the in-state benefits exceed the burdens on interstate commerce. By contrast, Congress may use the Interstate Commerce Clause to expressly preempt any state laws that regulate activities within the scope of Congress’s broad interstate-commerce authority.[62]

1. McCulloch’s Structural Principle
   Applied to Section 3

This Part applies McCulloch’s structural, collective-action principle to the issue of who can enforce Section 3 of the Fourteenth Amendment. It begins by analyzing whether states have the authority to disqualify candidates for President or Vice President. After answering that question in the negative, it considers whether the constitutional prohibition on such state disqualifications should be enforced by a rule or a standard. It then explains why the analysis offered here is consistent with the Constitution’s conferral upon the states of certain other powers bearing on the national political process. It next asks whether state officials or state courts are permitted to disqualify candidates for state or local office for violating Section 3 (yes), and whether state officials or state courts are allowed to disqualify candidates for Congress (maybe). This Part concludes by emphasizing two points: the structural approach taken in this Article does not depend on the (erroneously) perceived need to avoid the “chaos” that would allegedly ensue if states were permitted to disqualify presidential candidates, and this Article’s structural approach permits voters to enforce Section 3 by voting against candidates for any office whom voters believe to be oath-breaking insurrectionists.

1. State Disqualifications of
   Presidential Candidates

The Court in Trump v. Anderson did not connect its structural intuition to the collective-action reasoning in McCulloch; the Court instead quoted McCulloch only as indirect authority for the proposition that states may not burden Congress’s disability-removal power regarding candidates for federal office[63]—a power conferred by the text of Section 3.[64] The Court’s failure to make full use of McCulloch is regrettable. Just as states may not interfere excessively with a function of the national governing process, so they may not interfere excessively with a function of the national political process. The McCulloch principle naturally extends from the national governing process to the national political process because the national governing process is a product of the national political process, and the states’ core police powers do not naturally extend to either. According to the McCulloch principle, states may not interfere excessively with the national government or with the processes by which the national government functions and is constituted.[65] Moreover, the political process is uniquely national in the case of the Presidency and the Vice Presidency—that is, they are uniquely national offices—because all states, and in modern practice all voters within states, play a role in determining who will run for those offices and ultimately ascend to them.

Given that uniquely national political process, the constitutional question is whether a state official or a state court possesses the authority to decide whether a major party candidate engaged in insurrection and so is disqualified under Section 3.[66] If the answer is yes, and if the state official or court determines that such a candidate is disqualified, then the country runs the risk that a presidential election will swing from the disqualified candidate to the other candidate even though an electoral college majority presumably disagrees that the disqualified candidate engaged in insurrection and wants the individual to be President. That would not happen if a Democratic state official or a liberal state court disqualified a Republican candidate for President in a state that the Democratic candidate was going to win anyway. Nor would it happen if a Republican state official or a conservative state court disqualified a Democratic candidate for President in a state that the Republican candidate was going to win anyway. But it could happen if a Democratic state official or a liberal state court disqualified a Republican candidate in a battleground state or in a state that the Republican candidate was going to win. And it could happen if a Republican state official or a conservative state court disqualified a Democratic candidate in a battleground state or in a state that the Democratic candidate was going to win.

The structural issue is whether the externalized (and internalized) costs of potentially swinging the election from one candidate to the other are greater than the internalized (and externalized) benefits of vindicating the views of (at most) an Electoral College minority that the otherwise-winning candidate engaged in insurrection.[67] Although it bears repeating that there is not a purely objective, scientific way to compare costs and benefits, the answer is almost certainly yes. Outside the context of the Civil War, there are very likely to be intense disagreements about whether a candidate engaged in insurrection—and those disagreements are very likely to be informed by partisan commitments for most (although not all) people. Moreover, no matter how great of an imposition sister states may deem the decision of another state’s official or court to disqualify a major party candidate, there is nothing that sister states can lawfully do about it. (Playing tit for tat by disqualifying the clearly qualified candidate of the other political party is not a lawful option.) In such circumstances, the externalized costs are potentially enormous. That is a cost-benefit collective-action problem.[68]

Even when, say, a red-state official disqualifies the Democratic candidate for President, one should not dismiss its significance. Undecided or less committed voters in swing states might be influenced by the news that a state official or judge had formally determined that the nominee had participated in an insurrection. Such voters might also be influenced by the concern that their votes for the Democratic candidate would be wasted because other states would likely follow the red state’s lead. Moreover, even putting aside material effects, the decision of the red-state official still changes the national political process by formally reducing the number of states in which the process is taking place, and the disruption only gets worse if disqualification decisions are made in other states.

The Court came relatively close to recognizing those points and articulated similar structural reasoning in Anderson v. Celebrezze.[69] In that 1983 decision, the Court invalidated Ohio’s early filing deadline that presidential candidates had to satisfy to have their names placed on the ballot for the general election. The Court held that Ohio’s deadline, in contrast to typical state filing deadlines, imposed an unconstitutional burden on the voting and associational rights of third-party candidate John Anderson and his supporters.[70] But the Court further observed that “[t]he Ohio filing deadline challenged in this case. . . . places a significant state-imposed restriction on a nationwide electoral process.”[71] The Court reasoned that, “in the context of a Presidential election, state-imposed restrictions implicate a uniquely important national interest” because “the President and the Vice President of the United States are the only elected officials who represent all the voters in the Nation.”[72] The Court also emphasized, in effect, that Ohio was externalizing costs onto sister states that were higher than the benefits it was internalizing:

[T]he impact of the votes cast in each State is affected by the votes cast for the various candidates in other States. Thus in a Presidential election a State’s enforcement of more stringent ballot access requirements, including filing deadlines, has an impact beyond its own borders. Similarly, the State has a less important interest in regulating Presidential elections than statewide or local elections, because the outcome of the former will be largely determined by voters beyond the State’s boundaries.[73]

In other words, Ohio was regulating in a way that produced spillover effects in other states. Moreover, those external costs were higher than any benefits it was producing for itself because the outcome of Presidential elections “will be largely determined by voters beyond the State’s boundaries.”[74] That last point is off the mark: the main concern should have been that Ohio could tip or otherwise influence the election, not that the election would be largely determined elsewhere (which is in tension with the Court’s correct observation that Ohio was producing spillover effects). But crucially, the Court recognized that what happened in Ohio would not stay in Ohio.

To be sure, one could reject the foregoing structural reasoning, as well as the realist recognition of likely intractable disagreement, by simply insisting that if it really is true that a major party candidate is legally ineligible for office because the candidate engaged in insurrection, then all the externalities are beside the point. One could further insist that if it really is not true, then a state court would commit reversible error in disqualifying the candidate. But what is true as a constitutional matter is a distinct question from who is authorized to decide what is true. And a structural rationale originating in McCulloch and requiring an assessment of interstate externalities addresses that question of institutional settlement.

One might ask whether the same structural concerns would arise if a federal official or a federal court were to disqualify a major party candidate for President. It is difficult, however, to identify a scenario in which a federal executive officer would have authority to disqualify a presidential candidate. It is also difficult to perceive a scenario in which a federal court could take such action, except pursuant to an act of Congress. Acting under such a statute, the federal court would be proceeding pursuant to legislation passed by the only body in the nation in which all states are represented, so no state would have a valid structural objection.[75] Absent such a statute, a federal-question-jurisdiction action is not straightforward to imagine. If a state official or court were to exclude a presidential candidate, that scenario would be Trump v. Anderson on the level of the U.S. Supreme Court, and the Justices could vindicate the candidate’s McCulloch claim that the state acted unconstitutionally. Regarding a suit against a state for failing to disqualify a presidential candidate under Section 3, members of the public would not be able to establish injury in fact sufficient to satisfy Article III standing requirements.[76] Perhaps another presidential candidate would be able to establish injury in fact, but it is not clear that it would be injury to a legally protected interest. In contrast to Section 1 of the Fourteenth Amendment, Section 3 does not obviously protect individual rights. More fundamentally from a structural, collective-action perspective, even if the other candidate did have standing, the decision would still be made by a national court with subject-matter jurisdiction granted by the national legislature, in which—it bears repeating—all states are represented. No state would be imposing its preferences or decision on the rest of the states in the Union.

1. Rules Versus Standards

One could also ask all manner of interesting line-drawing questions here. Would McCulloch’s principle apply to someone with little support and no chance of securing the nomination? Maybe—a bright-line rule would be easy to administer. Such a rule would also avoid tasking courts with determining what a sufficient level of political support, or a sufficient likelihood of victory, looks like. In addition, an all-or-nothing rule would avoid giving the two major political parties special treatment in constitutional adjudication.

On the other hand, the same reasoning that allows one to distinguish Trump, the Republican front-runner and then nominee and now President, from someone running for the Invented-Yesterday Party nomination who is only twenty-nine years of age might allow a court to distinguish Trump from, say, someone who is on the GOP ballot in only a few states or maybe even someone who appears to have virtually no support. An appropriate standard might prohibit states from enforcing Section 3 against presidential candidates who enjoy substantial support within their political party. But this Article is content to leave open such questions, which implicate well-worn debates about the relative virtues of rules versus standards, and instead make two points.

First, to repeat, application of the McCulloch principle can consider questions of degree. That is why it matters whether presidential candidates enjoy substantial support within their party. That is also why the theory articulated in this Article does not threaten the longstanding authority of states to impose voter eligibility requirements and presidential eligibility requirements that fall within a range of reasonableness relative to what other states impose. For example, a reasonable filing deadline for presidential candidates is different from a very early deadline relative to sister states. Second, to the extent that a standard is preferable to a rule, Congress can draw the line more assertively than the federal courts because Congress enjoys superior democratic legitimacy. That is, certain state efforts to enforce Section 3 against presidential candidates should not be struck down by the federal courts even though Congress could forbid them. More generally, Congress has greater authority than the federal courts to determine that states may not interfere with certain functions of the national political process.

1. Other State Powers

Applying McCulloch’s structural principle to block state enforcement of Section 3 at least against presidential candidates with substantial support within their own party is consistent with the roles of the states in prescribing “[t]he Times, Places and Manner of holding Elections for Senators and Representatives” as expressly conferred in the first clause of Article I, Section 4.[77] That clause does not cover presidential elections, although state decisions under the provision typically determine the time, place, and manner of holding presidential elections in states given the Electors Clause discussed below.[78] More importantly from a structural perspective, state determinations under that provision do not ordinarily cause significant spillover effects in other states, so there is no collective-action objection under McCulloch to their making such determinations. But if a state were to make such a decision that applied to a presidential candidate and that externalized significant costs onto sister states, the decision would be subject to judicial review to determine its consistency with the McCulloch principle—specifically, whether the state was externalizing costs that exceeded the benefits it was internalizing.

Prohibiting states from enforcing Section 3 at least against presidential candidates with substantial support is also consistent with state power, under the second clause of Article II, Section 1,[79] to disqualify presidential candidates who do not meet the citizenship, age, and residency requirements set forth in the fifth clause of Article II, Section 1.[80] To repeat, proper application of McCulloch’s basic principle in modern times requires sensitivity to questions of degree and impact. In the ordinary case, a state decision not to list a candidate because of a determination that the candidate does not qualify under Article II imposes no costs on other states and their voters. That is in significant part because deciding whether someone is a natural-born citizen, at least thirty-five years old, and a U.S. resident for at least fourteen years will typically not (even if it sometimes may) generate disagreements among states about the facts or the law. But where such a decision does generate disagreements, and where a state is disqualifying a candidate for a major party nomination for President, the externalized costs are potentially large for the reasons already stated. In other words, there may be circumstances where states are constitutionally barred under McCulloch from enforcing presidential eligibility requirements. By contrast, any presidential candidate with substantial support is nearly always going to generate heated disagreements among states regarding whether the candidate previously engaged in insurrection and regarding the appropriate process for making that determination.

But can a structural limitation on enforcement of Section 3 by states be reconciled with the Electoral College method of electing the President, according to which “[e]ach State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors” for President?[81] Professor William Baude opines that the Court’s holding in Trump v. Anderson not only “lacked any real basis in text and history” but “also is at odds with the basic structure of the Electoral College, in which states have primary authority to decide how their slates of electors are chosen.”[82] Supporting Professor Baude’s view is the distinct possibility that there would be large spillover effects in sister states if certain states changed their method of allocating electors from following the popular vote in the state to empowering the state’s legislature to make the appointments. Yet, the constitutional text permits states to make that choice.

One response is that the power of state legislatures to designate the method of choosing presidential electors is conferred by clear constitutional text, and there is no further mystery about it: clear constitutional text trumps structural principles.[83] To be sure, one could respond that, because Section 3 is self-executing, state power to enforce it in all instances is also clear text. But that is not obvious. The Electoral College more clearly permits states to appoint presidential electors than Section 3 allows states to disqualify alleged oath-breaking insurrectionists with serious political support who aim to be President. For a methodological pluralist (who considers multiple modalities of constitutional argument), as opposed to a strict textualist or originalist,[84] there is no contradiction between arguing both that Section 3 is self-executing and that a structural principle of constitutional law prohibits states from enforcing it in certain circumstances. To say that Section 3 is self-executing means only that any institution or officer with the authority to make a decision to which the section is relevant may apply it even absent implementing legislation. It does not mean that that same institution or officer can enforce the section if some other principle of law deprives it of the authority to make the relevant decision.

Another response is that, even as a matter of textual argument, the second clause of Article II, Section 1, is not a blank check for state legislatures to do anything they choose. One should not assume that that textual power authorizes state legislatures to decide, for example, that only candidates older than forty years of age can receive the electors’ vote. In fact, one should not assume that a state legislature can constitutionally play bait and switch with the state’s electorate after the voters have exercised the franchise. A state legislature’s power here is, like other unquestionable state powers, subject to constitutional limitations that derive from other provisions of the Constitution or from general structural principles like McCulloch supremacy. A state legislature that tipped a presidential election from one political party to the other by changing its method of appointment of electors after eligible voters in the state had cast their votes might well violate the McCulloch principle articulated in this Article. The cost externalization would obviously be large, and the benefit internalization would arguably be modest, given that the state legislature would be rejecting the will of the majority of the state’s electorate after it had already been expressed.

Before an election, sister states can lawfully do something about another state’s decision to change the definition of “the state” for Electoral College purposes from a popular majority to a state legislative majority. Specifically, sister states can respond in kind. If they do not do so, then at least some of them presumably do not believe that the first state is externalizing significant costs. (Other possibilities include that a sister state does not think playing bait and switch in response would be democratically defensible or likely to change the outcome of the election in the state.) By contrast, as noted earlier, sister states have no lawful way of responding if another state disqualifies a major party candidate under Section 3.

1. Other State Disqualification Decisions

As adverted to by the Court in Anderson v. Celebrezze,[85] decisions by a state to disqualify candidates for state or local office under Section 3 are at the opposite end of the structural spectrum from decisions by a state to disqualify candidates for President and Vice President under Section 3. A state decision to disqualify a candidate for state or local office has no significant effects beyond the political process of the state making the disqualification decision,[86] so its making the decision raises no serious structural concern.[87] There is no collective-action objection, therefore, to states enforcing Section 3 against candidates for state or local office—and, in fact, there is an historical practice associated with their doing so.[88] Past practice may illuminate the constitutional structure by analogy to the functioning of a machine. If one wants to understand how the constitutional system is supposed to function, it makes sense to investigate how the system has in fact functioned.[89] As Justice Felix Frankfurter wrote in the Youngstown Steel Seizure Case, “The Constitution is a framework for government. Therefore the way the framework has consistently operated fairly establishes that it has operated according to its true nature.”[90]

Decisions by states to disqualify candidates for the U.S. Senate or House of Representatives are structurally situated in between the two categories of disqualification decisions analyzed above. On one hand, like a state decision to disqualify a candidate for President or Vice President, a state decision to disqualify a candidate for the Senate or House is national in scope in the sense that it can affect which political party obtains control of the relevant chamber of Congress—and such control is profoundly important to both parties and to the tens of millions of their members. Notably in that regard, the Court in U.S. Term Limits, Inc. v. Thornton quoted Joseph Story’s Commentaries on the Constitution of the United States for the proposition that members of Congress “owe their existence and functions to the united voice of the whole, not of a portion, of the people.”[91]

On the other hand, like a state decision to disqualify a candidate for state or local office, a state decision to disqualify a candidate for the U.S. Senate or House is state or local in scope in the sense that only voters within the state are entitled to vote in that election. It is therefore not obvious that there is any collective-action problem with a state secretary of state or state court ruling a candidate for the U.S. Senate or House ineligible on Section 3 grounds. Perhaps the federal courts should not disturb a state decision to disqualify a candidate for the U.S. Senate or House even as Congress could prohibit states from doing so.

Whatever is the correct answer to that difficult question, this Article would not reach it until a state purported to disqualify a congressional candidate under Section 3. Trump v. Anderson,[92] with its potentially enormous implications for who would win the 2024 presidential election, was a poor vehicle for the Court to decide that states cannot invoke Section 3 to disqualify congressional candidates.[93] Regarding Colorado’s disqualification of Trump, a presidential candidate with very substantial support, it more than suffices to draw the proper structural inference from the fact that all states, and all voters within states, play a role in determining who will run for the office of President and ultimately occupy that office. Just as “a part” may not tax “the whole” because the whole is not represented in the part,[94] so a part may not make presidential eligibility decisions that significantly undermine the capacity of the whole to determine who will represent it in the White House. As Justice Barrett wrote, “States lack the power to enforce Section 3 against Presidential candidates,” and “[t]hat principle is sufficient to resolve this case.”[95]

1. “Chaos” Theory and Popular
   Enforcement of Section 3

That structural rationale for disabling state officials and state courts from enforcing Section 3 against presidential candidates like Trump does not depend on a perceived need to avoid the “chaos” that every Justice imagined would ensue if states could make conflicting Section 3 disqualification determinations regarding presidential candidates.[96] That is an important feature of the McCulloch, collective-action theory of Section 3 enforcement, because the Justices’ “chaos theory” is both entirely consequentialist and likely incorrect. There would have been no chaos if the Justices had been willing to decide whether Trump had (or had not) engaged in insurrection and so was (or was not) disqualified from being President again. Any alleged chaos caused by different state determinations at different times would have ended as soon as the Court resolved that question—uniformly for the nation. In other words, the “chaos theory” reflects—indeed, presupposes—a Court that is unwilling to make such a politically impactful determination for prudential reasons, not legal ones. By contrast, the structural, collective-action theory of Section 3 enforcement offers a venerable legal rationale—it originates with McCulloch—and it suffices to explain why Trump v. Anderson is correctly decided.

Finally, it is worth observing that the McCulloch principle did not disable the people of Colorado from expressing the conviction that Trump was ineligible to be President under Section 3 of the Fourteenth Amendment. Likewise, the principle did not disable the people of Maryland from acting on the belief that the Second National Bank was unconstitutional. Just as Marylanders could have refused to patronize the Bank based on their constitutional objections to its existence, so Coloradans could have refused to vote for Trump based on their belief that he is an oath-breaking insurrectionist. What the McCulloch principle bars is a decision by state executive officials or state courts not to let the state’s voters cast their ballots one way or the other.

III. Two Objections

This Part anticipates two objections that have not already been addressed. One is legal, and the other is prudential.

1. A Legal Objection

First, and most importantly, one might ask what the difference is between a multi-state collective-action problem that constitutionally disables states from acting and one that does not. For example, when a state brings a criminal case against a presidential candidate (as New York and Georgia did against candidate Donald Trump),[97] the spillover effects can potentially be large in terms of public perception of the candidate, especially if the candidate is found guilty.[98] But it seems implausible to therefore conclude that such a prosecution and conviction are unconstitutional on structural, McCulloch grounds (although execution of the sentence, depending on what it is—see below—might be unconstitutional on precisely those grounds). Why is a state criminal prosecution different from a Section 3 disqualification by a state?

It is correct that the McCulloch principle is not universally applicable to all interstate spillover effects in which externalized costs exceed internalized benefits. Such spillover effects must instead be deemed relevant to some constitutional provision or process that is best interpreted in collective-action terms. To repeat an earlier point, the emphasis of the McCulloch Court on the national governing process extends to the national political process because the national political process produces the national governing process, and it is not a core function of the state’s reserved police powers to regulate either—which is why the Constitution expressly grants the states certain powers within the sphere of the national political process.[99] But like the national governing process, the national political process is not a completely open-ended constitutional category. Constitutional provisions like Section 3 that determine eligibility to run for President obviously fall within that category for reasons discussed in Part II.

By contrast, it is not obvious how the operation of state criminal law against a presidential candidate is part of the national political process, particularly when (in the author’s view) a structural, McCulloch rationale would preclude the execution of a state criminal sentence (for example, incarceration) that interfered with a duly elected President’s ability to serve. In addition, a state’s ability to enforce its criminal laws through investigations, prosecutions, convictions, and the imposition of sentences is a central component of its police powers, even if the state could not execute certain sentences against an elected President. More generally, although one could reasonably debate wherein lie the outer limits of the national political process to which the structural, McCulloch principle applies, it should not be controversial that there must be an outer limit to the category so that collective-action reasoning does not eviscerate the state’s police powers.

1. A Prudential Objection

Second, and moving from a legal objection to a prudential one, Americans who believe that Trump’s return to the White House means the end of the Constitution are unlikely to be impressed with the structural, collective-action argument offered in this Article. They may insist that the Constitution is not a “suicide pact,” as Justice Robert Jackson once famously observed.[100] They may further insist that the analysis offered here makes it one, particularly given Trump’s role in the storming of the Capitol on January 6, 2021,[101] and the unlikelihood (in their view) that anyone else will be deemed disqualified under Section 3 in the future even if Congress does pass enforcement legislation.[102]

When the stakes are high, prudential reasoning frequently appears in American constitutional discourse, so one should not necessarily reject out of hand its invocation in this consequential context. But taking the objection on its own terms, a persuasive prudential analysis would need to accurately predict the likelihood that Trump’s re-election would mean the end of the Constitution, as well as what that end would look like. Neither task seems straightforward. It is not obvious that Trump’s re-election would (now, will) mean the end of the Constitution. Moreover, a complete prudential analysis would need to account for the potential consequences of defeating Trump through a U.S. Supreme Court decision involving Section 3, as opposed to the expected value of defeating him at the polls. The likelihood of widespread political violence and broader democratic instability might have been high—potentially very high—if Section 3 had been used to defeat him. But it is not clear that anyone can know the answer to such prudential questions, which is one reason not to let them completely crowd out legal analysis.

But there is a more fundamental reason. The prudential objection should not just be taken on its own terms. Attempting to get the law right is crucial to upholding the rule of law, which exists in significant part to restrain politicians like Donald Trump. To the extent that the objection under consideration is an “all the laws, but one” argument, it is an extra-constitutional argument.[103] A key value of getting the law right is to help decisionmakers figure out—rationally, not emotionally—whether they really believe that the country is at the point at which the Constitution will come to an end.

Conclusion

Trump v. Anderson should have been decided on the narrow but deep legal ground that the structural principle articulated in McCulloch v. Maryland more than two centuries ago bars state courts and state executive officials from enforcing Section 3 of the Fourteenth Amendment against presidential candidates like Donald Trump. The judgment in Trump v. Anderson is best understood as resting on such a federalism analysis. The remainder should be distinguished as dicta or otherwise rejected.[104]

All the Justices shared the intuition that something was structurally amiss about what Colorado had done in the case. But they were unable to develop that intuition into a persuasive federalism rationale. The case was resolved in a hurry for understandable reasons, so perhaps the Justices can be forgiven for glimpsing but failing to grasp the constitutional reasoning that best explains and justifies the Court’s judgment.[105]

Even so, the Justices’ failure to do so was unfortunate. Given the bitter partisanship that plagues American life in contemporary times, there would have been a public benefit to showing that a decision laden with major political consequences can be grounded in fundamental constitutional principles that the Court recognized and applied more than two centuries ago. To be sure, most Americans do not read Supreme Court opinions. But their contents are nonetheless conveyed to the public through mainstream and social media.[106]

There is a broader structural implication here regarding the relationship between the original Constitution and the post-Civil War Constitution. The original Constitution primarily concerns how states relate to the federal government and to sister states, which is why the author has called it “the Collective-Action Constitution.”[107] By contrast, Section 1 of the Fourteenth Amendment primarily concerns how states relate to their own inhabitants on certain basic questions of constitutional rights, which is why the author has called that section part of “the Reconstruction Constitution.”[108] But the prospect of state enforcement of Section 3 of the Fourteenth Amendment against at least certain presidential candidates indicates that structural, collective-action reasoning can also be relevant to the proper interpretation of parts of that amendment. As Professor Graber explains, Sections 2, 3, and 4 of the Fourteenth Amendment had meanings when they were written and ratified that were far more settled—and were deemed far more important—than the meaning and significance of Section 1.[109] Relatedly, Sections 2 through 4 were structural provisions that reflected the goal of the Republican Framers in the Thirty-Ninth Congress “to prevent ‘rebel rule’ by empowering and protecting the persons, white and Black, who remained loyal to the Union during the Civil War.”[110]

 

         *    Professor and Co-Director of the Center on the Structural Constitution, Texas A&M University School of Law. For helpful feedback, I thank William Baude, Stuart Benjamin, Joseph Blocher, Curtis Bradley, Katherine Mims Crocker, Mark Graber, Jill Hasday, Margaret Lemos, Gerard Magliocca, Darrell Miller, and, especially, H. Jefferson Powell.

        [1].    144 S. Ct. 662, 664–65 (2024) (per curiam).

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

        [3].    Trump, 144 S. Ct. at 665–66.

        [4].    See id. at 672 (Sotomayor, Kagan, & Jackson, JJ., concurring in the judgment) (suggesting that state enforcement of Section 3 would violate “our Nation’s federalism principles” without identifying what those principles are).

        [5].    Id. at 671 (per curiam) (emphasis added). Examining the language of the per curiam opinion with care, some scholars have suggested that the Court did not actually hold that only Congress can enforce Section 3, nor did it hold that it may do so only through Section 5 legislation. See, e.g., Derek T. Muller, Administering Presidential Elections and Counting Electoral Votes After Trump v. Anderson, 60 Wake Forest L. Rev. 327, 342 (2025) (“The notion that ‘only’ Congress can enforce Section 3 and that it ‘must’ do so by legislation are potentially significant limitations on the federal government—if these characterizations are true. But parsing the per curiam opinion makes it much harder to agree that the concurring opinion has accurately characterized the per curiam opinion.”). From start to finish, however, the per curiam opinion emphasizes that it is Congress’s responsibility to enforce Section 3 against federal officeholders and candidates. Moreover, the opinion identifies Section 5 of the Fourteenth Amendment as the means of congressional enforcement. That said, the per curiam may not be best read as excluding congressional enforcement of Section 3 through other means, including the refusal to seat a member of Congress, the impeachment process, or the counting of electoral votes. The Justices who joined the per curiam may not have been considering and rejecting such alternative means of congressional enforcement.

        [6].    Trump, 144 S. Ct. at 671 (per curiam).

        [7].    It is equally clear that Section 3 covers the Presidency and former Presidents regardless of whether they previously held any state or federal office, but the Court did not address those issues, and regarding them this Article is content to cite the work of other legal scholars. See, e.g., William Baude & Michael Stokes Paulsen, The Sweep and Force of Section 3, 172 U. Pa. L. Rev. 605, 721–30 (2024); Mark A. Graber, Section Three of the Fourteenth Amendment: Is Trump’s Innocence Irrelevant?, 84 Md. L. Rev. 1 (2024). Professor Graber writes:

None of the many lawyers who sat in the 39th Congress or who wrote commentaries on the Fourteenth Amendment after the drafting pointed out that because of a legal technicality Section 3 did not disqualify a past or present President who engaged in an insurrection or rebellion but never held any previous state or federal office. No one has ever advanced a commonsense reason why such an exemption should exist.

Id. at 36–37 (footnote omitted). For a contrary view, see Josh Blackman & Seth Barrett Tillman, Is the President an “officer of the United States” for Purposes of Section 3 of the Fourteenth Amendment?, 15 N.Y.U. J.L. & Liberty 1 (2021).

        [8].    Baude & Paulsen, supra note 7, at 623. Moreover, Section 3’s language parallels many earlier-drafted provisions, including the qualifications clauses of Articles I and II. See U.S. Const. art. I, § 2, cl. 2 (“No Person shall be a Representative who shall not have attained to the Age of twenty five Years, and been seven Years a Citizen of the United States, and who shall not, when elected, be an Inhabitant of that State in which he shall be chosen.”); id. art. I, § 3, cl. 3 (“No person shall be a Senator who shall not have attained to the Age of thirty Years, and been nine Years a Citizen of the United States, and who shall not, when elected, be an Inhabitant of that State for which he shall be chosen.”); id. art. II, § 1, cl. 5 (“No person except a natural born Citizen . . .  shall be eligible to the Office of President; neither shall any Person be eligible to that Office who shall not have attained to the Age of thirty-five Years, and been fourteen Years a Resident within the United States.”). Section 3’s language also parallels that of Section 1 of the Thirteenth Amendment and Section 4 of the Fourteenth Amendment, neither of which have ever been treated as inoperative until enforced by congressional legislation. See U.S. Const. amend. XIII, § 1 (“Neither slavery nor involuntary servitude, except as a punishment for crime whereof the party shall have been duly convicted, shall exist within the United States, or any place subject to their jurisdiction.”); id. amend. XIV, § 4 (prohibiting questioning “the [v]alidity of the public debt of the United States,” including debts incurred “in suppressing insurrection or rebellion,” but barring the payment of any debts “incurred in aid of insurrection or rebellion against the United States, or any claim for the loss or emancipation of any slaves”).

        [9].    Graber, supra note 7, at 8, 12–13. See id. at 16 (“Nothing in the text or history of the Fourteenth Amendment provides any reason for thinking that any member of Congress or politically active citizen other than Salmon Chase . . . thought that the Thirteenth Amendment, Section 1 of the Fourteenth Amendment, and Section 4 of the Fourteenth Amendment were self-executing, but that Section 3 required implementing legislation.”).

      [10].    11 F. Cas. 7, 26 (C.C.D. Va. 1869).

      [11].    Graber, supra note 7, at 13.

      [12].    Professor Magliocca posed this devastating question that Chase’s opinion left unanswered: “How could Section Three be self-executing for Jefferson Davis but not self-executing for Black defendants in the same place at the same time?” Gerard N. Magliocca, Amnesty and Section Three of the Fourteenth Amendment, 36 Const. Comment. 87, 105 (2021). Chase had earlier agreed with Davis’s argument that Section 3 was self-executing and imposed a punishment on him, and that this penalty barred his federal prosecution for treason. Id. at 100, 102. Griffin’s Case involved Chase’s review of a grant of habeas relief to a Black defendant by federal judge John Underwood because the defendant was tried and sentenced by a state judge who was disqualified under Section 3. Id. at 102. Chase reversed on the ground that Section 3 was not self-executing, so that the state judge was not disqualified, given the absence of implementing legislation by Congress at the time of the state trial. 11 F. Cas. at 26–27. On the problems with Griffin’s Case and the potential motivations of its author, see Baude & Paulsen, supra note 7, at 644–59, and Magliocca, supra, at 100–11. For a contrary view, see Josh Blackman & Seth Barrett Tillman, Sweeping and Forcing the President into Section 3, 28 Tex. Rev. L. & Pol. 350 (2024).

      [13].    Civil Rights Cases, 109 U.S. 3, 20 (1883) (emphasis added).

      [14].    Id.

      [15].    Trump v. Anderson, 144 S. Ct. 662, 667 (2024).

      [16].    See Baude & Paulsen, supra note 7, at 625 (“The power to enforce adds to the substantive prohibition—it is not a subtraction from or suspension of it.”).

      [17].    109 U.S. at 20.

      [18].    Mark A. Graber, Punish Treason, Reward Loyalty: The Forgotten Goals of Constitutional Reform After the Civil War 92, 94, 162 (2023). “This potential for government by disloyal rebels,” Professor Graber emphasizes, “was the problem Republicans were trying to solve when they discussed, framed, and eventually passed the Fourteenth Amendment.” Id. at 93–94. By embedding the disqualification of former rebels in the Constitution, and thus beyond tampering by a congressional majority, the Amendment’s proponents meant Section 3 to play a vital role in “preventing former rebels from regaining control over state and national politics.” Id. at 91. Reading Section 3 to require congressional action before it can be enforced gets the point of the provision backward. If Section 3’s language indicated that the provision was not self-executing, its animating purpose might not matter. But the language rebuts the conclusion that Section 3 is not self-executing.

      [19].    For discussions of the modalities of constitutional interpretation, see Philip C. Bobbitt, Constitutional Fate: Theory of the Constitution (1982) (coining the term “modalities” and identifying six of them: historical, textual, structural, prudential, doctrinal, and ethical); Philip C. Bobbitt, Constitutional Interpretation (1991) (applying those six modalities); Jack M. Balkin, Memory and Authority: The Uses of History in Constitutional Interpretation 18–20 (2024) (identifying eleven modalities: text, structure, purpose, consequences, judicial precedent, political convention, custom, natural law, ethos, tradition, and honored authority); and Richard H. Fallon, A Constructivist Coherence Theory of Constitutional Interpretation, 100 Harv. L. Rev. 1189, 1244–46 (1987) (identifying five modalities: text, original intent, theory, precedent, and value). See also Curtis A. Bradley & Neil S. Siegel, Constructed Constraint and the Constitutional Text, 64 Duke L.J. 1213, 1239–43 (2015) (critically analyzing Bobbitt’s list of modalities). For an overview of theories of constitutional interpretation, see Daniel Farber & Neil S. Siegel, United States Constitutional Law 63–80 (2d ed. 2024).

      [20].    17 U.S. (4 Wheat.) 316 (1819).

      [21].    Id. at 435–36.

      [22].    Trump v. Anderson, 144 S. Ct. 662, 672 (2024) (Sotomayor, Kagan, & Jackson, JJ., concurring in the judgment) (using the phrase “federalism principles” three times).

      [23].    See generally, e.g., William Baude & Michael Stokes Paulsen, Sweeping Section Three Under the Rug: A Comment on Trump v. Anderson, 138 Harv. L. Rev. 676 (2025); Aziz Z. Huq, Structural Logics of Presidential Disqualification: An Essay on Trump v. Anderson, 138 Harv. L. Rev. 172 (2024); Ilya Somin, A Lost Opportunity to Protect Democracy Against Itself: What the Supreme Court Got Wrong in Trump v. Anderson, 2024 Cato Sup. Ct. Rev. 319 (2024).

      [24].    Trump, 144 S. Ct. at 671 (Barrett, J., concurring in part and concurring in the judgment).

      [25].    17 U.S. (4 Wheat.) 316 (1819).

      [26].    Daniel A. Farber, The Story of McCulloch: Banking on National Power, in Constitutional Law Stories 33, 44 (Michael C. Dorf ed., 2004). For inflation calculations from 1819 to 2025, see Ian Webster, Inflation Calculator,
Official Inflation Data, Alioth Finance (July 5, 2025), https://www.officialdata.org/ us/inflation/1819?amount=100, https://www.officialdata.org/us/inflation/1819?amount= 15000 (enter the “Start year” as 1819 and the “End year” as 2025).

      [27].    H. Jefferson Powell, Targeting Americans: The Constitutionality of the U.S. Drone War 23 (2016).

      [28].    See generally Neil S. Siegel, The Collective-Action Constitution (2024) [hereinafter Siegel, The Collective-Action Constitution] (arguing that, as embodied in McCulloch, the primary structural purpose of the U.S. Constitution is to empower the federal government to solve collective-action problems for the states and to prevent states from undermining federal solutions or causing such problems); see id. at 25–52 (analyzing both holdings in McCulloch through a collective-action lens).

      [29].    See U.S. Const. art. I, § 10, cl. 2 (discussed infra notes 49–50 & accompanying text).

      [30].    McCulloch, 17 U.S. (4 Wheat.) at 425–26.

      [31].    U.S. Const. art. VI, § 1, cl. 2.

      [32].    McCulloch, 17 U.S. (4 Wheat.) at 426–27, 431.

      [33].    277 U.S. 218, 223 (1928) (Holmes, J., dissenting).

      [34].    Marshall wrote:

But is this a case of confidence? Would the people of any one state trust those of another with a power to control the most insignificant operations of their state government? We know they would not. Why, then, should we suppose, that the people of any one state should be willing to trust those of another with a power to control the operations of a government to which they have confided their most important and most valuable interests?

McCulloch, 17 U.S. (4 Wheat.) at 431.

      [35].    Id. at 430. This section of the Article mixes arguments that Marshall presented as based on the Constitution with arguments that he presented at pages 428–30 as grounded in “just theory,” id. at 430, even though he bracketed his theory arguments at the beginning and end as distinct from the “test of the Constitution,” id. at 428. From a modern perspective, his theory discussion can properly be understood as part of his constitutional analysis.

      [36].    567 U.S. 519 (2012); see Robert D. Cooter & Neil S. Siegel, Not the Power to Destroy: An Effects Theory of the Tax Power, 98 Va. L. Rev. 1195 (2012) (developing an effects theory of Congress’s taxing authority that the Court largely adopted in NFIB).

      [37].    McCulloch, 17 U.S. (4 Wheat.) at 428.

      [38].    Id.

      [39].    Id. at 435.

      [40].    Id. at 429–30.

      [41].    304 U.S. 144, 152 n.4 (1938).

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

      [43].    For a discussion of the three categories of collective-action problems discussed in this and the following three paragraphs, see Siegel, The Collective-Action Constitution, supra note 28, at 5, 62–72, 89–92.

      [44].    Technically, with cooperation problems, no Nash equilibrium (which is the basic solution concept in game theory) is Pareto efficient. With coordination problems, at least one Nash equilibrium is Pareto efficient. “As a result, cooperation problems generally present a more compelling case for government intervention than coordination problems.” Id. at 87.

      [45].    See, e.g., Robert D. Cooter, The Strategic Constitution 32–35 (2000) (discussing cost-benefit efficiency); Kaldor-Hicks Efficiency, in A Dictionary of Finance and Banking (Oxford 6th ed. 2018) (ebook).

      [46].    Siegel, The Collective-Action Constitution, supra note 28, at 42, 182, 398 (noting Rhode Island’s consistent opposition to structural reform during the 1780s).

      [47].    U.S. Const. art. I, § 10, cl. 2.

      [48].    Id.

      [49].    See, e.g., Nat’l Pork Producers Council v. Ross, 143 S. Ct. 1142 (2023) (reaffirming the antidiscrimination principle unanimously and reaffirming the balancing test for nondiscriminatory burdens on interstate commerce by a vote of six to three).

      [50].    U.S. Const. art. IV, § 1.

      [51].    U.S. Const. art. IV, § 2, cl. 1.

      [52].    U.S. Const. art. IV, § 2, cl. 2.

      [53].    U.S. Const. art. IV, § 3, cl. 1.

      [54].    U.S. Const. art. IV, § 3, cl. 2.

      [55].    U.S. Const. art. IV, § 4.

      [56].    See Siegel, The Collective-Action Constitution, supra note 28, at 313–69 (analyzing the collective-action logics animating the provisions of Article IV).

      [57].    See, e.g., The Federalist No. 43, at 275 (James Madison) (Clinton Rossiter ed., 1961) (quoting Montesquieu as writing that “Greece was undone as soon as the king of Macedon obtained a seat among the Amphictyons”). Another principal motivation for the Republican Form Clause was Shay’s Rebellion. In a memorandum he wrote to himself while preparing for the Constitutional Convention, James Madison had Shay’s Rebellion in mind when he decried the “want of Guaranty to the States of their Constitutions & laws against internal violence.” James Madison, Vices of the Political System of the United States (April 1787), in 2 The Writings of James Madison 363 (Gaillard Hunt ed., 1904). Shay’s Rebellion was a popular insurgency in Massachusetts that began in late August 1786 and ended in early February 1787. The insurgents violently demanded, but did not receive, paper money and relief from taxation and debt. “The principal lasting effect of Shay’s Rebellion was that it persuaded a sufficient number of staunch Massachusetts federalists that the Confederation needed reform to enable it to protect their state against a future popular insurgency.” George William Van Cleve, We Have Not a Government: The Articles of Confederation and the Road to the Constitution 242 (2017).

      [58].    Panhandle Oil Co. v. Mississippi ex rel. Knox, 277 U.S. 218, 223 (1928) (Holmes, J., dissenting).

      [59].    Id.

      [60].    17 U.S. (4 Wheat.) 316, 404–05 (1819).

      [61].    Siegel, The Collective-Action Constitution, supra note 28, at 459–61, 484.

      [62].    For discussions of the dormant commerce doctrine and the Interstate Commerce Clause, see id. at 171–217; Farber & Siegel, supra note 19, at 117–75; supra note 49 and accompanying text (discussing the voting patterns in a recent dormant commerce clause decision).

      [63].    144 S. Ct. 662, 669 (2024) (per curiam) (quoting McCulloch, 17 U.S. (4 Wheat.) at 436, for the proposition that “[s]tates have no power . . . to retard, impede, burden, or in any manner control, the operations of the constitutional laws enacted by Congress”).

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

      [65].    See McCulloch, 17 U.S. (4 Wheat.) at 427 (stating that, in interpreting the Constitution, “no principle, not declared, can be admissible, which would defeat the legitimate operations of a supreme government,” and that “[i]t is of the very essence of supremacy, to remove all obstacles to its action within its own sphere, and so to modify every power vested in subordinate governments, as to exempt its own operations from their own influence”).

      [66].    Although the Founders did not anticipate political parties, the two-party system became part of the constitutional design after ratification of the Twelfth Amendment, which requires separate ballots for President and Vice President to prevent another election in which the winners of the presidency and the vice-presidency come from rival parties. Accordingly, reasoning from within the two-party system is constitutional reasoning; it makes structural sense to take that system into account.

      [67].    Where interstate externalities exist, it is possible that a cost-externalizing state is also internalizing some costs (e.g., by rejecting the views of those voters in the state who believe that a presidential candidate did not engage in insurrection) and externalizing some benefits (e.g., by vindicating the views of those voters in other states who believe that the candidate did engage in insurrection). That possibility can be handled analytically by comparing net externalized costs (i.e., externalized costs (EC) minus externalized benefits (EB)) with net internalized benefits (i.e., internalized benefits (IB) less internalized costs (IC)). If net externalized costs exceed net internalized benefits, there is a cost-benefit collective-action problem. Notably, to conclude that net externalized costs exceed net internalized benefits is equivalent to concluding that total costs exceed total benefits for the nation. To put the point symbolically, the expression (EC – EB > IB – IC) is arithmetically equivalent to the expression (EC + IC > IB + EB). In the case of a state official or court that swings a presidential election by disqualifying a major party candidate, the claim in the text is equivalent to the claim that the total costs for the nation exceed the total benefits.

      [68].    Professor Ilya Somin argues that allowing state courts or officials to enforce Section 3 actually solves a multi-state collective-action problem:

If a combination of partisan bias and voter ignorance leads to the election of a dangerous insurrectionist to high office, that too is a collective action problem, arising from the fact that most individual voters have little incentive to seek out relevant information and use it wisely. Moreover, individual states may have little or no incentive to address the problem of voter ignorance by means other than Section 3 disqualification, since much of the harm caused by ignorance within one state will be borne by people in other states. Thus, widespread availability of state-level Section 3 remedies is itself [a] way to alleviate interstate collective action problems.

Somin, supra note 23, at 342 (footnotes omitted). It is not enough, however, to identify a potential collective-action problem facing the states. To be structurally relevant, that collective-action problem must relate to some constitutional provision or principle that is appropriately understood in collective-action terms. Insofar as Professor Somin is implying that the relevant provision is Section 3, there is nothing in Professor Graber’s book cited supra note 18, or in any other historical source of which the author is aware, supporting the idea that the 1866 Republican Congress, which feared a future Democratic majority, was worried about white Southern voters’ being uninformed about what that future Democratic majority would do when it returned to power.

      [69].    460 U.S. 780 (1983).

      [70].    Id. at 795.

      [71].    Id.

      [72].    Id. at 794–95.

      [73].    Id. at 795 (footnotes omitted). Quoting Cousins v. Wigoda, 419 U.S. 477, 490 (1975), the Court in Anderson v. Celebrezze wrote that “[t]his Court, striking down a state statute unduly restricting the choices made by a major party’s Presidential nominating convention, observed that such conventions serve ‘the pervasive national interest in the selection of candidates for national office, and this national interest is greater than any interest of an individual State.’” Anderson, 460 U.S. at 795.

      [74].    Id.

      [75].    See Siegel, The Collective Action Constitution, supra note 28, at 39–43, 484 (explaining that Congress, where all states and all individuals are represented, is the government institution in the United States with the most democratic legitimacy to determine whether a cost-benefit collective-action problem exists and how to address it).

      [76].    See, e.g., FDA v. Alliance for Hippocratic Medicine, 144 S. Ct. 1540, 1554 (2024) (“As Justice Scalia memorably said, Article III requires a plaintiff to first answer a basic question: ‘What’s it to you?’”) (quoting Antonin Scalia, The Doctrine of Standing as an Essential Element of the Separation of Powers, 17 Suffolk U. L. Rev. 881, 882 (1983)).

      [77].    U.S. Const. art. I, § 4, cl. 1 (“The Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof; but the Congress may at any time by Law make or alter such Regulations, except as to the Places of chusing Senators.”). For discussion of the importance of the congressional backstop, see Siegel, The Collective-Action Constitution, supra note 28, at 369.

      [78].    See infra note 79. The fact that a state’s time-place-and-manner decisions apply to its own representatives—with respect to whom it will fully internalize the costs of poor processes—provides some assurance about the processes that it uses for presidential elections. By contrast, a state’s disqualification decision under Section 3 can apply only to a presidential candidate. By analogy, it is not worrisome from a collective-action perspective for a state to subject federal institutions within its borders to nondiscriminatory property taxes because the same taxes at the same rates apply to state and private institutions within the state. In that situation, the whole is virtually represented by the part.

      [79].    U.S. Const. art. II, § 1, cl. 2 (“Each State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors” for the President.).

      [80].    U.S. Const. art. II, § 1, cl. 5 (“No person except a natural born Citizen  . . .  shall be eligible to the Office of President; neither shall any Person be eligible to that Office who shall not have attained to the Age of thirty-five Years, and been fourteen Years a Resident within the United States.”).

      [81].    U.S. Const. art. II, § 1, cl. 2.

      [82].    William Baude, A Principled Supreme Court, Unnerved by Trump, N.Y. Times (July 5, 2024), https://www.nytimes.com/2024/07/05/opinion/supreme-court-trump.html.

      [83].    See Siegel, The Collective-Action Constitution, supra note 28, at 29, 395 (endorsing the position that clear constitutional text prevails over structural inferences and constitutional purposes).

      [84].    On methodological pluralism in constitutional interpretation, see Farber & Siegel, supra note 19, at 80.

      [85].    See supra note 73 and accompanying text (quoting the Court’s opinion).

      [86].    One might question whether state elections are entirely self-contained, given the existence of two national political parties that affect, and are affected by, state and local elections. Perhaps state elections are not entirely self-contained, but the general point set forth in the text still seems correct, especially compared with presidential elections.

      [87].    There is nothing anomalous about a state court’s being authorized to enforce Section 3 against state officials or candidates but not against federal officials or candidates. State courts can grant habeas relief to state prisoners but not to federal prisoners. See Tarble’s Case, 80 U.S. (13 Wall.) 397, 405–10 (1872). Similarly, state courts can issue writs of mandamus to state officials but not to federal officials. See McClung v. Silliman, 19 U.S. (6 Wheat.) 598, 603–05 (1821).

      [88].    Trump v. Anderson, 144 S. Ct. 662, 667–69 (2024) (per curiam). By contrast, before Colorado disqualified Trump, no state had ever disqualified a presidential candidate under Section 3. See id. at 669. On the other hand, it is not clear that a state had any occasion to do so given the lack of disqualified individuals who ran for President from 1868 to 2024.

      [89].    Cf. Jack M. Balkin, Living Originalism 142 (2011) (noting that structural principles “explain how the Constitution works in practice and how it should work”).

      [90].    Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 610 (1952) (Frankfurter, J., concurring).

      [91].    514 U.S. 779, 803–04 (1995) (quoting 1 Joseph Story, Commentaries on the Constitution of the United States § 627, p. 435 (3d ed. 1858)).

      [92].    144 S. Ct. 662 (2024) (per curiam).

      [93].    Id. at 667 (holding broadly that “States may disqualify persons holding or attempting to hold state office” but “have no power under the Constitution to enforce Section 3 with respect to federal offices, especially the Presidency”). The context was apparently so distracting that the per curiam did not even register Justice Barrett’s potential disagreement with the Court’s conclusion that states are prohibited from enforcing Section 3 against all federal officeholders and candidates. Compare id. at 671 (“So far as we can tell, [the four Justices writing separately] object only to our taking into account the distinctive way Section 3 works and the fact that Section 5 vests in Congress the power to enforce it.”), with id. at 671 (Barrett, J., concurring in part and concurring in the judgment) (“I agree that States lack the power to enforce Section 3 against Presidential candidates.”).

      [94].    McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 435–36 (1819).

      [95].    Trump, 144 S. Ct. at 671 (Barrett, J., concurring in part and concurring in the judgment). The concurrence in the judgment jointly authored by Justices Sotomayor, Kagan, and Jackson did not object to the Court’s conclusion that states may not enforce Section 3 against any federal officeholder. Instead, those three Justices objected to the Court’s conclusion that Section 3 is not self-executing and may be enforced only by congressional legislation properly enacted under Section 5 of the Fourteenth Amendment. See id. at 673–75 (Sotomayor, Kagan, & Jackson, JJ., concurring in the judgment). The Introduction to this Article discussed the issue of self-execution.

      [96].    See Trump, 144 S. Ct. at 671 (“Nothing in the Constitution requires that we endure such chaos—arriving at any time or different times, up to and perhaps beyond the Inauguration.”); id. at 672 (Sotomayor, Kagan, & Jackson, JJ., concurring in the judgment) (“Allowing Colorado to [decide whether a presidential candidate is disqualified under Section 3] would, we agree, create a chaotic state-by-state patchwork, at odds with our Nation’s federalism principles.”).

      [97].    See Politico Staff, Tracking the Trump Criminal Cases, Politico (Nov. 6, 2024), https://www.politico.com/interactives/2023/trump-criminal-investigations-cases-tracker-list (tracking the two state and two federal criminal cases then pending against former President Donald Trump).

      [98].    That said, Trump’s New York conviction appeared to exemplify how little a state criminal conviction can matter politically.

      [99].    See supra notes 78–82 and accompanying text (discussing those provisions).

    [100].    Terminiello v. Chicago, 337 U.S. 1, 37 (1949) (Jackson, J., dissenting) (“There is danger that, if the Court does not temper its doctrinaire logic with a little practical wisdom, it will convert the constitutional Bill of Rights into a suicide pact.”).

    [101].    See, e.g., Mary Clare Jalonick et al., Trump ‘Lit that Fire’ of Capitol
Insurrection, Jan 6 Committee Report Says, PBS News (Dec. 23, 2022), https://www.pbs.org/newshour/politics/trump-lit-that-fire-of-capitol-insurrection-jan-6-committee-report-says (“The House Jan. 6 committee’s final report asserts that Donald Trump criminally engaged in a ‘multi-part conspiracy’ to overturn the lawful results of the 2020 presidential election and failed to act to stop his supporters from attacking the Capitol, concluding an extraordinary 18-month investigation into the former president and the violent insurrection two years ago.”).

    [102].    Given the extent to which the deeply disturbing events of January 6, 2021, appear to have become accepted, normalized, excused, or even justified within the base of the Republican Party, one should probably be cautious, not confident, in predicting what the future will hold with respect to the continued relevance of Section 3.

    [103].    Abraham Lincoln, Message to Congress in Special Session, in 4 Collected Works of Abraham Lincoln 421, 430 (Roy P. Basler ed., 1953) (“[A]re all the laws, but one, to go unexecuted, and the Government itself go to pieces, lest that one [concerning which political branch has the authority to suspend the writ of habeas corpus] be violated?”). It turns out, however, that President Lincoln primarily relied upon a constitutional argument, not an extra-constitutional one, in justifying his decision at the outset of the Civil War to authorize the military to suspend the writ without any action by Congress, which was out of session. See Bradley & Siegel, supra note 19, at 1256–59 (analyzing Lincoln’s arguments during that episode, including under the Habeas Suspension Clause, U.S. Const. art. I, § 9, cl. 2).

    [104].    Section 3 issues, like all constitutional questions, must be justiciable to be resolvable by a federal court. But Trump v. Anderson implicitly rejects the argument that Section 3 poses only nonjusticiable political questions: to meet due process concerns, state disqualifications of would-be state officers presumably could end up in state court, and the Court presumably thinks that it could review the state court’s judgment by certiorari. Even if other justiciability issues such as Article III standing significantly reduce the number of situations in which Section 3 might be relevant to a federal court decision, it is difficult to imagine that there are no cases in which a Section 3 question could be reached. For example, a candidate disqualified by a state officer would presumably have a cause of action against that officer under 42 U.S.C. § 1983, and the candidate could bring suit in federal court. Again, the Court could have the final word. As explained supra in the text following note 96, concerns about disuniformity are a red herring.

    [105].    One might add that it is unfair to blame the Court for not relying on an argument that was not briefed by the parties. But one could respond that the Justices should be able to work through the problem on their own, resting as it does on fundamental principles of constitutional law. They had the intuition but did not develop it well.

    [106].    See, e.g., Baude, supra note 82 (writing an opinion editorial that is critical of the Court’s reasoning and result in Trump v. Anderson).

    [107].    See generally Siegel, The Collective-Action Constitution, supra note 28.

    [108].    See id. at 355–91.

    [109].    See Graber, supra note 18, at xxxviii–xl. Section 2 of the Fourteenth Amendment required a proportionate reduction in representation in the House of Representatives for states that denied the vote to Black men. U.S. Const. amend. XIV, § 2. Regarding Section 4, see supra note 8.

    [110].    Graber, supra note 18, at xxvii.