Twenty-Third Amendment Problems Confronting District of Columbia Statehood – Derek T. Muller

Posted by on Sep 3, 2021 in Per Curiam

Download PDF

Twenty-Third Amendment Problems Confronting District of Columbia Statehood

Derek T. Muller*


The 117th Congress is the latest to consider statehood for the District of Columbia, and prospects of statehood grow ever closer after the House of Representatives approved H.R. 51.[1] The Senate is evenly divided, but a Democratic Vice President holds the potential tie-breaking vote, and filibuster reform remains a possibility. But there remains a problem facing statehood: the Twenty-Third Amendment.

If the legislation succeeds, the District of Columbia becomes a new state and leaves behind a new, substantially smaller District named “Capital.” The Twenty-Third Amendment guarantees that the new District would have three electoral votes, no matter how few people reside in it. Proposals like H.R. 51 do not adequately address this issue. Statehood, if it proceeds, should be conditioned on repeal of the Twenty-Third Amendment. Potential alternative statutory solutions to the Twenty-Third Amendment—which do not exist in the present bill—present constitutional, legal, and practical problems.[2]

In one sense, the scope of this Article is modest. It does not weigh in on the constitutionality of admitting the District of Columbia as a state. It does not address the policy question of whether it ought to become a state. It only addresses the Twenty-Third Amendment and related, practical voting rights problems. But these problems are serious and vexing; problems that present legislation is inadequate to address.


I. The Twenty-Third Amendment guarantees that the new District would have three electoral votes, no matter how few people reside in it.

The text of the Twenty-Third Amendment states:

Section 1. The District constituting the seat of Government of the United States shall appoint in such manner as the Congress may direct:

A number of electors of President and Vice President equal to the whole number of Senators and Representatives in Congress to which the District would be entitled if it were a State, but in no event more than the least populous State; they shall be in addition to those appointed by the States, but they shall be considered, for the purposes of the election of President and Vice President, to be electors appointed by a State; and they shall meet in the District and perform such duties as provided by the twelfth article of amendment.

Section 2. The Congress shall have power to enforce this article by appropriate legislation.

This Amendment guarantees the District three presidential electors. Congress directed the “manner” of appointing electors after the Amendment was ratified in 1961.[3] It provided a popular election for slates of electors, and the presidential ticket receiving the most votes would have all three of its electors chosen on behalf of the District.[4]

Under proposed legislation, the new “District constituting the seat of Government of the United States” would be known as “Capital.”[5] Capital would be entitled to three electoral votes.

The boundaries of Capital are set out in H.R. 51.[6] They roughly—but not exactly—map onto Census Tract 62.02. That tract had 33 inhabitants in the 2010 census.[7] Per the 2019 American Community Survey, the tract had 58 inhabitants.[8] The actual population of a future Capital will likely differ. A subset of that population is of voting age. The President typically does not change residency to the White House, but the President and family members might do so, which would heavily influence this new federal enclave.[9]

A small group of prospective voters who happen to reside in this federal enclave would now have three electoral votes all to themselves. Critics of the Electoral College point to disproportionate voting power in states such as Vermont and Wyoming, each guaranteed three electoral votes despite having disproportionately smaller populations than states such as Texas and California. But such disparities would be dwarfed by an enclave about 1/20,000th the size of Vermont, an enclave entitled to receive three electoral votes.

This is a serious problem. In the event the District becomes a state, there would be 541 electoral votes.[10] Three electoral votes would be about 0.5% of the total number of electors. Exceedingly narrow Electoral College margins are not unheard of, and three votes may make the difference in an election.[11] Presidential elections should not turn on a handful of Capital residents.

The new District, then, will have a handful of inhabitants and be entitled to three electoral votes. The next question asks how to resolve the issue.


II. Proposed statehood legislation does not adequately address the Twenty-Third Amendment and related voting issues.

H.R. 51 addresses problems concerning new District voting rights in Sections 221–24.[12] First, it would allow Capital inhabitants to vote in their last state of domicile; second, it would repeal the Office of District of Columbia Delegate; third, it would repeal a federal law relating to timing, transmission, and counting of electoral votes; and finally, it would allow for expedited repeal of the Twenty-Third Amendment.[13] These provisions are inadequate.

First, there is no guarantee of repeal of the Twenty-Third Amendment. The expedited procedures would allow swift floor action on an amendment to repeal the Twenty-Third Amendment.[14] But there is no assurance that the required two-thirds of each chamber of Congress would vote to repeal.[15] Nor is there any guarantee that the legislatures of three-fourths of the states would approve the repeal before the next election, if ever.[16]

Why might Members of Congress or state legislatures choose not to ratify the repeal? If there is a genuine legal controversy about whether the District could become a state,[17] both supporters and opponents of statehood have an incentive to wait for the legal process to play out. If a federal court finds that statehood is unconstitutional, then District residents would prefer to retain the Twenty-Third Amendment. Litigation takes time.[18]

Amending the Constitution is a hard thing to do, and it has been happening with increasing rarity. The Constitution has been amended once in the last 50 years, and that was to ratify an amendment approved by Congress in 1789.[19] Before that, an amendment was ratified 50 years ago, when the Twenty-Sixth Amendment was ratified on July 1, 1971.[20] Only one amendment has ever been repealed.[21] While past performance is no indication of future success, the United States has been in a lull period of amending the Constitution. Merely wishing for future events to occur is not an adequate legislative solution. Enacting a statute, then hoping for Congress and the states to ratify a constitutional amendment, is unwise. Increased uncertainty for even one presidential election would be problematic.

Second, Section 223 of H.R. 51 is misleading in its scope. It is entitled, “Repeal of law providing for participation of seat of government in election of president and vice-president.”[22] But this amendment does not “repeal” the law “providing for participation” in the presidential election. It repeals 3 U.S.C. § 21, a provision added to clarify the Electoral Count Act of 1887 after the enactment of the Twenty-Third Amendment.[23]

The Electoral Count Act provides the rules for participation in presidential elections, such as the day on which “states” participate in presidential elections, the manner in which states send their electoral votes to Congress, and the manner in which Congress counts the electoral votes. 3 U.S.C. § 21 was added to clarify that “state” in the Electoral Count Act included the District of Columbia. It solely regulates the timing, transmission, and counting of electoral votes.

The District of Columbia’s participation in presidential elections is currently codified elsewhere.[24] That law is unchanged in H.R. 51 and would remain in effect in Capital upon its admission.[25] H.R. 51 does not change presidential elections in Capital. Indeed, H.R. 51 goes out of its way to amend some portions of the Elections Code without abolishing the rules pertaining to presidential elections.[26] In other words, the title of Section 223 does not do what it purports to do.

In contrast to the claim, the section repeals what is principally a conforming amendment to the Electoral Count Act. Among other things, the Electoral Count Act instructs Congress on the method of counting electoral votes, the circumstances in which members of Congress may object to counting of votes, and the procedures for handling objections.[27] The Twenty-Third Amendment entitles the District to three electoral votes, regardless of what federal law says. Capital would still hold a presidential election, choose three electors, and send those electors to Congress. The Electoral Count Act would no longer instruct Congress how to count them.

But on January 6, Congress would still have three electoral votes from Capital when it convenes to count electoral votes—votes constitutionally authorized for the District to select. Congress might then ignore the Electoral Count Act, as it would confront a circumstance falling outside its scope. A situation encouraging Congress to create new mechanisms in counting electoral votes seems unwise.[28]

Third, Section 221 may be unconstitutional. Section 221 is designed to reduce the likelihood that there would be any voters in Capital. It compels states to permit “absent Capital voters”—eligible voters in Capital who were previously domiciled in another state—to register in their former states and request absentee ballots for federal elections.

As mentioned earlier, it is likely that 30 to 60 people would reside in Capital, and only some of them would be eligible voters. In a way, it is negligible to try to press those voters back to their former states.

Congress likely lacks the power to create voter qualifications, but states have broad power over the qualifications of voters, including “reasonable citizenship, age, and residency requirements.”[29] Congress’s power to dictate qualifications for eligible voters, even in federal elections, is much more contested.

Congress has the power to “make or alter” regulations pertaining to the “times, places and manner of holding elections for Senators and Representatives.”[30] But states determine who is eligible to vote in those elections. For the House, “the electors in each state shall have the qualifications requisite for electors of the most numerous branch of the state legislature.”[31] The same is true for the Senate.[32] There is no federal power to fix voter qualifications.

Congress’s power in presidential elections is ostensibly even less. There is no analogous “times, places and manner” clause, although “Congress may determine the time of choosing the electors.”[33] Instead, “Each state shall appoint, in such manner as the Legislature thereof may direct, a number of electors . . . .”[34] All states currently hold popular elections for the selection of presidential electors, and these states define who is eligible to vote in those elections.

The Supreme Court has held that Congress’s authority to regulate presidential elections might be broader than the Constitution’s language suggests.[35] And the Constitution does forbid states from denying or abridging the right to vote on account of race, color, or previous condition of servitude;[36] sex;[37] failure to pay a tax;[38] or age, for those over the age of 18.[39]

So where is Congress’s power to establish voter qualifications—that is, to compel a state to allow people who moved out of the state and into Capital to vote back in their former state? Supreme Court precedent on weak footing supplies some hints.

In 1970, Congress amended the Voting Rights Act with, inter alia, two provisions. First, it established the minimum age of voters in federal elections at 18 (in most states at the time, it was 21).[40] Second, it effectively reduced residency requirements in presidential elections to 30 days before the election; and if a resident moves elsewhere within those 30 days, the resident may cast a vote in the presidential election in her former state.[41]

In Oregon v. Mitchell, the Supreme Court upheld these regulations. The age requirement in federal election was upheld.[42] A similar age requirement in state election was found unconstitutional, which was the impetus for enacting the Twenty-Sixth Amendment.[43] The reason for upholding the law was elusive. No single reason commanded a majority of the Court—indeed, a majority of the Court affirmatively rejected each constitutional basis for the law. One arguable basis was the Elections Clause;[44] another, the Fourteenth Amendment.[45] But the Supreme Court concluded in 2013 when weighing a claim under the Elections Clause: “That result, which lacked a majority rationale, is of minimal precedential value here.”[46]

The Court in Oregon v. Mitchell also upheld Congress’s authority to set minimum residency requirements on the basis of the right to travel.[47] A strong majority of the Court endorsed Congress’s power to help address the right to travel across parts of the country, here prohibiting states from treating new arrivals from other states differently.[48]

In 2013, in Arizona v. Inter Tribal Council of Arizona, Inc.,[49] a seven-justice majority of the Court endorsed a more limited understanding of congressional power: “Prescribing voting qualifications, therefore, ‘forms no part of the power to be conferred upon the national government’ by the Elections Clause, which is ‘expressly restricted to the regulation of the times, the places, and the manner of elections.’ The Federalist No. 60, at 371 (A. Hamilton).”[50] A strong majority of the Supreme Court has pressed back on Oregon v. Mitchell. It opens questions about whether Congress has much authority, if any, to dictate voter qualifications to states in federal elections.

Furthermore, Section 221 may threaten the votes of tens of thousands of military and overseas voters. Section 221 language tracks identical language in the Uniformed and Overseas Citizens Voting Act of 1986 (“UOCAVA”).[51] UOCAVA has never faced a serious legal challenge, no doubt in part due to the politically-fraught position of a plaintiff requesting the federal courts to strip military and overseas voters of federal voting opportunities.[52] Section 221, if enacted, would assuredly be challenged by some state that, under Section 221, would resist voter registration of an absentee Capital voter. An “absentee Capital voter” might attempt to register to vote in a former state, be denied, and sue for the right to register. Or a state, facing an absentee Capital voter, might seek injunctive relief to prevent enforcement of Section 221. A federal court—perhaps ultimately the Supreme Court—might repudiate Congress’s power to enact Section 221.[53] And with an on-point precedent, a litigant or a state might be emboldened to challenge UOCAVA.[54]

I use “may threaten” deliberately. It is possible, of course, that this litigation fails for a myriad of reasons. Perhaps no one sues. Perhaps federal courts and the Supreme Court ultimately conclude that Congress does have the authority to compel states to keep voters, who moved out of the state into Capital long ago. Perhaps no one would try to apply a holding of Section 221 to UOCAVA. And it is also possible that states would each voluntarily accede to UOCAVA’s rules even after it were found unconstitutional.

Simply put, these are, in my judgment, legitimate areas of uncertainty. H.R. 51 inadequately addresses them.


III. Statehood for the District of Columbia should be conditioned on repeal of the Twenty-Third Amendment.

Requiring repeal before statehood is hardly a new concern.[55] But the existing repeal mechanisms are inadequate, and a repeal could take different forms. One might be a straight repeal of the Twenty-Third Amendment, similar to the text of the Twenty-First Amendment: “The twenty-third article of amendment to the Constitution of the United States is hereby repealed.” But repealing, then awaiting statehood, seems like a suboptimal solution for District residents.

An alternative amendment might condition repeal of the Twenty-Third Amendment in the event the number of inhabitants in the District, constituting the seat of government, falls below 10,000. If the bulk of the present District of Columbia becomes a state or is retroceded, the conditional repeal would take effect, and the Twenty-Third Amendment would cease to apply to the remaining federal enclave.

Relatedly, a constitutional amendment could expressly authorize Congress to establish voter qualifications for inhabitants in the District constituting the seat of government in federal elections after repeal. This provision would obviate concerns that Section 221 might be unconstitutional.

In its current form, H.R. 51 allows for an expedited consideration of a repeal of the Twenty-Third Amendment, but only if the text of that amendment is “solely” dedicated to repealing the Twenty-Third Amendment.[56] Alternative amendments like those described would not benefit from any expedited consideration. And as mentioned, there is no guarantee of a repeal.


IV. Potential alternative statutory solutions to the Twenty-Third Amendment present constitutional, legal, and practical problems.

It is worth noting that none of the alternatives discussed below exists in H.R. 51. They are merely the stuff of conjecture. And because there are an unlimited number of alternative proposals that might arise, this Article addresses three that have received material attention.[57] First, Congress could award the electoral votes to nobody. Second, Congress could award them to the winner of the Electoral College. Third, Congress could award them to the winner of the “national popular vote,” the aggregation of the vote across the states. Each faces problems.

First, if Congress decided not to appoint electors by, say, repealing relevant provisions of the D.C. Code, Congress would be derelict in its duty. The Court has repeatedly noted that, in the context of elections, “shall” places a duty upon states.[58] The Constitution provides that “Each state shall appoint . . . a number of electors”[59] and that “The District . . . shall appoint” electors.[60] One construction of these mandates is that the District has a duty to appoint electors. Congress’s power—to “direct” the “manner” of appointing electors—is to provide the framework for how the District goes about that appointment.[61] And even if Congress decided not to award electors in the next presidential elections, Capital’s three electoral votes would be waiting for the next Congress. Congress could at any time change its mind and decide that the federal enclave should appoint electors. It would be a significant temptation and hardly an optimal solution.

Second, Congress might enact a new law awarding Capital’s electors to the winner of the Electoral College. But this is unconstitutional because it is impossible to comply with the Constitution’s terms. Presidential electors must give their votes on the “same” day “throughout the United States.”[62] To award a slate of presidential electors to the winner of the Electoral College would require waiting until after all the electors cast votes. Even then, controversies might arise in Congress if a state submitted multiple slates, or in the event a controversy arose over the legitimacy of the appointment of an elector in the first place.[63]

A more generous reading of this alternative proposal might be that sometime after Election Day, the apparent winner of the Electoral College would be awarded Capital’s electors. But that date, used to determine presidential transitions, is uncertain and in recent years yielded controversy.[64] And it would incentivize states to drag their feet or offer competing slates of electors to influence the outcome of Capital’s electors.

This proposal, and the proposal that Congress award Capital’s electors to the winner of the national popular vote, suffer from another problem. The proposals emphasize Congress’s power to “direct” the “manner” of appointing presidential electors. That power is undoubtedly broad.[65] But it is only one clause of a two-clause directive. The Twenty-Third Amendment provides, “The District . . . shall appoint,” and “Congress” “may direct” the “manner.”[66] The first is the who. The second is the how.

If Congress directs that it will award Capital’s electors based on the popular vote of the inhabitants of France, it would be hard to say that “the District” has appointed the electors. In the same way, if Congress chooses a manner that awards electors based on what happens in the rest of the United States, it is hard to say that “the District” has appointed anyone. Every choice of presidential electors in American history has been based on some sort of election within the state, not outside the state.[67]

Supporters of these alternative proposals have cited the Supreme Court’s recent opinion in Chiafalo v. Washington.[68] In Chiafalo, the Court approved a fine levied by the State of Washington on electors who cast votes inconsistent with state law, which required electors to vote for the candidate they had pledged to support.[69] The electors’ candidate had won the statewide popular vote. The Court emphasized that Washington had broad discretion over how to handle electors’ discretion, and the Constitution did not take away the power to do what Washington did.[70]

Chiafalo certainly speaks broadly about how states may control electors. But it says nothing about the predicate question: the selection of electors in the first place. Once a state holds a popular election, and the people choose a presidential candidate who carries the popular vote in a state, electors can be bound to vote for that candidate. But Chiafalo does not authorize legislatures to compel electors to vote for anything the legislature wants. The legislature cannot violate the Equal Protection Clause or add qualifications to presidential candidates.[71] The power to compel electors may not extend to compelling electors to vote for a dead candidate.[72] Likewise, if Congress cannot direct the power of appointment outside the District, then Chiafalo is simply inapplicable.

The third proposal, awarding Capital’s electors to the winner of the “national popular vote,” suffers from many legal and practical problems, written about extensively elsewhere in the context of the National Popular Vote Compact.[73] To start: how and when does Congress determine the winner of the “national popular vote”? How does it handle litigation and recounts—or a state that refuses to recount because the margin in the home state is wide but the margin nationwide is narrow?

There are major Equal Protection problems, too. Voter eligibility rules vary (or may vary in the future) from state to state—incarcerated felons can vote in some states, but ex-felons are prohibited in others. Voting procedures vary, from strict photo voter identification laws to no identification laws at all. Polling places are open in Hawaii well after the polls close in Kentucky. While some variance of election procedures is inevitable, other variations are so broad that they implicate the Equal Protection Clause.[74]

Ballot access standards vary from state to state. For instance, 2016 Green Party nominee Jill Stein and 2020 independent candidate Kanye West, among others, appeared on the ballot in only some states. How can we assemble a “national popular vote” when Americans aren’t even looking at the same ballot?

If it’s a single constituency election—like a nationwide popular election for president—we have to have a uniform set of rules. A system that creates a “national popular vote” is really just adding up the votes from 51 separate presidential elections. It may well violate the Equal Protection Clause, and it certainly presents significant practical problems.

Again, none of these proposals has been introduced in Congress, and they are not in the text of H.R. 51 as it has been presented before you today. But even these proposals, presented as “solutions” to the Twenty-Third Amendment dilemma, are wanting.

* * *

The Twenty-Third Amendment problems facing H.R. 51 are significant. Anomalies will exist.  If statehood (or retrocession) is the goal, the best path forward is a constitutional amendment that would repeal the Twenty-Third Amendment contingent on future statehood. That amendment should also expressly empower Congress to address the voting status of inhabitants of any federal enclave. Alternative solutions—both in the bill and in ideas floated about in the public domain—are inadequate to address constitutional, legal, and practical problems.

In a way, it may seem odd that the Twenty-Third Amendment, which empowered the electoral power of the District of Columbia, is now a barrier to statehood via ordinary legislation. But constitutionalizing a subject crystalizes it. The Constitution limits ordinary legislation. The only way to undo what the Constitution has done is through an amendment.



* Bouma Fellow in Law and Professor of Law, University of Iowa College of Law. Special thanks to Kevin Kim and Mark McDermed for their help on this piece. This Article is based upon testimony given to the U.S. Senate Committee on Homeland Security & Governmental Affairs, “Examining D.C. Statehood,” June 22, 2021.

[1] H.R. 51, 117th Cong. (2021). An identical bill was introduced in the Senate, S. 51, 117th Cong. (2021).

[2] For example, if retrocession is considered as an alternative proposal, the same analysis applies.

[3] Act of Oct. 4, 1961, Pub. L. 87-389, 75 Stat. 817 (amending Act of August 12, 1955, relating to elections in the District of Columbia).

[4] See id. See also D.C. Code § 1-1001.10(a)(2) (2021).

[5] H.R. 51, 117th Cong. § 111 (2021).

[6] H.R. 51, 117th Cong. § 112 (2021).

[7] See Population by Race and Hispanic or Latino Origin in the District of Columbia – Census Tracts: 2010, DC Office of Planning, [].

[8] ACS Demographic and Housing Estimates, Census Tract 62.02, District of Columbia, DC, United States Census Bureau (2019), [].

[9] See Charlie Savage and Emily Cochrane, White House Is Said to Quietly Push Change to D.C. Statehood Bill, N.Y. Times (May 13, 2021), (“It is not clear how many, if any, potential voters would be left there. The only residence in the rump federal enclave would be the White House; presidential families traditionally choose to vote in their home states, but nothing forces them to do so. In theory, homeless people might also claim residency in the envisioned enclave.”).

[10] One aside­—if an objective of H.R. 51 is to oversee the ultimate repeal of the Twenty-Third Amendment, it should also ensure that the size of the House is an odd number, not an even number, in part to prevent the possibility of a tie in the Electoral College. At present, the potential for a tie exists in the Electoral College because there are 538 votes, which evenly divides in half into 269. In the event no candidate wins a majority of the Electoral College, the election is sent to the House of Representatives to select a president between the tied candidates, and each state receives one vote. See U.S. Const. amend. XII. There are 538 votes because there are 100 senators (always an even number, as there are two senators in each state); 435 members of the House (presently fixed by statute); and the District of Columbia is guaranteed the smallest state’s allotment of electors, which, since the Twenty-Third Amendment’s ratification, has been three. An even number plus an odd number plus an odd number always yields an even number. H.R. 51 would permanently increase the size of the House by one. See H.R. 51, 117th Cong. § 102(d) (2021). That would yield 102 senators, 436 members of the House, and three new District electors, for a total of 541, an odd number. If the Twenty-Third Amendment is repealed, that figure would drop back down to 102 plus 436, or 538, an even number. Increasing the size of the House to 437—or to some other odd number—would be preferable, because it would eliminate the remote possibility of a tie.

[11] In 2000, George W. Bush received 271 votes in the Electoral College, needing 270 votes to secure a majority. In 1876, Rutherford Hayes received 185 votes in the Electoral College, the minimum number of votes to secure a majority.

[12] H.R. 51, 117th Cong. §§ 221–24 (2021).

[13] Id.

[14] Id. § 224.

[15] See U.S. Const. art. V.

[16] If the District were to become a state, it would take 39 states to ratify an amendment, for there would be 51 states and it takes three-fourths of the states to ratify an amendment. Id. But if there is a dispute about whether the new District were a state, one might wonder whether there were just 50 states requiring 38 legislatures’ approval. Or, if the new District’s legislature approved the amendment, there might be disputes about whether it could validly approve an amendment in the first place while a dispute remained pending.

[17] This testimony takes no position on merits or the likelihood of the outcome of that litigation. It simply notes that it is a non-trivial possibility.

[18] But see Letter from Law Professors to Congressional Leaders 4 (May 22, 2021), [hereinafter May 22 Letter] (Re: Washington, D.C. Admission Act and the likelihood of a constitutional challenge) (“None of the other 50 States has reason to seek to retain three electors for a largely unoccupied seat of government.”).

[19] U.S. Const. amend. XXVII.

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

[21] U.S. Const. amend. XXI, § 1 (repealing U.S. Const. amend. XVIII).

[22] H.R. 51, 117th Cong. § 223 (2021).

[23] See id.

[24] See D.C. Code § 1-1001.10(a).

[25] See H.R. 51, 117th Cong. § 114 (2021)

[26] See id. at § 222 (amending portions of the District of Columbia Elections Code of 1955 to strike the election of a “delegate” to the House of Representatives, but not the remainder of the Code).

[27] 3 U.S.C. § 15.

[28] See May 22 Letter, supra note 15, which argues that Section 223 “provid[es] for the repeal of the provision of federal law that establishes the current mechanism for District residents to participate in presidential elections.” Signed by a cohort of law professors, this is inaccurate and confuses 3 U.S.C. § 21, D.C. Code § 1-1001.10(a) with the Twenty-Third Amendment. Section 223 repeals the provision making the District a “state” for purposes of the timing, transmission, and counting of electoral votes. It does not “repeal” the “mechanism” for “participat[ion],” which is codified at D.C. Code § 1-1001.10(a), and which is not altered by H.R. 51.

See also id, where the letter inaccurately states that 3 U.S.C. § 21 “provid[es] that the District residents may select presidential electors.” Again, 3 U.S.C. § 21 relates to the timing, transmission, and counting of electoral votes, while D.C. Code § 1-1001.10(a) provides for the participation of District residents in presidential elections.

The May 22 letter confuses 3 U.S.C. § 21 as linked to the Twenty-Third Amendment, instead of Congress’s related power under the Twelfth Amendment (amending Article II of the Constitution) to count electoral votes. Repealing 3 U.S.C. § 21 would still entitle Capital, under D.C. Code § 1-1001.10(a), to elect presidential electors. Capital would no longer be obligated to vote on the “Tuesday next after the first Monday in November,” Electoral Count Act, 3 U.S.C. § 1, but it would still be required to do so under D.C. law, D.C. Code § 1-1001.10(a)(2). Some provisions of federal law would necessarily apply to Capital’s electors. First, Capital’s electors would be required to vote on the “first Monday after the second Wednesday in December.” 3 U.S.C. § 7. Even with the repeal of 3 U.S.C. § 21, the Constitution mandates that Congress determines “the day on which [electors] shall give their votes,” and that “day shall be the same throughout the United States.” U.S. Const. art. II, § 1, cl. 4. The Constitution also requires that “the votes shall then be counted” “in the presence of the Senate and House of Representatives.” U.S. Const. amend. XII. This power requires no implementing legislation. Contra 1 Annals of Cong. 486, 16-18 (1789) (Joseph Gales ed., 1834) (reporting on the convening of Congress to count electoral votes). Congress can, of course, act pursuant to implementing legislation, and it has acted pursuant not simply to the Electoral Count Act, but also pursuant to a concurrent resolution. See, e.g., S. Con. Res. 1, 117th Cong. (Jan. 3, 2021).

Respectfully, the May 22 letter appears to defend a hypothetical bill that does not exist in H.R. 51. Media commentary makes the same mistake, too. See Savage & Cochrane, supra note 9 (noting that in the bill, “legal procedures for appointing any electors would be rescinded”).

[29] Kramer v. Union Free Sch. Dist. No. 15, 395 U.S. 621, 625 (1969).

[30] U.S. Const. art. I, § 4, cl. 1.

[31] U.S. Const. art. I, § 2, cl. 1.

[32] U.S. Const. amend. XVII.

[33] U.S. Const. art. II, § 1, cl. 4.

[34] U.S. Const. art. II, § 1, cl. 2.

[35] See Burroughs v. United States, 290 U.S. 534, 544–45 (1934).

[36] U.S. Const. amend. XV.

[37] U.S. Const. amend. XIX.

[38] U.S. Const. amend. XXIV, § 1.

[39] U.S. Const. amend. XXVI, § 1.

[40] Voting Rights Act Amendments of 1970, Pub. L. No. 91–285, 84 Stat. 314 (codified as amended at 52 U.S.C. § 10701 (1975)); see also Oregon v. Mitchell, 400 U.S. 112, 117 (1970).

[41] Voting Rights Act Amendments of 1970, Pub. L. No. 91–285, 84 Stat. 314 (codified as amended at 52 U.S.C. § 10502 (1970)); see also Mitchell, 400 U.S. at 147–50.

[42] See Mitchell, 400 U.S. at 118.

[43] See id.

[44] See id. at 119–24.

[45] See id. at 126–28.

[46] Arizona v. Inter Tribal Council of Ariz., Inc., 570 U.S. 1, 16 n.8 (2013).

[47] See Brian C. Kalt, Unconstitutional but Entrenched: Putting UOCAVA and Voting Rights for Permanent Expatriates on a Sound Constitutional Footing, 81 Brooklyn L. Rev. 441, 475–77 (2016).

[48] Id.

[49] 570 U.S. 1 (2013).

[50] Id. at 17 (emphasis removed).

[51] Compare 52 U.S.C. § 20302(a)(1) & (2) (2017) with H.R. 51, 117th Cong. § 221(a)(1)(A) & (B) (2021).

[52] See Kalt, supra note 47, at 500–01 (2016).

[53] See id. (describing constitutional problems with UOCAVA).

[54] The May 22 letter claims, “no potential litigant would suffer the constitutionally cognizable injury required to establish standing to bring a court challenge to such legislative action” that might require District electors to cast votes for the candidate who receives the most electoral votes. See May 22 letter, supra note 18, at 5–6 n.11. Part of the justification is that Capital residents would be able to vote elsewhere under Section 221. But the May 22 letter elides over whether there are any infirmities with Section 221 and how litigation might proceed.

[55] See, e.g., Mike DeBonis, Is the D.C. statehood bill constitutional?, Wash. Post (Sept. 15, 2014), [] (citing comments of Roger Pilon “that the 23rd Amendment would need to be repealed, lest the tiny number of residents left in the federal enclave stand entitled to three presidential electoral votes”).

[56] H.R. 51, 117th Cong. § 224 (2021).

[57] See, e.g., the May 22 letter, supra note 18; Savage & Cochran, supra note 9.

[58] See, e.g., Arizona v. Inter Tribal Council of Ariz., Inc., 570 U.S. 1, 8 (2013); U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779, 804–05 (1995). See also Chiafalo v. Washington, 140 S. Ct. 2316, 2329 (2020) (Thomas, J., concurring in the judgment).

[59] U.S. Const. art. II, § 1, cl. 2 (emphasis added).

[60] U.S. Const. amend. XXIII, § 1 (emphasis added).

[61] For arguments surrounding Congress’s duty, consider Peter Raven-Hansen, The Constitutionality of D.C. Statehood, 60 G.W. L. Rev. 160, 184–89 (1991) (weighing the “close” debate).

[62] U.S. Const. art. II, § 1, cl. 4.

[63] See 3 U.S.C. § 15 (1948).

[64] See, e.g., Andy Card and John Podesta, The life-threatening costs of a delayed transition, Wash. Post (Nov. 10, 2020), []; Letter from Emily W. Murphy, Administrator, U.S. General Services Administration, to the Honorable Joseph R. Biden, Jr. (Nov. 23, 2020), [].

[65] See, e.g., McPherson v. Blacker, 146 U.S. 1, 35 (1892) (describing the power as “plenary”).

[66] U.S. Const. amend. XXIII, § 1.

[67] See Norman R. Williams, Why the National Popular Vote Compact Is Unconstitutional, 2012 B.Y.U. L. Rev. 1523, 1540, 1572–73, 1581–83 (2012).

[68] 140 S. Ct. 2316 (2020). See, e.g., Jessica Bulman-Pozen & Olatunde Johnson, The Electoral College Shouldn’t Get in the Way of D.C. Statehood, Take Care  (July 7, 2020), [].

[69] Chiafalo, 140 S. Ct. at 2322.

[70] Id. at 2324–25.

[71] Id. at 2324 n.4.

[72] Id. at 2328 n.8.

[73] Much analysis in this section draws from Derek T. Muller, The Electoral College and the Federal Popular Vote, 15 Harv. L. & Pol’y Rev. 129 (2020), []; Michael Morley, The Framers’ Inadvertent Gift: The Electoral College and the Constitutional Infirmities of the National Popular Vote Compact, 15 Harv. L. & Pol’y Rev. 81 (2020), []; Derek T. Muller, Invisible Federalism and the Electoral College, 44 Ariz. St. L.J. 1237 (2012); Norman R. Williams, Reforming the Electoral College: Federalism, Majoritarianism, and the Perils of Subconstitutional Change, 100 Geo. L.J. 173 (2011).

[74] See Morley, supra note 73 (discussing Equal Protection Clause concerns); see also Bush v. Gore, 531 U.S. 98 (2000) (per curiam).

Read More »

The Twenty-Seventh Amendment: Meaning and Application – GianCarlo Canaparo and Paul J. Larkin, Jr.

Posted by on Sep 2, 2021 in Per Curiam

Download PDF


The Twenty-Seventh Amendment: Meaning and Application

GianCarlo Canaparo* & Paul J. Larkin, Jr.**­



By 2021, lawyers have become accustomed to giving the term “law” a host of different meanings depending on its source. A law had one meaning under pre-Norman law, when it was one of the set of unwritten traditions embraced by decentralized local tribes or the written “dooms” issued by various local chieftains.[1] That term could take on a different meaning at English common law when Parliament was in its infancy and the royal courts were the sou­­rce of most of the “law” known throughout the kingdom.[2] In this nation, “law” was the product of a similar trial-and-error process of definition[3] before Congress overtook the courts as the principal federal lawmaking body.[4] Of course, today the regulatory state issues far more “laws” than Congress does[5] and its assorted rules, regulations, and even letters might or might not have the same legal status as an act of Congress.[6] The bottom line is this: The term “law” could have an entirely different understanding today depending on whether an edict is found in the Constitution of the United States,[7] in the United States Code,[8] in the Code of Federal Regulations,[9] or in an agency’s memorandum.[10]

With that in mind, consider our most recent constitutional amendment, the Twenty-Seventh. Like many of its kin, that Amendment is a bit of a troublemaker. Not a big one, mind you, but a bigger one than its simple text might suggest. That text is this: “No law, varying the compensation for the services of the Senators and Representatives, shall take effect, until an election of Representatives shall have intervened.”[11]

Ironically, the Twenty-Seventh Amendment was one of the first proposed in 1789 with the first draft of the Bill of Rights.[12] It wasn’t until 1992—203 years later—that the requisite number of states ratified it. That action caused an immediate flurry of scholarly and political debate about several issues, such as whether the Amendment had expired, the implications for Article V of a delayed ratification,[13] and whether the amendment process was a useful means of obtaining desired political outcomes.[14]  Those who argued that it had expired based their conclusions primarily on dicta from Dillon v. Gloss,[15] in which the Supreme Court held that Congress could set a reasonable time period in which states must ratify the Eighteenth Amendment.[16] In dicta, the Court declared that “ratification must be within some reasonable time after the proposal.”[17] Contemporaneity, it stated, is necessary to “reflect the will of the people in all sections at relatively the same period, which of course ratification scattered through a long series of years would not do.”[18] Some scholars considered the related question of whether the Court could or should resolve Article V questions or whether those issues were political questions best left to Congress.[19] Finally, for Professor Bernstein, the Twenty-Seventh Amendment raised concerns about “amendment politics” and the “willingness of right-wing politicians to reach for Article V as if it were a fire-ax on the wall.”[20]

Those debates have simmered down, but disputes about the Amendment’s text and application still pop up.[21] There has been litigation,[22] some scholarly engagement,[23] and political dispute over the Amendment’s meaning and application.[24] Disputes over the Amendment likely will continue to surface and require additional guiderails. This Essay attempts to provide them by ascertaining the Amendment’s meaning and implications according to its text and history.

This Essay proceeds in three Parts, each of which focuses on issues of interpretation and implementation raised by components of the Amendment. Part I examines the term “law” and concludes that it has the same technical meaning that the word does for purposes of the Bicameralism and Presentment requirements of Article I: namely, bills passed by Congress and signed by the President (or repassed over his veto).[25] The Amendment does not apply to any other government action. We also examine an important implication of this conclusion, namely, the Amendment cannot be used as a basis for holding an act of Congress unconstitutional, with rare exceptions. Taken together with the Ascertainment Clause, however, the Amendment does prohibit Congress from delegating its compensation-setting power to any other body. Part I also discusses the problems raised by trying to find a uniform meaning of what should be an elementary term—“law”—in a charter devoted to the rule of law. Part II examines the meaning of “varying” and “compensation,” and concludes that a law varies compensation when it changes the terms of an existing law enacted pursuant to the Ascertainment Clause.[26]  Part III examines the meaning of “shall take effect” and “an election . . . shall have intervened” and concludes that the former refers to the moment when members are entitled to receive a quantifiable change to their compensation, and the latter means that a new Congress has been seated.

I. The Meaning of “Law”

The Twenty-Seventh Amendment provides that “no law” may take effect until a House election has intervened.[27] The Constitution does not define “law” and the word has more than one meaning in that document. Article I, Section 7, for example, provides a technical definition: a bill passed by both houses of Congress and signed by the president (or repassed over his veto).[28] But the First Amendment’s declaration that “Congress shall make no law. . . .”[29] prohibits much more than Article I “Laws” as do the words “No State shall make or enforce any law” in the Fourteenth Amendment.[30] Article I, Section 8, Clause 10, meanwhile, refers to the “Law of Nations.” [31]

It turns out that the debate over the meaning of the term “law” in the Twenty-Seventh Amendment raises a marvelously complex issue of constitutional law. The reason is that, over time, the Supreme Court has followed different processes of legal interpretation that can lead to very different outcomes depending on the methodology that one adopts. A common-law method of interpretation, for example, would deemphasize the text of the Constitution and accentuate the Supreme Court’s rulings construing the charter’s terms, as well as the policies that a particular provision sought to protect and foster, all in the name of giving effect to the subjective intent of the people who drafted and approved the Constitution.[32]  By contrast, a textually oriented approach to constitutional interpretation would highlight the text that the Framers and Ratifying Conventions considered and approved in recognition of the facts that words have generally well understood meanings for particular people at specific times and that this geographically and temporally defined range of meanings supplies the universe of legitimate interpretations that the text may have.[33] Still other methods of interpretation might lead to any number of meanings.

Professor Adrian Vermeule, for example, seems to pick out of thin air a theory interpreting the Twenty-Seventh Amendment based on a thought experiment by John Rawls, which Vermeule argues, explains the Amendment’s purpose.[34] Of course, as far as the Framers were concerned, Vermeule’s Rawlsian theory of interpretation might as well have been Martian because a novel theory created by someone born 134 years after the Convention of 1787 had adjourned did not play, and could not have played, any role in the Framers’ understanding of the words they put on paper.

Because of the Supreme Court’s inconsistent approach to defining “law,” to pick one interpretive method is to take a position that is at odds with some body of case law. Here, for example, we take a textually oriented approach to defining “law” (and the other terms of the Amendment) that reveals a meaning irreconcilable with the Court’s reading of the same word in the First Amendment. The unavoidable conclusion is that either we are wrong or the Court is wrong. We have opinions about that question, but this is not the time or place to set out a grand unified theory of the meaning of a “law.”[35] Suffice it to say, we acknowledge that the originalist approach we follow here is at odds with and has implications for the Supreme Court’s interpretation of that term in other contexts. But that is a question for another day. Here and now, we turn our attention solely to the meaning of that word in the Twenty-Seventh Amendment. As it turns out, there are very good reasons why, in that context, “law” does not and should not bear the same expansive definition the Court has given it in other settings.

That conclusion, however, is not necessarily intuitive when we consider that, originally, James Madison intended for the “compensation amendment” to accompany the First Amendment in the Bill of Rights.[36] It may be tempting, therefore, to interpret “No law . . . shall take effect” in a similar manner to “Congress shall make no law.”  Both phrases, after all, are the only constraints included in the original Bill of Rights that expressly limit lawmaking.[37] Indeed, in several Twenty-Seventh Amendment disputes to-date, one party has argued that “law” means something other than a bill passed by both houses and signed by the president.[38] Those arguments suggest an understanding of “law” that is more aligned with the expansive meaning of that term that the Supreme Court has given it in the First Amendment.

There are two good reasons, however, not to give “law” in the Twenty-Seventh Amendment a loose meaning like it has in the First Amendment. The first reason is that the Twenty-Seventh Amendment’s text and history foreclose it. The Amendment was not born of nothing and poured out over blank parchment like light over the dark face of the deep.[39] Rather, it followed and was meant to amend the Ascertainment Clause, which says that compensation for members of Congress shall be “ascertained by Law.”[40] Indeed, any examination of the meaning of the Twenty-Seventh Amendment must start with the Ascertainment Clause; the Amendment cannot be divorced from it.[41]

It would be a mistake to give an expansive reading to the term “law” in the Twenty-Seventh Amendment. The history of the Ascertainment Clause makes clear that, “Law” has its technical Article I meaning—a bill passed by both houses and signed by the president.[42] There had been a fierce debate among the Framers about how members of Congress ought to be paid.[43] One faction believed the states ought to pay their representatives, while the other thought that the nation, through the federal government, ought to do so.[44] The debate seems to have assumed, as a matter of course, that if the federal government paid the members, that pay would be set by law.[45] This makes sense because the Constitution the Framers were debating gave the power over the federal purse to Congress.[46] Moreover, the Framers were acutely aware of the battles between Parliament and the English Crown over spending authority and realized that “control over the power of the purse was the foundation of Parliament’s ability to resist the authority of the king.”[47] That power, James Madison also noted, was in House members “alone.”[48] Accordingly, the Ascertainment Clause’s use of the word “Law” means, very simply, that members’ compensation must be set by the legislative process defined by Article I.

It follows that “law” in the Twenty-Seventh Amendment should have the same meaning as it does in Article I because that Amendment fixes conditions on the legislative process and timing of the laws enacted pursuant to the Ascertainment Clause.[49] In this way, the Twenty-Seventh Amendment is like the Origination Clause[50] and the bicameralism and presentment requirements.[51] All of these provisions establish procedural requirements that must be satisfied before an act of Congress may either become law or take effect.

The second reason that the Court’s expansive definition of “law” in the First Amendment need not apply to the Twenty-Seventh Amendment is that the amendments serve different purposes. Unlike the First Amendment, the Twenty-Seventh is concerned “with an issue of governmental structure rather than substantive individual right[s].”[52] It is more accurate to say the Twenty-Seventh Amendment is concerned with legislative procedure than governmental structure, but the point stands.

The conclusion that “law” within the Twenty-Seventh Amendment takes on the narrow technical definition of Article I resolves two of the debates about the Amendment.[53] The first was raised in Boehner v. Anderson.[54] There, Representative John Boehner challenged part of the Ethics Reform Act of 1989 that created automatic annual cost of living adjustments (COLAs) for members of Congress.[55] He argued, in relevant part, that each COLA was a new “law” for purposes of the Amendment, and that they were therefore unconstitutional because half of them (those in odd-numbered years) took effect without an intervening election.[56] The court acknowledged that “[t]he Constitution does not define a law except to say (at least implicitly) that it is the product of the legislative process.”[57] By that measure, the only law at issue was the Ethics Reform Act itself, which established the COLAs starting in 1991, after an intervening election.[58] The only other court to reach this issue came to the same conclusion.[59]

The most recent debate over the Twenty-Seventh Amendment is also resolved by giving “law” its narrow Article I meaning. In 2021, the House passed rules requiring Representatives to use metal detectors when entering the House Chamber enforced by fines that would be deducted from their salaries.[60] Several members objected on the grounds that the fines violated the Twenty-Seventh Amendment.[61] This argument fails because the House rules are not Article I laws, and therefore fall outside the scope of the Twenty-Seventh Amendment.[62] This conclusion is bolstered by Section 5 of Article I, which permits each House to “determine the Rules of its Proceedings, punish its members for disorderly Behaviour . . . .”[63] That the Constitution draws a line between House rules and laws supports the argument that the one is not the other. Finally, a deduction from a member’s pay—whether because of a violation of a congressional rule, a parking ticket, or an increase in the federal income tax—does not implicate the accountability concerns that underlie the Twenty-Seventh Amendment.[64]

That “law” takes its technical definition carries an important, but heretofore overlooked, implication: in general, the Amendment may not be used to strike down laws that violate it. Instead, it should be read as providing a default timing rule that is incorporated by any law varying compensation that lacks a similar timing term or that replaces the timing term in a law that purports to vary compensation before an election has intervened. This is implied by the Amendment.  It does not prevent passage of a law that would violate its terms; instead, it merely prevents that law from taking effect “until an election of Representatives shall have intervened.”[65] Unlike the Origination Clause and the bicameralism and presentment requirements, the Twenty-Seventh Amendment imposes its requirement after a law has fulfilled the Constitution’s procedural requirements. Other limitations on government action, both procedural and substantive—including the Origination Clause, the bicameralism and presentment requirements, the Bill of Attainder Clause,[66] the Emoluments Clause,[67] the First Amendment, and so forth—either state or imply that violations are void from the outset. Not so with the Twenty-Seventh Amendment. It permits enactment of laws that violate it but delays their effect until its timing condition is satisfied.

Two examples illustrate how the Amendment works. Consider first a law that changes members’ salaries but that sets no date upon which that change shall take effect. The law is missing a required term: A date after the next election. A straightforward reading of the Amendment does not lead to the conclusion that the law should be struck down, only that the earliest moment at which it may take effect is after the next House election. Moreover, a court would avoid striking it down as unconstitutional because a plausible constitutional explanation is present.[68] That is, the Amendment provides the default term for any statute that omits a provision setting the effective date of the change in compensation. Thus, the change in salary effected by any such law would simply be deferred until immediately after the next election.

Next, consider a bill that explicitly changes members’ salaries effective before the next election. The Amendment does not prevent that bill from becoming law. So again, a court faces the choice between striking down the law or superseding its timing provision with the Twenty-Seventh Amendment’s. The latter option accords both with the text of the Amendment and the avoidance doctrines.[69]

The takeaway from all this is that the Amendment generally cannot be used to strike down laws that vary compensation before an intervening election.[70] An exception to that rule might exist, however, for a law that makes an immediate change to compensation in response to a discrete event such that substituting the Amendment’s timing provision for the existing one would entirely frustrate the law’s purpose.[71] For example, in 2013 the House Rules Committee proposed a bill that would withhold Congress’ pay if members failed to pass a budget by April 15 and release the pay either when they passed a budget or at the end of the current Congress.[72] It is impossible to include the Amendment’s timing provision without frustrating the purpose of the bill. If members failed to pass a budget, their pay could not be withheld until after the next election, but by then their salaries would be restored regardless of whether the Congress had passed a budget. Thus, using the Twenty-Seventh Amendment as a source of default language would entirely frustrate Congress’s intent. In that limited circumstance the Amendment might be used to strike down a law.

One final observation and implication. As the Ascertainment Clause informs our understanding of the Twenty-Seventh Amendment, so too does the latter inform our understanding of the former. The purpose of the Amendment is to make Congress politically accountable for the salary decisions it makes pursuant to the Clause.[73] It follows from this purpose that Congress may not escape this political accountability by delegating its power (or obligation) to set members’ compensation. To delegate compensation-setting authority to any other body conflicts with the text of both provisions, which require an Article I law, and with the purpose of the Amendment because it would relieve Congress of the political accountability that the Amendment creates.[74]

The history of salary laws accords with that understanding. Congress set members’ compensation by law for the first 179 years after the states ratified the Constitution.[75] It was not until the Postal Revenue and Salary Act of 1967 that Congress first attempted to delegate its salary-setting authority.[76] The Salary Act empowered a commission to recommend congressional salaries to the President and provided that if the President accepted the salaries, they would take effect unless one House of Congress expressly disapproved of them.[77] Before the Twenty-Seventh Amendment was ratified, that delegation was upheld by a district court on no firmer basis than that the “growing complexity of all governmental functions” made it necessary to accommodate a “flexible approach.”[78] That decision cannot be squared with text of the Ascertainment Clause, and it cannot be squared with the subsequently ratified Twenty-Seventh Amendment, which underscores that Congress may not escape political accountability for its compensation decisions.[79]

II. The Meaning of “Varying” and “Compensation”

We begin with “compensation” because when considering whether and when compensation has been varied, it helps first to exclude whatever is not compensation.

As with “law,” the meaning of “compensation” becomes clear when we examine the Twenty-Seventh Amendment in light of the Ascertainment Clause. Because the Twenty-Seventh Amendment imposes a limitation on laws passed under the Ascertainment Clause, “compensation” simply means: Anything included in a law enacted pursuant to that clause. At first this may seem circular, but it is not. At no point in history could the Twenty-Seventh Amendment have pre-existed a law enacted pursuant to the Ascertainment Clause. The first such law was enacted in 1789 while the Twenty-Seventh Amendment was still being debated as the second amendment in the proposed Bill of Rights.[80] Even if the Twenty-Seventh Amendment had been ratified in 1791 with the Bill of Rights, there would already have been a compensation law on the books. Of course, because it was ratified 202 years after its proposal, there were scores of such laws on the books. The words “varying compensation,” therefore, assume the pre-existence of a law enacted pursuant to the Ascertainment Clause and must be defined in reference to it. In sum, a law varies compensation when it varies the terms of a previous compensation law.

What does it mean to vary? The history of the Twenty-Seventh Amendment makes clear that that word encompasses both increases and decreases to compensation.[81] The possibility that members of Congress would increase their salaries to enrich themselves was obvious to the Framers. But they were also aware of the practice in Britain, and in the States under the Articles of Confederation, of lowering representatives’ wages for political gain and of enacting property requirements that excluded challengers who lacked independent means.  The Framers aimed to avoid those problems.[82] On the question of increases or decreases, therefore, the answer is both.

The more difficult question, however, is: When does a law vary compensation—when it is enacted or when the change goes into effect? This issue was raised in Boehner v. Anderson.[83] The Ethics Reform Act provided automatic annual COLAs, one of which was due to be paid on January 1, 1994.[84] On March 4, 1993, however, Congress passed another law that cancelled the upcoming COLA without an intervening election.[85] The plaintiff argued that the law varied members’ compensation as of January 1.[86] The defendants argued that no variance in compensation occurred because the second law merely “extend[ed] the period during which their compensation remain[ed] unchained.”[87] The court did not reach the issue because it was improperly raised for the first time on appeal.[88]

For his part, Professor Vermeule argues, as a normative matter, that

The Twenty-seventh Amendment vesting rule should be that legislators’ “compensation” vests when statutory changes are enacted, not when they become effective. This does not mean that, once the Congress has enacted a scheduled adjustment, the adjustment may never be blocked, but merely that it may not be blocked after the election of the very Congress in which it will become effective. A statute enacted in 2001 that schedules a COLA for January 4, 2005 may be blocked by a supervening statute at any time before the congressional elections in November 2004.[89]

This is so, he argues, because the Amendment is designed to prevent sitting members of Congress from knowingly adjusting their own salaries.[90] His argument is normatively compelling and aligns with the Amendment’s originally intended purpose.[91] Consider the example he gives:

In Year 1, Congress enacts a future decrease, scheduled to go into effect on January 4 of Year 3, the beginning of the following Congress; there is an intervening election in Year 2; and on the afternoon of January 3, the new Congress enacts, as its first measure, a law blocking the next day’s decrease. On a “veil of ignorance” approach, this should be held a straightforward violation of the Amendment. Legislators voting on the January 3 statute presumptively suffer from the very decisionmaking distortion that provoked the Amendment; the new Congress has, in effect, knowingly adjusted its own salary.[92]

But does this argument align with the text? The text separates “varying” from “taking effect” suggesting that a law may vary compensation without taking effect. We find further support for this reading if we, again, view the Twenty-Seventh Amendment in light of the Ascertainment Clause.[93] The Twenty-Seventh Amendment was intended to modify the Ascertainment Clause by imposing a timing condition on laws enacted under it.[94] If compensation means the provisions of any such law, then “varying” means any increase or decrease to the provisions of that law. Thus, the most natural reading of the Amendment is that a law varies compensation whenever a new law changes a preexisting compensation law—that is, upon enactment.

That conclusion is bolstered by comparing the Twenty-Seventh Amendment to the Presidential Compensation Clause[95] and the Judicial Compensation Clause.[96] The first fixes the president’s compensation during his four-year term, and the second provides that judges’ salaries cannot be decreased during their service. Both explicitly prohibit certain changes in compensation during a period of time. The Twenty-Seventh Amendment does not. It allows changes to occur but stops them from taking effect until an election has intervened.

Reading the Amendment this way reveals that both parties in Boehner were wrong.  The law blocking the COLA did vary compensation because it changed the terms of a prior compensation law, but it complied with the Amendment’s timing condition because the variance occurred the moment the law was enacted. This reading also provides a second way to resolve the debate over House rules that deduct fines from Representatives’ salaries if they fail to use metal detectors.[97] Not only are the rules not “laws,” they also do not vary compensation because they do not change the terms of previous compensation laws. They are fines for misconduct, as are parking tickets. Neither one changes a member’s salary.

III.  The Meaning of “shall take effect” and “shall have intervened”

The previous conclusion—that a law varies compensation when it is enacted—largely resolves the question of what “shall take effect” means. The Amendment describes two discrete actions, “varying” and “taking effect.” If a law varies compensation when it is enacted regardless of whether members have seen a change to their paychecks, then “taking effect” must refer to that moment when members are entitled to receive a quantifiable change to their compensation. The court in Boehner adopted this reading, equating “taking effect” with the “operation” of a law varying compensation.[98] The court said that this operation occurs “at the earliest [when] the new Congress has been seated.”[99]

That assumes, however, that the words “an election . . . shall have intervened” mean not only that an election has occurred, but also that the new members have taken their seats. An alternative reading of the phrase “shall have intervened” might mean simply that an election has occurred. It turns out that the Boehner court’s assumption aligns with the best reading of the Amendment for three reasons. First, before the age of trains, automobiles, and planes, Congress adjourned several months before each election and did not return until several months after them in order to accommodate the long travel times to the furthest reaches of the Republic.[100] Second, James Madison—author of the Amendment—understood “an election . . . shall have intervened” to mean “a change in the Legislature.”[101] He used the latter phrase in a speech describing the Amendment even while presenting the text as currently written, suggesting that he viewed those two phrases interchangeably.[102] Finally, the alternative—that an election intervenes the moment it concludes but before new members take their seats—renders nonsensical his and others’ descriptions of the Amendment’s purpose and effect.

James Madison’s apparently casual substitution of “change in the Legislature” for “an election . . . shall have intervened” shows that he, at least, viewed the Amendment as delaying the effect of laws varying compensation until new members are seated. The legislature, after all, has not changed until newly elected members take their seats. Unfortunately, Madison’s statement is one of only a very few that illuminates the Framer’s understanding of the Amendment. When, on August 14, 1789, the Committee of the Whole House turned its attention to it, the debate was exceedingly brief.[103] Only four members discussed it, one of whom was Madison who responded to a comment from another member that was not related to the Amendment.[104] Of the two who commented on the Amendment, one had just three sentences to say about it; the other had but two. Theodore Sedgwick of Massachusetts, seems not to have fully understood the Amendment because his argument—that it might be used by “designing men” to gain popularity and exclude poor but meritorious challengers by lowering their own salaries—is a critique better suited to the Ascertainment Clause itself than to the Amendment.[105] Jacob Vining of Delaware supported the Amendment because it would avoid “the disagreeable sensation, occasioned by leaving it in the breast of any man to set a value upon his own work . . . .”[106]

None of that brief debate speaks directly to the meaning of “intervene,” but it does reinforce our understanding that the Framers intended the Amendment to prevent sitting members of Congress from enriching themselves or gaining a political advantage by controlling their own salaries.[107] Even Representative Sedgwick’s poorly aimed criticism supports this view because the Amendment attempts to solve the problems he identified, which flow from the Ascertainment Clause. If, as seems clear, the Framers understood the Amendment as preventing sitting members from exploiting changes to their compensation, then “an election . . . shall have intervened” must mean that a new Congress has been seated. If it does not, then for that brief period of time between an election and a “change” in the legislature,[108] sitting members may change their own compensation and benefit from it. They might even lower compensation so that successful challengers who lack independent means might have to relinquish their newly won seats.


The Twenty-Seventh Amendment sparks more debates than its simple text suggests it should, and few courts or scholars have spent much time parsing its text to help resolve those debates. This Essay aims to do that by interpreting the Amendment according to its text and history. That interpretation leads to the following conclusions: “law” has its technical Article I definition (a bill passed by both houses and signed by the president), “varying compensation” means any increase or decrease to the terms of pre-existing laws enacted pursuant to the Ascertainment Clause, “shall take effect” refers to the moment when members are entitled to receive a quantifiable change to their compensation, and “an election . . . shall have intervened” means that a new Congress has been seated. The best reading of the Amendment views it as a narrow provision that sets a timing condition on the operation of laws varying members’ compensation; it does not apply to other government conduct such as House rules that deduct fines from members’ salaries for misconduct. It does, however, inform our understanding of the Ascertainment Clause as prohibiting Congress from delegating its power to set compensation to any other body. The Amendment does not prohibit the passage of any compensation laws. Rather, it simply delays laws varying compensation from taking effect until an intervening House election has occurred. Accordingly, the Amendment cannot be used to strike down laws that do not comply with its timing requirement except in rare circumstances. Instead, the law is best read as providing a default timing term to laws varying compensation that either lack such a term or include a term that does not comply with the Amendment. One or more of those conclusions resolves the existing debates about the Twenty-Seventh Amendment and, hopefully, will do the same for any debates that arise in the future.



* Legal Fellow, The Heritage Foundation; J.D., Georgetown Law Center, 2014; B.A., University of California at Davis, 2011.

** John, Barbara & Victoria Rumpel Senior Legal Research Fellow, The Heritage Foundation; M.P.P. George Washington University, 2010; J.D. Stanford Law School, 1980; B.A. Washington & Lee University, 1977.

We are grateful to John G. Malcolm, Adeola Olowude, Zack Smith, and Amy Swearer for helpful comments on an earlier draft of this Essay. We are also indebted to Nicole Imhof and Lauren McCarthy for their research assistance. The views expressed in this Essay are our own and should not be construed as representing any official position of The Heritage Foundation. Any mistakes are ours (although we are willing to accept volunteers).

[1] See, e.g., Paul J. Larkin, Jr., The Lost Due Process Doctrines, 66 Cath. U. L. Rev. 293, 329 (2017) (“English King Ethelbert drafted the first written code in approximately 600 A.D. Consisting of only ‘ninety brief sentences,’ Ethelbert’s code—composed of dooms (‘decrees’) not leges (‘laws’), because the concept of ‘law’ was as yet unknown in England—was essentially a tariff, a schedule of fines, payable in money known as the wergild, that a wrongdoer was obliged to give to the victim of a crime or his kin, principally for murder, mayhem, other acts of violence, or cattle-thievery. The hoped-for goal was to forestall violent retaliation and intertribal warfare.”) (footnotes omitted); id. at 327–29.

[2] See, e.g., Arthur R. Hogue, Origins of the Common Law (Liberty Fund 1986) (1966); John H. Langbein et al., History of the Common Law (2d ed. 2009); Theodore F. T. Plucknett, A Concise History of the Common Law (Little, Brown & Co. 5th ed. 1956) (1929).

[3] See, e.g., Oliver Wendell Holmes, The Path of the Law, 10 Harv. L. Rev. 457, 460 (1897) (“The prophecies of what the courts will do in fact, and nothing more pretentious, are what I mean by the law.”). See generally Benjamin N. Cardozo, The Nature of the Judicial Process (Dover Pubs. 2012) (1921); Oliver Wendell Holmes, The Common Law (Dover Pubs. Rev. ed. 1991) (1881).

[4] The day when Congress was the primary federal law-making body, however, is now in the rear-view mirror; the administrative state annually promulgates far more rules than Congress passes statutes. See, e.g., Rachel Augustine Potter, Bending the Rules: Procedural Politicking in the Bureaucracy 14 (2019) (“By some estimates, more than 90 percent of American law is created by administrative rules issued by federal agencies.”); Cornelius M. Kerwin & Scott R. Furlong, Rulemaking: How Government Agencies Write Law and Make Policy 2 (4th ed. 2011) (“Increasingly, rulemaking defines the substance of public programs. It determines, to a very large extent, the specific legal obligations we bear as a society. Rulemaking gives precise form to the benefits we enjoy under a wide range of statutes. In the process, it fixes the actual costs we incur in meeting the ambitious objectives of our many public programs.”); infra note 5.

[5] See, e.g., Kisor v. Wilkie, 139 S. Ct. 2400, 2446–47 (2019) (Gorsuch, J., concurring in the judgment) (“Now, in the 21st century, the administrative state wields vast power and touches almost every aspect of daily life. Among other things, it produces reams of regulations—so many that they dwarf the statutes enacted by Congress. As of 2018, the Code of Federal Regulations filled 242 volumes and was about 185,000 pages long, almost quadruple the length of the most recent edition of the U. S. Code. And agencies add thousands more pages of regulations every year.”) (footnotes and internal punctuation omitted); Kevin R. Kosar, Reasserting Congress in Regulatory Policy, in 2 Unleashing Opportunity: Policy Reforms for an Accountable Administrative State 19, 19 (Yuval Levin & Emily MacLean eds., 2017) (“In recent years, Congress has enacted approximately 50 statutes annually on significant subject matter; the executive branch proposes 2,700 new regulations and finalizes another 4,000 rules each year.”) (footnote omitted); Paul J. Larkin, Jr., & GianCarlo Canaparo, Gunfight at the New Deal Corral, 19 Geo. J.L. & Pub. Pol’y (forthcoming 2021) (manuscript at 20 & n.78) (on file with authors).

[6] See, e.g., Perez v. Mortg. Bankers Ass’n, 575 U.S. 92, 96 (2015) (agency rules properly issued pursuant to delegated authority “have the ‘force and effect of law”’) (quoting Chrysler Corp. v. Brown, 441 U.S. 281, 302–03 (1979)); see generally Paul J. Larkin, Jr., Agency Deference after Kisor v. Wilkie, 18 Geo. J.L. & Pub. Pol’y 105 (2020) (discussing whether an agency’s interpretation of its own rules is entitled to deference and, if so, what type and how much).

[7] See, e.g., City of Boerne v. Flores, 521 U.S. 507, 524, 536 (1997) (“The power to interpret the Constitution in a case or controversy remains in the Judiciary.”); Miranda v. Arizona, 384 U.S. 436, 442 (1966) (“These precious [Sixth Amendment] rights were fixed in our Constitution only after centuries of persecution and struggle. And in the words of Chief Justice Marshall, they were secured ‘for ages to come, and . . . designed to approach immortality as nearly as human institutions can approach it.’”) (quoting Cohens v. Virginia, 19 U.S. (6 Wheat.) 264, 387 (1821)).

[8] See, e.g., Little Sisters of the Poor Saints Peter & Paul Home v. Pennsylvania, 140 S. Ct. 2367, 2380–81 (2020) (“Our analysis begins and ends with the text. . . . It is a fundamental principle of statutory interpretation that absent provisions cannot be supplied by the courts . . . . By introducing a limitation not found in the statute, respondents ask us to alter, rather than to interpret, the [act].”) (citations and internal punctuation omitted); Bostock v. Clayton Cty., 140 S. Ct. 1731, 1738 (2020) (“This Court normally interprets a statute in accord with the ordinary public meaning of its terms at the time of its enactment. After all, only the words on the page constitute the law adopted by Congress and approved by the President.”).

[9] See, e.g., Chevron U.S.A. Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837 (1984) (describing the legal effect to be given to an agency’s interpretation of a statute it is responsible for implementing).

[10] See, e.g., Kisor v. Wilkie, 139 S. Ct. 2400 (2019) (describing the legal effect to be given to an agency’s interpretation of one of its own regulations).

[11] U.S. Const. amend. XXVII.

[12] Which is why there is no merit to the argument, often made by free speech or free exercise advocates, that the First Congress put those guarantees ahead of the other Bill of Rights provisions because the Founding Generation considered those rights more important than the rest. What became the First Amendment was actually third in the list.

[13] U.S. Const. art. V (describing the process for amending the Constitution).

[14]  For a very thorough history of the Twenty-Seventh Amendment, from proposal through ratification and early legal challenges, see Richard Bernstein, The Sleeper Wakes: The History and Legacy of the Twenty-Seventh Amendment, 61 Fordham L. Rev. 497 (1992). See also Edward Dumbauld, The Bill of Rights and What It Means 161, 188, 195, 204–5 (Praeger 1979) (discussing how what would become the Twenty-Seventh Amendment came to be included in the original Bill of Rights and providing a detailed overview of the debates about the Amendment’s late ratification and what that meant for Article V). See generally, for further discussion: William W. Van Alstyne, What Do You Think About the Twenty-Seventh Amendment?, 10 Const. Commentary 9 (1993); Stewart Dalzell & Eric J. Beste, Is the Twenty-Seventh Amendment 200 Years Too Late?, 62 Geo. Wash. L. Rev. 501 (1994); Don J. DeBenedictis, 27th Amendment Ratified: Congressional Vote Ends Debate Over 203-Year-Old Pay-Raise Proposal, 78 A.B.A. J. 26 (1992); Christopher M. Kennedy, Is There a Twenty-Seventh Amendment? The Unconstitutionality of a “New” 203-Year-Old Amendment, 26 J. Marshall L. Rev. 977 (1993); Michael Stokes Paulsen, A General Theory of Article V: The Constitutional Lessons of the Twenty-seventh Amendment, 103 Yale L.J. 677 (1993); JoAnne D. Spotts, The Twenty-Seventh Amendment: A Late Bloomer or A Dead Horse?, 10 Ga. St. U. L. Rev. 337 (1994); Richard L. Berke, 1789 Amendment Is Ratified but Now the Debate Begins, N.Y. Times, May 8, 1992, at A1; Don Phillips, Proposed Amendment, Age 200, Showing Life, Wash. Post, Mar. 29, 1989, at A23; Kimberly Wehle, Can Members of Congress Carry Firearms On The Capitol Complex?, The Hill (Feb. 8, 2021, 9:30 AM), []; see also Yaniv Roznai, Unconstitutional Constitutional Amendments—the Migration and Success of A Constitutional Idea, 61 Am. J. Compar. L. 657, 657 (2013) (raising the (in our view, nonsensical) question, without mentioning the Twenty-Seventh Amendment, of whether a constitutional amendment can be unconstitutional).

[15] 256 U.S. 368 (1921).

[16] Id. at 375.

[17] Id.

[18] Id.

[19] See, e.g., Walter Dellinger, Constitutional Politics: A Rejoinder, 97 Harv. L. Rev. 446 (1983); Kennedy, supra note 14; Spotts, supra note 14.

[20] Bernstein, supra note 14, at 552.

[21] See, e.g., Boehner v. Anderson, 30 F.3d 156 (D.C. Cir. 1994) (rejecting the plaintiff’s argument that automatic cost of living adjustments were “laws”); Schaffer v. Clinton, 240 F.3d 878 (10th Cir. 2001) (avoiding the same claim on procedural grounds).

[22] See, e.g., Boehner, 30 F.3d at 161–62 (holding that automatic cost of living adjustments established by statute were not separate “laws” within the meaning of the Amendment); Shaffer v. Clinton, 54 F. Supp. 2d 1014 (D. Colo. 1999), aff’d on other grounds sub nom. Schaffer v. Clinton, 240 F.3d 878 (10th Cir. 2001) (same as Boehner, although the district court’s holding was vacated when the Tenth Circuit reversed on the ground that no plaintiffs had standing).

[23] See, e.g., Jonathan D. McPike, Merit Pay and Pain: Linking Congressional Pay to Performance, 86 Ind. L.J. 335, 365–66 (2011) (considering whether paying members according to “incentive contracts” would violate the Amendment); Adrian Vermeule, The Constitutional Law of Official Compensation, 102 Colum. L. Rev. 501, 516–21 (2002) (examining the issues raised in Boehner v. Anderson, 30 F.3d 156 (D.C. Cir. 1994), as a “veil of ignorance” mechanism).

[24] See GianCarlo Canaparo, Representatives Go On Offensive Against Pelosi’s Mask Rule, Cite 27th Amendment. Will It Work?, Daily Signal (July 29, 2021),; Andrew Solender, Pelosi to Fine House Members Up To $10,000 for Evading Metal Detectors, Forbes (Jan. 13, 2021, 9:09 PM), [] (describing an argument by Representative Thomas Massie that fines imposed for breaking House rules violated the Twenty-Seventh Amendment); Carey Vanderborg, 27th Amendment In Constitution Violated By GOP Budget Bill? House To Vote Wednesday, International Business Times (Jan. 23, 2013, 11:27 AM), []; Letter from Representative Andrew S. Clyde to Representative Theodore E. Deutch, Chairman, House Committee on Ethics (Feb. 23, 2021), [] (arguing that fines for violating House rules requiring members to pass through metal detectors violate the Amendment).

[25] U.S. Const. art. I, § 7, cl. 2 (“Every Bill which shall have passed the House of Representatives and the Senate, shall, before it become a Law, be presented to the President of the United States . . . .”).

[26] U.S. Const. art. I, § 6, cl. 1 (“The Senators and Representatives shall receive a Compensation for their Services, to be ascertained by Law, and paid out of the Treasury of the United States.”).

[27] Id. amend XXVII.

[28] See id. art. I, § 7; INS v. Chadha, 462 U.S. 919 (1983). To ensure that Congress cannot evade that intentionally onerous procedure through legislative shenanigans (such as by labeling a “Bill” as something else) and create a “Law” by labeling a proposal as something other than a “Bill,” Article I expressly applies to any “Bill” and “[e]very Order, Resolution, or Vote” requiring the approval of both chambers other than an “Adjournment.” U.S. Const. art. I, § 7, cl. 3. That is why Congress cannot escape the Twenty-Seventh Amendment by voting itself a pay raise and calling it a “bonus.”

[29] U.S. Const. amend I.

[30] Id. amend XIV. With respect to the expansive definition of “Congress shall make no law,” see, for example, Gitlow v. New York, 268 U.S. 652, 666 (1925) (establishing the foundation of the Incorporation Doctrine, whereby the First Amendment—and later other amendments—would be held to apply against the states via the Fourteenth Amendment’s Due process Clause); Sherbert v. Verner, 374 U.S. 398, 409–10 (1963) (holding that the First Amendment applied to a state administrative agency’s decision denying unemployment benefits to an employee fired for refusing to work on the Sabbath); New York Times Co. v. Sullivan, 376 U.S. 254, 292 (1964) (holding that the First Amendment prevents a state court from entering a civil verdict in favor of a public official suing for defamation unless the official can prove actual malice); Hustler Magazine v. Falwell, 485 U.S. 46, 57 (1988) (extending the holding of New York Times to intentional infliction of emotional distress); Church of the Lukumi Babalu Aye v. City of Hialeah, 508 U.S. 520, 547 (1993) (applying the First Amendment to a city ordinance); McCreary County v. ACLU of Kentucky, 545 U.S. 844, 881 (2005) (holding that the First Amendment prohibited county officials from displaying the Ten Commandments on public land). With respect to the Fourteenth Amendment, the Court has read expansively the words “No State shall make or enforce any law” but, at the same time, has nearly nullified the clause in which those words appear by reading narrowly the words “privileges or immunities.” See, e.g., Bradwell v. Illinois, 83 U.S. (16 Wall.) 130, 133 (1873) (applying the Privileges or Immunities Clause to a state supreme court’s decision to deny a woman admission to the state bar, but holding that the right to practice a profession is not a “privilege or immunity”); United States v. Cruikshank, 92 U.S. 542, 552–53 (1876) (applying the Clause to a criminal judgment entered after a guilty verdict but also holding that the Privilege or Immunities Clause did not bar a state from violating the rights listed in the First and Second Amendments); Slaughterhouse Cases, 83 U.S. (16 Wall.) 36, 80–83 (1873) (narrowly construing the Fourteenth Amendment Privileges or Immunities Clause).

[31] U.S. Const. art. I, § 8, cl. 10.

[32] See generally Henry P. Monaghan, Constitutional Common Law, 89 Harv. L. Rev. 1 (1975).

[33] See, e.g., Nixon v. United States, 506 U.S. 224, 233–38 (1993) (interpreting the word “sole” in the Senate Impeachment Clause as precluding judicial review of the Senate’s decision to remove a federal official from office); United States v. Jacobsen, 466 U.S. 109, 125 (1984) (limiting the scope of the Fourth Amendment to “searches” and “seizures”); United States v. Marion, 404 U.S. 307, 325–26 (1971) (construing the term “accused” in the Sixth Amendment to require that a formal charge have been filed against someone); see generally Antonin Scalia, A Matter of Interpretation: Federal Courts and the Law (Princeton University Press, 1998).

[34] See Vermeule, supra note 23, at 511 (arguing that the Amendment represents a “veil of ignorance” rule that increases the chance that Congress will enact a normatively good law because it lacks the information necessary to make a bad one); Adrian Vermeule, Veil of Ignorance Rules in Constitutional Law, 111 Yale L.J. 399, 400 (2001); see also Adrian Vermeule, Our Schmittian Administrative Law, 122 Harv. L. Rev. 1095, 1149 (2009) (arguing, in similar fashion, that American administrative law can be explained by the legal theories of Carl Schmitt).

[35] For example, consider the debate between H.L.A. Hart, on the one hand, and Lon Fuller and Ronald Dworkin, on the other, regarding the prerequisites for an edict to be deemed a “law.” Hart, a legal positivist, would classify as a law any directive issued by the governing authority in a polity. Considerations of morality did not enter into the question. See H.L.A. Hart, Positivism and the Separation of Law and Morals, 71 Harv. L. Rev. 593, 593 (1958). By contrast, Fuller and Dworkin demanded that a “law” satisfy a series of requirements in order to be entitled to the moral force that is implied by that term. See Lon L. Fuller, The Morality of Law (3d ed. 1969); Lon L. Fuller, Positivism and Fidelity to Law—A Reply to Professor Hart, 71 Harv. L. Rev. 630, 630 (1958); Ronald Dworkin, Law’s Empire 15–20 (1988). The debate is a fascinating one, but it is beyond the scope of this Essay.

[36] See Bernstein, supra note 14, at 522 (referring to the Amendment as the compensation amendment pre-ratification because it was included in the original Bill of Rights as the second amendment).

[37] Of course, other amendments in the Bill of Rights have the effect of limiting lawmaking, but only these two explicitly mention it. Cf. U.S. Const. amends. II (“shall not be infringed”), III (“[n]o Soldier shall . . . be quartered”), IV (“shall not be violated”), V (“no person shall be”), VI (“the accused shall enjoy the right”), VII (“the right of trial by jury shall be preserved”), VIII (“Excessive bail shall not be required, nor excessive fines imposed, nor cruel and unusual punishments inflicted.”), IX (“The enumeration in the Constitution, of certain rights, shall not be construed to deny or disparage others retained by the people.”), X (“[t]he powers not delegated . . . are reserved”).

[38] See Boehner v. Anderson, 30 F.3d 156, 163 (D.C. Cir. 1994) (rejecting the plaintiff’s argument that automatic cost of living adjustments were “laws”); Schaffer v. Clinton, 240 F.3d 878, 886 (10th Cir. 2001) (avoiding the same claim on procedural grounds); Solender, supra note 24 (describing an argument by Representative Thomas Massie that fines imposed for breaking House rules violated the Twenty-Seventh Amendment).

[39] Genesis 1:2–3 (“And the earth was without form, and void; and darkness was upon the face of the deep. And the Spirit of God moved upon the face of the waters. And God said, Let there be light: and there was light.”).

[40] U.S. Const. art. I, § 6, cl. 1; see also Bernstein, supra note 14, at 502. The House committee tasked with framing the proposed amendment did it this way:

ART. I, Sec. 6 — Between the words “United States” and “shall in all cases” strike out “they,” and insert, “But no law varying the compensation shall take effect until an election of Representatives shall have intervened. The members.”

H.R. Select Comm. Rep. (July 28, 1789), reprinted in Creating the Bill of Rights: The Documentary Record from the First Federal Congress 30 (Helen E. Veit et al., eds., 1991).

[41] That understanding of the Amendment is reflected clearly in the three proposals—from Virginia, New York, and North Carolina—that prompted Madison to include it in the original Bill of Rights. See Bernstein, supra note 14, at 514. New York’s proposal most clearly expresses this relationship:

That the compensation for the Senators and Representatives be ascertained by standing Laws; and that no alteration of the existing rate of Compensation shall operate for the Benefit of the Representatives, until after a subsequent election shall have been had.

Id. (reprinting the proposal).

[42] See id. at 502–08 (recounting the history of that provision and its bearing on what would later become the Twenty-Seventh Amendment).

[43] Id.

[44] Id.

[45] See id.

[46] See U.S. Const. art. I, § 9, cl. 7 (“No Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law[.]”); Paul J. Larkin, Jr. & Zack Smith, “Brother, Can You Spare a Million Dollars?”: Resurrecting the Justice Department’s “Slush Fund,19 Geo. J.L. & Pub. Pol’y (forthcoming 2021) (manuscript 9–12) (on file with authors).

[47] Todd David Pearson, Protecting the Appropriations Power: Why Congress Should Care About Settlements at the Department of Justice, 2009 BYU L. Rev. 327, 329–30 (2009).

[48] The Federalist No. 58, at 356–57 (James Madison) (Clinton Rossiter ed., 1961).

[49] See sources cited supra note 19.

[50] U.S. Const. art. I, § 7, cl. 1 (“All Bills for raising Revenue shall originate in the House of Representatives; but the Senate may propose or concur with Amendments as on other Bills.”).

[51] Id. cl. 2.

[52] Akhil Reed Amar, The Bill of Rights As a Constitution, 100 Yale L.J. 1131, 1145 (1991); see also Bernstein, supra note 14, at 512–16, 530 (explaining that the inclusion of two “structural amendments,” to the Bill of Rights was a political compromise that James Madison thought necessary to bridge the divide between Federalists and Anti-Federalists and describing those two amendments as “ha[ving] nothing to do with rights.”).

[53] It is not the only way to resolve these debates as we discuss in Part II.

[54] 30 F.3d 156 (D.C. Cir. 1994).

[55] Pub. L. No. 101–194, 103 Stat. 1716 (1989) (codified at 2 U.S.C. § 31(2) and 5 U.S.C. § 5318 note)

[56] Boehner, 30 F.3d at 161. As Professor Vermeule notes, that argument amounted to “an attempt to resurrect, in Twenty-Seventh Amendment guise, the implausible version of the Ascertainment Clause that would require each change in congressional pay to be enacted by separate statutes.”  Vermeule, supra note 23, at 519.

[57] Boehner, 30 F.3d at 161 (citing U.S. Const. art. I, § 7).

[58] Of course, the Twenty-Seventh Amendment had not been ratified yet, but the court did not have to decide if it applied retroactively because the Ethics Reform Act satisfied its requirements. The question of retroactivity is now safely behind us, and we need not spend time on it.

[59] Shaffer v. Clinton, 54 F. Supp. 2d 1014, 1023–24 (D. Colo. 1999) (giving “law” its Art. I, § 7, definition and saying of the COLAs that “a formula to create annual salary adjustments does not create a new law with each adjustment since the adjustments are not contingent upon passage by Congress and the President’s signature.”). This holding was vacated on appeal when the Tenth Circuit concluded, contra the district court, that no plaintiff had standing. See Schaffer v. Clinton, 240 F.3d 878, 886 (10th Cir. 2001).

[60] H.R. Res. 73, 117th Cong. (2021).

[61] See, e.g., Letter from Representative Andrew S. Clyde to Representative Theodore E. Deutch, Chairman of the House Committee on Ethics (Feb. 23, 2021) (available at []) (arguing that the fines violate the Twenty-Seventh Amendment); Solender, supra note 24 (describing the same argument made by Representative Massie).

[62] It also fails, as we explain below, because the fines imposed do not vary Representatives’ compensation.

[63] U.S. Const. art. I, § 5, cl. 2.

[64] See United States v. Hatter, 532 U.S. 557 (2001) (holding that Social Security and Medicare taxes do not violate the Article III guarantee that the salary of federal judges “shall not be diminished during their Continuance in Office”).

[65] U.S. Const. amend. XXVII.

[66] Id. art. I, § 9, cl. 3.

[67] Id. art. I, § 9, cl. 8 (“No Title of Nobility shall be granted by the United States[.]”).

[68] See, e.g., Blodgett v. Holden, 275 U.S. 142, 148 (1927) (Holmes, J., concurring) (“[T]he rule is settled that as between two possible interpretations of a statute, by one of which it would be unconstitutional and by the other valid, our plain duty is to adopt that which will save the Act.”); Edward J. DeBartolo Corp. v. Fla. Gulf Coast Bldg & Constr. Trades Council, 485 U.S. 568, 575 (1988) (citing NLRB v. Catholic Bishop of Chicago, 440 U.S. 490, 499–501, 504 (1979)) (“[W]here an otherwise acceptable construction of a statute would raise serious constitutional problems, the Court will construe the statute to avoid such problems unless such construction is plainly contrary to the intent of Congress.”). For a thorough discussion of these rules of avoidance see Adrian Vermeule, Saving Constructions, 85 Geo. L.J. 1945, 1949 (1997).

[69] One might object that this sounds like the sort of activism—judges rewriting a law according to some set of subjective values—that textually-oriented scholars usually find objectionable. But that criticism finds no purchase here for the simple reason that this substitution is neither subjective nor imposed on a judge’s whim. It is, instead, a specific term mandated by the Constitution. Reading the Amendment as providing a default term, therefore, amounts to nothing more unusual than a reiteration of the principle that the Constitution is supreme over statutes.

[70] It would, however, support a claim for injunctive relief to prevent an untimely change in compensation from taking effect until an election has intervened—assuming (which is dubious) that a plaintiff with Article III standing could bring such a lawsuit.  Similar types of lawsuits have not gained much traction. See, e.g., Ariz. Christian Sch. Tuition Org. v. Winn, 563 U.S. 125 (2011); Hein v. Freedom from Religion Found., Inc., 551 U.S. 587 (2007); Frothingham v. Mellon, 262 U.S. 447 (1923) (all ruling that a taxpayer cannot challenge the constitutionality of a law on the ground that it would make an impermissible use of his taxes).

[71] See DeBartolo, 485 U.S. at 575 (“[T]he Court will construe the statute to avoid such problems unless such construction is plainly contrary to the intent of Congress.”) (emphasis added).

[72] Carey Vanderborg, 27th Amendment in Constitution Violated By GOP Budget Bill? House To Vote Wednesday, Int’l Bus. Times (Jan. 23, 2013, 11:27 AM), [].

[73] See generally Bernstein, supra note 14. James Madison said that the Amendment would prevent changes in compensation from inuring to “the particular benefit of those who are concerned in determining the value of the service.” 1 Annals of Cong. 458 (Joseph Gales ed., 1834) (1789).

[74] See David Schoenbrod, Power Without Responsibility: How Congress Abuses The People Through Delegation 167–70 (1993) (arguing for courts to police delegations in part because they “interfere[] with democratic accountability”); Neomi Rao, Administrative Collusion: How Delegation Diminishes the Collective Congress, 90 N.Y.U. L. Rev. 1463, 1512 (2015) (discussing how “delegations erode the accountability of members of Congress” and “the lawmaking procedures of Article I, Section 7.”).

[75] See Pressler v. Simon, 428 F. Supp. 302, 303–04 (D.D.C. 1976), vacated sub nom. Pressler v. Blumenthal, 431 U.S. 169, (1977) (recounting the history of Congressional compensation laws and calling the Postal Revenue and Salary Act and the Adjustment Act “a major break with tradition.”).

[76]  Postal Revenue and Federal Salary Act of 1967, Pub. L. No. 90-206 § 225, 81 Stat. 613, 642-45 (1967) (codified as amended at 2 U.S.C. § 351).

[77] Id.

[78] Pressler v. Simon, 428 F. Supp. at 306. The Supreme Court later summarily affirmed this decision without opinion, but in concurrence, then-Justice Rehnquist noted that the affirmance “does not necessarily reflect this Court’s agreement with the conclusion reached by the District Court on the merits of the Ascertainment Clause question” because the district court also decided that the plaintiff lacked standing. Pressler v. Blumenthal, 434 U.S. 1028, 1028–29 (1978) (Rehnquist, J., concurring).

[79] Recently, the Supreme Court rejected the same argument that the “complexities of modern society” justify departing from the text, history, and tradition of the Constitution in the context of the Fifth Amendment. See Cedar Point Nursery v. Hassid, 141 S. Ct. 2063, 2077–78 (2021).

[80] See Bernstein, supra note 14, at 533 (citing 6 Documentary History of the First Federal Congress 1789-1791, at 1833–35 (Charlene Bangs Bickford & Helen E. Veit eds., 1986)).

[81] See id. at 500–05, 526–27; see also Vermeule, supra note 23, at 520 (“Self-interest, however, may push in the direction of decreasing salaries as well as in the opposite direction, for in the former case the political benefits of conspicuous self-denial may dominate purely financial losses.”).

[82] Bernstein, supra note 14, at 500–05, 526–27.

[83] 30 F.3d at 162–63.

[84] Id. at 162.

[85] Id.

[86] Id.

[87] Id.

[88] Id. at 163.

[89] Vermeule, supra note 23, at 521 (emphasis in original).

[90] Id.

[91] See supra note 73.

[92] Vermeule, supra note 23, at 520–21.

[93] See supra text accompanying notes 39–41, 49–51.

[94] See sources cited supra note 40.

[95] U.S. Const. art. II, § 1, cl. 7 (“The President shall, at stated Times, receive for his Services, a Compensation, which shall neither be increased nor diminished during the Period for which he shall have been elected . . . .”).

[96] Id. art. III, § 1 (“The Judges, both of the supreme and inferior Courts . . . shall, at stated Times, receive for their Services, a Compensation, which shall not be diminished during their Continuance in Office.”).

[97] See supra text accompanying notes 28–31.

[98] Boehner, 30 F.3d at 162.

[99] Id.

[100] See United States House of Representatives, Session Dates of Congress, [] (showing the adjournment dates of all Congresses).

[101] 1 Annals of Cong., supra note 73, at 458 (Madison: “I have gone, therefore, so far as to fix it, that no law, varying the compensation, shall operate until there is a change in the Legislature; in which case it cannot be for the particular benefit of those who are concerned in determining the value of the service.”).

[102] See id.

[103] Id. at 526–27 (quoting Debates in the House of Representatives (Aug. 14, 1789), in The Congressional Register, Aug. 14, 1789, reprinted in Creating the Bill of Rights: The Documentary Record from the First Federal Congress, supra note 40, at 149–50) (reprinting the entire “listless and desultory discussion”).

[104] See id. at 527.

[105] Id. at 526–27. The problems he identifies are created by the Ascertainment Clause, not the Amendment. The Amendment attempts to solve them by ensuring that no Representative who votes for a change in compensation can know for certain whether he or she will remain in Congress when it goes into effect.

[106] Id. at 527.

[107] Id. at 500–05, 526–27.

[108] Id. at 458.

Read More »

Per Curiam Preface – Sec. E. Spencer Abraham

Posted by on Sep 1, 2021 in Per Curiam

Download as PDF


When the Harvard Journal of Law & Public Policy published its first edition, the opportunities to publish conservative legal scholarship at American law schools were essentially non-existent. In 1977, several classmates and I approached the Harvard Law School administration about starting a journal to publish and foster debate about conservative ideas, ideas conspicuously absent from the legal conversation we encountered in the classroom. The administration refused. They claimed that Harvard’s stable of student-run journals did not endorse political viewpoints and were facially neutral. Although technically true, the dearth of conservative articles in those journals for many years underscored the absence of opportunities for conservatives to share their perspectives in the marketplace of legal argumentation. And Harvard was by no means unique within the academy in this de facto silencing of heterodox views.

In the face of this stiff resistance, a small group of HLS students persisted. With the support of generous benefactors; the guidance of Clifford Taylor, former Chief Justice of the Michigan Supreme Court and the Journal’s first academic advisor; the leadership of the Journal’s Co-Founder and first Editor-in-Chief, Steve Eberhard; and the tireless work of many others, the Journal was up and running. We published our first edition in the spring of 1978. And several years later, in 1982, after the Journal published the proceedings of a small conference of conservative law students, that group decided to make the Harvard Journal of Law & Public Policy its official publication. For nearly four decades since that initial collaboration, and continuing today, every member of the Federalist Society continues to receive the Journal as a benefit of a “FedSoc” membership.

In the intervening 40 years, the Harvard Journal of Law & Public Policy has grown from an idea in the minds of a handful of law students to the nation’s leading conservative and libertarian legal journal.[1] U.S. Senators, captains of industry, jurists at every level including the U.S. Supreme Court, and many others have all cut their teeth serving on the Journal’s staff. The Journal has published the writings of the greatest conservative legal minds of our day and has provided an opportunity for the next generation to engage with that scholarship.

The Journal was conceived as a home for conservative scholarship in print, and we look forward to continuing that important task well into the future. Today, however, the same hostility that descends on any challenge to progressive orthodoxy has crept into the digital marketplace of ideas as well. I am very pleased that JLPP: Per Curiam is emerging to combat this old threat that is surfacing in a new form.

The Harvard Journal of Law & Public Policy is uniquely qualified to provide an alternative to progressive platforms. The Journal was, and remains, as Steve Eberhard and I referred to it in its first volume, vox clamantis in deserto.[2] The Journal exists as a bulwark against the threat of a uniform progressive ideology establishing unchallenged dominance over the legal academy. JLPP: Per Curiam will be an essential part of continuing that mission in a digital age.

Despite the Journal’s growth and success since 1978, its original mission of providing an alternative to unvaryingly progressive scholarship is as relevant today as it was then. Throughout the nation in academia, government, entertainment, and elsewhere, diversity of opinion and freedom of thought are under threat. This situation is not new. What is new is the incursion of this ideological intolerance into the promised bastion of free expression: the Internet.

Today, online speech platforms all too often see themselves as judges of truth. Those who challenge this dogma risk their ideas being labeled misinformation, chilling dissent. Our “moral arbiters” also wield “deplatforming” as a tool to outright silence those with whom they disagree, removing the pen from the writer’s hand. In this sense, the Internet resembles the landscape of legal scholarship in the late 1970s.

However, in an important way, the Internet is different. The low cost of Internet distribution makes self-publication accessible. As a result, ideas distributed online are less dependent on institutional gatekeepers than ideas published in print. This dynamic is exemplified by conservative legal blogs, which have amassed substantial followings. “Less dependent,” however, should not be understood as independent. The availability of self-publication is not a perfect defense against ideological intolerance, because self-publication cannot replace the amplifying effect of a well-established platform, particularly one with decades of rigorous scholarship behind its name. Today, the Journal launches JLPP: Per Curiam to stand in the digital gap.

JLPP: Per Curiam will be a home for legal scholarship that dares to challenge progressive orthodoxy online. It will provide a digital platform for the top conservative legal minds of our day and a source of high-level legal argumentation for the large audience of our readers—including law students, lawyers, members of the judiciary, and policymakers—who seek an alternative to the existing sources of legal discussion that still predominate in today’s classrooms, faculty lounges, and government offices.

At the time of the Journal’s founding, it was our hope to select and publish a small number of thoughtful articles and notes each year. Of course, in 1978 it would have been difficult to imagine the Internet. Today, though, as one of the most widely circulated legal journals in the United States—and the only conservative one—JLPP: Per Curiam will take that platform online, where it can feature an even larger number of conservative voices and views that are urgently needed in today’s legal debate. Shorter articles, published more frequently, will make the Journal even more impactful than the triannual print edition alone. Additionally, JLPP: Per Curiam will meet readers online where they increasingly live and work, further multiplying the Journal’s impact. Everyone connected with the Journal, past and present, is excited by this development, and we have great expectations for its success.

JLPP: Per Curiam is a significant undertaking for the Journal. It is impossible to commemorate such an event without thanking those who made it possible. As I briefly mentioned earlier, that includes those first students who organized the Journal, met with the Harvard administration, and served on our first masthead. I also wish to express my gratitude to the generous individuals and foundations whose investments made the Journal’s early volumes possible. Thanks also are due to Clifford Taylor, former Chief Justice of the Michigan Supreme Court, our first advisor. He stepped forward when faculty members at Harvard Law School would not. His courage deserves mention here.

My thanks would be incomplete without a special recognition of my co-founder and the Journal’s first Editor-in-Chief, Steve Eberhard. As I wrote several years ago, “Steve was an exceptional student, a patriot, and a true friend. His untimely death robbed the conservative movement and the nation of a great man who would surely have been one of its most brilliant and courageous leaders.”[3] This advancement of the Journal into the digital age, and all of our future innovations, are possible only because Steve stepped into the breach all those years ago. He is still missed today.

Finally, I congratulate and thank those who are working today to establish JLPP: Per Curiam. Your efforts will expand the reach of conservative legal scholarship at a critical time in our nation’s history and in a vitally important medium. I wish your team the best of luck in this new endeavor!


E. Spencer Abraham




[1] Journals and Publications, HARV. L. SCH., [] (last visited Aug. 10, 2021).

[2] E. Spencer Abraham & Steven J. Eberhard, Preface, 1 Harv. J.L. & Pub. Pol’y, at vii, viii (1978) (meaning “the voice of one crying out in the wilderness,” from Isaiah 40:3, Matthew 3:3, Mark 1:3, Luke 3:4, John 1:23 (Vulgate)).

[3] E. Spencer Abraham, A Founder’s Reflection’s Reflections: The Journal at Forty Years, 44 Harv. J.L. & Pub. Pol’y, 446 (2018).

Read More »

An Introduction to JLPP: Per Curiam – Eli Nachmany and Alexander Khan

Posted by on Aug 31, 2021 in Per Curiam

Download PDF

An Introduction to JLPP: Per Curiam

Eli Nachmany, Editor-in-Chief of the Harvard Journal of Law & Public Policy

Alexander Khan, Director of JLPP: Per Curiam


Since 1978, the Harvard Journal of Law & Public Policy has been America’s law journal—the premier forum for conservative and libertarian legal scholarship. Three times a year, JLPP’s subscribers find that small, iconic, crimson book in their mailboxes, featuring essays from top legal scholars, commentators, and public officials. Copies of JLPP have made their way into law libraries, White House offices, Supreme Court chambers, and the halls of Congress.

But the nature of legal scholarship is changing. While JLPP has been able to meaningfully impact the conversation in various areas of law and public policy over the years, its current format focuses on the publication of scholarship that takes a longer view. The amount of time between our Journal’s acceptance of an article for publication and that article’s actual publication can be up to a year. This has allowed us to publish thoughtful and scholarly articles throughout our many years. Articles that have shaped legal thinking and, at times, the political landscape. What this means, however, is that we have missed out on shorter, more timely commentary.

So, today, we take the next step in JLPP’s history and launch JLPP: Per Curiam, an online counterpart to our print edition. The creation of Per Curiam will allow JLPP to participate in the day-to-day legal debates that animate much of our public discourse, providing a platform for conservative and libertarian content. It will enable JLPP to serve as a home for cutting edge commentary, reaction, opinion, and shorter scholarship. To be sure, we will continue to publish our print edition three times a year, showcasing the rigorous legal scholarship for which JLPP has come to be known. But now, with Per Curiam, JLPP adds a new dimension.

To launch Per Curiam, we are honored to present an online symposium celebrating Justice Thomas’s 30th Anniversary on the Supreme Court of the United States. This symposium includes essays and video interviews with former clerks of Justice Thomas, many of whom are currently serving on the federal bench. Perhaps the best part is that multiple participants in today’s symposium were once JLPP editors, whether as students at Harvard Law School or as student editors from other law schools helping out with our yearly national symposium issue. While the symposium includes a significant amount of reflection on Justice Thomas’s impact on the law, it also highlights the kind of man he is, from the perspective of those who have had the great fortune of working for him as law clerks.

In addition to this set of essays and video interviews, we have some terrific content already planned for Per Curiam. But our inbox is open. We invite submissions from law professors, judges, practitioners, and law students on issues of law and public policy from a conservative or libertarian perspective. We anticipate that Per Curiam will keep JLPP at the heart of the legal conversation, and we look forward to making this the best year in JLPP’s history to date.

Read More »