In most presidential elections, the Electoral College is part of the forgotten plumbing of our constitutional system.^1×1. See Robert M. Alexander, Representation and the Electoral College 93 (2019) (observing that the winners of the popular vote and Electoral College have differed only six times). But in 2016, a small group of “faithless” electors sought to change the outcome by voting for someone other than the candidate who carried the popular vote in their state.^2×2. Chiafalo v. Washington, 140 S. Ct. 2316, 2322 (2020). This effort failed, and in Chiafalo v. Washington,^3×3. 140 S. Ct. 2316. the Supreme Court bolstered states’ power to prevent faithless voting, holding that states may enforce a presidential elector’s pledge to support his party’s nominee.^4×4. See id. at 2320, 2322. The Court’s reasoning was grounded in the concept of constitutional liquidation — the idea that when the Constitution’s text is ambiguous, meaning can be settled by well-established practice.^5×5. See id. at 2326–28; NLRB v. Noel Canning, 134 S. Ct. 2550, 2560 (2014) (“[I]t ‘was foreseen at the birth of the Constitution, that difficulties and differences of opinion might occasionally arise in expounding terms & phrases necessarily used in such a charter . . . and that it might require a regular course of practice to liquidate & settle the meaning of some of them.’” (omission in original) (quoting Letter from James Madison to Spencer Roane (Sept. 2, 1819), in 8 The Writings of James Madison 447, 450 (Gaillard Hunt ed., 1908))). See generally William Baude, Constitutional Liquidation, 71 Stan. L. Rev. 1 (2019). In some ways, Chiafalo represented the perfect case for constitutional liquidation: both political and epistemic concerns weighed in favor of sustaining settled practice. But by relying on history rather than buttressing its conclusion with other constitutional values, the Court ignored the level-of-generality problem that plagues constitutional liquidation as an interpretive tool.

Americans do not elect the President directly. Rather, after the popular vote is tallied, each state chooses its presidential electors in accordance with state law, and those members of the Electoral College select the President.^6×6. See U.S. Const. art. II, § 1, cl. 2; Chiafalo, 140 S. Ct. at 2321. In the twentieth century, some states began to enact measures to ensure that electors’ ballots would reflect the state vote.^7×7. Chiafalo, 140 S. Ct. at 2321. Thirty-two states have pledge laws that impose that duty by law.^8×8. Id. Fifteen states go further and either remove and replace electors who violate their oaths or impose a monetary fine on such electors.^9×9. Id. at 2322 & n.2.

In November 2016, three Washington electors who had pledged to support Hillary Clinton violated their oaths.^10×10. Id. at 2322. Though Secretary Clinton carried the state vote, the electors cast their votes for Colin Powell.^11×11. Id. By voting for Secretary Powell, the electors hoped to encourage other electors to vote for an alternative candidate and punt the presidential election to the House of Representatives.^12×12. Id. Under Washington law at the time, an elector who did not vote for his party’s candidate was subject to a fine of up to $1,000.^13×13. Id. The State fined the three electors, and the electors contested the penalties, arguing that the Washington statute violated both the First and Twelfth Amendments.^14×14. Verbatim Report of Proceedings at 48, In re Guerra, No. 17-2-02446-34 (Wash. Super. Ct. Dec. 8, 2017). In an oral hearing, the state trial court ruled against the electors, explaining that the law passed muster since “[t]he State is not adding a qualification, nor is the State here requiring specific performance of the pledge.”^15×15. Id. at 49.

The Supreme Court of Washington affirmed.^16×16. In re Guerra, 441 P.3d 807, 807 (Wash. 2019). First, the court rejected the electors’ argument that the imposition of a fine unconstitutionally interfered with a federal function. While electors do perform a federal function, Article II “grants to state legislatures plenary power to appoint electors and determine the manner in which their appointment shall be made, and the fine falls within that broad grant of authority.”^17×17. Id. at 813. Second, the court dismissed the electors’ argument that the Constitution granted them discretion to exercise judgment. In Ray v. Blair,^18×18. 343 U.S. 214 (1952). the Supreme Court held that a pledge requirement — though one without a fine — was constitutional.^19×19. Id. at 231. Extending Ray, the Washington Supreme Court explained that “[i]n the same way that the Twelfth Amendment does not prevent an elector from pledging himself, it does not prevent a state from requiring its electors pledge to vote for its party candidate.”^20×20. Guerra, 441 P.3d at 816. In a separate case raising a similar question, Baca v. Colorado Department of State,^21×21. 935 F.3d 887 (10th Cir. 2019), rev’d, 140 S. Ct. 2316 (2020) (per curiam). the Tenth Circuit reached the opposite conclusion.^22×22. See id. at 902. The Supreme Court took up both cases to address the split.^23×23. Chiafalo, 140 S. Ct. at 2323.

The Supreme Court affirmed the Washington Supreme Court’s decision.^24×24. Id. Writing for eight Justices, Justice Kagan^25×25. Justice Kagan was joined by Chief Justice Roberts and Justices Ginsburg, Breyer, Alito, Sotomayor, Gorsuch, and Kavanaugh. explained that “[t]he Constitution’s text and the Nation’s history both support allowing a State to enforce an elector’s pledge to support his party’s nominee — and the state voters’ choice — for President.”^26×26. Chiafalo, 140 S. Ct. at 2323–24. Justice Kagan began with the text of the Constitution. Article II gives states the power to appoint electors “in such Manner as the Legislature thereof may direct.”^27×27. U.S. Const. art. II, § 1, cl. 2. And this “power to appoint an elector (in any manner) includes power to condition his appointment — that is, to say what the elector must do for the appointment to take effect.”^28×28. Chiafalo, 140 S. Ct. at 2324. Thus, Justice Kagan concluded that “absent some other constitutional constraint,” Article II enables states to penalize faithless voting.^29×29. Id. Chiafalo was decided alongside Colorado Department of State v. Baca, 140 S. Ct. 2316 (2020) (per curiam). Baca considered the closely related question of whether states could remove and replace a faithless elector. See Baca v. Colo. Dep’t of State, 935 F.3d 887, 902 (10th Cir. 2019), rev’d, 140 S. Ct. 2316. In a one-sentence per curiam opinion, the Court held that they could, explaining that states have the power to replace an elector “for the reasons stated in Chiafalo v. Washington.” Baca, 140 S. Ct. at 2316.

Justice Kagan rejected the electors’ argument that the text of the Constitution compels discretion. Article II names the members of the Electoral College as “Electors,”^30×30. U.S. Const. art. II, § 1, cl. 2. and the Twelfth Amendment demands that the electors “vote by ballot.”^31×31. Id. amend. XII. According to the electors, the “plain meaning” of these terms demands “freedom of choice.”^32×32. Chiafalo, 140 S. Ct. at 2325 (first quoting Consolidated Opening Brief for Presidential Electors at 31, Chiafalo, 140 S. Ct. 2316 (2020) (No. 19-465); and then quoting id. at 29). Justice Kagan rejected this claim, explaining that “although voting and discretion are usually combined, voting is still voting when discretion departs.”^33×33. Id. For example, Justice Kagan noted, a person who “always votes in the way his spouse, or pastor, or union tells him to” still “votes.”^34×34. Id. Justice Kagan also contrasted the Constitution’s “barebones” description of the Electoral College with state constitutions at the time of the Founding.^35×35. Id. at 2324; see id. at 2324–25. Unlike the Federal Constitution, certain state constitutions were explicit that electors for state offices should select those who “they, in their judgment and conscience, believe best qualified for the office.”^36×36. Id. at 2325 (quoting Md. Const. of 1776, art. XVIII). Finally, Justice Kagan explained that even if the Framers expected that electors would exercise discretion, the drafters’ expectations were not binding because “[w]hether by choice or accident, the Framers did not reduce their thoughts about electors’ discretion to the printed page.”^37×37. Id. at 2326.

Next, Justice Kagan turned to history. Quoting James Madison, Justice Kagan explained that historical practice can “liquidate & settle” the meaning of disputed constitutional provisions.^38×38. Id. (quoting Letter from James Madison to Spencer Roane, supra note 5, at 450). First, Justice Kagan reviewed the history of the Twelfth Amendment.^39×39. See id. at 2326–28. In 1796, the nation’s first competitive election, the Electoral College as designed broke down.^40×40. See id. at 2320, 2327. In that election, Federalist John Adams came in first, and his rival, Democratic-Republican Thomas Jefferson, came in second.^41×41. See id. And so “leaders of the era’s two warring political parties . . . became President and Vice President respectively.”^42×42. Id. at 2320. In 1800, the Electoral College failed again when Jefferson and Aaron Burr tied in the electoral vote.^43×43. See id. at 2320–21, 2327. The Twelfth Amendment “made party-line voting safe” by forcing electors to cast separate votes for President and Vice President.^44×44. Id. at 2327; see U.S. Const. amend. XII. As Justice Kagan explained, “the new procedure allowed an elector to ‘vote the regular party ticket’ and thereby ‘carry out the desires of the people.’”^45×45. Chiafalo, 140 S. Ct. at 2327–28 (quoting Ray v. Blair, 343 U.S. 214, 224 n.11 (1952)). By the nineteenth century, courts, legal commentators, and state election law all “recognized the electors as merely acting on other people’s preferences.”^46×46. Id. at 2327; see id. at 2327–28. Finally, Justice Kagan noted that faithless electors are a historical anomaly and account for less than one percent of all electoral votes cast since the Founding.^47×47. See id. at 2328. Thus, Justice Kagan concluded that history supported the constitutional understanding that states may sanction faithless electors.^48×48. Id. at 2326–28.

Justice Thomas concurred in the judgment.^49×49. Justice Thomas was joined in part by Justice Gorsuch. Though Justice Thomas agreed that states can fine faithless electors, he disagreed with the Court’s effort to ground that power in Article II.^50×50. See Chiafalo, 140 S. Ct. at 2329 (Thomas, J., concurring in the judgment). Justice Gorsuch did not join this part of the concurrence. First, Justice Thomas explained that the Court’s reading of Article II “stretch[ed] the plain meaning of the Constitution’s text” since “determining the ‘Manner’ of appointment certainly does not include the power to impose requirements as to how the electors vote after they are appointed.”^51×51. Id. at 2330 (quoting U.S. Const. art. II, § 1, cl. 2). Next, he explained that the Court’s reading of Article II was inconsistent with its prior interpretation of identical language in Article I. Article I provides that “[t]he Times, Places and Manner of holding Elections for Senators and Representatives, shall be prescribed in each State by the Legislature thereof.”^52×52. Id. (alteration in original) (quoting U.S. Const. art. I, § 4, cl. 1). In U.S. Term Limits, Inc. v. Thornton,^53×53. 514 U.S. 779 (1995). the Court held that this provision merely gives Congress the narrow power to “issue procedural regulations” rather than “the broad power to set procedural qualifications.”^54×54. Chiafalo, 140 S. Ct. at 2330 (Thomas, J., concurring in the judgment) (quoting Thornton, 514 U.S. at 833). Yet the Court’s opinion in Chiafalo “appear[ed] to take the exact opposite view.”^55×55. Id.

Justice Thomas argued that the Court’s decision should instead have rested on the Tenth Amendment. Quoting his dissent in Thornton, Justice Thomas explained that the meaning of the Tenth Amendment is that “‘[w]here the Constitution is silent about the exercise of a particular power[,] that is, where the Constitution does not speak either expressly or by necessary implication,’ the power is ‘either delegated to the state government or retained by the people.’”^56×56. Id. at 2334 (alterations in original) (quoting Thornton, 514 U.S. at 847–48 (Thomas, J., dissenting)). And since “nothing in the text or structure of Article II and the Twelfth Amendment contradict[ed] the fundamental distribution of power preserved by the Tenth Amendment,” the power to regulate electors remained with the states.^57×57. Id.

The Chiafalo Court embraced constitutional liquidation, a principle of construction that has always been in the background of constitutional law but is rarely discussed explicitly. On the one hand, the Court’s opinion highlights the virtues of liquidation, including departmentalism and epistemic humility. But by relying too heavily on history, the Court gave short shrift to the level-of-generality problem that emerges when trying to apply the lessons of history to the present. Since history did not provide a definite answer in this case, the Court should have explained how its reading of history is consonant with other constitutional values.

The concept of constitutional liquidation can be traced back to James Madison, who claimed that “‘a regular course of practice’ can ‘liquidate . . . the meaning of’ disputed or indeterminate ‘terms & phrases.’”^58×58. Id. at 2326 (majority opinion) (quoting Letter from James Madison to Spencer Roane, supra note 5, at 450); see also The Federalist No. 37, at 236 (James Madison) (Jacob E. Cooke ed., 1961) (“All new laws . . . are considered as more or less obscure and equivocal, until their meaning be liquidated and ascertained by a series of particular discussions and adjudications.”). In other words, constitutional liquidation treats settled historical practice as the rough equivalent of judicial precedent and applies history’s “holding” to the case at hand.^59×59. Baude, supra note 5, at 52; see id. at 37, 51–52 (discussing the analogy between constitutional liquidation and judicial precedent). It’s helpful to contrast constitutional liquidation with originalism. Originalist opinions use history to make arguments about the original meaning of the Constitution’s text.^60×60. See Antonin Scalia, Essay, Originalism: The Lesser Evil, 57 U. Cin. L. Rev. 849, 856–57 (1989). For example, an originalist might refer to Founding-era documents to explain how the text of the Second Amendment should be understood.^61×61. See District of Columbia v. Heller, 554 U.S. 570, 581 (2008). But in Chiafalo, the Court used history in a very different way. Instead of using history to shed light on the meaning of the Constitution’s text, the Court recognized that history has normative power in its own right.^62×62. See Chiafalo, 140 S. Ct. at 2326–28. Though historical practice has always played a role in constitutional interpretation, explicit references to constitutional liquidation were uncommon until recently — the Court’s first detailed treatment of liquidation appeared in NLRB v. Noel Canning^63×63. 134 S. Ct. 2550 (2014). in 2014.^64×64. See id. at 2560; Baude, supra note 5, at 6–7. In Chiafalo, the Court invoked this interpretive tool again. As a result, Chiafalo represents an important point to pause and reflect on liquidation’s merits.

In some respects, Chiafalo was the perfect case for constitutional liquidation. As Professor William Baude explains, one of liquidation’s great virtues is departmentalism: since activity in the legislative and executive branches defines settled practice, liquidation allows the elected branches to participate in constitutional interpretation.^65×65. See Baude, supra note 5, at 35–36. In Chiafalo, relying on historical practice created space for state legislatures — and, indirectly, voters — to participate in constitutional construction.^66×66. See Chiafalo, 140 S. Ct. at 2328 (citing practices established by state legislatures). Justice Kagan adverted to departmentalism in her closing line, explaining that the states’ power to sanction faithless electors “accords . . . with the trust of a Nation that here, We the People rule.”^67×67. Id. Chiafalo represents an ideal case for this sort of shared approach to constitutional interpretation. The Supreme Court’s authority hinges in large part on popular acceptance of the Court’s holdings,^68×68. See, e.g., Richard H. Fallon, Jr., Judicial Supremacy, Departmentalism, and the Rule of Law in a Populist Age, 96 Tex. L. Rev. 487, 493–94 (2018). and few cases attract more controversy than those wading into electoral politics.^69×69. Consider the lasting controversy surrounding Bush v. Gore, 531 U.S. 98 (2000) (per curiam), or the more recent debate surrounding Department of Commerce v. New York, 139 S. Ct. 2551 (2019), and Rucho v. Common Cause, 139 S. Ct. 2484 (2019). See, e.g., Richard L. Hasen, The Supreme Court May No Longer Have the Legitimacy to Resolve a Disputed Election, The Atlantic (Feb. 3, 2020), https://www.theatlantic.com/ideas/archive/2020/02/supreme-court-elections/605899 [https://perma.cc/VS3X-7C4H]. By relying on constitutional liquidation, the Court incorporated the popular will into the act of judicial interpretation and deftly bypassed difficult questions about countermajoritarian decisionmaking.

Constitutional liquidation also has the virtue of epistemic humility.^70×70. See Baude, supra note 5, at 44–47. By grounding constitutional interpretation in historical practice, liquidation incorporates the “accumulated wisdom of many generations.”^71×71. Id. at 44–45 (quoting David A. Strauss, Common Law Constitutional Interpretation, 63 U. Chi. L. Rev. 877, 892 (1996)). Of course, the Constitution’s meaning should not always be fixed by deference to the crowd. For example, in cases implicating the Equal Protection Clause, historical traditions might reflect systematic bias rather than accumulated wisdom.^72×72. Cf. id. at 45–46 (describing the arguments against considering tradition). But legislators’ judgment about faithless electors is unlikely to be colored by these sorts of biases. Because it’s nearly impossible for legislatures to predict ex ante whether elector discretion would benefit their preferred political party, the issue lacks partisan valence. Tellingly, forty-five states and the District of Columbia submitted an amicus brief supporting states’ right to bind electors.^73×73. See Brief for South Dakota & 44 States & the District of Columbia as Amici Curiae in Support of Colorado & Washington at 2–3, Chiafalo, 140 S. Ct. 2316 (2020) (No. 19-465).

Though constitutional liquidation has clear virtues, the Court may have leaned too heavily on this interpretive tool. Roughly speaking, constitutional liquidation treats history like judicial precedent.^74×74. See Baude, supra note 5, at 37, 52–53. And like a judge applying judicial precedent, a judge applying past practices to current events must determine that precedent’s scope. In other words, liquidation hinges on judges’ ability to determine history’s holding.^75×75. See id. at 52. This is easier said than done, since history does not wear its holding on its sleeve. Instead, there are two ways to determine the scope of a historical precedent: First, judges might look at patterns of practice and ask which generalizable rules are compatible with that history.^76×76. See id. Second, judges might examine historical debates and ask whether the reasons given by other constitutional actors support a given rule.^77×77. See id. But in the case of faithless electors, both of these tests are less determinate than the Court acknowledged.^78×78. But see Keith E. Whittington, Originalism, Constitutional Construction, and the Problem of Faithless Electors, 59 Ariz. L. Rev. 903, 938 (2017) (arguing that “[h]istorical traditions do not suggest that pledged presidential electors are free to cast their ballots for someone else if they believe that their party’s nominee is an unwise choice to be [P]resident”).

Historical practice alone does not conclusively answer whether states can bind electors in exceptional circumstances.^79×79. For an illuminating discussion of the application of constitutional liquidation to the question of elector discretion, see Rebecca Green, Liquidating Elector Discretion, 15 Harv. L. & Pol’y Rev. (forthcoming 2020) (manuscript at 15–16) (on file with the Harvard Law School Library). Generalizing from history, Justice Kagan explained that presidential electors are “trusty transmitters”^80×80. Chiafalo, 140 S. Ct. at 2326. of the state vote and that “[t]he history going the opposite way is one of anomalies only.”^81×81. Id. at 2328. But arguably, that same history is consistent with the electors’ position.^82×82. See Green, supra note 79 (manuscript at 10) (arguing that “elector discretion can be both rare and settled practice”). The presidential electors agreed with Justice Kagan that faithless voting should be an anomaly. As the electors explained in their brief, “[t]hey believed that the exceptional circumstances of the 2016 election counseled that they act contrary to their pledges.”^83×83. Consolidated Opening Brief for Presidential Electors, supra note 32, at 52. Moreover, though faithless electors are a rarity, a number of historical practices are consistent with the idea that electors are constitutionally vested with discretion. Congress has never rejected a faithless elector’s vote in the final tally.^84×84. Id. at 46–47. Thirty-five states either impose no restrictions on electors or require only a simple pledge,^85×85. See Chiafalo, 140 S. Ct. at 2322 (explaining that fifteen states have enacted “sanctions-backed pledge law[s]”). and the ballots that many states provide electors seemingly anticipate elector choice.^86×86. See Green, supra note 79 (manuscript at 14–15). Finally, though faithless electors have never changed the outcome of a presidential election, they have impacted the vice presidency. In 1836, twenty-three Virginia electors declined to vote for vice-presidential nominee Richard Mentor Johnson.^87×87. Alexander, supra note 1, at 132. The electors withheld their votes because they objected to Vice President Johnson’s marriage to an African American woman. Id. Instead, the Senate had to use the Twelfth Amendment’s contingency procedures to vote him into office.^88×88. Id.

When trying to determine history’s holding, we might look beyond patterns of practice and examine elected officials’ reasoning and deliberations.^89×89. See Baude, supra note 5, at 52. Just as a judicial precedent’s rationale may help determine the scope of a holding, so too may evidence from political deliberations shed light on the “holding” of history.^90×90. Id. But like patterns of practice, historical reasoning in this case is ambiguous. To be sure, there is a long line of political officials explaining that, whatever the Framers’ original intent, electors lack discretion: President Taft described the Electoral College as “an instrumentality for registering the people’s vote,”^91×91. William Howard Taft, Liberty Under Law 12 (1922). and Justice Story explained that “electors are now chosen wholly with reference to particular candidates, and are silently pledged to vote for them.”^92×92. 3 Joseph Story, Commentaries on the Constitution of the United States § 1457, at 321 (Boston, Hilliard, Gray & Co. 1833).

On the other hand, there are at least some indicia that elected officials who have considered the matter do not believe presidential electors can be stripped of discretion. First, there have been multiple proposed amendments either binding electors to the state vote or abolishing the Electoral College entirely^93×93. See Alexander, supra note 1, at 183–89 (surveying various reform proposals). — if states had the power to remove and replace faithless electors, a constitutional amendment might be unnecessary. Second, the text of the Twentieth Amendment seemingly presumes that electors have discretion. The Twentieth Amendment provides that the Vice President should take office if the President dies after the Electoral College votes, but it’s silent on how to handle the death of the presumptive nominee before the Electoral College meets.^94×94. Reply Brief for Presidential Electors at 20, Colo. Dep’t of State v. Baca, 140 S. Ct. 2316 (2020) (No. 19-518); see U.S. Const. amend. XX. The drafters of the Twentieth Amendment instead assumed that electors had the discretion to handle this sort of edge case.^95×95. Reply Brief for Presidential Electors, supra note 94, at 20. Third, in the lone case where an anomalous electoral vote was challenged in Congress, at least some representatives evinced an understanding that electors are entrusted with irrevocable discretion. In 1969, a Republican elector cast his vote for George Wallace rather than for Richard Nixon. In the corresponding debate, Senator Sam Ervin argued that “[t]he Constitution is very plain” that Congress cannot “take what was an ethical obligation and convert it into a constitutional obligation.”^96×96. 115 Cong. Rec. 203 (1969) (statement of Sen. Sam Ervin); see also Consolidated Opening Brief for Presidential Electors, supra note 32, at 47 (discussing the 1969 debate). Finally, surveys of electors themselves are also telling: in presidential elections since 2004, anywhere from seven to twenty-one percent of electors reported that they gave some consideration to defecting.^97×97. Alexander, supra note 1, at 147. Twenty-one percent of electors considered defecting in 2016. Id. Though more electors considered defecting in 2016 than in earlier elections, 2016 is not an anomaly: between 2004 and 2012, seven to twelve percent reported that they considered defecting. Id.

In short, neither patterns of practice nor historical reasoning can decisively reject the presidential electors’ claim that the Constitution grants them discretion. This problem of determining history’s holding closely resembles the “level of generality” problem in defining the scope of fundamental rights.^98×98. See, e.g., Jack M. Balkin, The New Originalism and the Uses of History, 82 Fordham L. Rev. 641, 654 (2013) (“[P]eople also disagree about how to describe and characterize tradition, and the level of generality at which we should understand and apply the teachings of the past . . . .”); Laurence H. Tribe & Michael C. Dorf, Levels of Generality in the Definition of Rights, 57 U. Chi. L. Rev. 1057, 1057–59 (1990). Relying on precedent alone cannot define the proper scope of constitutional guarantees like the Equal Protection Clause or the First Amendment. For a given line of precedent, there are a myriad of possible generalizable rules, each of which may produce different outcomes in future cases.^99×99. See Tribe & Dorf, supra note 98, at 1074–75. Similarly, in Chiafalo, there are multiple potential “holdings” that are plausibly consistent with history. In both cases, “[t]he selection of a level of generality necessarily involves value choices.”^100×100. Id. at 1058. But these value choices do not leave judges unmoored from any sort of guiding principles. Rather, courts faced with the level-of-generality problem often engage in common law–style reasoning, seeking “unifying principles to link disparate decisions”^101×101. Id. at 1069. through a process of “interpolation and extrapolation.”^102×102. Id. at 1059.

A similar approach could have buttressed the Court’s approach to history in Chiafalo. Two principles support the level of generality with which the Court interpreted the history of faithless electors. The first is the observation that popular sovereignty is a cornerstone of our constitutional structure. As Justice Kagan emphasized in her closing line, in our constitutional system, “We the People rule.”^103×103. Chiafalo, 140 S. Ct. at 2328. Supreme Court precedent reinforces the notion that “the right to have one’s vote counted” is fundamental.^104×104. Reynolds v. Sims, 377 U.S. 533, 554 (1964) (quoting United States v. Mosley, 238 U.S. 383, 386 (1915)). The second principle is pragmatism, especially when filling in the “construction zone” where the Constitution’s text permits multiple readings.^105×105. See Lawrence B. Solum, The Interpretation-Construction Distinction, 27 Const. Comment. 95, 108 (2010) (describing the “construction zone”). As multiple Justices acknowledged during oral argument, allowing faithless electors to change the outcome of presidential elections risks chaos.^106×106. See Transcript of Oral Argument at 21, 33, Chiafalo, 140 S. Ct. 2316 (2020) (No. 19-465), https://www.supremecourt.gov/oral_arguments/argument_transcripts/2019/19-465_c0n2.pdf [https://perma.cc/5Q94-TY5C]. On the other hand, the electors argued that there is “risk on both sides” since binding electors could create uncertainty if the nominee died before the Electoral College votes. Id. at 34. The “avoid chaos principle of judging,” as Justice Kavanaugh termed it in oral argument,^107×107. Id. at 33. Analogously, the Court’s decision in U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779 (1995), which held that states cannot impose term limits on members of Congress, rests on a “pragmatic understanding” of how our governing institutions should work. Stephen Breyer, Making Our Democracy Work 85 (2010); see id. at 85–87. might help us pick among competing understandings of history. In short, when history can be read at multiple levels of generality, seeking consistency with constitutional principles and precedents can help select among various possibilities.^108×108. In Michael H. v. Gerald D., 491 U.S. 110 (1989), Justice Scalia advocated for a different approach to solving the level-of-generality problem. According to Justice Scalia, the right level of generality at which to interpret a fundamental right is “the most specific level at which a relevant tradition protecting, or denying protection to, the asserted right can be identified.” Id. at 127 n.6 (plurality opinion). But the Michael H. test is inapposite to constitutional liquidation. The question in a case like Chiafalo is what to do when history itself admits multiple readings. In such a case, there simply is no “most specific level” at which history can be read.

In Chiafalo, the Court highlighted that constitutional liquidation is an important tool in the Court’s interpretive toolbox and one that can bring together members of the Court from across the political spectrum. What’s more, in contrast to many other interpretive tools, liquidation has the twin virtues of departmentalism and epistemic humility. But though Chiafalo highlighted constitutional liquidation’s virtues, it also exemplified its flaws. Like other forms of interpretation, constitutional liquidation suffers from a level-of-generality problem; even well-settled historical practices are consistent with a wide variety of interpretations. To address this uncertainty, the Court ought to have explained how other constitutional values aligned with its preferred interpretation of history.