Harvard Law Review Harvard Law Review Harvard Law Review

State Government

Executive Adjudication of State Law

The Founding Fathers, once having thrown off the yoke of King George, devised a system of government based on the separation of powers. Students of history, the Framers realized that a constitution’s structure, more than any specific clause or guarantee, chiefly determined whether a governing document would secure individual liberty.1×1. See, e.g., The Federalist No. 48, at 310 (James Madison) (Clinton Rossiter ed., 2003). To that end, the Constitution divides power twice over. Horizontally, the federal government is broken into three coequal branches. Vertically, the Constitution splits power among the federal and state governments, a system of “dual sovereignty” where each sovereign checks the other.2×2. Gregory v. Ashcroft, 501 U.S. 452, 457 (1991).

Within this dual-sovereign system, the federal government is vested with certain enumerated powers, and the states keep the rest.3×3. See The Federalist No. 45, supra note 1, at 289 (James Madison). Figuring out which sovereign is supposed to do what job can be straightforward from the face of the Constitution. For instance, the federal government is responsible for the military,4×4. U.S. Const. art. I, § 8, cl. 12; id. art. II, § 1, cl. 1; id. art. II, § 2, cl. 1. while states administer elections.5×5. Id. art. I, § 4. But when the Constitution is silent as to a given responsibility, the Supreme Court has turned to history and structural reasoning to figure out how a question should be resolved along the vertical separation of powers.6×6. Coyle v. Smith, 221 U.S. 559 (1911); John F. Manning, Federalism and the Generality Problem in Constitutional Interpretation, 122 Harv. L. Rev. 2003 (2009).

This approach makes good sense: not everything in a deal needs to be in fine print for it to count. We follow this intuition all the time. Imagine you’re playing Texas Hold’em. Any rulebook would have the basics on how to play: flushes beat straights; minimum bets match the big blind. Suppose, though, that you have a friend who folds before the flop and, after three clubs hit the table, exclaims: “I would’ve had a flush!” Assuming this isn’t in any rulebook (and don’t bring rulebooks to games), yelling out your hand would still be barred because it reveals what cards are out there and warps betting. It is as prohibited as betting below the big blind, even if the latter is in writing and the former isn’t.

This Note argues that the federal government might be breaking one of the implicit terms of the constitutional deal in how it adjudicates claims arising under state law.7×7. A variant of this argument was raised by Mr. Schor in Commodity Futures Trading Commission v. Schor, 478 U.S. 833 (1986). See Brief for Respondent, Schor, 478 U.S. 833 (Nos. 85-621, 85-642), 1985 WL 669415, at *34. But the Court dismissed it, in large part because Schor “identifie[d] no historical support.” Schor, 478 U.S. at 858. This Note attempts to remedy that deficiency. Here’s the basic idea: Congress is presently operating beyond its constitutional authority under Article I when it provides for non–Article III tribunals to adjudicate state law because doing so contravenes the original deal’s vertical separation of powers, and is therefore never “proper” under the Necessary and Proper Clause.

Today, state law actions can go to state courts, federal courts (sometimes), and certain executive branch non–Article III tribunals (sometimes). The third category, this Note argues, poses serious vertical separation of powers problems. States have a freestanding sovereign interest in which tribunals determine rights and obligations under their law, and the historical record demonstrates that they guarded this interest vigorously during the Constitution’s drafting and ratification. Indeed, there’s substantial evidence that the states specifically agreed to a deal where state law claims could be heard only by state courts or, in defined cases, federal courts exercising judicial power. Executive adjudication falls outside this deal, whose terms are baked into the Constitution and insulated from unilateral edits by the federal government.

To date, in analyzing executive adjudication of state law, the Court has used a balancing test to figure out whether these executive tribunals encroach too much on the federal courts’ territory and thus violate the horizontal separation of powers.8×8. Schor, 478 U.S. at 851. At times they do, at times they don’t.

But this puts the cart before the horse. For there to be a non–Article III tribunal within the executive branch that can adjudicate state law in the first place, Congress must have the power to create one. And because there’s no “Tribunals Clause” in the Constitution, Congress must rely, in part, on the Necessary and Proper Clause. However, this clause, while the fount of federal power in many respects, also limits the federal government in certain ways. As Professors Gary Lawson and Patricia Granger have explained, federal laws are not “proper” if they “usurp or expand the constitutional powers of any federal institutions or infringe on the retained rights of the states or of individuals.”9×9. Gary Lawson & Patricia B. Granger, The “Proper” Scope of Federal Power: A Jurisdictional Interpretation of the Sweeping Clause, 43 Duke L.J. 267, 271 (1993).

This sets up the Note’s two-step doctrinal claim. First, the history and structure of the Constitution reveal a tacit limitation on the federal government’s ability to adjudicate state law. Second, this limit manifests in the Necessary and Proper Clause, which bounds congressional authority along both the horizontal and vertical separation of powers. In short, executive adjudication of state law is unconstitutional because Congress lacks the power to create such tribunals, independent of whether doing so also crosses some functionalist line within Article III.

Part I lays the doctrinal background. It first traces how executive non–Article III tribunals gained the ability to adjudicate state law claims, and then it turns to two relatively recent examples — the anticommandeering doctrine and state sovereign immunity — where the Court has relied on the vertical separation of powers to enforce an implicit constitutional limit on federal authority. Part II turns to the history. The takeaway here is that Article III was the product of compromise; in exchange for a network of federal courts, the Federalists agreed to strictly limit the ability of the new federal government to act upon state laws and state judiciaries. Part III puts it all together, ultimately making the case that executive adjudication falls outside this bargain. Applying the structural-federalism framework of Part I to the history of Part II, it contends that Congress lacks the ability to create executive non–Article III tribunals that can adjudicate state law.

I. Non–Article III Tribunals and Structural Federalism

A. Non–Article III Tribunals and State Law

Article III has three main features: it vests “[t]he judicial Power of the United States . . . in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish”; describes what kinds of cases the federal courts can hear; and guarantees that federal judges shall have life tenure and salary protections.10×10. U.S. Const. art. III, §§ 1–2. A plain reading of these three items would seem to imply that federal judges must exclusively adjudicate whatever actions fall under Article III jurisdiction.11×11. See David P. Currie, The Constitution in the Supreme Court: The First Hundred Years 1789–1888, at 121 (1985). But virtually nobody has adopted this literal position.12×12. Craig A. Stern, What’s a Constitution Among Friends? — Unbalancing Article III, 146 U. Pa. L. Rev. 1043, 1043 (1998); see also id. at 1043 n.1 (collecting sources).

To make a long story short, the Court has permitted a host of non–Article III tribunals (that is, tribunals with judges who lack Article III’s salary and tenure protections) to adjudicate claims that Article III courts would otherwise hear.13×13. See generally Paul M. Bator, The Constitution as Architecture: Legislative and Administrative Courts Under Article III, 65 Ind. L.J. 233 (1990). State courts, which were always assumed to have concurrent jurisdiction with the federal courts over federal law claims, are the most obvious example.14×14. Martin v. Hunter’s Lessee, 14 U.S. (1 Wheat.) 304, 338–52 (1816). The Court has also held that military courts, because of a similar historical pedigree, pass muster.15×15. Dynes v. Hoover, 61 U.S. (20 How.) 65, 78–79 (1858). And the same is true for courts of state-like quasi sovereigns such as the District of Columbia or the territories.16×16. Palmore v. United States, 411 U.S. 389, 407 (1973); United States v. Coe, 155 U.S. 76, 85 (1894). After that, the picture gets murkier.17×17. See generally William Baude, Adjudication Outside Article III, 133 Harv. L. Rev. (forthcoming Mar. 2020) (manuscript on file with the Harvard Law Review). The quagmire here stems from the fact that Article III exclusively vests the “judicial Power” in the national judiciary, and if the political branches could divert all matters to other tribunals, they’d have the power to circumvent the third branch. Accordingly, when the Court concludes that a non–Article III tribunal has been given too much judicial business, it often holds the tribunal is exercising “the judicial power of the United States” in violation of Article III.18×18. E.g., Stern v. Marshall, 564 U.S. 462, 503 (2011). In light of this, there needs to be some limiting principle in place to satisfy the separation of powers. Identifying this marker, however, is the doctrinal $64,000 question.

The main piece of the puzzle for us to focus on falls under the label of “executive adjudication.”19×19. See Caleb Nelson, Adjudication in the Political Branches, 107 Colum. L. Rev. 559, 609–13 (2007). Professor Caleb Nelson helpfully frames the issue as follows: “Under what circumstances does the Constitution permit adjudicative decisions made by Congress or executive agencies to enjoy the sort of finality that is typically associated with judicial decisions?” Id. at 563. From time to time, Congress has created bodies either within administrative agencies (like the FTC) or otherwise outside of Article III (like bankruptcy courts) that have adjudicated matters that could have gone to the federal courts.20×20. See Bator, supra note 13, at 236–39. These tribunals are different in kind from state, territorial, or D.C. courts because they sit within the Executive, and thus cannot exercise “judicial power.”21×21. See Baude, supra note 17 (manuscript at 25–26) (describing traditional view). Executive adjudication of state law, therefore, is when an executive tribunal interprets state law, decides rights and obligations thereunder, and, in a large number of cases, enters a final judgment binding the parties.22×22. It’s generally the case that legislative courts have the power to enter final judgments while only some agencies, but not all, can do the same without an intervening court. See Martin H. Redish, Legislative Courts, Administrative Agencies, and the Northern Pipeline Decision, 1983 Duke L.J. 197, 214–19. For the point that this distinction makes little difference here, see id.

For a concrete example, consider the facts of Commodity Futures Trading Commission v. Schor.23×23. 478 U.S. 833, 835–38 (1986). Under the Commodity Exchange Act (CEA), traders can bring actions against brokers for failing to follow certain parts of the law.24×24. Id. at 836–37. Congress also said that the Commodity Futures Trading Commission (CFTC), an administrative agency, can decide these claims.25×25. Id. at 836–38. Back in the wild 1980s, Schor (a trader) filed a CEA claim against ContiCommodity (a broker) in the CFTC, and Conti then filed a state law counterclaim against Schor in the CFTC.26×26. Id. at 838–40. Can the agency adjudicate the state law claim, consistent with the text and structure of the Constitution, or must a court decide that part of the action?

Answering this question under current law is not easy. Scholars make sense of how the Court has separated kosher from unkosher modes of executive adjudication a lot like how bubbes prepare gefilte fish — everyone makes it a little differently, and there are some unsavory parts, but it all ends roughly the same.27×27. See John Harrison, The Relation Between Limitations on and Requirements of Article III Adjudication, 95 Calif. L. Rev. 1367, 1378 (2007). The historic distinction is that federal non–Article III tribunals can only adjudicate “public rights,” but not “private rights.”28×28. For a more thorough treatment of this, see Nelson, supra note 19. Of course, there are other ways to think about the non–Article III problem. E.g., Baude, supra note 17. The “core” private rights are the rights to life, liberty, and property.29×29. Nelson, supra note 19, at 567. Public rights “belong to the body politic” and cover interests like title to public land or use of public roads.30×30. Ann Woolhandler & Caleb Nelson, Does History Defeat Standing Doctrine?, 102 Mich. L. Rev. 689, 693 (2004); see also Nelson, supra note 19, at 566. Conceptually, the difference between the two types of rights turns on their source; public rights involve some form of government action, while private rights trace from the “unalienable rights of each individual.”31×31. Lansing v. Smith, 4 Wend. 9, 21 (N.Y. 1829) (Walworth, C.). Traditionally, political branch tribunals could just decide claims concerning public rights because private rights could be affected only by an exercise of judicial power, which these tribunals aren’t capable of wielding.32×32. Thomas M. Cooley, A Treatise on the Constitutional Limitations Which Rest upon the Legislative Power of the States of the American Union 92 (Boston, Little, Brown & Co. 1868); Nelson, supra note 19, at 563 (noting executive officers “cannot validly” exercise judicial power).

So far, so good: when you are dealing with private rights, you need an exercise of judicial power and that can’t come by way of executive adjudication.33×33. There is an exception to this general rule for military courts. Nelson, supra note 19, at 576. Indeed, virtually all Founding-era private rights were the product of state law.34×34. See Henry M. Hart, Jr., The Relations Between State and Federal Law, 54 Colum. L. Rev. 489, 491–92 (1954) (explaining state law covered “gamut of problems cast up” in everyday life). And, consistent with this Note’s ultimate claim, it was the uniform practice for the first hundred years of the Republic that courts would exclusively adjudicate these kinds of cases.35×35. Roscoe Pound, Executive Justice, 55 Am. L. Reg. 137, 140 (1907).

For better or worse, however, the Court has gradually permitted a whittling away of this line, allowing more and more executive tribunals to settle these sorts of private rights cases. To put it bluntly, it’s not entirely unfair to say that the Court has guarded the “judicial power” the same way that Monty Python’s Black Knight guarded his bridge, losing limb after limb but insisting that “it’s just a flesh wound” all the while.36×36. See generally Adrian Vermeule, Law’s Abnegation (2016). And, to continue this analogy just a little bit more, one of those main limbs was private rights actions arising under state law.37×37. For sure, states can create public rights too. But the logic of the “public rights” exception holds that the sovereign that created the law has some say in how and where it is adjudicated. See Crowell v. Benson, 285 U.S. 22, 85 (1932) (Brandeis, J., dissenting). So the federal government wouldn’t be able to rely on this exception to divert public rights state law claims. In all events, this distinction doesn’t matter too much for the core thesis laid out in Part III, which turns instead on the source of the law and the type of adjudicative forum. See infra section III.A, pp. 1436–40.

The landmark case of Crowell v. Benson38×38. 285 U.S. 22. is a useful starting point. There, an employee (Knudsen) brought a federal workers’ compensation claim against his employer (Benson) before an executive agency, the United States Employees’ Compensation Commission (ECC).39×39. Id. at 36–37. This was clearly a private rights claim: it concerned private party liability and, if Knudsen won, the deprivation of Benson’s money.40×40. Id. at 38–39. The Court nonetheless held that the ECC scheme did not violate the Constitution even though the case involved executive adjudication of federally created private rights.41×41. Id. at 54. This because the ECC was tasked only with fact-finding, and courts retained the ability to decide all questions of law.42×42. Id.

In the near-century after Crowell, the Court has split into two schools of thought about how much further this sort of executive adjudication could go when it comes to cases arising under state law.

The formalist approach, best captured by Justice Brennan’s plurality opinion in Northern Pipeline Construction Co. v. Marathon Pipe Line Co.,43×43. 458 U.S. 50 (1982). keeps intact the traditional bar on executive adjudication of state law claims.44×44. Id. at 71–72 (plurality opinion). There, the Court asked whether a bankruptcy court (a legislative court) could “constitutionally be vested with jurisdiction to decide [a] state-law contract claim” made as a counterclaim against a party not otherwise part of the bankruptcy proceedings.45×45. Id. at 87 n.40. Tribunals like bankruptcy courts are often called “legislative courts,” but they nonetheless are modes of executive adjudication. Akhil Reed Amar, Marbury, Section 13, and the Original Jurisdiction of the Supreme Court, 56 U. Chi. L. Rev. 443, 451 n.43 (1989). There was no majority opinion. But Justice Brennan’s plurality held that the bankruptcy court could not have jurisdiction because “[p]rivate-rights disputes . . . lie at the core of the historically recognized judicial power.”46×46. Northern Pipeline, 458 U.S. at 70; see id. at 90 (Rehnquist, J., concurring in the judgment). The plurality distinguished Crowell on the ground that there are really two types of private rights: those created by Congress and those created by some other source of law (namely, state statutory or common law).47×47. Id. at 83 (plurality opinion). With the second category, there is a historically grounded prohibition on adjudication by a federal non–Article III tribunal, no matter how much judicial review exists on the back end.48×48. Id. at 69 n.23. At bottom, this class of cases is given to Article III courts by the Constitution, and adjudication by another branch would violate the horizontal separation of powers.49×49. Id. at 69–70; see also Stern v. Marshall, 564 U.S. 462, 504–05 (2011) (Scalia, J., concurring).

The leading case for the functionalist view is Schor, authored by Justice O’Connor, which held that executive agencies could constitutionally adjudicate state law claims.50×50. Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 853 (1986); see also Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 568, 584, 594 (1985). As noted, a trader (Schor) brought a CEA claim against a broker (ContiCommodity) before the CFTC (an executive branch agency),51×51. Schor, 478 U.S. at 837–38. and Conti brought a state contract law counterclaim against Schor as part of the same action.52×52. Id. at 838; see also id. at 848–49 (discussing party consent). The Schor Court noted that this counterclaim involved “a ‘private’ right for which state law provides the rule of decision” and was thus “of the kind assumed to be at the ‘core’ of matters normally reserved to Article III courts.”53×53. Id. at 853. But, departing from Justice Brennan, Justice O’Connor held that the CFTC could decide the case because (i) the claim was bound up as part of a single dispute, (ii) the CFTC’s jurisdiction related only to a narrow class of claims that implicated its expertise, and (iii) some form of judicial review existed on the back to enforce the order.54×54. Id. at 855–56. On this view, “a given congressional delegation of adjudicative functions to a non–Article III body must be assessed by reference to the purposes underlying the requirements of Article III,” and executive adjudication of Conti’s small counterclaim didn’t cross this practical Article III line.55×55. Id. at 847; see also id. at 851 (detailing multifactor balancing test).

So where do things stand? Well, Schor stands for the proposition that the Constitution does not contain a per se bar on executive adjudication of state law claims. At the same time, there seems to be a consensus that Congress cannot create a federal non–Article III tribunal to hear only state law cases.56×56. See, e.g., Richard H. Fallon, Jr., Jurisdiction-Stripping Reconsidered, 96 Va. L. Rev. 1043, 1124 (2010); Harrison, supra note 27, at 1383; Mila Sohoni, Agency Adjudication and Judicial Nondelegation: An Article III Canon, 107 Nw. U. L. Rev. 1569, 1583 (2013). On these points, though, the Court hasn’t offered a clear way of figuring out the space in between. The best thing we have is Chief Justice Roberts’s attempt to distill present case law in Stern v. Marshall,57×57. 564 U.S. 462 (2011). where he seemed to cabin Schor’s adjudicatory blessing to cases where the state law claim either (i) “flows from a federal statutory scheme,” or (ii) is “‘completely dependent upon’ adjudication of a claim created by federal law,” and in any event, (iii) is decided by a tribunal with delineated rather than general jurisdiction.58×58. Id. at 493–94 (quoting Schor, 478 U.S. at 856).

Simply put, the Stern Court explained that there must be some federal nexus for a federal non–Article III tribunal to adjudicate claims arising under state law — exactly why this is so and, moreover, what constitutes a sufficient nexus are largely still questions for another day.59×59. Despite Justice O’Connor’s brief downplaying of the rights framework in Schor, it remains the dominant mode of analysis for evaluating political branch adjudication. See, e.g., Oil States Energy Servs., LLC v. Greene’s Energy Grp., LLC, 138 S. Ct. 1365, 1373 (2018); Stern, 564 U.S. at 494; Granfinanciera, S.A. v. Nordberg, 492 U.S. 33, 54–55 (1989); see also Nelson, supra note 19, at 565.

B. Structural Federalism as a Bar on Federal Power

The above cases involve history, structural reasoning, and the separation of powers. The below cases concern the same, but in the vertical separation of powers context. While they don’t immediately concern executive adjudication of state law, they’re key to this Note for a couple of reasons. First, these “structural federalism” cases establish the framework that this Note takes up: they use history and structure to identify a limit on federal power, and then enforce that limit as a term of the original constitutional deal. Second, they indicate the sources that matter for identifying these tacit terms: namely, English tradition, the Constitutional Convention, state ratifying conventions, and early practice.60×60. To be sure, a small army of scholars have mounted an attack against this approach. See, e.g., John F. Manning, The Supreme Court, 2013 TermForeword: The Means of Constitutional Power, 128 Harv. L. Rev. 1, 4 (2014). Fully engaging with this view is beyond the scope of this Note. For our purposes, it’s enough to say that today’s Supreme Court has not betrayed an affinity for this position. See, e.g., Franchise Tax Bd. v. Hyatt, 139 S. Ct. 1485, 1494–95 (2019).

1. Anticommandeering. — The anticommandeering doctrine holds that the federal government cannot force states or state officials to implement federal legislation. This might seem curious once we take stock of the federal government’s wide-reaching ability to affect state policy through Spending Clause legislation or preemption.61×61. See South Dakota v. Dole, 483 U.S. 203, 212 (1987) (Spending Clause); Hodel v. Va. Surface Mining & Reclamation Ass’n, 452 U.S. 264, 290 (1981) (preemption). But in Printz v. United States,62×62. 521 U.S. 898 (1997). for example, the Court held that the Brady Act — which tried to force state law enforcement to carry out part of the gun law’s background check provisions — was unconstitutional.63×63. Id. at 902–04, 933. In particular, the Court held that the law exceeded Congress’s authority because the federal government has the power to bind only individuals and, in turn, lacks the power to directly bind states (or their agents).64×64. Id. at 918–20. Notably, in both Printz and like cases, the Court has eschewed any reliance on the Constitution’s text to support this axiom.65×65. Id. at 905 (justifying focus on history and structure because of lack of on-point text). The doctrine instead principally flows from history and structural reasoning.

First, the historical argument. “Under the Articles of Confederation, Congress lacked the authority in most respects to govern the people directly” and had to instead rely on states as intermediaries to implement federal policy.66×66. New York v. United States, 505 U.S. 144, 163 (1992). This was quite inefficient, and the Framers sought to remedy this defect in the new national government.67×67. See The Federalist No. 15, supra note 1, at 105 (Alexander Hamilton). The Constitutional Convention thus “opted for a Constitution in which Congress would exercise its legislative authority directly over individuals rather than over States.”68×68. New York, 505 U.S. at 165. We know this because the Convention decisively rejected the New Jersey Plan, which would’ve replicated the Confederation’s state-dependent structure, in favor of the Virginia Plan.69×69. Id. And, according to contemporary records, the Framers saw the Virginia Plan as marking the creation of an entirely new form of government.70×70. Id. at 165–66 (collecting statements by Framers).

Second, the structural argument. As noted, the Constitution created a system of “dual sovereignty”: the federal government was entrusted with a certain set of powers, and the states retained everything else as residual sovereigns.71×71. Printz v. United States, 521 U.S. 898, 918–19 (1997) (quoting Gregory v. Ashcroft, 501 U.S. 452, 457 (1991)); see The Federalist No. 39, supra note 1, at 241–42 (James Madison). The object of this system was to create a “double security” for individual liberty, whereby the “different governments [would] control each other, at the same time that each [would] be controlled by itself.”72×72. The Federalist No. 51, supra note 1, at 320 (James Madison). The Court has held that this commitment to dual sovereignty can’t be squared with the view that the federal government may commandeer state governments in service of federal ends.73×73. Printz, 521 U.S. at 922–23. If the federal government could utilize its own powers and co-opt the powers of the states, it could aggrandize all authority within the federal sphere.74×74. Id. at 919–22 (“The power of the Federal Government would be augmented immeasurably if it were able to impress into its service — and at no cost to itself — the police officers of the 50 States.” Id. at 922.).

The Court has accordingly said that the history of the Constitution, confirmed by the document’s structure, holds that the federal government lacks the power to force the states to implement federal law. For sure, there’s no part of Article I that says the “foregoing powers only apply to individuals.” But it would also be unfaithful to the original deal struck between the states and the federal government to read each instance of constitutional silence as an invitation for expanded federal power. Therefore, in order to maintain the vertical separation of powers, the Court must look to structure and history. And it did so here.75×75. Of course, not everyone is sold on this. See, e.g., Vicki C. Jackson, Federalism and the Uses and Limits of Law: Printz and Principle?, 111 Harv. L. Rev. 2180 (1998). Most importantly, once having identified this limitation, the Court held that Congress could not transgress it through the Necessary and Proper Clause because doing so would not be “proper” since it would “violate[] [a] principle of state sovereignty” enshrined in the constitutional deal.76×76. Printz, 521 U.S. at 923–24.

2. State Sovereign Immunity. — Being sovereign generally means never having to say you’re sorry. Indeed, the doctrine of sovereign immunity, which says that a sovereign can’t be sued unless it has consented to suit, flows from the English maxim that “the King can do no wrong.”77×77. See Erwin Chemerinsky, Against Sovereign Immunity, 53 Stan. L. Rev. 1201, 1201 (2001); see also United States v. Lee, 106 U.S. 196, 205 (1882) (detailing English tradition). The Framers retained this common law concept (regal infallibility aside), and the Court has thus held that the federal government and states enjoy a baseline of immunity from suit as sovereigns.78×78. Fed. Mar. Comm’n v. S.C. State Ports Auth., 535 U.S. 743, 751–52 (2002).

As with anticommandeering, the Court has relied on history and structure in developing this doctrine. Unlike anticommandeering, however, the Constitution does have express text addressing the subject. The Eleventh Amendment, read naturally, bars federal courts from hearing any suit brought against a state by an out-of-stater or a foreign citizen.79×79. See U.S. Const. amend. XI. And, under ordinary interpretive rules, when a law provides for something in some cases, it usually means that it doesn’t provide for that thing in other cases.80×80. For the leading argument that courts should follow the literal text, see John F. Manning, The Eleventh Amendment and the Reading of Precise Constitutional Texts, 113 Yale L.J. 1663 (2004). But, underscoring the import of history and structure to constitutional reasoning, the Court has rejected this view because it doesn’t fully capture the original deal that the states agreed to.

The state sovereign immunity story really begins with the case of Chisholm v. Georgia,81×81. 2 U.S. (2 Dall.) 419 (1793). decided just four years after the Constitution was ratified.82×82. Id. There, the Chisholm Court held that the Constitution uprooted whatever sovereign immunity states enjoyed before the Union and, in particular, that a citizen of one state could sue another state in federal court.83×83. Id. at 450, 466, 467, 477. By all accounts, the case rocked the nation.84×84. 1 Charles Warren, The Supreme Court in United States History 96 (1922). And the young country soon responded by ratifying the Eleventh Amendment.85×85. Edelman v. Jordan, 415 U.S. 651, 662 (1974).

The question then became whether the amendment simply overturned Chisholm or marked a broader view of state sovereign immunity. Starting with Hans v. Louisiana,86×86. 134 U.S. 1 (1890). There, the Court held that a citizen of one state was barred from suing that state without its consent. Id. at 16–18. the Court took the latter view. In so doing, the Court has held that a state’s constitutionally protected immunity is broader than the Constitution’s text.87×87. Manning, supra note 80, at 1667. And in support of this view, the Court has looked to the drafting and ratification of the Constitution. For instance, during state ratification debates, Antifederalists charged that the proposed Constitution “implicitly waived the States’ sovereign immunity against private suits in federal courts.”88×88. Franchise Tax Bd. v. Hyatt, 139 S. Ct. 1485, 1495 (2019) (emphasis omitted). But leading Federalists forcefully denied this.89×89. Alden v. Maine, 527 U.S. 706, 715–18 (1999) (collecting quotes). For example, James Madison pledged at the Virginia Convention that the Constitution did not change the baseline of sovereign immunity states enjoyed following independence, and it would “not [be] in the power of individuals to call any state into court.”90×90. 3 The Debates in the Several State Conventions on the Adoption of the Federal Constitution 533 (Jonathan Elliot ed., Washington, D.C., 2d ed. 1836) (James Madison) [hereinafter Elliot’s Debates]; see also 2 The Records of the Federal Convention of 1787, at 423–25, 434–35 (Max Farrand ed., 1911) [hereinafter Farrand’s Records]. At bottom, states voted for the Constitution in reliance on these promises.91×91. Alden, 527 U.S. at 718–19 (documenting examples). As such, the argument goes, when Congress passed the Eleventh Amendment later on, it “acted not to change but to restore the original constitutional design” as seen at ratification.92×92. Id. at 722.

In service of this “original constitutional design,” the Court has repeatedly held that a range of suits not contemplated by the text of the Eleventh Amendment are barred by a state’s sovereign immunity.93×93. E.g., Hyatt, 139 S. Ct. at 1492 (different state’s citizens in other state’s courts); Fed. Mar. Comm’n v. S.C. State Ports Auth., 535 U.S. 743, 760–61 (2002) (private party in federal agency); Alden, 527 U.S. at 754 (state’s own citizen bringing federal claim in that state’s court); Monaco v. Mississippi, 292 U.S. 313, 331–32 (1934) (foreign nation in federal court); Smith v. Reeves, 178 U.S. 436, 448–49 (1900) (federal corporation in federal court). In each case, the Court has made clear that its holding was pegged to what was “implicit in [the Constitution’s] structure and supported by historical practice.”94×94. Hyatt, 139 S. Ct. at 1498; see also Alden, 527 U.S. at 713. Moreover, in parallel with expanding the baseline of states’ immunity, the Court has also held that the federal government largely lacks the power to abrogate state sovereign immunity. In fact, the Court has made clear that Congress lacks any power to abrogate state immunity, except narrowly under the Bankruptcy Clause or the Fourteenth Amendment’s Enforcement Clause due to their unique histories.95×95. Seminole Tribe v. Florida, 517 U.S. 44, 65 (1996); see also Cent. Va. Cmty. Coll. v. Katz, 546 U.S. 356, 378–79 (2006); Fitzpatrick v. Bitzer, 427 U.S. 445, 456 (1976). As above, the Court has relied on history and structure, pointing to the original understanding of the constitutional design and the structural intuition that “Article I cannot be used to circumvent the constitutional limitations placed upon federal jurisdiction.”96×96. Seminole Tribe, 517 U.S. at 73; see id. at 72–73; see also Alden, 527 U.S. at 732. For an excellent defense of the import of history and structure in constitutional reasoning and, in particular, the notion of the Constitution as a bargain, see Steven Menashi, Article III as a Constitutional Compromise: Modern Textualism and State Sovereign Immunity, 84 Notre Dame L. Rev. 1135 (2009). To that end, the Court has held that the Necessary and Proper Clause accounts for these original bounds and accordingly limits congressional power.97×97. Alden, 527 U.S. at 732.

Where does this leave us? With respect to the horizontal and vertical separation of powers, history and structure matter. On that front, when the Constitution’s background and design reveal a limitation on federal power consistent with the original deal struck by the people and their representatives, it’s incumbent on courts to enforce that limit, even if unenumerated. That is just sticking to the deal; no more, no less.

II. The Deal as to Adjudicating State Law

The creation of the federal judiciary, like that of the Constitution as a whole, was the product of compromise between the representatives of the people and their respective sovereigns (the original thirteen states).98×98. See Jack N. Rakove, Original Meanings 15 (1996); Lawrence C. Marshall, Commentary, Fighting the Words of the Eleventh Amendment, 102 Harv. L. Rev. 1342, 1353 (1989); see also Manning, supra note 6, at 2040–47; Menashi, supra note 96, at 1151–55. And who got to adjudicate state law claims when was a big part of that. As we’ll see below, the general bargain here was courts-for-jurisdiction. As to state law, the Founders devised a chicken-or-fish system: The vast majority of state law claims would remain in state courts and, in carefully defined cases, litigants could choose the federal courts. And as we’ll see in Part III, executive adjudication is neither chicken nor fish, and does not square with the hard-fought terms of this compromise.

Part II has four pieces. It first describes the context of the Constitutional Convention, focusing on colonial courts and state courts under the Articles of Confederation. The next few sections look at the deal struck between Federalists and Antifederalists regarding the federal judiciary, starting with the creation of the lower federal courts and then the narrowing of their jurisdiction to protect state interests and law.

A. The Need for an Independent Federal Judiciary

The Constitution was drafted against a backdrop of the Founders’ experience with political interference with judicial independence and impartial adjudication. This was the modus operandi of the British Crown and, to a lesser degree, state legislatures under the Articles of Confederation. These experiences inform not only the creation of the federal courts, but also the view that the original constitutional plan didn’t contemplate nonjudicial adjudication of state law claims.

1. Pre-Revolution America. — For much of England’s history, judges occupied a precarious position. On the one hand, judges served at the pleasure of the Crown.99×99. See Joseph H. Smith, An Independent Judiciary: The Colonial Background, 124 U. Pa. L. Rev. 1104, 1105–10 (1976) (describing status of common law judges). On the other hand, the Crown was loath to violate crystallizing norms of judicial independence.100×100. Compare 5 William Holdsworth, A History of English Law 346 (1927) (describing judicial independence under Elizabeth I), with id. at 352 (describing lack thereof with Stuarts).

By the time colonists touched down in Jamestown, however, the Crown was waging a sustained campaign against judicial independence. For one, the Tudors and Stuarts increasingly made use of a system of tribunals outside of the regular common law courts and within the Executive.101×101. Philip Hamburger, Is Administrative Law Unlawful? 134–35 (2014). The Star Chamber was the most prominent — a prerogative court of general jurisdiction that sat within the King’s Privy Council.102×102. See id. at 40–42; James T. Barry III, The Council of Revision and the Limits of Judicial Power, 56 U. Chi. L. Rev. 235, 238 (1989). Unlike ordinary common law courts, the Chamber “existed to defend the crown’s actions under the royal prerogative” and, critically, because it “existed solely by the King’s authority, the common perception was that no method existed by which to challenge the King’s actions.”103×103. Archibald Cox, The Independence of the Judiciary: History and Purposes, 21 U. Dayton L. Rev. 565, 568 n.3 (1996). As you might expect, this brand of executive branch adjudication quickly became synonymous across the empire with political oppression.104×104. See Hamburger, supra note 101, at 171–72.

Monarchs also sought to influence the judiciary through the specter of removal. Especially during the Stuart period, the Crown fired judges who ruled against it.105×105. Cox, supra note 103, at 569 & n.10. And other judges got the cue. As such, they “tended to become identified with the party and policy of the king” and became his “civil servants,” not “independent expositors of the law.”106×106. 5 Holdsworth, supra note 100, at 352.

The British were not fans of this system and eventually took up measures to ensure an independent judiciary. First, ahead of the English Civil War, the Parliament abolished the Star Chamber and all other forms of prerogative adjudication.107×107. Hamburger, supra note 101, at 138–39. The Privy Council (also known as the King’s Council) accordingly lost its power to adjudicate domestic matters outside of common law courts.108×108. James E. Pfander & Daniel D. Birk, Article III and the Scottish Judiciary, 124 Harv. L. Rev. 1613, 1652 (2011). Second, the Act of Settlement of 1701 secured that judges could be removed only for cause and by the Parliament (not the King).109×109. Todd D. Peterson, Restoring Structural Checks on Judicial Power in the Era of Managerial Judging, 29 U.C. Davis L. Rev. 41, 47–48 (1995). By the eighteenth century, “the complete independence of the bench was therefore permanently established.”110×110. Theodore F.T. Plucknett, A Concise History of the Common Law 61 (Little, Brown & Co. 1956) (1929).

The same could not be said for the colonies. In fact, neither of the reforms just listed carried beyond Britain’s shores. The Privy Council retained judicial and legislative authority over the colonies, meaning that it could “disapprove of statutes by veto and could also invalidate a colonial statute . . . [by] deciding a case.”111×111. John P. Frank, Historical Bases of the Federal Judicial System, 13 Law & Contemp. Probs. 3, 5 (1948); see also Pfander & Birk, supra note 108, at 1652 (“Privy Council review entailed an element of hierarchy; it proceeded on the assumption that the Council had the final word on the matter under review and that the colonial courts were duty-bound to carry its decrees into effect.”). And “[t]he Act of Settlement did not apply to the colonies,” so colonial judges still “served at the pleasure of the Crown” and, at times, had their salaries turn on royal whim.112×112. David P. Currie, Separating Judicial Power, 61 Law & Contemp. Probs. 7, 9 (1998); see also Smith, supra note 99, at 1114–17 (discussing removal of New York Judge Lewis Morris).

Colonial judges, in short, were seen as “appendages or extensions of royal authority embodied in the governors, or chief magistrates.”113×113. Gordon S. Wood, The Origins of Judicial Review Revisited, or How the Marshall Court Made More out of Less, 56 Wash. & Lee L. Rev. 787, 790 (1999). This judicial regime was anathema to most Americans. Practically, judges were seen as midwives for British policy, not impartial adjudicators of law.114×114. See John Ferejohn, Independent Judges, Dependent Judiciary: Explaining Judicial Independence, 72 S. Cal. L. Rev. 353, 378 (1999). A big reason for this was that, because colonial governments lacked bureaucracies, colonial courts had a lot of nonjudicial functions.115×115. William E. Nelson, The Eighteenth-Century Background of John Marshall’s Constitutional Jurisprudence, 76 Mich. L. Rev. 893, 903 (1978) (observing courts “collected taxes, supervised the construction and maintenance of highways, issued licenses, and regulated licensees’ businesses”). Philosophically, Americans also saw the British regime as fatally flawed. Indeed, the British judiciary was taken as part of the executive branch.116×116. The Earl of Clarendon to William Pym, No. III, Bos. Gazette, Jan. 27, 1766, reprinted in 3 The Works of John Adams 477, 480–82 (Charles Francis Adams ed., Boston, Charles C. Little & James Brown 1851); Jack N. Rakove, The Original Justifications for Judicial Independence, 95 Geo. L.J. 1061, 1063 (2007). To many an American mind, this was the surefire recipe for tyranny. As Montesquieu, who informed the Founders’ thinking, put it: “[T]here is no liberty, if the judiciary power not be separate from the legislative and executive” since “[w]ere it joined to the executive power, the judge might behave with violence and oppression.”117×117. 1 Montesquieu, The Spirit of Laws 152 (Thomas Nugent trans., rev. ed. 1900); see also The Federalist No. 78, supra note 1, at 465 (Alexander Hamilton) (endorsing this proposition).

This two-tiered system — an independent judiciary for Britain and a dependent one for America — was a key driver of the Revolution.118×118. Wood, supra note 113, at 790–91; see also 18 A Report of the Record Commissioners of the City of Boston, Boston Town Records, 1770–1777, at 102 (Boston, Rockwell & Churchill 1887) (“How alarming must it then be to the Inhabitants of this Province, to find so wide a difference made between the Subjects in Britain and America, as the rendering the Judges here altogether dependent on the Crown for their support.”). In fact, the Declaration of Independence listed as a grievance against King George that he “made judges dependent on his will alone, for the tenure of their offices, and the amount and payment of their salaries.”119×119. The Declaration of Independence para. 11 (U.S. 1776).

2. Revolutionary America. — It should be of little surprise then that most of the original states moved to create independent judiciaries of their own once they declared independence. Eleven of the thirteen states gave judges tenure protections.120×120. Joseph R. Grodin, Developing a Consensus of Constraint: A Judge’s Perspective on Judicial Retention Elections, 61 S. Cal. L. Rev. 1969, 1970 n.5 (1988) (describing appointments). Many of the states also guaranteed salary levels for state judges.121×121. See Smith, supra note 99, at 1153–56. Generally, state judges enjoyed a level of job security similar to what Article III judges enjoy today.122×122. Brian T. Fitzpatrick, The Constitutionality of Federal Jurisdiction-Stripping Legislation and the History of State Judicial Selection and Tenure, 98 Va. L. Rev. 839, 878 (2012) (“The vast majority of state judges enjoyed the exact same tenure as federal judges. Indeed, this was true not only of tenure, but other structural protections as well, including selection by appointment rather than election and protection against salary diminution.”).

The original state constitutions had two features that would come to shape the federal judiciary later on. First, they manifested a clear concern about political involvement in adjudication: namely, executive interference with the judiciary.123×123. Currie, supra note 112, at 9–10; Smith, supra note 99, at 1153–56. Second, they reflected a sharp break from the British model of the separation of powers.124×124. Rakove, supra note 116, at 1064–66. In fact, many states expressly provided for articles that required the separation of powers.125×125. Id. at 1064 (observing the “constitutionalists of 1776 were avowed Montesquieans”).

Ahead of the Constitutional Convention, the Founders had the benefit of seeing these theories tested in practice. While state constitutions mostly concerned executive interference with the judiciary, the early experience of the states mainly involved legislative pressure. Legislatures in the British model were generally checks on unilateral executive lawmaking rather than independent lawmakers — and while that began to change in eighteenth-century Parliament, the colonies had not yet caught up.126×126. Id. at 1066. The new state legislatures therefore had to start passing lots of laws. But quantity outstripped quality.127×127. Wood, supra note 113, at 791. As one author put it: “[E]very new law . . . acts as rubbish, under which we bury the former.”128×128. Rudiments of Law and Government, Deduced From the Law of Nature 35 (Charlestown, John M’Iver 1783); see Wood, supra note 113, at 791.

This flurry of legislative activity, coupled with state constitutions that included weak executives, created a concern among Americans that legislatures may in practice pose the greatest threat to individual liberty.129×129. The Federalist No. 48, supra note 1, at 306 (James Madison) (describing “impetuous vortex” of legislatures); see also Gordon S. Wood, The Creation of the American Republic, 1776–1787, at 409 (2d ed. 1998). The sheer mass of legislation raised the concern that legislatures would soon dominate the government by way of paper blizzard, creating impenetrable statutory schemes that placed citizens under the specter of indefinite liability.130×130. See Irving R. Kaufman, The Essence of Judicial Independence, 80 Colum. L. Rev. 671, 683–87 (1980); see also The Federalist No. 10, supra note 1, at 79 (James Madison). This led, as Professor Gordon Wood explains, “more and more Americans [to look] to the once-feared judiciary as a principal means of restraining these wild and rampaging popular legislatures.”131×131. Gordon S. Wood, The Origins of Judicial Review, 22 Suffolk U. L. Rev. 1293, 1304 (1988). For sure, in states where judicial protections were weaker, state legislatures interfered with some decisions.132×132. The most notable example of this came out of Connecticut, where the state legislature had the power to overturn court decisions. See Plaut v. Spendthrift Farm, Inc., 514 U.S. 211, 223 (1995). But for most cases, citizens relied on courts to bring order to legislative schemes and, moreover, the Founders began to appreciate the practical import of an independent judiciary to balance the aggression of both political branches.133×133. Ferejohn, supra note 114, at 378; Letter from William Plumer to William Coleman (May 31, 1786), in 11 Publications of the Colonial Society of Massachusetts 383, 384 (1910).

These intuitions regarding judicial independence were also confirmed by the Founders’ experience with political branch adjudication under the Articles of Confederation. The national judiciary under the Confederation was ad hoc and ineffective.134×134. See Henry J. Bourguignon, The Federal Key to the Judiciary Act of 1789, 46 S.C. L. Rev. 647, 650–54 (1995) (“To assure, however, that the new national government would not become as ineffective as that under the Articles of Confederation, the founding fathers determined that the federal Constitution and laws, as well as treaties, would be the supreme law of the land.” Id. at 650.). The Articles provided for only three narrow areas of national jurisdiction — piracy, admiralty, and state border disputes — but there were no federal trial courts.135×135. Frank, supra note 111, at 8; see also Michael G. Collins, The Federal Courts, the First Congress, and the Non-Settlement of 1789, 91 Va. L. Rev. 1515, 1525–26 (2005). Accordingly, the Confederation needed to rely on state trial courts, and appeals were generally handled, at first, by tribunals within the Continental Congress itself.136×136. See Bourguignon, supra note 134, at 652 n.20. In short, this Founding-era system of political branch adjudication conspicuously went uniformly poorly.137×137. See Deirdre Mask & Paul MacMahon, The Revolutionary War Prize Cases and the Origins of Diversity Jurisdiction, 63 Buff. L. Rev. 477, 490–94 (2015) (describing Committee on Appeals for prize cases); William F. Swindler, Seedtime of an American Judiciary: From Independence to the Constitution, 17 Wm. & Mary L. Rev. 503, 514–17 (1976) (same for boundary disputes).

How does all this relate to executive adjudication of state law? The takeaway so far is that the Founders believed all adjudication wasn’t created equal and, as they learned from the English model and early state practice, political branch adjudication is markedly distinct from the impartial exercise of judicial power by established courts. For that reason, as Professors Nathan Chapman and Michael McConnell have explained, the Constitution’s guarantee of due process was originally taken to “ensure[] that the executive would not be able unilaterally to deprive persons within the nation of their rights of life, liberty, or property except as provided by common law or statute and as adjudicated by independent judicial bodies.”138×138. Nathan S. Chapman & Michael W. McConnell, Due Process as Separation of Powers, 121 Yale L.J. 1672, 1807 (2012) (emphasis added); see also id. at 1801–04. Put plainly, sending a case to a judge was fundamentally distinct from sending it to a political branch.

B. The First Half of the Deal: The Lower Federal Courts

This section describes what the Federalists won in the constitutional bargain: the creation of a robust network of lower federal courts. The next sections will get into what they gave up for it: exacting limitations on the federal government’s ability to affect state law and state courts.

Following their experience with executive overreach into the colonial courts and the import of state judges counteracting overzealous state legislatures, the Founders entered the Constitutional Convention committed to making an independent national judiciary on equal footing with the other branches of the federal government.139×139. See The Federalist No. 78, supra note 1, at 463 (Alexander Hamilton); see also Akhil Reed Amar, A Neo-Federalist View of Article III: Separating the Two Tiers of Federal Jurisdiction, 65 B.U. L. Rev. 205, 225–28 (1985). In light of this, there were two areas of consensus among the delegates with respect to the federal judiciary. First, there needed to be a supreme national tribunal of some form.140×140. Robert N. Clinton, A Mandatory View of Federal Court Jurisdiction: A Guided Quest for the Original Understanding of Article III, 132 U. Pa. L. Rev. 741, 761 (1984). Second, federal judges should have structural protections to guarantee their independence: namely, they should receive life tenure subject to good behavior and their salaries must not be diminished by the political branches.141×141. See Lawrence Gene Sager, The Supreme Court, 1980 TermForeword: Constitutional Limitations on Congress’ Authority to Regulate the Jurisdiction of the Federal Courts, 95 Harv. L. Rev. 17, 61 (1981) (describing origins of these protections). Beyond this, the consensus frayed.

Three states submitted plans of government during the Convention and each provided for a federal judiciary in a different manner.142×142. Sidney Teiser, The Genesis of the Supreme Court, 25 Va. L. Rev. 398, 416–19 (1939). As you might recall from the state sovereign immunity discussion, the Virginia Plan won out over the New Jersey Plan (and also the South Carolina Plan).143×143. Clinton, supra note 140, at 762–63. Compared to the other options, the Virginia Plan had the broadest conception of a federal judiciary; it required the creation of lower federal courts where the other plans principally gave the national tribunal only appellate jurisdiction over state courts, and it gave the federal courts a much wider scope of constitutionally mandated jurisdiction while the other plans cut much more narrowly on this front.144×144. Compare 1 Farrand’s Records, supra note 90, at 21–22 (Virginia Plan), with 3 id. at 611–15 (New Jersey Plan), with 2 id. at 134–37 (South Carolina Plan).

But the views that animated the states’ rights–oriented New Jersey Plan quickly manifested in amendments to the Virginia Plan.145×145. See Wythe Holt, “To Establish Justice”: Politics, the Judiciary Act of 1789, and the Invention of the Federal Courts, 1989 Duke L.J. 1421, 1462 (“Yet there was by no means a consensus on the deficiencies of state courts, and only the Virginia Plan explicitly required lower federal courts other than admiralty courts.”). Shortly, two fault lines formed.146×146. For more on this, see Richard H. Fallon, Jr., et al., Hart and Wechsler’s the Federal Courts and the Federal System 20 n.2 (7th ed. 2015) (collecting sources). One involved federal jurisdiction, which we’ll pick up next.147×147. As a general matter, delegates’ views on this matter varied widely. Compare Bourguignon, supra note 134, at 688 (“Even those opposed to federal district courts in general conceded the need for district courts with admiralty jurisdiction.”), with Clinton, supra note 140, at 791 (describing move to give Congress plenary power over federal courts’ jurisdiction). The other was whether there should be lower federal courts at all, or whether state courts should handle everything instead.

One day after the Convention adopted the Virginia Plan, opponents of the lower federal courts successfully passed a motion to eliminate them.148×148. 1 Farrand’s Records, supra note 90, at 125; Holt, supra note 145, at 1462–63. John Rutledge, who led the charge against inferior federal tribunals, argued that “State Tribunals are most proper to decide in all cases in the first instance” and “the right of appeal to the supreme national tribunal [was] sufficient to secure the national rights & uniformity of Judgmts.”149×149. 1 Farrand’s Records, supra note 90, at 119, 124; see also id. at 125 (Pierce Butler). This outlook prevailed five states to four, with two undecided.150×150. Frank, supra note 111, at 10. Those in favor of a stronger federal government, led by Madison and James Wilson, responded in what is labeled the “Madisonian Compromise”: the agreement that the Constitution would not compel the creation of lower federal courts, but would rather give Congress the power to create them later.151×151. 1 Farrand’s Records, supra note 90, at 127. While some tried to oust lower federal courts permanently, the Compromise passed eight to two.152×152. Id. at 125.

The lower federal court issue was thus left for another day. From there, two schools of thought emerged about the propriety of the Compromise and the existence of lower courts, which played out in the state ratifying conventions and the debates over the Judiciary Act of 1789.153×153. Ch. 20, 1 Stat. 73.

Federalists believed that both practical and principled reasons militated in favor of lower federal courts.154×154. See Clinton, supra note 140, at 810–28. First, a federal judiciary composed solely of a single supreme court would not be an effective counterweight to the other branches.155×155. Bourguignon, supra note 134, at 655–57. Second, such a structure would be hard to administer.156×156. 1 Farrand’s Records, supra note 90, at 124 (James Madison). Third, state courts weren’t suited to handling national issues nor accountable to the national government.157×157. Speech of William Paterson (June 23, 1789), in William R. Casto, The First Congress’s Understanding of Its Authority over the Federal Courts’ Jurisdiction, 26 B.C. L. Rev. 1101 app. B at 1134 (1985) (describing dependence of state courts on state governments). Fourth, state courts couldn’t be trusted to impartially decide federal issues.158×158. E.g., 1 Farrand’s Records, supra note 90, at 203 (Edmund Randolph) (“The Executive & Judiciary of the States, notwithstanding their nominal independence on the State Legislatures are in fact, so dependent on them, that unless they be brought under some tie [to] the Natl. system, they will always lean too much to the State systems, whenever a contest arises between the two.”).

Antifederalists maintained, by contrast, that the risks posed by an expanded federal judiciary far outstripped any purported benefit offered by having lower federal courts.159×159. See Collins, supra note 135, at 1530–33. First, lower federal courts were entirely unnecessary because state courts were more than competent to do the job and any errors of federal law could be fixed by the Supreme Court.160×160. E.g., 3 Elliot’s Debates, supra note 90, at 325 (Patrick Henry) (underscoring state judges’ “firmness to counteract the legislature” and “oppose unconstitutional acts” while raising skepticism as to whether the federal courts would be “as well constructed, and as independent of the other branches, as our state judiciary”); 2 Farrand’s Records, supra note 90, at 45 (Pierce Butler). Second, additional federal courts would be expensive and inconvenient.161×161. E.g., 4 Elliot’s Debates, supra note 90, at 155 (Samuel Spencer) (“It must be unnecessary for the federal courts to do it, and would create trouble and expense which might be avoided.”). Third, lower federal courts would conflict with state courts and invariably supplant them at the expense of state sovereignty.162×162. E.g., 2 Farrand’s Records, supra note 90, at 45–46 (Luther Martin) (“They will create jealousies & oppositions in the State tribunals, with the jurisdiction of which they will interfere.”). The final point was salient, as Antifederalists like George Mason didn’t hide: “The Judiciary of the United States is so constructed and extended, as to absorb and destroy the judiciaries of the several States; thereby rendering law as tedious, intricate and expensive, and justice as unattainable, by a great part of the community . . . .”163×163. Id. at 638; see also Brutus, I, N.Y. J., Oct. 18, 1787, reprinted in The Anti-Federalist 108, 112 (Herbert J. Storing ed., 2d ed. 1985) (“These courts will be . . . totally independent of the states, deriving their authority from the United States, and receiving from them fixed salaries; and in the course of human events it is to be expected, that they will swallow up all the powers of the courts in the respective states.”).

The Federalists prevailed. The ratified Constitution included the Madisonian Compromise, and the Judiciary Act of 1789 created a robust, three-tiered federal court system that consisted of a district court in each state; three circuit courts; and a six-justice supreme court.164×164. Holt, supra note 145, at 1486.

But this is only half the story. As Professor Wythe Holt has laid out, the federal judiciary came out of a bargain where “[s]evere limitations and restrictions were to be placed upon the federal courts’ jurisdiction” in exchange for “a highly articulated three-tiered system of national courts.”165×165. Id. at 1485–86. To be clear, this characterization is not some ex post framing of what happened to take place. As we turn to next, the Founders understood that a bargain was taking place on these terms, and the public followed Congress to see if the promised deal would indeed happen.166×166. Charles Warren, New Light on the History of the Federal Judiciary Act of 1789, 37 Harv. L. Rev. 49, 53–65 (1923) (“The fact is that the final form of the Act and its subsequent history cannot be properly understood, unless it is realized that it was a compromise measure, so framed as to secure the votes of those who, while willing to see the experiment of a Federal Constitution tried, were insistent that the Federal Courts should be given the minimum powers and jurisdiction.” Id. at 53.). This point is important because the original deal sketched out here is consonant with the original public meaning of the Constitution as a whole. See generally Andrew S. Oldham, The Anti-Federalists: Past as Prologue, 12 N.Y.U. J.L. & Liberty 451 (2019).

C. The Second Half of the Deal: State Courts and State Law

After the Convention, Antifederalists believed that the federal judiciary, with its potentially innumerable courts and boundless jurisdiction, would destroy state courts.167×167. E.g., Clinton, supra note 140, at 801; see Brutus, XV, N.Y. J., Mar. 20, 1788, reprinted in The Anti-Federalist, supra note 163, at 182, 186 (“[N]othing could have been better conceived to facilitate the abolition of the state governments than the constitution of the judicial.”); see also Warren, supra note 166, at 55–56 (detailing proposed amendments). But after the Judiciary Act, their tune changed because they were able to cabin the federal judicial power.168×168. E.g., 7 Annals of Cong. 815 (1802) (Joseph Gales ed., 1834) (statement of Joseph H. Nicholson) (“In a Government like ours, extending over a large tract of country, and composed of sovereign States . . . it was rightly judged that its Judicial powers should not extend to any other cases of Judicial cognizance, than those which might be deemed somewhat of a general nature . . . .”); see also Holt, supra note 145, at 1484.

Before turning to those restrictions, though, one analytical point is crucial. To get a fair sense of the full original constitutional deal, it’s important to read Article III and the Judiciary Act together, even though the former is part of the Constitution and the latter is a statute. Indeed, when interpreting the Constitution, what the First Congress did — and what backdrop its members understood themselves to be operating against immediately following ratification — is given special import.169×169. E.g., Marsh v. Chambers, 463 U.S. 783, 790 (1983). And this is especially true for the Judiciary Act of 1789 for a few reasons.

For one, as a descriptive matter, Article III and the Judiciary Act were taken as a package item from the beginning. As a local newspaper put it: “[A] bill of rights, and new and additional checks in the judiciary department, are almost universally agreed, as well by the honest friends, as the avowed opponents of the government, to be essential improvements” that would follow ratification.170×170. Independent Gazetteer (Phila.), Mar. 23, 1789; see Warren, supra note 166, at 55–56. Alexander Hamilton similarly promised that any defects in the national judiciary were surely going to be statutorily cured when the First Congress convened after ratification.171×171. See The Federalist No. 80, supra note 1, at 480 (Alexander Hamilton).

Specifically, people took the Judiciary Act as the second part of a bargain contemplated at the state ratification conventions between Federalists and Antifederalists. Antifederalist objections to the judiciary were prominent enough during the ratification conventions that Federalists thought there was a decent chance the entire draft Constitution would fail in the states.172×172. Holt, supra note 145, at 1471. To secure passage of the Constitution, therefore, its defenders promised future action; in particular, they pledged that the federal judicial power would be conditioned by later legislation, and the states, like with sovereign immunity, voted for ratification based on that understanding.173×173. See Henry J. Friendly, The Historic Basis of Diversity Jurisdiction, 41 Harv. L. Rev. 483, 508 (1928); see also Warren, supra note 166, at 54 (“Not only was the Judiciary Act a compromise, but its final form was closely tied up with . . . the fate of the various Amendments to the Judiciary Article of the Constitution which were being debated in Congress . . . .”). These promises included guaranteeing civil juries, lowering costs, and limiting the scope and use of lower federal tribunals.174×174. Holt, supra note 145, at 1471. And the First Congress did the bulk of this through the Judiciary Act. As such, the Court has explained that the Act “was passed by the first Congress assembled under the Constitution, many of whose members had taken part in framing that instrument, and is contemporaneous and weighty evidence of [Article III’s] true meaning.”175×175. Wisconsin v. Pelican Ins. Co., 127 U.S. 265, 297 (1888).

The notion that the Judiciary Act must inform our view of Article III and the original conception of the federal judiciary is confirmed by the structure of the Constitution. Article III works differently than Articles I or II because much of the judiciary’s power turns on action by the political branches.176×176. Bourguignon, supra note 134, at 667. We’ve already seen that the existence and composition of the lower federal courts is up to Congress. Further, under the Exceptions Clause, Congress has the power to limit (or even eliminate) wide swaths of the federal courts’ jurisdiction.177×177. U.S. Const. art. III, § 2, cl. 2. See generally Tara Leigh Grove, The Exceptions Clause as a Structural Safeguard, 113 Colum. L. Rev. 929, 934–38 (2013). As a general rule, a federal court accordingly must have both constitutional and statutory authority to hear a given case.178×178. E.g., Home Depot U.S.A., Inc. v. Jackson, 139 S. Ct. 1743, 1746 (2019). Justice Chase put this all in a nutshell in 1799: “[T]he political truth is, that the disposal of the judicial power, (except in a few specified instances) belongs to congress.”179×179. Turner v. Bank of N. Am., 4 U.S. (4 Dall.) 8, 10 n.1 (1799).

To gild the lily, the Court has also said as much. Take Ames v. Kansas,180×180. 111 U.S. 449 (1884). which confirmed the proposition that state courts are allowed to have concurrent jurisdiction over cases that could otherwise go to federal courts.181×181. See id. at 469–70; see also Amar, supra note 139, at 234. As noted, the text of Article III is entirely silent on this point. And a number of Antifederalists had argued that the Constitution actually barred concurrent jurisdiction.182×182. Clinton, supra note 140, at 802 & n.193. But the Ames Court — relying on the history of the Convention, the text of the Judiciary Act, and early historical practice — concluded that Article III provided for concurrent state court jurisdiction over federal matters.183×183. Ames, 111 U.S. at 469–70. That is, in order to settle a constitutional question that turned on the design of the federal judiciary, the Court relied heavily on the Judiciary Act.

For these reasons, the rest of this Part relies on both the Constitution and the Judiciary Act as it pieces together the second half of the original deal. Turning to federal adjudication of state law, the next two sections look into the two heads of federal jurisdiction that mainly implicate state law claims: federal question and diversity jurisdiction. The upshot is that the bargain struck between Federalists and Antifederalists held that state courts would remain the default forum for state law claims and federal courts would have limited, concurrent jurisdiction over a defined band of cases.184×184. See Turner, 4 U.S. at 9 (“The jurisdiction of the state Courts is general; but the jurisdiction of the federal courts is . . . in the nature of an exception from the general jurisdiction of the state Courts.”). This deal, forged through Article III and the Judiciary Act, does not readily permit a third mode of settling state law disputes and animates this Note’s ultimate claim about executive adjudication.

1. Federal Question Jurisdiction. — Following the initial debates over the federal judiciary, the Constitutional Convention passed a resolution that gave the putative national courts jurisdiction over “cases arising under laws passed by the general Legislature, and to such other questions as involve the National peace and harmony.”185×185. 2 Farrand’s Records, supra note 90, at 39. The Committee of Detail would later refine this mandate and the Constitution would eventually state that the “judicial Power shall extend to all Cases . . . arising under this Constitution, the Laws of the United States, and Treaties made, or which shall be made, under their Authority.”186×186. U.S. Const. art. III, § 2, cl. 1; see also Clinton, supra note 140, at 772–86. This head of jurisdiction, labeled “arising under” or federal question jurisdiction, gives federal courts the power to hear cases involving federal law.

Of course, there are plenty of federal questions that only involve federal law. For instance, the Alien and Sedition Acts were federal laws that violated the First Amendment. But at the time of the Convention, the impetus for federal question jurisdiction had something different behind it: Who decides if state and federal law conflict and, if they do, who decides which sovereign’s law prevails? As we’ll see, the answer to this question was judges. Indeed, when it came to policing the boundaries of our dual sovereign system, the Founders turned to the judiciary. To that end, the history of federal question jurisdiction affirms the intuition that the adjudication of claims implicating state law was for the courts.

(a) Article III. — Federal question jurisdiction emerged out of the failed proposals for a congressional veto over state laws and a related “Council of Revision.” A faction of the Constitutional Convention, led by James Madison, was concerned that state legislatures would undermine the Union by passing laws that contravened the Constitution.187×187. See Jack N. Rakove, Judicial Power in the Constitutional Theory of James Madison, 43 Wm. & Mary L. Rev. 1513, 1545–47 (2002). On this view, the “necessity of a general Govt. proceed[ed] from the propensity of the States to pursue their particular interests in opposition to the general interest” and this “propensity [would] continue to disturb the system, unless effectually contro[lled].”188×188. 2 Farrand’s Records, supra note 90, at 27 (James Madison). To control state legislatures, the Virginia Plan originally proposed two measures. First, it gave Congress the power “to negative all laws passed by the several States, contravening in the opinion of the National Legislature the articles of [the] Union.”189×189. 1 id. at 21. Second, the Plan provided that “the Executive and a convenient number of the National Judiciary [would] compose a council of revision with authority to examine every act of the National Legislature before it shall operate, & every act of a particular Legislature before a Negative thereon shall be final,” and the Council’s negative would be final unless overridden.190×190. Id. Charles Pinckney offered a similar plan where no state law could take effect without Congress. Sager, supra note 141, at 46 n.79. Put plainly, Congress could veto any state law while the Council retained a defeasible veto over Congress.191×191. Amar, supra note 139, at 223 n.69.

To be sure, the Founders also assumed that the national judiciary, via judicial review, would be able to set aside federal or state laws that violated the Constitution.192×192. See Edward S. Corwin, The Establishment of Judicial Review II, 9 Mich. L. Rev. 283 (1911). The question here was whether that was enough to combat whatever threat existed from state legislatures.193×193. Fallon et al., supra note 146, at 11. Tellingly, neither measure survived the Constitutional Convention.

The congressional negative was the prime tool offered to bolster the federal government’s control over state law.194×194. Anthony J. Bellia, Jr., The Origins of Article III “Arising Under” Jurisdiction, 57 Duke L.J. 263, 299 (2007). For those like Madison, Congress needed this power because states could “pass laws which will accomplish their injurious objects before they can be repealed by the [General Legislature] or be set aside by the National Tribunals.”195×195. 2 Farrand’s Records, supra note 90, at 27 (James Madison); see id. at 30 (James Wilson). On the other hand, most delegates took Roger Sherman’s view that such a scheme would be divisive and unnecessary because “the Courts of the States would not consider as valid any law contravening the Authority of the Union, and which the legislature would wish to be negatived.”196×196. Id. at 27 (Roger Sherman). For invalid laws that slipped through, opponents of the congressional negative said they could either be struck down by the national judiciary or preempted by the national legislature.197×197. Id. at 28 (Gouverneur Morris). At the end of the day, the congressional veto was voted down by a vote of seven states to three.198×198. Id.

Relatedly, the Convention also voted against the Council of Revision in favor of an executive veto over federal legislation that could be overridden by a two-thirds majority of each congressional chamber.199×199. Clinton, supra note 140, at 770. For us, the key part of this debate is the issue of whether the judiciary should’ve been part of the Council and its veto power.200×200. For a more comprehensive treatment of this subject, see Barry, supra note 102.

Proponents of the original Council, which would’ve been composed of members of the executive and judicial branches, argued that its dual-branch nature was necessary.201×201. E.g., 1 Farrand’s Records, supra note 90, at 139 (James Madison). First, an executive-only veto scheme would not be an adequate counterbalance to the legislature.202×202. 2 id. at 74 (James Madison). Second, incorporating the judiciary into the lawmaking process would improve lawmaking.203×203. 1 id. at 139 (James Madison). Although this meant that judges could review laws they had a hand in passing, the costs outweighed the benefits in light of “the perspicuity, the conciseness, and the systematic character” that they’d lend to the lawmaking process.204×204. Id.; see also Rakove, supra note 116, at 1067–68. Third, and most relevant, the Council’s design would not compromise judicial independence.205×205. Barry, supra note 102, at 252. This view turned on a flexible view of the separation of powers; pointing to the British model, where judges were part of the lawmaking process, Council supporters held that results mattered more than structural purity.206×206. E.g., 2 Farrand’s Records, supra note 90, at 75 (Gouverneur Morris).

Opponents of the Council urged that it would undermine the fundamental order of the Constitution.207×207. See Clinton, supra note 140, at 770–71. First, the entire idea of an independent judicial branch turned on judges deciding cases, not setting policy.208×208. 2 Farrand’s Records, supra note 90, at 75 (Caleb Strong); Barry, supra note 102, at 255; see also 1 Farrand’s Records, supra note 90, at 97–98 (Elbridge Gerry). Second, the proposal ultimately risked the independence of the federal judiciary writ large because judges would be biased in favor of laws they reviewed while on the Council.209×209. 2 Farrand’s Records, supra note 90, at 79 (Nathaniel Gorham). Third, and most importantly, the Council violated the separation of powers. Taking up a formalist mantle, opponents maintained that intermingling federal judges and the work of the political branches would be a per se “improper mixture of powers” in the teeth of the Constitution’s structure.210×210. 1 id. at 140 (John Dickinson); see also Barry, supra note 102, at 241–42 (“Other Framers . . . rejected Blackstone’s view of the separation of powers in favor of the more idealistic interpretation of Montesquieu.” Id. at 242.).

The opponents of the Council prevailed. Following a formalist rather than functionalist view of the separation of powers and the judiciary — one that far more resembles the approach of the Northern Pipeline plurality than the Schor Court — the resolution lost eight to three.211×211. 2 Farrand’s Records, supra note 90, at 298; see also Barry, supra note 102, at 257, 259–60.

So with the original framework offered by the Virginia Plan voted down, what came in to take its place? The Founders turned to the judiciaries. Luther Martin introduced, and the delegates soon adopted, a preliminary version of the Supremacy Clause, which expressly guaranteed the supremacy of federal law over state law and bound state judges to abide by this hierarchy.212×212. Compare 3 Farrand’s Records, supra note 90, at 286–87, with U.S. Const. art. VI, cl. 2. Further, the delegates made sure that the federal courts could have jurisdiction — that is, “arising under” jurisdiction — over cases implicating the Supremacy Clause.213×213. Bellia, supra note 194, at 301; see also Kaufman, supra note 130, at 687. Because state courts were assumed to have concurrent jurisdiction with federal courts (recall the Madisonian Compromise), this meant that federal question cases would go to either state courts or, if created, lower federal courts, and appeals could then eventually end up in the Supreme Court.214×214. Bellia, supra note 194, at 300–01. Together, as Professor Jack Rakove has illustrated, these measures “confirmed the status of the Constitution as fundamental law, [and] also made the enforcement of its essential division of power between the Union and the States an inherently judicial function.”215×215. Rakove, supra note 116, at 1068–69.

This was a big turn, and the decision reveals an important calculation about who in the federal government was intended to decide state law questions. Whether a state law conflicts with federal law, or whether a federal statute exceeds the Constitution’s limitations on federal power, are fundamentally interpretive questions. That is, one must first decide the meaning of the relevant laws in order to figure out if a conflict really exists (and, if so, who wins). In rejecting the congressional negative, the Founders concluded that this interpretive function must be performed by judges rather than political actors.216×216. The Federalist No. 81, supra note 1 (Alexander Hamilton). This because judges alone could be tasked with assessing the meaning of a given state law in light of certain federal interests (or the other way around) in an impartial manner that gave fair weight to each sovereign’s interest.217×217. See Amar, supra note 139, at 223–25.

For this reason, the Founders also scrapped the Council of Revision.218×218. See Barry, supra note 102, at 257–60. Of course, the federal government can try to get rid of state laws it doesn’t like through preemption. But whether the federal government has the constitutional power to preempt a given law, or whether a state law really conflicts with a federal statutory scheme, is a judicial question — one that the national judiciary would be in an appropriate place to answer once removed from the lawmaking process.219×219. The Federalist No. 39, supra note 1, at 242 (James Madison) (identifying Supreme Court as tribunal that would decide “controversies relating to the boundary between” sovereigns). As James Madison put it to Thomas Jefferson, the Convention “intended the Authority vested in the Judicial Department as a final resort in relation to the States, for cases resulting to it in the exercise of its functions.”220×220. Letter from James Madison to Thomas Jefferson (June 27, 1823), in 9 The Writings of James Madison 137, 142 (Gaillard Hunt ed., 1910).

Turning to the ratification conventions, debates about federal question jurisdiction were bound up in the now-familiar objections to the existence of the federal judiciary as a whole. Antifederalists had two interrelated qualms. For one, “arising under” jurisdiction was arguably phrased so broadly that it’d inevitably extend to everything, giving federal courts general rather than limited jurisdiction.221×221. 3 Elliot’s Debates, supra note 90, at 565 (William Grayson). Also, the federal government would resolve federal questions in favor of federal interests.222×222. Rakove, supra note 116, at 1069–70. And that wouldn’t be fair for the cases involving state law.223×223. See id. at 1070.

The Federalist response was twofold. First, Antifederalist hand-wringing was misplaced because federal question jurisdiction required a clear nexus to federal law.224×224. Clinton, supra note 140, at 811; see also Casto, supra note 157, at 1104 & n.32. Second, and most of all, federal interests would not have outsized importance in such cases because those issues would be settled by the judiciary, rather than within a political branch.225×225. See Clinton, supra note 140, at 812–14. For instance, North Carolina’s William Davie noted in the state’s debates that if federal law “supersede[s] the laws of particular states,” the Supremacy Clause’s “great object can only be safely and completely obtained by the instrumentality of the federal judiciary.”226×226. 4 Elliot’s Debates, supra note 90, at 156–57 (William Davie).

These assurances were enough to get the Constitution through ratification. But Federalists would have to make good on them shortly.

(b) The Judiciary Act of 1789. — The Federalist concessions on this score manifest across two parts of the Judiciary Act, resulting in a system where state courts would handle the bulk of federal question cases and state laws would be invalidated only under specific circumstances.

First, in a surprising development, the Judiciary Act cut the lower federal courts out of pretty much all federal question cases.227×227. William R. Casto, An Orthodox View of the Two-Tier Analysis of Congressional Control over Federal Jurisdiction, 7 Const. Comment. 89, 92–93 (1990). Recall that the Constitution gives Congress the power to adjust the jurisdiction of the lower federal courts on the ground that the power to create courts comes with it the power to define their jurisdiction.228×228. E.g., Sheldon v. Sill, 49 U.S. (8 How.) 441, 448–49 (1850). As such, the First Congress (i) entrusted state courts with exclusive original jurisdiction over the vast majority of federal question cases, (ii) limited the Supreme Court’s appellate jurisdiction to cases where a federal claim was upheld (rather than denied), and (iii) gave lower federal courts exclusive jurisdiction over certain seizure cases and criminal matters.229×229. Fallon et al., supra note 146, at 22, 25. Under this setup, state courts (and often state courts alone) would be the ones that decided the lion’s share of federal question issues for the nation.230×230. Warren, supra note 166, at 62, 70.

This regime was a massive boon to the Antifederalists. For sure, many Federalists had argued that only federal judges could be trusted with handling federal question cases.231×231. Bourguignon, supra note 134, at 694. But there were enough Antifederalists in the First Congress, led by Senator Richard Henry Lee, who felt that state courts were suited to handle cases involving the Supremacy Clause and that giving this responsibility primarily to state judges would ensure an equitable balance of power among sovereigns.232×232. Holt, supra note 145, at 1480–81, 1484–87, 1487 n.232 (“Federal questions, which at the time were expected to consist of questions of the unconstitutionality of state laws, would arise most naturally in the midst of state-law cases, and the compromise recognized this by giving the trial jurisdiction of these issues to the state courts.” Id. at 1487 n.232).

Second, of a part with Federalists’ intimation that state law cases settled by the federal government would be decided by federal judges, the Founders took the conferral of “arising under” jurisdiction to include only cases that turned on a question of federal law, not federal interest. This is apparent from early Court precedent. As Professor Anthony Bellia has shown, “the Supreme Court explicated the Arising Under Clause in the first few decades following ratification to mean that a federal court may exercise jurisdiction over cases in which an actual federal law was determinative of a right or title asserted in the proceeding before it.”233×233. Bellia, supra note 194, at 269. Notably, this meant that the Court didn’t “assume for federal courts a constitutional jurisdiction to vindicate federal interests divorced from the governing requirements of an identifiable federal law.”234×234. Id. at 270; see also id. at 341.

This understanding of federal question jurisdiction is important because it confirms the point above that the Founders were quite precise regarding who can adjudicate state law issues and under what conditions. Particularly, the above betrays a view of the constitutional plan that holds federal adjudication of state law is a defined judicial function. To be clear, on this view, in order for the Court to hear a federal question case involving state law, (i) a federal law must be determinative of the right sought in the case, and (ii) a state law must conflict with that right.235×235. Id. at 328–29. It simply did not matter whether a state law is “incidental to” or “dependent upon” — to borrow language from the Schor Court — a general federal interest or scheme.236×236. Commodity Futures Trading Comm’n v. Schor, 478 U.S. 833, 856 (1986). What mattered was whether there was a clash of rights capable of creating a case for judicial resolution.

The inseparability of the federal government’s ability to act on state law and the exercise of judicial power is further laid bare by the Court’s 1816 landmark decision of Martin v. Hunter’s Lessee.237×237. 14 U.S. (1 Wheat.) 304 (1816). There, the Court addressed a claim arising under the Jay Treaty (thus a federal question) that turned on an antecedent question of Virginia law.238×238. See Herbert Wechsler, The Appellate Jurisdiction of the Supreme Court: Reflections on the Law and the Logistics of Direct Review, 34 Wash. & Lee L. Rev. 1043, 1051 (1977). Justice Story held that the Court could review the state law decision of Virginia’s top court because it was necessary for the ultimate resolution of the Jay Treaty claim.239×239. Hunter’s Lessee, 14 U.S. at 338–39. The Court justified its ability to reach the state law point entirely on the fact that the judiciary was given the judicial power over bona fide cases or controversies.240×240. Id. at 338. On that point, as Professor Herbert Wechsler distilled, Hunter’s Lessee stands for the proposition that where “the existence or the application of a federal right turns on a logically antecedent finding on a matter of state law, it is essential to the Court’s performance of its function that it exercise an ancillary jurisdiction to consider the state question.”241×241. Wechsler, supra note 238, at 1052. At bottom, this had nothing to do with the government’s power to make treaties or the nation’s interest in foreign affairs. Rather, the only hook that allowed the federal government to have the power to interpret Virginia law was that Martin’s claim constituted a “case” and the Constitution provided that the federal judicial power could extend to deciding the question presented.

In sum, the story of federal question jurisdiction reveals a compromise concerning federal power that is deeply protective over state law. Through rejecting the congressional veto and limiting the jurisdiction of the lower federal courts, the Founders sketched together a system that accounted for the supremacy of federal law but also protected the integrity of state judiciaries. As a result, when it came to cases implicating state law, not only was the judiciary the only federal branch able to act, but the scope of its power to do so was also limited by congressional restrictions on the original jurisdiction of lower federal courts and the plain meaning of the Arising Under Clause. In all events, adjudication of state law was a judicial function, be it in state or federal court.

2. Diversity Jurisdiction. — Most state law claims have nothing to do with federal law. When Bluto sues Otter in a typical tort action, the Supremacy Clause isn’t at issue since it’s a purely state law case. But if Bluto lives in a different state than Otter, his suit may qualify for diversity jurisdiction — the provision that gives the federal courts jurisdiction over certain claims “between Citizens of different States.”242×242. U.S. Const. art. III, § 2, cl. 1.

There isn’t a settled explanation for why diversity jurisdiction is in the Constitution.243×243. See Robert L. Jones, Finishing a Friendly Argument: The Jury and the Historical Origins of Diversity Jurisdiction, 82 N.Y.U. L. Rev. 997, 999–1026 (2007). Indeed, the records of the Constitutional Convention contain virtually zero debate on the matter.244×244. Fallon et al., supra note 146, at 17. That said, once the draft Constitution reached the states, diversity jurisdiction became one of the biggest controversies around and remained a key point of dispute through the Judiciary Act. Skipping ahead, diversity jurisdiction too was the product of a deliberate compromise that cabined the federal government’s ability to decide rights and obligations under state law.

(a) Article III. — Diversity jurisdiction was a national lightning rod as soon as it reached the states by way of their conventions.245×245. Friendly, supra note 173, at 499; Warren, supra note 166, at 56. And the ferocity of this opposition, coupled with the concessions it elicited, frame diversity jurisdiction’s later taming in the Judiciary Act.

In debating the draft Constitution, Antifederalists argued that diversity jurisdiction, which again could send garden variety state law claims to the federal courts, would be a gateway to the gutting of state judiciaries.246×246. The Objections of the Hon. George Mason to the Proposed Federal Constitution (Thomas Nicholas 1787), reprinted in Pamphlets on the Constitution of the United States 327, 329–30 (Paul Leicester Ford ed., Brooklyn, N.Y., 1888) (claiming federal judiciary will “absorb and destroy the judiciaries of the several states”). Once you take diversity jurisdiction, the argument went, and add to it the fact the draft Constitution also provided for alienage and state-versus-foreign-citizen jurisdictions, there was really nothing of import left for the state courts besides cases between citizens of the same state.247×247. Agrippa, V, Mass. Gazette, Dec. 11, 1787, reprinted in Essays on the Constitution of the United States 66, 67 (Paul Leicester Ford ed., Brooklyn, N.Y., 1892). Eventually, the prestige and authority of state courts would whittle away until people found them useless.248×248. See 1 Farrand’s Records, supra note 90, at 124 (John Rutledge); see also F. Andrew Hessick, Consenting to Adjudication Outside the Article III Courts, 71 Vand. L. Rev. 715, 744–45 (2018).

Of a part, this diminution of state courts would jeopardize the integrity of state law. Previewing an issue that would launch a thousand Erie-themed law review articles, Antifederalists maintained that federal courts sitting in diversity would apply federal law (either statutory or general common law) rather than state law.249×249. Agrippa, supra note 247, at 66. As such, the attack on state sovereignty was twofold: federal courts would steadily reduce the role of state courts and, as part of adjudicating an increasing swath of their cases, would displace state law in favor of a federal legal regime.250×250. See 2 Elliot’s Debates, supra note 90, at 551 (claiming that “extension of the federal jurisdiction [could] . . . sap those rules of descent and regulations of personal property”); see also Patrick J. Borchers, The Origins of Diversity Jurisdiction, the Rise of Legal Positivism, and a Brave New World for Erie and Klaxon, 72 Tex. L. Rev. 79, 97 (1993).

Perhaps jolted by this animosity, Federalists initially gave a faint reply.251×251. Friendly, supra note 173, at 487–88 (collecting quotes). Madison conceded that as to “disputes between citizens of different states, I will not say it is a matter of much importance” and “[p]erhaps it might be left to the state courts.”252×252. 3 Elliot’s Debates, supra note 90, at 533 (James Madison). Edmund Randolph similarly did “not see any absolute necessity for vesting [the federal judiciary] with jurisdiction in these cases.”253×253. Id. at 572 (Edmund Randolph). So with friends like these, why did diversity jurisdiction stick around? The conventional narrative has been that federal courts could provide a neutral forum to protect out-of-state litigants against local bias.254×254. See, e.g., Bank of the United States v. Deveaux, 9 U.S. (5 Cranch) 61, 67 (1809). But this has been largely replaced. Many today support Judge Friendly’s thesis that “the desire to protect creditors against legislation favorable to debtors was a principal reason for the grant of diversity jurisdiction.”255×255. Friendly, supra note 173, at 496–97; see Jones, supra note 243, at 1017 n.79 (collecting scholars). Others qualify this position by pegging diversity jurisdiction to worries over unwieldy state juries.256×256. Jones, supra note 243, at 1003–06.

The main point is that Federalists, for a combination of reasons, wanted a backstop of federal diversity jurisdiction to ensure that a class of state law cases could be heard outside of state courts. But to overcome Antifederalist objections, proponents of diversity jurisdiction insisted that this class would be defined and limited, only picking up cases with a true national importance (namely, high-dollar debt cases).257×257. See Borchers, supra note 250, at 96; see also id. at 91, 94. Indeed, Madison insisted that “ninety-nine out of a hundred” cases would remain with the state courts and the “number of cases within the jurisdiction of [federal] courts [would be] very small when compared to those in which the local tribunals [would] have cognizance.”258×258. 3 Elliot’s Debates, supra note 90, at 537–38 (James Madison); see also Holt, supra note 145, at 1471–78; Roger Sherman, Observations on the New Federal Constitution, New Haven Gazette, Dec. 25, 1788, reprinted in Essays on the Constitution of the United States, supra note 247, at 237, 240–41. Against this backdrop, the Judiciary Act delineated what state law matters the states were willing to cede to the federal government and under what terms.259×259. See Friendly, supra note 173, at 503–08.

(b) The Judiciary Act of 1789. — After the Constitution was ratified, some Antifederalists again tried to abolish diversity jurisdiction, this time through legislation.260×260. See id. at 502; see also Collins, supra note 135, at 1523–33. That didn’t work out. But the Judiciary Act came pretty close at the end of the day once one accounts for all of its restrictions. In the words of Richard Henry Lee, the leading Antifederalist on the Senate Judiciary Committee, writing to Patrick Henry, “I have endeavored successfully in the Judiciary Bill to remedy, so far [as] a law can remedy, the defects of the Constitution in that line.”261×261. Holt, supra note 145, at 1484 (alteration in original) (quoting Letter from Richard Henry Lee to Patrick Henry (Sept. 14, 1789)). This Antifederalist remedy was accomplished through three sections.

First, the vast majority of diversity cases would be excluded from federal jurisdiction by a steep amount-in-controversy requirement.262×262. Id. at 1487–88; see also 10 Annals of Cong., supra note 168, at 897 (Jan. 7, 1801) (statement of Rep. John Nicholas) (setting number amount as above average land tax debts). Specifically, suits would need to have at least $500 in dispute and, if they did, actions could go to the circuit courts (staffed by two Justices and one district court judge) rather than the district courts.263×263. Holt, supra note 145, at 1486. This threshold amount kept the bulk of state law claims purely in state court.264×264. Id. at 1487–88, 1488 nn.233–34 (collecting statistics on torts and contracts); see Bourguignon, supra note 134, at 669 (finding the amount would result in “leaving most cases in state courts”). For instance, Professor William R. Casto concluded that the “five hundred dollar amount in controversy limitation also effectively barred virtually all common law tort actions from the federal trial courts.”265×265. Casto, supra note 157, at 1113; see id. at 1113 n.93; see also Amar, supra note 139, at 260.

Second, the First Congress narrowed the definition of what it meant to be “diverse parties” for jurisdictional purposes. The Judiciary Act limited diversity jurisdiction to suits “between a citizen of the State where the suit is brought, and a citizen of another State.”266×266. Judiciary Act of 1789, ch. 20, § 11, 1 Stat. 73, 78. This meant that either the plaintiff or defendant needed to be a citizen of the state where the suit was brought.267×267. See Warren, supra note 166, at 79. The Act also added an “assignee clause,” which prohibited one person from assigning a promissory note to another in order to establish diversity.268×268. Casto, supra note 157, at 1114. The aim of these statutory provisions was to ensure that diversity could not be manufactured.269×269. See id. at 1114 & n.98; Holt, supra note 145, at 1497 & n.265.

Third, and perhaps most consequentially, section 34 of the Judiciary Act required federal courts sitting in diversity to apply state rather than federal law.270×270. Bourguignon, supra note 134, at 675. Of course, this reading is subject to debate, compare Warren, supra note 166, at 81–88, with Swift v. Tyson, 41 U.S. (16 Pet.) 1, 18 (1842), but Professor Charles Warren’s position is the conventional wisdom, see Borchers, supra note 250, at 103 & nn.187–88. Recall that Antifederalists were worried that Congress’s power to “appoint [federal] courts necessarily involve[d] in it the right of defining their powers, and determining the rules by which their judgment shall be regulated.”271×271. Agrippa, VI, Mass. Gazette, Dec. 14, 1787, reprinted in Essays on the Constitution of the United States, supra note 247, at 69, 69. Section 34 was a response to that concern.272×272. See Warren, supra note 166, at 83. At bottom, in the words of Professor Charles Warren, the provision made “it perfectly certain that the Federal Courts were simply to administer State law.”273×273. Id. To be sure, plenty of Federalists wanted federal courts sitting in diversity to apply federal law along a set of general common law principles.274×274. See Borchers, supra note 250, at 97 nn.138–39 (collecting sources). But in order to gain sufficient Antifederalist support for passage, Federalists gave up on this wish during the ratification debates.275×275. See, e.g., 3 Elliot’s Debates, supra note 90, at 556–57 (John Marshall). And, at least until Justice Story’s opinion in Swift v. Tyson,276×276. 41 U.S. (16 Pet.) 1 (1842). section 34 was taken to codify this commitment, as the Court explained in M’Niel v. Holbrook,277×277. 37 U.S. (12 Pet.) 84 (1838). “to make the rules of decisions in the courts of the United States, the same with those of the states.”278×278. Id. at 89–90; see Warren, supra note 166, at 84–85, 88.

In short, as with federal question jurisdiction, the lesson of diversity jurisdiction is as follows: Through the combination of Article III and the Judiciary Act of 1789, the Founders fashioned a compromise that assiduously defined and limited the circumstances when federal power could be used to decide rights and obligations under state law. Against a backdrop of blowback in the state ratifying conventions and the concomitant promises made to secure the Constitution’s adoption, the First Congress narrowed the number of diversity cases that could pass to the federal courts, the qualifications for diversity jurisdiction, and the law that would decide those disputes. When the dust settled, the status quo mostly held: for most cases, state courts would be the only act in town.

III. The Problem with Executive Adjudication of State Law

This Part tries to put it all together. The basic idea is that the history and structure of the Constitution reveal an implicit bar on the federal government’s ability to create executive branch tribunals that can adjudicate state law claims. And this matters when it comes to locating the constitutional source of congressional power to create such tribunals.

A. The Constitutional Bar

Executive adjudication is different in kind from what happens in Article III courts. To start, agencies have a different mandate than the “balls and strikes” model of the federal judiciary. Since Chenery II,279×279. SEC v. Chenery Corp., 332 U.S. 194 (1947). it has been black letter law that agencies may develop policy through adjudication, much as they do in rulemaking.280×280. See id. at 201–04. See generally Russell L. Weaver & Linda Jellum, Chenery II and the Development of Federal Administrative Law, 58 Admin. L. Rev. 815 (2006). “Judges” within agencies are also either officers or employees of the executive branch, which means that they lack Article III’s salary and tenure protections and instead, to varying and debated degrees, can be removed by the President.281×281. See, e.g., John Harrison, Addition by Subtraction, 92 Va. L. Rev. 1853, 1862 (2006). As a result, they lack complete independence, the judiciary’s single most important characteristic. For these reasons, when it comes to issues involving both federal and state law, agencies have a clear (and documented) institutional incentive to resolve questions in favor of federal interests.282×282. See Hamburger, supra note 101, at 234–35; Hessick, supra note 248, at 745. Moreover, a number of the other structural doctrines that the Founders accounted for to limit judicial power — such as standing, ripeness, and mootness — don’t apply to federal agencies, pulling an even wider range of matters into their ambit.283×283. See F. Andrew Hessick, Cases, Controversies, and Diversity, 109 Nw. U. L. Rev. 57, 94–95 (2014); Hessick, supra note 248, at 745 n.182. But see Woolhandler & Nelson, supra note 30, at 693–712. To be sure, much of the same describes legislative courts, which lack the same structural protections and also have a greater ability to render final judgments.284×284. Redish, supra note 22, at 214. That said, it’s true that in some instances tribunals like bankruptcy courts present their own set of wrinkles. See, e.g., Thomas E. Plank, Why Bankruptcy Judges Need Not and Should Not Be Article III Judges, 72 Am. Bankr. L.J. 567, 607–09 (1998). But they mainly relate to whether a tribunal can qualify as an “adjunct.” See F. Andrew Hessick, Federalism Limits on Non-Article III Adjudication, 46 Pepp. L. Rev. 725, 731, 753–54 (2019); see also David J. Meltzer, Legislative Courts, Legislative Power, and the Constitution, 65 Ind. L.J. 291, 303 n.61 (1990) (explaining why this distinction makes little difference for our purposes).

The point is that executive adjudication is its own form of adjudication, markedly distinct from federal, let alone state, courts. This is not good. The history and structure of the Constitution, informed by the Judiciary Act of 1789, suggest that creating this third path for state law claims conflicts with an implicit constitutional limit on federal power.

First, the historical argument. Remember that the creation of any lower federal courts was controversial enough at the Constitutional Convention that delegates needed to defer on the issue through the Madisonian Compromise. In fact, a critical mass of Antifederalists and delegates at state ratifying conventions were deeply concerned that any alternative judicial avenue for cases that had traditionally gone to state courts, let alone a non-judicial track nestled within one of the political branches, would invariably crowd out state judiciaries and, in turn, disrupt expectations forged under state law. The tradeoff offered to Antifederalists to tolerate these potentially intrusive federal courts was that their jurisdiction would be curtailed. Under this regime, state courts would stay as the default option. Critically, this meant that, with the exception of an enumerated set of state law cases that could go to the federal courts, everything else would remain with the original baseline, which was only state court adjudication. There was no third option.

Revealingly, if the Founders wanted to try to send state law claims to executive tribunals, they knew how to do so. Federal non–Article III tribunals have existed since the Founding to serve various purposes. As Professor Jerry Mashaw has catalogued, adjudication within the Executive has been around since the First Congress.285×285. Jerry L. Mashaw, Creating the Administrative Constitution 74–75 (2012). But each of these Founding-era executive tribunals appears to have adjudicated only federal public rights cases; for example, claims for government-backed debts or title to federal land.286×286. See Jennifer L. Mascott, Constitutionally Conforming Agency Adjudication, 2 J. Reg. Compliance 22, 46 n.195 (2017). And this makes sense once we take stock of the fact that adjudication is not synonymous with the exercise of judicial power; indeed, plenty of executive functions, like the distribution of Social Security benefits, are done with procedures that seem adjudicative.287×287. Gary Lawson, The Rise and Rise of the Administrative State, 107 Harv. L. Rev. 1231, 1246 (1994) (describing difference between adjudicatory actions and judicial power). But there don’t seem to be any Founding-era examples of executive adjudication of private right state law cases — which is exactly what you’d expect in the event this Note’s thesis is right.288×288. See Hamburger, supra note 101, at 191–92; Jerry L. Mashaw, Federal Administration and Administrative Law in the Gilded Age, 119 Yale L.J. 1362, 1410–11 (2010); Nelson, supra note 19, at 608–09; see also supra note 37.

On that point, the absence of historical analogs to the adjudicative scheme like the one in Schor is not surprising in light of the history. The Founders’ experience with the Privy Council and colonial courts motivated their rejection of England’s blended version of the separation of powers and fostered a skepticism toward political involvement in judicial functions. As the congressional veto episode makes plain, even if the Founders thought executive adjudication was appropriate in some settings, they believed that cases arising under state law (or the intersection of federal and state law) were exclusively for judiciaries to decide.

Second, the structural argument. Article III contemplates a two-track system for cases arising under state law. State courts are the default forums while federal courts have concurrent jurisdiction in narrow circumstances. As to the latter, the Founders fashioned a comprehensive scheme as to exactly how and when federal power could extend to state law: the judicial power was vested exclusively in the national judiciary; the very existence of federal courts was up to Congress; judges were made independent from the other branches through salary and tenure protections; and federal court jurisdiction was meticulously defined.289×289. See Hessick, supra note 284, at 737–40.

It is a heavy lift to hold that executive adjudication of state law — with non–Article III tribunals staffed by federal employees unprotected by Article III deciding cases that could otherwise go to state courts — neatly folds into this structure.290×290. For a fuller discussion of this point, see id. at 732–43. Put plainly, this practice is essentially the equivalent of a landlord spending months on a meticulous rental agreement with a specific tenant but also allowing that tenant to sublet the apartment to an unaccountable third party without any strings attached. For what it’s worth, if the Constitution contained this backdoor for federal adjudicatory power, nobody told the Federalists who were griping about the Judiciary Act’s limits on national judicial authority.291×291. See, e.g., Holt, supra note 145, at 1517.

Looking back at the features of the structural federalism cases, this argument against executive adjudication of state law finds close doctrinal company. For one, as with both anticommandeering and state sovereign immunity, this Note’s core thesis closely tracks Federalist promises made to Antifederalists to pass the Constitution.292×292. E.g., Franchise Tax Bd. v. Hyatt, 139 S. Ct. 1485, 1495–96 (2019). In fact, this Note arguably has greater support than some of the federalism cases because these pledges were codified in a source of positive law through the Judiciary Act. Also, as with the anticommandeering doctrine, there is an absence of Founding-era analogs to executive adjudication of state law — something the Court has held is particularly telling when a given legislative scheme doesn’t naturally fit within the Constitution’s structure.293×293. E.g., Printz v. United States, 521 U.S. 898, 907–09 (1997). And on that structural point, much as the rejection of the New Jersey Plan proved significant in cases like New York v. United States294×294. 505 U.S. 144 (1992). and Printz, the rejection of the congressional veto regime in favor of the Supremacy Clause lends important, concrete support here.295×295. See Printz, 521 U.S. at 919–22; New York, 505 U.S. at 164–65. In short, this Note is of a part with the sources and methods of the structural federalism cases. And provided those cases establish a way we’re supposed to interpret the Constitution, rather than an ad hoc way to carve off pieces of federal power on federalism grounds, then the executive adjudication of state law seems to have a hard time passing scrutiny.

So how should we think about all of this? The history and structure of the Constitution indicate that Congress cannot create a tribunal within the executive branch to adjudicate state law claims consistent with the vertical separation of powers. For sure, this Note’s claim could perhaps just rest on those grounds; indeed, doctrines like executive immunity rest on history and structure alone.296×296. See Hyatt, 139 S. Ct. at 1498–99. However, there’s likely a sounder textual hook here. As noted, for there to be executive adjudication of state law, Congress must first have the power to create the relevant tribunal — something it cannot do without the Necessary and Proper Clause. But it’s not “Proper,” as originally understood, to commit state law claims to executive adjudicators because doing so violates a principle of the vertical separation of powers and, in turn, the state sovereignty embodied in the history and structure of the Constitution.297×297. See Lawson & Granger, supra note 9, at 330–33; see also The Federalist No. 33, supra note 1, at 200 (Alexander Hamilton) (“But it will not follow from this doctrine that acts of the larger society which are not pursuant to its constitutional powers, but which are invasions of the residuary authorities of the smaller societies, will become the supreme law of the land.”).

Doctrinally, the Court has already taken up this approach with anticommandeering and state sovereign immunity, first identifying a limit along the vertical separation of powers based on history and structure, and then enforcing that limit through the Necessary and Proper Clause to cabin a general power otherwise available to Congress under Article I.298×298. See Alden v. Maine, 527 U.S. 706, 732–33 (1999); Printz, 521 U.S. at 923–24. The same framework holds here: Congress has the power to create some non–Article III tribunals, but not these ones. Put otherwise, executive tribunals with jurisdiction over state law actions are unconstitutional first and foremost not because they cross some ill-defined line within Article III and thereby suddenly exercise “the judicial power of the United States” (although they do), but because Congress simply lacks the Article I power to create them in that specific form.

The inevitable conclusion from this is that Schor was wrongly decided. Of course, plenty of people have reached this conclusion one way or the other.299×299. See, e.g., Bator, supra note 13, at 256–60. But grounding this view in Article I supplies surer footing because it accounts for the vertical separation of powers in a way that focusing just on Article III does not. Recall that both the dissent in Schor and the various majority opinions in the cases where the Court has struck down a federal scheme involving non–Article III adjudication of state law have relied exclusively on the horizontal separation of powers.300×300. See, e.g., Wellness Int’l Network, Ltd. v. Sharif, 135 S. Ct. 1932, 1951–52 (2015) (Roberts, C.J., dissenting). To be sure, Article III alone can resolve a number of questions involving non–Article III tribunals.301×301. See Glidden Co. v. Zdanok, 370 U.S. 530, 562–64 (1962). But it has proven an awkward fit as the sole explanation for why such tribunals cannot hear certain state law claims. First, the Constitution’s primary forum for state law adjudication is a non–Article III tribunal (state courts). Second, in terms of the federal interest, it’s intuitively odd to draw the line at state rather than federal law claims (rather than vice versa). And third, Article III’s Vesting Clause doesn’t convincingly supply any internal principle for parsing acceptable from unacceptable executive adjudication, as the Stern Court’s attempt to fashion one potentially seems to reveal.302×302. See Stern v. Marshall, 564 U.S. 462, 504 (2011) (Scalia, J., concurring).

This Note’s Article I argument provides a clear-cut proposition about the scope and source of federal power that stands in contrast to the sort of mercurial line-drawing that has otherwise plagued this issue. For sure, Article III is instructive here — indeed, the two Articles work in tandem. What went into the creation of the federal judiciary and its carefully defined jurisdiction, codified in Article III and refined by the Judiciary Act, are essential pieces of evidence for the constitutional deal that shapes this Note’s vertical separation of powers claim. But Article III doesn’t give the whole story. Instead, as with the structural federalism cases, the Constitution’s dual-sovereign nature betrays a limit on federal power, and the Necessary and Proper Clause accounts for that limit by bounding congressional power to what is “Proper” — that is, what’s consistent with each axis along the separation of powers.303×303. This vertical separation of powers perspective also reveals how foundationally incomplete the Schor decision was, even on its own terms. First, Schor’s balancing test is problematic not only because it’s mushy and imprecise, but also because it’s myopic. See Schor, 478 U.S. at 850–51. Indeed, it ignores the constitutionally protected interest that states have as separate sovereigns. Second, the Schor Court critically erred when it collapsed judicial and non-judicial adjudication. Justice O’Connor held that the fact the state law claim was “resolved by a federal rather than a state tribunal could not be said to unduly impair state interests” because “it is established that a federal court” could have heard the case. Id. at 858. But the historical record forecloses this point.

At bottom, as noted at the beginning, the Constitution divides power horizontally and vertically. And our system of dual sovereignty was based on the view that each dimension was equally important, working together like a structural Lennon & McCartney. But when it comes to executive adjudication of state law, the Court looked at this division more like Simon & Garfunkel, with one plainly more important than the other. This Note has attempted to remedy that core misapprehension.

B. Some Doctrinal Implications

If the above holds any water, then we should give greater weight to the vertical separation of powers when we think about executive adjudication of state law. This last section touches on a few doctrinal implications of that viewpoint. To be sure, these points only skim the surface; each of these topics can merit a far deeper treatment than what’s here.

1. The Appellate Review or Adjunct Solution. — There’s a view that pretty much anything can start in a non–Article III tribunal as long as there’s a court on the back end. This intuition has manifested in two related doctrines. For one, as we saw in Crowell, the Court has held that these sorts of tribunals can handle cases as long as they act as “adjuncts,” rendering preliminary decisions on questions of fact and law, but leaving any binding judgment to the courts.304×304. Hessick, supra note 284, at 731. Similarly, some maintain that non–Article III tribunals can adjudicate the vast majority of cases as long as courts retain adequate appellate jurisdiction over their decisions.305×305. See, e.g., Richard H. Fallon, Jr., Of Legislative Courts, Administrative Agencies, and Article III, 101 Harv. L. Rev. 915, 974 (1988). This is generally called the “appellate review model.”306×306. See generally Thomas W. Merrill, Article III, Agency Adjudication, and the Origins of the Appellate Review Model of Administrative Law, 111 Colum. L. Rev. 939 (2011).

Despite the first glance appeal of these positions, they likely do not salvage executive adjudication of state law. First things first, the Court has already suggested as much in the last three cases that involved a bankruptcy court’s (recall, a legislative court) handling of a state law claim.307×307. See Stern, 564 U.S. at 469; Granfinanciera, S.A. v. Nordberg, 492 U.S. 33, 55–56 (1989); N. Pipeline Constr. Co. v. Marathon Pipe Line Co., 485 U.S. 50, 56, 87 (1982) (plurality opinion). But the federalism-oriented approach described here offers additional reasons why. For one, as Professor F. Andrew Hessick has compellingly explained, federal adjuncts are designed to favor federal interests and “determinations of adjuncts often are binding on courts and dictate the outcome in a case.”308×308. Hessick, supra note 284, at 754; see also id. at 741–42, 753–54. As we have seen, this is the exact concern raised repeatedly by the Antifederalists at the Constitutional Convention. Likewise, even if sufficient judicial review exists on the back-end for some non–Article III adjudicatory scheme, a core problem remains: the watering down of the original jurisdiction of state courts.309×309. Id. at 745. Indeed, these sorts of tribunals are easier to make than Article III courts, face fewer doctrinal constraints, and operate under different incentives.310×310. Id. at 746–47.

For these reasons, they are a distinct type of forum compared to federal or state courts and will take cases that otherwise would work their way through state judiciaries.311×311. See id. at 737–39. As such, even if later oversight is demanding or tribunals are cast one way or the other as adjuncts, there’s a freestanding constitutional defect that would problematically linger — the diminishment of state courts. Because the preservation of state court jurisdiction drove the design of the federal judiciary, the theories noted here likely don’t bolster executive adjudication of state law.312×312. This also would seem to caution against the idea that parties could consent to have their state law action adjudicated by an executive tribunal. See Wellness Int’l Network, Ltd. v. Sharif, 135 S. Ct. 1932, 1939 (2015). Indeed, States have a freestanding interest in how their citizens’ rights and obligations are determined under their law (even if those citizens don’t care) as well as a stake in the integrity of the vertical separation of powers (which is presently ignored in the Court’s consent jurisprudence). Cf. Clinton v. City of New York, 524 U.S. 417, 450 (1998) (Kennedy, J., concurring); West Virginia v. EPA, 362 F.3d 861, 868 (D.C. Cir. 2004). On the other hand, it’s also possible that consent can transform a dispute from one that requires judicial resolution to one that doesn’t. See Sharif, 135 S. Ct. at 1960–70 (Thomas, J., dissenting). Likewise, these same points would seem to frame how to think about arbitration. See id. at 1968 n.6.

2. Agency Preemption of State Law. — Under current doctrine, agencies can preempt state law.313×313. See, e.g., Geier v. Am. Honda Motor Co., 529 U.S. 861, 886 (2000). In particular, either an agency can interpret its organic statute to preempt state law (and have that interpretation receive deference) or it can promulgate binding regulations based on a delegation of power from Congress (which wins out via the Supremacy Clause).314×314. Ernest A. Young, Executive Preemption, 102 Nw. U. L. Rev. 869, 881–96 (2008). And, as noted, agencies can exercise this preemptive power through either rulemaking or adjudication.315×315. See SEC v. Chenery Corp., 332 U.S. 194, 202–03 (1947). On “greater includes the lesser” grounds, there’s an intuitive first blush claim that these powers can provide for executive adjudication of state law.

But, on closer inspection, the power to preempt is really different in kind than the power to adjudicate. We’ve already seen this when it comes to the history of Article III. In fact, the Founders made sure that preemption decisions would be filtered through the separation of powers, rather than confined to a single branch. This was the cardinal lesson of the Convention’s choice to vote down Madison’s congressional veto. Under the adopted system, federal law would trump state law, but when there was doubt as to whether two schemes impermissibly clashed, state or federal judges would be the ones to ultimately decide which sovereign’s statute would prevail. For this reason, although Congress certainly has the power to preempt state law, nobody thinks, as we have seen, that it has the “lesser” power to create a tribunal in the Executive (or in Congress for that matter) to decide only state law cases.

Executive adjudication of state law potentially empowers agencies to circumvent this separation of powers framework. Specifically, treating the power to preempt as including the power to adjudicate leads to a situation where an agency can both (i) promulgate a federal regulation (like Congress), and (ii) define the meaning of state law as either broad enough to be preempted or narrow enough to be irrelevant (like a court). Such a regime rings a lot in the congressional veto proposed by Madison where Congress would’ve had the power to make law and also determine when state laws conflicted with those federal schemes. And the same structural intuition that led the Founders to reject that proposal cautions against accepting a “greater includes the lesser” argument here.

3. Foreign and International Tribunals. — Intermittently across American history, foreign or international courts have adjudicated certain domestic claims.316×316. Baude, supra note 17 (manuscript at 22–23). Most prominently, following a number of armed conflicts, the nation has set up mixed commissions to settle claims between Americans and foreign nationals.317×317. Henry Paul Monaghan, Article III and Supranational Judicial Review, 107 Colum. L. Rev. 833, 850–53 (2007). A good example is what followed the Jay Treaty. Under Article VI of the Treaty, an obviously non–Article III commission composed of Americans and British was put together to adjudicate outstanding debt claims.318×318. Baude, supra note 17 (manuscript at 16). Critically, a number of these cases were claims arising purely under American state law.319×319. Monaghan, supra note 317, at 854.

Do these episodes tank the historical claim of this Note? I hope not. Instead, due to their unique history and status, private claims against foreign nationals likely occupy a relatively sui generis doctrinal space. For one, as Professor Henry Monaghan has explained, these claims are better understood as public rights cases because they are bound up in issues of sovereign consent and, indirectly, sovereign immunity.320×320. Id. at 865–76. In particular, due to an established common law norm, these kinds of claims come with an inherent “congenital ‘espousal infirmity.’”321×321. Id. at 866. Because of this particular backdrop, the federal government has long claimed the power, confirmed by practice, to settle or even extinguish domestic claims against foreign countries or their citizens.322×322. See Dames & Moore v. Regan, 453 U.S. 654, 685–86 (1981). But such a power would be inconceivable for American claims that have formed under domestic law. For those reasons, these kinds of tribunals are better understood as an exception that proves the rule proposed here.

* * *

Like a well-working poker game, a well-functioning constitution requires participants to stick to both its written and unwritten rules. For us, that means keeping to every term along both the horizontal and vertical separations of power. But when it comes to executive adjudication of state law, the Supreme Court has failed to follow this injunction. Instead, the Court has focused exclusively on the horizontal dynamic between executive tribunals and the federal courts. This is a problem because it skips over the antecedent vertical question of federal power — a question that, in this instance, is dispositive. Indeed, the history and structure of the Constitution strongly indicate that Congress lacks the ability in the first place to empower executive tribunals to adjudicate state law. This because, at heart, the Founders came to a hard-fought compromise as to exactly when federal power could be used to affect rights and obligations arising under state law, and executive adjudication falls outside the terms of that original plan. As such, under current law, the federal government is taking more than it won in the original bargain. In short, it’s breaking the constitutional deal.