Harvard Law Review Harvard Law Review Harvard Law Review

Federal Arbitration Act

State Courts and the Federalization of Arbitration Law


In matters involving commercial arbitration, the nation’s laboratories of democracy have been shut down. The Supreme Court’s string of sweeping preemption decisions concerning § 2 of the Federal Arbitration Act1×1. 9 U.S.C. §§ 1–14. (FAA) has disabled many efforts at state regulation of arbitration. These decisions have also been roundly decried as inconsistent with federalism and congressional intent.2×2. See James Dawson, Comment, Contract After Concepcion: Some Lessons from the State Courts, 124 Yale L.J. 233, 233 (2014); Jeffrey W. Stempel, Tainted Love: An Increasingly Odd Arbitral Infatuation in Derogation of Sound and Consistent Jurisprudence, 60 U. Kan. L. Rev. 795, 795–96 (2012). Ironically, though, the state court–friendly jurisdictional provisions of the FAA ensure that state courts retain primary responsibility for interpreting and applying the Act.3×3. See Nitro-Lift Techs., LLC v. Howard, 568 U.S. 17, 17 (2012) (per curiam). Responding to the damage the Court’s FAA decisions have inflicted on state judicial and law enforcement systems, state courts have flouted those decisions in ways ranging from open defiance to arbitrary limitations on their reach.4×4. See Salvatore U. Bonaccorso, Note, State Court Resistance to Federal Arbitration Law, 67 Stan. L. Rev. 1145, 1156–65 (2015); Dawson, supra note 2, at 235–40.

This system of adjudicating FAA disputes has proven discombobulating for the federal system, and the result is a dynamic at odds with some basic commitments of the constitutional order. This Note argues that, absent legislative change, and despite powerful critiques of the Supreme Court’s interpretation of § 2, state courts should apply the FAA as a faithful federal court would. First, Article VI of the Constitution was drafted and ratified in contemplation of the type of recalcitrance state courts have displayed toward the Court’s FAA decisions. Second, the Court’s FAA case law and stare decisis doctrine clearly indicate that the Court’s current interpretation of the FAA is here to stay. In this environment, state courts that flout the Court’s FAA decisions contribute to maladministration of the FAA while doing little to preserve space for the operation of state law. Third, state court resistance to the FAA has compounded, rather than rectified, disruptions to the federal-state balance. It has contributed to uneven, unpredictable, and forum-dependent administration of federal law, led to opaque judicial decisionmaking, and undermined the legitimacy of the national legal system.

Part I lays a doctrinal and normative foundation. It first outlines the sweep of the Supreme Court’s FAA preemption cases, then offers observations on the disparate treatment of FAA issues between state and federal courts, and finally describes the stresses that the FAA has placed on states’ judicial and law enforcement systems. Part II documents the many methods state courts have used to dodge federal preemption. Part III makes the case against a “strict constructionist” approach to state court interpretation of the Court’s FAA decisions, instead advocating full faithfulness.5×5. Some caveats: This Note is not an empirical piece. It takes as given the prevailing observation in the literature that state courts are inventively avoiding the preemptive reach of the FAA and focuses on the normative implications of the current regime. Further, because Chapter 2 of the FAA, which applies to foreign arbitrations, confers much broader jurisdiction to federal courts, its provisions are outside the scope of this Note. See 9 U.S.C. §§ 203, 205.

I. The Awkward State-Federal Dynamic

Congress passed the FAA unanimously in 1925.6×6. See Margaret L. Moses, Statutory Misconstruction: How the Supreme Court Created a Federal Arbitration Law Never Enacted by Congress, 34 Fla. St. U. L. Rev. 99, 110 (2006). Historical context suggests, and most scholars agree, that Congress intended the law to be purely procedural.7×7. See Christopher R. Drahozal, In Defense of Southland: Reexamining the Legislative History of the Federal Arbitration Act, 78 Notre Dame L. Rev. 101, 103 (2002) (listing scholarly views). The statute contains no express preemption provision,8×8. Volt Info. Scis., Inc. v. Bd. of Trs. of Leland Stanford Junior Univ., 489 U.S. 468, 477 (1989). The Court’s reading of § 2 has become a sort of federal common law detached from the statutory text. Cf. Richard H. Fallon, Jr., et al., Hart and Wechsler’s The Federal Courts and the Federal System 635 (7th ed. 2015) (defining federal common law). and it is unlikely that the pre–New Deal Congress relied on a modern understanding of its Commerce Clause power in enacting the FAA.9×9. See Moses, supra note 6, at 109–10. Despite these observations, the modern Supreme Court has interpreted § 2 of the Act as a substantive commitment to a federal pro-arbitration policy that preempts state laws contrary on their face or in application.10×10. See AT&T Mobility, LLC v. Concepcion, 563 U.S. 333, 341–43 (2011).

A. Preemption of State Law Under the FAA

Section 2 of the FAA is the basis of the Supreme Court’s expansive preemption decisions. The statute provides that agreements to arbitrate “in any maritime transaction or a contract evidencing a transaction involving commerce . . . shall be valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract.”11×11. 9 U.S.C. § 2. While scholars and dissenting Justices have insisted that § 2 was designed to apply only in federal court,12×12. See, e.g., Preston v. Ferrer, 552 U.S. 346, 363 (2008) (Thomas, J., dissenting); Allied-Bruce Terminix Cos. v. Dobson, 513 U.S. 265, 284–85 (1995) (Scalia, J., dissenting); Southland Corp. v. Keating, 465 U.S. 1, 22–23 (1984) (O’Connor, J., dissenting); Moses, supra note 6, at 112. the Court has imbued the statute with a broad-reaching substantive commitment to enforcing arbitration agreements in both state and federal courts.13×13. See Southland, 465 U.S. at 10, 15; Moses, supra note 6, at 112. In doing so, the Court has effectively nullified any wisdom that state legislatures or courts might bring to bear on the increasing prevalence of arbitration clauses in contracts.

This section traces the Court’s expansion of the FAA, dividing the case law into four categories: (1) first-generation cases; (2) second-generation cases; (3) cases on procedural requirements in the formation of arbitration agreements; and (4) cases on separability doctrine. The first two categories track both substantive and chronological components of the case law: first-generation cases, generally decided before second-generation ones, address situations to which the FAA arguably extends explicitly, while second-generation cases address the FAA’s effect on the arbitration process.14×14. See Sarah Rudolph Cole, The Federalization of Consumer Arbitration: Possible Solutions, 2013 U. Chi. Legal F. 271, 271–72, 272 n.4; Christopher R. Drahozal, Federal Arbitration Act Preemption, 79 Ind. L.J. 393, 395 (2004). The latter two categories are treated separately because they do not squarely fit within the first two.

1. First-Generation Cases. — Southland Corp. v. Keating,15×15. 465 U.S. 1. a dispute over an arbitration clause in a franchise agreement,16×16. See id. at 3–4. began the process of federalizing state contract law. Chief Justice Burger, writing for a majority, held that “[i]n enacting § 2 of the [FAA], Congress declared a national policy favoring arbitration and withdrew the power of the states to require a judicial forum for the resolution of claims which the contracting parties agreed to resolve by arbitration.”17×17. Id. at 10. The Court asserted that Congress enacted § 2 pursuant to its Commerce Clause authority.18×18. Id. at 11. The Court also maintained that if the FAA applied in federal but not state court, the law would “encourage and reward forum shopping.”19×19. Id. at 15.

In dissent, Justice O’Connor read § 2 more narrowly, arguing that the statute’s legislative history conclusively established that it applies only in federal courts.20×20. See id. at 25 (O’Connor, J., dissenting). She discussed the text of other FAA provisions, such as §§ 3 and 4, that expressly apply only in federal (not in state) courts.21×21. See id. at 29. She also contended that Congress passed the FAA “specifically to rectify forum-shopping problems created by this Court’s decision in Swift v. Tyson.”22×22. Id. at 34 (citing 41 U.S. (16 Pet.) 1 (1842)). Southland also ushered in a curious contradiction: § 2 is the only provision of the United States Code that creates substantive rights that are enforceable — in most circumstances — only in state courts. Paul D. Carrington & Paul H. Haagen, Contract and Jurisdiction, 1996 Sup. Ct. Rev. 331, 381. As Justice O’Connor suggested in her Southland dissent, the existence of this divide is itself strong evidence that the Southland majority misinterpreted § 2. See Southland, 465 U.S. at 30 n.19 (O’Connor, J., dissenting).

Following Southland, several courts adopted an evasive device. These courts read the “involving commerce” phrase in § 2 to require parties to a contract with an arbitration clause to have actually contemplated an interstate arrangement.23×23. Allied-Bruce Terminix Cos., Inc. v. Dobson, 513 U.S. 265, 269–70 (1995) (collecting cases). In Allied-Bruce Terminix Companies v. Dobson,24×24. 513 U.S. 265. the Court rejected this test and expressly declined to overrule Southland,25×25. See id. at 272, 277–78. holding that § 2 exercises Congress’s Commerce Clause authority to its limit.26×26. Id. at 277. Thus, what mattered was not party intentions but whether the agreement involved interstate commerce in fact.27×27. Id. at 281. In a later case, the Court reversed a state court ruling that an arbitration agreement between an Alabama lender and an Alabama construction company did not involve interstate commerce.28×28. See Citizens Bank v. Alafabco, Inc., 539 U.S. 52, 53–54, 58 (2003) (per curiam).

2. Second-Generation Cases. — Perhaps the landmark second-generation case is AT&T Mobility LLC v. Concepcion.29×29. 563 U.S. 333 (2011). There, a consumer agreement mandated arbitration of any disputes that arose between the parties, but prohibited class proceedings.30×30. Id. at 336. The Concepcions nonetheless filed suit, pointing to a California Supreme Court decision that held class action waivers in adhesive consumer contracts unconscionable unless the party seeking arbitration demonstrated that bilateral arbitration was an adequate substitute for the deterrent effects of class actions.31×31. Id. at 337–38 (citing Discover Bank v. Superior Ct., 113 P.3d 1100, 1110 (Cal. 2005)).

The Supreme Court held that § 2 preempted the California rule.32×32. See id. at 344. Writing for the majority, Justice Scalia contended that “[r]equiring the availability of classwide arbitration interferes with fundamental attributes of arbitration and thus creates a scheme inconsistent with the FAA.”33×33. Id. That conclusion followed for two reasons, said the Court. First, the “principal advantage” of arbitration is procedural informality, an advantage that would be lost if arbitrators had to decide the ancillary issues that attend class litigation.34×34. Id. at 348. Second, the majority asserted that the lack of availability of an appeal from an arbitral award makes “[a]rbitration . . . poorly suited to the higher stakes of class litigation.”35×35. Id. at 350.

A later decision indirectly curtailed the ability of state courts to refuse enforcement of class action waivers. In American Express Co. v. Italian Colors Restaurant,36×36. 570 U.S. 228 (2013). the parties had entered an agreement providing for arbitration of disputes but prohibiting class arbitration.37×37. See id. at 231. Italian Colors opposed a motion to compel arbitration based on what the Court called the “effective vindication” theory.38×38. Id. at 235. Because the cost of litigating the claim would by far exceed individual recovery, Italian Colors argued, the class action waiver was invalid as a prospective waiver of a right.39×39. See id. The Court disagreed, citing Concepcion and again emphasizing the benefits of informality.40×40. Id. at 238 (citing Concepcion, 563 U.S. at 348). Although the Italian Colors action began in federal court, the Supreme Court later summarily vacated a state court decision applying the effective vindication theory to an arbitration agreement.41×41. See CarMax Auto Superstores Cal., LLC v. Fowler, 571 U.S. 1189, 1189 (2014) (mem.); Fowler v. Carmax, Inc., No. B238426, 2013 WL 1208111, at *7 (Cal. Ct. App. Mar. 26, 2013).

3. Separability Doctrine. — In Buckeye Check Cashing, Inc. v. Cardegna,42×42. 546 U.S. 440 (2006). the Court rejected an argument that the question of an arbitration clause’s severability from its container contract43×43. The term “container contract” refers to an agreement containing an arbitration provision. David Horton, Infinite Arbitration Clauses, 168 U. Pa. L. Rev. 633, 639 (2020). was one of state law. The Court held that, under the doctrine of separability, challenges to the validity of arbitration clauses are of two types: those that target the arbitration agreement specifically, and those that challenge the contract as a whole.44×44. Buckeye, 546 U.S. at 444. And if the challenge is to the agreement as a whole, the separability principle required that the arbitrator rather than a court consider the merits of the challenge, state law notwithstanding.45×45. See id. at 445–46. Two years later, the Court held that the FAA preempted a state law requiring an administrative agency rather than an arbitrator to consider a challenge to the validity of a contract containing an arbitration agreement. Preston v. Ferrer, 552 U.S. 346, 354 (2008). In Rent-A-Center, West, Inc. v. Jackson,46×46. 561 U.S. 63 (2010). the Court extended the separability doctrine to cover “delegation clauses” — clauses in the container contract that provide for arbitration of the arbitration agreement’s validity — even when a party specifically challenges the validity of the arbitration clause.47×47. See id. at 71–73.

4. Procedural Requirements in the Formation of Arbitration Agreements. — The Court’s preemption cases have also reached affirmative procedural requirements at the formation stage of arbitration agreements. In Doctor’s Associates, Inc. v. Casarotto,48×48. 517 U.S. 681 (1996). the Court invalidated a state rule mandating that arbitration clauses be set out in all capital letters on the first page of a contract, holding that the rule “singl[ed] out arbitration provisions for suspect status.”49×49. Id. at 687. Later, the Court held that the FAA preempted a state court–imposed requirement that a power of attorney agreement explicitly authorize the representative to enter an arbitration agreement.50×50. Kindred Nursing Ctrs., Ltd. v. Clark, 137 S. Ct. 1421, 1428 (2017).

B. Disparity and Its Causes

The Court’s interpretation of the FAA has eroded at least two officially held commitments of the legal order: first, that states remain important, if not primary, policymaking institutions in common law areas such as contract, tort, and property;51×51. See Patterson v. McLean Credit Union, 491 U.S. 164, 183 (1989). and second, that state courts are as trustworthy as federal courts when applying federal law in the first instance.52×52. See Burt v. Titlow, 571 U.S. 12, 19–20 (2013). While many courts have resisted the Court’s FAA preemption holdings, the scholarly consensus is that state courts are generally more willing than federal courts to limit or disregard those rulings. See, e.g., Aaron-Andrew P. Bruhl, The Unconscionability Game: Strategic Judging and the Evolution of Federal Arbitration Law, 83 N.Y.U. L. Rev. 1420, 1432–36 (2008). Of course, the Constitution does not assume perfect state court compliance with federal commands — hence, the need for a supreme federal tribunal with appellate jurisdiction over state courts. See Martin v. Hunter’s Lessee, 14 U.S. (1 Wheat.) 304, 347 (1816); infra p. 1194. But the text of Article III and the case law suggest that in general, state and federal courts ought to be treated as equally trustworthy arbiters of federal law in the first instance. The result is an arbitrary and flawed system of adjudicating FAA disputes that places stress on both the states’ policymaking apparatuses and the consistency of the national legal system, with no stable way to balance the tension. Yet despite state court interpretations of the FAA, in other contexts, available evidence suggests that state courts apply federal law as faithfully as federal courts do.53×53. See, e.g., Brett Christopher Gerry, Parity Revisited: An Empirical Comparison of State and Lower Federal Court Interpretations of Nollan v. California Coastal Commission, 23 Harv. J.L. & Pub. Pol’y 233, 285 (1999); Michael E. Solimine & James L. Walker, Constitutional Litigation in Federal and State Courts: An Empirical Analysis of Judicial Parity, 10 Hastings Const. L.Q. 213, 214–15 (1983). But see, e.g., William B. Rubenstein, The Myth of Superiority, 16 Const. Comment. 599, 599–600 (1999). So what explains the unique behavior of state courts in the FAA context? Insights from the literature on parity offer some answers that help contextualize and explain the peculiar institutional forces and other pressures on state courts to resist FAA preemption.

Article III of the Constitution mandates the existence of a federal Supreme Court but leaves the creation of lower federal courts to congressional discretion.54×54. U.S. Const. art. III, § 1. The decision to confer this discretion was the product of what is today known as the “Madisonian Compromise.”55×55. Fallon et al., supra note 8, at 8. As many have observed, a corollary of the Madisonian Compromise is that state courts are constitutionally competent to adjudicate federally created rights in the first instance.56×56. See, e.g., id. at 301. This argument has spawned a voluminous literature on the “parity” of state and federal courts.57×57. See generally, e.g., Symposium, Federalism and Parity, 71 B.U. L. Rev. 593 (1991). While some parity scholarship has focused on the extent to which state and federal courts should be considered constitutionally equivalent adjudicators of federal rights, this Note focuses on the empirical dimension of parity — the extent to which state and federal courts tend to reach the same outcomes in cases involving federal rights. And it operates on the premise that state courts are much less likely than the Supreme Court, and at least somewhat less likely than lower federal courts, to hold state arbitration policies preempted by the FAA.58×58. For a more complete defense of this proposition, see infra section III.C, pp. 1201–05.

The parity literature aids understanding of the state court response to FAA preemption by elucidating the influences that uniquely operate on, and motivate, state judges. In a seminal article, Professor Burt Neuborne endorsed a “weak” parity thesis, the crux of which held that federal courts were, on average, more solicitous of arguments affirming individual constitutional rights than are state courts.59×59. See Burt Neuborne, The Myth of Parity, 90 Harv. L. Rev. 1105, 1105–06 (1977) [hereinafter Neuborne, Myth]. Neuborne subsequently accepted the description of his thesis as an argument for “weak” disparity. Burt Neuborne, Parity Revisited: The Uses of a Judicial Forum of Excellence, 44 DePaul L. Rev. 797, 797 (1995). Neuborne posited three distinct reasons for this discrepancy. First, federal judges are generally more technically competent than state judges.60×60. See Neuborne, Myth, supra note 59, at 1121–24. Second, federal judges possess “a series of psychological and attitudinal characteristics” that make them more likely to enforce federal constitutional rights.61×61. Id. at 1124. Third, the independence of the federal judiciary better insulates it from any majoritarian pressures that elected state judges might feel.62×62. Id. at 1127–28. Today, these differences lie at the foundation of a dual, unequal, and highly forum-dependent system of deciding FAA disputes.

But the influences Neuborne identified do not always, or even often, seem to affect state court decisionmaking. So why is the FAA different? Return to Neuborne’s three observations. FAA preemption issues arise primarily in commercial disputes that tend to be relatively complex. And that complexity will likely disfavor the party on whom the burden falls to show the invalidity of a familiar, often democratically enacted state rule.63×63. Cf. id. at 1123. Moreover, state courts are unique in that they are guardians of not one but two constitutions: that of the state they serve and that of the federal government.64×64. David S. Schwartz, State Judges as Guardians of Federalism: Resisting the Federal Arbitration Act’s Encroachment on State Law, 16 Wash. U. J.L. & Pol’y 129, 130 (2004). Yet state courts are also more susceptible to majoritarian influence and less bureaucratically connected to the Supreme Court than are federal courts.65×65. See Neuborne, Myth, supra note 59, at 1124–25, 1127–28. Thus, state judges naturally feel more compelled to legitimate their state constitutions and repel perceived threats to state judicial systems. It is this fact, perhaps, that most affects state court treatment of the FAA. Undoubtedly, the Supreme Court’s interpretation of the FAA has caused outsized impacts not just on state laws, but also on the sphere of state judicial control.

C. The Burdens of FAA Preemption on States

Applying these lessons, this section identifies and considers three types of burdens that aggressive interpretation of the FAA has imposed on states. These burdens include: the subversion of fundamental state policies, the diminishment of state — especially state court — regulatory authority in an area of traditional state control, and the shifting of regulatory burdens to states’ nonjudicial public institutions. In each case, the onus places stress on state judicial systems’ abilities to preserve and organically develop the state judicial role in the constitutional scheme.

1. Subversion of Fundamental State Policies. — State courts often view themselves as the last and best barrier between federal interests and fundamental policies enshrined in state constitutions.66×66. See, e.g., State Constitutions: Freedom’s Frontier, Cato Inst. (2016), https://www.cato.org/policy-report/novemberdecember-2016/state-constitutions-freedoms-frontier [https://perma.cc/YR68-74GA]. This view can lead to destabilization of the federal-state balance in areas like arbitration, where the federal and state governments have divergent priorities. And indeed, a stark divide exists between the state and federal judicial systems on the importance of ensuring private litigants’ access to a judicial forum.

The federal Constitution guarantees due process, but does not mandate or even favor access to a public judicial forum. Several Supreme Court Justices have suggested that such a right does not exist.67×67. See, e.g., Bounds v. Smith, 430 U.S. 817, 840 (1977) (Rehnquist, J., dissenting); id. at 833–34 (Burger, C.J., dissenting). The FAA case law itself illustrates this lack of preference: in FAA cases, the Court routinely assumes that litigation and arbitration are equally acceptable methods for resolving disputes.68×68. David Horton, Clause Construction: A Glimpse into Judicial and Arbitral Decision-Making, 68 Duke L.J. 1323, 1329 & n.33 (2019). This assumed equality has become “a pillar of contemporary arbitration law.”69×69. Id. at 1329.

By contrast, a substantial majority of state constitutions guarantee some form of an individual right of access to courts.70×70. See Constitutional Access to Justice ProvisionsJudicial Administration State Links, Nat’l Ctr. for State Cts., https://www.ncsc.org/topics/judicial-officers/judicial-administration/state-links [https://perma.cc/P5XR-CEMD]. Far from assuming parity between judicial and private resolution, these provisions, taken literally, would seem to treat state judicial systems as superior to arbitration. Many state constitutions also mandate availability of a remedy for legal injuries.71×71. See id. While federal courts sometimes cite the Marbury v. Madison72×72. 5 U.S. (1 Cranch) 137 (1803). mantra that the existence of a right necessarily implies the availability of a remedy,73×73. Id. at 163. states — but not the federal courts74×74. Fallon et al., supra note 8, at 330. — have taken that principle more literally, elevating it to the status of constitutional enactments. Yet decisions like Concepcion and Italian Colors disable state courts from fully enforcing these state constitutional rights.

2. Diminishment of State Authority in an Area of Traditional Control. — It is hornbook law that state law governs contract-based disputes unless preempted by a federal statute.75×75. See Erie R.R. Co. v. Tompkins, 304 U.S. 64, 78 (1938); Cole, supra note 14, at 281. In many contexts, the Court has interpreted federal statutes narrowly to avoid encroaching on state contract law.76×76. See, e.g., Patterson v. McLean Credit Union, 491 U.S. 164, 183 (1989). These interpretations assume the desirability of states retaining a policymaking role in developing the law of contracts absent a clear legislative command to the contrary. They also tend to enable the realization of benefits possible only in a decentralized regulatory system.77×77. Cf. New State Ice Co. v. Liebmann, 285 U.S. 262, 311 (1932) (Brandeis, J., dissenting). And state courts, in line with the common law tradition, have played an outsized role in the development of state contract law.

The breathtaking scope of the FAA does not fit this picture. One observer describes the situation, without much hyperbole, as “the federal colonization of state contract law.”78×78. Dawson, supra note 2, at 233. The Court’s readings of the FAA have been especially disillusioning for state courts because they have transformed what state courts have viewed as one of their primary functions: developing contract law. When an FAA issue arises — which, under current law, is often — state courts are forced to be law takers rather than lawmakers in their typical common lawmaking guise.

3. Burden Shifting to States’ Public Institutions and Citizenry. — The Court’s FAA preemption decisions have limited states’ options for regulating various social and economic phenomena. The remaining options place the regulatory burden on states’ institutions or citizenries.79×79. Cf. Matthew J. Stanford & David A. Carrillo, Judicial Resistance to Mandatory Arbitration as Federal Commandeering, 71 Fla. L. Rev. 1397, 1418 (2019).

Consider the few remaining state regulatory avenues. Because nonjudicial branches of state government are not bound by arbitration agreements in the way that private parties before a court are,80×80. See, e.g., Rent-A-Center, Inc. v. Iowa C.R. Comm’n, 843 N.W.2d 727, 741 (Iowa 2014). state attorneys general may step in to rectify judicial underenforcement of state contract regulations.81×81. See Margaret H. Lemos, Privatizing Public Litigation, 104 Geo. L.J. 515, 577–78 (2016). But this mode of enforcement is not always efficient. Arbitration clauses and class arbitration waivers are so pervasive82×82. See Judith Resnik, Diffusing Disputes: The Public in the Private of Arbitration, the Private in Courts, and the Erasure of Rights, 124 Yale L.J. 2804, 2808 (2015). that it will inevitably prove difficult for public institutions with scarce budgets to effectively monitor and enforce states’ remedial aims. The added burden on states’ executive branches might prompt states to retain private counsel, an expensive option riddled with ethical conundrums.83×83. See, e.g., David A. Dana, Public Interest and Private Lawyers: Toward a Normative Evaluation of Parens Patriae Litigation by Contingency Fee, 51 DePaul L. Rev. 315, 315 (2001); Lemos, supra note 81, at 520–21. Moreover, the vast majority of state attorneys general are elected, and thus the availability of justice will turn in part on political calculation.84×84. See William P. Marshall, Break Up the Presidency? Governors, State Attorneys General, and Lessons from the Divided Executive, 115 Yale L.J. 2446, 2448 n.3 (2006). Another option, adopted by California’s Private Attorneys General Act85×85. Cal. Lab. Code §§ 2698–2699.5 (West 2014). (PAGA), is to empower injured private parties to sue in the name of the state. But this arguably unconstitutional approach86×86. See infra section II.3, pp. 1195–96. still burdens litigants by requiring that a significant part of the recovery go to the state.87×87. Iskanian v. CLS Transp. L.A., LLC, 327 P.3d 129, 147 (Cal. 2014).

Responding to these difficulties, there are growing calls for state legislatures to regulate the arbitration process to make it more closely resemble judicial process.88×88. See Alyssa S. King, Arbitration and the Federal Balance, 94 Ind. L.J. 1447, 1466–79 (2019). Like state courts, state legislatures have defied federal preemption case law by enacting state statutes in plain conflict with the FAA. See Sarah R. Cole, Uniform Arbitration: “One Size Fits All” Does Not Fit, 16 Ohio St. J. Disp. Resol. 759, 786–88 nn.93–100 (2001). But given the willingness of the Court to derive procedural prohibitions from § 2, it is questionable whether such proposals would be upheld. In any case, the Court’s FAA cases have reallocated burdens to the states themselves, and any costs of underenforced state norms of judicial access in contract disputes are borne by the states’ citizenries.

II. State and Federal Court Responses to FAA Preemption

Responding to the burdens that the Court’s interpretations of the FAA have placed on them, state courts continue to avoid the expansive reach of § 2 in numerous ways, some of them creative and others less subtle. Most of these tricks have been recounted elsewhere.89×89. See Dawson, supra note 2, at 235–40; Bonaccorso, supra note 4, at 1159–65. They include: (1) open defiance; (2) finding that an arbitration agreement does not “involve” interstate commerce; (3) permitting private parties to evade arbitration by initiating qui tam or analogous actions; (4) holding that a choice-of-law clause in an arbitration agreement incorporates state law contrary to the FAA; (5) defining “arbitration” under § 2 narrowly under state law; and (6) holding arbitration agreements void for unconscionability or duress. This Note makes two additional contributions to that list: (7) state courts’ unfaithful application of the Supreme Court’s separability precedents, and (8) state court–created procedural requirements for entry into a valid arbitration agreement. It is important to locate each of these methods of avoiding application of the FAA at different places on the spectrum of plausibility: some have been explicitly rejected by the Court, while others are merely dubious or questionable under Supreme Court precedent.

1. Open Defiance. — Some state courts have not hidden their derision for the Supreme Court’s preemption cases. Recall Casarotto. Before reversal by the Supreme Court, the Montana Supreme Court upheld a state law regulating arbitration agreements.90×90. See Casarotto v. Lombardi, 886 P.2d 931, 939 (Mont. 1994). Mincing no words, Justice Trieweiler authored a concurring opinion deriding federal judges for impinging on Montana law.91×91. See id. at 939 (Trieweiler, J., specially concurring). After reversal, two Montana justices dissented from their court’s remand of the case, calling the Supreme Court’s interpretation of § 2 “legally unfounded, socially detrimental and philosophically misguided.”92×92. Bruhl, supra note 52, at 1433. Later, the Alabama Supreme Court’s firebrand Chief Justice Roy Moore dissented from a judgment, arguing that the FAA did not apply in state courts despite Southland.93×93. See Selma Med. Ctr., Inc. v. Fontenot, 824 So. 2d 668, 677–78 (Ala. 2001) (Moore, C.J., dissenting). Concepcion spurred similar state court responses. In the year following Concepcion, the Supreme Court summarily reversed two state court decisions in plain conflict with the Concepcion holding.94×94. Bonaccorso, supra note 4, at 1147–48. In one of those decisions, a West Virginia state judge had criticized the Court’s “tendentious reasoning” in FAA cases.95×95. Id. (citing Brown ex rel. Brown v. Genesis Healthcare Corp., 724 S.E.2d 250, 278 (W. Va. 2011), rev’d sub nom. Marmet Health Care Ctr., Inc. v. Brown, 565 U.S. 530 (2012) (per curiam)).

2. Finding that a Contract Containing an Arbitration Clause Does Not “Involve” Interstate Commerce. — In Allied-Bruce, the Court held that § 2 of the FAA extended Congress’s Commerce Clause power “to the full.”96×96. Allied-Bruce Terminix Cos. v. Dobson, 513 U.S. 265, 277 (1995). In Citizens Bank v. Alafabco,97×97. 539 U.S. 52 (2003). the Court reminded observers that it indeed meant “to the full,” holding that a debt-structuring deal to be performed in Alabama between an Alabama lender and an Alabama construction company involved interstate commerce.98×98. See id. at 58. Of course, that holding was consistent with Supreme Court precedent interpreting Congress’s commerce power to be nearly limitless.99×99. See Wickard v. Filburn, 317 U.S. 111, 128–29 (1942). Yet since those decisions, some state courts have held that arbitration agreements do not “involve” interstate commerce.100×100. Bonaccorso, supra note 4, at 1159–60. In 2019, a New York court held that an agreement between a general contractor and an in-state resident for improvements to the resident’s property did not “involve” interstate commerce under § 2.101×101. Smith v. Nobiletti Builders, Inc., 177 A.D.3d 807, 810 (N.Y. App. Div. 2019).

3. Qui Tam or Analogous Actions. — Because arbitration is a creature of contract,102×102. Am. Express Co. v. Italian Colors Rest., 570 U.S. 228, 233 (2013). the existence of an arbitration agreement between two parties does not bar court actions initiated by a state agency, as long as the state was not a party to the agreement. Indeed, the Supreme Court and numerous state and federal courts have held that states may use their regulatory power to correct legal wrongs directly, even where an arbitration agreement would bar the victimized private party from initiating a lawsuit.103×103. See, e.g., EEOC v. Waffle House, Inc., 534 U.S. 279, 293–96 (2002); Commonwealth ex rel. Herring v. Net Credit Fin. Sols. of Utah, LLC, No. CL-2018-6258, 2019 WL 8883723, at *3 (Va. Cir. Ct. May 1, 2019).

Some state courts have stretched this principle to its limits. In California, PAGA authorizes private suits on behalf of the state for violations of the civil labor code. The theory of the statute is that the state cannot afford to bring administrative actions for all violations, and thus harmed employees may aid the state by initiating actions in the name of the state.104×104. Bonaccorso, supra note 4, at 1163. Judgments in PAGA litigation are binding on the state, and much of any recovery won in PAGA litigation is funneled into state coffers, but employees may collect a portion of any civil penalties imposed by a state court.105×105. See Iskanian v. CLS Transp. L.A., LLC, 327 P.3d 129, 151–52 (Cal. 2014). Although upheld by the California Supreme Court,106×106. See id. at 151. PAGA can be viewed either as a legitimate enforcement scheme or as a formalistic workaround of federal law, since an employee’s right to recover is itself conditioned on the terms of her employment contract, which might contain an arbitration clause.107×107. See Bonaccorso, supra note 4, at 1165. In a case arguably more consistent with Supreme Court precedent, the Third Circuit held that a state’s enforcement of its securities law was preempted by the FAA where the sole complainant was a party to an arbitration agreement.108×108. See Olde Disc. Corp. v. Tupman, 1 F.3d 202, 204, 209 (3d Cir. 1993).

4. A Choice-of-Law Clause in an Arbitration Agreement Incorporates State Law Contrary to the FAA. — In Volt Information Sciences, Inc. v. Board of Trustees of Leland Stanford Junior University,109×109. 489 U.S. 468 (1989). the Supreme Court permitted a state court to stay arbitration pending related litigation, holding that the FAA did not preempt state law where the arbitration agreement included a choice-of-law clause in favor of state procedural law.110×110. Id. at 470. But in later decisions, the Supreme Court pushed back on several state courts’ use of choice-of-law clauses to evade application of the FAA’s pro-arbitration policy.111×111. See DirecTV, Inc. v. Imburgia, 136 S. Ct. 463, 466 (2015); Mastrobuono v. Shearson Lehman Hutton, Inc., 514 U.S. 52, 58 (1995); see also Bonaccorso, supra note 4, at 1161–62. Since then, several state courts have invoked the Volt principle, see, e.g., Saheli v. White Mem’l Med. Ctr., 230 Cal. Rptr. 3d 258, 263–64 (Ct. App. 2018), but there is not widespread evidence that state courts continue to faithlessly limit the FAA by applying Volt.

5. Defining “Arbitration” Narrowly Under State Law. — Currently, courts are split on whether the meaning of “arbitration” in § 2 is defined according to federal common law or state law.112×112. Compare, e.g., Bakoss v. Certain Underwriters at Lloyds of London Issuing Certificate No. 0510135, 707 F.3d 140, 143 (2d Cir. 2013) (applying federal law), with, e.g., Hartford Lloyd’s Ins. Co. v. Teachworth, 898 F.2d 1058, 1061–62 (5th Cir. 1990) (applying state law). If the answer is state law, state courts might increasingly attempt to graft procedural requirements onto the definition of “arbitration,” thus finding any procedurally defective proceedings beyond the scope of the FAA.113×113. See Dawson, supra note 2, at 239–40; Christopher R. Drahozal, FAA Preemption After

Concepcion, 35 Berkeley J. Emp. & Lab. L. 153, 172–73 (2014). One California court concluded that a proceeding with a biased arbitrator was not an “arbitration” and thus that the FAA was inapplicable.114×114. See Cheng-Canindin v. Renaissance Hotel Assocs., 57 Cal. Rptr. 2d 867, 874 (Ct. App. 1996).

6. Holding Arbitration Agreements Void for Unconscionability and Duress. — Before Concepcion, state courts began hearing, and accepting, more arguments that arbitration agreements were unconscionable.115×115. See Bruhl, supra note 52, at 1437–42. But Concepcion held that unconscionability may not be applied in a way that treats arbitration differently from other matters of contract.116×116. See AT&T Mobility, LLC v. Concepcion, 563 U.S. 333, 341 (2011). State courts continue to invalidate arbitration agreements as unconscionable, not always consistently with Supreme Court instruction.117×117. For an especially dubious example of a state court not enforcing a class arbitration waiver post-Concepcion, see Brewer v. Mo. Title Loans (Brewer II), 364 S.W.3d 486, 487 (Mo. 2012) (en banc), especially Justice Price’s dissent, id. at 503 (Price, J., dissenting). For other examples, see Glob. Client Sols., LLC v. Ossello, 367 P.3d 361, 371 (Mont. 2016); and Figueroa v. THI of N.M. at Casa Arena Blanca LLC, 306 P.3d 480, 493 (N.M. App. 2012). And many state courts have accepted duress defenses to arbitration agreements.118×118. See Dawson, supra note 2, at 241 & n.52.

Several observers have noted the relationship between separability doctrine and defenses based on unconscionability or duress.119×119. See, e.g., Bruhl, supra note 52, at 1470–74; Dawson, supra note 2, at 239–40 & n.38. In Concepcion, Justice Scalia alluded to evidence that state courts were invoking unconscionability disingenuously in cases involving arbitration.120×120. Concepcion, 563 U.S. at 342–43. There might be good reason to apply separability differently to unconscionability defenses going to the entire contract, including the arbitration clause, than to duress defenses affecting both the arbitration clause and the container contract. Duress, unlike unconscionability, is a pure formation issue. And Buckeye expressly avoided deciding the question of whether courts may consider formation defenses going to the entire contract in the first instance. Buckeye Check Cashing v. Cardegna, 546 U.S. 440, 444 n.1 (2006). Perhaps because it can be difficult to prove that particular state court applications of unconscionability or duress unduly single out arbitration for skepticism, the Court has doubled down on separability.121×121. See Bruhl, supra note 52, at 1474–79. A rigorous separability doctrine ensures that more threshold questions are decided by arbitrators rather than hostile state courts. Of course, that very fact has led state courts to evade or ignore separability.

7. Unfaithful Application of Separability Doctrine. — State and federal courts are split on whether contract defenses relating to both the contract as a whole and its arbitration clause may be heard by a court.122×122. Dawson, supra note 2, at 241 n.47 (listing conflicting cases). Although many state courts have held that defenses to arbitration clauses must challenge the arbitration clause exclusively, some state supreme courts have simply ignored the separability issue and proceeded to invalidate the arbitration clause on a ground also pertinent to the container contract.123×123. See Narayan v. Ritz-Carlton Dev. Co., 400 P.3d 544, 549–51 (Haw. 2017); Brewer II, 364 S.W.3d 486, 492–96 (Mo. 2012).

Several state courts have applied Rent-A-Center to delegation clauses. Before Rent-A-Center, some state courts refused to sever delegation clauses from arbitration agreements, reasoning that arbitrators have a financial interest in finding a dispute to be arbitrable.124×124. Horton, supra note 68, at 1345. After the Supreme Court extended separability to delegation clauses, a few state courts dubiously held that alleged delegation clauses were insufficiently clear delegations to arbitrators.125×125. See Glob. Client Sols., LLC v. Ossello, 367 P.3d 361, 375–76 (Mont. 2016) (McKinnon, J., dissenting) (arguing the majority “manufacture[d] an ambiguity,” id. at 375, in the delegation clause and pointing to other contractual language evidencing an agreement to arbitrate arbitrability); Pinela v. Neiman Marcus Grp., Inc., 190 Cal. Rptr. 3d 159, 169 (Ct. App. 2015) (holding arbitration agreement invalid despite delegation clause providing that “[a]ny dispute concerning this Agreement . . . is subject to arbitration under this Agreement and shall be determined by the arbitrator”). In one egregious case, the West Virginia Supreme Court held that a clause providing that “arbitrator(s) shall determine all issues regarding the arbitrability of the dispute” was not a clear delegation.126×126. Schumacher Homes of Circleville, Inc. v. Spencer, 774 S.E.2d 1, 16 (W. Va. 2015) (Loughry, J., dissenting); accord id. at 14 (majority opinion). The Supreme Court summarily vacated the judgment.127×127. Schumacher Homes of Circleville, Inc. v. Spencer, 136 S. Ct. 1157 (2016) (mem.).

8. Formation Requirements. — The Court has struck down several states’ attempts to impose special formation requirements on arbitration agreements.128×128. See supra section I.A.4, p. 1189. Some states still have statutes on the books that impose similar requirements.129×129. See, e.g., Ohio Rev. Code Ann. § 2711.23 (West). It is not clear whether all or any such statutes remain valid after Casarotto and its progeny.

III. A Nationalist Critique of State Court Resistance to FAA Preemption

The Court’s FAA preemption cases have engendered uniquely broad consensus about their incorrectness. But to describe this problem is not to resolve it. While the Supreme Court’s § 2 cases have severely constrained states’ abilities to regulate wide swaths of economic life, state court recalcitrance has subverted important normative ideals of the national legal order. Emphasizing the real burdens that the FAA imposes on states, the Court’s critics have suggested various ways in which state courts should seek to skirt the reach of FAA precedents, calling for something of a “strict constructionist” approach to interpreting these precedents.130×130. Schwartz, supra note 64, at 142; see also Bonaccorso, supra note 4, at 1170–72; Dawson, supra note 2, at 247. Whatever the wisdom of such calls might have been before today, they are no longer prudent or viable. The Court, rather than retreating in the face of sharp critiques, has doubled down on FAA expansion. Innumerable private contracts have been entered in reliance on the Court’s reading of the statute. These factors, along with the especially strong pull of stare decisis in statutory interpretation cases and the context of state court defiance, would complicate any attempt by the Court to retreat. At this juncture, state court insubordination exacerbates the FAA’s harms to the constitutional order while doing little to properly recalibrate the federal-state balance.

The undesirability of this situation should be acknowledged. But ultimately, the most appropriate response for state courts is to bite the bullet and enforce the FAA as a faithful federal court would. Constitutional text and history, the whole range of post-Southland developments, and the legitimate aims of the national legal system all support this course of action. The many and serious problems caused by the Court’s rewriting of § 2 should, and can, be resolved by other institutions — institutions that do not share the features and obligations unique to the judicial function.

A. Constitutional Context

From the republic’s birth, ensuring the fealty of state judges to federal commands has been a matter of constitutional concern. Article VI of the Constitution highlights the Framers’ particular interest in requiring state court compliance with federal law. Its Oaths Clause mandates that “Members of the several State Legislatures, and all executive and judicial Officers” take an oath to support the Federal Constitution.131×131. U.S. Const. art. VI, cl. 3. Article VI’s Supremacy Clause goes a step further by singling out state judges, bluntly providing that “the Judges in every State shall be bound” by federal law.132×132. Id. art. VI, cl. 2. The Framers’ special concerns about state courts derived from the nation’s abysmal experience with state courts under the Articles of Confederation. After the new nation concluded the Treaty of Paris ending the Revolutionary War, state legislatures and courts found innovative ways to frustrate the aims of the federally negotiated peace.133×133. See Wythe Holt, “To Establish Justice”: Politics, the Judiciary Act of 1789, and the Invention of the Federal Courts, 1989 Duke L.J. 1421, 1440 & n.58, 1441, 1446–49, 1450–51 & n.94. State court disputes relating to the Treaty of Paris continued well after the ratification of the Constitution. Martin v. Hunter’s Lessee,134×134. 14 U.S. (1 Wheat.) 304 (1816). a case famous for its holdings on federal court jurisdiction and the supremacy of federal law, arose from a Virginia state court’s outright refusal to consider and apply the treaty’s provisions.135×135. See id. at 323–24.

Like state court treatment of the federally guaranteed contract rights of the Treaty of Paris, the state response to FAA-created contract rights has been solicitous of local desires at the expense of federal interests. Opposition to the Treaty of Paris stemmed from arguably legitimate grievances against the emerging merchant class and the wartime atrocities committed by the British.136×136. See Holt, supra note 133, at 1435–37. But then, as now, constitutional text and decisional law impose on state courts an obligation to take federal law as they find it and apply it faithfully, their own views to the contrary notwithstanding.137×137. See, e.g., U.S. Const. art. VI, cl. 2–3; Burt v. Titlow, 571 U.S. 12, 20 (2013).

B. Stare Decisis and Post-Southland Developments

Starting from the position that Southland was wrongly decided and that its applications and extensions have proved damaging, the argument for continued state court resistance to Southland must have more to stand on. While it is of course true that “stare decisis is not an inexorable command,”138×138. South Dakota v. Wayfair, Inc., 138 S. Ct. 2080, 2096 (2018) (quoting Pearson v. Callahan, 555 U.S. 223, 233 (2009)). the national legal system’s aspiration to finality is a vital consideration,139×139. Cf. Paul M. Bator, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners, 76 Harv. L. Rev. 441, 451–52 (1963). and is especially so in the FAA context.

Applying stare decisis is a messy business and inescapably involves normative ideas not all will share. Doctrinally speaking, however, two propositions are clear, settled, and relevant. First, whatever the import of stare decisis generally, there is a hefty rule of stare decisis in statutory interpretation cases.140×140. See Kimble v. Marvel Ent., LLC, 576 U.S. 446, 456 (2015); William N. Eskridge, Jr., Overruling Statutory Precedents, 76 Geo. L.J. 1361, 1362 (1988). The justification for that rule rests, at bottom, on a separation of powers rationale: Congress has the authority to amend or repeal the law if the Court misreads legislative instruction, and imposing a “super strong” burden on revisiting the Court’s reading of a statute encourages Congress to perform its functions of overseeing the court system and ensuring its will is effectuated.141×141. Amy Coney Barrett, Statutory Stare Decisis in the Courts of Appeals, 73 Geo. Wash. L. Rev. 317, 327 (2005); see also id. at 323, 325. Second, the force of precedent is “at [its] acme” in cases involving property and contract rights.142×142. Kimble, 576 U.S. at 457 (quoting Payne v. Tennessee, 501 U.S. 808, 828 (1991)). Courts do not operate in a vacuum, and private parties in the commercial world often arrange their affairs according to extant understandings of their rights and obligations.

These considerations, coupled with the Court’s post-Southland readings of the FAA, lead to two conclusions. First, the Court has long since passed the point of no return to a purely procedural reading of § 2. The necessary justification for a retreat would need to be doubly weighty: it would need to explain why the Southland error is so grave that it overcomes the heightened burdens associated with statutory- and contract-rights holdings. Virtually none of the Court’s decisions over the last few decades, except arguably Volt, provide the basis for limiting or dispensing with Southland’s substantive commitments. A retreat in the face of widespread state resistance would signal acquiescence to political and judicial pressure — a perception the Court increasingly strives to avoid.143×143. See, e.g., Jeffrey Rosen, John Roberts Is Just Who the Supreme Court Needed, The Atlantic (July 13, 2020), https://www.theatlantic.com/ideas/archive/2020/07/john-roberts-just-who-supreme-court-needed/614053 [https://perma.cc/PF8W-PZS5]. Second, state courts gain little, and do damage, by misapplying the FAA. Rather than paring back § 2 in the face of criticism, the Court’s response has been to double down. While a “strict constructionist” reading of the Court’s FAA preemption decisions might have been defensible years ago, today it is not. Few gaps in the doctrine remain left to exploit, and state court resistance contributes to the perception and reality of uneven administration of the law.

Of course, the error of Southland and its progeny is, in a sense, of constitutional dimension — as is any error of statutory interpretation.144×144. See Puerto Rico v. Franklin Cal. Tax-Free Trust, 136 S. Ct. 1938, 1949 (2016). And from time to time, the Court has recognized and rectified shocks to the federal system, even where Congress could have intervened but chose not to.145×145. See generally South Dakota v. Wayfair, Inc., 138 S. Ct. 2080 (2018); Erie R.R. Co. v. Tompkins, 304 U.S. 64 (1938). But the Court has never effected a 180-degree reversal in a context mirroring its relentless four-decade project of expanding the FAA, sometimes in unanimous decisions.146×146. See, e.g., Henry Schein, Inc. v. Archer & White Sales, Inc., 139 S. Ct. 524, 528 (2019). The only arguably comparable situation is the Court’s overruling of Abood v. Detroit Board of Education, 431 U.S. 209 (1977), in Janus v. AFSCME, Council 31, 138 S. Ct. 2448, 2460 (2018). But Janus concerned a separate constitutional provision (the First Amendment), the reliance interests induced by Southland and its progeny might exceed even the strong reliance interests induced by Abood, and the Court did not as consistently and relentlessly expand Abood as it has Southland. Until or unless it does, state courts that flout the clear reach of the Court’s decisions will only compound the constitutional imbalance.

C. National Commitments

The post-Southland pattern has generally followed a predictable cycle: state courts invent new ways to push back on the statute’s expanding reach, the Court rebukes those attempts at pushback, and state courts begin the process anew. Repetition of this pattern has significantly decreased available opportunities for limiting or evading the FAA in good faith. Yet many state courts continue to ignore or work around the law. Today such attempts produce little in terms of promoting the concrete interests of state laws or state judicial systems. The more obvious state courts are about defying federal law, the more likely they are to be summarily147×147. See, e.g., CarMax Auto Superstores Cal., LLC v. Fowler, 571 U.S. 1189, 1189 (2014) (mem.). or unanimously148×148. See, e.g., Marmet Health Care Ctr., Inc. v. Brown, 565 U.S. 530, 534 (2012) (per curiam); Citizens Bank v. Alafabco, Inc., 539 U.S. 52, 58 (2003) (per curiam). reversed. Even if an unfaithful state court decision avoids the Court’s intervention, the decision is likely only to delay, rather than prevent, federalization of the state rule.149×149. There are many examples of this pattern. Consider just a couple: In Allied-Bruce, the Court rejected the “contemplation of the parties” test many state courts had adopted to limit the reach of Southland. Allied-Bruce Terminix Cos. v. Dobson, 513 U.S. 265, 278 (1995); see also id. at 269–70 (collecting cases). And after several state courts applied the “effective vindication” theory in response to Concepcion, they were forced to backtrack after the Court’s decision in Italian Colors. See, e.g., Machado v. System4 LLC, 993 N.E.2d 332, 333 (Mass. 2013). In the meantime, resistance to the FAA frustrates legitimate aspirations of the national legal system. These normative commitments include uniformity and predictability, transparency, and promotion of the legal system’s legitimacy.

1. Uniformity and Predictability. — Parity is many things to many people — a dogma, an aspiration, an empirical question. But because parity remains an assumption of the national judicial system,150×150. See sources cited supra note 52. it is worth taking seriously as an imperative. It is fairly clear, if not definitively proven, that state courts are less likely to enforce arbitration clauses than lower federal courts.151×151. See supra note 52. There is clear evidence that the Supreme Court is more likely than state courts to funnel disputes out of the court system and into arbitration. See supra Part I, pp. 1185–94. There is also strong, though not conclusive, evidence that lower federal courts have interpreted the FAA in ways that counter the anti-arbitration influence of state courts. Professor Aaron-Andrew Bruhl finds one example of a more pro-arbitration trend in the federal courts in lower federal courts’ treatment of separability issues. Bruhl, supra note 52, at 1479–80, 1480 n.244. The disparity between state and federal adjudication of FAA rights threatens the chief virtues of empirical parity in the legal system: consistency and predictability of outcomes. State court resistance to the FAA makes enforcement of litigants’ federally recognized rights highly forum-dependent, despite the Southland rationale that a substantive reading of § 2 would help avoid forum shopping.152×152. Southland v. Keating, 465 U.S. 1, 15 (1984). In inventing novel ways to skirt the increasing federalization of state contract law, state courts have sanctioned differential treatment of similarly situated litigants. Such treatment is antithetical to the ideals of the fair and impartial administration of justice.

Some would contend that this type of judicial bias is simply part of a federalist judicial system, and that correcting it is a fanciful aspiration. Professor Michael Wells, for instance, asserts that the ideal of the unbiased judge who eschews even “personal qualities acquired in part from the institutional features of the court” — like the qualities a state judge possesses by virtue of serving on a state court — is an unrealistic and thus unproductive aspiration.153×153. Michael Wells, Behind the Parity Debate: The Decline of the Legal Process Tradition in the Law of Federal Courts, 71 B.U. L. Rev. 609, 643 (1991); see also id. at 641–44. But this critique unduly minimizes the documented capability of human beings — and, by extension, judges — to recognize and correct for their biases, a project to which this Note is devoted.154×154. See, e.g., Adam Hoffman, Can Science Help People Unlearn Their Unconscious Biases?, Smith. Mag. (July 2, 2015), https://www.smithsonianmag.com/science-nature/can-science-help-people-unlearn-their-unconscious-biases-180955789“>https://www.smithsonianmag.com/science-nature/can-science-help-people-unlearn-their-unconscious-biases-180955789/”>https://www.smithsonianmag.com/science-nature/can-science-help-people-unlearn-their-unconscious-biases-180955789 [https://perma.cc/2RUK-6AT5].

2. Transparency. — State courts’ resistance to FAA preemption is often pretextual, as illustrated by the stratagems detailed above. Faithful adherence to FAA precedent would eliminate the need for pretext and result in more transparent decisionmaking. Although most would consider transparency in legal reasoning to be a worthy ideal, whether transparency is an obligation,155×155. See Richard H. Fallon, Jr., Essay, A Theory of Judicial Candor, 117 Colum. L. Rev. 2265, 2292–306 (2017). especially an absolute obligation, is relatively controversial.156×156. See Micah Schwartzman, Essay, Judicial Sincerity, 94 Va. L. Rev. 987, 988 (2008). Although fulsome treatment of these issues is impossible here, a more faithful state court approach to FAA cases would likely produce more transparent decisionmaking. Increased transparency in FAA cases is desirable for many reasons, which fall under two categories: reasons intrinsic to the judicial process and reasons external to that process.157×157. See id. at 989–91. To say that judicial transparency is superior to judicial opacity is not to say that state courts’ open defiance of the Court’s FAA decisions is desirable. Still, the virtue of candid insubordination is that it can be easily identified and corrected. See, e.g., supra p. 1195.

Extrinsic justifications for judicial candor focus on the benefits that candor produces. For instance, a requirement of reason-giving in judging “serves a vital function in constraining the judiciary’s exercise of power.”158×158. David L. Shapiro, Essay, In Defense of Judicial Candor, 100 Harv. L. Rev. 731, 737 (1987). But because extrinsic proposals rest on contestable empirical judgments, the intrinsic justification for FAA candor is more powerful. The intrinsic line of argument focuses on the judiciary’s, and society’s, power of force over litigants.159×159. See Schwartzman, supra note 156, at 990. Judges, as stewards of that power, have a fiduciary responsibility to disclose the real reasons for their decisions because people have a right to know why they are bound by law, and thus why their freedom has been curtailed.160×160. See Ethan J. Leib et al., Essay, A Fiduciary Theory of Judging, 101 Calif. L. Rev. 699, 738–40 & n.205 (2013). On this view, the nature of judicial process itself imposes on judges an obligation of sincerity.161×161. Schwartzman, supra note 156, at 990.

Lamentably, many state courts have been opaque in their avoidance of FAA preemption. Consider Smith v. Nobiletti Builders, Inc.,162×162. 177 A.D.3d 807 (N.Y. App. Div. 2019); see supra p. 1195. a case in which a New York appellate court held an arbitration clause invalid under state law, blithely concluding that the FAA did not preempt state law because the contract did not involve interstate commerce.163×163. Smith, 177 A.D.3d at 810. The court’s conclusion was so implausible in light of settled law that it suggests the court did not disclose — because it did not want to — the real reasons for its decision.164×164. See supra section II.2, p. 1195. Consider also the instances in which state courts have simply ignored separability issues before beginning analysis of an arbitration agreement’s validity.165×165. See supra section II.7, pp. 1197–98. These state courts underperform their obligation to “make it intelligible to a reasonable reader who was acquainted with relevant law . . . how they could regard the reasons that they adduce in support of a decision as legally adequate under the circumstances.”166×166. Fallon, supra note 155, at 2293. Insufficient explanations are also missed opportunities to candidly evaluate the current system for adjudicating FAA cases: transparency holds potential to help create conditions for thoughtful dialogue, and constructive give-and-take, between state and federal courts.167×167. Cf. David S. Han, Transparency in First Amendment Doctrine, 65 Emory L.J. 359, 362 (2015). Further, transparency might assist Congress by illuminating undesirable systemic issues in FAA adjudication should it choose to intervene.

3. Promoting the Legitimacy of the Legal System. — The rule of law depends on the legitimacy, both real and perceived, of its commands, and state courts’ rejection of binding law undermines that legitimacy in several ways. Professor Richard Fallon identifies three different concepts of legitimacy: legal, sociological, and moral.168×168. Richard H. Fallon, Jr., Legitimacy and the Constitution, 118 Harv. L. Rev. 1787, 1790 (2005). Implausible state court applications of the Supreme Court’s FAA decisions implicate all three categories.

While legal legitimacy is hard to define, legitimacy and legality are not one and the same: illegitimacy is a charge typically directed only at the most condemnable judicial decisions.169×169. See id. at 1794–95. State courts that have openly defied FAA preemption have intimated that Southland is deserving of such disapprobation.170×170. See supra section II.1, pp. 1194–95. But those decisions tended to ignore the commands of stare decisis and federal supremacy, both of which are constitutionally rooted, and thus legally legitimate, doctrines.171×171. See Richard H. Fallon, Jr., Essay, Stare Decisis and the Constitution: An Essay on Constitutional Methodology, 76 N.Y.U. L. Rev. 570, 579–81 (2001). Ironically, then, such decisions are probably more susceptible to charges of illegitimacy than the Supreme Court’s preemption cases.

Moreover, legal institutions are sociologically legitimate if the public perceives them as worthy of adherence “for reasons beyond fear of sanctions or mere hope for personal reward.”172×172. Fallon, supra note 168, at 1795. Stare decisis promotes so-ciological legitimacy by “assuring the public that it is ruled by law so conceived.”173×173. Charles Fried, Commentary, Constitutional Doctrine, 107 Harv. L. Rev. 1140, 1156 (1994). In disobeying the twin commands of stare decisis and federal supremacy, state court resistance to the FAA fosters the troubling idea that law changes based on the ideologies of legal actors.

Finally, widespread state court insubordination undermines continuing moral justifications for the existence of the current constitutional order.174×174. See Fallon, supra note 171, at 588–89. By displaying open resistance and rendering the law of arbitration less predictable, state courts have contributed to the destabilization of the national legal system’s workability. If, as most would agree, “a good legal system requires reasonable stability,”175×175. Id. at 585. then state court decisions that attempt to reopen closed questions, or implausibly apply settled law, impair a key component of the national legal system’s claim to continuing moral legitimacy.


Getting things right matters, but it is not the only aim of a mature and cohesive legal system. For nearly forty years now, Southland and its progeny have endured harsh scholarly and judicial criticism, but to no avail. While a judicial retreat or a strict constructionist approach to interpreting § 2 might have been plausibly defensible at one time, they are no longer so. A substantive FAA is now our law, and state courts remain primarily responsible for applying it. The only remaining question is whether, absent congressional participation, state courts will accept the lamentable but unavoidable federalization of state contract law.