Saving state law bad-faith claims from preemption

Most lower courts continue to void state law remedies for ERISA plan participants. But a few federal district court judges are reevaluating the act’s preemptive effect on state law insurance bad-faith claims in light of recent U.S. Supreme Court analyses.

Donald T. Bogan
Reprinted with permission of TRIAL (April 2003)
Copyright the Association of Trial Lawyers of America

Extracting congressional intent from the Employee Retirement Income Security Act’s (ERISA)1 “antiphonal” preemption clauses continues to frustrate the U.S. Supreme Court. In a recent case, Justice David Souter, writing for the majority, noted that ERISA’s express language “seems simultaneously to preempt everything and hardly anything.”2

Applying early Supreme Court precedent, most courts have held that ERISA preempts state law bad-faith claims arising from ERISA-governed health care and disability benefit plans, even though the act itself does not substitute comparable federal remedies. The righteousness and strength of the crusade for consumer protections, however, offer hope that courts may yet reinvigorate the bad-faith remedy in cases where plan participants struggle to receive fair treatment from ERISA-plan insurers and HMOs.3

ERISA expressly supersedes “any and all state laws [that] . . . relate to any employee benefit plan” (the preemption clause), except “any law of any state which regulates insurance” (the savings clause). However, self-insured plans are not “deemed” to be insurance companies or insurance contracts for purposes of any state laws that purport to regulate insurance (the deemer clause).4

In 1987, during the era of expansive ERISA preemption, the Supreme Court delighted the insurance industry by ruling that the statute preempted a state common law bad-faith claim filed against an ERISA-plan insurer. In Pilot Life Insurance Co. v. Dedeaux, the Court held that Mississippi’s common law bad-faith remedy “relate[d] to” an ERISA plan, and, therefore, fell within the scope of the preemption clause.5 The Court also found that the bad-faith law was not saved from preemption as a law that regulates insurance because, in Mississippi, the bad-faith remedy is available for any aggravated breach of contract, not just the breach of an insurance contract.6

The Pilot Life Court applied a three-factor test—developed under cases interpreting the phrase “business of insurance” in the McCarran-Ferguson Act 7—to define which state laws regulate insurance. The Court explained that a state law “regulates the business of insurance” when the law affects the spreading of policyholder risk, affects an integral part of the insurer/insured policy relationship, and specifically targets the insurance industry.8

The Court found that Mississippi’s law affected the insurance industry but did not specifically target it; did not affect an integral part of the insurer/insured policy relationship any more than other applications of a state’s general contract law; and did not spread policyholder risk.9 Following this narrow, anticonsumer interpretation of ERISA’s express savings clause, the Court further increased the burden on working families by recognizing a new inference of preemption arising from the statute’s civil enforcement provision—§502.

That section details who can sue under the statute and what relief is available.10 The Court concluded that the “comprehensive” remedies in §502 provided evidence that Congress intended ERISA to broadly preempt state law, suggesting that the remedies listed were the only ones available in actions arising from an ERISA-governed employee benefit plan.11

After that decision, until 1999, each federal circuit court that addressed ERISA’s preemptive effect on state common law bad-faith or statutory unfair-settlement-practices claims filed against an ERISA-plan insurer held that the statute superseded the st‚te law remedies, either because the state law was not expressly saved from preemption or because the remedy conflicted with §502.12

Changing tide

As the Supreme Court began to retreat from its expansive view of ERISA preemption—beginning in 1995 with New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Insurance Co.13—some federal judges began to limit the statute’s preemption powers.14 Several district courts have now ruled that ERISA does not preempt the bad-faith or unfair-settlement-practices laws of various states.15

Each of these decisions noted that the savings-clause analysis in Pilot Life was unique to that case because of the Mississippi law involved. The courts issuing these decisions have found that bad-faith laws affect an integral part of the insurer/insured policy relationship and that when the remedy can be pursued only against the insurance industry, the laws are saved from preemption because they do regulate insurance.16

Unfortunately, none of these decisions has addressed Pilot Life’s suggestion that ERISA may impliedly preempt state law bad-faith actions because the remedy conflicts with §502. And this new line of cases presents questions that the Supreme Court has not yet answered: What happens when a state law that regulates insurance also creates a remedy aimed solely at the insurance industry? Does ERISA’s implied preemption of state law remedies trump the statute’s express exception to preemption for state insurance regulations?17

In 1999, in a footnote to its decision in UNUM Life Insurance Co. v. Ward, the Supreme Court acknowledged the evolving argument to limit Pilot Life.18 But the preemption question in Ward did not trigger a full exploration of the interplay between implied preemption under §502 and the exception to preemption for laws that regulate insurance in §514. Then, this past year, it appeared that the Court might conclusively decide the §514 versus §502 question in Rush Prudential HMO, Inc. v. Moran.19

Moran involved ERISA’s possible preemption of an Illinois HMO reform law. Under that law, if an HMO denied coverage for treatment recommended by a patient’s physician, the patient could demand an independent review of the coverage denial. Further, the Illinoiu statute required the HMO to abide by the independent reviewer’s conclusion if the reviewer sided with the patient.

Rush Prudential asserted that the Illinois law was not saved from ERISA preemption as a law that regulated insurance because HMOs are not insurers and because the law effectively provided the patient a state law remedy in conflict with ERISA §502. In Moran, as in Ward, the Court did not reach the §514 versus §502 conflict because it found that the independent-review statute did not provide an alternative remedy.

In dicta, however, the Moran Court mused that if ERISA’s express savings clause conflicts with implied preemption under §502, the inference of preemption arising from that section is so strong that it would trump even ERISA’s express savings-clause language.20

I suggest that the Moran dicta is incorrect and that it conflicts with the Court’s 1983 decision in Franchise Tax Board v. Construction Laborers Vacation Trust, the first Supreme Court opinion to discuss implied preemption of state law remedies under ERISA §502.21 Recognizing the Court’s continued movement to limit the scope of ERISA preemption of state law in non-pension-plan cases,22 the extent of the statute’s possible preemption of state law bad-faith and unfair-settlement-practices claims remains ripe for review—particularly in view of the conflict between Franchise Tax Board and the Moran dicta.

If a state common law bad-faith or unfair-settlement-practices claim applies only against the insurance industry and affects an integral part of the insured/insurer relationship, such that the state law falls within ERISA’s savings-clause exception, is the state law remedy nevertheless impliedly preempted by §502?

For the reasons outlined below, the clear answer is no.

15 arguments

Section 502 is not the sole remedy. In Franchise Tax Board, the Supreme Court ruled that implied preemption under §502 is limited by the express savings clause.23 In that case, the California tax collection authority sued in state court to enforce its statutory right to levy against an ERISA vacation trust that held assets of several delinquent California taxpayers. The defendant asserted that ERISA expressly superseded (under §514) the California law allowing the state to levy. The defendant also removed the action to federal court under the complete preemption doctrine, arguing that §502 provided the exclusive remedies arising in connection with an ERISA-governed employee benefit plan.

The Court did not decide the express-preemption question. Instead, it held that the federal court did not have removal jurisdiction under the exception to the well-pleaded complaint rule—established in Avco Corp. v. Aero Lodge No. 735, International Association of Machinists & Aerospace Workers24—because, the Court found, §502 did not provide the sole remedy for all claims arising in connection with an ERISA plan.25

Specifically, the Court held that entities which are not allowed to sue under §502 (like California’s Franchise Tax Board) may sue under state law.26 Similarly, the Court concluded that state law remedies protected by the savings clause would defeat implied preemption under §502 because the savings clause expressly limits whatever inference of preemption may arise from that section.27

Conflict preemption is appropriate. Franchise Tax Board instructs that §502 completely preempts only state law remedies that fall within the scope of that section. This suggests that courts should apply “conflict” preemption analysis, rather than “field” preemption analysis, in determining implied preemption under ERISA §502.

Under the conflict-preemption doctrine, a federal statute preempts only state laws that directly conflict with or frustrate the purpose of the federal act; supplemental and complementary state laws are not preempted. Under field-preemption principles, a federal law will totally occupy the field governed by the federal statute, preempting even complementary state laws that serve the very same purposes as the federal act.28

State law tort claims for extracontractual damages do not fall within the scope of §502 (and therefore do not conflict with it) because that section essentially provides only contract remedies for ERISA plan participants when they contest the denial of a claim for benefits. A state law tort remedy does not directly conflict with the contract remedy afforded in §502. Instead, it complements and supplements the §502 remedy in accord with ERISA’s overall purpose to protect employee benefit plan participants.29

In Roark v. Humana, Inc., the Fifth Circuit recently applied a similar rationale to hold that a state law negligence claim filed against an ERISA-governed HMO is not removable to federal court because the tort claim is not impliedly preempted under §502.30 Applying complete preemption analysis, the Roark court found that §502 provides primarily contract remedies for plan participants.

Because the plaintiff’s negligence claim against the HMO did not duplicate the remedies available under §502, the Roark court held, the state claim did not fall within the scope of ERISA’s civil enforcement provision and was not completely preempted.

The Moran dicta are incorrect. The Moran Court said, “Although we have yet to encounter a forced choice between the congressional policies of exclusively federal remedies and the ‘reservation of the business of insurance to the states,’ we have anticipated such a conflict, with the state insurance regulation losing out if it allows plan participants ‘to obtain remedies that . . . Congress rejected in ERISA.’”31 The Court then cited Pilot Life as the case in which it had anticipated the conflict.

The Moran Court’s reference to Pilot Life is incorrect, and it undermines the credibility of the Court’s dicta conclusions. Pilot Life did not present the §514 versus §502 conflict, because the Court specifically held that the state law at issue was not a law that regulates insurance. Since the Court found that Mississippi’s bad-faith law did not regulate insurance, the savings clause was not implicated. Consequently, Pilot Life presented no conflict between the savings clause and §502.

Similarly, in the companion case to Pilot Life—Metropolitan Life Insurance Co. v. Taylor—the state laws at issue, common law contract and tort claims, were not laws that regulate insurance.32 Consequently, in Taylor, as in Pilot Lifes the Court never addressed any conflict between implied preemption under §502 and ERISA’s express savings clause.

Other than the footnote reference in Ward,33 the only time the Supreme Court had discussed the potential conflict between savings clause and §502 before Moran was in Franchise Tax Board. There, the Court wrote that implied preemption under §502 was limited by ERISA’s express savings-clause exception to preemption for state laws that regulate insurance. The persuasiveness of the Moran dicta is seriously undermined by the Court’s inapposite citation to Pilot Life and by its failure to cite Franchise Tax Board.

The Moran Court improperly relied on Ingersoll-Rand. The Moran Court cites Ingersoll-Rand Co. v. McClendon34 for the premise that if a state law remedy conflicts with ERISA’s civil enforcement scheme, the remedy will be preempted. But Ingersoll-Rand did not involve an insurance law, so the conflict within ERISA between §502 and the savings clause was not present in that case. Also, the claim at issue there was a wrongful discharge claim under state law. ERISA provides a wrongful discharge remedy in its remedial provisions. Consequently, the state law duplicated an ERISA remedy.

ERISA preemption is a statutory construction exercise. Preemption involves a conflict between a state law and federal law. However, when a preemption question arises because one section of a federal statute conflicts with another section of the same statute, the problem should be remedied by applying standard statutory-construction rules.

Where a reasonable inference (of preemption, in this case) derived from the reading of one section of a federal statute (§502) conflicts with the express language of another section of the same statute (the savings clause in §514), statutory construction principles direct that the express language controls—that is, ERISA’s express exception to preemption for laws that regulate insurance (including remedies laws aimed solely at the insurance industry) trumps the implied preemption of state law remedies arising from §502.35

Pilot Life (hould not be extended beyond its scope. Moran’s dicta would extend Pilot Life beyond its holding, despite the Court’s clear trend toward limiting the extent of ERISA preemption in cases not involving pension plans. The Pilot Life Court relied on the U.S. solicitor general’s amicus curiae brief to advance the §502 implied preemption theory.

In Ward, the Court appeared to limit Pilot Life to its facts, citing the solicitor general’s brief filed in Ward, which explains and limits the solicitor general’s §502 implied-preemption position as presented in Pilot Life.36 Moran’s dicta would extend the scope of Pilot Life without even citing Ward.

Mississippi’s law was unique. Unlike the Mississippi bad-faith remedy at issue in Pilot Life, state bad-faith laws typically apply only to the insurance industry.37 Further, bad-faith and unfair-settlement-practices laws affect an integral part of the insurer/insured policy relationship. These laws clearly “regulate insurance.” They define the specific duties the insurer owes to the insured, and their provisions are incorporated into the policy language under state insurance law.

Courts should accept the plain meaning of the savings clause. The savings clause is not ambiguous. ERISA expressly states that it does not preempt state laws that regulate insurance: “[N]othing in this subchapter shall be construed to exempt or relieve any person from any law of any state which regulates insurance. . . .”38 Courts must assume that the ordinary meaning of the statutory language accurately expresses the legislative purpose.39

The states’ historic police powers were not meant to be superseded. ERISA preemption in non-pension-plan cases must be tempered by the presumption “that the historic police powers of the states were not to be superseded by the federal act unless that was the clear and manifest purpose of Congress.” 40 Insurance regulation and health care regulation have historically been dominated by the states.

Section 514 trumps §502. Section 514 says that “nothing in this subchapter shall be construed” to preempt state insurance regulation. Section 502 is part of the same subsection as §514. Consequently, the plain-meaning application of the savings clause compels the conclusion that §502 shall not be construed to preempt state insurance regulation. The solicitor general’s amicus curiae brief in Ward makes this point.

ERISA does not supersede the McCarran-Ferguson Act’s dictates. The McCarran-Ferguson Act states that “no act of Congress shall be construed to invalidate . . . any law enacted by any state for the purpose of regulating the business of insurance. . . .”41”Correspondingly, ERISA includes a second savings clause within §514 stating that the act shall not be construed to supersede any other federal law.42 Consequently, ERISA does not supersede the McCarran-Ferguson Act’s provision that regulating insurance is the business of the states. The McCarran-Ferguson Act directs that ERISA shall not be construed to invalidate state insurance regulations, including state laws that provide remedies for insurers’ bad-faith conduct.

Courts should adopt a broad reading of the savings clause to further consumer protection goals. In the early, seminal cases interpreting the McCarran-Ferguson Act, the Supreme Court interpreted the phrase “‘regulates’ the ‘business of insurance’” narrowly because insurers were trying to use the act to avoid complying with consumer-friendly antitrust statutes.43 The savings clause’s “regulates insurance” language must be construed broadly to further consumer protection goals.44 Additionally, ERISA’s savings-clause exception to preemption for laws that “regulate insurance” is broader than the McCarran-Ferguson Act’s definition of laws that “regulate the business of insurance.”45

Congress did not intend §502 to provide exclusive remedies. In Pilot Life, the Supreme Court suggested that §502 impliedly preempts state law remedies because Congress intended that section to provide the exclusive remedies available for claims connected with an ERISA employee benefit plan. To support this holding, the Court relied on a statement in ERISA’s legislative history that Congress expected the courts to develop a federal common law “to deal with issues involving rights and obligations under private welfare and pension plans.”46 If Congress intended courts to develop a federal common law to further define parties’ rights and responsibilities under ERISA, then obviously Congress did not intend §502 to provide exclusive remedies.

ERISA’s primary goal is to protect consumers. The Supreme Court often states that Congress enacted ERISA to provide a uniform system of limited remedies so that employers would be encouraged to continue to offer employee benefit plans.47 That view of ERISA’s goals puts the secondary purpose of trying to reduce the administrative burden for employers who voluntarily offer employee benefit plans ahead of its primary purpose, protecting plan participants.

If one did not read ERISA’s preamble or all of its legislative history, and read only the Court’s statements about uniform regulations, one would think that Congress enacted ERISA to make the world safe for large corporate employers and their insurers. On the contrary, ERISA’s preamble and legislative history confirm that Congress enacted ERISA to reform the private-pension industry in order to help ensure that plan participants received the pension benefits their employers and unions led them to believe they would receive on retirement.48

Allowing bad-faith claims won’t necessarily affect employers. Employers occasionally complain that if state law bad-faith remedies are characterized as insurance regulations—and so presumably saved from preemption—they would significantly expand employers’ potential liability. The argument goes like this: If courts recognized expanded consumer remedies, employers would just cease to offer such voluntary benefits, ultimately harming consumer interests.49 But allowing plan participants to pursue bad-faith remedies would not necessarily affect employers adversely. Bad-faith laws target plan insurers, not the plan itself or the plan sponsor.

Typically, under fully insured plans, the insurer retains the power to process claims submitted by plan participants. If extracontractual damages are recoverable only against plan insurers, the employers who sponsor plans—but do not control the claims-processing function—would not be exposed to liability.50 Further, if plan insurers that are subject to bad-faith liability attempt to pass along the costs of their extracontractual liability to employers in the form of increased premiums, employers can shop around to obtain insurance from other carriers that do not engage in bad-faith conduct.

The trial lawyer’s role The law is never finally decided until it is decided correctly. Most courts continue to deny plan participants their state law remedies against ERISA plan insurers, which the plan members would otherwise enjoy if ERISA had not been enacted.

However, the irony of insurers using ERISA—lauded as a great consumer-protection achievement—to avoid state consumer-protection laws frustrates principled judges. Trial lawyers must educate the courts about ERISA’s real and primary purpose: to secure fair treatment for employee benefit plan participants.

It appears that a meaningful Patient Bill of Rights will not emerge from Congress any time soon. While it remains an uphill battle to convince at least one federal circuit court, and then the Supreme Court, that Congress did not intend ERISA to limit plan participants’ rights, careful analysis of the statute and its legislative history confirms that Congress did not intend ERISA to supersede state bad-faith remedies that target the insurance industry.


1. Pub. L. No. 93-406, 88 Stat. 829 (1974) (codified as amended at 29 U.S.C. §§1001-1461 (1994 & Supp. IV 1998) and in scattered sections of the Internal Revenue Code).

2. Rush Prudential HMO, Inc. v. Moran, 122 S. Ct. 2151, 2158-59 (2002).

3. The distinctive feature of both the bad-faith and the unfair-settlement-practices claims that empower the consumer is the availability of extra-contractual damages when the consumer can prove aggravated misconduct by the insurer. See generally STEPHEN S. ASHLEY, BAD FAITH ACTIONS: LIABILITY AND DAMAGES (2d ed. 1997).

4. 29 U.S.C. §1144 (1994 & Supp. IV 1998).

5. 481 U.S. 41 (1987).

6. Id. at 50. Most states that recognize the tortious-breach-of-contract remedy limit its application to claims against the insurance industry. See ASHLEY, supra note 3, at 1-4.

7. 59 Stat. 33 (codified as amended at 15 U.S.C. §1011 (2001)).

8. See Pilot Life Ins. Co., 481 U.S. 41, 48-49. A commonsense analysis is used first to ascertain which state laws regulate insurance under the savings clause. The three-factor test is intended to supplement this. See UNUM Life Ins. Co. v. Ward, 526 U.S. 358, 367 (1999).

9. See Pilot Life Ins. Co., 481 U.S. 41, 50-51.

10. 29 U.S.C. §1132.

11. See Pilot Life Ins. Co., 481 U.S. 41, 51-54, 57.

12. See, e.g., Kanne v. Connecticut Gen. Life Ins. Co., 867 F.2d 489 (9th Cir. 1988), cert. denied, 492 U.S. 902 (1989); Anschultz v. Connecticut Gen. Life Ins. Co., 850 F.2d 1467 (11th Cir. 1988). But see Franklin H. Williams Ins. Trust v. Travelers Ins. Co., 50 F.3d 144 (2d Cir. 1995) (finding removal improper under §502 where state insurance law is saved from preemption under §514).

13. 514 U.S. 645 (1995).

14. See Donald T. Bogan, ERISA: The Savings Clause, §502 Implied Preemption, Complete Preemption, and State Law Remedies, 42 SANTA CLARA L. REV. 105 (2001).

15. See, e.g., Colligan v. UNUM Life Ins. Co., No. Civ. A. 00-K-2512, 2001 WL 533742 (D. Colo. Apr. 23, 2001), abrogated by Gilbert v. ALTA Health & Life Ins. Co., 276 F.3d 1292 (11th Cir. 2001); Lewis v. Aetna U.S. Healthcare, Inc., 78 F. Supp. 2d 1202 (N.D. Okla. 1999). But see Chamblin v. Reliance Standard Life Ins. Co., 168 F. Supp. 2d 1168 (N.D. Cal. 2001); Coffman v. Metropolitan Life Ins. Co, 138 F. Supp. 2d 764 (S.D. W.Va. 2001). The Eastern District of Pennsylvania is currently embroiled in a dispute among its judges as to whether recent Supreme Court decisions indicate that bad-faith claims should be protected from ERISA preemption. Compare Rosenbaum v. UNUM Life Ins. Co., No. Civ. A. 01-6758, 2002 WL 1769899 (E.D. Pa. July 29, 2002) with Zimnoch v. ITT Hartford, No. Civ. A. 99-6594, 2000 WL 283845 (E.D. Pa. March 14, 2000). Similarly, federal district judges in Alabama recently sparred over the issue until the Eleventh Circuit ruled that ERISA preempts state law bad-faith claims. See Gilbert, 276 F.3d 1292.

16. See, e.g., Colligan, No. Civ. A. 00-K-2512, 2001 WL 533742, at *2-3; Lewis, 78 F. Supp. 2d 1202.

17. See Selby v. Principal Mut. Life Ins. Co., No. 98 Civ. 5283 (RLC), 2000 WL 178191 (S.D.N.Y. Feb. 16, 2000).

18. 526 U.S. 358, 377 n.7 (1999).

19. 122 S. Ct. 2151.

20. Id. at 2164-65.

21. 463 U.S. 1 (1983).

22. See Donald T. Bogan, Protecting Patient Rights Despite ERISA: Will the Supreme Court Allow States to Regulate Managed Care? 74 TULANE L. REV. 951 (2000).

23. 463 U.S. 1.

24. 390 U.S. 557 (1968).

25. Franchise Tax Bd., 463 U.S. 1, 24-25. On remand, the California Court of Appeals held that ERISA did not expressly preempt the state law authorizing the tax board’s right to levy. 251 Cal. Rptr. 597 (Ct. App. 1988).

26. Only plan participants and beneficiaries, plan fiduciaries, or the U.S. Secretary of Labor may sue under §502.

27. See Franchise Tax Bd., 463 U.S. 1, 25.

28. See Silkwood v. Kerr-McGee Corp., 464 U.S. 238, 248 (1984); see also Bogan, supra note 22, at 960-63.

29. See Humana, Inc. v. Forsyth, 525 U.S. 299 (1999).

30. 307 F.3d 298, 308-11 (5th Cir. 2002).

31. Moran, 122 S. Ct. 2151, 2165.

32. 481 U.S. 58, 62 (1987).

33. See Ward, 526 U.S. 358, 376, n.7.

34. 498 U.S. 133 (1990).

35. See NORMAN J. SINGER, STATUTES AND STATUTORY CONSTRUCTION §46.05 (6th ed. 2000); Bogan, supra note 14, at 153-55.

36. Ward, 526 U.S. 358, 376, n.7; see Hill v. Blue Cross & Blue Shield, 117 F. Supp. 2d 1209, 1211-1212 (N.D. Ala. 2000), overruled by Gilbert, 276 F.3d 1292.

37. See ASHLEY, supra note 3, §1:02, at 1-4 (2d ed. 1997).

38 29 U.S.C §1144 (b)(2)(A).

39. See Shaw v. Delta Air Lines, Inc., 463 U.S. 85, 97 (1983).

40. See New York State Conf. of Blue Cross & Blue Shield Plans, 514 U.S. 645, 655 (citations omitted).

41. 15 U.S.C. §1012 (b).

42. 29 U.S.C. §1144 (d).

43. See Union Labor Life Ins. Co. v. Pireno, 458 U.S. 119, 126-28 (1982); Group Life & Health Ins. Co. v. Royal Drug Co., 440 U.S. 205, 231-33 (1979).

44. Royal Drug Co., 440 U.S. 205, 231-33: see also Metro. Life Ins. Co. v. Massachusetts, 471 U.S. 724 (1985).

45. See Moran, 122 S. Ct. 2151, 2159 & 2059 n.14, quoting Royal Drug Co., 440 U.S. 205, 211.

46. Pilot Life, 481 U.S. 41, 56.

47. See, e.g., Moran, 122 S. Ct. 2151, 2166; Ingersoll-Rand Co., 498 U.S. 133, 137, 142.

48. See Bogan, supra note 22, at 963-77.

49. See Moran, 122 S. Ct. 2151, 2166.

50. Presumably, ERISA’s deemer clause will continue to protect self-insured plans from extracontractual damages claims. See FMC Corp. v. Holliday, 498 U.S. 52 (1990).

Donald T. Bogan is a law professor and the director of clinical education at the University of Oklahoma, Norman