Erie Step Zero
Erie Step Zero
Alexander A. Reinert 0 1
0 Thi s Article is brought to you for free and open access by FLASH: The F ordham Law Archive of Scholarship and History. It has been accepted for inclusion in Fordham Law Review by an authorized editor of FLASH: The F ordham Law Archive of Scholarship and History. For more information , please contact
1 Benjamin N. Cardozo School of Law
Courts and commentators have assumed that the Erie doctrine, while
originating in diversity cases, applies in all cases whatever the basis for
federal jurisdiction. Thus, when a federal court asserts jurisdiction over
pendent state law claims through the exercise of supplemental jurisdiction in
a federal question case, courts regularly apply the Erie doctrine to resolve
conflict between federal and state law. This Article shows why this common
wisdom is wrong.
To understand why, it is necessary to return to Erie’s goals, elaborated
over time by the U.S. Supreme Court. Erie and its progeny are steeped in
diversity-driven policy concerns: concerns about unequal treatment based
on state citizenship and forum shopping figure most prominently. The first
concern, while salient in any diversity of citizenship case, simply has no
application in cases in which subject matter jurisdiction is founded upon
“arising under” jurisdiction. And the second Erie-derived principle,
avoiding forum shopping, while relevant to federal question cases, has a
different federalism timbre in diversity cases. In diversity cases, forum
shopping for certain substantive rules may deprive state courts of the
opportunity to adjudicate claims that involve state law through and through.
In jurisdiction founded on a federal question, by contrast, litigants are
encouraged to resort to the uniformity, experience, and solicitude of federal
courts; certain kinds of forum shopping are overtly welcomed, if not
encouraged. Thus, this Article shows that Erie applies, but differently, in
cases founded on federal question jurisdiction. In so doing, it provides a new
framework—what I call, borrowing from administrative law scholarship,
“Erie Step Zero”—for considering Erie questions in their proper
jurisdictional context, ensuring that federal law is not unnecessarily
displaced by a reflexive application of Erie in any case in which a state law
claim is presented.
* Professor of Law, Benjamin N. Cardozo School of Law. I am grateful to Richard
Bierschbach, Helen Hershkoff, Margaret Lemos, Max Minzner, and Kevin Stack for their
comments on this Article. In addition, I received very useful feedback from participants in
the First Annual Civil Procedure Workshop, in particular Stephen Burbank, Brooke Coleman,
Allan Erbsen, and Judith Resnik.
1. 304 U.S. 64 (1938).
2. Id. at 79–80.
3. One notable exception will be discussed below. See infra Part I.A. See generally Peter
Westen & Jeffrey S. Lehman, Is There Life for Erie After the Death of Diversity?, 78 MICH.
L. REV. 311 (1980) (arguing that the Erie doctrine should apply differently in federal question
and supplemental jurisdiction claims).
4. Erie, 304 U.S. at 78 (emphasis added).
5. See infra note 50 and accompanying text. To be clear, by “Erie case” I mean every
case in which the Court relied upon the Erie doctrine to resolve a choice-of-law question. In
some federal question cases, the Court has explicitly declined to apply the Erie analysis. See
infra notes 68–92 and accompanying text.
Court’s insight when reviewing the numerous cases in which Erie and RDA
questions arise in the federal question context. In the Court’s absence,
however, lower courts unfortunately have failed to fill the void with any
framework for deciding this question.
Despite the collective unwillingness to address this question in more detail,
it is an important one that can arise in numerous contexts outside of what we
might call a “true” Erie dilemma—where conflict between state and federal
law arises in cases in which the sole basis for jurisdiction is diversity of
citizenship. First, and certainly most common, are those cases in which state
law claims are raised because they share a “common nucleus of operative
fact”6 with a claim over which there is an independent basis for arising under
jurisdiction.7 For the most part, courts have assumed that Erie applies
without any modification to these pendent state law claims. The second
category of cases are undoubtedly much smaller and involve claims created
by state law but implicating an essential and substantial question of federal
law that is sufficient to create a basis for federal question jurisdiction.8
Courts have given almost no thought to how Erie applies in these cases.
In this Article, I provide a framework for approaching all of these difficult
Erie questions. The first step is recognizing that the questions we should ask
may depend on the jurisdictional context in which the case arises. For
instance, in diversity cases, we will usually at some point consider whether
the choice between state and federal law is outcome determinative in light of
“the twin aims of the Erie rule.”9 These twin aims—avoiding forum
shopping and inequitable administration of laws based on the citizenship of
the parties—have particular resonance in diversity of citizenship jurisdiction.
But they do not necessarily translate well to cases in which federal question
jurisdiction provides the jurisdictional hook. On the other hand, Hanna v.
Plumer’s10 insistence that we think of outcome determinativeness from an ex
ante rather than ex post perspective11 is sensible regardless of the
Taking these two insights together, therefore, my proposed framework
asks about forum shopping and inequity from an ex ante perspective but
recognizes that arising under jurisdiction serves different purposes than
diversity jurisdiction. Whereas diversity jurisdiction makes federal courts
available to resolve state law claims in a forum free from bias based on state
citizenship, arising under jurisdiction makes federal courts available to
resolve federal law issues because federal judges might be more experienced,
solicitous, and knowledgeable about federal law and because it serves the
federal system’s interest in the uniform application of federal law.12 When
state law questions arise in federal question cases, they are almost always
intimately connected to the federal issues: perhaps they arise through a state
law claim that is part of the federal question claim’s “common nucleus of
operative fact,”13 perhaps they arise through a state law claim that contains
an embedded and significant federal issue,14 or perhaps it is necessary to
resolve state law issues to resolve a federally created legal claim.15 In all of
these cases, concerns about forum shopping should be focused on those cases
in which the choice of a federal forum is motivated by something other than
the recognized and beneficial reasons for which we make federal courts
available to resolve federal law. Similarly, concerns about inequity should
not be focused on citizenship but on the differences in treatment between
state law claimants who do or do not have access to federal courts. Finally,
to the extent Hanna suggested that the balancing of interests introduced by
Byrd v. Blue Ridge Cooperative16 was unnecessary or unimportant,
examining Erie in the context of federal question jurisdiction should bring
that balancing back to the fore.
Of course, this issue will be most salient when the choice of law at stake
involves an issue that applies throughout the case, as opposed to an issue that
is claim specific, or when applying state law will effectively interfere with an
important federal interest. Often, where a rule of decision as to a legal issue
can be applied to one claim without interfering with the adjudication of
another claim, state law can be applied to state claims and federal law to
federal claims. But where a legal issue cuts across claims either directly or
indirectly, there will be more compelling reasons to apply federal rather than
state law. My proposed framework will be most useful in both framing and
resolving the conflicts present in each of these situations.
What I suggest here is a different take on a problem that Peter Westen and
Jeffrey Lehman addressed more than thirty years ago.17 Westen and Lehman,
12. See Grable, 545 U.S. at 312 (“The doctrine captures the commonsense notion that a
federal court ought to be able to hear claims recognized under state law that nonetheless turn
on substantial questions of federal law, and thus justify resort to the experience, solicitude,
and hope of uniformity that a federal forum offers on federal issues.”).
13. See, e.g., United Mine Workers v. Gibbs, 383 U.S. 715, 725 (1966).
14. See, e.g., Grable, 545 U.S. at 313.
15. See Indiana ex rel. Anderson v. Brand, 303 U.S. 95, 100 (1938) (making an
independent determination of state law in a case founded on the Contract Clause, and not
deferring wholesale to the decision of the state’s highest court, “in order that the constitutional
mandate may not become a dead letter”).
16. 356 U.S. 525 (1958).
17. See generally Westen & Lehman, supra note 3. Since Westen and Lehman’s article,
however, no one has attempted to systematically analyze the potentially different role that Erie
might play in federal question cases. And as I discuss in more detail below, Westen and
however, were concerned with assessing the validity of federal common law
rules that might govern federal question disputes, not with the choice between
federal and state law, because of their view that the Supremacy Clause
required application of any valid federal rule, whether sourced from
judgemade law or elsewhere.18 And when Westen and Lehman addressed the
question of supplemental jurisdiction, they only discussed ancillary claims
and focused on differing federal and state law rules that could be applied to
each claim in a case in isolation.19 As I discuss in detail below, the problem
is far more difficult when application of a state law rule cannot be confined
solely to the state law claim, either directly or indirectly.
In Part I of this Article, I lay out the treatment of this topic to date, from
both a scholarly and jurisprudential perspective. The orthodox view, rarely
challenged on its face, is that Erie applies in exactly the same way in federal
question cases as it does in diversity cases. As I show in Part II, the basis for
this position is questionable: scholars and courts cite to many of the same
sources, but a close consideration of them leads, at best, to ambiguity. In Part
III, I return to Erie and its progeny to show that many of the concerns driving
Erie were rooted in the specific context of diversity jurisdiction. Some of
these concerns have broad application in the federal question context, but the
specifics might look different. In Part IV, I propose a framework that
contextualizes the Erie analysis with respect to the different goals of diversity
and federal question jurisdiction. Critical to this framework is recognizing
that the first question in resolving a choice-of-law dispute between federal
and state law is what I call “Erie Step Zero”—one must situate the dispute in
its jurisdictional context before moving to the more familiar Erie choice of
law analysis. Just as in administrative law, where Chevron Step Zero
emerged to assist courts in determining whether the traditional Chevron
analysis applies at all to agency interpretations of governing statutes,20 Erie
Step Zero offers a methodology for courts to determine whether
straightforward Erie analysis should apply to particular choice-of-law
questions. After introducing the Step Zero concept, I then demonstrate some
areas of law in which this framework might prove particularly useful.
Lehman’s article does not sufficiently attend to the many complexities present in supplemental
jurisdiction cases and cases in which original jurisdiction is exercised through § 1331 over a
fundamentally state law claim. See infra notes 43–49 and accompanying text.
18. See Westen & Lehman, supra note 3, at 314.
19. See id. at 386 n.218.
20. The “Step Zero” moniker originated in a 2001 law review article. See Thomas W.
Merrill & Kristin E. Hickman, Chevron’s Domain, 89 GEO. L.J. 833, 836 (2001). It
subsequently was the subject of a close study and critique by Cass Sunstein. See Cass R.
Sunstein, Chevron Step Zero, 92 VA. L. REV. 187 (2006). It continues to occupy the time and
attention of administrative law scholars and judges, even though the Supreme Court has never
used the term. See Michael Herz, Chevron Is Dead; Long Live Chevron, 115 COLUM. L. REV.
1867, 1906 (2015).
In this part, I survey the scholarly and judicial approach to Erie’s
application outside of the diversity of citizenship context. Part I.A shows that
academics and commentators have almost uniformly stated that Erie doctrine
applies in the same way to all state law claims in federal court, whatever the
basis for federal court jurisdiction. Judicial decisions, reviewed in Part I.B,
are generally consistent with the academic consensus, at least on its face. In
application, however, I show that some courts have questioned the
orthodoxy, although none has provided a coherent framework for applying
Erie to nondiversity state law claims.
A. Survey of Scholarly Commentary
It is now almost universally assumed, within both the academy and the
judiciary, that the Erie doctrine applies to all state law claims, regardless of
whether the basis for jurisdiction in federal court is founded on diversity or
federal question.21 Judge Henry Friendly described the view that Erie
applied only to diversity cases as an “oft-encountered heresy,”22 although he
acknowledged that the Supreme Court’s jurisprudence, at least as of 1964,
had contributed to the blasphemy.23 The Wright and Miller treatise
succinctly states that the view that Erie applies only to diversity cases “is
simply wrong.”24 The logic behind this proposition has appeal. As Adam
Steinman has noted, to the extent that one believes Erie addressed a
constitutional problem—federal courts’ power to declare substantive law—
its logic is not limited to diversity cases, and neither the RDA nor the Rules
Enabling Act (REA) makes a distinction between jurisdictional bases.25
Thus, many commentators assert that Erie applies to all state law claims
presented to a federal court, whether through diversity jurisdiction or
21. See, e.g., Sergio J. Campos, Erie as a Choice of Enforcement Defaults, 64 FLA. L. REV.
1573, 1586 n.73 (2012); Abbe R. Gluck, Intersystemic Statutory Interpretation: Methodology
as “Law” and the Erie Doctrine, 120 YALE L.J. 1898, 1926 (2011) (“It is important to
remember that the Erie doctrine applies in federal-question and federal constitutional cases,
just as it does in diversity cases, provided that an analytically separate question of state law is
presented.”); Geri J. Yonover, Ascertaining State Law: The Continuing Erie Dilemma, 38
DEPAUL L. REV. 1, 22 (1988) (“It is also clear that Erie’s mandate impinges in areas of pendent
and federal question, as well as diversity, jurisdiction.”).
22. See Henry Friendly, In Praise of Erie—And of the New Federal Common Law, 39
N.Y.U. L. REV. 383, 408 n.122 (1964) (stating that the idea that “the Erie decision was limited
to diversity cases” is an “oft-encountered heresy” and that “[t]he correct view” is that “the Erie
doctrine applies, whatever the ground for federal jurisdiction, to any issue or claim which has
its source in state law”).
23. See id.
24. 19 CHARLES ALAN WRIGHT, ARTHUR R. MILLER & EDWARD H. COOPER, FEDERAL
PRACTICE AND PROCEDURE § 4520 (3d ed. 2016) (“It frequently is said that the doctrine of Erie
Railroad Company v. Tompkins applies only in diversity of citizenship cases; this statement
simply is wrong.” (footnote omitted)).
25. See Adam N. Steinman, What Is the Erie Doctrine?: (And What Does It Mean for the
Contemporary Politics of Judicial Federalism?), 84 NOTRE DAME L. REV. 245, 311–12
supplemental state law claims in federal question cases.26 Moreover, many
others assume that Erie also applies to all state law issues presented in cases
triggering federal question jurisdiction, whether those issues arise through
supplemental claims, state law claims implicating a substantial federal issue,
or federally created claims in which a state law issue arises on the
periphery.27 In the words of an oft-cited Second Circuit case, under Erie it
26. Joseph P. Bauer, Shedding Light on Shady Grove: Further Reflections on the Erie
Doctrine from a Conflicts Perspective, 86 NOTRE DAME L. REV. 939, 985 n.2 (2011) (noting
that Erie applied in diversity cases, supplemental jurisdiction cases, and when there are
statelaw-created counterclaims, cross-claims, or impleaders); Donald Earl Childress III, When Erie
Goes International, 105 NW. U. L. REV. 1531, 1545 n.96 (2011) (“For purposes of
completeness, I note that the Erie and Klaxon rules also apply to state-law claims heard in a
federal question case through the exercise of pendent or supplemental jurisdiction.”); Donald
L. Doernberg, The Unseen Track of Erie Railroad: Why History and Jurisprudence Suggest a
More Straightforward Form of Erie Analysis, 109 W. VA. L. REV. 611, 618 n.33 (2007) (“[I]t
is clear that today RDA is viewed as a choice-of-law statute for diversity cases and when state
law otherwise governs of its own force, as when supplemental jurisdiction brings state law
claims before a federal court in a federal-question case.” (citation omitted)); JoEllen Lind, The
End of Trial on Damages?: Intangible Losses and Comparability Review, 51 BUFF. L. REV.
251, 285–86 (2003) (stating that Erie applies in federal question cases if the source of law is
state law and application does not undermine federal policies).
27. Roy L. Brooks & Sharon A. Cheever, The Federal Loan Guarantee Program: A
Unified Approach, 10 J. CORP. L. 185, 221–22, 228 n.312 (1984) (noting that Erie applies to
nonfederal issues arising in federal question cases); see also 13D CHARLES ALAN WRIGHT ET
AL., FEDERAL PRACTICE AND PROCEDURE § 3567.1 (3d ed. Supp. 2016) (“Supplemental claims
are governed by state substantive law.”); Campos, supra note 21, at 1632 n.73 (stating that
Erie applies to state law questions in federal question cases); Daniel C.K. Chow, Limiting Erie
in a New Age of International Law: Toward a Federal Common Law of International Choice
of Law, 74 IOWA L. REV. 165, 167 n.4 (1988) (noting that Erie applies to any issue of state law
that has not been expressly or impliedly displaced by federal substantive law); Michael C.
Dorf, Facial Challenges to State and Federal Statutes, 46 STAN. L. REV. 235, 284 & nn.221–
22 (1994) (stating that federal courts follow state law in diversity cases, supplemental claims
in federal question cases, and cases such as due process cases in which the court must first
resolve an issue of state law); Martha A. Field, Sources of Law: The Scope of Federal
Common Law, 99 HARV. L. REV. 881, 912 n.141 (1986) (noting that Erie applies to diversity
cases and state law issues in federal question cases); Gluck, supra note 21, at 1926 (“In other
words, if one agrees that, in the straight diversity case, state methodology should be used to
interpret state statutes, then once we distill the fact that many nondiversity cases present
essentially identical questions of state statutory interpretation, Erie makes clear that the same
approach should apply.”); Abbe R. Gluck, The Federal Common Law of Statutory
Interpretation: Erie for the Age of Statutes, 54 WM. & MARY L. REV. 753, 785 n.122 (2013)
[hereinafter Gluck, The Federal Common Law of Statutory Interpretation] (“The Erie doctrine
applies in federal question and federal constitutional cases, just as it does in diversity cases,
provided that an analytically separate question of state law is presented.”); Alfred Hill, State
Procedural Law in Federal Nondiversity Litigation, 69 HARV. L. REV. 66, 70–71 (1955) (“This
aspect of Erie is clearly applicable in the federal courts today whether jurisdiction rests on
diversity of citizenship, on the presence of a federal question, or on some other nondiversity
ground.”); John Martinez, Taking Time Seriously: The Federal Constitutional Right to Be
Free from “Startling” State Court Overrulings, 11 HARV. J.L. & PUB. POL’Y 297, 319 n.95
(1988) (stating that Erie’s application depends on the source of the right sued upon, not upon
the basis for jurisdiction); Calvin R. Massey, State Sovereignty and the Tenth and Eleventh
Amendments, 56 U. CHI. L. REV. 61, 78 (1989) (“Even in federal question cases involving state
law issues, Erie commands that state decisional law be observed.”); Richard A. Matasar,
Rediscovering “One Constitutional Case”: Procedural Rules and the Rejection of the Gibbs
Test for Supplemental Jurisdiction, 71 CALIF. L. REV. 1399, 1489 n.401 (1983) (citing Gibbs
for the proposition that questions of state law are governed by Erie in federal question cases);
Adam N. Steinman, Kryptonite for CAFA?, 32 REV. LITIG. 649, 684 n.186 (2013) (“Erie and
is the “source of the right sued upon,” not the basis for jurisdiction, that
determines which law to apply.28
This consensus, however, neglects the critical question of how Erie applies
outside of diversity. To say that Erie applies in federal question cases does
not necessarily foreclose the possibility that Erie could apply differently in
different jurisdictional contexts. In the two decades after Erie was
announced, in fact, many commentators assumed that its role in federal
question cases was limited, although they did not explore the depths of its
role in supplemental jurisdiction claims. For instance, Sergei Zlinkoff,
writing four years after the decision, noted that the underlying policy of
Erie—avoiding discrimination on the basis of citizenship—did not apply in
federal question cases and carried this principle over to supplemental
jurisdiction because of the concern that applying multiple sources of law to a
single lawsuit would create needless complexity.29 Six years later, James A.
Gorrell and Ithamar D. Weed found it unsurprising that Erie had been mostly
ignored in federal question cases because they could not present problems of
a type similar to Black & White Taxicab & Transfer Co. v. Brown & Yellow
Taxicab & Transfer Co.30 and other cases that Erie targeted.31 Like Zlinkoff,
Gorrell and Weed recognized the problem of deciding what law to apply in
supplemental jurisdiction cases but assumed that state law should apply to be
consistent with Erie.32 Along the same lines, Henry Hart read the Erie
the REA are not categorically limited to diversity cases and can apply in federal question cases
as well.”); Steinman, supra note 25, at 311–12 (criticizing the “oft-stated assumption” that
Erie applies only in diversity cases); Arthur D. Wolf, Preliminary Injunction Standards in
Massachusetts State and Federal Courts, 35 W. NEW ENG. L. REV. 1, 41 (2013) (“State
substantive law claims, governed by the Erie doctrine, may arise in a federal court in at least
three types of cases or controversies: (1) where a state claim is asserted based on diversity
jurisdiction; (2) where a state claim is asserted based on supplemental jurisdiction; and (3)
where a state claim contains an essential federal element within general federal question
jurisdiction.”); Geri J. Yonover, A Kinder, Gentler Erie: Reining in the Use of Certification,
47 ARK. L. REV. 305, 310–14, 310 n.22 (1994) (stating that Erie applies to diversity cases and
state law claims brought pursuant to supplemental jurisdiction, in addition to state law
questions “incidental to resolution of the federal question presented”).
28. Maternally Yours v. Your Maternity Shop, 234 F.2d 538, 540 n.1 (2d Cir. 1956);
Harold L. Korn, Professor, N.Y.U. Sch. of Law, Remarks on the Continuing Effect of State
Rules of Evidence in the Federal Courts (June 21, 1969), in A Discussion of the Proposed
Federal Rules of Evidence Before the Annual Judicial Conference Second Judicial Circuit of
the United States, 48 F.R.D. 39, 69 (1969) (“[A]pplication of the Erie doctrine depends on the
source or nature of the right being asserted, rather than on the basis of the federal court’s
jurisdiction, so that although its usual application is in diversity cases, it is neither applicable
to all such cases nor limited to them.”).
29. See Sergei S. Zlinkoff, Erie v. Tompkins: In Relation to the Law of Trade-Marks and
Unfair Competition, 42 COLUM. L. REV. 955, 987–88 (1942) (“The application of different
sources of governing law to various issues in the same lawsuit will create endless
complications and result in fine-spun distinctions that will make those generated by Swift v.
Tyson look like child’s play.”); id. at 989 (arguing that “all issues over which federal
jurisdiction is sustainable, other than upon the basis of diversity, be governed by federal law,
regardless of whether or not diversity is also present in the particular case”).
30. 276 U.S. 518 (1928).
31. See James A. Gorrell & Ithamar D. Weed, Erie Railroad: Ten Years After, 9 OHIO ST.
L.J. 276, 300–01 (1948).
32. See id. at 302–04.
Court’s concern with forum shopping to be limited to diversity cases.33
Others also acknowledged, even if they did not agree, that Erie’s role outside
of diversity cases was at least up for grabs.34
More recent scholarship also considers the possibility that the Erie analysis
could be more complex in federal question cases. Allan Erbsen, for instance,
has suggested that the difference between interests in state law claims arising
in diversity and state law claims arriving through supplemental jurisdiction
might merit different interpretive methods: forum shopping may be less of a
concern in the latter instance, for example.35 And Erbsen and others have
recognized that federalism concerns may play out differently in the two
categories of cases.36 Some scholars have made a similar argument where
federal arising under jurisdiction is exclusively vested in federal courts.37
33. See Henry M. Hart, The Relations Between State and Federal Law, 54 COLUM. L. REV.
489, 513 (1954) (noting that forum shopping concerns would evaporate if in-state citizens
could remove based on diversity jurisdiction).
34. See Note, Federal Modification of State Law: Erie and the Bankruptcy Statute of
Limitations, 62 YALE L.J. 479, 479–80 (1953) [hereinafter Note, Federal Modification of State
Law] (observing that there is “no automatic guide” for whether federal courts must absolutely
adhere to state law in nondiversity cases, using cases under the Bankruptcy Act as an example);
Note, Rules of Decision in Nondiversity Suits, 69 YALE L.J. 1428, 1434 (1960) (observing that
in a series of post-Erie cases, the Court seemed to “permit a greater intrusion of federal
decisional rules in cases involving federal statutes or the United States as a litigant”); Note,
Toward a Workable Rule of Contribution in the Federal Courts, 65 COLUM. L. REV. 123, 128
(1965) (observing that on the day Erie was announced, the Court in Hinderlider v. La Plata
River & Cherry Creek Ditch Co., 304 U.S. 92 (1938), asserted power to make federal
substantive law in state law cases filed in state court).
35. See Allan Erbsen, Erie’s Four Functions: Reframing Choice of Law in Federal
Courts, 89 NOTRE DAME L. REV. 579, 658 n.286 (2013) (stating that federal courts are required
to apply Erie in federal question cases, but also recognizing that different interests in federal
question cases might call for a different theoretical approach, which the author describes as “a
fascinating question that merits further scholarship”); cf. William Baude, Beyond DOMA:
Choice of State Law in Federal Statutes, 64 STAN. L. REV. 1371, 1410–11 (2012) (arguing
against the extension of the Klaxon doctrine to federal question cases because of the differing
role of forum shopping concerns).
36. See Erbsen, supra note 35, at 658 n.286 (“Moreover, supplemental jurisdiction cases
might warrant special treatment because federal courts may approach state law differently
when it is ‘woven into’ a federal question case rather than when it is part of a ‘head of federal
jurisdiction which entails a responsibility to adjudicate the claim on the basis of state law,’
such as the Diversity Clause.” (quoting McNeese v. Bd. of Educ., 373 U.S. 668, 673 & n.5
(1963))); Gregory Gelfand & Howard B. Abrams, Putting Erie on the Right Track, 49 U. PITT.
L. REV. 937, 942 n.11 (1988) (distinguishing between “pure” Erie cases and quasi-Erie cases
in which state law issues are raised through supplemental jurisdiction or because federal law
incorporates state substantive law, and stating that federalism concerns may differ between the
two situations, resulting in a different application of the line between substance and
procedure); cf. Kurt M. Saunders, Plying the Erie Waters: Choice of Law in the Deterrence
of Frivolous Appeals, 21 GA. L. REV. 653, 663 n.33 (1987) (“There is sufficient authority to
support the proposition that the doctrine applies in all actions regardless of the basis of federal
jurisdiction unless a sufficient federal interest exists to mandate the application of federal
37. See, e.g., Gelfand & Abrams, supra note 36, at 978 n.129 (“In the context of diversity,
Erie’s forum shopping concern is a reaction to the comparison between a diversity case and
an identical case in which both litigants happen to be from the same state and a federal court
is unavailable. Since the United States is always a party to Federal Tort Claims Act cases,
there is no comparable state-court-only case by which to measure forum shopping induced
unfairness.”); Allan R. Stein, Erie and Court Access, 100 YALE L.J. 1935, 1967 n.161 (1991)
Other scholars have acknowledged the possibility that jurisdictional
context matters without taking a final position.38 Abbe Gluck, for example,
has criticized federal courts for “explicitly disregard[ing] state interpretive
principles in federal question cases,” even as she acknowledged the
potential—but minority—view that the Erie question might have different
application in federal question cases.39 And very few scholars have
embraced the radical proposition that Erie only applies in the diversity
context, without much, if any, analysis.40 John Hart Ely, for instance,
declared that the RDA (and by implication, Erie) “is not applicable in
(arguing that the Erie analysis has little weight in cases in which federal courts have exclusive
jurisdiction, thereby taking forum shopping and inequitable administration off the table).
There is, for example, some disagreement about Erie’s application in bankruptcy cases.
Compare Robert B. Chapman, Bankruptcy, 53 MERCER L. REV. 1199, 1307 n.811 (2002)
(collecting cases on division in courts between whether to apply federal or state choice-of-law
rules in bankruptcy cases), and Louis M. Guenin, Choice of Law in Usury, 109 BANKING L.J.
71, 81–82 (1992) (arguing that analysis is “more complicated” where usury arises in federal
question cases, particularly bankruptcy), with Alex Mills, Federalism in the European Union
and the United States: Subsidiarity, Private Law, and the Conflict of Laws, 32 U. PA. J. INT’L
L. 369, 418 (2010) (“The principle in Erie was quickly understood to apply not only to
diversity cases, but also to incidental issues of private law which arise when a federal court
has taken jurisdiction under other grounds—for example, mandating application of state
property law to determine proprietary issues which may arise when considering questions of
federal law, such as bankruptcy.”).
38. Martha Dragich, Back to the Drawing Board: Re-Examining Accepted Premises of
Regional Circuit Structure, 12 J. APP. PRAC. & PROCESS 201, 226–27 (2011) (“The
determination as to which cases call for the application of state law is also a matter of debate,
but may hinge on the particular grant of jurisdiction that allows the case to be heard in federal
court.”); Note, Federal Modification of State Law, supra note 34, at 484–85 (suggesting that
policy concerns driving Erie may not be applicable outside diversity context); Note, The Law
Applied in Diversity Cases: The Rules of Decision Act and the Erie Doctrine, 85 YALE L.J.
678, 680 n.6 (1976) (recognizing arguments about the power of federal courts to create
common law in nondiversity cases, but declining to address the issue in detail).
39. See Gluck, The Federal Common Law of Statutory Interpretation, supra note 27, at
785 n.122 (“Unlike in diversity cases, federal judges do not aim to ‘stand in the shoes’ of state
judges in federal question cases and that difference could potentially justify federal judges
approaching state statutory questions in federal question cases from a federal, rather than state,
interpretive perspective. But this is not the dominant doctrinal view with respect to other
questions of state law that arise in federal question cases and so would require an argument
for why statutory interpretation methodology should receive special treatment.”); see also
Gluck, supra note 21, at 1926.
40. See, e.g., Phillip I. Blumberg, The Transformation of Modern Corporation Law: The
Law of Corporate Groups, 37 CONN. L. REV. 605, 612 n.27 (2005) (stating that Erie requires
federal courts to apply state law doctrine on piercing in diversity cases and “single-factor”
doctrine in federal question cases); Steven S. Gensler, Bifurcation Unbound, 75 WASH. L. REV.
705, 740 (2000) (arguing that a state court presumption against bifurcation should play a minor
role in federal court in part because Erie does not apply to federal question cases); Tara L.
Haluch, Treatment of Americans with Disabilities Act Claims When the Plaintiff Is Deceased:
A Call for Uniformity, 48 EMORY L.J. 733, 760 n.187 (1999) (assuming that the
proceduralsubstantive distinction did not apply to Americans with Disabilities Act cases because they
are federal question cases); Korn, supra note 28, at 74–77 (arguing that state law privileges
should be recognized in diversity cases but not federal question cases); Peter L. Strauss, On
Resegregating the Worlds of Statute and Common Law, 1994 SUP. CT. REV. 429, 536 (arguing
that Erie’s prohibition of federal judge-made common law “is an artifact of diversity
jurisdiction,” which cannot be extended to federal courts sitting in federal question cases
without achieving “a stunning revision of constitutional understandings”).
nondiversity cases.”41 The case that Ely relied upon, however, involved a
federal question case in which the cause of action was created by federal
Despite the scattered recognition that Erie might apply differently in
federal question cases, however, Westen and Lehman, writing in 1980, are
the only scholars to date who have attempted to articulate how Erie might
function in federal question cases.43 The authors, predicting an end to
diversity jurisdiction based on then-pending legislation, wrote the article with
the hope of articulating a function for Erie in federal question and
supplemental jurisdiction claims, albeit one that was “significantly different”
from the doctrine’s function in diversity cases.44 The bulk of the article is
devoted to a reconceptualization of Erie in all types of cases and a general
defense of a relatively broad role for federal judge-made common law.45 As
to the specific issue covered here—the role that Erie plays in federal question
cases—the Westen and Lehman article does not offer much in the way of
specifics.46 They note, as I do here, that an Erie analysis focused on potential
discrimination based on citizenship does not resonate well in federal question
cases, and they suggest at least a general outline of an approach to the
problem.47 But when they focus on how Erie might work in practice, they
address only ancillary claims—not supplemental claims generally nor state
law claims that implicate federal interests that are sufficiently important so
as to justify the assertion of § 1331 jurisdiction—and they explicitly decline
to address cases in which one cannot neatly confine choice of law questions
to each claim in isolation.48 Thus, putting aside the merits of Westen and
Lehman’s broader claims about the meaning of Erie,49 their article leaves
much to be desired in terms of the specific question addressed here.
To sum up, with the exceptions noted above, the landscape of Erie
scholarship tends toward an overriding assumption that (1) Erie applies to all
state law issues (and some federal issues) arising in federal court regardless
of the jurisdictional context and (2) Erie looks the same in all contexts. And
most importantly, even scholars who question the orthodox view have
generally failed to address how courts should resolve choice-of-law questions
in the supplemental jurisdiction context or where state law claims are subject
to original federal question jurisdiction because they involve an embedded
substantial federal issue.
B. Judicial Application of Erie Outside Diversity Contexts
The orthodoxy that characterizes Erie scholarship is reflected in the case
law as well, at least on the surface. This is striking because Erie’s role in
federal question cases has only come up in lower federal courts, albeit with
some frequency. By contrast, every Supreme Court case to actually apply an
Erie analysis to a choice-of-law question has been founded on diversity
jurisdiction.50 Nor has the Court ever directly applied the Erie analysis to
claims based on supplemental jurisdiction.51 This has not deterred courts
Lehman, supra note 3, at 384 (arguing that “fairness value in diversity is abridged if even a
single lawsuit comes out differently because of judge-made rules of procedure”).
50. Without citing to every case in which the Supreme Court has applied Erie in a diversity
context, in the 205 Supreme Court cases that have cited to Erie, only one decision appears to
apply Erie analysis to a state law question arising in a federal question context. See Wichita
Royalty Co. v. City Nat’l Bank of Wichita Falls, 306 U.S. 103, 107 (1939). In Wichita Royalty,
a state law counterclaim was raised in response to a federal claim, and the Court stated, without
elaboration, that Erie required application of Texas law to the counterclaim. See id. Since that
decision, while the Court has discussed Erie in the federal question context, it has yet to apply
Erie to the choice-of-law analysis in those cases. See, e.g., DelCostello v. Int’l Bhd. of
Teamsters, 462 U.S. 151, 159 n.13 (1983) (stating that Erie does not compel federal courts to
borrow state law to fill in interstitial gaps in federal statutes, but acknowledging that when
Congress directs courts to borrow from state law, “Erie will ordinarily provide the framework
for doing so”); Burks v. Lasker, 441 U.S. 471, 476 (1979) (noting that Erie does not control
in cases premised on a federal cause of action); Blonder-Tongue Labs., Inc. v. Univ. of Ill.
Found., 402 U.S. 313, 324 n.12 (1971) (noting that whether nonmutuality doctrine applied in
federal question cases was a question of federal law and Erie did not apply); Urie v. Thompson,
337 U.S. 163, 174 (1949) (finding that the meaning of negligence in an FELA case is a
question of federal law and not governed by Erie); Vanston Bondholders Protective Comm.
v. Green, 329 U.S. 156, 162 (1946) (stating that Erie does not apply in bankruptcy cases);
Clearfield Tr. Co. v. United States, 318 U.S. 363, 366 (1943) (holding that Erie does not apply
in state law claims in which the United States is a party); Sola Elec. Co. v. Jefferson Elec. Co.,
317 U.S. 173, 176 (1942) (stating that Erie does not apply in a case controlled by federal
statute); D’Oench, Duhme & Co. v. FDIC, 315 U.S. 447, 455 (1942) (acknowledging but not
resolving the possibility that the Klaxon-Erie choice-of-law doctrine might not apply in federal
question cases). But see Agency Holding Corp. v. Malley-Duff & Assocs., Inc., 483 U.S. 143,
164 n.2 (1987) (Scalia, J., concurring) (“Contrary to the DelCostello Court’s claim, however,
neither our decision in [Erie] nor the Rules of Decision Act scholarship underlying
it . . . remotely established that that statute applies only in diversity cases.”); Anderson v.
Abbott, 321 U.S. 349, 365 (1944) (implying that state law that related to a federal question
suit would be enforceable pursuant to Erie in the absence of any conflict with federal policy
announced by Congress).
51. The Court has made offhand references to the role of Erie and state law in
supplemental claims without actually conducting any Erie analysis. See, e.g., Felder v. Casey,
487 U.S. 131, 151 (1988); United Mine Workers v. Gibbs, 383 U.S. 715, 72
). I address
from stating with near certainty that “[t]he principles set forth in Erie and its
progeny apply equally to diversity and federal question cases.”52 Or, put
differently, “it is not the basis of jurisdiction that requires application of the
forum state’s choice of law rules, but the source of the claim in question.”53
On its face, this looks simple. Many courts assert that state substantive
law applies in diversity cases and to supplemental state law claims in federal
question cases54 but that federal law governs adjudication of claims created
in detail below the widespread misinterpretation of the language in Gibbs. See infra notes 138–
41 and accompanying text.
52. See, e.g., Park v. City of San Antonio, No. SA-03-CA-0491-RF, 2006 WL 2519503,
at *14 n.67 (W.D. Tex. Aug. 16, 2006); see also Vess v. Ciba-Geigy Corp. USA, 317 F.3d
1097, 1102 (9th Cir. 2003) (“Erie applies irrespective of whether the source of subject matter
jurisdiction is diversity or federal question.”).
53. In re Rezulin Prods. Liab. Litig., 210 F.R.D. 61, 69 n.53 (S.D.N.Y. 2002); see, e.g.,
Maternally Yours, Inc. v. Your Maternity Shop, Inc., 234 F.2d 538, 540 n.1 (2d Cir. 1956)
(“[I]t is the source of the right sued upon, and not the ground on which federal jurisdiction is
founded, which determines the governing law.”); Monmouth Cty. Corr. Inst. Inmates v.
Lanzaro, 643 F. Supp. 1217, 1221 (D.N.J. 1986), aff’d in part, modified in part, 834 F.2d 326
(3d Cir. 1987); Cinnamon v. Abner A. Wolf, Inc., 215 F. Supp. 833, 835–36 (E.D. Mich.
1963); In re Hilt, 175 B.R. 747, 753 (Bankr. D. Kan. 1994) (relying on the Erie doctrine and
citing the Wright and Miller treatise).
54. Mason & Dixon Intermodal, Inc. v. Lapmaster Int’l LLC, 632 F.3d 1056, 1060 (9th
Cir. 2011) (“When a district court sits in diversity, or hears state law claims based on
supplemental jurisdiction, the court applies state substantive law to the state law claims.”);
Barton v. Clancy, 632 F.3d 9, 17 (1st Cir. 2011) (“A federal court . . . exercising supplemental
jurisdiction over a state law claim must apply state substantive law.”); Equals Int’l, Ltd. v.
Scenic Airlines, 35 F. App’x 532, 534 (9th Cir. 2002) (finding that a state law regarding
attorney’s fees applied to a state law claim supplemental to a federal copyright claim);
Penobscot Indian Nation v. Key Bank of Me., 112 F.3d 538, 557 n.27 (1st Cir. 1997) (relying
on Erie to apply state substantive law to pendent state law contract claims in federal question
cases); Zerman v. Ball, 735 F.2d 15, 19–20 (2d Cir. 1984) (applying, in a securities fraud case,
state choice-of-law rules to determine the law governing the effect of a choice-of-law clause
in a securities contract); Rohm & Haas Co. v. Adco Chem. Co., 689 F.2d 424, 429 (3d Cir.
1982) (requiring a court to look to applicable state choice-of-law principles in a state trade
secret case brought pendent to a federal patent action); Sys. Operations, Inc. v. Sci. Games
Dev. Corp., 555 F.2d 1131, 1136 (3d Cir. 1977) (applying Klaxon to pendent jurisdiction
claims); Spottedcorn v. Advanced Corr. Healthcare, Inc., No. CIV-11-1096-C, 2011 WL
6100653, at *1 n.2 (W.D. Okla. Dec. 7, 2011) (discussing choice of law issue in terms of a
requirement of attaching affidavits to complaints); Shannon v. Koehler, No. C 08-4059-MWB,
2011 WL 10483363 (N.D. Iowa Sept. 16, 2011) (applying federal evidence rules barring the
admissibility of a settlement offer rather than the state law that would have permitted
admission in a state law claim supplemental to a federal § 1983 claim); Arvie v. Dodeka, LLC,
No. H-09-1076, 2011 WL 1750242 (S.D. Tex. May 4, 2011) (involving a potential conflict
between Federal Rule of Civil Procedure 14 and state law regarding the designation of
responsible parties in a state law claim supplemental to a federal Fair Debt Collections
Practices Act claim); Sheasly v. Orr Felt Co., No. CV 10-956-PK, 2010 WL 4273230, at *2
(D. Or. Oct. 25, 2010) (“Whether Sheasley’s lawsuit is construed as a federal question action
involving supplemental jurisdiction over a state-law claim or as a diversity action, this court
will apply federal procedural law at all times, federal substantive law in analyzing the
[Employee Retirement Income Security Act] claim, and Oregon substantive law in analyzing
the perceived-disability claim.”), aff’d, 10-CV-956-PK, 2010 WL 4961807 (D. Or. Dec. 1,
2010); id. at *3 (noting that state law governed the application of a forum selection clause as
it related to the state law claims, whether the basis for jurisdiction was diversity or
supplemental jurisdiction to the federal claim); In re Lease Oil Antitrust Litig. (No. II), No.
MDL 1206, 2009 WL 5195977, at *4 (S.D. Tex. Dec. 22, 2009)
(noting that it was commonly,
but erroneously, assumed that Erie only applied in diversity actions when in fact it applied
by the federal sovereign.55 In perhaps the most difficult area, courts have
rarely addressed the choice-of-law analysis where § 1331 jurisdiction arises
because a federal question is embedded in a state law claim.56 Other courts
regardless of the basis for subject matter jurisdiction to all issues governed by state law)
and vacated sub nom. All Plaintiffs v. All Defendants, 645 F.3d 329 (5th Cir. 2011); Tillman
v. Georgia, 466 F. Supp. 2d 1311, 1316–17
(S.D. Ga. 2006)
(applying state law to determine
when an action is commenced in a state law claim that was pendent to a federal question
claim); Fondrk v. Westmoreland County, No. Civ.A 04-0900, 2006 WL 1699471, at *2–3
(W.D. Pa. June 20, 2006) (applying third-party beneficiary law to a state law claim
supplemental to a § 1983 claim); Crawford v. Potter, No. Civ. 1:04CV00303, 2005 WL
2452092, at *3 n.8
(M.D.N.C. Oct. 4, 2005)
(applying state law rules of contract interpretation
to a supplemental claim to Title VII); SEC v. Elmas Trading Corp., 683 F. Supp. 743, 748–49
(D. Nev. 1987) (relying on Erie to apply state contract law in federal question claim), aff’d,
865 F.2d 265 (9th Cir. 1988).
55. See Corbello v. DeVito, 844 F. Supp. 2d 1136, 1
155 (D. Nev. 2012
) (making this
observation in a copyright action), rev’d and remanded, 777 F.3d 1058 (9th Cir. 2015); West
v. N.M. Taxation & Revenue Dep’t, No. CIV 09-0631 JB/CG, 2011 WL 5223010 (D.N.M.
Sept. 30, 2011) (“When addressing federal questions, the Court does not have to conduct the
same analysis that [Erie] requires for diversity cases, to determine whether federal law and
procedures applies.”); McFadden v. Deutsche Bank Nat’l Tr. Co., No. 2:10-cv-03004 JAM
KJN PS, 2011 WL 3606797, at *6
(E.D. Cal. Aug. 16, 2011)
(holding that an anti-SLAPP
motion may be brought to strike state law claims pending in federal court, whether via diversity
jurisdiction or as claims supplemental to federal question claims but not claims “based on
federal law”), report and recommendation adopted, No. CIV-S-10-cv-3004-JAM-KJN-PS,
2011 WL 4765845
(E.D. Cal. Oct. 7, 2011)
; D.V. v. Westmoreland Cty. Children’s Bureau,
No. 07-829, 2008 WL 612744, at *2 (W.D. Pa. Mar. 3, 2008)
(“[H]ere the jurisdiction of this
Court arises under 28 U.S.C.A. § 1331, the general federal question statute, obviating the need
to undertake the Erie choice of law analysis and requiring the application of federal procedural
and substantive law.”)
; Brown v. Plainfield Cmty. Consol. Dist. 202, 522 F. Supp. 2d 1068,
1077 n.4 (N.D. Ill. 2007) (“It hardly bears repeating that the Erie doctrine applies to diversity
cases, while this case is premised on federal question jurisdiction.”); Atl. Mut. Ins. v. Napa
Transp., Inc., 399 F. Supp. 2d 523, 525
(regarding prejudgment interest in
Carmack Amendment claim), aff’d, 201 F. App’x 19 (2d Cir. 2006); Mobile Gas Serv. Corp.
v. Utils. Bd. of Daphne, No. Civ.A. 04-039
WL 2428329, at *2
(S.D. Ala. Sept.
(finding that attorney’s fee claims are governed by federal law in antitrust actions);
Boonma v. Bredimus, No. Civ.A.3:05-CV-0684-D, 2005 WL 1831967, at *6–8 (N.D. Tex.
July 29, 2005) (stating that state choice of law rules would apply to state law claims, whether
brought under diversity or supplemental to federal question cases and that federal choice of
law rules would apply to federal question claims); cf. Berman v. McManus, No. 2:11-cv-00635
MCE KJN PS, 2012 WL 1455088, at *2 & n.5 (E.D. Cal. Apr. 26, 2012) (noting that a state
anti-SLAPP law applies to state law claims brought pursuant to diversity or supplemental to
federal question claims but not claims based on federal law); Bulletin Displays, LLC v.
Regency Outdoor Advert., Inc., 448 F. Supp. 2d 1172, 1181–82 (C.D. Cal. 2006) (holding
that, even though California’s anti-SLAPP act was substantive, it did not apply to federal
question cases because Erie has no application to federal question claims, “only to state law
claims in diversity actions and pendent state law claims in federal question cases”); Martinez
v. Humana, Inc., No. Civ.A. C-04-95, 2005 WL 1923532, at *7–8
(S.D. Tex. Aug. 10, 2005)
(rejecting the “make whole” doctrine for subrogation in federal question ERISA cases, but
citing the Fifth Circuit case, Sunbeam-Oster Co. Grp. Benefits Plan v. Whitehurst, 102 F.3d
1368, 1374 (5th Cir. 1996), accepting the possibility that the issue might be different in a
diversity case); Reed v. Norfolk & W. Ry. Co., 635 F. Supp. 1166, 1167 (N.D. Ill. 1986)
(applying the federal statute of limitations in a Federal Employers’ Liability Act case and
describing the contention that Erie applies as borderline frivolous); In re Centerstone
Diamonds, Inc., No. 2:09-bk-23945-PC, 2014 WL 1330186, at *4–5 (Bankr. C.D. Cal. Apr.
2, 2014) (holding that a state anti-SLAPP law does not apply to federal question cases).
56. Amoco Chem. Co. v. Tex Tin Corp., 925 F. Supp. 1192, 1201 n.9 (S.D. Tex. 1996)
(applying state choice-of-law rules to a state law claim with an embedded federal issue—
assuming that Erie and Klaxon apply); Grynberg Prod. Corp. v. British Gas, PLC, 817 F. Supp.
might simply state without elaboration that Erie applies in federal question
cases without explaining how.57
But this explanation of the role of Erie in nondiversity cases has actually
almost never been so simple in application. First, the Supreme Court itself
has displaced state law with federal common law even when state sovereignty
would appear to be at a high point in diversity cases. In Banco Nacional de
Cuba v. Sabbatino,58 for example, the Court applied a federal common law
doctrine to a diversity case, even though it acknowledged that state law likely
would have resulted in the same outcome.59 The Court felt “constrained to
make it clear” that, notwithstanding the Erie doctrine, “an issue concerned
with a basic choice regarding the competence and function of the Judiciary
and the National Executive in ordering our relationships with other members
of the international community must be treated exclusively as an aspect of
federal law.”60 Indeed, the Court suggested that Erie might generally have
limited application when issues affecting international relations arise in
Similarly, in Howard v. Lyons,62 a diversity action for defamation, the
Court applied a federal common law defense because the statements were
made by an officer of the federal government acting in the course of his
duties.63 The question implicated the “effective functioning” of the
government, making it a “peculiarly federal concern” that should not be
subject to the “vagaries of the laws of the several States.”64 A more recent
example comes from Boyle v. United Technologies,65 a diversity case in
which the Court recognized a military contractor defense under federal law
that would not have been available under the state law that, had Erie applied,
should have controlled in the case.66 The Court in Boyle found a “‘significant
conflict’ . . . between an identifiable ‘federal policy or interest and . . . state
Outside of the diversity context, in federal question cases, the Supreme
Court has also found Erie inapplicable and federal common law controlling.
The same day that Erie was decided, for example, the Supreme Court
recognized its power to make federal common law in a state law dispute
between two Colorado private parties because the case concerned
interpretation of an interstate water compact.68 Similarly, in Clearfield Trust
Co. v. United States,69 the Court had to determine whether state law should
govern commercial paper obligations for a check drawn on the Treasurer of
the United States through the Federal Reserve Bank of Philadelphia. Because
of the concern for uniformity in defining the rights and obligations flowing
from a check backed by the federal fisc, the Court held that Erie did not apply
and that it was free to apply federal common law to the dispute.70
Cases like Clearfield Trust can be understood as providing an exception to
the general Erie rule when a “distinct federal interest” is at stake.71 And in
the decade or so after Erie, these cases were relatively common.72 In
D’Oench, Duhme & Co. v. FDIC,73 the Court declined to decide whether the
rule of Klaxon Co. v. Stentor Electric Manufacturing Co.74—an extension of
the Erie doctrine—applied where jurisdiction is not based on diversity of
citizenship, instead holding that federal law controlled the question of
liability because of the federal policy to protect a federal corporation from
certain misrepresentations.75 In his concurrence, Justice Robert Jackson
strived to provide a more detailed explanation of the Court’s reasoning.76 He
took the position that Klaxon did not control, because (1) the case was not
founded on diversity jurisdiction, (2) Congress did not direct the Court to
look to state law to answer this question, and (3) application of federal
common law is always necessary to “implement the federal Constitution
and statutes.”77 Neither the majority opinion nor any of the concurrences
argued that federal statutes explicitly required application of the rule adopted
by the Court—Justice Jackson rested his views on the “policy” behind the act
creating the FDIC.78
Following the lead of Clearfield Trust, in case after case, the Court has
declined to rely upon Erie in federal question cases that involve state law
issues, whether in tort cases,79 corporate law,80 calculation of damages,81 or
for statute of limitations questions,82 to give just a few examples. The
explanations that the Court has relied upon for applying federal common law
in all of these cases, regardless of their jurisdictional basis, revolve around a
conclusion that the issues involved are “matters essentially of federal
character” rather than the issues of local interest protected by Erie83—that
application of state law will undermine interests in uniformity84 or the federal
fisc.85 At times, the Court explicitly has stated that Erie does not apply in
federal question cases, at least where there are no state law claims asserted
through supplemental jurisdiction.86 The logic of Clearfield Trust does not
apply in every case that implicates federal interests, but these cases help
frame the relevant considerations. In Miree v. DeKalb County,87 for
example, the plaintiff brought a diversity case regarding an aircraft crash.
Although the Court recognized a general federal interest in the regulation of
air travel, and although the United States was a party to the contract at issue
in the case (but not a party to the case), the Court found that state law
governed the question of whether the defendants were liable on a third-party
beneficiary theory under the contract.88 Key to the resolution of Miree was
the Court’s conclusion that it would have no “direct” impact on the United
States or the federal fisc.89 United States v. Kimbell Foods, Inc.90 is another
more recent example of a case in which federal law was found to govern the
creation of substantive law but in which the Court found that absent
congressional lawmaking, the best course was to adopt state law as a default
rule.91 The Court arrived at its decision after determining that there was no
demonstrated need for national uniformity, that application of state law
would not “frustrate . . . federal interests,” and that disregarding state law
would unseat settled expectations.92
One could read all of these cases as interpretations of the RDA’s language
directing the application of state law rules of decision “in cases where they
apply” but not “where the Constitution or treaties of the United States or Acts
of Congress otherwise require or provide.”93 And scholars have recognized
that in some cases, the Supreme Court has implied that the RDA has no
relevance to federal question cases.94 But when doing so, the Court has not
explicitly addressed the question of whether it is applying the RDA but
finding that federal statute controls or whether it is just ignoring the RDA
because application of state law would undermine enforcement of a federal
right.95 As Henry M. Hart once observed, distinguishing among these
options is difficult because the RDA “seems almost perversely
uninformative.”96 In any event, however one views the role of the RDA (and
I will say more about this later), the Supreme Court has regularly found that
Erie’s principles either do not apply or are limited in federal question cases
and other cases in which federal interests are at stake. But the Court has
never actually conducted an Erie analysis for state law claims that themselves
are the basis of § 1331 jurisdiction or that are brought as supplemental claims
to an original federal question claim.
Lower courts have demonstrated a similar tendency to resolve choice of
law questions in tension with the orthodox statement that Erie applies equally
in federal question cases. For instance, the Foreign Sovereign Immunities
Act provides federal jurisdiction over state law claims when those claims are
asserted against foreign powers. On the simple rule enunciated above, Erie
should call for the application of state substantive law to resolve those claims,
but courts considering the question often turn first to federal common law to
application of federal common law to resolve the issue presented here would promote no
federal interests even approaching the magnitude of those found in Clearfield Trust.”).
89. See id.
90. 440 U.S. 715 (1979).
91. See id. at 740.
92. See id. at 728–29, 734, 739–40; De Sylva v. Ballentine, 351 U.S. 570, 580–81 (1956)
(looking to state law for the definition of the word “children” as used in Copyright Act because
the law of domestic relations is an area of traditional state regulation); Joshua M. Koppel,
Federal Common Law and the Courts’ Regulation of Pre-Litigation Preservation, 1 STAN. J.
COMPLEX LITIG. 171, 189–90 (2012).
93. 28 U.S.C. § 1652 (2012).
94. John T. Cross, State Choice of Law Rules in Bankruptcy, 42 OKLA. L. REV. 531, 553–
95. See id. at 553–54.
96. See Hart, supra note 33, at 504.
determine which substantive law governs.97 And, just as in Erie scholarship,
some courts have carved out certain areas—for example, bankruptcy and
choice of law in transfers—in which the simple rule applying Erie to all
conflicts is contested. In bankruptcy cases, some courts have found that Erie
principles do not apply as a formal matter, even as the courts decided cases
using Erie’s guideposts.98 Others have found that Erie principles apply
directly.99 And others have stated that federal law applies without
Transfers also bring up difficult issues. In Van Dusen v. Barrack,101 the
Supreme Court held that when a diversity case is transferred from one district
to another, choice of law should be determined by reference to the state in
which the transferor court sits.102 The justification for this rule is founded in
the logic of Erie—as the Court stated, “[W]e should ensure that the ‘accident’
of federal diversity jurisdiction does not enable a party to utilize a transfer to
achieve a result in federal court which could not have been achieved in the
courts of the State where the action was filed.”103 Less well examined is
what should happen when a case is transferred from one federal court to
another but the case is based on federal question jurisdiction. At least two
lower courts have found that in such a case, there is no reason to apply Erie
principles—instead, because jurisdiction is founded on a federal question, the
law of the transferee court may be applied to the dispute.104 However, several
97. See, e.g., Schoenberg v. Exportadora de Sal, S.A. de C.V., 930 F.2d 777, 782 (9th Cir.
1991); Barkanic v. Gen. Admin. of Civil Aviation of the People’s Republic of China, 923 F.2d
957, 959–61 (2d Cir. 1991) (choosing to apply state choice of law principles but only after
applying federal common law and determining that using state law would best accord with
congressional intent in enacting the Foreign Sovereign Immunities Act).
98. See, e.g., In re Gillissie, 215 B.R. 370, 377–78 (Bankr. N.D. Ill. 1997) (“The Court is
not exercising any jurisdiction based on the diversity of the parties under 28 U.S.C. § 1332,
(which diversity of citizenship does not exist in any event as all parties to these disputes are
Illinois residents), but rather is exercising its bankruptcy jurisdiction under 28 U.S.C. § 1334
and the automatic reference rule of the District Court, Local General Rule 2.33(A), by which
all bankruptcy cases and related adversary proceedings are automatically referred to the
bankruptcy judges comprising the adjunct unit of the District Court known as the Bankruptcy
Court via 28 U.S.C. § 151. Thus, the Court is not obliged to decide the matters at bar as would
a federal district trial court using state substantive law in a diversity case, as Erie requires,
because the federal questions under §§ 363(h), 522 and 544(b) must be decided utilizing
federal bankruptcy jurisdiction along with the Illinois state exemption and fraudulent
conveyance avoidance issues.”); see also In re Crist, 632 F.2d 1226, 1229 (5th Cir. 1980)
(holding, in a federal bankruptcy proceeding where disposition of a federal question requires
reference to state law, that federal courts are not required to apply state choice of law rules but
are instead free to apply the law considered relevant to the pending controversy).
99. See, e.g., In re Hilt, 175 B.R. 747, 753 (Bankr. D. Kan. 1994) (relying on the Erie
doctrine and the Wright and Miller treatise); In re Conner Corp., 127 B.R. 775, 777–78
(finding that Erie generally applies to issues of state law in bankruptcy but
that Clearfield Trust calls for the application of federal law to commercial paper issues that
govern the rights and duties of the United States).
100. See, e.g., In re Powerburst Corp., 154 B.R. 307, 310 (Bankr. E.D. Cal. 1993) (applying
federal choice of law rules in resolving a contract law issue).
101. 376 U.S. 612 (1964).
102. See id. at 636–37.
103. See id. at 638.
104. See In re Korean Air Lines Disaster of Sept. 1, 1983, 829 F.2d 1171, 1175 (D.C. Cir.
1987) (“There is no room in the federal system of review for rote acceptance of the decision
that questions about the scope of the Federal Rules of Civil Procedure arise
in diversity cases, but not federal question cases—the distinction was
unnecessary to resolution of the case, however.116 A court in the District of
Connecticut held that federal choice-of-law rules should apply where
jurisdiction was founded on both diversity and the presence of a federal
question,117 in part because the case involved application of a law that was
enacted prior to the Erie decision.118 The court accepted that this could mean
that results would differ if the case were sitting solely in diversity but argued
that the Supremacy Clause would not tolerate applying state law.119
The assumption that Erie applies regardless of the grounds for jurisdiction
was not always seen as obvious.120 Part of the confusion rested on the
question of whether Erie was a decision founded on constitutional law or
simply an interpretation of the RDA.121 The Supreme Court itself in 1947
stated, in the estoppel context, that federal courts sitting in diversity must
follow state law and state policy but that this is not necessarily the case for
federal courts sitting in federal question jurisdiction.122
II. EXAMINING THE BASES FOR THE ORTHODOX VIEW
As established in Part I, the scholarly and judicial community is almost
completely united in adherence to the position that Erie doctrine applies in
the same way to state law claims, whatever the basis for federal court
jurisdiction. And when one examines the relevant scholarship and
jurisprudence in detail, the sources for this common and accepted wisdom
are nearly always the same. To support the claim that pendent state law
claims in federal question cases are governed by Erie, most scholars and
courts look to United Mine Workers v. Gibbs,123 the Supreme Court’s seminal
founded on a federal question through the Edge Act, 12 U.S.C. § 632 (2012), but in which the
claim was created by state law; jurisdiction had also been alleged under the alternative ground
116. See Fitzgerald v. Gann Law Books, Inc., 956 F. Supp. 2d 581, 586 (D.N.J. 2013)
(suggesting that a question about the scope of the Federal Rules of Civil Procedure arises in
diversity cases, not federal question cases).
117. See Quadrini v. Sikorsky Aircraft Div., United Aircraft Corp., 425 F. Supp. 81, 88
(D. Conn. 1977) (holding that federal choice of law rules should apply in a case in which the
court exercised federal question and diversity jurisdiction).
118. See id. at 88 (noting that when the relevant federal law was passed, Swift was still good
119. See id. at 85 (“I assume that whatever substantive law is found to be applicable in the
exercise of this Court’s federal question jurisdiction should govern the lawsuit, even though a
contrary result might be reached if only diversity jurisdiction existed. It seems to be a fair
inference from the Supremacy Clause that whatever decision a federal court is impelled to
reach in the exercise of its federal question jurisdiction must prevail over contrary results that
would be reached if the Court were functioning essentially as a state court in the exercise of
120. In re Longhorn Sec. Litig., 573 F. Supp. 278, 280 & n.1 (W.D. Okla. 1983) (describing
“jurisprudential uncertainty in the post [Erie] universe”).
121. See Seaboard Fin. Co. v. Davis, 276 F. Supp. 507, 512–13 (N.D. Ill. 1967).
122. See Angel v. Bullington, 330 U.S. 183, 192 (1947) (“Of course, where resort is had to
a federal court not on grounds of diversity of citizenship but because a federal right is claimed,
the limitations upon the courts of a State do not control a federal court sitting in the State.”).
123. 383 U.S. 715 (1966).
supplemental jurisdiction decision.124 In support of the view that Erie
requires the application of state law in certain federal question cases,
regardless of which sovereign created the cause of action, many scholars and
judges turn to Hart & Wechsler,125 the Wright and Miller treatise,126
Commissioner v. Bosch,127 or United States v. Standard Oil.128 As will be
discussed below (and as already has been discussed in the case of Standard
Oil and related cases),129 however, not one of these sources adequately
explains or even supports the views expressed by the vast majority of scholars
The treatises are relatively uninformative. The Wright and Miller treatise
states succinctly that the view that Erie applies only to diversity cases “is
simply wrong,”130 citing to a Second Circuit decision that itself cites to Hart
and Wechsler.131 The Wright treatise also cites to Gibbs and Bosch, among
other opinions.132 Hart and Wechsler observe that the understanding that
Erie applies regardless of the basis of jurisdiction “has gained general
acceptance.”133 The two cases principally relied upon, Gibbs and Bosch,
require more discussion but do not clarify matters.
A. United Mine Workers v. Gibbs
The Supreme Court’s decision in Gibbs is frequently cited by both
courts134 and commentators135 in support of the proposition that Erie applies
to state law claims brought in federal court through pendent jurisdiction. In
Gibbs, the principal question involved whether subject matter jurisdiction
even existed for state law claims between nondiverse parties when those
claims related in some way to claims that were created by federal law and
amenable to arising under jurisdiction.136 The Court famously held that if
the state law claims shared a “common nucleus of operative fact” over claims
for which there was an independent basis for subject matter jurisdiction, a
federal court could exercise supplemental jurisdiction.137
While confirming the jurisdictional power of federal courts to assert
pendent jurisdiction, the Gibbs Court also gave lower courts discretion to
decline to exercise that power. Explaining the reason for this discretion, the
Court stated that if “considerations of judicial economy, convenience and
fairness” do not justify the assertion of pendent jurisdiction, “a federal court
should hesitate to exercise jurisdiction over state claims, even though bound
to apply state law to them.”138 To support the concluding clause, the Court
cited Erie, and it is this sparse language that many scholars and courts rely
on for the proposition that Erie translates seamlessly from diversity cases to
claims present in federal court via supplemental jurisdiction.
It should be clear from this discussion that the language relied upon did
not purport to apply Erie to pendent state law claims but was phrased in the
abstract. This on its own may be insufficient to question the reliance placed
on Gibbs, but the Court’s treatment of the substantive issues in the case,
virtually ignored by the commentators, raises serious questions. In Gibbs, it
must be remembered, the jury found against the plaintiff on his federal law
claims but awarded damages for state law claims for conspiracy and an
unlawful boycott. After the Court found that the district court permissibly
retained jurisdiction over the state law claims, it reversed liability under state
law. In so doing, the Court applied a federal statute, section 6 of the
NorrisLaGuardia Act, to limit the reach of state law, without any Erie or RDA
analysis.139 Indeed, the Court acknowledged that section 6 essentially
established a higher burden of proof for such claims brought against labor
unions, whether sourced in state or federal law.140 Thus, even in Gibbs, the
Court did not simply consider itself bound to apply state law, and only state
law, to pendent state law claims, despite the abstract language relied upon so
heavily by courts and commentators.141
B. Commissioner v. Estate of Bosch
If Gibbs is the case cited for the proposition that Erie applies with equal
force in supplemental jurisdiction cases, Bosch, a 1967 tax case, plays the
same role for those who insist that Erie applies with equal force in federal
question and diversity cases.142 A close read of Bosch, however, suggests
the contrary. In Bosch, the Supreme Court addressed the validity of a tax
deduction under 26 U.S.C. § 2056(b)(5), which permits “marital” deductions
against the estate tax.143 Central to the case was whether the decedent’s wife
validly executed an instrument related to the disposition of an inter vivos
trust.144 A state trial court, in a separate proceeding, had determined that the
execution was valid under state law, and the Supreme Court was therefore
compelled to determine the extent to which this state law determination was
binding in the federal tax court proceeding.145
To answer this question, the Court turned first to congressional intent
regarding the federal tax provision at issue, not to Erie.146 The Court
examined the legislative history regarding the enactment of the marital
deduction and noted that it “used very guarded language in referring to the
very question involved here,” stating that “‘proper regard,’ not finality”
should be given to state court interpretations of inter vivos instruments.147
The Court then turned to the RDA, and the associated Erie doctrine, to note
that its reading of congressional intent was “in keeping” with the policies
enunciated therein.148 And it finally closed with the language that most
commentators have focused on over the years:
This is not a diversity case but the same principle may be applied for the
same reasons, viz., the underlying substantive rule involved is based on
state law and the State’s highest court is the best authority on its own
law. . . . In this respect, [a federal court hearing a case involving
application of a federal statute in which an underlying issue of state law is
presented] may be said to be, in effect, sitting as a state court.149
This language, taken out of context, might suggest that Erie controls all
questions of state law presented in federal question cases, but the discussion
preceding it suggests the contrary. Moreover, the Court followed this key
language with the observation that its approach in this case “would be fair to
the taxpayer and protect the federal revenue as well,”150 suggesting the
relevance of these variables to its consideration of the question.
The dissenting opinions in Bosch also suggest that a straightforward
application of the Erie doctrine did not resolve the issue. Justice William
Douglas, writing alone, argued that the Court was, in fact, departing from the
Erie doctrine by permitting a federal court to ignore a state court decision
simply because it was not announced by the state’s highest court.151 Under
his interpretation of the Court’s prior decisions, unless there was reason to
suspect fraud or otherwise question the validity of the state court
determination, Erie required its application to a case like Bosch.152 Justices
John Marshall Harlan II and Abe Fortas, by contrast, found Erie and the RDA
to provide “relevant guidance” but read the Court as adopting a rule founded
not solely on Erie but also on federal interests in tax collection.153 The risk
of the majority’s view, according to these dissenters, could be “widely
this view, the Court stated, “We look at the problem differently,” understanding it instead as
a question of legislative intent. Id.
147. Id. at 464 (quoting S. REP. NO. 80-1013, pt. 2, at 4 (1948)).
148. See id.
149. Id. at 456.
150. See id. (emphasis added).
151. See id. at 466 (Douglas, J., dissenting).
152. Id. at 471 (“[W]here, absent [fraud, collusion, or other indications of unreliability], a
state court has reached a deliberate conclusion, where it has construed state law, the federal
court should consider the decision to be an exposition of the controlling state law and give it
effect as such.”).
153. In explaining their disagreements with both the majority opinion and Justice
Douglas’s dissent, Justices Harlan and Fortas noted that the latter placed too much reliance on
the importance of “uniformity in the administration of law within each of the States,” while
the former placed too much weight on the “hazards to the federal fisc from dubious decisions
of lower state courts.” Id. at 477–79 (Harlan, J., dissenting).
destructive both of the proper relationship between state and federal law and
of the uniformity of the administration of law within a State.”154
Thus, Bosch did not, as many commentators and courts suggest, adopt
Erie’s rule and apply it wholesale to state law issues raised in federal question
cases. Rather, the Court looked to both congressional intent regarding the
particular substantive federal statute at issue and salient federal interests to
conclude that, in the particular context of Bosch, interpretations of state law
should be determined by reference to the decisions of the state’s highest
III. ERIE AND THE NUANCED
To this point, I have shown that there is an orthodox view that Erie controls
in federal question cases, that courts and commentators have generally failed
to examine what this means when state law and federal law are intertwined,
and that the bases for the orthodoxy are insufficient. To move to a different
framework, however, requires a return to Erie and its progeny. All of the
Erie jurisprudence discussed herein revolves around the meaning of the
RDA, which provides:
The laws of the several states, except where the Constitution or treaties of
the United States or Acts of Congress otherwise require or provide, shall be
regarded as rules of decision in civil actions in the courts of the United
States, in cases where they apply.155
The RDA was originally enacted as section 34 of the Judiciary Act of 1789,
and has remained essentially unchanged since that time.156 There was
recognition from the start that vesting concurrent jurisdiction over state law
claims in both federal and state courts could present the possibility of
conflicts.157 Much ink has been spilled regarding whether Erie itself
correctly divined the RDA’s drafters’ intended resolution of such
conflicts,158 but, for the purposes of this Article, I take Erie’s interpretation
of the RDA as a given.
The central questions regarding application of the RDA have revolved
around when a federal court must apply state law and which state law it must
then apply. In Swift v. Tyson,159 a case founded on diversity jurisdiction, the
Court held that federal courts are bound to apply state law but only to the
extent that the state law was found in positive law. Where state law was
found only in the common law decisions of state courts, by contrast, the Swift
Court held that there was no requirement that federal courts follow those state
court decisions.160 As the Swift Court conceptualized it, state court decisions
“are, at most, only evidence of what the laws are, and are not, of themselves,
laws.”161 Swift thus famously ushered in the era of federal common law.162
For nearly one hundred years, federal courts were free to apply their own
interpretation of common law principles, even in diversity cases, but in Erie,
the Supreme Court overruled Swift.163 For the Erie Court, Swift was wrong
because it was based on a misinterpretation of the RDA, created perverse
outcomes, and was based on an unsupportable vision of the power of federal
courts.164 Thus, after Erie, it was commonly understood that, whatever the
source of state law, the RDA required that state law be applied, where
But therein, of course, lay the rub: Just when could a federal court rely on
federal law, whether created by statutes, the newly christened Federal Rules
of Civil Procedure, or even federal common law? Erie’s progeny clarified
the answer to this question. First, in Guaranty Trust Co. v. York,165 another
diversity case, the Court held that, to the extent that a federal court was
choosing between clashing federal and state law, the RDA only required the
application of state substantive law.166 This nuance was absent in Erie,
except to the extent that Justice Stanley Reed’s concurrence predicted the
problem,167 because the state law applied in Erie—relating to the obligations
owed by a railroad company to a trespasser or invitee—was indisputably
substantive. But in Guaranty Trust, the Court had to address whether the
conflict at issue—relating to a statute of limitations—was substantive, in
which event the RDA required the application of state law, or procedural,
which would permit a federal court to apply its own law, whatever the
source.168 The Court ultimately suggested that, to the extent that a choice
between federal and state law would change the outcome of the litigation, it
must be deemed substantive and the state law must be applied.169
Guaranty Trust’s outcome-determinative test was overly simplistic, in
large part because parties were unlikely to quarrel over the choice between
159. 41 U.S. 1 (1842).
160. Id. at 18–19.
161. Id. at 18.
162. Jonathan T. Molot, The Rise and Fall of Textualism, 106 COLUM. L. REV. 1, 60 (2006).
163. See Erie R.R. v. Tompkins, 304 U.S. 64, 79–80 (1938).
164. Molot, supra note 162, at 71–80.
165. 326 U.S. 99 (1945).
166. Id. at 107–08.
167. See Erie, 304 U.S. at 91–92 (Reed, J., concurring).
168. Guaranty Tr., 326 U.S. at 107–08.
169. Id. at 109.
federal and state law unless it changed the outcome of the case. In three
decisions after Guaranty Trust, all announced in 1949, the capaciousness of
the outcome-determinative test proved troublesome, especially as it related
to the clash between state laws and the Federal Rules of Civil Procedure.
Thus, in Ragan v. Merchants Transfer & Warehouse Co.,170 the Court held
that Federal Rule of Civil Procedure 3’s statement that an action
“commenced” with its filing did not determine the tolling of the statute of
limitations in a diversity case, where Kansas law provided that an action
“commenced” for the purpose of tolling when it was served on the defendant.
And in Cohen v. Beneficial Industrial Loan Corp.,171 the Court held that the
RDA required that a federal court apply a New Jersey statute requiring a
plaintiff in a shareholder derivative suit to post a bond even though the
applicable Federal Rule of Civil Procedure contained no such requirement.
Similarly, in Woods v. Interstate Realty Co.,172 the Court held that a
Tennessee corporation could not bring a diversity action in a federal court in
Mississippi if, by virtue of its failure to qualify to do business in that state,
the Mississippi state courts would bar such an action. In all three of these
cases, the Court applied Guaranty Trust’s outcome-determinative test to
determine the Erie question, and in all three cases, the Court displaced federal
The Court added another layer to the Erie analysis in Byrd v. Blue Ridge
Cooperative,173 another diversity case in which the Court displaced state law
in favor of conflicting federal law. In Byrd, the critical question was whether
the plaintiff was considered a “statutory employee,” in which case he was
prohibited from suing his employer in tort and was limited to recovery
delineated by South Carolina’s workers’ compensation scheme.174 South
Carolina vested fact-finding authority as to this question to a judge, whereas
federal law would permit the parties to invoke their Seventh Amendment
right to a jury trial on the question.175 In Byrd, the Court equivocated as to
whether the conflict between state and federal law was
“outcomedeterminative” as that term had been used in Guaranty Trust176 but
nonetheless found that resolving the Erie question involved the additional
step of considering whether federal interests in applying federal law
outweighed countervailing state interests.177 In Byrd itself, the Court found
that there was essentially no articulated state interest that justified vesting
fact-finding authority with the judge, while the federal interest resided in the
Seventh Amendment, making the balancing inquiry relatively
170. 337 U.S. 530 (1949).
171. 337 U.S. 541 (1949).
172. 337 U.S. 535 (1949).
173. 356 U.S. 525 (1958).
174. Id. at 527–28.
175. Id. at 533–37.
176. Compare id. at 537 (suggesting that the difference was outcome determinative), with
id. at 539–40 (suggesting the opposite).
177. Id. at 540.
178. Id. at 538–39.
After Byrd, the Erie analysis involved a three-step process. First a federal
court was required to determine whether a conflict existed between federal
law and state law. If no conflict existed, then no further Erie inquiry was
necessary. If a conflict was found, the court asked whether it was
outcomedeterminative. A negative answer resulted in application of federal law, on
the theory that the conflict only implicated substantive (and hence
state-lawderived) rights if it was outcome determinative. Even outcome-determinative
conflicts, however, required a third inquiry—Byrd’s balancing of federal
versus state interests to determine whether, even in the face of an
outcomedeterminative conflict, federal law should govern.179
One could conceptualize this inquiry as serving at least two, somewhat
different, purposes. First, the resort to Byrd balancing could be a means of
operationalizing the RDA’s instruction that state law applies in federal court
“except where the Constitution or treaties of the United States or Acts of
Congress otherwise require or provide.”180 The difficulty with this
conception is that the Erie analysis applies whether the source of state law is
positive law or judge-made common law.181 A competing conception of the
work that this Erie analysis accomplishes is that, by asking whether there are
significant state interests at stake in what appears to be an
outcomedeterminative conflict, the Court is essentially double-checking the intuition
that outcome-determinative conflicts are substantive. On this account, the
absence of state interests behind a rule that is outcome-determinative likely
indicates that it does not vindicate substantive judgments.
To this point, the Court had addressed Erie in cases that involved conflicts
between state law and federal law of all kinds, both positive law and common
law, without any suggestion that the source of federal law affected the
analysis. But in Hanna v. Plumer,182 the Court introduced two new
innovations in the Erie analysis, one of which is central to this Article. First,
and of less interest for these purposes, the Court clarified that, where the
source of federal law in any federal-state conflict is created by statute or a
Federal Rule, the “relatively unguided” Erie analysis was not applicable.183
Instead, a reviewing court must ask whether the positive law forecloses
application of the competing state law and, if so, whether the Constitution
(and in the case of Federal Rules of Civil Procedure, the Rules Enabling Act)
authorizes the federal positive law in question.184
More importantly for the purposes of this Article, however, the Hanna
Court also further clarified that the Erie-Guaranty Trust-Byrd analysis
governed conflicts between state law and federal judge-made law.185 Most
significantly, the Court stated that the outcome-determinative test introduced
179. For an example of Byrd balancing, see Wallis v. Pan American Petroleum Corp., 384
U.S. 63, 67–69 (1966), where the Court declined to rely on federal common law in a diversity
action because there was “no significant threat to any identifiable federal policy or interest.”
180. 28 U.S.C. § 1652 (2012) (emphasis added).
181. Erie R.R. v. Tompkins, 304 U.S. 64, 78 (1938).
182. 380 U.S. 460 (1965).
183. Id. at 470–72.
184. Id. at 471.
185. Id. at 468–69.
by Guaranty Trust was not as simple as had been suggested. The question
was not whether choosing state or federal law in the particular case would
alter the outcome—as previously noted, if it were, then nearly every conflict
presented to a court would pass the test. Instead, the Court stated that courts
must answer the outcome-determinative question in light of Erie’s twin aims:
reducing both forum shopping and inequitable outcomes based on the
citizenship of the parties.186
The Court’s reasoning was founded on the relationship between diversity
jurisdiction and the Erie doctrine. First, the Court emphasized that the Swift
doctrine had undermined the very purpose of diversity jurisdiction, which
was to level the playing field between citizens and noncitizens of a state.187
Because Swift made choice of law turn on whether the case was brought in
federal or state court, it gave the out-of-state citizen-plaintiff the power to
determine choice of law with the initial filing decision.188 Thus, Erie’s
concern, highlighted by Hanna, about the inequitable administration of the
laws was directly tied to the mechanics of diversity jurisdiction. Erie’s
concern about forum shopping, also highlighted by Hanna, was less directly
tied to diversity jurisdiction. Nonetheless, Erie’s concern seemed to be that
parties had manipulated their own citizenship to create diversity jurisdiction
so as to take advantage of a particular rule that would be applied under the
Swift doctrine.189 Nowhere did the Court in Hanna or Erie discuss concerns
about forum shopping in the context of federal question cases.
Since Hanna, little has changed with the Erie analysis, at least in its
outline. Disagreements have arisen about how to apply some of the
principles developed from Erie through Hanna, but the principles themselves
remain constant. The result the Court arrived at in Gasperini v. Center for
Humanities, Inc.,190 for example, may have been unorthodox, but its analysis
was basically consistent with what has been described so far.191 And Shady
Grove Orthopedic Associates, P.A. v. Allstate Insurance192 involved not the
“relatively unguided” Erie analysis, which is the focus of this Article, but
interpretation of the Rules Enabling Act analysis introduced in Hanna.
187. Id. at 467.
188. If the out-of-state plaintiff wished to take advantage of the state rule, she would file in
state court, preventing the in-state defendant from removing. 28 U.S.C. § 1441(b)(2) (2012)
(prohibiting removal on diversity grounds if the defendant is a citizen of the state in which suit
was brought). If the out-of-state plaintiff sought the benefit of the federal rule, she would file
in federal court, where the defendant could not shift the case to state court.
189. Erie R.R. v. Tompkins, 304 U.S. 64, 73 (1938) (discussing Black & White Taxicab &
Transfer Co. v. Brown & Yellow Taxicab & Transfer Co., 276 U.S. 518 (1928)); see also
Hanna, 380 U.S. at 467 n.8 (citing Black & White Taxicab, 276 U.S. at 518).
190. 518 U.S. 415 (1996).
191. The Court first applied Hanna’s outcome-determinative test before it turned to Byrd
balancing. Id. at 428–32.
192. 559 U.S. 393 (2010).
IV. INTRODUCING ERIE STEP ZERO
To this point, I have shown that there is an orthodox view that Erie applies
to all state court claims that arise in federal court, whatever the basis for
jurisdiction, but that courts and commentators rarely have paid attention to
whether Erie analysis should function in precisely the same way across
different jurisdictional contexts. Moreover, the Erie doctrine itself has been
justified by policies and concerns that are hard to translate outside of the
diversity context. The purpose of this part is to situate Erie analysis in the
nondiversity context and provide some examples of how Erie might function
differently outside of § 1332’s diversity jurisdiction.
In the conventional account, regardless of jurisdictional context, Erie
analysis consists of a number of steps, beginning with asking whether the
area in which state law and federal judge-made law conflict can be
characterized as obviously substantive or arguably procedural.193 To take
account of the insight I am advancing here, one must start earlier, at “Erie
Step Zero,” at which we must situate the conflict in its jurisdictional context.
For the purposes of this analysis, there is no need to revisit Erie analysis in
its purest form—where state law arises solely in the context of diversity
jurisdiction. Traditional Erie analysis is well equipped to address these
conflicts. Instead, there are two different jurisdictional contexts that I will
focus on here—state law claims that arise as supplemental claims in a federal
question case and state law claims over which a district court may assert
original jurisdiction because the claim implicates a substantial federal
interest. In these contexts, we are presented with three choices: apply Erie
analysis as if the case were a diversity case, not consider Erie (or the RDA)
at all, or find a way to translate the concerns of Erie and its progeny into the
federal question context.
The first and second options are not compelling. As to the first, as I have
discussed above, if conducting Erie analysis requires consideration of the
twin aims of Erie as defined by Hanna, many complications would ensue.
For Hanna may seek to reduce the incentive to forum shop as much as
possible, but the premise of federal question jurisdiction is that some forum
shopping is normatively acceptable, if not good.194 We make federal courts
available to vindicate federal interests because we think that federal courts
have more experience applying federal law, are more solicitous of federal
interests, and will strive for uniformity in application.195 Similarly, to the
extent that Hanna focuses on the potential effect that choice of law will have
on inequitable treatment according to citizenship, this concern does not exist
in the context of federal question jurisdiction.196
As to the second possibility, simply ignoring Erie principles and the RDA
in nondiversity cases is unsupported by the language of the RDA and the
jurisprudence discussed above.197 The difficulty, however, lies in crafting a
middle ground between these two extremes.
To do so, it is worthwhile to return to Hanna, which can guide the inquiry
in two distinct ways. First, Hanna clarifies that when one looks at outcome
determinativeness, one should be looking ex ante rather than ex post—
disputes over choice of law will always be outcome determinative ex post
because the parties would not be arguing about them otherwise. Second,
Hanna suggests that we should construct our choice of law rules to reduce
inappropriate forum shopping and to minimize inequitable treatment of
Translating Hanna into the federal-question context, then, requires that our
outcome-determinative test consider factors other than citizenship and the
risk that fundamentally state law claims will be driven into federal court to
take advantage of substantive law derived from federal sources. Rather than
inequity based on citizenship, we should be focused on potential inequity
based on the happenstance that a litigant is able to bring a state law claim into
a federal forum simply because it overlaps with the factual and legal contours
of a federal question claim or simply because it implicates the need to resolve
a substantial federal issue.198 We also would want to avoid adjudicating state
law claims differently simply because a litigant has chosen to frame them as
supplemental claims to a federal question anchor if the litigant also could
have brought them independently as diversity claims.199 And we might
196. Marsha S. Berzon, Securing Fragile Foundations: Affirmative Constitutional
Adjudication in Federal Courts, 84 N.Y.U. L. REV. 681, 704 (2009) (“A federal court
enforcing legal principles established by the federal Constitution or by federal statute is not
generating general common law in the forbidden Erie sense. . . . Its doing so does not raise
the worry that litigants with similar claims in state courts may receive substantively different
outcomes than those in federal court, the ‘mischievous results’ that prompted Erie in the first
place.” (footnote omitted)).
197. One could argue that state law does not “apply” in federal question cases and that
therefore the RDA is inapplicable, but this begs the question.
198. Patrick Woolley, The Sources of Federal Preclusion Law After Semtek, 72 U. CIN. L.
REV. 527, 557 (2003) (arguing that Erie is better characterized as a policy against “treating
any litigant who can take advantage of federal jurisdiction differently from those who
199. Note, The Competence of Federal Courts to Formulate Rules of Decision, 77 HARV.
L. REV. 1084, 1087–88 (1964) (“If such were the rule, not only could a party determine the
outcome of litigation by a choice of federal or state forum but under some circumstances he
could determine it by a choice of federal question or diversity jurisdiction.”). A district court
decision in a 1942 trademark case is the first, and only, opinion to consider this question in
the unique jurisdictional context where jurisdiction over state law unfair competition claims
could be founded on both diversity and supplemental jurisdiction. See Nat’l Fruit Prod. Co. v.
Dwinell-Wright Co., 47 F. Supp. 499, 501–05 (D. Mass. 1942), aff’d, 140 F.2d 618 (1st Cir.
1944). That opinion began by noting that Erie required a close analysis of the choice of law
to be applied to the unfair competition claims because the Supreme Court had not yet applied
Erie beyond diversity cases. The court recognized the arguments in favor of using federal
common law in an unfair competition claim pendent to a trademark claim and did not dispute
further distinguish federal question cases where exclusive jurisdiction is
present because in that case it is less likely that a litigant has exercised the
choice to bring supplemental state law claims in federal court simply to take
advantage of choice of law dynamics.200
As to forum shopping, we should be on the lookout for cases in which
parties choose federal court for reasons other than those recognized as valid
for the purposes of arising under and supplemental jurisdiction. The
motivation for bringing pendent state law claims in federal court may have
nothing to do with forum shopping but instead may relate to efficiency goals
or rules of preclusion, which create an incentive to bring the claims together
in the same jurisdiction.201 Indeed, if the federal question claim is the moving
force in the litigation, the litigant who invokes supplemental jurisdiction may
fundamentally be motivated by the desire to “resort to the experience,
solicitude, and hope of uniformity that a federal forum offers on federal
issues.”202 This choice is one to be supported, not one that should arouse
suspicion. And to the extent that forum shopping might be present, it would
likely be filtered out by a district court’s application of 28 U.S.C.
§ 1367(c)(2), which authorizes a district court to decline the exercise of
supplemental jurisdiction when the pendent claim “substantially
predominates” over the federal question claim.203 When federal question
jurisdiction is exercised over a state law claim that requires resolution of a
significant issue of federal law, § 1367 will not operate to weed out cases in
which state law substantially predominates, but Grable & Sons Metal
Products, Inc. v. Darue Engineering & Manufacturing204 and its progeny will
likely function in the same way.205
After one has conducted this outcome-determinative test, modified for
federal question jurisdiction, there still is the potential that Byrd balancing
should play a role. To be sure, if a court decides that the choice between
its power to do so, but ultimately it concluded that in the absence of statutory direction, and
given that the pendent claim also was brought pursuant to diversity jurisdiction, it would apply
state law. Id. at 503–05. The district court was no doubt influenced by its confidence that
Massachusetts state courts had developed a sophisticated common law for treatment of unfair
competition claims, but a fair reading of the court’s analysis leaves the impression that this
was not dispositive. Id. at 504.
200. Stein, supra note 37, at 2006 n.161 (making the argument in the context of exclusive
201. Erbsen, supra note 35, at 658 n.286.
202. Grable & Sons Metal Prods., Inc. v. Darue Eng’g & Mfg., 545 U.S. 308, 312 (2005).
203. Some ancillary claims, such as third-party claims, may look differently in this light.
In those cases, neither rules of preclusion nor mandatory joinder rules require that such claims
be brought in federal court, although efficiency considerations may favor including them.
204. 545 U.S. 30
205. In Empire Healthchoice Assurance, Inc. v. McVeigh, 547 U.S. 677 (2006), the Court
described Grable as a case in which the federal issue was “dispositive of the case.” Id. at 700.
In McVeigh, by contrast, the primary issues were “fact-bound and situation-specific,” causing
the Court to question why it should “place such a nonstatutory issue under the complete
governance of federal law, to be declared in a federal forum.” Id. at 701. Similarly, in Gunn
v. Minton, 133 S. Ct. 1059 (2013), the Court made clear that a federal issue must be substantial
both from the parties’ perspective and, more importantly, from the perspective of the “federal
system as a whole.” Id. at 1066. Therefore, cases that are heard under the Grable gloss on
§ 1331 will by definition involve important federal interests.
federal and state law is not outcome determinative, then it will be free to
apply the federal rule. If a court decides that the choice is outcome
determinative, however, it should move to Byrd balancing to determine
whether federal interests are sufficiently important to outweigh the state
interest in applying its rule. For if one is concerned that Hanna’s analysis
gives short shrift to federal interests in diversity cases,206 one should
presumably be at least as concerned when we move to the federal question
I concede that as a bottom-line matter, this analysis may result in courts
declaring that some rules are “substantive” for diversity purposes but not for
federal question purposes, which will be troubling to some. Some scholars
already recoil against the related idea of having “dual readings” of the Federal
Rules of Civil Procedure depending on whether the basis for jurisdiction is
diversity or federal question.208 But the Supreme Court itself has
acknowledged at times that the line between substance and procedure might
differ depending on whether jurisdiction is based on diversity or the presence
of a federal question.209
Lower courts have implied the same at times,210 but there is little
consistency in how courts have addressed the question. Many courts,
applying the intuition that Erie applies regardless of the basis for jurisdiction,
have concluded that state law controls both the federal and supplemental state
law claims, even if they would have disregarded state law were the federal
206. Martin H. Redish & Carter G. Phillips, Erie and the Rules of Decision Act: In Search
of the Appropriate Dilemma, 91 HARV. L. REV. 356, 360 (1977) (criticizing Hanna precisely
because it shifted focus from the balance between federal and state interests prioritized in Byrd
to avoidance of forum shopping and litigant inequity).
207. Cf. id. at 384 (“It would be unwise, however, to purchase a possibly greater degree of
predictability through use of the Hanna test where the price is the exaction of inescapable
costs on the federal system.”).
208. See Jennifer S. Hendricks, In Defense of the Substance-Procedure Dichotomy, 89
WASH. U. L. REV. 103, 125 (2011).
209. Levinson v. Deupree, 345 U.S. 648, 652 (1953). In Levinson, the Court held that
federal practice regarding amendments to pleadings applied to an admiralty case but
recognized that the issue might be analyzed differently were jurisdiction founded on diversity.
Id. (“Whether, if this were a diversity case, we would consider that we are here dealing with
‘forms and modes’ or with matters more seriously affecting the enforcement of the right, it is
clear that we are not dealing with an integral part of the right created by Kentucky.”).
210. For example, when lower courts heard Telephone Consumer Protection Act (TCPA)
cases under diversity jurisdiction, as they had assumed that the TCPA did not create a federal
cause of action, they applied Erie to choice of law disputes. See, e.g., Gottlieb v. Carnival
Corp., 436 F.3d 335, 342 (2d Cir. 2006); Holster v. Gatco, Inc., 485 F. Supp. 2d 179, 183–84
(E.D.N.Y. 2007), aff’d, 618 F.3d 214 (2d Cir. 2010), overruled on other grounds, Shady Grove
Orthopedic Assocs., P.A. v. Allstate Ins., 559 U.S. 393 (2010). After the Supreme Court
clarified that the TCPA does create federal question jurisdiction, see Mims v. Arrow Fin.
Servs. LLC, 565 U.S. 368 (2012), however, lower courts felt free to disregard Erie, see, e.g.,
Bailey v. Domino’s Pizza, LLC, 867 F. Supp. 2d 835, 839–40 (E.D. La. 2012). Even outside
of the TCPA context, courts have recognized the possibility that a state law could be
substantive for diversity but not federal question contexts. See, e.g., Doe v. City of Chicago,
883 F. Supp. 1126, 1134 (N.D. Ill. 1994) (discussing the role of a state law affidavit of merit
claim brought on its own.211 And some, while acknowledging the force of
the argument that there is a difference between the jurisdictional contexts,
ask if providing a different approach for such questions is justifiable as a
practical matter.212 But the lower courts have been forced to address this
question without sufficient guidance from the Supreme Court or
commentators. Therefore, this Article offers some examples of when my
proposed analysis will and will not affect choice-of-law conclusions.
Before turning to some specific examples of how Erie Step Zero might
have back-end choice-of-law consequences, let me first suggest the
circumstances in which my framework will not alter outcomes. For example,
in the supplemental jurisdiction context, if it is possible to isolate and apply
state law to state claims and federal law to federal claims, then traditional
Erie analysis should likely control. If, however, the choice of law relates to
an issue that cuts across claims or if applying state law will indirectly
interfere with or prejudice the adjudication of federal claims, then this
analysis offers reasons to question the traditional Erie analysis. In those
circumstances, dépeçage213 may not be available and a court will have to
choose whether to follow state or federal law. Some examples follow.
A. Erie Step Zero and Privilege Law
One of the most common areas in which a conflict between state and
federal law has arisen in federal question cases with supplemental state law
claims is privilege law.214 Indeed, in the debate regarding the adoption of
Federal Rule of Evidence 501, the Senate Judiciary Committee was
concerned about how the Rule 501 would be applied in federal question cases
211. Bradley v. City of Ferndale, 148 F. App’x 499, 511 (6th Cir. 2005) (looking to state
law to determine if denial of immunity is subject to interlocutory appeal where a state law
claim is supplemental to a § 1983 claim); Aliotta v. Nat’l R.R. Passenger Corp., 315 F.3d 756,
759 (7th Cir. 2003) (applying state law to a jury instructions issue involving a state law claim
in which there was federal question jurisdiction); In re Larry’s Apartment, L.L.C., 249 F.3d
832, 838 (9th Cir. 2001) (“[A] federal court sitting in diversity applies state law in deciding
whether to allow attorney’s fees when those fees are connected to the substance of the case.”);
Mangold v. Cal. Pub. Utils. Comm’n, 67 F.3d 1470, 1478 (9th Cir. 1995) (applying state law
to determine the fee award where a plaintiff in a civil rights case prevailed on federal and state
causes of action); Alvarado v. Fed. Express Corp., Nos. C 04-0098 SI, C 04-0099 SI, 2008
WL 2340211, at *4 (N.D. Cal. June 5, 2008) (“If the jury had based its verdicts solely on Title
VII, federal law would govern the award of fees. Because the jury verdicts were based on
both federal and state law claims, however, the Erie doctrine dictates that state law governs
the issuance of attorney fees.”); Chin v. DaimlerChrysler Corp., 461 F. Supp. 2d 279, 283
(D.N.J. 2006) (applying state law regarding attorney’s fees in a case in which state law claims
were supplemental to claims under the Magnuson-Moss Warranty Act).
212. See, e.g., Erbsen, supra note 35, at 658 n.286 (acknowledging possible difference in
interpretive approaches, but inviting analysis of whether “such confusing differences are
necessary in practice”).
213. Dépeçage is a choice of law technique in which rules from different legal systems are
used for different aspects of a lawsuit. See, e.g., Corporacion Venezolana de Fomento v.
Vintero Sales Corp., 629 F.2d 786, 794 n.8 (2d Cir. 1980).
214. Kaminsky, supra note 109, at 948 (recognizing that the question of privileges becomes
more difficult when there are supplemental state law claims in a federal question case).
with pendent state law claims.215 But Rule 501 did not resolve the problem,
stating unhelpfully that where state law “supplies the rules of decision,”
privilege claims should be governed by state law.216 The intent was to
incorporate Erie analysis into Rule 501 because there was some question
about whether state law as to privileges would be considered “substantive”
under Hanna.217 This works fine in a diversity case but not in a federal
question case with pendent state law claims, because states recognize
privileges from disclosure that could limit both the information disclosed
during pretrial discovery and the evidence presented at trial.218 If state law
privileges were applied during pretrial discovery to bar access to information
relevant to a federal question claim, it could substantially interfere with
enforcement of a federal right. If it were enforced during trial, such that
relevant statements were not admissible at all, even on federal law claims, or
certain statements were admissible on certain claims but not others, it would
either directly or indirectly interfere with distinct federal interests.
Using the framework proposed here, however, would clarify how courts
should resolve these issues. For example, if the state law privilege barred
access to evidence or discovery that was relevant only to the state law claim,
it could be enforced in full—in this circumstance, there would be no conflict
between applying state and federal privilege law. To the extent that there is
a conflict, however, one could rely on the modified Erie analysis to resolve
it. Do we expect, ex ante, that a federal forum was chosen so as to take
advantage of the federal privilege laws in the adjudication of the state law
claim? Is there a concern about inequity in the application of state privilege
law because of the happenstance that in this case the state law claims share
common facts with a federal question claim? Even if the answers to these
questions suggest that the choice of law is outcome determinative, how
should the federal interests be balanced against those of the state? Without
suggesting an answer to these questions, if enforcement of the state law
privilege law will have an impact throughout the entire case, it is more likely
that a court should choose to enforce federal common law.
As it turns out, most courts have come to a similar conclusion. Some
courts have held that federal law applies across the board to both the federal
claim and pendent state law claims.219 Others have held that the federal law
of privilege governs where the evidence sought is relevant to both federal and
state law claims, an outcome more consistent with my proposed
framework.220 But more difficult questions remain, particularly in the
context of state law claims for which there is original jurisdiction because of
an embedded substantial federal issue. In those cases, it may be impossible
to apply state privilege law to part of the claim and federal privilege law to
the rest, but it may also be difficult to define which sovereign provides the
rule of decision. Applying my framework, however, it seems unlikely that
one should even conclude that the choice between federal and state law is
outcome determinative in the relevant sense, for the very reason that federal
jurisdiction is available for the claim is because of the benefits of having a
federal forum to adjudicate an important federal issue.
B. Erie Step Zero and Affidavits of Merit
Many states require that an affidavit of merit be filed contemporaneously
with a complaint or shortly after a complaint has been filed when a claim
relates to medical malpractice.221 Most courts have concluded that these
requirements are substantive for Erie purposes and therefore have applied
them in diversity cases.222 Yet courts have not required that an affidavit be
filed to support a claim based on federal law that involves deficient medical
care, such as an Eighth Amendment violation for deliberate indifference.223
Where state law claims for deliberate indifference are joined to a federal
question claim, my proposed framework provides a simple solution: an
affidavit of merit may be required under state law to pursue the state law
claim but not to pursue a claim based in federal law. Even though many
state’s affidavits of merit define the cases to which the rule applies broadly
enough to encompass constitutional claims for deliberate indifference,224
applying the state law across the board would significantly interfere with
C. Erie Step Zero and Cost Shifting
Several courts have held, in the supplemental jurisdiction context, that the
Erie doctrine requires that state law govern awards of attorney’s fees in cases
claim, based both on interpretation of Rule 501 and principles of uniformity); In re
Combustion, Inc., 161 F.R.D. 51, 53–54 (W.D. La.), aff’d, 161 F.R.D. 54 (W.D. La. 1995)
(finding that federal law applied across the board in part because the relevant federal statutes
both provided for exclusive jurisdiction).
220. See supra note 108.
221. See Meryl J. Thomas, Note, The Merits of Procedure vs. Substance: Erie, Iqbal, and
Affidavits of Merit as MedMal Reform, 52 ARIZ. L. REV. 1135, 1139 (2010) (surveying state
statutes and reporting that more than a third of states have adopted this requirement).
222. See, e.g., Larca v. United States, 302 F.R.D. 148, 158 (N.D. Ohio 2014) (collecting
223. See, e.g., Jackson v. Grondolsky, No. 09-5617 (NLH/KMW), 2012 WL 960450,
at *13 (D.N.J. Mar. 20, 2012); Doe v. City of Chicago, 883 F. Supp. 1126, 1134 (N.D. Ill.
224. See, e.g., OHIO REV. CODE ANN. § 2305.113 (West 2015).
where a judgment is based on both federal and state claims.225 The
framework I propose here would raise questions about this approach. For it
is well recognized that, in areas such as Title VII, for instance, fee- and
costshifting provisions are part and parcel of effective enforcement of the
statutory regime.226 Thus, if applying state fee-shifting rules to federal law
claims undermined those substantive goals, it may be inappropriate to do so.
It may be that in cases that have common facts, costs and fees attributable to
a state law claim are indistinguishable from those attributable to a federal law
claim. But if that is the case, one still needs a valid reason to choose state
law over federal law to govern cost shifting.
D. The Boundaries of Erie Step Zero
There are other examples in which Erie Step Zero might apply, such as
commencement of an action for statute of limitations purposes,227 the
availability of remedies such as an anti-SLAPP motion,228 application of
forum non conveniens doctrine,229 or the impact of settlement releases and
assignability of interests.230 Consider the dispute at issue in Town of Newton
v. Rumery231: there the Court addressed whether a release-dismissal
agreement purporting to bar a person from filing a civil rights action in return
for a prosecutor’s dismissal of pending criminal charges was invalid as
against public policy. The Court rejected the First Circuit’s per se rule
invalidating such agreements and held that the bar should apply in that case
because the would-be plaintiff had voluntarily entered into the agreement,
there was no evidence of prosecutorial misconduct in obtaining the
agreement, and no indication that enforcement of the agreement would
“adversely affect the relevant public interests.”232 The Court, however, left
open the question of whether such agreements might be unenforceable in
other cases.233 And in a case like Rumery itself, in which the plaintiff had
brought § 1983 and pendent state law claims, a difficult question could arise
if we assume that state law would apply to the state law claims and federal
law—invalidating the agreement—would apply to the § 1983 claims.234
Nonetheless, there will be limits to our willingness to permit federal court
to displace state law in the examples I have provided. For instance, mere
recourse to the argument that uniformity is better when federal law is
involved should not typically be enough, absent a showing of how
disuniformity would interfere with the effective operation of federal law.235
And when we examine the potential for conflict between federal and state
interests, we might look for more than inconsistency and instead look for a
more direct clash.236 We might also be concerned if applying federal law
would “preclude the execution of state laws by state authority in a matter
normally within state power.”237 Balancing of state and federal interests will
almost always be central to the inquiry.238
232. Id. at 397–98.
233. Id. at 398 & n.10; see, e.g., Vallone v. Lee, 7 F.3d 196, 198 (11th Cir. 1993)
(distinguishing Rumery because the plaintiff presented evidence that he had been coerced into
signing the agreement); see also Livingstone v. North Belle Vernon Borough, 12 F.3d 1205,
1210 (3d Cir. 1993) (distinguishing Rumery on voluntariness and public interest grounds).
234. In Livingstone, the court assumed, in the absence of briefing, that state contract law
would govern the state law claims but concluded that the outcome would be the same under
both federal and state law. See Livingstone 12 F.3d at 1209 n.6. In a Northern District of Ohio
case, the parties did not address a similar question, and the court made no assumption about
which law would apply to the state law claims. See Kinney v. City of Cleveland, 144 F. Supp.
2d 908, 910 n.1 (N.D. Ohio 2001). A court in the District of Puerto Rico appears to have
assumed that its analysis of waiver law for the purposes of § 1983 was sufficient to permit
supplemental jurisdiction claims to go forward as well. Valle Colon v. Municipality of
Maricao, No. 09-02217(PG), 2011 WL 1238437, at *13
(D.P.R. Mar. 23, 2011)
. And a
Northern District of Oklahoma court relied on the potential merit and value of the plaintiff’s
federal and supplemental state law claims to distinguish Rumery and invalidate the waiver as
applied to all claims. See Spradlin v. City of Owasso, No. 12-CV-497-JED-FHM, 2014 WL
1664974, at *4 (N.D. Okla. Apr. 25, 2014).
235. Helen Hershkoff, Shady Grove: Duck-Rabbits, Clear Statements, and Federalism, 74
ALB. L. REV. 1703, 1720–22 (2011) (criticizing the Court’s implicit “clear statement” rule
applied in Shady Grove and proposing alternative approaches to balancing federal interest in
uniformity against “important state substantive concerns”); Theresa C. O’Loughlin, Adopting
State Law as the Federal Rule of Decision: A Proposed Test, 43 U. CHI. L. REV. 823, 840–41
(1976) (proposing that courts should look at whether a lack of uniformity would interfere with
“the functioning of a specific federal program” by making it administratively impracticable or
if disuniformity would conflict with the federal program).
236. O’Loughlin, supra note 235, at 838 (arguing that federal law should control if state
law operates as “a direct negation of particular purposes of the federal program”; otherwise,
courts should balance federal and state interests).
237. Davies Warehouse Co. v. Bowles, 321 U.S. 144, 155 (1944) (declining to use a
uniform federal rule in a rate regulation case).
238. Kaminsky, supra note 109, at 939 (arguing that federal courts should balance the
particular federal interests involved in a given case against the rationale of the relevant state
evidentiary privilege when deciding which privilege law to apply in federal question
litigation); O’Loughlin, supra note 235, at 829–30.
At the same time, however, it should be clear that my proposal does not
necessarily run afoul of federalism concerns. Even in diversity cases, the
Court has recognized that federal common law can displace state law.239
There also is federal power to displace state law to the extent it is necessary
to make the exercise of supplemental jurisdiction more effective. Take
Justice Antonin Scalia’s opinion for the Court in Jinks v. Richland County.240
In that case, a South Carolina county argued that 28 U.S.C. § 1367(d), which
tolls the statute of limitations for state law claims brought initially in federal
court over which a district court declines to exercise supplemental
jurisdiction, was unconstitutional because it displaced a state statute of
limitations.241 The Supreme Court held that § 1367(d) was within Congress’s
power, despite the fact that it affected a right that was “substantive” within
the meaning of Erie, for two reasons: (1) it “promotes fair and efficient
operation of the federal courts and is therefore conducive to the
administration of justice” and (2) “it eliminates a serious impediment to
access to the federal courts on the part of plaintiffs pursuing federal- and
state-law claims that ‘derive from a common nucleus of operative fact.’”242
My proposal here takes the intuitions from Jinks and suggests that that power
could be exercised, through the RDA, in federal question cases.
This reasoning, of course, does not eliminate the separation of powers
problem implicitly raised by Erie and its progeny when judges stray further
from the explicit dictates of a statute or from the dictates of the Rules
Enabling Act.243 There are forceful arguments against an expansive judicial
role in creating federal common law, particularly when state sovereignty is
threatened as a result. And this approach to Erie questions has had impact
beyond traditional choice-of-law conflicts. The view that Erie compels
application of state law in federal question cases has undoubtedly contributed
to the general skepticism of federal court lawmaking power even outside of
traditional Erie conflicts. Thus, whether it be filling gaps in a statutory
framework in federal question cases,244 inferring a damages remedy from the
239. See supra notes 58–67.
240. 538 U.S. 456 (2003).
241. The petitioner in Jinks initially filed a § 1983 claim in federal court with supplemental
state law claims. Id. at 460. The district court granted summary judgment on the § 1983 claims
and declined to exercise jurisdiction over the state law claims. Id. Petitioner refiled in state
court but had her jury verdict reversed by the South Carolina Supreme Court, which reasoned
that § 1367(d) could not toll South Carolina’s statute of limitations, because it interfered with
the state’s sovereignty. Id.
242. Id. at 462–63 (quoting United Mine Workers v. Gibbs, 383 U.S. 715, 725 (1966)).
243. Stephen B. Burbank, The Rules Enabling Act of 1934, 130 U. PA. L. REV. 1015, 1106
(1982) (“Nothing could be clearer from the pre-1934 history of the Rules Enabling Act that
the procedure/substance dichotomy . . . was intended to allocate lawmaking power between
the Supreme Court as rulemaker and Congress.”); see also Stephen B. Burbank,
Interjurisdictional Preclusion, Full Faith and Credit and Federal Common Law: A General
Approach, 71 CORNELL L. REV. 733, 757–59 (1986) [hereinafter Burbank, Interjurisdictional
244. See Kamen v. Kemper Fin. Servs., 500 U.S. 90, 98 (1991) (“Our cases indicate that a
court should endeavor to fill the interstices of federal remedial schemes with uniform federal
rules only when the scheme in question evidences a distinct need for nationwide legal
standards or when express provisions in analogous statutory schemes embody congressional
Constitution,245 or implying a private right of action from a federal statute,246
federal courts are increasingly hesitant to exercise what might be associated
with the common-law-making power exercised in the pre-Erie period.247
Despite the force of these arguments, substantial criticism has been levied
against the argument that any lawmaking power exercised by the federal
courts is inconsistent with separation of powers principles.248 And even
scholars who argue for an extremely formal and limited role for federal
common law—stemming from a principle such as “judicial federalism”—
acknowledge that the RDA’s “require or provide” language is broad enough
to tolerate some federal common law.249 Or, as Professor Steve Burbank put
policy choices readily applicable to the matter at hand. Otherwise, we have indicated that
federal courts should ‘incorporat[e] [state law] as the federal rule of decision,’ unless
‘application of [the particular] state law [in question] would frustrate specific objectives of the
federal programs.’” (alterations in original) (citations omitted) (quoting United States v.
Kimbell Foods, Inc., 440 U.S. 715, 728 (1979))); see also Camps Newfound/Owatonna, Inc.
v. Town of Harrison, 520 U.S. 564, 615 (1997) (Thomas, J., dissenting) (tracing the Erie
doctrine to reluctance to fill in gaps); Mark J. Loewenstein, Implied Contribution Under the
Federal Securities Laws: A Reassessment, 1982 DUKE L.J. 543, 573 n.156 (acknowledging
that state law might inform the law of contribution in federal securities law cases, even if Erie
does not apply, because federal courts may still look to state law for the rule of decision when
the federal statute is silent); Mark Moller, The Checks and Balances of Forum Shopping, 1
STAN. J. COMPLEX LITIG. 107, 142 (2012) (noting that the Court has relied on Erie to cast doubt
on federal courts’ power to fill in interstices in federal question cases, “unless federal courts’
authority to do so is clearly contemplated by a federal statutory scheme”). But see Lind, supra
note 26, at 285–86 & nn.146–51 (citing examples of federal question cases in which courts
declined to adopt state law comparability review of damages awards because of a conflict with
federal common law).
245. Carlson v. Green, 446 U.S. 14, 39 (1980) (Rehnquist, J., dissenting) (analogizing the
jurisprudence of Bivens v. Six Unknown Named Agents of Fed. Bureau of Narcotics, 403 U.S.
388 (1971), to federal common-law-making power “repudiated” by Erie); Berzon, supra note
196, at 703–04 (arguing that the conflation of Bivens jurisprudence with pre-Erie federal
common law misreads Erie because when enforcing the Constitution or federal statutes, there
is no need to worry over whether “litigants with similar claims in state courts may receive
substantively different outcomes than those in federal court”).
246. Sosa v. Alvarez-Machain, 542 U.S. 692, 726 (2004) (linking Erie to a “significant
rethinking” of the role federal courts play in implying private rights of action); id. at 730–31
(suggesting that “more expansive common law power related to 28 U.S.C. § 1331” might
be inconsistent with Erie)
; Cannon v. Univ. of Chi., 441 U.S. 677, 742 (1979) (Powell, J.,
dissenting) (arguing that Erie compels the abandonment of the implied right of action doctrine
of Cort v. Ash, 422 U.S. 66 (1975)); Wheeldin v. Wheeler, 373 U.S. 647, 651 (1963) (“As
respects the creation by the federal courts of common-law rights, it is perhaps needless to state
that we are not in the free-wheeling days antedating Erie.”).
247. Looking just at Bivens jurisprudence, for example, the Supreme Court has refused
since 1980 to recognize any new Bivens cause of action. See generally Alexander A. Reinert
& Lumen N. Mulligan, Asking the First Question: Reframing Bivens After Minneci, 90 WASH.
U. L. REV. 1473 (2013).
248. See, e.g., William N. Eskridge, Jr., Dynamic Statutory Interpretation, 135 U. PA. L.
REV. 1479, 1498–502 (1987) (“[T]he structure of the Constitution, the apparent expectations
of the Framers, and two hundred years of Supreme Court practice establish the authority of
federal courts to make law, subject to legislative override.”).
249. Thomas W. Merrill, The Common Law Powers of Federal Courts, 52 U. CHI. L. REV.
1, 31–32 (1985) (“A more cautious conclusion would be that the Rules of Decision Act is at
least consistent with what is conventionally thought of as ‘interpretation’—enforcing the
intentions of the draftsmen of the text.”). For Merrill, however, judicial displacement of state
law by federal common law is only appropriate when “inability to impose a federal common
law rule could frustrate specifically intended federal policies.” Id. at 34. Merrill identifies
it slightly differently, “If the Constitution or acts of Congress, fairly read,
provide for or require federal common law, state law does not apply” within
the meaning of the RDA.250 In that sense, the proposal I present here could
be viewed as one method for determining when federal law requires or
provides for federal common law.
For too long, courts and commentators have assumed that Erie analysis
applies to RDA problems in nondiversity cases without fully elaborating on
what that means. My goal in this Article is to show that we need a different
framework for addressing RDA problems in federal question cases, one that
takes some of the broad insights from Erie and its progeny but translates them
across jurisdictional lines. Erie Step Zero provides this structure, focusing
on the specific forum shopping, inequity, and balancing concerns that can be
implicated outside of the diversity context so as to ensure that courts do not
reflexively displace federal law in any case in which a state law claim is
presented. Although my proposed framework may not result in radically
different outcomes in federal question cases, I hope that it provides a better
guide to decision making.
6. See United Mine Workers v . Gibbs , 383 U.S. 715 , 725 ( 1966 ).
7. See 28 U.S.C. § 1367 ( 2012 ); Gibbs, 383 U.S. 715 . These state law claims share a common nucleus with a parallel federal law claim asserted in the plaintiff's complaint or may enter a federal question case as compulsory or permissive counterclaims raised by the defendant . See, e.g., Michelle S. Simon, Defining the Limits of Supplemental Jurisdiction Under 28 U .S.C. § 1367: A Hearty Welcome to Permissive Counterclaims, 9 LEWIS & CLARK L. REV . 295 , 303 ( 2005 ) (explaining that compulsory counterclaims by definition fall within Gibbs's “common nucleus of operative facts” test); see also Global NAPs, Inc . v. Verizon New Eng. Inc., 603 F.3d 71 , 76 - 77 , 88 ( 1st Cir . 2010 ) (concluding that § 1367(a)'s language is broader than the “transaction or occurrence” language in Federal Rule of Civil Procedure 13(a)).
8. See generally Gunn v. Minton , 133 S. Ct . 1059 ( 2013 ) ; Empire Healthchoice Assurance, Inc . v. McVeigh , 547 U.S. 677 ( 2006 ); Grable & Sons Metal Prods., Inc. v. Darue Eng'g & Mfg., 545 U.S. 308 ( 2005 ).
9. See Hanna v. Plumer , 380 U.S. 460 , 468 - 69 ( 1965 ).
10. 380 U.S. 460 ( 1965 ).
11. See id. at 468-69.
41. See John Hart Ely, The Irrepressible Myth of Erie , 87 HARV. L. REV. 693 , 737 n. 226 ( 1974 ).
42. See Holmberg v. Armbrecht , 327 U.S. 394 ( 1946 ) (“The considerations that urge adjudication by the same law in all courts within a State when enforcing a right created by that State are hardly relevant for determining the rules which bar enforcement of an equitable right created not by a State legislature but by Congress .”).
43. See generally Westen & Lehman, supra note 3.
44. See id. at 312-14.
45. See id. at 330- 76 . Martin Redish provided a scathing response to these aspects of the Westen and Lehman article . See Martin H. Redish , Continuing the Erie Debate: A Response to Westen and Lehman, 78 MICH. L. REV. 959 ( 1980 ).
46. Redish , supra note 45, at 959 ( referring to this portion of Westen and Lehman's article as “almost . . . an afterthought”).
47. See Westen & Lehman, supra note 3, at 378- 80 . Their approach is based on their reconceptualization of Erie as not involving a choice between state and federal law but instead simply an assessment of whether there is valid and pertinent federal law on point . See id. at 314-16 . It is precisely this characterization of Erie that Redish spends considerable time criticizing . Redish, supra note 45 , at 960-64.
48. See Westen & Lehman, supra note 3, at 386-87, 386 n.218.
49. Redish is certainly correct for noting that Westen and Lehman's conception of Erie appears to rest on a misreading of Hanna's modification to Guaranty Trust's outcomedeterminative test . Compare Redish, supra note 45 , at 966-67 ( noting that Westen and Lehman appear to assume that the outcome-determinative test is simply whether, ex post, a choice between state and federal law will result in a different outcome), with Westen & 1338 , 1361 - 62 ( E.D. Tex . 1993 ) (holding that Texas pleading law applied to the question of what constituted a “well-pleaded complaint”).
57. FDIC v. Countrywide Fin. Corp., No. 2 : 12 -CV-4354 MRP (MANx) , 2012 WL 5900973, at *14 n.28 ( C.D. Cal . Nov. 21 , 2012 ) (citing Steinman , supra note 25 , at 311- 12 for the proposition that “some portions of the Erie doctrine likely apply” in federal question cases).
58. 376 U.S. 398 ( 1964 ).
59. See id. at 425.
60. See id.
61. See id. Sabbatino was overruled by statute . See Westen & Lehman, supra note 3 , at 339-40; see also Foreign Assistance Act of 1964 , Pub. L. No. 86 - 633 , § 30 ( d ), 78 Stat. 1009 , 1013 (codified as amended at 22 U.S.C. § 2370 (e)( 2 ) ( 2012 )).
62. 360 U.S. 593 ( 1959 ).
63. See id. at 597 (“ We hold that the validity of petitioner's claim of absolute privilege must be judged by federal standards, to be formulated by the courts in the absence of legislative action by Congress .”).
64. See id.
65. 487 U.S. 500 ( 1988 ).
66. See Doernberg, supra note 26 , at 655-56.
67. See Boyle, 487 U.S. at 507 ( quoting Wallis v . Pan Am. Petroleum Corp ., 384 U.S. 63 , 68 ( 1966 )).
68. See Hinderlider v. La Plata River & Cherry Creek Ditch Co. , 304 U.S. 92 , 101 , 110 ( 1938 ).
69. 318 U.S. 363 ( 1943 ).
70. See id. at 366-67.
71. See Doernberg, supra note 26 , at 648-50 (discussing Clearfield Trust in detail).
72. Four years after Erie, the Court stated that Erie does not control a federal question case and that state law must “yield” when it clashes with federal law and policy . See Sola Elec . Co. v. Jefferson Elec. Co., 317 U.S. 173 , 176 ( 1942 ) ; see also Deitrick v . Greaney , 309 U.S. 190 , 200 - 01 ( 1940 ) (holding that where a lawsuit is based on federal law, Erie has no application because “[t]he extent and nature of the legal consequences of this condemnation, though left by the statute to judicial determination, are nevertheless to be derived from it and the federal policy which it has adopted”). Ten years after Erie, scholars could observe that the Clearfield Trust line of cases could be interpreted as indicating that Erie is limited to diversity cases, even if the Court had not justified them explicitly on those grounds . See, e.g., Gorrell & Weed, supra note 31 , at 295.
73. 315 U.S. 447 ( 1942 ).
74. 313 U.S. 487 ( 1941 ).
75. See D'Oench , Duhme & Co. at 456-59.
76. See id. at 465 ( Jackson , J ., concurring) (“I think we should attempt a more explicit answer to the question whether federal or state law governs our decision in this sort of case than is found either in the opinion of the Court or in the concurring opinion of Mr . Justice Frankfurter.”).
77. See id. at 470-72.
78. See id. at 473-75.
79. See United States v. Standard Oil Co ., 332 U.S. 301 , 305 ( 1947 ) (applying federal common law because the case involved government-soldier relations). The Court of Appeals had applied state law, relying on the Erie doctrine, id . at 303 n. 4, but the Court, extending the logic of Clearfield Trust, held that where a case involved the “duties imposed upon the United States and the rights acquired by it,” federal law should govern, id . at 305.
80. See Burks v. Lasker , 441 U.S. 471 , 476 ( 1979 ) (holding that Erie does not apply to resolve questions of the power of disinterested directors to terminate a derivative suit because “we proceed on the premise of the existence of a federal cause of action”).
81. See Norfolk & W. Ry . Co. v. Liepelt, 444 U.S. 490 , 493 ( 1980 ) (holding that federal common law governs assessment of damages in Federal Employers' Liability Act cases ).
82. See Holmberg v. Armbrecht , 327 U.S. 392 , 395 ( 1946 ). The Holmberg Court distinguished Guaranty Trust as a case involving the “duty of a federal court, sitting as it were as a court of a State, to approximate as closely as may be State law in order to vindicate without discrimination a right derived solely from a State,” rather than a federal court tasked with enforcing an equitable right created by Congress . Id.
83. See Boyle v. United Techs., 487 U.S. 500 , 511 - 12 ( 1988 ); Burks, 441 U.S. at 477 ( using federal common law because rules “impact significantly upon the effectuation of federal rights”); Howard v . Lyons , 360 U.S. 593 , 597 ( 1959 ); Standard Oil, 332 U.S. at 307.
84. See Standard Oil, 332 U.S. at 309-10; Holmberg, 327 U.S. at 395.
85. See Standard Oil, 332 U.S. at 310-11; Clearfield Tr. Co. v. United States , 318 U.S. 363 , 366 - 67 ( 1943 ).
86. Burks , 441 U.S. at 476 (“ Since we proceed on the premise of the existence of a federal cause of action, it is clear that 'our decision is not controlled by [Erie],' and state law does not operate of its own force.” (quoting Sola Elec . Co. v. Jefferson Elec. Co., 317 U.S. 173 , 176 ( 1942 ))).
87. 433 U.S. 25 ( 1977 ).
88. See id. at 29 (“ While federal common law may govern even in diversity cases where a uniform national rule is necessary to further the interests of the Federal Government . . . the
124. See , e.g., Childress, supra note 26 , at 1545 n.96 (citing Gibbs and stating that Erie applies to state law claims heard in federal court via supplemental jurisdiction); Dorf , supra note 27, at 284 & nn. 221 - 22 (citing Gibbs, 383 U.S. at 725) (stating that federal courts follow state law in diversity cases, supplemental claims in federal question cases, and cases involving due process where the court must first resolve an issue of state law); Matasar , supra note 27, at 1489 & n.401 (citing Gibbs for the proposition that questions of state law are governed by Erie in federal question cases ); Yonover, supra note 27, at 310 n.22; see also Fondrk v. Westmoreland County , No. Civ.A. 04-0900 , 2006 WL 1699471, at *2 ( W.D. Pa . June 20, 2006 ); Kline v . Sec. Guards, Inc., 159 F. Supp . 2d 848 , 851 (E.D. Pa . 2001 ).
125. See , e.g., Campos, supra note 21 , at 1586 n.73; Chow, supra note 27, at 166 n.4; Gluck, supra note 21, at 1926 n.87; Gluck , The Federal Common Law of Statutory Interpretation, supra note 27 , at 785 n.122.
126. See , e.g., Gluck, supra note 21 , at 1926; see also Ohio Valley Envtl . Coal., Inc. v. Marfork Coal Co., 966 F. Supp . 2d 667 , 678 n.6 (S.D. W. Va . 2013 ) (citing the Wright and Miller treatise for the proposition that it “simply is wrong” to maintain that Erie applies only in diversity cases); Monmouth Cty . Corr. Inst. Inmates v. Lanzaro , 643 F. Supp . 1217 , 1221 (D.N .J. 1986 ) (citing the Wright and Miller treatise), aff'd in part, modified in part, 834 F .2d 326 ( 3d Cir . 1987 ); In re Hilt, 175 B.R. 747 , 753 ( Bankr. D. Kan . 1994 ) (same).
127. 387 U.S. 465 ( 1967 ) ; see , e.g., Saunders, supra note 36 , at 722 n. 33 ( citing Bosch , 387 U.S. 456 ); see also Fondrk, 2006 WL 1699471, at *2 ( citing Bosch , 387 U.S. 456 ); Hernas v . City of Hickory Hills , 507 F. Supp . 103 , 105 (N.D. Ill . 1981 ) (citing Bosch, 387 U .S. at 461).
128. 382 U.S. 301 ( 1947 ) ; see , e.g., Calvin R. Massey , Abstention and the Constitutional Limits of the Judicial Power of the United States, 1991 BYU L . REV. 811 , 845 (citing Standard Oil , 332 U.S. at 307).
129. See supra notes 79-96 and accompanying text (noting that the Court has consistently declined to apply the Erie doctrine in federal question cases even where state law claims and issues were present).
130. See supra note 24.
131. Maternally Yours v. Your Maternity Shop , 234 F.2d 538 , 540 n. 1 ( 2d Cir . 1956 ) (“But despite repeated statements implying the contrary, it is the source of the right sued upon, and not the ground on which federal jurisdiction over the case is founded, which determines the governing law.” (citing HENRY M. HART & HERBERT WECHSLER , THE FEDERAL COURTS AND THE FEDERAL SYSTEM 690-700 ( 1953 ))).
132. 19 WRIGHT & MILLER, supra note 24, § 4520 nn.6 , 9 & 11 .
133. RICHARD H. FALLON, JR . ET AL., HART AND WECHSLER'S THE FEDERAL COURTS AND THE FEDERAL SYSTEM 589 (7th ed. 2015 ).
134. See , e.g., Fondrk v . Westmoreland County , No. Civ.A. 04-0900 , 2006 WL 1699471, at *2 ( W.D. Pa . June 20, 2006 ); Kline v . Sec. Guards, Inc., 159 F. Supp . 2d 848 , 851 (E.D. Pa . 2001 ).
135. See supra note 124 and accompanying text .
136. United Mine Workers v. Gibbs , 383 U.S. 715 , 721 ( 1966 ).
137. Id . at 725. Supplemental jurisdiction doctrine evolved substantially from Gibbs until Congress enacted 28 U .S.C. § 1367 . Judicial Improvement Act of 1990 , Pub. L. No. 101 - 650 , tit. 3, § 310 , 104 Stat. 5089 , 5113 (codified at 28 U.S.C. § 1367 ( 2012 ) ) . The Gibbs nucleus test was essentially codified in § 1367(a ).
138. See Gibbs, 383 U.S. at 726 ( emphasis added) (citing Erie R .R. v. Tompkins, 304 U.S. 64 , 64 ( 1938 )).
139. Id . at 735-42.
140. See id. at 736-37.
142. See , e.g., Jack Moore , United States v. White: The Second Circuit Validates an IRS Role in Policing State Probate Practice , 56 BROOK. L. REV. 669 , 681 - 82 ( 1990 ); Saunders, supra note 36, at 663 n. 33 (“ There is sufficient authority to support the proposition that the doctrine applies in all actions regardless of the basis of federal jurisdiction unless a sufficient federal interest exists to mandate the application of federal common law.” (citing Comm'r v . Estate of Bosch , 387 U.S. 456 ( 1967 ))); Yonover, supra note 27, at 310 n.22.
143. See Bosch, 387 U.S. at 459-61.
144. Id . at 473; see also Moore, supra note 142 , at 681 n.75.
145. See Bosch, 387 U.S. at 464.
146. Indeed, the Court recognized that some lower courts had found that the question was parallel to, if not controlled by, Erie principles . See id. at 463 ( citing Faulkerson's Estate v . United States , 301 F.2d 231 ( 7th Cir . 1962 ), as an example of this view) . After identifying
154. Id . at 480. Thus , Harlan and Fortas argued that state court determinations of state law should be controlling, regardless of their procedural level unless the state court proceeding was not “genuinely” adversarial . Id. at 481.
155. 28 U.S.C. § 1652 ( 2012 ).
156. It was codified as part of the U.S. Code in 1948, at which time Congress made clear that it applied to equitable claims as well as legal claims . See Act of June 25 , 1948 , ch. 646 , 62 Stat . 944 (codified at 28 U.S.C. § 1652 ).
157. See Wythe Holt, “To Establish Justice” : Politics, the Judiciary Act of 1789, and the Invention of the Federal Courts, 1989 DUKE L .J. 1421 , 1487 n.232.
158. Compare Charles Warren, New Light on the History of the Federal Judiciary Act of 1789 , 37 HARV. L. REV. 49 , 86 ( 1923 ) (providing the argument relied upon by the Erie Court, based on legislative history of the RDA), with W. RITZ, REWRITING THE HISTORY OF THE JUDICIARY ACT OF 1789: EXPOSING MYTHS , CHALLENGING PREMISES , AND USING NEW EVIDENCE 126-48 ( 1990 ) (“The one thing that can be said with assurance is that Section 34 was not intended to apply exclusively to diversity proceedings; that it was not intended to direct the application of the law of particular states in diversity proceedings; and that it was not intended to apply to suits in equity . In short, on its historical basis, Erie is dead wrong .”), and Holt, supra note 157, at 1506-07 (arguing in support of Ritz's interpretation) .
193. See supra Part II.
194. See , e.g., Sidney Schenkier , Ensuring Access to Federal Courts: A Revised Rationale for Pendent Jurisdiction , 75 NW. U. L. REV. 245 , 254 - 55 ( 1980 ) (stating that one purpose of Reconstruction was to provide a federal forum for resolution of federal claims); see also Doe v . City of Chicago , 883 F. Supp . 1126 , 1134 (N.D. Ill . 1994 ) (stating that forum shopping is not a concern in arising under cases).
195. Grable & Sons Metal Prods., Inc. v. Darue Eng'g & Mfg., 545 U.S. 308 , 312 ( 2005 ).
215. Id . at 958-60 ( detailing rulemakers' concern that Federal Rule of Evidence 501 might apply differently depending on the jurisdictional basis of the claim ).
216. FED . R. EVID. 501 ; see also In re Yasmin & Yaz (Drospirenone) Mktg ., Sales Practices & Prods. Liab. Litig., No. 3 : 09 -md-02100 - DRH-PMF, 2011 WL 1375011, at *1 ( S.D. Ill . Apr. 12 , 2011 ) (MDL action) .
217. See H.R. REP . NO. 93 - 650 , at 9 ( 1973 ) (stating that privilege rules should be considered substantive for Erie purposes ); see also In re Grand Jury Investigation, 918 F.2d 374 , 379 n. 6 ( 3d Cir . 1990 ); Laxalt v . McClatchy , 116 F.R.D. 438 , 446 (D. Nev . 1987 ); Theodore Campagnolo, The Conflict Between State Press Shield Laws and Federal Criminal Proceedings: The Rule 501 Blues , 38 GONZ. L. REV. 445 , 500 ( 2002 ); Korn, supra note 28, at 75.
218. Rule 501 applies to all stages of a civil proceeding, not just at trial . See FED. R. EVID . 1101 ( c ).
219. See Tucker v. United States , 143 F. Supp . 2d 619 , 622 - 25 (S.D. W. Va. 2001 ) (applying federal privilege law to a Federal Tort Claims Act claim and to a pendent state law
225. See supra note 211.
226. See , e.g., SEAN FARHANG , THE LITIGATION STATE: PUBLIC REGULATION AND PRIVATE LAWSUITS IN THE U .S. 4 , 62 ( 2010 ) ; Margaret H . Lemos, Special Incentives to Sue, 95 MINN. L. REV. 782 , 793 - 94 ( 2011 ) ; cf . Chambers v. Nasco, Inc., 501 U.S. 32 , 55 - 56 ( 1991 ) (concluding that the district court acted within its discretion in a diversity case by invoking its inherent power and assessing attorney's fees and expenses as a sanction for a party's bad faith litigation conduct, and rejecting an Erie challenge to the district court's action) .
227. Although state law governs commencement in diversity cases, see Walker v. Armco Steel Corp ., 446 U.S. 740 , 753 ( 1980 ), federal law governs commencement in federal question cases, see West v . Conrail , 481 U.S. 35 , 39 n. 4 ( 1987 ). While it may be possible to apply state and federal law simultaneously in some supplemental jurisdiction cases, a court would have to make a choice in those cases in which § 1331 jurisdiction is asserted over a state law cause of action.
228. Globetrotter Software , Inc. v. Elan Comput. Grp., Inc., 63 F. Supp . 2d 1127 , 1129 - 30 ( N.D. Cal . 1999 ) (applying anti-SLAPP law to state law claims asserted pendent to federal question claims but not to federal question claims).
229. The Supreme Court has, for instance, declined to decide whether forum non conveniens analysis is a matter of state or federal law in diversity cases . See Piper Aircraft Co. v. Reyno , 454 U.S. 235 , 248 n. 13 ( 1981 ). But it has at least implied that it should be considered procedural even if state courts might displace federal forum non conveniens law in admiralty and other cases . See Am. Dredging Co. v. Miller , 510 U.S. 443 , 454 & n.4 ( 1994 ).
230. Caleb Nelson , The Persistence of General Law , 106 COLUM. L. REV. 503 , 528 - 31 ( 2006 ) (arguing that “general law,” labeled as federal common law, is operative in cases implicating interstate disputes, rights and obligations of the national government, customary international law , and maritime law).
231. 480 U.S. 386 ( 1987 ).