Erie's four functions: reframing choice of law in federal courts.

Author:Erbsen, Allan
Position:III. Erie's implementation: Rethinking doctrinal Puzzles through Conclusion, with footnotes, p. 617-659

Parts I and II focused on Erie's functions; this Part focuses on its form. The morass of cases constituting Erie jurisprudence address several recurring problems that pose enduring puzzles. This Part illustrates how a more nuanced account of Erie's components can provide a new perspective on puzzles that have confounded courts and commentators for decades. For each puzzle, courts need to know whether their task is to create federal law, interpret federal law, prioritize conflicting laws, or adopt non-federal law. Courts likewise must recognize the difference between choosing an authoritative government, an authoritative institution within that government, and a rule that the institution would endorse.

The following sections analyze seven puzzles implicating all four of Erie's components. The goal is not to provide definitive answers, although I do offer some. Instead, the discussion exposes common themes underlying ostensibly dissimilar puzzles and places each puzzle in a context that helps to identify solutions.

First, analyzing multiple puzzles in the same article highlights how confusion in each area arises for similar reasons. This similarity is often difficult to discern because judicial opinions and scholarship usually address individual puzzles in isolation. Each problem might therefore appear to be difficult for idiosyncratic reasons. The analysis in this Part reveals that many puzzles arise from or become more difficult because of a shared pathology: a failure to parse Erie into its components. Each puzzle may raise some unique issues, but a more precise account of Erie's distinct components would place those issues in a context that can lead to better solutions.

Second, the analysis in this Part situates each puzzle within one of Erie's components and shows how that context can lead to doctrinal reform. A more precise account of where each puzzle fits within the broad Erie landscape helps determine what questions courts should ask and how to formulate coherent answers.

  1. Creation and the Puzzle of General Law: When Can Federal Common Law Incorporate or Choose General Law Rules that Could Not Apply of Their Own Force?

    Erie's central holding that "[t]here is no federal general common law" walks a semantic tightrope between the concepts of "federal common law" and "general common law." (128) Shaking the rope highlights how Erie failed to articulate a normative theory that could shape the role of general law in the federal system. The potential utility of such a theory is evident in debates about whether federal common law should apply in some diversity cases and about the status of customary international law in United States courts.

    The Court in Erie could not reject "federal common law" because federal courts routinely apply such law in multiple contexts. Judges and scholars often do not agree on when to apply and how to craft federal common law. But there is a broad consensus that courts can create federal common law in at least some cases that need a federal solution, but for which the Constitution, legislative action, and executive action have not directly supplied an answer. (129) Examples include rules governing interstate and foreign relations, the federal government's proprietary interests, and admiralty. (130)

    Likewise, the Court could not deny the existence of "general common law" because the Court itself had been diligently consulting such law since the Founding. Justice Brandeis doubted whether general law was a stable and legitimate source of judicial authority. (131) However, he could not deny general law's historical and contemporary salience as a source of ideas and guidance. (132) Indeed, on the same day the Court decided Erie, Justice Brandeis filed another majority opinion applying a "federal common law" rule of "equitable apportionment" for resolving interstate water disputes. (133) That judicially created rule apparently borrowed from general law, which did not seem to disturb the Court. (134)

    The continued vitality of general law is consistent with Brandeis's endorsement of Justice Holmes's argument that "law in the sense of which courts speak of it today does not exist without some definite authority behind it." (135) A central premise of Holmes's critique was that prior decisions had deemed general law "obligatory" rather than merely available. (136) Erie eliminated the obligatory component of general law, mooting further discussion of what role, if any, general law might play in a federal system. By this deft maneuver, Erie avoided saying that general law did not exist and held instead that general law was not binding in diversity cases. (137)

    Given that "federal common law" and "general common law" both survived Erie, the Court was careful to condemn only "federal general common law." Yet that novel term apparently had never appeared in any published judicial opinion before Erie. (138) The ambiguous term raises a question: What exactly did Erie reject? If general law survives in some form, but not as "federal general common law," what role can general law continue to play? The answer is not clear because Erie did not carefully distinguish between rules that incorporate general law and rules that adopt general law of its own force. Closer attention to the difference between Erie's creation and adoption inquiries might have highlighted the need for a normative theory explaining when federal courts can create rules that incorporate general law.

    One possible reading of Erie is that its rejection of "federal general common law" means that federal courts can never incorporate general law into federal law. Yet that conclusion is implausibly broad for three reasons.

    First, the Supreme Court has often expressly relied upon general law after Erie, notably in maritime cases. (139) The wisdom of this reliance is debatable. (140) But there is little doubt that if a federal court may craft federal common law in a particular case, then it has discretion to shape the rule (within limits). (141) This discretion includes authority to incorporate general law principles. (142) At that point the incorporated general law becomes federal law that applies in state courts under the Supremacy Clause (143) and can be the basis for federal question jurisdiction. (144)

    Second, the Constitution itself incorporates (or at least references) the general law of nations in clauses governing foreign relations, (145) This general law is a source that federal courts consult to explain both the Constitution's original public meaning and its modern application. (146)

    Third, many statutes incorporate general law as a means of filling gaps in coverage or implementing a vague governing standard. A rare explicit example is a statute criminalizing "piracy as defined by the law of nations." (147) Implicit reliance on general law is more common. (148)

    In sum, reading Erie to reject the incorporation of general law into federal law is implausible. General law principles permeated federal law before Erie and survived Erie's repudiation of "federal general common law."

    Another possible interpretation of Erie is that even if federal courts can incorporate general law into federal law, they cannot adopt general law of its own force. On this view, Swift was wrong because it admitted relying on general law, which it grandiosely claimed--citing Cicero and Lord Mansfield--was law throughout the "world." (149) Rejecting this aspect of Swift is appealing to a generation instilled with the positivists' attempt to ground law in sovereign authority and the legal realists' skepticism of law's pretentions to uniformity. (150)

    But this anti-adoption interpretation is less helpful than it may seem for two reasons. First, Swift may not have adopted general law and instead relied on general law because that was the law that New York courts would have relied upon. If so, Erie's non-adoption rule would not address the holding in Swift. Second, even if Swift did adopt general law, adoption can easily be recharacterized as incorporation. Indeed, it is possible to imagine a hypothetical variation of Swift that would survive Erie's anti-adoption rule and would be unconstitutional only if there is a normative argument against incorporation. Yet, as the following discussion illustrates, Erie never advanced a normative argument against incorporating general law into federal common law for purposes of diversity litigation.

    On one reading, Swift merely applied New York's local choice of law rule, which in turn selected general law. (151) The propriety of Swift would then hinge on whether a federal court could interpret the general law differently than the courts of the state whose choice of law rule it was enforcing. (152) That inquiry requires considering whether state courts were adopting general law or instead incorporating general law into their own law. Justice Story either overlooked or finessed this question, noting only that state courts would "deduce" the rule of decision from general law. (153) This phrasing is opaque in a legal environment where courts routinely borrow each other's ideas. The key question is what status the idea has after it is borrowed, rather than its status before it was borrowed. Justice Story focused only on where New York found rules, not on what those rules became when New York relied on them.

    Moreover, Justice Story began his analysis with an assumption about New York law that distorted his conclusion about general law. His opinion observed that New York's highest court had never endorsed the outlier holder in due course exception that Swift rejected, (154) but then assumed that the exception would not apply in federal court even if the state's highest court had endorsed it. (155) The problem is that because the highest court had not in fact endorsed the exception, lower courts had no occasion to decide if a higher court decision would bind them...

To continue reading