Printer Friendly

The twin aims of Erie.

III. THE TWIN AIMS OUTSIDE DIVERSITY

To sum up, I have sought to justify the role the twin aims play in diversity cases in terms of federal policies grounded in the diversity statute. The question remains whether they play a role outside of diversity. Because the twin aims have their source in federal jurisdictional policies, not concerns about state interests, there is no reason to think that they must apply simply because a federal court entertains a state law cause of action. The question must be answered on the basis of the statute giving the federal court jurisdiction. Conversely, one cannot conclude that the twin aims do not apply to a federal court simply because it entertains a federal cause of action, for the federal jurisdictional policies in the statute giving the federal court jurisdiction might implicate the twin aims. It is now time to consider these matters.

A. Federal Causes of Action

The Supreme Court has held that federal courts have no duty to ensure uniformity with forum state procedure when the plaintiff sues under federal law. (177) For example, a federal court entertaining a federal cause of action calculates prejudgment interest according to a uniform federal common law rule, (178) even though a federal court sitting in diversity must borrow the method that would be used by a forum state court. (179) Likewise, in Semtek the Supreme Court made it clear that the claim-preclusive effect of a dismissal by a federal court of a federal cause of action is governed by a uniform federal common law rule, (180) even though the claim-preclusive effect of a dismissal of a state law action by a federal court sitting in diversity should be borrowed from forum state law. (181)

Similarly, a federal court entertaining an action under a federal statute that lacks a limitations period has no duty under the twin aims to incorporate the forum state's period into a federal common law rule. (182) To be sure, federal courts find such incorporation convenient. (183) Furthermore, since Congress can be understood to have omitted a limitations period with this federal judicial practice in mind, there is arguably a presumption in favor of such borrowing, a presumption that must be overcome by some significant federal interest. (184) But the reason for borrowing is not the twin aims--that is, the desire for uniformity with how a forum state court would treat the federal action. Concern with a forum state court's treatment of the federal action is entirely absent.

The inapplicability of the twin aims in federal question actions is sometimes justified on the ground that states lack any interest in their rules applying in federal court in such cases. (185) But such a justification presumes that the twin aims are connected to Erie's goal of respect for state lawmaking power, which we now know to be false. The issue is not state interests, but the federal interests standing behind the statute giving federal courts jurisdiction. So the question remains: Why isn't uniformity with forum state procedure recommended by the federal question statute?

1. Federal Question Jurisdiction

To answer this question, we must look to the purposes of federal question jurisdiction. (186) Three are usually offered: 1) federal judges are more experienced in federal law and so are more likely to apply it correctly; 2) federal courts are necessary to enforce the supremacy of federal law against a possibly hostile state judiciary; and 3) lower federal courts are needed, in addition to the Supreme Court, to resolve any disuniformity that results from state court interpretation of federal law. (187)

Can we generate an argument that the twin aims apply in federal question actions on the basis of these purposes? If state courts were understood as the primary fora for federal causes of action, with federal question jurisdiction created solely to address the problem of state courts' incompetent, hostile, or disuniform interpretation of federal law, one might argue that Congress limited federal courts' power to create procedural common law by the twin aims. (188) Under this reading, Congress gave federal courts' power over federal causes of action for the narrow purpose of addressing these deficiencies with state courts. This was not a license to create procedural common law to serve other federal interests.

No one has ever suggested that federal courts entertaining federal causes of action are bound by the twin aims in this manner. Why not? The reason must be that everyone assumes that Congress did not consider state courts as the presumptive fora for federal causes of action. Instead, it conceived of federal courts' relationship to federal causes of action as analogous to the relationship that state courts have to state causes of action. Federal courts do not entertain federal causes of action solely to avoid incompetent, hostile, or disuniform interpretation of federal law by state courts. They have a more fundamental entitlement to entertain actions under federal law, an entitlement that carries with it the power to create procedure to vindicate any legitimate federal interest.

The idea that federal courts are not subsidiary fora for federal causes of action is arguably evident in the ratification debates. As Madison noted, no one was uncomfortable about federal question jurisdiction: "With respect to the laws of the Union, it is so necessary and expedient that the judicial power should correspond with the legislative, that it has not been objected to." (189) William R. Davie expressed the same view: "I thought, if there were any political axiom under the sun, it must be, that the judicial power ought to be coextensive with the legislative. The federal government ought to possess the means of canting the laws into execution." (190) Granted, these statements might be understood as expressing the view that federal question jurisdiction exists for a narrow purpose only, namely to enforce the supremacy of federal law against unwilling states. So understood, federal courts' power over procedure might be limited by this purpose. But there appears to be a more basic notion at work, one tied to the idea that the judicial power should be coextensive with the legislative. A court has a fundamental entitlement to apply the law of its own sovereign. Just as a Pennsylvania court is entitled to entertain Pennsylvania causes of action, a federal court is entitled to entertain federal causes of action. This is a reason to give federal courts jurisdiction over federal causes of action even if state courts interpreted and enforced federal law ably, willingly, and with perfect uniformity. It follows from this entitlement that a federal court entertaining a federal cause of action should have the power to create procedural law to effectuate any forum purpose, just as state courts do when entertaining actions under their own law.

Of course, the force of this entitlement should not be exaggerated, for federal question jurisdiction did not generally exist in the lower federal courts for almost a century. As we shall see, this is some evidence that state and federal courts were conceived of as coequal fora for federal causes of action. But it is not likely that when Congress created federal question jurisdiction in 1875, it conceived of federal courts as subsidiary to state courts. It intended federal courts to have procedural power unrestricted by the twin aims.

2. Reverse-Erie

Having come to the conclusion that everyone has assumed--that the twin aims do not bind a federal court entertaining a federal cause of action--we must now address the more difficult question of whether they bind a state court entertaining a federal cause of action. This requires us to consider the purposes for which Congress retained concurrent state court jurisdiction for federal causes of action.

Arguments for a reverse-twin aims approach, under which state courts borrow federal procedure, are based on the notion that federal courts are the presumptive fora for federal causes of action, and state court jurisdiction is derivative--existing to overcome some deficiency with exclusive federal jurisdiction. Tasked with accepting jurisdiction to satisfy this purpose, state courts are obligated to regulate their procedure accordingly.

Assume, for example, that Congress understood federal courts as the presumptive fora for federal causes of action, but retained concurrent state court jurisdiction to keep the federal docket from being overwhelmed or because it wanted to protect some interest of the parties in having a choice between a state and federal forum for their federal dispute. Given these purposes, state courts would have an obligation to borrow federal procedure. If there were a significant difference between state and federal procedure, forum shopping would frustrate the federal purposes standing behind the grant of concurrent state court jurisdiction. For example, if Congress gave state courts jurisdiction over federal actions to relieve pressure on the federal docket, procedural disuniformity would frustrate this purpose because parties disfavored by state procedure would choose the federal forum, flooding the federal courts. Or if Congress gave state courts jurisdiction to protect the parties' discretion to choose between a federal and state court, a difference between federal and state procedure would restrict this freedom of choice.

What is more, even if forum shopping is set aside, the second of the twin aims would apply. We are assuming that Congress gave state courts jurisdiction over federal causes of action solely to address a problem with the presumptive federal forum. It would not follow from this limited grant of jurisdiction that state courts had the procedural power they have when entertaining causes of action under their own or sister state law.

But a very different theory of state court jurisdiction for federal actions is possible. Congress may have retained such jurisdiction, not to solve a particular problem with exclusive jurisdiction in the presumptive federal forum, but because it conceived of state courts as having the same entitlement to entertain federal actions that they have to entertain actions under their own or sister state law. Under this equality approach, the twin aims would not bind state courts. State and federal courts entertaining federal causes of action would each be free to create procedural law to satisfy forum purposes, no matter how much vertical disuniformity might result. It would not matter, for example, that a state had procedural rules that discouraged parties from suing on federal causes of action in the state's courts, for the failure of the parties to seek out state jurisdiction would frustrate no federal jurisdictional purpose. The state's procedural rule would be acceptable, despite this forum shopping, as long as its purpose was neutral concerning federal and state causes of action. States could not, of course, create procedural rules that had as their goal discrimination against federal law. (191)

So which is correct, the equality or reverse-twin aims approach? We cannot answer the question by reference to the legislative history of the Judiciary Act of 1875, which is silent on this, and pretty much every other, matter. (192) But the notion that state courts have a fundamental entitlement to entertain federal causes of action is plausible, and there is some evidence for such a reading in the ratification debates. (193) Hamilton argued, for example, that Article III's grant of federal judicial power over actions arising under federal law did not divest state courts of their preexisting power to entertain such actions. To be sure, federal causes of action did not preexist Article III. (194) But, he noted, prior to Article III a New York state court had the power to "lay[] hold of all subjects of litigation between parties within its jurisdiction, though the causes of dispute are relative to the laws of the most distant parts of the globe. Those of Japan, not less than of New York, may furnish the objects of legal discussion to our courts." (195) New York state courts had an entitlement to entertain causes of action under any sovereign's law--including the law of a sovereign, like the United States, that had yet to come into being.

To the extent that Congress retained state court jurisdiction for federal actions in recognition of this state entitlement, rather than to overcome some deficiency with exclusive jurisdiction in federal courts, it would have allowed state courts to retain plenary power over procedure. (196) A New York state court entertaining a federal cause of action should have the same procedural power that it has when entertaining a Pennsylvania or Japanese cause of action. (197)

Of course, it is within Congress's authority to override state courts' entitlement to entertain federal actions. It can create exclusive federal subject matter jurisdiction. (198) And it can choose to retain state court jurisdiction over federal actions solely to address a problem with federal jurisdiction--for example, to release the pressure on the federal docket or to protect the parties' freedom of choice between state and federal court. If so, the twin aims would apply to state courts. The question is solely one of discerning congressional intent. But the fact that state courts were generally the sole fora for federal causes of action until 1875 makes it unlikely that the creation of federal question jurisdiction suddenly made state courts subsidiary to federal courts. Jurisdiction in state court was probably retained to give state courts the opportunity to exercise their preexisting entitlement to entertain federal actions. Given this purpose, the equality approach, under which the twin aims do not apply, would follow. (199)

If, as I believe, the twin aims are inapplicable to state courts, this makes a big difference to reverse-Erie cases. Many of these cases have been wrongly decided.

In determining the role of the twin aims in reverse-Erie cases, it is important to distinguish between a state court being obligated to use a federal rule because of the federal jurisdictional purposes expressed in the twin aims and because the federal rule is bound up with the federal action, in the sense that the failure to apply the federal rule will frustrate the action's substantive regulatory purposes. This is the reverse-Erie equivalent of the distinction between a federal court being obligated to use a state procedural rule because of the twin aims and because it is bound up with the state law cause of action in the manner discussed in Byrd.

Of course, distinguishing between the two in a reverse-Erie context isn't easy, because the Supreme Court is the final arbiter concerning both. In diversity cases, in contrast, the state supreme court has the final say on whether a state rule is bound up with the state's cause of action, and the Supreme Court is the final authority on the effect of the twin aims.

Nevertheless, I think the two types of federal interest--the first driven by jurisdictional purposes tied to the grant of concurrent jurisdiction to state courts and the second driven by the substantive regulatory purposes of the federal law under which the plaintiff sues--can be distinguished. In deciding which is implicated, it is helpful to consider the scope of the federal cause of action if it were entertained by the courts of a foreign nation, to which federal jurisdictional interests could not extend.

Consider Central Vermont Railway Co. v. White, (200) in which the Supreme Court held that a federal rule under which the burden of proving contributory negligence is on the defendant should be used by a state court entertaining an action under FELA. This is probably an example of a federal substantive interest, in which failure to use the federal rule would frustrate the purpose of the federal cause of action. State courts entertaining causes of action under sister state law generally use the sister state's rule concerning the burden of proof, on the theory that sister state officials consider the rule sufficiently important to follow sister state causes of action into other court systems. (201) Thus it is plausible that the Supreme Court, if asked, would hold that the federal rule concerning the burden of proof follows FELA actions into the courts of foreign nations.

With the distinction between federal jurisdictional and substantive interests in mind, it does not look, at first glance, as if the twin aims play a role in reverse-Erie cases. (202) The usual reason the Supreme Court gives for state courts' duty to apply a federal rule when entertaining a federal cause of action is that the rule is bound up with the action, in the sense that the failure to apply the rule will frustrate the action's substantive purposes. (203)

An example is Atlantic Coast Line Railroad Co. v. Burnette, (204) in which the Supreme Court held that a state court entertaining a FELA action was bound to use FELA's limitations period rather than the forum state's longer period. (205) In favor of its conclusion, the Court cited Davis v. Mills, (206) which adopted the choice-of-law rule that another sovereign's statute of limitations should be respected by the forum if the limitations period is part of the cause of action upon which the plaintiff sues. (207) This strongly suggests that the Court understood FELA's statute of limitations to be folded into FELA actions, following such actions into other court systems. So understood, the FELA statute of limitations would follow FELA actions even into the courts of other nations. (208)

Two other examples are Dice v. Akron, Canton & Youngstown Railroad Co., (209) and Brown v. Western Railway of Alabama. (210) In Dice, the Court held that a state court entertaining a FELA action was bound to use the federal rule, giving an issue to the jury, on the grounds that the right to trial by jury was "too substantial a part of the rights accorded by the Act to permit it to be classified as a mere 'local rule of procedure." (211) And in Brown, it held that a Georgia rule that construed allegations against the pleader would "impose unnecessary burdens upon rights of recovery authorized by federal laws." (212) The "federal right," it argued, "cannot be defeated by the forms of local practice." (213)

The notion that federal rules must apply in state court because they are bound up with the federal law under which the plaintiff sues can also be found in more recent cases, such as Felder v. Casey. (214) In Felder, the Court held that a federal civil rights action brought in state court in Wisconsin could not be dismissed because of a failure to comply with Wisconsin's notice-of-claim statute. (215) Once again, the Court appealed primarily to interference with the regulatory interests standing behind federal civil rights laws. (216) Thus, it argued that Wisconsin's statute "so interferes with and frustrates the substantive right Congress created that, under the Supremacy Clause, it must yield to the federal interest." (217) This suggests that had the plaintiff sued the defendant for violation of federal civil rights in the court of another nation, the foreign court's application of a notice-of-claim statute would also be impermissible (by American lights).

I think it is clear, however, that in many reverse-Erie cases--including some of those discussed above--the Supreme Court has misdescribed why it thinks state courts should use federal rules. The real reason is the twin aims. (218) Consider, for example, Engel v. Davenport, (219) in which the Supreme Court held that a state court was obligated to use the federal statute of limitations for FELA, instead of the state's shorter limitations period, on the ground that the federal limitations period "is one of substantive right, setting a limit to the existence of the obligation which the Act creates." (220) From a horizontal choice-of-law perspective, this sounds odd. Even when a sister state's statute of limitations is substantive, the forum is usually thought to be permitted to apply its own shorter period, (221) on the ground that the dismissal is without prejudice. The plaintiff is free to sue elsewhere. (222) The dismissal does not affect the right, only the ability to get a remedy within that forum.

It is puzzling, therefore, for the Court to hold that it is a substantive part of a FELA action that a state court take jurisdiction of the action, rather than dismissing it without prejudice. In what sense are the purposes of FELA frustrated by the fact that the action proceeds elsewhere (say, in federal court)? It seems much more likely that reason for the Court's conclusion is the twin aims. Federal interests tied to the grant of concurrent jurisdiction to state courts, not the regulatory purposes of FELA, are at issue. The state court is understood as obligated to take the FELA action when a federal court would. It cannot have procedural law that would substantially affect the choice of a federal or state forum. This explains why courts entertaining reverse-Erie cases occasionally speak of a "desirable uniformity in adjudication of federally created rights" (223) and "the federal interest in uniformity." (224)

The same point applies to Felder. In her dissent, Justice O'Connor rightly argued that Wisconsin's notice-of-claim statute did not burden the plaintiff's federal right, for he could have chosen a federal forum, where the statute would not have applied. (225) Indeed, under Wisconsin law dismissals for failure to satisfy the statute are without preclusive effect. (226) The plaintiff was still free to sue again in federal court.

Because she believed that Wisconsin's statute did not frustrate the regulatory purposes of federal civil rights law, Justice O'Connor could not see why the statute should be preempted. (227) In short, she adopted the equality approach--in which federal and state courts are considered coequal fora for the litigation of federal causes of action. Just as federal courts entertaining federal causes of action are not restricted in the purposes for which they may create procedural common law, state courts (and legislatures) are not restricted in the purposes for which they may create procedural law for their courts, even when these courts are entertaining federal actions. States' only duty when creating procedural law is to respect the substantive content of the federal action itself and to not have procedural rules that single out federal causes of action for disfavorable treatment. (228)

In contrast, the majority in Felder--although speaking of state courts' duty not to create procedural law that "interferes with and frustrates the substantive right Congress created" (229)--really adopted the reverse-twin aims approach, according to which state courts' uniformity with federal procedure is important even when the regulatory purposes of the federal cause of action are not frustrated. (230) States must create procedural law in a manner limited by the twin aims. (231)

The battle between the equality and the reverse-twin aims approaches can be seen in the most recent reverse-Erie case: Haywood v. Drown. (232) Haywood concerned a New York statute according to which any civil action for damages against prison personnel for torts committed within the scope of their employment had to be brought in the New York Court of Claims. (233) Because this court could not entertain federal civil rights suits, the effect of the statute would be to remove state court jurisdiction for federal civil rights suits brought against prison personnel. A majority of the Court held that the New York statute was preempted. (234)

Justice Thomas's dissent exemplified the equality approach. The New York statute, he argued, would not frustrate the federal right, for the plaintiff would remain free to sue in federal court. (235) He therefore saw no reason that the New York statute should be preempted. New York was free to create procedural rules on the basis of New York interests, so long as these rules did not have as their purpose discrimination against federal law, (236) and the substantive purposes of federal actions were not undermined.

I do not want to suggest that it is beyond Congress's power to command state courts to take jurisdiction of federal causes of action, state rules on the matter notwithstanding. It is even within Congress's power to preempt state procedure for such actions, to the extent that they diverge from the procedure that would be used by federal courts. The question is solely one of discerning congressional intent. But I think that Thomas was probably right: Congress, in giving state courts concurrent jurisdiction for federal civil rights actions, (237) did not intend that the twin aims apply there.

B. State Causes of Action in Federal Court

Let us now move on to whether the twin aims apply to federal courts entertaining state law actions under jurisdictional statutes other than 28 U.S.C. [section] 1332(a). I will begin with a general argument that the twin aims should be used whenever a federal court entertains an action under state law, no matter what the source of jurisdiction.

Consider, once again, why the twin aims apply in diversity. A federal court sitting in diversity was understood by Congress as a subsidiary forum in the state in order to address a particular deficiency with the state's courts, namely bias against non-domiciliaries. Procedural uniformity with forum state courts fosters this purpose by ensuring that disadvantageous federal procedure does not discourage those worded about state court bias from seeking a federal forum, and that those who are not worried about state court bias do not waste federal judicial resources by choosing a federal forum because of its advantageous procedure.

But even if there were no possibility of forum shopping, for example, if diversity cases had exclusive federal jurisdiction, the fact that federal courts were understood by Congress as subsidiary fora means that any federal procedural common law that significantly deviates from the procedure of the forum state still needs substantial justification. Congress did not give federal courts jurisdiction over diversity actions to come up with federal procedural common law to satisfy any conceivable federal interest.

It is highly probable that all other forms of federal jurisdiction for state law actions are, like diversity, created by Congress to address particular deficiencies with the presumptive state fora. Congress creates federal jurisdiction for state law actions for reasons, and these reasons must be that something about state court jurisdiction is inadequate. Given this fact, the power of federal courts to create procedural common law that deviates from the presumptive state fora must be justified by the purposes for which federal jurisdiction was created, or other countervailing federal interests of sufficient strength.

I am not suggesting that it is impossible for Congress to give to federal courts entertaining state law actions the plenary power over procedure that state courts have when entertaining their own or sister state actions. (238) But in the absence of evidence that this was Congress's intent, a federal court entertaining a state law action should be understood as a federal extension of a state court system. Thus, any deviation from the procedure that would be used in that state's courts needs a substantial justification. Notice that the purpose of this justification is not to overcome the state's interest in extending its procedural laws to federal courts. The federal court's subsidiary role is imposed by Congress, not a state.

A number of scholars have argued that CAFA allows federal courts to use independent federal choice-of-law rules, rather than the choice-of-law rules of the state where the federal court is located. See Samuel Issacharoff, Settled Expectations in a World of Unsettled Law: Choice of Law After the Class Action Fairness Act, 106 COLUM. L. REV. 1839, 1840 (2006); Linda Silberman, The Role of Choice of Law in National Class Actions, 156 U. PA. L. REV. 2001, 2002 (2008). I doubt, however, that they think the twin aims do not apply at all in CAFA actions. They would not say, for example, that federal courts have the power to apply an independent federal common law time limit to state law actions in federal court under CAFA. CAFA is not a license to federal courts to effectuate any federal policy they might come up with in a class action setting.

The argument that Klaxon does not apply in CAFA is probably based, not on the view that the twin aims do not apply at all in CAFA cases, but on the view that Klaxon is contrary to the particular purposes for which federal jurisdiction was created in CAFA. CAFA was enacted because of distrust of the manner in which state courts were handling nationwide class actions. See Class Action Fairness Act of 2005, Pub. L. No. 109-2, [section] 2(a)(4) (codified as amended at 28 U.S.C. [section] 1711 (2006)); Burbank, supra note 140, at 1950. According to this argument, CAFA overrides Klaxon because state courts' tendency to manipulate choice of law to aid certification was one of the (alleged) abuses that CAFA was meant to address. S. REP. No. 109-14, at 26-27 (2005). Under such an argument, however, CAFA overrode Klaxon only with respect to forum state choice-of-law rules that enhance the probability of certification. Burbank, supra note 140, at 1950-51. An example is a rule that subjects all the members of the class to a single state's law (such as the law of the defendant's principal place of business). E.g., Int'l Union of Operating Eng'rs Local 68 Welfare Fund v. Merck & Co., 894 A.2d 1136, 1153 (N.J. Super. Ct. App. Div. 2006). In other respects, Klaxon would still apply. I cannot address the role in CAFA of Klaxon, and the twin alms generally, further here, however.

With this general argument in mind, let us turn to two specific jurisdictional statutes, starting with supplemental jurisdiction.

1. Supplemental Jurisdiction

Do the twin aims apply to state law actions brought in federal court under supplemental jurisdiction, as has been widely assumed? (239) To repeat, if they are, the reason cannot be respect for state lawmaking power, for the twin aims are unconcerned about state interests. (240) Assume a federal court in New York takes a Pennsylvania cause of action under supplemental jurisdiction. If the twin aims apply, the court should borrow New York's statute of limitations, if that is what a New York state court would use, even if New York has no interest in extending its statute to federal courts within the state. Whether the twin aims apply in supplemental jurisdiction must be answered, not by looking to state interests, but by looking to the purposes of the supplemental jurisdiction statute. (241)

Let us begin with the purposes of supplemental jurisdiction in diversity cases. Consider a Nevadan sued by a Californian for state law negligence in California state court. Fearing state court bias, she removes the action to federal court. In the absence of supplemental jurisdiction, an impleader she has against a Nevadan who was a joint tortfeasor in the accident will have to be litigated in state court. This is inefficient, given that the evidence in the state and federal cases would be largely the same. Indeed, faced with the costs of duplicative litigation, she might fail to remove in the first place, despite her worries about bias. (242) It is to address these problems that Congress created supplemental jurisdiction in diversity cases. (243) Supplemental jurisdiction is efficient and supports the underlying purposes of diversity jurisdiction.

It follows from these purposes that the twin aims should apply to the impleader. With supplemental jurisdiction in place, it is likely that the negligence action and the impleader will be litigated together in federal or state court. Given this fact, the twin aims must apply to both actions in federal court or the purposes of diversity jurisdiction will be frustrated. Assume that the federal court applied a uniform federal common law limitations period to the impleader. If this period is shorter than that used by a California state court, the Nevadan might refrain from removing the negligence action to keep the impleader from being dismissed as time barred, despite her worries about state court bias concerning the negligence action. And if the federal common law period were longer, she might remove the negligence action, even though she had no worry about state court bias, in order to take advantage of the longer period for the impleader. (244) To ensure that her choice of a federal forum for the negligence action is made for the right reason, procedural uniformity between federal and forum state court for both actions is required.

Furthermore, even if concerns about forum shopping are set aside, the second of the twin aims would apply to the impleader just as much as the negligence action. The California state court remains the presumptive forum for both actions. Congress granted supplemental jurisdiction to the impleader to overcome the inefficiency of separate litigation and because it supported the purposes of diversity jurisdiction. There is no reason to think that this was a license to federal courts create procedural common law for the impleader to vindicate any conceivable federal interest. If the federal court created a short federal common law limitations period for the impleader as a means of weeding out actions with stale evidence, the Nevada defendant could argue that she was being treated unfairly, for the federal court would have deviated from the rules of the presumptive state forum in a manner that was not justified by the purposes of supplemental jurisdiction or by countervailing federal interests of sufficient strength.

Now let us consider the twin aims for supplemental jurisdiction actions in federal question cases. Here too the justification for supplemental jurisdiction is efficiency. Assume a California plaintiff sues a California officer in federal court in California because his arrest violated his federal civil rights. In the absence of supplemental jurisdiction, he would have to sue the officer for state law battery concerning the arrest in state court, even though the evidence presented in both actions will be largely the same.

Once again, it follows that the twin aims should apply to the battery action. The justification for the twin aims is particular strong if my argument in Part III.A.2 is faulty and the twin aims apply in a reverse-Erie context. Assume Congress wants vertical procedural uniformity in federal question cases, to avoid forum shopping. A California state court entertaining the federal civil rights action must borrow federal procedural law to ensure that the parties are not discouraged or overly encouraged from litigating there. But since the civil rights action and battery action will litigated together, (245) avoiding forum shopping is possible only if there is vertical procedural uniformity concerning the battery action too. Otherwise a party might avoid, or unjustifiably seek out, state court jurisdiction for the federal civil rights action because of the procedure that would be applied to the battery action. Concerning the battery action, however, the duty of ensuring procedural uniformity would fall on federal rather than state courts.

But assume, as I have argued, that forum shopping is not a concern with federal causes of action. The second of the twin aims still applies. Congress gave the battery action supplemental jurisdiction to avoid the inefficiency that would occur if it were litigated separately, not because it conceived of the federal court as a coequal forum for the action. If the federal court applied a short federal common law limitations period to the battery action the plaintiff could legitimately complain that he was being treated unfairly, for the content of this federal procedural common law rule would not be justified by the purposes of supplemental jurisdiction or by countervailing federal interests of sufficient strength. (246)

2. Bankruptcy

Finally, let us consider the thorny question of the twin aims in bankruptcy jurisdiction. In certain areas, federal bankruptcy jurisdiction is exclusive. (247) But bankruptcy jurisdiction over state law actions is generally concurrent, (248) and federal courts can abstain from hearing such actions in order to allow them to proceed in state court. (249) Our question is whether the twin aims apply to state law actions that are pursued in bankruptcy court. It should be noted that in my discussion I will speak of a bankruptcy court. generally, ignoring the difference--crucial in other contexts--between a bankruptcy court and a federal district court sitting in bankruptcy.

Discussion of the twin aims in bankruptcy has been directed almost entirely to the question of whether uniform federal choice-of-law rules should be used by bankruptcy courts or whether, per Klaxon, they should use the choice-of-law rules of the forum state. As we have seen, the applicability of Klaxon in diversity cases was a product of the twin aims. A uniform federal choice-of-law rule would "constantly disturb equal administration of justice in coordinate state and federal courts sitting side by side." (250) Klaxon is unrelated to respect for state lawmaking power under Erie, because a forum state's choice-of-law rules do not extend, and do not purport to extend, to federal courts. (251)

It is worth emphasizing, however, that bankruptcy courts do have an obligation under Erie to respect state lawmaking power. (252) Assume that choice of state law is not an issue, because only one state's law could, as a constitutional matter, be applied to the facts. Unless it has been preempted by federal law, bankruptcy courts are required to respect that state's law, (253) which includes respecting state procedural rules that are bound up with that law. (254) The mere existence of bankruptcy jurisdiction does not mean that preemption has occurred. As the Supreme Court put it in Butner v. United States, "Congress has generally left the determination of property rights in the assets of a bankrupt's estate to state law.... Unless some federal interest requires a different result, there is no reason why such interests should be analyzed differently simply because an interested party is involved in a bankruptcy proceeding." (255) Bankruptcy courts' Erie obligations also apply in cases where more than one state's law may constitutionally be applied to a matter, for unless state law has been preempted they are obligated to apply the law of one of those states.

Our question, however, is not Erie, but the twin aims. Does a bankruptcy court entertaining a state law action have a reason to borrow the rules that would be applied by a forum state court, even when these rules do not or cannot extend to federal court of their own force? If the twin aims apply in bankruptcy, a bankruptcy court will incorporate into federal procedural common law not merely the forum state's choice-of-law rules but many other rules that would be used by a forum state court. It would have a reason to borrow such rules, unless significant countervailing federal interests recommend a uniform federal common law rule.

Federal courts' treatment of the role of the twin aims in bankrupcty has been unsatisfactory. Consider In re Gaston & Snow, (256) in which the Second Circuit held that Klaxon controls in bankruptcy. The reason, the court argued, was that state law cannot be preempted by federal common law absent a compelling federal interest: "The ability of the federal courts to create federal common law and displace state created rules is severely limited.... Before federal courts create federal common law, 'a significant conflict between some federal policy or interest and the use of state law must first be specifically shown.'" (257)

This wrongly treats the twin aims as if they are tied to Erie's concern for the lawmaking power of the states. As we have seen, the twin aims are not about respect for state interests. They recommend incorporating state law standards into federal procedural common law to vindicate federal jurisdictional interests, without regard for whether the forum state supreme court would want its rule to extend to federal courts. Since state law does not apply in federal court, it cannot be displaced, whether or not a uniform federal common law rule is used.

Federal courts that have concluded that the twin aims and Klaxon are irrelevant in bankruptcy have not done much better. In In re Lindsay, (258) the Ninth Circuit held that Klaxon did not control because the state law issue faced by the bankruptcy court arose in an area where bankruptcy jurisdiction was exclusive. "In federal question cases with exclusive jurisdiction in federal court," the court reasoned, "the risk of forum shopping which is avoided by applying state law has no application, because the case can only be litigated in federal court." (259)

First of all, the very decision by the debtor about whether to declare bankruptcy might be motivated in part by the federal procedural common law available in bankruptcy court. (260) But more fundamentally, the second of the twin aims could still apply to bankruptcy courts even if forum shopping is not a concern. We have already come to this conclusion in connection with diversity. (261) If federal courts had exclusive jurisdiction over diversity cases, it would not follow that they could create a short common law time limit for state laws actions to address problems of stale evidence. Those disadvantaged by the time limit could still legitimately complain that federal courts were inequitably administering the laws. By creating diversity jurisdiction, Congress meant to provide a forum in each state free of pro-domiciliary bias--not to empower federal courts to create procedural common law to serve any conceivable federal interest.

The same point appears to be true of bankruptcy. Like diversity jurisdiction, federal bankruptcy jurisdiction was created as an alternative to the presumptive state fora in order to address a deficiency in state court jurisdiction. In the case of bankruptcy jurisdiction, the deficiency is a collective action problem. Each creditor would prefer to be the first to bring an independent state court action against the debtor, in order to get relief before the debtor's assets are exhausted. To allow for an efficient and equitable distribution of these assets-and to protect the debtor herself--it is crucial that all litigation by and against the debtor be controlled by one court. (262) Congress, in creating bankruptcy jurisdiction, intended to solve this problem with state court jurisdiction. It is unlikely that it thereby licensed federal courts to create procedural common law to serve unrelated federal purposes. If federal bankruptcy policies or other sufficiently strong countervailing federal interests do not recommend otherwise, federal courts lack a sufficient reason to deviate from the rules that would be used by the courts of the state where the action would have been brought absent bankruptcy jurisdiction. (263)

But there is a complication in bankruptcy that makes it difficult to satisfy the twin aims' demands. Bankruptcy proceedings will generally be brought in the district of the debtor's residence (for an individual) or state of incorporation, principal place of business, or location of assets (for a business). (264) The bankruptcy court in that district will have jurisdiction over all of the debtor's property, no matter where it is located, and nationwide service of process is available. (265) Thus, it can entertain a state law action even though the action could not have been entertained by a forum state court.

An example is In re Gaston & Snow. (266) The Boston law firm of Gaston & Snow entered involuntary bankruptcy proceedings in the Southern District of New York, where it had a branch office. The bankruptcy estate administrator filed an adversary proceeding against Robert Erkins, a resident of Idaho, to recover $1.7 million for legal services performed for Erkins by lawyers at Gaston & Snow's Boston office. The Second Circuit held that the relevant limitations period should be six years, because that is what would be applied by a New York state court, even though no action against Erkins could have been brought in state court in New York. (267)

As Gaston & Snow reveals, the twin aims cannot always be satisfied by borrowing the rules that would be used by a forum state court. The federal court entertaining the action against Erkins cannot be understood as subsidiary to New York state courts. If it is subsidiary to the courts of a state, the relevant state is the one where the action would have been brought had bankruptcy not been declared. But this state is difficult to determine. Although it is likely to have been Idaho, it might have been Massachusetts, where personal jurisdiction over Erkins would also have been possible. Indeed, theoretically the court of any state, including New York, might have entertained the action, if the summons and complaint had been served upon Erkins within the state's borders, (268) or Erkins had consented to personal jurisdiction, and the court did not dismissed the action on forum non conveniens grounds.

What effect should the broad personal jurisdictional scope of a bankruptcy court have on the applicability of the twin aims? In considering the matter, it is best to begin with cases where there is a high level of certainty about where the action would have proceeded outside bankruptcy.

An example is when the state law action was filed pre-bankruptcy and the bankruptcy court only subsequently got jurisdiction. (269) In such a case, the twin aims clearly recommend borrowing the rules that would be used by the courts of the state where the action was originally filed, as long as those courts would not have dismissed for lack of personal jurisdiction or forum non conveniens. (270) After all, if the bankruptcy court used different procedural law than the filing state's courts, there would be a danger of vertical forum shopping. The debtor might declare bankruptcy or avoid bankruptcy because of the procedural common law used in bankruptcy court. Granted, declaring bankruptcy, unlike invoking diversity jurisdiction, is not generally discretionary. But there are inevitably areas where the debtor has some room for choice. Furthermore, even though the opportunities to forum shop are narrower in bankruptcy than they are in diversity, the costs of forum shopping are greater. If a defendant who is not worried about state court bias removes a diversity case to federal court, all that happens is that federal judicial resources are wasted on a case unrelated to the purposes of diversity. If a debtor declares bankruptcy to take advantage of federal procedure, in contrast, a whole web of financial relations is disturbed. In addition, because the bankruptcy court can abstain from taking jurisdiction of a previously filed state law action, uniformity between the bankruptcy court's procedure and the procedure in the court of the filing state can avoid strategic decisions concerning requests for abstention.

But even when forum shopping is set aside, there is an argument for procedural uniformity with the courts of the filing state. The filing state's courts are the presumptive fora for the state law action. Unless the purposes of bankruptcy or other countervailing interests of sufficient force suggest otherwise, the bankruptcy court lacks a reason to deviate from the rules that those courts would use.

There are other cases, besides those involving pre-bankruptcy filing, in which it is clear where the action would have been brought had bankruptcy not occurred, namely when the parties are all residents of a state and the transaction being litigated occurred there. In such cases, the presumptive fora for the action are that state's courts. Had there been no bankruptcy jurisdiction the action would almost certainly have been brought there. (271) Here the twin aims recommend that the federal procedural common law used by a bankruptcy court borrow from the rules that would be used by the courts of that state. (272) But as one moves beyond cases in which the action was filed pre-bankruptcy or the parties' residence and the transaction at issue are all located in one state, it becomes much more difficult to identify the state where the action would have been brought absent bankruptcy. Do the twin aims still apply in such cases?

First of all, I think it is clear that the second of the twin aims is no longer relevant. The second aim applies in diversity cases because a federal court is conceived of as subsidiary to the courts of the forum state. This conception of a federal court can be meaningfully extended to a bankruptcy court when it is clear in which state the state law action would have been brought but for bankruptcy. But if all we know is that the courts of a number of states would have entertained the action in the absence of bankruptcy, the notion that a bankruptcy court is an extension of a particular state court system evaporates. The bankruptcy court is more analogous to a federal court sitting in diversity in the District of Columbia, particularly before 1970, when there was no "state" court system in the District parallel to "federal" courts. (273) Federal courts in the District were not understood as the extension of any state court system, even though one might have come up with a list of the states where the action would have been brought had there been no jurisdiction in the District. Because federal courts in the District were not understood as the extension of a state court system, they were free to apply federal procedural common law of their own devising to state law actions, without any duty to borrow from a state's law, provided that state rules bound up with the cause of action upon which the plaintiff sued were respected. (274)

One might argue, however, that the first of the twin aims is still relevant and recommends that a bankruptcy court borrow a rule from one of the states that could have entertained the action. (275) Assume, for example, that the courts of every state that could have entertained the action would have used a three-year limitations period. There would be a concern about forum shopping if the bankruptcy court used a longer or shorter period. Furthermore, if half of the state courts would have used a two-year period, and the other half a three-year period, there would be a danger of forum shopping if the bankruptcy court chose a one-year or four-year period.

Although, as an abstract matter, this approach has some attractions, it is administratively difficult for a bankruptcy court to come up with a list of those states where the action could have been brought but for bankruptcy, especially since the list can be expanded significantly by including states that might get personal jurisdiction over the defendant through in-state service or consent. Furthermore, even when the candidate states and their rules are known and a rule is chosen from one of those states, if the candidate states' rules differ from one another, there is no assurance that forum shopping will not still occur. Fearing a suit against him in a state with a three-year limitations period, the debtor might declare bankruptcy because he knows the bankruptcy court will chose a candidate state with a two-year period.

Because the second of the twin aims is irrelevant, the first is weakened, and the administrative costs are often high, I believe that in general a bankruptcy court that is not certain about the particular state where the action would have been brought outside bankruptcy should be released from the twin aims' demands. (276) It is free to come up with federal procedural common law rules for state law actions as it sees fit, although it can, of course, borrow from forum state law or any other state's law for reasons of convenience, just as federal courts do when federal statutes lack limitations periods. (277) That said, if it is administratively convenient and the procedural rule at issue is one--such as a limitations period or a choice-of-law rule--that generates serious forum shopping worries, bankruptcy courts would be advised do their best to choose the rule that would be used by the courts of one of the states where the action would most likely have been brought in the absence of bankruptcy.

It is important to remember that the above discussion concerns only the role of the twin aims in bankruptcy. Other elements of the relatively unguided Erie analysis must still be addressed. For example, even if the twin aims do not apply, a bankruptcy court can be limited in its power to create procedural common law, because a state has bound up a procedural rule with the cause of action. (278) And even if the twin aims do apply, countervailing federal interests can recommend a uniform federal procedural common law rule.

One unrecognized area where countervailing federal interests should sometimes override the twin aims' demands concerns statutes of limitations. When a state court dismisses a state law action on statute of limitations grounds, the dismissal generally does not have claim preclusive effect. The plaintiff is free to sue on that action in another jurisdiction that has a longer period. (279) The matter is different in bankruptcy, where the federal court's dismissal of the action is usually fatal. The action cannot be brought again anywhere. This is a reason for the bankruptcy court to choose the longer of the available periods, unless a shorter period has been bound up into the cause of action upon which the plaintiff sues. (280)

Finally, let us consider the role of Klaxon in bankruptcy. (281) It follows from my argument above that when one is certain where the action would have been brought outside of bankruptcy, for example, because it was filed before bankruptcy was declared, the twin aims recommend that the bankruptcy court use the choice-of-law rules of that state. On the other hand, if there is more than one state where the action could have been brought, it is generally free of the twin aims' demands and so can use federal choice-of-law rules.

But are there countervailing federal interests--in particular, interests fled to the purposes of bankruptcy--that recommend a uniform federal choice-of-law approach even when the court knows where the action would have been brought had bankruptcy not occurred? Dicta from Justice Black's opinion Vanston Bondholders Protective Community v. Green, (282) suggests as much. (283) Vanston concerned whether a creditor had a claim against the debtor for interest on unpaid interest. (284) In the end, the Supreme Court held that bankruptcy policy preempted any state law right to such interest. (285) But in his opinion Black suggested that had the Court been required to answer the horizontal choice-of-law question, the matter should have been decided by federal choice-of-law rules, rather than rules borrowed from the forum state's courts. (286) Because the interests of a number of states were involved, they would have to be balanced by the federal court in a manner that did not look to the choice-of-law rules of the forum state:
 [O]bligations ... often have significant contacts in many states,
 so that the question of which particular state's law should measure
 the obligation seldom lends itself to simple solution. In
 determining which contact is the most significant in a particular
 transaction, courts can seldom find a complete solution in the
 mechanical formulae of the conflicts of law. Determination requires
 the exercise of an informed judgment in the balancing of all the
 interests of the states with the most significant contacts in order
 best to accommodate the equities among the parties to the policies
 of those states. (287)


For this reason, he argued, bankruptcy jurisdiction is not the same as diversity jurisdiction: "In determining what claims are allowable and how a debtor's assets shall be distributed, a bankruptcy court does not apply the law of the state where it sits. Erie R.R.V. Tompkins ... has no such implication." (288)

Black's argument is flawed. Bankruptcy jurisdiction is not the only situation in which federal courts face competing state interests when deciding questions of choice of law. This is true in diversity cases too. And yet in diversity cases Klaxon demands that conflicts between states' interests be decided by the forum state's rule. As we have seen, Klaxon might keep federal courts from balancing state interests properly (289) But using the forum state's rule is nevertheless required, because the Supreme Court concluded in Klaxon that resolving competing state interests equitably is not a purpose for which federal courts were given diversity jurisdiction.

The same considerations should apply in bankruptcy. It is true that the choice-of-law problem faced by a bankruptcy court is not merely that there are competing state interests with respect to a single claim, but that there are competing claims against (and sometimes on behalf of) the debtor. Concerning each claim a number of states may have competing interests. But the fact that a bankruptcy court is reconciling competing state law claims is not, on its own, a reason to abandon the view that it is subsidiary to state courts for the adjudication of those claims. The purposes of bankruptcy would in no way have been frustrated had all state law claims first been litigated in state courts, using their own choice-of-law rules, with the results then presented to the bankruptcy court for reconciliation according to federal bankruptcy law. The only reason not to use Klaxon in bankruptcy is that the state where the state law action would have been brought absent bankruptcy is not known.

Up to this point, I have been speaking of whether Klaxon applies when a bankruptcy court entertains a state law cause of action brought against or on behalf of the debtor. But in order to apply federal bankruptcy law, a bankruptcy court also has to face state law issues, such as questions of the scope of the debtor's estate, (290) that are not tied to a particular state law cause of action. Choice-of-law questions can arise concerning these issues as well. With respect to these issues a federal choice-of-law rule clearly should be used, for the court once again usually has no idea where they would have been litigated had bankruptcy not occurred. (291) (Indeed, in many cases, the state law issue would have never come up outside of bankruptcy. (292) The twin aims are therefore irrelevant.

To sum up, if the bankruptcy court is certain where the action would have been brought but for bankruptcy, the twin aims recommend that the federal procedural common law it uses borrow from the rules of that state. In other cases, the twin aims do not apply, although if it is administratively convenient the bankruptcy court would be advised to choose a rule that would be used by the courts of one of the states where the action would most likely have been brought in the absence of bankruptcy jurisdiction, as a means of discouraging forum shopping. Finally, whether or not the twin aims apply, the bankruptcy court must also consider other factors of the relatively unguided Erie analysis in coming to a conclusion about what procedural rule to use.

CONCLUSION

In this Article, I have offered a comprehensive explanation of the twin aims by grounding them, not in Erie concerns about federalism and state interests, but in the federal policies standing behind the statute that gives the federal court jurisdiction. In the end, I have concluded that the twin aims are implicated, to some extent, in every state law action entertained by a federal court. In contrast, they do not apply to federal courts entertaining federal causes of action. Nor do they apply to state courts entertaining federal causes of action, although the question is subject to vigorous (if veiled) disagreement among the current members of the Supreme Court.

As we have seen, the twin aims can recommend that a federal court use state rules whatever the relevant state supreme court would say about the matter. The reason is the unique nature of the lower federal court system, which was created by Congress and endowed with jurisdiction over state law actions to address perceived deficiencies in the preexisting state court systems. Because federal jurisdiction over state law actions is seen as subordinate to the jurisdiction of state courts, a federal court's discretion when creating procedural common law is limited by the purposes for which it was granted jurisdiction. The twin aims are fundamentally about separation of powers, not federalism.

One important benefit of this conception of the twin aims is that it can ease the pressure faced by federal courts in relatively unguided Erie cases. Such cases are commonly understood as essentially concerning the division of lawmaking power between the federal government and the states. Seen in this light, the fluid and unpredictable manner in which they are decided looks problematic. (293) Under my reading, in contrast, federal courts are often simply making a decision about which federal common law rule will best serve competing federal interests. On the one hand, there is the interest--expressed in the twin aims--recommending uniformity with forum state procedure. But this must be balanced against other federal interests recommending a federal common law rule that is uniform among federal courts. So understood, the prospect of a mistaken decision is less serious. Rather than being an infringement upon state sovereignty, it is simply a badly designed federal common law rule, with its costs felt only by the federal government.

The twin aims and countervailing federal interests are not, however, the only consideration a federal court must face in a relatively unguided case. Federal interests aside, a state may have a legitimate interest in extending its rule to federal courts. Although, it has not been the goal of this Article to determine the role that state interests should play in relatively unguided Erie cases, I would like to end with a few words about where I believe research on this matter should be directed.

One problem is how a federal court can determine whether a state is in fact interested in its rules applying in federal court, in the sense that the Byrd bound-up test seeks to capture. Unless the question is certified, state courts will have no occasion to opine about when the state's rules follow the state's cause of action into sister state and federal courts. (294) They have a reason to speak only of what rules they should use. Whether their rules should be used in other court systems is a problem that must be dealt with by those courts. As a result, questions about whether the Byrd bound-up test is satisfied will be dealt with only by courts that cannot provide authoritative answers. The only way around this problem is the onerous process of certification, which no federal court, to my knowledge, has ever used in a relatively unguided Erie case. (295) Given the difficulty determining the extent of state interests, federal courts would arguably be better off giving up the Byrd test and relying on presumptions that are based on more readily discernible characteristics of the rules at issue. (296)

But even if one assumes perfect knowledge about whether the Byrd test is satisfied, there remain questions about whether it accurately captures a state's power to extend its rules to federal courts. The test appears over-inclusive, for it suggests that a state can displace federal procedural common law at will, simply by binding up a contrary state rule with the state law cause of action upon which the plaintiff sues. For example, a state could displace federal procedural common law concerning service by binding up the state's rules on service with a state law cause of action.

The Byrd test not only appears over-inclusive, it also appears under-inclusive. Can't a state legitimately extend its rules to federal courts without binding them up with a state law cause of action? Assume, for example, that the New York Court of Appeals has held that its attorney-client privilege law should be used by a federal court in Pennsylvania entertaining a Pennsylvania cause of action, because the relevant communications were made by the defendant in New York City to a member of the New York bar. (297) Isn't this too an example in which New York has legitimately extended its rules to federal court?

Finally, there is the problem of deciding how conflicts between federal procedural common law and applicable state law should be resolved. Justice Brennan's language in Byrd suggests that federal procedural common law must always yield to a state rule that is bound up with a state cause of action. But as the federal interests standing behind the federal procedural common law rule become stronger and the state interests become weaker, Brennan's rule of priority starts sounding implausible, particular as a statement of federal courts' constitutional obligations. (298)

Eventually these and other questions must be faced. But before the role of state interests in relatively unguided Erie cases can be ascertained, the very different role of the twin aims in such cases needs to be explained and justified. Such has been the purpose of this Article.

(1) Erie R.R. Co. v. Tompkins, 304 U.S. 64 (1938).

(2) Id. at 69.

(3) Id. at 70.

(4) Swift v. Tyson, 41 U.S. (16 Pet.) 1 (1842).

(5) Id. at 19.

(6) Erie, 304 U.S. at 78-80.

(7) Guar, Trust Co. v. York, 326 U.S. 99 (1945).

(8) Id. at 110.

(9) Hanna v. Plumer, 380 U.S. 460 (1965)

(10) Id. at 471.

(11) Id. at 468.

(12) Id. York was decided before Hanna introduced the twin aims test for relatively unguided cases. But it is clear that the decision in York would have been the same had the twin aims been used. See, e.g., Walker v. Armco Steel Corp., 446 U.S. 740, 752-53 (1980); Graziano v. Pennell, 371 F.2d 761, 764 (2d Cir. 1967).

(13) Swift v. Tyson, 41 U.S. (16 Pet.) 1, 19 (1842); see also Udell v. The Ohio, 24 F. Cas. 497, 497-98 (S.D.N.Y. 1851) (No. 14,321a) ("The local statutes will be enforced in the United States courts in appropriate cases according to their effect in the state where enacted, and consequently the expositions of the state tribunals are to be received as the highest evidence of their design and import."); Stewart Jay, Origins of Federal Common Law: Part Two, 133 U. PA. L. REV. 1231, 1265-66 (1985) ("[F]ederal courts [are] obligated to follow state statutes regarding general law matters and 'settled' constructions given the statutes by the state courts.").

(14) Pagan v. Merchs. Transfer & Warehouse Co., 337 U.S. 530, 533-34 (1949).

(15) Cohen v. Beneficial Indus. Loan Corp., 337 U.S. 541, 555-57 (1949).

(16) Chamberlain v. Giampapa, 210 F.3d 154, 161 (3d Cir. 2000).

(17) Stoner v. Presbyterian Univ. Hosp., 609 F.2d 109, 110-11 (3d Cir. 1979); Edelson v. Soricelli, 610 F.2d 131, 134-35 (3d Cir. 1979).

(18) Woods v. Holy Cross Hosp., 591 F.2d 1164, 1168-70 (5th Cir. 1979).

(19) Scottsdale Ins. Co. v. Tolliver, 636 F.3d 1273, 1280 (10th Cir. 2011).

(20) Burke v. Smith, 252 F.3d 1260, 1265-66 (11th Cir. 2001).

(21) Mangold v. Cal. Pub. Utils. Comm'n, 67 F.3d 1470, 1478-79 (9th Cir. 1995).

(22) Vishipco Line v. Chase Manhattan Bank, N.A., 660 F.2d 854, 865-67 (2d Cir. 1981).

(23) Commercial Union Ins. Co. v. Walbrook Ins. Co., 41 F.3d 764, 773-74 (1st Cir. 1994).

(24) Woods v. Interstate Realty Co., 337 U.S. 535 (1949).

(25) Id. at 536 n.1 (internal quotation marks omitted).

(26) Interstate Realty Co. v. Woods, 168 F.2d 701,704-05 (5th Cir. 1948), rev'd, 337 U.S. 535 (1949).

(27) Woods, 337 U.S. at 539 (Jackson, J., dissenting); see also Peter Westen & Jeffrey S. Lehman, Is There Life for Erie After the Death of Diversity?, 78 MICH. L. REV. 311, 357-58 (1980) (explaining Justice Jackson's concerns with reference to the decision in Woods).

(28) See infra note 105.

(29) See infra note 50.

(30) Westen and Lehman also argue that federal jurisdictional purposes recommend incorporating state law into federal common law. See Westen & Lehman, supra note 27, at 373-74. But their argument reaches well beyond the twin aims to any situation in which federal courts use state law. Id. at 356-59. That position has been greeted with considerable hostility. See, e.g., Martin H. Redish, Continuing the Erie Debate: A Response to Westen and Lehman, 78 MICH. L. REV. 959, 961 (1980) (criticizing Westen's and Lehman's position on incorporation of state law into federal law). In contrast, this Article provides a novel and detailed account of the federal purposes standing behind the twin aims in particular. It also considers the twin aims' role in a variety of jurisdictional contexts not addressed by Westen and Lehman, including bankruptcy and reverse-Erie.

For another suggestion that forum state law is incorporated into federal procedural common law in diversity cases, see Kermit Roosevelt III, Choice of Law in Federal Courts: From Erie and Klaxon to CAFA and Shady Grove, 106 NW. U. L. REV. 1, 15 (2012). Roosevelt too does not address the particular federal purposes standing behind the twin aims, nor their role outside diversity.

(31) See, e.g., UAW v. Hoosier Cardinal Corp., 383 U.S. 696, 703-04 (1966). They will generally do so unless the state limitations period will "frustrate or significantly interfere with federal policies." Reed v. United Transp. Union, 488 U.S. 319, 327 (1989).

(32) Moviecolor Ltd. v. Eastman Kodak Co., 288 F.2d 80, 83 (2d Cir. 1961) (Friendly, J.) (" [S] election of a period of years [is] not ... the kind of thing judges do....").

(33) E.g., UAW, 383 U.S. at 706.

(34) See, e.g., DelCostello v. Int'l Bhd. of Teamsters, 462 U.S. 151, 158-62 (1983); Hemmings v. Barian, 822 F.2d 688, 690 (7th Cir. 1987).

(35) For a discussion of the various types of federal procedural common law, see Stephen B. Burbank, Procedure, Politics, and Power: The Role of Congress, 79 NOTRE DAME L. REV. 1677, 1681 (2004). By federal procedural common law, I mean here law created by federal courts decisionally (rather than through prospective court rules). I also mean law that can always be overridden by federal statutes. For examples of the substantial literature on the possibility of federal courts' having powers over procedure in defiance of Congress, see Amy Coney Barrett, Procedural Common Law, 94 VA. L. REV. 813 (2008); Linda S. Mullenix, Unconstitutional Rulemaking: The Civil Justice Reform Act and Separation of Powers, 77 MINN. L. REV. 1283 (1993); Robert J. Pushaw, Jr., The Inherent Powers of Federal Courts and the Structural Constitution, 86 IOWA L. REV. 735 (2001).

(36) Semtek Int'l Inc. v. Lockheed Martin Corp., 531 U.S. 497 (2001).

(37) Id. at 497.

(38) Id. at 508.

(39) Id.

(40) Redish, supra note 30, at 961.

(41) See infra Part I.B-C.

(42) See infra Part I.

(43) See infra Part II.

(44) See infra Part II.A.

(45) See infra Parts III.A.1, B.1.

(46) Haywood v. Drown, 556 U.S. 729 (2009). See infra Part III.A.2.

(47) See infra Part III.B.2.

(48) Hanna v. Plumer, 380 U.S. 460, 473-74 (1965). Of course, the twin aims also do not apply if the matter is governed by the United States Constitution.

(49) Id. at 472; see also Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 32 (1988) (reasserting this congressional authority).

(50) See Paul D. Carrington, "Substance" and "Procedure" in the Rules Enabling Act, 1989 DUKE L.J. 281, 294; Thomas D. Rowe, Jr., Not Bad for Government Work: Does Anyone Else Think the Supreme Court Is Doing a Halfway Decent Job in Its Erie-Hanna Jurisprudence?, 73 NOTRE DAME L. REV. 963, 973 (1998); Patrick Woolley, Erie and Choice of Law After the Class Action Fairness Act, 80 TUL. L. REV. 1723, 1768 (2006).

(51) 28 U.S.C. [section] 2072(a) (2006).

(52) This does not mean that ambiguous Federal Rules should not be read to avoid frustrating the twin aims. Shady Grove Orthopedic Assocs., P.A. v. Allstate Ins. Co., 130 S. Ct. 1431, 1441 n.7 (2010).

(53) 28 U.S.C. [section] 2072(b).

(54) Compare Shady Grove, 130 S. Ct. at 1442-43 (majority opinion), with id. at 1448-50 (Stevens, J., concurring) (illustrating different conceptions of the limitation).

(55) Id. at 1447-48 (majority opinion); id. at 1448 (Stevens, J, concurring).

(56) Hanna v. Plumer, 380 U.S. 460, 468 (1965).

(57) The Court initially offered an outcome-determinative test in Guar. Trust Co. v. York, 326 U.S. 99, 109 (1945).

(58) Hanna, 380 U.S. at 468-69 (" [T] he difference between the two rules would be of scant, if any, relevance to the choice of a forum.").

(59) See supra note 12 and accompanying text.

(60) See supra notes 14-23 and accompanying text. Two recent examples are Liggon-Redding v. Estate of Sugarman, 659 F.3d 258 (3d Cir. 2011) (borrowing state rule requiring certificate of merit before filing malpractice action) and Scottsdale Ins. Co. v. Tolliver, 636 F.3d 1273 (10th Cir. 2011) (borrowing state rule allowing defendant to collect attorney's fees if defendant prevails after offer of judgment is refused).

(61) Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 496 (1941).

(62) See Richard D. Freer, Some Thoughts on the State of Erie After Gasperini, 76 TEX. L. REV. 1637, 1652 (1998). But see Walker v. Armco Steel Corp., 446 U.S. 740, 753 (1980) (claiming that inequity but not forum shopping test was violated because actual parties in case did not engage in forum shopping). For the persuasive argument that the forum shopping test should look beyond the actual parties' decisions, see Patrick Woolley, The Sources of Federal Preclusion Law After Semtek, 72 U. CIN. L. REV. 527, 548-49 (2003).

(63) ERWIN CHEMERINSKY, FEDERAL JURISDICTION [section] 5.3.5, at 316 (4th ed. 2003). This assumes, of course, that state interests do not recommend that state law be used in federal court. See infra Part I.B.3.

(64) Byrd v. Blue Ridge Rural Elec. Coop., 356 U.S. 525, 537 (1958).

(65) Id. at 538.

(66) Gasperini v. Ctr. for Humanities, Inc., 518 U.S. 415, 431-32 (1996).

(67) Semtek Int'l Inc. v. Lockheed Martin Corp., 531 U.S. 497, 508-09 (2001).

(68) E.g., Esfeld v. Costa Crociere, 289 F.3d 1300, 1307 (11th Cir. 2002); Moling v. O'Reilly Auto., Inc., 763 F. Supp. 2d 956, 975-76 (W.D. Tenn. 2011).

(69) Byrd, 356 U.S. at 535.

(70) Id. at 536.

(71) Swift v. Tyson, 41 U.S. (16 Pet.) 1 (1842).

(72) Erie R.R. Co. v. Tompkins, 304 U.S. 64, 74-75 (1938).

(73) Id. at 78.

(74) For my reading of the constitutional sources of the decision in Erie, see Michael Steven Green, Horizontal Erie and the Presumption of Forum Law, 109 MICH. L. REV. 1237, 1241-51 (2011) [hereinafter Green, Presumption] and Michael Steven Green, Erie's Suppressed Premise, 95 MINN. L. REV. 1111, 1115-61 (2011) [hereinafter Green, Suppressed].

(75) See Erie, 304 U.S. at 78 ("[N]o clause in the Constitution purports to confer ... a power [to create federal general common law] upon the federal courts."); see also Stephen B. Burbank, Interjurisdictional Preclusion, Full Faith and Credit and Federal Common Law: A General Approach, 71 CORNELL L. REV. 733, 756 n.102 (1986) (acknowledging the absence of such authority); John Hart Ely, The Irrepressible Myth of Erie, 87 HARV. L. REV. 693, 702-03 (1974) (same).

(76) For many, Erie and the Rules of Decision Act articulate separation-of-powers and not just federalism restrictions on federal courts' common law powers. See generally Curtis A. Bradley & Jack L. Goldsmith, Customary International Law as Federal Common Law: A Critique of the Modern Position, 110 HARV. L. REV. 815 (1997) (describing the limiting effect Erie and the Rules of Decision Act put on federal courts' common lawmaking power concerning customary international law); Curtis A. Bradley et al., Sosa, Customary International Law, and the Continuing Relevance 0fErie, 120 HARV. L. REV. 869, 878-81 (2007) (describing Erie as a separation-of-powers limit on federal courts' ability to create federal common law); Bradford R. Clark, Separation of Powers as a Safeguard of Federalism, 79 TEX. L. REV. 1321 (2001) (arguing that Erie's limitation on federal courts' powers to create common law illustrates how separation of powers protects federalism); Martin H. Redish, Federal Common Law, Political Legitimacy, and the Interpretive Process: An "Institutionalist" Perspective, 83 NW. U. L. REV. 761 (1989) (arguing that the Rules of Decision Act prohibits federal common law in the absence of a statutory directive). As a result, a new Erie doctrine, concerning separation of powers rather than federalism, has emerged. For a criticism of this new Erie doctrine, see Craig Green, Repressing Erie's Myth, 96 CAL. L. REV. 595, 615-55 (2008); Louise Weinberg, The Curious Notion that the Rules of Decision Act Blocks Supreme Federal Common Law, 83 NW. U. L. REV. 860, 861-69 (1989). For the purposes of this Article, the term "Erie" will be reserved for the federalism constraints on federal courts that were the focus of Brandeis's opinion.

(77) E.g., Boyle v. United Techs. Corp., 487 U.S. 500, 512 (1988) (recognizing limited federal common law immunity from state tort liability for injuries for defective products created by government contractors); Hinderlider v. La Plata River & Cherry Creek Ditch Co., 304 U.S. 92, 110 (1938) (describing apportionment of water from the La Plata River between Colorado and New Mexico as "a question of 'federal common law' upon which neither the statutes nor the decisions of either State can be conclusive"). See generally Jay Tidmarsh & Brian J. Murray, A Theory of Federal Common Law, 100 NW. U. L. REV. 585, 620-27 (2006) (describing the different enclaves in which federal common law may be made).

(78) Brandeis was, after all, the author of Hinderlider, 304 U.S. 92. See supra note 77.

(79) See United States v. Little Lake Misere Land Co., 412 U.S. 580, 591 (1973); Ely, supra note 75, at 713; Martha A. Field, The Legitimacy of Federal Common Law, 12 PACE L. REV. 303, 306 (1992).

(80) See Semtek Int'l Inc. v. Lockheed Martin Corp., 531 U.S. 497, 509 (2001).

(81) 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, 316-20 (2008).

(82) Green, Suppressed, supra note 74, at 1121-27. Indeed Georgia still holds such a view. Id. at 1126-27.

(83) Green, Presumption, supra note 74, at 1279-80.

(84) Henry M. Hart, Jr., The Relations Between State and Federal Law, 54 COLUM. L. REV. 489, 513 (1954) (describing forum shopping as "a relatively minor consideration which Brandeis mentioned only in passing").

(85) Cities Serv. Oil Co. v. Dunlap, 308 U.S. 208 (1939) (cited in Byrd v. Blue Ridge Rural Elec. Coop., 356 U.S. 525, 535 (1958)). Although the Byrd test, as I understand it, is tied to the likely decisions of the relevant state's supreme court, figuring out what this decision would be is exceptionally difficult. See Michael Steven Green, Law's Dark Matter, 54 WM. & MARY L. REV. 845, 869-84 (2013). For example, Dunlap cited no case in which the Texas Supreme Court, or any other Texas state court, said that Texas's rule on the burden of proof should follow Texas actions into other court systems. The reason is that (absent certification) no Texas state court would have an occasion to opine about whether a federal or sister state court should use the Texas rule. Texas state courts are concerned solely with the rules that they should use. The question of whether other courts should use the rule would be faced only by those other courts when entertaining an action concerning Texas property.

Indeed, with one exception, the courts of a sovereign that creates a rule concerning the burden of proof will never have the occasion, absent certification, to say whether the rule satisfies the Byrd test. Significantly, the one exception was in fact cited in Dunlap. This was Cent. Vt. Ry. Co. v. White, 238 U.S. 507, 512 (1915) in which the Supreme Court, taking a FELA case on appeal from the Vermont state court system, held that the federal rule (under which the burden of proving contributory negligence was on the defendant) was intended by Congress to follow FELA actions into state courts. See Garrett v. Moore-McCormack Co., 317 U.S. 239, 243 (1942) (holding that the federal rule concerning burden of proof should be used in a Jones Act action in state court). Reverse-Erie cases--that is, cases in which a state court entertains an action under federal law--are significant because the United States Supreme Court can take the case on appeal from the state court system. In contrast, a state supreme court cannot take an issue about its law on appeal from federal courts or from the courts of sister states. It can get the issue only through certification. Thus only in reverse-Erie cases do we have regular statements by a sovereigu's courts about which of its rules are bound up with a cause of action. But see infra Part III.A.2 (arguing that in many reverse-Erie cases in which the Supreme Court holds that a federal rule should be used by a state court, the real reason is not that the federal rule is bound up with the federal cause of action, but the twin aims).

(86) See RESTATEMENT (FIRST) OF CONFLICT OF LAWS [section] 595 (1934); RESTATEMENT (SECOND) OF CONFLICT OF LAWS [section] 133, cmt. b (1969).

(87) Dunlap, 308 U.S. at 212.

(88) See infra Conclusion.

(89) E.g., Gasperini v. Ctr. for Humanities, Inc., 518 U.S. 415, 427 (1996); Windy City Metal Fabricators & Supply, Inc. v. CIT Tech. Fin. Servs., Inc., 536 F.3d 663, 670 (7th Cir. 2008).

(90) E.g.,Jinks v. Richland Cnty., 538 U.S. 456, 465 (2003); Hollander v. Brown, 457 F.3d 688, 692 (7th Cir. 2006); Phelps v. McClellan, 30 F.3d 658, 661 (6th Cir. 1994).

(91) Hemmings v. Barian, 822 F.2d 688 (7th Cir. 1987).

(92) Id. at 689-90 (citing Guar. Trust Co. v. York, 326 U.S. 99 (1945)).

(93) Redish, supra note 30, at 961.

(94) Harbor Ins. Co. v. Cont'l Bank Corp., 922 F.2d 357 (7th Cir. 1990).

(95) Id. at 359.

(96) Id. at 362. Ultimately, Posner concluded that the factual record did not allow a determination of whether the doctrine applied. Id. at 365.

(97) Posner appeared to conclude that federal enacted law--especially FED. R. Cry. P. 8(e) (2)--was not dispositive on the matter. Id. at 364. For a concurring view, see Robert H. Sitkoff, "Mend the Hold" and Erie: Why an Obscure Contracts Doctrine Should Control in Federal Diversity Cases, 65 U. CHI. L. REV. 1059, 1080-84 (1998).

(98) Harbor Ins. Co., 922 F.2d at 363.

(99) Id.

(100) United Farm Family Mut. Ins. Co. v. Frye, 887 N.E.2d 783 (Ill. App. Ct. 2008).

(101) It is conceivable, however, that the doctrine is meant to be used by Illinois courts (including when entertaining contract actions under the law of other jurisdictions) and to follow Illinois contract actions into other court system. None of the cases Posner cited actually addresses the matter directly of course, for no state court has a reason to say whether one of its rules follows its causes of action into other court systems. As I put it in Green, supra note 85, the question is about "dark matter."

(102) Sitkoff, supra note 97, at 1084-89.

(103) For another criticism of Posner as examining a case solely in the light of Erie, without considering the twin aims, see Green, Presumption, supra note 74, at 1278-79.

(104) Guar. Trust Co. v. York, 326 U.S. 99, 108-09 (1945).

(105) The Supreme Court has not determined whether a forum state may prefer its longer procedural statute of limitations over an applicable sister state limitations period. The closest it has come is Sun Oil Co. v. Wortman, 486 U.S. 717 (1988). But Sun Oil told us only that the court had the power to apply its statute of limitations in the absence of competing sister state law, for in his opinion in Sun Oil Justice Scalia noted that the sister states at issue did not want their statutes of limitations to follow their causes of action into other court systems. Id. at 729 n.3. The Supreme Court has held that a state court may prefer its shorter procedural statute of limitations over a sister state's applicable limitations period. Wells v. Simonds Abrasive Co., 345 U.S. 514, 517-18 (1953). In this scenario, however, the dismissal usually allows the plaintiff to sue again in another forum. See RESTATEMENT (SECOND) OF JUDGMENTS [section] 19(f) (1982); RESTATEMENT (SECOND) OF CONFLICT OF LAWS [section] 110 cmt. b (1969). It is arguable, therefore, that the application of the forum's procedural law does not really conflict with sister state law. See RUSSELL J. WEINTRAUB, COMMENTARY ON THE CONFLICT OF LAWS [section] 9.2B (6th ed. 2010).

(106) Level 3 Commc'ns, Inc. v. Fed. Ins. Co., 168 F.3d 956 (7th Cir. 1998).

(107) Id. at 956-57.

(108) In United Farm Family Mut. Ins. Co. v. Frye, 887 N.E.2d 783 (Ill. App. Ct. 2008), an Illinois state court applied Illinois's doctrine of mend the hold to a contract action under Indiana law. Id. at 787-88. The court did not consider Indiana law. Id. This suggests that it thought Indiana law should always yield to Illinois law on the matter. It is conceivable, however, that the court would have held that Illinois's doctrine should yield to a contrary Indiana law if it had been shown that Indiana officials wanted their law to follow Indiana contract actions into other court systems.

(109) Again, Posner's mistake caused no harm in the particular case, for Illinois and Nebraska law appeared to be the same on the matter.

(110) In fact, it is unlikely that Allstate Ins. Co. v. Hague, 449 U.S. 302 (1981) would have been satisfied by the application of New York law to Erie's duty of care. Under Allstate, to satisfy the Due Process Clause of the Fourteenth Amendment and the Full Faith and Credit Clause, a forum state needs "significant contact or significant aggregation of contacts, creating state interests," such that "choice of its law is neither arbitrary nor fundamentally unfair." Id. at 308. I set aside such worries here.

(111) For a further discussion, see Green, Presumption, supra note 74, at 1274-80.

(112) For a very different view of Klaxon, see Roosevelt, supra note 30, at 16-23. As Roosevelt sees it, state courts sometimes understand their choice-of-law rules as territorial limits on the scope of their causes of action that follow those causes of action into federal and sister state court. A federal and sister state court cannot apply Pennsylvania's law to certain facts unless the Pennsylvania Supreme Court, when deciding the matter for lower Pennsylvania state courts, would hold that Pennsylvania law applies. Klaxon, in turn, expresses federal courts' constitutional obligation under Erie to respect state interests by following a state's choice-of-law rules when determining the territorial scope of the state's law. For a similar view of state choice-of-law rules, see Lea Brilmayer, Methods and Objectives in the Conflict of Laws: A Challenge, 35 MERCER L. REV. 555, 563 (1984); Larry Kramer, Return of the Renvoi, 66 N.Y.U.L. REV. 979, (986-87) (1991); Larry Kramer, Rethinking Choice of Law, 90 COLUM. L. REV. 277, 303 (1990); Kermit Roosevelt III, Resolving Renvoi: The Bewitchment of Our Intelligence by Means of Language, 80 NOTRE DAME L. REV. 1821, 1840-41, 1863-64, 1889-90 (2005). For a similar but less plausible conception of Klaxon than Roosevelt's, see infra note 275.

I criticize the Brilmayer-Kramer-Roosevelt approach at Green, supra note 85, at 871-84; Michael Steven Green, Choice of Law as General Common Law: A Reply to Professor Brilmayer, in THE ROLE OF ETHICS IN INTERNATIONAL LAW 125 (Donald Earl Childress III ed., 2011). It is unlikely that a state supreme court would hold that its choice-of-law rules bind federal and sister state courts when determining whether the state's law can be used. But even if Roosevelt is right that state courts want their choice-of-law rules to be used by federal and sister state courts, I see no evidence that Klaxon is about respecting such state interests. Klaxon concerns the twin aims, not Erie.

(113) Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 496 (1941).

(114) I am skeptical that a state would consider its choice-of-law rules to extend beyond its court system. For a discussion, see Green, supra note 85, at 171-84.

(115) It is constitutionally permissible for a state court to prefer forum interests over the stronger interests of sister states. See Nevada v. Hall, 440 U.S. 410, 422 (1979); Pac. Emp'rs Ins. Co. v. Indus. Accident Comm'n, 306 U.S. 493, 500 (1939).

(116) Henry M. Hart, Jr., The Relations Between State and Federal Law, 54 COLUM. L. REV. 489, 512-15 (1954) (criticizing "triviality" of avoiding forum shopping and recommending Klaxon be abandoned).

(117) A possible example is Gasperini v. Ctr. for Humanities, Inc., 518 U.S. 415 (1996), in which Justices Scalia and Ginsburg came to different conclusions about whether, under a relatively unguided analysis, a federal district court in New York entertaining an action under New York law should use the New York or federal standard for reviewing a jury's award of damages as excessive. Ginsburg, who understood New York's standard as substantive, id. at 429, concluded that it should be used by the district court, id. at 438-39. In contrast, Scalia, who argued that New York's standard was procedural, concluded that even under a relatively unguided analysis the district court should use the federal standard. Id. at 463-68 (Scalia, J., dissenting). In coming to different conclusions about how the relatively unguided analysis should turn out, their differing views about New York's interests clearly played a role.

(118) See supra Part I.C.

(119) Allan R. Stein, Erie and Court Access, 100 YALE L.J. 1935, 1955 (1991); see also id. at 1941.

(120) Freer, supra note 62, at 1650.

(121) Ely, supra note 75, at 722-23.

(122) See Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 38-41 (1988) (Scalia, J., dissenting); In re Air Crash Disaster Near Chi., 803 F.2d 304, 314 (7th Cir. 1986); Olympic Sports Prods., Inc. v. Universal Athletic Sales Co., 760 F.2d 910, 913 (9th Cir. 1985); Feinstein v. Mass. Gen. Hosp., 643 F.2d 880, 889 (1st Cir. 1981); Stoner v. Presbyterian Univ. Hosp., 609 F.2d 109, 111 (3d Cir. 1979).

(123) 28 u.s.c. [section] 1652 (2006). This is the Act in its current form, which is not different from the original form in any respect relevant here.

(124) WILFRED J. RITZ, REWRITING THE HISTORY OF THE JUDICIARY AcT Or 1789, at 148 (Wythe Holt & L.H. LaRue eds., 1990); see also Patrick J. Borchers, The Origins of Diversity Jurisdiction, the Rise of Legal Positivism, and a Brave New World for Erie and Klaxon, 72 TEX. L. REV. 79, 107-10 (1993); Suzanna Sherry, Overruling Erie : Nationwide Class Actions and National Common Law, 156 U. PA. L. REV. 2135, 2137-38 (2008). Some argue that the Rules of Decision Act was compatible with Swift, because the general common law was not thought to be the sort of "law" to which the Act applied. E.g., JULIUS GOEBEL, JR., HISTORY OF THE SUPREME COURT OF THE UNITED STATES: ANTECEDENTS AND BEGINNINGS TO 1801, at 502-03 (1971); William A. Fletcher, The General Common Law and Section 34 of the Judiciary Act of 1789: The Example of Marine Insurance, 97 HARV. L. REV. 1513 (1984). These writers assume, however, that the Act did limit federal courts' common lawmaking power, by obligating them to respect state common law that was local, as well as state statutes. E.g., id. at 1514. Ritz, as I understand him, argues that the Act did not even put a duty on federal courts to prefer those state laws over federal common law. To be sure, there was such an obligation--one that followed from the constitutional structure of the American legal system--but the Act had nothing to say about the matter.

(125) So understood, the Act would be redundant, since those obligations would already exist due to constitutional limits on federal lawmaking power. See Hawkins v. Barney's Lessee, 30 U.S. (5 Pet.) 457, 464 (1831) ("[Section 34 of the Judiciary Act is] no more than a declaration of what the law would have been without it...."); Agency Holding Corp. v. Malley-Duff & Assocs., 483 U.S. 143, 162 (1987) (Scalia, J., concur ring); Mason v. United States, 260 U.S. 545, 559 (1923); Alfred Hill, The Erie Doctrine in Bankruptcy, 66 HARV. L. REV. 1013, 1024-31 (1953).

(126) See Louise Weinberg, Federal Common Law, 83 Nw. U. L. REV. 805, 816 (1989) ("The Rules of Decision Act says that state law shall furnish the rules of decision in federal civil actions, in cases in which they apply. But what about cases in which they do not apply?" (footnote omitted)).

(127) Stephen B. Burbank, Afterwords: A Response to Professor Hazard and a Comment on Marrese, 70 CORNELL L. REV. 659, 661 & n.18 (1985); Alfred Hill, State Procedural Law in Federal Nondiversity Litigation, 69 HARV. L. REV. 66, 76-90 (1955).

(128) See Westen & Lehman, supra note 27, at 365-73; Peter Westen, After "Life for Erie"--A Reply, 78 MICH. L. REV. 971, 982-89 (1980).

(129) 28 U.S.C. [section] 1652 (2006) (emphasis added).

(130) Westen, supra note 128, at 989.

(131) And in situations in which more than one applied, common law choice-of-law rules would be used. Fletcher, supra note 124, at 1514-15.

(132) Rowe, supra note 50, at 985 n.85 ("The great generality of the Act's 'in cases where they apply' phrasing ... gives little if any guidance as to when they should apply, leaving just how to make the 'relatively unguided Erie choice' up to judicial interpretation...." (citation omitted)).

(133) Alienage jurisdiction concerns suits between citizens of a state and citizens or subjects of a foreign state. U.S. CONST. art. III, [section] 2; 28 U.S.C. [section] 1332(a)(2). My reading in this Article applies equally to both.

(134) See Bank of U.S. v. Deveaux, 9 U.S. (5 Cranch) 61, 87 (1809); Borchers, supra note 124, at 79-80; Susan Randall, Sovereign Immunity and the Uses of History, 81 NEB. L. REV. 1, 66-70 (2002).

(135) The power is derived from Article III, Section 2 of the United States Constitution, which extends the judicial power to controversies "between Citizens of different States" and "Citizens [of a State] and foreign ... Citizens or Subjects."

(136) 28 U.S.C. [section] 1335(a)(1).

(137) Thomas D. Rowe, Jr., Abolishing Diversity Jurisdiction: Positive Side Effects and Potential for Further Reforms, 92 HARV. L. REV. 963, 1007-08 (1979).

(138) 28 U.S.C. [section] 2361.

(139) 28 U.S.C. [section][section] 1332(d)(2), 1453(b).

(140) Stephen B. Burbank, Aggregation on the Couch: The Strategic Uses of Ambiguity and Hypocrisy, 106 COLUM. L. REV. 1924, 1950 (2006).

(141) Strawbridge v. Curtiss, 7 U.S. (3 Cranch) 267, 267-68 (1806).

(142) See Hertz Corp. v. Friend, 130 S. Ct. 1181, 1192 (2010); Burford v. Sun Oil Co., 319 U.S. 315, 336 (1943) (Frankfurter, J., dissenting); Borchers, supra note 124, 79-80; John P. Frank, Historical Bases of the Federal Judicial System, 13 LAW & CONTEMP. PROBS. 3, 22-28 (1948). This is true even though the diversity statute is poorly drafted to effectuate that purpose: it allows a New Yorker to sue a Pennsylvanian in federal court in New York, even though any pro-domiciliary bias in New York state court would be to her benefit, and allows a New Yorker to sue a Pennsylvanian in federal court in Delaware, even though a Delaware state court biased against non-domiciliaries would find no one to favor.

(143) As we shall later see, this assumption is not justified in connection with bankruptcy. See infra Part III.B.2.

(144) The problem of horizontal forum shopping, which would occur in the absence of diversity, is not a concern to which the twin aims are addressed. See Klaxon Co. v. Stentor Elec. Mfg. Co., 313 U.S. 487, 496 (1941) ("Whatever lack of uniformity [the Klaxon rule] may produce between federal courts in different states is attributable to our federal system, which leaves to a state, within the limits permitted by the Constitution, the right to pursue local policies diverging from those of its neighbors."); Michael Steven Green, Erie's International Effect, 107 Nw. U. L. REV. COLLOQUY 165, 171-72 (2012).

(145) Under 28 U.S.C. [section] 1367(d), if a state law action that had supplemental jurisdiction in federal court is dismissed and is subsequently brought in state court, that court must treat the relevant statute of limitations as tolled during the time the action was pending in federal court, and for 30 days after dismissal, unless state law provides for a longer tolling period. Section 1367(d) has been upheld as "necessary and proper for carrying into execution Congress's power '[t]o constitute Tribunals inferior to the supreme Court,' and to assure that those tribunals may fairly and efficiently exercise '[t]he judicial Power of the United States.'" Jinks v. Richland Cnty., 538 U.S. 456, 462 (2003) (citation omitted). For a discussion of federal regulation of state court procedures in a broader context, see Anthony J. Bellia, Jr., Federal Regulation of State Court Procedures, 110 YALE L.J. 947 (2001).

(146) Louise Weinberg has come closest. Louise Weinberg, The Power of Congress over Courts in Nonfederal Cases, 1995 BYU L. REV. 731, 766 (1995) (noting federal grants of concurrent jurisdiction to state courts for state law actions might generate "adjudicatory policies" that become supreme in state courts).

(147) Tellingly, Weinberg is inclined to understand state courts' possession of concurrent jurisdiction for diversity actions as a positive federal grant, in which state courts sit "as courts of the nation." Id. at 760.

(148) THE FEDERALIST No. 82 (Alexander Hamilton).

(149) One exception is that--as participants in the new federal order--they may not create procedural rules that discriminate against sister state or federal causes of action. See Hart, supra note 116, at 507. Hughes v. Fetter, 341 U.S. 609 (1951), is commonly understood as holding that a state court may not discriminate against sister state law. See Lea Brilmayer & Stefan Underhill, Congressional Obligation to Provide a Forum for Constitutional Claims: Discriminatory Jurisdictional Rules and the Conflict of Laws, 69 VA. L. REV. 819, 825-26 (1983); Larry Kramer, Same-Sex Marriage, Conflict of Laws, and the Unconstitutional Public Policy Exception, 106 YALE L.J. 1965, 1980-86 (1997). Testa v. Katt, 330 U.S. 386 (1947), is commonly understood as holding that a state court may not discriminate against federal law. See H. Jefferson Powell, The Oldest Question of Constitutional Law, 79 VA. L. REV. 633, 643 n.50 (1993); Laurence H. Tribe, Intergovernmental Immunities in Litigation, Taxation, and Regulation: Separation of Powers Issues in Controversies About Federalism, 89 HARV. L. REV. 682, 692 n.62 (1976).

(150) By using the term "legitimate," I intend to include two potential types of restriction on state courts' power over procedure. Restrictions of the first type arise when the plaintiff sues under sister state law. In such cases, the forum state's power over procedure might be limited when it conflicts with applicable sister state law. See supra note 105. Restrictions of the second type include limits that state legislation might have upon the procedural common law powers of state courts (limitations that would surely not be as restrictive as the twin aims).

(151) The same point is true of the New York legislature.

(152) See supra note 105.

(153) See the examples cited in Donna Mae Endreson, Note, Wisconsin's Borrowing Statute: Did We Shortchange Ourselves?, 70 MARQ. L. REV. 120, 122-27 (1986). Borrowing statutes vary greatly in their details but they generally borrow the statute of limitations of the state where the cause of action arose, which is (or should be) the state under whose law the plaintiff sues.

(154) Indeed, if the sister state did think its limitations period were applicable, the borrowing statute would be unnecessary.

(155) See Trzecki v. Gruenewald, 532 S.W.2d 209 (Mo. 1979) (en banc).

(156) Another justification for a deviation from forum state rules is that it is insignificant, as evidenced by the fact that it will not generate significant forum shopping if parties are given a choice between federal and state court.

(157) Henry J. Friendly, The Historic Basis of Diversity Jurisdiction, 41 HARV. L. REV. 483, 498-99 (1928).

(158) See supra text accompanying notes 119-120.

(159) Woods v. Interstate Realty Co., 337 U.S. 535, 538-40 1949); Guar. Trust Co. v. York, 326 U.S. 99, 112-19 (1945).

(160) Shady Grove Orthopedic Assocs., P.A. v. Allstate Ins. Co., 130 S. Ct. 1431, 1447-48 (2010) (citing Bradford R. Clark, Erie's Constitutional Source, 95 CAL. L. REV. 1289, 1302, 1311 (2007)).

(161) 326 U.S. 99 (1945).

(162) See supra Part I.C.

(163) The only other option would be to refuse to take jurisdiction of the action, which federal courts are forbidden to do under such circumstances. See Meredith v. Winter Haven, 320 U.S. 228, 234 (1943).

(164) Byrd v. Blue Ridge Rural Elec. Coop., 356 U.S. 525, 538 (1958).

(165) See, e.g., In re Air Crash Disaster Near New Orleans, La. on July 9, 1982, 821 F.2d 1147, 1157-58 (5th Cir. 1987) [hereinafter In re Air Crash New Orleans], vacated on other grounds, Pan Am. World Airways, Inc. v. Lopez, 490 U.S. 1032 (1989).

(166) Esfeld v. Costa Crociere, S.P.A., 289 F.3d 1300, 1309 (11th Cir. 2002).

(167) In re Air Crash New Orleans, 821 F.2d at 1159.

(168) Esfeld, 289 F.3d at 1311.

(169) Id. at 1312.

(170) See 14D CHARLES A. WRIGHT ET AL., FEDERAL PRACTICE AND PROCEDURE [section] 3828.5 (2009).

(171) Indeed, the expansive manner in which countervailing federal interests are identified suggests that federal courts might have the power to create their own uniform common law time limit for diversity actions after all, and that the twin aims, rather than having their source in separation of powers, are purely self-imposed. For an expansive account of federal courts' power to create federal common law, in which most declared limits are actually prudential and self-imposed, see Weinberg, supra note 126. More restrictive accounts can be found in Thomas W. Merrill, The Common Law Powers of Federal Courts, 52 U. CHI. L. REV. 1 (1985); Redish, supra note 76.

(172) For a general discussion of this problem, see Green, supra note 85. Of course, the same problem arises concerning Byrd's bound-up test. See supra note 85.

(173) In Wayman v. Southard, 23 U.S. (10 Wheat.) 1, 49 (1825), Chief Justice Marshall stated that there is no "original inherent power in the State legislatures, independent of any act of Congress, to control the modes of proceeding in suits depending in the Courts of the United States...." Indeed he thought that the absence of such power is "one of those political axioms, an attempt to demonstrate which, would be a waste of argument not to be excused." Id. It is unlikely that Marshall would have rejected Justice Brennan's position in Byrd v. Blue Ridge Rural Elec. Coop., 356 U.S. 525 (1958)--that is, that a state can extend its procedural rules to federal courts by binding them up with the state's causes of action. Marshall probably meant only that a state cannot seek to regulate federal procedural more directly, by extending its rules solely to federal courts within the state.

(174) See Air Products & Chem., Inc. v. Fairbanks Morse, Inc., 206 N.W.2d 414, 419 (Wis. 1973).

(175) See infra Part III.A.1.

(176) 337 U.S. 535 (1949); see also Westen & Lehman, supra note 27, at 357-58.

(177) Holmberg v. Armbrecht, 327 U.S. 392, 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.").

(178) See Ford v. Uniroyal Pension Plan, 154 F.3d 613, 616-20 (6th Cir. 1998); Koch v. Koch Indus., 996 F. Supp. 1273, 1279 (D. Kan. 1998).

(179) Commercial Union Ins. Co. v. Walbrook Ins. Co., 41 F.3d 764, 773-74 (lst Cir. 1994).

(180) Semtek Int'l Inc. v. Lockheed Martin Corp., 531 U.S. 497, 508 (2001).

(181) Id.

(182) Justice Stevens has argued that the Rules of Decision Act requires federal courts to borrow the forum state's statute of limitations when a federal statute lacks a limitations period of its own. See DelCostello v. Int'l Bhd. of Teamsters, 462 U.S. 151, 172-74 (1983) (Stevens, J., dissenting). For a criticism of such a reading of the Act, see supra Part II.A. Justice Scalia has gone even further and argued that the forum state's statute of limitations is not borrowed, but applies of its own force to federal causes of action in federal court. See Agency Holding Corp. v. Malley-Duff & Assocs., 483 U.S. 143, 157-65 (1987) (Scalia, J., concurring). The notion that a forum state's limitations period applies of its own force to federal statutes lacking such a period is implausible. Certainly, Scalia does not suggest that the forum state's supreme court would actually hold that the state's statute of limitation applies in federal courts in such cases. Indeed it is questionable whether states have the power to extend their statute of limitations to federal courts in this fashion. See supra note 173.

(183) E.g., North Star Steel Co. v. Thomas, 515 U.S. 29, 34-36 (1995); Graham Cnty. Soil & Water Conservation Dist. v. United States ex rel. Wilson, 545 U.S. 409, 415 (2005). 28 U.S.C. [section] 1658 (2006) establishes a four-year federal limitations period for federal statutory causes of action, but it does not apply to federal statutes enacted before December 1, 1990. For a broader interpretation of [section] 1658's effect, see Abner J. Mikva & James E. Pfander, Essay, On the Meaning of Congressional Silence: Using Federal Common Law to Fill the Gap in Congress's Residual Statute of Limitations, 107 YALE L.J. 393, 395 (1997).

(184) North Star Steel 515 U.S. at 33-34.

(185) See Hill, supra note 127, at 91-96.

(186) 28 U.S.C. [section] 1331 (2006). Although I will speak here of federal question jurisdiction in particular, my analysis will apply to other statutes giving federal courts jurisdiction over actions under federal law. E.g., E.g., U.S.C. [section] 1343 (giving federal jurisdiction over federal civil rights actions). With the exception of a one-year trial in 1801, 2 Stat. 89, 92 (1801), federal question jurisdiction was not introduced in federal trial courts until 1875. Judiciary Act of 1875, ch. 137, 18 Stat. 470. For a political history of the 1875 Act, see Howard Gillman, How Political Parties Can Use the Courts to Advance Their Agendas: Federal Courts in the United States, 1875-1891, 96 AM. POL. SCI. REV. 511 (2002); Tara Leigh Grove, The Structural Safeguards of Federal Jurisdiction, 124 HARV. L. REV. 869, 888-99 (2011). There is virtually no legislative history concerning the creation of federal question jurisdiction. James H. Chadbourn & A. Leo Levin, Original Jurisdiction of Federal Questions, 90 U. PA. L. REV. 639, 642-45 (1942).

(187) Grable & Sons Metal Prods., Inc. v. Darue Eng'g & Mfg., 545 U.S. 308, 312 (2005); Merrell Dow Pharm. Inc. v. Thompson, 478 U.S. 804, 826-28 (1986) (Brennan, J., dissenting); Donald L. Doernberg, There's No Reason for It; It's Just Our Policy: Why the Well-Pleaded Complaint Rule Sabotages the Purpose of Federal Question Jurisdiction, 38 HASTINGS L.J. 597, 647 (1987);John F. Preis, Reassessing the Purposes of Federal Question Jurisdiction, 42 WAVE FOREST L. REV. 247, 248-49 (2007).

(188) Of course, given that the Conformity Act was in place at the time, Act of June 1, 1872, ch. 255, [section] 5, 17 Stat. 196, this obligation would have been largely pleonastic. But it would have meant something when the Conformity Act was repealed by the Rules Enabling Act, 48 Stat. 1064 (1934).

(189) 3 THE DEBATES IN THE SEVERAL STATE CONVENTIONS ON THE ADOPTION OF THE FEDERAL CONSTITUTION 532 (Jonathan Elliot ed., J.B. Lippincott Co. 1941) (1836) [hereinafter DEBATES]. For a detailed discussion of the ratification debates concerning Article III arising under jurisdiction, see Anthony J. Bellia Jr., The Origins of Article III "Arising Under" Jurisdiction, 57 DUKE L.J. 263, 304-17 (2007). Bellia emphasizes the importance of arising under jurisdiction as a means of enforcing the supremacy of federal law.

(190) 4 id. at 158; see also THE FEDERALIST No. 80 (Alexander Hamilton) ("If there are such things as political axioms, the propriety of the judicial power of a government being coextensive with its legislative may be ranked among the number."). For a recent expression of the presumption in favor of federal jurisdiction for federal causes of action, see Mims v. Arrow Fin. Seros., LLC, 132 S. Ct. 740, 748-49 (2012).

(191) See supra note 149.

(192) See supra note 186.

(193) 4 DEBATES, supra note 189, at 141 (comments of Governor Johnston).

(194) THE FEDERALIST No. 82 (Alexander Hamilton) ("But this doctrine of concurrent jurisdiction, is only clearly applicable to those descriptions of causes, of which the State courts have previous cognizance.").

(195) Id.

(196) For a similar theory of state court jurisdiction for federal causes of actions, see Anthony J. Bellia, Jr., Congressional Power and State Court Jurisdiction, 94 GEO. L.J. 949, 969-72 (2006).

(197) Indeed, the entitlement with respect to federal causes of action is probably stronger, because state courts are arguably coequal interpreters of federal law with the lower federal courts, in the sense that they have no duty to abide by the latters' interpretations. See Lockhart v. Fretwell, 506 U.S. 364, 376 (1993) (Thomas, J., concurring) ("The Supremacy Clause demands that state law yield to federal law, but neither federal supremacy nor any other principle of federal law requires that a state court's interpretation of federal law give way to a (lower) federal court's interpretation."); Anthony J. Bellia, Jr., State Courts and the Making of Federal Common Law, 153 U. PA. L. REV. 825, 839 (2005).

(198) E.g., Employee Retirement Income Security Act of 1974, 29 U.S.C. [section][section] 1001-1461.[section]

(199) For a similar account of the equality approach, see Bellia, supra note 145, at 980. Bellia argues that the equality approach is exemplified in the Supreme Court's current reverse-Erie jurisprudence. Id. at 983-85. I disagree, for I see the effect of the twin aims.

(200) Cent. Vt. Ry. Co. v. White, 238 U.S. 507, 512 (1915).

(201) See RESTATEMENT (FIRST) OF CONFLICT OF LAWS [section] 595 (1934); RESTATEMENT (SECOND) OR CONFLICT OF LAWS [section] 133, cmt. b (1969).

(202) For such an account of reverse-Erie cases, see Bellia, supra note 145, at 959-63, 976-85. The account that I offer here is closer to Kevin M. Clermont, Reverse-Erie, 82 NOTRE DAME L. REV. 1, 36-37 (2006), which sees vertical uniformity having played a role in the Supreme Court's reverse-Erie jurisprudence.

(203) Engel v. Davenport, 271 U.S. 33, 38 (1926) ("This provision is one of substantive right, setting a limit to the existence of the obligation which the Act creates."); Felder v. Casey, 487 U.S. 131, 138 (1988).

(204) Atl. Coast Line R.R. Co. v. Burnette, 239 U.S. 199 (1915).

(205) Id. at 201.

(206) Davis v. Mills, 194 U.S. 451 (1904).

(207) Id. at 454.

(208) I am not saying, however, that the foreign nation's court would be obligated to prefer the federal limitations period over its own procedural law.

(209) Dice v. Akron, Canton & Youngstown R.R. Co., 342 U.S. 359 (1952).

(210) Brown v. W. Ry. of Ala., 338 U.S. 294 (1949).

(211) Dice, 342 U.S. at 363.

(212) Brown, 338 U.S. at 298.

(213) Id. at 296.

(214) Felder v. Casey, 487 U.S. 131 (1988).

(215) Id. at 138.

(216) See id.

(217) Id. at 151.

(218) See Clermont, supra note 202, at 36-37.

(219) Engel v. Davenport, 271 U.S. 33 (1926).

(220) Id. at 38.

(221) RESTATEMENT (FIRST) OF CONFLICT OF LAWS [section][section] 603, 605 (1934).

(222) RESTATEMENT (SECOND) OF JUDGMENTS [section] 19(f) (1982); RESTATEMENT (SECOND) OF CONFLICT OF LAWS [section] 110 cmt. b (1971).

(223) Brown v. W. Ry. of Ala., 338 U.S. 294, 299 (1949).

(224) Shaw v. Leatherberry, 706 N.W.2d 299, 310 (Wis. 2005); see also Clermont, supra note 202, at 36-37. As Clermont notes, the problem of disuniformity is not merely vertical, between the federal and forum state court, but also horizontal: states will apply their different procedures to the same federal cause of action.

(225) Felder v. Casey, 487 U.S. 131,160-61 (1988) (O'Connor, J., dissenting).

(226) Patzer v. Bd. of Regents of Univ. of Wis. Sys., 763 F.2d 851,855 (7th Cir. 1985).

(227) Felder, 487 U.S. at 160.

(228) For a classic critique of the Supreme Court's aggressive reverse-Erie jurisprudence made from the perspective of the equality approach, see Hart, supra note 116, at 508.

(229) Felder, 487 U.S. at 151.

(230) Id.

(231) Unless, perhaps, countervailing state interests are sufficiently substantial. Johnson v. Fankell, 520 U.S. 911 (1997), appears to be a case in which the decision would be the same under either the equality or the reverse-twin aims approach, due to the strength of countervailing state interests. In Johnson, the Supreme Court unanimously held that Idaho state courts entertaining federal civil rights actions were not required to provide interlocutory appeals of denials of qualified immunity, although that was the practice in federal courts. Id. at 922-23.

(232) Haywood v. Drown, 556 U.S. 729 (2009).

(233) Id. at 732; see N.Y. CORRECT. LAW. [section] 24 (McKinney 2003).

(234) Haywood, 556 U.S. at 740.

(235) Id. at 766 (Thomas, J., dissenting).

(236) See supra note 149. The majority in Haywood also concluded that the New York statute, although facially neutral, did indeed discriminate against federal civil rights actions. 556 U.S. at 736-37. My critique of Haywood is not directed to this part of the opinion.

(237) Patsy v. Bd. of Regents of Fla., 457 U.S. 496, 506-507 (1982).

(238) A possible candidate is the Class Action Fairness Act of 2005 (CAFA), which provides federal jurisdiction over large state law nationwide class actions in which there is minimal diversity. See 28 U.S.C. [section] 1332(d)(2). Some have argued that CAFA gives federal courts the power to create federal substantive common law. Suzanna Sherry, Overruling Erie: Nationwide Class Actions and National Common Law, 156 U. PA. L. REV. 2135, 2136-36 (2008). Our question, however, is whether the twin aims apply to a federal court entertaining a state law action under CAFA. The legislative history in CAFA disavows any attempt to change federal courts' obligations under Erie. See S. REP. No. 109-14, at 49 (2005); H.R. REP. No. 108-144, at 26 (2003). But it is likely that what was being referred to was federal courts' Erie obligations to respect applicable state law. It is not clear that the twin aims were being contemplated.

(239) See, e.g., Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 40 (1988) (Scalia, J., dissenting); Godin v. Schencks, 629 F.3d 79, 91-92 (1st Cir. 2010).

(240) Federal courts usually argue mistakenly that the twin aims apply in supplemental jurisdiction due to respect for state law under Erie. See, e.g., A.I. Trade Fin., Inc. v. Petra Int'l Banking Corp., 62 F.3d 1454, 1463 (D.C. Cir. 1995); Terry v. June, 420 F. Supp. 2d 493, 500 (W.D. Va. 2006).

(241) 28 U.S.C. [section] 1367 (2006).

(242) Robert A. Schapiro, Polyphonic Federalism: State Constitutions in the Federal Courts, 87 CAL. L. REV. 1409, 1422 (1999).

(243) United Mine Workers of Am. v. Gibbs, 383 U.S. 715, 726 (1966); Miller Aviation v. Milwaukee Cnty. Bd. of Supervisors, 273 F.3d 722, 731-32 (7th Cir. 2001); Richard D. Freer, Compounding Confusion and Hampering Diversity: Life After Finley and the Supplemental Jurisdiction Statute, 40 EMORY LJ. 445, 449 (1991).

(244) See Westen & Lehman, supra note 27, at 385-87.

(245) Under the transactional standard for claim preclusion, joinder would be required. See RESTATEMENT (SECOND) OF JUDGMENTS [section] 24 (1982).

(246) Cf. Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 40 (1988) (Scalia, J., dissenting) ("The decision of an important legal issue should not turn on the accident of diversity of citizenship ... or the presence of a federal question unrelated to that issue." (citation omitted)). This is not to say that a uniform federal procedural common law rule can never be applied to a state law action with supplemental jurisdiction. But the reason must be that there are federal interests of sufficient strength to overcome the twin aims. For example, as a practical matter it is necessary that the same rule concerning the attorney-client privilege be applied to federal and state law actions brought together before a federal court. Having been exposed to a piece of evidence in connection with one cause of action, the finder of fact cannot help but consider it in connection with the other. Thus, federal interests recommend using the federal rule for both state and federal actions. See yon Bulow ex rel. Auersperg v. von Bulow, 811 F.2d 136, 141 (2d Cir. 1987). I thank Anthony Bellia for identifying this issue.

(247) 28 U.S.C. [section] 1334(a), (e)(1) (2006).

(248) Id. [section] 1334(b). There is also the possibility of supplemental jurisdiction for state law actions in bankruptcy cases under [section] 1367. For a critical discussion, see Susan Block-Lieb, The Case Against Supplemental Bankruptcy Jurisdiction: A Constitutional, Statutory, and Policy Analysis, 62 FORDHAM L. REV. 721 (1994).

(249) Id. [section] 1334(c). In some cases, abstention is mandatory. Id. [section] 1334(c)(2).

(250) 313 U.S. 487, 496 (1941).

(251) See supra text accompanying notes 112-116. For a discussion of alternative views of Klaxon, see supra note 112 and infra note 275.

(252) See, e.g., Alfred Hill, The Erie Doctrine in Bankruptcy, 66 HARV. L. REV. 1013 (1953); Thomas E. Plank, The Erie Doctrine and Bankruptcy, 79 NOTRE DAME L. REV. 633, 643-48 (2004).

(253) Butner v. United States, 440 U.S. 48, 55 (1979); Plank, supra note 252, at 663-78.

(254) Raleigh ex rel. Estate of Stoecker v. Ill. Dep't of Revenue, 530 U.S. 15, 20 (2000) (holding that a bankruptcy court must respect the state rule concerning burden of proof for state law claims against a debtor).

(255) Butner, 440 U.S. at 54-55; see also Travelers Cas. & Sur. Co. v. Pac. Gas & Elec. Co., 549 U.S. 443, 450 (2007) ("the 'basic federal rule' in bankruptcy is that state law governs the substance of claims" (quoting Raleigh, 530 U.S. at 20)).

(256) In re Gaston & Snow, 243 F.3d 599 (2d Cir. 2001).

(257) Id. at 606 (quoting Atherton v. FDIC, 519 U.S. 213, 218 (1997)). In re Merritt Dredging Co., 839 F.2d 203 (4th Cir. 1988), also treats the matter as linked to Erie and respect for state law.

(258) In re Lindsay, 59 F.3d 942 (9th Cir. 1995). The question of state law at issue in Lindsay was fled to the whether a conveyance by the debtor was fraudulent under 11 U.S.C. [section] 548.

(259) In re Lindsay, 59 F.3d at 948.

(260) See Douglas G. Baird, Loss Distribution, Forum Shopping, and Bankruptcy: A Reply to Warren, 54 U. CHI. L. REV. 815, 824-28 (1987); John T. Cross, State Choice of Law Rules in Bankruptcy, 42 OKLA. L. REV. 531, 526-29 (1989).

(261) See supra Part II.B.2.

(262) In re Xonics, Inc., 813 F.2d 127, 131 (7th Cir. 1987); see also Block-Lieb, supra note 248, at 811-14.

(263) See Cross, supra note 260, at 576-78.

(264) See 28 U.S.C. [section] 1408 (2006). But see 28 U.S.C. [section] 1409 (limiting venue for actions by trustee of small value to district where defendant resides). These restrictions still leave significant discretion in choosing a venue, particularly when the debtor is a business. On horizontal forum shopping in Chapter 11 bankruptcy, see Theodore Eisenberg & Lynn M. LoPucki, Shopping for Judges: An Empirical Analysis of Venue Choice in Large Chapter 11 Reorganizations, 84 CORNELL L. REV. 967 (1999).

(265) 28 U.S.C. [section] 1334(e)(1) (2006); FED. R. BANKR. P. 7004(d) (2010).

(266) In re Gaston & Snow, 243 F.3d 599 (2d Cir. 2001).

(267) A complicating factor is that under New York law, the six-year period was to be used only if the plaintiff was a New York resident, and Gaston & Snow was not a resident of New York within the meaning of the statute. But the argument that the six-year period did not apply was not raised below and so was waived on appeal. Id. at 608.

(268) See Burnham v. Super. Ct. of Cal., 495 U.S. 604 (1990) (holding this service to be sufficient for personal jurisdiction).

(269) The case might be transferred from one federal court to the district with jurisdiction over the bankruptcy case under 28 U.S.C. [section][section] 1404(a) or 1412 or the practical equivalent of transfer might occur by the filing of a proof of claim in the federal court with jurisdiction over the bankruptcy case. See In re Coudert Bros. LLP, 673 F.3d 180, 182 (2d Cir. 2012). For an action filed in state court, removal to the federal court in the district where the state court is located would precede transfer. 28 U.S.C. [section] 1452.

(270) Coudert Bros., 673 F.3d at 185, 191 (holding that Connecticut limitations period rather than New York's period should be used for an action against the debtor, under either Connecticut or United Kingdom law, originally filed in state court in Connecticut, but ultimately entertained by bankruptcy court in New York). Cf. Ferens v. John Deere Co., 494 U.S. 516 (1990) (holding that federal courts sitting in diversity enter-raining a state law action transferred from a federal court in another state should use the transferor's state's choice-of-law rules); Van Dusen v. Barrack, 376 U.S. 612 (1964).

Another example of the twin aims applying due to pre-bankruptcy filing of the state law action is Nuveen Mun. Trust v. Withum Smith Brown, P.C., 692 F.3d 283 (3d Cir. 2012). A legal malpractice action was brought in federal court in New Jersey under diversity, but dismissed because it failed to satisfy New Jersey's Affidavit of Merit (AOM) statute, which requires that a plaintiff in a professional malpractice action provide an affidavit by an expert stating that there is evidence that the action has merit. Federal courts sitting in diversity in New Jersey borrow the AOM statute due to the twin aims. See Chamberlain v. Giampapa, 210 F.3d 154, 161 (3d Cir.2000). After the Third Circuit remanded the case to reconsider the existence of diversity jurisdiction, the district court held that it had bankruptcy jurisdiction over the action. On a second appeal, the Third Circuit concluded that the AOM statute still should be used, despite the different jurisdictional source. This makes good sense, for the fact that the plaintiff originally brought suit in federal court in New Jersey is strong evidence that the action would have proceeded in New Jersey state court in the absence of bankruptcy.

(271) Granted, it is conceivable that a court in another state might have entertained the action if the defendant has been served with the summons and complaint there and the court did not dismiss on forum non conveniens grounds. We cannot be completely certain of the state where the action would have been brought had bankruptcy not occurred. But the same lack of certainty arises in diversity cases, since it is always possible that the plaintiff's choice between federal and state court was diagonal--between a federal court in one state and a state court in a different state. See supra text accompanying notes 143-45. The twin aims are justified when the level of certainty about the presumptive state forum is sufficiently high and when a bankruptcy court entertains a state law action where the parties and transaction are located in a single state, the level of certainty is arguably as high as it would have been in a diversity case.

(272) See In re Johnson, 453 B.R. 433 (Bankr. M.D.Fla. 2011) (holding, on the basis of the twin aims, that a Florida statute forbidding a plaintiff to plead punitive damages until he offers evidence showing a reasonable basis for that relief should be used by a bankruptcy court in Florida entertaining an action under Florida law by bankruptcy trustee on behalf of a Florida debtor against Florida defendant).

(273) District of Columbia Court Reform and Criminal Procedure Act of 1970, title I, Pub. L. No. 91-358, 84 Stat. 473, 475 (1970).

(274) See Williams v. Rawlings Truck Line, Inc., 357 F.2d 581, 583-85 (D.C. Cir. 1965) (using federal common law choice-of-law rules when sitting in diversity, without deference to any state's rules); Tramontana v. S.A. Empresa De Viacao Aerea Rio Grandense, 350 F.2d 468 (D.C. Cir. 1965) (same).

(275) John Cross adopts such a position concerning choice-of-law rules. See Cross, supra note 260, at 572-78. A bankruptcy court should use the choice-of-law rules that would be used by the courts of one of the states that would have entertained the action but for bankruptcy. Cross mistakenly justifies his approach, however, by appeal to state interests. As he puts it, Klaxon follows from the Rules of Decision Act, because "the choice of law rules of a state are part and parcel of the substantive rights afforded litigants under the laws of that state." Id. at 559. Although the personal jurisdictional scope of a bankruptcy court removes the duty to use the forum state's choice-of-law rules, Klaxon and the Act still put a duty on the court to use the choice-of-law rules of a state that could have gotten jurisdiction.

This reading of Klaxon and the Act is mistaken. Assume that only New York or Pennsylvania law may be constitutionally applied to an event. A plaintiff sues concerning the event in diversity in federal court in Vermont. Under Klaxon, the federal court should use Vermont's choice-of-law rules. But Vermont's choice-of-law rules cannot possibly be understood as part and parcel of the substantive rights afforded litigants under Vermont law. Vermont can create no substantive rights. Nor can Vermont's choice-of-law rules be understood as part and parcel of the substantive rights afforded litigants under New York or Pennsylvania law, since Vermont has no control over those states' laws. Cross's mistake is precisely the one I argue against in this Article--interpreting Klaxon, and the twin aims generally, in terms of states' interests in their rules applying in federal court.

Cross's reading of Klaxon should be distinguished from Roosevelt's. See supra note 112. For Roosevelt, only certain choice-of-law rules are substantive, whereas for Cross all apparently are. Furthermore, for Roosevelt it follows from Klaxon that a federal court choosing between the laws of Pennsylvania and New York has a duty to look to Pennsylvania's substantive choice-of-law rules concerning the applicability of Pennsylvania law and New York's substantive choice-of-law rules concerning the applicability of New York law. His emphasis is on the choice-of-law rules of the state whose law the federal court is considering applying, not the choice-of-law rules of the courts of a state with jurisdiction.

(276) I would recommend a similar solution for state law actions in federal court under statutory interpleader, where nationwide service of process is also available. 28 U.S.C. [section] 1335(a) (1). See supra text accompanying notes 136-38. The Supreme Court has mistakenly held that a federal court sitting in statutory interpleader should borrow from the forum state's procedural law, even though a forum state court may not have been able to get jurisdiction. See Griffin v. McCoach, 313 U.S. 498, 503-06 (1941); see also Equitable Life Assurance Soc'y v. McKay, 837 F.2d 904, 905 (9th Cir. 1988); Travelers Indem. Co. v. Moore, 642 F.Supp. 1119, 1124 (C.D. Ill. 1986). Similar problems arise in cases in which a party was served pursuant to the "100-mile bulge" provisions of FED. R. Cir. P. 4(k) (1) (B), where federal courts have also borrowed forum state law even though a forum state court could not have gotten jurisdiction. See Sprow v. Hartford Ins. Co., 594 F.2d 412, 415-17 (5th Cir. 1979).

(277) See supra text accompanying notes 31-34 and 182-184.

(278) See Raleigh ex rel. Estate of Stoecker v. Ill. Dep't of Revenue, 530 U.S. 15, 20-26 (2000) (holding that a bankruptcy court must respect state rule concerning burden of proof for state law claim against debtor).

(279) See RESTATEMENT (SECOND) OF JUDGMENT [section] 19 cmt. f; Rochelle Cooper Dreyfuss & Linda J. Silberman, Interjurisdictional Implications of the Entire Controversy Doctrine, 28 RUTGERS L.J. 123, 145-47 (1996).

(280) Although no bankruptcy court, to my knowledge, has ever mentioned this concern, I do not think it is an accident that they tend to favor the longest of the limitations periods that would have been applied by the states that could have gotten jurisdiction over the action. See In re Coudert Bros., 673 F.3d 180 (2012) (choosing the longest period); In re Segre's Iron Works, Inc., 258 B.R. 547 (Bankr. D. Conn. 2001) (same); In re SMEC, Inc., 160 B.R. 86, 89-91 (M.D. Tenn. 1993) (same) ; see also In re Ovetsky, 100 B.R. 115 (Bankr.. N.D. Ga. 1989) (same). Indeed, even In re Gaston & Snow, 243 F.3d 599 (2d Cir. 2001) follows this principle. The court rejected the four-year limitations period of the state with the most significant relationship in favor of the forum state's six-year period. Although the action could not have been entertained by a forum state court, the federal court noted that another state where the action could have been brought, Massachusetts, also had a six-year period. Id. at 608 n.7.

(281) John Cross has argued against uniform federal choice-of-law rules in bankruptcy on the grounds that diverging from the choice-of-law rules that would be used by a state court that had personal jurisdiction over the parties would frustrate the reasonable expectations of the parties concerning which state law applied to their transaction. See Cross, supra note 260, at 535. Such concerns about party expectations are misplaced. First of all, where personal jurisdiction may later be available may not itself be able to be anticipated by the parties at the time of the transaction. Second, federal choice-of-law rules are modeled on the Second Restatement. See In re Vortex Fishing Sys., Inc., 277 F.3d 1057, 1069-70 (9th Cir. 2002); In re Lindsay, 59 F.3d 942, 948 (9th Cir. 1995). The Second Restatement considers party expectations when determining which state's law to apply. RESTATEMENT (SECOND) OF CONFLICTS [section] 6 (1971). Indeed, for federal courts in states that still use the First Restatement, it is Klaxon that is more likely to frustrate the parties' expectations, for the First Restatement generally decides what state's law should be used on the basis of triggering events, such as the place of the harm, that can fail to track the expectations of the parties. See RESTATEMENT OF THE CONFLICT OF LAWS [section] 377 (1934); Ala. G. S. R. v. Carroll, 11 So. 803 (Ala. 1892) (applying Mississippi law to determine applicability of the fellow-servant rule to wrongdoing of corporation's employee in Alabama, because harm from wrongdoing occurred in Mississippi). Third, worries about the parties' reasonable expectations are not a serious concern in any case, because any choice-of-law rule, whether federal or state, that clearly violated these expectations would be unconstitutional under the Due Process Clauses of the Fifth or Fourteenth Amendments. See generally Allstate Ins. Co. v. Hague, 449 U.S. 302, 326-31 (1981) (concerning Fourteenth Amendment Due Process limits in state court).

(282) Vanston Bondholders Protective Comm. v. Green, 329 U.S. 156 (1946).

(283) Vanston has sometimes been relied upon by courts that favor federal choice-of-law rules in bankruptcy. See, e.g., In re SMEC, Inc., 160 B.R. 86, 89-91 (M.D. Tenn. 1993); In re Kaiser Steel Corp., 87 B.R. 154, 158 (Bankr. D. Colo. 1988).

(284) Vanston, 329 U.S. at 156.

(285) Id. at 163.

(286) Id. at 161-62.

(287) Id.

(288) Id. at 162.

(289) See supra text accompanying notes 110-116.

(290) See 11 U.S.C.A. [section] 541 (2006).

(291) A number of writers have come to this conclusion. See, e.g., William Baude, Beyond DOMA: Choice of State Law in Federal Statutes, 64 STAN. L. REV. 1371, 1411 (2012) (addressing federal choice-of-law rules concerning validity of marriage in context of bankruptcy); Jackie Gardina, The Perfect Storm: Bankruptcy, Choice of Law, and Same-Sex Marriage, 86 B.U.L. REV. 881, 922-30 (2006) (same).

(292) The question of the choice-of-law rules that should be used by a bankruptcy court in connection with state law issues necessary to apply bankruptcy law (as opposed to choice-of-law rules for free-standing state law causes of action) is part of a more general problem of choice of law in federal question cases. For a discussion, see Green, Presumption, supra note 74, at 1281-85; Martha A. Field, Sources of Law: The Scope of Federal Common Law, 99 HARV. L. REV. 881, 958-60 (1986); Paul J. Mishkin, The Variousness of "Federal Law": Competence and Discretion in the Choice of National and State Rules for Decision, 105 U. PA. L. REV. 797, 808-10 (1957); Radha A. Pathak, Incorporated State Law, 61 CASE W. res. L. REV. 823 (2011); Note, Applicability of State Conflicts Rules When Issues of State Law Arise in Federal Question Cases, 68 HARV. L. REV. 1212 (1955).

(293) For an example of such criticism in connection with Gasperini v. Ctr. for Humanities, Inc., 518 U.S. 415 (1996), see Earl C. Dudley & George Rutherglen, Deforming the Federal Rules: An Essay on What's Wrong With the Recent Erie Decisions, 92 VA. L. REV. 707, 708 (2006); C. Douglas Floyd, Erie Awry: A Comment on Gasperini v. Center for Humanities, Inc., 1997 BYU L. REV. 267, 269-70 (1997);Jennifer S. Hendricks, In Defense of the Substance-Procedure Dichotomy, 89 Wash. U. L. Rev. 103 (2011). Understood as a choice between federal and state law, Gasperini must have been wrongly decided, for the amalgamation of federal and state standards the Court settled on could not possibly have been the result of a principled view about the division of federal and state lawmaking power.

(294) For a general discussion of this problem, see Green, supra note 85.

(295) In both Shady Grove Orthopedic Assocs., P.A. v. Allstate Ins. Co., 130 S. Ct. 1431 (2010), and Gasperini v. Ctr. for Humanities, Inc., 518 U.S. 415 (1996), members of the Court had profound disagreements about whether a New York rule was bound up with the New York actions brought in federal court. Compare Shady Grave, 130 S. Ct. at 1455-60 (Stevens, J., concurring in judgment), with id. at 1469-72 (Ginsburg, J., dissenting). Compare Gasperini, 518 U.S. at 428-29 (Ginsburg, J.), and id. at 446-47 (Stevens, J., dissenting), with id. at 464-65 (Scalia, J., dissenting). Despite this disagreement, no one thought to certify the question to the New York Court of Appeals.

(296) See, e.g., Shady Grove, 130 S. Ct. at 1457-60 (Stevens, J., concurring in judgment) (adopting a rebuttable presumption that a rule that is formally designated by a state as procedural is not intended to follow the state's causes of action into other court systems).

(297) See, e.g., Ford Motor Co. v. Leggat, 904 S.W.2d 643 (Tex. 1995).

(298) As Hanna made clear, Congress's power to establish (and, if it wishes, to disestablish) the lower federal court system--augmented by the Necessary and Proper Clause--gives it the power to regulate matters coming before a federal court that are arguably procedural that is, matters "which, though falling within the uncertain area between substance and procedure, are rationally capable of classification as either." Hanna, 380 U.S. 460, 472 (1965); see also Stewart Org., Inc. v. Ricoh Corp., 487 U.S. 22, 32 (1988) (reasserting this congressional power). Hanna suggests a rule of priority under which federal procedure trumps conflicting state law.

[c] 2013 Michael Steven Green. Individuals and nonprofit institutions may reproduce and distribute copies of this Article in any format at or below cost, for educational purposes, so long as each copy identifies the author, provides a citation to the Notre Dame Law Review, and includes this provision in the copyright notice.

Michael Steven Green, Dudley W. Woodbridge Professor of Law, William & Mary Law School. I would like to thank Pete Alces, Anthony Bellia, Lea Brilmayer, Kevin Clermont, John Cross, Neal Devins, Scott Dodson, Abbe Gluck, Tara Grove, Nate Oman, Thomas Plank, Richard Posner, Kim Roosevelt, Bill Van Alstyne, and Tobias Wolff for helpful comments.
COPYRIGHT 2013 University of Notre Dame Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2013 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:III. The Twin Aims Outside Diversity through Conclusion, with footnotes, p. 1904-1937
Author:Green, Michael Steven
Publication:Notre Dame Law Review
Date:Apr 1, 2013
Words:22862
Previous Article:The twin aims of Erie.
Next Article:Sustained dissent and the extended deliberative process.
Topics:

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters