An American Law Institute project on the conflict of laws is preparing to bring forth a new Restatement on the subject. The issues most hotly debated behind the scenes are those involving choice of law, a somewhat technical legal specialty with a well-earned reputation for impenetrability. Despite the theoretical difficulty of the topic, the drafting of the new Restatement (Third) has been the cause of intense interest on the part of the bench and bar. Selection of the applicable law—while deeply influenced by theoretical considerations—has immense practical consequences because of the recurrence of the issue in contemporary litigation.

The leading modern school of thought on choice of law is an approach known as “interest analysis.” Interest analysis was the product of the legal realist movement, a fact that was partially responsible for its widespread influence. In modern choice-of-law theory, the selection of the applicable law is structured upon the assumption that choice of law is not significantly different from the ordinary processes used in interpreting domestic substantive law. In reality, however, questions of jurisdiction are different from questions of substance because of the presence of two or more independent voices.

The presence of independent voices creates problems because modern choice-of-law theory purports to respect the different “interests” of the states whose law might be applied. But what interests do states really have? Are these subjectively determined (the interests that states think that they have) or are they objective (conceptual constructs devised by others, on behalf of and paternalistically imposed upon the state)? This is a conundrum for realists. But if the realist foundations of modern choice-of-law theory are to be respected, objective interests do not exist; jurisdictional realism must be reconfigured to avoid imposing contentious theoretical assumptions on those state decision-makers who create the law.


The history of American choice-of-law theory is inextricably bound up with the history of American legal realism. As legal realism swept through American law schools, it transformed almost everything that stood in its path. Although the rewriting of conflict of laws1 was somewhat slow in getting started, that subject was no exception.2 It was legal realism that freed an entire generation of conflict-of-laws scholarship from the bonds of conceptualist territorialism and that assigned to choice of law a new, important function: the furtherance of state substantive policies.3No wonder the choice-of-law reformers’ devotion to realist jurisprudence; they owed it their emancipation.

So committed were the reformers to legal realism in principle, however, that they failed to see how far they strayed from legal realism in practice. The self-consciously modern model that they proposed replaced the conceptualism of the theorists of old with a new methodology of statutory construction and interpretation.4 Instead of applying formalistic rules based on respect for a sister state’s territorial sovereignty and the need for decisional uniformity, choice-of-law decision-makers were supposed to identify the “interests” of the various states through the same sort of purposive reasoning that the legal realists had brought to bear on other legal subjects.5 In this way, the choice-of-law process would contribute to the furtherance of state substantive policies, the sole legitimate objective of the choice-of-law process.

The reformers dismissed all criticism of their approach as contrary to the spirit of realism and modernity.6 But two challenges, in particular, refused to go away. The first was the problem of the silent statute: Most substantive laws say nothing about their intended multistate scope, so how can statutory construction provide answers in choice-of-law disputes?7 Second, statutory construction does not explain why states were encouraged to determine de novo a foreign law’s scope of application. This is the problem of the other state’s interests:8 Given that the courts of one state are not authorized to make determinations concerning the laws of the others, how are states supposed to identify those laws’ applicability? And given that states adopt different approaches to interpretation of their own laws than to interpretation of the other state’s laws, how can the modern theory describe the two of them identically?9

It is worth pausing to explain what we mean by “modern” choice-of-law theory. The American Law Institute has recently begun work on a Restatement (Third) of Conflict of Laws.10 The committee faces the daunting task of updating the Restatement (Second), compiled in 1971, with nearly five decades of academic and legal progress in a field that Dean William Prosser once described as “a dismal swamp, filled with quaking quagmires.”11The core of the new Draft Restatement is its two-step framework, in which courts first determine the “scope” of the statute providing the cause of action in a case and then apply rules to resolve cases in which there are multiple state scope claims.12 This new approach innovates on the theories of Professor Brainerd Currie, who pioneered the governmental interest analysis approach, which was largely omitted from the Second Restatement.13 All choice-of-law scholars in the last fifty years have recognized that there are situations in which multiple states’ laws apply to a given set of facts. This recognition divides the field of conflicts neatly into early, prescriptive theories (like those of Professor Joseph Beale) and newer, realist-influenced theories, like Currie’s. But the “modern” theories discussed in this Article are a specific subset of these newer approaches. Common to both the modern theories of interest to us is the idea, shared by the Draft Restatement and the theories of Currie, that the “process of determining whether a state’s law is relevant is simply the typical method of interpretation used to determine the scope of a law in purely domestic cases.”14 In particular, the Draft Restatement’s modern approach builds on Currie’s central insight: that “choice-of-law questions should be resolved so as to promote . . . the governmental policies said to underlie the laws in question.”15 As Professor Kermit Roosevelt III, the reporter for the new Restatement, has described, the use of statutory interpretation in “trying to decide whether the facts of the case fall within the scope of a statute” asks “whether application of the statute to those facts would promote the purposes behind the statute.”16 It is the theory of the Draft Restatement, and all other theories that share the core features just described, that this Article lumps together under the label of “modern.”

If compatibility with legal realism is the test for choice-of-law theory, then the so-called modern choice-of-law theory scores rather low. It was realism, after all, that recognized that the law of a state is what its own courts say it is—and not what another state’s court might think that it is or ought to be.17 And it was realism that argued that law was open textured and thus that judicial creativity was unavoidable.18 Can it be called interpretation when the judge injects extrinsic norms or policies to fill gaps left by legislative silence?19 There is heated disagreement over the point at which the exercise of judicial discretion ceases to be explicable as judicial interpretation and starts, instead, to look like judicial usurpation.20 Where on this spectrum does choice-of-law theory fall? Modern theorists’ characterization of what they were doing as “statutory construction” defused many objections, but at the cost of introducing much confusion: What exactly does this mean?

The problem with modern choice-of-law theory lies in its failure to adapt realism to the special circumstances of situations involving more than one authoritative decision-maker. Jurisdiction is the study of multistate problems—problems, that is, involving two or more independent voices. The central paradox of a realist theory of jurisdiction lies in a state’s inability to identify authoritatively another state’s authentic needs, concerns, and interests;21 and legal decision-making cannot take account of what cannot be identified. Modern choice-of-law theory was grounded on the conviction that choice of law is no different from any other legal subject;22 that it simply extended realist insights from domestic law into the multistate context.23 But choice of law is all about the special jurisprudential circumstances of multistate authority; it is about the need to recognize and accommodate what one has no authority to identify or define.

Indeed, foregrounding realist principles reveals the key mistake of modern choice-of-law theories: they impose their own value judgements on states, rather than recognizing states’ authority to define interests of their own—essentially, the problem of the “silent statute.”24 This mistake manifests in two main ways. First, these theories tend to place arbitrary limitations on the values that may be taken into account in making decisions about states’ own interests, and thus the multistate scope of their own laws.25 And second, these theories fail to respect states’ choice-of-law decisions and the interests that they rely on—the problem of “the other state’s interests.”26 These problems do not necessarily compel the conclusion that the field of conflicts of law is not suited for the application of realist principles. Instead, they merely suggest that incorporating realism into choice of law requires some consideration of the unique characteristics of the multistate situation.27 In particular, a realist theory of conflicts requires state respect for the differing opinions of sister states, as realism recognizes that in the absence of the “correct” answers assumed by prescriptive theories, authoritative statements of law are those identified by the tribunals with the authority to pronounce them.

We start with an account of how the theorists of old envisioned the foundations of the choice-of-law process and then describe the starkly different assumptions on which the realist revolution relied. The two persistent challenges to modern choice-of-law theory are then explained: the problem of the silent statute and the problem of the other states’ interests. The modern theorists’ inability to answer these challenges, we go on to show, is a sign that they have lost touch with the realist principles that once inspired them. “Real” realism would look very different from the supposedly “modern” choice-of-law methods that have come to dominate discussions of contemporary choice-of-law theory. Real realism should be brought back, we then argue, and we conclude with some suggestions about what real realism in the jurisdictional context would look like.

I.  The Realist Revolution in Conflict of Laws

Few topics evoke more enthusiasm in the typical choice-of-law expert than the subject’s historical origins, which date back almost to the dawn of human civilization.28 Bold and curious minds have entertained the challenges of conflicts of law since at least the time of the early Egyptians.29 The Greeks, it is said, maintained separate courts for cases which involved foreigners.30 Roman law enshrined a respect for autonomy in the enjoyment of local custom, even when generally applicable Roman law would dictate a different result.31 From medieval times32 to the great classical publicists of private international law,33 conflict of laws has maintained a firm grip on the legal imaginations of the theoretically minded.

It was in the United States that choice-of-law theory attained its highest form. The federalist structure of the United States created a wealth of opportunities to consider conflicts questions. Not only do the laws of the several states often conflict, but state law often differs from federal law, and both state and federal law are inconsistent with international law and the laws of other nations. Unsurprisingly, then, courts across the nation have long grappled with conflicts principles in important cases.34

Choice of law today still maintains its historic reputation: not only terminologically and conceptually dense, but also gratuitously confusing—a subject, in short, better left to academics. Participants from all of the major schools of thought share in the responsibility for this notoriety, but the tone was first set by the earliest theorists and the blame, accordingly, has fallen mainly on them. The earliest American theorists assumed the posture of philosophers of state sovereignty, purveyors of logical truth, and arbiters of all legal rights and obligations.35 Beale, the main author of the First Restatement of the Conflicts of Law, was their acknowledged leader.36

A.    The First Restatement and the Theorists of Old

For most of the early twentieth century, conflict of laws was dominated by Beale.37 The author of a major treatise on the subject, Beale systematized choice of law and unified its various branches under a single legal rubric: vested rights. 38The metaphysical nature of his enterprise made him one of the two chief protagonists in the immediate past century of choice-of-law drama. The other was Beale’s chief critic, Professor Currie, whose ideas will be central to our analysis of the modern choice-of-law revolution.

Beale will long be remembered, although not particularly fondly. As the progenitor of American choice of law, his staying power has been substantial; in certain discrete applications such as the law of real property or decedents’ estates, some of the principles he enunciated are still applied today.39 But this is not the source of his continued fame; it was his remarkable theoretical ambition—some might say audacity—that earned him a place in any serious historical account of his subject. Credited with the popularization of conflict of laws in American law schools, Beale’s work heavily influenced the First Restatement of the Conflict of Laws (for which he served as the reporter).40 His approach to conflicts of law epitomized the old-fashioned legal formalism.41 Dictatorial in tone, it was a logically self-contained system of rules that celebrated the supposed uniqueness of choice of law.42 His theory of “vested rights” was inspired by a jurisprudence considered “remarkably reflective[,] even if ultimately unconvincing.”43

Beale’s scholarship was based on the principle of territorialism, from which a self-contained, deterministic system of rules could be derived that, when followed, would point to the one correct outcome in any conflicts case. 44Because states were assumed to have sovereignty over their own territorial space, figuring out the “correct” answer to any conflicts question reduced neatly to the problem of figuring out the location of the “last act” necessary to complete the cause of action.45 This was the state where the right at issue would have vested.46 The First Restatement reflected this idea, consisting largely of a collection of rules that could be applied to different types of legal disputes in order to identify and determine the location of this last act.47

There were also attempts to justify the First Restatement rules in terms of the practical benefits that they supposedly ensured.48 Chief among the so-called choice-of-law values that the Bealean system was said to promote were predictability, uniformity, avoidance of forum shopping, and interstate harmony. All resulted from the supposed ability of the Bealean system to generate the same result regardless of where the litigation was brought. The assumption was that all states would follow the First Restatement, including not only the general principles and methodology of the First Restatement, but also all matters of detail that might affect the outcome.

Beale had vastly overestimated, however, both the ability and also the willingness of judges to conform their decisions to what the Restatement required. Among the persnickety rules that judges were obliged to follow, for example, was a special rule for death by slow-acting poison: the applicable law was that of the state “where the deleterious substance [took] effect” rather than the place where it was administered or the place where the victim eventually died.49 And Beale’s reasons for requiring such results could be less than satisfying. He proposed, for example, that judges charged with identifying the last act were bound to follow “general common law”—a set of “untethered” norms that was presumed to exist independently of any body of man-made legal rules.50 All in all, the probability that nationwide uniformity of result could be achieved through such questionable methods was not high. As it turned out, uniformity of result imposed a far more thoroughgoing standard of loyalty to the First Restatement than normal judges could realistically satisfy. The necessary loyalty did not materialize, and the First Restatement’s longing for uniformity went unfulfilled.

Beale acknowledged that substantive laws might differ from one state to another. But he believed that choice-of-law outcomes were different. Choice-of-law questions supposedly had a single right answer that could be ascertained by studying the universal logic underlying the metaphysical nature of sovereignty.51 Beale’s highly conceptual analysis claimed a monopoly on truth, treating alternative analyses as categorically unacceptable. It was doomed before it even got off the ground.

B.    The Triumph of Legal Realism

The “conflicts revolution” that flourished during the mid-twentieth century began almost as soon as the First Restatement was published.52 Criticism of the First Restatement was swift and sharp.53 Some of the criticisms were directed at the ways that the First Restatement failed to meet its own announced objectives.54 Critics were quick to point out that the First Restatement did not deliver the certainty it promised. They noted the arbitrariness of the rules that Beale selected, including his almost complete disregard of nonterritorial factors.55 This arbitrariness undercut the functional rationalizations for the First Restatement’s rules—their uniformity and predictability were compromised by the fact that the reasoning of virtually any case could be manipulated in order to avoid applying rules the judge found unpalatable.

Although some such critiques were rooted in perceived errors in the detailed rules enumerated in the First Restatement, others reflected a rejection of the First Restatement’s overall approach.56 The dogmatism of Bealean metaphysics was a lightning rod for academic criticism from members of the newly flourishing legal realist movement.57 Beale’s method—the derivation of fixed, nationwide common law rules—was claimed to be incompatible with modern methodologies of statutory interpretation and thus anomalous in a democracy.58 In particular, applying Beale’s rules was said to frustrate state interests, and this was not a suitable role to entrust to judges.59 Realist choice-of-law theorists accordingly focused their efforts on ridding the world of Bealean metaphysics.

History was definitely on the side of the antiformalists, as there was little that could be done to salvage the Bealean enterprise. The First Restatement’s public endorsement of “general common law” was revealing in a way that the First Restatement’s Reporter had not intended. Its articulation came, ironically, within a few years of the Supreme Court’s definitive rejection of that concept in Erie Railroad Co v Tompkins.60 The pre-Erie notion of a universal common law of contracts, torts, property, etc., has now been widely debunked as the infamous “brooding omnipresence in the sky,”61 symptomatic of an outmoded way of understanding the nature of law. References to it in the First Restatement are stark reminders of the First Restatement’s investment in an obsolete jurisprudence; Beale had put a lot of money on the wrong horse.

The volume and pace of the academic criticism that followed was so great that work soon began on a new Restatement, more attuned to the contemporary climate of legal realism.62 The conflicts revolution sparked by these critiques reshaped American choice-of-law jurisprudence over the next half century.63 If he were here today, Beale would probably not recognize the subject. All aspirants to the dominance once enjoyed by Bealean metaphysics now seem to pay homage to legal realism. Currie’s governmental interest analysis, today a mainstream choice-of-law theory, self-consciously celebrates its legal realist roots.64 Other modern theories—the Second Restatement,65 the Third Restatement of Conflict of Laws (now an American Law Institute Project),66 the “better law approach,”67 and comparative impairment68—all purport to have accepted at least the basic realist creed, and sometimes much more. Although they draw somewhat different lessons from realist jurisprudence, all claim to accept the basic tenets of legal realism; all reject the pseudo-logical metaphysics of Bealean thought.

C.    The Realist Foundations of Modern Choice-of-Law Theory

Modern choice of law is a carefully thought out and intellectually impressive analysis of the nature of law as it applies in multistate problems. Although its early proponents did not set out to do so, realism and its multistate corollary, modern choice-of-law theory, can be reduced to a relatively concise set of principles. Discussed below are four foundational assumptions, each following from the ones prior to it. The first two are germane to legal realism generally and the latter two are applications of legal realist philosophy to the choice-of-law context.

1.   Two realist premises.

Two assumptions address the holdings of legal realism generally, as it developed to deal with legal questions of all sorts; they concern the general status of legal norms. The first assumption recognizes law as an outgrowth of human purpose; the second provides that because there is no objectively correct law, the human decision-maker’s pronouncements should be treated as correct by definition. These assumptions are sharply different from the ones on which Bealean traditionalism was grounded.

a)   Created, not discovered.  The principle of legal realism that is most central to the current enterprise is the first of these: law is purposive. Law is crafted by human beings, by individual human judges and legislators, for any number of reasons.69 They are not handed down by God on tablets; they are not a “brooding omnipresence in the sky.”70 Realists believed, first and foremost, that law was decided, and not derived from some higher truth by acts of logic and science. Law is a product of human choice, designed to accomplish human purposes.71 Law, that is to say, is created and not discovered;72 it is the product of human activity and choice, formulated through the exercise of legislative or judicial discretion.73 Legal decisions represent the exercise of decision-making power; they are not metaphysical deductions from religious precept, natural law, or ineluctable logic.74

b) Definitionally correct.  The second general assumption is so close to the first that it effectively functions as a corollary. It holds that, as there is no objective standard of correctness against which one can measure a legal decision, all acts of legal authority must be treated as correct by definition.75 Because law, say the realists, is partly a product of human idiosyncrasy, it should be expected that different groups of people will make different laws.76 Law receives its authority from its genesis in a particular culture and its articulation by particular people in particular authoritative institutions.77 If law were the manifestation of universal first principles and deductions from them, then authority would depend on its correctness in application of those universal principles. But since (according to legal realism) no universal first principles exist, it should be expected that law will vary from one political community to another, and that a particular law’s authority will be a function of the exercise of decision-making discretion by that community’s decision-makers.78

The second assumption, accordingly, is that a decision-maker is by definition the final authority on rules of law.79 When a particular legal question needs to be decided in a particular political community, the authority to decide that question is vested in the decision-makers of that state.80 If two different communities are faced with apparently identical problems, they are entitled to reach different results. In the eyes of the people of one community it may appear as though the people in the other got it wrong. But since the exercise of decision-making discretion is part of the process of formulating the state’s law, every state should defer to the decisions of every other state as correct by definition.81 States are not entitled to second-guess one another’s legal decisions.

2.   Two assumptions about application.

Two additional assumptions are applications of legal realism generally to the relationship between domestic law and multistate principles relating to choice of law. These are the unity of substantive matter and choice of law, and the shared methodology of statutory construction.

a)   Unity of subject matter.  Domestic substantive law and multistate choice of law are, in the modern view, not two distinct subjects. Domestic legal principles apply to multistate problems in the same way as, and to the same degree as, they apply to domestic problems. Beale had been a choice-of-law exceptionalist—he believed that conflict of laws was intrinsically different from substantive law.82 Beale argued that the values that underlay choice of law were specific to choice of law: they included respect for sister state sovereignty, uniformity of decision-making, avoidance of forum shopping, predictability of result, and so forth.83 Modern choice-of-law theory rejects such exceptionalism, seeking to explain choice-of-law theory and to decide choice-of-law disputes in the same way that domestic substantive legal decisions are made.84

The unity of domestic substance and choice of law was important for historical reasons. The choice-of-law revolution gained momentum slowly—considerably more slowly than the acceptance of realism generally—and was not fully underway until the late 1950s.85 Realist theory was therefore already well-grounded as a general matter by the time that it was considered for application to choice-of-law disputes; the choice-of-law revolution was an extension of existing realist principles into new territory.86 The application of realism to choice of law, for this reason, was treated as a question of whether an accepted theory of domestic law might apply in the multistate context.

The modern choice-of-law theorists accordingly shaped their arguments to emphasize the essential identity of domestic and multistate legal problems. One of their central points was that choice of law and substantive law were not significantly different.87 Although legal realism was not developed with choice of law in mind, its principles were considered fully applicable to choice of law; bringing legal realism to bear on the process of selecting the applicable law did not present any unique difficulties.88 It was natural to argue that legal realism would be as applicable to choice of law as it would be to any other set of legal issues.

The first realist assumption of modern choice-of-law theory is, therefore, the assumption of unity of subject matter. It states that choice of law is part of a single unified subject matter that also includes domestic substantive law. This first application of realism to choice-of-law theory says that domestic and multistate problems are two sides of the same coin, and should be approached the same way.

b) The methodology of statutory construction.  The final realist assumption in modern choice-of-law theory is that the shared methodology is the familiar process by which statutes are interpreted.89 This is one of the central tenets of modern choice of law: choice-of-law decisions can be made through the ordinary processes of statutory construction and interpretation. It is attributable to the work of the second great figure in twentieth century conflicts of law: Professor Brainerd Currie.90

Currie’s was the first of the modern critiques of Bealean conceptualism to bear positive fruit, and by far the most important.91 Currie started with a critical account of Bealean metaphysics. Choice-of-law values of the sort that motivated Beale were (in his opinion) irrelevant distractions because the sole legitimate function of choice of law was the enforcement of substantive norms.92 The chief problem with the First Restatement’s “metaphysical apparatus,”93 wrote Currie, was that it “operate[d] to nullify state interests.”94 He went beyond this critical account, however, to develop a positive theory of “governmental interest analysis,”95 which has had enormous impact on the foundations of the field.

Currie’s governmental interest analysis bears the unmistakable stamp of modern legal realism.96 His objective was to create a system of choice of law that would further the purposes underlying the substantive rules that vied for application.97 Courts, wrote Currie, should apply their state’s law whenever doing so would serve a state interest.98 At the core of his critique and of his proposed reform was his definition of state interests as determined through the “familiar [process] of [statutory] construction or interpretation.”99 Significantly, his approach to the derivation of states’ policy interests through the regular processes of statutory interpretation comports with the realist insistence that conflicts law should be considered part and parcel of substantive law.

In Currie’s system, the necessary first step in deciding a choice-of-law dispute was always to decide which states had “interest[s]” in having their law applied.100 This allowed the court to determine whether the dispute was a so-called false conflict;101 Currie argued that there are situations in which only one state has a legitimate interest in having its law applied.102 As to these false conflict cases, he proposed that the only reasonable solution is to apply the law of the interested state.103 Currie’s identification of the false conflict was immediately lauded as a major contribution to the theory of multistate decision-making.104

Cases of the other two possible types—true conflicts and unprovided-for cases—were to be decided through the automatic application of forum law.105 The only time that it is appropriate to apply the other state’s law is in situations when a court determines that the forum state does not have a legitimate interest served by the application of its own law but the other state does.106 Because choice of the applicable law ultimately turns on whether a case is a false conflict, a true conflict, or an unprovided-for case, the initial determination of interests is the single most important step in the process. Currie’s methodological claim that interests should be identified by the usual process of statutory construction is therefore the single most important premise of his entire theory.

One of the most useful features of Currie’s writings was the sample applications of his theory that appear periodically within his more theoretical discussions.107 These practical examples closely examined factual permutations of particular cases, making what would otherwise have been a purely abstract account much more useful and accessible to the bench and bar. One of these was his discussion of Milliken v Pratt,108 a case involving a “married women’s contract statute” that prohibited married women from entering into contracts as surety for their husbands—largely on the grounds that abuses were likely given that wives were presumed to be under the excessive influence of their husbands.109

Milliken generated sixteen factual permutations because there were two states (Maine and Massachusetts) and four potentially relevant factors: the place of contracting, the domicile of the married woman, the domicile of the creditor, and the place where the action was brought.110 Currie sorted the various permutations into the categories of false conflict, true conflict, and unprovided-for case by determining which of the two states had an “interest” in having its law applied.111 These interests, claimed Currie, were generally determined by whether the application of a state’s law would provide the local party with the benefit that the state legislature had in mind in designing the statute.112 Currie in this way gave concrete meaning to the principle of determining interests through statutory construction.

Currie’s definition of state interests won him considerable attention, some of it unfavorable.113 We will argue below that the basic source of doubts about this definition lies in the two purported applications of realism to choice-of-law theory already mentioned. The first of these is the assumption that domestic law and conflicts law are not importantly different; the second is the assumption that the proper methodology for both is the familiar process of statutory construction and interpretation. These assumptions, like the choice-of-law approach that Currie purported to derive from them, are not only wrong, but—more significantly—wrong in ways that cannot be squared with contemporary realist thought. Despite Currie’s claim to have finally brought realist thought to bear on the choice-of-law process, it is the realists, more than anybody, who have good reasons to reject Currie’s analysis. We turn, next, to the task of explaining what those good reasons are.

II.  Modern Choice-of-Law Theory: A Realist Critique

Professor Currie’s theory has had an enormous impact in this nation’s courts and in the ivory tower.114 His scholarship laid the groundwork for much of the contemporary conflicts-of-law literature; indeed, his name is virtually synonymous with the transition to modern choice-of-law theory.115 To a substantial degree, the acceptance of his ideas is linked to the widespread acquiescence in his claim to have followed realist principles—in particular, his claim to have done so by bringing to bear the method of statutory construction. Without the imprimatur of modern legal realism, Currie’s method might not have attained its considerable high visibility and intellectual respectability.

A.    Critical Responses

Currie’s connection to legal realism is therefore of considerably more than historical interest. Yet it is striking that few if any of his critics have taken him on over this essential, foundational matter. The contemporary choice-of-law landscape consists of several groups: interest analysis orthodoxy (composed of the “fundamentalists” and the “loyal opposition”),116 the Third Restatement school (the “reformers”),117 and the “interest analysis deniers.”118

Most of Currie’s critics have focused on perfecting the finer points of Currie’s theory,119 leaving unchallenged the underlying theoretical claims and creating an impression of near-universal acquiescence in Currie’s realist aspirations. These members of the loyal opposition disagree mainly on questions of implementation.120 Together with the dyed-in-the-wool, uncritical fundamentalists, they form the backbone of modern interest analysis orthodoxy. A small group of free-thinking interest analysis reformers is currently forging a road to a Third Restatement with potential for greater intellectual independence. Thoroughgoing unbelievers—what might be called interest analysis deniers—are hard to find.121

1.   Arranging the deck chairs versus manning the life boats.

The academic literature on choice-of-law theory has been dominated by the loyal opposition. These are academics who, while arguing that that the underlying structure of modern choice-of-law theory is basically sound—or at least the best available—perceive some minor defects. Professor Russell Weintraub, for instance, spotted only “warts” marring an otherwise healthy body of learning: “One thing should be clear from the outset: There are warts on interest analysis, just as there are flaws in any method of resolving legal disputes. [But the] question is always whether, in the light of costs and benefits, one approach is preferable to another.122

The warts a loyal critic might spot, for example, have to do with reconciling interest analysis with the goal of moderation in interstate relations.123 Some critics saw reconciliation as surrender to weighing or balancing interests (which Currie had originally rejected).124 Or they might dispute one of Currie’s sample applications; it is possible to disagree over the proper resolution of individual disputes while accepting modern theory’s basic principles.125 To the interest analysis deniers who reject the basic enterprise, this sort of tweaking looks unnervingly like rearranging the deck chairs on the Titanic.126

Currie’s sample applications have given critics much to disagree with. For example, one of Currie’s positions that has received considerable critical attention is his insistence that states have interests in applying their law to the benefit of their own citizens.127 When Currie wrote his sample analyses of particular cases, he claimed that under the “usual processes of statutory construction and interpretation” it was obvious that pro-plaintiff laws generated interests only when there was a local plaintiff and pro-defendant laws generated interests only when there was a local defendant.128 This struck many observers as problematic, and some found it unconstitutional as a violation of the Equal Protection Clause or the Privileges and Immunities Clause.129

Other criticisms dealing more with implementation than with the basics focus on Currie’s resolution of true conflicts. While most scholars were convinced by Currie’s approach to false conflicts, in which only one state was supposedly interested, many commentators took issue with his handling of true conflict situations.130 In true conflicts, Currie argued, the law of the forum should be the rule of decision.131 He also recommended applications of forum law in unprovided-for cases, in which no state had an interest in having its law applied; this recommendation too was criticized.132

Several alternative modern theories have been developed to reflect these partial rejections of Currie’s original analysis. One such notable modern theory is Professor William Baxter’s comparative impairment.133 Baxter’s theory proposed to create overall better outcomes by resolving true conflicts in the direction that would do the least violence to the interests of other states.134 Baxter’s theory has been used by a number of recent academic treatments as grounding for a law and economics or game theory approach to choice of law.135

Professor Robert Leflar, on the other hand, proposed five nonhierarchical choice-influencing considerations which might be considered when deciding true conflicts: predictability of results, maintenance of interstate and international order, simplification of the judicial task, advancement of the forum’s governmental interests, and application of the better rule of law.136 Unsurprisingly, perhaps, most states have concluded that their own law was better than other states’. 137 They have also tended to treat Leflar’s last factor as dispositive of choice of law, discounting the others as less important.138 Leflar is an exception to the generalization that most modern scholars have been sympathetic to Currie’s methodological claim about the usual processes of statutory construction and interpretation; Leflar was skeptical about whether it was possible to ascertain anything of significance from that methodology.139 Yet his approach has been grafted onto Currie’s methodology as constituting a solution to the true conflict and unprovided-for cases.

A final, particularly noteworthy alternative modern approach is that of the Second Restatement, which “incorporated a curious mélange of academic proposals.”140 Its basic philosophy is captured by a list of considerations meant to help courts in making the central determination of which state has the “most significant relationship.”141 These include the needs of the interstate and international system; the relevant policies of the forum; the relevant policies of the other interested states and the relative interests of those states in the determination of a particular issue; the protection of justified expectations; the basic policies underlying the particular field of law; and certainty, predictability, and uniformity of result.142

A Third Restatement is now in the making; with the experience of a half century of successes and failures, the Restatement (Third) sets out to do a better job than its predecessor of meeting the needs of our multistate legal system.143 The reporter, Professor Kermit Roosevelt III, has contributed to the methodological discussion with a level of care and insight not always displayed by participants in this debate.144 A thoughtful Reporters’ Memorandum lays out the basic structure of what is called the two-step method, the first step of which is identification of what orthodox interest analysts always used to call state interests.145

Step One proposes the concept of “scope” as the replacement for state interest.146 The scope of a substantive rule is its multistate extension; it is the set of disputes to which the substantive rule was designed to apply.147 Despite the change in terminology, the result is in many respects the same: a state interest might be said to exist if and only if a case falls within the scope of a statute and making this determination involves interpretation of the statute, using normal methods of statutory interpretation.148 Although the Third Restatement seems destined to disagree with orthodox interest analysis in some important respects, we believe that the two ideas can be usefully combined, and we call the resulting requirement the “interest/scope determination.”149

What ties all these alternative proposals together is their basic acceptance of the idea that the function of courts deciding choice-of-law cases is to further state substantive policies, which are defined through a process of interpreting state substantive laws.150 This idea, however, is deeply problematic, as is revealed by consideration of the arguments of interest analysis’ more trenchant critics.151 With the partial exception of some provisions in the tentative drafts of the Third Restatement, all of the alternative modern theories that accept the basic methodology share with interest analysis its basic foundational flaws.152

Only a small number of critics have rejected the entire interest-based enterprise, a theory that (to these critics) seems as misguided as Professor Beale’s.153 These “interest analysis deniers” refuse to accept the modern approach to choice of law at its most fundamental level.154 While the loyal opposition is busy rearranging the deck chairs on the Titanic, the interest analysis deniers are rounding up the passengers and shouting orders to abandon ship.

2.   Two persistent challenges and their realist credentials.

Those in the academy who challenge modern theory on its more fundamental level have been quite skeptical about the supposed derivation of state interests through statutory construction.155 They question whether it can really be called statutory construction when the legislature had no intent about multistate applicability and when there is no mention of any choice-of-law considerations in the statute or the legislative history. Symptomatic of this skepticism are two related questions that challenge the conventional realist version of modern choice of law: (1) Can statutes or common law substantive rules actually provide all of the guidance necessary to make a choice-of-law decision? (2) What should be done when a state views its own interests differently from the modern, orthodox assumptions about when an interest exists? These are the problems of the silent statute and the other state’s interests.

a) Realism and the question of the silent statute.  A persistent thorn in the side of modern choice-of-law theory is the obvious fact that most statutes do not, on their faces, give any guidance on choice-of-law issues.156 Nor is there typically any indication of geographical extension in the legislative history.157 This has been pointed out again and again, since almost the advent of governmental interest analysis.158 In what way, ask the critics, do domestic policies tell us anything about the multistate scope of a substantive law? How can you draw choice-of-law conclusions from a statute that makes no apparent reference to choice of law? Can you spin straw into gold?

The statutory interpretation claim seems implausible on its face. The substance of the statute defines which events must be proven for substantive recovery to be warranted. But not all of these substantively relevant factors are significant for choice-of-law purposes.159 Typically, a court identifies a single one of these substantive connections as the trigger factor, which points the court toward the state whose contacts with the dispute dominate the others for selection of the applicable law.160 Relevance for substantive purposes, which many events, people, and property possess, is different from relevance for choice-of-law purposes—which is typically a characteristic of only one factor. Choice-of-law relevance is something extra, something beyond substantive relevance; not every substantively relevant event can have it.161 It is the job of choice-of-law analysis—supplied by judges, scholars, and other experts—to identify what this something extra is. But in most cases, there is nothing in the statute that helps to distinguish one substantively relevant factor from another for purposes of designating the applicable law.162

Proponents of the statutory interpretation theory do not explain how judges are to carry out their assignment, but simply argue that it must be possible to do so in the choice-of-law context because judges do so all the time in domestic cases.163 Professor Russell Weintraub writes, for example, that “[w]hen choice-of-law analysis focuses on the reasons underlying putatively conflicting domestic rules, it simply mirrors the form of intelligent analysis employed in all fields of law.”164 Interest analyst Professor Larry Kramer labels the problem of what to do when the statute does not literally address the question “a common problem of unforeseen or uncontemplated circumstances,” adding that “it is black letter law that such problems can be resolved by ascertaining the statute’s purpose and extrapolating from that purpose to the particular question.”165

In domestic cases, it is said, judges are able to decide cases in the face of seeming legislative silence; in choice-of-law cases judges should be able to do the same.166 The answer to the choice-of-law question (apparently) is really in there—in the statute and its underlying policies; the judge need only look. All that judges need to do to determine whether the state has an interest is to follow their usual game plan. In domestic cases (it is argued) judges spin straw into gold as a regular matter; they should be able to spin straw into gold just as easily in choice-of-law disputes.

The analogy to domestic statutory interpretation, however, proves exactly the opposite of what the modern theorists have in mind. Judges are successful at resolving domestic cases because they assume that when there is no answer in the statute, they are empowered to consider values, empirical facts, and policy concerns that are extrinsic but relevant to the statute.167 The fact that a judge must decide a case even when a statute is silent on the matter does not show that the answer was really in the statute all along. What it shows is that we recognize the authority of judges to simply do the best they can, including by taking into account extrinsic considerations when necessary.168 Put simply—and unsurprisingly—there are times when judges do not find the sought-after answer in the text of the statute and instead must make law.

Consider a judge facing a novel torts issue: Are manufacturers of self-driving cars automatically responsible for defects in the cars’ computer programs under a theory strict liability, or must negligence be shown?169 There may be statutes dealing with design defects in the state where the case arises, but the precise problem is one that the legal decision-makers of yesteryear are unlikely to have confronted and addressed. If upon exhausting all authoritative sources (statutes, binding precedents, etc.) the judge finds no answer, then she is likely to resolve the issue by creating new law that is compatible with existing authority but also influenced by extrinsic materials that are not themselves legally binding. In the case of self-driving cars, it is easy to imagine what sorts of sources these might be.

Legal authorities write scholarly articles making recommendations about how to deal with autonomous vehicles;170 engineering and economics professors contribute what they have to say;171 and even psychologists and philosophers might offer their expert opinions.172 Empirical studies, statistical analysis of various kinds of experimental evidence, and historical investigation of how analogous problems have been dealt with in the past might all be relevant.173 The judge does indeed resolve the problem, but the fact that this problem got resolved does not mean that the resolution was drawn from any instructions contained in a statute.

One of the great revelations of the legal realist movement was the daunting number of questions upon which existing law is truly silent. Professor H.L.A. Hart captured this phenomenon with his famous reference to the “open texture” of the law.174 When a statute is ambiguous or doesn’t mention some particular issue, part of what a court must do is “determine whether the policies underlying [the law] support its application to the particular facts of the case.”175 But when legislation does not supply an answer, and when there is nothing else authoritative (for example, a constitutional provision) addressing the question, the judge must nevertheless make a decision. When there is a gap in a domestic statute, judges fill that gap.176 The fact that the judge in a domestic case was able successfully to fill an apparent gap does not mean that the material that filled the gap was found in the statute or the legislative history. And the process by which the solution was reached can be characterized as statutory construction only if that phrase is understood expansively, to include decisions in accordance with additional nonstatutory sources.177

Similarly, judges deciding choice-of-law disputes will sometimes—most of the time, perhaps—fill the legislative silence with something other than statutory language or legislative history. Academics write articles theorizing about the best way to select the law to be applied in large part because there are gaps in existing law to fill; it is for this same reason that judges consider choice-of-law values such as uniformity and predictability, and advocates include moral arguments, legislative facts, and economic or sociological materials in their briefs. These writings largely deal with the question of what to do in the face of legislative silence.

Gaps, indeed, are far more likely to occur in the choice-of-law context than in the domestic substantive context—legislatures rarely act with multistate scope considerations in mind. You can call the enterprise in question “statutory construction and interpretation” or anything else that you want to; the label is not what matters. But finding an answer requires injecting into the analysis some measure of community values, empirical evidence that may not have been available to the legislature, extrinsic concerns about unintended consequences, and other sorts of considerations that may have played no role in the original enactment by the legislature. And these can include choice-of-law values such as avoiding intrusions into other states’ sovereignty,178 minimizing forum shopping through the pursuit of decisional uniformity,179 maximizing predictability,180 or anything else that the state’s authoritative decision-makers think appropriate to take into account.

When a statute is silent on its face, various explanations are possible. Perhaps the legislature chose not to take a position on the issue in question.181 In some cases, the legislature may simply have been unable to reach agreement on the particular question and chose to leave the matter unresolved rather than try to fight it out.182 Or the legislature might have chosen to delegate the matter to the courts or agencies because at the time of the adoption of the statute there was not enough empirical evidence on key issues. Or perhaps it simply failed to notice the particular question. Interpreting the statute does not necessarily mean finding a definitive answer in the statute. There may be no answer in the statute, and the judge must simply find a way to cope. The position of modern choice-of-law theory—that there is necessarily an answer somewhere in the statute—effectively assumes that even if the legislature wants to leave a question to the courts, there is no way for it to do so.

The astute reader may have noticed that little of any real use has been said to this point about what statutory interpretation actually means. Perhaps this is not surprising; no particular definition currently enjoys universal (or even near-universal) approval in general usage. There are probably as many theories of statutory interpretation as there are choice-of-law theories.183 As one well-established authority on domestic statutory construction puts it, “American courts have no intelligible, generally accepted, and consistently applied theory of statutory interpretation.”184 The confusion and uncertainty about how to use the term in the present context certainly count as supporting evidence for this proposition.

This is a problem to which we will return.185 What matters for present purposes is that instructions to apply the usual methods of statutory construction are of no practical use whatsoever.186 One might as well tell a biochemist working in a medical research laboratory to find a cure for cancer by applying the usual scientific method. While certainly it would be unwise for the biochemist to conduct research in a manner inconsistent with this recommendation, such advice is entirely unhelpful as a way of advancing cancer research.

b) Realism and the question of the other state’s interests.  Modern choice-of-law theory assumes that even when the statute is silent, the ordinary process of drawing inferences from legislative purpose, if applied with sufficient determination, will give a definitive answer. This is certainly true in some cases, but not necessarily in all. Moreover, it cannot simply be taken as an article of faith; the amount of guidance that a statute provides can only be determined on a case-by-case basis, by examining the individual statute. Judges fill gaps whenever statutes and judicial precedents do not answer the legal problem of the parties before them.187 Making state policy—including through the exercise of judicial discretion—is part of their job.

The possibility of gaps is troublesome, however, when a court finds itself construing the statute of another state; making state policy for other states is not part of their job.188 Modern approaches to choice of law require a court to determine the multistate extension of foreign law in every choice-of-law dispute; a judge is obliged to decide both the interests of the forum and the interests of the other states in question.189 According to modern choice-of-law theory, the forum court examines the interests of the other state de novo—that is, without according any deference to the other state’s own views on the matter. Yet the rationale for judicial creativity applies only when a state interprets its own law or determines its own interests. A state court has no mandate to revisit other states’ laws and arrive at its own conclusions about the interests they reveal.

A judge’s treatment of local law is necessarily different from a judge’s treatment of foreign law. Statutory construction means something different when the court is construing its own state’s statutes than when it is construing the statutes of another state. The court has lawmaking power only for its own state’s law.190 This is as true when the question is whether a state has an interest in the choice-of-law sense as when the issue involves a disputed question of substantive law. Judges possess a greater degree of decision-making authority in cases that are wholly domestic to the forum than in cases wholly domestic to a foreign state. And if determination of interests is a question of statutory construction, then no state has authority to determine the interests of another.191

The difference between the process of applying local law and the process of applying foreign law is particularly apparent when the cause of action was judicially created. A common law cause of action arising under forum law can be treated with much greater freedom of judgment than a common law cause of action arising under the law of another state.192 In the former context, the judge can reinterpret, restrict, overrule, or expand the cause of action; in the latter context, the judge has no such power. As with statutory causes of action, this is true regardless of whether the case raises choice-of-law issues or purely substantive issues. A judge’s discretion in determining the multistate extension—the interest or scope—of local law is more considerable than the judge’s discretion in determining the multistate extension of foreign law.193 This phenomenon has been noted frequently in analogous jurisdictional situations, such as when a federal judge is charged with applying the law of the state in which she sits.194

While requiring the judge to carry out both of these different functions, however, modern choice-of-law theory provides only a single set of analytical tools: the usual processes of statutory construction and interpretation.195 Something is missing. If there are two situations that involve different theories of interpretation, they cannot both be explained simply as the usual processes of statutory construction and interpretation. While giving central importance to the methodology of statutory construction, modern choice-of-law theory has almost nothing to say about what that means—including the undeniable fact that it seems to mean different things for local and foreign law. The concept of statutory construction is treated as simple and obvious, when in reality the concept is anything but.

B.    Closet Textualists?

The pattern that emerges from these two persistent challenges is, at first, perplexing. Consider at the outset how things actually work. In the real world of American legal practice, domestic interpretation and multistate interpretation are quite different from one another. In the real world, the domestic process of interpretation is by far the more creative of the two decision processes. The court is a policymaker for its own state and is empowered to create new law, especially when doing so is necessary to fill gaps. Such gaps are assumed to be common occurrences; in filling them, judges in the real world take into account community values, empirical evidence, historical practice, and so forth. When applying another state’s law, however, judges are supposed to exercise no creative judgment, but rather implement mechanically whatever a relevant state court has decided to do with the issue.

Consider, in contrast, the alternative universe of modern American choice-of-law theory. In the largely counterfactual world of modern American choice-of-law theory, judges use the same method for interpreting their own state’s interests as they do for the other state’s interests. Judicial creativity is discouraged for both sorts of interpretation equally because in both contexts it is conclusively presumed that there are answers tucked away somewhere in the statute, patiently waiting for the judge to find them. Choice-of-law interests (according to modern choice-of-law theory) are identified without reference to choice-of-law values, and decisions that rely on choice-of-law values must be discounted because courts are never supposed to consult anything but the statute’s substantive provisions.

It is worth defining with more particularity what we mean by strict textualism in this context. The prescription of textualist theories generally is that courts ought to apply the plain meaning of a statute if one is availing.196 Over the last several decades, a new textualism has developed that even more strictly cleaves to the text of a statute, rejecting the use of other sources—primarily legislative history—in interpretation.197 More recently, a newer strand of textualism has emerged that eschews even the traditional role for canons of statutory interpretation.198 This newest development is likely responsive to the observation of some textualist commentators that judges use these canons as vehicles in which they can smuggle their own values to impermissibly defeat the primacy of text.199 It is this last, most extreme form of textualism that we refer to here as strict textualism.

What is perplexing about this pattern is how the modern American choice-of-law theorists have managed to sell the idea that they are the realists. The world of modern American choice-of-law theory is, in actuality, unreal. This is true in at least three respects: (1) whether domestic and multistate interpretation are the same; (2) whether there are gaps in what the statute’s substantive provisions, standing alone, can answer; and (3) whether judicial creativity can be brought to bear to resolve these gaps. The remarkable fact is that, in these respects, modern American choice-of-law theory more nearly exemplifies the theory of statutory interpretation commonly called strict textualism than it does legal realism.

Strict textualism is the theory of interpretation best positioned to explain one particular aspect of modern choice-of-law theory: the commitment to substantive values taken from the statute, and nothing but substantive values taken from the statute.200 The resemblance between the modern theorists’ methodological position and strict textualism lies in the insistence that deference to the legislature (the principle of legislative supremacy) requires that judges not exercise any creative judgment. Currie’s designation of statutory interpretation as the official methodology suggests an intention to honor legislative wishes, combined with a commitment to honestly putting aside one’s own personal values in favor of the decisions of the elected branches.201 From this perspective, the motivating principle underlying the limitation to substantive policies seems to be some form of antipathy toward judicial activism. Currie’s periodic references to the role of courts in a democracy support this interpretation.202

The resemblance between the principles motivating interest analysis and the principles animating strict textualism is striking. But whether such textualist tendencies will (or should) continue to be a prominent theme in the choice-of-law realism of the future is another question entirely. The resemblance is unlikely to contribute to modern choice-of-law theory in any positive way; the parallels are, in fact, a reason for modern theorists’ concern. For at least three reasons, the analogy between modern choice-of-law theory and strict textualism creates more problems for the modern theory than it solves.

The first reason concerns whether strict textualism is compatible with legal realism. Textualism is ordinarily thought to be quite different from purposivism, an important underlying principle of the realist movement.203 Textualism is associated with conceptual approaches to legal authority, because it treats texts as having intrinsic and immutable meaning independent of modern social and political context. At the very least, strict textualism and modern choice-of-law theory make strange bedfellows.

Second, strict textualism is not commonly accepted, while the pretension to general methodological acceptance is an important part of the modern theory’s appeal.204 By stressing the familiar methods of statutory interpretation, Currie seemed to be referring to methods that were already widely used domestically.205 The modern choice-of-law theorists’ enthusiasm for their methodology seems to be based at least in part on an assumption that it is accepted unproblematically throughout the American legal system.206 Surely it is perfectly unobjectionable (they assert) to use a method that is taken for granted on a day by day basis in domestic law.207 Doesn’t the general acceptance of domestic decision methods confirm that interpretation of a statute need not be based entirely on the statute’s express language or on something tangible found in the historical record of its enactment? This sort of argument is severely undercut if the recommended model of statutory interpretation is actually controversial—let alone if the model recommended for choice of law is actually a minority view on its home turf.

Now, of course, one might well be skeptical of whether common acceptance should be accorded much importance. All kinds of things go on in domestic courts, some of which are perfectly fine and others of which are not. For years our courts cheerfully tolerated eugenics and racial segregation.208 Closer to home, the formalist conceptualism employed by Beale would still pass the test of judicial acceptance today in a lot of places—with flying colors. Some states, after all, still use the First Restatement.209

But the general unpopularity of strict textualism is important here. Choice-of-law theorists are unlikely to predict that strict textualism is about to be adopted for general use in statutory interpretation in domestic cases. To adopt it for choice-of-law purposes would create an unexplained (and probably unexplainable) discontinuity between domestic and multistate decision-making.

Finally, and ironically, modern choice-of-law theory itself could hardly qualify under a strict textualist approach. Paradoxically, Currie’s own examples provide the clearest evidence of reliance on extrastatutory standards for whether an interest exists.210 The examples that Currie provided to illustrate the application of governmental interest analysis all seem to depend on assumptions that have no textual support.211 For example, Currie’s writing contains numerous examples of how the theory was to be applied in practice, typically with explicit indications about what legislative policy supposedly consisted of.212 When states reached results that were inconsistent with what Currie viewed as their interests, Currie did not hesitate to denounce them. He chastised not only local courts but also legislatures for contemplating any definition of interests other than the one he proposed.213 His followers have mostly followed suit.214

In sum, the resemblance between modern choice-of-law theory and textualism should be disconcerting for anyone committed to a purposive understanding of the nature of law. Strict textualism has, if anything, less appeal in choice of law than in its domestic applications. There is no reason to transplant it into choice of law when (from the point of view of the average observer) it has contributed so little to the environment it currently inhabits. The future of choice of law lies in the opposite direction.

We turn our attention, accordingly, to the general question of what should be done to redirect choice of law down a more promising path. A small group of revisionists has been laying out alternatives. With all of the problems already identified, realism still has a future for conflicts of law.

III.  A Realist Future for Choice of Law

Interest analysis is at a crossroads. Its orthodox followers keep the traditional faith. They insist that Bealean choice-of-law methods be rejected because they interfere with furtherance of legislative purposes, but they then insert as replacements methods of their own that have little or no demonstrated connection to legislative purposes either.215 Through this process, definitions of state interest come to reflect the values of academics rather than the values of the community and its decision-makers.

This hijacking of the determination of state interests is deeply problematic in a school of thought dominated by self-declared legal realists.216 Professor David Cavers cautioned his readers about the potential reification of Professor Currie’s concepts, warning of the misleading “air of substantiality” exuded by the term “interests.”217 But this is precisely what interests have become: “externally determined and supposedly objective concept[s]” that are “imposed on state legislatures and state judges by scholars.”218

A small number of more sophisticated interest analysts have recognized the incompatibility between orthodox notions of interest and fundamental principles of legal realism. They recognize the need to choose between (1) an objective theory of interests conforming to orthodox notions of benefitting locals that requires rejection of contrary state policies and (2) a subjective theory of interests that reflects actual legislative wishes and that in at least certain circumstances takes actual state choice-of-law preferences and values into account. These new modern theorists diverge from orthodox interest analysis by choosing the latter of these alternatives. Because the draft provisions of the Third Restatement make the same election, this school of thought might be called the Restatement (Third) School of subjective interests.219 Although ultimately not completely successful, these fresh and unorthodox modern theorists aim to return choice-of-law theory to its intellectual roots in real, authentic realism.

The final Part of this Article first identifies what changes modern choice-of-law theory needs in order to make real realism possible. It assesses the successes and failures of the Restatement (Third) School—those more sophisticated new realists who have developed their own brand of subjective interest theory. And a new approach is proposed. In the context of conflict of laws, realism must be compatible with basic principles relating to jurisdiction: it must be jurisdictional realism. This means that the theory must be able to account for, and work within, an environment in which many different independent voices make simultaneous appearances.220 The Article then sketches out some countervailing concerns that skeptical readers might be entertaining about these suggested changes and addresses them.221

A.    Real Realism

From a realist point of view, the most theoretically defensible position on questions of interest and scope would be for all states and all academic observers to accept as authoritative whatever the other state says about its own interests—a subjective definition.222 Orthodox interest analysts have hesitated to embrace such an approach.223 It opens the door to interests defined in terms of vested rights, territoriality, or anything else that the state might happen to find convincing. Generally speaking, therefore, scholars who adhere most closely to the teachings of Currie have embraced an objective view of choice-of-law policies,224 in which “interest analysts treat what they see as a suboptimal definition of interests as entitled to no credibility at all.”225

Under objective approaches, scholars examine jurisprudence and propound factors that they believe should animate the decisions of judges, or that they believe tend to make good policy.226 Determination of state interests is de novo, without crediting any prior determination by the state whose interests are at stake. By expounding a universal system of considerations for use by forum courts in analyzing the interests of other states, these modern scholars have stumbled into the very Bealean trap that they so criticized—it’s just that they mandate a network of nationwide policies rather than a nationwide network of rules.227 This feature of Currie’s brand of realism must be addressed if realism is to have a future in conflict of laws.

The difference that the substitution of subjective for objective interests would make is substantial. Substituting subjective for objective interests does not change the basic framework of the analysis; there are still the equivalents of false conflicts, true conflicts, and the like. However, the placement of a particular case into one of these categories will be different if subjective interests are used, rather than if objective interests are used. A dispute that Currie would consider to be a true conflict might instead be perceived as a false conflict, or it might turn into an unprovided-for case.

Take for example a products liability dispute in which the parties are both from State Alpha, which has a pro-recovery law. The injury occurred in State Beta, which would not allow recovery. Assume that State Beta applies the First Restatement of Conflicts and that all of its precedents involving local State Beta accidents accordingly hold that the applicable law should be State Beta law. Assume also that State Alpha applies orthodox interest analysis, and that it is the forum. Under orthodox interest analysis, the case is a false conflict and State Alpha law should apply regardless of where suit is brought. Under a subjective theory of interest, the case is a true conflict because Alpha is the place of common domicile but Beta is the place of the accident.

For purposes of comparing the consequences of the objective and subjective conceptions of interest, we sketch a framework for choice-of-law decision-making in which the only difference between the two points of comparison is whether subjective or objective definitions are used. The framework deals with the first step in what have come to be known as two-step models—that is, choice-of-law approaches in which the court first determines the existence of an interest in the various contending states and thereafter, in the second step, decides how to resolve the dispute based on the interests it has identified.

Figure 1 is a two-by-two matrix representing decision-making under orthodox interest analysis. Figure 1 does not take account of what the two states’ choice-of-law rules say; interests are determined objectively. The upper-left-hand cell (Cell 1) holds all fact patterns in which neither state has an interest; these are the unprovided-for cases.228 Under interest analysis, these would be cases in which the defendant is from State Alpha and the plaintiff is from State Beta.229 In Cell 2 we find cases in which State Alpha has an interest and State Beta does not. Cell 2 includes cases in which both parties are from State Alpha; these are false conflicts.230 Cell 3 contains true conflicts—that is, disputes in which both states have interests in the application of their law. Under interest analysis, these would be cases in which the defendant is from State Beta and the plaintiff is from State Alpha. In Cell 4 we find cases in which State Beta has an interest and State Alpha does not. As in Cell 2, the Cell 4 parties are from the same state; however, in Cell 4 the common domicile is State Beta. Cell 4 cases are all false conflicts in which the law of State Beta applies.231

Figure 1


No Interest of

State Alpha

Interest of

State Alpha

No Interest of 

State Beta

  1. Unprovided-for case
  1. False conflict: State Alpha law applies

Interest of State Beta

  1. False conflict: State Beta law applies
  1. True conflict


When an analogous matrix is employed with a subjective definition of interest/scope, the result is Figure 2.232 In Figure 2, State Alpha is once more the forum. We assume that State Alpha can determine its interests or the scope of its law by directly consulting local statutes, policies, norms, etc. As the forum, State Alpha’s exercise of this power is methodologically unproblematic because it has full authority to define its interests as it wishes. The left-hand column in the matrix represents all cases in which State Alpha comes to the conclusion that the particular fact pattern at issue does not fall within its interests/scope. The right-hand column, conversely, includes all cases in which State Alpha comes to the opposite conclusion.

When it comes to the position of State Beta, in contrast, methodological problems arise. State Alpha (the forum) does not always have recourse to all of the relevant information regarding State Beta. The best it may be able to determine is whether State Beta would apply its own law. This it determines from examining choice-of-law decisions in State Beta; it looks for the closest precedents in the State Beta case law and presumes that State Beta would follow its own established judicial authority. If the State Alpha court sees that State Beta would apply its own law, this supports the conclusion that State Beta has an interest/scope (subjectively speaking). If State Beta would not, then this indicates either that State Beta has no interest or that it would subordinate its interests in the present cases to the needs of the interstate system.


Figure 2


No Interest/Scope of State Alpha

Interest/Scope of

State Alpha

State Beta would not apply its own law

  1. State Alpha has no interest; State Beta either does not or is refraining out of deference.
  1. State Alpha has an interest; State Beta either does not or is refraining out of deference.

State Beta would apply its own law

  1. State Alpha has no interest; State Beta has an interest and would apply its own law.
  1. State Alpha has an interest; State Beta also has an interest plus it would apply its own law.


Aside from the substitution of subjective for objective interests, the basic structure of the analysis is parallel to the structure of the analysis in Figure 1.233 Cells 2 and 4 are analogues to Currie’s false conflicts,234 and Cells 1 and 3 are analogous to unprovided-for cases and true conflicts, respectively.235 As with standard interest analysis, Cells 2 and 4 contain the easy cases. Cells 1 and 3 present greater difficulty, but the problem does not arise out of whether the method employs a subjective or objective theory of interests/scope. Rather, it arises from the question of what to do when the preferences of the various states do not mesh neatly together. The conflict is intrinsic and present regardless of whether a subjective or objective theory of interests/scope is adopted.

The resolution of individual cases is different in Figure 1 and Figure 2, because a particular fact pattern may be a true conflict under a subjective theory and a false conflict under an objective theory, or the converse. This parallelism is nevertheless important because, as has frequently been pointed out, one of the great attractions of interest analysis is its identification and resolution of false conflicts.236 These are win-win cases in which both states will be satisfied with the resolution. It turns out that false conflicts are, in effect, a special case of a more general principle. The general principle is that if the other state’s choice-of-law rules lead to a result that is compatible with the forum’s determination of its own interests, then there can be no objection to application of the law so identified. At the underlying structural level, the objective and subjective approaches are quite similar, even though they would result in quite different results in a large number of cases.

Interest analysis and the modern theories derived from it have not been the trusty disciples of legal realism that they claimed to be. But modern choice-of-law theory’s lapses do not necessarily compel the conclusion that choice of law is not suit-able realist terrain. What they suggest, instead, is that applying realism to choice of law requires some awareness of the special characteristics of the multistate situation. It is possible to avoid at least some of the mistakes that the modern theorists made, simply by keeping in mind the basic realist objectives and what they mean for choice of law.

The chief mistake that modern choice-of-law theorists make is to impose their theory’s own particular value judgments on the states, rather than recognizing the states’ authority to define their interests on their own.237 There are, in effect, two ways in which this usually happens. The first consists of the placement of arbitrary limitations on the values that may be taken into account in making decisions about the state’s own interests or the multistate scope of its own laws. It is related to the problem of the silent statute. The second consists of failure to respect the other state’s choice-of-law decisions and the interests that they rely on. It is the problem of the other state’s interests. Both of these types of mistake are traceable to problems with the assumptions that the early conflicts realists made in applying legal realism to the subject matter of choice of law.

1.   The problem of the silent statute.

The problem of the silent statute does not and should not arise in an authentic realist approach to choice-of-law decision-making. For realism, it is not a problem that a particular statute does not contain the answer to questions of the appropriate scope of multistate application. The realist recognizes the judge’s individual role in the decision of particular cases and the formulation of generalized legal norms.238 Since there is no expectation that every domestic substantive statute will provide the specifications for its own choice-of-law boundary—any more than there is an expectation that the answers to other questions will always be provided in the statute—the silent statute is not an embarrassment.239 The judge simply decides the question of multistate extension by taking into account any existing norms or policies that may be relevant to the problem at hand, together with any appropriate evidence that the parties have brought to his or her attention.240

The silent statute problem in modern theory arises because of the artificial and unsustainable assumption that “the usual processes of statutory construction and interpretation” leave no room for consideration of values distinctive to the choice-of-law process.241 In making a decision about multistate applicability, it is claimed, domestic substantive policies are the only ones that should be taken into account; they are the only genuine policy preferences.242 In determining its interests, the state is supposed to disregard everything other than the simple, narrow concern that the modern theorists thought appropriate—for example, whether the party who would benefit from the application of local law is a local person.243 State Alpha’s courts are not to consider, for example, whether State Alpha might prefer to limit the application of local law to situations in which the application of local law would not unfairly surprise the out-of-state party, or to avoid application of local law when that would disrespect the territorial sovereignty of other states.244

This artificial limitation to domestic substantive policy considerations is completely unnecessary. Consider the following line of reasoning:

1. Gaps in substantive statutes or precedents should be filled by considering the policies underlying that substantive law.

2. Therefore, gaps in choice-of-law statutes or precedents should be filled by considering the policies underlying the substantive law.

Modern theory starts with Point 1 and treats it as leading inexorably to Point 2. But the more appropriate line of reasoning runs not to Point 2 but to Point 3:

1. Gaps in substantive statutes or precedents should be filled by considering the policies underlying that substantive law.

3. Gaps in choice-of-law statutes or precedents should be filled by considering the policies underlying the principles of choice of law.

Parallelism would require that choice-of-law gaps and ambiguities be resolved first and foremost by reference to choice-of-lawvalues and policies, not substantive values and policies; this would be the best way to carry forward the insights of legal realism.245

This way of integrating substantive considerations with choice-of-law considerations fits a familiar pattern. Compare, for example, the way that statutes of limitations and the merits of a cause of action fit together. A necessary requirement for the plaintiff to recover is that the statute of limitations has not run.246 That requirement is necessary for recovery in the same way that the merits of the case are necessary requirements for recovery. One might even say that the same methodological principles apply to both substantive issues and statutes of limitations—that is, the usual processes of statutory construction and interpretation. But if there is a gap or ambiguity in the statute of limitations, one does not immediately and exclusively try to fill that gap with inferences from the substantive rule of law involved in the case. It would be strange to insist that a definitive answer about the limitations period should always be discernible from studying the substantive law. Instead, resolution of competing claims about statutes of limitations requires consideration of policies germane to statutes of limitations. The logic moves from Point 1 to Point 4:

1. Gaps in substantive statutes or precedents should be filled by considering the policies underlying that substantive law.

4. Gaps in choice-of-law statutes or precedents should be filled by considering the policies underlying the reasons for adopting statutes of limitations.

This is not to say that the decision-maker should ignore arguments based on the substance of the cause of action. When such arguments provide guidance they should be considered. But there is no reason to assume that these are the only relevant considerations, or that standing by themselves substantive policy arguments will produce a definitive answer. The same could be said for other procedural issues such as the requirements for bringing a class action, venue requirements, or pleading requirements. Substantive law is not necessarily the best source of guidance for resolving ambiguities in such procedural rules.

In this respect, statutes of limitations are closely analogous to choice-of-law rules. Choice of law places limits on the cause of action’s spatial existence, just as statutes of limitations place limits on its temporal existence. This does not mean that choice of law is an impermissible intrusion on the substantive law, any more than it means that statutes of limitations are. Choice-of-law values are like the values promoted by the statute of limitations. There is no reason not to reflect them in the decision-making process.

2.   The problem of the other state’s interests.

The second persistent challenge is the problem of the other state’s interests.247 A court in State Alpha that is charged with deciding the applicable law must, under orthodox interest analysis, determine both its own (State Alpha) interests and also the interests of the other states (States Beta, Gamma, etc.).248 When it comes time to decide the interests of the other state whose law is vying for application, the State Alpha court (according to orthodox modern theory) is not to take into account that other state’s definition of its own interests.249 Under interest analysis, a determination by State Alpha that its interests are or are not implicated in some particular case is simply not authoritative.250

This problem and the problem of the silent statute are obviously related. As with the problem of the silent statute, an academic value judgment251—not authoritative politically, but justified on conceptual grounds—is used to answer a question that should instead be addressed by legislative preference. The effect, potentially, is to override the preferences of the politically authoritative decision-maker in favor of a theoretical construct with no pedigree as law. There are any number of ways that such a result could be explained or rationalized, but legal realism is not one of them.

As with the problem of the silent statute, the difficulty starts with incorrect assumptions in modern theory’s translation of basic realist principles into choice-of-law application. The problem here is with the first assumption about application (unified subject matter): it overlooks the salient differences between the domestic context and the jurisdictional context. This premise was a direct consequence of the claim that realism could and should be extended to cover choice of law, as well as other subjects. No effort was made to identify the differences between multistate and domestic problems; no effort was made to accommodate realism to the different situation that choice of law presented. However, these differences in situation must be identified if progress is to be made toward a distinctly jurisdictional form of realism.

3.   Toward a distinctly jurisdictional realism.

In both of these respects, the fault lies with the general assumptions that the early conflict-of-laws writers made about legal realism and its application to choice of law. It was assumed that realism could be applied straightforwardly, with the basic principles invariant, without the need for any adjustments to satisfy special circumstances of the multistate context.252 For this reason, the emphasis was on the unified subject matter and methodology, and on the (supposed) lack of basis for any different treatment. The two assumptions about application—unity of subject matter and methodology of statutory construction—deny any need to translate realism from one context to another.253 There was, supposedly, only one context.

a) The multiplicity of voices.  Choice of law is, however, different from other areas of law. What realists must take into account in the multistate context is that there are two or more legal decision-makers, each with an authoritative, independent voice.254 Because the voices are independent, it is only by happenstance that they reach agreement; they could have disagreed. In the domestic context, there is potential for disagreement because the answers that are given to particular legal disputes are contingent.255 But the presence of multiple voices generates the capacity to transform a situation of contingency—in which there is genuine potential for variability—into one of actual disagreement.

The distinctiveness of choice of law lies in the presence of these multiple voices. When there is only one voice, what it says is definitionally true. So long as only one voice is in operation, there is no chance of contradiction. But when one entity is talking about the interests, policies, or preferences of a second, the first is not authoritative. In a context of multiple independent voices, care is required in order to avoid taking one of these voices as authoritative for the other.

The significance of the presence of multiple voices is profound. It vastly complicates the assumption that legal decisions are, by definition, correct. The decisions that an authoritative legal entity arrives at on its own behalf are true because they were chosen by that entity; they are not true because of consistency with some objective standard but rather because they are treated, by definition, as valid choices.256 But this cannot be said when one state attempts to take account of the interests, preferences, or beliefs of other states. When one state is charged with making the choice between its preferences and another’s, the first cannot be assigned the responsibility of determining what the second wishes to see happen.

The central paradox of jurisdictional realism consists of the inability, in principle, to take into account something that one cannot identify or define. The difficulty arises whenever there is more than one voice in a system, and one party must make decisions based on what it considers the other party’s interests or preferences to be. It is thus endemic to modern choice-of-law theory, according to which the forum is called upon routinely to determine the other state’s interests. The central mistake of modern choice-of-law theory is to treat this determination as a matter of objective fact—of right and wrong. This is a mistake that can be corrected.

Jurisdictional realism is built upon the assumption that jurisdictional problems are distinctive and call for some accommodation of realist foundations. Modern choice-of-law theory has never squarely addressed that need. But the basic structure of modern choice-of-law theory can be adapted to meet the objections raised above and to conform more faithfully to realist principles. The essential alteration would be a substitution of a subjective definition of state interests or scope for the currently dominant objective one.257

Realism counsels state respect for the differing opinions of sister states; under realism, each state realizes that the choices of its sister states are for those sister states to make.258 This is true even when its sisters base their decisions on disfavored premises: old-fashioned choice-of-law values, for example, or the belief in vested rights. What ought to be done to address the problems of modern theories seems clear. In assessing state interests or the scope of a state substantive law, the state’s subjective position should be determinative. Interests and scope do not have independent objective existence. Taking this into account requires openness to one another’s choice-of-law rules. A state’s choice-of-law decisions can be as informative about its preferences as any other decisions that it makes, even when they reflect traditional values.

b) The new Draft Restatement’s solution.  Probably the most promising avenue for progressive development of choice of law at current writing is the American Law Institute’s project on Conflict of Laws. As noted earlier, the Restatement’s drafters self-consciously set out to learn from the half century of mistakes by other modern choice-of-law theories.259 One subject that it has considered in some depth is the objective and subjective definitions of interests or scope.260

The Draft Restatement (Third) recognizes the arguments for using subjective interests/scope unreservedly in principle. But it is of two minds when the time comes for putting them into practice. Symptomatic of both its unconditional recognition in theory and its simultaneous ambivalence in practice is the treatment it recommends for state choice-of-law decisions.

The Draft Restatement bases its recognition of subjective interests/scope on the assumption that each state decides for itself what fact patterns fall within the reach of its substantive laws. Determinations of scope are in essence just interpretations of a substantive statute, and these are indisputably subject to the state’s lawmaking power. The Reporters’ Memorandum accordingly observes, “The Third Restatement takes the view that interests are subjective, that is, that the scope of a state’s law is a question of the substance and meaning of that law and, subject to constitutional constraints, is within the authority of that state’s lawmakers.”261 The consequence of the subjective view is that a state’s determinations of the scope of its laws are authoritative:

In general, whether a state’s law attaches legal consequences to certain transactions or events is deemed to be a question under that state’s law, with respect to which the courts and legislature of that state are authoritative. . . . The consequence of this view is that when a state expressly determines the scope of its law, other states must respect that determination.262

However, the Draft Restatement limits the application of this insight substantially. For reasons of simplicity of application, for example, the Restatement categorically disqualifies judicial precedent from having any effect in determining scope.263 It asserts that only statutory specifications of scope are authoritative. The Reporters’ Memorandum states:

Instructing courts to look to other states’ choice-of-law rules to determine the scope of their law would therefore impose a difficult task and depart substantially from existing practice. With respect to statutory specification of scope, however, the task is easier and practice appears to be relatively uniform: courts do follow the limits announced in other states’ statutes.264

This proviso excludes all judicially developed choice-of-law principles, generally. That means, in particular, that choice-of-law rules cannot be brought to bear in determining a substantive norm’s geographical scope, and thus its entitlement to respect from other states as a subjective expression of interest.

The Draft Restatement confirms this exclusion in so many words. According to the Reporters’ Memorandum, for a choice-of-law determination to be authoritative in other states as a subjective expression of a state interest, it must be derived from a statute and must “expressly determine[ ] the scope.”265 The Reporters’ Memorandum confirms that “[a] choice-of-law rule (such as territorialism, or interest analysis, or the Neumeier rules) does not expressly determine the scope of state law.”266 The basis for the Draft Restatement’s conclusions about which things “expressly determine scope” and which do not is obscure. The notion seems to be that choice-of-law rules only function once the interests or scope have been determined, as a way of determining the priority of various competing interests. The Draft Restatement itself candidly acknowledges that judges cannot be expected to determine with confidence whether particular provisions expressly determine the scope:

[D]istinguishing between scope and priority in a choice-of-law decision is probably too difficult to do reliably, because it is not a distinction that all courts consciously consider in making choice-of-law decisions.267

One thing that is surely correct is the admission that making the distinction is “probably too difficult to do reliably.” We will return in a moment to the problems that this categorical exclusion of unknown (but probably large) numbers of choice-of-law considerations causes.

The added complexity that this distinction invites might give some readers pause about whether the move to subjective interests is worth undertaking. Uncertainty about what the determination of subjective interests requires might explain some of the contemporary disinclination to take up a subjective approach. We conclude this Article by examining the problems arising out of difficulties in determination of subjective interests along with two other reasons for hesitation about making this fundamental change.

B.    Hesitations

The Draft Restatement’s candid concession of the difficulty in applying certain of its distinctions is not the only reason for reluctance to adopt a subjective theory of scope or interests. Difficulty in assessing the meaning of choice-of-law determinations is only one of the reasons typically cited for hesitation.

Another of these reasons is that respect for other states’ choice-of-law rules raises the specter of renvoi, and renvoi is anathema to almost all commonly recognized schools of conflicts thinking.268 Finally, if multistate actors are denied the authority to second guess one another’s determinations of interests/scope, and academic judgments about what interests exist are rejected on the grounds that they purport to determine objective value, it is unclear what role is left for normative thought. Is this conclusion inevitable? And if so, what does it mean for the critical arguments raised in this Article?

Each of these sets of concerns will be examined below. All deserve to be taken seriously, but none is fatal to the current project. They suggest some steps that can be taken to avoid the problems of our predecessors. Although these steps are only preliminaries, this final Section suggests, at least, that authentic jurisdictional realism is possible.

1.   Difficulty in determination of subjective interests.

Let us assume that State Alpha (the forum) is deciding a case with connections both to itself and to State Beta. The State Alpha court knows or can determine State Alpha’s own interests; that is one of its primary job responsibilities, after all. But it is committed to also considering the needs and desires of State Beta—State Beta’s interests, if you will. When it understands what those interests are, it will be in a better position to decide which law to apply. The proposal to use choice-of-law decisions to ascertain subjective interests rests on the observation that State Alpha may be better situated to understand the needs and preferences of State Beta if it examines State Beta’s statutes and also State Beta precedents on choice of law with fact patterns resembling the current dispute.

An initial hesitation concerning adoption of the theory based on subjective interests is whether it is reasonable to impose on judges these additional difficulties of implementation. States will not always report their subjective interests in so many words. If interest analysis were to be modified to attend to subjective interests, what evidence would be taken as indicative that such an interest exists? Will judges be dependent, as economists are, on some theory of revealed preferences?269 In particular, is a choice-of-law rule really a useful indicator of a state belief that an interest exists?

But the degree of ambiguity is vastly exaggerated. It is not necessary to adhere to all of the complexities that the Draft Restatement does, with unneeded distinctions that only serve to increase the need for factual information that may not be available. The Draft Restatement distinguishes for example between “choice-of-law rules” rules of “priority” and principles that “expressly determine [ ] scope”; only the latter are supposed to be recognized by other states.270 Doing away with these distinctions makes it unnecessary to determine in which of these boxes some particular decision fits and thus resolves one potential ambiguity.

Moreover, the Draft Restatement provides that when something is termed a choice-of-law rule, it is ineligible for consideration in the subjective interests calculation.271 The consequence of putting a legal principle into this category is to further exacerbate the ambiguity, by disqualifying what might otherwise be probative on the question of subjective interest. The arbitrary treatment of these different subgroups thus doubly complicates the decision process. It is unclear why something gets characterized as only a choice-of-law rule—the definition of choice-of-law rules is so uncertain. But then, once it is given this label, it cannot be consulted in determining the substantive statute’s scope.

There is no reason to exclude a piece of otherwise probative evidence simply because it has been characterized as a choice-of-law rule and does not reflect academic categories of scope or priority. If State Beta characteristically applies its law to cases like the one before the court, then State Beta must certainly be of the view that it has an interest. If a line of precedents shows that when State Beta decides a case with a comparable fact pattern it applies its own law—well, what could be more probative than that? If State Beta characteristically applies its own law to a particular fact pattern of its own volition in its own courts, then this is presumably what it wants. If State Beta’s available precedents all go the opposite way—they end by not applying State Beta law—looking at them still tells the State Alpha judge something important. These precedents show that State Beta is of the view that even if it does have an interest in the particular fact pattern, it considers that interest less important than interstate harmony.

There is no guarantee that the information thus obtained will be sufficient for resolving all the choice-of-law disputes that arise. But the object is not to give a guarantee, still less to exclude from consideration everything but the other state’s subjective preferences. The point is simply that if State Alpha locates such information, there is no reason that State Alpha should not be free to use it—along with everything else.

The Draft Restatement has no basis for saying that State Beta’s choice-of-law rules should not be factored into the determination of scope, still less that State Beta’s pronouncement of its interests should be discounted if they are based on choice-of-law rules. It might as well tell states—as Currie did—that they may not consider conflicts values because only substantive values matter for determining interests.272 If one accepts that interest is a species of statutory construction, what matters is what the state that authored the statute has decided. States ought to be able to use choice-of-law rules to assist with scope determinations if they want to. A state that defines its interests in terms of its choice-of-law rules is well within its prerogatives.

What matters for scope and interests is not whether some other state concludes that the decision whether to apply the statute was influenced by choice-of-law rules rather than express determinations of scope. Regardless of the reason that State Beta’s law does not apply, we may know from precedent that it does apply or that it does not applyThat is enough for State Alpha to go on. For State Alpha to refuse to consider some State Beta precedent on the grounds that it is a choice-of-law rule and not a scope determination is to repeat the mistake of Professor Beale. Such overly rigid—but easily manipulable—conceptual apparatuses have not fared particularly well over the long haul. Our historical experience with “brooding omnipresence” has not been positive.

The bottom line, fundamentally, is that the drafters of the Third Restatement have substituted their own ideas, as did their predecessors, for what the legislature had in mind—and have done so ostensibly in the name of the usual processes of statutory interpretation and construction.273 The animus that supposedly motivated the determination of scope was the recognition that scope, as an interpretation of state substantive law, is a subjective concept that takes its meaning from the preferences of the state’s authoritative legal decision-makers. But the new Draft Restatement imposes upon the state’s laws a conceptual distinction that is to all appearances utterly unknown to the state’s elected representatives and judges. It specifies that choice-of-law rules have no effect on a statute’s scope and are therefore not entitled to the recognition in other states that scope rules get.274 No effort is made to rationalize this consequence in terms of legislative understandings or expectations; it seems unlikely that legislatures pass statutes with these conceptual distinctions in mind.

The Draft Restatement, in these respects, has failed to internalize its commitment to subjective interests “all the way down.” It is mired, along with its predecessors, in the Bealean swamp.

2.   Renvoi-phobia.

A second reason not to take choice-of-law rules into account concerns the supposed logical problems caused by doing so. By arguing that one actor in the interstate system is not entitled to second guess the values and interests of the others—that each state’s determination of interstate values and choice-of-law rules is entitled to respect—we seem to be leaving an opening to the notorious choice-of-law problem of renvoi.275 The doctrine of renvoi, which allows the use of a state’s choice-of-law rule to determine whether its law should be applied, is rejected by virtually all of the major theories of choice of law.276 However, the ultimate horrible in the renvoi parade—the supposed circulus inextricabilis277—is not as frightful as it is made out to be.

The renvoi issue concerns whether a reference to the law of another state means a reference only to a state’s substantive rules or also to the other state’s choice-of-law rules.278 Treating it as a reference to both creates the possibility of certain paradoxical results. Take, for example, a typical products liability case in which a lawnmower designed and manufactured in one state injures a consumer in another state. Assume that Alpha (the forum) is the state of the individual plaintiff’s domicile and the state where the defendant manufacturer was incorporated; that Beta is the state where the injury occurred; and that Gamma is the state where the defective lawnmower was designed and manufactured. If Alpha’s choice-of-law rules require application of the law of the place of injury, and Beta’s choice-of-law rules require the law of the state where the negligent conduct took place, then State Alpha will refer to State Beta law, which will then refer to the law of State Gamma.

That result, by itself, is not so bad. But what if the choice-of-law rules of State Beta instead refer back to the law of State Alpha? Or what if State Alpha choice of law refers to the rules of State Beta, which in turn refer to the law of State Gamma, whose choice-of-law rules then point back to the law of State Alpha? These two fact patterns exemplify the notorious infinite hall of mirrors; they result in endless recursion with no apparent resolution.279 The cycle starts with the law of State Alpha, which refers to the law of State Beta, and then back to the law of State Alpha; the cycle then repeats. Logically, there is no clear answer to what should be done.

The problem seems to be the inclusion of choice-of-law rules within the reference to the “law” of the state that is selected. If State Alpha’s choice-of-law rules refer to the “law” of State Beta, this reference is ambiguous. It might include State Beta’s choice-of-law rules but it need not. If it does not refer to Beta’s choice-of-law rules, the ongoing chain of reference does not arise. The easiest way to solve the problem is therefore to deny that the word “law” includes choice-of-law rules and limit its application to a reference to State Beta substantive law. And this solution is adopted by most generally recognized approaches to choice of law.280

This solution would appear to require the disregard of choice-of-law rules in determining a state’s interests/scope. It seems to require that when State Alpha starts to analyze the interests of State Beta, or the scope of its laws in multistate situations, that State Beta’s choice-of-law rules not be consulted. If this appearance is correct, then rather than consulting State Beta’s choice-of-law rules, the forum would have to perform its own determination of whether a State Beta interest/scope existed. This means that State Beta’s interests would be considered by State Alpha as objectively existing, not subjectively. State Alpha would not be obliged to follow the actual determination of State Beta’s courts and legislature on that matter. An objective approach to the determination of interests/scope avoids the renvoi problem while the subjective determination of interests makes renvoi possible. In short, in order to avoid the problem of renvoi, choice-of-law rules would have to be disregarded when calculating interests or scope. Or so the argument goes.

But this argument is faulty. If we adopt a subjective definition of interests/scope, the state’s choice-of-law rules are brought to bear in only a limited way, which does not lead to the infinite hall of mirrors that the fearmongers posit. It is only when choice-of-law statutes are used more expansively, for other purposes, that this supposed infinite regress can come about, and such uses are not required by jurisdictional realism.281 The other purpose to which another state’s choice-of-law rules might be put would be deciding what to do if that state’s law is not selected. The subjective definition of interests/scope does not entail any particular alternative result if it is determined that the laws of the particular state do not apply. Using State Beta’s choice-of-law rules to determine whether State Beta satisfies the interest/scope requirement is different from using it as a basis for deciding which other state’s law to apply when State Beta has no interest in having its law applied.

Consider again our hypothetical products liability case. If State Beta has a choice-of-law rule that requires application of the law of the place of the negligent design or manufacture, then in a case in which design and manufacture occurred in State Gamma, State Beta (by its own admission) does not have an interest in having its law applied. But this does not mean that State Beta choice-of-law rules should be followed to determine that State Gamma or State Alpha law should apply instead. For the same reason that State Alpha is not empowered to determine the interests of State Beta, State Beta’s choice-of-law rules are not authoritative about the interests of States Alpha or Gamma. State Beta choice-of-law rules are authoritative only in regard to the existence or nonexistence of its own state interests.

Thus, in the lawnmower example, the infinite hall of mirrors problem does not arise. The forum can consider State Beta’s choice-of-law rules to determine that State Beta substantive law does not apply, but it would not consider State Beta’s choice-of-law rules as effective to require a reference to the law of State Gamma or a reference back to the law of State Alpha. A state’s choice-of-law rules are consulted only for the purpose of determining whether that state’s laws purport to apply. They do not address the question of which law applies if the answer is no.

3.   Value neutrality.

A final concern involves whether realism leaves any room for normative analysis in choice of law. In what way is normative analysis useful to choice of law? Is choice of law to be reduced to value-neutral manipulation of whatever it is that states think that they want? And if so, then what is the point of academic writing about choice of law—or any other subject? Are the arguments in this Article, moreover, as guilty of value-laden reasoning as Beale’s and Currie’s?

If we say that states assessing their own interests may take into account whatever they choose, and that states in assessing the interests of another state are bound to follow the interests/scope in whatever way that the other state does, then there seems to be no place left for normative critique. If this is the argument of modern choice-of-law theorists, then this should be made clear. Not only does it seem commonly assumed that normative analysis is worthwhile—a lot of choice-of-law theorists engage in activity that could hardly be characterized as anything else282—but it is not unreasonable to think that this is the way it ought to be. (This, itself, is of course hardly a value-neutral position.)

Jurisdictional realism does not, however, require value neutrality. What concerns the realist is for one multistate actor to treat a legal decision as ineffective because it is (supposedly) normatively incorrect. A state’s decision to impose its law should not be dismissed as ineffective to create an interest or to determine scope simply because it is contrary to some multistate actor’s normative judgment.283 But there is no objection to states being influenced by normative arguments, or to academic reformers making normative arguments in order to influence state decision-making. People develop normative arguments to guide their own behavior; states might do the same. Norms are employed by many multistate actors for a number of legitimate purposes.

From the discussion in the text above, one might be forgiven for thinking that our goal was to denounce academic theorizing that brings to bear the values of the academic author. But this is not the case; it is to be expected that writers on the subject would have normative positions and urge state institutions to adopt them. Currie’s folly was not that he made value judgments, but the way that he brought these value judgments to bear on legal decisions made by authoritative state decision-making organs. His chief sin was to declare that these decisions were ineffective because they did not reflect the state’s true interests, as he understood them.284

Making normative arguments is of considerable value in choice of law. Currie was entitled to argue that states ought to take an interest in the application of local law when a local defendant or plaintiff was benefited.285 He was entitled to argue that states have no real interest in applying their laws when they benefit a foreigner.286 He was entitled to argue that states ought to conceive their interests differently than they do; that states ought to construe their statutes in certain ways; and that states ought to apply one methodology rather than another when deciding choice-of-law disputes. If they did not, however, the proper remedy was not to declare that State Alpha should disregard State Beta’s perceptions of its own interests. The proper remedy—at least for a legal realist—is to try to convince State Beta that its perceptions ought to be changed.

Normative judgments are particularly useful for at least three purposes: formulating proposals, making predictions, and developing presumptions. We have already mentioned that decision-makers may be influenced by normative arguments, and that people use normative arguments to achieve that objective.287 Normative arguments, in other words, can be used as the basis for proposals. For example, advocates in court, candidates for office, members of the media, and concerned voters tell decision-makers how they ought to decide, and whyTheir proposals are designed to change other actors’ behavior. This is as true when the decision-makers are faced with choice-of-law decisions as when they are faced with decisions about substantive law.

It is also reasonable for State Alpha’s courts to take normative judgments under consideration in anticipating what State Beta would say if it addressed the issue—to use value judgments as the basis for predictions. Predictions about other states’ behavior can be important in the determination of other states’ interests. When State Alpha is seeking to ascertain the interests of State Beta, and evidence about how State Beta actually defines its own interests is not available, State Alpha might rely on normative judgments to formulate its guess. State Alpha, that is to say, imputes to State Beta the definition of interests that State Alpha thinks is best.

It should be kept in mind, however, that these predictions are at most presumptions. State Alpha has no authority to determine State Beta’s interests in any binding way. Even its best prediction may turn out to be mistaken; should this turn out to be the case, State Alpha’s presumption about State Beta’s interests would have to be corrected.


Modern conflicts-of-law scholarship can only be understood against the backdrop of Professor Beale’s vested rights theory and the First Restatement. Beale’s critics, including Professor Currie, used the tools of legal realism to shred Beale’s carefully constructed choice-of-law rules, the embodiment of the previous generation’s formalism. And Currie developed a new approach to conflicts that took the coveted place that Beale had occupied with courts and scholars.

But the two conflicts titans of the twentieth century stood in parallel in more ways than one. It proved harder for Currie to steer clear of his own conceptual commitments than it had been for him to tear down Beale’s assumptions in the first place. Beale had his “vested rights”; but Currie had his “interests.” Beale masqueraded as a logician, but Currie adopted the pose of statutory interpreter. Beale wore his conceptualism on his sleeve while Currie kept his conceptualism up his.

Considering Currie’s appropriation of realist critiques and realist-inspired conflicts theories, it seems both fair and natural to expect both him and his successors to avoid the Bealean mistakes. But Currie and his followers failed to heed the realist lessons they taught Beale. Can modern conflicts scholars use realism to dismantle Beale, but then put their own Beale equivalent in its place? We think not.

Internal consistency is not too much to ask. But all is not lost for modern scholars; with attention to potential pitfalls, modern choice-of-law theory can reconcile with realism. Reconciliation would come at a cost, however; academics would have to relinquish the prerogative of passing judgment on what states think their interests are. And that’s a price some modern theorists might not pay.

“You’ve let the modern choice-of-law approaches off too easy!” some may say. Perhaps that’s fair. This Article is only interested in theoretical consistency. We have not asked if realism is right or wrong; the question of whether modern choice-of-law theory as a whole is right or wrong has also not been on the table. The issue raised above was simply: Does modern choice-of-law theory fulfill its promise of bringing legal realism to conflict of laws? Our recommendations have limited themselves to identifying the minimum changes necessary to bring modern choice-of-law theory in line with realist objectives.

This doesn’t mean that we are realists, and we certainly aren’t particular fans of modern choice of law. We have our doubts about two-step theories that supposedly start with identification of interests or scope and only then move on to face the choice between competing states. But that’s beside the point. This Article raises just one question: Can modern choice-of-law theory be reconfigured as jurisdictional realism? There probably are some who would say that this lets modern choice-of-law theory off too easy, because the more important question is whether modern choice-of-law theory can be reconfigured to be true. Yes, that would be a more important question. But it would also be more difficult to answer.

  • 1. The phrases “choice of law,” “conflicts of law,” and “conflict of laws” are often treated as interchangeable. To the extent that the first of these is distinguishable from the second and third, it denotes theory and application regarding selection of the proper rule of decision. The second and third of these include not only selection of the proper rule but also multistate problems of personal jurisdiction and judgments law.
  • 2. See generally Kermit Roosevelt III, Legal Realism and the Conflict of Laws, 163 U Pa L Rev Online 325 (2015).
  • 3. See Michael S. Green, Legal Realism, Lex Fori, and the Choice-of-Law Revolution, 104 Yale L J 967, 969 (1995).
  • 4. See, for example, Larry Kramer, Rethinking Choice of Law, 90 Colum L Rev 277, 319–44 (1990).
  • 5. See Brainerd Currie, The Constitution and the Choice of Law: Governmental Interests and the Judicial Function, 26 U Chi L Rev 9, 9–10 (1958). 
  • 6. See Green, 104 Yale L J at 976–77 (cited in note 3).
  • 7. See Part II.A.2.b.
  • 8. Lea Brilmayer, The Other State’s Interests, 24 Cornell Intl L J 233, 239 (1991).
  • 9. See Abbe R. Gluck, Intersystemic Statutory Interpretation: Methodology as “Law” and the Erie Doctrine, 120 Yale L J 1898, 1962 (2011).
  • 10. See The American Law Institute Announces Four New Projects (ALI, Nov 17, 2014), archived at
  • 11. William L. Prosser, Interstate Publication, 51 Mich L Rev 959, 971 (1953). See also William L. Reynolds, Legal Process and Choice of Law, 56 Md L Rev 1371, 1371 (1997) (“Choice of law today, both the theory and practice of it, is universally said to be a disaster.”).
  • 12. Lea Brilmayer and Daniel B. Listwa, Continuity and Change in the Draft Restatement (Third) of Conflict of Laws: One Step Forward and Two Steps Back?, 128 Yale L J F 266, 270 (2018). See also Restatement of the Law (Third) Conflict of Laws: Council Draft No 2, § 5.01, cmt b at 114–15 (ALI Sept 12, 2017) (Draft Restatement):

    Resolving a choice-of-law question requires two analytically distinct steps. First, it must be decided which states’ laws are relevant, in that they might be used as a rule of decision. This is typically a matter of discerning the scope of the various states’ internal laws: deciding to which people, in which places, under which circumstances, they extend rights or obligations. Second, if state internal laws conflict, it must be decided which law shall be given priority.

  • 13. Brilmayer and Listwa, 128 Yale L J F at 270–71 (cited in note 12).
  • 14. Kermit Roosevelt III and Bethan R. Jones, The Draft Restatement (Third) of Conflict of Laws: A Response to Brilmayer & Listwa, 128 Yale L J F 293, 303 (2018).
  • 15. Brilmayer and Listwa, 128 Yale L J F at 270 (cited in note 12).
  • 16. Roosevelt and Jones, 128 Yale L J F at 303 (cited in note 14).
  • 17. See, for example, Ernest G. Lorenzen, Territoriality, Public Policy and the Conflict of Laws, 33 Yale L J 736, 738 (1924) (“State A has, of course, no power to impose its own policy upon any other sovereign state. Only some supra-state authority could do this.”). It is important to distinguish the realist insight that a state has exclusive authority to decide its own laws from Beale’s theory of “territorial rights.” See Joseph H. Beale, 1 A Treatise on the Conflict of Laws § 97.1 at 417 (Baker, Voorhis 1935). Beale’s theory derived from “territoriality,” a kind of sovereignty, a normative prescription that a state’s laws “affected and bound all property and persons within it and all contracts made within its boundaries.” See Laura Kalman, Legal Realism at Yale, 1927–1960 25 n 104 (North Carolina 1986). Beale’s realist critics, on the other hand, recognized that multiple states’ laws could legitimately be said to apply to many fact patterns. See id at 25–26. But the realists’ shift required them to reconceptualize what the law was: policy decisions made by the appropriate authority. There were no right or wrong answers. But there were wrong answerers—a judge in State A would exceed her authority in deciding the policy of State B, which fell outside of the realist framework for “law.” See Lorenzen, 33 Yale L J at 738 (cited in note 17).
  • 18. See H.L.A. Hart, The Concept of Law 120–28 (Oxford 3d ed 2012).
  • 19. See Part II.A.2.b.
  • 20. See Part II.B.
  • 21. Brilmayer, 24 Cornell Intl L J at 242 (cited in note 8).
  • 22. See, for example, Brainerd Currie, Notes on Methods and Objectives in the Conflict of Laws, 1959 Duke L J 171, 177–78.
  • 23. Id at 178.
  • 24. See Part III.A.
  • 25. See Part III.A.1.
  • 26. See Part III.A.2.
  • 27. See Part III.A.3.
  • 28. See Hessel E. Yntema, The Historic Bases of Private International Law, 2 Am J Comp L 297, 300–01 (1953) (describing a papyrus that may contain the earliest known reference to conflicts of law preserved in a crocodile mummy). See also generally Friedrich K. Juenger, A Page of History, 35 Mercer L Rev 419 (1984) (describing the history of conflicts of law).
  • 29. Yntema, 2 Am J Comp L at 300–01 (cited in note 28).
  • 30. Id at 300.
  • 31. Id.
  • 32. See generally, for example, Fredric Cheyette, Choice of Law in Medieval France, in Morris D. Forkosch, ed, Essays in Legal History in Honor of Felix Frankfurter 481 (Bobbs-Merrill 1966).
  • 33. See generally, for example, George A. Zaphiriou, Choice of Forum and Choice of Law Clauses in International Commercial Agreements, 3 Intl Trade L J 311 (1978).
  • 34. See, for example, Swift v Tyson, 41 US 1, 8 (1842).
  • 35. Kermit Roosevelt III, The Myth of Choice of Law: Rethinking Conflicts, 97 Mich L Rev 2448, 2455–56 (1999).
  • 36. Id at 2449 n 6.
  • 37. Until the publication of the First Restatement in the 1930s, conflicts law received little academic attention. Celia Wasserstein Fassberg, Realism and Revolution in Conflict of Laws: In with a Bang and Out with a Whimper, 163 U Pa L Rev 1919, 1942 (2015) (“Until the First Restatement, the area of conflict of laws had barely been taught in American law schools and the subject was not extensively discussed in the literature.”). One of the most notable pre–First Restatement explorations of conflicts of law is Justice Joseph Story’s Commentaries on the Conflict of Laws, which was “something like the nineteenth century’s version of a Restatement.” Kermit Roosevelt III, Brainerd Currie’s Contribution to Choice of Law: Looking Back, Looking Forward, 65 Mercer L Rev 501, 503 (2014).
  • 38. Beale, 1 Treatise on the Conflict of Laws at § 1.1 (cited in note 17).
  • 39. See, for example, Restatement (First) of Conflict of Laws § 218 (1934) (“Substantial Validity of Conveyance of Interest in Land”); id at § 183 (internal affairs of corporations); id at §§ 603–05 (statutes of limitations).
  • 40. Ernest G. Lorenzen and Raymond J. Heilman, The Restatement of the Conflict of Laws, 83 U Pa L Rev 555, 555 (1935).
  • 41. Id at 556.
  • 42. Joseph Beale seems to have envisioned the rules that he developed as a higher law, one that embodied universal truth. He claimed that the objective of choice of law was the enforcement of vested rights that had to be determined by locating the state in which “the last act” necessary to complete the cause of action took place. See Beale, 2 A Treatise on the Conflict of Laws § 348.1 at 1250–51 (Baker, Voorhis 1935). The place where the last act occurred supplied the applicable law; the Bealean system has thus become known as “the territorialist system” of choice of law. See Lorenzen, 33 Yale L J at 737–38, 742 (cited in note 17).
  • 43. Lea Brilmayer, Conflict of Laws: Foundations and Future Directions18 (Little, Brown 1991).
  • 44. Roosevelt, 163 U Pa L Rev Online at 326 (cited in note 2). See also Restatement (First) of the Law of Conflict of Laws § 1 (1934) (“No state can make a law which by its own force is operative in another state; the only law in force in the sovereign state is its own law.”).
  • 45. Lea Brilmayer and Raechel Anglin, Choice of Law Theory and the Metaphysics of the Stand-Alone Trigger, 95 Iowa L Rev 1125, 1129 (2010).
  • 46. Brilmayer, Conflict of Laws at 20–21 (cited in note 43).
  • 47. Id. See, for example, Restatement (First) of Conflict of Laws §§ 332–54 (1934) (discussing rules applicable to contract conflict-of-laws questions).
  • 48. See Lorenzen, 33 Yale L J at 746–47 (cited in note 17).
  • 49. Restatement (First) of Conflict of Laws § 377, cmt a, note 2 (1934).
  • 50. Lea Brilmayer, Untethered Norms after Erie Railroad Co. v. Tompkins: Positivism, International Law, and the Return of the “Brooding Omnipresence”, 54 Wm & Mary L Rev 725, 728–29, 748–49 (2013). See also Restatement (First) of Conflict of Laws § 377, cmt a (1934).
  • 51. See, for example, Roosevelt, 163 U Pa L Rev Online at 326 (cited in note 2).
  • 52. Fassberg, 163 U Pa L Rev at 1920–21 (cited in note 37).
  • 53. Brilmayer, Conflict of Laws at 22 (cited in note 43). See also, for example, Lorenzen and Heilman, 83 U Pa L Rev at 556 (cited in note 40).
  • 54. Brilmayer, Conflict of Laws at 24 (cited in note 43).
  • 55. Id at 24.
  • 56. Id at 31 (“[N]o amount of tinkering with the Restatement would have been adequate. As soon as one problem was resolved, another would have arisen to take its place.”).
  • 57. Id at 31–33.
  • 58. See Currie, 1959 Duke L J at 176 (cited in note 22) (“[A]ssessment of the respective values of the competing legitimate interests of two sovereign states, in order to determine which is to prevail, is a political function of a very high order. This is a function which should not be committed to courts in a democracy.”); Currie, 26 U Chi L Rev at 77 (cited in note 5) (“[The] choice between the competing interests of coordinate states is a political function of a high order, which ought not, in a democracy, to be committed to the judiciary.”).
  • 59. Currie, 26 U Chi L Rev at 77 (cited in note 5).
  • 60. 304 US 64, 78 (1938) (“There is no federal general common law.”).
  • 61. Southern Pacific Co v Jensen, 244 US 205, 222 (1917) (Holmes dissenting).
  • 62. Fassberg, 163 U Pa L Rev at 1921 n 13 (cited in note 37) (“Work on [the Second] Restatement began less than twenty years after the First Restatement was published and almost twenty years were needed for its preparation.”).
  • 63. Id at 1921.
  • 64. See Currie, 26 U Chi L Rev at 12–14 (cited in note 5).
  • 65. Brilmayer, Conflict of Laws at 67 (cited in note 43). The Second Restatement, for example, recognizes that statutes rarely speak directly to the conflicts questions that judges have to answer, and recommends a number of policies to be subjectively weighed in deciding questions. Restatement (Second) of Conflict of Laws § 6 (1971).
  • 66. Roosevelt and Jones, 128 Yale L J F at 303 (cited in note 14).
  • 67. Note, Bundled Systems and Better Law: Against the Leflar Method of Resolving Conflicts of Law, 129 Harv L Rev 544, 555 (2015) (arguing that “the better law approach” owes a “debt to legal realism”).
  • 68. Note, Comparative Impairment Reformed: Rethinking State Interests in the Conflict of Laws, 95 Harv L Rev 1079, 1087 n 40 (1982).
  • 69. Kalman, Legal Realism at Yale at 4–8 (cited in note 17). Early realists such as Justice Oliver Wendell Holmes Jr wrote that judges made decisions according to their political, economic, and moral biases and then back-justified them by writing opinions. Id at 7; Edward S. Robinson, Law and the Lawyers 108 (Macmillan 1935). But see Joseph William Singer, Legal Realism Now, 76 Cal L Rev 465, 471 (1988) (“This vision of opinions as nothing but post hoc rationalizations seriously misrepresents what most legal realists argued. The most convincing legal realists argued that the reasoning demanded by judicial opinions substantially constrained judges.”). Within conflicts, on the other hand, early realists focused on demonstrating that conflicts questions were decided by judges seeking to craft good policy. Lorenzen, 33 Yale L J at 744–45 (cited in note 17).
  • 70. Jensen, 244 US at 222 (Holmes dissenting).
  • 71. Kalman, Legal Realism at Yale at 4–8 (cited in note 17).
  • 72. See Singer, 76 Cal L Rev at 474 (cited in note 69) (“The legal realists wanted to replace formalism with a pragmatic attitude toward law generally. This attitude treats law as made, not found.”).
  • 73. See, for example, Walter Wheeler Cook, The Logical and Legal Bases of the Conflict of Laws, 33 Yale L J 457, 487 (1924):

    [T]he danger in continuing to deceive ourselves into believing that we are merely “applying” the old rule or principle to “a new case” by purely deductive reasoning lies in the fact that as the real thought-process is thus obscured, we fail to realize that our choice is really being guided by considerations of social and economic policy or ethics, and so fail to take into consideration all the relevant facts of life required for a wise decision.

  • 74. Karl N. Llewellyn, The Bramble Bush: On Our Law and Its Study 69–73 (Oceana 1951).
  • 75. See Singer, 76 Cal L Rev at 470 (cited in note 69) (“[T]he realists argued that, because of the indeterminacy of abstract concepts and the manipulability of precedent, it was almost always possible to appeal to competing and contradictory rules to decide any interesting contested case.”).
  • 76. See Brilmayer, 24 Cornell Intl L J at 234–37 (cited in note 8) (outlining the reasons that substantive laws differ from state to state—including disagreements about appropriate policy objectives and how they are best achieved).
  • 77. See Anthony D’Amato, The Limits of Legal Realism, 87 Yale L J 468, 472 (1978) (“Hart has analogized the legal system to a game with an official scorer whose rulings are final.”); id at 474–75 (“Hart’s concept of law posits the existence of a ‘rule of recognition’ that serves to identify officials, courts, the jurisdiction of courts, and the general system for the creation, modification, and extinction of rules of law that directly affect the conduct of the average person’s life.”).
  • 78. See Brilmayer, 24 Cornell Intl L J at 234–37, 240–42 (cited in note 8).
  • 79. See D’Amato, 87 Yale L J at 474–75 (cited in note 77).
  • 80. But see Oliver Wendell Holmes Jr, The Path of the Law, 10 Harv L Rev 457, 467 (1897) (“I think that the judges themselves have failed adequately to recognize their duty of weighing considerations of social advantage. The duty is inevitable.”) (emphasis added).
  • 81. The principle that one sovereign cannot authoritatively announce the contents of another sovereign’s laws is exemplified by the Supreme Court’s practice of leaving questions of state law to be resolved by the states themselves. See, for example, Expressions Hair Design v Schneiderman, 137 S Ct 1144, 1156 (2017).
  • 82. In particular, Beale divided law into two categories: the particular, substantive law of a state and the general common law. Beale, 1 Treatise on the Conflict of Laws § 3.1 at 20 (cited in note 17). He viewed conflicts questions as questions of general common law, which was a single, shared philosophical system that transcended state boundaries. Id at § 3.4 at 23.
  • 83. See Brilmayer, Conflict of Laws at 20–22 (cited in note 43).
  • 84. See Currie, 1959 Duke L J at 178 (cited in note 22):

    This process is essentially the familiar one of construction or interpretation. Just as we determine by that process how a statute applies in time, and how it applies to marginal domestic situations, so we may determine how it should be applied to cases involving foreign elements in order to effectuate the legislative purpose.

  • 85. Brilmayer, Conflict of Laws at 43 (cited in note 43).
  • 86. See Kalman, Legal Realism at Yale at 25–35 (cited in note 17).
  • 87. Currie, 1959 Duke L J at 178 (cited in note 22).
  • 88. Indeed, critiques of Beale’s formalist theories arguably played a significant role in the shaping and development of early legal realism. See Kalman, Legal Realism at Yale at 25–30 (cited in note 17).
  • 89. Currie, 1959 Duke L J at 178 (cited in note 22).
  • 90. Id. See also Part II.A.1.
  • 91. See, for example, Kramer, 90 Colum L Rev at 278 (cited in note 4) (“Brainerd Currie’s ‘governmental interest analysis’ probably remains the dominant choice of law theory among academics.”) (citations omitted).
  • 92. Currie, 1959 Duke L J at 174–75 (cited in note 22).
  • 93. Id at 174.
  • 94. Id at 175.
  • 95. Currie, 26 U Chi L Rev at 9–10 (cited in note 5).
  • 96. But see Green, 104 Yale L J at 970 n 18 (cited in note 3) (noting that “[a]side from Currie’s acceptance of Cook’s local law theory, there is little evidence of strong tendencies toward legal realism in Currie’s work”). Still, for reasons discussed here, we feel that it is appropriate to characterize Currie’s theory as a realist approach—as did Green after examining the evidence to the contrary. Id. See also, for example, Currie, 1959 Duke L J at 177–78 (cited in note 22) (“In the meantime, we would be better off if we would admit the teachings of sociological jurisprudence into the conceptualistic precincts of conflict of laws.”).
  • 97. See, for example, Brainerd Currie, Married Women’s Contracts: A Study in Conflict-of-Laws Method, 25 U Chi L Rev 227, 236 (1958) (“Why not face the fact that the place of making is quite irrelevant; why not summon public policy from the reserves and place it in the front line where it belongs?”).
  • 98. Currie, 1959 Duke L J at 178 (cited in note 22) (“If the court finds that the forum state has an interest in the application of its policy, it should apply the law of the forum.”).
  • 99. Id.
  • 100. Id. See also Brilmayer, Conflict of Laws at 62–63 (cited in note 43) (providing an overview of Currie’s theory); Kermit Roosevelt III, Conflict of Laws 43–70 (Foundation 2010) (same).
  • 101. Currie, 1959 Duke L J at 178 (cited in note 22).
  • 102. Id.
  • 103. Id.
  • 104. See, for example, Roosevelt, Conflict of Laws at 50 (cited in note 100) (“The false conflict is generally considered the crowning glory of interest analysis.”).
  • 105. Currie, 1959 Duke L J at 178 (cited in note 22):

    If the court finds that the forum state has an interest in the application of its policy, it should apply the law of the forum, even though the foreign state also has an interest in the application of its contrary policy, and, a fortiori, it should apply the law of the forum if the foreign state has no such interest.

    See also Currie, 25 U Chi L Rev at 261 (cited in note 97) (“The sensible and clearly constitutional thing for any court to do, confronted with a true conflict of interests, is to apply its own law. In this way it can be sure at least that it is consistently advancing the policy of its own state.”).

  • 106. Currie, 1959 Duke L J at 178 (cited in note 22).
  • 107. See, for example, Currie, 25 U Chi L Rev at 236 (cited in note 97). See also Brainerd Currie, The Verdict of Quiescent Years, in Selected Essays on the Conflict of Laws 584, 606 (Duke 1963):

    Like all statutory construction, such a decision is essentially legislative in character; the Court is trying to decide as it believes Congress would have decided had it foreseen the problem. . . . The profound difference consists in the fact that the process employed by the Court in these cases is avowedly that of statutory construction, and legislative correction is positively invited.

  • 108. 125 Mass 374 (1878).
  • 109. See Currie, 25 U Chi L Rev at 228 (cited in note 97).
  • 110. Id at 231–32; Herma Hill Kay, Remembering Brainerd Currie, 2015 U Ill L Rev 1961, 1961–62.
  • 111. Currie, 25 U Chi L Rev at 251–52 (cited in note 97). Currie first explored fourteen cases, which “present no real conflicts problem” or “are problems which cannot be solved by any science or method of conflict-of-laws.” Id at 251 (emphasis in original). He then proceeded to discuss the remaining cases under the rubric of “true conflicts.” Id at 259–63.
  • 112. Id at 234–37 (discussing the policies underlying Maine’s and Massachusetts’s statutes).
  • 113. See Lea Brilmayer, Governmental Interest Analysis: A House without Foundations, 46 Ohio St L J 459, 463–64 (1985).
  • 114. See generally Symeon C. Symeonides, The Choice-of-Law Revolution Fifty Years after Currie: An End and a Beginning, 2015 U Ill L Rev 1847.
  • 115. See id at 1850.
  • 116. See, for example, Alan Reed, The Anglo-American Revolution in Tort Choice of Law Principles: Paradigm Shift or Pandora’s Box?, 18 Ariz J Intl & Comp L 867, 868 (2001) (“[T]he contemporary call for a drastic reappraisal represents an unnecessary obfuscation of prevailing orthodoxy that has operated perfectly satisfactorily since the nineteenth century.”).
  • 117. See, for example, Roosevelt and Jones, 128 Yale L J F at 298–99 (cited in note 14) (arguing that “the Draft Restatement still represents progress in choice of law”).
  • 118. See, for example, Brilmayer, 46 Ohio St L J at 461 (cited in note 113) (arguing that “the foundations of interest analysis, as originally conceived, are fatally flawed”).
  • 119. See Kramer, 90 Colum L Rev 277, 278–79 (cited in note 4) (describing Currie’s critics).
  • 120. See Part II.A.1.
  • 121. But see, for example, Brilmayer, 46 Ohio St L J at 480 (cited in note 113); Lea Brilmayer, Jurisdictional Due Process and Political Theory, 39 U Fla L Rev 293, 297 (1987); Lea Brilmayer, Rights, Fairness, and Choice of Law, 98 Yale L J 1277, 1278 (1989). See generally Perry Dane, Vested Rights, “Vestedness,” and Choice of Law, 96 Yale L J 1191 (1987); Alfred Hill, The Judicial Function in Choice of Law, 85 Colum L Rev 1585 (1985); Friedrich K. Juenger, Conflict of Laws: A Critique of Interest Analysis, 32 Am J Comp L 1 (1984); Harold L. Korn, The Choice-of-Law Revolution: A Critique, 83 Colum L Rev 772 (1983); Maurice Rosenberg, The Comeback of Choice-of-Law Rules, 81 Colum L Rev 946 (1981).
  • 122. Russell J. Weintraub, A Defense of Interest Analysis in the Conflict of Laws and the Use of That Analysis in Products Liability Cases, 46 Ohio St L J 493, 493 n 4 (1985).
  • 123. See Currie, 26 U Chi L Rev at 11–12 (cited in note 5) (counseling a restrained and moderate reassessment of state interests to avoid true conflicts).
  • 124. Id.
  • 125. For example, a state might be held to have interests not only when the benefitting party was a local person but also when some activity occurred within its territory, on the theory that interests can be either compensatory or deterrent.
  • 126. See, for example, Brilmayer, 46 Ohio St L J at 476–80 (cited in note 113).
  • 127. For examples of the works of interest deniers, see Brilmayer, 98 Yale L J at 1278 (cited in note 121); John Hart Ely, Choice of Law and the State’s Interest in Protecting Its Own, 23 Wm & Mary L Rev 173, 177 (1981); Douglas Laycock, Equality and the Citizens of Sister States, 15 Fla St U L Rev 431, 446–47 (1987). Several scholars were less critical, while still not accepting of Currie’s approach. See, for example, John Bernard Corr, Interest Analysis and Choice of Law: The Dubious Dominance of Domicile, 1983 Utah L Rev 651, 673; Mark P. Gergen, Equality and the Conflict of Laws, 73 Iowa L Rev 893, 923–25 (1988).
  • 128. Brilmayer and Listwa, 128 Yale L J F at 271 (cited in note 12).
  • 129. See, for example, Brilmayer, 46 Ohio St L J at 475 (cited in note 113); Ely, 23 Wm & Mary L Rev at 177 (cited in note 127); Laycock, 15 Fla St U L Rev at 446 (cited in note 127) (“In my judgment, Currie’s view is inconsistent with the constitutional commitments to equality and national unity, and it violates the privileges and immunities clause [sic].”). The overlap between those who question the constitutionality of Currie’s approach to interest analysis and those who question its very foundations is worth noting.
  • 130. Brilmayer, Conflict of Laws at 62–63 (cited in note 43). Currie eventually backed away somewhat from his conclusion that the default rule of application of the forum state’s law was the correct approach to resolving these cases, but he never resolved the issue in a sufficiently unambiguous and nonarbitrary way for many of his successors. Id.
  • 131. Id.
  • 132. For a criticism of Currie’s treatment of the unprovided-for case, see, for example, Robert Allen Sedler, Interstate Accidents and the Unprovided For Case: Reflections on Neumeier v. Kuehner, 1 Hofstra L Rev 125, 138 (1973) (arguing for a “common policy” approach to unprovided-for cases).
  • 133. William F. Baxter, Choice of Law and the Federal System, 16 Stan L Rev 1, 19 (1963).
  • 134. Id at 19–20.
  • 135. For a general discussion of the application of game theoretical concepts to choice of law, see Brilmayer, Conflict of Laws at 191–223 (cited in note 43).
  • 136. Robert A. Leflar, Conflicts Law: More on Choice-Influencing Considerations, 54 Cal L Rev 1584, 1586–87 (1966).
  • 137. See, for example, William A. Reppy Jr, Eclecticism in Choice of Law: Hybrid Method or Mishmash?, 34 Mercer L Rev 645, 691 & n 216 (1983) (noting that in Minnesota, which followed a version of Leflar’s approach, only one case had held out-of-state law to be the better than forum law).
  • 138. Id at 694–95.
  • 139. Robert A. Leflar, Choice-of-Law Statutes, 44 Tenn L Rev 951, 954 (1977) (“The term ‘statutory construction’ is no more than a pretentious disguise for application of the court’s conflicts law.”).
  • 140. Fassberg, 163 U Pa L Rev at 1928 (cited in note 37). See also Brilmayer, Conflict of Laws at 67 (cited in note 43) (“The Second Restatement is an interesting blend of different conflicts approaches, which attempts to incorporate the best features of the various modern theories while avoiding their defects.”).
  • 141. Restatement (Second) of Conflict of Laws § 145, cmt e (1971).
  • 142. Brilmayer, Conflict of Laws at 67 (cited in note 43), citing Restatement (Second) of Conflict of Laws § 6 (1971).
  • 143. Restatement of the Law Third, Conflict of Laws (ALI), archived at
  • 144. See generally, for example, Herma Hill Kay, Larry Kramer, and Kermit Roosevelt, Conflict of Laws: Cases, Comments, Questions (West 9th ed 2013); Roosevelt, Conflict of Laws (cited in note 100); Kermit Roosevelt III, Choice of Law in Federal Courts: From Erie and Klaxon to CAFA and Shady Grove, 106 Nw U L Rev 1 (2012); Kermit Roosevelt III, Current Debates in the Conflict of Laws, 153 U Pa L Rev 1815 (2005); Roosevelt, 97 Mich L Rev 2448 (cited in note 35).
  • 145. Kermit Roosevelt III, Laura Elizabeth Little, and Christopher A. Whytock, Reporters’ Memorandum, in Restatement of the Law Third: Conflict of Laws, Preliminary Draft No 2 xiii, xv (ALI Aug 12, 2016) (Reporters’ Memorandum).
  • 146. Kermit Roosevelt III and Bethan Jones, Symposium on the Third Restatement of Conflict of Laws: What a Third Restatement of Conflict of Laws Can Do, 110 AJIL Unbound 139, 142 (2016), archived at
  • 147. Id at 143 n 19.
  • 148. See Reporters’ Memorandum at xv (cited in note 145):

    How is the first step to be performed? Brainerd Currie’s claim was that determining the scope of a state’s law was simply the familiar process of interpretation: just as courts commonly do in a purely domestic case where the scope of a statute is unclear, we should identify the purposes behind a law and decide whether those purposes would be furthered by applying it to the facts of the case. This means, Currie said, that the first step of the choice-of-law process is not distinctive to choice of law at all; it is simply a question of determining the substance and meaning of the relevant states’ laws.

  • 149. One important point of disagreement is whether “scope” and “interests” are subjective or objective concepts. See Reporters’ Memorandum at xvi–xvii (cited in note 145):

    [Referring to] a question that divided interest analysts: are state interests subjective (i.e., within the control of a state’s lawmakers) or objective (i.e., to be determined by judicial or academic analysis, regardless of what state lawmakers say)? Some scholars argued that interests were objective. See, e.g., Peter Westen, Comment, False Conflicts, 55 Cal. L. Rev. 74, 85 (1967). Others claimed that Currie’s statements about his methodology implied that they were subjective. See, e.g., Lea Brilmayer, Conflict of Laws 81 (1995) (stating that “if [interests] are to reflect legislative policies, then they must be subjective because the legislature is free to determine what policies it wants to have”); Larry Kramer, Return of the Renvoi, 66 N.Y.U. L. Rev. 979, 1011 (1991) (arguing that understanding interests as objective “is inconsistent with the premise that choice of law is a process of interpreting the relevant laws”).

    The Second Restatement did not address this question explicitly, but it did offer a hint: it said that the existence and intensity of state interests might be discerned by looking at a state’s choice-of-law decisions. See § 8, Comment k. That takes a step towards the subjective view, because it grants that a state’s views as to whether or not it is interested should have some weight.

    The Third Restatement takes the view that interests are subjective, that is, that the scope of a state’s law is a question of the substance and meaning of that law and, subject to constitutional constraints, is within the authority of that state’s lawmakers. 

    For further comparison of the Third Restatement to existing approaches, see Part III.A.3.b.

  • 150. See note 128 and accompanying text. See also Kramer, 90 Colum L Rev at 279 (cited in note 4) (“Currie’s . . . basic insight about the applicability of standard methods of statutory interpretation was sound.”).
  • 151. See Part II.A.2.
  • 152. See Part III.A.
  • 153. Id.
  • 154. Id.
  • 155. See Brilmayer, 46 Ohio St L J at 466–76 (cited in note 113); Dane, 96 Yale L J at 1259 (cited in note 121):

    [T]here would be no good reason, in a system of government that considers itself bound by the rule of law, to interpret a substantive statute that is silent on choice of law as requiring courts to pursue the legislature’s interests in ways that violate vestedness. Just as courts routinely interpret statutory silence on the issue of retroactivity to imply only prospective effect, courts should—and ever so quietly do—interpret statutory silence to imply that they should not intrude on the normative systems established by foreign sovereigns.

    Among the group of earlier writers whose intellectual roots antedated Currie’s writings and who never accepted interest analysis are David Cavers, Willis Reese (the Reporter for the Second Restatement), Alfred Hill, Maurice Rosenberg, and Albert Ehrenzweig. See David F. Cavers, The Choice-of-Law Process 100 (Michigan 1965); Albert A. Ehrenzweig, A Treatise on the Conflict of Laws 349 (West 1962); Hill, 85 Colum L Rev at 1596 (cited in note 121); Rosenberg, 81 Colum L Rev at 959 (cited in note 121) (“[Reese] was right not to embrace the theory that the interest-analysis approach is the best answer to the choice-of-law problem.”).

  • 156. Elliott E. Cheatham, Sources of Rules for Conflict of Laws, 89 U Pa L Rev 430, 448 (1941) (“The statutes of a state, like the common-law rules of a state, are for the most part formulated without regard to Conflict of Laws. The ordinary statutes and the ordinary common-law rules of a state are normally referred to and applied, however, in a Conflict of Laws case.”); id at 449–50 (“Most statutes are formulated with regard to only the ordinary or internal situations and on the problems of Conflict of Laws they are silent.”); Kramer, 90 Colum L Rev at 293 (cited in note 4) (“Unfortunately, the great majority of laws are silent with respect to extraterritorial reach, and determining their prima facie applicability is more difficult.”).
  • 157. Lea Brilmayer, Interest Analysis and the Myth of Legislative Intent, 78 Mich L Rev 392, 399 (1980); Ely, 23 Wm & Mary L Rev at 193–94 (cited in note 127) (“Legislators probably seldom think about the choice-of-law implications of what they are doing. . . . What we don’t find, in either the statutes or the legislative history, are expressions of intention to protect only locals.”).
  • 158. Rosenberg, 81 Colum L Rev at 959 (cited in note 121). See also Juenger, 32 Am J Comp L at 29 (cited in note 121).
  • 159. See Currie, 25 U Chi L Rev at 231–32 (cited in note 97) (winnowing down the statutes to four relevant factors). Although oftentimes multiple underlying facts are relevant to scope, we use the singular “trigger factor” without loss of generality to refer to the conjunction of these factors into one larger consideration.
  • 160. See id at 232.
  • 161. If every substantively relevant factor possessed choice-of-law relevance, then virtually every case would be a true conflict, and interest analysis’s proudest boast (the solution of large numbers of false conflicts) would come to naught. Every case would be a true conflict because to be a choice-of-law case, there must be relevant events in more than one state. If there are relevant events in more than one state and every in-state event gives rise to an interest, then every case will have at least two interests: the ones created by those relevant events.
  • 162. See note 156.
  • 163. See, for example, Kramer, 90 Colum L Rev at 294 (cited in note 4) (“The court thus faces a common problem of statutory construction, to wit, unforeseen or uncontemplated circumstances.”) (emphasis added); id at 300 (characterizing this critique as “not an objection to a choice of law method alone, or even primarily,” but “to a conventional method of statutory construction”); Weintraub, 46 Ohio St L J at 494 (cited in note 122).
  • 164. Weintraub, 46 Ohio St L J at 494 (cited in note 122). Weintraub’s assessment of the situation is clear from the title of one of his articles dating to 1984. See generally Russell J. Weintraub, Interest Analysis in the Conflict of Laws as an Application of Sound Legal Reasoning, 35 Mercer L Rev 629 (1984).
  • 165. Kramer, 90 Colum L Rev at 300 (cited in note 4).
  • 166. See note 163.
  • 167. Henry M. Hart Jr and Albert M. Sacks, 2 The Legal Process: Basic Problems in the Making and Application of Law1413–17 (Cambridge 10th ed 1958); Richard Posner, The Federal Courts: Crisis and Reform 286–87 (Harvard 1985).
  • 168. See Kramer, 90 Colum L Rev at 300 (cited in note 4) (“We know that this omission should not be understood to mean that there are no limits, but we do not know what limits lawmakers wanted to impose.”).
  • 169. See, for example, John Villasenor, Products Liability and Driverless Cars: Issues and Guiding Principles for Legislation (Brookings Institution, Apr 24, 2014), archived at
  • 170. See generally, for example, John Frank Weaver, Robots Are People Too: How Siri, Google Car, and Artificial Intelligence Will Force Us to Change Our Laws (Praeger 2014); Lindsey Barrett, Note, Herbie Fully Downloaded: Data-Driven Vehicles and the Automobile Exception, 106 Georgetown L J 181 (2017); Sophia H. Duffy and Jamie Patrick Hopkins, Sit, Stay, Drive: The Future of Autonomous Car Liability, 16 SMU Sci & Tech L Rev 453 (2013); Kevin Funkhouser, Paving the Road Ahead: Autonomous Vehicles, Products Liability, and the Need for a New Approach, 2013 Utah L Rev 437.
  • 171. See generally, for example, Noah J. Goodall, Can You Program Ethics into a Self-Driving Car?, 53 IEEE Spectrum 28 (2016).
  • 172. See generally, for example, Sven Nyholm and Jilles Smids, The Ethics of Accident-Algorithms for Self-Driving Cars: An Applied Trolley Problem?, 19 Ethical Theory & Moral Prac 1275 (2016).
  • 173. See, for example, Jack Boeglin, The Costs of Self-Driving Cars: Reconciling Freedom and Privacy with Tort Liability in Autonomous Vehicle Regulation, 17 Yale J L & Tech 171, 195 (2015) (analogizing self-driving cars to taxis); Brian Paden et al, A Survey of Motion Planning and Control Techniques for Self-Driving Urban Vehicles, 1 IEEE Transactions on Intelligent Vehicles 33 (2016).
  • 174. Hart, The Concept of Law at 124 (cited in note 18).
  • 175. Brilmayer, 24 Cornell Intl L J at 240 (cited in note 8).
  • 176. See Julius Cohen, Judicial “Legisputation” and the Dimensions of Legislative Meaning, 36 Ind L J 414, 415 (1961). See also Frank H. Easterbrook, Statutes’ Domains, 50 U Chi L Rev 533, 545 (1983) (discussing the effects of the “clear statement” principle on judicial gap filling).
  • 177. See Jerome Frank, What Courts Do in Fact—Part One, 26 Ill L Rev 645, 662 (1932).
  • 178. See Laycock, 15 Fla St U L Rev at 447 (cited in note 127); Dane, 96 Yale L J at 1211 (cited in note 121) (discussing a principle “positing the equality of sovereigns”).
  • 179. See Brilmayer, 78 Mich L Rev at 427 (cited in note 157).
  • 180. See Brilmayer, 46 Ohio St L J at 460 (cited in note 113) (citing discussions of reasonableness, fairness, and predictability as choice-of-law values).
  • 181. See Easterbrook, 50 U Chi L Rev at 541 (cited in note 176).
  • 182. Id at 540 (“Almost all statutes are compromises, and the cornerstone of many a compromise is the decision, usually unexpressed, to leave certain issues unresolved.”).
  • 183. For an insightful introduction that cites the views of many prominent scholars of statutory interpretation, see William N. Eskridge Jr, Interpreting Law: A Primer on How to Read Statutes and the Constitution 3–11 (Foundation 2016).
  • 184. Hart and Sacks, 2 The Legal Process at 1201 (cited in note 167).
  • 185. See Part II.B.
  • 186. One corollary problem with the use of the two-step approach in the Draft Third Restatement is the deep disconnect between the “ordinary processes of statutory construction” and the purpose of a restatement of law. Brilmayer and Listwa, 128 Yale L J F at 291 (cited in note 12). The new Restatement will seek to provide rules, derived from surveys of state cases, that judges can apply to resolve future conflicts cases. Id at 268. But if the scope of each individual statute is to be determined in each case through the regular processes of statutory interpretation, it is unclear how the Restatement can be of any use to a court—it provides general principles regarding state scope, but not an individual analysis of the scope of each of the statutes defining each cause of action in each state. The Restatement thus “offer[s] neither the authority of a state court to declare the scope of its state statute as a matter of law nor the sort of particularized analysis that grounds a persuasive argument from statutory interpretation.” Id at 276.
  • 187. For discussion of a contrary point of view about what judges ought to do, however, see Part II.B. The practice of going beyond the text of a statute to consider extraneous factors such as legislative intent and history is not uniformly accepted. Some scholars and judges have sought to cabin the circumstances in which judges are permitted to “gap fill[ ].” Easterbrook, 50 U Chi L Rev at 545 (cited in note 176). Others, most notably the late Justice Antonin Scalia, argued that judges ought never depart from the text of a statute. See generally Antonin Scalia and Brian A. Garner, Reading Law: The Interpretation of Legal Texts (Thomson/West 2012). For a thoughtful and thorough discussion of this debate between textualists and purposivists, see Eskridge, Interpreting Law at 3–11 (cited in note 183).
  • 188. See Brilmayer, 24 Cornell Intl L J at 234 (cited in note 8).
  • 189. Currie, 1959 Duke L J at 178 (cited in note 22) (“If necessary, the court should similarly determine the policy expressed by the foreign law, and whether the foreign state has an interest in the application of its policy.”).
  • 190. See Lorenzen, 33 Yale L J at 738 (cited in note 17) (“State A has, of course, no power to impose its own policy upon any other sovereign state. Only some supra-state authority could do this.”).
  • 191. Id. See also Brilmayer, 24 Cornell J Intl L at 234–37 (cited in note 8).
  • 192. Brilmayer, 24 Cornell J Intl L at 234 (cited in note 8) (“When a judge applies the law of another jurisdiction, his or her creative leeway is fairly circumscribed.”).
  • 193. This is a consequence of the relative infrequency with which legislative commands or history speak directly to choice-of-law concerns. See notes 156–59 and accompanying text.
  • 194. See Charles E. Clark, State Law in the Federal Courts: The Brooding Omnipresence of Erie v. Tompkins, 55 Yale L J 267, 284 (1946), citing Richardson v Commissioner, 126 F2d 562 (2d Cir 1942); Peter Westen and Jeffrey S. Lehman, Is There Life for Erie after the Death of Diversity?, 78 Mich L Rev 311, 315 (1980).
  • 195. See Part I.C.2.b.
  • 196. See William N. Eskridge Jr, Abbe R. Gluck, and Victoria F. Nourse, Statutes, Regulation, and Interpretation 349 (West 2014); William D. Popkin, A Dictionary of Statutory Interpretation 263 (Carolina Academic 2007).
  • 197. See Eskridge, Gluck, and Nourse, Statutes, Regulation, and Interpretation at 366–68 (cited in note 196).
  • 198. See John F. Manning, What Divides Textualists from Purposivists?, 106 Colum L Rev 70, 91–95 (2006) (arguing that canons of statutory interpretation are legitimate only inasmuch as they bear on the semantic meaning of a text). This distinction is enormously significant, transforming the canons from broad, interpretive presumptions into a tool more like a dictionary, to be used only to help understand the semantic content of words rather than something like the purpose of a statute.
  • 199. See William N. Eskridge Jr, Textualism, the Unknown Ideal?, 96 Mich L Rev 1509, 1545–46 (1998); Bradford C. Mank, Textualism’s Selective Canons of Statutory Construction: Reinvigorating Individual Liberties, Legislative Authority, and Deference to Executive Agencies, 86 Ky L J 527, 565 (1997).
  • 200. See Eskridge, Interpreting Law at 3 (cited in note 183).
  • 201. Currie in fact sometimes cited the duties of courts in democracies as the reason for judicial implementation of legislative policy—for decision according to his theory of state interests. Currie, 1959 Duke L J at 176–77 (cited in note 22).
  • 202. See, for example, id at 176 (“[A]ssessment of the respective values of the competing legitimate interests of two sovereign states, in order to determine which is to prevail, is a political function of a very high order. This is a function which should not be committed to courts in a democracy.”); Currie, 26 U Chi L Rev at 77 (cited in note 5) (“[The] choice between the competing interests of coordinate states is a political function of a high order, which ought not, in a democracy, to be committed to the judiciary.”).
  • 203. William D. Popkin, Statutes in Court: The History and Theory of Statutory Interpretation 144 (Duke 1999) (“[T]o the extent [legal realism] had anything to say about statutory interpretation, it appeared to support a full-bodied purposivism that was optimistic about the lawmaking collaboration between judges and legislatures.”).
  • 204. We are, of course, “all textualists now.” Elena Kagan, The Scalia Lecture: A Dialogue with Justice Kagan on the Reading of Statutes 8:28 (Harvard Law Today, Nov 17, 2015), online at (visited Apr 23, 2019) (Perma archive unavailable). But while textualism of various stripes may have won near-universal acceptance, the strict textualism discussed in this Section certainly has not—and would likely be almost unrecognizable to many jurists who are properly counted as within the broad coalition of modern textualists.
  • 205. Currie, 1959 Duke L J at 178 (cited in note 22).
  • 206. See Kramer, 90 Colum L Rev at 301 (cited in note 4) (“But if the objection is that judges cannot determine statutory purposes, or that their decisions are necessarily arbitrary, I disagree. Experience demonstrates that courts generally are able to determine the underlying purposes of laws based on language, structure, legislative history (if such history exists), and background.”).
  • 207. Id.
  • 208. See generally Adam Cohen, Imbeciles: The Supreme Court, American Eugenics, and the Sterilization of Carrie Buck (Penguin 2016); Michael J. Klarman, From Jim Crow to Civil Rights: The Supreme Court and the Struggle for Racial Equality (Oxford 2004).
  • 209. See Fassberg, 163 U Pa L Rev at 1931 (cited in note 37) (noting that roughly 20 percent of jurisdictions still use the First Restatement).
  • 210. This argument cannot be met by protesting that judicial consideration of extrinsic materials is limited to filling existing gaps and resolving existing ambiguities with material acceptable to the community at large. What counts as a gap that can be filled with community judgments and reliably established empirical evidence is open to debate. One person’s community norms may be another person’s imposition of personal value judgments. One person’s gap might be another person’s deliberate exclusion of whatever is not explicitly included. The line is not clear and where one places it depends on one’s views on the role of judges and their authority to take extratextual values into account. The location of the dividing line between legitimate judicial decision and impermissible judicial legislation is a question of degree and a matter of debate. Thus in any given case, the textualists would insist on demonstration of the connection between the statutory language and the interest analyst’s conclusion about the existence of state interests. This would be difficult to show, and it would not be possible to assume that statutory interpretation would supply an answer in all choice-of-law disputes, as modern theory presumes.
  • 211. See, for example, Currie, 25 U Chi L Rev at 233 (cited in note 97):

    Massachusetts, in common with all other American states and many foreign countries, believes in freedom of contract, in the security of commercial transactions, in vindicating the reasonable expectations of promises. It also believes, however, that married women constitute a class requiring special protection. It has therefore subordinated its policy of security of transactions to its policy of protecting married women.

  • 212. Almost invariably, Currie assumed that a statute was designed to assist local people and to be inapplicable when that goal was not served. Id at 234 (“What married women? Why, those with whose welfare Massachusetts is concerned, of course—i.e., Massachusetts married women.”) (emphasis in original). Thus, a pro-plaintiff law gave rise to an interest when the plaintiff was from the state that had such law; the pro-defendant law conversely gave rise to an interest only when the defendant was a local.
  • 213. Brilmayer, 78 Mich L Rev at 400 (cited in note 157) (citing instances of Currie criticizing courts and legislatures that disagreed with him).
  • 214. See, for example, Weintraub, 46 Ohio St L J at 495 (cited in note 122) (stating that it can never be rational, when “the place of injury has no other contacts with the parties,” for a state to apply its law to an unintentional tort on the basis that it occurred within the state’s territory because doing so “never will advance the purpose of its rule”).
  • 215. See Part II.A. See also Brilmayer and Listwa, 128 Yale L J F at 279–84 (cited in note 12) (describing how the Restatement encourages deference to nationwide trends in the scope-determination step, rather than an actual analysis of the policies behind the individual statute in the individual state).
  • 216. The principle that one sovereign cannot authoritatively announce the contents of another sovereign’s laws is exemplified by the Supreme Court’s practice of leaving questions of state law to be resolved by the states themselves. See, for example, Expressions Hair Design v Schneiderman, 137 S Ct 1144, 1156 (2017).
  • 217. Cavers, The Choice-of-Law Process at 100 (cited in note 155).
  • 218. Brilmayer, Conflict of Laws at 99 (cited in note 43).
  • 219. See, for example, Roosevelt, Conflict of Laws: Cases, Comments, Questions at 6–7 (cited in note 144). The new Restatement position is found in the Reporters’ Memorandum at xvii (cited in note 145) (“The Third Restatement takes the view that interests are subjective, that is, that the scope of a state’s law is a question of the substance and meaning of that law and, subject to constitutional constraints, is within the authority of that state’s lawmakers.”).
  • 220. This approach stands in contrast to those proposed by other staunch critics of the modern theory. Professor William L. Reynolds, for example, proposed precisely the opposite approach:

    Choice of law should not be treated as a separate intellectual and legal discipline, as American courts and scholars have done for the past century and a half. Rather, choice-of-law questions should be treated in the same way as other ordinary questions requiring the reconciliation of conflicting precedents or conflicting statutes. In the ordinary case involving conflicting and competing rules of decision, the court first identifies the policies that animate those laws; the court then explains why, in light of the facts, one or the other of the policies should control; the court also explains why the other competing policies were not winners and why arguably applicable precedents were not followed.

    Reynolds, 56 Md L Rev at 1372–73 (cited in note 11). 

    Reynolds, after critiquing the entirety of post-Bealean conflicts scholarship, essentially concludes that the search for a coherent, utile theory of conflicts is hopeless. Id at 1372. As will be made clear in this Part, we are not so dour. Further, Reynolds’s proposal that “case[s] should be resolved in the same way that a purely domestic matter would be resolved,” id at 1403 (emphasis omitted), suffers from all of the same problems as the modern theory’s reliance on the panacea of the regular processes of statutory interpretation. Worse yet, Reynolds’s proposal leaves no room for the possibility that factors such as comity and sovereignty will enter the analysis—which would surely lead to decisions that no modern conflicts scholar would ever defend. Id.

  • 221. These are, of course, not the only possible critiques of the approach taken by the Third Restatement. See generally, for example, Brilmayer and Listwa, 128 Yale L J F (cited in note 12); Reynolds, 56 Md L Rev 1371 (cited in note 11); Louise Weinberg, A Radically Transformed Restatement for Conflicts, 2015 U Ill L Rev 1999.
  • 222. See Brilmayer, 24 Cornell Intl L J at 235 (cited in note 8): 

    On issues of foreign substantive law, in contrast, a local judge has no role as authoritative expositor. While the judge may have to make educated guesses about the content of foreign law, the local judge is acting merely as a proxy for the foreign lawgiver. In consequence, the local judge has no authority to overrule the other state’s determination of its own law.

  • 223. See id at 239–42.
  • 224. Id at 241; Brilmayer, Conflict of Laws at 73 (cited in note 43).
  • 225. Brilmayer, Conflict of Laws at 100 (cited in note 43). Indeed, this substitution of academically defined values for the actual policies of a given state was the focus of a recent dialogue about the new Restatement. See Brilmayer and Listwa, 128 Yale L J F at 270–71 (cited in note 12); Roosevelt and Jones, 128 Yale L J F at 303 (cited in note 14). Roosevelt and Jones defended the Restatement’s approach, arguing that “[its] presumptions are supported by the cases of which [they] are aware” and that directing states to defer to convergent patterns identified from surveys of state cases “is exactly what a restatement is supposed to do.” Id at 308–09.
  • 226. See Restatement (Second) of Conflict of Laws § 6(2) (1971); Leflar, 54 Cal L Rev at 1586–87 (cited in note 136).
  • 227. Brilmayer, Conflict of Laws at 96 (cited in note 43): 

    The reason that the renvoi problem arises in Currie’s theory as it did in Beale’s is that renvoi is endemic to the very structure of the choice of law problem. It is not peculiar to Beale’s vested rights system because it arises out of the concept of deference to other states, a concept common to both Currie and Beale.

  • 228. See Roosevelt, Conflict of Laws at 56–59 (cited in note 100); Brainerd Currie, Survival of Actions: Adjudications versus Automation in the Conflict of Laws, 10 Stan L Rev 205, 229 (1958); Roosevelt, 97 Mich L Rev at 2462–63 (cited in note 35).
  • 229. Roosevelt, Conflict of Laws at 57–59 (cited in note 100).
  • 230. See id at 50–56; Brilmayer, Conflict of Laws at 47 (cited in note 43).
  • 231. Brilmayer, Conflict of Laws at 47 (cited in note 43).
  • 232. Interestingly, other models of partial subjective interests might also be possible. Figure 2 models the situation in which the interests of both the forum and the other state are defined subjectively. The forum, in other words, defines its interests subjectively and also respects the other state’s subjective definition of interests. But it is possible that the forum might define its own interests subjectively while imposing its value judgment on the other state. The argument for doing so would be that if State Beta wants respect for its interests from State Alpha (the forum) then it should have to satisfy State Alpha that its interests are legitimate. And legitimate means adequate under State Alpha’s own standards, as it applies them to itself.
  • 233. This parallelism between the objective theory depicted in Figure 1 and the subjective theory depicted in Figure 2 leads to two interesting observations. First, interest analysis is something akin to a special case of the subjective interest model. If State Beta follows standard interest analysis in defining its own interests, then Figures 1 and 2 should resemble one another. Second, the interest analysis principle that false conflicts should be decided under the law of the only interested state has an analogue in the subjective theory of interests/scope. That analogue is that when State Beta would apply the same law as State Alpha, that consensus law should be applied. State Alpha first determines whether it has an interest/scope in having its law applied; it then asks what State Beta would do with the case. If the answers are the same, then there is no objection to applying that state’s law.
  • 234. See notes 98–104 and accompanying text.
  • 235. See notes 105–06 and accompanying text.
  • 236. See, for example, Roosevelt, Conflict of Laws at 50 (cited in note 100) (“The false conflict is generally considered the crowning glory of interest analysis.”).
  • 237. Brilmayer, 24 Cornell Intl L J at 241 (cited in note 8).
  • 238. See notes 64–81 and accompanying text.
  • 239. Id.
  • 240. See notes 75–78 and accompanying text.
  • 241. Brilmayer and Listwa, 128 Yale L J F at 271, 277–79 (cited in note 12). See also Part I.C.2.b.
  • 242. See note 105 and accompanying text.
  • 243. See note 112 and accompanying text.
  • 244. See, for example, Herma Hill Kay, A Defense of Currie’s Governmental Interest Analysis, 215 Recueil des Cours 9, 173 (1989); Bruce Posnak, Choice of Law: Interest Analysis and Its “New Crits”, 36 Am J Comp L 681, 687 n 42 (1988).
  • 245. See, for example, Bernkrant v Fowler, 360 P2d 906, 909 (Cal 1961); People v One 1953 Ford Victoria, 311 P2d 480, 482 (Cal 1957) (taking the need for predictability into account in applying governmental interest analysis).
  • 246. See Black’s Law Dictionary 1636 (West 10th ed 2014).
  • 247. See Part II.A.2.b.
  • 248. Currie, 1959 Duke L J at 178 (cited in note 22).
  • 249. See note 244 and accompanying text.
  • 250. Id.
  • 251. See note 225.
  • 252. See Part I.C.
  • 253. See id.
  • 254. This is a simple consequence of the application of realism’s respect for the decisions of judges regarding indeterminate statutes. See Part I.C.1.b, combined with a notion of comity.
  • 255. See Brilmayer, 24 Cornell Intl L J at 234–37 (cited in note 8).
  • 256. See notes 77–79 and accompanying text.
  • 257. This Article examines only the effect of the subjective/objective debate on the so-called “two-step” model. See Roosevelt, Conflict of Laws at 43 (cited in note 100); Reporters’ Memorandum at xv–xvi (cited in note 145). For a radically different model of choice-of-law decision-making, see Brilmayer and Anglin, 95 Iowa L Rev at 1168 (cited in note 45) (suggesting that a solution, instead of objective factors, might be a “verbal formulation such as center of gravity” test).
  • 258. See notes 77–79 and accompanying text.
  • 259. See Part II.A.1.
  • 260. Draft Restatement at 112–13.
  • 261. Reporters’ Memorandum at xvii (cited in note 145).
  • 262. Id.
  • 263. The decision to honor state scope determinations when they are found in statutes, but not when they are found in judicial decisions, is reminiscent of the pre-Erie distinction between state statutory law, which was treated as binding in federal courts sitting in diversity under the Federal Judiciary Act—and state common law, which was held not to be binding on federal judges. Section 34 of the Federal Judiciary Act provides that “The laws of the several states, except where the constitution, treaties or statutes of the United States shall otherwise require or provide, shall be regarded as rules of decision in trials at common law in the courts of the United States in cases where they apply.” Judiciary Act of 1789, 1 Stat 92, codified at 28 USC § 725. That distinction was discredited in ErieErie, 304 US at 78. Note that this example also arises in a jurisdictional context, with its characteristic problems of multiple voices. See Part III.A.3.a.
  • 264. Reporters’ Memorandum at xviii (cited in note 145).
  • 265. Id at xvii.
  • 266. Id at xvii: 

    For the purposes of the Third Restatement, an express determination is limited to a statutory restriction on the scope of a state’s law, i.e., something like “Charities shall be immune for tort claims based on negligence with respect to charitable activities performed wholly within this state.” A choice-of-law rule (such as territorialism, or interest analysis, or the Neumeier rules) does not expressly determine the scope of state law.

  • 267. Id at xviii.
  • 268. See Part III.B.2.
  • 269. The theory of revealed preferences holds, roughly speaking, that consumer preferences can be derived from information about the consumer’s buying habits. See Paul A. Samuelson, Consumption Theory in Terms of Revealed Preference, 15 Economica 243, 243 (1948) (“[T]he individual guinea-pig, by his market behaviour, reveals his preference pattern—if there is such a consistent pattern.”).
  • 270. Reporters’ Memorandum at xvi–xvii (cited in note 145).
  • 271. Id at xvii–xviii.
  • 272. See Part I.C.
  • 273. See Brilmayer and Listwa, 128 Yale L J F at 270–71 (cited in note 12).
  • 274. Draft Restatement at 112–13 (cited in note 12).
  • 275. For the classical doctrine, see Ernst Otto Schreiber Jr, The Doctrine of the Renvoi in Anglo-American Law, 31 Harv L Rev 523, 524 (1918) (“A jural matter is presented which the conflict-of-laws rule of the forum refers to a foreign law, the conflict-of-laws rule of which, in turn, refers the matter back again to the law of the forum. This is renvoi in the narrower sense.”); Ernest G. Lorenzen, The Renvoi Theory and the Application of Foreign Law, 10 Colum L Rev 190, 190 (1910). For important contemporary discussions one can do no better than Kramer, 66 NYU L Rev at 1005 (cited in note 149) and Roosevelt, Conflict of Laws at 22–25 (cited in note 100).
  • 276. See, for example, Restatement (First) of Conflict of Laws § 7 (1934); Brainerd Currie, Notes on Methods and Objectives in the Conflict of Laws, in Selected Essays on the Conflict of Laws 177, 184–85 (Duke 1963) (rejecting application of renvoi).
  • 277. John Pawley Bate, Notes on the Doctrine of Renvoi in Private International Law 49 (Stevens and Sons 1904).
  • 278. See Kermit Roosevelt III, Resolving Renvoi: The Bewitchment of Our Intelligence by Means of Language, 80 Notre Dame L Rev 1821, 1837 (2005); Kramer, 90 Colum L Rev at 279 (cited in note 4).
  • 279. See note 248.
  • 280. The Draft Restatement, however, proposes the recognition of choice-of-law rules for this purpose, at least on some occasions, and a small but seemingly growing number of choice-of-law theorists appears willing to consider the possibility. See Part III.A. Certain earlier authors had made somewhat similar suggestions, although they limited the suggestion to cases in which both states adhered to governmental interest analysis. See, for example, Arthur Taylor von Mehren, The Renvoi and Its Relation to Various Approaches to the Choice-of-Law Problem, in Kurt H. Nadelmann, Arthur T. von Mehren, and John N. Hazard, eds, XXth Century Comparative and Conflicts Law: Legal Essays in Honor of Hessel E. Yntema 380, 393 (A.W. Sythoff-Leyden 1961).
  • 281. In effect, a subjective consideration of the application of another state’s law solves the infinite recursion problem posed by renvoi by replacing an evaluation of the other state’s law with the result of that consideration, which the other state’s courts have already derived. This allows the recursion to bottom out, eliminating the risk of infinite, paradoxical evaluations.
  • 282. See notes 128–29 and accompanying text.
  • 283. See Part II.A.2.b.
  • 284. See note 213 and accompanying text.
  • 285. See note 128 and accompanying text.
  • 286. See id.
  • 287. See Part III.A.