The Draft Restatement (Third) of Conflict of Laws: A Response to Brilmayer & Listwa
abstract.This Essay responds to Lea Brilmayer and Dan Listwa’s criticisms of the Draft Restatement (Third) of Conflict of Laws. We appreciate their engagement. As a general matter, we disagree about the nature and purpose of restatements. More specifically, we disagree about the history and aims of the Restatements of Conflict of Laws. Brilmayer and Listwa’s main criticism—that the drafters of the Restatement (Third) are not authoritatively interpreting any single state’s law and therefore can be only persuasive authority as to the content of a state’s law—could apply to all restatements. But since this Draft Restatement, like other restatements, draws its rules from decided cases, the criticism makes little sense even on its own terms.
In 1934, the American Law Institute (ALI) published the Restatement (First) of Conflict of Laws, the brainchild of Reporter Joseph Beale. From the very beginning, the Restatement (First) was attacked by critics—both fiercely and successfully. Its formalistic, deductive approach was out of step with the increasingly realist perspective in American law, and it was not long before the academic revolution in choice of law spread to the courts.1 With courts increasingly rejecting the dogmas of the Restatement (First), the ALI went back to work in 1953, beginning a Restatement (Second).2
The Reporter of the Restatement (Second), Willis Reese, was aware that choice of law was undergoing a revolution. He did not believe it possible, amid such tumult, to restate determinate rules of the sort that characterized the Restatement (First).3 Instead, after over seventeen years of work,4 he offered a flexible approach that more or less told courts the relevant factors to consider and set them loose to see what they did.5 His hope was that judicial practice would converge on outcomes in particular categories of cases that would allow the drafting of rules that were narrower than those of the Restatement (First) and sensitive to state policies in a way that the earlier territorial rules of the Restatement (First) were not.6
Like the Restatement (First), the Restatement (Second) was the target of criticism—much of it coming even before it was published in 1971.7 While the Restatement (Second) was a success in terms of adoption by states,8 over the years, academic criticism continued, with professors focusing on the failure of the Restatement (Second) to provide meaningful guidance to courts.9 From one perspective, this criticism is justified: the Restatement (Second) requires judges to do a lot of work, and the outcomes of cases are often unpredictable to parties trying to plan their activities. But from Reese’s perspective, it was less justified. Reese did not, after all, intend the Restatement (Second) to be an approach that would be used forever.10 Instead, its framework was more of a means of collecting data that would allow future Reporters to write better rules.11 Academics lost sight of this point, if they ever understood it.12
Forty-three years after the publication of the Restatement (Second), the ALI commissioned a third.13 The aim of the Restatement (Third) is to complete the project Reese began with the Restatement (Second): to survey the practice of courts using the Restatement (Second) and other modern approaches; to consider codifications in the United States and elsewhere; and to determine when practice converges sufficiently to allow the formulation of appropriately narrow and policy-sensitive rules. Like its predecessors, the Restatement (Third) has received criticism even before its completion. Among the critics are Professor Lea Brilmayer and Mr. Dan Listwa. We are honored by their engagement with the Draft Restatement (Third) and their commentary is thought-provoking. On the whole, though, we find it puzzling and unpersuasive.
Brilmayer and Listwa structure their critique around the themes of change and continuity. They suggest that the Draft Restatement (Third) maintains continuity with its predecessor by retaining what they call the “rules/exceptions structure” of the Restatement (Second). They suggest that the Draft Restatement departs from the Restatement (Second) by endorsing “governmental interest analysis” and the modern approach to choice of law while abandoning considerations such as uniformity, predictability, and ease of application. And they suggest that these two elements—rules and policy analysis—cannot coexist.
Every part of this critique is mistaken. Brilmayer and Listwa are wrong about the structure and philosophy of the Restatement (Second). They are consequently wrong about the relationship between the Restatement (Second) and the Draft Restatement. And they are independently wrong about the relationship between rules and policy analysis. Finally, they offer an analysis of the Draft Restatement’s escape clause that is baffling and contradicts their earlier arguments. We address these issues in turn.
Brilmayer and Listwa correctly note that the Restatement (First) contained rules intended to bind judges. Those rules were, it is now generally conceded, bad rules. Self-contained and derived from a territorialist premise,14 the Restatement (First) often dictated arbitrary and unsatisfying results because its rules were insensitive to the content of state laws. Consequently, choice-of-law decisions were made without regard to relevant state policies.15 When Joseph Beale offered an explanation for his rules, it tended to be that no other rules were possible. Faced with the charge of dogmatism, he responded “Does not the Bar desire dogmatic statements?”16 It is not surprising, then, that judges often employed various escape devices to avoid the Restatement (First)’s unwavering rigidity.17
The Restatement (Second) abandoned those rules. In their place, it offered what its chief Reporter, Reese, called “an approach”—a list of factors that should be considered in making a choice of law decision.18 The factors included both what might be called “right answer” factors—factors that would lead a court to the correct decision if what it was trying to do was maximize aggregate state policy satisfaction over a long series of cases—and what might be called “systemic” factors—factors that make for a good choice-of-law system, rather than a good individual decision. These factors include uniformity, predictability, simplicity, ease of application, and so on. Including both sets of factors as guidance for decisions in individual cases was a little strange, because the way to satisfy the right answer factors is to select the right law, while the way to satisfy the systemic factors is more related to selecting the right choice of law methodology. The best way to produce predictability, simplicity, and ease of application is probably to adopt a rule.19
This observation was not lost on Reese, however. Reese justified the Restatement (Second)’s “eclectic”20 nature as reflecting the fluidity and change then animating American choice-of-law jurisprudence.21 Particularly in the areas of tort and contract, Reese considered it preferable to adopt an approach “too fluid and uncertain in application than to take one’s chances with a precise and hard-and-fast rule that may be proved wrong in the future.”22 This preference for standards over rules is unusual for a restatement, a text whose drafting has been described by the ALI as “the quest to determine the best rule.”23 But Reese did not expect the Restatement (Second) to last indefinitely. His hope, expressed in both The Second Restatement of Conflict of Laws Revisited24 and his earlier article, Conflict of Laws and the Restatement Second,25 was that experience applying the Section 6 factors would produce patterns of decisions that could be captured by rules. It was for this reason that he characterized the Restatement (Second) as a transitional work.26
The Restatement (Third), we submit, is best understood as an attempt to fulfill Reese’s vision. The methodology of its Reporters is to look at current choice-of-law decisions under the Restatement (Second), other modern approaches, foreign-country systems, and even the practice of territorial states, and identify categories of cases where the results are consistent enough to be stated in the form of rules.27 When sufficient convergence exists—and often it does28—a rule can deliver right answers while also satisfying the systemic factors. It may not capture all results, and in rare cases the Restatement (Third)’s rules may sacrifice right answers.29 But as long as the number of errors is small enough, the Draft Restatement still represents progress in choice of law. As Reese wrote, “[p]erfection is not for this world.”30
Brilmayer and Listwa apparently31 disagree with this characterization of the Restatement (Second) and its relation to the Draft Restatement (Third). Where we see a structural change—albeit one that Reese anticipated and hoped for—between the Restatement (Second) and Draft Restatement (Third), they see continuity, arguing that the Restatement (Second) did not embody an approach but instead had a rule-based structure similar to the Draft Restatement. Faced with Reese’s explicit description of Section 6 as an approach, they downplay the significance of that section by calling it an escape clause.32 We find this puzzling. The universal directive of the Restatement (Second) is to apply the law of the state with the most significant relationship. Section 6 tells judges what to think about in deciding which state has the most significant relationship. Section 6 is, therefore, the heart of the Restatement (Second), as virtually every commentator understands.33 Indeed, Section 6 analysis controls essentially every judicial application of the Restatement (Second).34 To call the analysis that drives the ultimate decision in virtually every case an escape clause strikes us as quite misleading.
It is true that some areas of the Restatement (Second) remained rule-governed.35 But we do not take Brilmayer and Listwa to be resting their characterization of the Restatement (Second) on these sections. They appear to claim instead that the Restatement (Second) is generally structured in terms of rules and exceptions and offer a citation to Reese’s article, Choice of Law: Rules or Approach, as support.36 In fact, in the passage they cite (containing the word “usually”), Reese did not speak of rules but rather “general principles.”37The Restatement (Second) itself repeatedly disavows the characterization of these black-letter sections as rules.38 So too did Reese.39
We do not mean to place too much emphasis on the use or disavowal of the word “rule,” which can be used to mean different things. (While the Restatement (Second) itself seems to take some pains to distinguish its standards, principles, and empirical appraisals from rules, Reese himself more than once referred to “the rule of most significant relationship.”40) The general point, however, is clear and not a topic of live debate among scholars. In conflict-of-laws scholarship, rules are associated with rigidity, guidance for judges, certainty, predictability, and ease of application. Standards or approaches are associated with flexibility, judicial discretion, unpredictability, and lack of certainty. Taking that perspective, it is almost universally agreed that the Restatement (Second) falls on the standard/approach side.41
In addition to mistaking change for continuity, Brilmayer and Listwa mistake continuity for change. They suggest that the Draft Restatement “casts its predecessor’s methodological caution to the winds” by focusing on state substantive policies rather than other Section 6 factors such as certainty, predictability, uniformity, and ease of application.42 This misunderstands the relationship between what we have called the right answer factors and the systemic factors. Promoting systemic factors is chiefly a matter of the form of a choice-of-law system: systemic factors are advanced by the use of rules. Promoting right answer factors is chiefly a matter of content: content that takes state policies into account delivers sensible outcomes. Rules that are sensitive to the policies of relevant states promote both sets of factors.43 The Draft Restatement explains this explicitly.44
Far from abandoning those systemic considerations, then, the Draft Restatement promotes them better than the Restatement (Second) did or could. The Restatement (Second) lists them as among factors to consider, but a judge in an individual case, consulting Section 6 to identify the state with the most significant relationship, has very little ability to promote simplicity, predictability, uniformity, or ease of application: she is applying an approach, not creating one. Reporters writing a restatement do have that ability: they can produce a system that delivers uniform results and is simple, predictable, and easy to apply. They can do that by writing straightforward rules and framing the analysis, where possible, in ordinary legal concepts rather than choice of law esoterica. That is what the Reporters for the Draft Restatement (Third) are doing.
The main substantive issue Brilmayer and Listwa raise, however, has less to do with the Restatement (Third)’s structural origins and more to do with its theoretical foundation. Part II of their Essay (“Institutional Competence”) argues that the Draft Restatement’s commitment to a jurisprudence of rules—and therein, its commitment to traditional values of predictability, uniformity, and ease of application—creates an unavoidable disharmony with the two-step model, which they claim rests on the “authority of a state court to declare the scope of its state statute as a matter of law.”45 We find this puzzling. Perhaps the best way to approach their criticism is to consider a few questions. First, what distinguishes modern from traditional approaches? Second, what is the standard objection to the modern move? And third, what is the standard rejoinder to that objection?
As to the first question, the distinctive nature of modern approaches can be described in a few interrelated ways that ultimately come down to the same thing. We can start with a maximally general account of choice of law: what we have called the two-step model. The two-step model understands the choice-of-law process as a matter of first identifying relevant state laws, and second, if multiple conflicting laws are relevant, selecting one of those laws.46 The traditional territorial approach can be described in this conceptual framework: the relevant law is the law of the state where the right to redress vested (where the tort, for instance, or the contract occurred).47 Since only one law is relevant, there is no need to pick among competing laws, and the territorial approach ends at the first step—though it does face some difficulties in identifying a single state where a multistate tort or contract occurred.48
One thing that distinguishes the modern approaches from the territorial approach is the modern approaches’ rejection of the idea that one and only one state’s law must be relevant. Instead, modern approaches generally grant that multiple states’ laws may be relevant. What makes a law relevant? Interest analysis says a law is relevant if the state is interested, that is, if application of the state’s law would promote the policies behind that law.49 The Restatement (Second) does not offer as much explicit guidance, but it certainly includes policy analysis that sounds quite similar to interest analysis, and indeed it uses the same vocabulary: it instructs judges to consider the policies of “interested states.”50 Other modern approaches such as California’s comparative impairment,51 Pennsylvania’s mixed approach,52 or New York’s Neumeier rules also use interest analysis as a first step.53 So, generally speaking, we can say that while the territorial approach identified a single relevant state by “deduction from territorial postulates,”54 modern approaches identify one or more by asking whether the policies behind the state’s laws are implicated by the particular facts of a multistate case.
There are two ways of developing this point a little further—again, we think they amount to much the same thing. One is to say that policies are within the state’s control. Another is to say that this 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. When, in a domestic case, a court is trying to decide whether the facts of the case fall within the scope of a statute, it often asks whether application of the statute to those facts would promote the purposes behind the statute.55 Understood this way, the first step of modern choice of law is conceptually recognizable as ordinary legal analysis.56
There are several interesting consequences here—the main one perhaps being that everyone else must listen to state courts and legislatures when they say that they are or are not interested, just as they must listen if a legislature defines “pedestrian” to include (or exclude) rollerbladers, or a state court of last resort so construes the word “pedestrian” in one of its own statutes. To put it simply, states are authoritative with respect to the scope of their laws in domestic cases, and the modern understanding of choice of law suggests that they are likewise authoritative in the multistate context. In the process of drafting the Restatement (Third) there has been debate about whether this is so, with some participants arguing that courts should be free to disregard the words of sister-state statutes and the interpretations of the courts of those states. This position seemed strange to us since it is inconsistent with the remainder of U.S. law, and there appears to be no support for it in the caselaw.57 Some traces of this discussion may be found in the Reporters’ Memoranda accompanying Preliminary Draft 2 and Council Draft 2.58 To the extent that Brilmayer and Listwa argue that the Draft Restatement should not tell judges to respect sister-state specifications of scope (we cannot figure out whether they do or not), they are urging the drafters to reject uniform caselaw on the basis of a theory they have not articulated. That does not seem proper for a restatement.
Whatever Brilmayer and Listwa think about the power of judges to contradict other states about the scope of those states’ law, however, they raise a different objection here. Here, they say that the “standard objection” to this account is that statutes often say nothing about multistate scope, and therefore the process of determining scope cannot be statutory interpretation. It is true that statutes often do not specify their multistate scope. Sometimes they do, however, and the fact that courts in such cases take the statements as binding59 suggests that the multistate scope of a statute is in fact a question about the content and meaning of the statute, as it obviously is in domestic cases.60
Regardless, we simply do not see how it follows that determining scope is not interpretation.61 Statutes do not explicitly specify lots of things: that is exactly why interpretation matters. Courts or other decisionmakers often have to decide whether a law grants rights to, or imposes obligations on, a particular person in particular circumstances. If a statute grants rights to pedestrians, does it do so to a person on rollerblades? When this question is answered in the domestic context, we call it interpreting the law, and there is no reason to suppose it magically becomes something else simply because another state is involved.62
Because we see no force in this objection, we are not sure what the response should be, except to say that the objection “but the statute doesn’t explicitly say that” can be applied to almost every instance of statutory interpretation.63 Brilmayer and Listwa offer another response, which they call “standard”: that courts have the last word on the meaning of their own state law, so they can authoritatively interpret it even if, in doing so, they are making it.
It is true that state courts have the power to make state common law. They do not have the power to make state statutory law, so their interpretive authority rests on a different ground there. In any case, we do not understand the relevance of this response. We have never seen it in the academic literature, so referring to it as “standard” seems odd. To the extent that it relates to Brilmayer and Listwa’s claim that the two-step model relies on “the authority of a state court to declare the scope of its state statute as a matter of law”64 we do not understand that claim either. Judges using the two-step approach spend at least half their time (more, if more than two states are involved) interpreting the laws of other states to determine their scope. It is simply not the case, within the two-step model or more generally, that only those with the power to make a law are entitled to interpret it.
Like other interpretive questions, the question of scope has to be answered in order to decide a case, and like other interpretive questions, it is a question about the content and meaning of the law. That means that some interpreters are authoritative and others are not, but it does not mean that what nonauthoritative interpreters do is not interpretation. Courts interpret the laws of other states.65 Scholars, too, can examine laws and argue for sensible interpretations. They can offer presumptions as to the scope of certain kinds of laws. They might consider, in the tort context, whether the primary purpose of the law is to deter wrongful conduct or to compensate injured parties. They can look to see if courts have given convergent interpretations of certain kinds of laws.66 They can recommend those convergent interpretations as best practices, to be adopted by state courts that find them appealing—in fact, that is what a Restatement does.67 Lots of different actors within the legal system do similar things all the time. That is not something that needs to be justified by lawmaking authority; it is a commonplace feature of our legal system.
At this point, it may also help to look at the actual issues. We will focus here on torts. What are the actual questions about the scope of state tort law? Basically, these questions distill down to what limits exist based on domicile or the location of the tort. Everyone agrees that state laws give claims to locals for things that happen inside the state. So the real questions are whether they also give rights to nondomiciliaries for things that happen inside the state, whether they give rights to locals for things that happen outside the state, and whether they give rights to nondomiciliaries for things that happen outside the state.
The Draft Restatement tends to presume relatively broad scope. Presumptively—because of course states can set the scope of their own law—it takes state laws to grant rights to everyone inside their borders. After all, discrimination by withholding rights from visiting out-of-staters is constitutionally suspect.68 It also takes state laws to create rights for domiciliaries outside the state’s borders. States probably want their domiciliaries to be able to benefit from their home law even for out-of-state events if no other law conflicts. (Importantly, that is all that the determination of scope does—it decides that the home law is relevant and available for selection as a rule of decision. If another law conflicts, then a second step, that of assigning priority, is required.69) But states probably do not want to extend the benefits of their law to nondomiciliaries for events outside the state—this is either unconstitutional, if no domiciliary is involved, or what the Reporter for the Restatement (Second) dismissively called making the state’s law “manna for the entire world.”70 These presumptions are supported by the cases of which we are aware.71 Brilmayer and Listwa express consternation that the drafters of the Restatement (Third) engage in “stylized” statutory construction by positing the “likely or generally accepted purposes” of particular laws and presumptively determining scope by reference to those purposes.72 But surveying judicial decisions to identify convergent patterns—what the Draft Restatement means by “generally accepted”—is exactly what a restatement is supposed to do.73 The idea that different kinds of laws have different purposes—in particular, the idea that some are predominantly concerned with regulating conduct and others with allocating losses—is one of the major advances of the last century of choice of law.74
It is possible, of course, that these presumptions will not always accord with the interpretations of state courts. When they do not, the error that the Draft Restatement would make with respect to scope would be extending scope too far. That error will be harmless if it ended up not choosing the law whose scope it has misconstrued. And because the Draft Restatement almost always resolves conflicts via territorial connecting factors, most such errors should in fact be rendered harmless at the second step: the Draft Restatement will choose the law of the place of the tort, which will include the case within its scope, rather than the law mistakenly presumed to extend extraterritorially. We can think of one significant category of cases in which that will not happen: cases where co-domiciliaries are involved in a tort outside their home state and the issue is one of loss-allocation. In such cases, the Draft Restatement selects domiciliary law.75 This will be an error if the scope of the state’s law is territorially limited.
Of course, if the law is a statute with an express limitation, the Draft Restatement directs courts that this limitation on scope must be respected and the statute cannot be used to govern extraterritorial torts, so that error will be avoided.76 The remaining possibility is that the scope is limited because the state’s courts have adopted a territorialist approach to choice of law and intended thereby to indicate that the state’s law is territorially limited.
Such cases will probably be relatively rare, but they may well arise. The error may present clearly enough for application of the Draft Restatement’s escape provision—the draft gives selection of a state’s law to govern matters outside its scope as an example of manifest inappropriateness sufficient to trigger the exception.77 Even if it does not, we have resolved these cases this way (selecting codomiciliary law despite the possibility of territorial scope) for two main reasons. First, absent explicit statement from the state’s courts, it is not possible to determine whether the territorial approach to choice of law means that the state’s law is territorially limited (a question of scope) or simply that conflicts are resolved by reference to territorial connecting factors (a question of priority). Second, it is not the practice of courts using modern approaches to take other states’ adoption of territoriality to indicate a binding restriction on the scope of their law.78 A restatement is, after all, supposed to restate, and so we have taken the practice of courts as our primary source of guidance. We seek primarily to explain what courts are doing and to allow them to continue to reach the same results more simply, easily, and consistently. This is, fundamentally, what the ALI handbook tells reporters to do.79
Last, Brilmayer and Listwa criticize the escape provision of the Draft Restatement in light of the draft’s view that states must respect other states’ statements about the scope of those other states’ laws. Their claim, as best we understand it, is that by suggesting that judges may depart from the Restatement’s rules if those rules select a law that by its terms excludes the parties or events, the escape provision creates “a time-consuming and error-prone process.”80
There are, we think, three points to be made in response. First, the inclusion of some kind of escape provision is an almost uniform feature of modern codifications.81 To the extent that Brilmayer and Listwa object to the existence of such a clause in general, they are going against virtually uniform practice. The form of such clauses varies; the Draft Restatement’s version has been drafted with care based on the experiences with other codifications and it gives guidance to users derived from those experiences.82
Second, if Brilmayer and Listwa’s concern is with the suggestion that divergent scope warrants invocation of the escape clause as a theoretical matter, they are taking the position that judges should be able to disregard express limitations on the scope of their own or sister-state law. As noted above,83 this position is contrary to every case of which we are aware, and Brilmayer and Listwa make no normative or theoretical argument in support of it. Rejecting uniform authority on the basis of nothing is not an appropriate approach for a restatement.
Third, if Brilmayer and Listwa object only that determining whether the scope of a state’s law excludes a particular case is too difficult,84 we think their concern is overblown. First, while it might be difficult to determine the scope of a state statute by looking at the state’s choice-of-law decisions, the Draft Restatement does not tell courts to do so, in part for the reason that such decisions may be ambiguous or unclear.85 Second, however difficult it might be to determine the scope of a state statute by reading the statute and perhaps some decisions interpreting it, courts have proven themselves capable of doing so.86 There are plenty of cases out there in which courts interpret sister-state law to determine its scope, distinguishing that enterprise from choice of law and noting that they cannot use a state’s law to attach legal consequences to persons or events outside its scope.87 What the Draft Restatement directs with respect to divergent scope is not some new and impossibly difficult task; it is a fairly simple process that courts already perform. Again, we do not see how ignoring that uniform practice would be appropriate for a restatement.
Brilmayer and Listwa’s commentary is thought-provoking but in the end it leaves us somewhat puzzled. They claim that the Draft Restatement has the same form as the Restatement (Second) (the continuity of the alleged rules/exceptions structure) but different content (outcomes derived from policy analysis without regard for systemic values). This is exactly backwards: the Draft Restatement has the same content (its rules direct the results on which courts using the Restatement (Second) have converged) but a different form (it has rules rather than standards).
Brilmayer and Listwa also suggest some inconsistency between the use of rules and the Restatement (Third)’s overarching theoretical framework and goals. We see no such inconsistency. What we are trying to do with the Restatement (Third) is two things. First, to bring greater predictability to choice of law through the more determinate rules. Second, to explain how choice of law generally works so that the commonalities between different approaches can be discerned and the field made more intelligible, particularly to nonspecialists. The theoretical perspective adopted for this purpose, the two-step model, is sufficiently abstract to accommodate all approaches of which we are aware (which is why it is appropriate for a restatement), and it aligns with ordinary legal practice. It says that courts must first figure out which states are trying to regulate (which states include the facts of the case within the scope of their law), and then, if conflicts exist, resolve them. The two-step model by itself does not direct any particular results; instead, those results are drawn from judicial decisions applying the model.88 The results the Draft Restatement sets out in rule form respect state policies (the modern insight) while also serving the systemic values of uniformity, simplicity, and predictability (the benefit of rules). If the cases did not converge, this would not be possible, but they do converge.
Admittedly, the Restatement (Third)’s Reporters are not imbued with the authority to make the scope of state laws. But this does not mean the Restatement (Third) cannot suggest rules based on presumptive interpretations, particularly when these interpretations are derived from patterns of decided cases.89 When the courts converge on particular answers in particular categories of cases, as our research thus far suggests, we believe we are doing a great service by encapsulating those answers in a rule that can be applied easily, cheaply, predictably, and consistently. In short, we believe the Draft Restatement (Third) already has successes to write about.90
Kermit Roosevelt III is a Professor of Law at the University of Pennsylvania Law School. Bethan Jones is an associate at Kellogg Hansen Todd Figel & Frederick. The views expressed in this response are those of the authors alone.