Jurisdiction and the Moral Impact Theory of Law

Abstract Positivists and interpretivists (Dworkinians) might accept that conceptual facts about the law—facts about the content of the concept of law—can obtain in the absence of communities with law practices. But they would deny that legal facts can obtain in such communities’ absence. Under the moral impact theory, by contrast, legal facts can precede all communities with law practices. I identify a set of legal facts in private international law—the law of jurisdiction—that concerns when a community's law practices can, and cannot, have the legal effects that the practices claim to have. This law is noncommunitarian, in the sense that it precedes the communities to which it applies. In this law's light, the legal effects of communities’ law practices are legally coordinated (or, at the very least, can be shown to legally conflict). Although interest in, and even commitment to, a noncommunitarian law of jurisdiction has receded among private international law theorists, I argue that some well-placed questions can elicit from all of us a commitment to this law. And this commitment is a reason to believe that the moral impact theory is correct.


Introduction
Anglophone philosophy of law has primarily revolved around the bridge principles that determine how legal effects are generated by events in a community's law practices. For the positivist, bridge principles are based solely in social facts about the law practices of the community at issue. 1 For the interpretivist (or Dworkinian), they are based in the interpretive confluence of such social facts and moral facts. One determines a community's bridge principles by arriving at the best moral justification of its law practices, in a manner that must fit, and so is constrained by, those practices. 2 Under both positivism and interpretivism, each community with law practices has its own bridge principles. There are no overarching bridge principles, independent of all communities' law practices, that take all communities' law practices within their scope. In the light of such overarching bridge principles, the legal effects of various communities' law practices would be legally coordinated or, at least, could be said to legally conflict. But positivism and interpretivism are pluralistic theories of law: law necessarily consists of separate and legally incommensurable systems.
Under the moral impact theory, by contrast, the legal effects generated by events in a community's law practices are the events' moral effects-the way that the events, when occurring in the context of those practices, change people's "moral profiles" (that is, their moral rights, duties, privileges, and powers). 3 Moral impact bridge principles are based solely in moral facts: they are the moral principles that determine the moral effects of any possible community's law practices. It follows that all communities have the same bridge principles. In their light, the moral (and so legal) effects of events in all possible communities' law practices are morally (and so legally) coordinated-or, at the very least, can be said to morally (and so legally) conflict. The moral impact theory is a monistic theory of law: there is necessarily only one moral (and so legal) system.
Although it is not represented among Anglophone philosophers of law, there is a fourth option. Under this monistic theory-which I believe is exemplified, if imperfectly, in the writings of Hans Kelsen-there are overarching bridge principles, independent of all communities, in the light of which the legal effects of events in communities' law practices are legally coordinated (or legally conflict). 4 But these principles consist of irreducibly legal facts. They cannot be cashed out in terms of social or moral facts. 5 The Anglophone debate between positivists, interpretivists, and moral impact theorists has focused on whether the legal effects of events within some community's law practices can be explained in terms of each theory's favored bridge principles. 6 But there is a more direct way to argue for monistic theories of law-to point to legal facts that obtain independently of any community's law practices.
Under positivism and interpretivism, there are no legal facts in the absence of a community with law practices. Monistic theories of law like the moral impact theory, by contrast, can accommodate legal facts that are independent of all communities and so can be understood as preceding them. 7 True, as the moral impact theory is usually formulated, legal facts need a community's law practices to exist. Law, under the official formulation of the theory, consists of the changes in people's moral profiles generated by a community's law practices. 8 Without such practices in place, there are no changes in moral profiles and so no legal facts.
But under the moral impact theory, fundamental moral principles determine how a community's law practices change people's moral profiles. Because these principles are independent of the communities to which they apply, moral facts that necessarily follow from these principles-including facts about how communities' law practices would (or would not) change people's moral profiles-also precede those communities. And moral impact theorists are free to describe these preexisting moral facts as legal.
Thus, if people are committed to legal facts that precede all communities-facts about how the legal effects of various communities' law practices are legally coordinated or legally conflict-that is reason to believe that a monistic theory of law like the moral impact theory (or Kelsenianism) is true.
Anyone familiar with the history of private international law in the United States would concede that people certainly were committed to such legal facts. In the nineteenth and early twentieth centuries American private international law theorists believed in a body of law, the law of jurisdiction, that was noncommunitarian, in the sense that it preceded all communities' law practices. 9 This law determined when a community's law practices could have legal effects. In its light the laws of all past, present, and future communities were legally coordinated-or their legal conflicts were revealed.
For example, imagine that Rex (the king of Ruritania), in the context of Ruritanian law practices, announces that members of another community, Borduria, should drive on the left while they are in Borduria. Under the law of jurisdiction, Rex's announcement cannot generate any legal effects on the Bordurians, because they are insufficiently connected to the Ruritanian community. This legal fact was understood as preceding Ruritanian, or any other community's, law practices-including our own (American) law practices and the law practices of a community of nations.
Past commitment to a noncommunitarian law of jurisdiction might be dismissed, however, if we are not committed to such law now. And it is true that attention to the law of jurisdiction among private international law scholars-and even acknowledgment of its existence-have receded. Their focus is now on positive conflicts law: each community's law of choice of law and the recognition of foreign judgments that is generated by its law practices. But I think I can, with some well-placed questions, show that we are all still committed to a law of jurisdiction.
One goal of this article is to reawaken interest in the law of jurisdiction in the field of private international law. But my primary goal is to explore its consequences for the philosophy of law. Speaking personally, I suspect the conclusion to draw is Kelsenianism, for I worry that the law of jurisdiction cannot be reduced to moral facts. But in this article, I will concentrate on how the law of jurisdiction can be used to defend the moral impact theory. 9 I rely on American private international law theorists because I am familiar with their work. But the same monistic views can be found outside the United States. ROXANA BANU, NINETEENTH-CENTURY PERSPECTIVES ON  My argument will be organized as follows. I start by describing the law of jurisdiction (at a very high level of abstraction), particularly as seen by American jurists in the nineteenth and early twentieth centuries. 10 This historical discussion substantially increases the article's length. But it is essential-and not merely in providing an example of past commitment to the law of jurisdiction. Because past private international law theorists were more sensitive to the distinction between the law of jurisdiction and positive conflicts law, their views are helpful in identifying the former. As they saw it, the law of jurisdiction is background law, in the light of which all positive law, including conflicts law, can be generated by a community's law practices.
I then argue that a commitment to the law of jurisdiction can, with the right questions, be elicited from us now. 11 For my part, I am committed to this law. But it is up to the reader to determine whether she agrees.
I then show how a commitment to the law of jurisdiction supports the moral impact theory, for the theory easily explains the law's existence and at least some views about its content. 12 I follow this with a discussion of whether positivist and interpretivist theories of law can accommodate the commitment. 13 Although I do not think they can, I do not purport to settle the matter. But I do think that the effect of such accommodation would be a significant transformation in communitarian theories of law as they are usually understood.

I. What the Law of Jurisdiction Is
The law of jurisdiction has two main elements. The first is an account of when a community is a "state," in the sense that it possesses a threshold capacity for law. 14 One might point to this element as sufficient to argue against communitarian theories of law. If facts about the requirements a community must satisfy to have law are legal facts (rather than being conceptual facts about the concept of law), monism follows: all past, present, and possible communities' laws are participants in an overarching legal system that precedes any community's law practices.
Despite its attractions, I will not offer such an argument here. My argument will instead rely on the second and more detailed element of the law of jurisdiction, which concerns when subjects have sufficient connection to a state such that an event, occurring within that state's law practices, can generate legal effects on those subjects. Because a state's citizens and officials have sufficient connections to the state to satisfy the law of jurisdiction, interesting jurisdictional questions usually arise only in an international context. It is for this reason that the law of jurisdiction has primarily been investigated in international law scholarship-particularly private international 10 Sections I and II. 11 Section III. 12 Section IV. 13 Sections V and VI. 14 It is usually thought that a state must have a defined territory generally subject to its authorities' control. RESTATEMENT (FIRST) OF CONFLICT OF LAWS (hereinafter RESTATEMENT (FIRST)) §2; RESTATEMENT (SECOND) OF CONFLICT OF LAWS (hereinafter RESTATEMENT (SECOND)) §3. Although there is good reason to be skeptical about this requirement, we can assume it here. law-even though it is necessarily implicated, albeit easily satisfied, in a fully domestic context as well.
If the law of jurisdiction is not satisfied concerning certain subjects, it is impossible for the event to have a legal effect on them. If it is satisfied, legal effect is possible, although it might still be the case that the event, when viewed in the light of the state's law practices, fails to have any legal effect on them. For example, if Rex's announcement is that Ruritanians in Ruritania ought to drive on the left side of the road, it would satisfy the law of jurisdiction, but it might still fail to have a legal effect because it follows from Ruritanian law practices that Rex is not authorized to regulate traffic matters.

A. Legislative Jurisdiction
To repeat, the law of jurisdiction concerns whether subjects have sufficient connection to a state such that an event within the state's law practices can generate legal effects on them. These events can be divided into two main types. First, there are those purporting to create general prospective legal regulation of subjects. Let us call these regulations laws and those who create them lawmakers. 15 Second, there are those purporting to create judgments, that is, particularized sets of legal obligations on parties as a resolution of their concrete dispute. Judgments are created by courts. 16 Legislative (or prescriptive) jurisdiction concerns when the subjects of an event that purports to create a law have sufficient connection to the state within whose law practices the event occurred. In the connection's absence, the event cannot have a legal effect on those subjects.
There are three general views about legislative jurisdiction. According to the traditional approach, it is always exclusive. There is only one state with legislative jurisdiction over a subject concerning a transaction, generally the state where the transaction occurred. 17 Assume, for example, that while they are in France, Jones (a New Yorker) enters into an oral agreement to perform services in France for Smith (a New Yorker), in return for $100. Also assume that under New York law contracts for services must be in writing, whereas under French law they can be entered into orally. 18 As traditionalists see it, French law exclusively applies to the agreement, making it a binding contract. The fact that the parties are New Yorkers does not permit New York to extend its law to the agreement.
Under the traditional approach, being told that the agreement between Smith and Jones is under the legislative jurisdiction of France answers the question of the parties' legal obligations. If France for some reason refused to extend its law to the agreement, These are assumptions only. I make no claims about whether this is in fact the law of New York or France. no other state would be able to fill in the gap. Indeed, to keep legal voids from arising, traditionalists generally assumed that legal regulation is nondiscretionary. A state's lawmakers must legally regulate everything within their sphere of lawmaking power. 19 In the absence of enacted law prohibiting or requiring beekeeping in France, necessarily beekeeping is legally permitted.
The second (modern) approach to legislative jurisdiction sees lawmaking power as concurrent in cases in which the transaction or its subjects have connections with a number of states. 20 Both New York and France can extend their law to the agreement between Smith and Jones-France because it is where the agreement was entered into and New York because it is the parties' domicile. 21 Of course, there are many views about the types of connection that give states legislative jurisdiction under the modern approach, but we can set aside these details here.
If legislative jurisdiction is concurrent, the exercise of lawmaking power should be discretionary. France might refrain from exercising its power to regulate beekeeping by Frenchmen in New York out of deference to New York's power. Under French law, such beekeeping might be neither legally prohibited, required, nor permitteda legal void might be allowed to exist so that it can be filled with New York law. To figure out which states' laws apply to a transaction, therefore, one must determine which of those states with legislative jurisdiction have chosen to exercise their power. And if more than one has done so, one must employ a rule of priority to resolve conflicts, to the extent that they can be resolved at all.
Notice that it is compatible with modern principles that legislative jurisdiction is exclusive when everything about a transaction and its parties is localized in one state. Moderns would agree that only Borduria has legislative jurisdiction over Bordurians driving in Borduria. By contrast, under the final and most controversial approach, legislative jurisdiction is unlimited-any restriction on a state's lawmaking power is self-imposed. 22 Rex's announcement can have a legal effect on Bordurians in Borduria. In my argument below, I will assume that the unlimited view of legislative jurisdiction is false, saving a discussion of its effect on my argument for the article's conclusion.

B. Personal Jurisdiction
Personal (or adjudicative) jurisdiction concerns when an event within a state's law practices that purports to be a judgment can have a legal effect on the parties. Some requirements for personal jurisdiction (such as reasonable notice 23 ) are independent of the connection between the parties and the forum state. But even if these other requirements are satisfied, a sufficient connection between the parties and the forum state must exist. The primary focus is on the defendant's connections, because adjudicative power over the plaintiff is thought to exist through the consent that occurred when she invoked the court's aid in obtaining relief. 24 Consent is also understood as a way that a court can get personal jurisdiction over a defendant. 25 The difficult question is how a state can get adjudicative power over a nonconsenting defendant.
Concerning personal jurisdiction, there are likewise three general views. The traditional approach is that, at any moment, only one state has personal jurisdiction over a nonconsenting defendant, namely the state where the defendant is at that point. Likewise, at any moment only one state has personal jurisdiction over the property of a defendant-the state where the property is located at that time. 26 Assume that in our example above, Smith paid Jones the $100 required by their agreement, but Jones refused to perform the services. Smith chooses to sue Jones in New York state court for breach of contract. Under the traditional approach, the New York court would have the power to issue a judgment binding on a nonconsenting Jones only by virtue of the presence of his person or property in New York. Furthermore, as long as Jones and all his property remain in New York, a French court would lack the power to issue a binding judgment on him concerning his dispute with Smith, unless he consented, even though their agreement occurred in France and was necessarily governed by French law.
The second (modern) approach allows for concurrent personal jurisdiction, at least concerning adjudicative power over the nonconsenting defendant's person. The fact that the agreement between Jones and Smith took place in France would give France personal jurisdiction over Jones, even if his person and all his property were located in New York at the time of the suit. 27 There are, once again, many views about the types of connection that give a state personal jurisdiction under the modern approach, but we can set aside the details here. 28 The third and most controversial view is that personal jurisdiction is unlimited. There are no restrictions on the ability of the courts of a state to issue judgments binding on a nonconsenting defendant except those imposed by the state itself. 29 Again, I will assume that the unlimited approach to personal jurisdiction is false in my argument below, saving its effect for the end of the article.

C. Enforcement Jurisdiction
A final form of jurisdiction is worth mentioning, although it will not be discussed in detail in what follows. This is enforcement jurisdiction, which concerns the ability of  International Shoe Co. v. Washington, 326 U.S. 310 (1945). 28 For some narrow situations where personal jurisdiction is thought to exist in the absence of any connection between the defendant and the forum state, see RESTATEMENT FOREIGN RELATIONS §423; Tracy L. Troutman, Jurisdiction by Necessity, 21 VAND J. TRANSNAT 'L L. 401 (1988). 29 RESTATEMENT FOREIGN RELATIONS, Introductory Note ("[I]t is not always clear whether the principles governing jurisdiction to adjudicate are applied as requirements of public international law or as principles of national law."). a state's officials to investigate, arrest, prosecute, punish, or otherwise enforce the law. Unlike personal and legislative jurisdiction, which empower actors within a state's law practices to generate legal effects, enforcement jurisdiction instead permits them to engage in coercive acts. Enforcement jurisdiction is generally understood as narrowly territorial: with a few exceptions, a state does not have enforcement jurisdiction beyond its borders, 30 absent the other state's consent. 31

II. What the Law of Jurisdiction is Not
Having discussed what the law of jurisdiction is, let us now consider what it is not. Of particular concern is positive law with which the law of jurisdiction might be confused.

A. It does not Depend on a State's Law Practices
The law of jurisdiction is distinct from what the law practices of a state take it to be. Under the law of jurisdiction, Rex's announcement that Bordurians ought to drive on the left in Borduria fails to generate any legal effects concerning its Bordurian subjects, even if Ruritanian law practices take Rex to have the power to legally regulate everyone in the world.
Traditionalists were clear that the law of jurisdiction did not depend on a state's law practices. Consider Justice Story, who sat on the United States Supreme Court from 1811 to 1845 and was instrumental in developing American conflicts law. Story saw principles of jurisdiction as "axioms" upon which "all reasonings on the subject must necessarily rest." 32 The same is true of Joseph Beale-the Harvard law professor (and first dean of the University of Chicago Law School), who dominated conflicts scholarship in the United States in the early twentieth century. As he put it, the law of jurisdiction is "theoretical law": a "body of principles worked out by the light of reason and by general usage, without special reference to the actual law in any particular state." 33 The noncommunitarian character of the law of jurisdiction for Story and Beale has been widely recognized in the conflicts literature. 34 A particularly vivid example of a commitment to the noncommunitarian character of the law of jurisdiction is Justice Field's famous opinion in Pennoyer v. Neff. 35 Pennoyer involved a challenge in a federal court in Oregon of the validity of a judgment issued by an Oregon state court against a Californian named Neff. Even though the judgment satisfied Oregon law practices, Field concluded that Oregon had insufficient connection to Neff's person and property for the judgment to have a legal effect on him. The point was not, or not only, that the judgment should not be enforced in federal court. Field insisted that the judgment was invalid everywhere, even in the Oregon state court system. The reason was not that the Oregon judgment violated the Fourteenth Amendment Due Process Clause, for the Amendment was not binding on the Oregon state court's actions, which took place before the Amendment's ratification. 36 The law of jurisdiction that Field thought invalidated the Oregon state court's judgment was the law of no state-not federal law, not Oregon law, not even international law that exists by virtue of the custom of nations. 37 It was instead an "elementary principle" 38 tied to "first principles of justice." 39 Because traditionalists thought that the law of jurisdiction was independent of any state's law practices, it is understandable that they were not bothered that their views about this law did not conform to views in other states. Such disagreement did not show that the law of jurisdiction was dependent on or relativized to a state. To the contrary, the fact that people in different states could disagree suggested that they were committed to there being a fact of the matter about whether there is jurisdiction, a fact that is independent of any state's law practices. If the existence of jurisdiction were dependent on or relativized to a state's law practices, there would be nothing for them to disagree about.
The traditional approach is unusual in the emphasis that it places on jurisdiction and its noncommunitarian character. The reason is that exclusivity of jurisdiction generates important consequences for courts addressing conflicts problems, especially problems of choice of law. As Beale saw it, the fact that the agreement between Smith and Jones was subject to the exclusive lawmaking power of France gave every state whose court might be entertaining their dispute a reason (although, as we shall see, not a legal duty) to use French law.
Matters are different if jurisdiction is concurrent. For moderns, recognizing that France has legislative jurisdiction over the agreement between Smith and Jones does not give a court a reason to use French law. After all, New York also has legislative jurisdiction and France may have refrained from exercising its lawmaking power out of deference to New York's. For moderns the focus of the choice-of-law analysis is not on legislative jurisdiction but on the positive law of the various states with such jurisdiction. 40 Because the focus of their choice-of-law inquiry is not on legislative jurisdiction, interest in this law-and even commitment to its existence-has receded. Of particular importance were the attacks of the American legal realists on Beale's noncommunitarian views concerning legislative jurisdiction. 41 My guess is that most conflicts scholars would now deny that there is a noncommunitarian law of 36 Id. at 733. 37 Wendy Collins Perdue, Sin, Scandal, and Substantive Due Process: Personal Jurisdiction and Pennoyer Reconsidered, 62 WASH. L. REV. 479, 505-507 (1987). 38 Pennoyer, 95 U.S. at 722. 39 Id. at 732. 40 It is for this reason that language about legislative jurisdiction was omitted in RESTATEMENT (SECOND). jurisdiction at all. There is nothing but the positive law of conflicts. But they are wrong. As I will later argue, I think that one can, with some well-placed questions, elicit from them-and all of us-a commitment to a noncommunitarian law of jurisdiction.

B. It is not Positive Jurisdictional Law
The law of jurisdiction is the background law determining whether it is possible for a state's law practices to generate legal effects concerning some subjects. It is not itself law generated by such practices, even if the generated law speaks of jurisdiction. I will reserve the term law of jurisdiction for the law that is the focus of my argument and call any law about jurisdiction that is generated by a state's law practices positive jurisdictional law.
For an example of positive jurisdictional law, consider the view, expressed in Ruritanian law practices, that Rex may legally regulate anyone in the world. Because this view is false, Rex's announcement that Bordurians should drive on the left in Borduria cannot generate any legal effect on its Bordurian subjects. But the false view about Rex's legislative jurisdiction would generate a legal effect on Ruritanian officials-who, after all, are connected to the Ruritanian community. They may now have a legal duty to enforce Rex's announcement, as best they can. As this example shows, views about jurisdiction expressed in a community's law practices, even if false, can generate positive jurisdictional law in connection with those who, under the actual law of jurisdiction, are sufficiently connected to the community.
When thinking of the law of jurisdiction, American lawyers will likely envision Supreme Court cases, like International Shoe, 42 that interpret the personal jurisdictional power of American courts under the Due Process Clauses of the Fourteenth and Fifth Amendments. But these decisions (and the Fourteenth and Fifth Amendments themselves) are positive jurisdictional law. To be sure, the Supreme Court's decision in International Shoe expressed modern views about concurrent personal jurisdiction. But it does not matter whether these views are correct, for their goal was to establish the legal duties of American courts in personal jurisdiction cases. Since these courts are sufficiently connected to the American community under any plausible view of the law of jurisdiction, the decisions clearly can have their intended legal effects.
Granted, it is likely that these cases also sought to determine the legal effects of the judgments of American courts on defendants. Assume that the New Yorkers Smith and Jones had entered into their contract in Oregon. As a result of International Shoe, the judgment of an Oregon state court entertaining Smith v. Jones can now bind Jones, even if he did not consent to jurisdiction and his person and all his property are in New York. International Shoe appears to have changed American courts' personal jurisdictional power over defendants.
But such cases do not show that the law of jurisdiction is positive law. The decisions were able to generate these legal effects, even if their views about jurisdiction 42 326 U.S. 310 (1945).
were wrong, because the subjects of the decisions (Americans) had sufficient connection to the United States under the actual law of jurisdiction. 43 It is only when the Supreme Court's decisions seek to determine the personal jurisdictional power of American courts over defendants with more tenuous connections to the United States-that is, situations where the actual law of jurisdiction might conceivably be violated-that the Supreme Court's decisions might fail in their goal. If the United States Supreme Court said that American courts have personal jurisdiction over Uzbeks in Uzbekistan who have no connection to the United States, the decision would not give those courts power over the Uzbeks, although it would still have a legal effect on the American courts themselves.

C. It is not Conflicts Law
The law of jurisdiction is also easily confused with conflicts law, that is, the law of choice of law and the recognition of foreign judgments. Consider our earlier example of the two New Yorkers, Smith and Jones, who enter into an oral agreement in France. Under French law the contract is valid. Under New York law it is not. As Story and Beale saw it, French law necessarily and exclusively applied to their agreement, giving Smith a legal right to Jones's performance. But assume a New York court entertaining their case chose to use New York law instead and, as a result, concluded that Jones has no legal duty to perform. As far as the law of jurisdiction is concerned, its judgment would be binding on the parties, provided that there was personal jurisdiction over them. Any conclusion that the judgment is invalid due to its choice of law would have to be based not in the law of jurisdiction, but in New York law practices. Indeed, according to Story and Beale, not only does a New York court with personal jurisdiction have the power to issue a binding judgment even if it uses New York law, French law practices cannot put upon the New York court a legal duty-even a pro tanto duty-to use French law. 44 Story and Beale gave different explanations for why New York law practices, despite being able to use any legal standard for Smith v. Jones, have a reason to use French law. For Story the reason was comity, 45 which includes a host of extralegal considerations, such as whether New York law would be applied by French courts had the situation been reversed. 46 Beale's answer was more straightforward: the reason is respect for preexisting legal rights. When Jones entered into an agreement with Smith in France, Smith's legal right to Jones's performance arose. That fact-that Smith's legal right had vested-gives New York law practices a reason (but no legal duty) to use French law for the case. 47 43 My argument here depends on a rejection of the dual sovereignty theory, according to which American states are, to some extent, legal communities independent of one another and of the federal community. Although I cannot provide the argument here, I believe that the fact that the Supreme Court's decisions concerning personal jurisdiction, even if in error, can have a legal effect on the citizens of a state is evidence that the dual sovereignty theory is false. 44 Green, supra note 41, at 32-40. 45 STORY, supra note 17, at 8. 46 Id. at 8; see also Donald Earl Childress III, Comity as Conflict: Resituating International Comity as Conflict of Laws, 44 UC DAVIS L. REV. 11, 26-28 (2010).
The same points applied to principles of personal jurisdiction, which did not have direct legal consequences concerning the recognition of foreign judgments. Assume that our New York state court applied New York rather than French law in Smith v. Jones. It therefore held the contract invalid, freeing Jones of any legal obligation to perform. If it had personal jurisdiction, its judgment is binding on the parties. But assume that after the New York judgment, Smith chose to sue Jones again in a French court, and that court had personal jurisdiction (because Jones was now in France or he had property there). The French court, ignoring the previous New York judgment, issues a new judgment applying French law and holding that Jones has a contractual duty to perform. As far as the law of jurisdiction is concerned, the French judgment would also be valid. 48 Any conclusion that the French judgment is invalid due to its refusal to recognize the New York judgment would have to be based not in the law of jurisdiction, but in French law practices. New York law practices cannot give the French court a legal duty to give the New York judgment effect in France. 49 Because, according to Story and Beale, it was up to a state's law practices, not the law of jurisdiction, what rules for choice of law and the recognition of foreign judgments their courts used, they were both clear that conflicts law was the positive law of a state. 50 In distinguishing between the law of jurisdiction and the positive law of conflicts, I have focused on the traditional approach to the law of jurisdiction. But the same distinction applies under the modern approach. Assume two Frenchmen enter into an agreement in France with performance in France. France has exclusive legislative jurisdiction over the agreement under the modern approach. But moderns can still insist that, as far as the law of jurisdiction is concerned, a New York court with personal jurisdiction over the Frenchmen can issue a binding judgment concerning their agreement even if it uses New York law. Legislative jurisdiction does not have direct effects on New York choice-of-law rules. All the law of jurisdiction demands is that the New York court making the choice-of-law decision have personal jurisdiction over the parties. If it does not, its decision is a nullity.
The fact that the principles of legislative jurisdiction do not have direct legal consequences for courts in making choice-of-law decisions can make them seem irrelevant. All that matters, it seems, is the forum's positive law of choice of law. But such a conclusion would be a mistake.
First, even if legislative jurisdiction were irrelevant, that would not show that the law of jurisdiction is irrelevant, for it would remain true that a court making a choice-of-law decision cannot have a legal effect on the parties in the absence of personal jurisdiction. Without personal jurisdiction the court's judgment, whatever its choice of law, is a nullity.
But, more fundamentally, legislative jurisdiction looks irrelevant only if one refuses to draw a distinction between preexisting legal duties created by laws and the 48 Clyde Spillenger, Risk Regulation, Extraterritoriality, and Domicile: The Constitutionalization of American Choice of Law, 1850-1940, 62 UCLA L. REV. 1240, 1269-1271(2015. 49 For Story it was comity and for Beale respect for vested rights that would give the French court a reason to give the New York judgment effect within its borders. Story, supra note 17, at 30, 37. 50 RESTATEMENT (FIRST) § §1, 2; 1 BEALE, supra note 33, at 112-113; Green, supra note 41, at 33-38. particularized legal duties generated by courts' judgments. And we do draw such a distinction.
Assume that under Ruritanian law practices, announcements by Rex create legal obligations on subjects but there are no Ruritanian courts that make particularized judgments about whether these announcements have been violated. (Rex doesn't make such judgments either.) The same point is true about the announcements by Basil (the king of Borduria) under Bordurian law practices. No judgments are created under those practices. Under the law of jurisdiction, an announcement by Rex that Bordurians ought to drive on the left in Borduria would still fail to have a legal effect on the Bordurians. And that can be understood only as concerning legislative jurisdiction.
Furthermore, even with courts, we draw a distinction between preexisting legal obligations created by laws and any particularized legal duties generated by courts' judgments, even if the latter fail to give effect to the former. This is evident in a fully domestic context. Assume that our New Yorkers Smith and Jones had entered into their oral agreement in New York. Under New York law, there is no contract and so Jones has no legal duty to abide by their agreement. But Smith chooses to sue Jones in a New York state court for breach of contract anyway and the court, misinterpreting the facts or the law, issues a judgment against Jones, putting a legal duty on him to perform. The court's decision is mistakenly affirmed on appeal. The following two legal facts both obtain: (1) Jones has no legal obligation to perform under New York contract law, and (2) he has a particularized legal obligation to perform due to the New York court's judgment.
Once we draw a distinction between obligations created by laws and those created by judgments, legislative jurisdiction is a meaningful limitation on a state's law practices. Just as the law of personal jurisdiction is necessarily implicated in every judgment issued by any court, for it is only in the light of this law that the court's judgment can have any legal effect on the parties, so the law of legislative jurisdiction is necessarily implicated in every lawmaking act issued by any lawmaker, for it is only in the light of this law that the act can have any legal effect on its intended subjects.
It is worth recapping at this point the reasons why it is so easy to miss the law of jurisdiction's existence. The first reason, which applies to both personal and legislative jurisdiction, is a tendency among both jurists and philosophers of law to emphasize the legal duties of courts and other officials, who will always have sufficient connection to a state's law practices under the law of jurisdiction. It is hard to notice the law of jurisdiction when one never sees it violated. Furthermore, the law practices of a state will inevitably speak of the officials' obligations concerning jurisdiction (thereby creating positive jurisdictional law), which makes it appear as if all there is to jurisdiction is such positive law.
The second reason, which likewise applies to personal and legislative jurisdiction, is that in most cases the subjects of a state's law practices are domiciliaries of the state who are present within its borders. Again, this means that the law of jurisdiction is trivially satisfied concerning them, making it hard to notice the law's existence. And again, there will be discussion of jurisdiction in the state's law practices (think International Shoe), creating positive jurisdictional law concerning the subjects. This makes it appear as if all there is to jurisdiction is such positive law.
The third reason, which especially applies to legislative jurisdiction, is the tendency (reaching pathological levels among the legal realists 51 ) to emphasize only the particularized legal duties on parties created by judgments, ignoring the preexisting legal duties created by laws. Because legislative jurisdiction does not have direct legal effects on courts making choice-of-law decisions, it starts looking irrelevant-or reducible to the positive law of choice of law. Although this error arises concerning legislative jurisdiction, a similar error can occur concerning personal jurisdiction. If a court forgets its own noncommunitarian obligation to have personal jurisdiction, the law of personal jurisdiction can also look irrelevant, since it has no direct legal effects on a court deciding whether to give foreign judgments domestic legal effect. The law of personal jurisdiction starts looking like nothing more than the forum's positive law concerning the recognition of foreign judgments.

D. It is not Positive International Law
Both Story and Beale spoke of the law of jurisdiction as international law. 52 But such a description likely referred solely to the fact that the law is international in its scope, not that it depended, in whole or in part, on international law practices. Indeed, the fact that both articulated the law of jurisdiction without concern for whether officials in other states agreed is evidence that they did not think of this law as dependent on such practices. The matter is different, of course, concerning conflicts law (that is, the law of choice of law and the recognition of foreign judgments). To the extent that states agreed on conflicts rules, this law could be understood as customary international law. 53 The fact that the jurisdiction of states can be expanded or contracted through international law practices also does not mean that the law of jurisdiction that preceded these changes is positive law. An example is the lawmaking power that Rex might have over Bordurians in Borduria because of Borduria's consent-whether in the form of a treaty between Rex and Basil (the king of Borduria) or because of Borduria's implied consent due to its participation (or acquiescence) in an established practice of nations. 54 To be sure, we might look at such positive international law of jurisdiction in the light of the law of jurisdiction, by asking how international law practices are able to have legal effect on the relevant subjects. We might say, for example, that Rex and Basil's treaty can have a legal effect on its Bordurians subjects because they have a connection to the community of nations (of which Ruritania and Borduria are  REV. 1915REV. , 1985REV. -1993REV. (2005. 52 STORY, supra note 17, at 9; Joseph Beale, The Jurisdiction of a Sovereign State, 36 HARV. L. REV. 241, 241-242 (1923). 53 Green, supra note 41, at 38-39. Thus, Beale says that "[i]t is only by international law, backed by the force of opinion of civilized nations, that the actual power of a sovereign can be restrained within the limits of legal jurisdiction." 1 BEALE, supra note 33, at 117. This suggests that the law of jurisdiction preexists its enforcement through international law. members). But it is the preexisting law of jurisdiction that would be our focus, not any positive international law that was created. The law of jurisdiction is the law standing in the background, in the light of which any community's law practiceseven international law practices-can have legal effect.

E. It is not the Content of the Concept of Law
To repeat, the law of jurisdiction is not created by law practices in a community but is instead the background law that is implicated whenever one determines whether it is possible for a community's law practices to have legal effect concerning some subjects. The law of jurisdiction is seen as necessary-it existed even before human communities with law practices arose. And it is discovered not by attending to communities' law practices, but through a priori reasoning. But heeding this warning brings up another, and more difficult, issue: whether the law of jurisdiction is law at all, rather than being facts about the content of the concept of law. Although Story and Beale tended to speak of the law of jurisdiction as law, an alternative conceptual reading of their views is possible.
Conceptual facts are commonly understood as necessary and so as obtaining in all possible worlds, even in worlds where the items falling under the concept do not exist. Furthermore, these facts are understood as discovered through a priori reasoning. An example is the following fact (F 1 ): If there is an unmarried human male, then there is a bachelor. This fact obtained before there were bachelors and can be known without investigating them. But it would be a mistake to conclude from the necessity and aprioricity of this concept-of-bachelor fact that facts about bachelors can also be necessary and a priori. Bachelor facts-such as the fact that their apartments tend to have many empty pizza boxes-require the existence of actual unmarried human males and are discovered by investigating such males. In their absence, there are no bachelor facts at all.
By the same token, communitarians will argue, simply because there are concept-of-law facts that are necessary and a priori, it does not follow that there are necessary and a priori legal facts. Legal facts cannot exist without an actual community with law practices, and they are discovered by investigating such practices. What I call the law of jurisdiction, if it exists, consists of concept-of-law facts, not legal facts. The demanded connection between the subjects of a lawmaking act and the community within whose law practices the act occurred is part of the content of the concept of law. It is not itself law.
Furthermore, the communitarian will insist, it is a concept-of-law fact that law is necessarily pluralistic. It consists of independent legal systems, each of which essentially depends on a particular community's law practices. The cognitive unity of the law of all communities under the concept of law does not mean that there is some legal unity to all that law, any more than the cognitive unity of all bachelors under the concept of a bachelor means that all bachelors are, in some sense, really one. 55 But consider another fact (F 2 ): If a one makes a promise, then one is morally obligated to keep it. 56 Some consider such facts about fundamental moral principles to obtain in all possible worlds and to be knowable a priori. 57 But F 2 is a moral fact, not a concept-of-morality fact. To be sure, both concept-of-morality facts and fundamental moral facts are seen as obtaining in all possible worlds. For this reason, one cannot distinguish them by pointing to circumstances where one set of facts obtains and the other does not. But that does not mean that there is no difference between them. If we imagine God extinguishing morality, all moral facts, including F 2 , would no longer obtain, while concept-of-morality facts still would. It would be in the light of the concept-of-morality facts that we would recognize morality's absence. Furthermore, as many see it, because all moral obligations-indeed, all moral rights, duties, privileges, and powers-follow from the same set of fundamental moral principles, morality is monistic. There is a moral unity to all possible moral obligations, no matter the community in which they arise.
The question is: Are facts about jurisdiction more like F 1 or F 2 ? Are they necessary and a priori concept-of-law facts about multiple legal systems or necessary and a priori fundamental legal facts about a monistic legal order? I will argue below that they are the latter. But a sign that one has properly identified the law of jurisdiction (rather than positive jurisdictional law) is that one begins worrying that it is not really law at all, but rather part of the content of the concept of law.
III. What Would We Say Now? I personally think past commitment to a noncommunitarian law of jurisdiction is a problem for communitarian theories of law. A theory of law is not adequate if it dismisses an entire generation of legal scholars-including a Supreme Court Justice and a law school dean-as incompetent at identifying law. But my guess is that these past views will be dismissed by communitarians as conceptually confused. It is a matter of importance, therefore, whether people currently believe in such law.
To highlight the law of jurisdiction, we need an example where a community's law practices take themselves to have legal effect but, under the law of jurisdiction, in fact do not. Given that the prevailing view is that jurisdiction is concurrent, we should choose an example where even the more generous requirements of concurrent legislative and personal jurisdiction are violated.
Let's start with legislative jurisdiction. Assume a world without communities with law practices. The very first such community, Ruritania, comes into being. Under Ruritanian law practices, Rex may legally regulate everyone in the world. A year later, a second, somewhat distant, community with law practices-Borduria-arises. Under Bordurian law practices, Basil may regulate everyone in the world. Basil chooses to announce only that people in Borduria ought to drive on the right. Rex announces that everyone, including Bordurians in Borduria, ought to drive on the left. Rex and Basil, although knowing of one another's existence, are not participants in any international law practices with one another. We now consider the following simple question: Do the Bordurians in Borduria have a legal duty to drive on the left 56 This obligation might be only pro tanto, in the sense that it can be defeated by a stronger obligation. (even if they might also have a legal duty to drive on the right)? Is it possible for Rex's announcement to have a legal effect on them?
We are not asking whether Rex's announcement can create legal duties on Ruritanian officials to enforce the announcement. It clearly can, since they have sufficient connection with the Ruritanian community. The question is whether the announcement can create legal duties on its Bordurian subjects.
We are also not concerned with whether the announcement can have a legal effect on a Bordurian who is subject to the personal jurisdiction of a Ruritanian court in a case involving driving in Borduria. In such a case the Bordurian would have sufficient connection with the Ruritanian community to make such a legal effect possible. The question is whether Rex's announcement can create legal duties independently of such an adjudication by a Ruritanian court.
We are also not concerned with whether Rex's announcement can be legally enforced. It is true that Ruritania would not have enforcement jurisdiction, since any coercive action by Ruritanian officials to ensure compliance with Rex's announcement would have to occur within the borders of Borduria. But legal duties can exist in the absence of enforcement jurisdiction. As moderns see it, Rex can make announcements that create legal duties on Ruritanians in Borduria even though Ruritania lacks enforcement jurisdiction to ensure that the announcements are complied with. Our question is whether he can do the same thing concerning the Bordurians.
Finally, our question is not whether the announcement can as a practical matter be enforced against its Bordurian subjects. It is true that when Bordurians continue to drive on the right there is probably nothing that Ruritanian officials will be able to do about it, even if they choose to flout the limits of enforcement jurisdiction. But legal duties can exist in the absence of practical ability to ensure compliance, as the legal duties created by Rex's announcements concerning Ruritanians in Borduria show. Indeed, there might be rough neighborhoods in the capital city of Ruritania where enforcement of Rex's announcements is ineffective. That does not mean that his announcements do not create legal duties in those neighborhoods. Our question, therefore, is whether Rex's announcement can create a similar legal duty on the Bordurians that exists in the absence of effective enforcement.
It is worth emphasizing that our question is whether Rex's announcement would have a legal effect on the Bordurians even if one were a participant in Ruritanian law practices. As a participant, one would say it has a legal effect on the Bordurians, but the question is whether one would be right.
I think the Bordurians have no legal duty to drive on the left. Ruritanians are simply wrong about the legal effect of their law practices, and I would be wrong if I were a participant in those practices. Furthermore, of the dozen or so people-with and without legal training, but with no background in the philosophy of law-whom I have asked the question (in various forms) over the past few years, all have agreed. In explaining why there is no legal effect, they employ principles (like "you can't tell people in other countries what to do") that they appear to have arrived at in an a priori fashion, without reliance on any community's law practices. In denying that the Bordurians have a legal duty to drive on the left, they are not saying that under some community's law practices the announcement will not be recognized as having a legal effect. The absence of legal effect does not depend on any community's law practices. They simply think that the Bordurians have no such legal duty, full stop.
It is worth noting that someone can deny that Rex's announcement has any legal effect on Bordurians in Borduria, while saying that "under Ruritanian law" it has such a legal effect, meaning that Ruritanian law practices take it to have such an effect. As an analogy, assume that, according to the moral views of Ruritanians, people ought not to touch chickens. It would not be unusual for someone to say that "under Ruritanian morality" people ought not to touch chickens. Such a locution is compatible with her conclusion that Ruritanian morality has no actual moral effect. And if she does conclude it has moral effect-for example, that one ought not to touch chickens in Ruritanians' presence-that could be due to moral principles (for example, of tolerance) that do not depend on the Ruritanian, or any other, community. By the same token, one cannot seize on people's talk about Bordurians' duties "under Ruritanian law" to show that they reject the law of jurisdiction. One must ask them what they think the legal effect of this Ruritanian law is. And I think their response will be that it has no legal effect (on the Bordurians) at all.
I have used a highly abstract hypothetical, rather than one involving, say, the United States and Uzbekistan. The reason is that if I spoke of an American law telling Uzbeks in Uzbekistan what side of the road to drive on, one might be inclined to answer the question taking into account actual American law practices or actual international law practices in which the United States and Uzbekistan participate. It might follow from these law practices that the American law has no legal effect even on American officials or on an Uzbek defendant subject to personal jurisdiction in an American court-that is, even when the law of jurisdiction is satisfied. Although our response to such an example, like our response to all legal scenarios, would implicitly rely on the law of jurisdiction, it would not help highlight the law's existence.
Let us now move on to personal jurisdiction. Here I will consider a case in which not merely modern views about concurrent personal jurisdiction are violated, but also the requirement of reasonable notice. Assume that a Ruritanian court hears about a dispute between a Bordurian tenant and her Bordurian landlord and, on its own initiative, decides to adjudicate the matter. Everything about the dispute arose in Borduria. The Bordurians have never been to Ruritania. No notice is given to the parties. Given that they both fail to appear, the court rules for the tenant and issues a default judgment for (the Ruritanian equivalent of) $100,000. Assume that everything the Ruritanian court did is in conformity with Ruritanian law practices. Under those practices, the judgment is taken to create a legal debt. Once again, we ask: Does the landlord legally owe the tenant $100,000? Is it possible for the Ruritanian judgment to have such a legal effect?
Again, the question is not about whether Ruritanian officials have a legal duty to enforce the judgment, as best they can, or whether the judgment will have a legal effect on the Bordurian landlord if he is subsequently subject to the personal jurisdiction of a Ruritanian court. The question is whether the judgment has a legal effect on the landlord in the absence of any of these connections to Ruritania. Nor is the question whether the judgment can be legally or practically enforced. The question is whether there is a legal effect in the absence of such enforcement.
Here too I think there is no legal debt and I've found that the small number of people I have asked agree. Ruritanian law practices are simply wrong about the judgment's legal effect, and I would be wrong if I were a participant in those practices. In saying that the judgment is without legal effect, we are expressing the same view as Justice Field in Pennoyer v. Neff. The fact that the judgment cannot have a legal effect on the landlord obtains everywhere-even in Ruritania-whatever Ruritanian law practices might say about the matter. In my experience, people are just as committed to a noncommunitarian law of jurisdiction as Justice Story and Joseph Beale. But it is up to the reader whether she agrees.
In what follows, I will assume that people have this commitment to the law of jurisdiction (which I will call the commitment). Our question is the consequences the commitment has in the philosophy of law. My primary emphasis will be on how, under the commitment, a community's law practices-even one's own community's law practices-can be wrong about their ability to have legal effect. The details of the connections required for legal effect (such as whether the traditional or the modern view is correct) will be left open.
The idea that a community's law practices can have no legal effect, even though they take themselves to, captures only the most basic way in which the legal effect of a community's law practices can diverge from what those practices say they are. The law of jurisdiction contains more detailed and contested aspects of how, when communities' law practices can generate legal effects, these effects are legally coordinated or legally conflict. Traditionalists understood exclusive spheres of personal and legislative jurisdiction as making legal conflict impossible. 58 Modern theories of concurrent jurisdiction appear to allow for legal conflicts, although one might argue that there are rules of priority in the light of which these conflicts are only apparent. The commitment takes there to be answers to these questions too, although we do not have to take a stand on what they are.
Although, as I put it, the question is the extent to which theories of law can explain a commitment to the law of jurisdiction, it is not necessary that a theory of law explain the commitment by showing that there is such law. For example, communitarian theories might accommodate the commitment by showing that it concerns the concept of law instead.
Because it can easily explain why there is indeed a law of jurisdiction, as well as predict at least some modern views about this law, I will start with the moral impact theory and then move on to positivism and interpretivism.

IV. The Moral Impact Theory
Under the moral impact theory, the legal effects generated by events in a community's law practices are the events' moral effects-the way that the events, when occurring in the context of those practices, change people's moral rights, duties, privileges, and powers. 59 The main challenge with understanding legal effects as moral effects is that legal standards can be morally arbitrary or even morally incorrect. But we are all aware of circumstances in which social facts create moral obligations to conform to morally arbitrary or incorrect standards. Although morality does not dictate what side of the road to drive on, if other people drive on the right, that social fact can give us a moral obligation to do the same (as a means of satisfying our moral duty to avoid harm). Or, having mistakenly used a morally incorrect standard for her son, a parent might be morally obligated by considerations of fairness to use the same standard for her daughter.
Greenberg's theory takes advantage of the varied and subtle ways that the use of morally arbitrary or incorrect standards in a community's law practices can generate moral effects. For example, in the absence of law practices concerning the enforcement of promises, someone entering into an agreement would often be uncertain about her moral duty of repair if she fails to keep her promise. Any standard of repair seems morally arbitrary. With law practices concerning contract law in place, however, the agreement will take place against the background of its standards, making her moral duty in the event of a breach more definite. 60 According to Greenberg, people's contractual obligations consist of the way that a community's law practices change their moral promissory obligations in this fashion. More generally, he argues that all legal facts consist of the moral changes similarly effected by law practices. 61 It is not my goal here to describe the details of the moral impact theory or to discuss Greenberg's arguments for the theory. For my purposes, it is enough to note how the theory can easily account for the commitment.
It is true that Greenberg presents his theory as if legal facts cannot obtain in the absence of a community's law practices. If law consists of changes in people's moral profiles generated by a community's law practices, law cannot exist in the absence of the relevant community. But he probably adopted this formulation because he, like other philosophers of law, was unaware of the noncommunitarian law of jurisdiction. 62 Furthermore, unlike communitarians, he can easily accommodate this law.
Greenberg's theory relies on fundamental moral facts, which do not depend on any community's law practices, concerning how any possible community's law practices can generate moral effects. 63 These preexisting moral facts are the theory's bridge principles, which take one from social facts about a community's law practices to their moral (and so legal) effects. Necessarily following from these bridge principles are moral facts about when the connections between subjects and a community are 59 Greenberg, supra note 3, at 1308-1319. 60 Id. at 1311-1312. 61 Id. at 1308-1319. 62 Mark Greenberg, How Facts Make Law, 10 LEGAL THEORY 157, 170 (2004) (describing as "uncontroversial" the view that "law practices are among the determinants of the content of the law," by which he likely means that such practices are always among the determinants, for all law). 63 Technically, the relevant facts are not solely moral. They consist of the conjunction of fundamental moral facts and contingent facts about human beings, such as their usual dispositions and the like. But what is important is that they do not depend on contingent facts about any community's law practices. I ignore this wrinkle. sufficient (or insufficient) for its law practices to possibly generate moral effects. And nothing keeps Greenberg from identifying these preexisting moral facts as legal.
Consider two Bordurians who enter into an agreement in Borduria with performance to occur in Borduria. Because their agreement will not take place in the background of Ruritanian law practices concerning contracts, these law practices cannot have an impact on their moral profiles. To recognize this moral fact is, in effect, to accept a principle of legislative jurisdiction, independent of any community's law practices. Ruritanian law practices cannot have a moral (and so legal) effect even if they take themselves to have such an effect.
Or consider Basil's announcement that people should drive on the right in Borduria. It is a moral fact, preexisting Borduria and all other communities, that the announcement can have a moral effect on its subjects. By solving the coordination problem, Basil can give them a moral duty to drive on the right (as a means of satisfying their moral duty to avoid harm to others). Of course, the fact that his announcement can have a moral effect does not mean that it does. When one considers more details about Bordurian law practices-such as the fact that Basil is not taken to have the power to regulate traffic matters-one might conclude that no moral effect can occur. But that is precisely how the law of jurisdiction is understood: it is a threshold determination of whether there is sufficient connection between subjects and a community's law practices for the practices to possibly have a legal effect on them.
Rex's announcement that Bordurians in Borduria should drive on the left, by contrast, cannot solve the coordination problem and so cannot have a moral effect on its subjects. It cannot have such an effect even if Ruritanian law practices take it to have such an effect. The moral impact theory can explain why Rex's lack of legislative jurisdiction is independent of Ruritanian, or any other community's, law practices.
But couldn't one argue that Rex's announcement, if widely and effectively communicated to Bordurians, would provide an alternative solution to the coordination problem and so could also have a moral effect on them? In one respect, the fact that our intuitions might shift in the light of these details is not a problem for the moral impact theorist. The theory is still able to explain a noncommunitarian law of jurisdiction, for our new conclusion that Rex possesses legislative jurisdiction still does not depend on Ruritanian law practices, but rather on moral facts concerning his ability to generate moral effects on the Bordurians. That said, if people remained committed to Rex not having legislative jurisdiction, even though a story about possible moral effect could be told, that would be a reason to question the moral impact theory. 64 This is simply one example of a more general vulnerability of the theory, which is that the legal effects of a community's law practices might diverge from their moral effects. If we came to the conclusion that the moral impact theory could not account for the law of jurisdiction, while still understanding this law as 64 That is not to say that Greenberg might not have a response, for example, by reference to the requirement that the moral effect be in the "legally proper" way. See, e.g., Greenberg, supra note 3, at 1323 n.72. Thanks to an anonymous reader for Legal Theory for identifying this possibility. independent of communities' law practices, we would be forced into a Kelsenian position, under which the law of jurisdiction consists of irreducibly legal facts.
But the moral impact theory predicts many views about jurisdictional limits, including modern views that jurisdiction is concurrent. Since the agreement entered into in France by the New Yorkers Jones and Smith can have occurred within the background of either New York or French law practices, both states would have legislative jurisdiction over their agreement. And as for our uncertainty about whether concurrent jurisdiction generates legal conflicts or whether competing legal effects are necessarily legally coordinated, the moral impact theory predicts that as well. The problem is the familiar one of whether moral dilemmas are possible 65 -in this case, moral dilemmas generated by the conflicting moral effects of events in different communities.
In addition, the moral impact theory predicts the distinction I drew earlier between the noncommunitarian law of legislative jurisdiction and the positive law of choice of law. 66 Assume a case with exclusive legislative jurisdiction under the moral impact theory, such as an exchange of promises between two Frenchmen in France with performance to occur in France. One of the Frenchmen fails to keep his promise, creating a dispute between them. Now assume a New York court gets personal jurisdiction over the Frenchmen-say, because they are in New York at the time of the litigation. A judgment by the New York court would change the moral profiles of the Frenchmen, even if the court failed to apply French law. That means that, under the moral impact theory, a court with personal jurisdiction can create a legally valid judgment despite its failure to apply the law of the state with exclusive legislative jurisdiction. The law of legislative jurisdiction does not determine valid law of choice of law. The same points apply, mutatis mutandis, concerning the distinction between the law of personal jurisdiction and positive law for the recognition of foreign judgments.
I have argued that the moral impact theory can explain the commitment, because under the theory a single set of bridge principles-consisting of fundamental moral facts-determine the legal effects of all possible communities' law practices. These principles, being necessary, precede the communities to which they apply. One might argue, however, that the moral impact theory can explain the commitment only with the addition of a controversial metaethical premise, namely moral realism, according to which moral statements describe states of affairs existing independently of human beings and their beliefs and attitudes. 67 Under my argument, Greenberg can explain why Rex's announcement has no legal effect, full stop (that is, independently of any contingent fact, including a community's practices and one's own views), only if it assumes the metaethical view that Rex's announcement has no moral effect, full stop (that is, independently of any contingent fact, including a community's practices and one's own views).
What if moral realism is false, and there are no "full stop" moral effects? Moral statements might describe states of affairs that fundamentally depend on human beings, 68 or even one's community's practices. 69 Or they might not describe states of affairs at all, but instead express certain noncognitive states of the speaker-including, possibly, her acceptance of her community's practices. 70 Although, when presenting his theory, Greenberg tends to assume moral realism, 71 he has argued that his theory is metaethically agnostic. 72 He can insist that only moral facts can adequately explain how a community's law practices generate legal effects, while leaving the metaethical status of those moral facts open. In general, metaethical positions do not purport to change people's everyday, or first-order, moral judgments. And it is only these first-order judgments -about the moral effects of a community's law practices-that his theory needs. 73 One might worry that my argument is more metaethically vulnerable than Greenberg's, however. Since he ignores inter-community scenarios, he can remain agnostic about the extent to which the moral principles that generate legal effects are, in some way, anti-realistically dependent on a community. My argument that the moral impact theory can explain the commitment, by contrast, must understand moral principles as metaethically transcending these communities.
Although I cannot explore the matter fully here, I think my argument can be as metaethically agnostic as Greenberg's. Our conclusion that Rex's announcement cannot have a moral effect on the Bordurians, full stop, is not a metaethical position but is a first-order moral judgment that this moral effect does not morally depend on contingent facts, such as a community's practices or one's own views. Such first-order moral judgments about the moral independence of moral effect from contingent facts is something any competent metaethical theory should be able to accommodate. 74 And it is only such first-order moral judgments that my argument needs. One way of putting this point is that the necessity of the fundamental moral 68 E.g., Mark Johnston, Explanation, Response-Dependence and Judgement-Dependence, in RESPONSE-DEPENDENT CONCEPTS 122 (P. Menzies ed., 1991). 69 The early H.L.A. Hart may have been an example. Even though he distinguishes legal and moral obligation, when he offers a generic communitarian account of obligation, the account appears to be applicable to moral obligation as well. H.L.A. HART, THE CONCEPT OF LAW (2d ed. 1994), at 83-88. True, an error theory might change one's first-order moral judgments by leading one to conclude that they should be abandoned. But even then, Greenberg's argument that only moral facts can explain the legal effect of a community's law practices would be unaffected. Combined with the error theory, one would conclude that both law and morality do not exist. principles on which my argument relies is a moral necessity-about the principles' moral independence from contingent facts. It is not a metaethical necessity, about which I can remain agnostic.

V. Positivism
Let us now consider the extent to which the commitment can be accommodated by communitarian theories of law. I will begin by offering an account of what I take to be the essence of positivist theories such as those offered by H.L.A. Hart and Scott Shapiro, emphasizing only those elements that are necessary to discuss the place that the commitment would have in them. Besides being limited in this way, my account will also be unusual in taking especial care to distinguish between the role that social facts about a community's law practices play in positivist theories and the role played by abstract objects. 75

A. The Basics of Legal Positivism
Let us understand a norm to be any standard that can be used to guide conduct. 76 An example of a norm (call it N bleach ) is one-ought-to-drink-bleach-every-morning. Norms, so understood, are abstract objects 77 whose existence does not necessarily give anyone a justifying reason for action. After all, there is-and always has been-an infinite number of norms, many of which (like N bleach ) would lead us to do crazy or evil things if we chose to be guided by them.
Let us understand a legal system as a complex abstract object based on a norm that I will call a metanorm. 78 Under a metanorm certain identified people engaging in norm-applying law practices (call them norm-appliers) ought to use norms that satisfy certain criteria. In addition, the subjects of the norms identified by the metanorm (call them norm-subjects) ought to conform to those norms. At least some of the criteria in the metanorm will refer to the decisions of certain people (call them norm-identifiers), although the criteria might also include other considerations, such as the norms' content. An example of a metanorm is MN Rex , according to which all norm-appliers ought to use the norms announced by Rex and all subjects of those norms ought to conform to them. 79 For each legal system, the laws of the system are all the norms that satisfy the criteria in the system's metanorm. Notice that these facts about a legal system are not social. They are necessary truths concerning the abstract object that is the legal system. It is a necessary truth about the legal system based in MN Rex Here I hope to accommodate what is common to Hart's idea of a rule of recognition (as well as other foundational secondary rules-such as rules of change and adjudication), HART, supra note 69, at 94, and Shapiro's idea of a master plan, SHAPIRO, supra note 76, at 166. 79 Of course, we are likely leaving out many details in MN Rex . What should norm-appliers do if Rex dies or abdicates or if a matter is not covered by one of his announcements? But we can set them aside here. that if Rex announces that people ought to drive on the left, norm-appliers ought to apply that norm in their law practices and people ought to conform to that norm.
Where social facts about a community come into play for positivists is in determining what legal system is assigned to a community, such that it is the legal system of that community. Very roughly, a legal system is assigned to a community if two main conditions are satisfied. First, the system's metanorm must be accepted by the community's norm-appliers. 80 Second, the norms identified as laws by the metanorm must generally (although not necessarily perfectly) be conformed to by the community's members. It is not necessary that these nonofficial members accept the identified norms. They may conform, say, solely to avoid sanctions for disobedience. When these two criteria are satisfied, the legal system is assigned to the community and all the norms identified by the system's metanorm are the laws of the community.
The abstract object MN Rex is the bridge principle that takes one from Rex's announcement that people ought to drive on the left to people's legal obligation to drive on the left. And since MN Rex is relevant to Ruritanian law practices solely because of social facts about those law practices (namely because MN Rex is accepted by Ruritanian norm-appliers and the norms it identifies are largely conformed to by Ruritanians), we can say that Ruritanian law ultimately depends solely on Ruritanian law practices. By the same token, it is because another metanorm, MN Basil , is accepted by Bordurian norm-appliers and the norms identified by MN Basil are largely conformed to by the Bordurian population that Bordurian law exists. In that sense, Ruritanian and Bordurian law constitute different and legally incommensurable systems. Each legal system is cognitively unified under the positivist's (pluralistic) concept of law, but there is no legal connection between the Ruritanian and Bordurian legal systems, because each fundamentally depends on a different metanorm.

B. Positivism and the Law of Jurisdiction
Let us now assume that Rex announces that Bordurians in Borduria ought to drive on the left. We already know that MN Rex should be assigned to the Ruritanian community. And given that the norm announced by Rex satisfies the criteria in MN Rex , it is valid Ruritanian law. Thus, the Bordurian subjects of the norm apparently have a legal duty to conform to it.
In general, positivists have not spent time considering cases in which the normsubjects of a community's metanorm are unconnected to the community. They assume that the norm-subjects are the community's members. 81 But there appears to be nothing in positivism that excludes norm-subjects who are completely unrelated to the relevant community. 80 I leave it open the type of acceptance by norm-appliers required. I also ignore the other social facts that are required to assign a legal system to a community under Hart's and Shapiro's theory of law, which I do not think make a difference to my argument. HART, supra note 69, at 116-117; SHAPIRO, supra note 76, at 170-173, 211-224. 81 SHAPIRO, supra note 76, at 167.

Jurisdictional Scope in Metanorms
Indeed, besides norm-subjects, there are two other areas of jurisdictional scope in metanorms, and in those areas as well positivists tend to assume a limitation to the community to which the metanorm is assigned. A metanorm speaks of certain norm-appliers as having an obligation to use norms that satisfy criteria that refer to the decisions of certain norm-identifiers. Positivists generally assume that the norm-appliers and norm-identifiers are members of the community to which the metanorm is assigned. 82 Because of these assumed limitations, the metanorms that are assigned to various communities do not conflict. No norm-applier has obligations under multiple assigned metanorms. No norm-identifier has lawmaking power under multiple assigned metanorms. And no one is a norm-subject of multiple assigned metanorms. But there appears to be nothing in the positivist theory of law that requires these limitations. The fact that positivists assume these limitations on jurisdictional scope should make us suspicious that they are importing the law of jurisdiction into their concept of law-thereby treating monistic legal facts as if they were conceptual facts concerning a pluralistic concept of law.
Although our primary worry is an expansive understanding of norm-subjects, it is helpful to briefly consider cases in which the limitations on the other two areas of jurisdictional scope in metanorms are abandoned. I have described MN Rex as obligating all norm-appliers (including those in other states, such as Borduria) to use the norms Rex announces. Once again, it would appear that if this metanorm is accepted by Ruritanian norm-appliers (even if it wasn't accepted by all the norm-appliers in the world), it would be assigned to the Ruritanian community, with the result that all norm-appliers in the world have legal obligations to use Rex's announcements in their law practices. My guess is that would be something that most people would also deny. They would say that Bordurian law-appliers have no legal duty to use Rex's announcements, even though Ruritanian law practices take them to have such a duty.
Relaxing the assumed limitations on norm-subjects and norm-appliers is aggressive-Ruritanian law starts telling non-Ruritanian subjects and officials what to do. Conversely, relaxing the limitations concerning norm-identifiers is yielding. The Ruritanian metanorm allows non-Ruritanians to tell Ruritanian subjects and norm-appliers what to do. An example would be a metanorm assigned to the Ruritanian community that obligates all norm-appliers to use Basil's announcements for events occurring in Borduria.
Indeed, one could imagine a MN Jurisdiction , assigned to the Ruritanian community, that distributes legal obligations to the norm-appliers and norm-subjects of all possible communities and assigns lawmaking power to the norm-identifiers of all possible communities in a way that perfectly replicates the law of jurisdiction. In that sense, the law of jurisdiction would be the law of Ruritania, but for the contingent reason that MN Jurisdiction was accepted by Ruritanian norm-appliers and norms identified by MN Jurisdiction were generally conformed to by Ruritanians. 82 Id. at 165-170. But in our hypothetical, the Ruritanian metanorm is not the law of jurisdiction. It instead gives Rex lawmaking power over Bordurian subjects in Borduria. The positivist's problem is that it would appear to follow from this metanorm and Rex's announcement that the Bordurians have a legal obligation to drive on the left in Borduria, whereas under the commitment they have no such obligation at all.
There are two strategies the positivist might take to accommodate the commitment. The first is to show how the commitment is about the law as the positivist understands it-it ultimately depends solely on some community's contingent law practices. The second strategy is to claim that the commitment concerns the content of the concept of law. Although I think each strategy fails, I will not try to settle the matter here. My goal will solely be to highlight some challenges each strategy faces.

The Law Strategy
To be about the law, the commitment must make use of a metanorm assigned to some community. But in the hypothetical, there is no such community available. Both the Ruritanian and the Bordurian metanorms violate the law of jurisdictioneach community's officials wrongly take its lawmaker to have power over everyone in the world. Nor can one rely on international law practices, for in the hypothetical no such practices exist.
The positivist might argue that the commitment is made not in the light of the law practices of a community in the scenario, but in the light of the law practices of a current community with which the speaker is connected. But it is hard to see how such an approach can explain the commitment, which takes the law of jurisdiction to be a limitation on lawmaking power whatever the law practices of any community (even one's own community) might say about the matter. Assume that the commitment is in the light of the law practices of the current international community, with which the speaker is connected. Under these practices, if Ruritania and Borduria existed, Rex's announcement would not be taken to have legal effect on the Bordurians. But, being contingent, these practices might have been different. They might have taken Rex's announcement to have a legal effect. And, according to the commitment, if they did, they would be wrong-and one would be wrong, as a participant in such practices, if one agreed.
Of course, if Borduria and Ruritania were members of this international community, it would follow that Rex's announcement, made in the context of these hypothetical international law practices that took Rex's announcement to have legal effect on the Bordurians, would indeed have legal effect. But the reason would not be because the law of jurisdiction depends on international law practices, but because the preexisting law of jurisdiction was satisfied by those practices: Bordurians would have a sufficient connection to the international community for events within its law practices to have a legal effect on them.
This can be seen by removing all connections between the Bordurians and the international community. Assume that Borduria has remained isolated from international law practices. It neither participates nor acquiesces in them. Again, Rex, in the context of these international law practices, announces that Bordurians should drive on the left in Borduria. Here I think that people's reaction would be the same as an attempt to ground Rex's power over the Bordurians in Ruritanian law practices. It does not matter that the law practices of the international community take Rex's announcement to have legal effect on the Bordurians. Because the Bordurians have insufficient connection to the international community, Rex's announcement can have no legal effect on them.
But the fact that a positivist is inclined to say that the commitment is made not from the perspective of a community in the scenario, but from the perspective of a current community with which the speaker is connected, suggests that she is making an anti-realist claim that the law of jurisdiction depends on the law practices of one's own community or, more likely, is the expression of one's acceptance of those practices. If the positivist is arguing that this anti-realism can accommodate our firstorder judgment, expressed in the commitment, that Rex's announcement has no legal effect on the Bordurians, full stop (that is, that the legal effect should not be thought to depend on any contingent fact, such as a community's law practices or one's own views), I would question whether the proffered position is positivist at all. It sounds like a form of first-order legal monism (whether of the moral impact or Kelsenian variety 83 ) combined with metanormative anti-realism.
Positivism is not such a metanormative position, but is an attempt to capture appropriate first-order legal judgments. According to the positivist, people speaking about the law should refer to multiple independent and legally incommensurable systems, with the bridge principles for each system depending solely on the relevant community's law practices. So understood, a community's law practices cannot be wrong about their legal effects (except to the extent, discussed below, that the concept of law is violated). But, according to the commitment, a community's law practices can be wrong about their legal effects. I question, therefore, whether a positivist can account for the commitment using the law strategy.

The Concept-of-law Strategy
The alternative, and to my mind more promising, approach for the positivist is to look to the content of the concept of law. The idea is that it is conceptually impossible for Rex's announcement to generate legal duties on the Bordurians, despite the fact that Ruritanian law practices take it to do so. This jurisdictional limitation is a concept-of-law fact, not a legal fact.
But what in the positivist's concept of law could lead to the conclusion that Rex's announcement cannot have a legal effect on Bordurians in Borduria? It cannot be the first requirement, that a metanorm is accepted by a community's norm-appliers. That is the requirement that expands jurisdictional scope, for a community's norm-appliers might accept a metanorm that gives Rex power over everyone in the world.
It must be the second requirement. As we have described it, a metanorm is assigned to a community if the norms identified by the metanorm are largely conformed to by the community's members. Positivists could argue that this formulation was based on their erroneous assumption that the norm-subjects of a candidate metanorm for a community are always the members of that community. Once that assumption is relaxed, the requirements for assigning a metanorm to a community 83 Indeed, Kelsen combined first-order legal monism with just such legal anti-realism. Green, supra note 4, at 291-297.
should arguably be as follows: a metanorm is assigned to a community if it is accepted by the community's norm-appliers and the norm-subjects of the metanorm largely conform to it. To the extent that a metanorm is accepted by a community's norm-appliers, but certain norm-subjects do not generally conform, the part of the accepted metanorm that gives the law-identifiers power with respect to those normsubjects is not part of the metanorm that is relevant in identifying the law of the community. Let us call this the subject-efficacy approach. It would follow from this approach that the part of MN Rex that allows Rex to regulate non-Ruritanians abroad would not be relevant in identifying the law generated by Ruritanian law practices, because non-Ruritanians abroad do not generally conform to Rex's announcements.
Although, again, I cannot explore this matter fully here, I think the subject-efficacy approach fails, for two reasons. The first is that it cannot account for people's intuitions concerning the legal effects of a community's law practices. The second is that, even if the approach could account for these intuitions, it cannot have its source in the concept of law as the positivist understands it.
Let's start with the first argument. It's true that the subject-efficacy approach can explain why Rex's announcement has no legal effect on Bordurians in Borduria, since they do not generally follow Rex's announcements. But can it explain why Rex's announcement would have a legal effect on all Ruritanians in Ruritania? Imagine rough neighborhoods within the capital city of Ruritania where Rex's announcements are systematically ignored. I'm confident that people would say that the Ruritanians in the neighborhood still have legal obligations under Ruritanian law. Why is subject-efficacy ignored here? The reason cannot be that Ruritanian norm-appliers take Ruritania's borders to be relevant in determining the scope of Ruritanian legal obligations, for if that were enough, the fact that they take Bordurians in Borduria to have such legal obligations would also be enough.
Or consider Ruritanians abroad. Because Ruritania lacks enforcement jurisdiction (as well as any practical ability to engage in enforcement), it is likely that Ruritanians will not systematically conform to Ruritanian law when it conflicts with the law of the place where they are located. And yet people who hold a modern theory of concurrent legislative jurisdiction would still say that Rex's announcements can create legal obligations on these Ruritanians. The subject-efficacy approach cannot explain such obligations. In short, the subject-efficacy approach is simply too crude to capture the circumstances where people would say that Ruritanian law practices can and cannot generate legal effects.
But let us assume that the subject-efficacy approach, or some other approach that claims to be derived from the content of the positivist's concept of law, can model people's intuitions about the ability of Ruritanian law practices to generate legal effects in cases with connections to multiple communities. There is a more serious problem. The decisions about whether legal effects exist in such cases are simply not the sort of thing that can be determined by the content of the concept of law as the positivist understands it.
The positivist's concept of law is pluralistic. A legal system that is assigned to a community is legally incommensurable with a legal system assigned to another community. The legal obligations of these various communities cannot legally conflict, although from a nonlegal perspective they might conflict practically. But as one starts answering detailed jurisdictional questions-such as why Rex's announcements cannot generate legal obligations on Bordurians in Borduria, but they can on Ruritanians in Borduria and Bordurians in Ruritania-it becomes clear that one is coordinating legal obligations created by Rex's and Basil's announcements in a manner that assumes that they can legally conflict.
Consider the following analogy. There is a particularly implausible form of moral communitarianism that is much beloved of college sophomores. Rather than being a form of moral anti-realism-which, as we have seen, is compatible with first-order moral judgments that assert morality's independence from contingent facts, including facts about a community's moral views-this is an alternative view about appropriate first-order moral judgments, according to which morality is pluralistic: it ultimately depends on a community's moral practices, such that each community has its own moral system, which cannot morally conflict with the moral system of another community.
The classic argument against this sophomoric theory is to demand an answer to jurisdictional problems. What happens when someone from community A is visiting community B? The sophomore's answer might be that she should conform to community B's moral system. What happens when community A's moral system tells people in community B what to do, even though what the Bs are doing has no effect on the As at all? The sophomore's answer might be that community A's moral system does not extend to the Bs in such a case.
By providing these answers, the sophomore has made it clear that he is a firstorder moral monist, not a pluralist. If he were a pluralist, his answers would be very different. When asked about the moral consequences of someone from community A visiting community B, he would observe that the question is ill-formed and that one must first specify whether one is answering it from the A or the B moral system. And when asked about the moral consequences of the A moral system telling Bs what to do, he would observe that the question provides its own answer: under the A moral system, Bs must do it, although under the B moral system they may not have to.
But the sophomore instead tried to coordinate the moral obligations of the two communities. And that means that he has a first-order monistic view of morality, according to which the existence of a community with moral practices is morally relevant in the light of preexisting moral principles (say, principles of tolerance) that do not depend on any community's moral practices.
The anti-sophomoric argument applies to the positivist who, in answering jurisdictional questions, seeks to coordinate legal obligations. In so answering, she reveals herself to be a first-order legal monist. I think that the positivist could credibly claim to be a legal pluralist only if, when asked about the legal consequences of Rex's announcement on Bordurians, she refused to engage in any legal coordination and insisted that the question was ill-formed. She would provide answers only from a legal system, whether the Bordurian, the Ruritanian, or some other. Without any need to coordinate, she would conclude that the Bordurians have a legal obligation to drive on the left. But, in so doing, she would fail to account for the commitment.
But the above argument depends on a highly restricted account of the basics of legal positivism. It might be that a more elaborate positivist theory could explain why the relevant jurisdictional limits follow as a conceptual matter. Consider, for example, Raz's view that law necessarily claims authority (even if it may not actually have it). 84 Raz's conception of actual authority is the normal justification thesis, under which authority usually exists because the subject is more likely to conform to the balance of reasons for action by obeying the authority's directive than by acting on his own determination of what the balance requires. 85 The law can have actual authority by aiding subjects in overcoming coordination problems, prisoner's dilemmas, and the like.
According to Raz, the demand that the law necessarily claims authority can put meaningful limitations on law's content. Even though a directive does not have to have actual authority to be law, it cannot be law if it cannot plausibly claim authority. And that means certain threshold requirements for authority must be met. An example is that the subjects of a directive that purports to be authoritative can determine its content without taking into account the underlying reasons for action that the directive purports to settle. 86 One might argue that similar threshold requirements for authority are violated by Rex's directive. But I think a Razian attempt to accommodate the commitment would face problems similar to those we have already encountered. Why can't Rex plausibly claim authority over Bordurians in Borduria on the ground that they would better abide by their reasons for action by listening to him rather than Basil? Plausible claims of authority can easily overlap in a manner that is incompatible with the commitment. To be sure, if legal authority required actual authority the commitment might be explained. But with the requirement of actual authority, we have left positivism aside.
I do not purport to offer the final word on the concept-of-law strategy here, however. Perhaps the positivist can add conceptual requirements for a law (such as Fullerian requirements of publicity and noncontradiction 87 ) that would explain the relevant jurisdictional limits, although I suspect that the addition of these requirements amounts to a retreat from positivism.

VI. Interpretivism
For the past fifty years or so, the primary competitor to positivism has been interpretivism. For the interpretivist, the bridge principles that take one from a community's law practices to their legal effects are based in the interpretive confluence of facts about that community's law practices and moral facts.
But one should not read interpretivism as a monistic theory of law, under which the bridge principles for all communities consist of the set of fundamental moral facts used in interpretation, with communities' law practices simply functioning as facts that are made morally relevant by those fundamental moral facts. The bridge principles for Ruritanian law practices are not moral facts, but rather moral facts as constrained by Ruritanian law practices, 88 with the result that Ruritanian legal obligations are normatively incommensurable with moral obligations. Morality and Ruritanian law are separate normative orders.
Furthermore, the legal obligations of different communities are legally incommensurable. Because the bridge principles that take one from Ruritanian law practices to their legal effects fundamentally depend on Ruritanian law practices (as morally interpreted), they are different from the bridge principles that take one from Bordurian law practices to their legal effects, which depend on Bordurian law practices (as morally interpreted). Ruritanian and Bordurian law are separate legal orders.
Like positivists, interpretivists have generally assumed that the subjects of a community's law are limited to the community's members. So what might interpretivists say of the commitment that Rex's announcement cannot generate a legal effect on Bordurians in Borduria?
Like the positivist, interpretivists can use the law strategy or the concept-of-law strategy. And one might think that the interpretivist is better positioned to explain the commitment using the law strategy than the positivist. The essence of the commitment is that a community's law practices can be wrong about their legal effects. And that is something that the interpretivist also accepts, for a moral interpretation of a community's law practices can yield bridge principles that are different from what a community's law practices take them to be. After the moral interpretation of Ruritanian law practices, we might conclude that Rex's announcement can have no legal effect on the Bordurians, even though the practices say they can.
But what if it is the character of Ruritanian law practices that, when morally interpreted, Rex's announcement does have such legal effect? To repeat, moral interpretation must fit the practices interpreted. Ruritanian practices might constrain moral justification in a way that interpretation would lead one to conclude that Rex has lawmaking power over the Bordurians. And that is incompatible with the commitment.
To the extent that the interpretivist insisted that moral interpretation of Ruritanian law practices, no matter what their content, must lead one to conclude that Rex has no power over Bordurians in Borduria, I think their position would start looking like the moral impact theory. All communities would have the same bridge principles, consisting of fundamental moral facts, and the law would simply be the moral effect of a community's law practices in those facts' light.
The alternative is to say that the commitment concerns concept-of-law facts. We don't have to engage in moral interpretation of Ruritania's contingent law practices to discover that Rex's announcement cannot generate legal effect on the Bordurians. An inquiry into the content of the (interpretive) concept of law tells us that it can't.
Here I think interpretivists will run into similar problems as positivists. First, interpretivism cannot explain the nuanced judgments people make about legal effect in interjurisdictional settings. According to interpretivists, legal obligations are associative-they obtain between the members of the relevant community. 89 Granted, that means that they can explain why Rex's announcements do not have a legal effect on 88 DWORKIN, supra note 2, at 66. 89 Id. at 195-202. Bordurians in Borduria, since these Bordurians aren't Ruritanians. But they are at a loss to explain why they have an effect on Bordurians visiting Ruritania.
And even if interpretivists could make sense of these nuanced judgments, they would be attempting to coordinate Ruritanian and Bordurian legal obligations, on the assumption that they can legally conflict. The anti-sophomoric argument would then kick in. The interpretivist would reveal himself as a monist malgré lui. Indeed, by speaking of legal obligations as associative, interpretivists have arguably already situated all communities' legal obligations within an overarching normative framework in a manner that is in tension with their avowed legal pluralism.
Toward the very end of his life, Dworkin himself abandoned interpretivism and started speaking of law monistically. Under this new theory of law, there is a "unity of value" that includes law. Law is "a branch, or subdivision, of political morality." 90 Thus, the later Dworkin would make the same criticism of his interpretive theory of law that I have suggested above. By describing how a community's law practices can generate associative (and so legal) obligations for the community's members, the early Dworkin thought he was articulating conceptual facts about the content of a pluralistic concept of law. What he was actually doing was articulating legal facts (which are also moral facts) within a monistic framework.

Conclusion
It has been my goal to use the law of jurisdiction to provide support for the moral impact theory. My argument supplements, but is different from, Greenberg's. The foundation of Greenberg's argument is that an adequate theory of law must account for what he calls the rational-relation requirement. The idea, roughly, is that it must be intelligible why law practices in a community generate legal facts. Greenberg claims that positivism-for which the bridge principles that take one from a community's law practices to their legal effects depend solely on that very community's law practices-fails to satisfy the requirement. The best way of making the relationship between law practices and legal facts intelligible, Greenberg argues, is by reference to moral facts.
The law of jurisdiction supports Greenberg's argument, because it speaks of how those social facts about a community's law practices, on which the positivist thinks legal facts ultimately solely depend, can themselves generate or fail to generate legal facts. The law of jurisdiction stands above those communities' law practices and speaks of their legal effects, thereby highlighting the demand that the relationship between a community's law practices and legal facts be intelligible.
That is not to say that a noncommunitarian law of jurisdiction proves that the moral impact theory is correct, for the content of this law might not be explicable in terms of moral facts. Traditionalists' theories of exclusive jurisdiction, for example, are hard to explain in moral terms. To the extent that a noncommunitarian law of jurisdiction cannot be explained morally, it would support a form of Kelsenianism.
In closure, let us now briefly turn to the unlimited theories of legislative and personal jurisdiction. What if they are right? What effect would that have on my argument? The answer is none. Assume that there are no limits on a state's personal or legislative jurisdiction beyond those self-imposed by the state's law practices. Even if that is the case, these legal facts about unlimited jurisdiction are independent of the law practices of any particular state. It is a legal fact that obtains for all states that Rex's announcement concerning Bordurians creates legal obligations on them. If, according to Bordurian law practices, Rex's judgment has no legal effect on the Bordurians, it nevertheless has a legal effect. There is a universe of difference between the communitarian's view that there is no noncommunitarian law of jurisdictionthat the very idea of such law is incoherent-and the view that there is a noncommunitarian law of jurisdiction that gives to communities' law practices an unlimited ability to generate legal effects. The unlimited theory of jurisdiction, although clearly wrong to my mind, is not a problem for my argument.