The ALI Principles on Transnational Intellectual Property Disputes: Why Invite Conflicts?

s those members of the innovation community who focus on procedural law know, the American Law Institute (ALI) 1 is engaged in a project to facilitate litigation of intellectual property disputes that cross national borders. 2 The enterprise owes its origins to the 1999 Draft of the Convention on Jurisdiction in Civil and Commercial Matters, negotiated at the Hague Conference on Private International Law. 3 By now, it has undergone several iterations. Columbia University Professor Jane Ginsburg and I used the Hague material as the starting point for proposing a stand-alone convention dealing not only with the general problems of international litigation, but also with issues that uniquely arise when intangible rights are at stake.

Our work, which was first presented at a Chicago-Kent College of Law symposium in October 2001, 4 attracted the attention of the ALI. After arranging a further presentation in April 2002, the Institute formally adopted the project as its own. It Revised drafts, now cast in the form of principles that courts may follow, rather than as a convention that nations must join, were presented to the Advisers in February of 2003 and 2004 and to a special session of the ALI membership in May 2004. A new draft will be presented to the Advisers in April 2005. It will be at least a year, probably longer, before the ALI will formally consider approving the final product. In the meantime, the Reporters are interested in broad input. Accordingly, we are grateful to Professor Sam Murumba for the valuable opportunity to discuss our work at Brooklyn Law School.

The current draft differs in many ways from the one initially unveiled in Chicago. Of particular importance, it goes beyond the issues of personal jurisdiction and enforcement of judgments-the issues that were at the heart of the Hague Convention-to cover choice of law. After explaining why a project tailored to intellectual property litigation is desirable and describing its key features, this paper discusses the decision to add principles on applicable law and the factors that were considered in making specific choices.

A convention on enforcement of foreign judgments has been in gestation at the Hague Conference for over a decade. Begun in 1992, the goal was to create an international analogue to the U.S. system of according full faith and credit to sister state judgments, and to the EU's Brussels Regulation, which establishes a regime for recognizing judgments within the European Union. 6 That is, member states were to agree to recognize and enforce any judgment rendered by another member state, so long as that judgment was predicated on a basis of personal jurisdiction approved under the convention. Conversely, members were to agree to refuse to recognize or enforce judgments predicated on jurisdictional bases prohibited by the convention. 7 Thus, the convention was to be comprised of essentially two parts: one would list the bases of jurisdiction that were approved or prohibited, the other would set out conditions of enforcement. 8 By 1999, a draft convention was promulgated, however it quickly ran into significant opposition. To some extent, the problems were substantive: states began with very different approaches to adjudicatory authority and that made it difficult to agree on specific jurisdictional provisions. 9 Other issues were technological: much of the drafting of the convention was completed before the advent of e-commerce. As a result, the negotiators barely considered a key source of future international disputes. Mainly, however, the issue was lack of enthusiasm. When the convention was first proposed, there was a strong perception that judgments rendered in the United States were difficult to enforce abroad, and that as a result, successful U.S. litigants were shortchanged and U.S. law was arguably under- 7. Unlike the Brussels Regulation, which has only white (approved) bases of jurisdiction and black (prohibited) bases, the Hague plan was to leave a grey area where members could decide for themselves whether to recognize or reject a judgment.

8. Recognition and enforcement are not the same thing. For example, a judgment can be recognized for purposes of determining whether another action can be filed without being at a point where it can be enforced for, say, money damages. However, for purposes of convenience, the terms "enforced" and "enforceable" will be used to cover both concepts unless specifically noted otherwise. 9. For example, EU legislation tends to emphasize institutional considerations such as predictability and consumer protection, while U.S. case law, which looks to the due process interests of each and every litigant, yields results that are more difficult to anticipate. See Rochelle C. Dreyfuss enforced. The U.S. bar was therefore willing to engage in a trade-off. It would sacrifice certain bases of jurisdiction that were opposed abroad (such as general doing business jurisdiction and tag jurisdiction, which are viewed elsewhere as exorbitant) and in exchange, enforcement of other U.S. judgments would become a near-certainty. 10 However, by the time the draft convention was promulgated, that trade-off was no longer considered desirable. Because the U.S. market had become extremely attractive to foreign capital investment, in most significant cases, foreign litigants had sufficient assets within the United States to satisfy judgments rendered against them. With less need to find internationally accepted predicates for jurisdiction, the enthusiasm within the United States for sacrificing familiar bases dissipated. And since the United States tends to enforce foreign judgments, and within the EU, the Brussels Regulation works well, there was a similar want of interest abroad.

As of this writing, the Hague Conference has suspended work on a general convention. Instead, it is taking a "bottom up" approach, which contemplates that agreements dealing with specific problems of international concern will be developed and that as experience with these is garnered, it will become easier to draft a general convention governing all private law disputes. Thus, there is now a proposal pending at the Hague for an instrument on exclusive choice of court agreements. This convention would make enforceable judgments rendered by a court chosen by business parties in a written contract. 11 In addition, the Hague Conference has urged practitioners and other interested parties to consider the wisdom and contours of agreements covering specific legal fields where international enforcement issues are particularly problematic. 10. See Draft Hague Convention, supra note 3, art. 18 (d), (f), (i). This would have been more of a sacrifice than might first meet the eye as even U.S. judgments enforced in the United States would be subject to the prohibition on tag and doing business jurisdiction if they involved litigants that were citizens of other convention states.

11. It can certainly be argued that intellectual property is one such field. On the copyright side, markets are now global. U.S. movies, television programs, and music have long enjoyed broad audiences abroad, and in recent years, foreign works have begun to appeal to Americans. Population shifts have produced diasporas-significant communities that consume works in one language while living in locations where another is spoken. At the same time, the Internet and other forms of digital transmissions have reduced the cost of reaching international markets and decentralized the mechanisms of distributing intellectual products. Many of these factors-along with the rise in international travel-have created a similarly global market for, and knowledge of, trademarks. 12 As to patents, the Internet is increasingly used to distribute patented software and to make offers to sell tangible embodiments of patented inventions to remote locations. 13 There are also patents that are explicitly drawn to the online environment. Some patents include socalled "divided" claims, which contemplate activity in more than one jurisdiction. 14 It is also becoming increasingly common for those investing in innovation to rely on fairly global exploitation, especially now that there are international instruments that make the acquisition of world-wide protection easier. These changes put significant pressure on both the enforcement of intellectual property rights and on effective defenses against infringement claims. A single Internet transmission can simultaneously produce copyright and trademark infringements on a world-wide basis. Similarly, a business method can be practiced on a server situated in one country, at the instigation of users located in a multiplicity of other states. As a result, rights holders now find that to fully protect their interests, they must sue in more than one jurisdiction, and shoulder the attendant risks of infringements occurring in places where the defendant lacks assets, where there are insufficient contacts to support adjudicatory authority, or where the law or the pace of legal proceedings is especially disadvantageous. Indeed, the ubiquity of the Internet's infrastructure gives the users of intellectual property unprecedented ability to choose to operate from precisely such locations. 16 And in some instances, servers can be artfully placed so that there is no one country where all the steps of a patent are practiced-and therefore, arguably, no infringement anywhere. 17 The globalization of intellectual property activities and communications also poses problems to potential defendants. Lacking the ability to predict when they will be subject to adjudicatory authority, consumers of intellectual property act at their peril when they utilize material protected anywhere. Further-

WHY INVITE CONFLICTS? 825 more, rights holders can use the necessity of successive actions to their advantage, to wear users down by bringing actions seriatim, hoping eventually to win in a big enough market to make the competitor's continuing activity everywhere unprofitable. This is a particular problem for small businesses that lack the legal and technical sophistication necessary to avoid becoming amenable to suit in foreign fora and the resources to fight multiple suits. To the extent that start-ups are especially responsible for innovation, these problems may significantly affect the public interest by chilling creativity and technological progress. 18 A convention of the type originally contemplated at the Hague would solve many of these problems. The Internet has created difficult personal jurisdiction cases for all courts, and these are further complicated by the intangible nature of the rights at issue. 19 Thus, it would be quite helpful to have an international agreement on which activities support the assertion of adjudicatory authority. Further, courts have taken different positions on their power to rectify and to stop (on both a permanent and temporary basis) injuries that occur outside their territories. 20 Clarity on that issue would thus also be welcome. 20. For example, in the United States, the single-publication rule, which requires the assertion of all claims for libel in a single action, is said to protect the interests of the media, see Keeton v. Hustler Magazine, Inc., 465 U.S. 770, 777 (1984); arguably a similar rule should protect media defendants in intellectual property actions. By contrast, however, the EU bars the assertion of extraterritorial libel claims in courts that are not situated at the residence of the defendant. See Shevill v. Presse Alliance S.A., [1995] 2 A.C. 18 (H.L. 1995) (Eng.). There is similar controversy over cross-border relief issues. See, e.g., Turner v. Grovit, 1 All. E.R. 960 (H.L. 2002) (Eng.); Grupo Mexicano de Desarrollo, S.A. v. Alliance Bond Fund, Inc., 527 U.S. 308 (1999); Symposium, But even more is possible. A convention tailored to the needs of the intellectual property community could also make the adjudication of international infringements efficient. It could facilitate cooperation among courts entertaining parts of the same series of transactions or create avenues for consolidating cases, thereby saving judicial resources on a world-wide basis. It would also reduce the private cost of enforcing intellectual property rights and improve the deterrent effect of the law. At the same time, an integrated system would prevent plaintiffs from harassing lawful users, safeguard free speech interests, and assure that materials in the public domain are genuinely available for use.

Admittedly, the ALI cannot fulfill all of these goals. Because the drafters do not represent states, its provisions will not be enacted directly into law. However, as a set of principles, the project can demonstrate how national courts could be used to create an efficient method for adjudicating international disputes. Thus, the hope is that states will be inspired to return to the bargaining table, where they can use the work as a template for action. The Principles can also have an impact as "soft law." In some cases, they could be followed by courts unilaterally or adopted through the consent of the parties-in their contract or at the time of litigation. A set of principles also creates a focus for future discussion by the intellectual property community. Indeed, it is heartening to see that similar projects are proceeding in other arenas. 21 Copyright's Long Arm: Enforcing U.S. Copyrights Abroad, 24 LOY. L.A. ENT. L. REV. 45 (2004); John Gladstone Mills III, A Transnational Patent Convention for the Acquisition and Enforcement of International Rights, 84 J. PAT. & TRADEMARK OFF. SOC 'Y 83, 85 (2002).

21. The International Association for the Protection of Intellectual Property [hereinafter AIPPI] has polled its membership on these issues and has adopted a Resolution proposing approaches to jurisdiction, choice of law, and enforcement of judgments that are generally consistent with the ALI Principles. See AIPPI, Report Q174: Jurisdiction and applicable law in the case of cross-border infringement (infringing acts) of intellectual property rights, AIPPI 2003 Y.B., at 827, available at http://www.aippi.org/reports/resol utions/Q174_E.pdf. There is also a group of intellectual property lawyers in Europe, led by the Max Planck Institute, working on an International Convention on Jurisdiction and Enforcement of Judgments, which deals with many of the same issues. Its principal author, Annette Kur, is an advisor on the ALI project.

Given this background, the contours of the ALI project are largely predictable. After defining the areas of intellectual property law to which the Principles will apply, the issues dealt with at the Hague are addressed. First, the jurisdictional provisions specify the sort of contacts a defendant must have with a state to become amenable to suit in its courts. These provisions also delimit the scope of the court's authority to hear and act upon claims arising within and without its territory. Second, the Principles set out rules on when, and on what terms, judgments based on these bases of jurisdiction should be enforced. However, the project also departs from the Hague approach in significant ways. It includes procedures for simplifying the adjudication of world-wide disputes, providing two methods, cooperation and consolidation. Further, as described more fully in the next section, it lays out criteria for determining the law to be applied in international contexts.

The Principles apply to all intellectual property rights, including not only copyright, patents, and trademarks, but also neighboring rights, trade secrets, domain names and rights stemming from concepts of unfair competition. The notion is to cast a broad net, so that courts enjoy maximum flexibility to structure litigation in ways that encourage efficiency. However, it is recognized that drawing lines can be difficult. Accordingly, it is contemplated that early in the litigation process, the court, helped by the parties, will determine whether, and to what extent, the Principles will apply.

As is standard in American jurisprudence, the Principles distinguish between bases of general and specific jurisdiction. A third type of jurisdiction, designed to improve efficiency, has also been added. Because the ALI project sets out multiple bases of adjudicatory authority but does not establish a preference among them, it differs sharply from the 1999 Draft Hague Convention and also from European practice. However, this approach was considered necessary to achieve the project's overall goal: a multiplicity of jurisdictional predicates creates the flexibility needed to situate each case in the court best able to provide complete justice to all the litigants. Nonetheless, it is also understood that one of the attractions of a set of Principles is that it can provide some assurance that jurisdiction will not be asserted inappropriately. Thus, the project does not opt for complete flexibility. Instead, it follows the Hague approach by defining certain bases of jurisdiction as prohibited. 22

The two provisions on general jurisdiction create authority to hear all claims against a defendant no matter where they arise. Both should be familiar to American lawyers: the defendant is subject to general jurisdiction at its habitual residence (domicile) and in any court where a general appearance is made. Thus, a defendant whose habitual residence is Germany, who engages in activity in France and Germany that leads to harm in France, Germany, the United States, and Japan, can be sued in Germany for the harm claimed in all four states. A general appearance in a court of any of the other countries will similarly create jurisdiction over claims to harm everywhere.

Unlike the general jurisdiction provisions, which are based entirely on general law, the three principles that deal with specific jurisdiction take into account the special needs of the intellectual property community.

The first provision (which in a sense lies midway between general and specific jurisdiction), expresses a position in favor of party autonomy. It makes defendants amenable to suit in any place agreed to in a choice of forum clause. 23 Some of the language in the current draft is essentially a placeholder for 22. As with the Draft Hague Convention, there is a residual grey area where every state has authority to make its own decisions on enforcement.

23. This has the flavor of general jurisdiction because the claims need not arise from the contract directly. See, e.g., Carnival Cruise Lines, Inc. v. Shute, 499 U.S. 585 (1991) (recognizing the enforceability of a forum selection clause in a consumer contract in a torts case for negligent operation of a vessel). However, unless the contract was badly drafted, the amenability to suit would be restricted to claims arising from the relationship created by the contract containing the forum selection clause.

WHY INVITE CONFLICTS? 829 changes that will likely be made to conform the Principles to the choice of forum convention currently under negotiation at the Hague. However, there is also language that will not likely be conformed to the new Hague instrument because it is intended to tailor the law to intellectual property transactions. In these transactions, there is a clear need for agreements that reduce the jurisdictional exposure of distributors of digitized information, but also a concern that nonnegotiated licenses (clickwraps and shrinkwraps) will overreach and require adjudication in a forum which is burdensome to the other side (and which applies law recognizing the enforceability of choice of forum agreements 24 ).

To deal with this problem, the Principles reject the Hague's distinction between consumer and business transactions, and instead differentiate between negotiated and nonnegotiated licenses. Judgments based on forum selection clauses in all negotiated agreements are enforceable, even when individuals are involved. However, for nonnegotiated agreements, forum selection clauses will be effective only if the forum chosen is reasonable under rules set out by the Principles themselves (as distinguished from the place where enforcement of the agreement is sought). If the agreement is reasonable when judged in light of the expectations, location, sophistication, and resources of the parties, in particular the weaker party, the interests of the relevant states, the availability of online dispute resolution, and the expertise of the court chosen, the forum selection will be honored; otherwise, it will be disregarded. In the end, the hope is that the risk of losing the benefits of forum selection clauses will encourage those who draft nonnegotiated agreements to make fair choices.

A second provision covers contract actions generally (that is, claims arising from contracts that do not contain enforceable forum selection clauses). This provision gives the courts of a state whose rights are in issue adjudicatory authority over the defendant. However, the court's power is limited to local 24. Currently, courts tend to use forum law to decide whether to enforce forum selection clauses, with the result that it is difficult to predict when they will be enforced and to know whether the nondrafting party will be sufficiently protected. See, e.g., Specht v. Netscape Communications Corp., 306 F.3d 17 (2d Cir. 2002); Evolution Online Sys., Inc. v. Koninklijke PTT Nederland N.V., 145 F.3d 505 (2d Cir. 1998).

claims. For example, a dispute over an agreement licensing world-wide trademark rights can be litigated in France, even if the defendant is not a domiciliary of France-but the only claims that can be adjudicated are those that deal with the French marks. As with forum selection clauses, this provision can be used for nonnegotiated agreements only when the contract is reasonable under criteria set out in the Principles.

The third head of specific jurisdiction involves one of the most controversial issues addressed by the Principles: jurisdiction over infringement actions. In a variation on the traditional approach in the United States, this provision creates a sliding scale, with the level of adjudicatory authority dependent on the nature of the forum state's connection with the dispute.

When the defendant has "substantially acted" in the state, the court's jurisdiction over the defendant extends to all claims of harm arising out of the defendant's in-state activity, no matter where the harm is felt. For example, a defendant who is habitually resident in Germany, who operates a server in Angola, where the plaintiff is habitually resident, and who uses that server to distribute infringing content to Portugal, Brazil, and Mozambique will be amenable to suit in Angola for all claims arising out of the activity in Angola, including claims pertaining to harm in Portugal, Brazil, and Mozambique.

When there is less connection to the state, the court's authority is more circumscribed and the ambit of the case is determined by whether the plaintiff is bringing the case in the forum where it is resident. It has, however, proved difficult to draw the line between activity that occurs as a result of the defendant's purposeful availment of the forum's benefits and activity that results from the unilateral actions of others. In the former situation, there is universal sentiment that jurisdiction is justified, whereas in the latter, there is a sense that asserting jurisdiction is inappropriate. In part, the problem is linguistic, for it is hard to describe what the defendant must be doing to be amenable to jurisdiction without involving the court in difficult determinations of intent. Various formulations have been considered, including "directing activity," "targeting the jurisdiction," and "endeavoring to direct."

To a large extent, however, the problem is normative. It revolves around questions of how much responsibility actors should bear to avoid jurisdictions in which they do not wish to be sued and whether it is reasonable to require the same avoid-

ance activities of all intellectual property users, no matter their size, wealth, and degree of technological and legal sophistication. For example, it is fairly clear that a German domiciliary who runs a website in Angola in the Portuguese language, which makes available music that appeals specifically to a Portuguese audience, should be subject to jurisdiction in Brazil for harm occurring there. It is less clear that this defendant should also be subject to jurisdiction in New York where, unknown to the defendant, there is a substantial diaspora of Portuguese speakers who migrated from Madeira and who found the website on its own. In the current draft, a conservative approach is taken. The court where the action is filed must scrutinize the defendant's activities to determine whether it is reasonable to believe it was directing the alleged infringement to the state. Since "directing" is defined as initiating or maintaining "contacts, business, or an audience ... on a regular basis," businesses with sporadic contacts will not be amenable to the court's power. Furthermore, the defendant is given an opportunity to avoid the court's authority by demonstrating that it took steps reasonable under the circumstances to avoid acting in the state. If the defendant is found to have directed infringement towards a state, then a plaintiff who is a resident of that state, may assert claims for all the harm resulting from the defendant's activity, no matter where that harm actually occurred. If the plaintiff is not a state resident, then only local harm may be asserted.

The remaining two bases of jurisdiction are designed to facilitate efficient adjudication. These grounds of jurisdiction may be unfamiliar to Americans, but they are based on the Brussels Regulation and are thus known to Europeans. 25 The first provision would expand the authority of the court where one defendant is habitually resident to include power over other defendants who are enmeshed in the same transactions. It applies only when the other nonresident defendants have some contact with the forum state and separate adjudication of claims against the various defendants would create a risk that the par-25. See Brussels Regulation, supra note 6, art. 6(1)-(2). BROOK. J. INT'L L.

[Vol. 30:3 ties will be subject to inconsistent outcomes.

In such cases, the scope of the lawsuit extends to all harms flowing from the joint activity, no matter where it occurs. The second provision pertains to third-party actions, and allows a local defendant to add parties who are liable to the defendant for all or part of the judgment the defendant suffers. Again, the court's power extends to harm flowing from the alleged activity, no matter where it occurs.

Finally, as with the Draft Hague Convention, the Principles list a series of bases that are considered inappropriate predicates for adjudicatory authority. These include jurisdiction based solely on nationality, temporary residence or presence, or service of process within the territory.

Consistent with U.S. law, the Principles draw a distinction between subject matter and personal jurisdiction. Local law supplies the rules on subject matter authority. However, to achieve efficiency, the Principles suggest that where possible under local law, and consistent with the scope of personal jurisdiction set out in the Principles, courts should extend their subject matter reach to cover all claims and counterclaims arising from the transaction that gave rise to the initial claims. Although there have been suggestions that efficiency should be forced on the parties by making related claims and counterclaims compulsory, the absence of well-developed doctrines of claim and issue preclusion in some parts of the world militate against that approach.

In addition, the Principles contemplate that courts will exercise their authority to hear declaratory judgment actions and to provide provisional relief. In the latter case, the Principles suggest that the court hearing the action exercise its competence to 26. Cf. Expandable Grafts P'ship v. Boston Scientific, B.V., Court of Appeal of the Hague (1999) F.S.R. 352, ¶ 19 (consolidating cases when the defendants are part of the same group of companies). The Principles recognize three types of inconsistency: redundant liability, judgments that undermine one another, and judgments to which the parties cannot simultaneously conform their behavior.

WHY INVITE CONFLICTS? 833 issue any protective order necessary, including those that cross national borders. Other courts are to limit their power to issue preliminary measures to actions that affect only their own territories.

As noted earlier, a key value of this project is its capacity to facilitate resolution of global disputes. The Principles offer two methods for simplifying such disputes, cooperation and consolidation, both of which draw on American and European methods of aggregation. 27 Both are thought to require some degree of supervision; the Principles use the lis pendens doctrine to choose the supervisor. Under this provision (and subject to an exception explained below), initial decisions on simplification are to be made by the court where the first of the related actions is filed (the court with "supervisory authority"). These decisions include whether to simplify, the method of simplification, and in the case of consolidation, the place of simplification. Since these decisions can be opportunities for delay, there are also provisions aimed at minimizing dilatory practices.

Initially, the court must decide whether the world-wide actions are closely enough connected to benefit from coordinated treatment. It is expected that such will be the case whenever two or more lawsuits in different countries arise from connected transactions.

The decision on how to simplify involves a choice between cooperation and consolidation. The Principles set out criteria for making this selection. These include such matters as whether there is a court with sufficient power over all of the litigants and enough authority to award the relief requested to make consolidation an option; whether there is a court with special expertise in the issues in contention; the impact of the decision on the resources of the parties; and the degree of cooperation that can be expected.

The cooperative approach is inspired by recent developments in international bankruptcy litigation, where the parties, with the aid of the courts where bankruptcy petitions are pending, develop a cooperative plan to coordinate the distribution of world-wide assets. 28 Although intellectual property disputes are significantly different from bankruptcy in that they are not zero-sum games, the litigants still have substantial incentives to cooperate. For example, cooperation will likely be appropriate in registered rights cases, particularly patent cases, where the laws are very different, and foreign (and in some cases, domestic) courts lack the capacity to order a patent office to act on a finding of invalidity. In such cases, litigation is best situated in each country in which rights are registered. At the same time, however, substantial benefits could be achieved if, before any trial commences, the parties agree to rely on a single examination of the inventor, choose to focus their disputes on the same embodiments of the accused device, and stipulate to the documents and practices that constitute the prior art. Although courts could still arrive at different decisions on validity or infringement, there is no real inconsistency because the laws applied are different and, in many cases, exploitation in one territory is (at least in theory) unaffected by exploitation elsewhere. 29 28. See, e.g., American Law Institute, TRANSNATIONAL INSOLVENCY 29. In practice this may not be so. Although the problem is not as dramatic as inconsistent judgments about whether a work can be distributed on the Internet, in fact, prohibiting the sale of patented articles in one jurisdiction can affect decisions on exploitation elsewhere because of factors such as economies in the scale of production and the demand for interoperable products.

In contrast, copyright cases involving Internet distribution may be better suited to the other approach, consolidation of all cases arising from a series of transactions in a single forum. In such cases, there are real risks that conflicting judgments will be entered-for instance, that one court will consider a transmission infringing while another court holds the same transmission noninfringing; or that more than one court will levy royalties on the identical communication. 30 Resources can be saved and inconsistency avoided if all cases are transferred to a single court, which can then determine how best to deal with the interests of the states involved. 31

If the court decides on cooperation, then it will develop a plan for adjudicating the world-wide dispute with input from the parties and the other courts involved. On the other hand, when a court decides to consolidate, then it must next select the place where the action will be heard. If the parties' contracts selected a unique forum, it will likely be chosen (subject to the usual caveat on nonnegotiated agreements). However, the goal is to situate the case in the court most closely connected to the parties and dispute, and most convenient to the witnesses. Preference is also given to a tribunal specialized in the field at issue (for example, a specialized patent court for a case involving only patent issues) and to a court in a state that belongs to the WTO and is therefore internationally accountable for its actions. 30. Two situations raising the problem of inconsistency are presented in the Grokster and iCraveTV litigation, where the defendants could easily have been exonerated in one country (for example, the Netherlands or Canada), while found liable in another (such as the United States). See Metro-Goldwyn-Mayer Studios, Inc. v. Grokster, Ltd., 380 F.3d 1154 (9th Cir.), cert. granted, 125 S. Ct

It has been suggested that the coordination approach, while valuable in theory, is vulnerable in practice because it provides infringers with multiple opportunities to engage in sharp practice and to delay adjudication. The Principles deal with these concerns in several ways. The decision to coordinate adjudication must be made early in the proceedings; if a decision is made not to coordinate or no decision at all is made, then each action can proceed where initially filed. Furthermore, no court is stripped of its authority. This is clearly the case under the cooperative approach; it is also true of consolidation because once a decision to consolidate is made, other courts suspend activity-they do not dismiss. If the consolidated case does not proceed in a reasonable time, then the individual actions can go forward. In addition, there is an important exception to the lis pendens rule: the court where a declaratory judgment action is filed is not treated as the court with supervisory authority. Instead, an intellectual property holder can file a coercive suit that, essentially, vetoes any attempt by the defendant to use a "torpedo action" (a declaration for a finding of noninfringement or invalidity filed in a court known for delay 32 ) to postpone adjudication.

To put this another way, the Principles improve upon the current system because the power to transfer cases carries with it the ability to choose a court that is expert and speedy. Moreover, the system as a whole reduces sharp practices by eliminating the benefits of forum shopping. Because there are many places where defendants are subject to adjudicatory authority, there is little advantage in situating activities or bringing declaratory actions in "information havens." By the same token, plaintiffs may not receive much benefit from suing in "information hells" because such cases are subject to transfer to a more

WHY INVITE CONFLICTS? 837 appropriate forum. As explained more fully below, control over applicable law further reduces the effects of forum shopping.

Although one of the ultimate goals of the project is to create a platform for enforcing judgments, this section has yet to receive focused attention by the Reporters or by their Advisers. To a significant extent, the current language is a placeholder. As much as possible, it will be conformed to any instrument that the Hague succeeds in promulgating. Even more important, the American Law Institute will likely expect the Principles to agree generally with its own Project on Recognition and Enforcement of Foreign Judgments: Analysis and Proposed Federal Statute. This work, which was formerly entitled the International Jurisdiction and Judgments Project, has been under consideration at the ALI for several years. It sets out uniform criteria for determining the recognition and enforcement of judgments in the United States; this Project is to be readied for adoption by the Institute in May 2005.

However, a few features of the current draft will probably endure. First, the Principles give the court where enforcement is sought responsibility to act as a check on the court that rendered the judgment. Most obviously, the enforcing court cannot enforce judgments predicated on prohibited bases of jurisdiction. In addition, the enforcing court must verify that the defendant received notice of the original action; in cases where the jurisdictional predicate is a general appearance, it must also verify that the defendant indeed waived objections to personal jurisdiction; if the rendering court was chosen in a nonnegotiated contract, the issue of reasonableness must be reviewed. Further, courts are to refuse to enforce judgments rendered in conflict with the Principles' lis pendens provisions and with decisions the court with supervisory authority makes on cooperation or consolidation. Thus, for example, if the first action was filed in France and that court decided to consolidate the worldwide dispute in Germany, then decisions rendered by any court other than the German court should not be enforced. Although this system of second-guessing may appear destabilizing of decisions and is certainly contrary to practice in many places, it is intended to compensate for the lack of hierarchical supervision present in other adjudicatory systems.

Second, the Principles include features that recognize the special import of intellectual property values, and the impact of intellectual property rights on culture, health, and well-being. Thus, the court where enforcement is sought is given some authority to vary remedies to conform the outcome to local needs. It can refuse to enforce noncompensatory awards unknown to its own law as well as awards that are grossly excessive when judged by domestic conditions. It can also decline to order injunctive relief when safety, health or local cultural policies are at issue. In addition, the Principles recognize a general, but circumscribed, exception for judgments contrary to local public policy.

As noted earlier, one of the main differences between the ALI Principles on the one hand, and other enforcement regimes and initiatives (full faith and credit, the Brussels Regulation, the Hague instruments, and the ALI Project on Recognition and Enforcement of Foreign Judgments), and one of their most innovative features, is the articulation of principles on applicable law. Further, the project provides an impetus for courts to apply the principles by denying enforcement to judgments that are based on choices of law "manifestly inconsistent" with the rules set out.

The inclusion of principles on applicable law may seem remarkable at first blush. Not only are conflicts rules missing from other enforcement regimes, they are also largely absent from the international intellectual property instruments currently in force. 33 Bill Patry has suggested that the reason for their omissions may be that until recently, there were so few cases involving multistate contacts that courts were never pre-33. Admittedly, there are a few provisions of international intellectual property law that arguably have choice-of-law overtones. Thus, the Nimmers have argued (unconvincingly) that national treatment provisions create choice of law rules on ownership. See Patry, supra note 12, at 413. Further, the Berne Convention refers to the "law of the country where protection is claimed." Berne Convention for the Protection of Literary and Artistic Works, art. 5(2), Sept. 9, 1886, 828 U.N.T.S. 221, S. Treaty Doc. No. 99-27. However, it is not clear whether this means the country where infringement occurred or the country where the case is being litigated. Accordingly, if this is a choice of law rule, it is one that is very poorly drafted.

WHY INVITE CONFLICTS? 839 sented with a choice: "there was only the law of the forum." 34 Alternatively, it may be that the paucity of global cases embedded concepts of territoriality so deeply into intellectual property jurisprudence, it was rarely evident that choices were being made. 35 However, as the prospect of international disputes has come to the fore, it has become increasingly clear that the situation has changed dramatically. 36 Thus, one of the reasons negotiations over a general convention at the Hague broke down was that Internet casesincluding especially cases involving transactions in intangible works-were beginning to proliferate. 37 Consideration of how these disputes would fare under the Hague's jurisdictional provisions demonstrated that there would surely be cases of overlapping adjudicatory authority. Further, it became clear that these overlaps would do more than draw litigants into tribunals far from their homes. When cases are litigated in far-flung fora, there is a real prospect that activity would be judged under law different (possibly unforeseeably different) from the law of the location at which the activity was conducted. 38 For example, 34. Patry, supra note 12, at 385. 35. Indeed, in an early presentation of the ALI Principles to the Advisers, a prominent jurist argued that there was no need for choice of law rules because the territorial principle was so obviously applicable.

36 38. This danger is reflected in the following question:

Regarding the Hague treaty and copyright and fair use on the Internet, what national laws would apply if I download an article, data, music or software from a European web site, to my US based computer, and make an unauthorized use, for teaching, reverse engineering, commentary, parody or some other use that would be fair use in the US, but possibly not fair use in Europe…Could I be sued in Europe for violating the European copyright laws? Would a judgment be collected against me in the USA? E-mail from James Love, Consumer Project on Technology, to Mary Streett, U.S. Dept. of Commerce (Sept. 29, 2000), available at http://lists.essent ial.org/pipermail/info-policy-notes/2000q3/000024.html; Cherie Dawson, Note, Creating Borders on the Internet: Free Speech, the United States, and International Jurisdiction, 44 VA. J. INT'L L. 637, 639 (2004); Nathan Garnett, Com-U.S. software producers began to realize that foreign use of their programs could expose their reverse engineering activities to adjudication under bodies of law that do not recognize a fair use defense.

Of course, one could hope that courts would use restraint when applying local law to foreign activity. For example, in F. Hoffman-La Roche Ltd. v. Empagran S.A., the U.S. Supreme Court recently held that U.S. antitrust law cannot be interpreted to cover foreign harm suffered by foreign defendants. 39 Noting that the decision to apply law extraterritorially requires heightened sensitivity to comity interests, Justice Breyer stated that courts are responsible for making sure that "conflicting laws of different nations work together in harmony-a harmony particularly needed in today's highly interdependent world." 40 Unfortunately, not every case has a fact pattern similar to the one in Empagran. In that case, it was assumed that foreign and domestic injuries were independent of one another. 41 International intellectual property cases are not always so easily teased apart. When the behavior in one place is necessarily intertwined with activity in another, even the highest regard for comity will produce overlapping prescriptive authority. As the Supreme Court of Canada recently noted in a case involving a transmission originating in Canada that was downloaded in the United States, "the answer lies in the making of international or bilateral agreements, not in national courts straining to find some jurisdictional infirmity in either state." 42 The bottom line is that without an international agreement, there are activities that will inevitably be subject to scrutiny under more than one body of law, leading to the possibility of unforeseen results, or even worse: to multiple liability; 43 to exposure to judgments mandating inconsistent behavior; and to the imposi-

tion of one country's cultural and industrial innovation policies on another sovereign state. 44 Nor is it always the case that restraint is the right answer. As noted earlier, there are patent claims that contemplate activity in more than one location; unless the law of some nation is applied extraterritorially, these inventions could be practiced without ever incurring infringement liability. 45 Furthermore, one way to deal with the messy problem of multi-jurisdictional infringements is to bring a single case in one jurisdiction and argue that its law should control activity occurring elsewhere. For example, there have been both copyright and trademark cases in which U.S. law has been applied to foreign activity on the theory that the extraterritorial activity affected U.S. markets. 46 The result was only rough justice (since the law at the location of the activity was not consulted), but that may be better than slow and expensive justice (for example, through suits in multiple locations).

Extraterritorial applications of law have other advantages as well. Thus, it has been forcefully argued that if the rights in each intellectual product were controlled by the law of a single jurisdiction, world-wide negotiations would be vastly facilitated. 47 For instance, it would be far easier to draft a global license covering the use of a U.S. movie in all media if there were no need to be concerned with the differences between U.S. law on work for hire and foreign rules mandating employee owner- 48 with a German law that bars the transfer of rights in undiscovered media; 49 or with France's elaborate moral rights doctrine. 50 Since facilitating such transactions would make information products more readily available, these arguments are not solely about the interests of rights holders and licensees. In fact, they have strong public policy overtones as well.

With these concerns in mind, this project has from its outset considered the issue of applicable law. Initially, the problem was dealt with through the back door, by making a judgment unenforceable if the law chosen was "arbitrary or unreasonable." 51 However, it quickly became evident that this formulation would be unworkable. Because there has been such a paucity of debate over choice of law rules in intellectual property cases, there is little shared understanding of what should count as a reasonable choice; if the issue remained open in the enforcing court, there was sure to be extensive relitigation. The current version of the project therefore confronts the problem headon by including provisions on applicable law.

Recent meetings of the Advisers have been partly devoted to working these rules out. As of this writing, it is fair to say that the hardest question is deciding between traditional notions of territoriality (which might enhance the appeal of the Principles for conservatively minded jurists) and a uniformity approach that would associate a work with a single nation's law (and break new ground in international intellectual property jurisprudence). As explained below, the Principles currently split the difference, depending on the practicalities of the situation; the relative advantages of adhering to, or departing from, tradition; and the national interests involved in the rule in issue.

As the draft now stands, the approach to these issues is largely territorial. For economic rights, this means that each of these matters is controlled by the law of each country in which an infringement occurs. Thus, for example, French law would apply to patent infringements in France, but U.S. law would apply when the same work is infringed in the United States.

As noted earlier, there was a strong temptation to apply the law of the country most connected to the work, no matter where infringement occurred, for example, to apply French law to all of these issues when a work is created in France by French domiciliaries. Not only would this simplify transactions, it would also maximize each nation's ability to encourage production within its territory and protect its creative citizens in the manner each regards as most appropriate. The decision to follow the traditional territorial approach was based on several considerations. This rule does the least damage to the ability of each state to influence the availability of intellectual products within its borders. By mimicking the outcomes that would obtain when litigation is pursued state-by-state (and each court uses its own state's law), this approach is likely to make the Principles more readily accepted. Furthermore, since states lack the capacity to alter foreign registrations, the uniformity approach would be difficult to apply to questions involving the validity of registered rights. Of course, a mixed system that treats registered rights differently from other forms of intellectual property rights could have been adopted. However, it was thought that this would be problematic because many economically important works implicate multiple intellectual property regimes. Computer games, for example, may include copyrightprotected animation and music, characters protected by rights of publicity, patented software, and marketing symbols that are protected by trademark law. 52 The Principles do, however, recognize several exceptions to territoriality. First, for noneconomic rights, the applicable law is that of the territory in which the author is habitually resident at the time the harm occurred. In a sense, this is a territorial 52. See e.g., Tom Loftus, Stars Seek More Control Over Video Games (Mar. 12, 2004), available at http://msnbc.msn.com/id/4223361. rule because it stems from the perception that noneconomic harm occurs where the author is found. However, the result is that one nation's law applies to all violations. For instance, if a copyrighted work authored by a Frenchman were utilized, moral rights issues would be analyzed under French law, irrespective of whether the work was used in France or in the United States. 53 The other exceptions are more conventional departures from territoriality. Where territorial law cannot be ascertained, the law of the forum applies. More important, in cases where the dispute is closely connected to a particular law, or to a preexisting legal relationship subject to another law, the applicable law will be that of the connection or relationship. Further, when use of protected works spills over national borders, and the decision is made to consolidate adjudication, applying every state's law may become unduly burdensome. In such cases, the court can choose to simplify the dispute by utilizing the laws most closely associated with the dispute. The criteria for choosing which laws will be applied include the locus of the plaintiff's and defendant's activities.

These exceptions may not be as great a departure from tradition as may first appear. Courts usually use their own forum's law in the case of uncertainty. In practice, parties facing multiple infringements often forgo adjudication in countries that are not closely connected to their prime business activities and needs. 54 Furthermore, it is not unknown for a plaintiff to sue a defendant in its largest market, hoping that a loss of that revenue will put the defendant out of business everywhere. In effect, the law of the largest market winds up controlling availability in all markets.

The final departure from territoriality is for agreements in which the parties choose to submit all or part of their dispute to the law of a single national law. As in other areas, the Princi-

WHY INVITE CONFLICTS? 845 ples favor party autonomy, but nonetheless circumscribe it in a variety of ways. First, the validity and maintenance of registered rights, the existence, scope and duration of rights, and the formal requirements for recordation cannot be varied contractually. Nonnegotiated agreements are subject to the usual scrutiny for reasonableness. Finally, these agreements cannot be enforced if to do so would affect the rights of third parties.

In this area, the decision has been made to adopt the uniformity approach as much as possible because assigning a single owner to world-wide rights greatly simplifies negotiations. Thus, the law that is applicable to ownership of rights created pursuant to a contract or preexisting relationship is that of the contract or relationship. In most other cases, the law chosen is that of the creator's residence at the time of the work's creation. Where the law thus designated does not provide a solution, initial title is determined by the law of the place where the work is first exploited. 55 Unfortunately, there are a few situations where the territorial rule appears unavoidable. For registered rights, the law of the country of regulation usually applies because regulation is controlled by local registries. However, in cases where the work was created pursuant to a contractual relationship, the law that governs the relationship controls, on the theory that the parties can be required to petition the place of registration for a change in title if that is what the court orders them to do. The territorial approach is also used for unregistered trademark rights. These rights arise directly out of local understanding of the source of the goods and services to which the marketing symbols are attached. Thus, the law that appropriately controls these rights is the law of the country in which the symbol at issue is conveying marketing information. 55. For example, rights of publicity are not recognized in the United Kingdom. If provisions were not made for cases in which the place of the creator's residence does not supply a solution, then a U.S. advertiser could use images of Prince William without authorization. Under the Principles, however, ownership of the right of publicity would be determined by the law of the United States if the images were first exploited there.

There are two issues that arise in connection with licenses and assignments. The first is "transferability": in some cases, intellectual property rights are inalienable or only partially alienable. As a result, there is a question on whether the right can be transferred at all. For rights that are transferable, the second issue is whether the parties took the steps necessary to effect the transfer.

On the first issue, transferability, the Principles follow the territorial approach. Thus, they provide that the transferability is controlled by the law of the state whose rights are at issue. For example, a global agreement that purports to transfer "all rights in all media" made before the discovery of DVDs is not effective to give the transferee rights to exploit the work on DVDs in Germany because (as noted above) German copyright bars transfers involving undiscovered media. This is so even if the license is effective under German law to transfer rights in known media (film, for example). The license will also operate to transfer DVD rights in the same work for exploitation in the United States.

As to the question of the effectiveness of transfers, here the Principles depart from territoriality and rely on the law of the agreement. 56 Most sophisticated parties will include a choice of law clause and this will be enforceable; in its absence, transfer will be judged under the law of the country most closely connected to the work, presumptively, the assignor's or licensor's habitual residence. As usual, nonnegotiated agreements (other than collective bargaining agreements) are to be scrutinized for reasonableness.

56. Transfer issues may not raise significant domestic policy concerns, see, e.g., Univ. of Mass. v. Robl, 2004 WL 1725418 Mass. Dist. Ct. (Aug. 2, 2004) (issue of ownership of patent rights allocated contractually does not raise a federal question). Note, however, that there is a lurking question on the law to be applied to characterizing a particular dispute as involving transferability or effective transfer.

There are at least two additional issues that call for further examination. The first is the use of mandatory rules. 57 Although unfamiliar to Americans, these rules are similar to public policy defenses to enforcement in that they operate as trumps. However, unlike public policy decisions, they apply ab initio-they go to the question of which law is used to adjudicate the case, and not to determine whether the outcome is acceptable in the place where enforcement is sought. Initially, some thought was given to barring the use of mandatory rules. However, in places that recognize them, the bar would require a departure from traditional practice and is thus not likely to be acceptable. Besides, the core territoriality principle may avoid much of the controversy because it always permits a state to apply its mandatory law to local infringements. If experience with the Principles leads to greater appreciation of the interests of other countries, the wisdom of their approaches, and the benefits of comity, then the extraterritorial application of mandatory rules will abate of its own accord.

The second under-developed issue is secondary liability. In the last few years, rights holders have begun to sue those who facilitate distribution of intellectual products, claiming that they are vicariously or contributorily liable for the infringements of users. 58 As it stands, the Principles do not break this issue out for special treatment. The theory is that since there cannot be secondary liability without primary liability, the law that governs primary liability should control. However, this rationale only goes part way to solving the problem, for once there is infringement, jurisdictions differ on the terms on which 57. See, e.g., CA Paris, 43 ch., 6 July 1989, R.I.D. A. 1990 they impose liability on third parties. 59 Of course, there is reason to think that part of the attraction of secondary liability actions is that they centralize litigation in one forum, thus eliminating the need to sue individual users in each of the countries where they are located. This project may render some of these secondary liability suits unnecessary in that the Principles envision joinder of individual users in a single action, orat the least-coordination of separate suits against them.

The Principles Governing Jurisdiction, Choice of Law, and Judgments in Transnational Disputes are intended to take up Justice Breyer's hope that "nations work together in [the] harmony ... needed in today's highly interdependent world." They recognize, however, that in many cases, courts cannot always assume interdependency away or cope with it on a unilateral basis. More is required, from the parties involved in interstate transactions, from the courts dealing with multinational disputes, and from interdependent nations themselves.

Although considerable work has gone into the drafting of these Principles, it is important to emphasize that the project is far from over. There are two lessons to be learned from the Hague's experience with a general convention on enforcing foreign judgments. The first is that input from all segments of the domestic and international bar is critical. The problems facing copyright, trademark, and patent holders are all somewhat different, as are the issues encountered in different parts of the world. The multinational composition of the Principles' Reporters and Advisers is an effort to consider these divergent viewpoints. The second lesson is that we are only beginning to fully appreciate the issues posed by a truly global marketplace. As the issues arising in an integrated economy are better understood, so too are the issues of an integrated system of dispute resolution. The hope is that this effort will be a model for adoption. But if it only serves as a starting point for debate, it will have served an important purpose.