547An Outsider’s View of the Brussels Ia, Rome I, and Rome II Regulations

Prof. Symeon C. Symeonides

Alex L. Parks Distinguished Chair in Law and Dean Emeritus, Willamette University, Oregon, USA; LL.B. (Priv. L), LL.B. (Pub. L.), LL.M., S.J.D., LL.D. (h.c.mult.), Ph.D. (h.c.), MAE

I. Introduction

The organizers of this Conference on “European Private International and Procedural Law and Third Countries” have kindly invited me to speak from the perspective of one such third country –the United States. Indeed, although a European (Greek) by birth and upbringing, I have lived and worked in the United States for more than four decades and, inevitably, my perspective is shaped by what I see from the American side of the Atlantic Ocean.

The first observation that comes to mind from that vantage point is that the European Union (EU), although technically not a federation, has been aggressively acting like one, at least as it concerns private international law (PIL). It has embarked on an ambitious campaign to codify and “federalize” the PIL of its Member States. The campaign has so far produced a whole series of EU Regulations that cover at least 80 percent of this subject. By contrast, the United States, which is a real and very old federation, has chosen not to federalize PIL, although the US Congress has the constitutional power to do so[1].

For the record, I have been and continue to be a proponent of codification and unification. I have advocated for federal legislation[2] and a new Conflicts Restatement at the national level[3]. I have also drafted and passed PIL codifications for two states 548and have the scars to prove it[4]. Consequently, I am favorably disposed towards the EU Regulations and, as my previous publications indicate, I admire many aspects of them.

At the same time, however, there are certain aspects of these Regulations which would have been much better if the drafting process had proceeded at a slower pace. This brief essay focuses only on those aspects, in three EU Regulations on PIL: Brussels Ia[5], Rome I[6], and Rome II[7]. This is not a global assessment of these Regulations. Such an assessment –which the space limitations of this Journal do not permit– would be generally favorable. Rather it is a brief discussion of only those aspects of these Regulations that can benefit from further improvement. These aspects are: (1) the treatment by Brussels Ia of defendants domiciled outside the EU and of the jurisdiction rules of third countries; (2) the extent to which Rome I protects or does not protect weak contracting parties from unfavorable choice-of-law clauses; and (3) the way in which Rome II resolves cross-border tort conflicts. Because of its limited focus, this essay may come across as more negative than it is intended to be.

Another risk, given the transatlantic vanguard point from which the essay is written, is that it may be taken as just another example of an “America-knows-best” mindset. Nothing would be further from the truth. To criticize European law on one point does not mean that American law is necessarily better on that point. It may or may not be. For example, in this essay I praise Rome I for protecting consumers, employees, and some other weak contracting parties and criticize it for not protecting some other weak parties, such as franchisees. At the same time, I note that American law is worse because it fails to legislatively protect most weak parties.

In any event, the goal of outside evaluations like this essay is not to proclaim a winner in comparing legal systems but to explore ways in which each system can benefit from the successes or the failings of the other.

549II. The Brussels Ia Regulation

The first of the three Regulations discussed here is also the oldest in origin. It began as the Brussels Convention in 1968[8] and was later converted into an EU Regulation in 2001[9] and finally recast in 2012[10]. This discussion focuses on the last iteration, the 2012 Recast, which is known as Brussels Ia[11].

From the EU perspective, the regime established by the Brussels Convention and its successor instruments is an unqualified success story, an admirable monument to the vision of European legal integration. It is certainly far more comprehensive than similar regional integration efforts in the Americas and elsewhere[12]. Indeed, its jurisdiction rules are much clearer and more efficient than the imprecise tests developed by the US Supreme Court in interpreting the Due Process clause of the Constitution in the last three generations[13].

In short, when one looks at the Brussels regime from the inside, one finds much more to praise than to criticize. However, from the outside perspective from which this essay is written, the success of the Brussels regime is tainted by the disappointing features which are discussed below.

1. Discrimination against Third-Country defendants

From this outside perspective, the most regrettable feature of the Brussels Ia regime is its discrimination against defendants domiciled outside the EU (hereinafter third-country defendants) for purposes of jurisdiction and, ultimately, recognition[14]. The drafters did a commendable job in singling out and blacklisting the exorbitant jurisdictional bases which were available under the laws of the Member States, such as the plaintiff’s nationality under Article 14 of the French Civil Code, or the presence of 550the defendant’s property under Section 23 of the German Code of Civil Procedure, or Article 40 of the Greek Code of Civil Procedure[15]. The black list was an excellent idea, except that it has that color only for some defendants –those domiciled in an EU Member State. Although the blacklisted bases may not be used against those defendants, they remain available against third-country defendants[16]. For them, the list is not black but green, since it gives the green light for suits against them.

Herein lies the first inconsistency. If the location of the defendant’s domicile was relevant for determining whether a jurisdictional basis is exorbitant, then the differentiation among defendants would point in the opposite direction. Using as an example Article 14 of the French Civil Code (a blacklisted rule that grants French courts jurisdiction, inter alia, when the plaintiff is a French national), it is more of an overreach, and more burdensome, to subject a Brazilian defendant to French jurisdiction under that article than to subject a Belgian defendant to the same. Obviously, however, geography was not the drafters’ rationale. We shall return to this point later. Suffice it to say that this feature of the Brussels Ia is almost unique in the international arena. It certainly stands in sharp contrast with the jurisdictional law of most countries in the Americas, which treat local and foreign defendants alike[17].

This feature of Brussels Ia is bad enough, but it is made worse by a second unique feature. The recognition part of Brussels Ia abandons the differentiation between defendants and treats all judgments rendered under the Regulation, including those based on a blacklisted basis, under the same relatively liberal standards of easy recognition and enforcement[18]. Moreover, the Regulation prohibits the state addressed from reviewing the jurisdiction of the state of origin and specifically precludes the application of the public policy exception “to the rules relating to jurisdiction”[19]. Thus, this second feature compels a Greek court to recognize a French Article-14 judgment against a Brazilian defendant, but prohibits the same court from recognizing a judgment rendered on the same jurisdictional basis against a Belgian 551defendant. Arthur von Mehren justifiably characterized this feature of the Brussels regime as “the single most regressive step that has occurred in international recognition and enforcement practice in [the twentieth] century”[20]. Even European authors, such as Trevor Hartley, castigate this “parochial attitude”[21], characterizing it as “a retrograde step” vis-à-vis the rest of the world[22]. As he points out, “[i]n Europe, Americans and other non-Europeans [defendants] are systematically discriminated against; in the United States, on the other hand, there is no discrimination against Europeans and other foreigners”[23].

One who wishes to defend the discrimination against third-country defendants can find some reasons to defend the first feature described above. For example, at the inception of the Brussels regime, one could argue that the European Community lacked institutional competence to force upon Member States uniform jurisdictional rules in cases involving third-country defendants[24]. Although that justification no longer exists, one could argue that, as matter of policy, the European Union has less of an interest in how Member States treat third-country defendants and thus has little justification to intrude into the procedural laws of those states. However, in its 2012 Recast proposal, the European Commission itself eliminated that argument by proposing that third-country defendants be subjected to the same jurisdictional rules that Brussels Ia provided for EU defendants[25]. Ironically, the rationale for this 552proposal was not a desire to eliminate the discrimination against third-country defendants but rather a desire to provide all EU plaintiffs with the same access to a forum, regardless of the defendant’s domicile[26]. In the end, this proposal survived in part. Articles 18(1) and 21(2) dealing with consumers and employment contracts, respectively, subject third-country defendants to the same jurisdictional bases as EU defendants, as do Articles 24 and 25, dealing with exclusive jurisdiction and exclusive forum selection clauses, respectively. These articles are discussed later.

Finally, one could defend the jurisdictional discrimination of third-country defendants by pointing out that other countries also have discriminatory rules, or have exorbitant jurisdictional bases that mistreat non-forum defendants (whether domiciled in sister states or foreign countries). For example, in the United States, tag jurisdiction is still constitutionally permissible under certain circumstances[27] and, until recently, the same was true for “doing business” as a basis for general jurisdiction[28]. The difference is that judgments rendered under these bases have no extraterritorial effect because the United States cannot force any other country to recognize them. Only in the context of an international convention or other similar supranational instrument is there an ability to bestow such an effect. In contrast to the two Inter-American Conventions on jurisdiction and recognition, which do not authorize such discrimination[29], the Brussels Ia regime is the only supranational regime that both authorizes exorbitant jurisdictional bases against third-country defendants and compels recognition of the resulting judgments in other contracting (i.e., EU) states. As Fritz Juenger noted, this is a “transgression upon comity and general decency”[30].

553For years, the pragmatic but often unspoken rationale for this discrimination was that it would give the EU the necessary bargaining power to negotiate with other countries. For example, and most notably, the EU could use that power to negotiate under the auspices of The Hague Conference of Private International Law, a worldwide comprehensive international convention on jurisdiction and recognition of judgments that would (or could) preclude recognition of judgments grounded on jurisdictional bases commonly regarded as exorbitant. Indeed, a comprehensive attempt for such a convention in the 1990s aimed for a double or mixed convention that would place exorbitant bases on a blacklist. Unfortunately, that attempt failed, producing only the much narrower Choice-of-Court Convention of 2005[31].

The process of drafting a broader Hague Convention began again in 2012, but disagreements on matters of direct jurisdiction were too sharp to allow a workable compromise. Consequently, it was decided early on that the new convention should not attempt to regulate direct jurisdiction, but should instead provide only for indirect jurisdiction as a condition of eligibility for recognizing the resulting judgments. Even on that matter, however, concessions were necessary. Unsurprisingly, the EU surrendered all the blacklisted exorbitant jurisdictional bases, but also some other bases, such as those that allow EU consumers and employees to sue in their home states. On their part, the United States surrendered “tag jurisdiction”[32] (which was also included in the blacklist of Brussels Ia), but also the infamous “doing business” basis of general jurisdiction against corporations, which was the major point of contention in the previous round that led to the 2005. However, this was not a real concession because in the meantime the US Supreme Court had essentially eliminated this jurisdictional basis[33].

In 2019, the Hague Conference approved a new convention, the Hague Convention on the recognition and enforcement of foreign judgments in civil or commercial matters. This Convention “deals with” exorbitant jurisdictional bases only by not including them in the agreed-upon indirect jurisdiction bases. Aiming for the broadest international consensus, the Convention provides a small number of narrow indirect jurisdiction bases, which cover only the most uncontroversial scenarios[34]. For example, with regard to non-contractual obligations, the Convention proposes an indirect jurisdiction rule that is drastically narrower than Article 7(2) of Brussels Ia. That provision covers all non-contractual obligations and, as interpreted by the ECJ, grants direct (and indirect) jurisdiction to an EU Member State in which either the conduct or the injury occurred. By contrast, Article 5.1(j) of the Convention 554covers only non-contractual obligations “arising from death, physical injury, damage to or loss of tangible property” and only if the conduct “directly” causing the injury occurred in the state of origin[35]. It excludes numerous other torts, such as those not involving or not connected to physical injury or involving loss of intangible property. Most notably, it excludes situations involving the torts covered by Article 5.1(j), but in which only the injury –but not the injurious conduct– occurred in the state of origin. One need only think of products liability to realize that these latter cases are far more numerous than the cases covered by the convention. In other words, this is a minimalist convention that does not rise above the lowest common denominator.

Interestingly, during the negotiations for the convention, the EU did not advocate for the adoption of a rule similar to Article 7(2) of Brussels Ia[36]. Apparently, the EU did not want to expose EU defendants to indirect jurisdiction in third countries whose only contact would be the occurrence of the injury. To be sure, the United States, and possibly some other countries, were unlikely to agree to an unqualified rule such as Article 7(2). However, had the EU pushed for such a rule, a compromise could have been reached, like the compromise of Article 5.1(g) of the Convention, which covers contractual obligations. That provision grants indirect jurisdiction to the state of performance of the contract (consistently with the Brussels regime) but also provides an escape for cases in which that state’s connection is not “purposeful and substantial”[37] (consistently with US jurisdictional principles).

In summary, because of the narrow scope of Article 5.1(j), a judgment rendered in a non-EU country that was the place of the injury but not the place of conduct will not circulate under the convention[38]. Likewise, a judgment rendered in an EU Member State under the same circumstances will also not circulate outside the EU under the convention; but it will circulate within the EU under Brussels Ia[39]. Thus, the long-awaited 555Hague Convention will not remedy the discriminatory treatment of third-country defendants that the Brussels Convention initiated and its successors continued. If the reason for this treatment was to serve as a bargaining chip for forcing or cajoling other countries to accept a balanced multilateral (direct or indirect) jurisdiction convention, that strategy has not paid off. Moreover, if past experience is any indication, the bilateral convention route does not appear any more promising.

2. Exclusive Jurisdiction

Article 24 of Brussels Iα (which is one of the articles that applies “regardless of the domicile of the parties”), grants exclusive jurisdiction, inter alia, to the courts “of a Member State”[40] in which an immovable is situated to adjudicate “rights in rem” (hereinafter “real rights”) in that immovable[41]. Article 26 complements Article 24 by prohibiting the courts of other EU Member States from entertaining actions involving real rights in the same immovable, as well as actions involving other subjects for which Article 24 assigns exclusive jurisdiction to the states designated therein[42].

To be sure, there is nothing wrong with this grant of exclusive jurisdiction. The problem arises from the fact that Article 24 confines this grant to the courts “of a Member State”[43]. Obviously, the EU cannot confer jurisdiction on the courts of third countries, but it could and should respect the equivalent claims of exclusive jurisdiction of third countries in analogous circumstances. The refusal of Brussels Ia to recognize these claims creates another invidious differentiation between immovables situated within and without the EU. For example, under Articles 24 and 26, a Greek court must refrain from entertaining an action involving real rights in a Bulgarian immovable, but the same court need not do so in a similar action involving real rights in a Serbian immovable[44]. Moreover, if the Greek court does entertain the latter action, the resulting judgment “shall be recognized in the other Member States” because it is a judgment permissibly “given in a Member State”[45].

556From the perspective of third countries, this denial of effet réflexe is another problematic feature of Brussels Ia. Koji Takahashi characterized this feature as “repugnant to comity,” reasoning that “[i]f the sovereignty of a Member State in which a connecting factor under Article [24] is situated warrants respect, there is no good reason why the same respect should not be accorded to the sovereignty of a non-Member State in which the same connecting factor is situated”[46]. It is unfortunate that the Commission failed to address this issue in its 2010 Recast proposal, although a carefully drafted provision proposed at that time by the Groupe européen de droit international privé (GEDIP) would have resolved this matter in an appropriate way. In the above hypothetical, the provision would require the Greek court to stay its proceedings if Serbia’s claim of exclusive jurisdiction were “analogous” to the claim under Article 24, and to decline jurisdiction once the Serbian court rendered a judgment entitled to recognition under Greek law[47].

In contrast to Brussels Ia, the Hague Convention of 2019 respects the exclusive jurisdiction of the situs state, even if that state is not a party to the convention, by imposing a negative obligation not to recognize a judgment that does not originate from the situs state. Article 6 of the Convention provides that a judgment that ruled on real rights in immovable property shall be recognized and enforced “if and only if the property is situated in the State of origin”[48]. Moreover, Article 15, which provides that the Convention “does not prevent the recognition or enforcement of judgments 557under national law,” is “[s]ubject to Article 6”[49]. The combination of these two articles means that, in the above hypothetical involving a Serbian immovable, a Greek judgment purporting to adjudicate real rights in that immovable will not be recognizable under the convention in any contracting state, even if Greece is (and Serbia is not) a contracting state. However, under Brussels Ia, the same judgment will circulate within the EU because it is a judgment permissibly “given in a Member State”[50].

3. Choice-of-Court Agreements

Article 25, another Brussels Ia article that applies “regardless of [the parties’] domicile” and the second article granting exclusive jurisdiction, suffers from the same problem as Article 24. Article 25 provides that, if the parties have validly agreed to confer jurisdiction to a court “of a Member State,” that court “shall have jurisdiction” and “[s]uch jurisdiction shall be exclusive unless the parties have agreed otherwise”[51]. Consequently, if that court is seized[52], it “has priority” to determine the validity and scope of the agreement[53] and the courts of other Member States must decline jurisdiction unless the agreed upon court declares that it has no jurisdiction under the agreement[54].

Article 25 deserves extensive discussion, especially regarding its problematic treatment of the choice-of-law question when raised before a court other than the one designated in the agreement[55]. However, for the purposes of this brief “external” 558review, the more relevant issue is the confinement of the Article 25 to agreements conferring jurisdiction on “a court or the courts of a Member State”[56] and the resulting exclusion of agreements conferring jurisdiction on the courts of third countries.

The question is whether this exclusion is eliminated or affected by the EU’s subsequent accession to Hague Convention on Choice-of-Court Agreements of 2005. This convention requires contracting states to enforce exclusive choice-of-court agreements falling within its scope and designating the courts of a contracting state (as well as judgments based on such agreements). Article 26.6 of the Convention provides that the convention “shall not affect” the application of the rules of a Regional Economic Integration Organisation that is a Party to this convention, such as the EU, “where none of the parties is resident in a Contracting State that is not a Member State of the Regional Economic Integration Organisation”[57]. Stated affirmatively, the Choice-of-Court Convention does affect, i.e., it displaces the relevant provisions of Brussels Ia only in cases involving exclusive choice-of-court agreements choosing a court in a Convention State and in which at least one of the parties is domiciled in a Convention State that is not an EU Member State.

At the time of this writing, four countries fall within this category: Mexico, Montenegro, Singapore, and the United Kingdom. Thus, if a domiciliary of one of those four countries enters into an exclusive choice-of-court (regardless of the other party’s domicile) which chooses a court of a Convention State, the agreement and the resulting judgment will be governed by the Convention and not Brussels Ia. However, Brussels Ia will continue to apply to the exclusion of the Convention if any of the above elements are changed. For example, if the agreement chooses a court in a non-Convention State, or if the parties are domiciled in non-Convention States, the Convention does not displace Brussels Ia, which in turn does not displace the laws of the Member States on this matter. In conclusion, the EU accession to the Choice-of-Court Convention obligates EU Member States to enforce an exclusive choice-of-court agreement choosing the courts of a non-Member State, but only if that state is a party to the Convention and agreement involves parties at least one of whom is domiciled in such a state.

Something similar will occur with the Hague Judgments Convention of 2019, which will go into effect in 2023 and has a broader scope than the 2005 Convention, inter alia, because it includes non-exclusive choice of court agreements. Article 23.4 of the 2019 Convention provides that this Convention “shall not affect” the application of the pre-existing rules of a Regional Economic Integration Organisation that is a Party to this Convention (such as the EU) “as concerns the recognition or enforcement 559of a judgment given by a court of a Contracting State that is also a Member State of the Regional Economic Integration Organisation”. Thus, the Convention will not affect the Brussels Ia rules that govern the recognition by one EU Member State of judgments rendered in another Member State. However, for the purposes of this essay, the relevant question is whether the Convention will affect the rules that govern the recognition of non-EU judgments which today is governed by the national laws of the Member States. The relevant provisions of the Convention on this question are the general provisions of Articles 4 and 7 and the specific provisions of Αrticles 5.1(m) and 7.1(d), which deal with choice-of-court agreements. Article 4 provides that “a judgment given by a court of a Contracting State… shall be recognized” and that recognition “may be refused only on the grounds specified in this Convention”, primarily in Article 7. Article 5.1(m) provides that a judgment is “eligible” for recognition if it was given by a court designated in non-exclusive choice-of-court agreement. At present, Ukraine is the only non-EU country that has acceded to the Convention, but this is sufficient for the Convention to enter into force. Thus, if in the future a Ukrainian court renders a judgment based on a non-exclusive choice-of-court agreement, that judgment would be eligible for recognition in Greece under Article 5.1(m).

Article 7.1(d) deals with the converse scenario –judgments rendered in violation of a choice-of-court agreement. It provides that recognition “may be refused” if the proceedings in the court of origin “were contrary to an agreement… under which the dispute in question was to be determined in a court of a State other than the State of origin”. According to the Explanatory Report, this provision applies when the choice of court agreement “validly excluded the jurisdiction of the court of origin, irrespective of whether the agreement is exclusive or non-exclusive” and “irrespective of whether the court chosen by the parties… was the court of a Contracting State or a third State”[58]. Thus, under Article 7.1(d), a Greek court “may” deny recognition to a Ukrainian judgment that was rendered in violation of a choice-of-court agreement choosing the courts of another country, regardless of whether that country was a Convention country. Obviously, the word “may” in Article 7 permits but does not require non-recognition. In this sense, the Hague Convention does not alter the present state of affairs under which the recognition of third-country judgments by EU states is governed by the national laws of those states. Article 15 of the Convention points in the same direction by providing that the Convention “does not prevent” the recognition of judgments under national law.

4. Lis Pendens and related actions

One of the improvements of the Brussels Ia Recast over its predecessor Regulation was the introduction in Articles 33 and 34 of soft discretionary lis pendens rules in favor of actions pending in third countries. These articles apply when the jurisdiction 560of a Member State court is based on Articles 4, or 7-9. By contrast, Articles 33 and 34 do not apply when the jurisdiction of the Member State court is based on other articles, such those that provide for jurisdiction in insurance, consumer, or employment contracts (Arts. 10-23), exclusive jurisdiction (Art. 24), or choice-of-court agreements (Art. 25).

Article 33 of the Recast provides in part that if, at the time a Member State court is seised under any of the jurisdictional bases of Articles 4 or 7-9, an action on the “same cause of action and between the same parties” was pending in a third state, the court of the Member State “may” stay the proceedings if: “(a) it is expected that the court of the third State will give a judgment capable of recognition… in that Member State; and (b) the… stay is necessary for the proper administration of justice”[59]. It is noteworthy that one of the factors for determining whether a stay is necessary for the proper administration of justice is “whether the court of the third State has exclusive jurisdiction in the particular case in circumstances where a court of a Member State would have exclusive jurisdiction”[60].

Article 34 provides a similar rule for “related” actions pending in a third country, even if they do not involve the same parties or cause of action. The article provides that if, at the time an action was filed in a court of an EU Member State under any of the jurisdictional bases authorized by Articles 4 or 7-9, a “related” action was pending in a third country, the court of the Member State “may” stay the proceedings if: “(a) it is expedient to hear and determine the related actions together to avoid the risk of irreconcilable judgments…; (b) it is expected that the court of the third State will give a judgment capable of recognition… in that Member State; and (c) the… stay is necessary for the proper administration of justice”[61].

Articles 33 and 34 represent a welcome and commendable gesture of good will toward third countries. It is unfortunate that this is virtually the only such gesture in the 2012 Recast.

5. Consumer or Employment Contracts

Another change introduced by the 2012 Recast was to extend against third-country defendants some of the jurisdictional rules previously applicable against EU defendants. Article 6 of the Recast reiterated the rule that third-country defendants may be sued under any of the jurisdictional bases (including the blacklisted ones) available under national law, but, in order to “ensure the protection of consumers and employees”[62], added the phrase “subject to Article 18(1) [and] Article 21(2)”. In this context, of course, “subject to” means in addition to the national-law bases.

561Article 18(1) provides that a consumer may sue the merchant in the consumer’s domicile, while Article 21(2) provides that an employee may sue the employer in the place “where or from where” the employee habitually works or the last place where he or she did so, or, in the absence of such a place, in the place “where the business which engaged the employee is or was situated”[63]. Employers cannot reasonably complain that the places listed in Article 21(2) lack a sufficient connection with the employer’s activities. The same is true with regard to merchants who are sued in the consumer’s domicile under Article 18(1). This is because this article incorporates the requirements of Article 17(1)(c), which provides that the merchant must pursue commercial or professional activities in the consumer’s domicile or direct such activities to that state[64]. In any event, the above articles do not treat third-country defendants any worse than EU defendants. In fact, the characteristic feature of the Brussels Ia provisions on consumer and employment contracts is that they are biased against both classes of defendants. This is a conscious and unapologetic policy choice of protecting the presumptively weak parties, a choice which I applaud.

Particular praise is also due for the policy reflected in Articles 19 and 23, which prohibit pre-dispute choice-of-court agreements unless they favor the consumer or employee, respectively. This courageous and wise policy choice stands in sharp contrast with the policy of the United States Supreme Court, which sees nothing wrong with pre-dispute forum-selection clauses (to say nothing of arbitration clauses, class-action waivers, and class-arbitration waivers) that, more often than not, deprive consumers and employees of their meaningful day in court[65]. As an American expert observed, “[t]he Court consistently has turned a blind eye and deaf ear on the problem of consumer forum-selection and arbitration clauses, instead merging consideration of consumer agreements with jurisprudence developed in the dissimilar context of sophisticated business partners freely negotiating at arm’s length”[66]. The result is that “forum-selection clauses will almost always provide defendants with a ‘heads I win, tails you lose’ forum preference”[67].

6. Brussels Ia: Conclusion

The Brussels Convention was then and remains today the most successful international convention of its kind. It converted an economic alliance of independent 562states into an area of mutual trust in civil and commercial litigation, laying the foundations for a legal quasi federation. In comparing the Convention with the jurisdictional regime of the US Federation, the great comparatist Fritz Juenger applauded the “impressive progress” of the European Community as reflected in the Convention’s “functional and pragmatic” approach to jurisdiction, which he found preferable to the “reliance on an ‘imprecise inquiry’”[68], which characterizes the jurisdiction law of the United States. In his view, the Convention demonstrated that “multi-state jurisdictional problems are amenable to rational solutions, and that national sovereignty need not inhibit the framing of workable rules”[69].

However, Juenger also castigated the Convention’s “discriminatory features” and suggested that, in the same way that US scholarship could be enlightened by the European experience in crafting clear and workable rules, “the Europeans might gain from paying attention to American ideas about fundamental rights and procedural fairness”[70]. He concluded that “[i]t would be deplorable if these two major systems, linked by political realities and a shared belief in the rule of law, were to disregard each other’s accomplishments”[71].

Almost four decades after Juenger wrote these words, his conclusions and admonitions retain their cogency. The European Union succeeded the European Community and continued the progress attained by the Brussels Convention by smoothing its rough edges and filling its gaps through the Brussels I Regulation and its Recast, as well as by enacting other successful Regulations. It is unfortunate that, despite all this progress, the “discriminatory features” of the Brussels regime remain largely intact.

At the same time, it cannot escape one’s attention that, more than five decades after the Brussels Convention, we live in a world in which discriminating against “the other” has become a political banner for some forces and threatened –or real– trade wars and retaliations initiated primarily by one particular political leader and soon followed by others, are becoming commonplace. Compared to the ugliness of that political reality, the discriminatory features of the Brussels regime are small potatoes. Nevertheless, if private international law is, as Werner Goldschmidt taught us, the derecho de la tolerancia[72], we must continue to speak up against intolerances, be they big or small, new or old.

563III. The Rome I Regulation

Like Brussels Ia, the Rome I Regulation of 2008 traces its origin to an earlier convention, the EEC “Rome Convention” of 1980 on the Law Applicable to Contractual Obligations[73]. Like the 1968 Brussels Convention, the Rome Convention is another success story as illustrated by the pervasive influence it has exercised on subsequent PIL codifications, not just in Europe but in other continents[74]. One of the Convention’s novelties, which today is taken for granted, was to separate for protective treatment consumer and employment contracts from all other contracts. The Rome I Regulation added contracts for the carriage of passengers and certain insurance contracts to the protected class. This differentiation enabled the drafters of these two instruments to protect the weak parties in the above enumerated contracts from unfavorable choice-of-law clauses and then to provide a very liberal regime for those clauses in all other contracts.

This was a very sensible differentiation, but several questions remain, including: (1) whether Rome I adequately protects the weak parties in the above enumerated contracts, and (2) whether other presumptively weak parties should have received the same, similar, or some protection. This section discusses only these two questions[75].

1. Consumers and Employees

For consumer contracts and individual employment contracts, Articles 6(2) and 8(1) of Rome I provide that a choice-of-law agreement may not deprive a consumer or an employee of the protection of the (simple) mandatory rules of the otherwise applicable law, which is hereinafter referred to as the lex causae. In consumer contracts, the state of the lex causae is the country in which the consumer has their habitual residence, provided that the other party pursues commercial or professional activities in that country or directs such activities to that country or to several countries including that country[76]. In employment contracts, the state of the lex causae is ordinarily the country in which (or from which) the employee habitually works, unless the contract is more closely connected with another country[77].

564In essence, Articles 6(2) and 8(1) allow the possibility of “double protection” under the chosen law and the lex causae. The consumer or employee can enjoy the protection of whichever of the two laws is more protective, and, in some instances, can enjoy the protection of both laws for different aspects of the contract. This may appear too generous to the consumer or employee, but the other contracting party may easily avoid it by simply not choosing a law other than the lex causae. Moreover, the above provisions may be a bit too generous to consumers and employees by guaranteeing the protection of all mandatory rules of the lex causae without requiring that those rules embody a strong public policy. This, however, is the policy choice made by the drafters of Rome I. One reason for respecting this policy is because, as a general proposition, it is better to err on the side of over-protecting, rather than under-protecting, consumers or employees.

2. Passengers and Insureds

In contracts for the carriage of passengers, Article 5(2) of Rome I limits the parties’ choice to the laws of the country of: (a) the passenger’s habitual residence; (b) the carrier’s habitual residence or place of central administration; or (c) the place of departure or destination[78]. With regard to insurance contracts, Article 7 differentiates between contracts covering “large” risks[79], wherever situated, and contracts covering other risks situated within the territory of an EU Member State (hereinafter referred to as "small-risk" contracts). Article 7(3) requires a specified geographic relationship with the chosen state only in the case of small-risk contracts[80]. The parties to such contracts may choose only the law of: (a) a Member State in which the insured risk is situated; (b) the country in which the insured has his habitual residence; or (c) in the case of life insurance, the law of the Member State of which the insured is a national[81].

To examine whether Rome I adequately protects passengers, let us consider the following hypothetical case. A carriage contract between a Greek passenger and a British air carrier headquartered in Jersey Island for a trip from Greece to Ireland contains a Jersey Island choice-of-law clause[82]. In such a case, the clause would meet 565the geographical requirements of Article 5(2) because Jersey is the place of the carrier’s central administration. Suppose now that Jersey law deprives the passenger of the protection provided by Greece’s mandatory rules. Should the choice-of-law clause be upheld in such a case?

In the absence of Article 5, the answer would be negative because this contract would qualify as a consumer contract and, under Article 6(2), the clause would be disregarded to the extent it deprives the consumer of the protection of Greek law, which would be the lex causae. However, as Article 6(1) expressly declares, Article 5 prevails over Article 6 for passenger contracts, and thus the passenger does not enjoy the protection that Article 6 provides for consumers[83]. The passenger can invoke the provisions of Articles 3(3), 3(4), 9(3), 9(2) and 21, but it is doubtful that any of them will lead to the avoidance of the choice-of-law clause. Specifically:

(1) Paragraph 3 of Article 3 would not help the passenger because not “all other elements relevant to the situation […] are located in a country other than the country whose law has been chosen”[84]. Here, the relevant elements are located not in a single country, as the quoted provision contemplates, but rather in four countries.

(2) For the same reason, paragraph 4 of Article 3 will also be unavailable to the passenger because not “all other elements […] are located in one or more Member States”[85]. One of those relevant elements, the carrier’s principal establishment, is located outside the EU. Thus, even if Jersey law violates existing mandatory rules of Community law, the contractual choice of Jersey law may not be disregarded.

(3) Paragraph 3 of Article 9 would help the passenger only if: (a) Greece (the country of the passenger’s habitual residence and place of departure) qualifies as the state of performance; (b) the pertinent Greek mandatory rules would qualify as “overriding” mandatory rules; and (c) those rules would render the performance of the contract unlawful. If, as is likely, any one of these conditions is missing, the chosen law must be applied[86].

(4) Paragraph 2 of Article 9 would help only if: (a) Greece is also the forum state (more on this later), and (b) its mandatory rules would qualify as “overriding” mandatory rules.

(5) Article 21 would help only if Greece is the forum state (see below) and the application of Jersey law would be “manifestly incompatible” with the Greek ordre public. Anything less than that would not defeat the chosen law.

566(6) Finally, there is no guarantee that Greece will be the forum state. The passenger may be prevented from suing in Greece if, for example, the contract contains an exclusive choice-of-court agreement mandating litigation in Jersey Island or another third country. Brussels Ia does not prohibit such an agreement.

In summary, the contractually chosen law will be applied in the above hypothetical even if it deprives the passenger of the protection provided by the mandatory rules of the country whose law would have been applicable in the absence of choice (lex causae). This result is regrettable. For all practical purposes, the passenger is a consumer who should be protected from the adverse consequences of a contractual choice of law for the same policy and practical reasons for which Article 6 protects other consumers. Article 5(3) purports to protect passengers by geographically circumscribing choice-of-law clauses. But as this hypothetical illustrates, geographical limitations are a poor substitute for substantive protections.

The same can be said about small-risk insurance contracts. In many of these contracts, the insured would qualify as a consumer had Article 7 not displaced Article 6. Article 7(3) attempts to protect insureds by limiting the countries whose laws can be chosen in a choice-of-law clause. However, as the above discussion of the passenger contract illustrates, a geographical limitation does not necessarily guarantee meaningful protection for the insured. For example, in a life insurance contract, a contractual choice of the law of the insured’s nationality satisfies the geographical requirement of Article 7(3)(c), but may well deprive the insured of the protection of the mandatory rules of a country that has a closer connection, such as the country in which the insured has their habitual residence and in which the insurance contract was applied for, issued, and delivered, and the risk materialized. For reasons explained above, the tools provided by Articles 3(3)-(4), 9, and 21, would probably not alter this equally regrettable result.

3. Other Weak Parties

In principle, the liberal treatment that Rome I accords to choice-of-law clauses in “all other contracts”, i.e., contracts other than the ones discussed above, is a good policy choice for commercial contracts in which the parties have a relatively equal bargaining power. However, some commercial contracts involve parties with drastically unequal power. A franchise contract is an example of such a contract. It does not qualify as a consumer contract under Article 6(1) because it is not a contract “outside the trade or profession”[87] of either the franchisee or the franchisor. Yet, in many franchise contracts, the franchisee is likely to be in as weak a bargaining position as most consumers. Recognizing this fact, many states of the United States have enacted statutes regulating franchises operating in their territory and usually involving franchisees domiciled there. Typically, these statutes prohibit waivers of 567the statutory provisions, either directly or through the contractual choice of another state’s law. Thus, when a contract purports to opt out of such a statute through a choice of another state’s law, courts routinely strike down the choice-of-law clause. Cases so holding are abundant[88]. In fact, more often than not, such clauses are disregarded even if the statute in question does not expressly prohibit them, at least when the case is litigated in the franchisee’s home state[89].

Unfortunately, Rome I does not seem to provide franchisees with sufficient protection against onerous choice-of-law clauses. The only provision of Rome I on franchise contracts is Article 4(1)(e), which provides that, in the absence of a choice-of-law clause, a franchise contract is governed by the law of the country where the franchisee has their habitual residence[90]. Thus, a contractual choice of the law of any other state will be upheld unless it exceeds the infrequently operable limitations of Articles 3(3)-(4), 9, and 21 of Rome I, which were discussed earlier.

Suppose, for example, that a franchise contract between Starbucks, a corporation headquartered in the State of Washington, and a Greek franchisee for a Starbucks franchise in Greece contains a choice-of-law clause selecting the law of the friendly Kingdom of Tonga, as well as a clause allowing the franchisor to unilaterally terminate the franchise under certain circumstances that would not be sufficient under either Washington law or Greek law. In such a case, the clause will easily pass the test of paragraphs 3 and 4 of Article 3 because one important contact is located in Washington and thus not all contract connections are located in a single country (Greece) or in one or more EU countries. The clause will also pass the test of Article 9, unless the chosen law: (a) violates the “overriding” mandatory rules of Greece, either in its capacity as the forum state (Art. 9.2) or as the state of performance (Art. 9.3), but in the latter case only if those rules render the contract illegal; or (b) is manifestly incompatible with the Greek ordre public (Art. 21). Nothing short of these two high thresholds will prevent the application of Tongan law. Moreover, an exclusive choice-of-court clause mandating litigation outside Greece will make the franchisee’s position even more onerous.

4. Contractual Choice of Law for non-contractual issues

Article 14 of Rome II compounds the problem of small commercial actors, such as franchisees, because it authorizes even pre-dispute choice-of-law agreements for non-contractual issues. Such agreements are enforceable if: (a) the parties are “pursuing a commercial activity”; (b) the agreement is “freely negotiated”; and (c) the choice of law is “expressed or demonstrated with reasonable certainty by the 568circumstances of the case”[91]. The only restrictions to these agreements are those of (a) the mandatory rules of a state in which “all the elements relevant to the situation… are located” in fully-domestic cases[92]; (b) the mandatory rules of Community law, in multi-state intra-EU cases[93]; and (c) the “overriding” mandatory rules and the ordre public of the forum state in all cases[94]. By limiting pre-dispute choice-of-law agreements to situations in which all the parties are “pursuing a commercial activity,” Rome II seeks to protect certain presumptively weak parties, such as consumers, employees, and certain –but not all– individual insureds. This limitation, however, leaves exposed a whole host of small commercial actors, such as small businesses.

Let us return to the example of the Starbucks franchisee in Greece and suppose that the “freely negotiated” Tongan choice-of-law clause was broadly phrased to include non-contractual obligations. Suppose further that Starbucks committed a tort against the Greek franchisee and that, unlike Greek or Washington law, Tongan law favors Starbucks. This clause would meet the initial requirements of Article 14, thus shifting the burden on the franchisee to prove that the clause would be unenforceable under paragraphs 2 and 3 of Article 14, or Articles 16 or 26 of Rome II.

Paragraphs 2 and 3 of Article 14 use identical language to that of paragraphs 3 and 4 of Article 3 of Rome I, respectively, except for one difference on the issue of the pertinent time: while the Rome I provisions speak of “all other elements relevant to the situation at the time of the [contractual] choice”[95], the provisions of Rome II speak of “all the elements relevant to the situation at the time when the event giving rise to the damage occurs”[96]. One could make a semi-plausible argument that the “relevant elements” contemplated by the Rome II provisions are the elements of the tort, rather than of the contractual relationship. If accepted, this argument could affect the outcome in a case in which “all” the elements of the tort are located in one country (here, Greece) even if the contractual relationship has contacts with another country. In any event, this argument would be unavailable in this case because the identity of the tortfeasor as a Washington corporation is a “relevant element” in both the contractual and the delictual relationship. The fact that this element is located outside Greece and outside the EU means that the contractual choice of Tongan law will satisfy the tests of paragraphs 2 and 3 of Article 14 of Rome II. Thus, the chosen law must be applied even if it “prejudices” the application 569of Greek mandatory rules (§ 2) or the mandatory rules of Community law (§ 3). The geographical contacts of the franchisor – franchisee relationship prevent the franchisee from invoking these provisions.

On the other hand, geography is not an obstacle to invoking Articles 16 and 26 of Rome II, if this case is litigated in Greece. If the franchisee can prove that the application of Tongan law violates the “overriding” mandatory rules of Greece (Art. 16) or is “manifestly incompatible” with the Greek ordre public (Art. 26), the franchisee will be able to avoid the application of Tongan law. However, because the threshold for applying either of these two articles is considerably high, the franchisee, or other similarly situated small commercial actors, will remain unprotected in all cases that do not meet this threshold. Moreover, an exclusive choice-of-court agreement (which Brussels Ia does not prohibit) that mandates litigation in another country will prevent the franchisee from suing in Greece.

One reason many European commentators may find this result unobjectionable is because they are used to the idea of applying the same law to the tort’s aspects of a case, as the one that governs the underlying contract between the same parties. Rome II preserves this idea. In stating the “manifestly closer connection” exception to the lex loci damni rule, Article 4(3) of Rome II provides that “[a] manifestly closer connection […] might be based in particular on a pre-existing relationship between the parties, such as a contract, that is closely connected with the tort/delict in question”[97]. One commentator has stated that “[e]ven if [the parties’] agreement would be invalid under Article 14(1), an action sounding in tort would still be governed by the law of their choice, as there is likely to be a closer connection between their contractual relationship and the tort at issue”[98]. This statement would be true only if the agreement chooses the law of a state that in fact has the “manifestly closest connection” with the case. However, Article 14 does not impose such a requirement. Indeed, Article 14 does not require any connection to the chosen state. Secondly, there is a difference between, on the one hand, applying the law of a given state because a court determines –after considering all the circumstances and exercising all proper discretion– that that state has a “manifestly closer connection,” and, on the other hand, applying a law solely because of a choice-of-law clause, which (in reality) is not at all negotiated in many cases. Rome II seems to recognize this difference, as well as the risk inherent in allowing pre-dispute choice-of-law clauses for non-contractual claims, by stating in Recital 32 that “[p]rotection should be given to weaker parties by imposing certain conditions on the choice”[99]. However, as the franchise example illustrates, Rome II 570does not always live up to this principle. As with some other freedom-laden ideas, Article 14 may well become the vehicle for taking advantage of weak parties, many of whom are parties to “commercial” relationships.

5. Rome I: Conclusion

The fact that the above comments are critical should not leave the impression that there is nothing to praise about Rome I, or that American law is superior. In fact, in my previous publications, I have praised the EU for having the political courage and legal acumen to devise special rules designed to expressly and legislatively protect some weak contracting parties, such as consumers, employees, passengers, and insureds[100]. As the above discussion has highlighted, these rules work quite well for consumers and employees, but not so well for passengers, insureds, and other weak parties, such as franchisees.

On this issue, American law is anything but superior. At the national level, the political will to protect weak contracting parties is in short supply. For example, the Restatement (Second), which is followed in most US states, does not differentiate between contracts involving weak parties and other contracts and thus does not even purport to protect weak parties. In 2001, a proposed amendment to the Uniform Commercial Code (UCC) would have protected consumers by differentiating between consumer contracts and business to business contracts. However, the amendment encountered stiff opposition by banks and other major businesses and it was withdrawn because, eight years later, only one state legislature (Virgin Islands) had adopted it[101]. The absence of federal choice-of-law legislation leaves the field to the individual states, several of which have enacted statutes designed to protect weak parties domiciled or acting in those states[102]. However, large commercial actors manage to evade those statutes through crafty combinations of choice-of-law clauses and choice-of-court agreements or, especially, arbitration clauses.

Fortunately, weak parties have a last line of defense. It consists of the courts, which generally are more sympathetic than legislatures and tend to strike down the most egregious abuses of bargaining power. Through their decisions, courts restore a certain balance to the system, but only ex post facto and only for those parties who have the means to seek judicial intervention and the ability to wait for the remedy. Can one hope that the United States can master the political will to act proactively and legislatively? Probably not in the near future. As one knowledgeable commentator observed, “US law is generally more pro-business and 571anti-regulatory” and “[f]or a variety of reasons having to do with legal culture and attitudes toward business regulation, … exceptions to party autonomy designed to protect weaker parties in a transaction are not likely to become prevalent in the United States”[103].

To be sure, an ideal legal system should combine pre-emptive legislative protection with remedial judicial protection of weak parties. But if such a combination is not possible and one must choose between the two, then, it may be preferable to have statutory rules pre-emptively protecting weak parties in most cases, even if those rules do not work well for some other parties, rather than not having any such rules. With this criterion, the Rome I regime is better than the American system. But being better does not exempt it from criticism nor does it mean that it cannot become even better.

IV. The Rome II Regulation

There is a lot to say about the Rome II Regulation[104], but again the space limitations of this Journal require me to limit my discussion here to only one aspect of it: How does Rome II resolve conflicts arising from cross-border torts, namely those in which injurious conduct or omission (what Rome II calls the “event giving rise to the damage”) occurs in one country and the resulting injury (what Rome II calls “damage”) in another? The general rule for these conflicts is found in paragraph 1 of Article 4, which mandates the application of the law of the country in which “the damage occurs”[105] (lex loci damni) “irrespective of the country in which the event giving rise to the damage occurred”[106]. This rule applies to all torts for which Rome II does not provide special rules for specific types of torts in Articles 5 through 14, including Article 7 on environmental torts[107].

The lex loci damni rule is subject to several exceptions, including the common habitual residence exception of paragraph 2 of Article 4, the “closer connection” exception of paragraph 3 of Article 4, the admonition to take account of “the rules of safety and conduct” of Article 17, the “overriding mandatory provisions” exception of Article 16, the ordre public exception of Article 26, and even the possibility that the parties may have entered into a pre-dispute or post-dispute choice-of-law 572agreement under Article 14, which was discussed in the previous section. The drafters’ assumption was that these exceptions would enable courts to avoid the lex loci damni rule whenever it produces problematic results. Unfortunately, but predictably[108], this assumption proved unrealistic because in most cases, these exceptions either do not fit the facts or they are too tight or difficult to apply. The cross-border torts discussed here are among these cases.

1. The two Patterns of Cross-Border Torts

For purposes of analysis, cross-border tort conflicts can be divided into two patterns, depending on the content of the relevant laws of the state of conduct and the state of injury:

(1) Pattern 1 consists of cases in which the state of conduct imposes a higher standard of conduct for the tortfeasor, or of financial protection for the victim, than the state of injury. In other words, the law of the state of conduct favors the victim (plaintiff) whereas the law of the state of injury favors the tortfeasor (defendant).

(2) Pattern 2, the converse of Pattern 1, consists of cases in which the state of conduct is more lenient on the tortfeasor by allowing a lower standard of conduct, or providing a lower standard of financial protection for the victim, than the state of injury[109].

The lex loci damni rule of Article 4(1) is content-blind[110]. It mandates the application of the law of the state of injury in both of the above patterns for better or worse –i.e., regardless of whether that law favors the tortfeasor or the victim. As the following discussion illustrates: (a) this indifference to the content of the applicable law is problematic in both Patterns 1 and 2, albeit for different reasons, and (b) in most of these cases, Rome II exceptions to the lex loci damni rule are inapplicable.

5732. Pattern 1: Conduct in a state with a pro-plaintiff law and injury in a state with a pro-defendant law

The recent German case Jabir and others v. KiK Textilien und Non-Food GmbH[111] is a good example of a Pattern 1 case in which none of the exceptions to the lex loci damni rule were available. The locus damni was in Pakistan, where a textile factory that lacked fire escapes and other basic safety features caught fire, killing 260 Pakistani workers, and injuring many others. Four of the victims brought suit in Germany against the factory’s main customer: a German retailer company, KiK Textilien, which was paying for more than 75% of the factory’s production capacity at the time of the fire. The plaintiffs argued that KiK was responsible for the fire, inter alia, because: (1) KiK had sufficient legal and factual control over the Pakistani owner of the factory, Ali Enterprises, to require and ensure that Ali provided safe conditions for its workers; (2) KiK’s Code of Conduct, which was part of its contracts with Ali, imposed this requirement, and (3) KiK’s literature had advertised this fact. If the court accepted plaintiffs’ arguments, this would have qualified as a cross-border tort because the conduct for which the plaintiffs sued KiK occurred in Germany, which was KiK’s corporate base and the place where its officers made –or failed to make– the relevant corporate decisions.

However, before reaching the merits, the plaintiffs had to overcome a preliminary but decisive obstacle, regarding the timelines of their lawsuit. Under the law of Germany, the state of the alleged wrongful conduct, their lawsuit was timely. However, under the law of Pakistan, the state of the resulting injury, their lawsuit was barred thus bringing this case within Pattern 1. Of course, Rome II does not differentiate between such patterns. Article 4(1) calls for the application of the lex loci damni (which would be Pakistan in this case) for better or worse. As one author wrote, the traditional lex loci rule in all its iterations (not only in Rome II) is “superficially ‘neutral,’ striking with even-handed ferocity now at plaintiffs, now at defendants”[112]. In this case, the rule struck at the plaintiffs. To make matters worse, Article 15 of Rome II states categorically that the applicable law includes “the rules of prescription and limitation”[113]. Thus, a straightforward application of Rome II would lead to the application of the Pakistani statute of limitation and, consequently, the dismissal of the plaintiffs’ action. This is exactly what the German court, the Landgericht of Dortmund, held.

The plaintiffs tried in vain to break out of the lex loci damni straitjacket of Article 4(1). However, none of the exceptions to the lex damni rule were available. Specifically:

(1) There was no common habitual residence between the plaintiffs and KiK and thus the exception of paragraph 2 of Article 4 was unavailable.

(2) Paragraph 3 of Article 4 was equally unavailable. The first sentence applies only if “the tort/delict” as a whole (as opposed to some aspects of or “issues” within it)[114] 574is “manifestly more closely connected” with a country other than that of the locus damni. The court found that because Pakistan was both “the place of the tort… and… the habitual residence of the plaintiffs”, the case had “a much closer connection to Pakistani law than to German law”[115]. The second sentence of paragraph 3 was also inapplicable because there was no “pre-existing relationship” between the plaintiffs and KiK, much less one that was connected with a country other than Pakistan.

(3) Article 14 was also inapplicable because, despite the plaintiffs’ contrary allegations, there was no express or implied post-dispute agreement between the plaintiffs and KiK that the dispute would be governed by German law[116].

(4) Even the omnibus ordre public exception of Article 26 was unavailable because Pakistan’s two-year limitation period was only slightly shorter than Germany’s three-year prescriptive period and thus did not “violate the plaintiffs’ (fundamental) right to effective legal protection”[117]. As the court noted, Germany had similarly short prescriptive periods for other claims[118].

(5) Finally, the court did not discuss Article 16 on overriding mandatory rules, or Article 17 regarding “rules of safety and conduct,” but it is clear that neither of these articles would be applicable in this case.

The Jabir case illustrates another problem encountered by tort victims from developing countries in suing the top actor in the tortfeasor’s supply or value chain, i.e., the party colloquially referred to as the “deep pocket”. The top actor is usually a business entity based in a developed country, which acts in developing countries through ostensibly independent local subsidiaries, subcontractors, suppliers, or other intermediaries. Such an arrangement can be motivated by several factors, but one of them is the top actor’s desire to insulate itself from tort liability vis-à-vis third parties, such as the employees of a supplier or other intermediary (as in Jabir), or residents of the local community in which the subsidiary operates (as in many environmental tort and human rights cases)[119].

If the Jabir case involved an environmental tort, it would be governed by Article 7 of Rome II, which allows the plaintiff to “base his or her claims” on the law of the state of conduct[120], i.e., to request and obtain the application Germany’s pro-plaintiff 575prescription law. The drafters of Rome II justified this pro-plaintiff rule by referring to the need for a “high level of protection” against environmental degradation and the “polluter pays” principle, both of which are part and parcel of the Community policy of protecting the environment articulated in Article 174 of the EC Treaty[121]. Article 7 implements this policy in cross-border conflicts by ensuring the application of the law of either the state of conduct or the state of injury, whichever is more protective of the environment, i.e., whichever favors the plaintiff. It assures this result by giving plaintiffs the option of requesting the application of the law of the state of conduct. Obviously, plaintiffs will exercise this option only in Pattern 1 cases, namely cases in which the state of conduct has a pro-plaintiff (pro-environment) law and not when the state of injury has a pro-plaintiff law. In other words, the plaintiff’s option is simply the vehicle for serving the pro-environment policy of the EU.

In 2004, a few weeks after the EU Commission released its proposal of what later became Rome II, I wrote that this option should be given to plaintiffs in all cross-border torts, not only environmental torts[122]. I reiterated the same view in an article written in 2007, a few weeks after the EU Council and Parliament adopted the Commission’s modified proposal[123]. I stated that “Rome II would have been a better system if the drafters had adopted the same logic when drafting the general rule of Article 4(1)” as they did in drafting Article 7 for environmental torts[124], namely giving tort victims a choice between the law of the state of conduct and the state of injury.

Developments since that time have demonstrated the need to extend the logic of Article 7 beyond environmental torts to other categories of cross-border torts. Recognizing this need, the Legal Affairs Committee of the European Parliament (JURI) proposed a similar pro-victim rule for one category of cross-border torts –those involving human rights violations– which would give victims even more choices than Article 7[125]. A subsequent Resolution of the EU Parliament[126] and a proposal 576by the EU Commission[127] have taken the same position. Likewise, two prestigious academic groups, the Group européenne de droit international privé (GEDIP)[128] and the European Law Institute (ELI)[129] have also proposed a similar pro-plaintiff rule for these conflicts.

I enthusiastically endorsed these proposals. The common denominator between cross-border environmental torts and cross-border violations of human rights is that both categories of conflicts involve a widely shared, arguably universal, and in any event EU-wide policy –protecting the environment and protecting human rights. But I also take the further step of arguing for an extension of this pro-plaintiff rule beyond human rights cases to encompass all cross-border torts, including for example cross-border infringement of personality rights[130]. The difficulty with my argument is that, unlike human rights cases in which a pro-plaintiff rule can be justified by the universal policy of protecting human rights, it is hard to argue that there is a similar universal policy of favoring plaintiffs in all cross-border torts.

The counterargument is that there is such a policy, and it is known as the favor laesi principle. Indeed, as I have documented in other publications[131], twenty-seven recent PIL codifications, following this principle, have adopted such a pro-plaintiff rule for all cross-border torts[132]. They authorize the application of the law of either the state of conduct or the state of injury, whichever favors the plaintiff. They do so by either choosing the more favorable of the two laws or allowing the plaintiff to choose between them.

However, my argument for such a rule is based on factors other than the favor laesi principle: the fact that Pattern 1 cases like the Jabir case (as opposed to the converse Pattern 2 cases) present what in the US PIL lexicon are known as false conflicts. To be sure, the term false conflict (and the very notion of public or state interests on 577which it is based), is anathema to most European PIL scholars[133], but it accurately describes what is, and what is not, at stake.

In Pattern 1 cases, only one state –the state of conduct– has something at stake; the other side –the state of injury– does not. The state of conduct has an undeniable interest in applying its pro-plaintiff law to police and deter conduct occurring within its territory and violating its law –even if the resulting injury occurs outside its borders. Indeed, the effectiveness of this law is undermined if it is not applied to out-of-state injuries. That is what is at stake.

On the other hand, the state of injury has no clear countervailing interest in applying its pro-defendant law because that law is designed to protect conduct within that state, not outside of it. That state has little at stake. If anything, it has every reason to welcome the application of a pro-plaintiff law for the benefit of plaintiffs who are injured and are usually domiciled in its territory.

In terms of fairness, there is nothing unfair in subjecting tortfeasors to the law of the state in which they acted. This is the state with which they voluntarily associated themselves and which is often their home state. Having violated the standards of that state, tortfeasors should bear the consequences of that violation and should not be allowed to invoke the lower standards of another state.

In conclusion, for those who subscribe to the notion that PIL cases implicate the interests of the involved states, in addition to the interests of the litigants[134], Pattern 1 cases present an easy choice-of-law question because only one of the involved states (the state of the injurious conduct) has an interest in applying its law –hence the American term false conflict. By contrast, Pattern 2 cases of the converse pattern, to which we now turn, are more difficult because each of the involved states has an interest in applying its law –hence the term true conflict.

3. Pattern 2: Conduct in a state with a pro-defendant law and injury in a state with a pro-plaintiff law

In Pattern 2 cases, the injurious conduct occurs in a state with a pro-defendant law (low or lenient standard) and the resulting injury occurs in a state with a pro-plaintiff law (high or strict standard). It is easy to see why in these cases the state 578of injury has an interest in applying its pro-plaintiff law. In most of these cases, the victim is domiciled in that state but, even if the victim is domiciled elsewhere, that state still has an interest to ensure reparation for injuries occurring there and caused by conduct it considers tortious, even if the conduct occurred in another state that does not consider it tortious. At the same time, the state of conduct also has an interest in applying its pro-defendant law to protect conduct occurring in its territory and considered lawful there. In most of these cases, the defendant is domiciled there but, even if the defendant is domiciled elsewhere, that state still has an interest in protecting the conduct itself.

Because each state has an interest in applying its law, Pattern 2 cases present a difficult choice-of-law dilemma. This is why the American PIL literature characterizes them as true conflicts. Still, a principled, consistent, and uniform solution must be found for these conflicts. A solution based solely on tradition, i.e., on the supposed longevity of the lex loci damni rule, is not functionally defensible. A solution motivated solely by the desire to favor victims (favor laesi) is morally attractive, but not sufficiently principled. An all-around principled solution is one that also considers the tortfeasor’s reasonable expectations and evaluates them by objective standards. This is why the application of the lex loci damni in Pattern 2 cases should not be left to an unchecked plaintiff choice, but should instead depend on objective foreseeability. I have argued for some time that, on balance, the application of the law of the state of injury in these cases is appropriate, but only if the occurrence of the injury in that state was objectively foreseeable[135].

Without hesitation, Rome II resolves the dilemma through the lex loci damni rule for all cross-border torts, including environmental torts, without regard to foreseeability. The fact that this rule is potentially unfair to defendants deserves attention because defendants have a right to expect legal fairness if not moral sympathy. One possible rationale for the lack of a foreseeability exception in Rome II is that the lex loci damni rule, which has been the traditional rule in most countries, did not allow such a defense. Firstly, adherence to tradition is a poor reason for perpetuating a potentially unfair rule. Secondly, even in the old days, the lex loci damni rule was not followed everywhere. For example, many countries that followed the lex loci delicti rule did not localize the locus delicti and thus allowed courts to choose between the locus commissi and the locus damni[136]. Thirdly, as noted earlier, many countries have recently adopted rules that allow either the plaintiff or the court to choose between the locus commissi and the locus damni[137]. Finally, the fact that the lex loci damni rule 579of Article 4(1) is subject to other exceptions does not cure the rule’s problems because, as discussed elsewhere[138] and illustrated by the Jabir case, these exceptions are often unavailable or not easily employable in cross-border torts.

By contrast, an objective foreseeability exception or proviso that is applicable in all cases will ensure that the application of the lex loci damni will not produce unfair results. The lack of such a proviso in Article 7 may be defensible either on the assumption that in environmental torts foreseeability is always present, or on the ground that such a proviso would weaken the EU’s strong pro-environment policy. The fact that neither justification is present in other cross-border torts creates a problem of fairness to defendants. Consequently, a system that aspires to be even-handed should make it possible to avoid the application of the lex loci damni when the occurrence of the injury at the locus damni was not objectively foreseeable. The rule proposed below addresses this issue.

4. Proposal for a new rule

Academic authors are used to criticizing the status quo –it is part of their job. But they also have an obligation to advance specific solutions for the problems they identify. In fulfillment of this obligation, I have proposed replacing Article 7 of Rome II with a new article that will be applicable to all cross-border torts and will be even-handed to both plaintiffs and defendants[139]. Using the less-than-perfect terminology of Article 7, the new article could provide as follows:

[New] Article 7. Cross-border torts

1. The law applicable to a non-contractual obligation arising out of conduct or omission in one country that causes damage to persons or property in another country is the law of the country in which the conduct or omission occurred.

2. However, the law of the country in which the damage occurred applies if:

(a) the occurrence of the damage in that state was objectively foreseeable, and

(b) the claimant expressly and timely requests the application of that law[140].

A recital or recitals could explain that: (1) the new article will apply to all cross-border torts other than those for which Rome II provides special rules, such as those 580of Articles 5 through 13[141]; (2) the timelines and formality of the claimant’s request under paragraph 2(b) will be determined by the law of the forum; (3) the request must encompass all claims and issues against each defendant, so that the claimant cannot engage in a self-serving dépeçage; and (4) Article 17 may not be used to exonerate the defendant[142]. Ideally, this should be included in the text of the article because it is a substantive rule but, if not, this would not be the first time that such a rule is sheltered in a recital.

As compared to present Article 7, the new article reverses the starting point of the choice-of-law process. Now, the starting point will be the lex loci commissi, which becomes the default rule, and the lex loci damni will be the exception, which can be invoked only by the plaintiff.

In Pattern 1 cases, in which the state of conduct has a pro-plaintiff law, the starting point will also be the ending point because plaintiffs will not invoke the lex damni exception. The law of the state of conduct will govern these cases under paragraph 1 of the proposed article.

In Pattern 2 cases (namely, cases in which the state of conduct has a pro-defendant law, and the state of injury has a pro-plaintiff law), the plaintiff will invoke the lex loci damni exception of paragraph 2 of the new Article 7 and, upon proof of objective foreseeability, that law will govern. Under the present text of Rome II [under both Art. 7 and Art. 4(1)], the lex loci damni governs regardless of foreseeability. Although it coincidentally satisfies the favor laesi principle, this result needs also to be defensible in terms of state interests and fairness to the parties. Regarding state interests, Pattern 2 cases present the true conflict paradigm because, as explained earlier, each state has an interest in applying its law. In other words, we have a draw. Consequently, we must rely on other factors to justify the application of the pro-plaintiff law of the state of injury. Objective foreseeability fulfils that role perfectly. It is a factor of sufficient weight to tip the scales in favor of applying the law of the state that experiences the impact of the injurious out-of-state conduct and provides a good retort to any argument of unfair surprise asserted by the defendant.

In conclusion, the application of a pro-plaintiff law is appropriate in Pattern 1 cases and (subject to foreseeability) in Pattern 2 cases. This is a good solution, not 581only –or not so much– because of the favor laesi principle, but also in terms of state interests and fairness to the parties. With this as a given, the next question is whether the choice of law should be made by the plaintiff rather than the court. A rule that directly allows plaintiffs to choose has distinct practical advantages. When the choice is assigned to the court, the court must determine whether, and explain why, one state’s law is more favorable than the other state’s law. Surprisingly, this is not always easy, and an erroneous determination would be a ground for appeal. On the other hand, giving the choice to the plaintiff obviates the need for a judicial answer to the question of whether a given law indeed favors the plaintiff. This is particularly helpful not only when the answer to that question is unclear, but also when a law favors one party on some issues and the other party on other issues. The proposed rule (as clarified in a recital) will avoid the possibility of an inappropriate dépeçage. A plaintiff will have to carefully weigh all the pros and cons of exercising or not exercising the right to choose. If the plaintiff exercises that right, the choice must be for all claims and issues against the defendant. If the choice proves ill-advised, it will not be appealable, and the plaintiff’s attorney should bear all the blame.

If such an article is acceptable in principle, some refinement will be necessary, and it should include answering two important policy questions. The first question is which party should bear the burden of proving foreseeability under paragraph 2(a). The second policy question is whether the new article should be subject to the common habitual residence exception of Article 4(2), the closer connection exception of Article 4(3), or both of those exceptions[143].

5. Rome II: Conclusion

In my first assessment of Rome II, published only a few weeks after its adoption, I characterized it as “a missed opportunity to do much better”[144] and explained that characterization in detail, after discussing all aspects of Rome II. I do not repeat that discussion in this essay, which is limited to only one aspect of Rome II –its treatment of cross-border torts. I do reiterate, however, that my less than enthusiastic appraisal of Rome II was “not for failing to emulate any American models, but rather for failing to take full advantage of the richness, sophistication, and progress of modern European PIL”, which a few years earlier had produced some excellent codifications, including the Belgian, German, and Swiss codifications[145]. One reason for the difference is that drafting and passing legislation for a multi-ethnic and plurilegal union like the EU is far more difficult than legislating at the national level, where it is much easier to attain the necessary consensus[146].

582In any event, although European PIL would have been better off with a “better” Rome II, the more realistic question is whether it would have been better off without Rome II at all. My answer is unhesitantly negative. If nothing else, and despite its flaws, Rome II has unified and thus equalized, the PIL of the EU Member States. This is a major political accomplishment. Although for some of these states this equalization amounted to regress, for many more states it represented a clear and consequential progress. From a transatlantic perspective, this unification or federalization, whatever its flaws, cannot be worse than the complete lack of common direction, much less unity, that has characterized modern American conflicts law for the last fifty years[147]. In a relatively short time, the EU has done much more –perhaps too much– to unify European PIL than American conflicts law could ever hope for.

Moreover, Rome II was not intended to be the last step in the arduous process of modernizing European PIL in tort conflicts. In fact, a review clause in Rome II anticipated the need for future changes and adaptations[148]. As noted earlier, one such change currently under consideration is to expand the scope of Article 7, the most successful provision of Rome II, to cover cross-border violations of human rights. This effort is commendable, especially as compared to the current stance of the US Supreme Court which has abdicated its responsibility to provide a meaningful forum to address such violations, to say nothing of Congress’s inaction[149].

In this essay, I have argued for a further expansion of the scope of Article 7 to encompass all cross-border tort conflicts and for an amendment that, in my view, will make the article more even-handed. I have made this proposal, not because of any illusions that academic writers like myself have a monopoly in expertise, much less wisdom, but because I believe that we have an obligation to propose concrete solutions rather than limit ourselves to criticizing the status quo.

[1] The power to federalize international conflicts derives from the Foreign Commerce Clause of the US Constitution (Art. I § 8), whereas the power to federalize interstate conflicts rests on the Full Faith and Credit Clause (Art. IV § 1). The latter clause obligates each state to give “full faith and credit” to the laws and judgments of sister states and authorizes Congress to enact laws prescribing the manner in which each state must discharge this obligation. However, Congress has exercised this power sparingly, having enacted only four brief statutes, three of which deal with family law matters. Thus, the bulk of American PIL is state rather than federal law, subject to very few federal restraints imposed by the federal Constitution.

[2] See Symeonides, S. C., The ALI’s Complex Litigation Project: Commencing the National Debate, 54 La L. Rev. 843 (1994).

[3] See Symeonides, S. C., The Judicial Acceptance of the Second Conflicts Restatement: A Mixed Blessing, 56 Md. L. Rev. 1246 (1997)· Symeonides, S. C., The Need for a Third Conflicts Restatement (And a Proposal for Tort Conflicts), 75 Ind. L.J. 437 (2000)· Symeonides, S. C., A New Conflicts Restatement: Why Not? 5 J. Priv. Int’l L. 383 (2009).

[4] See Symeonides, S. C., The Conflicts Book of the Louisiana Civil Code: Civilian, American, or Original? 83 Tul. L. Rev. 1041 (2009)· Symeonides, S. C., Codifying Choice of Law for Contracts: The Oregon Experience, 67 RabelsZ. 726 (2003)· Symeonides, S. C., Oregon’s New Choice-of-Law Codification for Tort Conflicts: An Exegesis, 88 Or. L. Rev. 963 (2009).

[5] See Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters (hereinafter “Brussels Ia”).

[6] See Regulation (EC) No 593/2008 of the European Parliament and of the Council of 17 June 2008 on the law applicable to contractual obligations (Rome I) (hereinafter “Rome I”).

[7] See Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II) (hereinafter “Rome II”).

[8] See Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters (Brussels Convention), 1972 O.J. (L 299) 32.

[9] See Council Regulation (EC) No 44/2001 on Jurisdiction and the Recognition and Enforcement of Judgements in Civil and Commercial Matters, O.J. L1 2/1 (2001).

[10] See Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters, O.J. L 351, 20/12/2012.

[11] Except as otherwise indicated, all citations hereinafter are to the 2012 Recast. The Recast is also discussed in Συμεωνίδης, Σ., Κανονισμός Βρυξέλλες Ια και τρίτες χώρες, 4 Lex&Forum (forthcoming 2022).

[12] See, e.g., Inter-American Convention on Extraterritorial Validity of Foreign Judgments and Arbitral Awards of 1979; Inter-American Convention on Jurisdiction in the International Sphere for the Extraterritorial Validity of Foreign Judgments of 1984; Agreement between the Government of Australia and the Government of New Zealand on Trans-Tasman Court Proceedings and Regulatory Enforcement of 2008; Riyadh Arab Agreement for Judicial Cooperation of 1983.

[13] See Juenger, F. K., Judicial Jurisdiction in the United States and in the European Communities: A Comparison, 82 Mich. L. Rev. 1212 (1984).

[14] See Brussels I, Arts. 4 and 6.

[15] See id. Art. 6.2.

[16] Id..

[17] See, e.g., Inter-American Convention on Extraterritorial Validity of Foreign Judgments and Arbitral Awards, Art. 2(d)· Inter-American Convention on Jurisdiction in the International Sphere for the Extraterritorial Validity of Foreign Judgments, Art. 1· Canadian Uniform Enforcement of Foreign Judgments Act, Arts. 8-9. The jurisdictional law of the United States also treats local and foreign defendants alike and provides the latter with the same due process protection as the former. One might criticize one aspect of that law, which allows discrimination against foreign plaintiffs by more easily denying them a forum through the application of the doctrine of forum non conveniens as compared to domestic plaintiffs. Such discrimination, however, is not automatic. It does not affect all foreign plaintiffs and depends on a case-by-case consideration of additional factors other than the plaintiff’s domicile.

[18] See Brussels I, Arts. 36, 39.

[19] See id., Art. 45.3.

[20] Von Mehren, A. T., Recognition and Enforcement of Sister-State Judgments: Reflections on General Theory and Current Practice in the European Economic Community and the United States, 81 Colum. L. Rev. 1060 (1981). See also Takahashi, K., Review of the Brussels I Regulation: A Comment from the Perspectives of Non-Member States (Third States), 8 J. Priv. Int’l L. 2 (2012) (“[I]t does seem unfair to non-EU-domiciled defendants to subject judgments against them which are founded on exorbitant bases of jurisdiction to the Regulation’s liberal regime of recognition and enforcement: it effectively means that the assets of a non-EU-domiciled defendant situated in each Member State are exposed to any exorbitant base of jurisdiction that a claimant can invoke in any EU Member State”)· Nadelmann, K. H., Jurisdictionally Improper Fora in Treaties on Recognition of Judgments: The Common Market Draft, 67 Colum. L. Rev. 995 (1957). But see Kerameus, K. D., Improving the Procedures for the Recognition and Enforcement of Foreign Judgments and Arbitral Awards, in W. Wedekind (ed.), Justice and Efficiency: General Reports and Discussions of the Eighth World Conference on Procedural Law, 226, 257 (1989)· Gaudemet-Tallon, H., Les frontières extérieures de l’espace judiciaire européen: quelques repères, in E Pluribus Unum–Liber Amicorum Georges A.L. Droz (Borrás, A. et al., eds.), 85-87 (1996).

[21] Hartley, T. C., The Brussels Regulation and Non-Community States, in Japanese and European Private International Law in Comparative Perspective (Basedow, J./Baum, H./Nishitani, Y., eds.), 25 (2008).

[22] Id. at 21.

[23] Hartley, T. C., International Commercial Litigation: Text, Cases and Materials on Private International Law2, 172 (2015).

[24] See Droz, G. A. L., Compétence judiciaire et effets des jugements dans le marché commun (1972).

[25] See European Commission, Proposal for a Regulation of the European Parliament and of the Council on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters, Brussels, 14.12.2010, COM(2010) 748 final, 2010/0383 (COD).

[26] See § 2.2.5.3 of Impact Assessment accompanying the Proposal for a Regulation of the European Parliament and of the Council on Jurisdiction and the Recognition and Enforcement of Judgments in Civil and Commercial Matters (Recast), Brussels, 14.12.2010, SEC(2010) 1547 final (“The full harmonisation of jurisdiction rules would achieve the desired objective of ensuring both full and equal access to justice to the European courts. Access to justice would also be fully transparent because all rules on international jurisdiction would be consolidated in one single document, the revised Regulation. This option would do away with the artificial distinction between defendants domiciled within the EU and outside the EU”).

[27] See Burnham v. Superior Court, 495 U.S. 604 (1990).

[28] In Goodyear Dunlop Tires Operations SA et al. v. Brown, 564 U.S. 915 (2011), and Daimler AG v. Bauman, 571 U.S. 117 (2014), the US Supreme Court essentially eliminated this jurisdictional basis in most cases. The Court held that this basis is constitutionally permissible only when the defendant’s activities in the forum state are “so continuous and systematic” as to be analogous to the contacts of a domiciliary defendant.

[29] See supra note 12.

[30] Juenger, F. K., Judicial Jurisdiction in the United States and in the European Communities: A Comparison, 82 Mich. L. Rev. 1211 (1984).

[31] See the Hague Convention on the Choice-of-court Agreements of 30 June 2005.

[32] See supra note 27.

[33] See supra note 28.

[34] See the Hague Convention on the Recognition and Enforcement of Foreign Judgments of 2 July 2019, Art. 5 (hereinafter referred to as the “Hague Convention”).

[35] See the Hague Convention, Art. 5.1(j).

[36] I served as a member of the “experts’ group” that drafted the preliminary text and also participated in the Special Commission meetings that finalized the text and the Diplomatic Conference that adopted it. The views expressed here are solely my own and do not reflect the views of any country or organization. For a brief discussion of the Hague Convention, see Συμεωνίδης, Σ., Ο χαμηλότερος κοινός παρονομαστής: Η Σύμβαση της Χάγης για την Αναγνώριση και Εκτέλεση Αλλοδαπών Αποφάσεων σε Αστικές ή Εμπορικές Υποθέσεις (2019), Lex&Forum 4/2021. 957.

[37] The Hague Convention, Art. 5.1(g).

[38] Unless, of course, the state of origin has indirect jurisdiction under another basis authorized by Article 5.1.

[39] Article 23.4 of the Hague Convention provides that it “shall not affect the application of the rules of a Regional Economic Integration Organisation that is a Party to this Convention [as is the EU] as concerns the recognition or enforcement of a judgment given by a court of a Contracting State that is also a Member State of the Regional Economic Integration Organisation where (a) the rules [like Brussels Ia] were adopted before this Convention was concluded...”.

[40] Brussels I, Art. 24.

[41] This provision is subject to an exception for proceedings involving “tenancies of immovable property concluded for temporary private use for a maximum period of six consecutive months”. Brussels Ia, Art. 24(1).

[42] Article 24 contains three additional rules assigning exclusive jurisdiction to the Member States designated therein for matters involving legal persons or associations, entries in public registries, and the registration or validity of patents, trademarks, designs, or other similar rights required to be deposited or registered. For the sake of simplicity, these rules are not discussed here, although the discussion in the text applies to these rules as well.

[43] Brussels I, Art. 24.

[44] See Mari, L./Pretelli, I., Possibility and Terms for Applying the Brussels I Regulation (Recast) to Extra-EU Disputes, 15 Y.B. Priv. Int’l L. 219 (2013-14) (“[T]he Brussels system does not stipulate that European judges must decline jurisdiction when the connecting factor –the one used to grant exclusivity to European judges– points to a non-European forum”).

[45] Brussels I, Art. 36.1.

[46] Takahashi, K., Review of the Brussels I Regulation: A Comment from the Perspectives of Non-Member States (Third States), 8 J. Priv. Int’l L. 9 (2012).

[47] See Groupe européen de droit international privé, Consolidated version of a proposal to amend Regulation 44/2001 in order to apply it to external situations, Art. 22bis (Bergen 2008, Padua 2009, Copenhagen 2010) available at [https://www.gedip-egpil.eu/documents/gedip-documents-20vce.htm] (“Where no court of a Member State has exclusive jurisdiction under Article 22, a court of a Member State before which proceedings are brought concerning a matter to which that Article applies and which has jurisdiction under another provision of this Regulation shall stay its proceedings if it is established that the courts of a non-Member State have exclusive jurisdiction under the law of that State on the basis of provisions analogous to those in Article 22”). A much softer version of this principle survived in Articles 33 and 34 of the Recast dealing with lis pendens and related actions. According to Recital 24, in the above hypothetical, the Greek court may consider Serbia’s analogous claim of exclusive jurisdiction and may take the actions specified in the GEDIP proposal, but only if an action involving the same immovable was already pending in Serbia when the Greek court was seized· See section II.4, infra.

[48] The Hague Convention, Art. 6 (emphasis added). The Explanatory Report notes that “Article 6 does not expressly exclude judgments on rights in rem over immovable property situated in non-Contracting States” and thus “a judgment given in State A, which is a party to the Convention, that rules on a right in rem over an immovable property situated in State B, which is not a party, does not circulate under the Convention”· Garcimartín Alférez, F. J./Saumier, G., Explanatory Report to the Hague Convention of 2 July 2019 on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters, § 240.

[49] The Hague Convention, Art. 15. The Explanatory Report states that this reservation does not prohibit recognition under national law· see Garcimartín Alférez, F. J./Saumier, G., Explanatory Report, § 241-242.

[50] Brussels I, Art. 36.1. See also the Hague Convention Art. 23.4 (reproduced in supra note 39), which provides that the Convention does not affect the rules of Brussels Ia.

[51] Brussels Ia, Art. 25.

[52] If the agreed-upon court is seized first, it has priority under Article 29, the general lis pendens provision of Brussels Ia. If that court is seized second, it has priority under the exception to the lis pendens provision of Article 31.2. However, if that court is not seized and the substantive validity of the agreement is contested before another court, the latter court will answer that question under the choice-of-law rules of the Member State designated in the choice-of-court agreement. See Brussels I, Art. 25.1, and Recital 20.

[53] Brussels I, Recital 22.

[54] Brussels I, Art. 31.2-3. Paragraphs 2 and 3 of Article 31 do not apply in consumer, employment, or insurance contracts when the consumer, employee, or the presumptively weak party in the insurance contract (the policyholder, insured, beneficiary, or injured party) is the claimant and the agreement is not valid under the relevant provisions of Brussels I. See Brussels Ia, Art. 31.4.

[55] For extensive comparative discussion of the question of which law governs the substantive validity of a choice-of-court agreement, see Symeonides, S. C., Choice-of-Court Agreements: American Practice in a Comparative Perspective, in US Litigation Today: Still a Threat for European Businesses or Just a Paper Tiger? (Bonomi, A./Schefer, K. N., eds.), 85-135 (2018). See also Nielsen, P. A., The New Brussels I Regulation, 50 Comm. Mkt L. Rev. 522-523 (2013).

[56] Brussels I, Art. 25.

[57] For a discussion of this rather peculiar provision, see Hartley, T./Dogauchi, M., Explanatory Report to the Hague Convention of 30 June 2005 on Choice of Court Agreements, §§ 291-310 (2005).

[58] Garcimartín Alférez F. J./Saumier, G., Explanatory Report, § 269.

[59] Brussels Ia, Art. 33.1.

[60] Brussels Ia, Recital 24.

[61] Brussels Ia, Art. 34.1.

[62] Brussels I, Recital 14.

[63] Brussels I, Art 21.1(b), through a cross-reference to Art. 21.2.

[64] Brussels I, Art. 17.1(c).

[65] See Symeonides, S. C., Oxford Commentaries on American Law: Choice of Law, 435-468 (Oxford UP 2016). For discussion of state statutes and court decisions that are friendlier to consumers and employees, see Symeonides, S. C., What Law Governs Forum-Selection, 78 La. L. Rev. 1119-61(2018).

[66] Mullenix. L. S., Gaming the System: Protecting Consumers from Unconscionable Contractual Forum-Selection and Arbitration Clauses, 66 Hastings L. J. 719, 719 (2015).

[67] Id. at 736.

[68] Juenger, F. K., Judicial Jurisdiction in the United States and in the European Communities: A Comparison, 82 Mich. L. Rev. 1212 (1984) [quoting Calder v. Jones, 104 S. Ct. 1482, 1487 (1984)].

[69] Id.

[70] Id.

[71] Id.

[72] Goldschmidt, W., Derecho internacional privado: derecho de la tolerancia; basado en la teoría trialista del mundo jurídico10 (2009).

[73] See EEC Convention on the Law Applicable to Contractual Obligations of the European Communities (1980).

[74] See Symeonides, S. C., Codifying Choice of Law Around the World: An International Comparative Analysis, 109-70 (Oxford UP 2014).

[75] For more extensive comparative discussion of Rome I, see Symeonides, S. C., Party Autonomy in Rome I and II from a Comparative Perspective, in Boele-Woelki, K./Einhorn, T./Girsberger, D./Symeonides, S. (eds.), Convergence and Divergence in Private International Law–Liber Amicorum Kurt Siehr, 513 (2010)· Symeonides, S. C., Party Autonomy and the Lex Limitativa, in Pampoukis, C. (ed.), In Search of Justice: Essays in honour of Spyridon V. Vrellis, 909 (2014)· Symeonides, S. C., Party Autonomy in International Contracts and the Multiple Ways of Slicing the Apple, 39 Brooklyn J. Int’l L. 1123 (2014).

[76] Rome I, Art. 6(1).

[77] Rome I, Art. 8(2-4).

[78] See Rome I, Art. 5(2).

[79] See Rome, Art. 7(2), defining “large risks” through a cross-reference to Article 5(d) of the First Council Directive 73/239/EEC of 24 July 1973.

[80] Another difference between the two categories is that, in the absence of a choice-of-law agreement, the lex causae is the insurer’s habitual residence in large-risk contracts, see Rome I, Art. 7(2), and the law of the member state in which the risk is located in small-risk contracts, see id. Art. 7(3).

[81] See Rome I, Art. 7(3), which also allows two additional choices in other types of insurance contracts. If the laws of the country in which the insured risk is located or the law of the insured’s habitual residence allow more choices or have more liberal limits to party autonomy, the parties may choose a law within those limits. Id..

[82] Jersey Island is a British Crown dependency that has its own legal system. It is not part of the United Kingdom.

[83] Rome I, Art. 6(1) provides that Article 6 applies “[w]ithout prejudice” to Article 5. Recital 32 states that “Article 6 should not apply in the context of those [carriage and insurance] contracts”.

[84] Rome I, Art. 3(3) (emphasis added).

[85] Rome I, Art. 3(4) (emphasis added).

[86] Even if all conditions are present, the application of the “overriding mandatory rules” is itself permissive, not mandatory. See Rome I, Art. 9(3).

[87] Rome I, Art. 6(1).

[88] For citations, see Hay, P./Borchers, P. J./Symeonides, S. C./Whytock, C. A., Conflict of Laws6, 1042-48 (West 2018).

[89] See id..

[90] This provision is subject to the closest connection exception to Art. 4(4).

[91] Rome II, Art. 14(1)(b). Another requirement is that the agreement “shall not prejudice the rights of third parties” Id..

[92] Rome II, Art. 14(2) (emphasis added)

[93] See Rome II, Art. 14(3).

[94] See Rome II, Arts. 16 and 26.

[95] Rome I, Art. 3(3) and (4) (emphasis added).

[96] Rome II, Art. 14(2) and (3) (emphasis added).

[97] Rome II, Art. 4(3).

[98] De Boer, T., Party Autonomy and its Limitations in the Rome II Regulation, 9 Y.B. Priv. Int’l L. 19, 27 (2008).

[99] Rome II, Recital 31.

[100] See supra note 75.

[101] See Scoles, E./Hay, P./Borchers, P./Symeonides, S., Conflict of Laws4, 983–87 (West 2004).

[102] See Hay, P./Borchers, P. J./Symeonides, S. C./Whytock, C. A., Conflict of Laws6, 1028-51, 1130-31 (West 2018).

[103] Borchers, P., Categorical Exceptions to Party Autonomy in Private International Law, 82 Tul. L. Rev. 1645, 1659-60 (2008).

[104] I have extensively discussed Rome II in Symeonides, S. C., Tort Conflicts and Rome II: A View from Across, in H.-P. Mansel, et. al. (eds.), Festschrift für Erik Jayme, 935 (2004)· Symeonides, S. C., Rome II and Tort Conflicts: A Missed Opportunity, 56 Am. J. Comp. L. 173 (2008) [hereinafter Symeonides, “Missed Opportunity”] and Symeonides, S. C., Rome II: A Centrist Critique, 9 Y.B. Priv. Int’l. L. 149 (2008)· Symeonides, The American Revolution and the European Evolution in Choice-of-Law: Reciprocal Lessons, 82 Tul. L. Rev. 1741 (2008).

[105] Rome II, Art. 4(1).

[106] Id..

[107] Except for Article 7, the other articles are not discussed here.

[108] See Symeonides, “Missed Opportunity”, 192-204, 211-15.

[109] Strictly speaking, one should separate standards of conduct for the tortfeasor from standards of financial protection for the victim because one state may have a lower standard of conduct but a higher standard of financial protection. Such differentiation between the two sets of standards would produce at least four different patterns of cross-border torts. See Symeonides, S. C., Oxford Commentaries on American Law: Choice of Law, 218-22, 238-47 (Oxford UP 2016)· Symeonides, S. C./Perdue, W. C., Conflict of Laws: American, Comparative, International4, 354-59 (West 2019). However, for the sake of simplicity and because Rome II eschews issue-by-issue analysis and dépeçage (at least officially) (see Symeonides, “Missed Opportunity”, 184-86), the following discussion assumes that a state that prescribes a higher standard of conduct for the tortfeasor also provides a higher standard of financial protection for the victim. For additional simplicity, the following discussion uses the terms “pro-defendant” and “pro-plaintiff” to describe the laws of the two states. It also assumes that the defendant is the tortfeasor (or those who can be held liable for the tortfeasor’s conduct), and the plaintiff is the tort victim.

[110] See Symeonides, “Missed Opportunity”, 181-83.

[111] Case No 7 O 95/15, Landgericht Dortmund, Date: 10.1.2019.

[112] Weinberg, L., Theory Wars in the Conflict of Laws, 103 Mich. L. Rev. 1631, 1645 (2005).

[113] Rome II, Art. 15(h).

[114] For a critique on this point, see Symeonides, “Missed Opportunity”, 196-203.

[115] Jabir and others v. KiK Textilien und Non-Food GmbH, at § 43.

[116] See id. § 42.

[117] Id. at § 44.

[118] See. id..

[119] For an extensive discussion of these issues, see Symeonides, S. C., Rome II et la responsabilité délictuelle transfrontière: une nécessaire refonte, Rev. critique dr. int’l priv. 529, 534-39 (2022) [hereinafter Symeonides, “La responsabilité délictuelle transfrontière”].

[120] Rome II, Art. 7.

[121] See Rome II, Recital 25 (“the principle that the polluter pays, fully justifies the use of the principle of discriminating in favour of the person sustaining the damage”).

[122] See Symeonides, S. C., Tort Conflicts and Rome II: A View from Across, in H.-P. Mansel, et. al. (eds.), Festschrift für Erik Jayme, 935, at 954 (2004) (“[T]here is a good deal of wisdom in the rules that allow the victim or the court to choose between the laws of the state of conduct and the laws of the state of injury in cases of cross-border torts. It is regrettable that the drafters of Rome II have chosen not to adopt a similar rule [for all cross-border torts] as they did with regard to environmental torts”).

[123] Symeonides, “Missed Opportunity” at 192.

[124] Id. at 210.

[125] See <Proposal for an EU Directive on Corporate Due Diligence and Corporate Accountability> (27 Jan 2021).

[126] See <European Parliament resolution of 10 March 2021 with recommendations to the Commission on corporate due diligence and corporate accountability (2020/2129 INL)>, Art. 20.

[127] See <Proposal for a Directive of the European Parliament and of the Council on Corporate Sustainability Due Diligence and amending Directive (EU) 2019/1937> (23 Feb 2022).

[128] See <Recommendation of the European Group of Private International Law (GEDIP) concerning the Proposal for a directive of 23 February 2022 on Corporate Sustainability Due Diligence, following up on its Recommendation to the Commission of 8 October 2021>.

[129] See <Report of the European Law Institute Business and Human Rights: Access to Justice and Effective Remedies (with input from the EU Agency for Fundamental Rights, FRA) (2022)>.

[130] I have developed this argument in Symeonides, S. C., Rome II and Cross-Border Violations of Human Rights, Infringements of Personality Rights, and Other Cross-Border Torts, 23 Y.B. Priv. Int’l Law. 171 (2021-22).

[131] See Symeonides, S. C., Private International Law: Idealism, Pragmatism, Eclecticism, 203-08 (The Hague Academy of International Law 2021) [hereinafter Symeonides, Idealism].

[132] Moreover, at least 25 other codifications contain express favor laesi rules applicable only to certain cross-border torts, such as products liability or infringement of personality rights.

[133] See, e.g., Kegel’s classic statements that “the state has no interest in the field of private law in applying its own law to the maximum exclusion of foreign law” (Kegel, G., The Crisis of Conflict of Laws, 112 Recueil des cours. 91, 184 (1964)), that “the state does not suffer, it is not hurt, if in some cases… foreign law is applied” (id. at 182-83), and that, in dealing with conflicts cases, the state “is playing the role of the judge, not of litigant. … It does not decide its own affairs but the affairs of others”. Id. at 182. I have respectfully expressed my disagreements to these statements. See, inter alia, Symeonides, S. C., The “Private” in Private International Law (9 Maastricht Law Series 2019).

[134] For an explanation and discussion of this view, see Symeonides, Idealism, 76-101· Symeonides, S. C., The Choice-of-Law Revolution Fifty Years after Currie: An End and a Beginning, 2015 U. Ill. L. Rev. 1847, 1850-67 (2015).

[135] For an early documentation and defense of this thesis, see Symeonides S. C., The American Choice-of-Law Revolution: Past, Present, and Future, 192-200, 228-36 (The Hague Academy of International Law 2006).

[136] See Symeonides, Idealism, 205.

[137] See supra at notes 131-32.

[138] See Symeonides, S. C., Rome II: A Centrist Critique, 9 Y.B. Priv. Int’l. L. 149, 155-66 (2008). Moreover, for environmental torts, the lex loci damni rule of paragraph 1 of Article 4 is not subject to the exceptions of paragraphs 2 and 3 of the same article because Article 7 refers these cases only to paragraph 1.

[139] This proposal is fully developed in Symeonides, “La responsabilité délictuelle transfrontière”, 550-56.

[140] In the interest of full disclosure, this article is modelled on Section 15.440(3) of the Oregon codification of 2009, which is discussed in Symeonides, S. C., Oregon’s New Choice-of-Law Codification for Tort Conflicts: An Exegesis, 88 Or. L. Rev. 963, 1022-1032 (2009). Recently, the proposed Third Conflicts Restatement, which is currently in the drafting stage, has adopted the same rule. See American Law Institute, Restatement of the Law Third, Conflict of Laws, Council Draft No 4, § 6.08 (September 4, 2020).

[141] The same solution should be followed if, in the future, Rome II includes a special rule for cross-border infringement of personality rights. See Symeonides, S. C., Rome II and Cross-Border Violations of Human Rights, Infringements of Personality Rights, and Other Cross-Border Torts, 23 Y.B. Priv. Int’l Law. 171, 202-10 (2021-22). For a more ambitious proposal for these conflicts, see Symeonides, S. C., Infringement of Personality Rights via the Internet: Jurisdiction and Choice-of-Law, 5 Lex&Forum (forthcoming 2022).

[142] See the ELI and GEDIP proposals, supra. With all its problems (see Symeonides, “Missed Opportunity”, 211-15), Article 17 will remain applicable to other torts that are governed by the law of a state other than the one in which the conduct occurred, such as those in which all parties have their habitual residence in one state, but both the conduct and injury occur in another state.

[143] For possible answers to these questions, see Symeonides, “La responsabilité délictuelle transfrontière”, 551-52.

[144] Symeonides, “Missed Opportunity”, 216.

[145] Id. at 174.

[146] I first became aware of these difficulties –compared to my experience in drafting legislation for Louisiana and Oregon– when I had the honor of chairing a few working groups of the EU Council for six hectic months in 2012. During that time, Cyprus held the rotating Presidency of the EU Council. Because I retain my Cypriot citizenship, I was eligible to serve in that position. During the Cyprus Presidency, we finalized the recast of the Brussels I Regulation, which was signed on 15 December 2012, and made significant progress on what later became the Regulations on a European Account Preservation Order (EAPO), matrimonial property regimes, and the property consequences of registered partnerships.

[147] The Third Conflicts Restatement, which is currently been drafted, promises to restore some of the unity and certainty that was lost during the otherwise necessary choice-of-law revolution. See Symeonides, S. C., The Third Conflicts Restatement’s First Draft on Tort Conflicts, 92 Tul. L. Rev. 1 (2017).

[148] See Rome II, Art. 30.

[149] See Symeonides, “La responsabilité délictuelle transfrontière”, 539-40 (discussing recent decisions of the US Supreme Court in restricting the reach of the Alien Tort Statute).