Protection against the breach of choice of court agreements: A comparative analysis of remedies in English and German courts

Abstract

In fixing the place and provider for the resolution of disputes in advance, choice of court agreements increase procedural legal certainty and the predictability of litigation risks. Hence, their protection is crucial. This article undertakes a functional comparison of the remedies for breach of exclusive choice of court agreements in English and German courts, painting a picture of different approaches to a common problem. English courts, now no longer constrained by EU law, employ an entire arsenal of remedies, most strikingly the anti-suit injunction and damages effectively reversing a foreign judgment. In contrast, German courts exercise greater judicial restraint, even though damages for the breach of a choice of court agreement have recently been awarded for the first time. Against this backdrop, two distinct but interrelated reasons for the diverging approaches are identified and analysed, the different conceptions of choice of court agreements and the different roles of comity and mutual trust.

Keywords:

• choice of court agreements
• breach of choice of court agreements
• enforcement
• anti-suit injunction
• damages
• conceptions of choice of court agreements
• comity
• mutual trust
• Hague Convention on Choice of Court Agreements
• German Federal Court of Justice
• comparative law

A. Introduction

Choice of court agreements have been labelled “one of the most important jurisdictional devices of modern times.”¹ Indeed, no international commercial contract “is complete without a clause which identifies … the methods to be used for the determination of disputes.”² Simply put, provisions of that kind “avoid the expense and delay of having to argue about these matters later.”³ In the commercial world and elsewhere, procedural legal certainty and predictability of litigation risks are valuable commodities. Fixing the place and provider for the resolution of disputes in advance (usually) prevents satellite litigation about where to litigate and forum shopping. Against this background, it is apparent that the means for protecting such dispute resolution agreements are crucial. At the same time, rigorous enforcement might not be appropriate in all cases, both as a matter of legal doctrine and where interests of a foreign legal order are implicated.

The problem examined in this article is as simple as it is serious: The breach of an exclusive choice of court agreement. It arises where two parties have provided for the jurisdiction of a designated court in disputes of a certain description (prorogation) to the exclusion of the jurisdiction of all other courts or arbitral tribunals (derogation), but one of them nevertheless initiates proceedings in a non-designated forum. Whether the agreement is exclusive in practice is a question of construction, but for the purpose of this article and its focus on remedies exclusivity is assumed. The breach of a choice of court agreement generates several interests for the defendant in terms of being protected from the immediate and further consequences of the breach. These interests may be vindicated either in the designated court, the forum prorogatum, or the non-designated court, the forum derogatum.

In light of this, an in-depth comparison of the respective remedies for breach of choice of court agreements available before courts in England and Germany is undertaken.⁴ Analysing and explaining the differences in approach as well as the underlying reasoning of courts in different legal systems is the comparative lawyer’s task. Moreover, the United Kingdom’s (UK) departure from the European Union (EU) has resulted in a disharmonisation of the conflict of laws rules, specifically those on jurisdiction and foreign judgements.⁵ This further reinforces the impetus for a comparative analysis. After almost 50 years, the UK has opted to uncouple itself from, inter alia, the so-called Brussels-Lugano regime on jurisdiction and recognition and enforcement in civil and commercial matters,⁶ falling back on the common law and statutory rules as well as international instruments.⁷ On the continent, the German Federal Court of Justice (Bundesgerichtshof) has recently awarded damages for breach of a choice of court agreement for the first time, further evincing the spirit of change and momentum currently prevailing in and surrounding the topic.⁸

At the outset, it must be noted that a comprehensive account of remedies for breach of choice of court agreements in all conceivable constellations and areas of law is beyond the scope of this article. Therefore, the focus lies on exclusive choice of court agreements in civil and commercial matters in cross-border cases, taken to be valid and assumed to cover the dispute in question. Moreover, the analysis proceeds on the premise that no special rules capable of overriding the effects of choice of court agreements apply. These include, for example, limits on subject-matter jurisdiction, rules providing for exclusive jurisdiction, and provisions relating to consumer and employment contracts.⁹

Against this background, the article is structured as follows: It begins by providing an overview of the different remedies for breach of choice of court agreements available in English and German courts (B.). On this basis, it proceeds to examine these remedies in more detail and compares them in light of their functions, discerning both similarities and differences (C.). Further analysing the two remedial approaches, the article finally provides two distinct but interrelated reasons for their diverging functional emphases, the different conceptions of choice of court agreements (D.) and the different roles of comity and mutual trust (E.).

B. Overview of the remedies in English and German courts

When asked to grant relief against the breach of a choice of court agreement, English courts employ four different remedies. Generally, they can decline jurisdiction, grant anti-suit injunctions, award damages, and refuse the recognition and enforcement of foreign judgments. Notably, while these remedies are generally rooted in autonomous English law, some of them are governed by the 2005 Hague Convention on Choice of Court Agreements (Hague Convention), which contains provisions on jurisdiction and the recognition and enforcement of judgments. Moreover, the Convention may have an impact on those remedies available under autonomous English law. The Hague Convention applies in international cases to exclusive choice of court agreements concluded in civil and commercial matters and designating the court(s) of a Contracting State.¹⁰ It has significantly gained in importance after the UK’s exit from the EU and the end of the transition period, with the Brussels-Lugano regime no longer applying in the UK.¹¹ The Convention originally entered into force for the UK on 1 October 2015 when the EU, as the relevant Contracting Party, brought it into effect.¹² This legal basis dissolved on Completion Day¹³ and the UK acceded to the Hague Convention in its own right with effect from 1 January 2021, adopting corresponding legislation to give it the force of law in England.¹⁴

In contrast, German courts have three remedies for breach of a choice of court agreement at their disposal. Generally, they can decline jurisdiction, award damages, and refuse the recognition and enforcement of a foreign judgment. Crucially, German courts do not grant anti-suit injunctions, though they have recently begun to grant anti-anti-suit-injunctions. The rules on jurisdiction and the recognition and enforcement of foreign judgments applicable in German courts are contained in several legal sources. Most importantly, in the field of civil and commercial matters, they consist of autonomous German law and EU law, particularly the Recast Brussels I Regulation. Given the supremacy of EU law,¹⁵ the Regulation significantly affects the availability and operation of those remedies which are ordinarily available under autonomous German law. Furthermore, the Lugano Convention applies in relation to the non-EU Contracting States Iceland, Norway, and Switzerland, and the Hague Convention in relation to the non-EU Contracting States Mexico, Montenegro, Singapore, Ukraine, and the United Kingdom.¹⁶ Save for one important exception, the reformed lis alibi pendens mechanism in cases involving an exclusive choice of court agreement, the Lugano Convention and the Recast Brussels I Regulation are substantially the same with regard to matters considered in this article. In fact, the Lugano Convention is modelled on the Brussels I Regulation.¹⁷ Given that the Lugano and Hague Conventions play a relatively minor role before German courts, the analysis will focus on those remedies which are available under autonomous German law and EU law.

C. Functional comparison of the different remedial approaches

Naturally, the ultimate function of both remedial approaches is to provide relief to the wronged party. However, each remedy can also be seen to serve a more specific function, accounting for one part of the overall solution to the problem. Before plunging into detail, it is sensible to take a step back and identify, at least for the purposes of this analysis, what a “remedy” is and what functions it can serve. There is a considerable amount of academic literature arguing for and against certain meanings of the concept.¹⁸ This article makes no attempt to discern a—let alone “the”—correct definition of remedy as a matter of theory. Instead, it merely seeks to provide clarity about what is meant when the word “remedy” is used. For the purposes of the analysis, a remedy denotes the relief a person can obtain from a court to counter an (threatened) infringement of her rights.¹⁹ In the present context, this is the right not to be sued in a derogated forum. The primary functions of judicial remedies include compelling performance of positive obligations, preventing a wrong, compelling the undoing of a wrong, declaring rights, compensation, restitution, and punishment.²⁰

Out of these six, four distinct functions can be identified with regard to the remedies for breach of a choice of court agreement. First, preventing proceedings before a derogated court from commencing, ie preventing a wrong, or from continuing, ie compelling the undoing of a wrong. Given that these two functions differ only in the point in time when they are engaged, rather than in their ultimate effect, they will be dealt with together as one function. Second, reimbursing for loss caused by the breach, ie compensation. Third, protecting parties from the effects of a decision rendered by a derogated court. While the identification of the first two functions is relatively straightforward, the third function requires a closer look, given that it is somewhat removed from the initial wrong, ie the breach of the choice of court agreement. In fact, it presupposes a breach that has already been committed and cannot be undone anymore. Yet, the function served is still best understood as preventing a (further) wrong. Doctrinally, this can be explained in two ways. The choice of court agreement could be construed so as to include an obligation not to seek recognition and enforcement of any decision obtained in breach.²¹ Alternatively, seeking the recognition and enforcement of such a decision is at any rate exacerbating the wrong already committed, thereby deepening the breach. Against this background, the following comparison analyses whether and to what extent these functions are served by the specific remedies available before English and German courts.

1. Preventing proceedings from commencing or continuing

Where one party has breached a choice of court agreement by initiating proceedings before a derogated court, both English and German courts can prevent these proceedings from continuing where they themselves are the derogated court. In this situation, an English court has a discretion to stay proceedings which should be exercised unless there is “strong cause” to allow the proceedings to continue.²² However, if the prorogated foreign court is that of a Contracting State to the Hague Convention, the English court must suspend or dismiss proceedings pursuant to Article 6 Hague Convention, unless one of the exhaustively listed narrow exceptions to the provision applies. Before German courts, choice of court agreements which prorogate a Member State court are governed by Article 25 Recast Brussels I Regulation, save in purely domestic cases.²³ If one party breaches a choice of court agreement by commencing proceedings before a derogated German court while another Member State court is prorogated, the German court must decline jurisdiction, provided the other party invokes the agreement.²⁴ If the prorogated forum is a non-Member State court, the same follows from the Hague Convention in relation to courts of Contracting States²⁵ and otherwise from autonomous German law.²⁶ Thus, both English and German courts can prevent the continuation of proceedings before them, serving the same function of compelling the undoing of a wrong.

In the converse situation, where one party has breached a choice of court agreement by initiating proceedings before a derogated foreign court, the picture becomes more complex. Starting with English law, a prorogated English court can restrain the party in breach from continuing these proceedings by way of granting an anti-suit injunction. The same applies where a party only threatens to breach the choice of court agreement. In this case, the anti-suit injunction serves to restrain the party from commencing proceedings. Under section 37(1) of the Senior Courts Act 1981, English courts have the power to grant anti-suit injunctions where it is “just and convenient to do so.” In Donohue v Armco Inc, the House of Lords held that where the parties have agreed on an exclusive choice of court agreement, an “English court will ordinarily [grant an anti-suit injunction] to secure compliance with the contractual bargain, unless the party suing in the non-contractual forum can show strong reasons for suing in that forum.”²⁷ Those strong reasons predominantly lie in the risk of parallel proceedings and inconsistent decisions, either because interests of parties other than those bound by the exclusive jurisdiction clause are involved or grounds of claim not subject to the clause are part of the dispute.²⁸

In this context, the question of the Hague Convention’s impact on the remedies available as a matter of autonomous English law arises. First, in obliging the derogated court to suspend or dismiss proceedings, Article 6 Hague Convention might be seen to preclude an anti-suit injunction where the objected-to proceedings are in a Contracting State to the Hague Convention, although this point has not been settled by authority.²⁹ Second, it has been suggested that the anti-suit injunction is not a device suitable to enforce public law rights such as “the right to have the provisions of the Lugano Convention applied properly”³⁰ or, more pertinently here, the Hague Convention. Rather, it could only be employed to enforce private law rights, such as those arising from a contract or an analogous legal relationship.³¹ The argument has been considered and rejected by the High Court in Clearlake Shipping Pte Limited v Xiang Da Marine Pte Limited when it granted an anti-suit injunction on the basis of Article 6 of the Hague Convention.³² It also found there was a contractual right not to be sued corresponding to the public law right afforded by the Convention.³³ Importantly, it considered itself bound by the Court of Appeal decisions in Samengo-Turner v J & H Marsh & McLennan (Services) Ltd³⁴ and Petter v EMC Europe Ltd³⁵ on the first aspect of enforcing a public law right.³⁶ In both cases, an anti-suit injunction was granted to enforce the right of an employee only to be sued at his Member State domicile in England, afforded by Chapter II section 5 of the (Recast) Brussels I Regulation, even though the parties had agreed on the exclusive jurisdiction of the courts in New York and Massachusetts respectively. Consequently, for now, the English courts grant anti-suit injunctions within the jurisdictional regime of the Hague Convention.

With regard to its function, the in personam nature of the anti-suit injunction begs the question to what extent it can actually prevent foreign proceedings from commencing or continuing. After all, the remedy is not directed “against the foreign court but against the part[y]” held to be in breach.³⁷ While English courts admit that anti-suit injunctions at least “indirectly affect the foreign court”,³⁸ their perception by foreign courts is often quite different. For example, the Higher Regional Court of Düsseldorf held that “the anti-suit injunction is capable of directly influencing the work of the German courts and has an effect which is, under certain circumstances, equivalent to that of an order directly addressed to the court.”³⁹ Moreover, anti-suit injunctions are usually unenforceable in foreign jurisdictions as they are considered objectionable as matter of public international law and thus contrary to public policy.⁴⁰ Nevertheless, the anti-suit injunction is a singularly effective remedy to prevent foreign proceedings from commencing or continuing, given that failure to comply with it amounts to contempt of court, punishable by the payment of a fine, the sequestering of assets, or even imprisonment.⁴¹ Thus, any party within the territorial jurisdiction of the English courts or with assets in England faces considerable risks when commencing or continuing foreign proceedings in the face of an anti-suit injunction.

Moving on from English courts, their German counterparts do not grant orders restraining a party from commencing or continuing proceedings before a derogated foreign court. In Turner v Grovit, the European Court of Justice (ECJ) ruled that the 1968 Brussels Convention on jurisdiction and the enforcement of judgments in civil and commercial matters (Brussels Convention) precludes injunctions restraining a party from commencing or continuing proceedings before a foreign Member State court.⁴² But even in cases outside the Brussels regime, German courts do not currently grant restraining orders of this kind. In a singular decision in 1938, long before any major jurisdictional harmonisation efforts in Europe, the Supreme Court of the German Reich (Reichsgericht) granted an injunction restraining a claimant primarily domiciled in Germany from continuing proceedings in Latvia.⁴³ The claimant had established a secondary domicile in Latvia—then known as a “divorce paradise”⁴⁴—for the purposes of bringing divorce proceedings against his wife there. At the time, Latvian courts applied the law of the claimant’s (secondary) domicile and Latvian divorce law was much more liberal than its German counterpart.⁴⁵ The Reichsgericht held that the initiation of proceedings amounted to an intentional infliction of damages on the respondent wife in a manner contrary to public policy, giving rise to a delictual claim for injunctive relief (section 826 of the German Civil Code).⁴⁶ Crucially, the case did not involve a choice of court agreement. Though its legal basis still exists under German law, the decision has since remained the only instance in which German courts have enforced it by means of an anti-suit injunction. Apart from this historical exception, orders restraining a party from commencing or continuing proceedings before a foreign court are simply alien to the German legal order.⁴⁷ In line with this, the Bundesgerichtshof has recently held—in an obiter dictum—that “a (legally enforceable) primary obligation” not to sue before a derogated court cannot be validly agreed.⁴⁸ Given that it is precisely such a primary obligation which would be enforced by way of an anti-suit injunction, this statement precludes the remedy as a matter of legal doctrine.

However, German courts have recently been confronted with a number of anti-suit injunctions granted by foreign courts to restrain the claimants from continuing proceedings in Germany. Thus compelled to engage with the remedy, they awarded anti-anti-suit injunctions in response. In a case regarding the infringement of standard-essential patents, the Higher Regional Court of Munich confirmed such an order on the basis of a proprietary claim for injunctive relief under s 823(1) in connection with s 1004(1) of the German Civil Code as “the only effective means of defence against an anti-suit injunction, which can safeguard the applicant’s ability to assert their legal position as patent holders in the patent infringement proceedings pending in Germany.”⁴⁹ Following suit, the Higher Regional Court of Hamm granted an anti-anti-suit injunction on the basis of a delictual claim for injunctive relief stemming from the claimant’s “right of access to justice under s 823(1) or in any case s 823(2) of the German Civil Code.”⁵⁰ It held that “the defendant’s pursuit of proceedings in the US for the purpose of preventing the continuation of proceedings pending before the District Court of Essen constituted an impermissible infringement of the right of access to justice and, concurrently, the legal sovereignty of Germany.”⁵¹ However, it must be emphasised that neither of these anti-anti-suit injunctions were granted on the basis of a choice of court agreement. Anti-anti-suit injunctions are granted to restrain a party from applying for an anti-suit injunction before a foreign court or, where an application has already been made, to order the party to withdraw it. As such, they may serve as a remedy for the breach of a choice of court agreement only if the (threatened) attempt to obtain an anti-suit injunction from a derogated court constitutes a breach. This is primarily a matter of construction of the agreement in question. It remains to be seen whether the German courts would grant an anti-anti-suit-injunction in case they construed the choice of court agreement in that way. At any rate, the cases make clear that the unavailability of anti-suit injunctions under German law does not preclude the availability of anti-anti-suit injunctions, at least in “exceptional case[s]”.⁵²

For now, when asked to restrain a party from commencing or continuing proceedings before a derogated foreign court, a prorogated German court remains idle, expecting that the derogated court will consider the choice of court agreement and consequently decline jurisdiction. However, it would be myopic to simply conclude that German courts fail to fulfil the function of preventing foreign proceedings from commencing or continuing. First, with regard to cases covered by the Recast Brussels I Regulation, it is crucial to appreciate the minimal need for judicial intervention in a compulsory system of jurisdiction enabled by the principle of mutual trust and applied uniformly across the Member States.⁵³ This greatly reduces the formation of jurisdictional conflicts which generate the need for orders restraining parties from commencing or continuing proceedings before a derogated court in the first place. Second, a derogated Member State court must ex officio stay proceedings before it as soon as the prorogated court has been seised and later decline jurisdiction when the prorogated German court has established jurisdiction in accordance with the choice of court agreement (Article 31(2)(3)). At least in theory, this eliminates the remaining risk of inconsistent findings in a system of common jurisdictional rules by providing for the Kompetenz-Kompetenz of the prima facie prorogated court to decide upon the validity and scope of the choice of court agreement, thus removing the need for a restraining order. In the less common situation in which the derogated court is subsequently seised, the same follows from the general lis alibi pendens rule (Article 29(1)(3)). Yet, it must be stressed that these theoretical considerations provide neither help nor solace should the derogated court—for whatever reason—ignore its obligations under the Regulation. Moreover, no comparable mechanism applies in cases outside the Recast Brussels I Regulation.

As a result, the remedial approaches adopted by English and German courts are only partly functionally equivalent. Where the Recast Brussels I Regulation applies, the function of preventing foreign proceedings from commencing or continuing is served in German courts not by granting injunctive relief but through a compulsory regime of common jurisdictional rules based on mutual trust. Specifically, the regime disincentivises the initiation of proceedings before a derogated court in the first place and instructs a seised but derogated court to decline jurisdiction. To that extent, there is a functional—if only theoretical—equivalence with the anti-suit injunction. However, where the Recast Brussels I Regulation is inapplicable, the German courts do not fulfil this function and to that extent, no functional equivalence exists. At the same time, it is important to bear in mind that each of the three functions identified above address a different aspect of the overall problem. This first function of preventing foreign proceedings from commencing or continuing operates at the earliest remedial stage of the breach of a choice of court agreement. The fact that it is not fulfilled to the same degree in the legal systems under comparison says little about whether and how the problem might be dealt with at a later stage.

2. Compensating for loss caused by the breach

Moving on to the function of compensation, both prorogated English and German courts now generally award damages for loss caused by the breach of a choice of court agreement on the basis of English and German law respectively.⁵⁴ However, the extent to which compensation is available differs significantly. Before English courts, the principle that damages may be awarded for breach of an obligation not to sue in a certain forum can be traced back to Ellerman Lines Ltd v Read.⁵⁵ In this case, which concerned the salvage of a steamship, the defendant sued and obtained a favourable judgment in the court of Constantinople in violation of an agreement to pursue any claim by arbitration in London. The judgment had already been partially satisfied by sale of the claimant’s arrested vessel when the Court of Appeal not only ordered the defendant to refrain from (further) enforcing the Turkish judgment but also awarded damages, inter alia, for the value of the ship.

In essence, two different heads of damages are now accepted. First, in Union Discount v Zoller, the Court of Appeal awarded damages for loss incurred in defending an action before a derogated foreign court where that court had declined jurisdiction and made no order as to costs.⁵⁶ Second, in Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG, it held that damages for breach of a choice of court agreement could also include loss in the form of an unfavourable foreign judgment, effectively reversing it on the merits.⁵⁷ This also holds true where the Hague Convention applies.⁵⁸ As will be shown below, there is no obligation to recognise and enforce a judgment by a court of a Contracting State which was obtained in breach of a choice of court agreement.⁵⁹

Turning to German law, until recently, German courts have not awarded damages for loss caused by the breach of a choice of court agreement at all. In a landmark decision in 2019, the Bundesgerichtshof, for the first time, awarded such damages where an action had been commenced in the United States (US) despite a choice of court agreement providing that “Bonn shall be the place of jurisdiction.”⁶⁰ The US District Court had dismissed the action for lack of jurisdiction but made no order as to costs. The Bundesgerichtshof held that, at least where the foreign court has declined jurisdiction on the basis of the agreement, costs incurred in defending the action in the derogated forum can be recovered.⁶¹ Though this qualification appears difficult to reconcile with awarding damages for loss in the form of an unfavourable foreign judgment, which naturally requires the exercise of jurisdiction, the court has not explicitly ruled it out.

Moreover, even though Article 25 Recast Brussels I Regulation applied to the choice of court agreement prorogating the (Member State) courts of Bonn,⁶² the principle of mutual trust was not engaged since the seised but derogated court was in the US, a non-Member State. It is thus unclear whether the Bundesgerichtshof would have reached a different decision had the case involved proceedings in another Member State. Crucially, this depends on whether the ECJ’s reasoning regarding anti-suit injunctions in Turner can be transposed to the award of damages for breach of a choice of court agreement. Interestingly, the (English) Court of Appeal decided against referring exactly that question to the ECJ in Starlight Shipping but was ultimately only able to delay its resolution by the European court.⁶³ In the same dispute, the Supreme Court of Greece (Areios Pagos) has recently requested a preliminary ruling on the issue, albeit in a slightly different guise. Petitioned to recognise and enforce one of the judgments resulting from the Starlight Shipping litigation,⁶⁴ it has referred to the ECJ the question of, inter alia, whether it is contrary to public policy within the meaning of Articles 34(no1) and 45(1) of the Brussels I Regulation to recognise and enforce a judgment awarding provisional damages for breach of a choice of court agreement.⁶⁵ At the time of writing, the decision is still pending, though Advocate General Richard de la Tour has argued for the question to be answered in the affirmative.⁶⁶

The Bundesgerichtshof, specifically addressing Turner, noted that the main consideration underpinning the decision, the principle of mutual trust, was not applicable in relation to non-Member States.⁶⁷ At first glance, this would seem to suggest that the applicability of the principle in relation to Member States might prove an obstacle to an award of damages. However, the court further held that, at least where the derogated court has declined jurisdiction, awarding damages could not be seen as reviewing the foreign court’s jurisdiction.⁶⁸ Given that this is one of the defining characteristics of the principle of mutual trust in the context of jurisdiction, the court might be prepared to award damages in this situation even where the Recast Brussels I Regulation applies. At the same time, the reasoning suggests that damages for loss in the form of an unfavourable foreign judgment on the merits would likely be seen as a prohibited review of the foreign court’s jurisdiction. In this case, the foreign court will necessarily not have declined jurisdiction.

In conclusion, while both English and German courts generally compensate for loss caused by the breach of a choice of court agreement, the circumstances in which compensation is available and the extent of it differ vastly. English courts are prepared to grant compensation by awarding damages for any such loss, including loss in the form of an unfavourable judgment from the foreign court. In contrast, German courts are more hesitant and predicate any award on the foreign court declining jurisdiction, thereby restricting it to loss incurred in defending the action in the derogated forum. To this extent, the compensatory function of awarding damages is less fulfilled. Yet, it bears repeating that each specific function only represents one part of the remedial approach, leaving the operation of the other functions and their potential to deal with the breach of a choice of court agreement at a later stage untouched.

Before moving on to the third function, awarding damages for breach of a choice of court agreement subtly fulfils an additional function, ie preventing foreign proceedings from commencing or continuing. Generally, the prospect of having to pay damages disincentivises the breach of obligations. In the present context, the availability of damages as a remedy thus indirectly encourages compliance with the choice of court agreement.⁶⁹ Moreover, the intensity of the disincentivising effect corresponds with the extent of the damages. Hence, the prospect of having to compensate any loss caused by the breach of a choice of court agreement has a stronger disincentivising effect than the prospect of having to pay damages only for loss incurred in defending an action in the derogated forum. Again, to this extent, the function of preventing foreign proceedings from commencing or continuing is less fulfilled by the remedies available before German courts than by those before English courts.

Moreover, the specific relief granted in Starlight Shipping produces a further disincentivising effect. Notably, the court gave judgment for “damages to be assessed” before the foreign proceedings had concluded, launching to some extent a “pre-emptive strike”.⁷⁰ Thus, already then, the defendant knew for certain that it would not only have to pay damages for any loss incurred by the claimant so far but also for any potential further loss, including an unfavourable foreign judgment. In addition, regardless of the outcome in the foreign court, the sum of damages would only increase if proceedings were not discontinued. By this means, English courts encourage swift conclusions to proceedings before derogated courts in a manner with no functional equivalent in the German system.

3. Protecting from the effects of a decision rendered by a derogated court

Lastly, the function of protecting from the effects of a decision rendered by a derogated court comes into operation when, in addition to breaching the agreement, a party has obtained a final and conclusive judgment on the merits from the derogated foreign court. In this case, an English court can refuse recognition and enforcement of the foreign judgment on the basis of the breach of the choice of court agreement under section 32(1) of the Civil Jurisdiction and Judgments Act 1982. Moreover, even where the foreign court has specifically ruled that there was no breach and that it had jurisdiction, the English court can refuse recognition since it is not bound by any such decision under section 32(3) of the Civil Jurisdiction and Judgments Act 1982. Notably, the Hague Convention applies only to the recognition and enforcement of judgements given by a court of a Contracting State designated in an exclusive choice of court agreement. It does not provide for the refusal of recognition and enforcement of judgements rendered by a non-designated court.⁷¹ Moreover, there is nothing in the Convention which binds an English court to another Contracting State court’s ruling as to its own designation.⁷² Instead, this is simply a case for section 32 of the Civil Jurisdiction and Judgements Act 1982 as well. ⁷³

Considering German courts next, their ability to refuse recognition and enforcement of a foreign judgment crucially depends on the operation of the Recast Brussels I Regulation. If the judgment was rendered by another Member State court, there is no direct remedy. Any such judgment is recognised and enforceable ex lege in all other Member States (Articles 36, 45), even when the proceedings leading to the judgment were in contravention of a choice of court agreement (Article 45(3)). This appears particularly striking considering that a violation of the heads of exclusive jurisdiction provided for in Article 24 does constitute a ground for refusing recognition and enforcement (Article 45 (1)(e)(ii)).

In contrast, under autonomous German law, the recognition and enforcement of foreign judgments is based on the “mirroring principle” in combination with reciprocity.⁷⁴ According to this principle, a foreign judgement is recognised and enforced if the foreign court has exercised a jurisdiction which mirrors the jurisdiction a German court would have exercised. Simply put, the foreign court must have had hypothetical jurisdiction under German law. Thus, a German court will refuse recognition and enforcement of a foreign judgment obtained in breach of a choice of court agreement on the basis that the foreign court did not have jurisdiction under German law (section 328(1)(no1) of the German Code of Civil Procedure). This holds true even where the foreign court specifically ruled that there was no breach of the choice of court agreement. Decisions of foreign courts as to their jurisdiction have no binding effect on German courts.⁷⁵ Crucially, this is because the subject matter of the foreign court’s decision and that of a German court are not considered to be identical.⁷⁶ While the foreign court has assumed jurisdiction according to its lex fori, the German court enquires whether the foreign court had hypothetical jurisdiction under German law. Moreover, the German courts are not bound by any finding of fact made by the foreign court as to its jurisdiction. Otherwise, the foreign judgment would be afforded preclusive effect in Germany without fulfilling the conditions for recognition and enforcement.⁷⁷ As a consequence, the German court will always examine afresh the question of whether there was a choice of court agreement and, in case of an affirmative answer, refuse recognition and enforcement of the foreign judgement as a whole.

Thus, both English and German courts can generally protect parties from the effects of a decision rendered by a derogated court by refusing recognition and enforcement. However, where the decision is rendered by a court of another EU Member State, a German court is barred from taking any action, seemingly failing to serve a protective function. Again, such a conclusion would be myopic insofar as it ignores that within the scope of the Recast Brussels I Regulation, the protection against decisions obtained in breach of a choice of court agreement is established at a much earlier stage than that of recognition and enforcement. Protection is provided by the principle of mutual trust and consists of prevention rather than of countermeasures ex post facto. First, the Recast Brussels I Regulation establishes a compulsory regime of common jurisdictional rules, applied uniformly across the Member States.⁷⁸ At least in theory, this renders the need for protection against the effects of a decision obtained in breach of a choice of court agreement minimal. Specifically, it significantly reduces the risk of a derogated court finding itself competent and rendering a judgment on the merits in the first place. Second, this risk is eliminated where the lis alibi pendens mechanisms of Articles 29(1)(3) and 31(2)(3) have been engaged by seising the prorogated court.⁷⁹ Yet again, it must be emphasised that where a derogated court—for whatever reason—ignores its obligations under the Regulation, no protection is offered to the wronged party.

4. Interim conclusion

What has emerged is a multifaceted picture of the remedial approaches and their distinct functions. More specifically, the English courts will deal with a breach by a combination of remedies fulfilling all of the four functions identified to the fullest extent. In contrast, the remedial approach of the German courts is rather less extensive, relying instead on judicial restraint and the protection of parties against decisions from a derogated court as a last resort. Taking a step back, both remedial approaches, each with its emphasis on different functions, illustrate how the ultimate function of providing relief for the breach of a choice of court agreement can be served. Moreover, far from marking the end of the comparison, the functional similarities and differences identified provide the basis, and indeed the reason, for further analysis.

D. The different conceptions of choice of court agreements

In Airbus Industrie GIE v Patel, Lord Goff noted that with regard to the law “concerned with the resolution of clashes between jurisdictions … [t]wo different approaches to the problem have emerged in the world today, one associated with the civil law jurisdictions of continental Europe, and the other with the common law world.”⁸⁰ Taking this statement as a point of departure, the following segment explores to what extent the remedial approaches of the English and German courts, with their different functional emphases, are attributable to their respective conceptions of the law of international jurisdiction, and in particular of choice of court agreements. The fundamental question underlying these conceptions is whether one regards choice of court agreements as procedural instructions for the courts to exercise or refrain from exercising jurisdiction or rather as instructions for the parties in the sense of substantive obligations which they assume to each other in relation to the jurisdiction of a court. The answer might well be a little more nuanced than the question suggests as these conceptions are not necessarily mutually exclusive. Nonetheless, the principal orientation towards one of these two paradigms largely determines the availability of remedies for breach of a choice of court agreement.

1. The conception in English courts

English courts have adopted a primarily contractual conception of choice of court agreements, attaching great weight to the principle of party autonomy. The position is perhaps best illustrated by Professor Adrian Briggs: “The truth, so far as English law is concerned, is that a very large amount of the law on jurisdiction … is dependent on the very private law notions of consent and obligation.”⁸¹ That is not to say an English court will not recognise the prorogating or derogating effect of a choice of court agreement but it will not do so invariably.⁸² Rather, the agreement will only influence the exercise of rather than determine the inherent jurisdiction of an English court.⁸³ It is only in this respect that the English courts’ approach to choice of court agreements is not fully in line with party autonomy. Moreover, an exception applies in cases covered by the Hague Convention pursuant to which English courts must recognise the prorogating and derogating effect of choice of court agreements under Articles 5 and 6. Apart from this limited effect on the jurisdiction of the English courts, choice of court agreements are primarily conceived as a source of private obligations relating to the initiation of proceedings in case of a dispute. More specifically, it is the obligation not to sue the other party in a non-contractual forum which an English court enforces by way of granting an anti-suit injunction, awarding damages, or refusing the recognition and enforcement of a foreign judgment.

In fact, the prevalence of the contractual conception of choice of court agreements can be observed in relation to each of these remedies. First, an “English court will ordinarily [grant an anti-suit injunction] to secure compliance with the contractual bargain”.⁸⁴ It will not do so in every case but the exception to the rule, which applies when “the party suing in the non-contractual forum can show strong reasons for suing in that forum”,⁸⁵ is equally in line with this contractual conception. The doctrine of privity posits that only the parties to a contract assume obligations under it.⁸⁶ A judge, who is not privy to the contract, is therefore not bound to give effect to it under any circumstances but is instead entitled to conclude that the agreement should not be given specific force.⁸⁷

Second, once it is accepted that a choice of court agreement contains an ordinary contractual bargain, it follows that damages are a suitable remedy for its breach. Indeed, it is implicit in the English decisions dealing with damages for breach of a choice of court agreement that the cause of action is a straightforward one for breach of contract.⁸⁸ Explicitly, these cases were concerned with issues including, for example, the interpretation of the choice of court agreements and the principle of mutual trust in Starlight Shipping⁸⁹ or the effect of the potential recognition and enforcement of the foreign judgment constituting the loss suffered in CMA CGM SA v Hyundai Mipo Dockyard Co Ltd.⁹⁰ Crucially, however, none of them questioned the basic premise that a choice of court agreement can serve as a source of substantive obligations which, if breached, give rise to a claim for damages.

Finally, the “priority given in English private international law to the parties’ bargain” is also reflected in the framework for refusing the recognition and enforcement of foreign judgments.⁹¹ Under section 32(1)(a) of the Civil Jurisdiction and Judgments Act 1982, a foreign judgment shall not be recognised or enforced if “the bringing of proceedings was contrary to an agreement under which the dispute in question was to be settled otherwise than by proceedings in the courts of that country”. However, if the foreign court was otherwise internationally competent to adjudicate, any final and conclusive decision would normally have to be recognised as res judicata, including any ruling that there was no breach of the choice of court agreement.⁹² In this case, section 32(3) of the Civil Jurisdiction and Judgments Act 1982 provides that the English court is not bound by any such foreign decision on the choice of court agreement. Exemplifying the contractual analysis of choice of court agreements, the English court’s refusal to recognise and enforce the foreign judgment is conceptualised as the enforcement of the contractual obligation not to sue before the foreign court, rather than as the review of another court’s jurisdiction according to English law.

In summary, the premise of the English courts’ remedial approach is the following: “Contracts are made to be performed, and the fact that they may be contracts about the jurisdiction of courts is nothing to the point.”⁹³ Notably, the rules of the conflict of laws are not immediately concerned with obligations that parties to a contract assume to each other. Rather, it is concerned with them only insofar as they influence the jurisdiction of a foreign court, the recognition and enforcement of a foreign judgment, and less pertinently here, the application of the law of a foreign state. By putting emphasis on the substantive obligations of the parties to the choice of court agreement inter se, the contractual analysis shifts the focus away from the interests of the foreign legal order and thus bolsters the principle of party autonomy in the English courts’ remedial approach to the breach of choice of court agreements.

2. The conception in German courts

In contrast, the German courts have adopted a primarily procedural and court-oriented conception of choice of court agreements, with the principle of party autonomy playing a more limited role. Both in cases governed by EU law and autonomous German law, choice of court agreements are generally conceptualised as procedural instructions for the courts and jurisdiction is exercised or declined in line with the parties’ agreement. In fact, until recently choice of court agreements were exclusively ascribed the effect of prorogation and derogation, vesting the competence to adjudicate in the designated court(s) and thereby displacing the jurisdictional rules otherwise applicable.⁹⁴ In this sense, the principle of party autonomy is promoted in the German courts’ conception of choice of court agreements, particularly since the rules on exercising or declining jurisdiction are non-discretionary in nature.

However, a procedural conception of choice of court agreements is at odds with the principle of party autonomy in other respects. The Bundesgerichtshof describes the nature of a choice of court agreement as “a substantive contract about procedural relations”⁹⁵ while others prefer the characterisation as a “procedural contract”.⁹⁶ Either way, both views have been marked by a “German doctrinal reluctance to understand a [choice of court agreement] as a source of private rights.”⁹⁷ As a consequence, the anti-suit injunction and damages are not obviously suitable remedies for its breach.

Remarkably, in its decision awarding damages for breach of a choice of court agreement for the first time, the Bundesgerichtshof has cautiously abandoned this reluctance, thereby challenging the procedural conception of choice of court agreements hitherto prevalent in Germany. After reiterating its established jurisprudence on the nature of a choice of court agreement as “a substantive contract about procedural relations”, it held that the subject matter of such a contract need not be limited to procedural relations.⁹⁸ Rather, the court ruled that parties, as a matter of their freedom of contract, may agree on further substantive obligations which are separate from the procedural effects of a choice of court agreement but can be established simultaneously with them.⁹⁹ This conceptual change paved the way for the court to award damages for loss incurred in defending foreign proceedings commenced in breach of a choice of court agreement where the derogated court had declined jurisdiction.

Notably, this appears to exclude not only damages for loss in the form of an unfavourable judgment on the merits by the foreign court but also to eliminate the possibility of an anti-suit injunction, at least where the foreign court has already decided to hear the case. Specifically with regard to the last point, the Bundesgerichtshof held that its finding of a secondary obligation to pay damages is not precluded by the fact that “a (legally enforceable) primary obligation” not to sue before a derogated court cannot be validly agreed.¹⁰⁰ Yet, it is precisely such a primary obligation which would be enforced by way of an anti-suit injunction. Thus, while marking a significant conceptual change, the decision’s practical implications are less extensive than one might suspect at first glance. The remedial approach of German courts still reflects a primarily procedural conception of choice of court agreements, emphasising their effects on the exercise of jurisdiction by courts while acknowledging their (limited) potential to generate substantive obligations of the parties inter se. In this way, a procedural conception of choice of court agreements is less conducive to the principle of party autonomy.

Moreover, the German courts’ approach to protecting parties from the effects of a decision from a foreign derogated court is equally underpinned by a procedural conception of choice of court agreements. It bears repeating that, where the Recast Brussels I Regulation applies, this protection is preventive in nature and realised through a compulsory regime of common jurisdictional rules. These non-discretionary rules translate the contents of choice of court agreements into procedural instructions for the courts, thus reflecting their procedural conception. In this sense, they can also be seen to promote the principle of party autonomy. In particular, Article 25 provides for the prorogating and derogating effect of choice of court agreements and the lis alibi pendens mechanisms of Articles 29(1)(3) and 31(2)(3) direct any non-designated Member State court to stay proceedings or decline jurisdiction in favour of the seised designated court. Yet, this does not take away from the fact that the breach of a choice of court agreement provides no defence to the recognition and enforcement of a Member State court's judgment (Article 45(3)). To that extent, the principle of party autonomy is trumped by the principle of mutual trust.¹⁰¹

Outside the scope of the Regulation, however, German courts do refuse the recognition and enforcement of foreign judgments. Crucially, this is in line with a procedural conception of choice of court agreements as well. Under the “mirroring principle”,¹⁰² refusal is based on the derogated foreign court’s lack of jurisdiction according to German law, which conceptualises choice of court agreements procedurally. As German law upholds party autonomy with regard to the exercise of jurisdiction, and the “mirroring principle” applies German law to the foreign court’s exercise of jurisdiction for the purposes of recognition and enforcement, a refusal based on the latter’s lack of jurisdiction is equally upholding party autonomy. At the same time, the notion that the choice of court agreement may also contain substantive obligations of the parties which they assume to each other with regard to jurisdiction of the court plays no role in this regard.

E. The different roles of comity and mutual trust

The second explanation for the different functional emphases of the remedial approaches of English and German courts is their diverging appreciation of what can broadly be described as comity. Generally understood, comity denotes the conduct and attitude of courts towards the interests of a foreign legal system, jurisdictionally connected to the same dispute, in administering justice within its own territory.¹⁰³ Put more floridly, it comprises “good neighbourliness, common courtesy and mutual respect between those who labour in adjoining judicial vineyards.”¹⁰⁴ With regard to jurisdiction and adjudication, comity requires a court to respect, and not interfere with, the integrity of foreign judicial proceedings and the resulting judicial orders, at least in so far as they apply to persons or property within the territory of the foreign state.¹⁰⁵

Comity is a concept primarily used in England and the wider common law world. In contrast, German law, like many other civilian legal systems, lacks a similarly established concept in its autonomous law. In the context of EU law, the Recast Brussels I Regulation is “based on the principle of mutual trust in the legal system and judicial institutions of each other Member State”.¹⁰⁶ It is important to note that the principle applies in German courts but no longer in English courts.¹⁰⁷ Viewed in the abstract, the principle of mutual trust governs the conduct and attitude towards the interests of other Member States, making it functionally equivalent to the principle of comity. In other words, it represents an EU-specific reflection of it. At the same time, the principle of mutual trust is considerably more rigorous in its demands for judicial restraint than comity. Though both principles are ultimately rooted in the consent given by states to respect each other’s territorial sovereignty, the principle of mutual trust is enshrined in the multilateral Conventions and EU/EC Regulations which make up the Brussels-Lugano regime and thus given contractual effect.¹⁰⁸ Moreover, even where the principle of mutual trust does not apply, it is submitted that the considerations underlying the concept of comity are equally present in the German courts’ remedial approach. Finally, regardless of whether courts recognise a bigger or smaller role for comity, they agree at the very least that its concern for the interests of the foreign legal order represents a hurdle to overcome.

1. Comity (and mutual trust) in English courts

Starting with the role of comity in English courts, it is first sensible to recap that they perform the function of preventing proceedings from commencing or continuing by declining jurisdiction when they are seised but derogated, and by granting anti-suit injunctions when they are prorogated. In the first scenario, there is no interference with proceedings before a foreign court. In the second scenario, though the anti-suit injunction is technically “directed not against the foreign court but against the part[y]” held to be in breach,¹⁰⁹ it has an effect—however disguised and indirect—on the foreign court.¹¹⁰ Yet, while it is well-established that comity generally exerts a restraining force against granting anti-suit injunctions,¹¹¹ English courts do not consider this to apply where the injunction is ordered to enforce a choice of court agreement. Instead, the dominant view seems to be that “comity […] has little if any role to play where anti-suit injunctive relief is sought on the grounds of breach of contract”,¹¹² with some going as far as stating that “the true role of comity” in these cases is to ensure that the parties’ agreement is respected.¹¹³

Similarly, English courts do not perceive comity as a serious restraining factor in compensating for loss caused by the breach of a choice of court agreement. Where a derogated foreign court has declined jurisdiction on the basis of the choice of court agreement but made no order as to costs, they see “no reason of comity why our courts should enforce a foreign jurisdiction’s policy perception in preference to our own” by declining to award damages.¹¹⁴ Indeed, an award of damages in these circumstances is entirely in line with comity given that it confirms the foreign judgment. But even in cases in which damages were awarded for loss in form of an unfavourable judgment on the merits by a derogated foreign court, comity seems to carry little relevance. Both the Court of Appeal in Starlight Shipping, and the Supreme Court by way of obiter dictum in the same litigation in The Alexandros T, briefly referred to the principle of mutual trust, then still applicable, but considered it not engaged.¹¹⁵ The underlying rationale of the decisions was that there can be no interference with the integrity of foreign proceedings where these have either already run their course or, if still on-going, the party continuing them is not restrained. The foreign decision is only taken as data for the assessment of loss.¹¹⁶ In fact, in Starlight Shipping, the award of damages was portrayed as an “acknowledgement of the Greek court’s jurisdiction.”¹¹⁷

Lastly, the theoretical bedrock of the recognition and enforcement of foreign judgments in English law is the “doctrine of obligation” rather than any notion of comity.¹¹⁸ Broadly speaking, there are two bases for recognition at common law:¹¹⁹ Presence of the person against judgment given was within the territory of the adjudicating court when the proceedings were instituted¹²⁰ and their acceptance of the adjudication of the foreign court, usually by agreement in advance¹²¹ or by submission after the proceedings are commenced.¹²² Comity provides a fundamental justification for recognition on the first basis, contributing to “the oldest and most deeply seated sense that sovereign acts are territorial, and when a sovereign has so acted, his act is to be respected, and if our courts are asked to do it, his judgments are to be recognized.”¹²³ However, in considering remedies for the breach of a choice of court agreement, only the second basis for recognition is of concern. If there is an agreement to accept the adjudication of a foreign court but one party initiates proceedings before another court instead, the English court’s refusal to recognise and enforce a resulting judgment does not implicate any considerations of comity. Since comity would not have encouraged the recognition and enforcement of the derogated foreign court’s decision, it cannot be offended by a refusal.¹²⁴ Moreover, as the refusal is territorially restricted in its effect, there can be no question of interference with the foreign legal order.

In conclusion, in the context of a breach of a choice of court agreement, English courts consider the relevance of comity to be slight, if anything. In practical terms, comity exerts no restraining force in the granting of any of the available remedies. Crucially, its limited relevance is predisposed in its conceptual foundation. The principle of comity, as understood by English courts, is based on the implied consent of one sovereign state to respect the territorial sovereignty of another.¹²⁵ This means that comity is not a matter of absolute obligation but instead operates only to the extent that consent can reasonably be implied.¹²⁶ On this premise, two further inferences can be drawn. First, in an international legal system based upon competing equal sovereignties, each country, through its courts, is its own arbiter of what comity demands. Second, comity is not imposed on states by some higher authority but is rather accepted on a voluntary basis. Consequently, this conceptual framework allows for a relatively lax treatment of comity, paving the way for the consideration of other principles to which higher importance is attached. Specifically, this concerns “the priority given to the parties’ bargain in English private international law” and the principle of party autonomy.¹²⁷ As has been shown in Part D, the English courts’ conception of choice of court agreements is primarily contractual. In viewing remedies for breach of a choice of court agreements as the largely orthodox enforcement of private obligations, their focus is predominantly on the agreement’s effects on the parties rather than the effects on the derogated court. In stark contrast, comity is primarily concerned not with private interests but with the interests of foreign legal systems and, by extension, their courts. As a result, considerations of comity are largely eclipsed by the contractual principle.

2. Comity and mutual trust in German courts

Moving on to the role of comity and mutual trust in German courts, it is recalled that they fulfil the function of preventing proceedings from commencing and continuing only partially. If German courts are seised but derogated, they will decline jurisdiction, provided that the choice of court agreement is invoked. Otherwise, their remedial approach is characterised by restraint, contrasting most sharply with the English courts’ anti-suit injunction. Where the Recast Brussels I Regulation applies, this is readily attributable to the principle of mutual trust. In Turner, the ECJ made clear that the principle of mutual trust lies at the heart of the anti-suit injunction’s incompatibility with the Brussels Convention. Specifically, it held that such an injunction “must be seen as constituting interference with the jurisdiction of the foreign court which, as such, is incompatible with the system of the Convention.”¹²⁸ Moreover, the court found that the anti-suit injunction implies an assessment of the appropriateness of bringing proceedings before a court of another Member State, noting that “[s]uch an assessment runs counter to the principle of mutual trust which … underpins the Convention and prohibits a court … from reviewing the jurisdiction of the court of another Member State.”¹²⁹

However, this cannot explain why German courts do not grant anti-suit injunctions in relation to proceedings in non-Member States where the principle of mutual trust does not apply. While the above-mentioned landmark decision by the Bundesgerichtshof was not directly concerned with anti-suit injunctions, it did raise the question whether the award of damages for breach of a choice of court agreement would be precluded by the ECJ’s decision in Turner.¹³⁰ Identifying the principle of mutual trust, and in particular the prohibition to review the jurisdiction of another Member State court, as the main reason for the incompatibility of the anti-suit injunction with the Brussels regime, it concluded that these considerations did not apply in relation to third countries such as the United States.¹³¹ Nevertheless, after noting that there could be no prohibited review “where the derogated court has declined jurisdiction” in any case, the court ruled that the choice of court agreement contained an obligation to pay damages “at least to the extent that the derogated court has recognised its lack of jurisdiction.”¹³² This suggests that the Bundesgerichtshof was concerned about interfering with the integrity of the foreign proceedings by contradicting the foreign court’s conclusion regarding the effect of the choice of court agreement, even where that court was not in another Member State. And if one had to find a label for that concern, it would be that of comity. Finally, if the court was reluctant to retrospectively interfere with the integrity of foreign proceedings already concluded, by way of awarding damages, it is likely that it would be at least as reluctant to interfere by way of restraining a party from commencing or continuing foreign proceedings. Crucially, at least where an anti-suit injunction during on-going foreign proceedings is sought, the foreign court will necessarily not have declined jurisdiction.

In parallel, the preceding analysis also sheds light on the role of comity and mutual trust in the German courts’ fulfilment of the compensatory function. After all, the Bundesgerichtshof decision was directly concerned with damages rather than an anti-suit injunction. It bears repeating one remarkable aspect: The court showed apparent concern for the potential impacts of its decision on the foreign court, regardless of whether the latter was in another Member State. Otherwise, nothing would have stood in the way of taking the inapplicability of the principle of mutual trust to its logical conclusion. The court could have simply interpreted the choice of court agreement as containing an obligation to pay damages in case of its breach, without restricting it to situations where the foreign court has declined jurisdiction. Instead, it seems that, for the Bundesgerichtshof, the question of whether an award of damages for breach of a choice of court agreement is in line with comity or mutual trust turns on the foreign court’s verdict regarding its own jurisdiction.

Finally, the relevance of comity and mutual trust in the protection of a party from the effects of a decision obtained from a derogated court differs vastly in cases outside and inside the scope of the Recast Brussels I Regulation. Under autonomous German law, comity plays no role in the recognition and enforcement of foreign judgments since it is not to be equated with either the mirroring principle or reciprocity.¹³³ In contrast, where the Recast Brussels I Regulation applies, “[m]utual trust in the administration of justice in the Union justifies the principle that judgments given in a Member State should be recognised in all Member States without the need for any special procedure.”¹³⁴ At the same time, it underpins the protection provided by the Regulation against the effects of a decision rendered by a derogated court, which operates at an earlier stage than that of recognition and enforcement.¹³⁵

Having considered the role of comity and mutual trust in relation to each of the available remedies, it has transpired that German courts are significantly more concerned about interfering with foreign judicial proceedings and resulting judicial orders than their English counterparts. In practical terms, comity and mutual trust are regarded as capable of precluding certain kinds of relief for breach of a choice of court agreement. First, this is due to the rather exceptional nature of the principle of mutual trust which is considerably more rigorous in its demands for judicial restraint than comity. In stark contrast to the consent-based conception of comity prevalent in English courts, the basis of the principle of mutual trust is contractual and thus obligatory in nature. It can be traced back to the 1968 Brussels Convention, an international treaty concluded between the six founding members of the European Communities (EC). While the word “mutual trust” does not feature in its text, the ECJ made clear that the Convention is “necessarily based on the trust which the Contracting States accord to one another’s legal systems and judicial institutions.”¹³⁶ Moreover, the adoption of the Brussels I Regulation in 2001 and its Recast version in 2012 on the basis of the pertinent European Treaties¹³⁷ embedded the principle of mutual trust on the supranational plane, further strengthening its obligatory force by linking it to the EC/EU. In fact, it is now regarded as one facet of the principle of sincere cooperation expressed in Article 4(3) of the Treaty on EU.¹³⁸

Crucially, the contractual basis of the principle of mutual trust both justifies and explains the limits of its restraining force. On the one hand, it imposes an obligation to “strictly obey to the provisions of the Regulation and to unhesitatingly enforce them through the national courts’,¹³⁹ particularly by not interfering with another Member State court’s exercise of jurisdiction.¹⁴⁰ In this sense, the wording is somewhat misleading as the mutual “trust” is enforceable. On the other hand, any such obligation applies only to the states which have ratified the Brussels Convention and/or the EU Treaties. In the latter case, the principle of mutual trust is even imposed by a higher authority, though an authority which has been vested with the corresponding competence by the states to which the principle applies. Moreover, mutual trust is confined to a compulsory system of common rules on jurisdiction and the recognition and enforcement of foreign judgments, rendering the potential for difference in opinion and thus the (perceived) need for interference minimal. Put simply, trusting countries within a relatively small group of countries with shared rules is significantly easier than extending the same trust to all sovereign states on the international plane.

However, there must be a further reason for the German courts’ judicial restraint, given that they are more reluctant in granting remedies for breach of a choice of court agreement than English courts even where the principle of mutual trust does not apply. In its landmark decision awarding damages for breach of a choice of court agreement, the Bundesgerichtshof seemed inspired by the principle of mutual trust, developing its reasoning against the background of Turner. More fundamentally, as has been shown in Part D, the German courts’ conception of choice of court agreements is primarily procedural, marked by a doctrinal reluctance to understand them as a source of private rights. It follows that a conception which is predominantly focused on the effects of a choice of court agreement on the prorogated and derogated courts is more susceptible to considerations of comity which is concerned with the interests of foreign legal systems. In other words, on a procedural conception of choice of court agreements, the mind of a judge turns more naturally to the potential effect of remedies on foreign courts rather than seeing them as operating merely in personam.

F. Conclusion

The comparison has revealed that the remedial approaches to the breach of a choice of court agreement in English and German courts generally perform the same functions, though in different ways and to different degrees. More specifically, English courts employ an entire arsenal of remedies, most strikingly including the anti-suit injunction and damages with the effect of effectively reversing a foreign judgment on the merits. In contrast, German courts exercise greater judicial restraint, relying instead on the principle of mutual trust in cases involving another EU Member State court and the refusal of recognition and enforcement to foreign judgments as a last resort. Against this backdrop, this article also serves as a source of information for parties about the potential benefits and drawbacks of choosing English and German courts in their choice of court agreements and the consequences of breaching them. On a purely pragmatic view and considering only the available remedies, parties will usually be well-advised to select the English rather than the German courts.

Moreover, two explanations for the diverging functional emphases of the respective remedial approaches have been developed. First, while English courts have adopted a primarily contractual conception of choice of court agreements, German courts tend towards a more procedural conception. Crucially, the former conception with its focus on private obligations lends itself to granting a wider range of remedies than the latter conception with its emphasis on courts. Second, the principle of comity and mutual trust play different roles in English and German courts. While functionally equivalent, comity and mutual trust are conceptually different, resulting in the latter being considerably more rigorous in its demands for judicial restraint than the former. But even where the principle of mutual trust does not apply, the German courts show more concern for the interests of foreign courts than their English counterparts, which corresponds to their primarily procedural conception of choice of court agreements. In light of this, a picture of two different approaches to a common problem has emerged, contributing to and facilitating an understanding of how legal systems converge and diverge in their pursuit of dispensing justice.

Acknowledgements

I would like to thank Dr Johannes Ungerer for his advice and the two anonymous reviewers for their constructive comments. Moreover, I would like to acknowledge the support provided by the Open Access Publication Fund of Humboldt-Universität zu Berlin.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Trevor Hartley, International Commercial Litigation (Cambridge University Press, 3rd edn, 2020), 198.

2 Fiona Trust & Holding Corp v Privalov [2007] UKHL 40; [2007] Bus LR 1719, [26] (Lord Hope).

3 Ibid.

4 For reasons of brevity, the author kindly asks that all references to England, including those to English courts and law, be read to include Wales.

5 The two principal EU Regulations on choice of law, Regulation (EC) No 593/2008 of 17 June 2008 on the law applicable to contractual obligations (Rome I) OJ L 177/6 (hereafter: Rome I Regulation) and Regulation (EC) No 864/2007 of 11 July 2007 on the law applicable to non-contractual obligations (Rome II) OJ L 199/40, have been retained, see The Law Applicable to Contractual Obligations and Non-Contractual Obligations (Amendment etc) (EU Exit) Regulations SI 2019/834, as amended by SI 2020/1574, reg 10–11.

6 Principally consisting of Regulation (EU) No 1215/2012 of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (recast) OJ L351/1 (hereafter: Recast Brussels I Regulation) and Convention of 30 October 2007 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters OJ L339/3 (hereafter: Lugano Convention).

7 The Civil Jurisdiction and Judgments (Amendment) (EU Exit) Regulations 2019 SI 2019/479, reg 82–91. For a full account of the consequences, see Andrew Dickinson, “Realignment of the Planets – Brexit and European Private International Law” (2021) Praxis des Internationalen Privat- und Verfahrensrechts 213.

8 Bundesgerichtshof, Judgment of 17 October 2019 – III ZR 42/19; BGHZ 223, 269 (all quoted passages from this judgment have been translated by the author).

9 For example: Civil Jurisdiction and Judgments Act 1982, s 30; German Code of Civil Procedure, s 24; Recast Brussels I Regulation, Art 24; Civil Jurisdiction and Judgments Act 1982, ss 15A–E; Recast Brussels I Regulation, Arts 10–23.

10 Hague Convention, Arts 1, 5, 6, 8.

11 Supra, n 6.

12 Status Table of the Contracting Parties to the Hague Convention, https://bit.ly/3PBKPxF (accessed on 19 June 2023).

13 European Union (Withdrawal Agreement) Act 2020, s 39(1).

14 Civil Jurisdiction and Judgments Act 1982, s 3D(1).

15 Case 26/62 Van Gend en Loos v Nederlandse Administratie der Belastingen [1963] ECR English Special Edition 1; Case 6/64 Flaminio Costa v E.N.E.L. [1964] ECR English Special Edition 585.

16 Though the EU Member State Denmark has also acceded to the Hague Convention in its own right, the Recast Brussels I Regulation – which applies between it and the other EU Member States based on the Agreement between the European Community and the Kingdom of Denmark on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, 19 October 2005, OJ L299/62 and updated 21 March 2013, OJ 2013 L79, 4 – takes precedence where both parties are resident in a EU Member State, Art 26(6) Hague Convention.

17 Andrew Dickinson, “Background and Introduction to the Regulation”, in Andrew Dickinson and Eva Lein, The Brussels I Regulation Recast (Oxford University Press, 2015), para 1.13.

18 For example: Peter Birks, “Rights, Wrongs, and Remedies” (2000) 20 Oxford Journal of Legal Studies 9–17; Rafal Zakrzewski, Remedies Reclassified (Oxford University Press, 2005), 43–61; Stephen A Smith, Rights, Wrongs, and Injustices: The Structure of Remedial Law (Oxford University Press, 2019), 6–8.

19 Andrew Burrows, Remedies for Torts and Breach of Contract (Oxford University Press, 4th edn, 2019), 1–2.

20 Ibid, 9.

21 cf Ellerman Lines Ltd v Read [1928] 2 KB 144 (CA), 152–153 (Scrutton LJ) (anti-enforcement injunction on the basis of an arbitration agreement); Bank St Petersburg OJSC v Archangelsky [2014] EWCA Civ 593; [2014] 1 WLR 4360, [29], [31] (anti-enforcement injunction on the basis of a choice of court agreement).

22 The Eleftheria [1970] P 94; Donohue v Armco Inc [2001] UKHL 64; [2002] 1 Lloyd’s Rep 425, [24].

23 Ulrich Magnus, “Prorogation of jurisdiction”, in Ulrich Magnus and Peter Mankowski (eds), Brussels Ibis Regulation – Commentary (Otto Schmidt, 2022), para 40.

24 This obligation is not expressly contained in the Regulation but follows partly from Art 26 and is otherwise “evidently intended to exist as part of the determination in Article 25”, Alex Mills, Party Autonomy in Private International Law (Cambridge University Press, 2018), 130; see also, Magnus, supra n 23, para 173.

25 Hague Convention, Art 6.

26 German Code of Civil Procedure, s 39.

27 Donohue, supra n 22, [24].

28 Ibid, [27].

29 Adrian Briggs, Civil Jurisdiction and Judgments (Routledge, 7th edn, 2021), para 28.20 (“it is likely that the English court will consider that this does not affect or restrict its power to grant relief”); see also Trevor Hartley, Choice-of-Court Agreements under the European and International Instruments (Oxford University Press, 2013), para 10.30 (“The Convention neither prohibits nor requires them: it is neutral”).

30 Briggs, supra n 29, para 28.06.

31 Ibid.

32 Clearlake Shipping Pte v Xiang da Marine Pte Ltd [2019] EWHC 1536 (Comm); [2019] 4 WLUK 616, [53]–[64].

33 Ibid, [61].

34 Samengo-Turner v J & H Marsh & McLennan (Services) Ltd [2007] EWCA Civ 723; [2007] 2 All E.R. (Comm) 813.

35 Petter v EMC Europe Ltd [2015] EWCA Civ 828; [2015] 7 WLUK 833.

36 Clearlake, supra n 32, [59]–[60].

37 SNI Aerospatiale v Lee Kui Jak [1987] AC 871 (PC), 892 (Lord Goff), relying on Bushby v Munday (1821) 5 Madd 297, 307; 56 ER 908, 918 (Sir John Leach VC) while noting that “[t]here are, of course, many other statements in the cases to the same effect”.

38 Ibid; see also British Airways Board v Laker Airways Ltd [1985] A.C. 58, 95 (Lord Scarman); Airbus Industrie GIE v Patel [1999] 1 AC 119 (HL), 131, 141 (Lord Goff).

39 Oberlandesgericht Düsseldorf, Decision of 10 January 1996 – 3 VA 11/95; [1997] ILPr. 320 (translation by the author, emphasis added). See also Laker Airways Ltd v Pan American World Airways, 559 F. Supp. 1124, 1128 (D.D.C. 1983) (“a direct interference”).

40 Anatol Dutta and Christian A Heinze, “Prozessführungsverbote im englischen und europäischen Zivilverfahrensrecht” [2005] 13 Zeitschrift für Europäisches Privatrecht 428, 433–34.

41 Patricia Londono, David Eady, A T H Smith, Lord Eassie, Arlidge, Eady & Smith on Contempt (Sweet & Maxwell, 5th edn, 2019), para 14-1.

42 Case C-159/02 Turner v Grovit [2004] ECR I-3565, [31]; The same reasoning applies in the context of the Brussels I (Recast) Regulation, Ulrich Magnus, “Introduction”, in Magnus and Mankowski, supra n 23, paras 1, 8. See also Case C-185/07 Allianz SpA v West Tankers Inc [2009] ECR I-663 (no anti-suit injunction to enforce an arbitration agreement).

43 Reichsgericht, Judgment of 3 March 1938 – IV 224/37; RGZ 157, 136–41.

44 Thomas Pfeiffer, Internationale Zuständigkeit und prozessuale Gerechtigkeit (Vittorio Klostermann, 1995), 767.

45 Ibid, fn. 238.

46 Reichsgericht, supra n 43, 138–40.

47 Jennifer Antomo, Schadensersatz wegen der Verletzung einer internationalen Gerichtsstandsvereinbarung? (Mohr Siebeck, 2017), 233, 243.

48 Bundesgerichtshof, supra n 8, [29].

49 Oberlandesgericht München, Judgment of 12 December 2019 – 6 U 5042/19; GRUR 2020, 379, [55], [69] (translation by the author); see also Landgericht München I, Judgment of 2 October 2019 – 21 O 9333/19; BeckRS 2019, 25536; Landgericht München I, Judgment of 25 February 2021 – 7 O 14276/20; GRUR-RS 2021, 3995; LG München I, Judgment of 24 June 2021 – 7 O 36/21; GRUR-RS 2021, 17662; Oberlandesgericht Düsseldorf, Judgment of 07 February 2022 – 2 U 27/21; GRUR-RS 2022, 1375 (quashing an anti-anti-suit injunction granted at first instance but confirming the general availability of such orders).

50 Oberlandesgericht Hamm, Judgment of 2 May 2023 – 9 W 15/23; BeckRS 2023, 10005, [7] (translation by the author).

51 Ibid.

52 Oberlandesgericht München, supra n 49, [72].

53 Turner, supra n 42, paras 24–25.

54 For the purposes of this article, English or German law is assumed to be the law applicable to the choice of court agreement. In practice, the determination is not as straightforward, especially since choice of court agreements fall outside the scope of the Rome I Regulation, Art 1(2)(e). For English law, see Lord Collins of Mapesbury and Jonathan Harris (eds), Dicey, Morris & Collins: The Conflict of Laws (Sweet & Maxwell, 16th edn, 2022), para 12-069. For German law, see Evgenia Peiffer and Marcus Weiler, “Vertraglicher Schadensersatzanspruch wegen Verletzung von Gerichtsstands- und Schiedsvereinbarungen – Teil I” (2020) Recht der internationalen Wirtschaft 321, 325.

55 Ellerman Lines, supra n 21.

56 Union Discount Co v Zoller [2001] EWCA Civ 1755; [2002] 1 WLR 1517.

57 Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG [2014] EWCA Civ 1010; [2014] 2 Lloyd’s Rep 544, [19]–[20]; see also Starlight Shipping Co v Allianz Marine & Aviation Versicherungs AG [2014] EWHC 3068 (Comm); [2015] 2 All E.R. (Comm) 747, [89] and the obiter dicta in Donohue, supra n 22, [36] (Lord Bingham), [48] (Lord Hobhouse), AES Ust-Kamenogorsk Hydropower Plant LLP v Ust-Kamenogorsk Hydropower Plant JSC [2013] UKSC 35, [2013] 1 WLR 1889 [25], and The Alexandros T [2013] UKSC 70; [2014] 1 All ER 590, [131]–[132], [135].

58 Briggs, supra n 29, para 222.

59 See infra, 222.

60 Bundesgerichtshof, supra n 8.

61 Bundesgerichtshof, supra n 8, [33].

62 Bundesgerichtshof, supra n 8, [17].

63 Starlight Shipping [2014] EWCA Civ 1010, supra n 57, [15]–[16].

64 Namely Starlight Shipping [2014] EWHC 3068 (Comm), supra n 57.

65 Case C-590/21 Charles Taylor Adjusting v Starlight Shipping, request for preliminary ruling of 23 September 2021.

66 Case C-590/21 Charles Taylor Adjusting v Starlight Shipping, Opinion of AG Richard de la Tour, [52]–[55].

67 Bundesgerichtshof, supra n 8, [30].

68 Bundesgerichtshof, supra n 8, [31].

69 Bundesgerichtshof, supra n 8, [47].

70 James Ruddell, “Monetary Damages for Wrongful Foreign Proceedings” (2015) Lloyd’s Maritime and Commercial Law Quarterly 9, 13.

71 It should be noted that the UK is currently considering becoming a Contracting State to the 2019 Hague Convention on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters, see http://bit.ly/3JngkY5 (accessed on 19 June 2023). Under Art 7(d) of that Convention, recognition and enforcement may be refused if the judgment was in breach of a choice of court agreement. The Convention would apply to judgments given by courts of the (other) Contracting States, notably including the Member States of the EU.

72 See Hague Convention, Art 8(2); Trevor Hartley and Masato Dogauchi, Convention of 30 June 2005 on Choice of Court Agreements, Explanatory Report (2013), para 166.

73 Adrian Briggs, The Conflict of Laws (Oxford University Press, 4th edn, 2019), 154.

74 The conditions in full are set out in the German Code of Civil Procedure, ss 328, 722(1), 723(2)(2).

75 Bundesgerichtshof, Judgment of 26 March 1969 – VIII ZR 194/68; BGHZ 52, 30, [36]–[40].

76 Ibid, [37].

77 Ibid, [38].

78 Turner, supra n 42, [25].

79 See supra, 218.

80 Airbus Industrie, supra n 38, 133.

81 Adrian Briggs, “The Impact of Recent Judgments of the European Court on English Procedural Law and Practice” [2005] 124 Zeitschrift für Schweizerisches Recht 231, 235.

82 See supra n 22; Bouygues Offshore SA v Caspian Shipping Co (Nos 1, 3, 4 and 5) [1998] 2 Lloyd’s Rep 461.

83 Briggs, supra n 29, para 29.06.

84 Donohue, supra n 22, [24] (emphasis added).

85 Donohue, supra n 22, [24].

86 Tweedle v Atkinson (1861) 1 B & S 393, 121 ER 762.

87 Briggs, supra n 81, 236.

88 Briggs, supra n 29, para 29.02; cf The Alexandros T, supra n 57; Starlight Shipping, supra n 57; Hin-Pro International Logistics Ltd v Compania Sud Americana de Vapores SA [2015] EWCA Civ 401; [2016] 1 All E.R. (Comm) 417; CMA CGM SA v Hyundai Mipo Dockyard Co Ltd [2008] EWHC 2791 (Comm); [2009] 1 Lloyd’s Rep 213 (concerning an arbitration agreement).

89 Starlight Shipping [2014] EWCA Civ 1010, supra n 57, [12]–[17].

90 CMA CGM, supra n 88, [37]–[40].

91 Edwin Peel, “How Private is English Private International Law?”, in Andrew Dickinson and Edwin Peel (eds), A Conflict of Laws Companion (Oxford University Press, 2021), 299, 304–306.

92 The Sennar (No 2) [1985] 1 WLR 490.

93 Adrian Briggs, Agreements on Jurisdiction and Choice of Law (Oxford University Press, 2008), para 6.05.

94 German Code of Civil Procedure, s 38(1); for EU law, see Briggs, Agreements, supra n 93, para 8.04.

95 Settled case law since Bundesgerichtshof, Judgment of 29 February 1968 – VII ZR 102/65; BGHZ 49, 384; most recently Bundesgerichtshof, supra n 8, [26].

96 Seminally: Gerhard Schiedermair, Vereinbarungen im Zivilprozess (Röhrscheid, 1935), 40, 100; Hendrik Schultzky, “§ 38 ZPO”, in Richard Zöller (ed), Zivilprozessordnung: ZPO (Otto Schmidt, 34th edn, 2022), para 4.

97 Briggs, Agreements, supra n 93, para 8.75.

98 Bundesgerichtshof, supra n 8, [26].

99 Ibid.

100 Ibid, [29].

101 Mills, supra n 24, 129; see also supra, 222–224.

102 See supra, 223.

103 Similarly: Andrew Dickinson, “Taming Anti-suit Injunctions”, in Dickinson and Peel, supra n 91, 77, 109.

104 British Airways Board v Laker Airways Ltd [1984] Q.B. 142 (CA), 185–86 (Sir John Donaldson M.R.)

105 Adrian Briggs, “The Principle of Comity in Private International Law” (2012) 354 Recueil des Cours 181.

106 Magnus, supra n 42, [2a]; Case C-116/02 Erich Gasser GmbH v MISAT Srl [2003] ECR I-14963, [72]; Turner, supra n 42, [24].

107 See supra n 7.

108 Briggs, supra n 105, 88, 134.

109 SNI Aerospatiale, supra n 37, 892 (Lord Goff).

110 See supra, 215.

111 SNI Aerospatiale, supra n 37, 892 (Lord Goff); Airbus Industrie, supra n 38, 133 (Lord Goff).

112 Enka Insaat ve Sanayi AS v OOO Insurance Company Chubb [2020] UKSC 38; [2020] 1 WLR 4117, [180], [184] (Lords Hamblen and Leggatt); The Angelic Grace [1995] 1 Lloyd’s Rep 87, 96. Both cases concerned arbitration agreements, but the cited rationale applies equally to choice of court agreements. See also Highland Crusader Offshore Partners LLP v Deutsche Bank AG [2009] EWCA Civ 725; [2010] 1 WLR 1023, [50].

113 OT Africa Line Ltd v Magic Sportswear Corp [2005] EWCA Civ 710; [2006] 1 All ER (Comm), [32] (Longmore LJ); see also Clearlake, supra n 32, [33] (Bryan J: “comity requires that [the] agreement is respected”, emphasis added)

114 Union Discount, supra n 56, [23]–[26] (Schiemann LJ).

115 Starlight Shipping [2014] EWCA Civ 1010, supra n 57, [15]–[16]; The Alexandros T, supra n 57, [39].

116 Most pointedly: The Alexandros T, supra n 57, [132].

117 Starlight Shipping [2014] EWCA Civ 1010, supra n 57, [16] (Longmore LJ).

118 Godard v Gray (1870) L.R. 6 Q.B. 139, 149–150; Schibsby v Westenholz (1870) L.R. 6 Q.B. 155, 159; Adams v Cape Industries Plc [1990] Ch. 433 (CA) 513.

119 Adrian Briggs, “Recognition of Foreign Judgments: A Matter of Obligation” (2013) 129 Law Quarterly Review 87, 90–92.

120 Adams, supra n 118.

121 Vizcaya Partners Ltd v Picard [2016] UKPC 5; [2016] 3 All ER 181.

122 Rubin v Eurofinance SA [2012] UKSC 46; [2013] 1 AC 236.

123 Briggs, supra n 105, 150.

124 Similarly: Ibid, 152–53.

125 James Edelman and Madeleine Salinger, “Comity in Private International Law and Fundamental Principles of Justice”, in Dickinson and Peel, supra n 91, 325, 327–28; Seminally: Joseph Story, Commentaries on the Conflict of Laws (Hillard, Gray, and Company, 1834), § 23.

126 Edelman and Salinger, supra n 125, 337.

127 See supra, 225–227.

128 Turner, supra n 42, [27].

129 Turner, supra n 42, [28].

130 Bundesgerichtshof, supra n 8, [30].

131 Bundesgerichtshof, supra n 8, [30].

132 Bundesgerichtshof, supra n 8, [31], [33].

133 See supra, 223.

134 Recast Brussels I Regulation, Recital 26.

135 See supra, 218.

136 Turner, supra n 42, [24].

137 Treaty establishing the European Communities, Arts 61(c), 65; Treaty on the Functioning of the European Union, Arts 67(4), 81.

138 Dickinson, supra n 17, para 1.68.

139 Magnus, supra n 42, [2a].

140 For example: Gasser, supra n 106; Turner, supra n 42; West Tankers, supra n 42.