Skip navigation |

The differing approach to commercial litigation in the ECJ and the courts of England and Wales

Speech by the Rt. Hon. Sir Anthony Clarke, Master of the Rolls




I must begin with a word of thanks and a word of apology. The thanks are due to David Gladwell, who of course runs the Court of Appeal (and thus all our lives). He is a member of the Friends of the Institute of Advanced Legal Studies and about six months ago asked me if I would address this august body. So here I am, although I am not sure if thanks is quite the right word. It is one thing to agree to give a talk and quite another to prepare it. Fortunately I have had the great help of John Sorabji in doing so, although he has sensibly gone on holiday rather than listen to me. Present or not, my thanks are due to him. The good bits are his, the errors are mine.

The word of apology is that the title to this lecture does not give any very clear picture of what it will contain. Its purpose is to highlight the differing approaches taken by the English courts on the one hand and the European Court of Justice ("the ECJ") on the other to the exercise of its jurisdiction in civil and commercial matters. In particular recent decisions in the ECJ have shown that its approach to anti-suit injunctions is markedly different from the approach adopted here, especially by the Commercial Court and the Admiralty Court. The cases are well known. They are of course Erich Gasser GmbH v MISAT Srl [2005] QB 1, Turner v Grovit [2005] 1 AC 101 and Owusu v Jackson (t/a Villa Holidays Bal Inn Villas) [2005] QB 801. These decisions and the rationale that underlies them to my mind have important implications for the future.

Before turning to an analysis of them and their implications I shall first say a word about the traditional English approach to questions of jurisdiction and its exercise, then turn to review the background and rationale underlying the Brussels Convention regime and its aims in an attempt to put the latest case developments in context and to highlight the different approach taken here and in Europe.

Approach of the English courts

I have spent much of my professional life both at the Bar and as a judge dealing with cases in which parties, usually defendants, have done their utmost to avoid having the dispute tried on the merits in England. Arguments of every kind have been deployed over the years to persuade courts that the interests of justice lie in the issues being determined elsewhere, although in very many cases the true position is that the defendant's real interest is to ensure (if at all possible) that the issues will in practice never be determined at all.

The English courts have drawn a distinction between due service of process, which is necessary to confer jurisdiction on the court and the exercise of that jurisdiction. Thus the court may decline to exercise its jurisdiction and grant a stay on a number of different bases, depending upon the circumstances.

As to jurisdiction, the English courts have traditionally accepted jurisdiction, at least in so far as actions in personam are concerned, on four bases: presence in the jurisdiction, consent, exercise of the discretionary power to assume jurisdiction over parties out of the jurisdiction under what was RSC Order 11 and is now CPR 6.20 and 6.21 and international convention.

As to presence, it is to be noted that it was traditionally presence and not domicile which founded jurisdiction. As to consent, many commercial parties have consented to the jurisdiction of the English courts. Such consent can of course be given in a number of ways. For example, litigants domiciled outside England may voluntarily submit to the court's jurisdiction by appointing an agent in England to accept service on their behalf: see Tharsis Sulphar Co v Société des Metaux (1889) 58 LJQB 435. In such cases where jurisdiction was established as of right, the court would only grant a stay of an action where the defendant could show that the proceedings were in some way vexatious of oppressive: see eg McHenry v Lewis (1882) 22 Ch D 397 at 408 and Lord Morris's excellent summary of the development of this line of authority in The Atlantic Star [1974] AC 436 at 455ff.

It was objected that this led to plaintiffs looking for a forum which would be most favourable to them and to undesirable forum shopping. Lord Denning's answer to that suggestion is to be found in his judgment in The Atlantic Star [1973] QB 364 at 381 - 382:

" No one who comes to these courts asking for justice should come in vain. ... This right to come here is not confined to Englishmen. It extends to any friendly foreigner. He can seek the aid of our courts if he desires to do so. You may call this 'forum shopping' if you please, but if the forum is England, it is a good place to shop in, both for the quality of the goods and the speed of service."

That has a certain ring to it, perhaps particularly appealing to those practising commercial law in London. However, it did not last. Due mainly to the efforts of Lord Goff, who had, as counsel, argued for their introduction in The Atlantic Star, the principles forum non conveniens as understood in Scots law were introduced into English law: see The Abidin Daver [1984] AC 398. As I am sure all present are fully aware, this doctrine seeks to ascertain which country is a dispute's 'natural forum' and to resolve an application for a stay or anti-suit injunction accordingly: see Spiliada Maritime Corporation v. Cansulex Ltd [1987] A.C. 460 per Lord Goff at 478. Where one party has brought proceedings in England concerning, for instance, the performance of a contract in the United States or Hong Kong it will try to ascertain which is the most appropriate forum to decide that dispute.

Why did the English courts adopt the Scottish approach? The answer is I think clear. It is an expression of the fundamental principle of doing substantial justice to the parties in litigation. This was recognised by Lord Goff in Spiliada. He said at page 474:

" In cases where jurisdiction has been founded as of right, i.e. where in this country the defendant has been served with proceedings within the jurisdiction, the defendant may now apply to the court to exercise its discretion to stay the proceedings on the ground which is usually called forum non conveniens. That principle has for long been recognised in Scots law; but it has only been recognised comparatively recently in this country. In The Abidin Daver [1984] A.C. 398, 411, Lord Diplock stated that, on this point, English law and Scots law may now be regarded as indistinguishable. It is proper therefore to regard the classic statement of Lord Kinnear in Sim v. Robinow (1892) 19 R. 665 as expressing the principle now applicable in both jurisdictions. He said, at p. 668:"the plea can never be sustained unless the court is satisfied that there is some other tribunal, having competent jurisdiction, in which the case may be tried more suitably for the interests of all the parties and for the ends of justice""

It is perhaps an example of the application of the principle stated by Lord Mansfield CJ as long ago as 1768 in Alderson v Temple (1768) 4 Burr. 235, where he said:

"The most desirable object in all judicial determinations, especially in mercantile ones, (which ought to be determined upon natural justice, and not upon the niceties of law,) is, to do substantial justice."

See also per Lord Diplock in Bremer Vulcan v South India Shipping Corporation [1981] AC at 977.

The English courts thus adopted an approach, rooted in the jurisprudence of a civil law jurisdiction, which sought to ensure that cases were tried in the most appropriate forum in which their dispute can be resolved. This is a pragmatic approach exercised on a case by case basis. However, as is of course well known, the framers of the Brussels Convention rejected it in favour of a simpler and more certain approach, to which I will return in a moment.

In cases where there is no exclusive jurisdiction clause and English jurisdiction is established as of right and the principles of forum conveniens apply, it is for the defendant to persuade the court that the interests of justice require a stay of proceedings in favour of proceedings elsewhere. In cases where there is no exclusive jurisdiction clause and the permission of the court is necessary in order to permit service of the proceedings out of the jurisdiction, the claimant must satisfy the court first that the case falls within one of the categories of case set out in CPR 6.20 and, secondly, that the English court is "the proper place in which to bring the claim": see CPR 6.21(1) and 2A. The burden is thus on the claimant and not on the defendant. This is consistent with the principles in Spiliada.

The position is somewhat different if there is an exclusive jurisdiction clause in an agreement between the parties. This is an area in which there is a stark difference between our approach and that of the ECJ. In this class of case (and indeed in the case where there is an exclusive arbitration clause) the English approach has traditionally been that, where the parties have agreed a particular jurisdiction or arbitration in a particular place, the court will refuse or grant a stay (or anti-suit injunction) in favour of litigation or arbitration in the agreed forum unless there are strong reasons for not doing so: see eg The El Amria [1981] 2 Lloyd's Rep 119 and many other cases. I have often wondered why the courts conferred on themselves a discretion to allow a party to proceed in breach of an agreed jurisdiction or arbitration clause in a contract. There are, after all, no other terms of a contract which the court can disapply by the exercise of a discretion. In all other types of case the courts would (at any rate in the old days) simply have said pacta sunt servanda. However, that is by the by.

Until dealt what may be a fatal blow in Gasser and Turner (at any rate in the European context), the English courts adopted a robust approach to applications for anti-suit injunctions. They were willing to grant injunctions to restrain proceedings by claimants in other proceedings in breach of English exclusive jurisdiction or arbitration clauses. There are now many examples of this. It is sufficient to refer to Continental Bank NA v Aekos Compania Naviera SA [1994] 1 WLR 588 and The Angelic Grace [1995] 1 Lloyd's Rep 87. In the first case the English court granted an anti-suit injunction restraining proceedings brought before a Greek court in breach of an exclusive jurisdiction clause. It considered the question whether the procedural rule set out in Article 21 of the Convention overrode Article 17. It held it did not. In doing so the Court of Appeal relied upon the principle that substantive law takes precedence over procedural law. That approach was identified as being embodied in Article 17, which gives precedence to the courts of the agreed jurisdiction and thus enunciates the principle of party autonomy: see [1994] 1 WLR at 598. Thus parties should be held to their bargains so far as jurisdiction agreements are concerned, just as they should be held to other parts of their agreements.

The Court of Appeal took the view that to take any other view would lead to what were described as 'ludicrous' consequences. The consequences which the court had in mind no doubt included a device that we all now know by the name of the Italian Torpedo, to which I will return in a moment. It is the device whereby, for purely tactical purposes, a litigant deliberately rushes to institute proceedings (perhaps for a declaration of non-liability) in a forum other than the one which has exclusive jurisdiction under a contract. The forum chosen is usually one which is renowned for slow moving civil process, which offers a wide scope for procedural skirmishing. The tactical purpose is of course to cause undue delay, unnecessary costs and thereby pressurise the other party to settle at an unduly low level. Its purpose is clearly antipathetic to the achievement of justice.

In The Angelic Grace the court was not faced with the breach of an exclusive jurisdiction clause, but rather with the breach of an arbitration clause. The court concluded that there was nothing in principle to distinguish the two types of breach. It thus concluded that the Brussels-Lugano regime did not preclude the court from granting an anti-suit injunction. Again the English court adopted an approach which gave greater importance to holding the parties to their bargain than to the operation of a formal procedural rule.

Perhaps the clearest and typically incisive and robust statement of this principle is to be found in the judgment of Millett LJ in The Angelic Grace. I quote it in detail because it is the high water mark of the English approach. He said at page 96:

"In my judgment, the time has come to lay aside the ritual incantation that this is a jurisdiction which should only be exercised sparingly and with great caution. There have been many statements of great authority warning of the danger of giving an appearance of undue interference with the proceedings of a foreign Court. Such sensitivity to the feelings of a foreign Court has much to commend it where the injunction is sought on the ground of forum non conveniens or on the general ground that the foreign proceedings are vexatious or oppressive but where no breach of contract is involved. In the former case, great care may be needed to avoid casting doubt on the fairness or adequacy of the procedures of the foreign Court. In the latter case, the question whether proceedings are vexatious or oppressive is primarily a matter for the Court before which they are pending. But in my judgment there is no good reason for diffidence in granting an injunction to restrain foreign proceedings on the clear and simple ground that the defendant has promised not to bring them.The Courts in countries like Italy, which is a party to the Brussels and Lugano Conventions as well as the New York Convention, are accustomed to the concept that they may be under a duty to decline jurisdiction in a particular case because of the existence of an exclusive jurisdiction or arbitration clause. I cannot accept the proposition that any Court would be offended by the grant of an injunction to restrain a party from invoking a jurisdiction which he had promised not to invoke and which it was its own duty to decline.

We should, it was submitted, be careful not to usurp the function of the Italian Court except as a last resort, by which was meant, presumably, except in the event that the Italian Court mistakenly accepted jurisdiction, and possibly not even then. That submission involves the proposition that the defendant should be allowed, not only to break its contract by bringing proceedings in Italy, but to break it still further by opposing the plaintiff's application to the Italian Court to stay those proceedings, and all on the ground that it can safely be left to the Italian Court to grant the plaintiff's application. I find that proposition unattractive. It is also somewhat lacking in logic, for if an injunction is granted, it is not granted for fear that the foreign Court may wrongly assume jurisdiction despite the plaintiffs, but on the surer ground that the defendant promised not to put the plaintiff to the expense and trouble of applying to that Court at all. Moreover, if there should be any reluctance to grant an injunction out of sensitivity to the feelings of a foreign Court, far less offence is likely to be caused if an injunction is granted before that Court has assumed jurisdiction than afterwards, while to refrain from granting it at any stage would deprive the plaintiff of its contractual rights altogether.

In my judgment, where an injunction is sought to restrain a party from proceeding in a foreign Court in breach of an arbitration agreement governed by English law, the English Court need feel no diffidence in granting the injunction, provided that it is sought promptly and before the foreign proceedings are too far advanced. I see no difference in principle between an injunction to restrain proceedings in breach of an arbitration clause and one to restrain proceedings in breach of an exclusive jurisdiction clause as in Continental Bank N.A. v. Aeakos Compania Naviera S.A., [1994] 1 W.L.R. 588. The justification for the grant of the injunction in either case is that without it the plaintiff will be deprived of its contractual rights in a situation in which damages are manifestly an inadequate remedy. The jurisdiction is, of course, discretionary and is not exercised as a matter of course, but good reason needs to be shown why it should not be exercised in any given case.

As appears later, the contrast between the views of Millett LJ (and the other members of the court, namely Neill and Leggatt LJJ) on the one hand and those of the ECJ on the other is striking (to put it no higher). The approach taken by the English court in these cases places greatest weight on what commercial parties have agreed between themselves in arriving at the conclusion that it had the jurisdiction to impose anti-suit injunctions. Its interpretation of the regime in both cases rested on a case by case analysis of the merits of the particular case. The court in both cases was implicitly asking the question as to what justice, substantive justice, required it to do. Should it consistently with substantive law hold the parties to their bargain as to jurisdiction or arbitration? Or should it interpret the Convention so as to give one party a procedural advantage that would not only permit it to evade substantive law but permit it to evade substantive justice through use of a procedural device? The English courts took the view that substantive matters took precedence and that commercial parties who had reached a bargain should be held to that bargain. Procedural law should not be used to permit that bargain to be frustrated. Its commitment to affording precedence to the achievement of substantive justice informed its approach to interpretation of the Convention.

I will return in a moment to the inroads which have been made into those principles by Gasser and Turner.

The fourth class of case to which I referred earlier was jurisdiction under an international convention. The only such convention with which we are concerned this evening is the Brussels Convention on Jurisdiction 1965 as subsequently amended. The relevant rules are now contained in the Council Regulation (EC) Regulation 44/2001. I will call the whole system the Brussels-Lugano regime.

The Brussels-Lugano regime

The regime was intended to provide those countries to which it applies with a single, uniform, jurisdictional system in the field of civil and commercial private international law. This aim was clearly set out in both the Brussels Convention's preamble and Rapporteur Jenard's report. It was intended to simplify, and render more expeditious, the procedure for the mutual recognition and enforcement of judgments of the courts of each contracting state in other contracting states under Article 220 (now Article 293) of the Treaty of Rome 1957. That was the Convention's aim. It was not however a freestanding one. Its aim furthered the implementation of a wider Community goal, which as Rapporteur Jenard [Footnote 1] records was described in a note sent by the Commission of the European Economic Community to the six member states requesting that they commence the negotiations that resulted in the Convention. The note stated that:

"a true internal market between the six States will be achieved only if adequate legal protection can be secured. The economic life of the Community may be subject to disturbances and difficulties unless it is possible, where necessary by judicial means, to ensure the recognition and enforcement of the various rights arising from the existence of a multiplicity of legal relationships. As jurisdiction in both civil and commercial matters is derived from the sovereignty of Member States, and since the effect of judicial acts is confined to each national territory, legal protection and, hence, legal certainty in the common market are essentially dependent on the adoption by the Member States of a satisfactory solution to the problem of recognition and enforcement of judgments."

The overarching aim of the Convention was thus to further the creation of a fully-functioning single internal market for the EEC pursuant to, for example, Articles 3 and 7 of the Treaty of Rome. This aim had two subsidiary, or what might better be described as, facilitative aspects. First it required an increase in legal certainty within the common market. As Jenard put it, the Convention sought to bring about "a genuine legal systematization . . . (to) . . . ensure the greatest possible degree of legal certainty." [Footnote 2] This would in turn facilitate an increase in legal protection of those individuals and corporations domiciled within the common market.

These three aims have been stressed by the ECJ throughout the past 38 years: see for instance: Owusu Bank Ltd v Bracco [1994] ECR I-117. Equally they were stressed in Jenard and Möller's Report on the Lugano Convention: see paragraphs 4 - 5 and 8 - 11. They were again emphasised in the preamble to Regulation 44/2001 in the following terms:

" (1) The Community has set itself the objective of maintaining and developing an area of freedom, security and justice, in which the free movement of persons is ensured. In order to establish progressively such an area, the Community should adopt, amongst other things, the measures relating to judicial cooperation in civil matters which are necessary for the sound operation of the internal market.(2) Certain differences between national rules governing jurisdiction and recognition of judgments hamper the sound operation of the internal market. Provisions to unify the rules of conflict of jurisdiction in civil and commercial matters and to simplify the formalities with a view to rapid and simple recognition and enforcement of judgments from Member States bound by this Regulation are essential."

Legal certainty was stressed in paragraphs 11 and 15 of the preamble.

The Brussels-Lugano regime's aims can thus be summarised as threefold: first, the creation of a single, straightforward and supranational code governing jurisdiction and the enforcement of judgments, which, secondly, increases legal certainty surrounding such matters in those nations which form part of its jurisdictional area and, thirdly, thereby assists the creation of a fully operative single internal market within those member states.

The regime was to achieve these aims in two connected ways: first, by providing a directly effective system of law, or as in the United Kingdom's case an indirectly effective one in the form of the Civil Jurisdiction and Judgments Act 1982 ("the 1982 Act"), which overrode each member state's own domestic law, except where recourse to such law was specifically provided for within the regime itself: see, for instance, Articles 1 (2), 4, 67 and 71 of the Regulation, the corresponding Articles of the Conventions and Sanicentral GmbH v Collin [1979] ECR 3423; and secondly, by ensuring common, autonomous, interpretation of its terms: see Mulox v Geels [1993] ECR I-4075 at 10 - 11. It was therefore not only a supranational jurisdictional regime, but one which operated according to supranational jurisprudence developed by the ECJ. It was inevitable that the ECJ would develop a European approach which would, at the very least differ from, and might well conflict with, the approach taken by national courts.

Its operation was likely (as has proved to be the case) to differ most starkly from the traditional English pragmatic approach because it operates along formalist lines. This is I think demonstrated by the three cases to which I referred earlier and to which I shall refer in a moment.

The English approach to the Brussels-Lugano regime

The regime became part of English law and is now set out in the Regulation, which has direct effect which is recognised in Civil Jurisdiction and Judgments Order 2001 (SI 2001/3929) and reflected in CPR 6.19. As a result, subject to certain restrictions, permission is not required in order to serve proceedings out of the jurisdiction in respect of a claim which the court has power to determine under the Regulation. There is no room for the doctrine of forum non conveniens, at any rate as between parties who are domiciled in a member state or member states, because the Regulation states which court has jurisdiction in each class of case. Moreover, it has detailed provisions as to which of one or more possible courts in different member states has jurisdiction. Its essential philosophy is that the court first seised has jurisdiction and other do not: see article 21 of the Convention (now article 27 of the Regulation).

It is not necessary for me to set out the detailed provisions of the Regulation here for two reasons. The first is that you all know them by heart and the second is that, if I did, this lecture would never come to an end. It is sufficient to note that the basic principle that jurisdiction depends on domicile is contained in Article 2 of the Convention, now Article 2 of the Regulation. Article 16 of the Convention (now Article 17 of the Regulation) provides for exclusive jurisdiction in particular classes of case. Article 17 of the Convention (Article 23 of the Regulation) provides for the exclusive jurisdiction of the courts of a state in the case where the parties have included an exclusive jurisdiction clause in their contract. Article 21 (now Article 27) provides that, where proceedings involve the same cause of action, any court other than the court first seised must decline jurisdiction. Article 22 (now Article 28) provides that where actions are brought in the courts of different contracting states any court other than the court first seised may stay its proceedings.

It is right to say that the English courts initially took a narrow view of the regime's ambit. They did so by adopting an interpretative approach to the Brussels-Lugano regime which sought to keeps its ambit within bounds set by what they took to be its primary aim: the creation of a single market between the member states. They took therefore an approach to the regime which sought to retain as broad an ambit of application for national law as was compatible with that aim.

This approach was clearly evidenced in the decision in the Court of Appeal in In re Harrods (Buenos Aires) Ltd [1992] Ch 72. In that case proceedings were brought in England in respect of the winding up of an English company. An issue arose as to whether Argentina was the more appropriate forum on the ground that the company carried out its business exclusively in Argentina. The Brussels Convention was (to use a Strasbourg expression) prima facie engaged as the company was the defendant in the action and was domiciled in England. However the Court of Appeal accepted an argument that the Convention did not apply because the question was whether the courts of a member state or the courts of a non-member state were the appropriate forum and the Convention only applied where the issue of jurisdiction was as between the courts of member states.

In the Court of Appeal both Dillon and Nicholls LJJ concluded that the Brussels Convention was limited in its ambit to governing international relations between member states amongst themselves, rather than governing member states international relationships per se. They did so by identifying the Convention's purpose, in Jenard's words, as the creation of:

". . . an autonomous system of international jurisdiction in relations between the member states. . ." (per Dillon LJ at 96.)

The single market was thus one that existed only as between member states. That its scope was limited in this way flowed from its other aim, namely the creation of a common jurisdictional basis for the mutual recognition and enforcement of judgments between member states. If that was the Convention's purpose under Article 220 of the Treaty of Rome, it was difficult to envisage how the Convention's jurisdictional rules could apply in litigation involving parties domiciled outside the member states. Bingham LJ adopted a similar interpretative approach, although he did so by placing reliance on the aim of the Treaty and the Convention to create for the various member states a single, jurisdictional unit between themselves alone and to do so by replacing prior bilateral agreements between several of the member states: see page 101 et seq.

The rationale behind the Court of Appeal's decision was subject to a number of criticisms, of which the two most serious are noted in Cheshire & North's Private International Law at page 264 as first, that it displayed a misunderstanding of the jurisdictional ambit of the Convention and, second, that it created uncertainty in the law. In Owusu the ECJ has held that the decision was wrong.

The three decisions

Before looking at the issues raised by the three decisions it is perhaps helpful to give an overview of them.

Erich Gasser GmbH v MISAT Srl [2005] QB 1

In Gasser proceedings were brought in both Italy and Austria arising out of an alleged breach of a commercial contract for the supply of children's clothing. The first set of proceedings was brought in April 2000 by MISAT before a court in Rome. In August 2000 Gasser commenced proceedings before a court in Feldkirch in Austria. Gasser argued that the Austrian court was the appropriate forum under the Convention because Austria was the place of performance of the contract within Article 5 (1) and because of a choice of jurisdiction clause in the contract, with the result that the Austrian court had jurisdiction under Article 17. MISAT argued that the Austrian court had no jurisdiction as it was domiciled in Italy and that Article 2 applied. It also contested the validity of the choice of jurisdiction clause and argued that the Roman court must determine that question as the court first seised. The Austrian court stayed the proceedings before it of its own motion in accordance with Article 21. Gasser appealed.

The Austrian appellate court referred two questions to the ECJ. It is the first of those which is of interest. It had two limbs: first, whether a court second seised could proceed to determine the case without waiting for the court first seised to determine jurisdiction where it, the court second seised, had exclusive jurisdiction under the contract; and secondly, if this were impermissible as a general rule, whether an exception to the effect of Article 21 could legitimately be made where the courts of the member state first seised were subject to excessive procedural delay. The ECJ gave its judgment in December 2003 and answered both questions in the negative: the procedural rule embodied in Article 21 took precedence over Article 17 and no exceptions could be made to its operation on the basis of internal aspects of a member state's procedural system: see the judgment at paragraph 46 - 47 and 70 - 73.

Turner v Grovit [2005] 1 AC 101

Shortly before Gasser was handed down substantially the same constitution of the ECJ heard Turner. The issue in the action was straightforward. Mr Turner, a solicitor, was employed by a group of companies as a legal adviser. During the course of his employment he was transferred to work in Madrid. He worked there for a very short period before he sought to terminate his employment. He then issued proceedings before an employment tribunal in England claiming unfair dismissal. Having dismissed an objection to the proceedings on the grounds of jurisdiction, the employment tribunal found in Mr Turner's favour. While those proceedings were continuing the defendants commenced proceedings against Mr Turner in Madrid alleging a breach of contract. Mr Turner applied for what is inelegantly known as an anti-suit injunction to restrain the defendants from proceeding with the Spanish action. An interim injunction was initially granted by the High Court in December 1998 but the judge refused to renew it in February 1999. Mr Turner appealed that refusal. The Court of Appeal allowed the appeal and reimposed the injunction.

Laws LJ gave the leading judgment. The Court of Appeal held that the English and Spanish proceedings were both concerned with the same subject matter, namely the termination of Mr Turner's contract, and that as the court first seised the English tribunal had exclusive jurisdiction under Article 21. More significantly perhaps the court held that the Spanish proceedings were brought merely to harass and intimidate Mr Turner. As Laws LJ put it:

". . . it is to my mind plain beyond the possibility of argument that the Spanish proceedings were launched in bad faith in order to vex the plaintiff in his pursuit of the application before the Employment Tribunal here."

He said that the court's power to grant an anti-suit injunction to restrain such proceedings derived from its inherent jurisdiction to prevent the abuse of process and that that power was not inconsistent with the Brussels Convention. On the contrary he held that:

". . . the carefully constructed system of mutual recognition between jurisdictions established by the Brussels Convention is built on the premise that the courts of one State only will hear the case, and its judgment may be enforced, without further consideration of the merits, in any of the other contracting States. To my mind it follows that where a party in the courts of one State seeks to vex and oppress his opponent by process against him in another State, directed to issues which are being or could be litigated within the proceedings in the first State, the case is to all intents and purposes the same as one where, within this jurisdiction, one party oppresses his adversary by the issue and prosecution of multiple actions."

Unassisted by the decision of the ECJ, I would have formed the view that that reasoning was impeccable. The defendants appealed to the House of Lords, which was of the preliminary opinion that, where a second set of proceedings was commenced in bad faith, there was no inconsistency between the grant of such an injunction and the provisions of the Brussels Convention: see [2005] 1 AC at 106. However it referred the question to the ECJ. The ECJ in a short judgment disagreed with both the Court of Appeal and the preliminary view of the House of Lords. The ECJ held that such injunctions were impermissible because they were incompatible with the Convention and they were incompatible with the Convention because they ran counter to the doctrine of mutual trust between member states, which underpinned the Convention.

Owusu v Jackson (t/a Villa Holidays Bal Inn Villas) [2005] QB 801

The third case raised the correct application of Article 2 of the Convention. Mr Owusu was domiciled in the United Kingdom. He hired a holiday villa in Jamaica from the first defendant, Mr Jackson. Whilst he was on holiday he suffered a serious injury, which resulted in him being left tetraplegic. The injury occurred while he was using a private beach owned and operated by the second defendant to which the terms of the holiday let gave him access. Mr Owusu brought a claim against Mr Jackson in England for damages for breach of contract. In addition he brought actions in tort against five Jamaican-based defendants. Four of the six defendants, including Mr Jackson, applied to the High Court for declaratory relief. The basis of their application was that the English court should decline jurisdiction or stay the action pursuant to the doctrine of forum non conveniens. They argued that Jamaica was the appropriate forum to hear the dispute because both the five Jamaican defendants and the relevant witnesses had closer links with Jamaica than with England and that the accident took place in Jamaica.

At first instance the judge refused to grant the order sought. He did so, following the ECJ's decision in Universal General Insurance Co (UGIC) v Group Josi Reinsurance Co SA [2001] QB 68 that jurisdiction under Article 2 of the Convention depended on the domicile of the defendant. The judge held that, since the first defendant was domiciled in the UK, the court could not decline jurisdiction or stay the proceedings against him and that it followed that the appropriate forum for the proceedings against the other non-UK based defendants was England even though the Convention did not apply to them and Jamaica was in one sense the more convenient forum. It was more appropriate for the English court to have sole jurisdiction for all proceedings. Otherwise there was a risk that separate proceedings in the two countries would lead to inconsistent judgments on the same or similar facts and issues.

The defendants appealed to the Court of Appeal, which referred the following questions to the ECJ:

"(1) Is it inconsistent with the Brussels Convention . . . where a Claimant contends that jurisdiction is founded on Article 2, for a court of a contracting State to exercise a discretionary power, available under its national law, to decline to hear proceedings brought against a person domiciled in that State in favour of the courts of a non-Contracting state:if the jurisdiction of no other Contracting State under the 1968 Convention is in issue;
if the proceedings have no other connecting factors to any other Contracting State?(2) If the answer to question 1 (a) or 1 (b) is yes, is it consistent in all circumstances or only in some and if so in which?"

Both the Advocate-General in his opinion and the ECJ in its judgment concluded that there was no scope for an application of the doctrine of forum non conveniens where a defendant was domiciled in a member state. It made no difference to the exclusive applicability of Article 2 that the dispute centred on a non-member state and the acts or omissions of other defendants who were domiciled in that state.

This decision was perhaps predictable in the light of the court's reasoning in Turner. In paragraph 35 of his opinion in Turner Advocate-General Ruiz-Jarabo Colomer drew the following comparison:

"The effects of restraining orders are similar to those produced by application of the doctrine of forum non conveniens, whereby a decision may be made not to hear actions which have been brought in an inappropriate forum. Restraining injunctions, however much they are addressed to the parties and not to a judicial authority, presuppose some assessment of the appropriateness of bringing an action before a specific judicial authority. However, save in certain exceptional cases which are not relevant here, the Convention does not allow review of the jurisdiction of a court by a judicial authority of another contracting state: . . . "

The rationale behind the decisions

The rationale underlying these decisions is stark in its simplicity: the Brussels-Lugano regime must be interpreted so as to further its three overarching aims, which I identified earlier, that is to say, it must be interpreted so as
to promote legal certainty thereby enhancing legal protection for those domiciled in the member states;
to place proper weight on the doctrine of mutual trust between member states' and their internal legal systems; and
thereby to enhance the development of the EU internal market.

Legal certainty

The importance of legal certainty is emphasised throughout the three decisions. The court's rejection of a power to derogate from Article 21 was rejected in Gasser on two grounds.

The first was that there was no explicit power in the Convention to derogate: see paragraph 71. The absence of an explicit power of course does not necessarily mean that a power could not be implied into the Convention. Although the Court did not address the question of implication, the second reason for rejecting the derogation demonstrates why it would not have been able do so: such a power would undermine the aim of ensuring the achievement of legal certainty which the Convention sought to ensure by creating a single, common jurisdictional regime: see paragraph 51. It is interesting to note that the importance of this aim was one which the Court felt no need to expand on in great detail because its status and importance were accepted as common ground between the parties: see paragraph 72 and its importance is implicitly acknowledged at paragraphs 44 - 45 of Advocate-General Lèger's opinion.

The role that legal certainty plays in guiding the Court's interpretation of the regime is also inherent in Advocate-General Ruiz-Jarabo Colomer's opinion in Turner. He said in paragraph 33:

"The Convention seeks to provide a comprehensive system, for which reason it is appropriate to ask ourselves whether a measure which has an impact on its field of application is compatible with the common rules which it establishes. The question must be answered in the negative."

A comprehensive system is one which tends towards certainty: it is of the widest application. Equally, and obviously, the creation of common rules applicable to all, in this case Convention member states, fosters certainty. The ability of States to arrogate a power outwith that common regime to decide questions of jurisdiction by, for instance, issuing an anti-suit injunction would in the Advocate-General's words lead to 'chaos'. It would be wholly antipathetic to legal certainty because, although the Advocate-General does not put it in these terms, it would in effect permit individual States to reintroduce national rules governing jurisdiction, something which the Convention explicitly aims to remove.

The Advocate-General's opinion on this issue was approved by the ECJ in paragraphs 29 and 30 of its judgment, where it clearly stresses that the regime's commitment to legal certainty must not be undermined by member states deploying any procedural mechanism in respect of litigation governed by the regime except those sanctioned by the regime itself. The regime is all-embracing.

Advocate-General Lèger's opinion in Owusu offers the most detailed exposition of the commitment to legal certainty. That aim is identified at paragraph 159 - 162 of his opinion in these terms:

"(159) In the terms of its preamble, the Convention aims 'to strengthen in the Community the legal protection of persons therein established'. Again according to the preamble, it is for that purpose that the Convention lays down, first, rules concerning the jurisdiction of courts common to the Contracting States and, second, rules to facilitate recognition of judgments and to establish an expeditious procedure for their enforcement.(160) The Court has clarified the meaning of that aim of the Convention, in particular with regard to the common jurisdictional rules which it contains. It has taken the view that the strengthening of the legal protection of persons established in the Community involves 'enabling the claimant to identify easily the court in which he may sue and the defendant reasonably to foresee in which court he may be sued'. The Court has also characterised those rules as 'guaranteeing certainty as to the allocation of jurisdiction among the various national courts before which proceedings in matters relating to a contract may be brought'.(161) Only jurisdictional rules meeting those requirements are capable of guaranteeing observance of the principle of legal certainty, which is also, according to settled case-law, one of the objectives of the Brussels Convention.(162) In my view, those two aims of the Convention, both that of strengthening legal protection for people established in the Community and that of ensuring legal certainty, mean that the application of Article 2 of the Convention cannot be made conditional on the existence of a dispute displaying connections with different Contracting States."

Those aims would be undermined by permitting an interpretation of Article 2 of the Convention which could give rise to litigation concerning the question of when it applied and when it did not. Most significantly, perhaps, the Advocate-General identified disputes as to intra-community jurisdiction as particularly problematic and such as would give rise to the greatest degree of legal uncertainty if such litigation were permitted: see paragraphs 164 - 168. To militate against such uncertainty he concluded that:

"(168) . . . in more general terms it is important to bear in mind that private international law is a discipline which it is far from easy to handle. The Brussels Convention is a specific response to a concern for simplification of the rules in force in the various Contracting States regarding jurisdiction of the courts, as well as recognition and enforcement. That simplification contributes, in the interest of everybody, to promoting legal certainty. It is also intended to facilitate the work of national courts in dealing with proceedings. It is therefore preferable not to introduce into the system created by the Convention elements which are liable seriously to complicate its operation."

Looking more specifically at the possible application of a forum non conveniens doctrine to the Brussels-Lugano regime, the Advocate-General concluded, with explicit reference to the ECJ's decision in Turner, that permitting its introduction into the regime would run counter to the aim of ensuring legal certainty and undermine its efficacy. He said at paragraphs 263 - 270:

"(263) . . . by allowing the court seised the opportunity to decline - in a purely discretionary manner - to exercise the jurisdiction which it derives from a provision of the Convention, such as Article 2, the doctrine of forum non conveniens seriously affects the predictability of the effects of the jurisdiction rules laid down by the Convention, in particular the rule in Article 2. . . . that predictability . . . constitutes the only way of ensuring observance of the principle of legal certainty and ensuring greater legal protection for people established in the Community, in accordance with the objectives pursued by the Convention. Any impact of that kind on the predictability of the jurisdiction rules . . . thus ultimately detracts from the effectiveness of the Convention.(264) In that connection, it is important to bear in mind that the Convention is largely inspired within the civil law system, which attaches particular importance to the predictability and inviolability of rules on jurisdiction. That dimension has a lower profile in the common law system, since the application of the rules in force is approached in a somewhat more flexible manner and on a case-by-case basis. In that way, the forum non conveniens doctrine fits easily within the common law system, since it grants the court seised the power to exercise a discretion in considering whether or not it is appropriate to exercise the jurisdiction vested in it. It is therefore clear that that doctrine is hardly compatible with the spirit of the Convention.(265) Quite apart from the foregoing general considerations, it is important to consider in greater detail the procedural consequences of implementing the forum non conveniens doctrine. In my view, such consequences would be difficult to reconcile with the objectives of the Convention which, let it be remembered, relate both to observance of the principle of legal certainty and to greater legal protection for people established in the Community.(266) As we have seen, as English law stands at present, the application of that doctrine entails a stay of proceedings, that is to say suspension of an action, which may operate sine die. That situation is inherently unsatisfactory in terms of legal certainty.(267) Moreover, in my view, instead of providing greater legal protection for people established in the Community, the forum non conveniens doctrine is more liable to undermine it. That is particularly true for claimants.(268) It bears repeating that it is upon a claimant seeking to escape the effect of the procedural objection in question that it is incumbent to establish his inability to secure a just outcome in the foreign forum in question. Here too, that situation is not satisfactory, in view of the real fear that that procedural objection may be invoked by certain defendants for the sole purpose of delaying the progress of proceedings against them.(269) Furthermore, where the court seised has finally decided to allow the plea of forum non conveniens, it is once again incumbent upon a claimant wishing to re-initiate proceedings to produce the evidence necessary for that purpose. Thus, it is for the claimant to establish that the foreign court does not ultimately have jurisdiction to hear the case or that he himself is not likely to secure a just outcome in that court or has not been able to do so. That burden of proof on the claimant may prove particularly heavy. In that respect, application of the forum non conveniens doctrine is therefore liable to have a considerable impact on the defence of his interests, so that it tends to detract from rather than reinforce the legal protection enjoyed by the claimant, contrary to the objective of the Convention.(270) Finally, in the event of the claimant not succeeding in producing the evidence in question to oppose a stay of proceedings (which could be pronounced sine die) or to recommence proceedings already suspended, the only possibility that would remain if he sought to pursue his claims would be to take all the steps needed to commence a new suit before the foreign court. It goes without saying that those steps have a cost and are likely considerably to prolong the time spent in the conduct of proceedings before the claimant finally has his case heard. Moreover, in that respect, the mechanism associated with the forum non conveniens doctrine could be regarded as incompatible with the requirements of Article 6 of the European Convention on the protection of human rights and fundamental freedoms."

The Advocate-General's approach was, in a much pared down form, adopted by the Court in its judgment in Owusu. In doing so it emphasised, at paragraphs 38 - 43, that the achievement of legal certainty was one of the regime's aims and that that would not be "fully guaranteed" if the doctrine of forum non conveniens was held to be compatible with the regime's operation. It stressed once again that that doctrine would undermine "the predictability of the rules of jurisdiction laid down by the Brussels Convention . . . and consequently to undermine the principle of legal certainty, which is the basis of the Convention."

It is more than apparent from the judgments and the opinions of the Advocates General in the three cases that the ECJ acknowledges the achievement of legal certainty as a fundamental instrument of interpretation. The commitment to certainty brooks no exceptions and guides the regime's interpretation and application. In so doing it has led the ECJ to accept an interpretation of the Convention which affords territorial range of very considerable scope. Its international jurisdiction now governs not only jurisdictional questions between the 29 member states but equally (one might say) their extra-member state relations.

Mutual trust

The second element which guided the ECJ in this line of authorities is the doctrine of mutual trust. That is the idea that each member state must place trust in the courts of other member states properly to carry out their obligations under the Convention. While there is nothing new about this doctrine - indeed, as Blobel and Spath rightly note in The Tale of Multilateral Trust and the European Law of Civil Procedure (2005) European Law Review, 30 (4) at 528, it can be traced back to the earliest days of the European institutions - its importance as a guiding principle to the interpretation and application of EC law had not previously been stressed so clearly. Blobel and Spath again rightly note that both Gasser and Turner 'stand out (as judgments) for the decisive weight . . .' the ECJ grants to the doctrine of mutual trust: ibid at 531. It played no part in Owusu.

Its importance is indeed emphasised in robust terms. Advocate-General Lèger relies on it in his opinion in Gasser, in rejecting the argument that Article 21 could be derogated from where the court first seised with a dispute is in a member state, such as those identified at paragraph 85 of his opinion i.e., "Italy, Greece and France", whose courts are known for excessive delay. He makes the point as follows:

"(88) . . . It does not really seem conceivable that it should be possible to refrain from applying article 21 of the Convention on the ground that the court first seised is established in a member state in whose courts there are, in general, excessive delays in dealing with cases. That would be tantamount to saying that the rules on lis pendens do not apply where the court first seised is established in one of certain member states.(89) Such an interpretation would be manifestly contrary to the scheme and the basis of the Brussels Convention. The Convention does not contain any provision to the effect that its rules, in particular those of article 21, should cease to apply because of the length of proceedings before the courts in another contracting state. Moreover . . . the Convention is based on the trust which the member states accord to each other's legal systems and judicial institutions: see R v Medicines Control Agency, Ex p Smith & Nephew Pharmaceuticals Ltd (Case C-201/94) [1996] ECR I-581 . . . It is on the basis of that trust that the Convention establishes a compulsory system of jurisdiction which all the courts within its purview are required to observe. It is also that trust which enables the contracting states to waive the right to apply their internal rules on the recognition and enforcement of foreign judgments in favour of a simplified mechanism for recognition and enforcement. It is therefore also the basis of the legal certainty which the Convention seeks to ensure by allowing the parties to foresee with certainty which court will have jurisdiction."

The Advocate-General's opinion was emphatically endorsed by the ECJ, which at paragraph 72 of its judgment attributes to mutual trust a key foundational role in the development of the Brussels-Lugano regime:

". . . it must be borne in mind that the Brussels Convention is necessarily based on the trust which the contracting states accord to each other's legal systems and judicial institutions. It is that mutual trust which has enabled a compulsory system of jurisdiction to be established, which all the courts within the purview of the Convention are required to respect, and as a corollary the waiver by those states of the right to apply their internal rules on recognition and enforcement of foreign judgments in favour of a simplified mechanism for the recognition and enforcement of judgments."

That foundational role, in tandem with the need to ensure legal certainty and the absence of an explicit power of derogation, left the Court in no doubt that the procedural rule enunciated in Article 21 was not one which could be derogated from where the court first seised was one from a legal system characterised by delay. Derogation on such a basis would tend to undermine the requisite trust that all member states must repose in each other's legal systems.

A more detailed examination of the doctrine's importance is to be found in Turner at paragraphs 26 - 34. First, the Advocate-General reviewed the argument put forward by the UK government that anti-suit injunctions were aimed at protecting an individual claimant from being harassed by 'obstructive procedural measures' arising from abusive litigation by defendants in other jurisdictions and thus protecting the integrity of UK proceedings and that their use would thereby achieve the regime aim of reducing the number of proceedings before the courts of various member states. In his opinion that argument failed because of the doctrine of mutual trust. He thought that anti-suit injunctions 'cast doubt' on the doctrine. That they did so was in his view 'decisive'. It was impermissible to undermine the doctrine of mutual trust. He noted that the commitment to mutual trust had most recently been emphasised in Recitals 16 and 17 of the Regulation. The use of anti-suit injunctions, which either directly or indirectly undermined mutual trust, was impermissible because it would undermine European judicial co-operation, "which presupposes that each state recognises the capacity of other legal systems to contribute independently, but harmoniously, to attainment of the stated objectives of integration". I pause here to note that mutual trust and judicial co-operation are viewed as facilitative of a wider aim, namely integration and a single internal market.

The Advocate-General's rejection of the validity of anti-suit injunctions was again endorsed by the European Court. In its short judgment it highlighted the importance afforded to the doctrine of mutual trust by emphasising the rationale of decision in Gasser and by concluding that:

"(25) It is inherent in that principle of mutual trust that, within the scope of the Convention, the rules on jurisdiction that it lays down, which are common to all the courts of the contracting states, may be interpreted and applied with the same authority by each of them: see, to that effect, Overseas Union Insurance Ltd v New Hampshire Insurance Co (Case C-351/89) [1992] QB 434, 458, para 23, and Gasser, para 48."

It is perhaps unsurprising that with such a ringing endorsement of the doctrine the Court concluded that to permit the court of one member state to find that proceedings brought in another member state were abusively brought was impermissible. Assessing the appropriateness of proceeding in another member state undermines mutual trust as it effectively undermines first, the regime for such review put in place by the Convention, and secondly, the ability of the courts of other member states to assess for themselves the proceedings brought before them. Impinging on the jurisdiction of another court cannot but undermine trust between member states.

Development of the single internal market

The third limb which has guided the ECJ's interpretation of the regime is the aim to facilitate the growth of the single internal market. Of the three limbs it is the least developed in this field.

It is only implicitly referred to in the Gasser judgment, where the Court alludes to the Brussels-Lugano regime itself being an embodiment of the single market ideal by creating a common, uniform jurisdictional regime which all courts of all member states have equal authority to administer. The ECJ's view is that to permit derogations from its rules, such as that embodied in Article 21, would run counter to that single juristic space: see paragraph 48. A similar point is made in the Advocate-General's opinion in Turner at paragraph 37, where it is noted that while member states are autonomous insofar as their national civil procedural codes are concerned they must ensure that any such rules comply with the underlying the regime. The regime thus sets uniform, common standards with which all member states must comply. It thus facilitates the creation of a 'true internal market' between the member states.

The creation of common rules governing the determination of issues of private international law by way of the Brussels-Lugano regime is further identified in the Advocate-General's opinion in Owusu: see paragraphs 189 - 212. Rejecting the UK government's argument that the Brussels Conventions sought to facilitate the growth of a common market only in respect of the recognition and enforcement of judgments delivered in member states, he emphasised the content of the second and eighth recitals to the Regulation's preamble. As he put it at paragraph 196:

"(196) . . . as emphasised in the second and eighth recitals in the preamble to the regulation, the jurisdictional rules contained in it - in view of the diversity of the existing national rules in this area and the resulting difficulties for the proper functioning of the internal market - seek to 'unify the rules of conflict of jurisdiction in civil and commercial matters', so as to arrive at 'common rules' in the Member States. This exercise of unifying jurisdictional rules forms part of an approach comparable to that provided for in Article 94 EC for the adoption of directives, since the aim of that substantive legal basis is 'the approximation of such laws, regulations or administrative provisions of the Member States as directly affect the establishment or functioning of the common market'."

The Advocate-General drew further support for the importance of the need to interpret the regime compatibly with the aim of facilitating the growth of the single internal market by reference to the interpretation of Article 95 of the EC Treaty, Directive 95/46 and Regulation 1408/71. In arriving at its judgment the Court at paragraph 34 expressly referred to this discussion when it emphasised that the Brussels-Lugano regime was intended to facilitate the development of the internal market so that: ". . . there is a real and sufficient link with the working of the internal market, by definition involving a number of member states."


Where are we now? It seems to me that a number of conclusions can be drawn from a contrast between the approach of the English court and of the ECJ on the other to two particular aspects of civil and commercial litigation. The first is the anti-suit injunction and the second is forum non conveniens. I take them in turn.

The anti-suit injunction

The contrast between the approaches can be seen from a comparison between the English approach exemplified by the judgment of Millett LJ in The Angelic Grace, which I quoted earlier and the approach of the ECJ in Gasser and Turner. The latter approach has been the subject of no little academic criticism. That criticism includes articles by Jonathan Mance (now of course Lord Mance) in LQR 2004, 120 (JUL), 357-365 and Adrian Briggs LQR 2004, 120 (OCT), 529-533.

Jonathan Mance recognises that the decision in Gasser is imbued with pure European principle but observes that it promises problems for legitimate claimants and opportunities for those unwilling to meet their obligations. Article 21 (now article 27) adopts a simple test of chronological priority for the court first seised. The effect of Gasser is to give that court priority over the agreed jurisdiction, so that, until the court first seised has decided that there is indeed a binding jurisdiction clause in favour of the courts of another member state, the latter must decline jurisdiction in favour of the former and the former must accept jurisdiction. He says:

"It is at the practical efficacy of Art.17 (now Art.23) that the European Court's decision in Erich Gasser seriously strikes. London is one important elective jurisdiction. However, the decision is of far from parochial concern, and may even affect commercial parties' willingness to agree to litigate (as distinct from arbitrate) in Europe."

Jonathan Mance then contrasts the views of Advocate General Leger with those of the ECJ as follows:

"The Advocate General was M. Philippe Léger. In a comprehensive and nuanced Opinion he drew the analogy between exclusive jurisdiction under Art.16 (now Art.22) and Art.17 (now Art.23). In paras [57], [62] and [66]-[68], he pointed out that, if Art.21 prevailed, it would "seriously compromise" the utility of Art.17 and the legal certainty to which it contributed. He went on: "67... In effect, ... the party who, in violation of his obligations resulting from the agreed choice of jurisdiction, has first initiated proceedings before a tribunal which he knows to be incompetent could abusively delay the resolution of the dispute on the merits when he knows this would be unfavourable to him ... 68. This consequence is shocking as a matter of principle and risks encouraging delaying tactics..."

The Advocate General also noted that the basic problem was one of tactical manoeuvring, not simply delay in some judicial systems. A party commencing proceedings in a country (State A) other than the country agreed (State B) would "use every internal means to delay the moment when the decision that this jurisdiction is incompetent becomes definitive" (para.[69]). The solution was to allow the court second seised to continue to exercise jurisdiction provided that it could establish the existence and application of the agreed choice of jurisdiction clause in a rigorous manner and beyond any possible doubt—any risk of contradictory decisions being thereby largely avoided) paras. [81-83].

The European Court reached very different conclusions: see [2004] 1 Lloyd's Rep. 222. As to question two, the court second seised must under Art.21 always defer to a court first seised, unless and until that court declares itself incompetent. There should be a clear and precise rule, in view of "the disputes which could arise as to the very existence of a genuine agreement between the parties" within Art.17. A court second seised is "in no case ... in a better position than the court first seised to *360 determine whether the latter has jurisdiction" (para. [48]). Practical implications were summarily dismissed:

"53...the difficulties ... stemming from delaying tactics ... are not such as to call in question the interpretation of any provision of the Brussels Convention as deducted from its wording and purpose".

The court answered question three even more shortly, though with the Advocate General's support. An exception to Art.21 based on excessive delay was contrary to the letter, spirit and aim of the Convention (para. [70]). The Convention was necessarily based on mutual trust, and sought to ensure legal certainty (para. [72]. The court impliedly rejected the United Kingdom's fall-back suggestion that the court of State B could determine jurisdiction under and exclusive clause where (i) suit was issued in state A in bad faith to block any suit in State B and (ii) the court in State A had failed to adjudicate upon its own jurisdiction within a reasonable time.

It may comfort theoreticians that the Community has rules of ideological purity and logical certainty. But the result can only be practical uncertainty with large scope for tactical manoeuvring. The irrebutable assumption that all national systems operate for the best shows the barrier on the Rhine between Strasbourg and Luxembourg concerns and thinking.

Jonathan Mance expresses the view that the reasoning of the ECJ is in five critical respects unpersuasive: see pages 360-2. I will refer briefly to only three. His first point is that the ECJ's judgment seems ambivalent whether article 21 is subject to any exceptions. What about article 16? He asks the question:

"If Art. 16 is an exception, why should Art. 17, resting on party autonomy, be different? .... According to the Schlosser Report (para 22), not mentioned by the court, a court must "also of its own motion consider whether there exists an agreement on jurisdiction which excludes the court's jurisdiction and which is valid in accordance with article 17"."

His third point is that, whereas in some circumstances the court first seised may be better placed to rule on the question of jurisdiction, the reverse applies where the parties have chosen a jurisdiction. His fourth point (which is closely related to his third) is this:

"Fourthly, the 'legal certainty' so esteemed by the court consists apparently in knowing 'clearly and precisely which of the two national courts is to establish whether it has jurisdiction under the rules of the Convention' (paras 51 and 72). But the parties' commitment, when contracting, was that the chosen court should assume its exclusive jurisdiction without delay and without either party having to engage in litigation elsewhere to achieve this."

In conclusion Jonathan Mance suggests that in contrast to the ECJ's absolutist approach, the Advocate General's careful opinion offered a measured compromise. I must say that I agree with that.

In his article Adrian Briggs focused on Turner, in which the court simply held that an anti-suit injunction was incompatible with the Convention, even in a case in which the injunction was granted by the court first seised. He said that "taken together with the court's wilful weakening of jurisdiction agreements (noted by ... Jonathan Mance ...), an effective and sophisticated tool of English commercial litigation has been decommissioned." Adrian Briggs expressed his views in strong terms. He referred to the Continental Bank case and observed that the effect of Gasser and Turner is that it cannot be reconciled with the Convention. He added:

"It is well known that many continental lawyers have a peculiar hostility to the anti-suit injunction. As an antidote to jurisdictional shenanigans its usefulness is second to none, but its roots did not lie in civilian legal systems. So it had to go, as the dullardism of the lowest common denominator asserted itself. In its place, we are to repose trust in the other states' legal systems and judicial institutions. This pious substitute for adjudication may be all very well for judges and diplomats, but it was not much use to Gasser or Turner, and will doubtless be just as useless for future litigants who inhabit the real world. No doubt the enlargement of the European Union will bring even more opportunities to trust foreign legal systems, and make even more redundant the summary and direct enforcement of jurisdictional agreements and jurisdictional rules."

I would not myself use such strong language but it does seem to me to be a pity that the ECJ has set itself against the anti-suit injunction with quite such determination. Why not adopt the sensible compromise suggested by Advocate General Leger quoted above (and not expressly commented upon by the ECJ), namely that the solution is to allow the court second seised to continue to exercise jurisdiction provided that it could establish the existence and application of the agreed choice of jurisdiction clause in a rigorous manner and beyond any possible doubt—any risk of contradictory decisions being thereby largely avoided?

I hope that one of these days the ECJ might be willing to consider some of these considerations. After all there is now a good deal of academic comment to the effect that its present approach can be said to legitimise the use of a procedural device whose purpose is to frustrate the proper determination of disputes. This is the device which has become known as 'the Italian Torpedo': see Delaygua, Choice of Court Clauses: Two Recent Developments [2004] ICCLR 15 (9) 288 at 295. It is an approach which can be compared with the French courts' approval of their own use of anti-suit injunctions: see Banque Worms v Brachot (cass. 1re civ., November 11, 2002; 2003 Rev. Crit. 816. [Footnote 3] The French introduction of the anti-suit injunction might be thought to support the conclusion that this type of injunction is not anti-pathetic to civilian jurisdictions. It might also help to persuade the ECJ to give serious consideration to its approval as part of the armoury of all member states' courts as a necessary procedural device in order to ensure that the Brussels-Lugano regime is not abused.

However, until the ECJ is persuaded to change its view that anti-suit injunctions are impermissible, it must be recognised that the law applicable in England (as in other member states) is as stated in Gasser and Turner. Thus the Continental Bank case can no longer be regarded as good law.

The question remains whether the English court is now prevented from granting anti-suit injunctions in cases to which the Convention (or Regulation) does not apply. This is a point which I have had to consider in a judicial capacity in the comparatively recent past. This was in a case which we held was outside the Convention because it was within the arbitration exception in Article 1.2 (d) of the Regulation. The case was Through Transport Mutual Insurance Association (Eurasia) Ltd v New India Assurance Co. Ltd [2005] 1 Lloyd's Rep 67 in the Court of Appeal.

The basis of the jurisdictional dispute before the court was relatively straightforward. Certain goods were to be transported by ship from Calcutta to Kotka in Finland. They were then to be transported to Moscow. They were however lost during the final leg of their journey. They were insured against loss and damage by the defendants, New India Assurance. The shipper claimed against that policy. That claim was subsequently compromised and the benefit of any action against the carrier was assigned to New India. The carrier and an associated company that was also involved in the shipment however declared bankruptcy. They were insured by the claimants, Through Transport. New India issued proceedings in Finland under a Finnish third parties' rights against insurers statute against Through Transport. After the proceedings were served, Through Transport sought to contest jurisdiction and did so by way of issuing proceedings in England for a declaration that the parties were bound by an arbitration clause, which required the parties to arbitrate the dispute in England, and an anti-suit injunction. An issue arose between the parties as to whether the arbitration clause applied. The English High Court held that the Regulation did not apply as it did not apply to arbitration proceedings and it thus granted the sought for declaration and anti-suit injunction. New India appealed.

One of the issues the Court of Appeal had to decide was whether it was bound to apply the Gasser decision and hold that, as the court second seised, it must stay the proceedings before it pending the outcome of a jurisdictional challenge in Finland. It seemed to the Court that, at least as a matter of principle, there was an argument that the Regulation required the court first seised to determine the issue of jurisdiction i.e., whether the proceedings before it fell under the arbitration exception (see paragraph 24). It appeared to us however that such an approach was inconsistent with the ECJ's decision in Marc Rich & Co AG v Societa Italiana PA (The Atlantic Emperor) [1992] 1 Lloyd's Rep. 342.

A strong argument was put to us that The Atlantic Emperor could not now be relied on to support the conclusion that the English court had the requisite jurisdiction. That argument was based on an application of Gasser to the effect that as the court first seised the Finnish court was the correct court to determine the issue of jurisdiction. We took the view that Gasser could be distinguished as in that case both parties accepted that both sets of proceedings in Italy and Austria fell within the regime. In contrast the question before the court here was whether the English proceedings fell within the regime at all. It was not therefore a question of how did the regime apply but whether it applied at all (paragraph 36). Given that distinction and the decision in The Atlantic Emperor we concluded that the correct approach would be for the English court, even as the court second seised, to determine whether the proceedings fell within the regime and then only if they did apply the regime's procedural rules.

Agreeing with the judge, we held on the facts that the proceedings were within the arbitration exception and thus outside the Brussels-Lugano regime. On that basis the question arose whether we should grant an injunction restraining the claimant from proceeding in Finland. That raised the question whether it would be wrong in principle to do so in the light of the decision in Gasser and Turner. Should the principle in those cases that no anti-suit injunction be granted in the interests of certainty and mutual trust equally be applied where proceedings are said to have been brought in one member state in breach of an arbitration clause?

We reached the conclusion that the answer to that question was no. The case was not like Gasser because there was no breach of an exclusive jurisdiction clause and the proceedings were outside the Convention because they were within the arbitration exception and it was not like Turner because it was not a case in which the sole question was whether the proceedings restrained were vexatious or oppressive and there was no breach of an arbitration agreement. We reached the conclusion, rightly or wrongly that, where proceedings are brought in breach of an arbitration clause the principles expressed by Millett LJ in The Angelic Grace remain applicable. I recognise that there is scope for argument as to whether that is correct or not. It appears that Adrian Briggs thinks that it is not. However, the view we took was that, once it is held that the proceedings are in breach of an arbitration agreement, there is nothing in the Convention to prevent the court from granting an injunction on the basis of Millett LJ's principles. It seemed to us that no question of mutual trust arose, there was no reason why the Finnish court should be offended by an injunction which would simply have enjoined the New India personally. We noted that, if the positions had been reversed, there would have been no question of the English courts being offended by an injunction granted elsewhere enjoining claimants from continuing with proceedings in England in breach of an agreement to arbitrate. This approach was followed by Colman J in The Front Comor [2005] 2 Lloyd's Rep 257.

In the event, however, we allowed the appeal against the grant of an anti-suit injunction in that case because we formed the view that in the particular circumstances of the case the proceedings in Finland were not brought in breach of contract and that, applying Turner, it would not be appropriate to grant an injunction. Let it never be said that the English courts do not loyally apply the decisions of the ECJ to questions to which they are applicable. I should however add that in The Front Comor Colman J, following The Jay Bola [1997] 2 Lloyd's Rep 279 (a case not cited to us in Through Transport), granted an injunction in a case where the proceedings restrained were brought by subrogated underwriters. Neither side sought to appeal to the House of Lords in Through Transport but I understand that there is to be a leapfrog appeal to the House of Lords in The Front Coma. Moreover I understand that the appeal is to be heard in the near future. I shall look forward with great interest to the result, especially to the speech of Lord Mance.

Forum non conveniens

Owusu had also come in for some academic criticism, notably by Adrain Briggs in LQR 2005 121 (OCT) 535-540. He says that this was the third time in 15 months that the ECJ has struck a blow against international commercial litigation. As to Owusu he says:

"And now the court has ruled that though an English court has been satisfied, clearly and distinctly, that a court in a non-contracting state is more appropriate for the trial of the action, and has further determined that no injustice would be done to the claimant if the English proceedings were stayed, a defendant domiciled in the United Kingdom may not seek, and a court may not grant, a stay on the ground of forum non conveniens."

Adrian Briggs notes that the effect of Owusu is to overrule Re Harrods (Buenos Aires) Ltd, to which I referred earlier, and says that the real, if collateral, victims of the ruling were the five Jamaican defendants, who had no connection with any member state but found that they were in effect dragged to England to defend themselves. Briggs describes the ECJ as being 'airily dismissive of their predicament'. As he puts it, the ECJ simply stated that such considerations were not such as to call into question the fundamental rule of jurisdiction contained in Article 2.

I must say that this does seem a startling result and appears to promote the principle of legal certainty above the more pragmatic and (it might be thought) just solution of arriving at a fair conclusion on the facts of a particular case. However, it is not perhaps for me to say. I leave you to judge.


I am sure that these three decisions have not provided a fatal blow to commercial litigation, where in England or elsewhere but they have certainly not helped. I hope that the ECJ will bear some of these considerations in mind when deciding future questions of this kind.


  1. Jenard Report, preliminary remarks. [back to footnote 1]
  2. Jenard Report, chapter 4.2 [back to footnote 2]
  3. Cited in Briggs LQR (2004) 120, Oct at 530. [back to footnote 3]

Please note: that speeches published on this website reflect the individual judicial office-holder's personal views, unless otherwise stated. If you have any queries please contact the Judicial Communications Office.


We use cookies to ensure that we give you the best experience on our website. If you continue without changing your settings, we'll assume that you are happy to receive all cookies on the Judicial Office website. To find out more about the cookies, see our privacy policy.