Evialis S.A. v S.I.A.T. & Ors

Introduction

In this country, the claimants are Evialis, who assert that the contract of insurance provides for English jurisdiction. They seek an indemnity (or damages) under the contract of insurance, declarations about the jurisdiction provisions of the contract and damages for their breach, and injunctions in respect of proceedings brought by the insurers in Italy.

The insurers contend that the contract provides for Italian jurisdiction. In the Italian proceedings, they seek orders declaring that nothing is due from them under the contact of insurance, or restricting the loss in respect of which they are liable. In this judgment I shall use the term “non-liability declarations” to include declarations limiting the insurers’ liability as well as declarations absolving them from liability altogether.

Section 3 of the Brussels Regulation

The effect of these provisions is that, subject to any effective jurisdiction agreement, Evialis’ claim for an indemnity under the contract of insurance could have been brought either in Italy (being the Member State in which the Insurers are domiciled) or in France (being the Member State in which Evialis is domiciled): see article 9(1). A claim by the underwriters against Evialis could have been brought only in France: article 12(1).

Background

The goods were shipped on 21 September 2000 on board the vessel “Irene” for carriage from Abidjan to Rouen. Upon discharge at Rouen on 23 October 2000, the goods were found to be damaged. Proceedings were brought in France against the Owners of the vessel, and Evialis were awarded damages in the sum of E407,508.32. The Owners have not appealed against that award, and their time for appealing expired on 7 March 2003.

The amount in dispute between the parties is modest because of Evialis’s success in the French proceedings. The litigation here and in Italy is about the costs of the proceedings and some costs in the French proceedings, amounting, as I was told, to some E45,000. Nevertheless, the parties wished to proceed to a hearing in order to determine in which jurisdiction the claim should be determined. Mr S Morris QC represented Evialis, and Mr S Males QC and Mr. M. Ashcroft represented the insurers.

The insurance

The open cover was written in Italian, but the translation of the relevant provisions produced by Evialis in these proceedings was not materially disputed. It was on an Italian form, Polizza Italiana di Assicurazione Merci Trasportate (Edizione 1983). It comprised a front page and signature page specific to the cover, General Conditions, certain Special Agreements, attached Institute Clauses and a number of appendices issued over the years.

The front page stated:

The General Conditions stated in article 1 that:

Throughout the General Conditions there were references to Italian law. Two examples will suffice to illustrate this: Ametra were to declare the “name of the vessel for the purposes of Article 523 of the Navigation Code”: Article 2. In the event of a claim Ametra or the person insured were under various obligations, and there was provision that if they were not complied with “Article 1915 and 1918 of the Civil Code will be applicable”. The Italian focus of the insurance was also reflected in the requirement that if a loss occurred where the insurers did not have loss adjusters or experts “the intervention must be requested of some other loss adjuster or qualified expert or of the Italian consulate…”.

At article 16, under the heading “The Competent Court”, the General Conditions provided as follows: “The competent court, at the choice of the plaintiff, is exclusively that of the court office of the place in which is located the head office of the insurer or the agency which is in charge of the policy or in whose offices the contract was concluded”. Ametra were specifically to approve this provision “for the purposes of Articles 1341 and 1342 of the Civil Code”, and Ametra provided their signature accordingly.

The Special Agreements contained by way of introduction before the numbered articles a statement that, “These Special Agreements have priority, in the event of discrepancy, over the above General Conditions.” Article 2 was headed “Cover conditions” and read:

The certificate relating to the cargo on the “Irene” was on a form printed in Italian, with an English translation subscribed to the Italian text. It certified insurance “under policy No 31799”. Its printed terms included the following: “Cover is subject to the following conditions, Known to the parties: Polizza Italiana di Assicurazione Merci Trasporte (Ed 1983) integrated as follows: classification clause in force – institute radioactive contamination exclusion clause ed 1.10.90 - Computer millennium clause (cargo) (JC 98/024) – CARGO ISM ENDORSEMENT (JC 98/019)”.

It is to be observed that some of these clauses are not in line with the open cover. Most importantly, the certificate contained the provision about an “address for legal proceedings” (to which I shall refer as “the service clause”). Mr Morris identified other discrepancies: the certificate and the cover referred to different versions of the Institute radioactive contamination exclusion clause, and of the Institute Classification clause (a third version, that of 1 July 1978, being appended to the Open Cover); and the typed provision in the certificate that damages were to be settled irrespective of percentage does not reflect the open cover.

Copies of all the Institute clauses (in English) referred to in the Special Agreements (subject to the discrepancy about the version of the classification clause) were appended to the open cover, and all include a clause headed Law and Practice which provides, “This insurance is subject to English law and practice”.

Legal Proceedings

On about 21 June 2002 the insurers brought their proceedings dated 14 June 2002 before the Tribunale di Genova, and they were served on Evialis in France on or about 27 June 2002. In their summons the insurers stated that “The Court of Genoa has jurisdiction and territorial authority within the sense of Article 16 of the General Conditions of the Policy number 31799”. They sought the following relief:

In their defence, which was filed on 23 December 2002, Evialis challenged the jurisdiction of the Tribunale di Genova. They said (i) that they were domiciled in France; (ii) the policy of insurance contained an exclusive jurisdiction clause and (iii) Article 16 of the General Conditions of the policy was not relevant because it applied only where the Italian jurisdiction was established, which was not the case.

The proceedings in this court were brought on 20 December 2002, after the insurers had brought the Italian proceedings but before Evialis had filed their defence in which they made their jurisdictional challenge. In them Evialis seek both an indemnity arising from the cargo damage and interest (claims to which I shall refer as “the indemnity claims”) and relief related to their contention that the English court and not the Italian court should determine the indemnity claims (“the jurisdiction claims”). Specifically they claim this relief:

The applications

The parties are in dispute about what (if any) jurisdiction clauses were incorporated into the agreement between them. This dispute has given rise to issues about whether the contract of insurance is contained in the open cover as well as the certificate, and what law governs this question and the contract of insurance generally. However, it is convenient first to mention one point that is not in dispute. Although the parties to the dispute are the insurers and Evialis, and not Ametra, the original assured, it is common ground that this does not affect the questions that I have to decide. Neither party contends that the position of Evialis with regard to whether there is a binding and effective jurisdiction agreement differs from that of Ametra. This common ground is in accordance with the decision of Moore-Bick J in Glencore International AG v Metro Trading Inc., [1999] 2 Ll L R 632, applying The Tatry, [1999] QB 515.

Evialis submit that under English law the contract of insurance is contained in the certificate and only those terms of the open cover that are properly incorporated into it are terms of the contract of insurance. This is the basis for their argument that the terms of the contract of insurance included the service clause but not article 16 of the General Conditions. Mr Morris seeks to establish the position under English law because he submits that English law is the governing law of the contract (and, no doubt, because he would say that, if the contract is governed by Italian law, English law should be applied in the absence of satisfactory evidence that Italian law differs from it). I shall have to consider in due course whether Mr Morris is correct in his submission as to the proper law. However, the insurers dispute the position under English law, arguing that the proper approach is to look at the open cover and the certificate together, both being contractual documents between the insurers and their insured, Ametra. I shall consider this issue first.

The insurers cite in support of their submission the judgment of Saville LJ in Credit Suisse Financial Products v Societe General d’Enterprises, [1997] CLC 168. In that case a contract relating to a bond put option was agreed by telephone and confirmed by a fax which the other contracting party signed and faxed back. The confirmation referred to an “ISDA Master Agreement”, which contained a governing law and jurisdiction clause, and stated that the agreement and each confirmation thereunder would be governed by English law. The Court of Appeal concluded that there was an effective jurisdiction agreement under article 17 of the Brussels Convention. Saville LJ, having referred to the decision of the ECJ in Estasis Salotti di Colzani Aimo e Gianmario Colzani v RUWA Polstereimaschinen GmbH, [1976] ECR 1831, said, “To my mind the question is simply whether the express reference in the written contract in the present case amounts to a “clear and precise” demonstration that the clause conferring jurisdiction was the subject of a consensus between the parties. I have no doubt at all that it does”. So too in this case, the insurers say, in order to ascertain there is a clear demonstration of a consensus between the insurers and Ametra, it is necessary to look at both the open cover and the certificate.

Evialis submit that the focus of the Salotti decision was not whether the jurisdiction clause was incorporated into the contract between the parties, but whether the agreement was effective under article 17 of the Convention. This is of some significance when I consider whether these questions are to be adjudicated by reference to the proper law of the contract, but it suffices at this point to observe the similarity of the approach under article 17 as a matter of community law and under English law: see AIG Europe SA v QBE International Insurance Ltd., [2001] 2 Ll L R 268 at para 26.

Next Mr. Morris refers to the well established rule that in order for arbitration clauses to be incorporated from charterparties into bills of lading, specific or clear words are required in the bill or in the charterparty: see Scrutton on Charterparties (20^th Ed.) p.79. I do not consider this a powerful analogy. Again I refer to AIG Europe SA v QBE International Insurance Ltd. (cit sup). It is one thing to incorporate terms from an entirely separate contract. It is another for a contract evidenced in a certificate to include terms from the open cover under which the certificate was issued.

Mr. Morris is also able to rely upon three cases concerned with insurance certificates served under open covers, namely De Monchy v Phoenix Insurance Company of Hartford, (1929) 34 Ll L Rep 201, MacLeod Ross & Co Limited v Cie d’Assurances Generales l’Helvetia, [1952] 1 Ll L.R. 12 and Tradigrain SA v SIAT SpA, [2002] EWCA 106 (Comm), [2002] 2 Ll L.R. 553.

In De Monchy v Phoenix Insurance, (cit sup) the House of Lords considered the relationship between an insurance policy and a certificate that had been issued under it. The appellant insurers argued that it was necessary to look at the policy in order to ascertain the contract of insurance. Visc Dunedin said (at p.205):

Mr Males points out that in the De Monchy case the certificate contained the words, “This certificate represents and takes the place of the policy, and conveys all the rights of the original policy holder (for the purpose of collecting any loss or claims) as fully as if the property was covered by a special policy…”. He argues that the question before the House of Lords was about the interpretation of a particular certificate and policy, and submits that the speeches in the House of Lords are not authority for any general proposition about the relationship between insurance policies and certificates issued under them.

That was not the view of Somervell LJ in MacLeod Ross & Co Limited v Cie d’Assurances Generales l’Helvetia (cit sup). That case concerned a police d’abonnement (or open cover) issued to a firm of forwarding agents that contained a jurisdiction clause, and an avenant d’assurance (or certificate of insurance) issued under it. The claim was made by a company that bought goods from a customer of the forwarding agents and became the holder of the certificate. The first question before the Court of Appeal was whether the clause conferring jurisdiction was a term of the contract between the parties. By the certificate, the insurers stated that in accordance with a declaration under the open cover they had insured the goods “on its general and special conditions”. There was a note at the bottom that read, “See back for extracts from printed conditions”. On the back were certain general conditions from the open cover, but not the jurisdiction clause. The Court of Appeal decided that the holder was not bound by the jurisdiction clause.

Somervell LJ (at p.15) said this in his judgment:

In his judgment in MacLeod Ross Denning LJ described De Monchy as “not directly in point” but “quite in line with” his decision, which was that there were two contracts, one between the insurers and the forwarding agents and one between the insurers and the holder of the certificate. Accordingly the question that Denning LJ asked himself was “What are the terms of the insurance contained in the certificate of insurance?” He answered this as follows (at p16): “When the insurance company printed conditions on the back [of the certificate] they presumably printed all those which concerned the holder, and omitted those which did not concern him. I do not think it would be right to hold him bound by other conditions which were not drawn to his attention, especially procedural conditions which do not affect the insurance itself but only the way of enforcing it.”

The third case referred to by Mr Morris was the decision of Colman J in Tradigrain v SIAT, (cit sup). That was another case of certificates issued in respect of declarations under an open cover. One question was whether a Hamburg jurisdiction clause in the open cover prevented the assured from suing in England. Mr Morris, as I understood his submission, suggested that the approach of Colman J was to look primarily to the terms of the certificate in order to answer that question. I accept that he does refer to the question in term of whether there was “incorporation of the Hamburg jurisdiction clause from the open cover”. However, I do not read his judgment as turning upon this point, and I do not consider that the Tradigrain case offers any real support for Mr Morris’ argument.

I am not convinced that this dispute really amounts to more than a question of emphasis. Given that Ametra and the insurers were parties to both the open cover and the certificate, I consider that the proper approach to identifying the contract of insurance is to consider both documents as potentially expressing the parties’ intention as to the terms upon which the insurance was effected. I do not consider that, in the absence of express words so providing, it is proper to approach the question on the basis that the certificate is the only contractual document determining the terms of the insurance and to have regard to the terms of the open cover only to the extent that they have been incorporated into the certificate. However, in determining the terms of the insurance contract by interpreting the two documents, it is likely that the certificate will be given more weight, since, being directed to the specific risk, it is likely to prevail over more general terms. As Visc Dunedin put it in De Monchy (cit sup), it “clinches the bargain as to the particular shipment” and as such is “the determinative of the two instruments”. It is true that there are references in the judgments to the terms of the cover being “incorporated” into the certificate, but this does not, in my judgment, require express reference to them in the certificate. On analysis, I do not consider that that requires more than that the parties evinced an intention that a term in the open cover should be of contractual effect in relation to the insurance of the particular risk, which they might do despite it not being referred to in the document that was specifically directed to that risk.

The governing law of the contract of insurance

Accordingly, Italian law would be applicable to the contract unless the parties chose the law of another country. Paragraph 2 of the Schedule provides that any such choice must “be expressed or demonstrated with reasonable certainty by the terms of the contract or the circumstances of the case”. The essential question, therefore, is whether there was so expressed or demonstrated a choice of English law by the parties, that is to say by Ametra and the insurers. In my judgment, that question is to be determined in accordance with the law of Italy, the law which would be applicable to the contract in the absence of an express choice of law: see Dicey and Morris The Conflict of Laws (13^th Ed.) at para 32-101. However, I would have reached the same conclusion whether I determined the question by the law of England or that of Italy.

Mr Morris submits, and I accept, that paragraph 2 of the Schedule contemplates that the choice of the parties can be made impliedly rather than expressly, but his argument is that here the parties chose English law as the applicable law because they incorporated into the contract of insurance the Institute Clauses, which in turn included the Law and Practice clause and the choice of English law for “this insurance”. Moreover, he says that the service clause shows that the parties decided that any litigation arising from the contract should or might be in the English courts, and so the parties’ apparent choice of English law is not so remarkable as to invite scepticism. Since in English law the certificate is at least the more “determinative” of the documents containing the contract of insurance, it is not in point that article 1 of the General Conditions state that the insurance is subject to Italian law.

Mr Males, on the other hand, says that the certificate does not incorporate the Institute clauses without qualification, but incorporates them as part of and under the umbrella of the Polizza Italiana (this being the connotation of the words “integrated as follows”). Hence, he argues, the Institute clauses are introduced into the contract of insurance only for that purpose, and to the extent and subject to the qualifications dictated by the open cover, and that there the relevant reference in the open cover to the Institute clauses is in article 2 of the Special Agreements (the reference in the General Conditions being barren because no Institute Clauses were specified on the front page). However, the terms of the open cover make it clear that the Law and Practice provision in the Institute clauses does not evince a choice of English law as the governing law of the contract of insurance. First, the front page of the open cover says that the General Conditions are supplemented by additional clauses and the Special Agreements “for the purposes of the limitation and duration of cover”. (This is reflected also in article 1 of the General Conditions itself). They –and so the Institute clauses that they introduce - are not introduced to select the governing law. Secondly, article 1 of the General Conditions expressly states that terms of the insurance were supplemented by “the clauses cited on the front page” – including article 2 of the Special Agreements – “subject to Italian law remaining applicable”.

However, this argument assumes that the clauses in type on the certificate (as well as those in print) are governed by the words “integrated as follows”, and are not governed directly by the words, “The following conditions, Known to the parties”. Mr. Morris disputes this, and so does not accept that insurers’ contention that the Institute clauses are introduced into the certificate as part and parcel of the “integration” of the Polizza Italiana. He points out that the service clause was never part of the Polizza Italiana, and the words “integrated as follows” cannot govern that typed provision, and that the certificate does not refer to the same versions of the Institute Clauses as the Open cover.

I accept Mr Morris’ argument thus far. However, it does not follow that the reference in the certificate to Institute clauses expresses a choice by the parties that the contract of insurance should be governed by English law. I do not understand Mr Morris to dispute that the open cover itself is governed by Italian law, and so the choice of English law as the governing law of one particular cover declared under it and recorded in the certificate would, on the face of it, be rather surprising, particularly as the certificate was issued by Italian insurers through an Italian broker to an Italian assured. The Law and Practice provisions were not specifically referred to in the certificate. It might be that the effect of the provisions was that the interpretation of the Institute clauses themselves was governed by English law and practice – that does not call for consideration on these applications. However that may be, I do not consider that the reference to the Institute clauses in the certificate evinces the parties’ intention to choose an English governing clause for the contract of insurance as a whole, still less does it express or demonstrate a choice “with reasonable certainty”. I conclude that the contract of insurance was governed by Italian law, and Evialis have not shown a good arguable case that it was not.

Italian Law

The evidence adduced by both parties included the Italian lawyers’ opinions as to how the contract of insurance should be interpreted. However, in argument both parties accepted that this evidence was not properly relied upon, and did not dispute the statement of principle found in Dicey and Morris (cit sup) at para. 9-019: “… the expert merely proves the foreign rules of construction, and the court itself, in light of these rules, determines the meaning of the documents”: see Rouyer Guillet & Cie v Rouyer Guillet & Co. Ltd, [1949] 1 All ER 244.

Thus, the evidence of Italian law properly went to three questions, namely (i) what were the proper principles of construction to be applied upon any questions of interpretation determined by the governing law; (ii) what documents comprised the contract of insurance, if Italian law is to be applied in considering this question; and (iii) how would the Italian courts regard an anti-suit injunction issued by this court so as to prohibit the insurers from pursuing the Italian proceedings.

Italian law: principles of contractual construction

Mr Morris says that the evidence of Mr Cigolini supports a further proposition. Mr Cigolini puts it in these terms:

Mr Bregante disputes this part of Mr Cigolini’s evidence. I cannot resolve this difference between the Italian lawyers, and need not finally do so. Mr. Cigolini’s opinion is clearly expressed, and on the face of it there is nothing improbable or remarkable about what he says, and no obvious reason to reject it. I accept that Evialis have a good arguable case that there is such a principle of Italian law.

Mr Bregante’s evidence is that Italian law would regard the certificate as an extension or application of the open cover and would not regard it as a separate contract. At least in the case of a “mandatory” or obligatory open cover such as this one, he says that the contract of insurance includes all terms and conditions of the cover and of the certificate of insurance. He states that the insurers’ liability to a holder of the certificate is determined by the certificate and “the terms contained, incorporated or referred to in the certificate of insurance (and obviously of the open cover incorporated therein)”.

Mr Cigolini does not, as I understand his evidence, dispute that as a matter of principle Italian law will adopt the approach described by Mr Bregante. However, he observes that it will depend on the particular case whether the parties intended terms from the open cover to be included in a particular insurance. He says that in his opinion, “… the relationship between an open cover policy and a certificate of insurance cannot be established in principle once and for all … The correct approach under Italian law would be to consider the terms of the specific open cover policy, and those of the specific certificate of insurance included.” Accordingly, while he acknowledges that the insurers and Ametra were bound by the open cover when insuring the goods on the “Irene”, he points out that, “when agreeing to issue the certificate of insurance for a specific transport, the parties were obviously free to agree new and different or more appropriate terms; and … only those terms contained, incorporated or referred to in the specific certificate of insurance are the terms which rule the relationship between the holder of the certificate of insurance issued “to order”, and cargo underwriters”.

I see no significant dispute between the expert witnesses on this point. The court will look at the open cover and the certificate, and form a view, as a matter of interpretation of the certificate, whether the parties agreed that the terms of the open cover, or some of them, should not apply to a particular contract of insurance.

The insurers rely upon the evidence of Prof. La China in support of their submission that it would be offensive to the Italian courts for this Court to make an order restraining the insurers from pursuing the proceedings there. He says that they would regard it as infringing the jurisdiction of the Italian courts. The evidence of Prof. La China is this:

This evidence is disputed, and not in accordance with that of Mr Cigolini. Moreover, Mr Morris submits that I should not give weight to it for two reasons.

First, he points out that Prof. La China does not cite judicial authority for his opinion, and says that this detracts from the value of his evidence. He drew my attention to the decision of Aikens J in Navigation Maritime Bulgare v Rustal Trading Ltd (The “Ivan Zagubanski”), [2002] 1 Ll L R 106, who was not persuaded by the opinions of French academic lawyers that an anti-suit injunction would be offensive to the French courts, observing that the opinions before them did not record the actual opinion of French judges or refer to any cases where those views have been expressed by French judges. Aikens J said that he would expect the French court to take the view referred to by Millett LJ in The “Angelic Grace”, [1995] 1 Ll L R 87 at p.96: “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”.

The second point made by Mr Morris is that the opinion of Prof. La China appears to be based upon his view that Italian courts would consider an order of the English court to be contrary to the Brussels Regulation.

I do not consider that these observations greatly detract from the weight to be given to the opinion expressed by Prof. La China. I do not find it surprising that he does not cite direct expressions of judicial opinion on a question such as this. The views of Aikens J and Millett LJ were expressed in straightforward cases of breach of an arbitration agreement. One might expect the foreign courts in those circumstances to be the less exercised about the breach being restrained. On any view the position about what the parties agreed about jurisdiction in this case is significantly more obscure.

It is true that Prof. La China’s opinion is based upon what the Italian courts would consider to be the proper interpretation of the Brussels Regulation. That does not detract from his point. Of course the Italian courts’ view about the proper ambit of their jurisdiction in a case such as this will reflect their understanding of the Regulation. The question which Prof. La China was considering was how, in a case which they regarded as being within the ambit of their jurisdiction, the Italian courts would regard an order from the English court restraining proceedings based not upon a jurisdiction deriving from the Regulation but from inherent powers which the English court regards as supplementing, or, as some might see it, departing from, that jurisdiction. The opinion of Prof. La China is that the Italian court might consider that the exercise of those powers would give rise to questions of public policy such as contemplated by article 34 of the Regulation: “A judgment shall not be recognised ... if such recognition is manifestly contrary to public policy in the Member State in which recognition is sought”.

Whatever may be the position when the English court makes an anti-suit injunction in such clear cases as Continental Bank v Aeakos [1994] 1 WLR 588, The “Angelic Grace”, (cit sup) and The “Ivan Zagubanski”, (cit sup), I accept, on the basis of the evidence adduced by the insurers, that there is a real risk that the Italian courts would regard an anti-suit injunction such as those sought by Evialis as an affront to the proper ambit of their jurisdiction.

Evialis argue that the service clause in the certificate is an agreement that the English court should have jurisdiction in respect of their proceedings against the insurers. If there is such an agreement, it would relate to a contract of insurance covering goods in transit, and so fall within article 13(5) of the Brussels Regulation: see article 14(1)(b); and accordingly it would not be deprived of legal force by article 23(5). There is no dispute that any such agreement would comply with the formalities of article 23. The dispute is whether there was any such agreement.

The question whether there was an agreement under article 23 is to be determined by community law, and not by the governing law of the contract of insurance: Powell Duffryn plc v Petereit, [1992] ECR 1-1745. It is well established that article 17 of the Brussels Convention imposes on the court the duty of examining “whether the clause conferring jurisdiction upon it was in fact the subject of a consensus between the parties, which must be clearly and precisely demonstrated”: Estasis Salotti di Clozani Aimo e Gianmario Colzani v RUWA Polstereimaschinen GmbH, [1976] ECR 1831 at para 7. Hence the approach of Saville LJ in the Credit Suisse Financial Products case (cit sup) to which I have referred. However, while the court will apply autonomous community law in order to decide whether article 23 of the Regulation is satisfied, in so far as the court needs to interpret the contract between the parties in order to do so, the court will, I take it, interpret it in accordance with its governing law.

I consider that it is highly likely that the service clause was incorporated into the contract between Evialis and the insurers. Indeed, I do not understand the insurers to dispute this. The question is whether, on its true construction, it provided for English jurisdiction for proceedings such as those brought by Evialis. My consideration of this question is not affected by the issue about the governing law of the contract, there being no evidence of relevant principles of interpretation under Italian law.

In interpreting the service clause, I can, in my judgment, have in mind as part of the background against which the certificate was issued that the insurance related to goods sold under a contract upon the terms of GAFTA 100, including clause 19 of those terms. This was at one point challenged by the insurers, but in the end Mr Males did not dispute Mr Morris’ submission that it is permissible to take this into account. It is part of the background known to, or presumed to be known to, both the parties. As for the insurers’ knowledge of this, their evidence was that they had no recollection of being aware before the certificate was issued that the underlying contract was subject to GAFTA terms, and that they assumed that the service clause was requested by Ametra “for the purpose of its documentary credit”. However, whether or not that is so, the insurers do not dispute Evialis’ evidence that Finargo were informed that the sale was subject to GAFTA terms. Finargo’s knowledge is to be attributed to the insurers for this purpose: firstly, they issued and signed the certificate for the insurers. Secondly, the Special Agreements provided that, “For the purposes of the terms stipulated by the insurance conditions, any communication made by the special transport agency [sc. Finargo], for and on behalf of the person insured, to the company, will be understood as having been made to the person insured. Similarly, any communication made by the person insured to the special transport agency will be understood as having been made to the company.”

There is no dispute that the service clause provides an address for the service of proceedings upon the insurers, and not an address for the service of proceedings on the assured or holder of the certificate. Evialis’ submission is that the service clause permits them to bring proceedings in England against the insurers. When they opened their case, they also argued that it provided for the English court to have exclusive jurisdiction, and at times during his oral submissions, Mr Morris also submitted that the provision (on its proper interpretation or by implication) precluded or restricted the insurers’ right to seek non-liability declarations. It is not entirely clear to me how far Mr Morris persists in these two arguments. However, I should consider three questions about the service clause: (i) whether it entitles Evialis to sue the insurers in England; (ii) whether it is in any sense an exclusive jurisdiction clause; and (iii) whether it applies to proceedings for non-liability declarations brought by the insurers against Evialis.

Evialis argue that the purpose of the service clause was to comply with article 19 of GAFTA 100, and the purpose of clause 19 of GAFTA 100 is that insurers who are not domiciled, and do not have a place of business, in the United Kingdom, should agree to be treated for the purpose of legal proceedings as if they had a domicile and an address for service here. It is submitted that the commercial purpose of this, and the parties’ intention, could only be that English legal proceedings might be served upon them.

The insurers dispute this. They submit that the clause is concerned only with where proceedings can be served, and that it is not directed specifically to English proceedings but to proceedings in any legitimate jurisdiction. They point to the literal wording of the provision, and also submit that in the context of this particular contract their interpretation attributes to the service clause a meaning that avoids any contradiction or tension between it and clause 16 of the General Conditions. Furthermore, they observe that the service clause does not meet the requirements of GAFTA 100 - it does not refer to domicile, and the insurers did not purport to accept a British domicile when they issued the certificate – and suggest that this undermines Evialis’ attempt to rely upon GAFTA 100 as an aid to construction of the service clause.

A literal reading of the service clause does, at first blush, lend some support to the insurers’ submission, but there are two possible responses to this point. First, the insurers’ argument assumes that the words “in UK” qualify the word “address”. The words can read as qualifying “legal proceedings”, and this interpretation might be suggested both by their position and by the fact that as a qualification to “address” they would be tautologous. Thus, it can be argued that the wording of the service clause itself suggests that the parties contemplated proceedings in the United Kingdom.

Secondly and in any event, Evialis say that the insurers’ interpretation lacks any realistic business sense in that it supposes that the parties contemplated that the insurers should provide an address for serving foreign proceedings in London; and so they argue that the service clause should give a more robust and commercially probable interpretation. Accordingly, they say that the clause evinces an intention that the insurers might be sued in England. I consider that there is much force in Evialis’ argument, and consider their interpretation to be correct. At all events, I conclude that they have a good arguable case (in the sense that I shall explain) that by the service clause the parties agreed that the assured or holder of the certificate might sue the insurers in England. I add, out of deference to the insurers’ observation that the service clause departs from what was contemplated by GAFTA 100, that these conclusions do not rely upon this background to the provision in the certificate.

I turn to the suggestion that the service clause is properly interpreted as conferring exclusive jurisdiction on the English courts in respect of claims against the insurers because the GAFTA clause contemplated that the insurers should accept a British domicile. There are a number of answers to this. First, in this respect the service clause itself does not reflect the GAFTA terms. Secondly, under the Brussels Regulation jurisdiction does not depend solely on the insurers’ domicile. Thirdly, for these purposes a corporation can have its seat, and so its domicile, in more than one Member State: section 42 of the Civil Jurisdiction and Judgments Act, 1982. I do not accept that the service clause confers any exclusive jurisdiction on the English courts. I add that Mr Morris accepted (and was right to accept) that when article 23 of the Brussels Regulation provides that, “Such jurisdiction shall be exclusive unless the parties have agreed otherwise”, it contemplates that the parties’ agreement might be either express or a matter of implication or inference.

As I have said, Mr Morris also argued at one point that the proper interpretation of the service clause was that it prohibits the insurers from bringing proceedings for non-liability declarations in Italy. I find this argument difficult to understand. Proceedings for non-liability declarations are not unusual either in this country or, as I understand it, in other member states. Mr Morris does not, I think, suggest that the service clause prohibits the insurers altogether from seeking such relief. Certainly I would regard it as unarguable that it does. Nor does it allow them to bring such proceedings in England. But whatever the effect of article 16 of the General Conditions, the insurers would have been entitled to seek such relief in France under article 12 of the Brussels Convention. There is nothing in the service clause that prevents them from doing so.

The burden is upon Evialis to show a good arguable case that the court has such jurisdiction to hear their claim: Canada Trust Co v Stolzenberg (No 2), [1998] 1 WLR 547, [2002] 1 AC 1. The term “good arguable case” is flexible, and I do not find it altogether easy to discern from the authorities quite what it connotes where, as here, the question is whether there is an effective jurisdiction clause (or that one has been clearly and precisely demonstrated). Of course, the view that the court forms upon a hearing of this kind might be in one sense the more tentative because there is not a full trial of the issue, but there remains the question how confident of its conclusion the court has to be, on such material as is before it, if it is to find that a good arguable case has been made out. On the one hand it has been said that the court must form the view that it is “highly likely” that if the matter were tried out the claimant would succeed in his argument on the jurisdiction clause (I P Metal Ltd v Ruote OZ SpA, [1993] 2 Ll L R 60 at p.63, Canada Trust Co v Stolzenberg (No 2), [1998] 1 WLR 547 at p.559D). On the other hand, it has been said that the test “requires less than proof on the balance of probabilities, but must nevertheless be strong enough to justify retaining the action here rather than declining jurisdiction in favour of the Court first seised in accordance with article 21 [of the Brussels Convention]”: Glencore International AG v Metro Trading International Inc, [1999] 2 Ll L R 632, 642. For my part, I consider that “highly likely” remains a proper elucidation of the test in this context, for the reasons explained by Waller J in the I P Metal Ltd. (cit sup).

It follows from my conclusion about the meaning of the service clause that I consider that Evialis have shown a good arguable case (in the sense that I have described) that the English court has jurisdiction under article 23 of the Brussels Regulation to hear their claim, and I reject the insurers’ application for an order declaring that the court has no jurisdiction, and that the claim form and its service be set aside.

The insurers’ alternative application is that the proceedings should be stayed under article 27 on the basis that the English court was seised of these proceedings (in the terms of article 30 of the Regulation) after the Italian court was seised of the Italian proceedings. Article 27 provides:

There is no dispute that the English and Italian proceedings are between the same parties. However, in response to this application Evialis raise three arguments that the insurers should not be granted a stay:

that the proceedings in England do not involve the same cause of action (within the meaning of article 27) as the Italian proceedings.

That, even if the English and Italian proceedings involve the same cause of action (so that, the Italian court being first seised, prima facie article 27 would apply), no order should be made because of the jurisdiction provision in the service clause.

That, even if the English and Italian proceedings involve the same cause of action and proceedings in Italy were brought first (so that prima facie article 27 would apply), no order should be made in view of the “quasi-exclusive” regime established in respect of matters concerning insurance and in view of article 35 of the Brussels Regulation.

The last two of these arguments proceed on the basis that the contract of insurance does not include a provision that gives the insurers the right to bring the Italian proceedings. I shall deal with this next. I do so, although the insurers submit that it is not necessary to do so, and that the dispute about this should be left for determination in the Italian proceedings. There is much force in this submission, and I agree that the Italian court is in a far better position than I am to decide these questions of Italian law and interpretation of a contract written in the Italian language. However, I should, I think, enter some way upon consideration of this matter in order to evaluate Evialis’ arguments against their proceedings being stayed and to examine their contention that the insurers have brought the Italian proceedings in a court that has no jurisdiction over them.

The insurers assert that the Italian courts have jurisdiction because of the agreement between the parties contained in article 16 of the General Conditions, and under article 23 of the Brussels Regulation. That argument is, like the question whether the English court has jurisdiction to entertain Evialis’ proceedings because of the agreement in the service clause, to be answered by applying community law, but any question of interpretation of the contract is to be determined by its proper law, the law of Italy.

As I see it, the question whether the parties to the contract of insurance reached a consensus that there should be Italian jurisdiction with sufficient clarity and precision to satisfy article 23 turns upon two main questions: what does clause 16 of the General Conditions mean? And was the clause a term of the contract of insurance?

Evialis suggest that clause 16 does no more than stipulate what is the appropriate forum within Italy for any Italian proceedings properly brought. Whether they are right about this depends upon the meaning of a contract that is not only governed by Italian law but in the Italian language. I look at the contract in a translation of what I am told are the relevant parts. I am not much assisted, I think, by the principles of Italian law about construing contracts that have been identified. I am not able to resort to the principle of interpretation contra proferentem in the absence of satisfactory evidence about who is to be regarded as the party putting forward the contract. My own impression of the words of clause 16 is that Evialis’ interpretation of them is unduly narrow, and the connotation of the reference to an exclusively competent court is that the parties were agreeing to the jurisdiction of the Italian courts

I have already concluded that Evialis have established a good arguable case that it is a principle of Italian law that jurisdiction agreements can be incorporated into contracts from another document only by specific reference and not by general reference. I am sceptical as to whether this principle assists Evialis in this case. The argument that clause 16 consensually confers jurisdiction of the Italian courts is not really one that it was incorporated into the certificate, but that the parties, having agreed upon Italian jurisdiction in the open cover, did not evince in the certificate an intention to set that agreement aside. If I understand Mr Cigolini correctly, that is indeed how he would expect the Italian court to go about investigating the contract between the parties.

If this approach is adopted, Evialis are able to argue that, if it is right to interpret the service clause as an agreement that the insurers may be sued in England, the terms of the certificate are inconsistent with a provision for exclusive Italian jurisdiction, and that the parties therefore cannot have intended that clause 16 should be a term in the contract of the insurance of this cargo. This argument undoubtedly has force as far as it goes, but I do not consider it goes far enough to answer the insurers’ contention. Certainly, if the service clause is to be given the effect that I have concluded it should have, then clause 16 cannot be a term of the contract without modification. However, there is no need to regard the service clause as doing more by way of modifying the agreement as to jurisdiction in the open cover than to provide the assured with a further option about where to proceed against the insurers.

I bear in mind that the insurers must, in order to establish that they can rely upon article 23 of the Brussels Regulation because of the parties’ agreement, demonstrate the agreement with the required precision and clarity. Nevertheless, my own tentative conclusion is that the insurers are able to do so, and, if it were necessary, I would be driven to conclude that Evialis do not have a good arguable case that the Italian proceedings contravene the Brussels Regulation. However, for the reasons that I have explained, I have been reluctant to embark upon this enquiry which the Italian courts are better able to conduct, and am able to base my conclusions on these applications on other grounds.

The insurers’ argument that the English and the Italian proceedings involve the same cause of action is simple. They refer to the judgment of Rix J in Glencore International A.G. v Shell International Trading and Shipping Co. Ltd., [1999] 2 Lloyds LR 692. That case involved consideration of Article 21 of the Brussels Convention. Rix J pointed out that the article laid down a triple requirement of the same parties, the same cause and the same “objet” (the last two requirements reflecting the French – and other – versions of the Convention). He said:

Thus, the insurers submit that when the central or essential issue in the proceedings here and in Italy is identified, in both it is found to be the liability of the insurers under the contract of insurance in respect of the damage to the goods on the “Irene”, and that, looked at realistically, the purpose of the claims about jurisdiction is to support and supplement the substantive claims.

However, Evialis argue that the English proceedings, being concerned to prevent the Italian proceedings, involve a different cause of action, relying upon three decisions at first instance: Continental Bank NA v Aeakos Compania Naviera SA, see [1994] 1 WLR 588 at p.595H, Toepfer International GmbH v Molino Boschi Srl, [1996] 1 Ll L R 510 and Toepfer International GmbH v Societe Cargill France, [1997] 2 Ll L R 98.

In each of these cases the only relief sought by the claimant in the English proceedings was to give effect to a jurisdiction clause, and there was no claim for substantive relief in the underlying dispute. In the Continental Bank case, the Bank claimed an injunction against the defendants to prevent them from pursuing Greek proceedings. At first instance Gatehouse J said that in his judgment “the defendants’ action in the Athens court and the plaintiff’s action in this court do not involve the same causes of action for the purpose of article 21, as interpreted in Gubisch Maschinenfabrik KG v Palumbo (Case 144/86) [1987] ECR 4861.” The Court of Appeal did not consider this point, deciding the case on the basis of the jurisdiction clause.

In Toepfer v Molino, the defendants had brought proceedings in Italy. The English claimants sought a declaration that the defendants were obliged to refer the dispute that was the subject of the proceedings to GAFTA arbitration, and an injunction to restrain the defendants from pursuing the Italian proceedings and ordering their discontinuance. The defendants applied to have the English proceedings dismissed or stayed under articles 21 and 22 of the Brussels Convention. Mance J said, in relation to the application under article 21, “... it is necessary to bear in mind possible differences between the claims for declaratory and injunctive relief” (p.513). Of the claim for an injunction, he said that it did not have the same object as the Italian proceedings. He considered that to be the case notwithstanding that the English claimants had raised in Italy a preliminary objection that the claims should be arbitrated, regarding that as, at most, only one aspect of the Italian proceedings. Of the claim for declaratory relief, Mance J said that although the object of the claim was less clear-cut, it was different from that of the Italian proceedings, and concluded that “The cause of action …is … not the same for reasons similar to those identified in considering the claims to injunctions” (p.513).

In Toepfer v Cargill, the claimants sought a declaration that certain disputes fell within a GAFTA arbitration clause, that the defendants were obliged to refer them to arbitration, and that proceedings in respect of them that the defendants had brought in France were in breach of the clause; and an injunction restraining the defendants from pursuing the French proceedings and ordering their discontinuance. It was argued on behalf of the claimants that the English proceedings and the French proceedings involved the same cause of action within article 21 of the Brussels Convention. Colman J decided that article 1(4) excluded the proceedings from the operation of the Convention. Accordingly, he did not need to decide the point about article 21, and did not hear full argument upon it. Nevertheless he did express the provisional view that the reasoning of Mance J in Toepfer v Molino was cogent and compelling, and he specifically said that he found compelling the view of Mance J that it was necessary to compare the object or cause of action in the English proceedings with the object or cause of action in the substantive proceedings in France and not with the issues raised by Toepfer’s jurisdictional objection raised in those proceedings.

The case went to the Court of Appeal, who referred two questions to the European Court for rulings: [1998] 1 WLR 379: Of the question about the causes of action within the meaning of article 21, Phillips LJ, giving the judgment of the Court of Appeal, said this (at p.387):

However, I do not consider that it would be right for me to treat these three first instance decisions as wrong. The question is whether they apply in this case. In none of these three cases did the claimant seek any relief in the English proceedings other than declarations and injunctions concerning the proper forum for the resolution of the underlying claim. No relief was sought in respect of the underlying claim. The insurers submit that it is a significant distinction from these cases that here Evialis are seeking an indemnity for their loss. They cite Dicey and Morris (cit sup) at para 12-058, where the view is expressed that where the English court is seised with only a claim for a final anti-suit injunction and no other relief on the merits of the claim, it is improbable that the two causes of action are the same, “but where the injunction is sought within the context of proceedings for substantive relief, the application of article 21 may be inescapable”.

For my part, I do not understand Dicey and Morris to be considering the position where the claimant in the English court is seeking both substantive relief on the underlying claim and an anti-suit injunction, but rather to be contemplating a claimant for substantive relief seeking an interim anti-suit injunction in support of his substantive claim on the merits. However, I do accept the insurers’ submission, that if the proper approach is to look at the English proceeding as a whole, to ask what is the central or essential issue and to consider whether that is the same as that in the Italian proceedings, the issue in both proceedings is the insurers’ liability in respect of the cargo damage. I therefore reject Evialis’ first argument against a stay of these proceedings.

Evialis contend that, even if the English and Italian proceedings involve the same cause of action, nevertheless no order should be made under article 27 (or 28) because of the jurisdiction clause whereby the parties agreed English jurisdiction. They rely upon Continental Bank v Aeakos (cit sup), in which the Court of Appeal concluded that since the English court had jurisdiction by reason of the parties’ agreement upon jurisdiction, the Greek court, though first seised, did not have jurisdiction, and that it was open to the English court to restrain proceedings in Greece by an anti-suit injunction.

The decision in Continental Bank is controversial: Toepfer v Cargill, [1998] 1 Ll L R 379 at p.386, Briggs and Rees on Civil Jurisdiction and Judgments (3^rd Ed.) at para 2.197. Not least of the criticisms is that no mention is made in the judgment of the decision of the European Court in Overseas Union Insurance Ltd v New Hampshire Insurance Co, [1992] QB 434, which, as is observed by Briggs and Rees, “makes evaluation of the reasoning rather problematic”. However, Continental Bank is binding upon me and unless it can properly be distinguished, I must apply it in this case. Mr Males seeks to distinguish it on two grounds: firstly because the governing regime in this case is the Brussels Regulation and not the Brussels Convention; and secondly because, he submits, on the facts of this case Continental Bank does not apply.

The insurers submit that it is significant that this case is to be decided under the Brussels Regulation because the Regulation contains recitals, which do not correspond to provisions in the Convention, and which emphasise that the rules governing jurisdiction are governed solely by the Regulation. The Regulation contains an effective mechanism for resolving cases of lis pendens, and is to be interpreted teleologically. Accordingly, while it is, of course, recognised that the articles of the Regulation largely correspond linguistically to those of the Convention, the recitals provide important guidance upon their interpretation.

On this basis, the insurers argue that the following recitals to the Regulation are indicative that it should not be so interpreted as to allow the English court to make orders such as were made in Continental Bank:

Recital 2, which reads:

Recital 6, which reads:

Recital 11, which reads:

Accordingly, while I accept Mr Males’ submission that it is right in interpreting the Regulation to have regard to its purposes and aims as set out in the recitals and to seek to give effect to them, I do not consider that they differ significantly from those of the Convention, under which regime the Continental Bank case was decided, or that a proper regard for the recitals upon which Mr Males relies undermines the reasoning that led the Court of Appeal to its decision.

Moreover, the insurers’ argument ignores recital (14) to the Regulation, which provides:

I reject the submission that the fact that this case is governed by the Regulation and not the Convention is a proper basis for distinguishing this case from Continental Bank, which therefore I regard as a binding precedent upon me in this case. Certainly its authority is such that I decline not to follow it.

Mr Males’ alternative submission is that this case is to be distinguished on its facts from Continental Bank, and other cases in which Continental Bank has been applied. There are three distinctions: (i) that the service clause is not an agreement for exclusive jurisdiction and does not prohibit the Italian proceedings; (ii) that the contract is not governed by English law; and (iii) that there is not a clear breach of the choice of forum clause.

The first point is based upon the interpretation of the service clause that I accept. I have concluded that it does not apply to proceedings by the insurers for non-liability declarations, and that it does not provide for the exclusive jurisdiction of the English court. The appropriate forum for the insurers’ proceedings for non-liability declarations was, when they brought the proceedings, either Italy, if article 16 of the General Conditions is effective and applicable to provide for Italian jurisdiction, or France under article 12 of the Brussels Regulation. (They could now bring a counterclaim in these proceedings under article 12(2), but Mr Morris did not rely upon this provision.)

Undeniably, it would be an extension of the decision in Continental Bank to apply it in these circumstances. However, the basis of the decision was respect for the parties’ autonomy, and Evialis submit that, assuming that the parties did not agree that the Italian courts should decide such claims as those in the Italian proceedings, the Court should respect the parties’ autonomy by rejecting the insurers’ application for a stay of these proceedings. I have already expressed the tentative conclusion that the assumption is wrong, but even assuming it to be correct, I reject Evialis’ submission.

Their argument is this: if the Italian jurisdiction provision is not valid and effective, it is a breach of the contract of insurance for the insurers to frustrate Evialis’ right, conferred by the service clause, to sue the insurers in this country by bringing proceedings which they are not entitled, under the Regulation, to bring. Mr Morris argues that the jurisdiction of the English court is exclusive, not in the sense that it is the only court that can properly hear the dispute about the indemnity claims (Evialis being entitled to sue in France or Italy, and the insurers being entitled to sue in France), but in the sense that the agreement conferred upon Evialis the right to sue the insurers in England and that the implication is that the insurers would not defeat that right by suing Evialis in an illegitimate jurisdiction.

In support of this argument, Evialis cite the decision of the Court of Appeal in Sabah Shipyard (Pakistan) Ltd v Islamic Republic of Pakistan, [2002] EWCA Civ 164. In that case, the parties to a guarantee payable on demand had entered into a non-exclusive English jurisdiction agreement, but the guarantor, the State of Pakistan, brought proceedings in Pakistan permanently to restrain any demand under the guarantee and so, effectively, to prevent any claim under it in England or elsewhere. The English court granted an injunction to restrain the proceedings in Pakistan. At para 38, Waller LJ said this:

I cannot accept Evialis’ argument that the insurers are in breach of an implied terms of the contract of insurance in bringing the Italian proceedings. The right that the parties agreed to confer upon Evialis was always subject to the possibility that the insurers would bring proceedings against them for non-liability declarations in another jurisdiction. The most that Evialis could properly have expected is that if the insurers did bring proceedings in another Member State, the machinery of the Brussels Regulation, and in particular articles 27 and 28, would be available to ensure that the proceedings were not brought illegitimately. The starting point of the Regulation is that the court first seised should decide whether it has jurisdiction over the dispute, and if so should proceed to adjudicate it. I decline to infer that the parties intended to go further and contemplated that the English courts should intervene outside the regime of the Regulation.

I add that Evialis’ submission was, as I have said, based upon the Sabah Shipyard case in which the parties’ contract was governed by English law, and was argued on the basis of English law about the implication of contractual terms. This argument had not been foreshadowed when the evidence of Italian law was prepared, and is not dealt with by the Italian lawyers. Since I have concluded that the contract of insurance is governed by Italian law, I should, in these circumstances, have felt uncomfortable about refusing a stay on the basis of a term that English law would imply, but in view of my conclusion about the position under English law, this difficulty does not arise.

In view of these conclusions, it is not necessary for me to consider Mr Males further arguments that Continental Bank has no application unless (i) the agreement as to forum is governed by English law, and (ii) there is a clear breach of the agreement as to forum. It is certainly the case that both Continental Bank and The “Angelic Grace”, [1995] 1 Ll L R 767, in which the Court of Appeal robustly adopted the same approach, were straightforward cases in which the relevant contracts were governed by English law. In Turner v Grovit, (cit sup at para 40) Lord Hobhouse mentioned that in the contract in the Continental Bank case the parties had expressly chosen English law. Mr Males also draws to my attention National Westminster Bank v Utrecht-America Finance Company, [2001] EWCA 658, [2001] 2 AER (Comm Cas) 7, in which the Court of Appeal expressed the view that it would not be appropriate to grant an interim injunction to restrain foreign (Californian) proceedings when it was no more than arguable that they were brought in breach of contract. However, while these are considerations that the court will weigh in deciding whether to grant an injunction or other discretionary relief, it does not seem to me that they are decisive considerations when the court second seised is deciding whether the defendant is entitled to a stay under article 27 of the Brussels Regulation.

Mr Morris’ third argument is that the claims in these proceedings are governed by the special jurisdictional rules for insurance in part II, section 3 of the Brussels Regulation, and that, just as with cases under the exclusive jurisdiction of article 22, so too in cases that fall under the “quasi-exclusive jurisdiction”, articles 27 and 28 are inapplicable. Both where there is exclusive jurisdiction and where there is “quasi-exclusive jurisdiction” under section 3 (and, in the case of consumer contracts, under section 4 of part II of the Regulation), a judgment is not to be recognised if it conflicts with the governing jurisdiction provisions, in contrast with the general rule under the Regulation that the jurisdiction of the court of the Member State of origin may not be reviewed: article 35. Therefore, it is submitted, it cannot be said in these cases that it is objectionable in principle for the court second seised to examine the jurisdiction of the court first seised. The Regulation contemplates that it should do so, and accordingly the English court should not decline jurisdiction or stay these proceedings. At least it should not do so without first examining the jurisdiction of the Italian court to entertain the Italian proceedings and being satisfied that the Italian proceedings comply with section 3.

The term “quasi-exclusive” that Mr Morris uses in advancing this argument is that of Briggs and Rees (cit sup) at para 2-197. In Overseas Union Insurance Ltd v New Hampshire Insurance Co, [1992] QB 434, the European Court, responding to the second and third questions, which concerned whether under article 21 of the Convention the court second seised was in all circumstances obliged to stay its proceedings as an alternative to declining jurisdiction, or whether it might examine the jurisdiction of the court first seised, gave the answer that the only alternative to declining jurisdiction for the court second seised was to stay its proceedings “without prejudice to the case where the court second seised has exclusive jurisdiction under the Convention and in particular under article 16 thereof”: see paragraph 26. In reaching this conclusion, the Court said this (at para 24): “Moreover, the case[s] in which a court in a contracting state may review the jurisdiction of a court in another contracting state are set out exhaustively in article 28 and the second paragraph of article 34 of the Convention [broadly the equivalents of article 35 and article 45(1) of the Regulation]. Those cases are limited to the stage of recognition or enforcement and relate only to certain rules of special or exclusive jurisdiction having a mandatory or public policy nature. It follows that, apart from those limited exceptions, the Convention does not authorise the jurisdiction of a court to be reviewed by a court in another contracting state”. Briggs and Rees comment (at para 2-187 fn 999) that the principle that judgments given by courts which contravene article 22 of the Regulation (equivalent to article 16 of the Convention) must be denied recognition “may extend also to cover jurisdiction under the insurance and consumer contract cases, where the judgment of the court must be denied recognition if it breached those jurisdictional rules; these may be seen as cases of quasi-exclusive jurisdiction”.

Similarly, Dicey and Morris (cit sup) at para 12-043 expresses the view that the reasoning whereby the European Court exceptionally entertained the possibility of the court second seised examining the jurisdiction of the court first seised in cases of exclusive jurisdiction, would justify extending the exception to other cases in which the judgment of the court first seised will not be recognised without examination of its jurisdiction, including insurance contracts covered by section 3.

I do not find support for this argument in the Overseas Union case. First, in its answer to the second and third questions, the European Court left open in cases of exclusive jurisdiction the question of the court second seised examining the jurisdiction of the court first seised when considering an application under article 21 of the Convention (or article 27 of the Regulation). It did not decide that even in cases of exclusive jurisdiction the court second seised should or might do so.

Secondly, the exception was left open only in cases of “exclusive” jurisdiction. The Court recognised that under the Convention (like the Regulation) the court in a contracting state will review the jurisdiction of another contracting state not only in cases of exclusive jurisdiction but also in cases of “special” jurisdiction (which I understand to include cases of insurance contracts), and stated that these cases, which are exhaustively stated at articles 28 and 34, are limited to the stage of recognition or enforcement. There is nothing in the judgment to suggest that the exception in the case of special jurisdictions arises at an earlier stage of the proceedings. The case was not concerned with the stage of recognition or enforcement, and so was not concerned with whether the case was one of special jurisdiction. It was simply observed (at para 20) that “nothing in the documents before the court suggests that the main proceedings fall within an exclusive head of jurisdiction laid down in the Convention, in particular in article 16 thereof”.

Evialis’ argument would involve supposing that when in paragraph 26 the Court uses the term “exclusive jurisdiction”, it is referring not only to what is described in the Convention as exclusive jurisdiction, but to other cases in which the jurisdiction of the court first seised will be examined at the recognition and enforcement stage, including contracts of insurance and consumer contracts. I cannot so read the judgment. The Court uses the term “exclusive jurisdiction” precisely and consistently to refer to cases covered by article 16 of the Convention. When referring to the ambit of article 28 of the Convention, it uses the term “special or exclusive jurisdiction”.

It is suggested that, if the court second seised is to review the jurisdiction of the court first seised at the stage of recognition or enforcement, there can be no objection to it reviewing it earlier. I cannot accept that. After all, the freedom of review at the recognition stage is limited by article 35(2): “In its examination of the grounds of jurisdiction referred to in [article 35(1), which provides that a judgment shall not be recognised if it conflicts with, inter alia, section 3], the court or authority applied to shall be bound by the findings of fact on which the court of the Member State of origin based its jurisdiction”.

It is also significant that, whereas the Regulation expressly provides for the possibility of actions coming within the exclusive jurisdiction of several courts (article 29: “Where actions come within the exclusive jurisdiction of several courts, any court other than the court first seised shall decline jurisdiction in favour of that court”), it does not deal specifically with cases within the special jurisdiction of several courts. The inference to be drawn, it seems to me, is that this situation is governed by articles 27 and 28.

No English authority has been drawn to my attention in which Evialis’ submission that article 27 does not apply to cases of special jurisdiction was considered, but if the argument is sound, it was overlooked in such cases as Denby v Hellenic Mediterranean Lines Co Ltd, [1994] 1 Ll L R 320 and Tradigrain SA v SIAT SpA, (cit sup). More significantly, Mr Males argues that if the submission is sound, the point was overlooked in the Overseas Union case. The court described as “redundant” the fourth question, which was directed to whether it made a difference that the contract was one of reinsurance of a quota share and whether such a reinsurance was within section 3. I accept Mr Males’ submission that the implication is that the court did not consider that the fact that a contract is governed by the section 3 regime is a passport for the court seised second to examine the jurisdiction of the court first seised.

I reject Evialis’ argument that because the claim is governed by section 3, a stay should not be ordered under article 27.

I conclude that none of the three arguments adduced by Evialis against a stay of their indemnity claims is sound.

However, Evialis have a further argument. They submit that, even if their indemnity claims are to be stayed under article 27, their jurisdiction claims should not be. They say that in these proceedings they pursue more than one cause of action within the meaning of the term in article 27, and that their jurisdiction claims, which do not involve the same cause of action as any claim in the Italian proceedings, comprise anything between one and four separate causes of action. This is because these claims have do not have the same “cause” and “objet” as the indemnity claims. Mr Morris developed his submission about the causes of action in these proceedings as follows:

I agree that, if the proper approach to article 27 is to distinguish different claims within proceedings so as to identify distinct causes of action, it is natural to adopt as the criterion for doing so that of whether claims have the same “cause” and “objet”, the criterion from other language versions of the Regulation for whether the proceedings in two jurisdictions involve the same cause of action. However, Mr Morris’ submission demonstrates the complications that will frequently arise if fragmentation of this kind is permitted. I am unable to accept that this is what article 27 requires.

This submission raises a number of questions about the proper approach to article 27, but it is convenient first to consider where it leads. If I accepted Evialis’ submissions, I should stay the indemnity claims in the English proceedings. Evialis would, nevertheless, invite me to order the insurers not to pursue the Italian proceedings on the underlying insurance claim. If I did so, and assuming the insurers obeyed the order, neither party would be able immediately to pursue proceedings to resolve the indemnity claim. I acknowledge that the impasse might be resolved by Evialis pursuing the jurisdiction claims. This might resolve it in one of two ways. One would be for Evialis to be granted a final injunction whereby the insurers are ordered to discontinue the Italian proceedings. It is to be assumed that the order would be observed and the proceedings discontinued, so that proceedings would no longer be pending in Italy and the stay on the indemnity claims would be lifted. Alternatively, if the English court hearing the jurisdiction claims held it a breach of contract for the insurers to have brought the Italian proceedings, the resultant stay of the indemnity proceedings might be said to have caused damages in that the indemnity proceedings could not be pursued, and their value might be recovered as damages for breach of the jurisdiction agreement. (I pass over the question whether in the latter eventuality the damages would be measured by the loss of a chance,) I consider it unlikely that it is the intention of the Brussels Regulation that a perfectly ordinary claim under an insurance contract should be trammelled in such complexities.

The insurers submit that in any event, if I were to conclude that the indemnity claims but not the jurisdiction claims should be stayed under article 27, I should stay the jurisdiction claims under article 28. It is convenient to consider this submission before Evialis’ argument that the jurisdiction claims should not be stayed under article 27 .

Article 28 of the Regulation provides that where “related actions” are pending in the courts of different Member States, any court other than the court first seised may stay its proceedings. Article 28(3) provides that, “For the purpose of this Article, actions are deemed to be related where they are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings”.

Mr Morris argued that the jurisdiction claims in the English proceedings are not related to claims in the Italian proceedings. He cited the judgment of Mance J in Toepfer v Molino (cit sup) at p.513-514. In that case, Mance J was considering Toepfer’s claims under an originating summons for a final injunction and declaratory relief and Molino’s applications under articles 21 and 22 of the Brussels Convention. In this case, although Evialis claims a final injunction in the proceedings, their application before me is for an interlocutory injunction.

Mance J considered that Toepfer’s claim for declaratory relief was related to Italian proceedings brought by Molino since both proceedings were concerned with whether claims raised in those Italian proceedings should be determined in London arbitration, that being one of the defences raised by Toepfer in the Italian proceedings. However, he considered that the claim for an injunction was not related to the Italian proceedings for three reasons: first because its purpose was to stop the Italian proceedings from continuing, and therefore it would defeat the purpose, and be inexpedient, for the claim to be heard with the proceedings that they were designed to stop. Secondly, it was to be assumed that if an injunction was granted, it would be obeyed and the Italian proceedings would not proceed; and “If no injunction is granted, caedit quaestio. Either way, the problem of irreconcilable judgments should not arise”. Thirdly, there was every reason to suppose that the Italian judgment would not be recognised in this country because of article 27 of the Convention.

In these proceedings, Mr Morris puts forward similar reasons that the jurisdiction claims in these proceedings are not related to the Italian proceedings. However, I do not find them convincing as providing an answer in this case. As Mance J recognised, and as is established by The Tatry, [1999] QB 515 (at para 53), the interpretation of “related actions” is to be a broad one, to cover all cases where there is a risk of conflicting decisions even if the judgments can be separately enforced and their legal consequences are not mutually exclusive. A broad commonsense approach is to be adopted, and over-sophisticated analysis is discouraged: Sarrio SA v Kuwait Investment Authority, [1999] AC 32 at p.41F.

Turning to the three reasons that Mance J gave in Toepfer v Molino, it is to be noted that the first applied only to the claims for injunctive relief, the fourth of the jurisdictional causes of action that Mr Morris tentatively identified. The same is true of Mance J’s second reason. Moreover, if in this case no injunctive relief is granted, there seems to me to remain a risk of inconsistent judgments, inconsistency for example as to the governing law of the contract of insurance. As for Mance J’s third reason, even if the judgment in the Italian proceedings is not recognised in England, it does not promote the good administration of justice that inconsistent judgments are reached. Moreover if they are inconsistent on questions of fact, the inconsistency would outflank article 35(2).

I conclude that these proceedings and the Italian proceedings are “related actions” within the meaning of the term in article 28.

I revert to the question whether Evialis are correct in their submission that the jurisdiction claims are to be separated from the indemnity claims for the purposes of article 27, and that, even if the indemnity claim is stayed, the jurisdiction claims should not be. This raises the following issues: (i) whether in applying article 27 the court should consider separately different causes of action in the proceedings; (ii) if so, whether the jurisdiction claims constitute “causes of action” within the meaning of article 27; and (iii) if so, whether the Italian proceedings involve the same cause(s) of action as jurisdiction claims in these proceedings.

In the English version of the Regulation, the provisions of articles 27 and 28 refer to “actions” and “proceedings”, and on the face of it do not contemplate parts of actions or proceedings being stayed. I acknowledge that in The Tatry, [1999] QB 515 at paras 34-36, it was held that proceedings might be stayed against some parties and allowed to continue against others. It is to be noted, however, that the European Court described the fragmenting of proceedings even to this extent as a “disadvantage”, and there is nothing to suggest that the Court contemplated further fragmentation of the kind that Evialis seek. I consider that this would be contrary to both the wording and the purpose of the Brussels Regulation, for the reasons explained by Saville LJ in The “Happy Fellow”, [1998] 1 Ll L R 13 at pp.17/18:

Mr Morris submitted that in Toepfer v Molino (cit sup) Mance J gave separate consideration to the claim for an injunction and that for a declaration, and cited this judgment as authority for his argument that the court should give separate consideration to the different causes of action in these proceedings, and decide of each whether or not the same cause of action is involved in the Italian proceedings. I am not entirely convinced that, when he was considering the application before him under article 21, Mance J was contemplating staying part of the action, although I acknowledge that appears to have been his approach when considering the application under article 22. In any event, he does not consider in his judgment whether or not the Convention contemplates fragmenting proceedings in this way, and I draw the inference the question was not argued before him.

I add that in Grupo Torras v SA v Al-Sabah, [1995] 1 Ll L R 374 at p.419, which Mance J decided before the judgment of the European Court in The Tatry (cit sup) and his own decision in Toepfer v Molino (cit sup), he said that articles 21 and 22 of the Convention required “an overall comparison between competing or related proceedings in different States”, and referred to “the undesirable practical implications…in tending to split up litigation between different States and causing inappropriate complexity”. He did not exclude the possibility that “an action might in exceptional circumstances fall to be regarded as containing separate proceedings...; or that nominally separate proceedings might in some circumstances fall to be regarded as if they were a single set of proceedings”. But it is clear that he thought that would be the proper approach only in unusual circumstances.

I reject Mr Morris’ submission that in a case such as this it is right to look at separate causes of action within the proceedings when determining an application under article 27. Like Mance J, I do not rule out entirely the possibility that in unusual circumstances this might be proper. But there are no features of this case that justify it.

Having reached this conclusion, I do not propose to consider the question whether in any event it is impermissible to characterise a jurisdictional dispute such as this as a “cause of action” within the meaning of the term in the Regulation. However, the facts of this case perhaps reinforce the reservations about this expressed by Phillips LJ in Toepfer v Cargill, (cit sup) which I have already set out.

The insurers have another answer to this argument. Evialis’ argument is based upon the premise that the Italian court became seised of the jurisdiction issues after these proceedings were brought. This depends upon whether it is right to say that the Italian courts became seised of proceedings involving the jurisdictional causes of action only when Evialis filed their defence on 23 December 2002 and challenged the jurisdiction of the Tribunale di Genova.

I shall assume in favour of Evialis that, if the court were to embark upon an examination of the application of article 27 to different causes of action within a set of proceedings, it would not take for all the causes of action a single date of first seisure of a set of proceedings, although, as Moore-Bick J observed in Glencore International AG v Metro Trading International Inc, (cit sup) at p.637, that course has much to commend it in terms of practical consequences. However, Brooke LJ in Fox v Taher, [1997] I L Pr 441 accepted the submission that “the answer to the question “when did the relevant lis between the relevant parties become definitively pending?” will depend upon which lis one is asking about”. This view was not espoused by the majority of the Court of Appeal (Morritt and Leggatt LJJ). However, I was not addressed upon these authorities, and I shall, as I say, assume in favour of Evialis that this is the approach to be adopted.

When, then, did the Italian court become seised of the proceedings involving the jurisdictional cause(s) of action? In their summons the insurers asserted the jurisdiction of the Tribunale di Genova on the basis of article 16 of the General Conditions of the Policy. Mr Morris says that this is no more than the equivalent of the statement that is to be made under CPR 6.19(3) and the Practice direction dealing with Service out of the Jurisdiction para 1.3. That may be so: there is no evidence of Italian law dealing with the point. But it does not seem to me to follow that, when the Italian proceedings were issued, the Italian court was not seised of proceedings involving the jurisdiction claims. Of course, the jurisdictional questions might or might not have proved contentious – often it is not clear what issues are contentious until a defence is served – but that does not mean that the proceedings did not involve jurisdictional questions. On the contrary, in as much as courts of Member States are obliged under the scheme of the Regulation to satisfy themselves that they have jurisdiction in a case such as this, the proceedings inherently involved the question of jurisdiction. If it is asked when the Italian court became seised of proceedings involving the jurisdictional questions, I consider that it did so when the Italian proceedings were brought in June 2002.

For these reasons I consider that Evialis have no answer to the insurers application for a stay of the jurisdiction claims under article 27 of the Brussels Regulation. Had I not so concluded, I would have ordered a stay under article 28, considering these proceedings, including the jurisdiction claims, and the Italian proceedings to be related actions. I order that these proceedings be stayed.

In Turner v Grovit (cit sup), the House of Lords referred to the European Court of Justice the question whether it is inconsistent with the Brussels Convention for the courts of this country to grant restraining orders against defendants who are threatening to commence or continue legal proceedings in another Convention country where the defendants are acting in bad faith with the intent and purpose of frustrating or obstructing proceedings properly brought before the English courts. The European Court has yet to determine this reference. I must proceed on the basis of the law as I understand it presently to be. It was authoritatively described by Lord Hobhouse in his speech in Turner v Grovit, in which the following principles are stated:

Evialis claim that the defendants should be restrained from pursuing the Italian proceedings and ordered to discontinue them. They seek an interim order restraining the insurers from pursuing the proceedings on the grounds that their claim raises a serious issue to be tried for three reasons: first, that in bringing the Italian proceedings, the insurers are in breach of the contract of insurance and in particular the provisions of the service clause; secondly, that in bringing the Italian proceedings, the insurers are contravening Evialis’ right to be sued only in accordance with the provisions of the Brussels Regulation; and thirdly, that the insurers’ conduct in bringing the Italian proceedings is vexatious, oppressive or otherwise unconscionable.

For reasons that I have explained, I do not consider that Evialis have a case that the insurers are in breach of the contract of insurance in bringing and pursuing the Italian proceedings that is sufficiently sustainable to justify an interim injunction.

I have also concluded that, on the evidence before me, Evialis have not shown a good arguable case that the Italian proceedings are not brought in accordance with the Brussels Regulation, or that France would be the only proper forum for such proceedings. However, whether or not this argument is correct depends upon the interpretation of the Italian contract of insurance and even if I had reached a different conclusion about this, I should not have made an interim injunction. I can state my reasons briefly:

First, the suggested justification for the injunction is not a need to protect a right to be sued or to sue in this country, but a right to be sued in France. Of course, the answer to this question has an impact on the English proceedings, but that is true of any question where a court seised before the English court has to decide whether proceedings are properly brought before it. The English court has no sufficient interest in protecting any right that Evialis might have to be sued in France rather than Italy to protect it with injunctive relief.

Secondly, the right that Evialis assert and seek to protect is a right arising under the Regulation. The Regulation itself provides a machinery for protecting the rights that it confers, and for determining the court that is to give effect to those rights and to enforce the corresponding obligations. I recognise, as Mr Morris emphasised, that the Brussels Regulation is directly applicable, and so creates rights and obligations between individuals under the EC Treaty, but this does not seem to me to justify granting injunctive relief outwith the machinery of the Regulation in order to enforce any rights that it confers.

In any case, the question whether the Italian proceedings are justified by article 16 of the General Conditions raises a question about the interpretation of an Italian contract that seems to me to be of some difficulty. It is appropriate that the Italian court, rather than this court, should determine it. In these circumstances, I should not exercise my discretion to grant an injunction.

I have, for reasons that I have explained, concluded that there is a real risk that the Italian courts would regard it as an affront for an English court to make an injunction in a case such as this, and for that reason too, I should have been disinclined to exercise my discretion in Evlialis’ favour.

However, Evialis put forward other reasons that the Italian proceedings are vexatious or oppressive or unconscionable, and so should be restrained. First, they rely on the fact that the insurers brought the Italian proceedings without responding to Evialis’ letter before action of 14 March 2002 and without sending a letter before action of their own. Evialis say that there is no good reason that the insurers could not simply have defended the English proceedings. Accordingly Mr Morris submits that the only purpose of the insurers’ conduct was to bring it about that the Italian court was first seised, and so to deprive the English court of the jurisdiction that it was agreed by the parties that it should have. However, this argument, as it seems to me, loses much of its force when it is recognised that the insurers, warned of impending English proceedings, did not rush to anticipate them by issuing a summons in Italy. They waited three months after the letter before action before bringing the Italian proceedings. Evialis’ delay in following up their letter before action is unexplained, but the chronology of events belies the suggestion that the insurers were intent upon frustrating Evialis’ wish to sue in England. I do not doubt that the insurers prefer to litigate in Italy for tactical purposes. That is not inherently improper, and I see nothing oppressive or unconscionable in the way that the insurers went about doing so.

Evialis’ second complaint is that the insurers’ case, as set out in the Italian proceedings, is inaccurate and potentially misleading in that the summons refers neither to the letter of 25 October 2001 in which Evialis presented a quantified claim nor to the service clause. Evialis’ suggestion is that the insurers should have referred to the former because they are asserting the one-year time bar, and to the latter given that they are asserting Italian jurisdiction despite the service clause. I am unconvinced that there is anything in these points. I am in no position to say whether or not the letter of 25 October 2001 provides an answer under Italian law to the time bar, but I see no reason to suppose that the insurers were under any requirement of Italian procedure or practice to anticipate this potential defence to their claim in their summons. Equally, I see no reason to suppose that, as a matter of Italian procedure or generally accepted practice, the insurers, having asserted Italian jurisdiction on the basis of article 16 of the General Conditions, should have referred to the service condition or any other matter upon which Evialis might rely in a jurisdictional challenge in the Genoan court.

I therefore do not consider that Evialis have an arguable case that the insurers’ conduct is unconscionable in the sense explained by Lord Hobhouse. Evialis have not demonstrated a basis upon which I could properly make an injunction. Even if they had done so, I should not have exercised my discretion in their favour for the reasons that I have stated.

I therefore order that there be a stay of these proceedings. Evialis’ application is dismissed.