Mahmudov v Sanzberro. Addressing libel tourism under Brussels Ia with a debatable reading of eDate’s Centre of Interests.

Mahmudov & Anor v Sanzberro & Ors [2021] EWHC 3433 (QB) tackles the issue of libel tourism. As Collins Rice J puts it [3]

underlying the contest of law is a contest of two mainstream policies embodied in modern defamation law: on the one hand, the need for the law to keep up with the borderless realities of the internet, and on the other the need for international libel to be dealt with by the courts best able fairly to do so (or, to put it less neutrally, to prevent ‘libel tourism’).

The case is held under Brussels Ia for the claim was introduced on 31 December 2020, ‘IP completion day’.  Parties mostly seem at loggerheads over the implications of CJEU C-509/09 eDate. Claimants suggest eDate establishes a stand-alone full jurisdictional gateway for the Member State where the aggrieved has his or its centre of interests – CoI. Defendant claims [19]

there is still a binary choice, as per Shevill: to sue either (a) where a defendant is domiciled or (b) where a completed tort (the harmful event) occurred. The effect of eDate, they say, is that claimants taking the second route in their CoI country can now get global relief rather than being limited to compensation for harm arising in that individual state. CoI is not jurisdictional in the pure sense of introducing a freestanding basis for bringing an action somewhere; it is jurisdictional only to the limited or secondary (but nevertheless important) extent of the nature and quantum of the relief that may be sought.

Parties oddly seem in agreement that Shevill v Presse Alliance (No.2) [1996] AC 959 reaffirmed ([11] in Mahumdov]

that what constituted the ‘harmful event’ was to be determined by the national court applying its own substantive law. In other words, the preliminary jurisdictional question for the High Court in a libel case brought against a non-domiciled defendant was whether a claimant could show to the requisite standard that all the components of a tort actionable in the UK were present

I find that debatable to say the least, and in fact that consensus has an important impact on the judge’s final conclusion, which rejects CoI as a stand-alone gateway: [28] the judge sides with the defendants for the claimant’s reading would imply ‘an autonomous meaning of the ‘place where the harmful event occurred’ ‘. The latter, many might argue, must be the implication of the CJEU’s overall application of Brussels Ia. At [34] Napag Trading is offered in support however the judge I feel in Mahmudov should  have made a clearer distinction (as the judge did in Napag Trading) between the EU-governed jurisdictional gateway for tort, and the (England and Wales) governed Civil Procedure Rules test for a ‘good arguable case’. As I note in my review of Napag Trading, these CPR rules may still form a formidable procedural hurdle, however properly distinguishing between them is important, among others for costs reasons.

Geert.

Ask me no questions, and I’ll tell you no lies. The CJEU on internet (libel) jurisdiction in Gtflix.

The CJEU held yesterday in Grand Chamber in C-251/20 GtflixTV – for the facts see my initial flag of the case here. I reviewed the Opinion of Hogan AG here. The AG need not have bothered for the Court entirely ignores the Opinion.

The AG had predicted, as had I, that the CJEU would not heed his calls (joining those of plenty of AGs before him) that the Article 7(2) CJEU Bier introduced distinction between Handlungsort and Erfolgort be abandoned or at least curtailed. The CJEU however also dismisses his suggestion that the case at issue, which involves defamation of competitors over the internet, does not engage the Bolagsupplysningen case-law (infringement of personality rights over the internet) but rather Tibor Trans on acts of unfair competition.

I do not see quite clearly in the Grand Chamber’s mention [28] that Gtlix did not request inaccessibility of the information in France: for Gtflix did request retraction.

Instead of qualifying locus damni jurisdiction, the CJEU squarely confirms its faith in the Mosaic consequences of Article 7(2) locus damni jurisdiction. Each court in whose district damage has occurred, will continue to have locus damni jurisdiction even if the claimant requests rectification of the information and the removal of the content placed online in the Handlungsort or centre of interests jurisdiction. Locus damni jurisdiction in my view extends only to the damage occurring in that district (for Article 7(2) determines territorial, not just national jurisdiction), albeit in current, internet related case the CJEU [38] would seem to speak of ‘national’ jurisdiction, linked to accessibility in the Member State as a whole.

Those courts’ locus damni jurisdiction is subject to the sole condition that the harmful content must be accessible or have been accessible in that Member State. Per CJEU Pinckney, an additional direction of activities to that Member State is not required (the recent High Court approach in Mahmudov on which I shall blog shortly, is at odds with that approach nota bene).

Grand Chamber judgments must not only be expected in cases where earlier authority is radically changed or qualified. It can also occur in cases where the CJEU wishes to reconfirm a point earlier made but stubbornly resisted in scholarship and lukewarmly embraced in national court practice.

Geert.

Athena Capital Fund v Secretariat of State for the Holy See. Thank Heavens for jurisdictional mercies (here inter alia involving lex fori prorogati and agency for choice of court).

Athena Capital Fund Sicav-Fis SCA & Ors v Secretariat of State for the Holy See [2021] EWHC 3166 (Comm) features as defendant the Secretariat of State of the Holy See  (not the Holy See itself), and relates to a fraud and embezzlement claim of property in Chelsea, London.

Defendant says that from the perspective of Claimants, the purpose and intention of bringing these proceedings is to try to influence the criminal process in Italy, and/or the publicity emanating from the criminal process.

For its jurisdictional challenge, defendant argues [81] i) The claim was not a “civil and commercial matter” within the meaning of A1(1) BIa; ii) one of the claimants was not a party to the relevant Sale and Purchase Agreement (SPA) and could not rely upon it [this was summarily dealt with [88] by suggesting an amendment of claim] and, more forcefully, (iii) Defendant was not a party to the SPA for the purposes of A25 BIa.

Salzedo J justifiably in my view held [84] that

whether the claim is a civil or commercial matter does not turn on the subjective intentions of the claimant as to the ultimate effect that a claim might have on its interests, but on an objective reading of the claim itself and the relief that it seeks from the court. On that basis, it is a claim for declarations against the Defendant concerning the Defendant’s entry into commercial transactions with the Claimants.

and that the transaction was not entered into by the Defendant in the purported exercise of public powers: [86]

The Transaction was one that any private person could have entered into if it had the requisite funds. Nothing that was essential to the Transaction required sovereign powers to enter it and nothing that the Defendant did or purported to do was in the exercise of public authority.

As for the defendant not being a party to the SPA, the context here is whether a party involved in the signing accepted the SPA and its choice of court as an agent of the defendant. The judge, confirming the parties’ consensus, points out that that agency issue befalls to be addressed by English law. It is not said why that is the case however it is of course the result of the amended A25 – as others before it, however, the court does not complete the lex fori prorogati analysis with the recital 20 in fine mandated renvoi. On the agency issue the judge holds there is a good arguable case that the relevant agent did bind the defendant.

Next [103] ff follows a CPR-heavy discussion on the amendment of the claim form, seeing as the claimants erroneously assumed [120] that BIa was not engaged as the Vatican is not party to Brussels Ia. At [123] the conclusion is that the claim form may be amended and that defendants’ time spent in dealing with the service out issues under the common law (a wasted exercise as BIa applied), may be met in the costs order.

Once the A25 point rejected, there would have been a most narrow window for any kind of stay, yet the defendants try anyways, with [129] a series of abuse and case management arguments. One particularly poignant one is that the proceedings would interfere with a criminal proceeding. After discussion the judge [159] dismisses the idea on the facts, seeing as none of the declarations sought would involve any assertion as to what does or does not amount to criminality as a matter of the law of the Vatican State.

[163] ff discusses the abuse of process issue which the defendants, I understand, presented more or less as being integrated into the criminal procedure element, discussed above. That was wise, for abuse of process, while entertained among others in Vedanta, is arguably noli sequitur in a BIa claim. [Support for the alternative view here was sought [172ff] in Messier-Dowty v Sabena SA[2000] 1 WLR 2040]

The case-management stay proper is discussed 192 ff with reference ia to Municipio, and Mad Atelier. The judge in current case is very aware of not re-introducing through the back door what CJEU Owusu shut the front door on. He summarily discussed the possibility anyway, only to reject it.

An interesting case.

Geert.

Bitar v Banque Libano-Francaise. A reminder that, under (acquired) EU law, one must not be domiciled in the EU for one to ‘direct’ one’s activities here and so trigger the consumer section.

Bitar v Banque Libano-Francaise S.A.L. [2021] EWHC 2787 (QB) discusses whether a Lebanese bank could be considered to have ‘directed’ its activities at the UK under the CJEU Pammer Alphenhof criteria, thus triggering the consumer section of the (acquired) EU Brussels Ia Regulation.

Kent DJ held it had: claimant’s arguments are at [29] ff, purporting to build evidence of a chain of marketing aimed at the expat Lebanese community. They show the importance of information put on websites, often made to look more glamorous by the addition of elements such as links to England and London in particular.  The judge is on point I find where he dismisses the singular relevance of the use of English etc in a world where every Tom Dick and Harry put that on their website. However he does conclude [65]

the website pages to which I have referred which were visible from the United Kingdom do indeed give the impression to a fair-minded observer—and I would say quite a strong impression—that the Bank was interested in obtaining custom from the expatriate Lebanese community in whichever part of the world not insignificant numbers of those who can be treated as falling within that expression were gathered and that in 2014 did include England. 

Geert.

EU private international law, 3rd ed. 2021, Heading 2.2.9.2.7, 2.270 ff

Chowdhury v PZU SA. Domicile for the purposes of the insurance title.

Chowdhury v PZU SA [2021] EWHC 3037 (QB) is worth a brief post on the determination of ‘domicile’ for the purposes of the insurance title of Brussels Ia (the very same Title and provisions which I discuss in Betty Tattersal). Ritchie J discusses whether the claimant was ‘resident’ in England and Wales for despite the insurance section talking of ‘domicile’, the Regulation refers for that notion to the residual rules of the Member States; and in England and Wales, domicile for natural persons, for private international law purposes, is determined by their ‘residence’.

The judge held, having summarised all relevant authority, that the earlier finding of residence absolutely stands [72]:

Claimant was a British citizen, with a British passport, who grew up in Worthing and was educated in England, worked in England, had his parents and family in England, had his friends in England, had rented flats in London, in Earls Court and in Putney, had his benefits paid in England, had his property by way of clothes and personal items in England and kept some of those at his parents’ house in Worthing, in his own room there.

That he gave up his rental accommodation in England was entirely due to him seeking medical treatment in Germany on account of the very tort he is suing for. Clearly that could not dislodge his English residence, despite the most likely temporary impact on physical stays in England.

Geert.

Kabab-Ji SAL (Lebanon) v Kout Food Group. The UKSC in my view unconvincingly on governing law of an arbitration agreement.

I am slowly getting through the in-tray with back cases, looking in this post at the UKSC judgment in Kabab-Ji SAL (Lebanon) v Kout Food Group [2021] UKSC 48. There is plenty of analysis on the case already out there, among Gilles Cuniberti and a team of CMS lawyers. I previously discussed the judgment appealed.

The SC dismissed the appeal and the judgment therefore stands: parties’ choice of English law for the underlying contract was found to also be an express choice of the law governing the arbitration agreement. I suggested that finding was optimistic on the facts of the case. Moreover it would seem at odds with the separability line previously towed by the English courts: there may be perfectly valid reasons for having a different lex causae for the underlying contract, the arbitration clause, the lex arbitri and indeed the lex curia. Identity should not be too readily assumed. The SC however would seem to have been swayed by the New York Convention’s call for straightforwardness in enforcement (here lies as the SC notes a difference with previous case-law which concerned the pre-enforcement stage).

Not only is the UKSC approach at odds with the French SC, as I noted in my review of the Court of Appeal judgment. I am also not convinced that for the enforcement stage the SC should insist one keeps things simple whilst at many other stages through its authority it encourages often convoluted argument.

Geert.

 

 

The CJEU holds back on vis attractiva securalia in Betty Tattersal (re: the insurance Section).

I cross-referred to the pending preliminary review in C-708/20 Betty Tattersall in my review of Flowers v Centro Medico. The CJEU held this morning in Betty Tattersall (which also means the Flowers case may now continue). I call the issue ‘vis attractiva securalia’ which is a term I made up but hopefully usefully summarises the issue. Domicile of the claimant, as readers of the blog will know, is generally of no consequence in EU jurisdictional rules. There are a few exceptions, in particular for the protected category of consumers (for employees, it is the place of habitual performance of the employment contract which is relevant, which often co-incides with their domicile), and for insurance contracts – but in the latter case, it turns out, not of benefit for the injured party.

Ms Tattersall argues that , in relation to EB, who is domiciled at Ireland and owns the holiday home which Ms Tattersall rented and in which she suffered a fall, that a claimant may bring an action against an insurer domiciled abroad under A13(3) BIa. A13 effectively piggy-backs claims of the injured party viz liability insurance, unto the by virtue of A11-12 extended forum possibilities for the insured, the policyholder or beneficiary of the insurance policy. In her view, the existence of a ‘dispute’ between the insurer and the insured regarding the validity or effect of the insurance policy is not necessary in that regard. The only requirement under A13(3), she suggests, is that such an action against the insured is provided for by the law governing direct actions against the insurer, in this case Spanish law. EB by contrast argues that Ms Tattersall’s claim is not an insurance claim and cannot become one merely because it was brought in the same action as the direct action against the insurer.

One can see why Ms Tattersall attempted to join EB into the English procedural bath: the insurer is an interesting defendant of course for it has deeper pockets; however EB’s insurer argue that the limitations and restrictions in the insurance policy meant that the policy did not extend to EB’s use of the property for the purpose of accommodating third parties on holiday against payment. Should that argument hold on the merits, Ms Tattersall’s claim will fall flat and she would have to sue EB separately, in Ireland.

The CJEU insists on the need to read the insurance Section with its specific purpose in mind: the protective effect intended by the Section is aimed at those considered to be in a weaker position vis á vis the insurer: the insured and the injured person are considered to be the weaker party in the contractual relationship, not the third party who is in a non-contractual relationship to both. Therefore [30]

to justify the application of the special rules of jurisdiction laid down in Section 3 of that regulation, the action before the court must necessarily raise a question relating to rights and obligations arising out of an insurance relationship between the parties to that action.

The Court does [35-36] acknowledge recital 16 BIa which refers to the objective of facilitating the proper administration of justice. It concedes that the involvement, by the injured person, of the insured, as a third party to the proceedings before the court seised, would make it possible to avoid the risk of the coexistence of two parallel sets of proceedings. Nevertheless, the emphasis must fall on the effet utile of A7(2): allowing the injured person to bring an action against the insured on the basis of A13(3) would amount to circumventing the forum delicti rules of A7(2). Each injured person could then bring an action against the insurer on the basis of A13(2), to benefit from the more favourable provisions of A10 to 12 in order, subsequently, to bring an action against the insured, as a third party to those proceedings, on the basis of A13(3).

The CJEU’s’ formal reply therefore is that

Article 13(3) of Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters must be interpreted as meaning that, in the event of a direct action brought by the injured person against an insurer in accordance with Article 13(2) thereof, the court of the Member State in which that person is domiciled cannot also assume jurisdiction, on the basis of Article 13(3) thereof, to rule on a claim for compensation brought at the same time by that person against the policyholder or the insured who is domiciled in another Member State and who has not been challenged by the insurer.

Readers may enjoy the strategic off the cuff claim engineering possibilities which I entertained with Sarah Crowter QC.

Geert.

The CJEU in Hrvatske Šume on contract or tort re claims of unjust enrichment. Confirmation that not all claims need to be either one or the other.

The CJEU held this morning in  C‑242/20 Hrvatske Šume. Gilles Cuniberti already has a summary of the judgment here and I reported on the AG Opinion here. The Opinion was in essence confirmed.

Of first note is that the CJEU unlike the AG does address the Article 24(5) Brussels Ia issue of exclusive jurisdiction for claims in ‘enforcement’ of a judgment. It holds that an action for restitution based on unjust enrichment does not come within A24(5)’s scope for [32] an action the subject matter of which is a claim for restitution based on unjust enrichment is not intended to obtain a decision in proceedings relating to recourse to force, constraint or distrain on movable or immovable property in order to ensure the effective implementation of a judgment or authentic instrument. This is the case even if that unjust enrichment arises from the fact that enforcement has been annulled.

On the A7(1)-(2) issue the Court first of all and justifiably dismisses the suggestion made ia by the European Commission that A7(1) and (2) BIa dovetail: ie that necessarily a claim which is not a contractual one, must (and in subsidiary fashion) be on in delict per A7(2): [53]: forum delicti requires a harmful event (reference in support is made to Austro-Mechana),  which is simply absent in cases of unjust enrichment.

A claim in restitution of unjust enrichment may in fact be contractual [47] if there is a pre-existing relationship that is closely linked to the claim. An obvious example [48] is that of the applicant relying as the basis of its right to restitution. on unjust enrichment closely linked to a contractual obligation which he or she regards as invalid, and which has not been performed by the defendant, or which the applicant considers he or she itself has ‘over-performed’.

Geert.

EU Private International Law, 3rd ed. 2021, 2.419 ff.

Heslop v Heslop. A reminder of the constraints of the Moçambique rule for rights in rem, and (obiter) on joining a pre-Brexit with a post-Brexit claim under the Withdrawal Agreement.

Heslop v Heslop & Anor [2021] EWHC 2957 (Ch) essentially queries whether Deceased testator actually had any estate or interest in Jamaican Property which she could pass by will.

Under the Moçambique rule (after British South Africa Co v Companhia de Moçambique [1893] AC 602) an English court will not, as a matter of its own limits to jurisdiction, by and large determine matters of title to foreign land. The purpose of the rule is the maintenance of comity and the avoidance of conflict with foreign jurisdictions. The rule has been discussed on the blog before and it finds its EU equivalent of course in Article 24  Brussels Ia.

After considering the rule and the facts of the case, Dray DM holds it is not triggered here for [51-52]

the relief sought (across the two claims) is relief of an in personam nature in a dispute between the two central protagonists, the Second Defendant (the asserted trustee) and the Claimant (the asserted beneficiary) under the asserted trust. The fact that the land in question is situated in Jamaica does not preclude this court from having jurisdiction to hear the claim. The proceedings do not involve any determination of rights in rem. They do not assert a property right which is by its nature enforceable against third parties and they do not purport to bind strangers/third parties. For instance, no possession order, effective against the world at large, is sought (and none could be granted by this court). Neither is any order directed to the Jamaican Land Registry claimed (ditto). The court is only asked to resolve a dispute between those before it, the proceedings being based on an alleged personal (trust) relationship between the Claimant and the Defendants.

Obiter he then [57ff] considers forum non conveniens (argued in fact by neither parties), with the complication [63] that the two claims before the court have not been consolidated and are thus separate claims, albeit proceeding together, and that the first claim was commenced before the end of the Brexit transition period whereas the second claim was commenced afterwards. The judge holds (again: obiter) [68] (seeing also that no consolidation has been sought) that the former claim needs to be assessed viz BIa and the latter viz the post-Brexit rules, [74 ff] that under BIa A24 is not engaged for the same reason as the Moçambique rule, and [72] that if it had been, he would have been minded to follow (with all the necessary caveats  Kennedy v National Trust for Scotland‘s reflexive application.

Geert.

EU Private International Law, 3rd ed. 2021, para 2.208.

Mann v TUIP. Foreign parties reminded of the need to seek proper legal advice, early.

Mann v Towarzystwo Ubezpieczen Inter Polska SA & Ors [2021] EWHC 2913 (QB) showcases the recalcitrance which parties can still demonstrate to object to service despite the EU Service Regulation 1393/2007, should they have the obstinance, and deep enough pockets, to do so. Daniel Matchett reviews the case here and I am happy to refer.

Master Thornett at 5 is particularly on point when he refers to the need, particularly for professional defendants, to seek advice on E&W CPR early:

I do not find Mr Grochowalski’s emphasis upon the First Defendant being a small niche company in the Polish insurance market and as had no previous experience of a claim against it brought in another jurisdiction relevant. I am satisfied that such a company could and should taken reliable advice to understand about the English proceedings from service in May 2017 if not previously upon their notification by the Claimants’ solicitors. This could have been done by a variety of means…

Of additional note to Daniel’s assessment I find is the jurisdictional challenge dismissed seemingly by Master Brown in 2017 for reasons I understand of the tardiness of the objection, and the unsuccessful current attempt to resurrect it in this later application. One assumes claimant may have argued the contract was a consumer contract, allowing her (and now that she has sadly passed away, her husband) from pursuing the case in her own domicile.  An objection to jurisdiction which, going by the little the judgment reveals of the contractual circumstances, may have had some merit had it been brought earlier (I emphasise however I do not have much to go by here).

Geert.

%d bloggers like this: