Flux européens

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Like all my posts, this post is entirely AI free

Convrgnt Value Engineering LLC v Kennedys Dubai LLP [2026] EWHC 1754 (Ch)  raised an interesting choice of law issue under the UK’s Unfair Contracts Act 1977, in combination with the ‘assimilated’ EU Rome I Regulation on the law applicable to contractual obligations.

Context is a retainer on legal services provided by the UAE office of an English law firm.

s27(1) UCTA reads

“Where the law applicable to a contract is the law of any part of the United Kingdom only by choice of the parties (and apart from that choice would be the law of some country outside the United Kingdom) sections 2 to 7 and 16 to 21 of this Act do not operate as part of the law applicable to the contract.”

Choice of law was made for English law. The issue, so as to determine whether UCTA may apply at all, is what law would have applied had no choice been made.

Issues for determination in current judgment are whether Kennedys Dubai is a “branch” within the meaning of A19(2) Rome I, displacing England as habitual residence under A4(1) Rome I; alternatively, whether A4(3) Rome I’s escape clause applies.

Article 19 Rome I reads

Article 19 – Habitual residence

1.   For the purposes of this Regulation, the habitual residence of companies and other bodies, corporate or unincorporated, shall be the place of central administration.

The habitual residence of a natural person acting in the course of his business activity shall be his principal place of business.

2.   Where the contract is concluded in the course of the operations of a branch, agency or any other establishment, or if, under the contract, performance is the responsibility of such a branch, agency or establishment, the place where the branch, agency or any other establishment is located shall be treated as the place of habitual residence.

3.   For the purposes of determining the habitual residence, the relevant point in time shall be the time of the conclusion of the contract.

Article 4 Rome I contains the general regime for determining the applicable law, outside the specific regimes for protected categories such as employees and consumers, when parties have not made choice of law themselves. The relevant entry under A4 Rome I for the case at issue is A4(1)b :

a contract for the provision of services shall be governed by the law of the country where the service provider has his habitual residence

and the escape clause of A4(3) reads

Where it is clear from all the circumstances of the case that the contract is manifestly more closely connected with a country other than that indicated in paragraphs 1 or 2, the law of that other country shall apply.

Kennedys Dubai’s argument is this [36].

The TOB included a choice of law clause, whereby the parties agreed that the Retainer would be governed by and construed in accordance with the law of England and Wales. Without that clause, by application of [A4 Rome I]… the relevant place of habitual residence was the UAE; pursuant to [A4(1)] (read together with [A19 Rome I] it is that which dictates that the applicable law is that of the UAE. Alternatively, it is said that the Retainer was manifestly more closely connected with the UAE than any other country, and, pursuant to Article 4(3), it is that which dictates the applicable law. If either of those contentions is made out, section 27 will operate to disapply the provisions of UCTA on which CVE relies.

[42] Parties agree that Kennedys Dubai’s habitual residence is the UK, that being the “place of central administration” for the purposes of A19(1). Kennedys Dubai relies on paragraph (2) of A19 to displace that result, contending that the Retainer was concluded in the course of the operations of a branch and/or that under the Retainer performance was the responsibility of such a branch. If that is correct, then since the branch is located in Dubai, Dubai is to be treated as the place of habitual residence and, had parties not chosen the law of England and Wales as lex voluntatis, it would be the law of Dubai which governs the Retainer pursuant to A4(1)(b).

Noting that there is no authority on ‘branch’ within Rome I, and only a little within the Brussels Ia Regulation, Shea DJ [46] gives an important “health warning” viz the consistency of interpretation between the Brussels Ia and the Rome Regulations:

The editors of Chitty on Contracts (36th Ed, at para 34-062) suggest that it is likely that the expression will be interpreted consistently with the same expression as found in Brussels Recast. However the editors of Dicey, Morris & Collins on the Conflict of Laws (16th Ed), at para 32-020 warn that the desire for consistency expressed in Rome I and Brussels Recast must be understood in the context of the fact that Rome I and Brussels I are pursuing differing objectives.

Regular readers of the blog will be aware that I very much stand with the editors on Dicey on this one, see eg here.

The judge again [46] refers to the Dicey editors also noting

“as the United Kingdom is no longer bound by the Brussels I Regulation or Brussels Convention, it is possible that the gravitational pull of those instruments will fade somewhat over time as the jurisprudence of the English courts develops”.

This of course is a complicating factor for litigation in the UK.

On habitual residence, the judge discusses the relevant authorities including Gefion (but not Rynair, which was not put before her it seems). She emphasises, per Gefion, the need for restrictive interpretation.

Kennedys Dubai is a limited liability partnership under the law of England and Wales, incorporated in England and Wales on 1 March 2010. It has always been regulated by the UK Solicitors Regulatory Authority which treated is as “an overseas practice” pursuant to the SRA Code of Conduct, on the basis that the Dubai Office was “a practice from an office outside England and Wales”.

Kennedys Dubai is a wholly owned subsidiary of Kennedys Law LLP (“KL”) which was incorporated in England and Wales on 15 March 2010 and was one of two designated members of Kennedys Dubai, the second being Kennedys Management Holdings Limited. Kennedys Dubai’s registered office was at London, which was also the physical office of Kennedys Law. Kennedys Dubai’s only physical office was at Dubai. KL’s financial statements at the relevant time state that Kennedys Dubai had “its place of business” in Dubai.

The people who have been carrying on the former partnership, Kennedys Legal Consultants, have agreed to transfer that business to Kennedys Dubai under the terms of a transfer agreement.

[61] the judge points to the obvious challenge: [Kennedys Dubai] “is not, as contemplated under the Ets A de Bloos formulation, subjected to the direction and control of any parent body. That very formulation requires the existence of two bodies. One will not suffice….Kennedys Dubai is not a branch of any entity or undertaking other than itself. More is needed.” In other words [63]: there is no ‘parent body’ and therefore there cannot be a ‘branch’.

In oral argument counsel for Kennedys argued that the Dubai office “appeared to third parties as being an easily discernible extension of the parent organisation, the defendant in London” (judge’s emphasis). The judge however holds [66]

The only sense in which Kennedys Dubai was “in London” was by way of its registered office. It is for that reason that the parties agree that England is “the place of central administration” for the purpose of determining Kennedys Dubai’s habitual residence under Article 19(1) (a conclusion which Kennedys Dubai says is overturned by the application of Article 19(2)). However I reject the proposition that the fact that London is Kennedys Dubai’s place of central administration entails that Kennedys Dubai is “in London” in any form in relation to which the Dubai office can sensibly be said to be an extension.

She is also right [67] to dismiss, for Article 19 purposes, any relevance for the Dubai office being “branded” (by reference to, by way of example, the footer of the paper on which correspondence was sent) as being part of “Kennedys Law worldwide”.  This ‘Kennedys Law Worldwide’ simply does not exist.

One feels the stronger avenue for A19 purposes would have been to argue that England was not the place of central administration of Kennedys Dubai – I am not privy to the reason why that point was conceded by Kennedys.

By a combination of A4(1)(b) and A19(1) therefore, the law governing the Retainer will be that of England and Wales, unless Kennedys Dubai can bring itself within A4(3) – a high bar as I have posted repeatedly on the blog – see eg Winrow v Hemphill (a Rome II case however Rome I and II are materially identical on this point) or Enka v Chubb at the UKSC.

[80] The factors are listed which Kennedys Dubai says bring the Retainer within the scope of A4(3):

The Retainer was concluded in the course of the operations of the Dubai Office. Responsibility for performance of the Retainer lay exclusively with the Dubai Office. The lawyers providing the services were physically located in Dubai. The dispute in respect of which Kennedys Dubai was retained was a Dubai dispute. The provision of the services was dependent on the fact that the Dubai Office and the fee earners were appropriately registered and licenced in Dubai. The client was based in the UAE. The services related to the Emaar Claim which (a) arose in respect of a construction project in the UAE; (b) concerned a contract itself governed by UAE law; (c) involved parties which were both incorporated and had licenses to trade in the UAE; and (d) was to be pursued in the UAE. The Retainer expressly addressed the fact that the Emaar Claim would require the services of local advocates to appear in the Dubai Courts, thus strengthening the connection to Dubai, not weakening it. The Retainer was drafted on the letterhead of the Dubai Offices, and provided for payment in local currency, payment which as understood by both parties would be made to the Dubai Office’s local bank account (as the evidence shows occurred).

The judge correctly holds [81] that this is not a numbers game:

“This issue must not be determined on the basis of which party can list the greater number of factors tending to support its view. I must look at the substance of those factors, in the context of the Retainer, to determine whether it is manifestly more connected to Dubai than it is to England and Wales.”

She concludes “with little hesitation” that it is: [82] ff:

the substantive elements of the Retainer are almost entirely connected to Dubai. The Emaar Claim was to be commenced in either the Dubai Courts, or in the DIAC Courts, or by way of arbitration in the event that agreement was reached with Emaar to this effect. The Retainer concerned a claim to be brought concerning a contract between two UAE entities involving a development which took place in the UAE. All the evidence in that claim would concern matters which occurred in the UAE. England and Wales had nothing to do with the Emaar Claim, whether as to law or evidence….

The fact that local advocates would have to be used if court appearances were required illustrates rather than detracts from the close connection of the Retainer with Dubai. Monies were to be paid in UAE currency, into a Dubai bank account. Further, the Dubai Office was physically located in Dubai. The fact that Mr Carnell also spent time in London is of little significance in times of easy global travel. Mr Sharma himself spent a lot of time in Switzerland during his mother’s illness, providing an apposite example of the irrelevance of the physical whereabouts of the principals of the contracting parties. Mr Carnell at paragraph 31 of his draft witness statement says that he anticipated that “the services under the Retainer would be provided solely by and from the Dubai office”, and that the dispute “was clearly only ever suitable for lawyers from the Dubai office”. He also states at paragraph 32: “everyone who acted on the case both lived and worked permanently in Dubai at all times”.

“The substance of the Retainer – its operative parts, its subject matter, its performance – spoke of Dubai or the UAE. The TOB contained the provisions that would be triggered in the event any one of a number of unlikely situations arose, and provided details of applicable statutory and regulatory regimes. They were connected to the law of England and Wales, but were nowhere near sufficient to displace the fact that the substance of the Retainer was manifestly connected with Dubai/UAE.”

In conclusion therefore if “the parties had not chosen the law of England and Wales, the law of Dubai would apply. Accordingly, section 27(1) of UCTA is triggered and the UCTA provisions regarding the reasonableness or otherwise of terms in contracts for the supply of services do not apply to the Retainer.”

The remainder of the judgment deals with the issues under the liability cap.

A judgment of illustration note for both A19 and A4 Rome I.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Like all my posts, this post is entirely AI free.

I have reported before on the frequent(ish) judgments of the United States Supreme Court discussing the possibilities, or lack thereof, of using the United States Congress’ Alien Tort Statute – ATS. Jesner v Arab Bank and Nestlé are I think the most recent posts, and my 2012 paper with Charlotte Luks gives overview of the earlier development of ATS in current context.

Readers will observe in these posts and the 2012 paper that the use of the ATS as a conduit for causes of action under what one could call business and human rights #bizhumanrights cases has been severely curtailed by SCOTUS over the years. Yet claimants continue to give it a go, in current case twinning the cause of action under ATS with one under the 2000 Trafficking Victims Protection Act TVPA.

As the FT summarise, in current case

Chinese and US citizens sued Cisco under the 1789 Alien Tort Statute alleging that Cisco helped the government in Beijing develop a robust surveillance system called the “Golden Shield” that facilitated the crackdown on Falun Gong members.

In CISCO SYSTEMS, INC., ET AL. v. DOE ET AL. No 24-586 Justice Barrett wrote an editorially most attractive, clear and concise judgment for the majority, holding the current Supreme Court

close the door that Sosa cracked to judicially created ATS liability. We do not disturb Sosa’s holding that the ATS is a jurisdictional statute; nor do we revisit its assumption that causes of action are available for torts corresponding to the Blackstone three. We conclude only that Sosa was overly optimistic in its prediction that there might be a narrow class of cases in which courts may create ATS actions without infringing on the prerogatives of the political branches. In truth, this class is a null set. What result for this case? Cisco argues that the Ninth Circuit erred in holding that aiding-and-abetting liability exists under the ATS for the torts alleged by the plaintiffs. Cisco is correct. Courts cannot create new rights of action to remedy violations of international law, so there is necessarily no liability for aiding and abetting such violations. Plaintiffs’ ATS claims against Cisco must be dismissed.

(footnote omitted). The ‘Blackstone three’ are explained earlier in the judgment:

Sosa reasoned that the First Congress would not have expected the ATS to “l[ie] fallow” until Congress or state legislatures enacted causes of action for violations of the law of nations. Id., at 719. Law-of nations offenses formed part of “the ambient law of the era,” and some “were understood to be within the common law.” Id., at 714, 720. William Blackstone’s legal treatise discussed three: “violation of safe conducts, infringement of the rights of ambassadors, and piracy.” Id., at 715 (citing 4 Commentaries on the Laws of England 68 (1769)).

The TVPA was also found not to harbour an ‘aiding and abetting’ cause of action.

Justice Barrett presents the judgment as logically following from earlier findings, calling out the Sosa precedent as internally incoherent, and current judgment correcting that inconsistency. Justice Sotomayor criticised the finding as not properly identifying its overturning precedent.

Which ever is right, the judgment confirms the difficulty of pursuing business and human rights claims in the US under the ATS.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Like all my posts, this post is entirely AI free.

In Case C-771/24 Hortis (in full Hortis GRC SA v JA, France Travail Île-de-France, formerly Pôle emploi Île-de-France) the CJEU held yesterday on the application of A6 Rome Convention, now A8 of the Rome I Regulation, as to the application of that Article’s ‘proper law of the contract’ approach. It generally follows an approach which IMO is consistent with the statutory language.

JA was employed as a ‘director’ by the Swiss IT company Hortis and was fired in 2012. JA lives in France and carried out his professional activities in France during the performance of the employment contract. The parties agreed that Swiss law would be the lex contractus.

Jurisdiction for the French courts (one assumes in accordance with the Lugano Convention; the proceedings were initiated prior to the entry into force of Brussels Ia, which dropped the requirement for the employer to be EU-domiciled for the employment title to kick in) was an issue in the lower courts but it is not part of the procedure before the CJEU.

The Court of Appeal held that French law, as the default law under Article 6 of the Rome Convention, trumps the Swiss lex voluntatis, because it is more protective of the employee: Swiss law lacks the requirement for a pre-dismissal meeting between employer and employee, and the requirement for a letter of dismissal to include a statement of reasons.

Appeal to that decision is now before the French Supreme Court. Authorities referred to in the referral decision of course include CJEU Schlecker and Joined cases C-152/20 (Gruber Logistics) and C-218/20 (Samidani Trans).

The core of the French Supreme Court’s doubt focuses on the ‘escape clause’ of Article 6, in fine, Rome Convention, now Article 8(4) of the Rome I Regulation [as the CJEU notes [30] Rome I’s “structure reflects the EU legislature’s intention to clarify, by using separate paragraphs, the different steps of the reasoning to be followed by national courts in order to determine the applicable law in the absence of a choice made by the parties”].

In the Rome Convention the text reads

.’..unless it appears from the circumstances as a whole that the contract is more closely connected with another country, in which case the contract shall be governed by the law of that country.’

The SC doubts whether the court seized must determine closer connections with the country of the lex voluntatis and, if so, disregard the rules of French
law which the employee seeks to have applied and which afford greater protection than those of the law chosen by the parties. The CJEU formules the doubt rather sharper in its judgment [30] than the SC did in its referral:

The referring court harbours doubts  as to the application of the mandatory rules of French law, which afford greater protection than those of Swiss law, in the light of the circumstances of the dispute before it, in which, first, Hortis and JA expressly chose Swiss law as the law governing the contract of employment at issue in the main proceedings and, secondly, JA habitually carried out his work in France. In that regard, it seeks to ascertain whether, pursuant to the last limb of Article 6(2) of the Rome Convention, it is required to examine whether that contract is more closely connected with Switzerland, taking into account also the connections resulting, in the performance of that contract, from the parties’ choice of the law applicable to it. If so, it enquires whether it must disregard the rules of French law.

In essence therefore, may nay must the escape clause set aside the more protective default law in favour of the chosen law, in those cases where it appears from the circumstances as a whole that the contract is more closely connected with the country of the chosen law?

The CJEU of course connects the issues, as it did viz the consumer title in ia VK v N1 Interactive, to the twin if sometimes competing objectives of predictability /legal certainty v protection of the weaker parties.

[33] it suggests it had actually already held on the question in CJEU Locatrans:

in paragraph 63 of its judgment of 11 December 2025,  Locatrans (C‑485/24, EU:C:2025:955), the Court held that ‘it is thus for the referring court to determine whether, in accordance with the last limb of Article 6(2) of the Rome Convention, it follows from all of the circumstances that the contract of employment at issue in the main proceedings is more closely connected with France than with Luxembourg, the law of which was chosen by the parties as the law applicable to that contract’. It follows that the Court has expressly recognised that that provision may require the referring court to give priority to the law chosen by the parties as the law applicable to the contract of employment concerned, where that contract is more closely connected with the country the law of which was thus chosen.

[36] therefore the Court’s formal finding is that

the last limb of Article 6(2) of the Rome Convention must be interpreted as meaning that, where the parties have chosen the law applicable to the contract of employment, national courts must give priority to that law by disregarding the mandatory rules, affording greater protection, of the law which the employee seeks to have applied and which would be applicable, in the absence of choice, under Article 6(2)(a) and (b), if, in the light of the circumstances as a whole, that contract of employment is more closely connected with the country the law of which was chosen by the parties as the law applicable to that contract.

[41] in determining the ‘proper law of the contract’, the ‘circumstances as a whole’ clearly do not rule out the lex voluntatis as a relevant factor.  [42] legal certainty is pushed as an important consideration of Rome Convention cq Regulation. [43] I think is slip-up:

Thus, the choice, made by the parties to a contract of employment, of the law applicable to that contract is not, in itself, a factor which may be taken into account in order to determine whether the contract is more closely connected with a country other than that in which the employee concerned habitually carries out his or her work or that of the place of business through which he or she was engaged. (emphasis added)

The better formulation is included in the operative part and in [45]

In order to ensure that the law of the country which best reflects the reality of the contractual relationship between the parties to a contract of employment governs that relationship, account must be taken of all the objective elements which define such a relationship, including those resulting from the choice, made by the parties to that contract, of the law applicable to it.

The balance between the various factors must be struck by the national court and in making that assessment, it is instructed to look at the actual presence of true consent (which of course the Regulation generally assumes is absent for the protected categories): [46]:

..in the context of that overall assessment, the weight attached to each of the elements of the employment relationship concerned may differ. It thus falls to the national court to weigh up, on the basis of their importance and relevance, each of the elements put before it and to assess which of them appears to be the most significant, irrespective of their number. In that respect, in order to guard against the employer misusing the connecting factors of the employment contract, the national court must examine, inter alia, whether those factors are the result of mutual agreement or whether they were imposed on the employee by the employer.

Of final note is that the CJEU finding here puts into the spotlights once again the absence of a ‘proper law of the contract’ approach /absence of an escape clause in the consumer title of Rome I.

A judgment of note.

Geert.

EU Private International Law, 3.39 ff.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Like all my posts, this post is entirely AI free.

Dexia SA v Comune di Torino [2026] EWHC 1401 (Comm) is the continuation of Dexia SA v Comune Di Torino [2025] EWHC 1903 (Comm) which I cross-refer to in an earlier post. The 2025 judgment is a declaratory one, incl on the validity of choice of court and breach of that clause by Torino, given the actions it has started in Italy.

Baker J in current case deals again with the issue of characterisation and applicable law, echoing issues in Banca Intesa which  I discuss here. The issues are considered not under the Rome I Regulation (assimilated law in the UK) rather under the 1980 Rome Convention,  with the statutory text however not materially different. He follows Dexia Crediop SpA v Comune di Prato [2017] EWCA Civ 428, the landmark authority. The critical question given Article 3(3) Rome Convention’s ‘purely domestic contracts’ rule (in the event all elements relevant to the “the situation” (notably not the same as ‘contract’) are domestic to one country, choice of law for a law other than that country, cannot trump the mandatory law of that country) is whether, apart from the ISDA law and jurisdiction clause “all the other elements relevant to the situation are connected with one country only”.

Of note at this stage is that claimant (‘Dexia’) is a French banking corporation which has succeeded to the rights and obligations of its former subsidiary in Italy, Dexia Crediop SpA, previously Crediop SpA. In other words the initial contract at issue, was concluded between two Italian corporations, but with choice of law for English law.

In Comune di Prato in the Court of Appeal,  the Court approved of the approach of Cooke J in Caterpillar Financial Services Corporation v SNC Passion [2004] EWHC 569 (Comm) [404] that “the inquiry is not limited to elements that are local to another country, but includes elements that point directly from a purely domestic to an international situation”. [134] in Prato, the CA singled out the use of the ISDA Master Agreement as a strong element in a finding of ‘international’ as opposed to purely domestic contracts.

In the case at issue, Baker J [81] holds

“the choice of law was not the only element relevant to the situation that disconnected it from Italy. Thus:

(i) the ISDA Master Agreement chosen was the ‘Multicurrency – Cross Border’ agreement rather than the ‘Local Currency-single Jurisdiction form’, so it contemplated more than one currency and the involvement of more than one country, as well as being in the English language;

(ii) the Transactions were part of a wider set of derivative agreements entered into with other banks, including JPM, a foreign bank (relative to the putative exclusive tie to Italy); and

(iii) the Transactions were the subject of two back-to-back hedges with Barclays Bank plc originally dated 12 April 2006 and amended on 24 June 2008, one of which was novated to Goldman Sachs International on 23 July 2015, and both Barclays Bank plc and Goldman Sachs International are foreign banks (relative to the putative exclusive tie to Italy).

English judges now firmly hold that the use of the multicurrency-cross border version of the ISDA Master Agreement, which in and of itself contemplates more than one currency and more than one jurisdiction, establishes an international element sufficient to rule out a Article 3(3) exception.

In doing so, they justifiably honour the party autonomy principle of Article 3 and Rome I in general, and pay heed to the need to apply Article 3(3) restrictively. However others will find that considerable weight put on the multicurrency form a touch excessive (and ofc in the case at issue, other elements, too, led to a disapplication of A3(3)). German courts, I understand, have a different approach to the issue and Rob Rooman, one of my PhD students and FWO fellow at Leuven, has a paper on the issues forthcoming.

Geert.

Handbook of EU private international law, 4th ed, 2024, 3.76 (5th ed forthcoming January 2027).

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Like all my posts, this post is entirely AI free.

Marion Ho-Dac reported a short while back on the French Supreme Court’s decisions 24-21.790 and 24-21.422 ECLI:FR:CCASS:2026:C100235. The facts of the case involve the Lebanese banking crisis, the jurisdictional implications of which have been reported on the blog before. The lower courts had held (and the SC does not second-guess that finding) that the activities of the bank where not directed at France and that the consumer title of Brussels Ia did also not otherwise apply.

In the absence of other jurisdictional anchors in Brussels Ia, jurisdiction therefore was left to be decided under residual French rules. As Marion reports, the SC has now extended its rule in employment contracts (jurisdictional arrangements pointing away from France, cannot deprive employees employed in France of the international and territorial jurisdiction of the French labour courts) to consumer contracts: French substantive consumer law ringfences jurisdiction for French courts. Marion cites the relevant provisions:

According to Article R. 631-3 of the French Consumer Code, “the consumer may bring proceedings either before one of the courts having territorial jurisdiction under the Code of Civil Procedure, or before the court of the place where he was resident at the time the contract was concluded or the harmful event occurred”. Furthermore, according to Article R. 632-1 of the same Code, “the court may, of its own motion, apply all the provisions of this Code in disputes arising from its application”.

In essence, the French SC dropped the direction of activities criterion of BIa and CJEU Pammer Alpenhof, leading to a more universal jurisdiction for French courts for parties qualifying as consumers and with habitual residence in France.

Now, I am not going to repeat professor Ho-Dac’s analysis, which is excellent both viz its French and EU PrivIntLaw angles. What I wanted to add (over and above emphasising Marion’s point that the issue needs to, and will be addressed in the review of Brussels Ia), is the difference between these cases and CJEU Roi Land Investments. In that case the CJEU held inter alia that more protective national rules (in Roi Land: a direct right to sue the “successor in title” of the employer) cannot trump Brussels Ia’s jurisdictional provisions for the  protected categories. However that is ‘more protective’ in cases where Brussels Ia’s protected categories title is engaged. In current case, the protective title is not engaged at all, neither are other parts of Brussels Ia, hence giving Member States free range in their jurisdictional space (subject to any relevant Hague or bilateral instruments, of course).

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Like all my posts, this one is entirely AI free.

In Case C‑43/25 SML Maschinengesellschaft mbH v AK, the CJEU reminded us end of March of the narrow opportunity which Article 16 of the European Insolvency Regulation – EIR offers to escape claw-back by the insolvency practitioner with use of an insulating lex contractus. Yet its ruling leaves open an important question on the lex cause for shareholder loans.

I realise that for the uninitiated, the intro above is a bit puzzling. Readers may want to refer to my post on CJEU Lutz or Nike to appreciate the context.

The Court essentially held that Article 16’s insulation mechanism does not extend to actions related to the relative ranking of claims in the collectivity.

Context is [31] “whether the Article may be relied on by a person who has received repayments in respect of a shareholder loan, which are regarded as detrimental to all of the creditors, in response to a request for restitution submitted by the insolvency administrator of the company that is the debtor of the loan where that request seeks to ensure compliance with the ranking of claims that is laid down in the legislation of the State of the opening of the insolvency proceedings.”

In other words the kind of debt at issue is a shareholder loan. These are commonly used as  way to finance a business. Arie van Hoe and Melissa Vanmeenen flagged some of the issues already in 2011 .

[33] the CJEU effectively handles the question of insulation with the use of Article 16 EIR, by simply referring to the need to apply Article 16 restrictively, as an exception to (now) Article 7 EIR, and [35] it concludes from that restrictive angle that “[Article 16’s] scope is limited to the situation envisaged in the latter provision, namely where what is at issue are the rules relating to the voidness, voidability or unenforceability of legal acts detrimental to all the creditors, and not the rules relating to the claims to be lodged against the debtor’s estate or their ranking.”

This is a pretty sec statement, made without AG Opinion. Sec, because while the shareholder’s action at issue did not indeed directly see to a pauliana scenario, the context of ranking is evidently one of (escaping) clawback.

As Arie notes in his review (in Dutch) of the judgment, the affirmative reply to the first question (the ranking request does not fall under A16), cancelled out an opportunity to respond to the second and further questions, These included the issue of determination of lex causae for shareholder loans: are these subject to Rome I’s lex voluntatis; or are they necessarily covered by lex incorporationis.

More questions are therefore to be expected…..

Geert.

EU private international law, 4th ed. 2024 (5th ed forthcoming 2027), Chapter 5.

Xandra Kramer, Stefaan Voet and Adriana Dori have edited an excellent volume on European Civil Procedure. To celebrate its launch, the editors are organising an on campus (Leuven) and online seminar, with authors speaking on topics related to the book.  Link to the event and registration is here.

Participation is free. This should be a welcome academic breather in between exam marking…..Hope to see many of you there.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Like all my posts, this one is entirely AI free.

Anyone with an interest in SLAPP (Strategic Lawsuits Against Public Participation – SLAPP) suits will of course be aware of the battle raging between Greenpeace and Energy Transfer. The former is an environmental and sustainability NGO of established pedigree. The latter the developer of the Dakota Access Pipeline  against which Greenpeace protests vigorously, supported ia by indigenous groups.

The procedural history particularly on the US side of the pond is excellently summarised by Hannah Buxbaum here. In short, Energy Transfer sued for defamation, tortious interference with business, and conspiracy under state law and won a considerable damages award in front of a jury, later effectively halved by the bench to a still massive USD 345 million. Greenpeace are seeking a retrial of that claim.

Meanwhile, Greenpeace itself sued in The Netherlands, alleging tortious liability on the basis of a range of public statements by and /or on behalf of Energy Transfer.

In these interlocutory debates, Energy Transfer argue that the Dutch courts lack jurisdiction, and subsidiarily, that they ought to apply lis pendens or, should the Dutch courts find that the retrial application in the US  bars res judicata, should stay the Dutch claim until the US case is final.

The background includes the EU’s anti-SLAPP Directive 2024/1069, summarised by Aberdeen’s Justin Borg Barthet here. Greenpeace argue the EU’s anti-SLAPP Directive may already be applied to the claim, despite the Directive’s implementation period not having been passed. It argues the Dutch abuse of process provisions, predating the Directive, already meet with the Directive’s requirements, and that consequently the Dutch rules may be applied with the specific provisions of the Directive in mind (CJEU Marleasing and CJEU Inter-Environnement on steroids, as it were: in Marleasing, the Court held that courts in the EU must apply national law as much as possible in line with EU Directives (which, unlike Regulations, lack direct applicability) and in inter-Environnement it expanded on what that means for national law prior to the Directive having been transposed into national law).

Application of the Directive not only would pre-signal unenforceability of the US findings per A16. It would also solidify jurisdiction of the Dutch courts: A17 reads

Jurisdiction for actions related to third-country proceedings

1.   Member States shall ensure that, where abusive court proceedings against public participation have been brought by a claimant domiciled outside the Union in a court or tribunal of a third-country against a natural or legal person domiciled in a Member State, that person may seek, in the courts or tribunals of the place where that person is domiciled, compensation for the damage and the costs incurred in connection with the proceedings before the court or tribunal of the third-country.

2.   Member States may limit the exercise of jurisdiction under paragraph 1 while proceedings are still pending in the third-country.

Meanwhile, a Dakota District court has denied an anti-suit injunction sought by Energy Transfer, yet the Dakota Supreme Court issued one (the Dutch court referred to the former but not to the latter finding): for all details on these, see Hannah’s post.

In Stichting Greenpeace Council v Energy Transfer LP et al ECLI:NL:RBAMS:2026:5461 the first instance court at Amsterdam has now held last Tuesday that it does have jurisdiction.

[4.4] ff the court finds that the anti-SLAPP Directive does not apply, Marleasing fashion, and cannot be used to apply the Dutch abuse of process rules in a way which would grant forum actoris (jurisdiction on the basis of the claimant’s domicile). The court rebukes the Dutch Government’s views expressed in its draft Act for transposition of the anti-SLAPP Directive as being optimistic.

However it then holds the residual private international law rules do already grant jurisdiction to the Dutch courts as the locus damni. [4.14] it refers to the clear Dutch parliamentary intention to apply Dutch CPR – Civil Procedure Rules in line with CJEU authority on the Brussels Ia Regulation in cases where, such as here, BIa does not apply.

[4.16] it emphasises that the defendant’s arguments on the actual relationship between parties must be taken into account yet that jurisdictional issues must not lead to a mini-trial, instead must be judged prima facie, to ‘good arguable case’ standard as it were.

[4.17] with reference to Universal Music, the court reiterates that the mere presence of a bank account does not suffice to establish jurisdiction (reference to CJEU Universal Music and Kolassa):  additional elements are required, and in the case of infringement of personality rights such as a libel claim, the court of the centre of the claimant’s interest may have jurisdiction (reference to CJEU eDate and Bolagsupplysningen).

[4.18] Greenpeace qualifies the damage as (i) financial loss, (ii) loss resulting from the deliberate obstruction of GPI in the pursuit of its charitable objectives, and (iii) damage to reputation. The court holds that Greenpeace does to prima facie standard show that its has led financial damage in The Netherlands as a result of this specific litigation and the corresponding legal costs (Energy Transfer had argued that the intense financial streams between Greenpeace and its US entities on a rolling basis, make it impossible to establish financial damage in The Netherlands).

[4.20] ff the court then also finds the presence of additional elements locating the damage in The Netherlands: [4.22] it accepts Greenpeace’s arguments that it had to dedicate significant manpower at ats Dutch international headquarters to deal with the US case, leading to those staff members not being available for other duties.

[4.24] the court holds that having established locus damni jurisdiction viz the US proceedings, it need not hold on the centre of interest jurisdictional gateway as a result of these proceedings, however it does hold on that gateway viz the alleged reputational damage resulting from Energy Transfer’s statements on the internet. [4.30] it holds that Greenpeace’s centre of interest is in The Netherlands, despite its global activities (the court refers ia to the absurd implication of Energy Transfer’s arguments on Greenpeace’s activities being global: for that would mean that globally active outfits can never have a COI).

[4.31] it dismisses Energy Transfer’s appeal to CJEU Shevill as being entirely behind the curve, given the later eDate and Bolagsupplysningen case-law, and [4.34] it rejects Energy Transfer’s argument per CJEU Marinari, that The Netherlands are merely a place of indirect damage: the damage manifests itself directly in The Netherlands.

[4.36] the appeal to lis pendens and /or res judicata is summarily dismissed: the claims are held to be entirely different claims even if they take place between the same parties.

[4.41] Permission to appeal is denied (it will be resurrected once the court finds on the merits, for which the parties are given a short additional period to file their submissions).

Of much note and just in time for my SLAPP slot at Monash University’s Prato Campus, in the climate litigation unit of my excellent colleague Joanna Kyriakakis.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so. Like all my posts, this one is entirely AI free.

I have been otherwise engaged for most of May and have a bit of a backlog on the blog – I shall try and tackle in in the upcoming days.

Acer Incorporated & Anor v Nokia Technologies OY [2026] EWCA Civ 564 is a good illustration of an issue I also raised at the first meeting of the experts on the reform of Brussels Ia: the relevance of qualification aka characterisation for jurisdictional purposes (and of course I was thinking of the impact of claim formulation and the room for forum shopping by same. In the EU, CJEU Wikingerhof has left many issues open).

The context in the case at issue is an application for a global licence on FRAND terms.

The issue Arnold LJ dealt with,, was whether a claim seeking to enforce Nokia’s obligation to licence the Challenged Patents on RAND terms, “relate[s] wholly or principally to property within the jurisdiction”, as required by the relevant jurisdictional gateway in England: CPR gateway 11. The Court of Appeal held it does, and thereby rejected the jurisdictional appeal.

[33] RAND Claims are contractual claims. In essence, the Claimants seek to enforce Nokia’s obligation to licence the Challenged Patents on RAND terms. It is common ground that a licence on RAND terms would be a global licence in respect of all corresponding patents worldwide (and, indeed, all other patents in the relevant Nokia Portfolio). Nokia contends that it follows that the subject matter of the RAND Claims does not “relate wholly or principally to property within the jurisdiction”, but to property situated very largely outside the jurisdiction.

Following Tesla Inc v InterDigital Patent Holdings Inc [2025] EWCA Civ 192 (currently before the UKSC), it is necessary to distinguish between the property on the one hand and the FRAND obligation which affects it on the other hand. RAND obligations attach to specific UK SEPs (standard essential patents) however they carry with them an obligation to grant a licence of global, and not merely UK, extent.

[36] In short, Nokia contends that the error in the reasoning is that it elevates form over substance. In form, such claims may be confined to UK patents, but as a matter of substance and reality they are global claims in respect of property most of which is outside the UK. Nokia says that implementer-led contractual claims such as this are therefore quite different to claims by SEP owners to enforce UK SEPs, where the (F)RAND obligation only arises by way of defence to the infringement claim.

Arnold LJ however in the briefest of terms, holds that he adheres to his reasoning in Tesla v InterDigital, and the other LJs agree. The pending UKSC appeal shall reveal whether the perception of extreme forum shopping possibilities, will hold out.

Nokia were also unsuccessful on lis pendens (the other forum here being arbitration: the terms of an interim licence are subject to adjustment by the decision of an arbitral tribunal as to what final licence terms are RAND).

However they were successful on a case-management stay.

A case of note.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Polish law requires a written form for the transfer of copyright and German law does not. [Said transfer may take various forms, whether by way of permanent assignment or temporary licence; but that distinction is not relevant to the conflict of laws analysis at issue]. That is the trigger for Case C-176/25 [Steizer], IU v BT in which Emiliou AG opined last week.

Given the difference between the two approaches, which is the law that governs the issue of the formal validity of that contractual licence, in a claim by the German (potentially exclusive) copyright holder ‘BT’ over photographs  of car floor mats (the facts are not that glamorous….) taken by a Polish photographer, where that validity is contested by a third party, ‘IU’, defendant in the course of infringement proceedings brought by BT seeing as IU used the same photographs of car floor mats to illustrate eBay sales?

IU argues that BT does not have standing to bring proceedings, because there was arguably no valid assignment from the photographer […] IU argues that the validity of the assignment of the exclusive rights to use the photographs is  governed by Polish law, per A4(2) Rome I: forum contractus determined by the habitual residence of the party required to effect the characteristic performance, arguably the Polish photographer, resident in Poland. A11(1)  Rome I [which the AG (44) calls ‘quasi-substantive (or ‘result oriented’) rules’: because A11 supports ‘favor negotis’: one of many potential laws may lead to the contract’s formal validity, in an effort not to let contracts stumble of conflict of laws formal validity roadblocks) then determines the law governing formal validity:

A contract concluded between persons who, or whose agents, are in the same country at the time of its conclusion is formally valid if it satisfies the formal requirements of the law which governs it in substance under this Regulation or of the law of the country where it is concluded.

Both alternatives in A11(1), in IU’s view, point to Polish law, and A53 of the Polish Law on copyright requires written form as defined in Article 78 of the Polish Civil Code for the assignment of rights of use in respect of copyright photographs.

By contrast, BT argues that German law applies as lex loci protectionis which , it says, is the relevant benchmark following A8(1) Rome II Regulation:

“The law applicable to a non-contractual obligation arising from an infringement of an intellectual property right shall be the law of the country for which protection is claimed. “

A8, BT suggest, applies to the act of disposal of the exclusive right of use (the Spaltungstheorie, or ‘splitting theory’, see below). Since an act of infringement took place in Germany, hence the right to use the photographs was also assigned in Germany,  BT asserts that the agreement on assignment of the rights of use is also governed by German law.

The ‘unitary theory’ in it seems mostly German speaking scholarship on Rome I and II, suggests all aspects of the right to use copyright works are governed by the principle of lex  loci contractus. This approach find support in A14(1) Rome I, according to which the property aspects of the assignment are uniformly governed by the contract law that applies to the underlying causal transaction.

The ‘splitting theory’ better IMO *bifurcation* (Spaltungstheorie) argues that only the act giving rise to obligations (titulus) is subject to the lex loci contractus, whereas the act of disposal (modus) is subject to the lex loci  protectionis.

The case therefore is very much one of applicable law shopping through claim qualification: if Rome I applies to this issue between BT and IU then BT will see its foundation for an infringement claim fall away. If Rome II applies then BT’s road to a successful infringement claim lies open.

The AG correctly imo opines that the formal validity of a contractual licence or assignment of copyright does not relate to ‘non contractual liability’, and that that issue should not be classified as ‘proprietary’, but rather as ‘contractual’.

Formal validity of a contractual licence or assignment of copyright does not relate to ‘non contractual liability’

This is the easy part of the Opinion, (52) ff.  As an example of Vorfrage (not a term the AG uses here), the issue of the validity of a contractual licence or assignment is preliminary to (and, thus, must logically be resolved by the court seised before) any discussion on whether the defendant infringed the copyright. The AG is absolutely right (54 – footnotes omitted)

the argument of BT that that validity issue should nevertheless be subsumed under the ‘non-contractual liability’ category when it arises in infringement proceedings, because it would be simpler for the court seised to determine the whole of the proceedings under the single substantive law designated by Article 8(1) of the Rome II Regulation, is misguided. In accordance with the general principles underpinning characterisation in conflicts of law, summarised in point 35 above, the conflict-of-law rules and, by extension, the law governing a given issue should depend on the nature of that issue, not the procedural context in which it is raised. Such an approach would entail the risk that, for instance, a given contractual licence or assignment be considered valid, under one set of substantive rules, when contested in infringement proceedings, but invalid, under another set of substantive rules, in the context of proceedings for determination of ownership (or counterclaim for invalidity of the assignment, and so on), and vice-versa. Such an outcome would be incompatible with the objectives of legal certainty and predictability underpinning the Rome I and Rome II Regulations.

The formal validity of a contractual licence or assignment of copyright is not a ‘proprietary’, but ‘contractual’ issue.

This is the part of the Opinion where the AG needs to cut through fog created by substantive law discussions. Copyright, like all intellectual property rights, is peculiar in that it creates intangible ideas into exclusive rights of ownership. Rome I, as the AG notes, implies parties should be free to choose the law applicable to their ‘contractual obligations’, yet they should not, by contrast, have that freedom with respect to the intrinsic characteristics of the copyright itself. (57, footnotes omitted)

The ‘proprietary’ category undoubtedly includes the primary issue of the transferability of the copyright concerned – that is, whether that right may be transferred at all from its initial owner (the author) to another person, in particular by effect of a contract.  Indeed, transferability  (and the extent thereof) constitutes an intrinsic characteristic of the right concerned. Accordingly, a copyright licence may only be granted, or assignment occur, through a contract if the law governing the right concerned permits it, irrespective of the law applicable to that contract.

In the case at issue, German law excludes the assignability of copyright, however it does allow exclusive licences such as the present one.

The AG suggests that if “the law governing a copyright allows its transfer by effect of a contract (which is the case here since, while German law excludes the assignability of copyright, it does allow exclusive licences such as the present one; [GAVC I do not see quite clearly why the AG does not involve Polish law in this analysis but that is probably linked to the lex loci protectionis issue and I am not an expert on intellectual property rights]), the subsequent issue of whether a contractual licence or assignment of that copyright needs to comply with certain requirements as to form (including, here, the requirement of a written document) relates, in substance, not to the right ‘as such’, but to the underlying contract.”

(73) any formidable exceptions of German law to the implications of the Polish lex contractus may be offset with a lois de police argument.

I have a feeling the CJEU wil cut to the chase more quickly but then of course AG Opinions are a good forum for more in-depth pondering.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

In Re Tesla, Inc. Derivative Litigation C.A. No. 2024-0631-BWD, the Delaware Court of Chancery the day before yesterday essentially held that a shareholder (majority) move of seat, with coinciding amendment of the corporation’s bylaws re choice of court for derivative action, implies that the newly mandated venue applies even to claims filed prior to the move taking effect.

The ratione temporis and conflit mobile question was at issue in Tesla’s successful motion to dismiss a number derivative shareholder claims, based on an (amended) forum selection clause.

As the FT Reports, “At stake were a series of investors’ lawsuits that had accused Musk of breaching his duties to Tesla’s shareholders. [Claimants] argued he had diverted resources from the EV maker to other companies in his universe, and that he had sold stock while in possession of negative information about Tesla.”

The Texas Forum Bylaw was publicly proposed with the Redomestication [change of seat, GAVC] on 17 April 2024; the derivative actions were filed on 24 May, 10 June, and 13 June, respectively. The Texas Forum Bylaw was adopted on 13 June, so by the time Defendants appeared in the Actions on 25 June and 16 September, the Texas Forum Bylaw was in effect.

The relevant Civil Procedure Rule is the US Federal Rule 12(b)(3), improper venue. Relevant authority is Ingres Corp v CA Inc: Forum selection provisions, including those contained in corporate bylaws, “are ‘presumptively valid’ and should be ‘specifically’ enforced unless the resisting party ‘[] clearly show[s] that enforcement would be unreasonable and unjust, or that the clause [is] invalid for such reasons as fraud and overreaching.’”

The judge rejected arguments that venue be determined at the time a complaint is filed, arguments which were based on federal precedent found not to apply to the case, but she also generally referred to Tesla not having exercised a unilateral right to alter its forum selection bylaw, nor having done so months after the claims were filed. The judgment concedes that Delaware courts sometimes do look to later points in time when determining venue, such as when the defendant appears or at the time a movant seeks transfer.

Vice Chancellor David emphasised however that

A stockholder does not have a vested right to litigate in a particular forum, even for claims arising from past conduct, because the contractual relationship “among the directors, officers, and stockholders formed within the statutory framework of the DGCL [Delaware General Corporation Law, GAVC]” is, “by design, flexible and subject to change in the manner that the DGCL spells out and that investors know about when they purchase stock in a Delaware corporation.”

and in essence

I am loath to second-guess Tesla stockholders’ chosen forum by purporting to weigh the advantages and disadvantages of Texas law and procedure relative to our own.

All in all an interesting comparative conflicts judgment.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

I fear this post is fairly anorak, meant for the die hard conflicts nerd. In X v Y ECLI:NL:RBMNE:2026:1482 (the usual anonymisation nonsense), essentially a claim by a Dutch consumer against a set of BVI companies, the Rechtbank Midden-Nederland, in interlocutory proceedings, ordered the BVI companies to surrender relevant data on the holder of a wallet in their crypto exchange. The wallet allegedly holds ia the crypto currencies financed by X, following a ‘boiler room’ fraud.

My reason for flagging it lies in the interesting approach of the court towards using, or not, Brussels I authority in its application of the residual space: national private international law that fills the gap where EU law does not apply. Of course whether or not to do so is the prerogative of the Member States: EU law has no bearing on it.

Here, the court [3.7] generally holds that seeing as what was A5(3) of the Brussels Convention and the Brussels Ia Regulation, served as a model for the Dutch residual rule of Article 6e CPR, CJEU authority may be used as guidance for the interpretation of the Dutch rules.

However it then [3.9] holds that the very qualification of a claim as in in tort, must not so use CJEU authorities, instead relying on lex fori: “It is not obvious that the definition of the term ‘tort’ should be aligned with the much broader interpretation given to this term by [the CJEU]”: one assumes it is thinking here of the CJEU Kalfelis formula, where the Court of Justice held [17] that ‘matters relating to tort, delict and quasi-delict’ “covers all actions which seek to establish the liability of a defendant and which are not related to a ‘contract’ within the meaning of Article [7](1).”

[3.11] the court refers to Dutch substantive lex fori to qualify the claim as one in tort.

Article 6e CPR allocates jurisdiction to the Dutch courts insofar as the “place where the harmful event occurred” took place in The Netherlands. Here the court [3.13] holds that CJEU authority can fully play, subsequently referring to CJEI Bier’s locus delicti commissi /locus damni distinction, and placing the locus damni in The Netherlands (in particular the loss of the opportunity, were the data not surrendered, of claiming compensation etc from the fraudsters.

I m not sure whether there is a PhD in the application of BIa to residual PrivIntLAw in the Member States. But there most certainly is a good paper in it.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Stichting Offlimits v X.AI llc,. X Corp, X Internet Unlimited Company ECLI:NL:RBAMS:2026:310, in which the courts at Amsterdam issued an injunction yesterday, echoes some of the themes of SEOK v Hammy Media. Of relevance to the blog of course are the jurisdictional issues.

X.AI is domiciled at Palo Alto, X Corp at Texas, and X Internet Unlimited Company ‘XIUC’ at Dublin.

[4.1] The Stichting’s claim serves two purposes, namely 1) to ensure that Grok and X no longer offer functionality that allows for the generation and  distribution of images depicting naked, existing individuals without their consent (‘non-consensual nude images’), and 2) to ensure that it is no longer possible to use Grok and X to generate and distribute child sexual abuse material ‘CSAM’.

The Stichting’s claim has a dual cause of action: the GDPR viz image generation of existing people; and tort (under Dutch law) viz CSAM (because it may also involve non-existing individuals. This has an impact on jurisdiction.

The court applies a textbook approach to jurisdiction [4.4] ff. It first assesses jurisdiction against XIUC (‘processor’ within the meaning of the GDPR): for the claim viz non-consensual nude images The Netherlands per A79(2) GDPR [as I detail here]:

Proceedings against a controller or a processor shall be brought before the courts of the Member State where the controller or processor has an establishment. Alternatively, such proceedings may be brought before the courts of the Member State where the data subject has his or her habitual residence, unless the controller or processor is a public authority of a Member State acting in the exercise of its public powers.

The Netherlands is Off Limits’ habitual residence. Territorial jurisdiction is determined by Dutch civil procedure rules – CPR and rests with the courts at Amsterdam.

For the claim viz CSAM also The Netherlands, as locus damni per A7(2) Brussels Ia.

[4.7] the court accepts jurisdiction viz the US entities on the basis of the joinder and anchor defendant rules in Dutch CPR (A8 Brussels Ia does not apply to non-EU domiciled defendants); the claims are clearly related to the Dutch claims.

[4.8] Applicable law is determined as the GDPR for the non consensual images claim and Dutch law as lex voluntatis per A14 Rome II.

Privacy lawyers will be interested to read the remainder of the judgment with the court being unimpressed with X’s argument that GROK has been fixed to avoid generation of both types of images. Claimant produced stills showing the exact opposite.

In an echo of  the right to be forgotten, the court then limits its injunction (a ban on offering GROK with the relevant functionality) viz the US entities re the non-consensual images, to offering the functionality to individuals domiciled in The Netherlands (because the US entities do not themselves offer services in The Netherlands). That limitation is not extended to XIUC: here the court orders XIUC to simply block the offer on X, of GROK with the relevant functionality.

Conclusion: the court in essence

[5.1.] prohibits X.AI from generating and/or distributing sexual imagery insofar as this involves the use of functionality whereby persons are partially or fully undressed without having given their express consent, insofar as this concerns persons domiciled in the Netherlands;

[5.2.] prohibits X.AI from producing, distributing, offering, publicly displaying and/or possessing sexual imagery in the Netherlands insofar as this involves the use of functionality whereby imagery is generated that qualifies as child pornography under Dutch law;

[5.3.] orders X.AI to confirm in writing to Offlimits that and how it has complied with the prohibitions under 5.1. and 5.2.;

[5.4.] orders X. AI to pay Offlimits a penalty of €100,000.00 for every day or part thereof that it fails to comply with (one of) the prohibitions under 5.1. or 5.2., or with the order under 5.3., until a maximum of €10,000,000.00 has been reached,

[5.5.] prohibits X from offering the functionality of Grok as part of the X platform for as long as Grok acts in breach of the prohibition under 5.1.;

[5.6.] orders X to confirm in writing to Offlimits that and how it has complied with the requirement under 5.4.;

[5.7.] orders X to pay Offlimits a penalty of €100,000.00 for every day or part thereof that it fails to comply with the prohibition in 5.5. or the requirement set out in 5.6. is met, up to a maximum of €10,000,000.00,

[5.8.] prohibits XIUC from offering the functionality of Grok as part of the X platform for as long as Grok acts in breach of the prohibitions set out in 5.1. and 5.2.;

[5.9.] orders XIUC to confirm in writing to Offlimits that and how it has complied with the prohibition under 5.6.;

[5.10.] orders XIUC to pay Offlimits a penalty of €100,000.00 for every day or part thereof that it fails to comply with the prohibition under 5.8. or the order under 5.9., up to a maximum of €10,000,000.00

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

When I first flagged the claim by Belgian farmer Hugues Falys v Total, I concluded with

“Re the jurisdictional issues see likely Total use of an argument which I flagged here: viz an emerging corporate strategy to deflect A7(2) forum damni jurisdiction, the argument that a causal link between the damage and the alleged shortcoming of the defendant needs to be shown in the claim form itself for it to ground jurisdiction.”

My short paper for Lex&Forum flags this and other issues related to climate claims and private international law; in my post on Lluya v RWE I note the shortcomings in the German Court of Appeal’s Article 7 Rome II applicable law analysis (I had earlier expanded on Article 7 here); the earlier post on Hagues Falys contains some updated bibliographical references; and Stefania Bariatti et al have a considerable volume here dealing with ESG principles relevant to the claims, including chapters on jurisdiction and applicable law.

The first instance court at Tournai /Doornik issued its interlocutory judgment yesterday, holding solely on jurisdiction and applicable law and concluding with an Article 30 Brussels Ia stay. Guillaume Croisant has posted early overview here.

In essence, the Belgian court holds it has jurisdiction on the basis of Article 7(2) of the Brussels Ia Regulation: ‘locus damni’ jurisdiction, and that Belgian law applies to the case.

Note of course immediately that the judgment, if upheld on appeal, does not of course mean that Belgian or EU courts have jurisdiction as locus damni against all oil majors wherever incorporated: Article 7(2) only applies to defendants with domicile in  the EU.

Total had essentially argued what I had predicted, namely an emphasis on the diffuse nature of the damage in climate claims. Its objection to jurisdiction for the Belgian courts on the basis of A7(2) Brussels Ia relies on the foreseeability of damage and on a restrictive interpretation of CJEU Bier. It emphasises the CJEU’s reasons for its foundational approach to Article 7’s split between Handlungsort [place of the (in)action leading to the damage] and Erfolgort: [place where the damage occurs (or may occur)]. These foundational principles, it argues, are foreseeability and proximity (of the courts with jurisdiction, to the place of damage).

The court disagrees that these foundational principles are infringed in a finding of jurisdiction in Belgium [more specifically, the local legal district: A7(2) assigns territorial and not just national jurisdiction] as locus damni. It notes p.10: precisely following CJEU Bier, Total can hardly be surprised to find itself being sued in a place other than its place of domicile; and proximity is of course met where the alleged damaged occurred slap bang in the legal district seized.

However it is on this point that I feel the court may be a touch optimistic in its application of A7(2) locus damni jurisdiction: (footnotes omitted)

TE soutient que le Tribunal doit préalablement examiner l’existence du lien causal entre le fait générateur et le dommage allégué, en ce sens qu’à défaut de preuve de lien étroit suffisant entre les deux, la competence du tribunal de céans doit être déclinée.

S’il paraît justifié de procéder à un examen prealable au stade de la competence, celui-ci ne saurait se muer en un procès avant le procès et cette question doit naturellement être soumise a un examen « prima faciae », ce qui suppose d’apprécier uniquement si les prétentions du demandeur ne paraissent pas invraisemblables.

En l’espece, il ne paraît pas invraisemblable que les EME [court shorthand for ‘climate emergencies’, GAVC] dont se plaint Monsieur FALYS aient causé le dommage qu’il soutient avoir subi et que ces mêmes phénomènes soient en lien causal avec des émissions de gaz a effet de serre.

Translated this would read

[Total] contends that the Court must first examine whether there is a causal link between the event giving rise to the claim and the alleged damage, in that, in the absence of sufficient evidence of a close link between the two, this Court must decline jurisdiction.

Whilst it appears justified to carry out a preliminary examination at the stage of determining jurisdiction, this must not turn into a trial before the trial, and this question must naturally be subject to a ‘prima facie’ examination, which involves assessing only whether the claimant’s claims do not appear implausible.

In the present case, it does not appear implausible that the climate emergencies complained of by Mr FALYS caused the damage he claims to have suffered and that these same phenomena are causally linked to greenhouse gas emissions.

The judgment cites one (scholarly) source for this contention. The court is certainly correct that there must not be a mini trial. However the CJEU with its mozaik, distributive jurisdiction approach per Shevill, does emphasise the  actual existence and demonstration of damage within the jurisdiction and one could also point to CJEU Mittelbayerischer Verlag – mutatis mutandis – as one of the recent cases circumscribing A7(2) scope. I am not saying that claimant did not in fact meet with a higher evidentiary standard. Yet with its ready acceptance and superficial burden of proof, the court imo takes the favor laesis (goodwill for the victim) implication of CJEU Bier too far.

Having established jurisdiction, the court most succinctly opts for Belgian law as the applicable law as a result of Article 7 Rome II. It does not discuss the many angles to Rome II which could have (and I understand, had actually been) raised by Total (in earlier submissions). Instead it held that Total had conceded the applicable law point should the court accept jurisdiction. I am not privy to all parties’ submissions and I was not at the hearings, yet if Total had not so in fact conceded then of course this is a clear ground for appeal.

Eventually the court decides to stay its case on the basis of Article 30 Brussels Ia’s ‘related cases’ provision. The court found sufficient connection with a parallel French proceeding against Total in Paris. I could not find further detail on that case (with claim number, “role n°” 22/03403) hence I cannot say much about it, yet of course an A30 stay is very much to the discretion of the judge seized. The Belgian judges note p.17 that while claimants are different, the objective of both the French and the Belgian proceedings are the same:

to secure an order requiring TOTAL ENERGIES to reduce its GHG emissions and fossil fuel production, and to refrain from making any new investments in projects related to these, on a global scale.

Given that both cases ultimately seek to compel the company to reduce its GHG emissions and fossil fuel production on a global scale, the judges hold that there is a genuine risk of irreconcilable outcomes, and seeing as Paris is expected to rule in mid (European) spring,they stay the Belgian case. The court emphasises in parting that this is merely to secure sound administration of justice as per Brussels Ia’s lis pendens requirements, that it reaffirms its findings on jurisdiction, and that is is not merely bowing to the Paris courts.

The judgment will be appealed at any event. I wonder whether upon appeal the same approach will be taken re burden of proof (or a reference made to the CJEU; current court saw no need at all); and whether there will be more engagement with Article 7 Rome II, which will be most welcome. Of course by the time of an appeal, the lis pendens issue will have become moot, lest any spring Paris judgment will in turn be appealed.

Geert.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

Probably because I am reporting this case in the middle of a very busy few months, the title of this post as not as punchy as one might have expected. However at least the post does what it says on the tin.

Thank you Ludo Veuchelen en Urs Rybi for alerting me yesterday to Eckart von Malsen’s post on the judgment of the first instance court at Paris in Sherpa v Yves Rocher. Yves Rocher has been found to have fallen short in their duty to ensure respect for the freedom of association in their supply chain.

Un grand merci to the clerks at the tribunal who sent my copy of the judgment most most swiftly.

Eckart summarises the main background as

Out of 81 former employees, only nine who had not signed a 2019 settlement agreement retained standing, together with Turkish union Petrol‑Is and French NGOs Sherpa and Action Aid. The court found that these nine workers were dismissed because of their trade union membership, that Yves Rocher had enough information to identify a serious risk to freedom of association, and that proper inclusion of this risk in the vigilance plan would have prevented the harm up to 2019.​

Yves Rocher is the French mother corporation of the Turkish group subsidiaries. In that respect the case differs from eg Dyson, which involved supply chain due diligence at arm’s length.

Of interest to the blog, are the conflict of laws elements. Yves Rocher argue that the law which applies to the employees non-contractual claim, is Turkish law as a result of the Rome II Regulation, it being the lex loci damni: the law of the place where the damage took place.

In its view, the French Droit de Vigilance, France’s main supply chain due diligence statute, was not intended as a lois de police or ‘overriding mandatory law’ rule per A16 Rome II. And that even if it was so intended, that that  displacement has no effect on the conflict of laws elements.

Yves Rocher also argue that the EU’s Corporate Sustainability Due Diligence Directive 2024/1760, CS3D for short, has no bearing on the interpretation of the French devoir de vigilance, and that even if it does, the EU’s Omnibus  Directive (now published as Directive 2026/470) which was being debated at the time of the hearings, most in turn have an impact seeing as at intended to remove (as I discussed at the time) and in the end did so remove A29(1) (7)’s provisions on civil liability and lois de police.

The result of the application of Turkish law notably would impact the statute of limitation. Defendants argue,all claims have been served out off time.

The court refers to CJEU HUK -Coburg for its discussion of lois de police under Rome II (and on a side-note, Da Silva Martins) concluding [56] that the CJEU applies a strict interpretation to A16 Rome II, requiring both a sufficient nexus with the forum, and a law considered to protect an interest fundamental to society in the forum.

The court then reviews the many references in the travaux of the Droit de Vigilance, to international human rights, OECD and other soft-law in the area, and [63] to one of the conclusions of the French Government when it proposed the law, that it should be ‘loi impérative’, designed to protect the weaker in society. [75] it concludes that the duty of care introduced by the Act [Article L225-102-2] is a lois de police within the meaning of Rome II.

[76] it refers to the sentiment of the CS3D echoing the intention of the Droit de Vigilance and [77] to Omnibus’ deletion of A29(1) and (7) not having taken away Member States’ continued possibility to regard their own due diligence laws, whether or not taken in application of the CS3D, as being lois de police.

The remainder of the judgment then firstly holds the French 5 year statute of limitation has not been exceeded, and it extensively deals with the question whether Yves Rocher have fallen short of their duties of vigilance. It concludes it has.

The judgment confirms what many including myself have suggested: Omnibus has not increased predictability for corporations. It has decreased it. Those who believe the deletion of A29(1) and (7) has somehow improved legal certainty for corporations, simply err.

Geert.

EU Private International Law, 4th ed 2025, Chapter 7.

Where does time go…(those who follow me on Linked-in know where it went in January).

I flagged Richard de la Tour AG’s Opinion in C‑791/24 TERVE  Production spol. s r. o. v Intesa Sanpaolo Holding International SA when it came out at the time. I am sorry I am only reviewng it now.

E.ON Czech Holding of course features prominently.

The case concerns a dispute re the protection of minority shareholders which did not vote for the delisting of a company (VÚB) in which they hold shares. Slovak law provides that, when such a delisting is decided upon, a takeover bid must be made and that, after the expiry of that bid, the company, or its majority shareholder which made the bid on its behalf, has the right to require the delisting of the remaining shares. Conversely, after the expiry of that period, the minority shareholders may require the purchase of their shares.

Intesa of Luxembourg initiated a share takeover bid, loco VÚB, and, at the end of the three-month validity period, exercised its right of squeeze-out, which resulted in the transfer of the remaining shares into its name. At the same time, TERVE, a minority shareholder opposed to the sale, to exercise its right of sell-out, sent Intesa a draft share purchase agreement, which Intesa did not approve. Intesa justified its refusal on the ground that, since TERVE had not approved its own squeeze-out proposal within the mandatory three-month period, TERVE’s remaining shares had been transferred to Intensa and TERVE therefore no longer had the standing to bring proceedings as a shareholder and had forfeited the opportunity to accept the squeeze-out offer.

The case concerns two types of claims:

The first cause of action seeks to substitute the majority shareholder’s approval of a draft share purchase agreement, presented by the minority shareholders, with a court ruling.

The AG opines this falls within Article 7(1) Brussels Ia’s forum contractus – not A7(2) forum delicti.

The question aims essentially to determine whether the fact that Intesa freely assumed the obligation to make a mandatory takeover bid to purchase
shares in lieu of the issuer of the shares (VÚB, a. s.), in a situation where that
voluntary assumption of the obligation to make a bid preceded and was a
condition for the respondent’s subsequent exercise of the right of squeeze-out, may at the same time be considered as a voluntary assumption of an obligation vis-à-vis Terve (and possibly also vis-à-vis the other shareholders of
VÚB, a. s, who did not vote in favour of delisting their shares) to purchase
shares pursuant to Paragraph 118j of the Slovak Securities Law – a contract whose forum contractus is Slovakia.

Intesa argues there is neither a contract nor A7(2) jurisdiction, the latter it argues requiring ‘civil liability’.

The AG refers to the usual suspects such as CJEU Feniks to come to a wide notion of contract for the purposes of Article 7(1), and dismisses the dense statutory context of corporate take-overs and squeeze-out as being determinant.

His reference (41) in particular to Intesa’s voluntary launch of the take-over bid instead of the corporation itself is confusing (and may indeed lead to the forum shopping consequences he signals. What he really seems to consider crucial (41) is that VÚB’s relationship with its shareholders is contractual in nature. That is the kind of test which I called the ‘ancestry’ test in particular when reviewing Sharpston AG’s Opinion in Ergo.

Unfortunately the AG’s answer to the forum contractus question means he does not entertain the forum delicti referral.

The other cause of action relates to a claim viz the invalidity of a resolution of a general meeting which decided to transfer the shares of the minority shareholders following the exercise of the right of squeeze-out by the majority shareholder. The response with respect to that claim determines the standing of those minority shareholders to bring proceedings, and the AG opines it is caught by Article 24(2)’s exclusive jurisdictional rule.

(47) the AG first suggests (as does the Commission seemingly) that this question is relevant only if, under Slovak law, the court hearing an application to substitute a majority shareholder’s approval of a sell-out proposal with a court ruling, can rule on the validity of a decision of a general meeting without the company whose decision is being challenged being a party to the proceedings. I assume this is the case for otherwise the decision may be impossible to be imposed upon that company.

(55-57) are the core paras (with reference before and in the paras to the classics such as BVG, Hassett and Doherty, Kerr: In the paras below, I have deleted the references

55.      Consequently, it is necessary to examine whether, in the context of a legal action seeking to substitute the approval of the sell-out proposal with a court ruling, the preliminary issue of the validity of the decision of the general meeting to transfer the remaining shares to the person exercising the right of squeeze-out can constitute the principal subject matter of the dispute.

56.      TERVE’s standing to bring proceedings depends directly on whether or not the resolution of the second general meeting is valid since, if that resolution is invalid, the transfer of the shares to Intesa is called into question and TERVE is still a shareholder of VÚB. However, if that resolution is valid, TERVE has lost its status of shareholder and can no longer exercise its right of sell-out.

57.      I therefore consider that the validity of the resolution of the second general meeting is an essential premiss for TERVE’s action to substitute Intesa’s approval of its draft agreement to purchase the remaining shares with a court ruling.  It is actually that resolution which establishes the right of sell-out for the other shareholders. The situation in the present case is not one in which the alleged invalidity of the decisions of the company’s organs is ancillary to the principal subject matter of the action. Accordingly, the Slovak courts have jurisdiction on the basis of the exclusive jurisdiction provided for in Article 24(2) of Regulation No 1215/2012.

(60) in its own  delisting proposal, Intesa referred to Slovak courts’ jurisdiction: hence being sued there can hardly have been unexpected.

I do hope the CJEU wil take the opportunity of the forum contractus question to refine its formula on Article 7(1).

Geert.

EU Private International Law, 4th ed. 2024, 2.192 ff, 2.405 ff.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

The Opinion of Advocate General Rantos in C-232/25 [Idziski] (at the time of writing not available in English) focuses on eDate and Bolagsupplysningen’s ‘centre of interests’ test in the context of Article 7(2) Brussels Ia jurisdiction for infringement of personality rights. (The case formally concerns A5(3) Brussels I however there is no material difference).

Applicants Z.R. (an indivdual) and Ś. (an organisation representing former soldiers’ interests) claims an infringement of personality rights by a series coproduced by the defendants, who are domiciled in Germany. The contested series traces the fates of a number of characters (a friendship group of young Germans) during World War II. Part of the story involving one of the main characters takes place in the Polish territories under German occupation. A crucial aspect of this plot strand is a meeting between this character and the underground military unit mentioned in the questions referred for a preliminary ruling; the soldiers belonging to the relevant military unit make clear their attitudes towards Jews during this meeting.

Names etc are anonymised in the proceedings however it does not take long to figure out the series concerned is Unsere Mütter, unsere Väter – aired on Polish TV but also available for streaming. Lower courts’ findings and orders including for public apologies are all being appealed and current reference by the Supreme Court enquires about the jurisdictional issue.

Claimants argue centre of interests per CJEU eDate, and argue there is identifiability, required per CJEU Mittelbayerisher Verlag. They distinguish that case by referring to the fact that the content infringing personality rights pertained to an entire nationality, and the applicant invoked membership of this nationality; in the present case, however, the plot of the film describes the behaviour of a narrower and closed group of individuals, namely the soldiers of the relevant military unit, and unambiguously and undeniably identifies this unit.

The alleged infringement of personality rights follows not from a posting on the internet (prima facie required per CJEU eDate and confirmed in CJEU Wintersteiger) rather from a cinema and film broadcast on terrestrial TV in a number of Member States, including Poland, whereby the applicants had their centre of interests in Poland. The Polish SC essentially wonders whether the ‘internet’ element is still relevant seeing as technological advances have led to a blurring of the boundaries between traditional forms of long-distance broadcasting and the distribution of the content via the internet.

The AG (53) firstly suggests the relevant technological context is the 2015 one, when the claim was formulated: not the 2026 one.  I for one would suggest his opinion should hold even in 2026 circumstances.

(56) He justifiably reminds us again of the need to interpret all exceptions to Article 4 Brussels Ia domicile jurisdiction restrictively, and in the specific context at issue, that not all territorially broadcasted content necessarily ends up online; reviewing it ad hoc would not serve predictability. (59) This leads to bifurcation: for the terrestrial part of the broadcast, CJEU Shevill is the relevant authority, and for the internet part, eDate. As for the identifiability, the AG essentially reminds the judge of the Mittelbayerischer Verlag criteria and (67) suggests a presumption against such identifiability.

The second preliminary question is relevant only if (as indeed the AG suggests) the answer to the first question likely leads to non-applicability of the eDate route hence only limited (A7(2) locus damni) jurisdiction for the   Polish courts. Might it then encompass at least some of the nonpecuniary claims lodged by the applicants: that specific information be displayed whenever the film is broadcast and that a corresponding statement be published. The AG sees no issue with that, provided the territorial consequences are limited.

(80) the AG ends with the remark that the mozaik approach is not ideal for claimants at issue and in similar cases, however he suggests, rightly in my view, that such is the consequence of long-standing authority and not something the CJEU rebus sic stantibus (statutory intervention being an issue for the legislator) should alter.

A sensible Opinion.

Geert.

This post has been written by Dr Mykola Lazeranko, postdoctoral fellow at KU Leuven Law. I am most grateful for his contribution. Of note is that Mykola’s aim is not to highlight whether the Russian courts are correct in their approach; rather to document their private international law methodology.

Geert.

Introduction

Mid-December 2025, news outlets across Europe and beyond reported that the Central Bank of the Russian Federation had initiated proceedings against Euroclear in the Moscow Arbitration Court, seeking approximately USD 229 billion in relation to frozen Russian assets (see Guardian, Reuters, VRT news, etc.). On 17 December 2025, the court issued a ruling accepting the statement of claim for consideration and scheduled a court hearing for 16 January 2026 (see here).

On 16 January 2026, the court issued several procedural rulings. In particular, it dismissed Euroclear’s application for strike-out (formally ‘application to have the statement of claim left without consideration’), it granted Euroclear’s application to adjourn the preliminary hearing, and it granted the application of the Central Bank of the Russian Federation for the proceedings to be conducted in camera on the grounds that the case involves secreta commercii.

The next hearing is scheduled for 4 March 2026.

While the European Commission earlier dismissed the claim as ‘speculative’ and groundless, the litigation forms part of a broader pattern: since 2022, Russian courts have witnessed a steady increase in claims against Euroclear arising from sanctions-related asset freezes. By mid-January 2026, just short of 200 claims against Euroclear had been filed across Russia.

This post examines the private international law (PIL) dimension of this litigation, using the core, ‘trigger’ case of Bank Saint Petersburg v Euroclear. Its aim is to analyze how the courts grounded their reasoning to reject arguments of Euroclear that EU and Belgian law is applicable, and how it built sanctions-related case-law against Euroclear based on Russian law.

The Trigger Case: Bank “Saint-Petersburg” v. Euroclear

The Euroclear-related litigation began with case No. А40-205635/2022, filed in September 2022 before the Moscow Arbitration Court by Bank “Saint-Petersburg” PJSC.The bank sought compensation for losses it attributed to Euroclear’s blocking of funds following the imposition of EU sanctions.

In February 2023, the court of first instance granted the claim, awarding the bank USD 107,085,768.65 and EUR 488,994.50 in damages, alongside litigation costs. On appeal, the claimant withdrew the euro-denominated portion, and in May 2023, the Ninth Arbitration Court of Appeal upheld the decision for the USD portion. It also clarified that payment should be made in Russian roubles according to the exchange rate on the date of payment.

From the outset, the issue of applicable law was central. Euroclear stated that the dispute fell under EU law and Belgian law, invoking the regulatory framework governing its activities and the contractual arrangements between Euroclear and the Russian National Settlement Depository (NSD).

Consolidation of Defendants: From NSD and Euroclear to Euroclear Alone

Initially, the claim was brought jointly against Russia’s National Settlement Depository – NSD and Euroclear. From a PIL perspective, such a combination of domestic and foreign defendants could have implications for both jurisdiction and applicable law.

However, during the proceedings the claimant withdrew its claim against NSD. The court further emphasised that NSD could not be a defendant in this case because the claimant and NSD were bound by an arbitration agreement providing for dispute resolution before the Arbitration Centre at the Russian Union of Industrialists and Entrepreneurs. Consequently, the claims against NSD were left without consideration, and Euroclear remained the sole defendant.

The court made it clear that the mere involvement of a foreign party does not provide grounds to disregard an arbitration agreement. Had the claims against NSD and Euroclear been brought separately, the court noted, those would not have fallen within its jurisdiction.

PIL: First Instance

 Public Order as a Consideration

Relying on Article 1193 of the Russian Civil Code, the court noted that foreign law must yield to Russian law when the consequences of its application would have obviously been in conflict with the fundamentals of law and order (public order, ordre public) of the Russian Federation. The court stressed that when deciding whether it is necessary to apply the ordre public exception, the court should proceed not from a contradiction between the content of the foreign rule and the fundamental principles of the legal order (i.e. not from a review of the substance of foreign law), but from the unacceptability, for the forum state, of the consequences resulting from the application of the foreign law provision.

The court also stated that according to the Presidium of the Supreme Arbitration Court (informational letter No. 156 of 26 February 2013), ‘public order’ encompasses fundamental legal principles of the highest imperative authority, universality, and particular social and public significance, forming the foundation of the state’s economic, political, and legal system. Among these principles is the prohibition of actions explicitly forbidden by peremptory norms of Russian law (Art. 1192 of the Civil Code) when such actions threaten the sovereignty or security of the state.

Applying this standard, the court held that EU sanctions preventing the transfer of funds to the bank were incompatible with fundamental constitutional principles. Under Article 55(3) of the Russian Constitution, restrictions on the exercise of rights by Russian legal entities may be imposed only by federal law. Russian legislation does not compel compliance with foreign sanctions, hence, judicial enforcement of EU sanctions would contravene Russian public order.

The court stated that liability for harm in Russian law is a private-law construct, grounded in principles of fairness, proportionality, and fault. Proportionality, directly mentioned in the aforementioned informational letter No. 156 of 26 February 2013 of the Presidium of the Supreme Arbitration Court, is considered part of Russian public order.

The court further emphasised the prohibition of abuse of rights. Drawing on the Resolution of the plenum of the Russian Supreme Court of  23 June 2015, No. 25 “About application of some provisions of the Section I of part one of the Civil code of the Russian Federation by courts”, it noted that exercising a right in a manner that causes harm to others, particularly through unlawful means, constitutes abuse. Material damage, including the diminution of economic value or the need for additional expenditures, falls within this category. The court cited the ruling of the Supreme Court of the Russian Federation dated 28 November 2017, No. 309-ЭС-13269, in case No. А07-27391/2016, indicating it as a source where it is noted that the breach of the prohibition on abuse of rights constitutes violation of the principles of public order of the Russian Federation.

The court also relied on statutory and constitutional authorities concerning sanctions: the Federal Law “On Measures (Countermeasures) in Response to Unfriendly Actions of the USA and (or) other Foreign States”, which identifies foreign sanctions targeting Russia as threatening territorial integrity and economic stability, and the ruling of the Constitutional Court of the Russian Federation dated 13 February 2018 No. 8-П, where the court held that a right whose exercise depends on compliance with sanctions imposed against Russia or its economic entities by any state, outside proper international legal procedures and contrary to multilateral treaties to which Russia is a party, cannot be protected by the courts.

Non-Contractual Nature of the Claim

Euroclear argued that Belgian law should govern because of the contractual arrangements with NSD. The court rejected this, noting that the claims were non-contractual in nature, arising from harm caused by Euroclear’s actions, and therefore outside the scope of any contractual choice of law. The applicable law must be determined according to the conflict-of-law rules governing obligations arising from harm.

Lex Loci Damni and Foreseeability

Article 1219 of the Civil Code provides that obligations arising from harm are governed by the law of the place where the harmful act occurred, or, if the harm occurs elsewhere, the law of the place where the damage materialised, provided the harm was foreseeable.

The court found that the damage occurred in Russia, and Euroclear knew or ought to have known that blocking the funds would harm Russian entities. Sanctions were expressly directed at Russia, and Euroclear had been informed of the impact.

The court referred to paragraph 52 of the Plenum Resolution of the Supreme Court of the Russian Federation No. 24 dated 9 July 2019, “On the Application of Private International Law Norms by Courts of the Russian Federation”, which provides that if a claim arises from harm caused by an act or other circumstance that occurred on the territory of the Russian Federation, or if the harm materialized on the territory of the Russian Federation, the court may apply Russian law to the relations between the parties.

Accordingly, the court stated that since the consequences of the harm caused by the actions (or inaction) of Euroclear occurred on the territory of the Russian Federation, and Defendant was aware of the location where the harm would materialize, the applicable law governing these relations is the law of the Russian Federation.

Closest Connection

Finally, the court applied the subsidiary rule of closest connection (Article 1186 of the Civil Code), considering that the legal relationship is most closely connected with the Russian Federation for the following reasons:

• • Bank Saint Petersburg, as a banking organization registered in the Russian Federation, is claiming damages (actual loss) in its favour to fulfill its obligations to a wide range of parties, including both individuals and legal entities;
  • The harm caused by the unjustified actions (or omissions) of Euroclear since March 2022 is substantial, affecting not only the claimant but also specific individuals and legal entities, as well as the economy of the state;
  • The issuers of the securities for which funds were not transferred to the Bank are predominantly Russian entities;
  • The agreement between the NSD and the Bank is governed by Russian law, and NSD’s obligations were not fulfilled due to Euroclear’s actions (or omissions) that were inconsistent with Russian legislation.

These factors, the court concluded, outweighed the arguments relied upon by Euroclear.

PIL: Appellate Review

 In its May 2023 judgment, the Ninth Arbitration Court of Appeal endorsed the reasoning of the first-instance court concerning the PIL, further systematising its approach.

Emphasising Articles 247–249 of the Arbitration Procedure Code and Supreme Court Plenum Resolution No. 23 (2017), the appellate court further assessed jurisdiction and connection to Russia:

• • The bank’s main activity is in Russia, oriented toward Russian clients.
  • Key evidence resides with NRD, Russia’s central depository, subject to Russian law and supervision.
  • The securities involved were issued by Russian entities, and Euroclear’s actions had a significant economic impact domestically.
  • Euroclear maintains accounts in Russia.

The court in its judgment indicated that with regard to the present case, the arbitration court of the Russian Federation has jurisdiction to hear the Bank’s claims on the basis of the following:

• • Bank “Saint-Petersburg” PJSC, as a banking institution registered in the Russian Federation, seeks recovery of losses in its own favour for the purpose of performing its obligations towards a wide range of persons, including both individuals and legal entities.
  • The claimant’s principal business activities are carried out in the territory of the Russian Federation and are oriented towards Russian individuals and legal entities.
  • The damage caused to the Bank by the unjustified actions (or omissions) of Euroclear Bank SA/NV, commencing in March 2022, is substantial and affects not only the claimant, but also its clients and the economy of the state.
  • The main evidence confirming the need to transfer funds in favour of Bank “Saint-Petersburg” PJSC is held by the NSD on the territory of the Russian Federation. NSD, as a Russian legal entity, possesses the key evidence in the present dispute.
  • The obligations arising in connection with the Eurobonds for which funds were not transferred to the Bank involve predominantly Russian issuers.
  • The agreement between NSD and the Bank is governed by Russian law, and NSD’s obligations were not performed as a result of actions (or omissions) of Euroclear Bank SA/NV that were inconsistent with Russian legislation.
  • NSD, as the party obliged to transfer funds to the Bank, is registered in the Russian Federation, holds the status of Russia’s central securities depository, and its activities are regulated by Russian law and supervised by the Central Bank of the Russian Federation.
  • Euroclear Bank SA/NV maintains bank accounts within the territory of the Russian Federation.

The appellate court stated that the court of first instance in its consideration of the case correctly established that the law applicable to the present dispute is the law of the Russian Federation, and that the arbitration court of the Russian Federation has jurisdiction to hear the Bank’s claim against Euroclear. As  indicated in its ruling, the occurrence of damage within the territory of the Russian Federation constitutes an additional basis for the jurisdiction of the Russian arbitration court.

Overall, the appellate court’s decision did not introduce new substantive arguments but, to some extent, supported and further formalised the PIL reasoning, creating a guide for the large number of Euroclear-related claims in Russia.

Conclusion

The Bank Saint Petersburg v Euroclear litigation demonstrates the Russian courts’ interpretation and application of PIL principles in a sanctions context. Key takeaways include:

1. 1. Public order (ordre public) considerations override foreign law where enforcement would conflict with fundamental constitutional or statutory principles.
   2. Non-contractual claims arising from harm are governed by the law of the place where harm occurs, rather than contractual choice-of-law clauses.
   3. Lex loci damni and foreseeability determine applicable law when harm has consequences in another state.
   4. Closest connection serves as a subsidiary test, weighing factors such as the parties’ domicile, the location of harm, and the economic and legal context.

In Euroclear-related disputes, the Russian courts have constantly rejected arguments delivered by Euroclear to apply EU and Belgian law, establishing Russian law as the governing law for sanctions-related claims, which represents the interplay between private international law and the enforcement of cross-border sanctions.

At present, the court register shows around 190 cases across the Russian Federation against Euroclear. All of these cases were initiated after the decision in Bank Saint Petersburg v Euroclear, and in the majority of them final judgments have already been rendered (as a rule, the claims against Euroclear are upheld). The reasoning supporting the jurisdiction of the Russian courts and the application of Russian law in all such cases mirrors that adopted in this ‘trigger’ case.

Mykola.

If you do use the blog for research or database purposes, citation would be appreciated, to the blog as a whole and /or to specific blog posts. Many have suggested I should turn the blog into a paid for, subscription service however I have resisted doing so. Proper reference to how the blog is useful to its readers, will help keeping this so.

I have given background to Limby v Dyson when I discussed the important jurisdictional decisions. Current post deals with case management issues held by Pepperall J in Limbu & Ors v Dyson Technology Ltd & Ors [2026] EWHC 38 (KB).

The judgment is crucial for the discussion of equality of arms in supply chain due diligence claims, an issue which as the judge in current case also points out, was part of the forum non considerations at the time of the jurisdictional challenge. The judge’s considerations [8] re access to justice, costs, realism viz law firms contingency fees (and a side note on the barristers in the case, who of course act as single, self-employed risk takers viz the recoverability of their time spent) are an example of realism, empathy, concern for the rule of law and appreciation of real-life impact.

A first important issue under consideration comes down to whether the judge would order the case to first proceed with lead claimants, having to prove their working conditions etc, or rather, as defendants suggested, cut to the chase (claimants’ counsel called it ‘answering the siren song) as to Dyson’s knowledge and control of the circumstances in which claimants found themselves, with factual circumstances assumed for the purpose of the liability trial. [31] ff the judge explains while he rejects the latter (incl the difficulty of any appeal judges having to hear an appeal on the basis of assumed rather than proven facts) and he distinguishes the case management in Municipio on this point. [44] ff the claim in unjust enrichment is not going to trial now, for here the judge finds it impossible to split the issues of liability for and the assessment of the extent of any unjust enrichment (the latter requiring costly forensic evidence).

Focusing on one more element of the case-management: the request, which was granted, for ‘specific’ disclosure of a number of specific documents, even prior to what is known as ‘first disclosure’. There is an asymmetry of information between the parties as to what was reported to and known by Dyson. The judge ordered disclosure of 5 specific documents (the existence of which was revealed in related libel proceedings against Channel 4 journalists).

A judgment of much note, and another round hard fought by claimants, underscoring their need for tenacity and for support by level-headed lawyers.

Geert.

EU private international law, 4th ed. 2024, Chapter 7.