Provisional text

OPINION OF ADVOCATE GENERAL

EMILIOU

delivered on 22 February 2024(1)

Case C‑339/22

BSH Hausgeräte GmbH

v

Electrolux AB

(Request for a preliminary ruling from the Svea hovrätt (Svea Court of Appeal, Stockholm, Sweden))

(Reference for a preliminary ruling – Area of freedom, security and justice – Judicial cooperation in civil matters – Jurisdiction and the recognition and enforcement of judgments in civil and commercial matters – Regulation (EU) No 1215/2012 – Exclusive jurisdiction – Proceedings concerned with the validity of patents – Article 24(4) – Scope – Infringement proceedings – Invalidity of the patents allegedly infringed raised as a defence – Consequences on the jurisdiction of the court seised of the infringement proceedings – Patent registered in a third State – ‘Reflexive effect’ of Article 24(4))

I.      Introduction

1.        The present request for a preliminary ruling of the Svea hovrätt (Svea Court of Appeal, Stockholm, Sweden) concerns the interpretation of Regulation (EU) No 1215/2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters. (2)

2.        By its questions, the referring court seeks clarification, first, about the jurisdiction of the courts of the Member States of the European Union, under that regulation, to hear actions concerned with the infringement of patents registered in other Member States, particularly where the validity of the patents allegedly infringed is challenged by the opposing party. As I will explain in this Opinion, significant uncertainty surrounds that question as a result, notably, of an ambiguous decision delivered by the Court a long time ago, namely the judgment in GAT. (3) The present reference provides the Court with the opportunity to confirm one of several possible readings of that decision.

3.        Secondly, the Court is invited to clarify whether Member State courts have jurisdiction to hear proceedings concerned with the validity of patents registered in third States. In that respect, the Court will have to address the delicate and long-standing issue of whether certain rules of the Brussels I bis Regulation apply to ‘external’ situations in the same way as they apply to ‘intra-EU’ conflicts of jurisdiction, or have a ‘reflexive effect’, as will be explained in this Opinion.

II.    Legal framework

A.      International law

4.        The Convention on the Grant of European Patents, which was signed in Munich (Germany) on 5 October 1973 and entered into force on 7 October 1977, in the version applicable to the facts in the main proceedings (‘the EPC’), establishes, as stated in Article 1, ‘a system of law, common to the Contracting States, for the grant of patents for invention’.

5.        Article 2(2) of the EPC provides that ‘the European patent shall, in each of the Contracting States for which it is granted, have the effect of and be subject to the same conditions as a national patent granted by that State’.

B.      The Brussels I bis Regulation

6.        Article 4(1) of the Brussels I bis Regulation states that ‘subject to this Regulation, persons domiciled in a Member State shall, whatever their nationality, be sued in the courts of that Member State’.

7.        Article 24 of that regulation, entitled ‘Exclusive jurisdiction’ provides, in its paragraph 4:

‘The following courts of a Member State shall have exclusive jurisdiction, regardless of the domicile of the parties:

…

(4)      in proceedings concerned with the registration or validity of patents …, irrespective of whether the issue is raised by way of an action or as a defence, the courts of the Member State in which the deposit or registration has been applied for [or] has taken place …

Without prejudice to the jurisdiction of the European Patent Office under [the EPC], the courts of each Member State shall have exclusive jurisdiction in proceedings concerned with the registration or validity of any European patent granted for that Member State’.

C.      Swedish law

8.        Paragraph 61, second subparagraph, of the Patentlagen (1967:837) (Patents Law) provides that ‘if an action concerning patent infringement is brought and the person against whom the action is brought claims that the patent is invalid, the question of invalidity may be considered only after an action to that effect has been brought. The court shall order the party claiming that the patent is invalid to bring such an action within a specific period’.

III. Facts, national proceedings, the questions referred and the procedure before the Court

9.        BSH Hausgeräte GmbH (‘BSH’) is the holder of European patent EP 1 434 512, protecting an invention related to vacuum cleaners, granted for (and accordingly validated in) Austria, France, Germany, Greece, Italy, the Netherlands, Spain, Sweden, Türkiye and the United Kingdom.

10.      On 3 February 2020, BSH brought an action against Aktiebolaget Electrolux (‘Electrolux’), a company registered in Sweden, before the Patent- och marknadsdomstolen (Patents and Market Court, Sweden). That action is based on Electrolux’s alleged infringement of EP 1 434 512 in the various States for which it had been granted. In that context, BSH seeks, inter alia, an injunction prohibiting Electrolux from continuing to use the patented invention in all those States and damages for the harm caused by that unlawful use.

11.      In its statement of defence, Electrolux contended that the Patent- och marknadsdomstolen (Patents and Market Court) should dismiss that action to the extent that it concerns the Austrian, French, German, Greek, Italian, Netherlands, Spanish, Turkish and UK parts of EP 1 434 512 (‘the foreign patents’). In that respect, Electrolux pleaded, inter alia, the invalidity of the foreign patents.

12.      Furthermore, Electrolux argued that, in the light of that defence, the Swedish courts have no jurisdiction to hear and determine the infringement proceedings to the extent that the foreign patents are concerned. In that respect, the infringement proceedings should be regarded as ‘concerned with the … validity of patents’ within the meaning Article 24(4) of the Brussels I bis Regulation and, pursuant to that provision, the courts of the different Member States where those patents have been validated are exclusively competent to hear the case in so far as ‘their’ patent is concerned.

13.      In response, BSH submitted that the Swedish courts have jurisdiction to hear the infringement proceedings pursuant to Article 4(1) of the Brussels I bis Regulation, as Electrolux is domiciled in Sweden. Article 24(4) thereof is not applicable, since the action brought by BSH is not, in itself, ‘concerned with the … validity of patents’ within the meaning of that provision. Moreover, pursuant to the second subparagraph of Paragraph 61 of the Patentlagen, where the defendant pleads, in such infringement proceedings, that the patent is invalid, the court seised must order him or her to bring a separate action to that effect before the competent courts. In this case, Electrolux should thus bring separate invalidity proceedings before the courts of the different States for which the foreign patents were granted. In parallel, the Patent- och marknadsdomstolen (Patents and Market Court) could determine the infringement matter in a provisional judgment and then stay proceedings pending a final judgment in the invalidity proceedings. Finally, with respect to the Turkish part of EP 1 434 512, BSH submitted that Article 24(4) of the Brussels I bis Regulation is, in any case, not applicable to patents issued by third States and therefore it cannot have any bearing on the jurisdiction of the Swedish courts.

14.      By decision of 21 December 2020, the Patent- och marknadsdomstolen (Patents and Market Court) dismissed the action in respect of the infringement of the foreign patents. While, at the moment of bringing the proceedings, the Swedish courts had jurisdiction to hear the action under Article 4(1) of the Brussels I bis Regulation, Article 24(4) of that regulation became applicable when Electrolux pleaded the invalidity of those patents as a defence. Under that provision, the courts of other States have exclusive jurisdiction to consider the validity issue, and as that issue is crucial for the outcome of the infringement action brought by BSH, the national court declared that it lacked jurisdiction over the proceedings to the extent that the foreign patents are concerned. That court also declined jurisdiction with respect to the Turkish patent, taking the view that Article 24(4) expresses an internationally accepted principle of jurisdiction, whereby only the courts of the State that granted a patent can adjudicate its validity.

15.      Subsequently, BSH filed an appeal against that decision before the Svea hovrätt (Svea Court of Appeal, Stockholm), maintaining that Article 24(4) of the Brussels I bis Regulation does not apply to actions for infringement of patents. Nevertheless, because Electrolux pleads invalidity as a defence, jurisdiction is distributed: the Swedish courts have jurisdiction under Article 4(1) to rule on the infringement matter, while the validity issue must be determined by the courts of the States of registration under Article 24(4). The Swedish courts are also competent with respect to the Turkish patent on the basis of Article 4(1) of that regulation. Indeed, the jurisdiction of the State where a defendant is domiciled is a recognised principle in international law. Electrolux maintained, for its part, that Article 24(4) applies to infringement proceedings in which invalidity is pleaded as a defence. The Swedish courts have no jurisdiction over the proceedings as a whole, since the infringement and validity issues cannot be separated.

16.      It is in those circumstances that the Svea hovrätt (Svea Court of Appeal, Stockholm) decided to stay the proceedings and to refer the following questions to the Court of Justice for a preliminary ruling:

‘(1)      Is Article 24(4) of [the Brussels I bis Regulation] to be interpreted as meaning that the expression “proceedings concerned with the registration or validity of patents … irrespective of whether the issue is raised by way of an action or as a defence” implies that a national court, which, pursuant to Article 4(1) of that regulation, has declared that it has jurisdiction to hear a patent infringement dispute, no longer has jurisdiction to consider the issue of infringement if a defence is raised that alleges that the patent at issue is invalid, or is the provision to be interpreted as meaning that the national court only lacks jurisdiction to hear the defence of invalidity?

(2)      Is the answer to Question 1 affected by whether national law contains provisions, similar to those laid down in the second subparagraph of Paragraph 61 of the [Patentlagen], which means that, for a defence of invalidity raised in an infringement case to be heard, the defendant must bring a separate action for a declaration of invalidity?

(3)      Is Article 24(4) of the [Brussels I bis Regulation] to be interpreted as being applicable to a court of a third [State], that is to say, in the present case, as also conferring exclusive jurisdiction on a court in [Türkiye] in respect of the part of the European patent which has been validated there?’

17.      The present request for a preliminary ruling, dated 24 May 2022, was lodged on the same day. Written observations have been submitted by BSH, Electrolux, the French Government and the European Commission, and the interveners were represented at the hearing that took place on 22 June 2023.

IV.    Analysis

18.      The present case concerns the jurisdiction of Member State courts to hear actions for infringement of European patents allegedly committed in multiple States. Before dissecting the questions referred to the Court, I find it appropriate to provide the reader, who may not be familiar with the intricacies of this complex area of law, with an overview of the relevant substantive and jurisdictional rules.

19.      Generally speaking, patents are intellectual property rights issued by States following registration procedures carried out by national patent offices, in accordance with the requirements of granting (or ‘patentability’) laid down in their national law. Such patents confer on its holder certain exclusive rights over the patented invention (essentially a trade monopoly), whose extent is defined by that law. Since, in principle, a State only has the sovereign power to regulate trade on its territory, the protection conferred is so circumscribed (an idea generally referred to as the principle of territoriality of patents). The inconvenience of that system is that a person seeking to protect a single invention in several States must apply for a patent in all of them, individually.

20.      The EPC was adopted to provide a (partial, as will be seen) solution to that inconvenience. That treaty, which binds 39 Contracting Parties, including the Member States and Türkiye, established an autonomous system for the grant of so-called European patents through a centralised registration procedure before the European Patent Office (‘EPO’), established in Munich. (4) In that respect, it lays down, inter alia, uniform patentability requirements. The EPO is tasked with examining applications for European patents in the light of those requirements. (5) When they are fulfilled, the EPO grants a European patent for (depending on the applicant’s wishes) one, several or all the Contracting Parties. (6) In the main proceedings, through that procedure, BSH obtained EP 1 434 512, which has been granted for several Member States and Türkiye.

21.      That said, despite what its name seemingly implies, a European patent is not a unitary title providing uniform protection of the invention in question across the States for which it has been issued. In fact, a European patent comes to life essentially as a bundle of national ‘parts’, assimilated to patents issued by the States in question. It accordingly needs to be ‘validated’ by the respective patent offices of those States. As such, the national ‘parts’ of a European patent are legally independent of one another. Each of them confers on the patent holder the same exclusive rights over the patented invention as an ‘ordinary’ national patent (7) and is equally limited to the national territory. Furthermore, in principle, (8) a European patent may solely be revoked ‘part’ by ‘part’, the revocation of a ‘part’ being effective for the territory of the corresponding State only. (9)

22.      Consequently, where a given invention is protected by a European patent, the unauthorised use of that invention by a third party may, on the one hand, result in the patent holder’s monopoly being breached in multiple States (namely the ones for which that patent has been granted). In the main proceedings, BSH accuses Electrolux of precisely that sort of ‘multistate’ patent infringement. On the other hand, as a European patent is not a unitary title, its infringement across several States is considered, from a legal point of view, as a bundle of national patent infringements, the breach of each of its ‘parts’ being appraised separately, in the light of the national law applicable to it. (10) In effect, BSH’s action against Electrolux constitutes a bundle of infringement claims resting on the various ‘parts’ of EP 1 434 512.

23.      Litigation over European patents, including actions for infringement, is also a matter left to the Contracting Parties and their national courts. (11) With respect to cross-border disputes, the EPC does not allocate jurisdiction between those courts either. (12) That question is to be determined in the light of the rules of private international law applied by the courts of those Contracting Parties.

24.      In the Member States of the European Union, jurisdiction with respect to cross-border patent disputes between private parties is determined by the rules of the Brussels I bis Regulation (13) where the defendant is, like Electrolux, domiciled in such a Member State.

25.      The jurisdictional regime laid down in that instrument (and in its predecessors) (14) (‘the Brussels regime’) operates, with respect to those disputes, under the following dichotomy.

26.      On the one hand, proceedings ‘concerned with the registration or validity of patents’ are governed by a special rule under Article 24(4) of the Brussels I bis Regulation, which confers exclusive jurisdiction on the courts of the Member State that granted the patent in question (‘the State of registration’). Where the registration or validity of a European patent is at issue, the courts of the different Member States for which that patent has been granted each have exclusive jurisdiction with respect to their national ‘part’. (15) That rule is mandatory: parties cannot depart from it by agreement. (16) Furthermore, where litigants bring their dispute before the ‘wrong’ court, that court is required, pursuant to Article 27 of that regulation, to decline jurisdiction of its own motion. (17)

27.      On the other hand, all other proceedings related to patents are governed by the general rules of the regulation. In principle, that includes infringement proceedings, since those ‘concern’ not the registration or validity of patents, but their enforcement. (18) Those rules afford litigants some jurisdictional leeway.

28.      While the courts of the State of registration do have jurisdiction over infringement proceedings pursuant to Article 7(2) of the Brussels I bis Regulation, (19) that jurisdiction is not exclusive, but optional. Such proceedings may accordingly be brought before other courts. In particular, the patent holder may seise the courts of the Member States where the defendant is domiciled, under Article 4(1) of that regulation. In the event of a ‘multistate’ infringement of a European patent, a patent holder has an obvious interest in doing so.

29.      Indeed, the jurisdiction of the courts of the Member State of registration, under Article 7(2) of the Brussels I bis Regulation, is territorially limited. In accordance with point 22 above, the courts of each State for which a European patent has been granted may only rule to the extent that their national ‘part’ and territory are concerned. (20) Consequently, the patent holder, seeking a comprehensive remedy, would have to start separate proceedings in all those States.

30.      By contrast, the jurisdiction of the courts of the Member State where the defendant is domiciled, under Article 4(1) of the Brussels I bis Regulation, is universal. Thus, it may extend to the infringement of the European patent committed in all the States for which it has been granted. (21) Those courts may award damages to compensate for the total loss suffered by the patent holder, or issue an injunction prohibiting the continuation of the infringement in all those States. In sum, that provision enables the patent holder to consolidate all his or her infringement claims before, and to obtain a comprehensive relief from, a single forum. In the present case, BSH has made use of that precise option and brought its entire action against Electrolux before the competent patent court in Sweden, where the latter is domiciled.

31.      Against that background, the referring court harbours doubts as to whether such a consolidation is, in fact, possible in the main proceedings and, if so, to what extent. Those doubts arise from the fact that Electrolux pleaded, as a defence against BSH’s claims, the invalidity of the various ‘parts’ of EP 1 434 512 on which those claims rest. (22) In the light of that defence, the referring court wonders whether, and to what extent, Article 24(4) of the Brussels I bis Regulation applies and ‘trumps’ Article 4(1) thereof. Under Article 24(4), those courts would have exclusive jurisdiction with respect to the Swedish ‘part’ only; Whereas other courts would have exclusive jurisdiction with respect to the foreign ‘parts’. Consolidation of the proceedings in a single forum would not be possible and their fragmentation would be inevitable, instead.

32.      Specifically, the first and second questions of the referring court, which it is appropriate to examine together, concern whether infringement proceedings fall within the scope of Article 24(4) of the Brussels I bis Regulation where the validity of the underlying patent is challenged by way of a defence. If that is the case, that court wonders whether, here, that provision deprives it of jurisdiction (to the extent that the foreign ‘parts’ of EP 1 434 512 are concerned) to hear and determine the infringement proceedings generally, or only the validity issue. I will examine that issue in Section A of the present Opinion.

33.      Assuming that the Court answers the first two questions to the effect that Article 24(4) is of relevance in a situation such as the one at issue in the main proceedings, the third question concerns whether that provision applies also with respect to the validity of the Turkish ‘part’ of EP 1 434 512. I will examine that issue in Section B of the present Opinion.

A.      The substantive scope of Article 24(4) of the Brussels I bis Regulation (first and second questions)

34.      As indicated above, the scope of Article 24(4) of the Brussels I bis Regulation appears to be clear. Under that provision, the exclusive jurisdiction of the courts of the State of registration covers only proceedings ‘concerned with the validity of patents’, not proceedings ‘concerned with’ other patent-related matters, including infringement.

35.      In reality, ambiguity lies beneath those terms. That dichotomy is sometimes blurred in practice. Indeed, while the validity of patents may be the object of dedicated actions (for revocation or cancelation), a patent’s invalidity may also be pleaded as a defence against, most notably, infringement claims. By doing so, the alleged infringer seeks to obtain the dismissal of those claims by vitiating the title on which they stand. (23) Electrolux pleaded such a defence in the main proceedings.

36.      The fact that the courts of the State of registration have exclusive jurisdiction over the first category of actions has always been certain. By contrast, whether, and to what extent, they also do in the second scenario is the subject of a long-standing debate.

37.      Since the early 1990s, when patent holders started to make good use of the possibilities of consolidating their infringement claims under the general rules of the Brussels regime, Member State courts have been confronted with the issue. The original wording of the contentious rule of exclusive jurisdiction, as laid down (at the time) in Article 16(4) of the Brussels Convention, did not provide guidance on the matter. Those courts came up with three main approaches:

–        First, some courts, especially in Germany, considered that the rule of exclusive jurisdiction in question does not apply when invalidity is pleaded as a defence in infringement proceedings. Courts outside the State of registration could, under the general rules of the convention, hear such proceedings and, in that context, adjudicate the validity of the patent(s) in question.

–        Secondly, other courts, notably in the United Kingdom, took the view that, when an invalidity defence is raised, infringement proceedings become ‘concerned with the validity of patents’ and accordingly fall within the exclusive jurisdiction of the courts of the Member State(s) of registration.

–        Thirdly, a last group of courts, including those in the Netherlands, considered that the contentious rule applied where invalidity is pleaded as a defence in infringement proceedings, but in a somewhat sophisticated way: only the validity issue falls within the exclusive jurisdiction of the courts of the State of registration; other courts could determine the infringement matter. (24)

38.      In 2006, the Court entered into the debate with its judgment in GAT. Notably, that case was not about infringement proceedings per se. It concerned an action by which a company sought a declaration from the German courts that it had not infringed two French patents held by a German company (‘negative declaration’), inter alia on the ground that those patents were invalid. Those courts suspected that Article 16(4) of the Brussels Convention could be relevant and referred a question in that regard to the Court. Nevertheless, the Court did not provide an answer tailored to those particular facts; instead, it ruled in general terms that the rule of exclusive jurisdiction (then) laid down in that provision ‘concerns all proceedings  relating to the … validity of a patent, irrespective of whether the issue is raised by way of an action or a plea in objection’. (25) Contrary to what BSH submits, that reply is general enough to cover, inter alia, the scenario of infringement proceedings in which an invalidity defence has been raised. The reference to ‘a plea in objection’ was evidently added to that effect. (26)

39.      A few years later, the legislature codified the judgment in GATin Article 24(4) of the Brussels I bis Regulation, by specifying, in the text of that provision, that the exclusive jurisdiction of the courts of the State of registration covers proceedings concerned with the validity of patents ‘irrespective of whether the issue is raised by way of an action or as a defence’.

40.      However, clarity on the matter was not achieved, to say the least. In fact, the answer provided in the judgment in GAT(and now in Article 24(4)) to the issue discussed in the present section raised more questions than it solved. Indeed, while that answer dealt an unwarranted fatal blow to the first approach listed above (1), the Court (and the EU legislature) left national courts and litigants debating about whether the second or third approach was correct instead (2).

1.      The unwarranted judgment in GAT

41.      Usually, where a single set of proceedings concerns two distinct matters (like, in the present case, infringement and validity), falling within the scope of mutually exclusive rules of jurisdiction (here, the general rules with respect to infringement; Article 24(4) of the Brussels I bis Regulation with regard to patent validity), the Court follows some pragmatic principles to determine which court(s) is (or are) competent to hear and determine them.

42.      First, for the purpose of determining which rules of jurisdiction apply, such proceedings are to be classified on the basis of the principal subject matter (or ‘object’) of the action brought by the claimant and any preliminary (or incidental) questions that may otherwise be raised, notably by way of a defence, should be left aside. (27)

43.      Secondly, the forum designated by the applicable rules of jurisdiction is competent to hear the entire proceedings, that is to say, not only the action, but also the defence, even though the latter concerns a matter usually reserved to a different judge. (28) Indeed, procedurally speaking, such a defence is an integral part of the action and, logically, follows the jurisdictional treatment applied to the latter.

44.      If the Court had followed those principles in the judgment in GAT, it would have adopted the first approach referred to in point 37 above. Indeed, with respect to infringement proceedings in which an invalidity defence has been raised, the principal subject matter (or ‘object’) of the action is, quite simply, infringement. By contrast, the validity issue is a prime example of a preliminary question. Since one cannot infringe an invalid patent, the judge will first have to determine the validity of the title on which the claimant relies in order, then, to resolve the principal question of whether the defendant’s acts breached the rights conferred by the title in question. According to those principles, such proceedings should have been governed, in the light of that subject matter and irrespective of that defence, by the general rules (then) laid down in the Brussels Convention. Furthermore, the courts designated as competent to hear and determine the infringement proceedings under those general rules, in particular the courts of the Member State where the defendant is domiciled, should also have had jurisdiction to determine that defence.

45.      Evidently, that is not what the Court did in the judgment in GAT. Rather, it took the view that the rule of exclusive jurisdiction for ‘proceedings concerned with the validity of patents’ applies even to proceedings in which that issue is raised merely by way of a defence. By doing so, the Court adopted an interpretation which is, to my knowledge, unique in the Brussels regime. In fact, that solution departs even from the ones the Court adopted, so far, under the other  rules of exclusive jurisdiction (now) laid down in Article 24 of the Brussels I bis Regulation. Following the principles discussed above, the Court considers that those other rules apply where an issue referred to therein constitutes the subject matter of the action only. (29) In several (but unfortunately not all) language versions of the regulation, that approach even stems from the very letter of that provision. (30) Moreover, in its judgment in BVG, delivered a few years after the judgment in GAT, the Court took the view that the rule of exclusive jurisdiction for proceedings concerned with the validity of companies or of the decisions of their organs (now laid down in Article 24(2) of that regulation) does not apply to proceedings in which such an issue is raised only by way of a defence. (31)

46.      The precise implications of the interpretation adopted in the judgment in GATare uncertain, as mentioned above, and will be discussed in Section 2 below. For present purposes, what is clear is that, contrary to what BSH submits, where infringement proceedings are brought outside the Member State of registration and an invalidity defence is raised, those courts are not allowed to adjudicate the validity of the patents concerned as a preliminary question.

47.      That being said, unfortunately, in my view, the rather brief reasoning of the Court in the judgment in GAT does not offer a convincing justification for that solution.

48.      The first argument put forward by the Court relates to the ‘position of [that rule of exclusive jurisdiction] within the scheme of the [Brussels Convention]’ (i.e. its primacy over the general rules of jurisdiction) and its ‘mandatory nature’. (32) That argument fails to convince me. (33) In fact, those elements more readily support the opposite interpretation.

49.      The rules of exclusive jurisdiction are exceptions in the Brussels regime. As such, they must be interpreted strictly. (34) Indeed, they are supposed to apply in a ‘few well-defined situations’ (35) only. Moreover, as the Court ruled in the judgment in BVG, a strict interpretation of those rules is ‘particularly necessary’ precisely because they have primacy over the general rules and are mandatory. (36) Where Article 24(4) applies, it deprives claimants of the choice of forum, which would otherwise be theirs, and may result in defendants being sued outside of the Member State of their domicile, where they would usually be better placed to defend themselves.

50.      By contrast, the interpretation adopted by the Court in the judgment in GAT can only be described as ‘wide’. (37) Certainly, the validity of patents is, in itself, a ‘well-defined’ issue. Nevertheless, it may be raised in an ‘ill-defined’ range of proceedings concerned with other matters. (38)

51.      The two other reasons given by the Court in the judgment in GAT concern the general objective of legal certainty pursued by the Brussels regime. (39)In essence, the Court explained that if the contentious rule of exclusive jurisdiction did not apply when the issue of the validity of patents is raised by way of a defence in infringement proceedings (and so on), and courts outside the State of registration, seised of such proceedings, were allowed to adjudicate that issue as a preliminary question, it ‘would have the effect of multiplying’ the number of courts that could rule on that issue. That, in turn, would ‘undermine the predictability of the rules of jurisdiction’ and ‘multiply the risk of conflicting decisions’ on the matter, all of which would undermine legal certainty. (40)

52.      I am not convinced by this either. When one ‘sees the whole picture’, those elements again more readily support the opposite interpretation. Admittedly, from one angle, the solution in the judgment in GAT prevents different courts from taking contradictory views of a patent’s validity. To that extent, it contributes to legal certainty. However, from another angle, the judgment in GAT has the potential to make the operation of the Brussels regime, with respect to infringement proceedings, precarious for patent holders.

53.      Whereas that regime normally allows a patent holder to bring such proceedings outside the State of registration, inter alia, before the courts of the Member State of the defendant, the solution in the judgment in GAT creates uncertainty as to whether those courts would be able to grant a relief against infringement or, at least, to do so within a reasonable time. Indeed, were an invalidity defence to be raised by the alleged infringer at some stage of the proceedings, those courts would not be able to simply adjudicate that defence and proceed with the case but, depending on how that solution is to be understood, would either lose their jurisdiction and have to dismiss the proceedings, or may have to stay proceedings until the courts of the Member States of registration have determined the patent’s validity (see, further, Section 2 below).

54.      Whatever the correct interpretation, the judgment in GAT makes the consolidation of infringement claims concerning the different ‘parts’ of a European patent before those courts an unattractive option. It encourages patent holders to start separate proceedings in the various States of registration of those ‘parts’ instead, since, at least, it is certain that the courts of those States are competent to rule on both the infringement and validity of ‘their part’ (as explained in points 26, 28 and 29 above). This creates, in turn, a risk that different courts take contradictory views on the same infringement dispute.

55.      Such uncertainty and/or complexity with respect to enforcement of patents is all the more undesirable, given that intellectual property is protected as a fundamental right, inter alia, in Article 17(2) of the Charter of Fundamental Rights of the European Union (‘the Charter’). Under that provision, patent holders are to enjoy a ‘high level’ of protection of their intellectual property rights in the European Union. The possibility to bring expedient civil proceedings and obtain a relief in the case of infringement is essential in that respect. That is also required by the fundamental right to an effective remedy, guaranteed in Article 47 of the Charter. In connection with that, I recall that Article 41(2) of the Agreement on Trade-Related Aspects of Intellectual Property Rights (‘the TRIPS Agreement’) (41) provides that ‘procedures concerning the enforcement of intellectual property rights … shall not be unnecessarily complicated or costly, or entail … unwarranted delays. This also concerns, in my view, the operation of the relevant rules of international jurisdiction.

56.      In any event, pursuant to the settled case-law of the Court concerning the interpretation of the rules of exclusive jurisdiction under the Brussels regime, (42) the only question that the Court in the judgment in GATneeded to answer (but that it did not discuss) was whether the specific objective pursued by the contentious rule of exclusive jurisdiction ‘required’ that that rule also apply to infringement proceedings in which an invalidity defence has been raised. In my view, it did not.

57.      From the outset, I would like to clear some confusion as to what that objective is. The explanation usually provided by the Court in that regard (and mentioned ‘in passing’ by the Court in the judgment in GAT) is that the rule in question pursues an objective of good administration of justice. In its view, the courts of the State of registration are ‘best placed’ to determine proceedings concerned with the registration or validity of patents, because of the ‘close link in fact and law’ that exists between such proceedings and that State. (43) However, in my Opinion, that is not the real raison d’être of that rule.

58.      Admittedly, as explained in points 19 and 21 above, a patent is governed by the law of the State of registration. There is some weight to the argument that, for instance, a German judge is ‘best placed’ to apply German patent law (because of the language, knowledge of that law, and so on). (44) Furthermore, because a patent is only protected in the State of registration, there is usually a factual proximity between disputes concerning that patent and the territory of the latter.

59.      However, those considerations only explain why the courts of the State of registration may hear and determine such disputes. They justify why, for instance, those courts have, under the Brussels regime, jurisdiction over infringement proceedings concerning their territory. (45) By contrast, such considerations do not reveal why, with respect to proceedings concerned with the registration or validity of patents, those courts should have jurisdiction to the exclusion of all others. (46) In particular, the patent law of the State of registration is not so unique that only the courts of that State would have the ability to comprehend it. (47) While it may be harder for them to do so, the courts of another Member State are perfectly capable of applying such a foreign law. To imply the contrary would be tantamount to questioning the very foundations of the Brussels regime (and the entire field of private international law). (48)

60.      The real raison d’être of the contentious rule lies in the fact that, as the Jenard report puts it, ‘the grant of a … patent is an exercise of national sovereignty’. (49) Indeed, the only compelling justification for providing such a rule of exclusive jurisdiction is the role (traditionally) played by State authorities in the issuing of such intellectual property rights, mentioned in point 19 above, (50) specifically the facts that national administrations are tasked with examining patent applications, granting them where the relevant requirements are fulfilled, and registering patents accordingly. However, the solution in the judgment in GAT was, in my view, not ‘required’ by that consideration either.

61.      Indeed, on the one hand, proceedings that have the registration or validity of patents as their object, call into question, by their very nature, the functioning of the administration of the State of registration. (51) The heart of the dispute is whether the relevant State authority (patent office) ‘did its job’ correctly. With an action for revocation, in particular, a claimant essentially asks the court to review whether that authority was right to grant the patent in the first place and, if it was not, to declare the patent invalid as a relief. Such a declaration has, by nature, erga omnes effect and may, as such, be invoked against the authority in question. The ruling delivered by the court may even instruct the latter to rectify its registers accordingly. Obviously, such rulings should only be delivered by the courts of the State of registration. Deference to the sovereignty of States applies here. States would find it unacceptable if the actions of their authorities were to be sanctioned by the courts of a foreign State and that the latter instruct them on how to manage their national registries. (52)

62.      On the other hand, infringement proceedings, in particular, do not call into question the functioning of the administration of the State of registration, even where the invalidity of the patent allegedly infringed is raised by way of a defence. There, that issue, as a preliminary question, is examined by the court, but only for the purpose of resolving the infringement matter. The only consequence that may ensue is the court dismissing the infringement claim. Such a ruling is concerned with the private interests of the litigants and, accordingly, generally has inter partes effects only. (53) It cannot encroach on the sovereignty of the State of registration because it has no effect on its administration nor does it purport to have one. The validity of the patent, from a legal point of view, is not affected. No instruction to that administration is being issued by a court of another sovereign State.

2.      The proper reading of the judgment in GAT

63.      For all those reasons, the judgment in GATis, in my view (and according to the majority of the literature I have consulted), (54) an unfortunate decision. If the solution laid down therein stood on that ruling alone, I would have advised the Court to overturn it and to declare instead that the rules of exclusive jurisdiction for proceedings concerned with the validity of patent does not apply where an invalidity defence is raised in infringement proceedings, to the extent that the judgment delivered by the court seised would produce inter partes effects only. (55)

64.      However, as indicated above, the EU legislature codified that judgment in Article 24(4) the Brussels I bis Regulation. (56) Thus, as EU law stands, the Court is ‘trapped’ in the solution that it initially adopted. It is left to choose, at the invitation of the referring court, between two possible readings of the judgment in GAT (and its codification), which correspond, respectively, to the second and third approaches listed in point 36 above.

65.      Pursuant to the first reading, defended by Electrolux, and which I shall describe as ‘broad’, where invalidity is pleaded as a defence in infringement proceedings, the courts of the Member State of registration are (or become) exclusively competent under Article 24(4) of the Brussels I bis Regulation to determine those proceedings. Any other court must decline jurisdiction pursuant to Article 27 of that regulation.

66.      According to the second reading, defended by the Commission and which I shall describe as ‘narrow’, where invalidity is pleaded as a defence in infringement proceedings, the courts of the Member State of registration have exclusive jurisdiction under Article 24(4) of the Brussels I bis Regulation to determine the issue of validity only. Other courts could have (or keep) jurisdiction, under the general rules of that regulation, to determine the infringement matter.

67.      As EU law stands, it does not provide clear indications as to which reading is supposed to be correct. First, while, as Electrolux submits, the ‘broad’ reading of the operative part of the judgment in GAT and of the wording of Article 24(4) of the Brussels I bis Regulation seems the most natural one, (57) those elements can also reasonably be read the ‘narrow’ way. Although the Court said, in that judgment, that the contentious rule of exclusive jurisdiction ‘concerns all proceedings relating to the … validity of a patent’, it did not indicate to what degree. This drafting is simply ambiguous. Secondly, the subsequent case-law of the Court does not support one reading or another of the judgment, as it contains contradictory indications in that regard. On the one hand, as Electrolux points out, the Court seemingly endorsed the ‘broad’ reading in its judgment in BVG. (58) On the other hand, as the Commission underlines, the Court apparently endorsed the ‘narrow’ reading in the judgment in Roche Nederland and Others. (59) Finally, the legislature did not express any position in that regard in the text of Article 24(4) or in a recital of the regulation. (60)

68.      Hence, to resolve the controversy, one must turn to the system established under the Brussels I bis Regulation as well as the objectives pursued by that instrument generally and by Article 24(4) thereof, specifically. In the light of those elements, the Court should reject the ‘broad’ reading of in the judgment in GAT (part (a)) and endorse the ‘narrow’ reading instead (part (b)). It should also provide some guidelines to national courts as to how that reading is to be put in practice (part (c)).

(a)    The critical flaws of the ‘broad’ reading

69.      In the first place, the ‘broad’ reading of the judgment in GATcan hardly be reconciled with the system established under the Brussels I bis Regulation. Under that system, as was intended by its drafters, the exclusive jurisdiction of the courts of the State of registration is an exception, circumscribed to ‘proceedings concerned with the registration or validity of patents’, while infringement proceedings, and the rest of patent disputes, may normally be brought before other courts.

70.      However, if the judgment in GATwere to be understood in the way suggested by Electrolux, the exception would, in fact, become the rule, as the Commission observes. Since invalidity defences are often raised in infringement proceedings, such proceedings would frequently fall within the exclusive jurisdiction of the courts of the State of registration. The application of the general rules of jurisdiction, and the options they confer on patent holders, would be limited to the cases where such a defence is not raised.

71.      Secondly, contrary to what Electrolux submits, since the classification of infringement proceedings and, thus, the rules of jurisdiction applicable to them would depend on whether or not an invalidity defence is raised (in contradiction with the principle stated in point 42 above), the ‘broad’ reading of the judgment in GATwould compromise the predictability and certainty of jurisdiction aimed at by the Brussels I bis Regulation. (61)

72.      Indeed, for jurisdiction over infringement proceedings to be predictable, patent holders should be able to easily identify the court before which they may bring such proceedings. However, under the ‘broad’ reading of the judgment in GAT, it would be hard for them to determine in advance whether such proceedings fall within the exclusive jurisdiction of the State of registration or may be brought before other courts, as they have no control over the defence strategy that the alleged infringer will adopt. (62)

73.      Furthermore, were the patent holder to choose to bring proceedings outside of the State of registration, for instance before the courts of the Member State where the alleged infringer is domiciled, the jurisdiction of those courts would be precarious. Indeed, it could disappear if the alleged infringer were to raise an invalidity defence. Those courts would have to decline to continue hearing the case. (63) If, under the procedural rules of the forum, such a defence may be raised not only at the beginning of the trial, but also at subsequent, later stages, including on appeal, proceedings that have been going on for months, or even years, could suddenly reach a dead-end. The alleged infringer could also strategically choose the moment to raise such a defence and effectively ‘torpedo’ the proceedings. As BSH and the Commission observe, the consequences for the patent holder would be dramatic. Indeed, as it stands, there is no possibility, under the Brussels I bis Regulation, for the courts of a Member State to transfer a case to the courts of another. The courts initially seised could only dismiss the proceedings, leaving the claimant the task of starting fresh proceedings in the State of registration.

74.      To add insult to injury, it may no longer be possible for the patent holder to do so. Indeed, the limitation periods applicable to the infringement claims may have elapsed in the meantime. In effect, the patent holder would be deprived, through no fault of their own, of all possibility of a remedy against the breach of their intellectual property rights. Such an outcome would be contrary to Article 17(2) and Article 47 of the Charter as well as Article 41(2) of the TRIPS Agreement.

75.      Even if it were still possible for the patent holder to do so, in case of ‘multistate’ infringement of a European patent, they would be bound to start infringement proceedings in all the States concerned to obtain a comprehensive relief. (64) It would not be possible to consolidate the claims before a single forum. A number of courts could get involved in what is essentially the same dispute, increasing the risk of conflicting decisions mentioned in point 54 above.

76.      Finally, contrary to what Electrolux submits, the ‘broad’ reading of the judgment in GATgoes beyond what is ‘required’ by the specific objective of Article 24(4) of the Brussels I bis Regulation, namely, as explained in points 60 and 61 above, to ensure that deference is given to the sovereignty of the State of registration. Even understood broadly, it would only be possible for that objective to ‘require’ (where an invalidity defence is raised in infringement proceedings) that the courts of that State have exclusive jurisdiction to adjudicate the validity issue, not the infringement matter.

(b)    The ‘narrow’ reading of the judgment in GAT is ‘the lesser evil’

77.      The ‘narrow’ reading of GATscores significantly higher on all the aspects discussed above. The principle remains that the general rules of the Brussels I bis Regulation govern infringement proceedings. As such, jurisdiction is predictable and certain for the patent holder. If they bring proceedings outside the State of registration and the alleged infringer raises an invalidity defence, the courts seised do not lose the competence to hear and determine the action. Those courts ‘simply’ cannot adjudicate the validity of the patent(s) in question, which, under the exceptional rule laid down in Article 24(4) of that regulation, can only be determined by the courts (65) of the State of registration. Furthermore, in the case of a ‘multistate’ infringement of a European patent, that reading allows for the partial consolidation of the claims before a single forum. Only the validity of the patent, if challenged, would have to be determined in the various States for which it was granted.

78.      As the Commission observes, that reading of the judgment in GAT implies that the Court admitted a derogation to the principle, under the Brussels regime, that jurisdiction to hear and determine an action extends to any potential defence (see point 43 above). That being said, such a derogation, although unique within that regime, is not completely unheard of. Indeed, similar derogations feature in the rules of territorial jurisdiction of the Member State for certain matters subject to exclusive jurisdiction under national law. (66)

79.      In practical terms, it follows that, when infringement proceedings concerning a patent registered in a Member State are pending before the courts of another Member State and an invalidity defence is raised, because those courts can neither adjudicate the validity issue nor (as the Brussels I bis Regulation stands) refer an interlocutory question on the matter to the authorities of the State of registration, it is up to the alleged infringer (if he or she had not already done so) to bring invalidity proceedings before those authorities so that the latter determine that issue. (67)

80.      Electrolux objects, not without merit, that the ‘splitting’ of the infringement and validity issues into two sets of proceedings brought in different Member States is questionable in terms of the administration of justice. Indeed, those issues are closely interrelated. (68) As explained in point 44 above, in principle, the preliminary issue of validity of the patent must be resolved in order to determine the main issue of infringement. Furthermore, besides the fact that the law of the State of registration applies to both issues, they essentially depend on the same element, namely the construction of the patent claims. (69)

81.      Nevertheless, in my view, while it is not always ideal from a practical point of view for the validity and infringement issues to be determined by different courts and/or authorities, (70) it is not impossible either. In fact, at national level, several Member States, such as, it would seem, Sweden, (71) adopted a ‘bifurcation system’ of patent adjudication, under which those issues are determined by different judges in separate, dedicated proceedings. (72)

82.      While validity and infringement must be ‘split’ in this way, it does not follow, as Electrolux posits, that, where an invalidity defence is raised in infringement proceedings, the court hearing those proceedings should, or even could, systematically ignore that defence, presume the patent to be valid and issue a final decision on the infringement matter, irrespective of the invalidity proceedings that may be pending in parallel in another Member State.

83.      Indeed, as BSH and the Commission submit, Article 24(4) of the Brussels I bis Regulation cannot be interpreted that way. Otherwise, an alleged infringer would be completely deprived of one of the most effective shields against bogus infringement claims. That would constitute an unacceptable limitation of their right of defence, which are guaranteed, inter alia, by Article 47 of the Charter, and which the Brussels regime seeks to ensure. (73)

84.      Moreover, in some cases, that could result in irreconcilable judgments being delivered. Indeed, on the one hand, the courts in charge of the infringement proceedings could recognise the infringement while, on the other, the authorities of the State of registration could subsequently declare the patent invalid. The latter could also confirm the validity of the patent, but under a narrow construction of its claims (which would normally exclude a finding of infringement), while the courts in charge of the infringement proceedings could recognise infringement under a wide construction of the patent claims (which would have led the validity judge to declare the patent invalid). (74)

85.      As I will explain further in the next section, there are circumstances where the courts in charge of the infringement proceedings would be entitled to presume the patent valid and rule accordingly, irrespective of an invalidity plea. However, in other circumstances, respect for the rights of defence will require those courts to wait until the validity of the patent has been determined by the authorities of the State of registration before they issue a final, concurring decision on infringement. (75)

86.      So, case management and/or procedural measures will sometimes have to be taken to ensure the coordination of the infringement and invalidity proceedings. In that regard, Electrolux observes that neither the Brussels I bis Regulation nor EU law in general provide for a solution in that regard. Specifically, Article 30(1) of that regulation could allow the courts in charge of the infringement proceedings to stay those proceedings until the authorities of the State of registration have determined the validity of the patent, but only if the latter have been seised first.  That provision provides no relief in the case of subsequent invalidity proceedings. Nevertheless, as BSH and the Commission submit, until the EU legislature has enacted provisions to that effect, (76) the courts seised of the infringement can, and sometimes should, apply the solutions provided under its procedural law (lex fori).

87.      Electrolux retorts that such a reliance on national procedural law poses a risk for the uniform treatment of cases and litigants in the Member States, as different courts may have different powers at their disposal or may apply them differently. However, in my view, that is another unavoidable downside of the judgment in GAT. Furthermore, the matter is not left entirely to national law. As I will explain further in the next section, EU law significantly frames the latter, ensuring a sufficient degree of uniformity.

88.      Finally, it is often argued that the ‘narrow’ reading of the judgment in GAT is not ideal for the effective enforcement of patents, either. The ‘splitting’ of the validity and infringement issues into two sets of proceedings increases the cost and inconvenience for the parties. The need for the courts in charge of the infringement proceedings, in some circumstances, to await an answer on validity from the authorities of the State of registration could potentially delay those proceedings, whereas it is usually a matter of urgency for the patent holder that such infringement be sanctioned and prohibited. (77) It may also incite alleged infringers to raise ‘torpedo’ defences or delay the bringing and conduct of the invalidity action in order to paralyse the infringement proceedings. While I generally agree with those objections (as it stems from Section 1), the fact remains that, of the two possible approaches left on the table after the judgment in GAT, it is the ‘lesser evil’. Furthermore, the issues outlined above can be curtailed with pragmatic measures, as discussed below.

(c)    Practical guidelines for national courts

89.      At the hearing, at the invitation of the Court, the interveners discussed how courts outside the State of registration should act when hearing infringement proceedings and an invalidity defence is raised. Although, as discussed above, that is mostly a matter for the procedural rules of those courts, the Court is, in my view, competent to set guidelines in that regard. Indeed, I recall that, pursuant to settled case-law, those procedural rules cannot impair the effet utile of the Brussels I bis Regulation, and must be applied accordingly. (78) The principles laid down in the TRIPS Agreement and in Directive 2004/48, as well as, in relation to the patent holder, the right to an effective remedy and, on the alleged infringer’s side, the rights of defence, both protected under Article 47 of the Charter, also frame national law in that regard.

90.      When an invalidity defence has been (properly) (79) raised by the alleged infringer, a solution, which is often advanced by commentators and debated before the Court, would be for the courts hearing the infringement proceedings – where their procedural rules give them the power to do so (which is, I assume, usually the case) – (80) to stay the proceedings until the validity of the patent in question has been determined by the authorities of the State of registration. (81)

91.      While that is indeed a solution, I agree with BSH that those courts should not grant a stay automatically. They must, in fact, carefully consider the matter before doing so, because of the (potentially significant) delays that such a measure would inevitably entail for the resolution of the infringement proceedings. A stay should only be grant where, in accordance with Article 3 of Directive 2004/48 and Article 41 of the TRIPS Agreement, it is proportionate and fair to do so, and those delays are ‘warranted’. Those courts must, therefore, be accorded the discretion to balance, on the one hand, the requirements of efficiency of procedure as well as the right to an effective remedy of the patent holder, and, on the other hand, the sound administration of justice and the rights of defence of the alleged infringer.

92.      Specifically, as BSH and the Commission submit, the courts hearing the infringement proceedings should assess the seriousness of the invalidity challenge first. In my view, Article 24(4) of the Brussels I bis Regulation does not prohibit them from taking a preliminary view on how the authorities of the State of registration would decide the matter. (82) In that regard, those courts should only consider granting a stay where that challenge has a genuine prospect of success. Indeed, because patents are granted after a prior control of the patentability requirement has been conducted by patent offices, they benefit from a presumption of validity. Therefore, the grounds put forward by the alleged infringer must seem, prima facie, sufficiently solid to call that presumption into question. Where that is not the case, those courts can assume that the patent is valid and rule on the infringement accordingly. Moreover, it would make little sense, in the light of efficiency of procedure and the right to an effective remedy of the patent holder, to delay the infringement proceedings in cases of frivolous invalidity defences. There is also no risk (or, at least, a negligible one) of irreconcilable decisions, as there is no (reasonable) chance that the authorities of the State of registration subsequently declare the patent invalid. (83) Such an assessment also limits the possibility of infringers raising spurious defences as dilatory tactics. (84)

93.      Where the invalidity defence is serious, the courts in charge of the infringement proceedings should grant a stay. Indeed, in such circumstances, the rights of defence would usually require it. (85) So would the sound administration of justice, as the risk of irreconcilable decisions discussed above would be significant. However, as the Commission submits, to ensure the efficiency of proceedings and to avoid, once more, dilatory tactics by the alleged infringer, those courts should set a deadline for the latter to bring invalidity proceedings in the State of registration (if they have not already done so). If the alleged infringer does not do so, those courts should lift the stay, assume that the patent is valid and rule on the infringement. Where such a stay has been granted, those courts should follow the progress of the invalidity proceedings and decide to maintain or lift the stay accordingly.

94.      Finally, for the duration of the stay, nothing prevents the courts hearing the infringement proceedings from ordering provisional measures, including protective ones, such as an interlocutory injunction prohibiting the continuation of the potentially infringing acts (here again, depending on the seriousness of the validity challenge). (86) Indeed, such a possibility was expressly preserved by the Court in the judgment in Solvay, (87) and it should be used where proportionate to safeguard the rights of the patent holder.

B.      The ‘reflexive effect’ of Article 24(4) (third question)

95.      It flows from Section A of this Opinion that, while the Swedish courts have jurisdiction, pursuant to Article 4(1) of the Brussels I bis Regulation, to hear the infringement proceedings brought by BSH, they cannot adjudicate the validity of the parts of the European patent allegedly infringed. Under Article 24(4) of that regulation, the courts of the various Member States where those parts have been validated are exclusively competent to determine that issue.

96.      That being said, given that BSH’s action is also based on the part of that European patent that was validated in Türkiye and that Electrolux challenges, the referring court wonders, by its third question, whether Article 24(4) of that regulation is ‘applicable to a court of a third [State]’, that is to say, whether, in the main proceedings, it ‘confer[s] exclusive jurisdiction’ on the Turkish courts over that issue.

97.      Taken literally, that question calls for an obvious answer. As a piece of EU law, the Brussels I bis Regulation binds the Member States. It determines the jurisdiction of their courts. That instrument cannot possibly confer any sort of jurisdiction on the courts of third States. The European Union has no competence in that regard. The jurisdiction of those third-State courts depends on their own rules of private international law.

98.      Nevertheless, to provide a useful answer to the referring court, the Court cannot limit itself to that self-evident finding. Read in the context of the main proceedings, it is clear that the third question concerns, in essence, not the positive, jurisdiction-conferring effect of Article 24(4) of the Brussels I bis Regulation, but its negative, jurisdiction-depriving effect. What is really at issue is whether that provision deprives Member State courts of the power to adjudicate the validity of third-State patents in the same way that those courts are deprived with respect to patents registered in other Member States.

99.      As the French Government observes, understood in that way, that question raises a cross-cutting issue, the relevance of which goes well beyond the scope of the present case. Indeed, it may arise in relation to any of the subject matters for which Article 24 of the Brussels I bis Regulation lays down a rule of exclusive jurisdiction. For instance, what if Member State courts are seised of a claim concerned with the validity of rights in rem over immovable property (a matter addressed in Article 24(1) thereof), but that property is located in China? The same issue may also arise in relation to exclusive choice of court agreements. When such an agreement designates the courts of a Member State, another provision of that regulation, namely Article 25, deprives all other courts of jurisdiction. However, what if the courts of a Member State are seised notwithstanding a similar agreement in favour of third-State courts?

100. The answer to that issue is, by contrast, very obscure. In fact, it has generated significant debate in academic writing and before the national courts. So far, the Court has not provided a clear and comprehensive answer. As I will explain in detail in the next sections, the complexity of the issue stems from the fact that, when it comes to its territorial scope, the Brussels regime suffers from what I would call a ‘design flaw’ (1), which requires deeper reflection as to the best way to ‘fill the gaps’ of that regime with regard to those scenarios (2).

1.      The ‘design flaw’ of the Brussels regime

101. The ‘design flaw’ mentioned above is the result of a paradox. On the one hand, it is clear that the territorial scope of the Brussels I bis Regulation extends to disputes having strong connections with third States. Indeed, under Article 4(1) of that regulation, interpreted in the light of the seminal judgment in Owusu, (88) that regulation is applicable, ratione territoriae, to any cross-border dispute where the defendant is, like Electrolux in the present case, domiciled in a Member State; it does not mean only ‘intra-EU’ disputes. Disputes which, beyond the seat of that litigant, are connected with third States are also covered, even where the subject matter is closely related to such a State or where there is a choice-of-court agreement in favour of third-State courts. (89) In principle, only where the defendant is not domiciled in the European Union is the matter excluded from the scope of that regulation, pursuant to Article 6(1).

102. On the other hand, the Brussels regime was not actually designed for disputes connected to third States. That regime was, for the most part, drafted with ‘intra-EU’ disputes in mind. Articles 24 and 25 of the Brussels I bis Regulation demonstrate that clearly. The wording of the first provision limits its scope to disputes, the subject matter of which is closely related to a ‘Member State’. The second refers to choice-of-court agreements regarding ‘the courts of a Member State’ only. The scenario of disputes having similar connections with third States was not envisioned when crafting those rules. As a result, that regime is generally silent on the effect, if any, that such connections should have on the jurisdiction of the courts of the Member States. (90)

2.      ‘Filling the gaps’ of the Brussels regime

103. Where a Member State court is seised of a dispute which, on the one hand, involves an EU defendant but, on the other, is strongly connected to a third State (because it concerns a subject matter closely related to that State, or because it is covered by an exclusive choice-of-court agreement in favour of its courts), the general silence of the Brussels I bis Regulation on the matter leaves open the question of how that court should proceed. Three possible answers can be derived from the literature and the debate that took place before the Court in the present case.

104. At one extreme lies a first answer, which was not supported by any interveners before the Court, pursuant to which Article 24 or Article 25 of the Brussels I bis Regulation would apply, by analogy, in that scenario. Accordingly, pursuant to the relevant provision, a Member State court would be deprived of jurisdiction to hear such a dispute and would be bound to dismiss the proceedings.

105. At the other extreme lies a second answer, which is supported by BSH, the French Government (reluctantly) (91) and the Commission, pursuant to which the general rules of that regulation would apply instead. As a result, amongst others, the courts of the Member State where the defendant is domiciled would have jurisdiction pursuant to Article 4(1) of the Brussels I bis Regulation. Furthermore, they would be bound to  exercise  it and thus to determine the case, save in some limited circumstances.

106. In between lies a third answer, which Electrolux supports. It corresponds to the ‘reflexive effect’ (effet réflexe) theory developed many years ago by Droz. (92) It boils down to saying that while a Member State court may have jurisdiction under that regulation over a dispute featuring such connections with a third State, it may decline to exercise it, where that ‘reflects’ the system established therein.

107. In my view, the two extreme answers must be rejected and the middle one endorsed. Indeed, as I will explain in the sections below, because Articles 24 and 25 cannot possibly apply in such circumstances (a), the general rules of the regulation do instead (b); however, a Member State court cannot be bound to exercise the jurisdiction it derives from those rules in such situations (c). I will then clarify the conditions under which it can legitimately decline jurisdiction (d).

(a)    Articles 24 and 25 cannot apply

108. Although part of the literature suggests the contrary, (93) it is clear, in my view, that Articles 24 and 25 of the Brussels I bis Regulation cannot apply, as such, to disputes having connections of the kind envisioned therein with third States.

109. Such an interpretation would be in direct conflict with the express wording of those two articles, which, as indicated above, are limited to disputes, the subject matter of which is closely related to a ‘Member State’, and choice-of-court agreements designating the courts of a ‘Member State’. (94) Furthermore, extending the scope of those provisions through analogical reasoning to similar scenarios involving third States would be incompatible with the principle of strict interpretation of exceptions, as BSH, the French Government and the Commission submit. Besides, the Court has already refused to do so. In the judgment in IRnova, it ruled that, since Article 24(4) of the Brussels I bis Regulation ‘does not envisage’ disputes concerned with the validity of third-State patents, ‘that provision cannot be regarded as applicable’ in such a situation (that reasoning being transposable to all the provisions of Article 24). (95) Similarly, in its judgment in Coreck Maritime, (96) the Court took the view, with respect to the rule equivalent to Article 25 laid down in the Brussels Convention, that ‘as the wording of … Article 17 … itself makes clear’, it ‘does not apply to [choice-of-court agreements] designating a court in a third [State]’.

110. Furthermore, as the French Government observes, the system established under Articles 24 and 25 of the Brussels I bis Regulation, pursuant to which Member State courts are obliged to relinquish jurisdiction in favour of the courts designated by those provisions, makes sense for ‘intra-EU’ disputes only. In those circumstances, pursuant to that regulation, where one court has no jurisdiction, another one does. That is not the case with respect to ‘external’ disputes. As I mentioned before, the jurisdiction of third-State courts depends on their own rules of private international law. While those courts usually consider themselves competent where the disputed matter is closely related to their territory, or where chosen in a jurisdiction agreement, that may not always be the case. If Member State courts were deprived of jurisdiction in such a situation, there would be a denial of justice. Furthermore, the strict, almost automatic obligation for the courts of the Member State to remit litigants to other courts under Articles 24 and 25 is justified by the ‘mutual trust’ that those States accord to each other’s judicial institutions. (97) That ‘trust’ does not extend to third States. It cannot be presumed that litigants would have a fair trial in such a State. It can even be excluded in certain cases.

(b)    The general rules of the regulation apply instead

111. As BSH, Electrolux, the French Government and the Commission submit, since Articles 24 and 25 of the Brussels I bis Regulation are not applicable with respect to disputes having connections of the kind envisioned therein with third States, the logical consequence, in the system established under that regulation, is that the general rules apply instead. It follows that, inter alia, the courts of the Member State where the defendant is domiciled have jurisdiction to hear such a dispute, under Article 4(1) of that regulation.

112. That systematic reading is supported by several of the official reports that accompany the Brussels instruments. (98)Most of all, it has been endorsed implicitly (but quite clearly) by the Court (Full Court) in Opinion 1/03 (New Lugano Convention). (99) For the present purposes, it suffices to recall that that decision was concerned with whether the European Union had exclusive competence to conclude the Lugano II Convention, which depended, in turn, on whether the latter would ‘affect’ the Brussels regime. (100) The Court considered that it would. It explained that, with respect to disputes that involve a defendant domiciled in a Member State but that are closely related to a third-State party to that convention, or covered by a choice-of-court agreement in favour of its courts, the future convention would confer exclusive jurisdiction on the third State, (101) whereas, under that regulation, the courts of that Member State would have had jurisdiction. (102)

(c)    Member State courts cannot be bound to exercise the jurisdiction they derive from the general rules of the regulation

113. While, notably, the courts of the Member State where the defendant is domiciled have jurisdiction over disputes having such connections with third States under Article 4(1) of the Brussels I bis Regulation, I disagree with the view, supported by BSH, the French Government and the Commission, that those courts are bound to exercise that jurisdiction, save in a narrow set of circumstances. In my view, such an approach is neither required by the text of that regulation and the related case-law (1), nor is it compliant with the objectives of the Brussels regime (2). The fact that international agreements may, in some situations, remedy the difficulties that such an interpretation would generate does not justify endorsing it (3), and nor does the alleged intent of the EU legislature (4).

(1)    The terms of the regulation and the related case-law

114. The main argument put forward by BSH, the French Government and the Commission is based on the text of the Brussels I bis Regulation. In their view, it flows from the terms of Article 4(1) thereof (‘Subject to this Regulation, persons domiciled in a Member State shall … be sued in the courts of that Member State’) that, in principle, jurisdiction under that provision is mandatory. It means that, where the courts of that State are seised, they are generally bound to hear and determine the case. Furthermore, in the judgment in Owusu, the Court took, with respect to the equivalent provision of the Brussels Convention (Article 2), the strict view that ‘there can be no derogation from the principle it lays down except in the cases expressly provided for’ by the Brussels regime. (103) In that case, the courts of the United Kingdom (then a Member State) had been seised of a dispute against, inter alia, a defendant domiciled in that State concerning a tort that had occurred in Jamaica. The first question referred by those courts was whether they were permitted to decline jurisdiction in favour of the courts of Jamaica, pursuant to the common law doctrine of forum non conveniens. (104) The Court replied in the negative, on the ground that the Brussels regime does not contain such an exception. (105)

115. For those interveners, since the Brussels I bis Regulation does not expressly provide exceptions for disputes whose subject matter is closely related to third States, or covered by a choice-of-court agreement in favour of third-State courts, it follows that the courts of the Member State, when seised of such a dispute and competent under Article 4(1) of that regulation, are bound to hear and determine the case. They could decline jurisdiction only in the specific circumstances envisioned by certain rules added by the EU legislature with the adoption of the Brussels I bis Regulation, namely Articles 33 and 34 thereof, that is to say, in the case of concurrent proceedings in third-State courts and only where the dispute was already pending before the latter when the Member State courts were seised. (106)

116. That is, in my view, a simplistic reasoning.

117. First, to deduct, from the mere absence of provisions dedicated to disputes strongly connected to third States in the Brussels I bis Regulation, that Member State courts are generally bound to ignore those connections and to rule on such cases is a flawed logic. Indeed, it disregards the fact that, as indicated in point 102 above, that instrument was not designed with those disputes in mind. That fact explains the general silence of the text on the matter, and why any positive consequences cannot, in my view, be drawn from it. (107)

118. Secondly, one also cannot reasonably deduce, from the mere fact that Articles 33 and 34 of the Brussels I bis Regulation now expressly permit Member State courts to decline jurisdiction over, inter alia, such disputes where concurrent proceedings are pending in the courts of a third State, that any possibility to do so is excluded in all other scenarios. Again, the text of that regulation is simply silent on that question. Indeed, nothing in the wording of those provisions or in the related recitals indicates that they are meant to regulate exhaustively the possibility for Member State courts to decline jurisdiction in favour of the courts of third States. (108)

119. Thirdly, the judgment in Owusu, in fact, provides no authority for the literal interpretation suggested by BSH, the French Government and the Commission. Admittedly, the Court formulated its statement about the only possible derogations to the mandatory effect of (now) Article 4(1) of the Brussels I bis Regulation being the ones expressly provided therein in unqualified terms. However, at the same time, it refused to answer the second question of the referring court, which concerned, specifically, whether Member State courts are bound to exercise jurisdiction also when seised of disputes in which the subject matter is closely related to third States, or notwithstanding choice-of-court agreements designating third-State courts, because those situations were not at issue in the main proceedings. Clearly, if that statement were meant to cover those situations as well, the Court would have replied to both questions together. Instead, it specifically excluded them from its judgment. (109)

120. Finally, at least two other decisions of the Court, not discussed by those interveners and delivered respectively before and after the judgment in Owusu, indicate (quite clearly, in my view) that Member State courts are, in fact, not bound to hear and determine disputes having such close connections with third States, despite the absence of express provisions to that effect in the Brussels regime.

121. In the judgment in Coreck Maritime, the Court, having decided that Article 17 of the Brussels Convention is not ‘applicable’ to choice-of-court agreements designating third-State courts, immediately specified that, if it is seised notwithstanding such an agreement, a Member State court ‘must … assess the validity of the [agreement] according to the applicable law, including conflict of laws rules, where it sits’. (110) The logical implication of that reasoning is that if that court finds that agreement to be valid, it may give effect to it and decline jurisdiction in favour of the designated courts.

122. In the judgment in Mahamdia, the Court took the view that the courts of a Member State, having jurisdiction over an employment dispute pursuant to the protective rules (then) laid down for such matters in the Brussels I Regulation, were not permitted to give effect to a choice-of-court agreement designating the courts of a third State. Crucially, however, it did so on the ground that the agreement at issue did not respect the limits laid down in the Brussels regime with respect to party autonomy in employment matters. (111) The logical implication is that, where such limits are respected (see, further, point 150 below), Member State courts are permitted to give effect to jurisdiction agreements in favour of third-State courts. (112)

123. Although those judgments concerned choice-of-court agreements, the general idea that emerges from them (that Member State courts are permitted, in certain situations, not to exercise jurisdiction, even in the absence of express provisions to that effect in the Brussels regime) is, in my view, transposable to disputes whose subject matter is closely related to third States.

(2)    The teleological and consistent interpretation of the regulation

124. Turning to principles, it seems to me that, while, as indicated above, it would be undesirable to completely deprive Member State courts of jurisdiction over disputes which have connections of the kind envisioned in Article 24 or Article 25 of the Brussels I bis Regulation with third States, it would be unjustifiable to oblige those courts to rule on them.

125. First, such an interpretation would be at variance with the very reasons, related to fundamental principles, underpinning Articles 24 and 25 of the Brussels I bis Regulation.

126. The raison d’être of (most of) the rules of exclusive jurisdiction laid down in Article 24 of the Brussels I bis Regulation is, I recall, the deference due to certain sovereign rights and interests. Disputes over rights in rem in immovable property (Article 24(1) of that regulation) touch upon the traditional sovereignty of States to control land within their borders. Those concerned with the validity of entries in public registers or patents (Article 24(3) and (4) thereof) question the functioning of the public administrations involved. Those surrounding the enforcement of judgments (Article 24(5) thereof) directly concern the right, reserved to States, to exercise enforcement power on their territory. States would not accept that foreign courts interfere with such matters. They can only be adjudicated by ‘their’ tribunals. (113) Remarkably, that policy was regarded as so important by EU legislature that it provided for Article 24 of the Brussels I bis Regulation to apply as soon as a Member State’s sovereign rights and interests are at stake, irrespective of the domicile of the defendant. (114)

127. Accordingly, I fail to see the logic behind the contention that the courts of a Member State are, on the one hand, prohibited to adjudicate on the validity of a title of land situated in another Member State or the suitability of enforcement measures taken by its authorities (and so on), but are, on the other hand, generally mandated to do exactly that where a third State is concerned. The notion that, in the main proceedings, the Swedish courts are prohibited from adjudicating the validity of, say, the German part of the European patent at issue, even as a mere preliminary question, but would have been obliged, if seised to that effect, to issue a declaration as to the validity of the Turkish one, is baffling; the same sovereign rights and interests would be respected in the first case, but completely dismissed in the other. (115)

128. Such an interpretation of the Brussels I bis Regulation would be questionable from the point of view of public international law. Although the issue is disputed, the majority view is that international law sets boundaries to States’ adjudicatory jurisdiction over civil matters. (116) Admittedly, as the Commission submits, where the defendant in a civil dispute is domiciled in a Member State, such a connection to its territory usually entitles the latter, in public international law, to adjudicate on the matter. However, where that dispute impinges upon the rights of another State, its claim to jurisdiction is paramount. The location of the defendant’s domicile could hardly serve as justification for the first State to interfere in the domestic affairs of the second. That could be seen as a breach of the principle of sovereign equality. (117) The Brussels I bis Regulation must be interpreted consistently with those fundamental principles. (118) The system laid down in that regulation is not isolated from the rest of the world and it cannot entirely invalidate the claims to exclusive jurisdiction of third States.

129. For its part, Article 25 of the Brussels I bis Regulation reflects a policy in favour of party autonomy. The EU legislature found it desirable to promote the ability of contractual partners to ‘choose their judge’. (119) By resolving the issue of which court(s) would adjudicate potential disputes arising from contracts in advance, choice-of-court agreements increase legal certainty and the predictability of litigation, which, in turn, fosters international commerce (hence the frequent use of that instrument by businesses). Here again, the EU legislature considered that policy so important that, where the courts of a Member State are designated in such an agreement, it mandated, in principle, (120)any other courts to cede jurisdiction to the courts chosen by the parties, and Article 25 to apply irrespective of the defendant’s domicile. (121)

130. Accordingly, I fail to see once more the logic behind the contention that Member State courts should, on the one hand, be bound to enforce choice-of-court agreements in favour of the courts of another Member State but, on the other hand, should generally ignore similar agreements designating third State courts. That would subvert the policy pursued in the Brussels regime. The will of the parties would be respected in the first case but dismissed in the other. If Member State courts would be required to assume jurisdiction notwithstanding such agreements, those instruments would lose their purpose of ensuring legal certainty. For example, a European Union-based company and a company established in the United States could not reach a binding compromise in favour of the courts of New York (United States). The United States company could freely break its commitment by suing before the courts of the Member State where the EU company is established. If seised first, those courts would be denied the permission to enforce the agreement in question. (122)

131. Such an interpretation of the Brussels I bis Regulation would, again, be questionable from the point of view of overarching norms, namely, this time, fundamental rights. Indeed, party autonomy is an expression of freedom of contract, which is protected under EU law by, inter alia, Article 16 of the Charter. (123) That freedom entails that, in principle, a State’s legal system gives effect to the will of the parties to a contract. If Member State courts were prohibited, under an extreme view of the ‘mandatory effect’ of the Brussels I bis Regulation, from giving effect to choice-of-court agreements in favour of third-State courts, that would entail a severe and, in my view, unjustifiable limitation to that freedom. (124) Accordingly, that interpretation cannot be endorsed by the Court. (125)

132. Secondly, that interpretation would also go against the general objectives of the Brussels regime. Notably, it seems to me that obliging the courts of the Member States, where competent under the rules of the Brussels I bis Regulation, to adjudicate on the validity of titles of property situated in third-State territories, or of third-State patents (and so on), or disputes covered by choice-of-court agreements designating third-State courts, would hardly contribute to the legal certainty that that regulation seeks to achieve in terms of jurisdiction.

133. That is obvious, in particular, with respect to choice-of-court agreements. Parties expect that only the courts chosen by them will hear and determine their potential dispute. That some Member State courts might be obliged to rule notwithstanding such agreement would shatter those expectations.

134. Moreover, as Electrolux submits, such a solution would hardly contribute to the sound administration of justice. Indeed, while the decisions that the courts of the Member States would issue on such disputes would be considered valid in the European Union, they would (precisely because they impinge on a sovereign matter or because they were delivered in breach of a choice-of-court agreement) most likely be ignored in the third States in question. (126) Plainly, a ruling on the validity of a property title, or of a patent, is of little value if it cannot be enforced in the State where the property is located or the patent registered. As for a ruling delivered in breach of a choice-of-court agreement, while the claimant may find some strategic advantage in bringing proceedings in the Member State where the defendant is domiciled (since usually the assets of the latter are located in that State), because the judgment would be denied effect in the third State initially chosen, nothing would prevent the same dispute from being re-adjudicated there by the other party. On top of that, in all those scenarios, irreconcilable decisions may end up being delivered by Member State and third-State courts on the same dispute.

(3)    International conventions do not provide a comprehensive answer

135. The French Government and the Commission observe that, in the absence of dedicated provisions in the Brussels I bis Regulation, a solution to the objections expressed above may be found in the international conventions dealing with jurisdiction in civil matters, which bind the European Union and its Members. Under certain conditions, (127) such conventions take precedence over the Brussels regime. Where those conventions are applicable, the jurisdiction of the courts of the Member States is governed by their rules instead. The two main instruments (128) in that regard are the Lugano II Convention and the Hague Convention of 30 June 2005 on Choice of Court Agreements. (129) The first contains provisions equivalent to Articles 24 and 25 of the Brussels I bis Regulation. The second mandates the enforcement, by the courts of the Contracting Parties, of exclusive choice-of-court agreements.

136. Admittedly, those international conventions provide for an ideal solution to the issues discussed above. They ensure that Member State courts respect the rights and interests of the third States involved, as well as those of private parties willing to entrust their disputes to certain third-State courts. They also guarantee reciprocity from the States concerned, as well as the recognition and enforcement, in all the Contracting Parties, of the decisions given by the courts of their counterparts.

137. However, the solution so provided is inevitably (very) partial. By nature, those conventions take precedence over the Brussels regime only in the cases where the third State concerned by the dispute, or whose courts were designated by a choice-of-court agreement, is a party to the convention in question. (130) In that regard, for instance, the Lugano II Convention only binds, besides the Member States, the European Free Trade Association States and the Swiss Confederation. It offers no solution where the patent at issue is, like in the present case, registered in Türkiye. Similarly, to date, only a handful of third States are bound by the Hague Convention. (131) Thus, only in some situations do such conventions provide a solution to the issues discussed above.

138. Obviously, the more third States participate in such international conventions with the European Union, especially the Hague Convention, the more relevant such conventions will become in international litigation and contribute to the certainty of the field. Nonetheless, pragmatically speaking, never every third State in the world, or almost the majority of them, will do so. Thus, multilateral solutions do not (and will never) alleviate the need for sound unilateral solutions in the Brussels regime. The obligation for Member State courts to adjudicate on disputes closely connected to third States is not one of those solutions.

(4)    The alleged ‘clear intent’ of the EU legislature

139. BSH, the French Government and the Commission nonetheless argue, or at least imply, that the ‘clear intent’ of the EU legislature, when adopting the Brussels I bis Regulation, was that Member State courts, where competent under that instrument, be bound to hear and determine disputes having strong connections with third States, save where Articles 33 and 34 of that regulation apply. While it is well aware of the shortcomings of that solution, enumerated above, and thus unsatisfied with the result, the French Government submits that it is not for the Court to correct the will of the legislature by way of interpretation.

140. I cannot but agree with the general idea behind that objection. Nonetheless, I do not think that it bears weight in this particular instance.

141. First, together with Advocate General Bobek in his Opinion in BV, (132)I am of the view that the intent of the EU legislature is, in principle, only conclusive where it is clearly expressed in the adopted legislation. As explained, inter alia, in point 118 above, that is not the case here. If the will of the legislature was that explained by the French Government and the Commission, it would have had ample opportunity to state it, at least, in a dedicated recital of the Brussels I bis Regulation.

142. Secondly, with respect to what happened during the legislative process that led to the adoption of that regulation, I also concur with Advocate General Bobek that, generally speaking, the Court should avoid undertaking ‘an almost archaeological excavation’ of the travaux préparatoires of an instrument and feel bound by events, ideas and intentions it (re)discovered that way, especially because such an exercise usually does not reveal a clear picture, but a complex and blurry one. (133)That is precisely the case here.

143. As the French Government and the Commission emphasise, it is clear from the relevant travaux préparatoires that the EU legislature was aware of the issue surrounding the application of the Brussels regime to ‘external’ situations. (134) Furthermore, while the initial proposal of the Commission only contained a new provision on concurrent proceedings, proposals to include rules providing for discretion to decline jurisdiction in cases where the subject matter closely related to a third State or a choice-of-court agreement designating third-State courts applied were put forward at several steps of the legislative process, both before the European Parliament (135) and the Council, notably by the French and UK delegations. (136)Evidently, those proposals were rejected by the legislature, as only the rules on concurrent proceedings were kept in the final text (which became Articles 33 and 34 of the Brussels I bis Regulation).

144. However, one should not jump to any conclusions. There is, in the publicly available documents, little (if any) explanation as to why the EU legislature rejected those proposals (137)or, most importantly for the present purpose, as to the consequences that the absence of such dedicated rules in the Brussels regime should, in its mind, have on the jurisdiction of Member State courts with respect to disputes closely connected to third States. Notably, I did not find any indication, let alone a clear statement, that the legislature sought, by not adding such rules, to deny the possibility for Member State courts to decline jurisdiction. In fact, the only internal document of the Council that I could find that substantially discusses the proposals of the French and United Kingdom Governments is a note from the German delegation, which indicates the contrary. Therein, that delegation declares its opposition to those proposals but, crucially, on the ground that ‘the Brussels I Regulation does not definitively regulate the international jurisdiction of courts in the Member States vis-à-vis courts in third States’ and that, consequently, it should, in particular, ‘continue to be left for the national law of the Member States independently to regulate the effect of an agreement conferring jurisdiction on the courts of a third State’. (138)That explanation echoes the judgment in Coreck Maritime, discussed in point 121 above.(139)

145. Besides, the refusal of the legislature to add such rules, save for Articles 33 and 34 of the Brussels I bis Regulation, needs, in my view, to be considered within its context. I recall that the original intent of the Commission, through its legislative proposal, was to proceed to a complete ‘internationalisation’ of the Brussels regime by extending it to defendants domiciled in third States. (140) However, the legislature rejected the idea, inter alia because of the impact that such an extension could have on the relations of the European Union with its international partners and the difficulties litigants would face in having the decisions delivered by the courts of the Member States recognised abroad. In that context, it seems that the legislature did not mean to impose a comprehensive solution on the issues of disputes closely connected to third States. Rather, it wanted to deal with the issue a minima, regulating only concurrent proceedings and leaving the rest to national law (for now). (141)

146. Finally, and in any event, I do not think it ever was in the EU legislature’s power to oblige the courts of a Member State to rule on matters impinging on the sovereign rights and interests of third States, or to require those courts to generally  deny choice-of-court agreements pointing to such a State. Evidently, public international law and Article 16 of the Charter, as higher norms, frame the legislature’s discretion when adopting an instrument of secondary EU law such as the Brussels I bis Regulation. I have explained, in points 128 and 131 above, why such a solution would be incompatible with those higher norms. Hence, while the legislature could legitimately choose to add, in that instrument, rules governing the conditions under which a court of a Member State can decline jurisdiction in favour of third-State courts in those situations, or not to add such rules, the implication of the latter choice cannot possibly be the one defended by the French Government and the Commission.

(d)    Member State courts may decline to hear such disputes, where they ‘reflect’ the system of the regulation

147. Together with Electrolux, I am of the view that, while the courts of a Member State may have jurisdiction, under the rules of the Brussels I bis Regulation, in disputes whose subject matter is closely related to third States or is covered by an exclusive choice-of-court agreement designating their tribunals, that regulation permits them to decline to hear the case. It is, quite simply, the only rational answer, achieving the purpose of that instrument and ensuring its consistency with higher norms.

148. For those particular disputes, an implicit derogation from the mandatory effect of Article 4(1) of that regulation must be acknowledged. As that instrument does not (yet) contain provisions conferring on the courts of the Member States the power to decline jurisdiction in such scenarios, it means that its silence is to be construed in the sense that it allows those courts to do so on the basis of their national law. That is what the Court hinted at in the judgment in Coreck Maritime.

149. That being said, Member State courts do not have an unfettered discretion to decline jurisdiction, based on their national law, over those disputes (as that last judgment, read in isolation, could imply). As Droz explained a long time ago, and as the overwhelming majority of courts (142) and academics (143) that have dealt with the matter have confirmed since, Member State courts may do so only where it ‘reflects’ the system established in the Brussels I bis Regulation. Specifically, they can do so in cases where, had there been a similar connection with a Member State, the court would have had to relinquish jurisdiction under Articles 24 or 25 of that regulation. (144)

150. That requirement of ‘reflectivity’ implies, first, that a court of a Member State may refuse to exercise jurisdiction over a dispute connected to a third State only where the matter in dispute falls within the substantive scope of one of the rules of exclusive jurisdiction laid down in Article 24, (145) or where that court is seised notwithstanding a choice-of-court agreement which fulfils the requirements laid down in Article 25 (146) (so that, in a similar ‘intra-EU’ dispute, the relevant article would have applied). Secondly, with respect to those agreements, following the judgment in Mahamdia, ‘reflectivity’ implies that such a course of action be taken only where the limits imposed on their effects in ‘intra-EU’ disputes are respected. A Member State court cannot give effect to such an agreement where the dispute falls under the exclusive jurisdiction of the courts of another Member State, or where that agreement does not comply with the provisions of the regulation which protect weaker parties (policy holder, consumer or employee). (147)

151. Such a requirement of ‘reflectivity’ is both logical and crucial. Indeed, if a Member State court may decline jurisdiction in an ‘external’ situation where, under Articles 24 or 25 of the Brussels I bis Regulation, it would have had to do the same had the situation been ‘intra-EU’, the consistency of the Brussels regime is ensured. It also permits the attainment of the policies pursued under those provisions in both cases. By contrast, there would be no reason for ‘external’ situations to be treated more favourably than ‘intra-EU’ disputes. If EU law did not frame the powers of national courts, the paramount rules that confer exclusive jurisdiction on Member States, or that protect weaker parties, could easily be circumvented by businesses by simply including third-State choice-of-court agreements in their contracts. Thus, those rules would lose a significant part of their effectiveness. (148)

152. Where those conditions are fulfilled, the requirement of ‘reflectivity’ does not go as far as requiring a Member State court to decline jurisdiction in the manner that Articles 24 or 25 of the regulation require in the cases to which they apply directly, that is to say, automatically. I have explained in point 110 above why that cannot be the case: a risk of denial of justice could ensue, in breach of Article 47 of the Charter. Accordingly, that court must be accorded some limited discretion, to verify that (i) the third-State courts concerned indeed have exclusive jurisdiction under their own rules of private international law and (ii) those parties may obtain an effective remedy there. If that is not the case, the Member State court seised should not decline jurisdiction; it should do so in the opposite scenario. Where the matter is not clear, it should stay proceedings until the parties have seised the third-State court in question (and it is clear that that court will adjudicate the matter) in proceedings that offers fair-trail guarantees. (149)

153. In the case in the main proceedings, it follows from the foregoing considerations that, while Article 24(4) of the Brussels I bis Regulation does not apply directly with respect to the issue of the validity of the Turkish patent, contested by Electrolux by way of a defence, that provision may have ‘reflexive effect’ on the jurisdiction of the Swedish courts. It means that those courts may use the powers available to them under their national law to decline to adjudicate that issue and to stay proceedings until the Turkish courts determine the validity of that patent, under the conditions explained above.

154. Contrary to what the French Government implies, recognising such an implicit derogation to the mandatory effect of Article 4(1) of the Brussels I bis Regulation does not go contra legem or, stated differently, ‘rewrites’ that regulation. That solution does not require the Court to depart from the text which, I recall, is silent on the issue under consideration and, thus, open to such a teleological and consistent interpretation.If anything, it may well incite the EU legislature to deal with the matter when revising that instrument.

155. Recognising such (limited) room for national law does not mean reversing the judgment in Owusu either, despite what BSH and the Commission argue. I recall that the Court specifically excluded the hypothesis of disputes closely related to third States, or choice-of-court agreements in favour of third-State courts, from its analysis. Accordingly, it can perfectly well recognise an exception, tailored to those situations, to its seemingly unqualified statement about the mandatory effect of (current) Article 4(1) of the Brussels I bis Regulation. (150) Furthermore, doing so reconciles that judgment with that in Coreck Maritime. In fact, the interpretation suggested in the present Opinion has the non-negligible advantage of conciliating all the decisions related to the issue under consideration (from the earliest, the judgment in Coreck Maritime, to the most recent, the judgment in IRnova), where any other interpretation would require the Court to depart from certain of those precedents.

156. Those interveners nevertheless retort that, if Member State courts were permitted to decline jurisdiction in the scenarios at issue, under the powers reserved to them under their national law, that would be contrary, if not to the letter, at least to the rationale of, the judgment in Owusu. Indeed, such a solution would create the same issues as the doctrine of forum non conveniens.

157. In my view, the comparison does not withstand close scrutiny.

158. The application of the forum non conveniens doctrine, at issue in the judgment in Owusu, would have given Member State courts a wide discretion not to exercise jurisdiction in any ‘external’ disputes, based on the appropriateness of the forum, taking into account a wide range of factors. Such a flexible, case-by-case approach was completely at odds with the spirit of the Brussels regime, based on clear-cut rules. It would have seriously affected the predictability of jurisdiction under that regime, something which, in turn, would have undermined the principle of legal certainty and the legal protection of persons established in the European Union (as the defendant would not have been able to reasonably foresee where he or she could end up being sued, and the claimant would have had no guarantee that the court seised, while designated by that regime, would actually hear the case). Finally, the uniform application of that regime in the Member States would have been affected, in so far as that doctrine is recognised only in a few of them. (151)

159. By contrast, here, I propose to recognise a narrow discretion for Member State courts to decline jurisdiction in specific circumstances where, and to the extent that, it ‘reflects’ the solutions applicable under Articles 24 and 25 of the Brussels I bis Regulation. Thus, the grounds on which a Member State court may do so (exclusive jurisdiction of a third State and choice-of-court agreement in favour of the courts of the latter) are limited and precise, as opposed to the forum non conveniens doctrine. Hence, recognising such discretion does not give rise to the kind of uncertainty that the application of that doctrine would have created. The Brussels regime admits granting courts a limited discretion where it serves its goals. (152) That is the case here.

160. Such discretion does not affect the predictability of jurisdiction. A reasonably well-informed claimant may foresee that he or she would have to sue (and a defendant that he or she might be sued) in relation to the subject matters envisaged in Article 24, in the courts of the third State concerned. It could also hardly come as a surprise to those litigants that proceedings have to be brought before the courts designated in a prior agreement they concluded. In fact, that solution enhances the predictability of jurisdiction as it contributes to the comparable treatment of similar ‘intra-EU’ and ‘external’ disputes.

161. It also enhances legal certainty, for it allows Member State courts, notably, to give effect to agreements designed to ensure it. The legal protection of persons established in the European Union is also enhanced. It avoids the defendant having to face, in the courts of his or her Member State, proceedings that would result in pointless decisions, or abusively started in breach of such an agreement. The legal protection of the claimant is also ensured, as he or she will still be able to bring proceedings before the courts of the Member States if he or she could not obtain justice before the third State courts concerned.

162. Finally, as for the uniform application of the rules of the Brussels I bis Regulation in the Member States, I observe that, where a few Member States applied the doctrine of forum non conveniens, it is widely recognised, in those States, that courts should not entertain proceedings relating to disputes whose subject matter is closely related to a third State, and that jurisdiction agreements designating foreign courts should in principle be enforced. (153) Thus, the courts of all the Member States generally have the power, in their national law, to decline jurisdiction in such circumstances. While the precise conditions under which they do so could, in principle, vary from one Member State to the other, I recall that EU law frames national law in quite a significant way, as explained in points 150 and 152 above, which ensures that such a solution is applied by Member State courts in a sufficiently consistent way.

V.      Conclusion

163. In the light of all the foregoing considerations, I propose that the Court of Justice answer the questions referred by the Svea hovrätt (Svea Court of Appeal, Stockholm, Sweden) as follows:

(1)      Article 24(4) of Regulation (EU) No 1215/2012 of the European Parliament and of the Council of 12 December 2012 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters

must be interpreted as meaning that where the courts of a Member State are seised of proceedings concerned with the infringement of a patent registered in another Member State, and an invalidity defence is raised by the alleged infringer, those courts have no jurisdiction to adjudicate the validity issue.

(2)      Article 24(4) of Regulation No 1215/2012

must be interpreted as meaning that that provision does not apply with respect to the validity of a patent registered in a third State. However, the courts of the Member States, where competent under another rule of that regulation, may not adjudicate that issue.

---

1      Original language: English.

---

2      Regulation of the European Parliament and of the Council of 12 December 2012 (OJ 2012 L 351, p. 1) (‘the Brussels I bis Regulation’).

---

3      Judgment of 13 July 2006 (C‑4/03, ‘the judgment in GAT’, EU:C:2006:457).

---

4      See Article 1, Article 2(1) and Articles 10 to 25 of the EPC. The regime laid down in the EPC is not part of EU law, as the European Union itself is not a party to that treaty.

---

5      See Articles 52 to 57 and 94 to 97 of the EPC.

---

6      See Articles 3 and 79 of the EPC.

---

7      See Article 2(2) and Article 64(1) of the EPC.

---

8      Articles 99 to 105 of the EPC set out an opposition and limitation procedure through which a person may seek, before the EPO, a centralised, European-wide revocation of a ‘European patent’ within nine months of its grant.

---

9      See Article 138 of the EPC. I underline that, following a tumultuous series of debates and proposals, a ‘European patent with unitary effect’ has been created through enhanced cooperation between several Member States. Pursuant to Regulation (EU) No 1257/2012 of the European Parliament and of the Council of 17 December 2012 implementing enhanced cooperation in the area of the creation of unitary patent protection (OJ 2012 L 361, p. 1), a ‘European patent’ granted by the EPO through the centralised EPC procedure may, under certain conditions, be attributed ‘unitary effect’ in respect of all the participating Member States. Such a patent confers on its holder uniform protection and rights across all those States. Furthermore, as a unitary title, it may only be revoked in respect of all those States. The new regime entered into force on 1 June 2023, however it is not relevant for the purpose of the present case; indeed, the title at issue is not a ‘European patent with unitary effect’.

---

10      See Article 64(3) of the EPC and Article 8(1) of Regulation (EC) No 864/2007 of the European Parliament and of the Council of 11 July 2007 on the law applicable to non-contractual obligations (Rome II) (OJ 2007 L 199, p. 40).

---

11      By contrast, litigation of ‘European patents with unitary effect’ was conferred on a special jurisdiction, the Unified Patent Court (see Agreement on a Unified Patent Court (OJ 2013 C 175, p. 1)).

---

12      The EPC does include a Protocol on Jurisdiction and the Recognition of Decisions in respect of the Right to the Grant of a European Patent, of 5 October 1973. However, as its title indicates, that protocol governs jurisdiction over a narrow matter, namely claims to the ‘right to the grant of a European patent’ (see Article 60 EPC), formulated during the granting procedure.

---

13      Those actions come under ‘civil and commercial matters’ within the meaning of Article 1(1) (see judgment of 8 September 2022, IRnova, C‑399/21,‘the judgment in IRnova’, EU:C:2022:648, paragraph 30).

---

14      The Brussels I bis Regulation replaced Council Regulation (EC) No 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (OJ 2001 L 12, p. 1) (‘the Brussels I Regulation’), which had itself replaced the Convention on jurisdiction and the enforcement of judgments in civil and commercial matters, signed at Brussels on 27 September 1968 (OJ 1978 L 304, p. 36) (‘the Brussels Convention’). I underline that the Court’s interpretation of those preceding instruments is transposable to the equivalent provisions of the Brussels I bis Regulation (see, inter alia, judgment in IRnova (paragraphs 29 and 37)). Accordingly, in this Opinion, I will only distinguish between those instruments where necessary.

---

15      The jurisdiction of the EPO under the centralised revocation procedure provided in the EPC (see footnote 8 above) is reserved by Article 24(4) of the Brussels I bis Regulation.

---

16      See Article 25(4) and Article 26(1) of the Brussels I bis Regulation.

---

17      Finally, a judgment delivered by a court in breach of that rule may be refused recognition in another Member State (see Article 45(1)(e) of the regulation).

---

18      See judgments of 15 November 1983, Duijnstee (288/82, EU:C:1983:326, paragraphs 23 to 25); in GAT (paragraphs 15 and 16); and in IRnova (paragraphs 40 and 48). See also explanatory report on the Brussels Convention, prepared by Mr P. Jenard (OJ 1979 C 59, p. 1, at p. 36) (‘the Jenard report’).

---

19      Article 7(2) provides that a person may be sued, in ‘matters relating to tort’ in the ‘courts for the place where the harmful event occurred’. In that respect, first, the breach, by a third party, of the exclusive rights that a patent confers on its holder constitutes a ‘tort’ within the meaning of that provision. Secondly, the ‘place where the harmful event occurred’ is located on the territory of the State of registration of the patent infringed. Indeed, since the rights stemming from a patent are only protected on the territory of the State that issued it, those rights can only be breached there (see, to that effect, judgment of 19 April 2012, Wintersteiger, C‑523/10, EU:C:2012:220, paragraphs 27 and 28).

---

20      See, by analogy, judgment of 22 January 2015, Hejduk (C‑441/13, EU:C:2015:28, paragraphs 36 and 37 and the case-law cited).

---

21      See Opinion of Advocate General Jääskinen in Pinckney (C‑170/12, EU:C:2013:400, point 46).

---

22      Except, it would seem, the Swedish part.

---

23      See judgment in GAT (paragraph 17). A defendant in infringement proceedings may also bring a counterclaim in invalidity, which is not a mere defence, but an independent claim. The alleged infringer seeks not merely the dismissal of the infringement action, but a separate ruling on the validity of the patent at issue. In effect, it is an action for invalidity brought during the infringement proceedings. Thus, it falls in the first category.

---

24      See, for reference to the case-law of the courts of various Member States, Fawcett, J.J. and Torremans, P., Intellectual Property and Private International Law, Oxford Private International Law Series, 2nd ed., §§ 7.22 to 7.26.

---

25      See judgment in GAT (paragraph 31 and the operative part). Or, stated differently, that it applies ‘whatever the form of proceedings in which [that issue] is raised, be it by way of an action or a plea in objection’ (see paragraph 25 of that judgment).

---

26      Furthermore, the scenario of infringement proceedings where invalidity is pleaded as a defence was expressly taken into account by the Court in several parts of its reasoning (see judgment in GAT (paragraphs 17 and 26)).

---

27      See, inter alia, judgments of 8 May 2003, Gantner Electronic (C‑111/01, EU:C:2003:257, paragraphs 25 and 26); of 12 May 2011, Berliner Verkehrsbetriebe (C‑144/10, ‘the judgment in BVG’, EU:C:2011:300, paragraphs 37 and 38); and of 16 November 2016, Schmidt (C‑417/15, EU:C:2016:881, paragraph 25).

---

28      See, to that effect, judgments of 9 November 1978, Meeth (23/78, EU:C:1978:198, paragraphs 7 to 9), and of 13 July 1995, Danværn Production (C‑341/93, EU:C:1995:239, paragraphs 13 to 16). See also Hartley, T., Civil Jurisdiction and Judgements in Europe, Oxford University Press, Oxford, 2017, § 9.39.

---

29      See, inter alia, judgments of 3 October 2013, Schneider (C‑386/12, EU:C:2013:633, paragraph 29); of 23 October 2014, flyLAL-Lithuanian Airlines (C‑302/13, EU:C:2014:2319, paragraphs 40 and 41); and of 10 July 2019, Reitbauer and Others (C‑722/17, EU:C:2019:577, paragraph 44). See also the Jenard report, pp. 34 and 39, and Gothot, P. and Holleaux, D., La Convention de Bruxelles du 27 septembre 1968 : compétence judiciaire et effets des jugements dans la CEE, Paris, Jupiter, 1985, p. 83, § 141.

---

30      Some versions, such as the German language one, indicate that the rules of exclusive jurisdiction apply to proceedings which have as their ‘subject matter’ or ‘object’ an issue mentioned therein. Other versions use vaguer terms.

---

31      See judgment in BVG (paragraphs 37 and 47).

---

32      Judgment in GAT (paragraphs 24, 26 and 27).

---

33      The fact that the contentious rule of exclusive jurisdiction is ‘mandatory’ says a priori nothing about its scope or why it should apply to any proceedings where the validity of patents is discussed, even simply as a preliminary question. The argument of the Court that otherwise a ‘claimant would be able, simply by the way it formulates its claims, to circumvent … [that] rule of jurisdiction’ (judgment in GAT, paragraph 27 (emphasis added)) carries weight only in relation to the particular facts of the judgment in GAT. It cannot explain why that rule should apply even where invalidity is raised by the defendant, for instance in infringement proceedings.

---

34      See, inter alia, judgment in IRnova (paragraph 39 and the case-law cited).

---

35      See recital 15 of the Brussels I bis Regulation.

---

36      See judgment in BVG (paragraph 32).

---

37      Judgment of 12 July 2012, Solvay (C‑616/10, EU:C:2012:445, paragraph 47).

---

38      Besides infringement proceedings, an invalidity defence may be raised against an action for failure to pay the fee provided in a licensing agreement, and so on.

---

39      See recital 15 of the Brussels I bis Regulation as well as, inter alia, judgment of 9 December 2021, HRVATSKE ŠUME (C‑242/20, EU:C:2021:985, paragraph 30).

---

40      Judgment in GAT (paragraphs 28 and 29).

---

41      Agreement which constitutes Annex 1C to the Agreement establishing the World Trade Organization (WTO), signed at Marrakesh on 15 April 1994 and approved by Council Decision 94/800/EC of 22 December 1994 concerning the conclusion on behalf of the European Community, as regards matters within its competence, of the agreements reached in the Uruguay Round multilateral negotiations (1986-1994) (OJ 1994 L 336, p. 1)

---

42      See, inter alia, judgment of 5 October 2017, Hanssen Beleggingen (C‑341/16, EU:C:2017:738, paragraph 32 and the case-law cited).

---

43      See, inter alia, judgment in IRnova (paragraphs 36 and 40).

---

44      See, to that effect, judgment in GAT (paragraph 22).

---

45      Indeed, despite what the Court implied in paragraph 48 of its judgment in IRnova, that ‘close link in fact and law’ exists also with respect to infringement proceedings. Infringement of a patent is also determined in the light of the law of the State of registration. The factual proximity between the dispute and the territory of that State is similarly present (see points 22 and 28 above). For those reasons, in its case-law on Article 7(2) of the Brussels I bis Regulation, the Court has ruled that the courts of that State are also ‘best placed’ to hear such proceedings (see, inter alia, judgment of 19 April 2012, Wintersteiger, C‑523/10, EU:C:2012:220, paragraphs 27 and 28).

---

46      See Usunier, L., Compétence judiciaire, reconnaissance et exécution des décisions en matière civile et commerciale. – Compétence. – Règles de compétence exclusives. – Article 24 du règlement (UE) n° 1215/2012, JurisClasseur Droit international, Fasc. 584-160, 2015, § 3.

---

47      Quite the contrary. The patent laws of the Member States have been harmonised by several international treaties, in particular the TRIPS Agreement. Furthermore, with respect to disputes over the validity of European patents, I recall that the patentability requirements (which are at the heart of such disputes) are laid down in the EPC. Thus, they are identical in all the national laws of the Contracting Parties.

---

48      In any case, even if the sound administration of justice were the objective pursued by Article 24(4) of the Brussels I bis Regulation, that objective could not ‘require’ the solution in the judgment in GAT, as it may lead to unsound administration of justice with respect to infringement proceedings (see points 53 and 54 above).

---

49      The Jenard report, p. 36.

---

50      Something that was noted, but not discussed, by the Court in the judgment in GAT (paragraph 23).

---

51      That explanation, while valid with regard to ‘ordinary’ national patents, is less compelling with respect to European patents. I recall that those patents are granted by the EPO, the intervention of national administrations in that process being limited to the formal requirement of ‘validation’ in their respective States.

---

52      See, inter alia, Gothot, P. and Holleaux, D., op. cit., pp. 88 and 89, § 155; Treppoz, E., ‘Répertoire de droit international – Contrefaçon’, Dalloz, §§ 5, 9, 10, 11 and 30; Fawcett, J.J. and Torremans, P., op. cit., p. 19; Gaudemet-Tallon, H. and Ancel, M.-E., Compétence et exécution des jugements en Europe : matières civile et commerciale : règlements 44/2001 et 1215/2012, Conventions de Bruxelles (1968) et de Lugano (1988 et 2007), LGDJ, Paris, 6th ed., 2018, p. 162, § 120; and Usunier, L., op. cit., §§ 3 and 63.

---

53      The argument of the Court in the judgment in GAT(paragraph 30) that, in some national rules of procedure, a finding of invalidity contained in a ruling on infringement proceedings has a declaratory, erga omnes effect and that the interpretation of Article 16(4) of the Brussels Convention could not vary depending on the effect of such a ruling under national law missing the mark. The Court could perfectly (and legitimately) have decided, instead, that a rule of procedural law conferring erga omnes effect on validity findings made in infringement judgments should be set aside, on the ground that its application impairs the effectiveness of Article 16(4) (see judgment of 15 May 1990, Hagen, C‑365/88, EU:C:1990:203, paragraphs 17 and 20).

---

54      See, inter alia, Szychowska, K., ‘Quelques observations sous les arrêts de la Cour de justice dans les affaires C‑4/03 GAT et C‑539/03 Roche’, Revue de Droit Commercial Belge, No 5, 2007; Kur, A., ‘A farewell to cross-border injunctions? The ECJ decisions GAT v. LuK and Roche Nedertland v. Primus and Goldenberg’, International Review of Intellectual Property and Competition Law, Vol. 37, 2006, p. 844; Treppoz, E., op. cit.; and Ancel, M.-E., ‘Brevet – L’arrêt GAT : une occasion manquée pour la défence de la propriété intellectuelle en Europe’, Communication Commerce électronique, 2007, no 5, étude 10.

---

55      See European Max Planck Group on Conflict of Laws in Intellectual Property, Draft Principles for Conflict of Laws in Intellectual Property, 25 March 2011, Article 2:401(2) and Article 4:202. See, in the United States, The American Law Institute, Intellectual Property: Principles Governing Jurisdiction, Choice of Law, and Judgements in Transnational Disputes, 2008, §§ 211(2), 212(4) and 213(2).

---

56      The reason behind that codification seems mostly circumstantial. It appears that, shortly after the judgment in GAT was delivered, those terms were added in what would become Article 22(4) of the Convention on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, signed on 30 October 2007 (OJ 2007 L 339, p. 3) (‘the Lugano II Convention’) in order to ensure consistency between that provision and the corresponding rule of jurisdiction laid down in the Brussels regime, as interpreted in that judgment (see explanatory report by Pocar, F. on the Lugano II Convention (OJ 2009 C 319, p. 1), paragraph 102). Then, when the Council approved that convention on behalf of the European Union, it committed itself to do the same when recasting the Brussels I Regulation, to ensure the parallelism between the two instruments (see Annex I to Council Decision 2009/430/EC of 27 November 2008 concerning the conclusion of the Convention on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (OJ 2009 L 147, p. 1).

---

57      It flows from those elements that Article 24(4) of the Brussels I bis Regulation does not cover the ‘issue’ of validity of patents per se, but any ‘proceedings’ in which that issue is raised. Infringement proceedings in which an invalidity defence has been raised should be regarded as ‘concerned with the validity of patents’ within the meaning of that provision, the seemingly logical consequence of that classification being that those proceedings fall within the exclusive jurisdiction of the Member State of registration.

---

58      In the judgment in BVG, the Court, evidently aware of the tension between the interpretations it provided with respect to the rule of exclusive jurisdiction (now) laid down in Article 24(2) of the Brussels I bis Regulation and the judgment in GAT, attempted to distinguish the two cases. For present purposes, it suffices to recall that, in that context, it summarised in the judgment in GAT as saying that the courts of the Member State of registration have ‘exclusive jurisdiction to adjudicate upon any dispute in which the patent’s validity is contested’ (judgment in BVG, paragraph 46 (emphasis added)).

---

59      Judgment of 13 July 2006 (C‑539/03, EU:C:2006:458). In that judgment, the Court excluded the possibility, for a patent holder, to consolidate claims against several infringers before the courts for the domicile of one of them under Article 6(1) of the Brussels Convention (now Article 8(1) of the Brussels I bis Regulation). Amongst the reasons given in that regard the Court stated that, even if such consolidation had been possible, it could not have prevented ‘a partial fragmentation of the patent proceedings’ where validity of the patents is contested, because ‘that issue … is a matter of exclusive jurisdiction … in favour of the courts of the … State [of registration]’ (paragraph 40) (my emphasis).

---

60      Neither does it stem from the travaux préparatoires of the regulation. The legislature did not reject a proposal to indicate that, where invalidity is pleaded as a defence in infringement proceedings, the courts seised with those proceedings would merely have to stay proceedings until the courts of the State of registration had determined the validity issue. Such a proposal was never put forward by the Commission nor suggested during the legislative process. In fact, it appears that the matter had been scarcely discussed, if at all, by the legislature.

---

61      See, to that effect, judgments of 8 May 2003, Gantner Electronic (C‑111/01, EU:C:2003:257, paragraphs 24 to 32); of 15 May 2003, Préservatrice Foncière Tiard(C‑266/01, EU:C:2003:282); and in BVG (paragraph 35).

---

62      Admittedly, a patent holder may sometimes be able to anticipate that the alleged infringer will raise an invalidity defence. In the present case, Electrolux submits that BSH was perfectly capable of doing so because their dispute over both the validity and infringement of the European patent in question has been going on for more than 10 years. However, that may not always be the case. Even where a patent holder could foresee such a defence being raised, they may not know it for certain. Deciding where to sue would become a gamble based on the probability of invalidity being raised.

---

63      Contrary to what Electrolux submits, I do not think that such a consequence would be compatible with Article 27 of the Brussels I bis Regulation either. I recall that that provision obliges a court of a Member State to dismiss proceedings when seised of ‘a claim which is principally concerned with’ (my emphasis), inter alia, the validity of a patent registered in another Member State. To consider that infringement proceedings are ‘principally concerned with’ validity where that issue has merely been raised as a preliminary question would be to stretch those words.

---

64      The argument of Electrolux that Directive 2004/48/EC of the European Parliament and of the Council of 29 April 2004 on the enforcement of intellectual property rights (OJ 2004 L 157, p. 45, and corrigendum OJ 2004 L 195, p. 16) has harmonised the procedural rules applying to infringement proceedings, so that patent holders enjoy an equivalent level of protection in all the Member States, misses the mark. Indeed, having to start multiple proceedings instead of a single one is, in itself, an issue (even if each of the courts seised offer adequate guarantees).

---

65      Or the authorities of that State, more generally. Invalidity proceedings can also be brought before the relevant patent office.

---

66      See, for instance, Article 49 of the Code de procédure civile (French Code of Civil Procedure).

---

67      Indeed, since it is the alleged infringer challenging the validity of the patent in question, it is only logical that the burden of bringing invalidity proceedings rests on them. Furthermore, positive declaratory relief concerning the validity of a patent is not available in every Member State (see Report from the Commission to the European Parliament, the Council and the European Economic and Social Committee on the application of [the Brussels I Regulation] (COM(2009) 174 final, 21 April 2009, pp. 6 and 7; ‘the 2009 Commission report’).

---

68      See judgment in  BVG (paragraph 46).

---

69      Patent claims define the boundaries of the patented invention and, as such, lay down what the patent does or does not cover. Therefore, the construction of those claims is decisive to determine whether the acts committed by the alleged infringer fall within the scope of the patent and, thus, infringe the latter. Patent claims are also decisive in determining whether the patent is valid or not. Indeed, one of the usual conditions for the grant of a patent over a given invention is the ‘novelty’ of the latter. To assess that requirement, the invention in question, as defined in those claims, is compared with the ‘prior’ state of the art (see, with respect to European patents, Article 54 and Article 69(1) of the EPC).

---

70      Again, that complexity could have been avoided by simply allowing courts outside the State of registration, when seised of infringement proceedings, to adjudicate the validity of the patent as a preliminary question (see Section 1 above).

---

71      The referring court explains that, pursuant to the second subparagraph of Paragraph 61 of the Patentlagen, where an invalidity defence is raised in infringement proceedings, the defendant must bring a separate invalidity action.

---

72      See Tang, Z.S., ‘Validity in patent infringement proceedings – A new approach to transnational jurisdiction’, The Queen Mary Journal of Intellectual Property, 2021, pp. 47 to 68.

---

73      See recital 38 of the Brussels I bis Regulation and judgment of 17 November 2011, Hypoteční banka (C‑327/10, EU:C:2011:745, paragraph 48).

---

74      That would also allow strategic manoeuvres by the patent holder. They could advance a narrow construction of the patent claims before the validity judge in order to avoid a finding that the patent covers prior art and, thus, is invalid, and advance a different, broader construction of those claims before the infringement judge for the purposes of bringing the alleged infringer’s acts within its scope.

---

75      To be clear, even understood that way, as Electrolux maintains, the ‘narrow’ reading of the judgment in GAT is still not ideal from the point of view of the rights of defence. To defend effectively themselves against bogus infringement claims, a defendant is not able to simply raise a defence in the course of the infringement proceedings, but must, on top of that, start invalidity proceedings in the State(s) of registration of the patent(s) (which is not only less convenient, but also significantly more expensive to do). However, the rights of the defence are not unfettered prerogatives (see, inter alia, judgment of 15 March 2012, G, C‑292/10, EU:C:2012:142, paragraph 49). The Court in the judgment in GAT (and the EU legislature afterwards) evidently considered that such a limitation of those rights was necessary to ensure respect for the exclusive jurisdiction of the State of registration.

---

76      See European Max Planck Group on Conflict of Laws in Intellectual Property, Principles for Conflict of Laws in Intellectual Property – Draft, Article 2:703(1).

---

77      See Hess, B., Pfeiffer, T. and Schlosser, P., Report on the Application of Regulation Brussels I in the Member States (Study JLS/C4/2005/03), Ruprecht-Karls-Universität Heidelberg, 2007, § 818.

---

78      See, inter alia, judgment of 15 May 1990, Hagen (C‑365/88, EU:C:1990:203, paragraphs 17, 19 and 20).

---

79      Obviously, in infringement proceedings, Article 24(4) of the Brussels I bis Regulation comes into play if the alleged infringer has raised such a defence in compliance with the conditions that the procedural rules of the forum may impose as to timing and form. To the extent that those conditions do not make it excessively difficult or impossible, in practice, for the defendant to do so, they are not objectionable under EU law.

---

80      Notably, the referring court explains that, under the Swedish rules of procedure, an infringement action can be stayed pending determination of the invalidity action.

---

81      See, inter alia, Opinion of Advocate General Geelhoed in GAT (C‑4/03, EU:C:2004:539, point 46); Pocar, F., op. cit., § 102; and Hartley, T., op. cit., § 12.34.

---

82      See, by analogy, judgment of 12 July 2012, Solvay (C‑616/10, EU:C:2012:445, paragraphs 49 and 50).

---

83      Those courts could also deliver a provisional, enforceable decision on infringement, subject to a contrary decision on validity.

---

84      See Opinion of Advocate General Geelhoed in GAT (C‑4/03, EU:C:2004:539, point 46), and Tang, Z.S., op. cit., pp. 47 to 68.

---

85      Of course all the circumstances should be taken into account. The invalidity proceedings may already be so advanced that it is not necessary to stay the infringement proceedings (and so on).

---

86      See Article 9 of Directive 2004/48.

---

87      Judgment of 12 July 2012 (C‑616/10, EU:C:2012:445, paragraphs 31 to 51).

---

88      Judgment of 1 March 2005 (C‑281/02, ‘the judgment in Owusu’, EU:C:2005:120, paragraphs 24 to 35).

---

89      Some experts defend the opposite view. In their view, while the domicile of the defendant is the main criterion of applicability of the Brussels regime, it flows from Article 6(1) of the Brussels I bis Regulation that, with respect to such disputes, only the situation of the property at issue (and so on) or the courts designated by the agreement is relevant. Therefore, where those factors point to third States, the dispute falls outside the scope of the regulation. Unfortunately, that is not what the text of that regulation actually says. Article 6(1) thereof provides that the principle pursuant to which national rules of jurisdiction apply where the defendant is not domiciled in the European Union is ‘subject … to Articles 24 and 25’. The clear meaning of those terms is that, in such a scenario, the dispute is nonetheless governed by the rules of that regulation when the elements referred to in those two provisions point to a Member State. They cannot be stretched to mean that that regime does not cover disputes against EU defendants when they concern immovable property situated in, or a choice-of-court agreement designating the courts of, a third State. The case-law of the Court confirms that interpretation (see, inter alia, judgments of 19 July 2012, Mahamdia, C‑154/11, ‘the judgment in Mahamdia’, EU:C:2012:491, paragraph 39, and in IRnova (paragraphs 25 to 28).

---

90      To be fair, where its predecessors were completely silent, the Brussels I bis Regulation touches on the matter (see point 115 below).

---

91      Indeed, the French Government considers that that solution is undesirable but imposed by the text of the regulation (see points 115 and 139 below).

---

92      See Droz, G., Compétence judiciaire et effets des jugements dans le Marché commun, Paris, Dalloz, 1972, §§ 164 to 169 and 204.

---

93      See, inter alia, Mayer, P., Heuzé, V. and Remy, B., Droit international privé, LGDJ, Paris, 12^th ed., 2019, § 360.

---

94      The same way, Article 27 of that regulation obliges a Member State court to relinquish jurisdiction to ‘the courts of another Member State’ only.

---

95      Judgment in IRnova (paragraphs 34 and 35).

---

96      Judgment of 9 November 2000 (C‑387/98, ‘the judgment in Coreck maritime’, EU:C:2000:606, paragraphs 17 and 19, respectively).

---

97      See judgment of 9 December 2003, Gasser (C‑116/02, EU:C:2003:657, paragraph 72).

---

98      See Jenard, P. and Möller, G., Report on the Lugano Convention (OJ 1990 C 189, p. 57), § 54, and Almeida Cruz, M., Desantes Real, M. and Jenard, P., Report on the Convention of San Sebastián (OJ 1990 C 189, p. 35) § 25.

---

99      Opinion of 7 February 2006 (‘Opinion 1/03’, EU:C:2006:81).

---

100      See judgment of 31 March 1971, Commission v Council (22/70, EU:C:1971:32, paragraphs 17 to 22).

---

101      See point 135 below.

---

102      Opinion 1/03 (point 153). The fact that Article 4(1) may apply where the disputes present, with a third State, the kind of connections envisioned in Articles 24 and 25 of the Brussels I bis Regulation stems also from Articles 33 and 34 of that regulation (see footnote 107 below).

---

103      See judgment in Owusu (paragraph 37).

---

104      According to the doctrine of forum non conveniens, known in common law countries, a court may decline to exercise jurisdiction on the ground that a court in another State would be a more appropriate forum for the trial of the action (see the judgment in Owusu (paragraph 8)). In that case, in the view of the United Kingdom courts, that was Jamaica.

---

105      The Court also rejected the application of the forum non conveniens doctrine on principle grounds (see point 156 below).

---

106      Those provisions cover the situation where any given claim (Article 33) or related claims (Article 34) is pending, in parallel, in a Member State court and a third State court. Based on the relevant provision, the Member State court, where competent under Article 4(1) of the Brussels I bis Regulation, may decline jurisdiction in favour of the third-State court, under certain conditions, including that it is ‘necessary for the proper administration of justice’. In that regard, recital 24 of that regulation hints at disputes closely connected to third States, by stating that ‘when taking into account the proper administration of justice, the court of the Member State concerned’ should consider, inter alia,  ‘whether the court of the third State has exclusive jurisdiction in the particular case in circumstances where a court of a Member State would have exclusive jurisdiction’. Therefore, where Article 33 or Article 34 apply, Member State courts can indirectly take those connections into account and decline jurisdiction, not on those grounds per se, but because of the lis pendens situation. However, as stated above, those provisions apply only where the third-State court has been seised first, not where it has been seised second.

---

107      Article 4(1) of the Brussels I bis Regulation is, in my view, also silent on the issue at hand. Indeed, that provision needs to be read in conjunction with Article 5(1) thereof. The first indicates that a defendant ‘shall’ be sued before the courts of the Member State of their domicile, and the second that they ‘may be sued in the courts of another Member State only by virtue of the rules set out [in the regulation]’. Thus, those provisions only deal with the allocation of jurisdiction between the Member States. They do not address the question of whether and, if so, in what circumstances, the courts of the defendant’s domicile may relinquish jurisdiction to the courts of a third State.

---

108      See Briggs, A., Civil Jurisdiction and Judgments, Informa Law, Oxon, 2015, 6th ed., pp. 316 to 362, at p. 345, and Mills, A., Party Autonomy in Private International Law, Cambridge University Press, Cambridge, 2018, p. 217. Notably, Articles 33 and 34 of the Brussels I bis Regulation do not feature in a closed section, entitled ‘Possibility to decline jurisdiction in favour of the courts of third States’. Instead, they are part of Chapter II, Section 9, of the regulation, entitled ‘Lis pendens – related actions’. Those provisions do not indicate either, for instance, that ‘a court of a Member State may only decline jurisdiction in favour of the courts of a third State where …’. Instead, they indicate that ‘where’ there is a situation of lis pendens, the court of the Member State may do so if the conditions they provide are fulfilled. The same goes for recital 24 of that regulation. The latter simply indicates that, in case of concurrent proceedings, to assess whether to decline jurisdiction the Member State court may take into account the close connections of the case with the third State at issue. It does not say that such connections may be taken into account only in that situation.

---

109      See judgment in Owusu (paragraphs 48 to 52). It is sometimes argued that Opinion 1/03 supports such a literal interpretation. However, it does not, for it simply does not address the question of whether Member State courts are obliged to exercise the jurisdiction they derive from the Brussels regime. Similarly, in the judgment in IRnova, the Court did not address the issue of whether, as Article 24(4) is not ‘applicable’ to disputes concerning the validity of patents registered in third States, the courts of the Member State where the defendant is domiciled would be bound to adjudicate on the matter. The Court did not have to do so, because that case was not about validity (see paragraphs 36 to 48 of that judgment).

---

110      Judgment in Coreck (paragraph 19 and the case-law cited).

---

111      See judgment in Mahamdia (paragraphs 60 to 66).

---

112      See, to that effect, judgment in Mahamdia (paragraph 65). See, also, Kistler, A.R.E., ‘Effect of exclusive choice-of-court agreements in favour of third States within the Brussels I Regulation Recast’, Journal of Private International Law, Vol. 14, No. 1, 2018, pp. 79 and 81 to 83; Hartley, T.C., op. cit., §§ 13.19 to 13.21. Thus, to accept the view suggested by BSH, the French Government and the Commission would mean accepting also the awkward notion that the Court (1) implicitly overturned the judgment in Coreck Maritime in that in Owusu and (2) followed an entirely moot reasoning in the judgment in Mahamdia.

---

113      See, with respect to Article 24(1) the Brussels I bis Regulation, Opinion of Advocate General Poiares Maduro in ČEZ (C‑343/04, EU:C:2006:13, points 35 to 39); with respect to Article 24(4) of that regulation, point 60 above; and with respect to Article 24(5) of that regulation, judgment of 26 March 1992, Reichert and Kockler (C‑261/90, EU:C:1992:149, paragraph 26). See, generally, the Jenard report, pp. 35 and 38; Usunier, L., op. cit., §§ 3, 43 and 59; Droz, G., op. cit., §§ 137 and 156; Gaudemet-Tallon, H. and Ancel, M.-E., op. cit., §§ 104 and 118; Hartley, T., op. cit., p. 212; and Gothot, P. and Holleaux, D., §§ 154, 155 and 158.

---

114      See recital 14 as well as Article 6(1) and Article 24 of the Brussels I bis Regulation.

---

115      Or, more precisely, could be taken into account, but only in the case of concurrent proceedings, and not if the Member State courts were seised even just one day before the third-State courts (see point 115 above), which would be incoherent, if not absurd. The ‘special relationship’ between the European Union and its Member States cannot justify such a difference of treatment. The rights of those States are taken into account, in Article 24, not because of their capacity as Members of the European Union, but simply because they are sovereign entities in the international order.

---

116      See, inter alia, Roorda, L. and Ryngaert, C., ‘Public International Law Constraints on the Exercise of Adjudicatory Jurisdiction in Civil Matters’, in Forlati, S. and Franzina, P. (eds), Universal Civil Jurisdiction – Which Way Forward?, BRILL, 2020, pp. 74 to 95, and Mills, A., ‘Rethinking Jurisdiction in International Law, The British Yearbook of International Law, 2014, Vol. 84, No. 1, pp. 187 to 239.

---

117      See, inter alia, Roorda, L. and Ryngaert, C., op. cit., pp. 77, 82 and 87; Mills, A., Party Autonomy in Private International Law, p. 233; Usunier, L., op. cit., §§ 1, 43 and 67. BSH and the Commission nonetheless argued that, if the courts of a Member State were, for instance, to declare invalid a patent granted by a third State, it would not impinge on the sovereignty of that State, because that decision would not be recognised there. In my view, that argument misses the point on two accounts. First, a third State would deny any authority to such decision precisely because it would be perceived as an interference in a sovereign matter. Secondly, because such a decision could never be enforced in that State, a fortiori Member State courts should not be mandated to rule in the first place (see point 134 below).

---

118      See, inter alia, judgments of 16 June 1998, Racke (C‑162/96, EU:C:1998:293, paragraph 46), and of 26 April 2022, Poland v Parliament and Council (C‑401/19, EU:C:2022:297, paragraph 70). Even if the Court erroneously took the view that such limits to adjudicatory jurisdiction do not exist in international law, as Electrolux submits, it would still be repugnant to international comity for a Member State court to rule on matters that impinge on a third State’s rights.

---

119      See recitals 14 and 19 of the Brussels I bis Regulation, as well as, inter alia, judgment of 7 July 2016, Hőszig (C‑222/15, EU:C:2016:525, paragraph 44).

---

120      Provided the conditions laid down in Article 25 are fulfilled and the limits imposed by the rules on protective jurisdiction are respected (see point 150 below).

---

121      See recital 14, Article 6(1) and Article 25 of the Brussels I bis Regulation.

---

122      See point 115 above. It would encourage an unhealthy ‘race’ to the courts, each party seeking to be the first to act to either enforce or, on the contrary, trump the choice-of-court agreement.

---

123      See judgment of 12 January 2023, TP (Audiovisual editor for public television) (C‑356/21, EU:C:2023:9, paragraph 74 and the case-law referred).

---

124      Such a limitation would not fulfil the conditions laid down in Article 52(1) of the Charter. First, it would not be ‘provided for by law’, for such a drastic outcome would stem from the absence of dedicated rules in the Brussels I bis Regulation. Secondly, it could hardly be regarded as ‘genuinely meet[ing] objectives of general interest recognised by the Union’ (see more in point 161 below). Thirdly, even if such an objective existed, the incoherence, if not the absurdity, of that limitation would make it impossible to defend. I recall that Member State courts would be permitted to give effect to choice-of-court agreements in favour of third State, in case of concurrent proceedings, but not if Member State courts were seised even just one day before the third-State courts (see point 115 above).

---

125      See, inter alia, judgment of 26 April 2022, Poland v Parliament and Council (C‑401/19, EU:C:2022:297, paragraph 70 and the case-law cited).

---

126      The refusal to recognise foreign judgment delivered in violation of the required State’s exclusive jurisdiction is a common measure in national law. It is also provided, intra-EU, in Article 45(1)(e)(ii) of the Brussels I bis Regulation, as well as in international conventions, such as the Convention on the Recognition and Enforcement of Foreign Judgments in Civil or Commercial Matters (see Article 5(3) and Article 6 of that convention).

---

127      See Article 71(1) and Article 73(1) and (3) of the Brussels I bis Regulation as well as judgment of 4 May 2010, TNT Express Nederland (C‑533/08, EU:C:2010:243).

---

128      By contrast, I recall that the EPC does not contain rules of international jurisdiction, save for one exception, not relevant for the present case (see point 23 above). Therefore, as all the interveners emphasised, a State, such as Türkiye, which is a party to the EPC, is to be treated like any other third State for the purposes of the rules of the Brussels I bis Regulation.

---

129      Convention approved by Council Decision 2014/887/EU of 4 December 2014 on the approval, on behalf of the European Union, of the Hague Convention of 30 June 2005 on Choice of Court Agreements (OJ 2014 L 353, p. 5) (‘the Hague Convention’).

---

130      See Articles 22 and 23 of the Lugano II Convention and Article 3(a) of the Hague Convention.

---

131      Namely China, Mexico, Montenegro, Singapore, Ukraine and the United Kingdom (see Hague Conference on Private International Law, Status table, 37; Convention of 30 June 2005 on Choice of Court Agreements; accessible at https://www.hcch.net/en/instruments/conventions/status-table/?cid= 98.

---

132      C‑129/19, EU:C:2020:375, points 117 to 123.

---

133      Opinion of Advocate General Bobek in BV (C‑129/19, EU:C:2020:375, point 123).

---

134      The question was singled out, in view of the future recasting of the Brussels regime, by several studies commissioned by the Commission (see, inter alia, Hess, B., Pfeiffer, T., and Schlosser, P., op. cit., §§ 360 to 362 and 388) and in the 2009 Commission report (p. 5). Furthermore, the Commission consulted a stakeholder on that question (see European Commission, 21 April 2009, Green Paper on the review of Council Regulation (EC) no 44/2001 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (COM(2009) 175 final), p. 4).

---

135      See European Parliament, Committee on Legal Affairs, doc 2010/0383(COD), 19 October 2011, Amendments 106, 112 and 113.

---

136      Council of the European Union, doc. 9474/11 ADD 8, 8 June 2011, Note from the United Kingdom delegation to Working Party on Civil Law Matters (Brussels I), pp. 7 to 15; doc. 9474/11 ADD 14, 16 June 2011, Note from the French delegation to Working Party on Civil Law Matters (Brussels I), pp. 8 to 18; doc. 8000/12, 22 March 2012, Note from the United Kingdom delegation to Working Party on Civil Law Matters (Brussels I), pp. 1 to 8; doc. 8205/12, 27 March 2012, Note from the French delegation to Working Party on Civil Law Matters (Brussels I), pp. 1 to 7.

---

137      The (then) future approval by the European Union of the Hague Convention features a few times (see Council of the European Union, doc. 9549/12, 4 May 2012, Note from the delegation of the United-Kingdom to Working Party on Civil Law Matters (Brussels I), p. 2) and may explain why the legislature chose not to add rules on choice-of-court agreements in favour of third States in the regulation.

---

138      Council of the European Union, doc. 13756/11 ADD 1, 9 September 2011, Note from German delegation to Working Party on Civil Law Matters (Brussels I), p. 3 (emphasis added).

---

139      An argument is put forward in the literature that the EU legislature indeed wanted to deny Member State courts the right to give effect to choice-of-court agreements in favour of third-State courts, because that would give an incentive to third States to adhere to the Hague Convention (see Beaumont, P., ‘The revived Judgments Project in The Hague’, Netherlands InternationaalPrivaatrecht (NIPR), 2014, Vol. 4, pp. 532 to 539). Admittedly, the author was a member of the working group of the Council that designed the Brussels I bis Regulation. However, such an expert’s statement can hardly be enough to establish the will of the EU legislature, especially where contradicted by statements from other delegations. In any event, the notion that the legislature could ever have intended to sacrifice party autonomy under the regulation in order to promote it at the international level is perplexing.

---

140      See the Commission’s proposal, Article 4(2).

---

141      See, inter alia, European Parliament, Committee on Legal Affairs, doc. A7-0320/2012, 15 October 2012, Report on the proposal for a regulation of the European Parliament and of the Council on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters (recast) (COM(2010) 748)); Council, doc. 9474/11 ADD 9, 10 June 2011, Note from the Belgian delegation to Working Party on Civil Law Matters, p. 3.

---

142      See, inter alia, in the United Kingdom, High Court, Queen’s Bench Division (Commercial Court), Konkola Copper Mines plc v. Coromin [2005] 2 Lloyd’s Rep. 555; High Court, Queen’s Bench Division (Commercial Court), Ferrexpo AG v. Gilson Investments [2012] EWHC 721 (Comm); in France, Cour de cassation (court of cassation), Chambre Civile 1, 2 April 2014, 13-11.192, FR:CCASS:2014:C100356. See, further Nuyts, A., Study on Residual Jurisdiction, General Report, 2007, pp. 73 to 80.

---

143      See, inter alia, Nuyts, A., ‘La théorie de l’effet réflexe’, in Le droit processuel et judiciaireeuropéen, Bruxelles, La Charte, 2003, pp. 73 to 89; Mills, A., Party Autonomy in Private International Law, p. 138; Harris, J., ‘Stay of proceedings and the Brussels Convention’, ICLQ, 2005, Vol. 54, pp. 933 to 950; Bachmeir, T. and Freytag, M., ‘Discretional elements in the Brussels Ia Regulation’, Journal of Private International Law, 2022, Vol. 18, No. 2, pp. 296 to 316; Kistler, A.R.E., op. cit., pp. 66 to 95; Fawcett, J.J. and Torremans, P., op. cit., § 7.218; Usunier, L., op. cit., § 14; De Verneuil Smith, P., Lasserson, B. and Rymkiewicz, R., ‘Reflections on Owusu: The Radical Decision in Ferrexpo’, Journal of Private International Law, 2012, Vol. 8, no 2, pp. 389 to 405; Dicey, A.V., Morris, J.H.C. and Collins, L.A.C., The Conflict of Laws, Sweet & Maxwell U.K., 2012, 15^th ed., §§ 12.021 to 12.024; See Goodwin, J., ‘Reflexive effect and the Brussels I Regulation’, Law Quarterly Review, 2013, Vol. 129, pp. 317 to 320; Takahashi, K., ‘Review of the Brussels I Regulation: A Comment from the Perspectives of Non-Member States (Third States)’, Journal of Private International Law, 2012, Vol. 8, no 1, pp. 8 to 11; Briggs, A., op. cit., §§ 2.305 to 2.308; Gaudemet-Tallon, H. and Ancel, M.-E., op. cit., § 106; and Gothot, P. and Holleaux, D., op. cit., §§ 37 and 142.

---

144      Of course, that is an imperfect ‘reflection’ of the solution stemming from Articles 24 and 25 for intra-EU disputes. There, any other court than those designated is deprived of jurisdiction and is obliged to relinquish jurisdiction.

---

145      It is not enough, for instance, that a dispute concerns a third-State patent; it must concern, specifically, its validity or registration, as envisioned in Article 24(2) of the Brussels I bis Regulation.

---

146      Apart, obviously, from the one concerning the designation of the courts of a Member State.

---

147      See Articles 15, 19 and 23 and Article 25(4) of the Brussels I bis Regulation.

---

148      See, to that effect, judgment in Mahamdia (paragraph 64).

---

149      See Kistler, A.R.E., op. cit., pp. 89and 90; Nuyts, A., ‘La théorie de l’effet reflexe’, § 6; Mills, A., Party Autonomy in Private International Law, pp. 138 and 233; Goodwin, J., op. cit., pp. 317 to 320; Gaudemet-Tallon, H. and Ancel, M.-E., op. cit., § 106; Bachmeir, T. and Freytag, M., op. cit., p. 309; and Usunier, L., op. cit., § 14.

---

150      By analogy, in its judgment of 27 June 1991, Overseas Union Insurance and Others (C‑351/89, EU:C:1991:279), the Court laid down, in unqualified terms, a principle governing the operation of the rules of lis pendens of the Brussels regime (‘in no case is the court second seised in a better position than the court first seised to determine whether the latter has jurisdiction’) (paragraph 23). At the same time, it expressly reserved the scenario of the court second seised having exclusive jurisdiction (see paragraph 20). Eventually, in its judgment of 3 April 2014, Weber (C‑438/12, EU:C:2014:212, paragraph 53 et seq.), the Court acknowledged an exception to that principle, precisely for the scenario that it had initially reserved.

---

151      See judgment in Owusu (paragraphs 38 to 43).

---

152      See Articles 30, 33 and 34 of the Brussels I bis Regulation. Furthermore, the (slight) degree of uncertainty coming with such a limited discretion is infinitely preferable to the absolute certainty, and absurdity, that would come with the solution consisting in obliging the courts of the European Union to adjudicate such cases.

---

153      See Nuyts, A., Study on residual jurisdiction, §§ 93 to 96 and 103.