The Brussels I Regulation (No 44/2001)
Chapter I of the Brussels I Regulation contains a single Article, which is unchanged. The scope is the same as that of the Brussels Convention. It should be remembered that meanwhile the Brussels II Regulation 2003 (No 2201/2003) and the Regulation for insolvency proceedings (No 1346/2000) have entered into force. The first provides rules for jurisdiction and the recognition and enforcement of judgments in matrimonial matters and matters of parental responsibility. The second regulates jurisdiction, recognition and applicable law in (cross-border) insolvency proceedings. Both areas are for this reason excluded from the scope of the Brussels I Regulation (as they were excluded as well from the scope of the 1968 Brussels Convention). The Brussels I Regulation is confined in its scope to proceedings and judgments regarding international legal relationships, including relationships that involve not two Member States, but one Member State and one non-Member State (third State). It applies automatically, whether or not it is invoked by the parties, and it applies only to civil and commercial matters, irrespective of the nature of the court. The Regulation does not concern revenue, customs or administrative matters, but it may apply to disputes between public administrative authorities and individuals, in so far as the authorities have not acted in the exercise of their public powers. The Regulation’s material scope is also delimited by means of a list of matters excluded from it. See also:
The formal scope of the Brussels I Regulation with respect to jurisdiction is largely defined in Articles 2, 3 and 4 BR I. These Articles specify in which events, provided that the lawsuit falls within the material scope of the Brussels I Regulation (civil and commercial matters), the Brussels I Regulation is applicable. According to these Articles, decisive is only whether the defendant has his domicile within the European Union, therefore in one of the EU Member States, irrespective of the nationality of the parties to the lawsuit and irrespective of the domicile of the plaintiff. This means that the Brussels I Regulation not only applies when both, the plaintiff and the defendant, are domiciled in the European Union, but also when the plaintiff resides in a third State, which is not bound by the Brussels I Regulation, provided that the defendant has his domicile in one of the EU Member States. In order to understand this rule, it is important to notice that the question of jurisdiction is always dealt with from the point of view of the court which is actually seised by the parties to the lawsuit. The seised court only has to answer the question whether it has jurisdiction to give a ruling. If it thinks that it has jurisdiction, it will accept the case. If it thinks that it has no jurisdiction, it will refuse to take the claim into consideration and proclaim a non-suit. The seised court does not investigate if there is a possibility that the courts of other States are allowed to deal with the lawsuit too. Where parties have seised a court outside the European Union (or EFTA-territory), that court will, of course, not apply the Brussels I Regulation, nor any other European law. It will determine on the basis of the rules of private international law of the State of which it forms a judicial organisation whether it has jurisdiction or not. As far as that State is a Contracting Party to an applicable International Convention, it will take the provisions of that Convention into consideration. This means that the plaintiff sometimes has the possibility to choose between a court in one or more EU Member States or in a third State ('forum shopping'). Where the plaintiff decides to bring his case to a court of a third State, that court will investigate pursuant to its own rules of private international law whether it is competent to rule on the claim. If it thinks it has jurisdiction, its judgment will be valid in that specific third State, and it may be recognized and enforced even in other States, possibly including some EU Member States, on account of International Conventions or the internal law of the State where recognition and enforcement is sought. Again, the Brussels I Regulation plays no part in this. So the Brussels I Regulation only has any influence when a court of a EU Member State is seised by one or both parties involved at the proceedings. But in that case, the seised court, insofar the legal claim falls within the material scope of the Brussels I Regulation (civil and commercial matters), always has to draw in that Regulation in order to determine whether it has jurisdiction or not. A court of a EU Member State, seised by one of the parties, therefore will first have to investigate where the defendant is domiciled.
As mentioned before, the above rules apply irrespective of the nationality of the parties involved in the proceedings and irrespective of the domicile of the plaintiff. If the plaintiff is domiciled as well in a EU Member State, yet in another one than where the defendant has his domicile, the application of the Brussels I Regulation, therefore of the above mentioned rules, shall not cause any problems. All Member States of the European Union have subjected themselves to these rules, whereas Article 69 of the Brussels I Regulation specifies that this Regulation supersedes a number of bilateral Treaties between various EU Member States (or Yugoslavia) listed in that Article, insofar they govern the same subject-matters as the Brussels I Regulation. This means that all EU Member States in their mutual relationships have to apply the Brussels I Regulation, so that the outcome will be consistent in all EU Member States, including the one where the plaintiff has his domicile. But problems may arise when the plaintiff is domiciled in a third State, therefore not within the European Union. Although the composers of the Brussels I Regulation and the European Court of Justice (ECJ 13 July 2000 'Group Josi v UGIC', C-412/98) have repeatedly made clear that the applicability of the Brussels I Regulation does not depend on the domicile of the plaintiff, this only seems to be appropriate to the extent that a EU Member State is not bound by any other International Convention on jurisdiction in civil and commercial matters.
This, however, is merely a hypothetical issue. Except for the Lugano Convention 2007, the EU Member States are not bound by any other International Convention on jurisdiction which in general, in a similar way as the Brussels I Regulation, covers all kind of civil and commercial matters. The relation between the Brussels I Regulation and the Lugano Convention 2007 is well defined and does not lead to any conflicts. So the before raised question doesn’t have to be answered in practice. Nevertheless, there are bilateral and multilateral Conventions that point out for a specific subject-matter which of the participating Contracting States has jurisdiction. Such International Conventions aren’t pushed aside by the Brussels I Regulation. Article 71, paragraph 1, of the Brussels I Regulation confirms explicitly that this Regulation does not affect any International Convention to which the EU Member States are Contracting Parties and which in relation to particular matters govern jurisdiction or the recognition or enforcement of judgments. Many EU Member States are a Contracting Party to various International Conventions on the field of, for instance, aviation, maritime law and freight and carriage. Such Conventions often contain rules for jurisdiction with regard to specific topics. Examples are the Warsaw Convention of 12 October 1929 and the Montreal Convention of 28 May 1999, both relating to international carriage by air, the International Convention on the Arrest of Ships, concluded at Geneva on 12 March 1999, the Convention on Limitation of Liability for Maritime Claims, concluded at London on 19 November 1976, the Athens Convention relating to the Carriage of Passengers and their Luggage by Sea of 13 December 1974 and the Convention for the Suppression of Unlawful Acts Against the Safety of Maritime Navigation, concluded at Rome on 10 March 1988. Where a subject-matter is covered by one of such Conventions, the Brussels I Regulation plays no part, not even when it concerns a case between citizens or residents from two or more EU Member States mutually. A similar provision like Article 71, paragraph 1, BR I for International Conventions that in general deal with jurisdiction in civil and commercial matters wasn’t necessary, simply because such International Conventions did not exist at the moment on wich the Brussels I Regulation entered into force, whereas EU Member States are not allowed to participate in possible future Conventions of such kind.
As already shown, the rules of jurisdiction are based on the defendant's domicile. Also the domicile of the plaintiff may be of importance for the application of several provisions of the Brussels I Regulation. Contrary to various International Conventions, particular those drawn up within the framework of the Hague Conference on Private International Law, and contrary to Regulation (EC) No 2201/2003 concerning jurisdiction and the recognition and the enforcement of judgments in matrimonial matters and the matters of parental responsibility (‘the Brussels II Regulation’), the Brussels I Regulation does not link the question of jurisdiction to the habitual residence of the defendant (or plaintiff), but solely to his domicile.
In view of the connection that the Brussels I Regulation makes between jurisdiction and the domicile of the defendant (and sometimes that of the plaintiff), it is of the greatest importance to determine that domicile. Despite of that, the Brussels I Regulation does not include a common definition of domicile. Article 59, paragraph 1, BR I merely states that in order to determine whether a party is domiciled in the Member State whose court is seised, that court has to apply its internal law. Therefore, the seised court has to examine its domestic law regarding the domicile of natural persons to establish if the defendant (or plaintiff) has his domicile in that State.
National law may have to be applied twice by the seised court where the defendant concerns a person, such as a minor or an adult placed under guardianship or fiduciary administration (or a married woman) whose domicile depends on that of another person, like his parents, legal guardian or legal administrator (or husband) or on the seat of an authority. For example, the national law of a minor first determines whether his domicile is dependent on that of another person. If it is, the national law of the minor similarly determines where that domicile is situated (e.g. where his parents or guardians are domiciled). If however, the domicile of the dependent person (minor) is under his national law not dependent on that of another person or on the seat of an authority, Article 59 BR I may be applied once, only to determine the domicile of the dependent person (minor) himself.
It follows that, if there is a conflict between the lex fori and the law of another Member State when determining the domicile of a party, the lex fori prevails, in the sense that the seised court has to apply its own internal law to determine where the defendant is domiciled. If from that internal law shows that the defendant is domiciled in the Member State whose court is seised, that court has jurisdiction, irrespective whether the defendant probably has a (second) domicile in another Member State according to the national law of that other Member State. For example, if a defendant, who is sued in a French court in matters relating to his business activities, is domiciled both in France, because he has his principal place of business there, and in Belgium, because his name is entered there in the official population registers, where the laws conflict, the French court must in first instance apply French law to determine whether the defendant is domicled in France, the EU Member State of the seises court. If it is established under that law that the defendant is in fact domiciled in France, the court does not need to take other laws into consideration. It does not have to investigate if the defendant is probably domiciled in Belgium too. It may rule that it has jurisdiction, because the defendant is domiciled (also) in France. This is justified on various grounds. First, to take the example given, a defendant, by establishing his domicile in a given country, subjects himself to the law of that country. Next, only if the lex fori prevails, the court is able to examine whether it has jurisdiction.
This rule will be applied in particular where a defendant is sued in the courts of a Member State in which he is not domiciled. If the jurisdiction of the court is contested, then, following the basic principles on jurisdiction, whether or not the seised court has jurisdiction on account of the Brussels I Regulation will vary according to whether the defendant is domiciled in another Member State or outside the European Union. Thus, for example, where a person domiciled within the European Union has concluded a contract which has to be performed in Belgium, may be sued in Belgium since Article 5, under (1), of the Brussels I Regulation confers jurisdiction on the courts of Member State where the obligation is performed, irrespective of the domicile of the plaintiff or defendant (unless an exception is applicable as in the case of an employment agreement). This rule, however, does not apply to a defendant who is domiciled outside the European Union. When the defendant lives, for instance, in the United States, the fact that the contract was performed in Belgium, does not bring along that the Belgian court may base its jurisdiction on the Brussels I Regulation. The Belgian court has to apply other rules of Belgian private international law to establish if it is allowed to admit the claim against this defendant. Thus if a defendant wishes to contest the jurisdiction of the Belgian court, he must prove that he is domiciled in a State that is not a member of the European Union. The seised Belgian court must determine on the basis of national Belgian law if the defendant is domiciled in Belgium, and if not, if he is probably domiciled in one of the other EU Member States, according to the national laws on this matter of those individual EU Member States. When the seised Belgium court has established that the defendant, according to these national rules of the EU Member States, is not domiciled within the European Union at all, it may not accept that it has jurisdiction pursuant to the Brussels I Regulation, and neither pursuant to Article 5, under (1), because this Article is only applicable if the defendant has his domicile in another EU Member State. Therefore, the Belgian court has to apply Article 4, paragraph 1, BR I, from which follows that, if the defendant is not domiciled in a Member State, the jurisdiction of the courts of each Member State shall, subject to Articles 22 and 23 of the Brussels I Regulation, be determined by the law of that Member State, in this case by the national law of Belgium, including the private international law of Belgium.
In most disputed cases
it will be necessary to determine where the defendant is domiciled. However,
when applying certain provisions of the Brussels I Regulation, in particular
Article
5, under (2) (matters relating to maintenance) and Article
9 (insured person or policyholder), Article
16 (consumers) and Article
23 (prorogation clause), also the domicile of the plaintiff is of
importance. The rules set out in Article
59 will be used in the same way to determine the plaintiff's domicile.
For this reason Article
59 does not specify either the defendant or the plaintiff since is
was felt desirable that the same provisions for determining domicile must
apply to both parties.
Article 60 BR I provides that, for the purposes of the Brussels I Regulation, a company or other legal person or association of natural or legal persons is domiciled at the place where it has its:
Article 60 BR I refers to companies or other legal persons and to associations of natural or legal persons. To speak only of legal persons would have been insufficient, since this expression would not have covered certain types of companies, such as the ' offene Handelsgesellschaft' under German law, which are not legal persons. Similarly, it would not have been sufficient to speak only of companies, since certain bodies, such as associations and foundations, would then not have been covered by the Brussels I Regulation. Article 60 BR I uses three independent criteria to determine whether a legal person is domiciled in a Member State or outside the European Union. It lists as alternatives the statutory seat, the central administration, or the principal place of business of the company or other legal person. The fact that these are listed as alternatives means that if just one of them is situated in a Member State, the legal person may be sued before the courts of that Member State, even if the others are in another EU Member State altogether or in a third State outside the European Union. This definition is open to a degree of forum shopping, even between EU Member States, which is also possible to some extent in relation to the domicile of natural persons. In justification it may be pointed out that if a legal person decides to keep its central administration in a place separate from its principal place of business, it chooses to expose itself to the risk of being sued in both places. The Brussels I Regulation does not define what is meant by the seat of a legal person or of a company or association of natural or legal persons. In determining the location of the seat, the seised court has to apply its own rules of private international law. Therefore, Article 60 refers to domestic rules of private international law with regard to the statutory seat of legal persons of the State of the court hearing the case. This may cause difficulties, because it’s possible that the statutory seat, according to domestic private international law, is not located in any Member State or not in a State where the legal person maintains property or has its head office. For this reason, two alternatives have been added: the place of the legal person’s central management or, as another option, the principal place of its business, so that a legal person may be linked as well to a Member State on the basis of factual elements.
With this, Article 60 of the Brussels I Regulation answers the need for a connecting factor that will ensure that if a legal person is incorporated in a Member State, or does business there, any dispute regarding its activities will fall within the jurisdiction of that EU Member State, so that the plaintiff will not be deprived of a court within the European Union. It also offers the plaintiff the possibility of suing in the courts of the place where the judgment will probably have to be enforced. None of the criteria considered would have answered these needs on its own. The statutory seat does offer a significant degree of certainty, since it is easy to identify, but it is often situated somewhere other than the location of the company’s assets, and does not lend itself to the enforcement of a judgment; it would, moreover, allow a legal person to have its central administration in a Member State, or to carry on its principal business there, while having its statutory seat elsewhere, and thereby escape the jurisdiction of the EU Member States. In turn, the central administration provides a link with a place that is useful for the purpose of enforcing judgment, but it is a factor internal to the legal person, often not immediately identifiable, which makes it harder to determine the court with jurisdiction. And if the central administration is located in a third State outside the European Union, this criterion would not allow the legal person to be sued in a EU Member State, even if it had its statutory seat or its principal place of business there. Finally, the principal place of business is certainly easier to identify and verify, but if taken as the only connecting factor it would not allow jurisdiction to be exercised against a legal person which had its principal place of business outside the European Union, even if that legal person had its statutory seat and central administration inside one of EU Member States and conducted a significant amount of business there. These considerations, taken together, underpin the choice of a broad definition that allows a company or other legal person to be summoned before a court in a EU Member State with which it has a significant connection, in the shape of its central administration, its principal place of business, or its statutory seat. The concept of the ‘statutory seat’, however, is not an appropriate connecting factor for a company or legal person in the United Kingdom or Ireland, where the legal systems refer instead to the place where a company is entered in the register that exists for the purpose, or to the place in which it was incorporated. The registration criterion allows for the fact that the rule concerns not just companies or firms as such but also any body that is not a natural person, so that a registered office is of greater relevance than a ‘seat’ indicated in the founding documents. Article 60, paragraph 2, BR I therefore specifies that for purposes of those two countries the term ‘statutory seat’ means the registered office or, if there is no registration, the place of incorporation, or, if there is no place of incorporation, the place under the laws of which the formation took place. This last reference to the law applied to determine the place of formation that is treated as the statutory seat takes account in particular of the case of a partnership in Scottish law, where the only criterion looked at is the law under which the partnership was formed, regardless of the place of formation. The working out of the concept of the domicile of companies and legal persons in Article 60 BR I was also guided by the desirability of harmonising the general criterion of jurisdiction regarding companies with the connecting factors used in Article 48 of the EC Treaty for the purpose of recognising the right of establishment of companies or firms in the territory of the Community: Article 48 lists the ‘registered office’, the ‘central administration’ and the ‘principal place of business’ within the Community. Even if the need addressed by Article 48 is different – its purpose is to identify the companies or firms entitled to operate in all Member States - it appeared justified to use the same connecting factors to allow companies or firms to be sued in the courts of one of the States bound by the Convention. In other words, if one of the connecting factors referred to in Article 48 is enough to make a company a Community company, enjoying the advantages conferred by that status, it should be treated as a Community company for all purposes, and should therefore be subject to the civil jurisdiction of the Member States in which it operates and is entitled to operate. The concept of domicile under consideration here relates to the forum generale of companies and legal persons, without prejudice to the definition of the domicile of a company for purposes of the forum speciale for particular categories of dispute, such as those which have as their object the validity of the constitution, the nullity or the dissolution of companies or other legal persons having their seat in a State bound by the Convention, or the validity of the decisions of their organs, which are the subject of Article 22, under (2), of the Brussels I Regulation. For disputes relating to insurance contracts, consumer contracts and individual contracts of employment, Articles 9, 15 and 18 of the Brussels I Regulation make specific provision. Nor does the concept explained above affect jurisdiction in disputes arising out of the operations of a branch, agency or other establishment of a company, which are covered by Article 5, under (5), of the Brussels I Regulation (Pocar Report C 319/8) The present text of Article 60 BR I also leaves unchanged the determination of the domicile of a trust, which it refers to the private international law of the court seised. While the application of this provision does not present particular problems in Member States whose legal systems recognise the trust as an institution, difficulties can arise in Member States in which this institution is unknown; in the absence of appropriate conflict rules for determining the domicile of trusts in the legal system of the court seised, the question may be made to depend on the law to which the trust is subject (Schlosser Report, paragraphs 109- 120).
The Brussels I Regulation determines the international jurisdiction of the courts of the Member States of the European Union in legal relationships with an international element. It does not define the concept of 'international element', since this may depend on the particular facts of the proceedings of which the court is seised. In general, the Brussels I Regulation will be relevant only if the parties to the lawsuit, therefore the plaintiff and the defendant, are living in another country. At least the country where the defendant is domiciled, has to be a EU Member State. The country where the plaintiff has his domicle, may be another Member State or a third State. As a result, proceedings instituted in the courts of a Member State (e.g. the Netherlands) which involves only persons domiciled in that same Member State (e.g. the Netherlands) will normally not be affected by the Regulation. Article 2 BR I simply refers matters back to the rules of jurisdiction in force in that State. The legal dispute between such parties is governed only by domestic law (e.g. Dutch law), whereas the national law on civil procedure of that State (e.g. the Code of Civil Procedure of the Netherlands) defines which type of court (e.g. within the Dutch judicial order) is allowed to render a judgement (national subject-matter jurisdiction) and in which territory such type of court has to be addressed (national territorial jurisdiction). It is possible, however, that an international element may be involved in proceedings of this type. This would be the case, for example, where the defendant was a foreign national, a situation in which the principle of equality of treatment laid down in the second paragraph of Article 2 BR I would apply. Although the defendant, who is domciled in a Member State, is not a national of that Meber State, he shall be governed by the rules of jurisdiction applicable to nationals of that State. Since the Brussels I Regulation is applicable in any EU Member State, it ensures that such legal proceedings with an international element (e.g. one of the involved parties is not of the nationality of the Member State whose court is seised) are treated exactly the same as if the person domiciled in that Member State was a national of that State. The Brussels I Regulation is not applied any differently in such cases. Legal proceedings between two persons domiciled in the same Member State may have an international element also where the claim relates to a matter over which the courts of another Member State have exclusive jurisdiction (Article 22 BR I), for instance when two Dutch citizens have a dispute about a property right in a French house, or where identical or related proceedings have been brought to the courts of another Member State (Articles 27 to 29 BR I) (Jenard Report OJ C 59/8). Where two parties domiciled in the Netherlands have a dispute about a right in rem in a house located in France, they cannot bring their case to a Dutch court. If they do so anyhow, the Dutch court must decline jurisdiction, since Article 22, under (1), BR I points out that such case may be handled only by the courts of France, irrespective of the domicile of the plantiff and/or the defendant. Furthermore, the Brussels
I Regulation not only covers international legal relationships that involve
two Member States, but also that involve one Member State and one non-Member
State. The European Court of Justice ruled that the Brussels I Regulation
(in particular Article
2 thereof) is applicable as well in circumstances where the plaintiff
and one of the defendants are domiciled in the same Member State (United
Kingdom), while the case between them, before the courts of that State
(United Kingdom), has certain connecting factors with a non-Member State
(Jamaica), but not with another Member State. The Court indicated that,
for the jurisdiction rules of the Brussels Regulation to apply at all,
the existence of an international element is required. But for the application
of the Regulation the international element doesn't necessarily require
the involvement of at least two Member States, in the sense that the plaintiff
and defendant must be domiciled in two different Member States or, when
they both live in the same Member State, in the sense that the subject-matter
of the proceedings (the case) must be connected with another Member State
(for example that the obligation was performed or the accident occurred
in another Member State than the Member State where plaintiff and defendant
both live). The involvement of a Member State and a non-Member State,
for example because the plaintiff and one of the defendants are domiciled
in the first State (the same Member State, like the United Kingdom) and
the events at issue occurred in the second State (the non-Member State,
like Jamaica), would, for the purpose of the application of the Brussels
I Regulation, also make the legal relationship at issue international
in nature (ECJ
1 March 2005 ‘Owusu v Jackson c.s.’ C-281/02). See also: It is clear that at the recognition and enforcement stage the Brussels I Regulation governs only international legal relationships since ex hypothesi it concerns the recognition and enforcement in one Member State of judgments given in another Member State (Jenard Report OJ C 59/8).
The Brussels I Regulation applies automatically. This principle is formally laid down in Articles 25 and 26 BR I which deal with the matter of examination by the courts of the Member States of their international jurisdiction. The courts must apply the rules of the Regulation whether or not they are pleaded by the parties. It follows from this, for example, that if a person domiciled in Belgium is sued in a French court on the basis of Article 14 of the French Civil Code, and contests the jurisdiction of that court but without pleading the provisions of the Regulation, the court must nevertheless apply Article 3 BR I and declare that it has no jurisdiction (Jenard Report OJ C 59/8). Under Articles 25 and 26 BR I the provisions concerning 'direct jurisdiction' are to be observed by the court of its own motion: in some cases, i.e. where exclusive jurisdiction exists, irrespective of whether the defendant takes any steps; in other cases only where the defendant challenges the jurisdiction. Similarly, a court must also of its own motion consider whether there exists an agreement on jurisdiction which excludes the court's jurisdiction and which is valid in accordance with Article 23 BR I. An obligation to observe the rules of jurisdiction of its own motion is by no means an unusual duty for a court in the original Member States (Schlosser Report OJ C 58/82-83). It does not necessarily follow from Articles 25 and 26 BR I that the courts must, of their own motion, investigate the facts relevant to deciding the question of jurisdiction that they must for example inquire where the defendant is domiciled. The only essential factor is that uncontested assertions by the parties should not bind the court. For this reason the following rule is reconcilable with the Regulation: a court may assume jurisdiction only if it is completely satisfied of all the facts on which such jurisdiction is based; if it is not so satisfied it can and must request the parties to provide the necessary evidence, in default of which the action will be dismissed as inadmissible. In such circumstances the lack of jurisdiction would be declared by the court of its own motion, and not as a result of a challenge by one of the parties. Whether a court is itself obliged to investigate the facts relevant to jurisdiction, or whether it can, or must, place the burden of proof in this respect on the party interested in the jurisdiction of the court concerned, is determined solely by national law. Indeed some of the legal systems of the Member States, for example Germany, do not require the court itself to undertake factual investigations in a case of exclusive jurisdiction even though lack of such jurisdiction has to be considered by the court of its own motion (Schlosser Report OJ C 58/82-83). See also:
The Regulation does not specify what is meant by 'civil and commercial matters', nor does it point to a solution of the problem of classification by determining the law according to which that expression should be interpreted. From the text of the Regulation, however, follows that civil and commercial matters are to be classified as such according to their nature, and irrespective of the character of the court or tribunal which is seised of the proceedings or which has given judgment. This emerges from Article 1 BR I, which provides that the Regulation shall apply in civil and commercial matters 'whatever the nature of the court or tribunal' (Jenard Report C 59/9). All proceedings relating to civil and commercial matters that are not expressly excluded, fall within the scope of the Brussels I Regulation (Schlosser Report OJ C 59/83). That's why, even before particular provisions were dedicated to employment agreements, also labour law and international employment relationships have always been acknowledged as a matter covered by the Brussels Convention 1968 (ECJ 13 November 1979 'Sanicentral v Collin', 25/79). The Regulation also applies irrespective of whether the proceedings are contentious or non-contentious (Jenard Report C 59/9). It is irrelevant as well, whether an action is brought 'against' a named defendant. It is true that in such a case Article 2 et seq. BR I cannot operate; but otherwise the Regulation remains applicable (Schlosser Report OJ C 59/83). The expression "civil or commercial matters” is taken from the fourth session of the Hague Conference on private international law of examining the Convention of 14 November 1896 in order to draw up common rules on a number of aspects of private international law relating to civil procedure. It reported as follows:
According to the European Court of Justice, the concept 'civil and commercial matters', used in Article 1 BR I, must be regarded as an independent concept which must be construed with reference first to the objectives and scheme of the Brussels I Regulation [Brussels Convention] and secondly to the general principles which stem from the corpus of the national legal systems. As the Regulation must be applied in such a way as to ensure, as far as possible, that the rights and obligations which derive from it for the Member States and the persons to whom it applies are equal and uniform, it must be interpreted solely in the light of the division of jurisdiction between the various types of courts existing in certain States; its ambit must therefore be essentially determined by reason of the legal relationships existing between the parties to the action or of the subject-matter of the action (ECJ 16 December 1980 'Netherlands State v Rüffer', 814/79). Article 1 BR I does not include actions brought by the agent responsible for administering public waterways against the person having liability in law in order to recover the costs incurred in the removal of a wreck carried out by or at the instigation of the administering agent in the exercise of its public authority. The term ‘civil and commercial matters’ does not cover disputes resulting from the exercise of public powers by one of the parties to the case, as it exercises powers falling outside the scope of the ordinary legal rules applicable to relationships between private individuals and there is all the more reason for such an assessment as regards a legal action for compensation deriving from operations conducted by armed forces, as such operations are one of the characteristic emanations of State sovereignty, in particular inasmuch as they are decided upon in a unilateral and binding manner by the competent public authorities and appear as inextricably linked to States’ foreign and defence policy. The question as to whether or not the acts carried out in the exercise of public powers that constitute the basis for such proceedings are lawful, concerns the nature of those acts, but not the field within which they fall. Since that field as such must be regarded as not falling within the scope of the Regulation, the unlawfulness of such acts cannot justify a different interpretation. Therefore Article 1 of the Regulation does not cover a legal action brought by natural persons in a Member State against another Member State for compensation in respect of the loss or damage suffered by the successors of the victims of acts perpetrated by armed forces in the course of warfare in the territory of the first State (ECJ 15 February 2007 ‘Lechouritou v Dimosio’, C-292/05).
It is to be noticed that even civil and commercial matters ruled by administrative or criminal courts fall within the scope of the Regulation. The Regulation applies in all not expressly excluded civil and commercial matters 'whatever the nature of the court or tribunal'. The distinction between civil and commercial matters on the one hand and matters of public law on the other is well recognized in the legal systems of the Member States and is, in spite of some important differences, on the whole arrived at on the basis of similar criteria. Thus the term 'civil law' also includes certain important special subjects which are not public law, especially, for example, parts of labour law. However, the borderline between the public law and the private law activities of public agencies is not rigidly prescribed in some of the legal systems. Public authorities have, within certain limits, a right to choose whether in carrying out their functions they wish to use the method of a 'sovereign act', i. e. an administrative contract, or merely to conclude a private transaction. In respect of those areas, where public authorities may act either under private or public law, it is not always easy to decide whether or not they have acted as private individuals. In practice a clear indication is often lacking. Relations between public authorities may also be governed either by private or by public law. If governed by public law, such relations are not subject to the Brussels I Regulation, even if, as in Italy, they are not considered a part of administrative law. However, relations of States and public corporations with each other would fall almost without exception within the sphere of private law, if they contain international aspects (and are not subject to public international law). It is hard to imagine how, for example, it would be possible for relations under public law to exist between two local authorities in different (Member) States. However, such relations could, of course, be established in future by Treaties (Schlosser Report OJ C 59/84-85). In the same way the Regulation covers civil proceedings brought before criminal courts, both as regards decisions relating to jurisdiction, and also as regards the recognition and enforcement of judgments given by criminal courts in such proceedings. Of course, as regards both, jurisdiction and recognition and enforcement, the Regulation affects only civil proceedings of which those courts are seised, and judgments on civil matters given in such proceedings. The Regulation in no way alters the penal jurisdiction of criminal courts and tribunals as laid down in the various codes of criminal procedure. A claim for compensation for loss to an individual resulting from a criminal offence, even though made in the context of criminal proceedings, is civil in nature unless the person against whom it is made is to be regarded as a public authority which acted in the exercise of its powers. That is not the case where the activity called in question is the supervision by a state-school teacher of his pupils during a school trip (ECJ 21 April 1993 ‘Sonntag v Waidmann c.s.’, Case C-172/91). Criminal proceedings and criminal judgments of all kinds are excluded from the scope of the Brussels I Regulation. This applies not only to criminal proceedings stricto sensu. Other proceedings imposing sanctions for breaches of orders or prohibitions intended to safeguard the public interest also fall outside the scope of civil law. Certain difficulties may arise in some cases in classifying private penalties known to some legal systems like contractual penalty clauses, penalties imposed by associations, etc. Since in many legal systems criminal proceedings may be brought by a private plaintiff, a distinction cannot be made by reference to the party which instituted the proceedings. The decisive factor is whether the penalty is for the benefit of the private plaintiff or some other private individual. Thus the decisions of the Danish industrial courts imposing fines which are for the benefit of the plaintiff or some other aggrieved party, certainly fall within the scope of the 1968 Convention [and therefore also of the Brussels I Regulation] (Schlosser Report OJ C 59/85).
See also:
Article 1, paragraph 2, BR I excludes various civil and commercial matters from the scope of the Brussels I Regulation. In the first paragraph Article 1 expressly specifies that the Brussels I Regulation does not extend itself to revenue, customs or administrative matters. So these issues can never be regarded as 'civil or commercial matters' itself within the meaning of the Brussels I Regulation. The term 'customs' does, however, not cover a claim by which a Member State seeks to enforce a guarantee contract (surety agreement) intended to guarantee the payment of a customs debt, where the legal relationship between the State and the guarantor, under that contract, does not entail the exercise by the State of powers going beyond those existing under the rules applicable to relations between private individuals, even if the guarantor may raise pleas in defence which necessitate an investigation into the existence and content of the customs debt (ECJ 15 May 2003 TIARD v Staat der Nederlanden, C-266-01; in a similar way: ECJ 5 february 2004 'Frahuil v Assitalia', C-265/02).
The Brussels I Regulation doesn't apply to all civil and commercial matters. Some matters are explicitly excluded. Most of these exclusions represent a genuine limitation of the civil and commercial matters covered, with their exclusion being necessitated for different reasons in every instance. This is the case as regards the relationships listed in point a (status or legal capacity of natural persons, rights and property arising out of a matrimonial relationship and succession), point b (bankruptcy, proceedings relating to the winding up of insolvent companies or other legal persons, judicial arrangements, compositions and analogous proceedings), and point d (arbitration). The exclusion contained in point c (social security) is justified both by the fact that social security comes under public law in some countries whilst it falls in the borderline area between private law and public law in others, and because social security matters are increasingly governed by secondary Community legislation (Evrigenis Report 1986 C 298/9-10). Article 1 BR I entails a formula which excludes certain matters from a principal rule to one which would have involved giving a positive definition of the scope of the Regulation. The solution adopted implies that all litigation and all judgments relating to contractual or non-contractual obligations which do not involve the status or legal capacity of natural persons, wills or succession rights in property arising out of matrimonial relationship, bankruptcy or social security must fall within the scope of the Regulation, and that in this respect the Regulation should be interpreted as widely as possible. However, matters falling outside the scope of the Regulation do so only if they constitute the principal subject-matter of the proceedings. They are thus not excluded when they come before the court as a subsidiary matter either in the main proceedings or in preliminary proceedings (Jenard Report C 59/10). See also:
Point a of the second paragraph of Article 1 BR I refers to the status or legal capacity of natural persons, rights and property arising out of a matrimonial relationship and succession. The exclusion of these matters from the scope of the Regulation was necessitated by their specific characteristics, which are reflected in the great variety of ways they are dealt with at national level in both substantive law and private international law. Their inclusion in the Regulation would have meant either that these specific characteristics would have had to be levelled out or, alternatively, that such matters would have been dealt with in a rather inconsistent manner from the point of view of international jurisdiction, although consistency is one of the main aims of the Regulation. Faced with this dilemma, the drafters of the predecessor of the Regulation, the 1968 Brussels Convention, preferred to exclude these relationships from its scope (Jenard Report C 59/10). Interpreting these exclusions, the Court of Justice of the European Communities has ruled that the enforcement of a judicial decision on the placing under seal or the freezing of the assets of the spouse as a provisional measure in the course of proceedings for divorce does not fall within the scope of the Regulation (ECJ 6 March 1980 'De Cavel v De Cavel', 120/79). The Court took the same view in the case of an application on the part of the wife for the Court to order the husband, as a provisional protective measure, to deliver up a document in order to prevent its use as evidence in a dispute concerning a husband's management of his wife's property, because the management was closely connected with the proprietary relationship resulting directly from the marriage bond (ECJ 31 March 1982 'C.H.W. v G.J.H', 25/81). Matters relating to maintenance, however, do fall within the scope of the Regulation, as is apparent from Article 5, point 2, BR I, which governs jurisdiction with regard to maintenance obligations. As was perhaps to be expected, problems have arisen from the common practice of linking maintenance claims with proceedings relating to the status of persons, and, in particular, with divorce proceedings. The European Court of Justice has ruled that the Regulation is applicable to an interim maintenance award under a divorce judgment (ECJ 6 March 1980 'De Cavel v De Cavel', 120/79). This point is expressly dealt with in the 1978 amendment to Article 5, point 2, of the 1968 Brussels Convention (Evrigenis Report 1986 C 298/10). Matters will relate to maintenance if the granted provision in divorce proceedings shows that it is designed to enable one spouse to provide for himself or if the needs and resources of each of the spouses are taken into consideration in the determination of its amount. Such a judicial decision will therefore fall within the scope of the Brussels I Regulation. If, on the other hand, the granted provision solely concerns the division of property between the spouses, it relates to matters regarding rights in property arising out of a matrimonial relationship and will, therefore, not be governed by or enforceable under the Brussels I Regulation. Where a judicial decision involves both matters, it may, in accordance with Article 48 of the Brussels I Regulation, be enforced in part if it clearly shows the aims to which the different parts of the judicial provision correspond. It follows that a decision rendered in divorce proceedings ordering payment of a lump sum and transfer of ownership in certain property by one party to his or her former spouse must be regarded as relating to maintenance and therefore as falling within the scope of the Regulation if its purpose is to ensure the former spouse's maintenance. The fact that in its decision the court of origin disregarded a marriage contract is of no account in this regard (ECJ 27 February 1997 'Van den Boogaard v Laumen', C-220/95).
See also:
Point b of the second paragraph of Article 1 excludes from the scope of the Regulation bankruptcies, proceedings relating to the winding up of insolvent companies or other legal persons, judicial arrangements, composition and analogous proceedings. These matters had to be excluded given that the Member States of the Community intended to draft, and meanwhile have drafted, a separate Community bankruptcy instrument, Council regulation (EC) No 1346/2000 of 29 May 2000 on insolvency proceedings. In relation to Article 22, point 2, BR I, which stipulates that, in proceedings which have as their object the dissolution of companies or other legal persons or associations of natural or legal persons, the courts of the Member State in which the company, legal person or association has its seat have exclusive jurisdiction, this exclusion may give rise to problems where the dissolution is a consequence of bankruptcy, winding up, judicial arrangement, composition, or analogous proceedings (Evrigenis Report 1986 C 298/10). The European Court of Justice has indicated which matters fall within the scope of the terms bankruptcy, proceedings relating to the winding-up and analogous proceedings (ECJ 22 February 1979 'Gourdain v Nadler', 133/78). See also:
Social security, which is excluded from the Regulation by point c of the second paragraph of Article 1 BR I, is regarded in some national legal systems as a matter of public law and in others as a mixed legal category on the borderline between public law and private law. Although it could perhaps be argued that this feature alone would be enough to exclude social security from the scope of the Regulation as defined in the first paragraph of Article 1, its express exclusion was nevertheless thought to be desirable. There were, however, other reasons as well for excluding social security from the scope of the Regulation, such as the fact that it is governed by the Treaties and by secondary Community legislation, and the fact that there are numerous bilateral social security agreements between the Community Member States. The drafters of the Brussels Convention and Regulation considered that this legal situation should not be disturbed by extending the Regulation to regulate social security. It should, however, be noted that this exclusion concerns relationships directly connected with the insurance aspect and, particularly, relationships between the insuring body and the insured party, his successors in title and the employer. Ancillary matters, such as direct claims of the injured party against the insuring body or subrogation of the insuring body to the claims of an injured party as against a third party responsible for the injury or damage, are in principle covered by ordinary legal rules and come within the scope of the Regulation (Evrigenis Report 1986 C 298/10-11) (see also ECJ 14 November 2002 ‘Gemeente Steenbergen v Baten’, C-271/00). See also:
Arbitration, a form of proceedings encountered in civil and, in particular, commercial matters, (Article 1, second paragraph, point d BR I) is excluded because of the existence of numerous multilateral international agreements in this area. Proceedings which are directly concerned with arbitration as the principal issue, e.g. cases where the court is instrumental in setting up the arbitration body, judicial annulment or recognition of the validity or the defectiveness of an arbitration award, are not covered by the Regulation. However, the verification, as an incidental question, of the validity of an arbitration agreement which is cited by a litigant in order to contest the jurisdiction of the court before which he is being sued pursuant to the Regulation, must be considered as falling within its scope (Evrigenis Report 1986 C 298/10). If, because of the subject-matter of the dispute, that is, the nature of the rights to be protected in proceedings, such as a claim for damages, those proceedings come within the scope of the Brussels I Regulation, a preliminary issue concerning the applicability of an arbitration agreement, including in particular its validity, also comes within its scope of application. It follows that the objection of lack of jurisdiction raised on the basis of the existence of an arbitration agreement, including the question of the validity of that agreement, comes within the scope of the Brussels I Regulation and that it is therefore exclusively for the court to rule on that objection and on its own jurisdiction, pursuant to Article 1, paragraph 2, point (d) and Article 5, point (3), of that Regulation (ECJ 10 February 2009 ‘Allianz and Generali v West Tankers’ C-185/07). Yet, the exclusion provided for in Article 1, paragraph 2, point (d), BR I extends to litigation pending before a national court concerning the appointment of an arbitrator, even if the existence or validity of an arbitration agreement is a preliminary issue in that litigation (ECJ 25 July 1991 ‘Marc Rich v Società Italiana Impianti’ C-190/89). See also:
Denmark is, at least officially, not involved in the Regulation, because it explicitly has stipulated that it will not be bound by European Regulations of this nature. Initially this had the result that the 1968 Brussels Convention still applied in legal relationships on civil and commercial matters between residents of EU Member States and Denmark. Afterwards, the European Union and Denmark have concluded a separate Agreement on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters. The purpose of this Agreement is to apply the Brussels I Regulation in relations between Denmark and the rest of the European Union and to achieve a uniform application and interpretation. Consequently, the European Court of Justice is given jurisdiction to interpret the Agreement at the request of a Danish court. So, although the Brussels I Regulation is not directly applicable to Denmark, it has effectively been extended to Denmark by a separate Agreement between the European Union and Denmark. On 27 April 2006, the Agreement was approved on behalf of the EU by Council Decision 2006/325/EC. It entered into force on 1 July 2007. See also:
|