International Environmental Litigation in EU Courts: A Regulatory Perspective

This article argues that the rules of European private international law, which frame international litigation in the courts of the Member States of the EU, fail in their pursuit of the cosmopolitan goals of EU environmental policy. The practical application of these rules is limited to the following two types of case: 1) the case of victims suing an operator whose actions in one country directly cause environmental damage elsewhere, and 2) the case of victims suing a European-based multinational corporation operating in an extraction or chemical industry whose overseas subsidiary, typically in a developing country, causes environmental damage. By arguably not accommodating claims by public authorities against foreign operators, including from other Member States, which are crucial in cases of pure environmental damage, and the cases of the second type in industries other than extraction and chemical, European private international law fails to achieve fully its regulatory potential. Furthermore, the rules of European private international law have the effect of raising the level of environmental protection solely within the EU and at its borders in the first type of case and shielding European multinational corporations from liability for the environmentally detrimental and degrading effects of their overseas operations in the second type of case. These rules are therefore an inadequate tool of global governance. Avenues for improving the law are mentioned.


Introduction
In the early 1990s the Movement for the Survival of the Ogoni People protested in Nigeria against the environmental damage allegedly caused by Royal Dutch Shell's oil operations in Ogoniland, in the Niger Delta. Several protesters were arrested by the Nigerian armed forces, charged with questionable crimes and executed. Amongst them was Dr Barinem Kiobel, whose widow, after obtaining asylum in the United States, together with a number of other former Ogoniland residents, brought a class action suit under the 1789 Alien Tort Claims Act ('ATCA'). 1 The Act provides that '[t]he district courts shall have original jurisdiction of any civil action by an alien for a tort only, committed in violation of the law of nations or a treaty of the United States'. The claimants argued that the Anglo-Dutch corporation and its Nigerian subsidiary were complicit in the Nigerian government's violations of the law of nations, including extrajudicial killing, torture, rape, arbitrary arrest and detention, theft and destruction of property.
In April 2013, after a protracted legal battle, the US Supreme Court decided unanimously in favour of the defendants. 2 The majority noted a strong presumption against the extraterritorial application of US law, and held that the Act did not cover overseas conduct of multinational corporations. According to Roberts CJ, who delivered the opinion of the Court, the presumption will be displaced 'where the claims touch and concern the territory of the United States…with sufficient force… [I]t would reach too far to say that mere corporate presence suffices.' 3 Alito J went further in his concurring opinion, in which environmental damage should as a priority be rectified at source and that the polluter should pay'. 20 Since many environmental problems are transnational, even global, in nature, it is logical that the relevant rules of European private international law, which may assist in their solution, should take a cosmopolitan approach. Indeed, many influential private international law scholars are of the opinion that Article 7 of Rome II pursues, and in an adequate manner, the cosmopolitan goals of EU environmental policy. 21 This rule has also been endorsed by the Hague Conference on Private International Law and the International Law Association. 22 Contrary to the majority opinion, this article argues that the relevant rules of European private international law fail to take a cosmopolitan approach. They are therefore an inadequate tool of global governance. There is a strong argument that the Brussels I Recast and Rome II exclude from their scope claims by public authorities against foreign operators and the resulting judgments, thus potentially hindering an important type of international 20  environmental litigation, which is crucial in cases of pure environmental damage, from taking place in Europe. The two Regulations address relatively adequately the type of case where the claim is brought by victims against an operator whose actions in one country directly cause environmental damage elsewhere. But the cases of this type that are brought in EU courts will almost always concern actions committed in the EU and/or environmental damage suffered in the EU. The rules of the Brussels I Recast and Rome II concerning this type of case therefore help to raise the level of environmental protection within the EU and at its borders, but not elsewhere. Crucially, it is argued that European private international law fails to deal with the globally more important and frequent cases of the same type as Kiobel.
It effectively shields European multinational corporations from liability for the environmentally detrimental and degrading effects of their overseas operations. For these reasons, the relevant rules of European private international law overall fail in their pursuit of the cosmopolitan goals of EU environmental policy. This conclusion raises the question how European private international law could more fully achieve its regulatory potential.
This article is divided into seven sections. Following this introduction, the second section outlines the regulatory potential of private international law with regard to the protection of the environment. The third section divides the relevant CJEU and English and other domestic cases into different types. It demonstrates that whatever regulatory potential European private international law may have, its practical operation is limited to only two types of case, namely the case of victims suing an operator whose actions in one country directly cause environmental damage elsewhere and the cases of the same type as Kiobel, but the latter only with regard to extraction and chemical industries. The reasons for the relatively limited practical operation of the relevant rules of European private international law are provided. The subsequent three sections explore how the rules of jurisdiction of the Brussels I Recast and the special choice-of-law rules for environmental damage of Rome II apply to the two types of case identified in the third section. Shortcomings in the current rules are identified. The seventh section concludes and examines the avenues for improvement.

Private international law and the regulation of the environment
It is nowadays widely understood that private law can be used as a regulatory tool that supplements or even replaces the 'command and control' and other types of regulation.
Private international law, a field of law of particular relevance for international environmental litigation, is no exception. 23 This section presents the regulatory challenges posed by transnational environmental cases and various regulatory responses to them. 24 The aim is to disclose the regulatory potential of private international law in this context.
International environmental litigation concerns cases of environmental damage that are connected with more than one country. In some cases, an operator's actions in country A directly cause environmental damage in country B. The main challenge in this type of case is that, since environmental standards 25 of countries differ, the operator will be subject to the exclusive application of the laxer standards of country A, even if the negative externalities of its activities are suffered in country B. In other cases, as illustrated by Kiobel, a parent company from country A has a subsidiary in country B that causes environmental damage in 23 For the regulatory function of private international law see R. Michaels, 'New European Choice-of-Law Revolution ' (2008) 25 The term 'environmental standards' is used here in a broad sense and includes both 'public law' standards such as 'command and control' type of regulation and 'private law' standards of tort law. the latter country. In this type of case, regulatory challenges are created by the fact that a multinational corporation, composed of the parent and its subsidiaries, operates its many parts in more than one country 'with the coherence of intent and implementation that resembles a single entity'. 26 Of course, the constituent parts of a multinational corporation can be connected into a single economic entity not just through bonds of ownership but also through contractual relations or even informal alliances. 27 The constituent parts, however, remain separate legal persons, falling under distinct and independent regulatory systems of the countries in which they are established and operate. Consequently, each legal person forming part of a multinational corporation is accountable, in principle, only to its local authorities and liable only to its own creditors. The main challenge in this type of case is that, through smart corporate organisation, the multinational corporation, as a single economic entity, may reap the profits generated by its constituent parts and simultaneously shift the risk of negative externalities of their activities on third parties by taking advantage of the constituent parts' separate legal personality, their limited liability, the territorial jurisdiction of local authorities, and regulatory failures in host countries. These failures consist in the inability or unwillingness of local authorities to regulate and oversee the multinational corporation's local activities because of the host country's socio-economic underdevelopment, low administrative capacity and technical expertise, information asymmetry, fear of driving away foreign investors, corruption, collusion with the corporation and the like. 28 These failures are often caused or exacerbated by a great disparity in economic power between the multinational corporation and its home country, on the one hand, and the host country, on the 26 M. Anderson other. 29 Under-regulation may therefore occur in both types of case, leading to a globally suboptimal allocation of resources.
In setting, monitoring and enforcing environmental standards, countries are typically guided by the interests and values of local individuals and communities, without considering and accommodating the interests and values of outsiders. This results in two additional regulatory challenges. 30 First, countries may turn the problem of under-regulation into an opportunity, and compete to attract foreign direct investment and try to increase the competitiveness of domestic businesses by lowering their environmental standards, especially when the negative externalities of local activities are suffered across the border. This 'regulatory competition' may, in turn, lead to a 'race to the bottom'. An extreme example of this is Papua New Guinea where lax environmental standards were coupled with legislation making it a criminal offence to sue multinational mining corporations operating and causing damage in that country. 31 Yet another challenge is that countries typically regulate activities taking place in, or affecting, their territory without regard to the regulation in other countries.
Consequently, operators may be subject to environmental standards of more than country, which may result in over-regulation. Both scenarios lead to a globally inefficient resources allocation.
These regulatory challenges would not exist if there were a global standard-setting, monitoring and enforcing authority or an effective global system of cooperation of national authorities with regard to the protection of the environment. But there is no such authority or 29 ibid. 30  and civil society pressure, and corporate self-regulation. But since these regulatory responses typically do not result in the imposition of legal obligations on multinational corporations and other operators, their regulatory potential is limited.
Another response to the regulatory challenges posed by transnational environmental cases lies in international environmental litigation. In purely domestic settings, tort law is used as a regulatory tool for the protection of the environment, usually as a supplement to 'command and control' and other types of regulation. 41 It pursues the objective of compensating victims of environmental damage for the harm they suffer as a consequence of the tortfeasor's actions, possibly also making funds available for environmental remediation, in particular if environmental NGOs are given standing to commence environmental litigation. The compensatory function of tort law is usually supported by liability insurance or environmental damage insurance. Furthermore, by forcing actual or potential tortfeasors to internalise the negative externalities of their activities, tort law also performs a specific and general deterring function. The tortfeasor is in a better position to bear the risk of its activities than those who are affected by them. For instance, it can improve its environmental performance by investing in its technologies and practices and take out insurance. On the other hand, it is highly unlikely that many victims of the tortfeasor's activities will have known of, insured themselves against or in other way managed the risk of those activities.
By shifting the risk to the best cost-avoider, tort law contributes to the achievement of the optimal allocation of resources. Where environmental damage is caused by a member of a corporate group, the fulfilment of tort law's regulatory function also depends on whether, and 41  to what extent, company law ascribes liability to the parent company for the environmental damage caused by its subsidiary. In transnational cases, the applicable tort law and company law depend on the applicable choice-of-law rules, which, in turn, depend on where the litigation takes place. By allowing the victim to commence proceedings in a certain jurisdiction and by leading to the application of a certain law, private international law also performs compensatory and deterring functions. International environmental litigation has other characteristics which support its regulatory function: it generates pressure on private actors and governments; it constitutes, sustains and energises transnational networks of civil society; it provides information and policy opinions to interested parties, transnational networks and broader publics; it encourages settlement. 42 The fact that international environmental litigation relies on private parties to bring claims with a regulatory impact contributes to the effectiveness of this regulatory response. In some countries, conditional or contingent fees are available. There, the victim may be able to obtain the services of top litigation lawyers who will also have an interest in the outcome of the litigation.
International environmental litigation, however, comes with a number of disadvantages. It can be slow, expensive, and operates on an ex post and case-by-case, and therefore selective and unsystematic, basis. There may be factual and legal problems arising out of the long latency of harm, the indeterminate nature of claimants and defendants, and the intersection of public law and civil liability. 43 In criticising the use of tort law as a regulatory tool, Cane notes that courts are neither expert in relevant areas of regulated activity nor politically responsive, that they rely more or less exclusively on litigants for information about regulated activities, and that court procedures in tort cases largely exclude participation The rules of private international law concerning the environment in general, and particularly those of EU law, should take a cosmopolitan approach for the following reasons.
Since environmental problems are transnational, even global, in nature, they require cosmopolitan solutions. 50 The fact that truly global solutions are precluded because of genuine disagreement among countries and the problem of free-riding shows that the objective of environmental protection can more adequately be pursued at a sub-global level, for example through rules of private international law in conjunction with the applicable tort and company laws. The EU is particularly well suited for providing cosmopolitan private international law rules. 51  individual Member States and, furthermore, prides itself with the highest standards in many policy areas. It is therefore not surprising that EU environmental policy expressly pursues cosmopolitan goals. 52 Europe is a home to many multinational corporations and other operators engaged in environmentally detrimental and degrading activities. The EU therefore has the power to affect the way such activities are conducted and thereby neutralise the effects of the measures adopted by host countries that lead to a 'race to the bottom', thus raising the global level of environmental protection. As a beneficiary of such activities, the EU should be responsible for their regulation, especially where there are regulatory failures in host countries. 53 This is indeed demanded by many EU citizens, who act as consumers and small investors. Requiring multinational corporations established in the EU to respect high environmental standards also makes long-term economic sense. 54 For example, in order to reduce their compliance costs, those corporations will have to invest in research and development. Since there is a global trend of raising environmental standards, such investment may pay off in the long term. Those corporations may be able to export advanced environmental technologies and practices, on the one hand, and enjoy greater productivity levels in comparison with their foreign competitors, on the other. 55 The argument that home state regulation represents an imperialist infringement of the sovereignty of disempowered responsibility, and the formation of political and cultural identity' which reinforces this field of law 'as a site of contestation and deliberation over questions of authority, responsibility and identity'). 52 55 Gunningham and Grabosky mention in Smart Regulation, above, n 24, at p 43, Germany as an example of a country where high environmental standards have been credited with not only improving the productivity of existing firms through technological and managerial improvements, but also the creation of entire new pollution control industries. These firms are in a competitive position to export their products and services as others catch up.
host states is not persuasive where the interests at stake, i.e. the protection of the environment, are truly global. 56 If its rules do take a cosmopolitan approach, private international law will have gone a long way towards adopting a 'planetary perspective' 57 and fulfilling its 'social mission '. 58 The examination of whether, and to what extent, European private international law achieves its regulatory potential starts with the exploration of the types of transnational environmental cases it is able to accommodate.

Types of transnational environmental cases and international environmental litigation in EU courts
International environmental litigation is not a new phenomenon in Europe. Two relevant cases have thus far been decided by the CJEU. The English courts have been particularly busy with this kind of litigation. Very important cases have recently been brought in the Dutch and Swedish courts. By focusing on the two CJEU cases and the experience of the English, Dutch and Swedish courts in the past two decades, this section aims to identify the types of transnational environmental case that are litigated in Europe, and to explain why some types of case do not appear in EU courts.  processing plants situated in South Africa, which were owned by local subsidiaries of an English-based corporation. The claimants commenced proceedings in England against the parent company arguing that they had contracted asbestiosis as a result of the defendant's negligence to protect them from the exposure to asbestos. In addition, claims were brought by four Italian claimants for the harm suffered while working at or living close to a factory These two cases are unusual in that no overseas subsidiaries were involved in the poisoning of the claimants, although the defendants in both cases argued that the waste had been dumped by local independent contractors, which had been appointed in good faith.
Interestingly, claimants in all the English cases, as well as in Chandler v Cape, 68 were represented by the same firm of London lawyers, Leigh Day. 69 Cases of international environmental litigation can be divided into several types by using the following criteria: the (de)localised nature of the actual or potential tort; the defendant's corporate organisation; the defendant's industry; the nature of the claimant (see table 1). Sometimes, as in Bier and Land Oberösterreich v ČEZ, the tort is delocalised, in the sense that the defendant's actions and/or damage are not confined to one country. But, as the English, Dutch and Swedish experience shows, cases where the elements of the tort are localised, i.e. confined to one country only, are more common. Depending on the defendant's corporate organisation, cases of international environmental litigation also fall into two types. In some cases, it is the defendant's acts or omissions that directly cause the damage, as in Bier and Land Oberösterreich v ČEZ. In others, a foreign subsidiary of the defendant is also involved. Since the foreign subsidiary often lacks funds or even ceases to exist before the claim is brought, claimants typically look for ways to ascribe liability to the parent company for the actions of the subsidiary. An option is to rely on the 'piercing of the corporate veil', 'enterprise liability', agency and related doctrines. Another option, regularly pursued in the English and recently in the Dutch and Swedish courts, is to argue that the defendant has breached a duty of care that it owes directly to the victims. Such cases are frequently referred to as 'foreign direct liability' cases. 70  industries. The minority of cases concern other industries such as the production of chemicals (Thor Chemicals), the production and distribution of nuclear energy (Land  organisation. The following sections analyse how, within the scope of their practical operation, the rules of European private international law apply to these two types of case in order to examine whether, and to what extent, this field of law is achieving its regulatory potential. The analysis starts with the rules of jurisdiction.

Adjudicatory jurisdiction and environmental damage
The  The Lubbe v Cape type of case may concern a claim brought in one Member State against a parent company domiciled in another. The court seised with the claim will not have general jurisdiction over such parent. Article 7(2) of the Brussels I Recast will be inapplicable, since the harmful event will have occurred in a third country. The only other rule of special jurisdiction of the Recast that the claimants could rely on is Article 7(3) which gives jurisdiction to a court seised with criminal proceedings over a civil claim for damages or restitution which is based on the criminal act giving rise to such proceedings. courts are loath to split up proceedings involving co-defendants, 134 it seems that claimants will easily satisfy these requirements. 135 In conclusion, the rules of jurisdiction of the Brussels I Recast deal satisfactorily with the type of international environmental litigation in which the claim is brought against an EU domiciliary acting in one Member State and directly causing environmental damage in another. In such cases, victims may access several forums, most importantly the courts for the place of the direct damage suffered by direct victims. This place usually coincides with the habitual residence of direct victims and the place where the damaged property is located.
If a claim for a delocalised tort is brought against a non-EU domiciliary, the jurisdiction of the Member State court seised with the claim depends on the traditional law of that Member State. The English traditional jurisdictional rules applicable in this type of case, for example, seem to be somewhat wider than the analogous jurisdictional rules of the Brussels I Recast.
These rules therefore also put victims in this type of case in a good litigational position. that the attitude of the CJEU, which is the ultimate interpreter of the Recast, will be the same. 138 The Lubbe v Cape type of case also sometimes concerns a claim against a subsidiary domiciled either in or outside the EU. Both the Brussels I Recast and the English traditional rules of jurisdiction provide for this scenario. In theory, the Lubbe v Cape type of case could also concern a claim against a non-EU parent. Such cases, however, do not arise in practice, possibly because it would be hard to obtain jurisdiction of the Member State courts in such cases.
The following text continues the examination of the regulation of the environment in The scope of Article 7 is wide, although there is some uncertainty in this respect. It clearly covers personal injury, property damage, economic loss and pure environmental damage. It is also clear that nuclear damage is excluded. 140 According to some scholars, Article 7 is wide enough to cover cases brought under the English law of negligence, trespass, nuisance, Rylands v Fletcher 141 and breach of statutory duty. 142 Others, however, interpret the scope of Article 7 more narrowly. 143 The Article 26 allows the court to disregard a provision of the foreign applicable law whose application is manifestly incompatible with the public policy of the forum.  The connecting factor of the place of the event giving rise to the damage avoids the potential application of different laws to claims of victims of the same tortious action. It is also often justified with the tortfeasor's interests in mindthe tortfeasor can be expected to know and comply with the law of the country of the event but not with the potentially multiple laws of the countries of the damage. 158  that it gives countries that already host polluting industries an incentive to decrease their environmental standards where the environmentally detrimental and degrading effects of those industries are exclusively or primarily felt across the border. 159 It also increases the incentives of operators moving to the country with the laxest standards and could lead to a 'race to the bottom'. 160 The lex loci actus can be a law unconnected and unfamiliar to the victim, whose application they do not expect. 161 In international environmental litigation there is no justification for favouring the interests of polluters over those of victims.

Some of the ideas that motivated the CJEU decisions in
The principle of the closest connection is also not an optimal solution. The most closely connected country will ordinarily be either the country of the damage or the country of the event giving rise to the damage. The downsides of these two connecting factors are therefore attributable to the principle of the closest connection.
With regard to party autonomy, there is a concern that the operator, typically a big corporation, might abuse its typically superior bargaining power and impose upon the potential victims of environmental damage the application of the law favourable for it. That In practice, the cost of ex ante choice-of-law agreements is prohibitive. Potential victims of an action might be numerous and difficult to determine in advance. The operator might therefore have to negotiate with a large number of potential victims. Such individual negotiations would seldom lead to identical outcomes. Some potential victims might refuse to negotiate. All of this leads to prohibitive cost of negotiation: Nicita and Winkler, above, n 30, 697. 163 Betlem and Bernasconi, above, n 21, 146; Nicita and Winkler ibid 698. According to von Hein, above, n 21, 1699: 'There may be legitimate procedural reasons to opt for the lex fori, even if it is less stringent than the law of another country, in order to have recourse to the highest civil court in countries such as Germany, where the concerns of the abuse of the operator's typically superior bargaining power, since the victim, after the risk materialises, is in a position to assess the pros and cons of the application of different laws. They will not easily give up their rights under the law(s) applicable by default, and will agree to the application of another law only if they consider that to be in their interest. A further concern is that party autonomy should not be used to undermine the rights and obligations of third parties. Oberösterreich v ČEZ 168 demonstrate, the environmental damage and the tortious action occur in different countries. Indeed, Article 7 produces beneficial effects in this type of case.
But both Bier and Land Oberösterreich v ČEZ, and presumably the vast majority of cases of this type that end up in EU courts concern intra-EU delocalised environmental torts.
Presumably, the minority of cases concern delocalised environmental torts where one of the elements of the tort occurs within and the other outside the EU. It is unlikely that the Member State courts will be seised with a claim concerning the Bier type of case where both elements of the tort occur entirely outside the EU. The exception is where a delocalised environmental tort is committed by an overseas subsidiary of a European parent company.
But this is the Lubbe v Cape type of case, to which this article now turns. In conclusion, the effect of Article 7 in the Bier type of case is to raise the level of environmental protection 166    to compensate them. Although it may be argued that limited liability is needed to ensure that the group will take on the enterprise risk, such argument is inapplicable where creating a subsidiary is done purely for business organisation purposes. 182 Even where a subsidiary is established as a vehicle for new and risky investments, the shifting of the risk of liability onto involuntary creditors is hard to justify in the light of the fact that they are typically uninsured and in a much worse position than the group to obtain insurance. 183 Furthermore, the risk of moral hazard speaks against the parent's limited liability in such situations. 184 The courts have never inquired whether this radical development required a reconsideration of the doctrine of limited liability. Although there have been calls for shifting the focus from the liability of individual members of a group to enterprise liability 185 such calls remain largely theoretical.
Limited liability, of course, is not absolute. In English company law, for example, exceptions include the piercing of the corporate veil and the common-law concepts of agency law. But these exceptions fail to provide an adequate solution in most cases. According to Blumberg, '[t]his is one of the most unsatisfactory areas of the law', '[w]ith hundreds of irreconcilable decisions and shifting rationales', that functions 'in an almost inscrutable manner behind conclusory metaphors such as "mere instrumentality", "sham", "adjunct", "agent, "alter ego", "puppet", or dozens of similarly murky terms'. 186 It is therefore not surprising that claimants in many cases of the Lubbe v Cape type attempt to avoid the piercing of the corporate veil and agency doctrines, and instead proceed on the basis of the parent company's direct tortious liability. Meeran, a partner in Leigh Day who was involved in many of the English cases described above, draws an interesting analogy between cases of this type and product liability cases: 'Save that one is dealing with "processes" rather than "products" an analogous duty to that owed by a manufacturer to consumers should be imposed ("process" liability).
Indeed here it is arguable that the proximity of a [transnational corporation] to 184  overseas employees of its subsidiaries [and the people in its vicinity] is closer than that of a manufacturer to consumers of its products. ' 187 Chandler 188 is the first, and so far the only, English case where a parent was held directly liable towards the victims of its subsidiary's operations.
The question whether the issue of a parent company's liability under the Chandler principles falls under the company law exclusion from the scope of Rome II is determined by autonomous interpretation. Guidance can be found in the CJEU case law on the definition of the concept of 'civil and commercial matters' for the purposes of the Brussels jurisdictional regime. It will be remembered that in Baten 189 the CJEU held that it was necessary to examine the basis and the detailed rules governing the bringing of a claim to determine whether it concerned a civil or commercial matter.
Much has been written about the nature of the Chandler type of liability in English law. Almost all scholars agree with Arden LJ that the case was in no way concerned with the piercing of the corporate veil. 190 In essence, a duty of care was imposed on the parent company because control represented an assumption of responsibility, which, in turn, satisfied the requirements of proximity and justice demanded by the duty of care analysis in the English law of torts. Chandler has been criticised for conflating the four elements of control, assumption of responsibility, proximity and fairness into one 'pragmatic' inquiry into circumstances in which a duty of care might exist, and introducing a degree of uncertainty with regard to each of them. 191 Nevertheless, one thing is clear. Control was the key factor for imposing a duty of care. The nature of the relevant control is therefore crucial for determining whether the Chandler type of liability concerns tort law or company law for the purposes of Rome II. As Joseph explains, the control test in direct liability cases is different than in piercing cases: 'The issue is not control by the parent over the subsidiary. Rather, the relevant control is that exercised by the parent over the conduct which gave rise to the tort at issue. Thus, the relevant "control test" focuses on the extent to which a parent is somehow in control of the causes of the tort, which will be linked to, but will not be the same as, the issue of a parent's control over its subsidiary.
Similarly, a parent corporation may attract direct liability if it undertakes to perform services for a subsidiary... For example, employees of subsidiaries have successfully sued parent companies on the basis that the parent undertook but failed to provide a safe working place on behalf of the subsidiary.' 192 Indeed, some of the relevant features of Chandler are that the parent had employed group medical and safety officers who had overseen the health and safety of employees across the group, that the contemporaneous documents, such as board minutes, showed that the parent had taken a direct interest in the working practices of the subsidiary that had employed the claimant, and that the parent and the subsidiary had shared directors who had been fully aware of what had been happening on the ground. 193 Looking from this perspective, the 191 Petrin ibid. 192 Joseph, above, n 8, 136 (footnotes omitted). 193 But see Muchlinski, above, n 7, 313 ('where decision-making is so centralised, that major policies could not have been formulated or put into operation without the direct involvement of the parent company, the parent ought to be answerable. In these circumstances the parent is likely to be aware, or ought to be aware, of the risk to potential claimants of such group actions, and to be sufficiently proximate to hold a duty of care towards Lowenfeld proposes a unique solution that favours the claimant: 199 if the forum is where the parent is established and its law imposes a liability on it in respect of the actions of its foreign subsidiary, then that law should be applied; if, on the other hand, the law of the country of the injury imposes enterprise liability but the law of the parent does not, then the former law should be applied.
In conclusion, it seems that the issue of whether a duty of care should be imposed on a parent company under the Chandler principles is to be classified as tortious for choice-of-law purposes, thus triggering the application of the choice-of-law rules of Rome II. In the unlikely case that this issue is classified as pertaining to company law, the English traditional choice-of-law rules will determine the governing law. Although not free from doubt, it seems that the law applicable to this issue under the traditional rules would be the law of the subsidiary. If so, English law, including the Chandler principles, would not apply in the Lubbe v Cape type of case brought in the English courts against an English-based parent company, but typically the law of a developing country. This law will typically contain laxer environmental and compensation standards in comparison with the law of the parent company's home country.

(2) Choice of law in tort
If the issue of whether a parent company is liable under the Chandler principles is classified as a tort law matter for the purposes of Rome II, the choice-of-law rules of that instrument will determine whether English law applies.
In the Lubbe v Cape type of case, important decisions concerning the operations of the subsidiary are typically taken by the parent company in its home country. Decisions vary from those concerning general issues of policy to those concerning day-to-day operations.
The notorious Bhopal litigation 200 represents one end of the scale, where the documents seem to have shown that the decision to shut off the refrigeration unit on the tank of methylisocyanate in Bhopal, India, which led to the warming up and explosion of the gas, was taken in the parent's US headquarters, and communicated by letter to the management of the Indian subsidiary. Typically, however, claimants allege that the parent's superior knowledge of the health and safety issues and knowledge of the subsidiary's unsafe system of work, coupled with its inaction, are the grounds for holding the parent liable. According to Article 7 of Rome II, the law applicable to environmental damage is by the default the lex loci damni, but the victim can choose the lex loci actus. In the Lubbe v Cape type of case, it is clear that the country of the damage is the country of the subsidiary. The key question is therefore whether the decisions concerning the operations of the subsidiary taken by the parent in its home country that start the chain of events resulting in environmental damage are to be considered as the relevant event for the purposes of Article 7.
Opinions are divided. Some scholars argue that the decisions taken by the parent company constitute the relevant event, 201 others that the relevant event is the tortious action of the subsidiary directly causing the environmental damage. 202 To answer the question, one should look closely at the choice-of-law rules of Rome II and at the case law on Article 7(2) of the Brussels I Recast. Article 7 of Rome II states that the law applicable to environmental damage is 'the law determined pursuant to Article 4(1)', or, if the victim so chooses, the law of the country of the event giving rise to the damage. According to Article 4(1), the applicable law for torts in general is the law of the country where the direct damage occurs regardless of where the indirect damage occurs. The focus on the direct damage is said to be required by the principles of legal certainty and foreseeability and by the need to 'ensure a fair balance between the interests of the person claimed to be liable and the person sustaining the damage'. 203 The parent company's decisions in its home country that start the chain of events resulting in environmental damage can be regarded as an 'indirect event' in the sense that it precedes the subsidiary's tortious action causing the damage directly. Should the distinction between direct and indirect damage for the purposes of Article 4(1) be applied by analogy to the question of the nature of the event giving rise to the damage for the purposes of Article 7, this 'indirect event' would be disregarded for choice-of-law purposes.
There are no CJEU cases directly on this point. The case law on Article 4(1) of Rome II and Article 7(2) of the Brussels I Recast is potentially relevant. It will be remembered that the CJEU cases dealing with the jurisdictional treatment of indirect damage and indirect victims demonstrate that only the place where the direct victim suffers direct damage is of jurisdictional relevance. 204 The only CJEU case on the interpretation of Article 4(1) of Rome II, Florin Lazar v Allianz SpA, 205 follows this line of cases and confirms that it is only the place where the direct victim suffers direct damage that is relevant for the purposes of Article 4(1) of Rome II. Should this distinction be applied by analogy to the question of the nature of the event giving rise to the damage for the purposes of Article 7 of Rome II, both the 'indirect event' (i.e. the parent company's decisions in its home country that start the chain of events resulting in environmental damage) and the actions of the 'indirect tortfeasor' (i.e. the parent whose decisions concern the operations of the subsidiary) would be disregarded for choice-of-law purposes. This logic seems to be supported by a recent CJEU decision in Melzer v MF Global UK Ltd. 206 Here, a German private investor brought a tortious claim in Germany against an English broker trading in futures on the basis that he had been solicited as a client in Germany by a German company. The defendant seems to have been sued both for its own alleged wrongdoing in England and for assisting the German company's alleged wrongdoing in Germany. 207 Furthermore, it seems to have been conceded that the only viable jurisdictional basis was with respect to the German company's alleged wrongdoing in Germany, which the defendant had allegedly assisted. 208 As Dickinson explains, this case 'appears to be an attempt to turn the tort upside downto treat a tort committed in London with facilitation from Germany, as one committed in Germany with facilitation from London'. 209 The CJEU interpreted Article 5(3) of the Brussels I Regulation, the predecessor of Article 7(2) of the Brussels I Recast, strictly and refused to allow the German courts to assume jurisdiction over the English co-perpetrator. Similarly, in the Lubbe v Cape type of case victims attempt to treat a tort committed in a developing country with facilitation from a developed country as a tort committed in a developed country. Should Melzer be applied by analogy to Article 7 of Rome II, this argument is bound to fail. In the first case the defendants were alleged to have unlawfully interfered with the claimant's contracts for the carriage of goods by arresting the vessel in Panama. The decision to arrest the vessel was taken by the defendant's London branch. According to the court: 'In one sense the decision of the branch of the Defendants in London can be said to have given rise to and to be the origin of the damage because the arrest is executed pursuant to that decision. However, it can also be said that the arrest is the event which gives rise to the damage and is the origin of the damage because without the arrest there would be no interference with the contracts of carriage.' 212 The court held that the event giving rise to the damage for the purposes of Article 5(3) of the Brussels I Regulation occurred in Panama. In Vava v Anglo American South Africa Ltd, 213 the court was concerned with the question whether the defendant, a South African subsidiary of an English-based parent company, was domiciled in England for jurisdictional purposes.
The claimants argued that the defendant's central administration was in England because the management entrepreneurial decisions relating to its business had been taken in England at the headquarters of the parent. The court, however, held that a company's central administration was not where such decisions were taken, regardless of whether they were taken by the company, its parent or anyone else. An important reason was that accepting the claimants' argument would mean that if such decisions were determined predominantly by the wishes of a bank or other institution on which it relied for its financial survival, then the defendant would have its central administration where the bank or institution took its decisions. These two cases show that the English courts interpret restrictively the relevant jurisdictional connecting factors in cases in which there are joint tortfeasors. If these cases are applied by analogy to the choice-of-law context in the Lubbe v Cape type of case, the tortious action of the subsidiary directly causing environmental damage will be considered as the relevant event for the purposes of Article 7 of Rome II.
It is also important to look at the High Court decision in RTZ v Connelly 214 and the second Court of Appeal decision in Lubbe v Cape. 215 Wright J, when dealing with the case after the decision of the House of Lords in RTZ v Connelly, 216 was clear that the cause of action had arisen in Namibia for the purposes of the common law choice-of-law rules. It was not the making of decisions in England that was crucial, but the concrete results that this produced in Namibia. 217 His Lordship also referred to a decision of the Court of Appeal to similar effect in Durham v T&N Plc. 218 In Lubbe v Cape, the second Court of Appeal 219 departed from the decision of the first Court of Appeal 220 that, for the purposes of the common law choice-of-law rules, the breach of the parent's duty of care had occurred in their boardroom in England and held that the obligations of the parent towards the employees of the subsidiary and the people in the vicinity were to be determined under South African law.
Although not made under Rome II, these decisions illustrate the likely answer to the question whether the decisions concerning the operations of the subsidiary taken by the parent in its home country are to be considered as the relevant event for the purposes of Article 7.
Instead of framing their claims in terms of environmental damage, victims may claim for the alleged violation of their human rights. Rome II does not have special choice-of-law rules for human rights claims. The general choice-of-law rules of Article 4 would apply and point to the application of the law of the country of the direct damage. Since the direct damage in the Lubbe v Cape type of case occurs in the subsidiary's country, the law of that country will govern. It is highly unlikely that that law will be displaced in favour of the law of the parent company's home country under the escape clause of Article 4(3).
In conclusion, regardless of how a claim for environmental damage in the Lubbe v Cape type of case is framed, the applicable law is almost certain to be the law of the country of the subsidiary, typically a developing country. This law will typically contain relatively lax environmental and compensation standards in comparison with the law of the parent company's home country. Consequently, even though victims of environmental damage are able to obtain the jurisdiction of the English courts over the parent established in that country, they will struggle to find lawyers willing to take on their case on a 'no win no fee' basis.

Conclusion
This article examines whether, and to what extent, the rules of European private international law, which frame international litigation in the Member State courts, contribute to the regulation of the environment. Contrary to the majority opinion, this article argues that these rules fail in their pursuit of the cosmopolitan goals of EU environmental policy. They are therefore an inadequate tool of global governance.
After outlining the regulatory potential of private international law with regard to the protection of the environment, this article observes that international environmental litigation in EU courts is effectively confined to two types of case. In the first type private claimants sue an operator whose actions in one country directly cause environmental damage elsewhere. In the second type of case the claim is brought by victims against a Europeanbased corporation operating in an extraction or chemical industry whose overseas subsidiary, typically in a developing country, causes environmental damage. Proceedings are commenced by private claimants only, arguably because European private international law excludes from its scope claims by public authorities against foreign operators. Claims by public authorities, however, are crucial in cases of pure environmental damage where no individual suffers actionable personal injury, property damage or economic loss or where such harm is thinly spread among a number of victims, as recognised by the Environmental Liability Directive. The second type of case is limited to extraction and chemical industries not just because of their characteristics and location but also because such industries tend to be run by hierarchical parent-subsidiary corporate groups. Other industries tend to be run by corporate groups with more open and flexible forms of corporate organisation or business networks, where the chances of ascribing liability to the parent company/the controlling enterprise(s) are so low that arguably no lawyer will take on a case on a 'no win no fee' basis.
The rules of European private international law also have the following regulatory effects. By allowing the victim of environmental damage to choose both the forum and the applicable law between the courts/law of the country of the damage and the courts/law of the country of the event giving rise to the damage, the Brussels I Recast and Rome II address relatively adequately the first type of case. But the cases of this type that are brought in EU courts will almost always concern actions committed in the EU and/or environmental damage suffered in the EU. By guaranteeing the victim in this type of case the choice of the law of an EU Member State, Rome II helps to raise the level of environmental protection within the EU and at its borders. Crucially, European private international law fails to deal with the globally more important and frequent cases of the second type. The Brussels I Recast allows the claimants in this type of case to commence proceedings in the EU against a parent company domiciled here, although it gives an amount of discretion to the Member State courts to stay their proceedings when parallel or related proceedings are already pending in a third country. The choice-of-law rules of Rome II, however, arguably lead to the application of the law of the country of the subsidiary, typically a developing country. This law will typically contain laxer environmental and compensation standards in comparison with the law of the parent's home country, which will make it hard for victims to obtain representation on a 'no win no fee' basis. Importantly, claimants in this type of case in the English courts will be unable to invoke the type of liability recently established in the English law of torts in Chandler v Cape. 221 European private international law therefore effectively shields European multinational corporations from liability for the environmentally detrimental and degrading effects of their overseas operations.
For these reasons, the rules of European private international law overall fail in their pursuit of the cosmopolitan goals of EU environmental policy. In fact, they often lead to outcomes that are contrary to the universally accepted environmental law principles on which EU environmental policy is based such as the principles that environmental damage should as a priority be rectified at source and that the polluter pays.
It is worrying to see similar developments occurring elsewhere. In April 2013, for example, the US Supreme Court effectively barred victims of gross violations of human rights and the environment outside the US by foreign corporations from bringing ATCA claims. A year earlier UK Parliament passed a bill overhauling the system of conditional fees to the detriment of claimants in international environmental litigations. Under the new rules, the claimants' lawyers can recover from the losing defendant only 'proportionate' (as opposed to 'necessary') legal costs and no success fees and litigation insurance premiums. 222 Coupled with the fact that, after the entry into force of Rome II, the assessment of damages is a matter for the law governing the tort, 223 often the law of a developing country, it is likely that these developments will have a chilling effect on international environmental litigation in England. 224 It should not be forgotten that the House of Lords in Lubbe v Cape 225 and Connelly v RTZ 226 regarded the shortcomings of the Namibian and South African legal systems which had a similar effect on international environmental and human rights litigation in these countries as leading to a denial of justice. It is likely that the future of international environmental litigation in England will depend on the willingness and ability of nongovernmental environmental organisations to bring and sustain this type of litigation, as is the case now in the Netherlands and Sweden. 227 The costs of such litigation, however, may turn out to be prohibitive.
In order to achieve fully its regulatory potential, the rules of European private international should be improved. The Brussels I Recast and Rome II should include within their scope claims by public authorities against foreign operators. The rules of the Recast that allow the courts to stay their proceeding when parallel or related proceedings are already pending in a third country should be applied restrictively. The victims of environmental interpretation is more likely. It is for this reason that the Brussels I Recast and Rome II should be amended to expressly include within their scope claims by public authorities against foreign operators 228 and to expressly allow the victim to choose the law of the parent's country. On their own, however, these improvements will not be enough. They should be complemented with adequate rules on the funding of international environmental litigation and liability of parent companies in corporate groups and dominant enterprise(s) in business networks for environmental damage caused by their subsidiaries, affiliates and cooperating enterprises. At one point, it seemed that English law would move in this direction. However, the proposed Corporate Responsibility Bill 2003 introduced in the House of Commons never made it through to the second reading. 229 The time is ripe for these issues to be taken up by the EU legislator.