Mass tort litigation in England for overseas environmental harm

08 April 2024

In a series of recent environmental disaster decisions, the UK Supreme Court and Court of Appeal have found that:

  • it is at least arguable that a UK parent company could be responsible in law for harm that occurs in the local jurisdiction of its subsidiaries; and
  • that the trial in relation to any such harm can be heard before the courts of England and Wales, notwithstanding that the harm happened thousands of miles away.

The English court’s readiness to accept jurisdiction is often remarked upon by overseas lawyers who might expect jurisdiction to be refused in such cases in favour of the local courts where the harm occurred.

The recent cases represent some of the largest multi-party litigation that the English court has ever seen, and it is a trend that many anticipate will increase. Indeed, the Supreme Court’s recent refusal to give permission to appeal in the Mariana case will mean it is the largest opt-in group action to be brought before the English courts, with over 700,000 claimants seeking redress in the trial which is scheduled to last for 11 weeks beginning in October 2024.1

In this article we explore the recent decisions and unpack the English court’s approach, which has wider ramifications for UK-based parent companies and how they conduct themselves overseas. We also address (briefly) how the English court approaches the case management of these multi-party collective redress cases which involve many complexities, including those associated with the participation of hundreds of overseas claimants and third-party funders.

Bringing claims in England (Jurisdiction)

The commonality between the recent high-profile decisions is that a group of people affected by a disaster local to them brings a claim in England against both a local company and its English parent company, alleging each company is liable to them in tort for the damage they have suffered.2 The fact that one of the defendants (the so-called "anchor" defendant) is based in England can found the jurisdiction of the English court to hear the claim, even though the damage occurred, and the claimants are resident, overseas.

Parent companies have sought to challenge the jurisdiction of the English court to hear these claims. It has been argued that the claimants have no real arguable case that these parent companies owe them any duty of care, and instead have only joined them to proceedings so as to benefit from the jurisdiction of the English court (with the perceived advantages for claimants including, for example, the possibility of extensive disclosure). It has also been argued that England is not the proper forum to hear such claims, and that instead the courts local to the damage and the subsidiary company are more appropriate and convenient.

These jurisdiction challenges have involved a resource-consuming series of appeals; Vedanta3 and Okpabi4, in particular, were both appealed to the UK Supreme Court. In both cases, the Supreme Court found that the cases should proceed to substantive trial in England (Vedanta subsequently settled out of court). In Okpabi, the lower courts had wrongly dismissed the claims as having no prospect of success against the English parent companies. The lower courts had erroneously been drawn into conducting a "mini-trial" on the evidence, whereas they should only have been considering whether the claim disclosed no arguable case.

There is now a noticeable trend of allowing this type of case to go to trial in England. Although this is undoubtedly significant, care should be taken not to overinterpret these judgments on jurisdiction. In permitting the cases to go to trial, the court is not going further than acknowledging that there is a more than fanciful argument to be heard, and thus the claimant should be permitted to have that argument determined at trial.

Role of parent company – anchor defendant

As we note above, the English parent company in these cases acts as an "anchor" defendant. It is the fact that the parent company’s jurisdiction of domicile is England which founds the English court’s jurisdiction to try the claim. This in turn forms the basis on which the claimant is permitted also to sue the local subsidiary in England. Where the case concerns allegations against both parties in respect of the same damage, the English court is asked to take jurisdiction over the dispute against both parties so as to avoid the risk of proceedings going ahead in both England and an overseas court and reaching different conclusions – known as the risk of irreconcilable judgments.

Permission of the English court is needed to serve the claim on the overseas subsidiary. This permission is usually sought on the basis that the subsidiary is a "necessary and proper party" to the claim made against the "anchor" defendant parent company. In Vedanta, the Supreme Court neatly summarised what a claimant needs to demonstrate in order to establish this:

“i) that the claims against the anchor defendant involve a real issue to be tried;

ii) if so, that it is reasonable for the court to try that issue;

iii) that the foreign defendant is a necessary or proper party to the claims against the anchor defendant;

iv) that the claims against the foreign defendant have a real prospect of success;

v) that, either, England is the proper place in which to bring the combined claims or that there is a real risk that the claimants will not obtain substantial justice in the alternative foreign jurisdiction, even if it would otherwise have been the proper place, or the convenient or natural forum.”5

It is important to note that the applications discussed above contesting the jurisdiction of the English court to hear these claims were decided under the regime in place before the UK left the European Union, which meant that the Recast Brussels Regulation (the Regulation) applied. Article 4 of the Regulation meant that the English court was prohibited from finding that England was not the appropriate forum to hear a case against an English domiciled company, even where another jurisdiction might also be suitable.

In Vedanta, the defendants argued that it was an abuse of EU law for the claimants to use the Regulation in this way, arguing that the claimants were only suing the parent company so as to found jurisdiction against the subsidiary. The lower court disagreed and its decision was upheld by the Supreme Court. Whilst it could be an abuse to sue a party solely to attract jurisdiction against another party, in this case the claimants had a genuine desire to obtain judgment against the parent company as well, notwithstanding that attracting jurisdiction against the subsidiary was a factor. This finding was a question of fact before the lower court and therefore not open to the Supreme Court to consider (as it can only consider questions of law).

The Supreme Court discussed in Vedanta the fact that prior to EU law on this subject, the English court used to deal with situations where it felt the overseas jurisdiction was more appropriate by both refusing permission to serve out and by staying the English proceedings against the would-be anchor defendant. We are now seeing revived application of these principles in the ESG sphere. For example, in Limbu v Dyson Technology Ltd6 the court concluded that England was not the appropriate forum for claims brought by Nepalese and Bangladeshi migrant workers alleging they had been subjected to forced labour in Malaysian factories run by the Dyson group’s Malaysian suppliers. Applying the long-standing test from Spiliada Maritime Corporation v Cansulex Ltd7, requiring the court “to identify the forum in which the case can be suitably tried for the interests of all the parties and for the ends of justice”, the court concluded that the claim should not be heard in England; Malaysia was clearly and distinctly the more appropriate forum and there was not a risk that substantial justice would not be obtained if the case were heard in Malaysia.

Role of parent company – duty of care

Whether or not the parent companies in these cases owed a duty of care to the claimants is a question of fact to be determined at the substantive trials. Nevertheless, the Supreme Court has, through these cases, made a number of observations as to how one might go about answering that question, indicating the following. 

  • Parent company duty of care is not a distinct category of liability in negligence. Whether a duty of care arises “depends on the extent to which, and the way in which, the parent availed itself of the opportunity to take over, intervene in, control, supervise or advise the management of the relevant operations…of the subsidiary”8, or, as stated in Okpabi, “there is no special test applicable to the tortious responsibility of a parent company for the activities of its subsidiaries”.9
  • It is incorrect as a matter of law to state that the setting out of group-wide policies or standards cannot, in itself, suffice to incur a duty of care. On the contrary, such policies “may be shown to contain systemic errors which, when implemented as of course by a particular subsidiary, then cause harm to third parties”.10
  • Focussing on "control" by the parent company can lead to error. As stated in Okpabi “control of a company and de facto management of part of its activities are two different things. A subsidiary may maintain de jure control of its activities, but nonetheless delegate de facto management of part of them to emissaries of its parent”.11 This means that the real question is which company is, as a matter of fact, managing relevant matters.
  • If a parent company holds itself out as having the requisite degree of control over its subsidiaries, for example through statements made in annual reports, this may be enough to incur liability regardless of the factual question of control because the company has undertaken publicly to have responsibility for certain matters.

A tension has therefore arisen for parent companies who need to demonstrate to the market that they have adequate oversight of foreign subsidiaries and their operations whilst at the same time restricting the risks arising where a duty of care crystalises.

Parent company willing to submit to overseas jurisdiction?

In certain of the cases recently brought on this topic, the parent companies have made offers to submit to the overseas jurisdiction to bolster their argument that they should not be sued in England but rather in the jurisdiction where the damage occurred. The weight of such an argument will be fact-specific, but it is an interesting departure from the primary argument that some have advanced that they should not be sued at all in respect of damage for which it is (at most) the subsidiary company that should be held liable.

In Vedanta, this was a significant factor in the Supreme Court’s comments relating to the risk of irreconcilable judgments. The point arose because the Claimants had highlighted this risk as potentially arising if they were not allowed to sue the local subsidiary in England as it might then pursue parallel proceedings in England (against the parent) and in Zambia (against the local subsidiary). The Supreme Court commented that Vedanta’s expression of willingness to submit to Zambia’s jurisdiction should have been taken into account by the judge below, because in effect it meant that risk of irreconcilable judgments should be discounted. This is because if the defendant is willing to be sued overseas, and can also be sued in England, it is up to the claimant to decide in which forum to bring proceedings. The claimant cannot then rely on the argument that there is a risk of irreconcilable judgments since it will have generated that problem itself by bringing proceedings in multiple forums where it could have seised just one. The Supreme Court determined that in such circumstances, the risk of irreconcilable judgments should not be considered a “trump card”12 and went so far as to state that, were it not for the real risk that substantial justice might not be available in Zambia (which we address in the section below), it would have found that England was not the proper place to bring the claim.13

Risk of substantial injustice

Where a defendant satisfies the English court that it should grant a stay because another, more suitable forum is available, it is open to the claimant to ask the court nevertheless to refuse that stay because there is a real risk that the claimant will not obtain justice in that other forum. That issue has been argued in these recent cases, and has often been persuasive, resulting in the English court accepting jurisdiction. For example, this is the factor alluded to above that led the Supreme Court to find that the English court should take jurisdiction in Vedanta notwithstanding that Zambia would otherwise have been the proper place for the proceedings.

Various factors might provide the necessary persuasive evidence that there is a real risk of substantial injustice in the other forum. In Vedanta, the factors included that funding was not available (in circumstances where the claimants could not afford to bring their claim without it), and that local legal teams would not be sufficiently well-equipped to handle disputes of the associated scale and complexity. It was noted that there is a general principle that the availability of legal aid will not usually be considered a factor, but exceptionally it can be taken into account if the claimant can establish that lack of funding will mean substantial justice cannot be achieved.14

The preceding authorities in which a lack of funding was considered relevant have generally also been mass harm cases (being about exposure to asbestos15 and exposure to radiation16). In addition, funding is often inherently intertwined with the availability of a sophisticated legal team to act, as the scale and complexity of such cases means that a suitable team of lawyers will be required if the case is to be brought in an effective manner. Considering that the Mariana case will be among the largest ever heard in England, the scale can be considerable indeed. Teams of lawyers will rarely take on such large cases without remuneration, and as such funding becomes essential.

In Mariana17 the Court of Appeal considered that it is important to identify the exact alternative to proceedings in England. In that case, it was unrealistic to suggest the claimants might proceed individually against the defendants in Brazil, so the only alternative proceedings in Brazil would be an Ação Civil Pública, which was not appropriate for a variety of reasons including that they can only be brought by particular bodies specified in law rather than by individual private claimants, and that this is an asymmetric procedure that can only generate a “generic sentence” establishing liability, and will only do so if it finds in favour of the claimants. If it does not find in favour of the claimants, then it will make no findings at all, and so importantly will not declare the defendants are not liable. Thus the claimants remain free to seek other forms of redress. The Mariana judgment also mentioned clear evidence presented to the Court of Appeal that neither legal aid nor third party litigation funding would be available to the claimants to support a case in Brazil.18

By contrast, in Limbu v Dyson the High Court was not persuaded that there was a risk substantial justice could not be achieved in Malaysia. The judge heard arguments on similar points to those raised in Vedanta regarding the availability of suitable fee arrangements and legal teams. But the court was persuaded by evidence that in fact suitable lawyers would be available and willing and able to act on workable fee arrangements.

Case management considerations – options for collective redress

Collective redress claims in England can take a few different forms. In this section, we consider these and the judicial comments that have been made about their availability for bringing mass tort claims of this sort. As will be seen, the selection of procedure can be crucial to the claim’s prospects of success.

Representative action

The English court’s procedural rules contain a specific rule permitting a representative to bring a claim on behalf of multiple parties where they all have the same interest in the claim. In practice, the requirement that all the represented parties have the “same interest” proved problematic for the claimants in the case of Jalla19, which concerned an oil spill in Nigeria.

The Court of Appeal confirmed in Jalla that the scope of representative actions is rather narrow. Because the represented parties were occupants of different parcels of land and suffered a range of different alleged damage, it would be necessary to consider in relation to each piece of land the issues of loss and damage, causation, and limitation. The interests of the different represented parties were therefore not the same and would require participation from the members of the class to resolve.

It seems unlikely that parties will make further attempts to use the representative action procedure to bring mass environmental harm claims following this decision, which commented that two other forms of collective redress, namely group litigation orders (GLOs) and use of a sample from amongst a large group of named claimants, are usually deployed instead in cases of mass environmental harm.


A GLO is an order granted by the court to enable claims giving rise to common or related issues of fact or law to be managed together. Both claimant and defendant are able to apply for a GLO, but there is no automatic entitlement to one and they are granted at the discretion of the court and require the permission of either the President of the King’s Bench Division, the Chancellor of the High Court or the Head of Civil Justice in the County Court as appropriate.

A GLO was sought in the case of Alame20, which is related to the Okpabi case concerning an oil spill affecting communities in Nigeria. The court has granted the GLO, but before doing so, the judge ordered that a significant amount of further information was required from the claimants before the GLO could be made. This included requiring details from each claimant of where they say they suffered damage, and when they say they were first impacted.21 This illustrates the procedural hurdles that can be required to be overcome for Claimants wishing to join a GLO. That said, the requirement for GLO claims to raise “common or related issues” is significantly lower than the “same interest” requirement for a representative action.

While commenting on the GLO procedure in Alame, Mrs Justice O’Farrell noted that the proceedings were also going to be transferred from the Technology and Construction Court (TCC) to the general Queen’s Bench Division (as it then was) despite opposition to this transfer by the defendants. This was because the QBD would be able to offer an allocated managing judge, which would lead to more efficient case management. The English court’s faith in its own case management capabilities for group litigation was also underlined in the Mariana case, where the Court of Appeal commented that “the TCC is well-known for its ability robustly and actively to case manage complex litigation, including group litigation”.22

Sample claimants

Whilst the Court of Appeal in Jalla found that it was not suitable to be tried as a representative action, Coulson LJ did note in his judgment that it is common in the TCC not only to use GLOs but also to use the process whereby all claimants are named as parties in their own right, and then a “representative sample of lead claimants” is selected through “careful sampling”, and their cases are used to decide the substantive issues.23 He cited as an example the Ocensa Pipeline Group Litigation24, in which four sample claims were used for the trial of a dispute in which Colombian farmers sued an English subsidiary of BP over damage allegedly caused by oil pipeline construction.

The cases discussed here, and the judicial comments about the English court’s ability to manage collective claims, suggest a strong willingness to hear these cases in England where there is a serious issue to be tried against an English company.

Conclusions on current landscape

The cases discussed in this article present a mixed picture. There is a clear trend of the English court accepting jurisdiction in mass environmental harm claims. Attempts to have them removed from England on grounds of jurisdiction have frequently failed, at least after appeals have been exhausted.

With that said, it is important to remember that these cases all remain at very early stages, notwithstanding how long ago the alleged damage in many of them occurred. This is testament both to the complexity of such claims and to the effort that has gone into resisting them proceeding in England. The oil spills cited in Alame and Okpabi, for example, occurred in 2011 – 2013 and, in the case of the latter, the litigation has taken over six years just to clear the jurisdiction phase. It is fair to say, therefore, that the lawsuits following such disasters can be very long running.

What cannot yet be stated with any certainty is how likely it is the English parent companies in these cases will be found liable. Allowing the cases to proceed to trial establishes only that there is a genuine issue to be tried. We now have the benefit of judicial comments in cases such as Vedanta and Okpabi as to the type of behaviours that might indicate an assumption of responsibility, but only a substantive trial judgment will be able to tell us more about the circumstances in which a parent company might face liability. The Mariana case will be closely followed for possible answers. Meanwhile, claimants may well be encouraged by the recent appellate court decisions confirming that this type of claim can and will be heard in England. In those circumstances, English parent companies whose subsidiaries are connected with environmental disasters can expect that, at least for now, the risk of these claims will continue.

1 Municipio De Mariana v BHP Group (UK) Ltd & Anor [2023] EWHC 1134 (TCC)

2 The cases have concerned an oil spill in Nigeria (Okpabi; Alame), discharge of toxic matter from a mine in Zambia (Lungowe), destruction caused by a burst dam in Brazil (Mariana), and serious violence and destruction following an election in Kenya (AAA v Unilever))

3 Vedanta Resources Plc and another v Lungowe and others [2019] UKSC 20

4 Okpabi and others v Royal Dutch Shell and another [2021] UKSC 3

5 [2019] UKSC 20 per Lord Briggs at [19]

6 [2023] EWHC 2592 (KB)

7 [1987] 1 AC 460

8 [2019] UKSC 20 per Lord Briggs at [49]

9 [2021] UKSC 3 per Lord Hamblen at [27]

10 [2019] UKSC 20 per Lord Briggs at [52]

11 [2021] UKSC 3 per Lord Hamblen at [147]

12 [2019] UKSC 20 per Lord Briggs at [84]

13 [2019] UKSC 20 per Lord Briggs at [87] and at [102]

14 Applying Connelly v RTZ Corp Plc (No.2) [1998] AC 854

15 Lubbe v Cape plc [2000] 1 WLR 1545

16 Connelly v RTZ Corp Plc (No. 2) 1998] AC 854

17 Municipio de Mariana v (1) BHP Group (UK) Ltd and (2) BHP Group Ltd [2022] EWCA Civ 951

18 The Supreme Court refused permission to appeal the Court of Appeal’s judgment (Permission to Appeal June 2023 – The Supreme Court), and the trial is now scheduled to take place in October 2024.

19 Jalla and another v Shell International Trading and another [2021] EWCA Civ 1389

20 Alame and others v Royal Dutch Shell and another [2022] EWHC 989 (TCC)

21 [2022] EWHC 989 (TCC) at [74]

22 [2022] EWCA Civ 951 at [135]

23 [2021] EWCA Civ 1389 at [50]

24 [2016] EWHC 1699 (TCC)