High Court allows case to proceed against defendant companies domiciled in England despite claims having more real and substantial connection with Brazil

The High Court has refused an application brought by two UK-domiciled companies for an order that it should not exercise its jurisdiction to try the claims against them. The court held that, while Brazil had a closer connection to the case, there was a real risk that the claimants would not obtain substantial justice there due to difficulties in obtaining legal representation for their claims: Catarina Oliveira da Silva v Brazil Iron Ltd [2025] EWHC 606 (KB).

This decision is significant for UK-based companies with operations elsewhere, particularly where questions of the control and direction of those operations arise. It joins a series of cases termed exceptional by the courts in which difficulties faced by impecunious claimants in obtaining funding for representation in their home jurisdiction is a significant factor in the jurisdictional analysis.

Another such case is Limbu v Dyson Technology Ltd [2024] EWCA Civ 1564 (considered in our recent blog post), in which the Court of Appeal found there was a serious risk that the claimants would not be able fund their claims in Malaysia. In that case, however, the court's decision that England was the appropriate forum was reinforced by other factors connecting the claims to England. In the present case, in contrast, it is clear that the court would have held that the claims should be tried in Brazil if it weren't for the funding difficulties.

The present case is also of interest in rejecting the defendants' attempt to address concerns about the claimants' ability to obtain justice in Brazil by offering various undertakings, including to pay certain of the claimants' disbursements such as technical experts' fees. The court commented that the discussion in Dyson leads to the "unsurprising conclusion that it is not fair for a defendant to have any degree of control over a claimant’s expert evidence, whether by paying for it or otherwise". Following these decisions, that seems an unpromising route for defendants looking to counter arguments based on funding difficulties in the foreign forum.

Background

Claims were brought by around one hundred residents of indigenous communities in Brazil against two UK-domiciled defendants. The claimants claimed to have suffered environmental damage to land and other damage as a result of pollution from a mine operated by the defendants' Brazilian subsidiary. They claimed that the subsidiary operated the mine under the control and direction of the defendants, and this gave rise to various strict and fault-based liabilities under Brazilian law.

The defendants applied for an order that the court should not exercise its jurisdiction to try the claims, and for a stay of proceedings, on the basis that the claims should be brought in Brazil.

Various proceedings for environmental damage were already underway against the Brazilian subsidiary, including a Civil Public Action in Brazil. However, the defendants to the present claim could not be included in this action and would only become liable in the event of the subsidiary's insolvency. An ordinary civil claim had also been issued by other parties in Brazil against the subsidiary.

Each side relied on extensive expert evidence and evidence of fact, predominantly from Brazilian lawyers experienced in environmental litigation. This included evidence as to whether the claimants could obtain legal representation and funding if the claims proceeded in Brazil.

Decision

The High Court (Bourne J) refused to stay the proceedings.

The principles to be applied were those set out in Spiliada Maritime Corporation v Cansulex Ltd [1987] 1 AC 460. Since the defendants had been validly served with the proceedings in England, the burden was on them to show that there was another available forum that was clearly and distinctly more appropriate (often referred to as Stage 1 of the Spiliada analysis). If this was the case, it was for the claimants to show by cogent evidence that there was a real risk that they would not be able to obtain substantial justice in Brazil (Stage 2).

Stage 1 – Brazil clearly and distinctly more appropriate

In submitting that the action had the most real and substantial connection with Brazil, the defendants argued that:

• The alleged wrongdoing occurred at the mine, which was in Brazil, and the alleged pollution occurred in Brazil.
• The claimants were in Brazil, as were witnesses who might give evidence about the mine and its impact.
• Most of the witnesses would only speak Portuguese.
• Most of the documents were likely to be in Brazil and in Portuguese.
• Although it was alleged that the mine was controlled by English companies, management acts did not necessarily take place in England. In any event, the core issue was the environmental damage caused in Brazil.
• There were already other proceedings regarding the mine underway in Brazil. These overlapped with the issues in the present litigation and meant a risk of irreconcilable judgments.
• Some of the claimants claimed to be part of the Quilombola community, a group with specific rights. A Brazilian court would be better placed to determine any social and cultural issues arising.
• A Brazilian court would also have knowledge of the region and the environment, as well as the customs of its population.
• Brazil has a civil law system, which is very different from England's common law system.

In reply, the claimants emphasised that this claim was concerned with the conduct and liability of companies domiciled in England, and the defendants had been validly served in England. In addition:

• The issue of the defendants' control over its subsidiary was central, and such control would have emanated from England.
• The control issue was likely to depend on documents created by the defendants in English and located in England.
• The defendants' defence would be co-ordinated from their offices in London (a relevant factor in Dyson).
• The extensive categories of factual and expert witnesses that the defendants predicted as being necessary were over-stated.
• Acts or omissions giving rise to the defendants' alleged liability as an indirect polluter took place in England, meaning that the place of commission of the tort was either a factor in the claimants' favour or neutral.
• While Brazilian law applied, the real issues in the case were factual.
• The importance of Quilombola status and the claimants' property rights had been over-stated. The right to bring a property-based damages claim was founded in possession, and this was a familiar concept to an English court.
• There was no risk of irreconcilable judgments. The claimants were not parties to the Brazilian Civil Public Action, and the defendants could not be parties to it. The control issue played no part in the Brazilian actions.

The court examined the various factors presented to it. Several of them leaned towards England, including the defendants' domicile and service in England, and the fact that evidence on the issue of control was likely to emanate from England and be in English. However, factors leaning towards Brazil prevailed:

• There was a risk of inconsistency of outcome due to the other proceedings underway in Brazil (a factor said in Lungowe v Vedanta Resources Plc [2019] UKSC 20 to be significant).
• The most important issues were likely to concern the operation and regulation of the mine and its impact on the claimants. Although control was a relevant issue, the claims concerned environmental damage in Brazil. This was also where the relevant witness and documentary evidence would be.
• Brazilian law and regulation applied to the disputes. Although the English courts were used to applying foreign law, differences between the systems favoured the jurisdiction of the Brazilian courts, as did issues around Quilombola status, which would be unfamiliar to the English courts.
• Convenience and expense in terms of the location of claimants and witnesses and the need for translation pointed towards Brazil.

The court therefore concluded that Brazil, not England, was the forum with which the action had the more real and substantial connection.

Stage 2 – Ability to obtain substantial justice in Brazil

In relation to Stage 2 of the Spiliada analysis, the court concluded that there was a real risk, proven by cogent evidence, that the claimants would not be able to obtain substantial justice in Brazil because they would not be able to fund legal representation of the kind necessary to litigate the claims to a proper conclusion. This was not a criticism of the Brazilian legal system, which the court described as sophisticated, but rather because there were particular features of the claims which created difficulty. These included:

• Their relatively small size, with recovery perhaps being in the range of £10,000 to £30,000 per claimant.
• The number of claimants, which meant a large group of individuals with whom lawyers would have to interact, set against a fairly low total claim value.
• The claimants were of limited means and lived in remote areas, making legal consultations difficult and expensive.
• The claims were complex, with expert evidence required in several areas, and loss and damage likely to vary between claimants.
• This complexity meant that the claims would probably take years to resolve and place significant demands on the legal representatives involved.

If the case were to proceed in England, the claims could go to trial, even though the claimants were unable to pay their legal costs and disbursements, on the basis of a "no win, no fee" arrangement with their lawyers. The question was therefore whether there was a real risk that sufficient funding for the claims would not be available in Brazil. The following factors were relevant to the analysis:

• The claimants were clearly not able to fund the claims themselves and had not found representation in Brazil. Although conditional fee agreements (CFAs) of a sort existed, the CFA funding model used in England did not.
• Even assuming that legal aid would be available for many of the claimants' overheads, including expert witness fees, it would not pay for lawyers' fees.
• The CFAs available in Brazil were effectively damages-based agreements capped at a certain percentage of the damages claim. In this context, the small size of the claims became particularly significant.
• The overheads incurred by the claimants' lawyers would depend on whether and to what extent the claims could be joined together, which was a matter of some uncertainty. Even if the Brazilian courts were open to group actions, there might still have to be a number of separate claims.
• The defendants had offered to meet the claimants' costs of instructing their own technical experts to assist in their interaction with a court-appointed expert (as these costs would be unlikely to be covered by legal aid). However, the court held that this was not satisfactory for the reasons identified in Dyson, including that it would involve a conflict of interest for the defendants, and the claimants' ability to instruct the relevant experts could be delayed by disputes about whether the costs were reasonable and necessary.
• Although the figures for overheads were speculative, they would undoubtedly impact on costs recovery, leaving little return for lawyers.
• Brazilian lawyers considering taking on the case on the basis of a CFA would first undertake a cost-benefit analysis, balancing the potential returns against the work and investment involved and the possibility that the claims would fail, leaving the lawyers out of pocket.

On this analysis, there was a real risk that the claimants would not find representation in Brazil, simply because the economics of litigating the claims in Brazil and England differed. Although there were various ways in which the claims might be funded in Brazil, they all had various shortcomings or uncertainties. The other undertakings offered by the defendants (eg to submit to Brazilian jurisdiction and pay any damages ordered by a Brazilian court) might address some of the uncertainty in the cost-benefit analysis a prospective lawyer would conduct, but did not remove the risk that no suitable lawyer in Brazil would consider the case to be economically viable.   

Finally, the judge reviewed case law supporting the proposition that the risk of a lack of representation in a particular forum might exceptionally persuade the court that the forum was not one in which the case might most suitably be tried in the interests of the parties and for the ends of justice. These cases are:

• Connelly v RTZ Corp Plc [1998] AC 854, in which the Court of Appeal held that justice could be done in England but not in Namibia, noting that "the availability of financial assistance in this country, coupled with its non-availability in the appropriate forum, may exceptionally be a relevant factor" to be taken into account.
• Lubbe v Cape Plc [2000] 1 WLR 1545, in which the House of Lords concluded that, on the evidence that legal aid would not be available for the action in South Africa and no local firm would take on the case on a CFA, the case could probably not proceed there and this would be a denial of justice.
• Lungowe v Vedanta Resources Plc [2019] UKSC 20, in which the Supreme Court held that there was a real risk that the claimants would not obtain substantial justice in Zambia due to there being no legal aid or CFA representation, and a lack of local expertise in the specialist litigation involved.  

The court observed that the present case was one of this exceptional kind, where there was not merely a difference in the availability of funding in the two jurisdictions, but a real risk that justice would not be obtained in the foreign jurisdiction.

On this basis, Brazil was not the appropriate forum for the case.