Dhan Kumar Limbu & Ors v Dyson Technology Limited & Ors [2024] EWCA Civ 1564

[2024] EWCA Civ 1564Case No CA-2023-002333IN THE COURT OF APPEAL (CIVIL DIVISION) ON APPEAL FROM THE HIGH COURT KINGS BENCH Clive Sheldon KC sitting as a Deputy High Court JudgeVenue Royal Courts of Justice Strand, London, WC2A 2LLDate Friday 13 th December 2024
Before
SIR GEOFFREY VOS, MASTER OF THE ROLLS LORD JUSTICE POPPLEWELL&LORD JUSTICE WARBY
Between
Dhan Kumar Limbu & othersClaimants
Marie Louise Kinsler KC, Edward Craven & Thomas Fairclough (instructed by Leigh Day) for AppellantsCharles Gibson KC, Adam Heppinstall KC & Freya Foster (instructed by Baker & McKenzie LLP) for RespondentsHearing Hearing dates: 26 th& 27 th November 2024
LORD JUSTICE POPPLEWELL:

Introduction

[1]This appeal concerns Nepalese and Bangladeshi migrant workers who have brought claims against three companies in the well-known Dyson group, alleging that they were trafficked to Malaysia and there subjected to conditions of forced labour, exploitative and abusive working and living conditions, and in the case of some of them, detention, torture and beating, in the course of manufacturing components and parts in the supply chain for the Dyson group. It is not about whether the claims are made out (it is accepted that they have a real prospect of success), but rather whether England or Malaysia is the appropriate forum in which the claims can and/or should be determined. The applicable principles are those established in the seminal decision of the House of Lords in Spiliada Maritime Corp v Cansulex Ltd [1987] AC 460, and neither side suggested otherwise.[2]The case is a mixed case involving both service in and service out. The first and second defendants (‘D1’ and ‘D2’ or collectively ‘Dyson UK’) are English companies, which were served as of right here at their place of business, and sought a stay on forum non conveniens grounds. The third defendant (‘D3’ or ‘Dyson Malaysia’) is a Malaysian company, which was served pursuant to the grant of permission to serve out by Master Gidden as a necessary and proper party, pursuant to CPR PD6B para 3.1(3), which D3 sought to set aside on the forum non conveniens grounds that England was not the proper place in which to bring the claim (CPR 6.37(3)).[3]The judge below, Clive Sheldon KC, as he then was, sitting as a Deputy High Court Judge (‘the Judge’) conducted a hearing over three days in July 2023 with very extensive evidence served on both sides. He concluded that Malaysia was the more appropriate forum for the claims to be heard and that there was no real risk of the claimants being unable to access justice there. The claimants appeal on both aspects of his decision.[4]Such a decision involves the judge considering a range of different factors which may point to one or other jurisdiction as more appropriate, and evaluating the written evidence. That is an exercise in which views may reasonably differ, both in assessing the individual aspects of the evidence, and in assessing the relative weighting of the various factors. The principles applicable on an appeal by way of review from the exercise of such an evaluative assessment are well-established. Absent some procedural unfairness or irregularity, an appeal court will only interfere where the lower court has made an error of principle, such as taking into account irrelevant matters or failing to take into account relevant matters, or has reached a conclusion which exceeds the generous ambit within which reasonable disagreement is possible and so is plainly wrong: see for example Carroll v Chief Constable of Greater Manchester Police [2017] EWCA Civ 1992 [2018] 4 WLR 32 at [42(13)]; Samsung Electronics Co Ltd v LG Display Co Ltd [2022] EWCA Civ 423, per Males LJ at [4]. Where such interference is justified, the court may make its own evaluative assessment afresh.

The parties

[5]The twenty-four claimants in this action are all nationals of Nepal or Bangladesh (or in one case the personal representative of the estate of such a national), who at various times between 2011 and 2022 were recruited from their home country and transported as migrant workers to two factories in Johor Bahru, a city at the southern end of the Malaysian Peninsular. Most worked at a factory operated by ATA Industrial (M) Sdn Bhd (‘ATA’); two worked at the factory of a sister company, Jabco Filter System Sdn Bhd (‘Jabco’). Nothing turns on the distinction between ATA and Jabco for the purposes of the present appeal and I will refer to them both simply as ‘ATA/Jabco’. The claimants lived in associated accommodation provided by ATA/Jabco nearby. ATA/Jabco are not parties to the action, but the defendants have indicated an intention to join them as necessary and proper parties should the appeal succeed.[6]All but six of the claimants have since returned to their native countries. Five remain in Malaysia. One is in the United Arab Emirates. All are extremely impoverished and speak little or no English or Malay.[7]The defendants are three companies within the Dyson group of technology and manufacturing companies, which has a worldwide turnover measured in billions of pounds. D1 and D2 were incorporated in England (in 1985 and 1991 respectively), and their principal place of business has been in England at all times since their incorporation. D1 and D2 operate the Dyson group website. The group website identifies D1 as the group’s UK trading company. D1 and D2 have at all material times operated from Dyson’s UK office in Malmesbury, Wiltshire, which was the headquarters for the whole Dyson group until late 2019 (i.e. for the majority of the period covered by the claimants' claims). In late 2019, the Dyson group chose to move its corporate head office to Singapore. However, the evidence suggests that the Dyson UK office remains the primary operational control centre for the Dyson group with at least 3,500 employees based there including most of the group’s senior management team and the key senior management and operational staff relevant to these claims. The UK office hosts a UK based sustainability team whose members are responsible amongst other things for developing and promulgating mandatory policies and standards to be observed in the supply chain for the Dyson group as a whole; implementing those standards and policies across the group; and monitoring, auditing, and ensuring compliance with those standards and policies. The group’s chief legal officer is employed in the UK by D2.[8]D3 is a company incorporated and domiciled in Malaysia. It was D3 which contracted with ATA/Jabco’s parent company for the manufacture of the Dyson components and products at the factories at which the claimants worked. The evidence did not reveal the size of D3 or the scale of its commercial activity, nor its corporate relationship with D1 and D2, save that it was said to be a “sister company”.

The claims

[9]The claims are set out in detail in the Particulars of Claim, the body of which comprises allegations referable to all claimants, followed by a Schedule setting out the facts and losses specific to each. The claims were summarised at [4] to [13] of the Judgment, from which I gratefully borrow with some further detail.[10]It is alleged that each of the claimants was recruited to work at the factories by recruitment brokers or agents working for ATA/Jabco which left them subject to debt bondage. They were forced to work substantial overtime, above their 12-hour shifts, in breach of section 60 of the Malaysian Employment Act 1955 (‘the 1955 Act’). They were refused annual leave, contrary to sections 60D and 60E of the 1955 Act. They were not paid the legal minimum wage, contrary to various Minimum Wage Orders. They were subjected to onerous production targets, and placed under considerable pressure to meet those targets and frequently punished if they failed to do so, including by way of intimidation and physical violence. They suffered unlawful deduction of wages contrary to section 20 of the 1955 Act.[11]They were required to live at ATA/Jabco accommodation, which was invariably insanitary, overcrowded and degrading. They were not able to leave the accommodation at will, and were forced to hand over their passports to ATA/Jabco personnel. It is alleged that although ATA/Jabco assured the claimants that their visas would be renewed, two of them were arrested and detained on the grounds that they did not have a permit. It is also alleged that one claimant, Mr. Limbu, who sought to expose the abusive working and living conditions to an assistant of Mr. Andy Hall, a British specialist in human and migrant rights, was arrested and assaulted by the Malaysian police when his provision of evidence came to light. It is alleged that this was facilitated by representatives of ATA/Jabco, and that a representative of ATA/Jabco threatened and intimidated Mr. Limbu into signing a statement saying that he had received a large sum of money from Mr. Hall for providing his information. It is alleged that another Claimant (Mr. Hossain), who had taken photographs of the factory and living conditions was also arrested and interrogated by police, having been taken there by ATA/Jabco personnel.[12]The defendants are alleged to exert a high degree of control over the manufacturing operations and working conditions at ATA/Jabco’s factories. The defendants are also alleged to have promulgated mandatory policies and standards concerning the working and living conditions of workers in the Dyson group’s supply chain, and implemented them through processes of training, supervision (including regular audits) and enforcement. The policy documents promulgated are identified as the following detailed documents issued by D1 and/or D2:(1) the Dyson Ethical and Environmental Code of Conduct, which prohibits forced labour, bonded labour, slavery, trafficking and withholding of personal ID documents, and imposes standards for working hours, wages, health and safety, and worker accommodation, which suppliers must establish management systems and processes to see are observed;(2) the Dyson Modern Slavery and Human Trafficking Statement 2020, promulgated on behalf of all Dyson group companies, which provides for risk assessments and audits to be carried out to ensure compliance within the supply chain; and sets out remediation mechanisms for non-compliance;(3) the Supply Chain Foreign Migrant Worker Recruitment and Employment Policy, which sets out minimum requirements for the recruitment and treatment of migrant workers by Dyson suppliers; and(4) the Dyson Supplier Accommodation Standard containing mandatory minimum standards and practical guidance in relation to accommodation for workers in the supply chain.[13]D1 and D2 are said to be responsible for promulgation and implementation of those policies based on a number of factors: their issue by D1 and/or D2; the inclusion in the Modern Slavery Statement to a specialist supply chain team based in part in the UK; and the advertisement for, and employment of, sustainability personnel by D1 and/or D2 with specific responsibilities for implementing, enforcing, monitoring and reporting on the supply chain policies. D3 is said to be responsible for promulgating and implementing the policies. This allegation against D3 is primarily advanced on the basis of the response by Baker & McKenzie LLP, the solicitors for D1 and D2, to letters before action, in which D1 and D2 alleged that D3 was responsible for setting the standards for worker welfare in supply chains in South East Asia, including ATA, and for implementing them by various means including audits and training in Malaysia. This reflects the evidence of the claimants’ solicitor, Mr Holland, which was that the claimants intended to bring the claim against Dyson UK alone until these letters seeking to invoke D3’s role, and that it was on the basis of this response that they sought to join D3 as a necessary and proper party. The claimants also rely on the fact that D1 and D2 advertised for four members of a sustainability team to be based at Dyson Malaysia.[14]As to knowledge, it is alleged that the defendants knew of the high risk of forced labour in Malaysian manufacturing from public sources; they must have known of the practices at ATA/Jabco through implementation of their audit and monitoring policies; and that Dyson UK were specifically told of the mistreatment at ATA/Jabco in a series of communications from Mr Hall from 9 August 2019.[15]On 25 November 2021, Dyson terminated its contractual relationship with ATA. In pre-action correspondence Baker & McKenzie explained that some of Mr Hall’s complaints about working and living conditions had been substantiated.[16]The causes of action advanced against the defendants are threefold:(1) in negligence for breach of a duty of care to take reasonable and effective steps to ensure that the claimants did not suffer the economic loss and personal injuries resulting from the abuse;(2) liability in tort for false imprisonment (by ATA/Jabco at the factories and living accommodation), intimidation (by ATA/Jabco at the factories) and assault (by threats of violence and actual violence by ATA/Jabco at the factories and in the case of four claimants by the Royal Malaysian Police (‘RMP’)); ATA/Jabco or RMP are labelled ‘the Primary Tortfeasors’;(3) a restitutionary claim for unjust enrichment.[17]Each of the causes of action is governed by Malaysian law, which with one exception is said to be the same as that in England and Wales (I shall return to the extent to which there are issues about this).[18]As to the negligence claim, the circumstances capable of giving rise to parent company liability for harm suffered by individuals as a result of a foreign subsidiary’s operations, which it is said can extend to harm caused by the subsidiary’s supply chain contractor, include what have been referred to as Vedanta routes 2, 3 and 4, being three of the non-exhaustive sets of circumstances in which the Supreme Court in HRH Emere Godwin Bebe Okpabi v Royal Dutch Shell Plc [2021] UKSC 3 [2021] 1 WLR 1294 held that such parent liability may exist. They are where the parent has provided defective advice and/or promulgated defective group-wide safety/environmental policies which were implemented as of course by the subsidiary; has promulgated group-wide safety/environmental policies and taken active steps to ensure their implementation; and has held out that it exercises a particular degree of supervision and control over its subsidiary, even if it in fact does not do so (see Okpabi at [26]-[27]). The breach is said to consist of(a) defects in the policies which contain systemic errors preventing their purported objective being carried out, and(b) failure to take steps to ensure implementation and enforcement of the policies.[19]As to the other tort claims, the joint liability of the defendants for the commission of those torts by the Primary Tortfeasors is said to arise because the defendants(a) assisted in the commission of the torts pursuant to a common design with the Primary Tortfeasors to commit the tortious acts and/or(b) procured the commission of those torts by inducement, incitement, or persuasion of the Primary Tortfeasors. The conduct said to amount to such assistance, inducement, incitement or persuasion overlaps with that relied upon for the negligence claim, namely that it is, or is to be inferred from, a failure to enforce the policies or take steps to prevent the abuse in the face of constructive or actual knowledge of it.[20]As to the unjust enrichment claim, the enrichment is said to have occurred in the first instance in Malaysia by the saving of expenditure on products which, but for the abuse, would have cost more; and the receipt of the manufacturing services performed by the claimants at the factories. The claims are for the enrichment of the defendants through the consequential increased trading profits, which will have been enjoyed at their places of business.[21]As to remedies, the claimants claim damages which include quantified economic losses such as for insufficient wages; damages for deprivation of liberty and personal injuries; and aggravated and/or exemplary damages. In relation to the unjust enrichment claim there are personal and proprietary remedies sought in respect of the profits made as a result of the abuse.

The law applicable to the applications and the appeal

[22]There was very little dispute between the parties as to the applicable legal principles. In a service in case, the burden is on the defendant to show that there is another available forum which is clearly and distinctly more appropriate. The burden reflects the fact that in such a case the claimant has served the defendant as of right which is an advantage which will not lightly be disturbed (Spiliada at p. 476F, 477E). In a service out case, the burden is on the claimant to show that England is clearly the appropriate forum, which is simply the obverse of the position in a service in case (Spiliada at p. 481E). In both cases appropriate forum means that in which the case may be tried more suitably for the interests of all the parties and the ends of justice (Spiliada p. 476C, Lungowe vVedanta Resources Plc [2019]UKSC 20 [2020] AC 1045at[66]). In determining the appropriateness of the forum, the court looks at connecting factors to determine with which forum the action has the most real and substantial connection (Spiliada at p. 478A). These include not only factors affecting convenience or expense, but also other factors such as governing law, the place where the parties reside or carry on business, and where the wrongful acts and harm occurred (Spiliada p. 478A-B, Vedanta at [66]). The risk of multiplicity of proceedings giving rise to a risk of inconsistent judgments is only one factor, although a very important one (Vedanta at [69]). In applying these connecting factors to cases involving multiple defendants, their relative status and importance in the case should be taken into account, such that greater weight is given to the claims against those who may be described as a principal or major party or chief protagonist: JSC BTA Bank v Granton Trade Limited [2010] EWHC 2577 (Comm) per Christopher Clarke J at [28].[23]For both service in and service out cases, if the court concludes that the foreign court is more appropriate by reference to connecting factors, applying the relevant burden of proof, the court will nevertheless retain jurisdiction if the claimant can show by cogent evidence that there is a real risk that it will not be able to obtain substantial justice in the appropriate foreign jurisdiction (Vedanta at [88]). Cogent evidence does not mean unchallenged evidence (Vedanta at [96]). This is often conveniently treated as a second stage in the analysis because it usually calls for an assessment of different evidence, but it does not involve a different question: if there is a real risk of denial of justice in a particular forum it is unlikely to be an appropriate one in which the case can most suitably be tried in the interests of the parties and for the ends of justice: Vedanta at [88]. In this case the parties and the Judge adopted that two-stage approach, labelling the first stage as “appropriate forum” and the second stage as “access to justice”. I will adopt the same structure, whilst keeping in mind that second stage factors may also be relevant to the first stage in what is juridically a single holistic exercise in seeking to identify where the case can most suitably be tried in the interests of the parties and for the ends of justice.

The defamation proceedings

[24]Also relevant are defamation proceedings brought by D1 and D2 in the High Court. Those proceedings are accurately described at [19] to [26] of the Judgment, from which I again gratefully borrow in giving a less detailed summary of the position at the time of the hearing before the Judge. I shall come to subsequent developments in the defamation proceedings in the course of dealing with one of the grounds of appeal.[25]The defamation proceedings were commenced on 25 February 2022, some three months before the issue of the Claim Form in the present proceedings, by D1 and D2 (and originally by Sir James Dyson, the eponymous founder of the Dyson group) against the broadcasters of a Channel Four television programme aired on 10 February 2022. The broadcast related to abuse of workers at ATA factories in Malaysia generally, and that of Mr Limbu as a whistleblower, and raised questions over why it hadn’t been picked up by Dyson. D1 and D2 alleged that the natural and ordinary meaning was that they were complicit in the systemic abuse and exploitation of workers at ATA; that they were complicit in the persecution and torture of Mr Limbu; and that they claimed to act in a responsible and ethical way, but when serious abuses of workers were brought to their attention these abuses were not properly investigated, but were ignored and tolerated for a prolonged period of time while they tried to cover them up and shut down public criticism. The claim by D1 and D2 relied upon the promulgation and implementation of the policies which are identified in the current proceedings. At the time of the hearing before the Judge, a defence had not yet been served by the broadcasters, but in correspondence responding to a letter before action they had indicated that they would advance a defence that the imputations conveyed by the statements complained of were true or substantially true. It is not necessary to give further details of the claim to make the point, which was not in dispute, that there was a very substantial overlap between the factual issues raised in the defamation proceedings and those in the current proceedings, not merely in relation to the allegations of abuse but also in relation to D1 and D2’s knowledge and complicity, and the promulgation, implementation, supervision and enforcement of its policies.

The Judgment

[26]Before the Judge, the defendants offered various undertakings which were recorded in the order (‘the Undertakings’), including undertakings by D1 and D2 to submit to the jurisdiction in Malaysia if the claimants brought “these claims” there; for the purposes of the Undertakings “these claims” were defined as those particularised in the Particulars of Claim. Paragraph 3 of the Undertakings included the following:
“3. Each of the First, Second and Third Defendants has undertaken to the Court that, if any of the Claimants bring these claims against the Defendants (or any of them) in Malaysia, in respect of those claims: … c. The Defendants will pay the reasonable costs necessary to enable the Claimants to give evidence in Malaysian proceedings including (if necessary) affidavit affirmation fees and other costs necessary for the Claimants to give remote evidence including travel and accommodation costs, costs associated with the provision/set-up of suitable videoconferencing technology, translation fees, and other costs associated with the logistics of giving evidence remotely; … e. The Defendants will pay for the Claimants’ share of the following disbursements to the extent reasonably incurred and necessary: (i) Court interpretation fees, (ii) Transcription fees, and (iii) Joint expert evidence; … f. The Defendants will not seek to challenge the lawfulness of any success fee arrangement entered into between the Claimants and their Malaysian lawyers.”
[27]In addressing appropriate forum at the first stage, the Judge identified what he considered the key connecting factors under six numbered headings, and addressed each at [84] to [121], before summarising his conclusions at [122]-[123]. His headings and overall conclusions were as follows.(1) Practical convenience for the parties and the witnesses: this is a neutral factor between England and Malaysia.(2) A common language for the witnesses: this too is a neutral factor.(3) The system of law governing the claims: the fact that this is Malaysian law is a factor which strongly favours Malaysia.(4) The place where the “issues” in the case took place: this is a factor which strongly points towards Malaysia as the proper forum because the “centre of gravity in this case is plainly Malaysia”.(5) The location and availability of documents: this factor slightly favours Malaysia.(6) The risk of multiplicity of proceedings and inconsistent judgments: (a) so far as contribution claims against ATA/Jabco and RMP are concerned, this factor is a relatively slight one in favour of Malaysia: there is no real risk of inconsistent judgments in respect of contribution proceedings against ATA/Jabco; there is such a risk in relation to contribution proceedings against RMP (although no such claim has yet been intimated by Dyson) because the claim would not be justiciable in England, but this is a relatively minor part of the claim. (b) so far as the defamation proceedings are concerned, the multiplicity of proceedings and risk of inconsistent judgments if the claims proceeded in Malaysia was a significant factor in favour of England, but its weight was diminished by the fact that there was a real risk that that would also arise if the current claim proceeded in England, because it was most unlikely that the defamation proceedings and these proceedings would be case managed together or even with a real eye on one another.[28]The Judge summarised his stage 1 conclusion in the following terms at [122] (the emphasis is that of the Judge):
“Taking all of these factors into account, my conclusion at the end of Stage 1 is that England is not the natural or appropriate forum and that Malaysia is another available forum which is clearly and distinctly more appropriate. The centre of gravity in this case is Malaysia: that is where the primary underlying treatment about which the Claimants complain took place, and is therefore the forum with “the most real and substantial connection” per Lord Goff in Spiliada at 478A. Malaysian law is also the governing law, and there are good policy reasons for letting Malaysian judges consider the novel points of law that are being raised in this claim within the context of their jurisprudence, rather than letting an English Court second guess what they might decide. In my judgment, these factors are not “dwarfed” by countervailing factors (per Lord Mance in VTB). The risk of irreconcilable judgments resulting from the defamation proceedings is an important factor, but it does not tilt the balance in favour of the English Court being the proper forum to determine the Claimants’ claim.”
[29]In relation to stage 2, the Judge had a considerable body of detailed evidence from well-qualified deponents on both sides as to the practice and procedure in Malaysia, and the availability and funding of suitable representation. He resolved a number of issues by reference to the cogency of the conflicting evidence. The principal determinations which remain relevant to the issues in the appeal are the following. The claimants would have a reasonable and well-founded fear for their safety if they gave evidence in Malaysia and accordingly any trial in Malaysia would involve them giving evidence remotely. The Malaysian court would permit them to do so, and the cost of making arrangements for them to do so was covered by the Undertakings. There was no real risk that the claimants could not find suitably qualified lawyers who would be prepared to conduct the case pursuant to a conditional fee agreement (‘CFA’) with a small upfront basic fee of about 1000-1500 MYR (the equivalent of about £200). Central to the Judge’s reasoning in rejecting the evidence adduced by the claimants on this issue was his view that the deponents had not recognised that “substantial justice” does not require that a party receive the same ‘Rolls Royce’ service (or as the Judge preferred to call it, ‘Tesla’ service) as would be available here, and that it was therefore irrelevant that the representation available would be of a lesser standard. Contrary to the claimants’ argument, there was no real risk that such a partial CFA would be unlawful. There were certain disbursements for which funding would have to be provided otherwise than by the CFA lawyers themselves. These were largely covered by paragraphs 3(c) and (e) of the Undertakings. It was accepted that there was a real risk that there would be a shortfall. Based on(a) the defendants’ model of how the proceedings would be case managed, involving a split trial with three test cases and three contested disclosure applications, and(b) the defendants’ schedule of disbursements not covered by their undertakings, the shortfall which would need to be funded from elsewhere was 1,916 MYR (£328). It was accepted on behalf of the defendants that the claimants themselves could not afford to make any financial contribution to the pursuit of a claim in Malaysia. The small shortfall would be made up by NGOs and there was no real risk that they would not do so. So far as the Undertakings were concerned, there was nothing improper in the defendants offering them: the giving of undertakings is not uncommon in jurisdictional disputes, and there was no conflict of interest in the Undertakings relating to the costs of pursuing the claims against them. Accordingly the claimants’ stage 2 argument failed.

The grounds of appeal

[30]The claimants advanced five grounds of appeal in respect of stage 1 and four grounds in respect of stage 2: Ground 1(a) The Judge failed to adopt the correct approach to the applications by D1 and D2, which was to address the factors relevant to the issue of appropriate forum in respect of the claim against them separately from, and before, his consideration of the factors relevant to the appropriate forum in respect of the claim against D3; and as a result he failed to take any (or any sufficient) account of the key connecting feature that D1 and D2 are domiciled in England and have been served here as of right. Ground 1(b) In considering “the centre of gravity” of the case the Judge failed to have any (or any sufficient) regard to the fact that the real dispute between the parties (and hence the likely focus of any trial) is principally concerned with the role, acts and alleged breaches of duty and enrichment in England of D1 and D2. Ground 1(c) The Judge erred in law in his treatment of the Malaysian law factor by taking into account “policy reasons” for letting Malaysian judges decide such questions, contrary to Lubbe v Cape Plc [2000] 1 WLR 1545. Ground 1(d) The Judge was plainly wrong to consider that there was a real risk of irreconcilable findings in relation to the defamation proceedings even if the current proceedings proceed in England on the basis that it was most unlikely that the High Court would case manage the proceedings to avoid or reduce the risk of such a possibility. Ground 1(e) The Judge erroneously failed to have any regard to the uncontested fact that the defendants’ defence of the claims would be coordinated and conducted from England by English employees and officers of D1 and D2. Ground 2(a) The Judge was wrong to conclude that there was no real risk the Claimants and NGOs would be unable to fund the disbursements necessary to pursue their case in Malaysia. In particular the Judge was wrong(i) to rely on the Undertakings;(ii) to accept the defendants’ calculations of the amounts of the necessary disbursements; and(iii) to conclude that there was no real risk of NGOs failing to fill any gap. Ground 2(b) The Judge was wrong to conclude that the claimants would be able to obtain representation from suitably qualified and resourced lawyers in Malaysia under a partial CFA because there is at least a real risk that (i) the proposed CFA arrangement would be unlawful; and (ii) the claimants would be unable to instruct suitably qualified and resourced lawyers prepared to represent them on such a basis. Ground 2(c) The Judge wrongly failed to have any regard to the fact that the claimants’ inability to attend a trial in person in Malaysia presents a real risk of injustice to the Claimants irrespective of the possibility of remote hearings. Ground 2(d) The Judge wrongly failed to have any regard to the real risk that the claimants will not be able to obtain disclosure of documents in Malaysia which are essential to establishing their case against the defendants.

Discussion

[31]I start with the question whether the Judge made an error of principle or was plainly wrong in the sense necessary before this court will interfere with his evaluative assessment. I will take each ground in turn. Ground 1(a)The Judge failed to address the factors relevant to the issue of appropriate forum in respect of the claim against them separately from, and before, his consideration of the factors relevant to the appropriate forum in respect of the claim against D3; and as a result he failed to take any (or any sufficient) account of the key connecting feature that D1 and D2 are domiciled in England and have been served here as of right.[32]Ms Kinsler KC treated the two parts of the ground as separate points, with the second arising independently of the first as well as consequentially.[33]I would reject the first part of the argument. In a mixed case such as the present, the court has to look holistically and in the round at the question of appropriate forum for both the service in and service out applications. That is not, as Ms Kinsler submitted, to ignore the burden of proof in the service in application, which is the obverse of the burden in the service out application (Spiliada at p. p481E); nor is it to elevate the risk of conflicting judgments to the status of a decisive factor when it is only one factor, albeit a very important factor. It is because the exercise is to identify where the case can most suitably be tried in the interests of the parties and for the ends of justice, as Lord Goff formulated the basic principle in Spiliada at p. 476C and as Lord Briggs emphasised in Vedanta at [68]:
“ … the court is looking for a single jurisdiction in which the claims against all of the defendants may most suitably be tried.”
Moreover it will usually be impractical to compartmentalise the two applications in the way Ms Kinsler suggested. If the stay application by D1 and D2 were to be taken first and separately, it would not be capable of resolution on its own, because one very important factor is multiplicity of proceedings and the risk of inconsistent judgments, which simply could not be applied whilst it was unresolved whether the application by D3 would result in the claims against it proceeding in England or Malaysia.[34]However, I would accept the second part of the argument, that the Judge failed to take any account of the important connecting feature that D1 and D2 are domiciled in England and have been served here as of right. The domicile of the parties was not one of the Judge’s headings and did not feature in his conclusory paragraphs. It is, however, an important factor. The reason it is an important connecting factor in relation to jurisdiction is because presence here is the basis for establishing the court’s jurisdiction, and domicile here connotes a degree of permanence and allegiance to the country’s institutions, including its courts, which means that the party can reasonably expect, and be expected, to meet claims against it in such courts in the absence of sufficient countervailing factors. That is why within the EU domicile remains the foundational factor for allocating jurisdiction in civil and commercial matters, subject to derogations. The importance of presence or domicile is at the heart of the difference in the burden of proof between service in and service out cases. In the latter case the assertion of jurisdiction is prima facie “exorbitant”, whereas in the latter it is prima facie “as of right”. That is why, as Lord Goff emphasised in Spiliada at pp. 476F, 477E, the burden in a service in case is on the defendant to point to a distinctly and clearly more appropriate forum, because the advantage to a claimant of pursuing a defendant in his place of domicile will not lightly be disturbed.[35]Mr Gibson KC advanced two arguments in response. The first was that the Judge had taken this into account. He had recorded the reliance on Dyson UK’s domicile when summarising the claimants’ submissions, and had said elsewhere that he had taken all the submissions into account. I cannot accept this argument. Had the Judge attached any significance to Dyson UK’s domicile as a factor he would have been bound to have said so in the course of his lengthy and detailed reasoning and his conclusions. He did not do so, and it did not fall within any of his headings.[36]Mr Gibson’s second argument was that the domicile of D3 in Malaysia provides an argument of equal and opposite weight to that of Dyson UK’s domicile in England, which neutralises domicile as a factor pointing towards England any more than towards Malaysia. However, the logic of the point founders on the fact that the claims against Dyson UK and Dyson Malaysia do not fall to be treated as of equal importance in this regard. The claim against D1 and D2 is the primary claim. The claim was intended to be pursued solely against Dyson UK. D3 was only added as a defendant in response to Dyson UK’s pre-action correspondence. Although the claim against Dyson Malaysia in many ways mirrors that against the Dyson UK defendants, that is solely on the basis of D1 and D2’s assertion that D3 was responsible for promulgating and implementing the policies, although the relevant policy documents which have been identified, not only by the claimants, but also by the defendants themselves in correspondence and in the defamation proceedings, are those promulgated by D1 and D2, not D3. D3 is a group company for whom the litigation will be coordinated and conducted, whether it takes place in England or Malaysia, by the English officers and employees of Dyson UK from England, where the group chief legal officer is based. This is common ground. The reality is that Dyson UK is the principal protagonist and Dyson Malaysia a more minor and ancillary defendant to the claim against D1 and D2. I would therefore conclude that the Judge’s failure to take into account the domicile of D1 and D2 in England as a factor in favour of retaining English jurisdiction is an error of principle. Ground 1(b) In considering “the centre of gravity” of the case the Judge failed to have any (or any sufficient) regard to the fact that the real dispute between the parties (and hence the likely focus of any trial) is principally concerned with the role, acts and alleged breaches of duty and enrichment in England of D1 and D2.[37]The “centre of gravity” is not itself a separate test and was used by the Judge as a heading under which to address various links between the issues in the case and England or Malaysia. It was under this heading that he addressed the location of conduct or events relevant to issues of duty, breach, harm and remedy, insofar as he addressed them at all.[38]In his conclusory paragraph [122] the Judge treated the sole reason why the centre of gravity was Malaysia as being that that was where there occurred the primary underlying abusive treatment about which the claimants complain. In the discussion under his fourth heading at paragraphs [102] to [104], in which he identified the “centre of gravity”, he referred to this conduct as fundamental to the claim, although he also mentioned in this context at [102] that Malaysia was where the harm occurred. This one-sided approach failed to take account of a number of factors which in the Judge’s taxonomy fell to be addressed under this heading of centre of gravity. The case against Dyson UK was the primary claim and it was necessary to focus particularly on the issues which would arise in relation to that claim. The promulgation of the policies took place in England and their relevance to the Vedanta routes to whether there existed a duty of care in negligence, including the allegation that they are flawed, points towards England. The allegation of breach by Dyson UK in failing to take steps to see that the policies were implemented in Malaysia, and failing to respond adequately to what was or ought to have been known about the abuse, which is at the heart of the allegations of breach for both the negligence and other tort claims, is an allegation of a failure occurring amongst the management in England and is alleged primarily to have occurred in England, although it will also focus to some extent on conduct in Malaysia. The complaints made by Mr Hall were made to Dyson UK and the alleged failure to take steps to act on them is primarily a failure of English personnel in England. The unjust enrichment of D1 and D2 ultimately took effect in England at their centre of trading, and the proprietary remedies claimed are of property rights over profits and products located in this country.[39]As Mr Gibson submitted, there are undoubtedly features under this heading which point to Malaysia. The suffering of loss and damage by the claimants is one; the existence and terms of audits in Malaysia is another; so too is the place where the alleged abuse occurred, as the Judge identified, although the Judge failed to recognise that there was at least a real prospect of the underlying abuse not being substantially in issue. The defendants envisage, in whichever jurisdiction, test cases for three claimants, whose treatment is likely to be influential on the outcome for all; that is the purpose of test cases. In pre-action correspondence, by which time D1 and D2 purported to have investigated the underlying allegations at least to the extent reported by Mr Hall, the focus of the response was not on the truth or otherwise of those allegations, but rather the extent to which any liability could attach to Dyson UK. A letter from solicitors acting for the Channel Four broadcasters dated 9 March 2022 recorded that Dyson UK had accepted Mr Limbu’s account regarding poor working conditions and allegations of ATA corruption; that they had been provided with details of his account, including by Baker & McKenzie actually interviewing him in October 2021; and that at no time had D1 or D2 sought to contradict his account. It seems to me likely that Mr Limbu would be one of the lead/test claimants. The Judge was right to observe that the defendants had not admitted the underlying allegations of abuse and that he could not assume that they would. But nor could he safely assume that they would form the fundamental issue in the test cases or the proceedings as a whole. It was more likely that the main focus of the trial would be where it had been put by D1 and D2 in pre-action correspondence, both in these proceedings and the defamation proceedings, namely on Dyson UK’s role and activity in England.[40]The failure by the Judge to take into account these other aspects of the location of conduct or events relevant to duty, breach, harm and remedy, led to a mistaken assessment of the centre of gravity which he premised, primarily, just on the location of the alleged underlying abuse. Had he had taken all these other matters into account he would have been bound to treat the centre of gravity, in the sense used by him, as pointing towards England, or at least as no more than neutral. This was, in my view, a further error of principle. Ground 1(c) The Judge erred in law in his treatment of the Malaysian law factor by taking into account “policy reasons” for letting Malaysian judges decide such questions, contrary to Lubbe v Cape.[41]I would reject the argument that the Judge made any error of principle on this ground. In Lubbe v Cape, Lord Bingham at p. 1561E-G and Lord Hope at p. 1567A-D made clear that resort could not be had to general public policy or public interest arguments which were outwith the scope of the Spiliada principles because they were not related to the private interests of the parties or the ends of justice in the particular case. The Judge’s reference to “policy reasons” in the context of the desirability of a Malaysian court deciding novel points of Malaysian law was not resorting to any general public policy arguments of that kind. He was saying no more than that there were good reasons for issues of foreign law to be decided by the foreign court, which is the underlying policy behind the relevant system of law being a relevant connecting factor in the forum non conveniens assessment: see Lord Mance in VTB Capital Plc v Nutritek International Corpn [2013] UKSC 5 [2013] 2 AC 337 at [46]. Ground 1(d) The Judge was plainly wrong to consider that there was a real risk of irreconcilable findings in relation to the defamation proceedings even if the current proceedings proceed in England on the basis that it was most unlikely that the High Court would case manage the proceedings to avoid or reduce the risk of such a possibility.[42]I agree that in this the Judge made a serious error of principle and was plainly wrong. Group litigation in this country, whether subject to a Group Litigation Order or not, is subject to particularly extensive and flexible case management, which may involve not only split trials and lead cases, but the determination separately of particular sets of issues (see Municipio de Mariana v BHP Group (UK) Ltd [2022] EWCA Civ 951 [2022] 1 WLR 4691 at [134]-[142]). Similarly, defamation proceedings are carefully case managed and rarely proceed directly from claim form to full trial without identification of particular issues as suitable for separate resolution, such as the meaning of the words complained of. I would find it very surprising if, in circumstances where the defamation proceedings and these proceedings were both taking place in the King’s Bench Division of the High Court in England at the same time, the court did not take steps to coordinate case management in the two cases with a view to minimising the risk of factual issues being tried twice with different evidence or argument so as to duplicate court time to the detriment of other court users, and give rise to a risk of conflicting decisions. Far from it being most unlikely that such coordination would take place, I would regard it as overwhelmingly likely, and that there would be a substantial likelihood that it would succeed in avoiding, or at least very much reducing, duplication of proceedings and the risk of inconsistent judgments. My conclusion on this question is reinforced by the concurring judgment of Warby LJ on this question whose experience in defamation cases far exceeds my own.[43]Mr Gibson’s first response is that we should disregard this ground of appeal because it has been overtaken by events. The defamation proceedings came to an end with the recent notices of discontinuance filed by D1 and D2 on 23 August 2024. That occurred after the broadcasters had served a defence in those proceedings advancing a defence of truth, and shortly after an order of Warby LJ that the defence could be relied on by the claimants in this appeal. No reason has been given by Dyson UK for such discontinuance. The timing and history is such that a cynical observer might infer that this was in whole or in part to obtain tactical advantage in this appeal. Mr Gibson suggests, however, that there are a number of legitimate reasons why D1 and D2 might now have decided to discontinue which are unrelated to the effect it might have on this appeal, and I will proceed on that basis.[44]The discontinuance does not, however, render the defamation proceedings irrelevant. The normal approach on an appeal by way of review is to consider whether the decision of the lower court was or was not wrong on the basis of the evidence adduced, and the submissions advanced, in the lower court: Sharab v HRH Prince Al-Waleed Bin Talal Bin Abdal-Azh Al-Saud [2009] EWCA Civ 353 per Richards LJ at [52]; Samsung Electronics per Males LJ at [1] and [5]. That remains the approach which we should adopt in determining whether the decision of the Judge involved an error of principle or was plainly wrong such that we may consider the evaluative assessment afresh in this court. If we do so conclude, and come to exercise the evaluative assessment for ourselves, it was common ground that we should do so on the basis of the current position, which is that the defamation proceedings will not be pursued. In determining whether the Judge made an error of principle or was plainly wrong, however, we must look at the position which existed at the hearing before him, and take the defamation proceedings into account. Mr Gibson submitted that the Court had already ruled to the contrary by reason of the Order of Andrews LJ on 22 May 2024 permitting the claimants to rely on a supplementary note in which they relied on the defence of truth having been pleaded in the broadcasters’ defence, when she observed that it was inappropriate to seek to preclude this court from being informed of relevant developments in the defamation proceedings; and the Order of Warby LJ of 16 August 2024 permitting the broadcasters’ amended defence to be adduced in evidence on the appeal and echoing what Andrews LJ had said about it being desirable that the court should be in the best position to decide for itself the extent of overlap between the proceedings. However, those orders and remarks did not purport to address the current issue, and are entirely consistent with the material being made available to the court because it would be relevant material if the court should conclude that it is required to conduct the evaluative assessment afresh for itself.[45]In the alternative, Mr Gibson sought to persuade us that the Judge had been right to treat the prospect of avoiding a duplication of proceedings and risk of inconsistent findings as most unlikely. The argument was maintained with panache, but I found it unconvincing not least because it assumed that the likely case management of both these proceedings and the defamation proceedings would be approached in an improbably inflexible way.[46]I would therefore conclude that this ground identifies a further error of principle by the Judge. Ground 1(e) The Judge erroneously failed to have any regard to the uncontested fact that the defendants’ defence of the claims would be coordinated and conducted from England by English employees and officers of D1 and D2.[47]The Judge recorded this as a matter upon which the claimants relied, and it is clearly a factor of some relevance as a connecting factor generally, as well as casting light on the fact that Dyson UK is the primary defendant. Mr Gibson argued that the Judge took this into account at [94] in the context of a discussion of the giving of instructions to legal representatives under the heading “practical convenience”. I do not consider that paragraph [94] can be read as doing so. The Judge did not mention this factor in that paragraph and it is not just a matter of “practical convenience” in giving instructions. The fact that litigation will be coordinated and conducted from one of the two rival fora, irrespective of the forum in which the litigation takes place, is a significant connecting factor with that forum. A further indication that the Judge did not have this point in mind at [94] is that he regarded the element as neutral between the different fora given the location of the witnesses and parties.[48]This too, therefore, was an error of principle. Ground 2(a) The Judge was wrong to conclude that there was no real risk the Claimants and NGOs would be unable to fund the disbursements necessary to pursue their case in Malaysia. In particular the Judge was wrong(i) to rely on the Undertakings;(ii) to accept the defendants’ calculations of the amounts of the necessary disbursements; and(iii) to conclude that there was no real risk of NGOs failing to fill any gap.[49]I start with the Undertakings. In the experience of the court they are unprecedented, and the researches of counsel have not identified anything similar (we were referred to Société Nationale Industrielle Aerospatiale v Lee Kui JAK [1987] AC 871, an anti-suit injunction case, in which the undertakings were not remotely comparable). As a mechanism for ensuring that the impoverished claimants are thereby enabled to meet disbursements necessary to conduct the claims in Malaysia, they seem to me to suffer from six serious flaws.[50]First, the Judge was wrong to suggest that they did not involve a conflict of interest. The interest of the claimants is that costs reasonably incurred in having to make disbursements of the kind identified in paragraph 3c and 3e of the Undertakings are funded in advance. The interest of the defendants is to take every step legitimately available to them to defeat the claims. Their conduct of the litigation to date fully justifies the Judge’s assessment that they will be “tough” and “difficult” opponents. Mr Gibson accepted that Dyson would take all steps available to it to defeat the claim provided they were lawful and ethical. Mr Gibson suggested that it would not be in the interests of Dyson for the claimants to be underfunded because that would enable the Claimants to come back to England and lift the stay. To my mind that was wholly unrealistic. Contested applications for the lifting of the stay in England would not be an appropriate way of resolving disputes arising on particular facts about the scope or application of the Undertakings in Malaysia. A dispute shortly before the trial, for example, about whether travel costs for a claimant to give evidence remotely were or were not reasonably necessary, would not allow justice to be done by requiring the claimant to pursue a contested application to lift the stay in England, for which in any event they would have no funding.[51]Nor did I find any more realistic Mr Gibson’s submission that the defendants could be expected to abide by “the spirit” of the Undertakings. Undertakings mean what they say and anyone with experience of commercial litigation knows that such undertakings are carefully crafted to define their extent. Dyson would be entitled to object if any request was outside the scope of the Undertakings on their proper construction, and it would be a legitimate pursuit of their interests to insist upon the letter of their Undertakings. They could not be expected to comply with some ill-defined “spirit” of the Undertakings which was not covered by their wording.[52]Secondly, disputes about whether costs were reasonable and necessary (it is to be noted that paragraph 3c and 3e of the Undertakings requires both) would need to be resolved in advance, and the claimants would be delayed in being able to make the disbursement until any such dispute were resolved. This itself would provide Dyson with a tactical advantage.[53]Thirdly, requests for disbursements in advance would likely require the claimants to waive legal professional privilege if reasonableness and/or necessity were to be challenged and investigated. This would be grossly unfair and give Dyson an improper litigation advantage. Mr Gibson sought to meet this difficulty by suggesting that if privilege were claimed, Dyson would simply have to pay whatever was sought however unreasonable or unnecessary it might be if it were investigated. That is not what the Undertakings say.[54]Fourthly, there is no satisfactory mechanism for resolving disputes as to what is reasonable or necessary. Mr Gibson submitted that the matter could be resolved by the Malaysian court. It is difficult to see why this is so: the Undertakings are given to the English court (and there has been no undertaking to give the same undertakings to the Malaysian court). There is no Malaysian law evidence suggesting that the Malaysian court would adjudicate upon undertakings given by a foreign defendant to a foreign court. It is obviously unsatisfactory for the English court to have to police the conduct of Malaysian proceedings. But in any event, whether such issues fall to be resolved in England or Malaysia, it is probable that the claimants will not have the means to fund the resolution of any such dispute. The premise on which the Undertakings were given, in relation to the specific disbursements they cover, was that such disbursements would not be funded by Malaysian lawyers acting for the claimants under a partial CFA. If so, there is no reason to suppose that such lawyers would fund the costs of disputes about the nature or extent of those disbursements being reasonable or necessary. In England too there would be no one to fund them. This again reveals that the conflict of interest affords the defendants an unfair litigation advantage.[55]Fifthly, the disbursements which are identified as the subject matter of the Undertakings are those which the defendants have identified as arising on their own case management model. This is really the gravamen of aspect (ii) of this ground. However, the proceedings are at an early stage and it is impossible to predict with confidence how case management will shape the procedural development of the case, or to predict with any certainty what costs and disbursements will be required. Mr Craven, who presented the oral argument on the Stage 2 grounds on behalf of the claimants with conspicuous skill, argued that the number of test cases, or disclosure applications, was a matter of speculation. Mr Gibson’s response to this was that these costs were to be borne by the lawyers conducting the case on the partial CFA. But what of disbursements, akin to those covered by the Undertakings, which will or might not be so covered? For example, paragraph 3e covers expert evidence only if it is joint expert evidence. Forensic accounting is one of the expert disciplines identified as necessary for the unjust enrichment claim, and it was common ground that it would be required for the purposes of the first part of a split trial because it goes to liability. It cannot be certain that a joint expert report will be ordered; and even if it is, why should the claimants not have the opportunity to take their own expert advice, as is common where a joint expert is appointed, and which Dyson with its ample resources will be able to do? That is in any event likely to be reasonably necessary for the claimants in order to plead out their particulars of the unjust enrichment. The Judge recognised that there might be the need for such expert evidence but doubted that it would need to be “extensive” and said that forensic accounting issues would not be very complex, being limited to the level of profit Dyson earned from the labour of the claimants and what a reasonable level of profit would have been. I am less sanguine about the scale and expense of the exercise; for my part I consider that there is a real risk that such an exercise undertaken by forensic accountants would be both complex and expensive.[56]The point, shortly put, is that confining the Undertakings to the disbursements which are currently identified as outside the scope of lawyer funded disbursements, on a present prediction of case management, simply does not take account of contingencies which are almost inevitable in litigation of this kind.[57]That leads on to the sixth flaw. The Undertakings are confined to disbursements in relation to the claims as currently particularised (“these claims”). It is highly likely that there will be amendments given that in a case of this kind the importance of disclosure of documents internal to Dyson is, as Lord Briggs JSC put it in Vedanta at [44], blindingly obvious. It would obviously be unsatisfactory and unfair if disbursements are only covered to the extent they fall within the current version of the Particulars of Claim which has of necessity been drafted on the basis of what the defendants have so far chosen to reveal and without sight of what will no doubt be many relevant documents; yet that is the extent of the Undertakings. And what if new claimants emerge who wish to join the group? If they are not covered by the Undertakings at all, which they are not, must they start their own separate proceedings here?[58]For all these reasons, the Judge was in my view plainly wrong to say that there was no conflict of interest, and to treat the Undertakings as a satisfactory mechanism to meet the disbursements which they covered, let alone other disbursements which the claimants might need to fund as the proceedings developed. Ground 2(b) The Judge was wrong to conclude that the claimants would be able to obtain representation from suitably qualified and resourced lawyers in Malaysia under a partial CFA because there is at least a real risk that(i) the proposed CFA arrangement would be unlawful; and(ii) the claimants would be unable to instruct suitably qualified and resourced lawyers prepared to represent them on such a basis.[59]I would accept Mr Gibson’s submissions that the Judge was entitled to reach the conclusions which he expressed on these issues, for the reasons he gave, as part of his evaluative assessment of the evidence. Addressed as a stage 2 issue, there is no basis for interfering with the Judge’s conclusion. However there is an aspect of the Judgment on these matters which is relevant to the overall assessment of the appropriate forum in which the case may most suitably be tried. As Mr Gibson was keen to emphasise, the Judge’s rejection of the evidence of the claimants’ witnesses was based on his conclusion that they failed to recognise that it would not amount to substantial injustice if the service provided by such lawyers was of a lesser standard than a ‘Tesla’ service. In a case such as the present, where there is a huge imbalance between the impoverished and vulnerable claimants and the well-resourced and commercially experienced defendants, and the allegations are of very serious human rights abuses, there is a particular need to ensure equality of arms in the conduct of litigation if justice is to be served. Indeed Mr Gibson accepted that equality of arms was a consideration. If the defendants can be expected to have the very high standard of legal service in Malaysia which their resources permit, but the claimants only a lesser standard, whereas in England the claimants will also be represented by experienced and well-resourced solicitors, as the evidence establishes that they will, that is a factor, although only one factor, which favours England as a more appropriate forum. This is not something which gives rise to a real risk of substantial injustice at Spiliada stage 2. However, it is a consideration in the overall assessment of the appropriate forum in which the case may most suitably be tried, because inequality of arms in one of the two fora is a factor pointing to the other as more appropriate. Ground 2(c) The Judge wrongly failed to have any regard to the fact that the claimants’ inability to attend a trial in person in Malaysia presents a real risk of injustice to the Claimants irrespective of the possibility of remote hearings.[60]Mr Craven did not suggest that the fact that witnesses are only able to give evidence remotely, or that parties or their representatives are only able to “attend” the trial remotely, is something which necessarily and of itself must involve substantial injustice. Rather, he submitted, it did so in the particular circumstances of this case because the vulnerable claimants would need to give evidence through translators, and have translators if they were to follow the proceedings; and it involved an inequality of arms in circumstances in which the representatives of the defendants would be able to attend in person and follow the proceedings; and their witnesses would be able to give evidence in person. I would accept that this is another inequality of arms factor which is relevant and favours England as a more appropriate forum. Ground 2(d) The Judge wrongly failed to have any regard to the real risk that the claimants will not be able to obtain disclosure of documents in Malaysia which are essential to establishing their case against the defendants.[61]Mr Craven did not press this point with any vigour and he was right not to do so. On the evidence before him the Judge cannot properly be criticised for treating disclosure from the defendants as equally available and effective in the Malaysian courts as in this jurisdiction.

Conclusion thus far

[62]For the above reasons I would treat the Judge as having made a number of errors of principle such that this court can and should exercise its own evaluation of the Spiliada factors and reach its own conclusion. This court’s evaluative assessment Funding

This court’s evaluative assessment

[63]It is accepted by the defendants that in order to pursue these claims in Malaysia there are certain disbursements which the claimants would have to fund from sources other than the lawyers prepared to act on their behalf under a partial CFA. Those disbursements are unlikely to be as limited as are envisaged by the Undertakings for the reasons I have explained, but even if they are, they are substantial. The evidence in relation to NGO funding suggests that they are not such as could reasonably be expected to be funded by NGOs. That is no doubt why the Undertakings were offered. The claimants are very poor and do not have the means to pay them, as was common ground. The Undertakings do not provide a satisfactory means of funding them, for the reasons I have identified. Accordingly the claimants will not be able to bring the claims in Malaysia. At the lowest there is a serious risk that that is the case. Whether this is addressed as a stage 1 or stage 2 issue, it points overwhelmingly in favour of England as the distinctly more appropriate forum. This is not in any sense a criticism of the Malaysian justice system, for which this court has the highest regard, but arises out of the particular and unusual features of the case.[64]That England is clearly and distinctly more appropriate is reinforced by a consideration of the other connecting factors.

The domicile of the parties

[65]This favours England, for the reasons I have already explained. It is the domicile of Dyson UK which is the principal protagonist, with Dyson Malaysia a more minor and ancillary defendant to the primary claim against D1 and D2.

Practical convenience

[66]As to documents, there are unlikely to be many, if any, held by the claimants themselves. There may be some documents in Malaysia relevant to the alleged abuse, but by its very nature such abuse is unlikely to be documented. There will be some relevant documentation in Malaysia evidencing the claimants’ employment and wages. The bulk of the documentation in the case is likely to be that relevant to the central issue of responsibility for the alleged abuse in the supply chain. That will predominantly be located at Dyson UK, where the policies were devised and promulgated and where the alleged failures to implement them will have taken place. So too will be documents relevant to what Dyson UK knew and how it responded. This will be counterbalanced to some extent by documentation held by D3 and/or ATA/Jabco in Malaysia, but my assessment is that overall the majority of the relevant documents will be in England.[67]As to attendance of the witnesses, the claimants are, for the most part, in neither England nor Malaysia. They can much more conveniently give evidence in England, where Leigh Day have confirmed they are able and willing to bring them to attend in person; whereas in Malaysian proceedings they would have to give evidence remotely, for which there is no satisfactory funding. For the five claimants in Malaysia, they would have to surface and risk deportation if they were to give evidence in Malaysia. The witnesses of Dyson UK will mostly, if not entirely, be in England, which is where it will be most convenient for them to give evidence, whereas witnesses from Dyson Malysia and ATA/Jabco will likely be in Malaysia. My assessment is that on the Dyson side there are likely to be more witnesses in England than Malaysia, but in any event those from the foreign jurisdiction could easily travel or give evidence remotely. Given the primacy of the claim against Dyson UK, the balance of convenience for the witnesses favours England.[68]As to attendance of the parties, the balance again favours England. The claimants can attend in person here whereas their attendance remotely in Nepal or Bangladesh, or by the five claimants inside Malaysia, is fraught with difficulty, not merely as a matter of remote access but by reference to translation to enable them to follow the proceedings. By contrast the litigation will be coordinated and conducted in England by Dyson UK on behalf of all the defendants, including Dyson Malaysia, and the representatives exercising that coordination and conduct can most easily attend in London.

Coordination and conduct of the litigation in London

[69]The fact that the litigation will be coordinated and conducted by Dyson UK on behalf of all the defendants, including Dyson Malaysia, wherever it takes place is a further connecting factor with England quite apart from its effect as a matter of practical convenience on the attendance of the relevant party representatives. Connection with the issues in the case: duty, breach, harm and remedy.[70]For the reasons I have already explained, these are mixed, but overall I would regard them as pointing more towards England than Malaysia, or at most neutral.

Governing Law

[71]Malaysian law is the governing law for all the claims. If and to the extent that Malaysian law is undisputed, this factor will be of little significance. The focus must be on issues of law which will be required to be resolved in England or Malaysia. In this case the disputes about Malaysian law are relatively narrow. The claims are common law and equitable claims, for which Malaysian law draws heavily on English decisions as well as other Commonwealth authorities. It is said that Malaysian courts draw more heavily on other Commonwealth authorities than do the English courts, but that is a matter which can readily be taken into account in the English court and the parties have not identified any Commonwealth decisions as likely to be of critical importance. In relation to the issue of duty for the negligence claim, the Malaysian Courts have not yet considered Vedanta or Okpabi, or considered their application in the context of liability for misconduct in a supply chain. That is a novel issue in English law as well. The Judge observed that there were differences between the two former Chief Justices whose statements were in evidence before him over whether “a Malaysian Court would be likely to apply Caparo over Vedanta and Okpabi” but it is a matter of debate as to whether there is any relevant difference in relation to the duty question in this case. No differences have been identified between Malaysian and English law in relation to the other tort claims. In relation to the unjust enrichment claim, two differences or potential differences are identified. The first is that whereas under English law in order to establish that the defendant's enrichment was unjust, the claimant must establish that it was vitiated by one or more of the specifically recognised "unjust factors", in Dream Property Sdn Bhd v Atlas Housing Sdn Bhd [2015] 2 MLJ 441 the Federal Court of Malaysia held that the unjustness of the enrichment depends on whether there is an "absence of basis" for the enrichment. This does not seem to be in issue, and Dream Property is itself reasoned by an analysis of English authorities. The second is that there may be an issue whether the unjust benefit has to flow directly from the claimant to the defendant. The Judge also observed that the underlying labour laws which the claimants complain have been breached by ATA/Jabco are Malaysian statutes, where there is presumably a large corpus of law with which the Malaysian courts are familiar (and will certainly be more familiar than their English counterparts), although there is nothing to suggest that these will give rise to controversy.[72]These differences would make it preferable, all other things being equal, for the Malaysian court to resolve them, especially in relation to the novel issue of the duty of care in the negligence claim. They are nevertheless issues which the English court is well equipped to deal with as a matter of expert evidence and using its own experience to analyse Malaysian and other Commonwealth authorities, given that Malaysian law is closely related to English law. As Lord Hodge DPSC observed in Perry v Lopag Trust Reg [2023] UKPC 16 [2023] 1 WLR 3494 at [12] and [15], where the foreign law is a common law system which applies the same or analogous principles and means of analysis as English law, there is considerable scope for the English judge to bring to bear their legal skills and experience in domestic law in determining and applying the foreign law.

Multiplicity of proceedings and risk of inconsistent judgments

[73]The Judge held that there was no real risk of multiplicity of proceedings or risk of inconsistent judgments vis-à-vis ATA/Jabco and only a slight risk, of not very great weight, in relation to RMP. I agree, and note that there was no Respondent’s Notice on these points. The defamation proceedings now no longer fall to be considered (although I observe that had they not been discontinued they would have been a very powerful pointer towards England as the appropriate forum). ‘Cambridgeshire factor’[74]Ms Kinsler sought to advance a different argument under ground 1(d), which was that there was a ‘Cambridgeshire factor’ in this case by reason of Dyson UK’s legal team having acquired a body of familiarity with the issues, and expertise in them, from the investigation and conduct of the defamation proceedings. This was not a point raised below and it would depend upon the extent to which the familiarity and expertise was built up as a result of the defamation proceedings, which the defendants would have had the opportunity to address in evidence had it been raised. I note that by the time of the broadcast on 10 February 2022 Leigh Day had already sent a letter before action in relation to the claims in these proceedings to Dyson UK to which Baker & McKenzie had responded, which casts doubt on whether the defamation proceedings can properly be considered the source for any familiarity or expertise in relation to the claims. In any event this is a far cry from the familiarity and expertise which applied to the Cambridgeshire factor in Spiliada. I would not attach any significance or weight to this supposed factor.

Equality of arms

[75]Both in terms of the standard of legal representation and the ability of the claimants to attend and give evidence in person, equality of arms favours England.

Conclusion

[76]The inability of the claimants to fund proceedings in Malaysia, and an assessment of the relevant connecting factors, make England clearly and distinctly the appropriate forum in which the case should be tried. I would allow the appeal.

LORD JUSTICE WARBY:

[77]I agree. I would add only these few observations in connection with the judge’s approach to the risk of irreconcilable findings. Concurrent actions involving overlapping issues and parties are a common feature of the litigation landscape. The English court will not take a blinkered approach to the case management issues that inevitably arise. It will strive to avoid or minimise duplication of effort and cost and, in particular, any risk of inconsistent outcomes or findings. That is no less true where one of the actions is for defamation. These are often accompanied by other, related actions. Various case management techniques have been deployed over the years, including an interim stay of proceedings, orders for sequential trials, and transfer from one Division to another. The Civil Procedure Rules laid new emphasis on the court’s case management role. Its ability to perform that role flexibly in defamation cases was enhanced in 2014 when Parliament abolished the presumption that these would be tried by jury. Even before that a concurrent trial by judge alone of common factual issues in two separate libel claims was directed and successfully conducted: see Mitchell v News Group Newspapers Ltd, Rowland v Mitchell [2014] EWHC 2615 (QB) and [2014] EWHC 4014 and 4015 (QB). The judge’s assessment that it was unlikely these claims would “be case managed together, or even with a real eye on one another” was plainly wrong. The strong probability is that the court’s approach would have been carefully co-ordinated.

SIR GEOFFREY VOS, MASTER OF THE ROLLS:

[78]I agree with both judgments.

Cited in 8 later judgments