1 The first plaintiff is incorporated in Norway and is the ultimate holding company of a multinational group of companies known as the Skaugen group. The Skaugen group provides marine transportation services in the oil and gas industry. ¹ The second plaintiff is a wholly-owned subsidiary of the first plaintiff and is incorporated in Singapore. It is one of the ship-owning arms of the Skaugen group. ²

The first defendant is incorporated in Germany and is the ultimate holding company of a multinational corporate group known as the MAN group. The MAN group, amongst other things, designs and manufactures marine engines. ³ The second defendant is a wholly-owned subsidiary of the first defendant and is incorporated in Norway. ⁴ It supports the business of the MAN group by maintaining contact with its customers in Norway. ⁵

2 In 2000 and 2001, the first plaintiff sourced six marine engines of a single specific model from the MAN group. The six marine engines were sold and shipped directly to shipbuilders in China, to be installed in six ships which were being built there for the Skaugen group. The plaintiffs’ case is that, in the course of promoting this model of engine to the first plaintiff, the defendants negligently or fraudulently misrepresented to the first plaintiff that that model of engine consumed fuel at no more than a specific rate under certain specified conditions.

3 The plaintiffs now bring this action in Singapore to seek damages from the defendants for the alleged misrepresentation. The question which arises before me on this registrar’s appeal is whether the Singapore courts have jurisdiction over the plaintiff’s misrepresentation claim. The assistant registrar held that the Singapore courts do not have jurisdiction. He agreed with the plaintiffs that its claim falls within Order 11 r 1(f)(ii) of the Rules of Court (Cap 322, R 5, 2014 Rev Ed). But he also agreed with the defendants that Singapore is forum non conveniens. That latter finding made this case “not a proper one for service out of Singapore” within the meaning of Order 11 r 2(2) of the Rules of Court.

4 The plaintiffs now appeal against the assistant registrar’s decision.

5 The issue before me now, quite simply, is whether the assistant registrar was correct. But the geographical reach of this case takes it beyond the anodyne procedural dispute over jurisdiction. The defendants authored the promotional material containing the representation in Germany. They delivered the material to the Skaugen group at sales and marketing meetings held either in Denmark or Norway. The six engines are the subject-matter of contracts of sale governed by English law, ⁶ entered into between the first defendant and the Chinese shipbuilders. Before delivering each engine, the defendants carried out field tests on the engine in Germany in the presence of the plaintiffs’ representatives. The six engines were delivered to China and installed in the six ships there. The ships have, since delivery to the Skaugen group, been owned and operated by various entities incorporated in the Cayman Islands, Hong Kong, and Singapore.

6 I have held that the plaintiffs have established a good arguable case that this action comes within both Order 11 r 1(f)(ii) and Order 11 r 1(p) of the Rules of Court. Order 11 r 1(f)(ii) is satisfied because there is a good arguable case that owners of the ships who have assigned their claims to the plaintiffs suffered loss and damage in Singapore. Order 11 r 1(p) is also satisfied because there is a good arguable case that the plaintiffs’ assignors’ causes of action arose in Singapore. Further, I have also held that the plaintiffs have established that Singapore is clearly the appropriate forum for the trial of this action because of the possibility of a transfer to the Singapore International Commercial Court (“SICC”) and also because the appropriateness of both alternative forums – Germany and Norway – has been overstated. I have therefore allowed the plaintiffs’ appeal.

7 The defendants have appealed to the Court of Appeal against my decision. I therefore set out my reasons.

8 The assistant registrar has set out a comprehensive summary of the facts underlying this action in his wide-ranging grounds of decision in IM Skaugen SE and another v MAN Diesel & Turbo SE and another [2016] SGHCR 6 (“Skaugen”). I therefore need set out only a brief summary.

9 In 2000 and 2001, the Skaugen group and GATX Corporation (“GATX”) entered into a joint venture to commission, own and operate six gas carrying ships. ⁷ To that end, the Skaugen group entered into six shipbuilding contracts with shipbuilders in China. The first plaintiff entered into the first four of the six contracts in July 2000. ⁸ These four contracts were almost immediately novated to a special purpose company known as Somargas Limited (“Somargas”). ⁹ The last two of these six contracts were entered into in May 2001 by another special purpose company known as Vintergas Limited (“Vintergas”). ¹⁰ Somargas ¹¹ and Vintergas ¹² are both Cayman Islands companies and are both owned equally by the Skaugen group and the GATX group.

10 In the summary of the facts which follows, I do not distinguish between the position of the four entities involved: the first plaintiff, GATX, Somargas and Vintergas. That is because the first plaintiff conducted all negotiations with the defendants on behalf of Somargas and Vintergas. ¹³ Further, all four of the following features of this case are, for all material purposes, identical: (a) the six shipbuilding contracts; (b) the six ships; (c) the terms on which the first four shipbuilding contracts were novated to Somargas; and (d) the ownership of Somargas and Vintergas.

11 The first plaintiff had the right, under the first four shipbuilding contracts, to approve the model of engine to be installed in those ships. ¹⁴ In mid-2000, the first plaintiff initiated a process to select the engine to be approved from those available on the market. ¹⁵ The first defendant participated in that process and promoted to the first plaintiff a particular model of four-stroke diesel engine which it designed and manufactured. ¹⁶ I shall refer to this engine as the “MAN Engine”. In the course of that process, in or around July 2000, the first defendant delivered to the first plaintiff a document known as the Project Planning Manual (“PPM”). The PPM is part of the defendants’ standard promotional and marketing materials. It is a general document which is not tailored to a specific customer or to a specific engine ¹⁷ but which sets out the general specifications of a range of marine diesel engines available from the first defendant. One of those engines is the MAN Engine. So the PPM does state that the MAN Engine consumes fuel at no more than a specific rate under certain specified conditions. ¹⁸

12 In or around November 2000, the defendants delivered to the shipbuilders a second document known as “6. Kraftstoffsystem Fuel System” (“FSI”). ¹⁹ The shipbuilders duly passed the FSI on to the plaintiffs. ²⁰ Like the PPM, the FSI sets out general information about a range of marine diesel engines produced by the first defendant and is not tailored to a specific customer or to a specific engine. ²¹ But the FSI, in effect, maintained that the MAN Engine consumed fuel at no more that the specific rate set out in the PPM under the same specified conditions. ²²

13 The first plaintiff eventually granted formal approval under the shipbuilding contracts for the MAN Engine to be installed in the first four ships. ²³ As for the last two ships, the first plaintiff caused Vintergas to enter into contracts with the shipbuilders which stipulated that the MAN Engine was to be installed in those two ships. ²⁴

14 As a result, the shipbuilders duly entered into six contracts to purchase six MAN Engines from the first defendant: four in September 2000 ²⁵ and a further two in June 2001. ²⁶ The technical specifications in all six of these contracts between the shipbuilders and the first defendant stipulated that the MAN Engine ultimately delivered was to consume fuel at no more than the rate specified in the PPM under the conditions specified in the PPM. ²⁷

15 Between May 2001 and June 2002, the six MAN Engines became ready to be handed over to the shipbuilders. ²⁸ As and when a particular engine was ready for handover, the first defendant conducted a field acceptance test (“FAT”) on that engine at the first defendant’s factory in Germany. ²⁹ Representatives of the Skaugen group and of the defendants were present to witness each of the six FATs. ³⁰ The purpose of each FAT was to: (i) run the MAN Engine in the laboratory under the specified test conditions; (ii) measure the rate at which that particular MAN Engine consumed fuel; and (iii) verify for those present that the engine’s actual fuel consumption met the contractually-stipulated specification. ³¹ The results of each FAT on each engine appeared to verify that each engine consumed fuel at a rate which was well below specification. ³²

16 Between October 2002 and October 2003, the shipbuilders duly delivered to the Skaugen group all six ships. ³³ Each ship had a MAN Engine installed in it which the first plaintiff believed met the fuel consumption specification.

17 Following delivery of the six ships in 2002 and 2003, a chain of entities related to the Skaugen and the GATX group owned each of the ships without noticing anything amiss with the rate at which the MAN Engines were consuming fuel. It is the plaintiffs’ case that all of these owners have assigned to the plaintiffs all of their claims in misrepresentation against the defendants arising from the facts which follow. ³⁴

18 In May 2011, the first defendant issued a press release reporting the interim results of its investigation into “possible irregularities” in the FATs conducted before handover on a broad class of marine engines produced by the first defendant over the preceding ten years. ³⁵ The MAN Engine fell within one of the classes of marine engine and the FATs done on the six MAN Engines fell within the period of time. The “possible irregularity” was that the first defendant’s employee conducting the FAT had manipulated, while the test was ongoing, the fuel consumption figures reported by the test equipment so that the reported fuel consumption deviated in the first defendant’s favour from the fuel consumption actually measured. ³⁶ This manipulation took place with the knowledge of the first defendant’s representative who was supervising the FAT as well as with the knowledge of the first defendant’s executive board. ³⁷ As the first defendant said in the press release: ³⁸

19 By a letter dated 25 May 2011 ³⁹ from the first defendant addressed generally to purchasers of the MAN Engine, the first defendant enclosed a copy of its May 2011 press release and informed the purchasers as follows: ⁴⁰

The first defendant says that it sent this letter, including the press release, to affected customers including the first plaintiff in or about May 2011, the date which it bears. The plaintiffs say that they did not receive this letter or its enclosure until June 2012. ⁴¹ The plaintiffs also say that they were not aware of the May 2011 press release in May 2011, and that they did not become aware of the press release until they received a copy of it with this letter in June 2012. ⁴²

20 By a letter dated 31 January 2012 ⁴³ addressed specifically to the first plaintiff, the first defendant gave the first plaintiff an update on the results of its ongoing investigations. ⁴⁴ The first defendant informed the first plaintiff in this letter, in effect, that there was a real possibility that the FATs on all six MAN Engines were manipulated to show that each MAN Engine met the fuel consumption specification when it actually did not. In this letter, the first defendant said: ⁴⁵

The first defendant says that it sent this letter by ordinary mail to the first plaintiff in or around January 2012. The first defendant accepts that it did not send the letter to the first plaintiff by fax until June 2012. ⁴⁶ The first plaintiff says that it did not receive this letter – whether by mail or by fax – until June 2012. ⁴⁷

21 In April 2012, ⁴⁸ in June 2012 and in September 2012, ⁴⁹ the first defendant admitted in writing to the Skaugen group that there were indications that the fuel consumption values for three of the six MAN Engines built for the Skaugen group were “externally influenced in an improper manner” during the FATs.

22 The first defendant’s press release of May 2011 initiated a criminal investigation by the German public prosecutor. In March 2013, the first defendant was found liable for the regulatory offence under German law of, in effect, failing in its duty to supervise its employees and failing to have effective compliance structures in place to prevent their misconduct. The German criminal court found that there had been “considerable manipulation” of the FATs over a long period of time and that the scheme was “planned and carried out with high criminal energy using the software modules programmed especially for this purpose”. ⁵⁰ As a result of these findings, the German court fined the first defendant €8.2m. ⁵¹

23 The plaintiffs’ case in brief is that: (i) the defendants’ representation to the first plaintiff in the PPM that the MAN Engine consumed fuel at a specific rate under certain specified conditions ⁵² was and is false; (ii) that the defendants made the representation either negligently or fraudulently in order to induce the first plaintiff to select the MAN Engine for installation under all six shipbuilding contracts; ⁵³ (iii) that the first plaintiff relied on the misrepresentation in selecting the MAN Engine for installation; and (iv) all of the owners of the six ships over the years have assigned to one or other of the plaintiffs the owners’ claims for all losses caused by the misrepresentation.

24 The defendants’ case in brief is that: (i) the PPM and the FSI do not contain any actionable representation made to the first plaintiff as to the rate at which the six MAN Engines actually delivered to the shipbuilders would consume fuel; ⁵⁴ (ii) any misrepresentation made on the occasion of the FATs occurred after the first plaintiff had selected the MAN Engine under the shipbuilding contracts and therefore could not have induced the first plaintiff’s decision to approve it for all six ships; ⁵⁵ (iii) the second plaintiff has no claim in its own right because it did not even exist when the representation was made and played no role in selecting the MAN Engine; ⁵⁶ and (iv) neither plaintiff has any claim by reason of the assignments because the defendants made no representation at any time to any of the assignors. ⁵⁷

25 The plaintiffs commenced this action in January 2015 and secured leave ex parte to serve the writ on the defendants outside Singapore in March 2015. The plaintiffs relied on three limbs of Order 11: (i) on Order 11 r 1(f)(i), on the basis that their claim is founded on a tort constituted, at least in part, by an act or omission in Singapore; (ii) on Order 11 r 1(f)(ii), on the basis that their claim is founded on or for recovery of damages for damage suffered in Singapore; and (iii) on Order 11 r 1(p), on the basis that their claim is founded on a cause of action arising in Singapore. The writ was served on the first defendant in Germany in June 2015. ⁵⁸ The first defendant entered an appearance in July 2015. The writ was served on the second defendant in Norway in September 2015. The second defendant entered an appearance in October 2015. ⁵⁹

26 The first defendant applied in August 2015 to challenge jurisdiction. The primary relief it sought was the following: (i) an order setting aside the writ; (ii) an order setting aside service of the writ out of the jurisdiction; and (iii) an order determining that this court has no jurisdiction over the plaintiffs’ dispute with the first defendant. In the alternative, the first defendant sought an order staying this action on grounds of forum non conveniens. ⁶⁰ The second defendant made a virtually identical application in November 2015 on its own behalf. ⁶¹ Although both defendants seek an order that the writ be set aside, neither defendant suggests any grounds to set aside the writ other than the plaintiffs’ failure to bring themselves within Order 11. That aspect of the defendants’ application therefore does not require independent analysis.

27 Both defendants are represented on their applications by the same counsel and advance the same arguments.

28  The defendants’ arguments, in brief, are as follows. First, both plaintiffs lack standing to bring this action. That is because the two plaintiffs have never owned any of the six ships, and thus did not themselves suffer any loss or damage. Second, the plaintiffs’ claim is not within any of the three heads of Order 11 on which the plaintiffs rely (see [25] above). ⁶² Third, even if the plaintiffs’ claim comes within one of those three heads, this case is not a proper one for service out of Singapore as required by Order 11 r 2(2) because Singapore is not clearly the appropriate forum for the trial of this action, with both Germany and Norway being more appropriate forums.

29 The assistant registrar set aside service of the writ and the ex parte order granting the plaintiffs leave to serve the writ out of the jurisdiction. ⁶³ His reasons for doing so are set out comprehensively in his grounds of decision in Skaugen. Those reasons can be summarised as follows. First, the plaintiffs’ claims have sufficient merit for leave to serve out of the jurisdiction to be granted. The plaintiffs have shown that the causes of action on which they rely have been assigned to one or other of them (Skaugen at [46]). Second, the plaintiffs’ claim comes within Order 11 r 1(f)(ii). They have shown that their claim is founded on damage sustained within Singapore caused by a tort “wherever occurring” (Skaugen at [106]). Finally, however, Singapore is not the appropriate forum for the resolution of this dispute. German law is the lex loci delicti (at [55] of Skaugen). Germany is therefore a more appropriate forum than Singapore. A German court would have the home court advantage in determining the plaintiffs’ claim under German law. Moreover, a German court would not have to apply Singapore’s double actionability rule, and would therefore determine the plaintiffs’ claims under German law alone (at [129] and [139] of Skaugen). The double actionability rule would require a Singapore court to apply both German law and Singapore law to determine the plaintiffs’ claims.

30 The plaintiffs have appealed ⁶⁴ against the assistant registrar’s decision. The defendants too have appealed against his decision, ⁶⁵ but only insofar as he found that the plaintiffs’ claims have sufficient merit and that those claims fall within Order 11 r 1(f)(ii) of the Rules of Court. The defendants also now seek to support the assistant registrar’s decision on the additional ground that the plaintiffs failed to make full and frank disclosure when they secured leave ex parte to serve the writ in this action out of the jurisdiction.

31 There are four issues to be decided in these appeals:

(a) whether the plaintiffs have standing to bring this action against the defendants (Issue 1);

(b) whether the plaintiffs failed to make full and frank disclosure when they applied ex parte for leave to serve the writ out of jurisdiction in May 2015 (Issue 2);

(c) whether this action comes within Order 11 r 1(f)(i), Order 11 r 1(f)(ii) or Order 11 r 1(p) (Issue 3); and

(d) whether this case is a proper one for service out of Singapore within the meaning of Order 11 r 2(2) (Issue 4).

32 Before I analyse each of these issues, I consider the burden on each of these issues and the standard to which that burden must be discharged.

33 The Court of Appeal held in Zoom Communications Ltd v Broadcast Solutions Pte Ltd [2014] 4 SLR 500 (“Zoom”) at [26] that there are three requirements for valid service of process out of the jurisdiction:

(a) the plaintiff’s claim must come within one of the heads of jurisdiction set out in Order 11 r 1 of the Rules of Court;

(b) the plaintiff’s claim must have a sufficient degree of merit; and

(c) Singapore must be the proper forum for the trial of the action.

34 Where a plaintiff applies under Order 11 for leave to serve a defendant outside Singapore, the burden of establishing all three requirements must lie on the plaintiff. That is axiomatic: it is the plaintiff who is the applicant and it is therefore the plaintiff who must bear the burden on the application. But it is also the position that where a defendant has been served outside Singapore under Order 11 and applies to set aside service, the burden even then remains on the plaintiff to establish all three requirements. That is so even though it is the defendant who is the applicant and the plaintiff who is responding to the application: Zoom at [75]–[76].

35 The standard to which a plaintiff must discharge its burden on the first Zoom requirement set out at [33] above is that of a good arguable case: Bradley Lomas Electrolok Ltd and another v Colt Ventilation East Asia Pte Ltd and others [1999] 3 SLR(R) 1156 (“Bradley Lomas”) at [14]. A “good arguable case” entails one side having “a much better argument on the material available”: Canada Trust Co v Stolzenberg (No 2) [1998] 1 WLR 547 at 555F-G.

36 The standard to which a plaintiff must discharge its burden on the second Zoom requirement is that the plaintiff must show merely that there is a serious question to be tried: Bradley Lomas at [14] and [19]. The standard on the second Zoom requirement is therefore less stringent than that on the first.

37 However, where a plaintiff relies on a head of jurisdiction under Order 11 r 1 which, in itself, requires the court to examine the merits of the plaintiff’s claim when considering the first Zoom requirement, the inquiry on the second Zoom requirement is subsumed in the first: Bradley Lomas at [18]–[20]. That has two consequences. First, the plaintiff must establish a good arguable case on the merits of its claim, and not merely a serious question to be tried. Second, there is no need for the plaintiff to address the second Zoom requirement separately. For example, in Bradley Lomas itself, the Court of Appeal observed that a plaintiff who establishes on the first Zoom requirement a good arguable case that its claim comes within Order 11 r 1(f)(i) does not need to show separately on the second Zoom requirement a serious question to be tried on the merits of its claim. A plaintiff in that situation, by satisfying the first Zoom requirement, has already established a good arguable case that its “claim is founded on a tort … constituted, at least in part, by an act or omission occurring in Singapore”.

38 That point is significant in the appeals before me. All of the three heads of jurisdiction on which the plaintiffs rely – Order 11 r 1(f)(i), Order 11 r 1(f)(ii) and Order 11 r 1(p) – require an examination of the merits of the plaintiffs’ claim. The plaintiffs must therefore show a good arguable case on the merits rather than merely a serious question to be tried. Further, the plaintiffs do not need to address the second Zoom requirement separately.

39 The standard to which a plaintiff must discharge its burden on the third Zoom requirement is to show that Singapore is clearly “the forum in which the case can be suitably tried for the interests of all the parties and for the ends of justice” (see Spiliada Maritime Corp v Cansulex Ltd (The Spiliada) [1987] AC 460 (“Spiliada”) at 476C and 480H to 481E).

40 The final point I make is that an appeal from the registrar to a judge in chambers is dealt with by way of rehearing. I am therefore entitled on these appeals to decide the defendants’ application de novo; I am in no way fettered by the registrar’s decision: ACES System Development Pte Ltd v Yenty Lily (trading as Access International Services) [2013] 4 SLR 1317 at [12].

41 With these principles in mind, I turn to consider in turn the four issues I have set out [31] above.

42 The issue of standing is an integral part of a plaintiff’s burden on the second Zoom requirement. Standing is not, on an application of this nature, usually analysed as a discrete issue therefore. The defendants, however, have chosen to challenge standing as a discrete issue. That is no doubt in order to put specific focus on the chain of assignments on which the plaintiffs found this action. As the parties have addressed standing discretely, I shall analyse it discretely.

43 The plaintiffs bear the burden of establishing a good arguable case on standing, just as they would on any other issue which goes both to the first Zoom requirement and to the merits of the case. To establish standing, the plaintiffs rely on their status as the assignees of all claims against the defendants from the current and all previous owners of the six ships. ⁶⁶ The defendants’ challenge to standing proceeds in two steps. First, they point out that the plaintiffs themselves have no cause of action against the defendants. Neither plaintiff was ever the owner of any of the six ships. Quite apart from anything else, therefore, neither plaintiff could have suffered any loss or damage by reason of the alleged misrepresentation. ⁶⁷ Second, the defendants argue that that the plaintiffs have not shown that the actual owners of the ships – who are the entities who actually did suffer loss by reason of the alleged misrepresentation – have assigned their claims to the plaintiffs. ⁶⁸

44 The assistant registrar has comprehensively set out in Skaugen (at [34]–[37]) the fairly complicated chain of transfers in ownership and operation of the ships. I therefore need do no more than summarise the chain to the extent that it is relevant for present purposes.

45 I begin by noting that, as pleaded, the plaintiffs’ claim for damages for loss caused by the alleged misrepresentation before June 2012 – which, on the plaintiffs’ case, is when they became aware that the representation as to fuel consumption was false ⁶⁹ – is in respect only of loss suffered by the current and past owners of the six ships. ⁷⁰ The plaintiffs’ claim, as pleaded, does not seek to recover damages for any loss suffered by the current and past operators of the six ships. I therefore begin by tracing the changes in ownership of the ships upon and after delivery.

46 All six ships were delivered between October 2002 and October 2003 to a Hong Kong company known as Somargas Ltd (“Somargas HK”). ⁷¹ Somargas HK is owned equally by the Skaugen group and the GATX group. ⁷² Somargas HK therefore became, upon delivery, the first owner of the ships. Somargas HK acquired the rights to the ships when Somargas and Vintergas novated the six shipbuilding contracts to Somargas HK while the ships were still under construction. Somargas HK owned the ships until February 2011. ⁷³

47 The second owner of the six ships was Somargas II Pte Ltd (“Somargas SG”). Somargas SG is a special purpose company incorporated in Singapore as a wholly-owned subsidiary of Somargas HK. ⁷⁴ Somargas SG owned all six ships from February 2011 until April 2013. Somargas SG also registered all six ships under the Singapore flag in March 2011. ⁷⁵

48 In April 2013, the GATX group became the sole owner of three of the ships through special-purpose, one-ship GATX subsidiaries. I shall refer to these subsidiaries as “the three GATX subsidiaries”. Somargas SG became a wholly-owned member of the Skaugen group and continued to own the three remaining ships. ⁷⁶

49 Somargas SG sold the three remaining ships between June 2013 and December 2014 as follows:

(a) In June 2013, Somargas SG sold a ship to a company known as SGPC1 Pte Ltd.

(b) In November 2014, Somargas SG sold a ship to a company I shall call “Gasmar”.

(c) Finally, in December 2014, Somargas SG sold the sole remaining ship to a company I shall call “Zhonghua”.

SGPC1 Pte Ltd is a Singapore company which is 35% owned by the first plaintiff through a wholly-owned subsidiary, Skaugen Marine Investment Pte Ltd (“Skaugen Marine”). ⁷⁷ Gasmar and Zhonghua are not members of the Skaugen group.

50 The plaintiffs’ case is that the plaintiffs have acquired, under one or other of two agreements, the claims of the current and every former owner of each of the six ships against the defendants. The first of the two agreements is a Claims Transfer Agreement entered into in June 2014 between GATX and the first plaintiff. ⁷⁸ Under this agreement, the first plaintiff acquired from GATX the latter’s “possible claim against [the first defendant] for compensation of all losses suffered by GATX and its subsidiaries as a result of any improper external influence and/or manipulation of the FATs on the main engines on any or all” of the six ships. In February 2016, the three GATX subsidiaries issued letters asserting that, as at the date of the Claims Transfer Agreement, they had transferred their claims to GATX and authorised GATX to transfer those claims on to the first plaintiff. ⁷⁹

51 The second agreement is an Assignment Agreement entered into in January 2015 between the second plaintiff as assignee and Somargas SG and Skaugen Marine as assignors. ⁸⁰ This agreement describes all three parties as constituting “current or former – direct or indirect – owners or charterers” of the six ships. The operative part of the Assignment Agreement assigns all of Somargas SG’s and Skaugen Marine’s “Claims” to the second plaintiff. “Claims” is defined as “claims for damages or other compensation against [the first defendant] and/or other companies within the [MAN] group of companies” for “losses as a consequence of the [six ships’] actual fuel consumption levels and [the first defendant’s] manipulation of FATs conducted with respect to the engines installed therein”. ⁸¹

52 The defendants challenge each chain in the link of transfers of the ships and assignments of the claims which I have outlined above. Their argument in essence is that if entities earlier in either chain do not have a valid claim against the defendants, those entities had nothing to pass on by assignment to entities later in the chain and, ultimately, to the plaintiffs. ⁸²

53 Based on this chronology, and taking the plaintiffs’ own case at its highest, the defendants’ first argument is that the only two entities which can conceivably be said to have any claims against the defendants are Somargas HK and Somargas SG. That is because they are the only entities which actually owned the ships – and therefore the only two entities who can conceivably be said to have suffered loss as a result of the misrepresentation – before the Skaugen group became aware that the representation was false. That took place at the very latest in June 2012. ⁸³ The plaintiffs cannot establish standing by relying on the claims of any other entity. That includes Somargas and Vintergas. Although those two entities may have relied on the alleged misrepresentation, they never owned any of the ships and so cannot have suffered any loss in reliance on the misrepresentation. ⁸⁴ That also includes the three GATX subsidiaries. They became owners of the ships after June 2012 and so would have been aware that the representation was false. ⁸⁵ That also includes GATX and Skaugen Marine. Neither of those entities owned any of the ships, ⁸⁶ nor is it the plaintiffs’ pleaded case that either of these entities relied on any misrepresentation.

54 Second, the defendants argue that the plaintiffs cannot derive any standing to sue the defendants from Somargas HK’s claims. Although Somargas HK did in fact own the ships upon delivery, the plaintiff has not pleaded that Somargas HK suffered any loss in reliance on any misrepresentation. ⁸⁷

55 That leaves Somargas SG as the only entity through which the plaintiffs can derive standing. But the defendants’ third argument is that Somargas SG has no standing to assert a claim against the defendants, whether derived from Somargas HK or in its own right. First, there is no evidence that Somargas HK transferred any claims it might have against the defendants to Somargas SG. ⁸⁸ Further, the only claims of its own which Somargas SG assigned to the second plaintiff were claims arising from the misrepresentation of the results of the FATs, and not claims arising from the misrepresentation in the PPM or the FSI. But the plaintiffs’ pleaded case does not allege that the misrepresentation of the results of each FAT was an actionable misrepresentation in its own right. Instead, the plaintiffs’ case is that the FAT results are merely confirmations of an earlier actionable representation as to fuel consumption. ⁸⁹ Finally, and in any event, even if Somargas SG did rely on any representation from anybody in relation to the ships, that representation must have come from Somargas SG’s transferor, Somargas HK, and not from the defendants. ⁹⁰

56 The plaintiffs respond as follows. All they have to do at this stage in the proceedings is show a good arguable case that one of the assignors has assigned to either of the plaintiffs a claim which it can assert against the defendants. ⁹¹ Somargas HK owned the ships from delivery until February 2011. When Somargas SG acquired the six ships from Somargas HK in February 2011, it also acquired all of Somargas HK’s assets and liabilities. ⁹² Those assets included Somargas HK’s rights of action against the defendants. ⁹³ Somargas SG then assigned those rights of action to the second plaintiff under the Assignment Agreement. ⁹⁴ Further, so far as GATX is concerned, it was authorised by the GATX subsidiaries to transfer their rights of action to the first plaintiff under the Claims Transfer Agreement. ⁹⁵ And so far as Skaugen Marine is concerned, the claim it assigned to the second plaintiff under the Assignment Agreement is a claim as the ship’s charterer, not as the ship’s owner. ⁹⁶

57 Having set out the chain of transfers and assignment, and the parties’ contentions, I now turn to consider whether the plaintiffs have made out a good arguable case on standing.

58 The defendants’ contentions put the focus in this inquiry squarely on Somargas HK and Somargas SG. Both companies became owners of the ships after delivery and before the representation as to fuel consumption was falsified, at the latest in June 2012. Both companies are, on the defendants’ own case, in a position to argue that they became owners of the ships without knowledge that the fuel consumption figures had been manipulated and also that they suffered loss thereby.

59 I consider first the position of Somargas SG. I accept that the plaintiffs have established a good arguable case both that: (i) Somargas SG has a cause of action in its own right against the defendants; and (ii) Somargas SG has a cause of action derived from Somargas HK against the defendants.

60 Somargas SG’s claim in its own right arises from its ownership of all six ships from February 2011 to April 2013, and before it transferred each of the six ships to other entities on various dates between April 2013 and December 2014. It has assigned those causes of action to the second plaintiff under the Assignment Agreement. This, to my mind, suffices in itself to establish a good arguable case that the plaintiffs have standing to bring this action.

61 But I accept also that the plaintiffs have established a good arguable case that Somargas SG has a cause of action against the defendants derived from Somargas HK. For this purpose, I accept – both as a matter of evidence and substance – that the plaintiffs have established a good arguable case that Somargas SG acquired Somargas HK’s claims against the defendants together with the six ships in February 2011.

62 As a matter of evidence, the plaintiffs rely on a statement on oath by Mr Morits Skaugen on behalf of the plaintiffs that Somargas HK transferred its claims against the defendants to Somargas SG together with the six ships and all its other assets. ⁹⁷ The defendants attack this aspect of the plaintiff’s case by pointing to the absence of any contemporaneous documentary evidence that Somargas HK did transfer its claims to Somargas SG together with the six ships. They say that it is incredible that a transfer of claims was not evidenced contemporaneously by way of a board resolution, a shareholders’ resolution, minutes of meeting or in some other documentary form. ⁹⁸ It is true that the only evidence of this transfer is a statement on oath. But the defendants are unable to adduce any evidence to the contrary, not even a statement on oath. A statement on oath is evidence even if it is unsupported by independent evidence. The lack of documentary evidence does not disprove the plaintiffs’ statement on oath that such a transfer took place. I therefore accept that evidence as being sufficient to establish a good arguable case on this aspect of the plaintiffs’ claim.

63 Further, as a matter of substance, I observe that the absence of documentary evidence is not fatal to the plaintiffs’ case that there was a transfer of Somargas HK’s claims to Somargas SG. An assignment of a chose in action must be documented only if the law governing the assignment requires it. What is the relevant governing law? The Claims Transfer Agreement is governed by New York law ⁹⁹ and the Assignment Agreement is governed by Norwegian law. ¹⁰⁰ Both before me and before the assistant registrar, neither party has suggested that any of the issues arising in the law of assignment should be determined under either system of foreign law. The assistant registrar held that, in the absence of any submissions by the parties on foreign law, the presumption of similarity of laws applies: see Skaugen at [40]. He therefore applied Singapore law to determine questions relating to assignment. He held that under Singapore law, an assignment can either be a legal assignment or an equitable assignment. He held, further, that an equitable assignment need not be in writing (Skaugen at [43]).

64  I agree broadly with the assistant registrar’s approach. The fact that Somargas HK’s assignment of its claims to Somargas SG is not documented does not suffice to reduce the plaintiffs’ case on this aspect below a good arguable case because the assignment could have been an equitable one: Skaugen at [43]. The defendants do not allege that an equitable assignment would be ineffective either for lack of notice or for failure to join the assignor to this action.

65 Similarly, both before me and before the assistant registrar, the defendants did not argue that the assignments to the plaintiffs are void as the assignment of bare rights of action in tort: Skaugen at [45]. Both parties appear to accept, therefore, that causes of action in tort – just like causes of action in contract – are assignable provided that the assignee has a genuine commercial interest in taking the assignment. ¹⁰¹ That is what the assistant registrar held, citing the seminal House of Lords decision in Trendtex Trading Corporation v Credit Suisse [1982] AC 679 (“Trendtex”).

66 I agree with the assistant registrar. The assignability of a cause of action in tort, however, requires a little further consideration. A leading local textbook observes that a bare right to litigate is not generally assignable and, further, that that includes a bare right to litigate a “personal tort”: Tan Yock Lin, Personal Property (Academy Publishing, 2014) (“Tan Yock Lin”) at para 18.014. The bar exists because the assignment of a bare right of action in tort facilitates a third party meddling officiously in the disputes of others, thereby tending to stir up unwanted litigation and contravening public policy in the same way as maintenance and champerty.

67  It might be said in favour of assignability in this case that the tort of misrepresentation is not a “personal tort” and therefore not caught by the bar on assignment of causes of action in tort. But the bar on assignment is not always framed in terms which limit it to “personal torts”. For example, dicta in the High Court of Australia’s decision in Poulton v The Commonwealth (1952) 89 CLR 540 (“Poulton”) suggests that the bar extends to the right to litigate any cause of action in tort (see Poulton at 602). That said, some Australian courts have recognised an exception to even this broad formulation of the rule where the assignee has a genuine and legitimate commercial interest in taking an assignment of the right.

68 In the decision of the High Court of Australia in Equuscorp Pty Ltd v Haxton [2012] CLR 498, Heydon J noted that Poulton predated the decision in Trendtex (at [156]–[157]) and suggested that the bar in Poulton does not apply where a genuine commercial interest underlies the assignment. In the same case, Gummow and Bell JJ took a similar view, holding that there is an exception to the bar where there is “a genuine and substantial commercial interest” (at [79]). It must be noted, however, that their comments were not made specifically in relation to causes of action in tort.

69 Similarly, the Queensland Court of Appeal in WorkCover Queensland v AMACA Pty Limited [2012] QCA 240 was prepared to find that even causes of action for personal injuries are assignable where the assignee has a pre-existing legitimate commercial interest in enforcing the right: at [17].

70 It therefore appears to me that causes of action in tort are assignable under Singapore law where the assignee has a genuine commercial interest in taking the assignment. That broader single exception encompasses within it the mischief contemplated by the bar on the right to assign the bare right to litigate a “personal tort”. The real mischief in all cases is the lack of a sufficient interest in the assignee taking the assignment rather than the difficult question of whether a particular tort can be characterised as a “personal tort” (see Simpson v Norfolk and Norwich University Hospital NHS Trust [2012] 1 All ER 1423).

71 Here, the causes of action in tort assigned under both the Claims Transfer Agreement and the Assignment Agreement are capable of being assigned. As far as the Skaugen assignors are concerned, they assigned their claims to the second plaintiff, who is an entity within the same group of companies. As far as the GATX assignors are concerned, until April 2013, the GATX group had – by reason of its joint venture with the Skaugen group to construct and operate the six ships as a fleet – an equal interest in the six ships and in the corporate entities which owned the six ships. The GATX assignors’ assignment to the first plaintiff is an extension and a natural consequence of their commercial relationship in the joint venture. The claims have, broadly speaking, moved with the transfers of the ships. They are now vested in the plaintiffs for a specific purpose: to enable the plaintiffs to attempt to recover compensation for a wrong done to the economic and commercial enterprise associated with the six ships. This is sufficient for the plaintiffs as assignees to have a genuine commercial interest in taking the assignments and pursuing the claims assigned.

72 I therefore accept that both plaintiffs have established a good arguable case on the issue of standing.

73 The defendants contend that the plaintiffs failed to make full and frank disclosure when they applied ex parte for leave to serve the writ out of jurisdiction. They say that the plaintiffs’ disclosure at the leave stage is wanting in four key points. ¹⁰² First, the plaintiffs failed to disclose that they had taken the unequivocal position in a parallel claim in Norway in respect of the same subject-matter as this action that the entire damage was suffered in Norway. That is contrary to their position now that at least part of the damage was suffered in Singapore. ¹⁰³ Second, the plaintiffs failed to disclose that, by their own case, no tortious acts or omissions were committed by the defendants in Singapore. Third, the plaintiffs gave the incorrect impression in their statement of claim that both plaintiffs had suffered losses in their own right, rather than being merely the assignees of claims for losses suffered by others. Fourth, the plaintiffs failed to disclose that no claims had been assigned to them by Somargas, Vintergas, or Somargas HK.

74 I do not accept that the plaintiffs failed to make full and frank disclosure, or that any such failure would on these facts justify setting aside the order made ex parte granting leave to serve the writ in this action outside the jurisdiction.

75 I reject the first of the defendants’ four points. The claim which the plaintiffs filed in Norway ¹⁰⁴ did not take the position that the entire damage arising from the subject-matter of this action was, in point of fact, suffered in Norway. What the plaintiffs pleaded in the Norwegian claim was as follows: ¹⁰⁵

Nowhere in this pleading do the plaintiffs state that the entire damage was suffered in Norway. The pleading asserts only that damage was suffered in Norway, not that the entire damage was suffered in Norway.

76 Further, reading the pleading in context, it is clear that the assertion being made is that “as a result of the case’s connection with Norway”, damage “must be regarded” to have taken place in Norway. In other words, what the plaintiffs did is to take a position that the result of applying Norwegian procedural law to the facts of the case pointed to Norway as the place in which the damage “must be regarded” as having been suffered. That is not the same thing as taking a position that damage factually occurred in Norway let alone that the entire damage factually occurred in Norway. That position taken in Norway has little or no bearing on a submission that some damage factually occurred in Singapore. I therefore find that, even though the plaintiffs’ position in the Norwegian proceedings was not disclosed to the court on the plaintiffs’ ex parte application for leave to serve the defendants out of the jurisdiction, that failure does not amount to material non-disclosure.

77 I reject also the defendants’ second and fourth points. These two points do not raise any facts which the plaintiffs failed fully and frankly to disclose. These two points are no more than two contested points on which the defendants simply take a different position from the plaintiffs. The plaintiffs believe, and have argued, that tortious acts and omissions did occur in Singapore and that there were valid assignments of all of these claims. The defendants taking a different view from the plaintiffs on a contested point does not suffice to establish non-disclosure of facts, much less to establish material non-disclosure of facts.

78 Finally, I reject also the defendants’ third point. It is true that the statement of claim as it stood at the time of the ex parte application concluded by pleading that the first plaintiff and the second plaintiff had each suffered loss and damage respectively in their own right. ¹⁰⁶ But the statement of claim also expressly pleaded that the first and second plaintiffs, “as assignees of claims, are entitled to bring a claim against the 1st and/or 2nd Defendants for the loss and damage incurred by all current and previous registered owners” of the six ships. ¹⁰⁷ It then went on to plead expressly and with particulars the entire chain of transfers, the Claims Transfer Agreement and the Assignment Agreement. There was therefore no material non-disclosure of the fact that the plaintiffs bring this action to recover damages as assignees of the underlying causes of action.

79 In any event, even if I accepted that there had been a failure to make full and frank disclosure in March 2015 at the ex parte application for leave to serve the defendants out of the jurisdiction, that would at best result in setting aside the specific order made on that application. The defendants do not suggest that any of the alleged failures to make full and frank disclosure in this case were so egregious as to amount to an abuse of the process of the court sufficient to preclude the plaintiffs making a fresh application for leave. That fresh application would have to be considered today on its merits and in light of all the evidence which is now before me. That evidence, of course, includes evidence of all of the facts which the defendants allege were not fully and frankly disclosed on the ex parte application. Setting aside the order of March 2015 on grounds of a failure to make full and frank disclosure would serve little purpose except to waste time and costs. The plaintiffs would remain entitled to re-apply for leave on the same evidence which is before me now and to have that fresh application adjudicated on its merits.

80 It is to those merits that I now turn.

81 The next Zoom requirement is that the plaintiffs establish a good arguable case that one of the heads of jurisdiction under Order 11 r 1 of the Rules of Court is satisfied. As I have mentioned, the plaintiffs rely on three heads: Order 11 r 1(f)(i), Order 11 r 1(f)(ii) and Order 11 r 1(p). The first two heads are founded specifically on tort. The third head is a general head and requires merely that the plaintiffs’ cause of action arose in Singapore.

82 In order to ascertain whether a claim founded on tort is justiciable in Singapore at all, it is necessary first to ascertain where the tort was committed. To answer this question, the court applies the lex fori, in this case Singapore law: Macmillan Inc v Bishopsgate Investment Trust Plc (No 3) [1996] 1 WLR 387 at 392C. If the answer to the question is that the tort was committed in a foreign jurisdiction, the court must then take the further step of analysing whether the tort satisfies the double actionability rule, ie that it is actionable both under the law of the jurisdiction in which the tort was committed, ie, the lex loci delicti, and also under the substantive law of the lex fori: Rickshaw Investments Ltd and another v Nicolai Baron von Uexkull [2007] 1 SLR(R) 377 (“Rickshaw Investments”) at [53]. The double actionability rule is subject to a flexible exception allowing the court to disapply one or both limbs if injustice or unfairness would otherwise result. This exception is, however, strictly applied lest the exception overwhelm the rule: Rickshaw Investments at [58].

83 It should also be noted that Singapore law’s adherence to the double actionability rule to determine justiciability appears to be an anachronism. Most common law jurisdictions have by now abandoned the rule. Canada’s Supreme Court abrogated the rule by judicial decision in 1994, in Tolofson v Jensen (1994) 120 DLR (4th) 289. England abrogated the rule by statute in 1995, when it passed the Private International Law (Miscellaneous Provisions) Act 1995 (c 41). Australia’s High Court abrogated the rule in 2002, in Regie Nationale des Usines Renault SA v Zhang (2002) 210 CLR 491.

84 In 2003, the Law Reform Committee of the Singapore Academy of Law called for the double actionability rule to be abolished in Singapore law, proposing a legislative regime mirroring the English statute: see Law Reform Committee of the Singapore Academy of Law, Reform of the Choice of Law Rule Relating to Torts (31 March 2003). The Court of Appeal in Rickshaw Investments (at [66]) held that abolition of the double actionability rule in Singapore could not be achieved by common law and noted the Law Reform Committee’s recommendation for statutory abolition. In the meantime, the Court of Appeal consoled itself by noting that, even if the law were not reformed as proposed, the flexible exception to double actionability remained available to ensure that a just and fair result could nevertheless be achieved in all cases.

85 I am bound by Rickshaw Investments to apply the double actionability rule to the plaintiffs’ causes of action. That is another way of saying that the plaintiffs cannot establish a good arguable case in tort if they cannot establish a good arguable case that they satisfy the double actionability rule. That is critical for their case for service out of the jurisdiction under Order 11 r 1(f)(i) and Order 11 r 1(f)(ii). Order 11 r 1(p), being a general head of jurisdiction not confined to tort, raises different considerations. I analyse how Order 11 r 1(p) interacts with causes of action founded on tort and the double actionability rule below.

86 Before I can even consider whether the plaintiffs are able to bring themselves within Order 11 r 1(f)(i) or Order 11 r 1(f)(ii), therefore, I must address the following questions:

(a) First, where was the tort committed?

(b) Second, if the tort was committed in a foreign jurisdiction, is the tort actionable under the lex loci delicti?

(c) Third, would the acts which constitute a tort actionable under the lex loci delicti also constitute a tort under Singapore law? and

(d) Fourth, does the flexible exception to the double actionability rule apply?

87  The first issue that must be addressed is where the tort took place. The Court of Appeal in JIO Minerals FZC and others v Mineral Enterprises Ltd [2011] 1 SLR 391 (“JIO Minerals”) at [90] observed that the test that is applied is one “which looks at the events constituting the tort and asks where, in substance, the cause of action arose”. This is known as the “substance test”: JIO Minerals at [90].

88 Applying the substance test to the tort of misrepresentation is especially prone to complication, precisely because the tort is founded on human communication. Modern technology now has the ability to amplify the temporal and spatial reach of human communication. The law can no longer assume, in formulating general rules, that the typical misrepresentation is communicated in a single place on a single occasion to a single person. A misrepresentation today can be communicated, in a literal sense, to the world at large, preserved permanently and be relied upon all over the world by a significant subsection of the world’s population.

89 But the complications associated with the tort of misrepresentation are not due only to advances in modern technology. They are due also to the interplay between the intrinsic and intangible nature of human communication and the constituent elements of the tort. A misrepresentation can quite easily be made in one jurisdiction, received in a second and relied upon in a third. The Court of Appeal itself recognised that each of these three elements plays a role in applying the substance test to the tort of misrepresentation. Thus, at [93], the Court of Appeal observed that there is “room for debate as to whether the place where the representation was made is more significant than the place where the representation was received”, and later, at [94], contrasted the place where the misrepresentation was received with the place where it was relied upon. The Court of Appeal’s analysis in JIO Minerals therefore suggests that applying the substance test to the tort of misrepresentation requires examination of all three elements: where a representation is made, where it is received and where it is relied upon.

90 With all this in mind, the Court of Appeal in JIO Minerals held that the following specific rules operate as a gloss on the ordinary substance test when determining the loci delicti in the tort of misrepresentation:

(a) Where the misrepresentation is received and relied upon in a single jurisdiction, then that jurisdiction is the loci delicti unless (at [93]) it is fortuitous that the misrepresentation was received and relied upon in that jurisdiction.

(b) Where the misrepresentation is made in one jurisdiction and is received and relied upon in another jurisdiction:

(i) if the misrepresentation is made to a specific person or to a specific class of persons, the loci delicti is the place where the representation is received and acted upon: JIO Minerals at [91] and [93], citing Cordoba Shipping Co Ltd v National State Bank, Elizabeth, New Jersey (The “Albaforth”) [1984] 2 Lloyd’s Rep 91 at 92, and Diamond v Bank of London and Montreal Ltd [1979] QB 333 at 345–346;

(ii) if the misrepresentation is made to an unspecified class of persons, the loci delicti is the jurisdiction in which the misrepresentation is made: JIO Minerals at [91], citing Cordova Land Co Ltd v Victor Brothers Inc [1966] 1 WLR 793 at 801.

(c) Where the misrepresentation is made in more than one jurisdiction, or is received in more than one jurisdiction, or is relied upon in more than one jurisdiction, it is the ordinary substance test which applies: JIO Minerals at [93]–[95].

91 The case before me is a case where it can plausibly be argued that the misrepresentation was made in more than one jurisdiction, was received in more than one jurisdiction and also was relied upon in more than one jurisdiction. It is therefore the ordinary substance test which I must apply, unassisted by any gloss.

92 It is appropriate at this juncture to set out in greater detail the content of the representation which the plaintiff alleges to be a misrepresentation.

93 The plaintiffs’ case is that the representation was made in the PPM. The PPM states that the fuel consumption of the MAN Engine under ISO conditions at a load of 85% is 180 g/kWh. ¹⁰⁸ The PPM was provided to the first plaintiff in July 2000. ¹⁰⁹ The first plaintiff’s case is that it relied on this representation when it approved the MAN Engine under the shipbuilding contracts for the first four ships. ¹¹⁰ This representation was not made to Vintergas and was made long before Vintergas entered into the shipbuilding contracts for the fifth and sixth ships. The plaintiffs’ case in relation to the two later ships is that the first plaintiff conducted all negotiations with the defendants on behalf of Vintergas and that the first plaintiff’s reliance on the representation in agreeing to approve the MAN Engine in the two shipbuilding contracts Vintergas entered in 2001 can therefore be attributed to Vintergas. ¹¹¹

94 The plaintiffs’ case, further, is that the representation in the PPM was maintained and confirmed on two subsequent occasions. ¹¹² The first such occasion is in the FSI. The defendants delivered the FSI to the shipbuilders, who then transmitted the FSI to the first plaintiff. ¹¹³ The FSI states that the fuel consumption of the MAN Engine at a load of 100% under ISO conditions with attached pumps is 193.64 g/kWh with a tolerance of +3%. ¹¹⁴

95 The second occasion on which the plaintiffs say that the defendants maintained and confirmed the representation in the PPM is in reporting the results of the manipulated FATs. Before delivery, each MAN Engine was put through a FAT conducted at the first defendant’s factory in Augsburg, Germany. The results of the FATs showed that each engine was consuming fuel at a rate which comfortably exceeded the specifications in the PPM and the FSI. ¹¹⁵ The six FATs – one FAT per engine – were conducted in May, August and November 2001 and in February, May and June 2002. ¹¹⁶

96 It is also important to keep in mind the chronology by which the plaintiffs came to realise that at least three of the MAN Engines installed in the six ships were not as fuel efficient as the representation had led them to believe. This is important because any misrepresentation would have ceased to be operative on the plaintiffs’ assignors from that time. I have summarised the chronology at [18] to [22] above. The key dates are May 2011, January 2012, April 2012 and June 2012.

97 I now apply the substance test with these facts and dates in mind. To begin, if the focus is on where the representation was made, it appears that Germany is the relevant location. The first defendant is a German company. It authored the PPM in Germany. The representation in the PPM might therefore be said quite obviously to have been made in Germany. The same applies for the occasions on which that representation was maintained and confirmed in the FSI and at the FATs, all of which have their location in Germany.

98 But if the question is where the representation was received, then four options present themselves. In relation to each of these four options, it is also possible to argue that that location is also where the representation was, for all intents and purposes, made. The argument is that a representation is not made when it is published in the abstract, but only when the representation is actually communicated to a putative plaintiff.

99 The first option as to the location of receipt is Norway. ¹¹⁷ The defendants gave the PPM to the first plaintiff in July 2000 ¹¹⁸ in the course of negotiations between the first plaintiff and the second defendant. Those negotiations took place primarily in Norway, at the second defendant’s sales office and the first plaintiff’s offices there.

100 The second option as to the location of receipt is Denmark. The first defendant says that the defendants gave the PPM to the first plaintiff in Denmark, as the negotiations between the first defendant and the first plaintiff took place there. ¹¹⁹

101 The third option as to the location of receipt is China. The defendants delivered the FSI, which on the plaintiffs’ case maintained and confirmed the representation in the PPM, ¹²⁰ to the shipbuilders who are located in China. The shipbuilders in turn transmitted the FSI to the first plaintiff’s office in China. ¹²¹ The plaintiffs say that the place of receipt was therefore China. The defendants accept that the shipbuilding project was ongoing in China, but also suggest Norway as an alternative, because that is where the first plaintiff had its principal place of business. ¹²² This misrepresentation therefore might plausibly be said to have been received in China, or, alternatively, Norway.

102 The fourth and final option as to the location of receipt is Germany. The FATs, which also maintained and confirmed the representation made in the PPM, were conducted in Germany and the results were reported in Germany. The misrepresentation might therefore have been received in Germany.

103 On the question of where the representation was relied or acted upon, several options again present themselves. Reliance might be said to have occurred the moment the first plaintiff accepted the representation in the PPM as being true. That occurred when that representation was apparently confirmed at the manipulated FATs. On that analysis, reliance took place where the FATs took place, ie in Germany. Alternatively, it might be said that reliance took place when the first plaintiff – first on behalf of itself in 2000 and then on behalf of Vintergas in 2001 – approved the MAN Engine for installation in the six ships. On that analysis, reliance took place where that decision was taken, probably Norway where the first plaintiff is located. Or it might be said that reliance occurred only when the ships were physically delivered to and accepted by Somargas HK. The location of reliance could then be China, where the shipyard is located, or Hong Kong where Somargas HK is incorporated. Or it might also be said, as the plaintiffs say, that reliance was and is continuing, and therefore continues to occur so long as loss continues to be suffered. In that case, Singapore enters the picture. Loss continued to be suffered by Somargas SG when it took ownership of the ships in February 2011 and put them under the Singapore flag in March 2011.

104 Applying the substance test, I take the view that Germany is the place where the tort occurred. It is true that the PPM and the FSI were both authored in Germany. But that is, to my mind, a marginal factor in favour of Germany. The weightiest factor pointing to Germany is that Germany is where the FATs took place. It is the representation at each FAT which are at the core of the plaintiffs’ loss, not the representation in the PPM or in the FSI. The representation in the PPM and the FSI could not be said to be untrue until the FATs took place. Further, it is only because the FATs were manipulated that the plaintiff was prevented from knowing that the FATs had proven the representation in the PPM and the FSI to be false. The misrepresentation as to the true fuel efficiency of each of the six MAN Engines was therefore made in Germany, received by the plaintiffs in Germany and relied upon in Germany.

105 It is true, of course, that the first plaintiff relied on the representation in the PPM and the FSI at a much earlier point in time: when it approved the MAN Engine for installation in the six ships in 2000 and 2001. But it was the six manipulated FATs which induced the first plaintiff to accept the six specific engines which the defendants actually supplied to the shipbuilders and which the shipbuilders actually installed in each of the six ships. On the plaintiffs’ case, if the FATs had not been manipulated, the FATs would have revealed the representation in the PPM and the FSI to be a misrepresentation, the plaintiffs would have rejected each engine, and the ships would have been fitted with other engines which consumed fuel at the rate represented in the PPM and FSI.

106 In contrast, the places where the representation in the PPM and the FSI was actually communicated to and received by the first plaintiff are too far removed from the substance of this tort to say that any one of them is in substance the place where the tort occurred. The place at which the PPM and FSI were handed over to the first plaintiff is, to my mind, fortuitous. Given that both the first plaintiff and the first defendant operate around the world, these documents could have been handed over to the first plaintiff in any jurisdiction in the world, or even in no jurisdiction in the world, eg on a private jet in international airspace or on the high seas. In that sense, Norway’s connection to the tort on the facts of this case is tenuous in that it is simply the first plaintiff’s principal place of business. So too, China’s and Singapore’s claim to be the location of the tort on the facts of this case is tenuous.

107 I hold, therefore, that Germany is the loci delicti. German law is therefore the lex loci delicti.

108 The question of whether the tort is actionable under the lex loci delicti was argued before the assistant registrar. He considered the expert opinions filed by each side, and took the view that it was. I agree.

109 Both parties’ experts agree that the plaintiffs’ claim would be actionable against the defendants in the German law of tort under §826 of the German Civil Code (“BGB”). They agree that the German court could take jurisdiction over such an action ¹²³ and that it would apply German tort law to determine the plaintiffs’ claim. ¹²⁴ They further agree that, if German tort law were to be applied, the plaintiffs would be entitled to recover damages against the first defendant pursuant to §826 BGB if they could establish that: (i) the manipulation of the FATs was intentional; (ii) that the defendants knew of the importance of the results of the FATs for users of the MAN Engine in the plaintiffs’ position; and (iii) the defendants acted recklessly by providing test results which could be affected by potential irregularities. ¹²⁵

110 On appeal before me, neither party has made any new arguments on German law. The plaintiffs have briefly made the point that German law is not the lex loci delicti, ¹²⁶ but that is a separate point that does not go to what the situation would be if, as I have just held, German law is indeed the lex loci delicti. The new expert evidence adduced before me pertains to whether German law prohibits the transfer of data overseas, thus hindering Singapore courts from considering all relevant evidence and in turn raising questions as to whether Singapore is the forum conveniens. That is an issue I deal with later.

111 There is a potential argument, however, that the plaintiffs’ claim is no longer actionable under German law because the German limitation period has expired. After the plaintiffs discovered that they had a potential cause of action against the defendants, the parties entered into a series of agreements whereby the defendants waived the benefit of the German limitation period. The last waiver expired on 31 March 2015. ¹²⁷ It could be argued that the lex loci delicti limb of the double actionability rule is not today satisfied because the plaintiffs’ claim is today no longer actionable under German law because it is time-barred.

112 That argument, to my mind, would be misconceived. That is no doubt why the defendants have not raised it. It is obvious to me that the lex loci delicti limb of the double actionability rule is to be applied as at the date on which the action is commenced in the forum, and not as at the date on which the forum considers the question of double actionability. If it were the latter, the outcome of the double actionability rule would no longer depend on any matter of substance but on accidents of scheduling. The plaintiffs commenced this action on 28 January 2015. ¹²⁸ That was well before the last waiver of the German limitation period expired. I accept that, for the purposes of the double actionability rule, the wrong of which the plaintiffs complain in this action are actionable under German law.

113 I turn now to consider whether the plaintiffs’ claim is actionable in tort under Singapore law. Before proceeding, it is necessary to clarify what the lex fori limb of the double actionability rule requires. This limb asks the counterfactual question whether the acts constituting the plaintiffs’ claim, if all of those acts were carried out in Singapore, would constitute a tort under Singapore law. It does not ask the factual question whether the defendants’ acts – done as they were here across a range of countries, with only some acts done in Singapore – constitutes a tort under Singapore law. This much is clear from well-established authority.

114 To begin, the leading textbook, Dicey and Morris on the Conflict of Laws (Lawrence Collins gen ed) (12th Ed, 1993) at pp 1487 to 1488 expresses the view that the lex fori limb requires assuming that all of the acts constituting the tort were done in the forum:

115 This is the position also in Singapore. In Halsbury’s Laws of Singapore vol 6(2) (LexisNexis, 2016 Reissue) (“Halsbury’s Laws of Singapore”) at para 75.374, the editor observes that “[for] a tort to be actionable under the law of the forum, the claim must be made out in all its elements under the domestic law of Singapore, as if the wrong had been committed in the forum” [emphasis added]. Similarly, Singapore Civil Procedure 2018 vol 1 (Foo Chee Hock JC gen ed) (Sweet & Maxwell 2018) indicates at para 11/1/30 that where the double actionability rule applies, the Court should give leave under Order 11 r 1(f) “only if the act complained of is one which, if done here, would be a tort, and which is also actionable according to the law of the foreign country where it has been done” [emphasis added].

116 This approach has also been applied in case law. In Kuwait Airways Corporation v Iraqi Airways Company and others (Nos 4 and 5) [2002] 2 AC 883 (“Kuwait Airways”), the Iraqi government took ten aircraft owned by Kuwait Airways Corporation (“KAC”) from Kuwait, flew them to Iraq and purported to vest ownership of the aircraft in Iraqi Airways Co (“IAC”). Through no act of its own, therefore, IAC found itself in possession and control of several aircraft which KAC claimed continued to be its own property. KAC brought a claim in England against IAC for conversion. The House of Lords applied the double actionability rule, first considering whether IAC’s acts amounted to conversion in English law and then considering whether IAC was liable under Iraqi law for the civil wrong of usurpation: Kuwait Airways at [37] and [45].

117 Lord Nicholl’s analysis of actionability under English law in Kuwait Airways is of chief concern here and is instructive. The acts by IAC which were said to constitute a tort took place entirely overseas, and only in Iraq. He identified the three material acts as: (i) being in possession and control of the aircraft; (ii) intending to keep the aircraft as its own; and (iii) treating the aircraft as its own. He considered that by doing those acts, IAC was asserting rights inconsistent with KAC’s rights as owner: Kuwait Airways at [43]. He then concluded that “IAC’s acts would have been tortious if done in this country” [emphasis added]: Kuwait Airways at [44].

118 The proper approach in the present case, therefore, is to consider whether the tort – which I have found as a matter of general substance occurred in Germany – would constitute a tort in Singapore law if the acts said to constitute the tort had been carried out entirely in Singapore. In Kuwait Airways, that counterfactual hypothetical was easy to construct and analyse. All of IAC’s acts had taken place in only one country, Iraq. Constructing the hypothetical simply involved swapping Iraq for England.

119 In this case, the acts complained of took place in a multiplicity of jurisdictions: at the very least, Germany, Denmark, Norway, China, Hong Kong and Singapore. It might be said that it is overly artificial simply to sweep away this multiplicity of jurisdictions and treat all of the acts as having taken place in Singapore. But that is, to my mind, what the lex fori limb of the double actionability rule requires. I therefore proceed on this limb by assuming that all the acts complained of took place in Singapore.

120 That does not mean, however, that I should ignore the fact that the acts complained of involved separate legal entities. The lex fori limb requires the court to assume, contrary to the facts, that all of the acts took place in Singapore but nothing else. But in applying the lex fori limb, the court must continue to acknowledge that distinct legal entities were involved. This approach allows me to be both consistent with the rule and also not be overly artificial. It would be wholly unrealistic to pretend that the representation was not only made, received and relied upon in Singapore, but also that it was also made by only one party and received by, relied upon by and damaged only one party. The interposition of separate legal entities on the facts of this case raises its particular difficulty in the context of the tort of misrepresentation. Even if it is assumed that all the acts which constitute the tort took place in Singapore, it remains the case that the same entity did not receive, rely upon and suffer loss by reason of the misrepresentation.

121 I now consider elements of both fraudulent and negligent misrepresentation. The assistant registrar has helpfully set out the relevant elements for each tort in Skaugen at [65]–[66], and I adopt them here:

122 I deal with the elements common to both torts first. The first common element is that there be a false representation of fact. I have no hesitation in finding that the plaintiffs have established a good arguable case on this aspect for the same reasons as the assistant registrar did below. The PPM and the FSI represented that the MAN Engine consumed fuel at a certain rate under certain specified conditions. The results of the FATs confirmed that the six specific MAN Engines constructed for the Skaugen group consumed fuel at no worse than the rate represented. The first defendant accepts that there are indications that the fuel consumption measurements for three of the six engines were externally influenced in an improper manner during the FATs with the result that the fuel consumption rate specified in the PPM and the contracts were exceeded for those engines. ¹²⁹ The three remaining MAN Engines were of an identical design. They were manufactured at the same factory in Germany. The FATs were carried out on them on the same compromised test equipment at that factory. The plaintiffs have therefore established a good arguable case that the other three engines similarly consume fuel at a rate which exceeds that set out in the PPM and confirmed in the FSI and the FAT results. Through a combination of concession and inference, the plaintiffs have therefore established a good arguable case that the representation in the PPM, repeated in the FSI and at the FATs, were false in point of fact for all six of the MAN Engines.

123 The next common element asks whether there was inducement and reliance. The plaintiffs’ case is that the actionable misrepresentation is the rate set out in the PPM as the rate at which the MAN Engine consumes fuel under the conditions also set out in the PPM. That representation was then maintained and confirmed by the FSI and in the results of the six FATs. I therefore agree with the assistant registrar that it would be artificial to consider the misrepresentation on a document-by-document basis.

124 The defendants make the point that the alleged misrepresentation was acted upon only by the first plaintiff, when it was induced by the FATs to accept the six engines. ¹³⁰ This is because the subsequent owners of the ships were not the intended recipients of the misrepresentation, ¹³¹ and also because, by the time Somargas SG came to own the ships it could no longer be said to be relying on a manufacturer’s representation made several years earlier, before the ships went into actual operation. By the time it became the owner of the ships, even if it did not know that the representation was a misrepresentation, it knew and was content with the rate at which the engines actually consumed fuel in actual daily operation. ¹³²

125 It seems to me, however, that subsequent owners of the six ships might also be said to have relied on the misrepresentation. I agree with the reasoning of the assistant registrar in Skaugen at [77] that the facts, viewed holistically, suggest that the defendants knew that the engines were to be installed on ships which were not to be owned by any one corporate entity in particular but to be deployed generally within the Skaugen group. So long as the entities which owned the ships and which have assigned their claims to the plaintiffs are part of the Skaugen group, I consider there to be a good arguable case that they were within the class of persons to whom the representation was made.

126 It is also not clear to me that the effect of the first defendant’s misrepresentation was displaced by knowledge of the ships’ actual fuel consumption. The misrepresentation was in relation to the rate at which the MAN Engine consumed fuel under certain specified conditions which the FATs were intended to replicate. The fact that the actual fuel consumption, once the engines were in operation, was different from the figure represented in the PPM does not by itself mean that the misrepresentation as to fuel consumption under the specified conditions became inoperative.

127 The last common element is damage. It seems to me that the plaintiffs have also established a good arguable case on this element. The ship owners who have assigned their claims to the plaintiff have suffered loss either because they had to purchase more fuel to operate the ships than they would have but for the misrepresentation, or because the market value of each ship is diminished by the confirmation that its engine’s fuel efficiency under the specified conditions is below the specification in the PPM, quite apart from the engine’s actual fuel efficiency under actual operating conditions.

128  As for the elements specific to each species of misrepresentation, I do not propose to go into these further because I substantially agree with the reasoning of the assistant registrar in Skaugen at [82]–[83].

129 I therefore conclude that the plaintiffs have established a good arguable case that the acts of the defendants which the plaintiffs allege to be tortious would have been actionable under Singapore law had they been committed in Singapore.

130 The double actionability rule is subject to a flexible exception. I hold that there is a good arguable case that the flexible exception does not apply. As explained by the Court of Appeal in Rickshaw Investments at [57], the exception comes into play when: (i) the parties and other connecting factors have nothing to do with the place where the tort was committed; (ii) have nothing to do with the place where the place the action is brought, or (iii) have nothing to do with both. If the place where the tort was committed is purely fortuitous, then the flexible exception does not require that the tort be actionable under the lex loci delicti. And similarly, if the action is brought in Singapore purely fortuitously, the flexible exception does not require the acts be actionable under the lex fori. Further, if both the place where the tort was committed, and Singapore as the forum, are purely fortuitous, it is possible that a third possible law might govern the action concerned.

131 Here, the tort was committed substantially in Germany. Singapore is also a place where current and former owners of the ships are located. They, together with all other owners of the ships, have assigned their claims to the plaintiffs. There does not seem to be any “clear or satisfying grounds” to displace either or both of these places for the purposes of the double actionability rule. That is the threshold for invoking the flexible exception: Rickshaw Investments at [65]. There is therefore a good arguable case that the flexible exception does not apply.

132 I now turn to consider the particular requirements of each of the two heads of jurisdiction under Order 11 r 1(f) on which the plaintiffs rely.

133 The first head on which the plaintiffs rely is Order 11 r 1(f)(i):

134 Before the assistant registrar, the plaintiffs argued that the words “a tort…which is constituted…by an act or omission in Singapore” are wide enough to cover both acts or omissions by a tortfeasor as well as acts or omissions by the victim of the tort. ¹³³ The plaintiffs therefore argued that the tort of misrepresentation in this case was constituted by their own acts of reliance. ¹³⁴ They submitted, further, that those acts of reliance took place in Singapore. Thus, the plaintiffs argued, Order 11 r 1(f)(i) was satisfied.

135 The defendants contended before the assistant registrar that Order 11 r 1(f)(i) considers only the acts or omissions by a tortfeasor which are said to constitute the tort and not the acts or omissions of the victim of the tort. The defendants relied on the decision of the English Court of Appeal in Metall und Rohstoff AG v Donaldson Lufkin & Jenrette Inc and another [1990] 1 QB 391 (“Metall und Rohstoff AG”) which interpreted the analogous provision in Order 11 of the English Rules of the Supreme Court 1965 (SI 1965 No 776) (UK). ¹³⁵ The defendants submitted to the assistant registrar that, if the only act or omission which is relevant under Order 11 r 1(f)(i) is the tortfeasor’s act or omission, then the only relevant act in this case is the defendants’ alleged misrepresentation. ¹³⁶ That, they submit, could not have taken place in Singapore on any view of the facts. The assistant registrar was persuaded by this argument, and ruled in the defendants’ favour on this point: Skaugen at [97]–[99]. If the same point had been presented to me in the same way, I would have ruled in the same way.

136 Before me, however, the plaintiffs advance a different point on Order 11 r 1(f)(i). Instead of relying on their own act of reliance as constituting the tort and bringing the case within Order 11 r 1(f)(i), they now rely on the defendants’ omission to rectify the misrepresentation as constituting the tort. Thus, the plaintiffs’ argument now is that the defendants failed to rectify their misrepresentation between July 2000 and June 2012 ¹³⁷ and that it is that omission which constituted the tort. They submit further that the omission is located in Singapore. For this latter submission, the plaintiffs rely on two things which happened between 2000 and 2012. First, in late 2004, the plaintiffs moved their operations to Singapore. ¹³⁸ And second, from February 2011 until April 2013, Somargas SG – a Singapore company – owned the six ships. ¹³⁹

137 I am not persuaded that the defendants’ omission to rectify the misrepresentation can be said to be located in Singapore for the purpose of Order 11 r 1(f)(i). I assume, only for the sake of argument, that the omission to rectify the misrepresentation was in this case an omission which constituted the tort. Even then, it seems to me impossible to say that that omission had its location in Singapore. I say that for two reasons. First of all, it appears to me that the most natural place in which an omission is located is where the person who has failed to act is located, not where the effect of the failure to act is felt. In this case, it would be far more natural to say that the omission took place in Germany than to say it took place in Singapore.

138 My second reason, however, is far more fundamental. An act, by its very nature, is positive. It therefore always has a factually-ascertainable location. This is true even if there is some difficulty in making a second-order finding of fact as to what precisely that location is. One example would be in determining in which country a man fires a gun if he has one foot on either side of an international border when he does so. But an omission is by definition a failure to act. An omission is by its nature negative. A failure to act has significance only because the law attaches a significance to it, usually because the failure to act comes in the face of a duty to act. Unlike an act, an omission has no independent, physical significance of its own and, accordingly, has no factually-ascertainable location. As the Victorian Court of Appeal has observed, “it makes no sense to speak of the place of an omission”: Puttick (as Executor of the Estate of Russell Simon Puttick) v Fletcher Challenge Forests Pty Ltd [2007] 18 VR 70 at [16]. An omission occurs both nowhere and everywhere.

139 In my view, a mere failure to correct an earlier misrepresentation to a person located in Singapore who has received or relied on that misrepresentation does not satisfy Order 11 r 1(f)(i). If it were otherwise, because of the negative nature of an omission, the scope of Order 11 r 1(f)(i) would be far too broad. That, in turn, would defeat the purpose of the Order 11 heads of jurisdiction, which is to ensure that jurisdiction over a foreign defendant is not too easily exercised. It is also the case that the plaintiff’s claim against the defendants as pleaded is based only on an actionable misrepresentation in 2000 and not on an actionable failure to remedy the misrepresentation between 2000 and 2012.

140 I therefore hold that the plaintiffs are unable to bring their claim within Order 11 r 1(f)(i).

141 In the alternative, the plaintiffs say that they are within Order 11 r 1(f)(ii) because they bring this action to recover damages for damage suffered in Singapore:

142 The plaintiffs say that the defendants’ tort caused damage to be suffered in Singapore for two reasons, correlated to the two types of damage which flow from the defendants’ misrepresentation. The first, and most obvious type of damage, is the additional fuel costs which the owners of the six ships incurred over the years because the MAN Engines were not as fuel efficient as they were represented to be. ¹⁴⁰ Somargas SG incurred this type of damage and is a Singapore company. The GATX subsidiaries also suffered this type of damage and are also Singapore companies. ¹⁴¹ This type of damage was therefore suffered in Singapore.

143 The second, and less obvious type of damage, is that the capital value of the six ships on the open market is depressed because any arm’s-length buyer will factor in the future fuel costs caused by the fuel-inefficient MAN Engines in valuing the ships. ¹⁴² Somargas SG suffered this type of loss when it acquired the ships from Somargas HK. The value for the six ships in that transfer was fixed by a straight-line depreciation of their purchase price rather than by assessing the ships’ actual value on the open-market. The open market value of the ships would have reflected the lower value of the ships by reason of their fuel inefficient engines. ¹⁴³ The GATX subsidiaries, in turn, incurred this type of damage when they acquired three of the ships from Somargas SG. Again, in that sale the three ships were valued on a straight-line depreciation rather than at their open market value. ¹⁴⁴ Somargas SG and the three GATX subsidiaries are incorporated in Singapore. On both types of damage, therefore, the plaintiffs say the damage was suffered in Singapore.

144 In response, the defendants say that the damage occurred when the misrepresentation was acted upon. That occurred when the MAN engines were accepted after the FATs were conducted, or, at the latest, when the ships were delivered to and accepted by Somargas HK. No entities who owned the ships after that point can be said to have suffered damage because none of those entities acted upon the misrepresentation. ¹⁴⁵ This would therefore exclude all of the Singapore entities, including Somargas SG, who are all further down the chain from Somargas HK. Further, the defendants argue that the inquiry is limited only to direct damage suffered in reliance on the alleged misrepresentation. ¹⁴⁶

145 I do not accept the defendants’ argument. On the facts of this case, the cause of action in tort which the plaintiffs now pursue was not complete when the first plaintiff accepted the engines in Germany. At that point, although there was in place misrepresentation and reliance, there was no damage whatsoever. Indeed, at any point in time from the time the first plaintiff accepted the engines after each FAT until Somargas HK took delivery of the ships, the plaintiffs could not have pursued the defendants for the cause of action which they are now pursuing. That is because the plaintiffs’ decision to accept the engines remained in point of fact and as a matter of law reversible up until that time. If the plaintiffs had found out the truth at any time before the shipbuilders tendered delivery of the ships to Somargas HK, they could have substituted fuel-efficient engines, holding the defendants liable for any delay, and avoided the two types of damage which they now claim they have suffered.

146 It is true that Somargas HK did suffer damage when it took delivery of the ships. By that act alone, Somargas HK became the owner of ships which had a lower market value than they would have had if the misrepresentation had never been made and if they had been installed with fuel efficient engines instead. But the damage suffered by Somargas HK upon delivery does not, in itself, preclude future damage. And it is certainly the case that the owners of the ships thereafter did suffer damage, at the very least by increased fuel expenditure which they incurred and which they would not have incurred if the misrepresentation had never been made.

147 Nor do the defendants’ contentions on the directness of damage really assist them. As I noted at [125] above, the plaintiffs have made a good arguable case that the misrepresentation was not only intended to be relied upon by the first plaintiff, but also by the subsequent owners of the ships. If this is the case, then these owners would also have suffered damage that results directly from the misrepresentation. Even if that is not the case, however, I agree with the assistant registrar’s reasoning in Skaugen at [103] that direct damage is not required under Order 11 r 1 (f)(ii). The text of Order 11 r 1(f)(ii) requires only that the claim is at least “…partly founded on…damage suffered in Singapore”. It nowhere speaks of a requirement that the damage suffered in Singapore must arise directly from the tort. The owners of the ships have suffered damage in the form of the increased fuel expenditure and the diminished capital value of the ships.

148 The plaintiffs have therefore established a good arguable case that their claim falls within Order 11 r 1(f)(ii).

149 I turn now to Order 11 r 1(p), which is the last head of jurisdiction on which the plaintiffs rely:

150 The plaintiffs’ argument on Order 11 r 1(p) has evolved. Before the assistant registrar, the plaintiffs argued that: (i) the substance test approved in JIO Minerals applies to determine whether a cause of action arose in Singapore within the meaning of Order 11 r 1(p); ¹⁴⁷ and (ii) that there was a good arguable case that that test applied to the facts of this case pointed to Singapore as the place where the plaintiffs’ cause of action arose. Before me, however, the plaintiffs advance a new argument. They thereby implicitly accept that the substance test does not point to Singapore, as both the assistant registrar and I have found.

151 The argument which the plaintiffs now advance on Order 11 r 1(p) runs as follows. The test to determine whether a claim is founded on a cause of action which arose in Singapore within the meaning of Order 11 r 1(p) is not the substance test approved in JIO Minerals but the test set out in the decision of the Privy Council in Distillers Co (Biochemicals) Ltd v Laura Anne Thompson [1971] AC 458 (“Distillers”). Under the Distillers test, a claim falls within Order 11 r 1(p) if the plaintiffs’ “cause of complaint” is located in Singapore. There is a good arguable case on the facts of this case that the “cause of complaint” for at least some of the plaintiffs’ assignors is located in Singapore because that is where those assignors are incorporated and accordingly where those assignors suffered the damage complained of in this action. ¹⁴⁸

152 The defendants’ submission is that the substance test from JIO Minerals applies to Order 11 r 1(p). When that test is applied to the facts of this case, the defendants submit, it does not on any view of the facts point to Singapore as the place where the cause of action arose. ¹⁴⁹

153 The crux of the plaintiffs’ case on Order 11 r 1(p) turns on the appropriate test to be applied. If the defendants are right and the substance test from JIO Minerals applies, the analysis which I have already undertaken establishes that Germany is not only the loci delicti but also the place where the causes of action arose. Conversely, if the plaintiffs are right and the Distillers test is a different test from the substance test, capable of yielding a different result, the plaintiffs’ failure to establish that Singapore is the loci delicti does not in itself mean that it cannot bring itself within Order 11 r 1(p).

154 Central to this aspect of the plaintiffs’ case is the Privy Council decision in Distillers. The first point I make is that the Court of Appeal in JIO Minerals in fact cited Distillers (at [90]) as authority for the substance test. It could thus be argued that the plaintiffs’ submission before me – distinguishing between the substance test and the Distillers test for the purposes of Order 11 r 1(p) – is based on a false dichotomy. However, JIO Minerals is not a case on Order 11 r 1(p). It is a case which considers double actionability in the context of an application to stay proceedings at common law on grounds of forum non conveniens. Distillers, on the other hand, is a case in which the Privy Council had to interpret and apply a statutory provision which is in pari materia with Order 11 r 1(p). Indeed, the statutory provision considered in Distillers and its successor procedural rule in New South Wales is very closely connected to our Order 11 r 1(p): see Singapore Civil Procedure 2018 vol 1 (Foo Chee Hock JC gen ed) (Sweet & Maxwell 2018), paragraph 11/1/40.

155 It therefore appears to me that JIO Minerals is not authority that the substance test used to determine the loci delicti when applying the double actionability rule is equally to be used to determine whether a claim is founded on a cause of action arising in Singapore within the meaning of Order 11 r 1(p). The two concepts are quite different. The place of a tort is a question of weighing the facts. A cause of action is a legal construct under Singapore law. Although determining whether a cause of action arose in Singapore also involves weighing the facts, it requires the court to weigh and view those facts not in isolation but through that legal construct. To put it another way, the inquiry as to the loci delicti starts with the facts and weighs them in order to identify the lex loci delicti and then to ascertain whether the facts in question give rise to a claim known to the lex loci delicti. The inquiry under Order 11 r 1(p) starts instead with the actual cause of action which the plaintiff has chosen to plead under Singapore law and asks whether that specific cause of action arose in Singapore.

156 I turn now to consider the facts of Distillers.

157 Distillers came to the Privy Council from the Court of Appeal of New South Wales. In Distillers, an English drug company manufactured a drug in England which contained thalidomide. Thalidomide causes birth defects when taken by pregnant women. The drug company sold the drug to a distributor in Australia. The drug company knew or ought to have known that the drug posed a risk to pregnant women. But it did not warn the distributor of the risk or place any warnings on the packaging in which it sold the drug to the distributor. The plaintiff and her mother were both domiciled in New South Wales. The plaintiff’s mother, while she was pregnant with the plaintiff, took the drug in New South Wales. The plaintiff was later born with serious birth defects.

158 The plaintiff sued the drug company in New South Wales in the tort of negligence. She served the writ on the drug company in England under s 18(4)(a) of New South Wales’ Common Law Procedure Act (No 21 of 1899) (NSW). That provision permits service of a New South Wales writ out of the jurisdiction if the judge is satisfied that “there is a cause of action which arose within the jurisdiction”. The drug company entered a conditional appearance and immediately applied to the New South Wales court to set aside service on the grounds that the plaintiff’s cause of action against the drug company did not arise in New South Wales.

159 The Privy Council considered (at 466C-D) that there were three possible theories as to how s 18(4)(a) should be interpreted. The first was that “cause of action” must mean “the whole cause of action, so that every part of it, every ingredient of it, must have occurred within the jurisdiction”. The second theory was that “it is necessary and sufficient that the last ingredient of the cause of action, the event which completes a cause of action and brings it into being, has occurred within the jurisdiction”. The third and final theory is that “the act on the part of the defendant which gives the plaintiff his cause of complaint must have occurred within the jurisdiction”. The Privy Council held (at 467A) that the third theory was correct.

160 The Privy Council dismissed the first theory out of hand as being too narrow. That view would deny the courts of a country jurisdiction unless all the ingredients of the cause of action also occurred within that country. The Privy Council considered that approach to be too restrictive for modern times.

161 The Privy Council also rejected the second theory. It held that the place where the last event which completed the cause of action occurred might be quite fortuitous, and therefore should not be the sole determinant of jurisdiction: Distillers at 468A. It would not be just and reasonable to ask the defendant to answer for his wrongdoing in whatever country in the world the plaintiff happened to be in when the tort became complete: Distillers at 468D.

162 This left the last theory, which proposes that a cause of action arises within the jurisdiction if the act on the part of the defendant which gives the plaintiff his “cause of complaint” occurs within the jurisdiction: Distillers at 468E. The Privy Council considered this to be an inherently reasonable rule, as the defendant would be called upon to answer for his wrong only in the courts of the country where he did the wrong (at 468F).

163 The Privy Council explained what it meant by “cause of complaint”. It observed at 468G that “[the] defendant does not merely by behaving negligently give the plaintiff any cause of complaint in law.” Rather, “[the] plaintiff has such cause of complaint if the defendant’s negligence has caused damage to the plaintiff.” The Privy Council recognised there could be considerable separation in time and place between the negligent act of the defendant and the resulting damage to the plaintiff: Distillers at 468H to 469A. Ultimately, however, the Privy Council did not have to deal with that difficulty on the facts of Distillers. The Privy Council accepted that the negligent act and the resulting damage both occurred in New South Wales. The drug company’s negligent act was the failure to warn pregnant women in New South Wales of the risks of taking the drug while pregnant. And the plaintiff’s damage obviously arose in New South Wales. The plaintiff’s cause of action therefore did arise in New South Wales.

164 It is apparent from the reasoning of the Privy Council that the negligent act in this case was actually an omission (at 469D). But, as I have pointed out at [138] above, it is a nonsense to speak of an omission, without more, as being localised in a specific place. On the facts of Distillers, what localised the drug company’s omission to warn in New South Wales were three plaintiff-centric factors: (a) New South Wales was where the plaintiff’s mother purchased the drug, bereft of a warning; (b) New South Wales was where the plaintiff’s mother consumed the drug while pregnant with the plaintiff, in ignorance of the risk which the drug posed to the plaintiff in utero; and (c) New South Wales was where, on the assumptions on which the case was decided, the plaintiff suffered damage in utero.

165 The plaintiff-centric nature of the test in Distillers is apparent from the Privy Council’s focus on the “cause of complaint”. According to the Privy Council, the “cause of complaint” is neither simply the defendant’s tortious act nor equally simply that which completes the cause of action. The “cause of complaint” is a combination of all of the factors, but which trains the court’s focus on the gist of the plaintiff’s complaint in the action. That would obviously include the type of harm which the plaintiff complains of and seeks to remedy.

166 The effect of Distillers, therefore, is to adopt a more plaintiff-centric approach on Order 11 r 1(p) than the substance test for the lex loci delicti limb of the double actionability rule set out in JIO Minerals. That is also the effect of the Privy Council’s emphasis on the plaintiff’s “cause of complaint”. The more plaintiff-centric approach is not surprising, because Order 11 r 1(p) requires the court to view the facts of the case through the cause of action which the plaintiff has sought to invoke. It is significant to me that Distillers (at 468E) mandates “look[ing] back over the series of events constituting [the tort] and ask[ing] the question, where in substance did this cause of action arise” [emphasis added]. The question posed by the Privy Council is not the more general and more factual question “where in substance did the tort take place”.

167 On the facts of this case, the plaintiffs and their assignors had no cause of complaint simply because the first defendant made a misrepresentation in the PPM and the first plaintiff relied upon it. As soon as the assignors suffered damage by reason of the misrepresentation, they would have a complete cause of action against the defendants in misrepresentation. But even then, their “cause of complaint” would not be fixed. It would vary, depending on when they discovered the misrepresentation and the type of damage they were seeking compensation for by the cause of action asserted. Thus, for example, if the first plaintiff had discovered the misrepresentation after the first plaintiff had accepted the results of an FAT on a particular MAN Engine but before that engine had been delivered and installed in one of the ships, its cause of complaint would be that the construction and delivery of the ships would be delayed while the plaintiff sourced substitute engines with the required fuel efficiency. The cause of action in this hypothetical scenario would be misrepresentation, ie the same cause of action as the one which the plaintiffs pursue in this action on behalf of their assignors. But the cause of complaint – or the gist of the plaintiff’s complaint – would be quite different.

168 The cause of complaint in this action includes the increased expenditure on fuel incurred by the plaintiffs’ assignors. That complaint is continuing in nature. The plaintiffs’ assignors continued to suffer that harm so long as they owned the ships and bore that increased expenditure. The plaintiffs’ assignors who suffered loss of this type include entities located in Singapore. The plaintiffs’ “cause of complaint” occurred in Singapore. And therefore their cause of action within the meaning of Order 11 r 1(p) arose in Singapore. The plaintiffs have thus made a good arguable case that Order 11 r 1(p) is satisfied.

169 Although it is not strictly necessary for me to do so, I should make clear that the approach I have adopted does not propose that the place where the last element of a tort occurs is the place where the cause of action based on that tort arises. That is simply the second theory which the Privy Council in Distillers rightly rejected as being capable of leading to arbitrary results. An example of those arbitrary results – based on an example given in Distillers itself (at 468A) – would be if the plaintiff’s mother had bought the drug in New South Wales from a New South Wales drug company, travelled to Singapore with the drug for a short holiday while she was pregnant, consumed the drug in Singapore causing damage to the plaintiff in utero in Singapore and then returned to New South Wales. On the second discredited theory considered in Distillers, the cause of action in this example would have arisen in Singapore, and only in Singapore. That is not the approach which Distillers advocated or which I have adopted here. Further, the fact that damage in this case was suffered in Singapore may not have been anticipated by the defendants – or even by the plaintiffs – at the time the MAN Engines were selected, tested, accepted and installed. But the fact that damage was suffered in Singapore is not fortuitous. The Skaugen group moved the ownership of the ships to Singapore in the course of their business, for genuine and long-term commercial reasons.

170 The final step in the analysis is to determine whether this was a proper case for leave to be granted to the plaintiffs to serve the writ out of Singapore within the meaning of Order 11 r 2(2). The test to be applied to determine this is the same test which is to be applied when considering whether to stay proceedings commenced in Singapore in which the defendant has been served within the jurisdiction, without leave and as of right, on grounds of forum non conveniens. Order 11 r 2(2) will therefore be satisfied if the plaintiff is able to establish that Singapore is clearly the appropriate forum in which to resolve the parties’ dispute (Spiliada at 481D).

171 Before I turn to apply the Spiliada test to the facts of this case, however, I deal with a preliminary issue raised by the defendants.

172 The defendants contend that, on an application to set aside service out of the jurisdiction, the Spiliada test is to be applied by reference only to the factual position which obtained when leave to serve out of the jurisdiction was granted. That contention means, on the facts of the defendants’ applications, that I would have to disregard events which took place after the plaintiffs were granted leave to serve the defendants in Germany and Norway respectively in March 2015. ¹⁵⁰

173 The event which occurred after March 2015 and which is relevant for present purposes is that, in August 2015, the plaintiffs commenced proceedings in Norway against the defendants on the very dispute which is the subject-matter of this action. ¹⁵¹ Those proceedings are relevant to the Spiliada test not just because it amounts to lis alibi pendens, but also because once the courts of Norway were seised of the dispute, the courts of Germany were obliged by Art 27 of the Lugano Convention (otherwise known as the Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters (Lugano) 28 ILM 620 (1989)) to cede jurisdiction over this dispute in favour of the court first seised ie, the courts of Norway. ¹⁵² In other words, once the plaintiffs commenced the Norwegian proceedings, the courts of Germany ceased to be available to the parties as a possible forum for resolving this dispute. That is an important factor because both the assistant registrar and I have found that the loci delicti is Germany and the lex loci delicti is accordingly German law.

174 Whether a particular case is a proper one for service of originating process out of the jurisdiction within the meaning of Order 11 r 2(2) is determined by the Spiliada test (Spiliada at 481A). Satisfying the Spiliada test is therefore an integral part of the plaintiff’s burden under Order 11, both when applying ex parte for leave and also when resisting a defendant’s inter partes application to set aside leave.

175 The substance of the Spiliada test is “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” (Spiliada at 480G). The forum identified by the Spiliada test is the forum conveniens or, in English, the appropriate forum. As the Court of Appeal observed in Zoom at [30], the substance of the Spiliada test is the same whether it is applied under Order 11 r 2(2) where a defendant is to be or has been served out of the jurisdiction with the leave of court or it is applied on a defendant’s application for a discretionary stay of proceedings where a defendant has been served within the jurisdiction as of right. The only difference between those two situations lies in who bears the burden on the application. In the former situation, the burden is on the plaintiff to satisfy the court that Singapore is clearly the appropriate forum (Spiliada at 481E). In the latter situation, the burden is on the defendant to satisfy the court that some other forum is clearly and distinctly more appropriate than Singapore (Spiliada at 477E).

176 The logical consequence of this approach is that, if a defendant fails in its application to set aside service out of the jurisdiction, that necessarily means that the plaintiff has discharged its burden under Order 11 r 2(2) and established that Singapore is indeed clearly the appropriate forum. Any application which the defendant brings at the same time and on the same evidence seeking a discretionary stay of the action on grounds of forum non conveniens must necessarily fail.

177 The Court of Appeal elaborates on this observation in Zoom at [78], noting that “if the substance of the Spiliada test does not differ according to the nature of the application, then a foreign defendant who seeks the setting aside of an overseas service leave order and prays as a fall-back for a stay of proceedings on improper forum grounds, as was done here, will find that his stay application is pointless”. It concludes its observations on this matter by pointing out that “it is in fact wholly unnecessary and likely counter-productive for a foreign defendant who does not wish to have his dispute with the plaintiff tried in the local court to make both a jurisdictional challenge and a stay application based on the same material” [emphasis in original]: Zoom at [79]. Of course, a standalone stay application brought at a later time and on different evidence may succeed: Zoom at [30].

178 As was the case in Zoom, each defendant here prays in its application both to set aside service and, alternatively, for a stay of this action on grounds of forum non conveniens. ¹⁵³ These two prayers of the defendants’ applications, having been filed at the same time and on the same evidential material, would ordinarily stand or fall together.

179 But the defendants’ preliminary issue adds another string to their bow. If the defendants are correct, their prayer to set aside service must be determined on the factual position as at March 2015, when the plaintiffs secured leave ex parte, while their prayer to stay this action on grounds of forum non conveniens must be determined on the factual position now. That difference in the underlying evidential material means that each prayer must be analysed separately and could yield different outcomes.

180 The defendants make this argument primarily on the authority of a decision of Hoffmann J (as he then was) in ISC Technologies Ltd and another v James Howard Guerin and others [1992] 2 Lloyds Rep 430 (“ISC Technologies”). In ISC Technologies, the plaintiffs issued a writ in England and obtained leave to serve the writ out of the jurisdiction on a defendant in the United States. That defendant then sought to set aside service under Order 12 r 8 of the English Rules of the Supreme Court 1965 (SI 1965 No 776) (UK) (“RSC”). Before Hoffmann J, the defendant argued that the application under RSC Order 12 r 8 was a rehearing of the ex parte application, and involved the exercise of a fresh discretion. He argued that Hoffmann J could therefore take into account events that had taken place after leave was granted in deciding whether or not to set aside service.

181 Hoffmann J disagreed. He held instead that the defendant’s application was to discharge the order granting leave to serve out, and the question was therefore “whether that order was rightly made at the time it was made”: ISC Technologies at 434. Therefore, the court being asked to set aside service could not receive evidence that had not been put before the court at the ex parte stage, and could not take into account changes to the factual position after leave was granted unless that was necessary to throw light upon the relevant considerations at the time leave was granted. Conversely, Hoffmann J observed that the position would be different where the defendant applied for a stay on grounds of forum non conveniens. In that case, the court ought to consider the factual position as it was on the date the application was heard.

182 The defendants also cite the case of Credit Agricole Indosuez v Unicof Limited and others [2004] 1 Lloyd’s Rep 196 (“Credit Agricole”). This decision of the English High Court, however, is not of much assistance. In Credit Agricole, the approach which the defendants before me now advocate was common ground between the parties (at [22]).

183 The issue of whether the approach in ISC Technologies should apply in Singapore was touched on tangentially in the case of William Jacks & Co (Singapore) Pte Ltd v Nelson Honey & Marketing (NZ) Ltd [2015] SGHCR 21 (“William Jacks”). ¹⁵⁴ In William Jacks, the plaintiff secured leave to serve a writ on the defendant outside the jurisdiction. The writ asserted a cause of action for breach of a specific contract for the sale of a specific shipment of honey. When the defendant applied to set aside leave, the plaintiff relied on a broader distributorship agreement between the parties and pointed to an exclusive jurisdiction clause in that agreement in favour of Singapore. The plaintiff accordingly attempted to oppose the setting-aside application on the basis that it was entitled to leave under Order 11 r 1(r). But that was not a ground on which the plaintiff relied when it had secured leave ex parte. And the plaintiff’s pleaded cause of action was not founded on the broader distributorship agreement which contained the exclusive jurisdiction clause.

184 The assistant registrar held that, in the absence of any finding that the plaintiff’s conduct was an abuse of process, the plaintiff was entitled to rely on the new head of jurisdiction under Order 11 r 1(r) and on the new cause of action arising from the distributorship agreement even though neither had been raised or relied upon at the ex parte stage. The assistant registrar also allowed the plaintiff to adduce additional evidence to support the new cause of action, holding that it would not make sense to allow the plaintiff to rely on the new cause of action but to then bar it from adducing fresh evidence on the setting-aside application to support it.

185 The case before me is an even stronger case than William Jacks. In this case, the plaintiffs do not rely on a new cause of action or on a new head of Order 11 r 1. They seek simply to argue, by reason of the Norwegian proceedings, that Germany is now unavailable as an alternative forum. The plaintiffs are not changing either their substantive cause of action against the defendants nor their case on Order11. They merely seek to rely on evidence of events that took place after they secured leave ex parte.

186 A further consideration of William Jacks is warranted. There is a line of English authority which holds that, in any action where leave to serve a defendant outside the jurisdiction is required, the plaintiff cannot add a new cause of action by amending the pleadings in that action but must instead commence fresh proceedings and seek fresh leave to serve the originating process in those proceedings on the defendant outside the jurisdiction: see Parker v Schuller [1901] 17 TLR 299 and Metall und Rohstoff AG.

187 There is a competing line of English authority which rejects this position. The leading case in this line is NML Capital Ltd v Republic of Argentina [2011] 2 AC 495 (“NML Capital”). In NML Capital, Lord Phillips held (at [74]) that “procedural rules should be the servant not the master of the rule of law” and that the objective of the rules is to “enable the court to deal with cases justly, and that this involves saving expense and ensuring that cases are dealt with expeditiously”. To that end, Lord Phillips observed that “where there is a valid basis for subjecting [an out-of-jurisdiction person] to the jurisdiction, it is not obvious why it should be mandatory for the claimant [seeking to invoke new grounds] to be required to start all over again rather than that the court should have a discretion as to the order that will best serve the overriding objective”: NML Capital at [75].

188 In deciding William Jacks as he did, the assistant registrar (at [19]) followed Lord Phillips’ approach. Lord Phillips’ point is, in essence, the same point which I made with regard to a failure to make full and frank disclosure which is not sufficiently egregious as to constitute an abuse of process precluding a fresh application for leave to serve out (see [79] above).

189 This is undoubtedly the correct approach, subject as it must be to the court’s overarching power to prevent an abuse of the process of the court. The other approach wastes costs and time for both the plaintiff and the defendant. For the plaintiff, it is a waste of costs and time to require the plaintiff to file a fresh originating process even where there is a good arguable case that the additional causes of action come within a head of jurisdiction in Order 11. For the defendant, it also wastes costs and time to require the defendant to take out a second setting aside application based on the fresh originating process.

190 I therefore agree with Lord Phillips’ approach in NML and with the assistant registrar’s approach in William Jacks. I hold that the court has a discretion to take into account evidence of events occurring after leave is granted ex parte when a defendant applies to set aside service out of the jurisdiction. It seems to me that time and costs will be wasted if a court assesses a setting-aside application only on the facts and matters as they existed at the time leave was granted. That would leave it open to the plaintiff, upon service being set aside, simply to seek leave to re-serve the writ or to issue a fresh writ and seek leave to serve that writ, fortified by the fact that there is now a new factor that tends in his favour. The defendant would then come to court again, presumably, to have the re-service or the fresh writ set aside. Everyone will then be made to go over the entire process again, when this could have been dealt with at the hearing to set aside the first writ.

191 It also seems to me that taking into account on a setting aside application supervening events transpiring after leave was granted ex parte does not go against the purpose of Order 11 of ensuring that both the court and the defendant are given full and proper notice of the basis on which the plaintiff claims the court should exercise jurisdiction over the defendant. A setting aside application is an inter partes hearing. In the course of that hearing, the defendant will be notified and informed of any supervening events on which the plaintiff intends to rely. If the court’s analysis is that the original order granting leave should not have been granted on the material then before the court, but that the material now before the court justifies leave being granted, any prejudice to the defendant can be addressed by an appropriate order as to costs.

192 Further, it cannot be said that allowing the court to take into account events after leave is granted is an approach which unjustly favours plaintiffs as a class over defendants as a class or vice versa. In the long run, the approach is likely to operate equally as between the two classes.

193 The court is not, of course, obliged to take into consideration any supervening events. If the plaintiff has engaged in sharp practice or conduct which causes substantive prejudice to the defendant, the court may disregard the supervening events and allow the setting aside application. If the plaintiff’s conduct amounts to an abuse of the process of the court, the plaintiff will be precluded from making a fresh application. If not, the plaintiff is at liberty to make a fresh application for leave.

194 Conversely, to follow ISC Technologies is to give the defendant a systemic advantage from the procedural sequence in which a setting aside and a stay application are heard. Where the supervening event is against the defendant’s case, the plaintiff will never be able to rely on it. The defendant simply asks the court to shut its eyes to the event on the setting aside application. That makes it more likely that the setting aside application will succeed. If so, the plaintiff will never get the chance to rely on the event on the stay application. But where the supervening event supports the defendant’s case, the defendant can always rely on it. Even if the setting aside application fails, the defendant can still ask the court to take the supervening event into account on the stay application. The approach in ISC Technologies confers an inevitable advantage on the defendant and an inevitable disadvantage on the plaintiff. It is not apparent why this state of unfairness should be the law.

195 One might say that, if the supervening event favours the plaintiff, the plaintiff’s case is no weaker than it was at the time it applied for leave, and thus it should be foreclosed from relying on new factors in its favour. After all, when it applied for leave it did not have the reinforcing factor of there being, for example, another appropriate forum being unavailable. But it seems to me that this does not compel the conclusion that the courts must deliberately ignore the change in circumstances that has led to a weak case becoming stronger. The conclusion should not be that if the case was weak then, it must still be weak now, when everyone knows that this not true.

196 I would add also that taking into consideration events occurring after the leave is granted will help align the substantive approaches under the setting-aside application and the stay application (ie, aside from burden of proof). As things stand, if ISC Technologies is right, the court will have to apply Spiliada twice in a case where there are supervening material events, once for each application. On the setting-aside application, the court pretends that the supervening events did not occur and assesses whether leave should have been granted. But on the stay application, the court takes into account the supervening events, and assesses whether Singapore is the appropriate forum. Moreover, in the case where the supervening events favour the defendant, it becomes apparent that the defendant is made to file the setting-aside application unnecessarily. It knows that its case is stronger on the stay application, but it has to file the setting-aside application anyway to indicate that it has not submitted to the jurisdiction: Zoom at [56]. So the application itself is superfluous, and the court’s consideration of it pointless. To my mind, this duplication of efforts ignores reality, and has the potential to create unjust results. It is also unnecessarily laborious, and disruptive of the current practice, which, as the Court of Appeal observed in Zoom at [80] is for the local court to “collapse the issue of the proper forum into one question considered in the round”. I would therefore not follow ISC Technologies.

197 Applying the above observations to the present facts, it seems quite unnecessary to force the court to pretend that Germany continues to be an available forum, when in reality it is not. I therefore consider that this is a relevant factor in the test to be applied, which is now the same on the setting-aside application and the stay application. I turn now to apply the Spiliada test.

198 At the first stage of the Spiliada test, as it applies under Order 11 r 2(2), the plaintiffs have the burden of showing that Singapore is clearly the appropriate forum: Zoom at [71]–[75], Spiliada at 481E. As the Court of Appeal recently observed in Rappo, Tania v Accent Delight International Ltd and another and another appeal [2017] 2 SLR 265 (“Accent Delight”) at [70], this inquiry is concerned with finding those incidents or connections that have “the most relevant and substantial associations with the dispute” [emphasis in original]. In so doing, the lodestar for the court is whether any of the connections point to a jurisdiction in which the case may be tried more suitably for the interests of all parties and for the ends of justice: Accent Delight at [72].

199 I begin with the four connecting factors I consider most significant, namely:

(a) The availability of Germany as an alternative forum;

(b) The governing law and possible transfer to the SICC;

(c) The availability of witnesses; and

(d) The availability of documents.

200 I will then turn to deal with other connecting factors that were less significant in my analysis, namely:

(a) The availability of Norway as an alternative forum; and

(b) What the plaintiffs, adopting the language of Lord Goff in Spiliada, have called the Cambridgeshire factor.

201 The first connection I consider is whether Germany is available as an alternative forum. I consider Germany as a forum first because the Court of Appeal in Rickshaw Investments indicated at [35]–[39] that the loci delicti is prima facie the appropriate forum for a claim in tort (“the Albaforth principle”). Having said that, the Court of Appeal did observe at [40] that this is “only the prima facie position and/or a weighty factor pointing in favour of that jurisdiction”. Building on that observation, I share the view of the assistant registrar that the weight to be given to the Albaforth principle is greatly attenuated when the exercise of determining the loci delicti is not a straightforward one. In this case, as my analysis above has shown, the overall conclusion that Germany is the loci delicti is by no means obvious and does not detract from the fact that significant acts of reliance upon the representation and much of the loss and damage arising from the representation occurred outside Germany. Accordingly, the claim of Germany to be the most appropriate forum simply by virtue of the Albaforth principle is not a strong one.

202 Further, it is the case that Germany is no longer available as a forum for the parties to resolve this dispute. The Norwegian proceedings which the plaintiffs commenced in August 2015 mean that, as matters stand today, any German court asked to determine this dispute is obliged to cede jurisdiction to the courts of Norway. The German courts are thus now bound by the “court first seised” rule in the Lugano Convention to decline jurisdiction over this dispute. ¹⁵⁵

203 The defendants argue that the fact that the German courts are no longer available should be disregarded because that has come about entirely as a result of the plaintiffs’ voluntary act in commencing proceedings in Norway. ¹⁵⁶ The assistant registrar agreed with the defendants, holding that the plaintiffs cannot assert unavailability of another court if the unavailability is due to their own acts: Skaugen at [125]. My view is that I cannot disregard the reality that this dispute cannot be litigated in the German courts. I say that for three reasons.

204 First, the plaintiffs cannot in my view be criticised for having chosen to commence proceedings in Norway. The defendants accept that Norway is an appropriate forum for the resolution of this dispute, even if they do not accept that Norway is the most appropriate forum. ¹⁵⁷ The first plaintiff has a real and substantial connection to Norway. So does the second defendant. The subject-matter of this dispute too has a real, though perhaps not substantial, connection to Norway. That is the place at which the plaintiff received the PPM and from which the first plaintiff took the decision to select the MAN Engine in reliance on the PPM. The plaintiffs’ decision to commence proceedings in Norway is both reasonable and bona fide. It cannot in my view be stigmatised as a tactical decision calculated to exclude Germany as an available forum.

205 The second reason that I cannot disregard the unavailability of Germany as a forum is because that is the result of the very design of the Lugano Convention. Both Germany and Norway, amongst other countries, have acceded to the Lugano Convention. Those two countries have thus adopted, as part of their own respective procedural law, the “court first seised rule” embodied in Art 27 of the Convention. The procedural law of the two countries therefore accepts that a plaintiff with a claim for which both Germany and Norway are an appropriate forum has a free choice as to whether to commence those proceedings in the courts Germany or in the courts of Norway. More importantly, the procedural law of the two countries also accepts that both the counterparty to those proceedings as well as the forum which the plaintiff does not choose are obliged to respect the plaintiff’s choice of forum. Again, in those circumstances, it does not appear to me that this court ought to disregard the very real consequences of the “court first seised principle”, at the very least where it is common ground that the court first seised is an appropriate forum and where I have found that the first plaintiff has reasonably and bona fide chosen that forum.

206 The third reason I cannot disregard the unavailability of Germany as a forum is the actual outcome of the Norwegian proceedings. In October 2016, the Norwegian Court of Appeal decided that, in its view, the plaintiffs’ dispute with the defendants should not be litigated at all – whether in Norway, Germany or Singapore. Instead, the court held that the plaintiffs are bound by the arbitration clauses in the contracts between the shipbuilders and the first defendant for the sale and purchase of the six MAN Engines to arbitrate this dispute in China, even though the plaintiffs are not parties to those contracts or to the arbitration agreements which they contain. The plaintiffs have appealed the decision of the Norwegian Court of Appeal to the Norwegian Supreme Court. But, until and unless the Norwegian Supreme Court reverses that decision, it binds the parties.

207 The final reason I cannot disregard the unavailability of Germany as a forum is because, at least on one view, the limitation period applicable to the plaintiffs’ claim against the defendants under German law has expired. The applicable limitation period in Germany is the long-stop limitation period of ten years pursuant to §199(3) of the German Civil Code (“BGB”). ¹⁵⁸ The last of the FATs took place in 2002. The German limitation period therefore expired, at the latest, in 2012. The defendants agreed a series of limitation waivers commencing in 2012. The last waiver expired without renewal on 31 March 2015. ¹⁵⁹ The defendants’ position is that the plaintiffs’ claim in Germany is now time barred. The plaintiffs do not accept this position. ¹⁶⁰ It is true that a plaintiff who fails to issue protective proceedings in the more appropriate forum cannot then argue that the expiry of the limitation period in that forum is a reason that it should be allowed to continue proceedings in the less appropriate forum it has chosen (see Spiliada at 483F-G). But in this case, as I have found, the plaintiffs acted reasonably and bona fide in commencing proceedings in Norway rather than Germany. Indeed, the plaintiffs assure me that the Norwegian proceedings were commenced defensively only, in order to avoid having the time bar expire both in Germany and Norway in a situation where it finds itself unable to proceed in Singapore. ¹⁶¹

208 I therefore conclude on this factor that significant weight must be given to the fact that Germany is no longer an available forum. That leaves a choice between only Singapore and Norway as the appropriate forum. But, in light of the Norwegian Court of Appeal’s decision to exclude itself as an available jurisdiction, only Singapore remains. That suffices to make Singapore clearly the appropriate forum in which the case can be tried more suitably for the interests of all the parties and the ends of justice.

209  I shall nevertheless analyse the remaining factors on the basis that the courts of both Germany and Norway continue to be available, in case I am wrong on this aspect.

210 The next key factor is the law governing the dispute. As Professor Yeo Tiong Min says in Halsbury’s Laws of Singapore vol 6(2) (LexisNexis, 2016 Reissue) at para 75.374, the double actionability rule is not only a jurisdictional rule but also a choice of law rule. I have found that the lex loci delicti is German law.

211 The defendants argue, as they did before the assistant registrar, that it would be more appropriate for a German court to apply German law, not only because that is its own law, but also because a German court would not have to contend with the double actionability rule and would have to apply only German law. Our double actionability rule would require a Singapore court to apply both German law and Singapore law. ¹⁶² The defendants also argue that this disadvantage cannot be mitigated simply by transferring the matter to the SICC. Additional time and costs will still be required to establish double actionability, because that principle of Singapore private international law applies even when the forum is the SICC. Further, the SICC does not have a German judge who can deal with issues of German law by way of submission rather than evidence. ¹⁶³ These arguments found favour with the assistant registrar, who considered that they together gave the German court the “home advantage”: Skaugen at [128]–[130].

212 The plaintiffs contend, however, that the SICC is competent to decide questions of German law with the benefit of submissions from suitably-qualified foreign counsel. ¹⁶⁴ The possibility of a transfer to the SICC therefore favours Singapore as the appropriate forum.

213 I agree with the plaintiffs that the possibility of a transfer to the SICC mitigates substantially the disadvantage of having to apply German law in the present dispute. I start by observing that the availability of the SICC is a factor I can take into account in my analysis. As the Court of Appeal indicated in Accent Delight at [121], a relevant factor in the forum non conveniens analysis is a possible transfer to the SICC. A transfer is available where the requirements of Order 110 r 12(4)(a) are met. First, the claims must be of an “international and commercial nature”; second, the parties must not seek any relief in the form of, or connected with, a prerogative order; and third, the High Court must deem it more appropriate that the case be heard in the SICC. There is a good arguable case that those requirements are satisfied here. The claims are indeed of an international and commercial nature; neither party has sought relief in the form of, or connected with, a prerogative order; and the fact that German law applies is a significant factor in favour of deeming it more appropriate for this case to be heard in the SICC than in the High Court.

214 Turning to the advantages of the SICC, Order 110 r 25 of the Rules of Court allows the SICC to order, on the application of a party, that a question of foreign law be determined on the basis of submissions instead of proof. This will substantially reduce the time and expense involved in pleading issues of foreign law. Similarly, the Court of Appeal also noted in Accent Delight at [122] that the fact that foreign law applies carries less weight in the forum non conveniens analysis if the Singapore courts, through the International Judges in the SICC, are familiar with and adept at applying that foreign law. Although I appreciate that there is currently no German judge appointed as an international judge eligible to sit on SICC cases, there are judges from civil law jurisdictions who are nevertheless equipped with the necessary skills and experience to deal with German law here. Indeed, a Japanese judge sits on the SICC bench, and Japan’s Civil Code has historically been influenced by the German Civil Code.

215 Further, it is not obvious that the cost and expense to be saved by litigating this dispute in Germany will be so great as to justify discontinuing the proceedings in Singapore, when no proceedings have even commenced in Germany at this time. I appreciate that this situation is the result of the plaintiffs’ choosing to commence proceedings in Norway. But it seems to me that on this separate factor of saving time and expense, I must take things as they stand. And as things stand, the plaintiffs’ Norwegian proceedings are ongoing, with no proceedings whatsoever in Germany. Additionally, it is not guaranteed that the plaintiffs will withdraw the Norwegian action and attempt to proceed in Germany, even if I hold that Singapore is not available as the appropriate forum. That would nullify the basis of preventing the plaintiffs from litigating this dispute in Singapore. Even if the plaintiffs were to withdraw the Norwegian proceedings and proceed in Germany, it seems to me that parties will have to appoint counsel in Germany ¹⁶⁵ and initiate proceedings in the German courts. This will incur additional expense of time and costs. I acknowledge, of course, that time and expense will also be incurred in appointing German counsel to assist the SICC. But it seems to me that this process will be made faster by the involvement of the local counsel already briefed on both sides, who are already quite familiar with this dispute.

216 I add that a dispute of this nature – where the factual and legal connections are distributed across jurisdictions as diverse and geographically divided as Norway, Germany, China, Hong Kong, and Singapore – is the archetypal dispute which might be better dealt with by an international panel of judges, as is available under the SICC, than by the judges of any one jurisdiction. That is a factor which weighs in favour of SICC against both the Norwegian Courts and the German courts.

217 My conclusion on this factor is that the advantages of having the matter heard in either Germany or Norway are overstated once the benefits of the SICC are taken into account. Hence the good arguable case in favour of a transfer to the SICC tends to favour Singapore as clearly the more appropriate forum to hear the dispute.

218 I turn now to the connections of the parties and the witnesses. The assistant registrar held that witness location would favour Germany, as the relevant personnel on both sides are based in Europe. As a result, it would be more convenient for the trial to be heard in Germany. The first defendant is based there. And Germany is closer to the second defendant’s and the first plaintiff’s place of business than Singapore: Skaugen at [130]. The defendants echo and amplify these concerns on appeal: they say that concerns of witness compellability and witness convenience both point to Germany as the appropriate forum. ¹⁶⁶ This is because the Singapore courts have no power, unlike German courts, to compel witnesses who are German citizens or residents. ¹⁶⁷ They say this is important because the key witnesses who have personal knowledge of the events material to this dispute are German nationals who speak German as their first language, and it would be more convenient for their evidence to be heard by a German court. ¹⁶⁸

219 The plaintiffs argue that the defendants have not shown any real inconvenience or difficulty with receiving witness evidence in Singapore. They say that the defendants have not identified any of these potential key witnesses who are German citizens, speak German as their first language, and reside in Germany, such that Germany should be favoured over Singapore as the appropriate forum. ¹⁶⁹

220 I agree with the plaintiffs on this point. Despite being asked to do so, the defendants cannot identify any of the key witnesses who cannot be compelled to give evidence in Singapore and who may be inconvenienced by having to do so. To put it simply, there was no evidence before me to show that witness convenience as a factor pointed in favour of Germany. ¹⁷⁰

221 It is also the case, as the Court of Appeal has observed, that the physical location of witnesses is no longer of vital or even material consideration in the Spiliada analysis because of the option of giving evidence by videolink: Siemens AG v Holdrich Investments Ltd [2010] 3 SLR 1007 (“Siemens”) at [11]. Although translation might prove more of a concern, it is an ambivalent or neutral point at best. The plaintiffs’ operations have variously been run out of Norway, China, Hong Kong, and Singapore. If Germany were the forum, translation would remain an issue as the evidence of the plaintiffs’ witnesses would have to be translated into German.

222 I would further add that in any event, the importance of the location and the compellability of witnesses depends on whether the main disputes will revolve around questions of fact: Rickshaw Investments at [19]. Without expressing any view on the merits of the case, I would simply note that the defendants have, by their own letter to the plaintiffs, admitted that deceptive FATs took place for at least three of the six MAN Engines. ¹⁷¹ Moreover, the defendants have not, in the material before me, identified any witnesses who might give any exculpatory evidence on this issue at trial. ¹⁷² It appears to me, therefore, that much of the focus at trial will be on the issue of loss and damage. Much of the evidence on that issue, both in terms of witnesses as well as documents, will be located in Hong Kong and Singapore, and at any rate outside Germany. Given all these factors, the non-availability and non-compellability of German witnesses becomes much less significant a factor.

223 In these circumstances, I consider that the defendants’ inability to point to any specific witnesses that they will have difficulty in compelling to give evidence at a trial in Singapore means that witness location does not favour Germany. Witness location therefore does not point away from Singapore as the appropriate forum.

224 The next point concerns the availability of documentary evidence. The defendants raise several contentions on this point. First, they say that the agreements are in German and have to be translated into English. ¹⁷³ Second, they say that some of the documentary evidence in the defendants’ possession contains the personal data of their former employees and other individuals and is therefore subject to German and EU data protection laws. That makes it much more difficult for the evidence to be transferred to Singapore. ¹⁷⁴ The transfer of protected data is subject to a “balance of interests” tests, balancing the interests of disclosure pursuant to the plaintiffs’ discovery obligations in Singapore, as against the legitimate interests of the affected persons in not having their data disclosed. ¹⁷⁵ This test has to be applied on a document-by-document basis, ¹⁷⁶ which will be both time-consuming and costly. Third, they raise the concern that the plaintiffs have admitted that they wish to use the Singapore action as means of obtaining common-law discovery to be deployed in other proceedings, which the defendants say is a collateral motive that ought not to be condoned. ¹⁷⁷

225 The plaintiffs’ response is that it is impossible to give any meaningful weight to the defendants’ submissions because they have neither particularised what these protected documents might be, nor how many of them are affected by German data protection law. ¹⁷⁸ In any event, the plaintiffs say, data protection issues can be resolved through various means such as anonymisation or pseudonymisation of the data or by sealing the court file. ¹⁷⁹ Alternatively, if the matter is transferred to the SICC, the defendants will be subject to less onerous discovery obligations since there is no process of general discovery in the SICC, with the SICC’s discovery rules being modelled on the IBA Rules on the Taking of Evidence in International Arbitration 2010. ¹⁸⁰ Further, the plaintiffs allege that the bulk of the documents and evidence to be adduced will be in relation to documents and evidence to be produced by the plaintiffs in Singapore, as these will go towards reliance on the misstatements and the loss and damage consequently suffered.

226 Let me deal with the point on translation first. It seems to me that the point on translation of documents is slightly in favour of Singapore. A parallel to Siemens can be drawn here. In Siemens, the Court of Appeal took the view that the fact that the relevant agreement and the documentary evidence in that case were all in English suggested that English was the lingua franca of the parties and that an English-speaking forum would thus be preferable (at [11]). The key documents exhibited in this case – the PPM, FSI and FATs – are either solely in English, or in both English and German. ¹⁸¹ All this suggests that Singapore, as an English-speaking forum, would be more suitable for resolution of this dispute.

227 To the extent that there are other documents which might need to be translated, it seems to me that the inconvenience to the parties is evenly balanced as between Germany and Singapore. The defendants’ documents – which go to the making of the representation – are likely to be in German and will have to be translated into English if the action proceeds in Singapore. But the plaintiffs’ documents – which go to the question of receipt, reliance and loss and damage – are likely to be in English and will have to be translated into German if the action proceeds in Germany. The point on translation therefore does not clearly assist either side or may even be said to slightly favour Singapore.

228 The thornier question concerns the German and EU data protection requirements. Both parties have adduced expert opinions as to the restrictions imposed by these laws. ¹⁸² What the experts have done in both these opinions is to set out how the laws apply generally to documents in general, and, to certain classes of documents in particular. But the experts do not appear to have been specifically asked to examine the documents in this dispute. In particular, the defendants’ expert confirms in his first report that “an evaluation of information in order to fulfill [sic] the requirements under German/European Data protection law … has not been conducted by [the first defendant] at this moment”. As a result, he could not “at present assess what kind of interests are affected by a disclosure of relevant information to the Singapore High Court”. ¹⁸³ He repeats this in his second report, filed in response to the report of the plaintiff’s expert, noting that “[with] regard to the required balancing of interests in accordance with data protection law, a screening of the documents has yet not been carried out.” ¹⁸⁴

229 I agree with the plaintiffs that the defendants have failed sufficiently to identify and particularise what specific documents would be subject to data protection restrictions which might in turn affect the defendants’ ability to transfer the documents to Singapore. The defendants accept that they bear the evidential burden on the issue of whether their documents will be subject to data protection restrictions (Spiliada at 476E). ¹⁸⁵ But there is simply no evidence before me as to how many documents are subject to these restrictions, nor as to the scale of the filtering or redaction or anonymisation exercise that will have to take place to comply with German data protection law. The affidavit on which the defendants rely to support their arguments on this point ¹⁸⁶ merely says that the defendants’ documentary evidence may comprise emails and other correspondence which might contain personal data; and that some documents may contain confidential information and thus be protected as business or trade secrets under German or EU law. ¹⁸⁷ Notably, it does not go so far as to say that the documents so protected will be “substantial” or “numerous” or anything of the sort. This is not enough to discharge the defendants’ evidential burden. Indeed, as the defendants’ own expert observed above, it appears that the defendants have not even attempted to assess or evaluate what information might be subject to German data protection laws. In these circumstances, there is an insufficient evidential basis for the defendants to say that German data protection law points to Germany as the appropriate forum.

230 The next point to be considered is that the plaintiffs seek to use common-law discovery available in the Singapore action to obtain documents for use in subsequent proceedings. ¹⁸⁸ The defendants object to this as a collateral motive which this court ought not condone. ¹⁸⁹ I take the view, however, that the common-law discovery available in Singapore, as compared to civil law jurisdictions such as Germany or Norway, is a legitimate advantage that the plaintiffs may make use of. Further, insofar as it is the plaintiffs’ intent to deploy this material in other litigation, they cannot do so unilaterally and must instead seek the court’s permission to relax their implied undertaking not to do so: Beckkett Pte Ltd v Deutsche Bank AG [2005] 3 SLR(R) 555 at [18]. This is therefore not a point to be taken against the plaintiffs.

231 Finally, and again without expressing any view on the merits of the claim, it appears to me that at least half of the evidence will be concerned with issues relating to the plaintiffs’ reliance and the extent of loss and damage suffered. That evidence lies with the plaintiffs in Singapore, not Germany. The first defendant’s letter to the first plaintiff accepts that there were deceptive FATs in relation to three out of the six MAN Engines. ¹⁹⁰ A large part of the defendants’ submissions before me have been concerned with the evidence of assignments being validly made, and of reliance by the plaintiffs or the parties which have assigned their claims to them. The inquiry into reliance upon the alleged misrepresentation naturally leads to the question of the extent of loss and damage suffered, and in turn, to the quantum of damages. This all strikes me as evidence that is more likely to be within the plaintiffs’ hands than the defendants’. Thus, the availability of documentary evidence in Germany is also not that significant a factor in the entire analysis.

232 For the reasons above, I find that on the factor of availability of documents, the points raised by the parties suggest that Singapore is the appropriate forum, not Germany.

233 The four factors I have just analysed concern the connections which I consider to be most material to determining in which forum this “case may be tried more suitably for the interests of all the parties and the ends of justice” (Spiliada at 476C). To my mind, they all point clearly to Singapore. For the sake of completeness, I will also address two other factors cited to me by the parties which I have not found to be of significant weight.

234 The defendants argue, in the alternative, that Norway is a more appropriate forum than Singapore. ¹⁹¹ This is because the plaintiffs have accepted that the Norwegian courts have jurisdiction; Norway is the place of business of the first plaintiff and its headquarters; certain acts of receipt and reliance took place in Norway; and the ongoing proceedings in Norway suggest that there is a risk of conflicting judgments if this action is also allowed to proceed in parallel in Singapore. ¹⁹²

235 The plaintiffs argue that the state of proceedings in Norway indicate that Norway is not available as an alternative forum. ¹⁹³ Even if it were, the plaintiffs further say that Norway has not been demonstrated to be a forum which will deliver efficient, expeditious and economical resolution of the dispute. ¹⁹⁴

236 The first point that has to be considered is whether the Norwegian court is available as an alternative forum. On 31 October 2016, the Norwegian Court of Appeal dismissed the plaintiffs’ claims in the Norwegian proceedings, holding amongst other things that the plaintiffs are considered to be bound by the arbitration clauses in the sales contract between the first defendant and the shipbuilders for the supply of the six MAN Engines. ¹⁹⁵ The plaintiffs filed an appeal to the Norwegian Supreme Court on 2 December 2016, ¹⁹⁶ and the outcome of that appeal was unknown at the time I rendered my decision. The decision of the Norwegian Court of Appeal is, of course, final until and unless it is reversed on appeal.

237 The effect of all this is that Norway appears now to be unavailable to the plaintiffs as an alternative forum. I do acknowledge that this ultimately depends on the outcome of the appeal to the Supreme Court of Norway. To that extent, the court of Norway are not unavailable to the plaintiff in the same way as the courts of Germany are unavailable. The plaintiffs’ appeal means that they are still attempting to pursue a resolution of the dispute through litigation in the Norwegian courts. As compared to that, no proceedings at all are pending before the German courts. And even so, if the German time bar has set in, no proceedings can ever be commenced or determined on the merits in Germany. But the current state of affairs in the Norwegian proceedings significantly diminishes the prospect of Norway serving as an alternative forum. It also diminishes the chances of conflicting judgments arising in Norway and in Singapore. On this point, therefore, there is something to be said that the objective interests of justice would favour the action proceeding in Singapore as compared to Norway.

238 Quite apart from all this, the other connecting factors assessed objectively also point to Singapore being the more appropriate forum than Norway. The points which are likely to be most heavily disputed in this action are issues such as the validity of the assignments, whether any of the assignors relied on the alleged misrepresentation, the degree of reliance, and the extent of loss and damage suffered by the assignors. As I have already found, the evidence on all those issues is far more likely to be in Singapore than in Norway. Only two key acts took place in Norway: (i) Norway is where the first plaintiff received the misrepresentation in the PPM, and (ii) the first plaintiff communicated its acceptances of the MAN Engine to the shipbuilders from Norway. Both of those acts are of only marginal relevance to ascertaining where this dispute can be tried suitably in the interests of the parties and in the interests of justice.

239 Conversely, many of the current and former owners of the ships were or are located in Singapore, and their losses would presumably be documented here. Further, the evidence is also that the Skaugen group focused on Singapore as the centre of its ship owning location and operations from 2004 to date. ¹⁹⁷ The availability of evidence that will likely be most pertinent to this dispute therefore favours Singapore, not Norway, as the appropriate forum.

240 A factor raised by the plaintiffs in support of Singapore being the appropriate forum is what Lord Goff referred to in Spiliada as the Cambridgeshire factor. This factor arises where there has been, or is, litigation in a jurisdiction involving very complex facts that in turn necessitates highly specialised expert evidence, and evidence and expertise has been built up in a particular jurisdiction on those facts. Where the Cambridgeshire factor is in play, it suggests that that jurisdiction is the prima facie appropriate forum: see Virsagi Management (S) Pte Ltd v Welltech Construction Pte Ltd and another appeal [2013] 4 SLR 1097 at [39], citing the observations of Prof Yeo Tiong Min in Halsbury’s Laws of Singapore vol 6(2) (LexisNexis, 2009) (“Prof Yeo”) at para 75.094.

241 The plaintiffs say that the Cambridgeshire factor arises on our facts because Singapore lawyers have been briefed on this matter, and Danish lawyers who handled an ICC arbitration in Hamburg concerning a different set of engines have similarly been briefed and will be able to apply to the SICC for rights of audience, such that time and costs will be saved overall rather than proceeding with litigation in either Germany or Norway. ¹⁹⁸

242 The defendants reject the plaintiffs’ argument based on the Cambridgeshire factor. They say that the facts here are nowhere as complex, the proceedings nowhere as advanced, and the expertise built up nowhere as vast, as in Spiliada where the House of Lords held that the Cambridgeshire factor was a legitimate factor and justified declining to set aside service. ¹⁹⁹

243 I agree with the defendants on this point. As Prof Yeo has observed, the Cambridgeshire factor is recognised to be a “highly exceptional factor and has rarely been applied since [Spiliada]”. The facts here are quite far removed from those in Spiliada. Why this is so can readily be appreciated once we understand how the Cambridgeshire factor arose in Spiliada itself.

244 In Spiliada, the plaintiff ship owners brought a claim against the defendant shippers for damage to their ship caused when the shippers loaded wet sulphur which caused severe corrosion and pitting to the ship. The judge at first instance in Spiliada, Staughton J, had heard a substantial part of another action brought by a different ship owner against the same shipper for the same type of damage caused in the same way: Spiliada at 485B. The ship in the other action was the Cambridgeshire. The Cambridgeshire action involved 15 counsel, each of whom was familiar with a substantial mass of documents: Spiliada at 467G. The shipper’s witnesses had already been brought over from Canada for the Cambridgeshire action; both sides of the Cambridgeshire action were represented by the same counsel as both sides of the Spiliada action; both sets of counsel had educated themselves on the various topics on which expert evidence would be called; and a number of English experts had been called: Spiliada at 470E–471A. Staughton J observed it would be wasteful in the extreme if all the talent, effort and money which the parties had already invested in having Cambridgeshire brought to trial were not put to use again to determine the Spiliada action, as would have been the result if British Columbia were to be held to be the appropriate forum instead of England: Spiliada at 471B.

245 The Court of Appeal held that Staughton J had given the Cambridgeshire factor too much importance. But on further appeal to the House of Lords, Lord Goff accepted that Staughton J was correct, on the particular facts before him, to have treated the Cambridgeshire factor as a legitimate factor in the forum non conveniens analysis and to have given it the weight that he did. In particular, Lord Goff spoke of the Cambridgeshire factor in these terms (at 485F–H):

246 Lord Goff therefore held that Staughton J had been right to take the view that having experienced teams of lawyers and experts available on both sides of the litigation who were already familiar with Cambridgeshire would contribute to efficiency, expedition and economy in the resolution of Spiliada itself, and moreover, advance the interests of justice: Spiliada at 486A–C. He further noted that although the plaintiffs in the Cambridgeshire action and in the Spiliada action were different, they were insured by the same insurers; those insurers had been financing both actions and thus were dominus litis; and those insurers had instructed solicitors for both the Cambridgeshire action and the Spiliada action: Spiliada at 486D–F. The Cambridgeshire action would therefore set the backdrop against which the Spiliada action would be decided or settled.

247 The facts here are quite far from those in Spiliada. Although the present facts are fairly complex, they do not appear to require highly specialised expert evidence. There have been no related proceedings in the Singapore courts that have involved the same legal team litigating substantially the same issues, albeit between different parties. There will in all likelihood have to be foreign counsel appointed to submit on German law if the matter is transferred to the SICC, and these foreign counsel may not be familiar with the matter. The plaintiffs therefore cannot derive any assistance from the Cambridgeshire factor.

248 Having concluded that the key connecting factors in the first stage of the Spiliada test point to Singapore as being clearly the appropriate forum, it is not strictly necessary to consider the plaintiffs’ further contention that they would suffer substantial injustice if they were shut out from proceeding in Singapore and required to proceed in Germany instead. This parallels the Court of Appeal’s analysis in Rickshaw Investments, in considering an appeal against an order granting a stay of proceedings. The Court of Appeal held that once the factors for the first stage pointed to Singapore as the appropriate forum, the court did not then have to consider whether the plaintiffs would also succeed in arguing on the second stage that they would suffer substantial injustice if compelled to litigate in the foreign jurisdiction instead of Singapore (at [91]).

249 Similarly, in the applications before me, the plaintiffs have succeeded on the first stage in establishing that Singapore is clearly the more appropriate forum than any other. There is thus no need for me to consider whether they would suffer substantial injustice if made to litigate elsewhere under the second stage of the Spiliada test. I therefore do not address the plaintiffs’ arguments that the lack of common law discovery, and the German time bar having set in, would operate to make them suffer substantial injustice if they were compelled to litigate in Germany.

250 Indeed, it would appear that, the plaintiffs having succeeded on the first stage of the Spiliada test, the burden now shifts to the defendants to show why they would suffer substantial injustice if this action were to proceed in Singapore. This shifting of the burden simply mirrors the approach that would be taken if this were an application by the defendants for a stay in a case where service had been effected within the jurisdiction as of right.

251 In a stay application, the defendants would bear the burden of proving that the connecting factors at the first stage of the Spiliada test pointed to some forum elsewhere which was clearly or distinctly more appropriate than Singapore, while the plaintiffs would then bear the burden at the second stage of the Spiliada test of showing that they would not obtain justice in the foreign jurisdiction: Spiliada at 477E and 478D. The converse of that test, to be applied in the Order 11 context, is that the plaintiffs bear the burden of proving that the connecting factors point to Singapore: Zoom at [75], and if the plaintiffs fail on the first stage, they must then go on to show that they would suffer substantial injustice in the foreign jurisdiction. But if the plaintiffs succeed on the first stage of the Spiliada test, the burden shifts to the defendants to show that they would suffer substantial injustice in Singapore. After all, it cannot be right that the plaintiffs who succeed on the first stage must also go on to show that the defendants would not suffer substantial injustice in Singapore. This goes too far and essentially asks the plaintiffs to make the defendants’ case. Nor does it seem to me that this is what the Court of Appeal had in mind when it said that in the Order 11 context, the plaintiff retains the burden of showing that Singapore is the proper forum “at all times”: Zoom at [76].

252 With these principles in mind, the relevant consideration here on the second stage of the Spiliada test is whether the defendants allege that they will suffer substantial injustice should the action be allowed to proceed in Singapore. The defendants make no such allegation and have produced no such evidence. I therefore do not need to consider any further the second stage of the Spiliada test.

253 For the sake of completeness, I also address briefly the defendants’ alternative prayer for a stay of this action on grounds of forum non conveniens. The defendants submit that if they fail to set aside service, the court must consider their stay application. The defendants accept that the substance of the test to be applied remains the Spiliada test. The only material change of approach is that the defendants now bear the burden of showing that there is another forum which is clearly and distinctly more appropriate for the trial of the action. ²⁰⁰

254 I consider that my application of the Spiliada test in determining whether this was a proper case for service out of the jurisdiction within the meaning of Order 11 r 2(2) precludes any separate consideration of the defendants’ alternative prayer for a discretionary stay on grounds of forum non conveniens. There might have been a point to applying the Spiliada test once again to the defendants’ alternative prayer for a stay if I had followed ISC Technologies and accepted that the court could not look at events occurring after leave had been granted to serve the writ out of jurisdiction on the setting-aside application but could do so on a forum non conveniens application for a stay. But for the reasons already stated above, I have not adopted that approach.

255 There is therefore no material difference between the two alternative prayers – one for setting aside and one for a discretionary stay – that could lead to a different outcome upon the application of the substantive Spiliada test. The Court of Appeal was alive to this unnecessary duplication in Zoom, and observed at [78] that where the “substance of the Spiliada test does not differ according to the nature of the application, then a foreign defendant who seeks the setting aside of an overseas service leave order and prays as a fall-back for a stay of proceedings of improper forum grounds…will find that his stay application is pointless”. I agree. The transfer of the burden on the application from the plaintiff to the defendant does not change the analytical framework set out above or the outcome of applying that framework to the facts of this case.

256 Singapore remains the forum conveniens. The defendants’ alternative prayer for a discretionary stay must fail.

257 For the reasons I have set out above, I have found that the plaintiffs have shown a good arguable case that they fall within the heads of jurisdiction under Order 11 r 1(f)(ii) and Order 11 r 1(p). Singapore is also the forum conveniens. I have therefore allowed the plaintiffs’ appeal in RA 163 of 2016 and dismiss the defendants’ appeals in RA 167 and 168 of 2016.