1 These grounds of decision deal with costs for Element Six Technologies Ltd v IIa Technologies Pte Ltd [2020] SGHC 26 (the “Judgment”) in Suit No 26 of 2016 (“S 26/2016”).

2 I highlight here the key facts relevant to the issue of costs. The plaintiff (“E6”), incorporated in the United Kingdom (“UK”), designs, develops, and manufactures synthetic diamond products. E6 owns Singapore Patent No 115872 (“SG 872”) and Singapore Patent No 110508 (“SG 508”). The first is a patent for the manufacture of synthetic diamond using chemical vapour deposition (“CVD”); the second is a patent to convert a coloured single crystal CVD diamond to another colour under suitable heat treatment conditions, a process referred to as annealing. The defendant (“IIa”), a Singapore company, deals in the manufacture of CVD diamond products and the colour treatment of diamonds. E6 contended that three Samples (referred to as “Sample 2”, “Sample 3”, and “Sample 4” respectively, and collectively as the “Samples”) that originated from IIa infringed SG 872. Samples 2 and 4 were CVD diamond optical applications, while Sample 3 was a pink CVD diamond gemstone, which E6 contended infringed SG 508 as well. IIa’s response was a threefold denial: first, it denied that the Samples originated from them; second, even if they did, the Samples were manufactured using IIa’s confidential process; and third, it counterclaimed that SG 872 and SG 508 were invalid.

3 In the Judgment, I found that IIa was responsible for manufacturing all three Samples. On the counterclaim for revocation, I found that SG 872 was valid while SG 508 was not. IIa’s confidential process did not withstand scrutiny. E6 was able to show that IIa had used SG 872 to manufacture the Samples, while the result on SG 508 was not conclusive in view of my finding as to its anticipation by prior art. Therefore, infringement was made out on SG 872 while the invalidity of SG 508 functioned as a defence to any infringement of SG 508.

4 I consider IIa’s offer to settle (“OTS”) as a preliminary matter. IIa contended that O 22A r 9(3) of the Rules of Court (Cap 322, R 5, 2014 Rev Ed) (“ROC”) applied on the basis of the OTS it served on E6 on 3 March 2017. The material terms of the OTS are as follows: ¹

5 In order for the costs implications under O 22A r 9 of the ROC to follow, a valid OTS must be an offer to settle that has “some semblance of reasonableness”: Wyno Marine Pte Ltd (in liquidation) v Lim Teck Cheng and Others [1998] SGHC 340 at [9], and should be “a serious and genuine offer” which contains “an element which would induce or facilitate settlement”: Man B&W Diesel S E Asia Pte Ltd and another v PT Bumi International Tankers and another appeal [2004] 3 SLR(R) 267 (“Man B&W Diesel”) at [8]. What would constitute such a serious and genuine offer must depend on the circumstances and issues of the case: Man B&W Diesel at [10].

6 In my view, IIa’s OTS was not a serious nor genuine offer aimed at compromise. The substance of the OTS was for E6 to cease its claim entirely, on the basis that IIa would not pursue its defence and counterclaim to revoke the patents which E6 was claiming IIa had infringed. Tongue-in-cheek, it offered a return to the status quo prior to the commencement of the action, with nothing gained from the litigation. The OTS effectively asked E6 to capitulate with nothing obtained or achieved. As the Court of Appeal recognised in Man B&W Diesel at [14], a common feature of the offers to settle across cases where both local and foreign courts have refused to grant indemnity costs “is … that in each instance the offeror effectively expected the other party to capitulate”. IIa would thereafter continue what E6 perceived as its commercial exploitation of E6’s patents, with no cost to itself and with no compensation to E6 to allow that conduct to continue.

7 Further, even if the OTS was an offer within the bounds of O 22A, in order for O 22A r 9(3) to apply, the defendant must show that “the plaintiff [has obtained] judgment not more favourable than the terms of the offer to settle”: O 22A r 9(3)(b) of the ROC. IIa made two arguments in its attempt to establish that E6 has not obtained a more favourable judgment.

8 First, IIa argued that if E6 had agreed to the OTS, E6 would have kept both SG 508 and SG 872 on the register, which was more favourable than the outcome of S 26/2016 at trial, which was that SG 872 was valid and was enforced while SG 508 was revoked. However, the purpose of a patent is the protection it affords. Although SG 508 was revoked, E6 succeeded in relation to SG 872 and obtained important consequential orders for itself, namely: an injunction, an order for delivery up and/or destruction of any infringing articles, and an inquiry into damages or an account of profits. In contrast, if the OTS had been accepted, both SG 872 and SG 508 would have been effectively unenforceable against IIa.

9 Second, IIa argued that if E6 had accepted the OTS, it would only have had to pay IIa’s costs from 3 March 2017 up to the date of acceptance. IIa cited Chan Sek Keong CJ’s decision in CCM Industrial Pte Ltd v Uniquetech Pte Ltd [2009] 2 SLR(R) 20 at [41]–[42] for the proposition that the court should take into account the duration in which the plaintiff could have accepted the OTS and the benefits that could have accrued in that time. This argument was misconceived. E6, having substantially succeeded in the suit, was not in danger of having to pay IIa’s costs. Further, IIa’s argument was rather circular. Order 22A r 9(3)(b) refers to the situation where a plaintiff “obtains judgment not more favourable than the terms of the offer to settle” at a moment in time prior to the issue of costs being determined, since it is the very issue of costs that O 22A r 9(3) regulates. The “judgment” referred to in O 22A r 9(3)(b) does not include any decision on costs, since there would not have been any such “judgment” at the time O 22A r 9(3)(b) is considered.

10 IIa’s OTS was, therefore, irrelevant to the issue of costs.

11 E6 won on all issues save for the validity of SG 508. It relied on the adopted by the Court of Appeal in Tullio Planeta v Maoro Andrea G [1994] 2 SLR(R) 501 (“Tullio Planeta”) as stated in Re Elgindata Ltd (No 2) [1992] 1 WLR 1207 (“Re Elgindata (No 2)”). I refer to this as the Tullio approach in these grounds of decision. It therefore argued that it was the overall winner and was entitled to costs with a 15% discount to reflect the fact that SG 508 was revoked. It further asked for costs on an indemnity basis, because of IIa’s unreasonable conduct. ²

12 IIa, on the other hand, advocated an issue-based approach. While it asked for “a single costs order”, premised on its having won the validity argument on SG 508, it first contended in its written submissions that it should obtain 50% of its costs, ³ then in oral argument contended that costs should be apportioned in equal parts between E6 and itself, with each side paying 50% of the costs to the other. At a point, it appeared to suggest, taking the two suggestions into consideration, that no order on costs should be made at all. ⁴

13 I held that E6 should receive costs of the whole action on the standard basis. My decision may be explained by reference to these issues, which are related:

(a) whether the issue-based approach or the Tullio approach was applicable in the present case;

(b) the conduct of IIa, arising out of which E6 asked for indemnity costs, and that of E6, which was also relevant to the issue of costs; and

(c) the importance of proportionality and reasonableness of the parties’ conduct in the court’s exercise of its discretion on costs.

14 E6 argued that costs should follow the event and, as the overall successful party in this part of S 26/2016, it should be awarded its costs in the suit so far. This argument was based on the principles summarised in Re Elgindata (No 2) (and cited in Tullio Planeta at [24]):

15 E6 recognised that it had, however, failed in terms of the validity of SG 508, and conceded that it could be deprived of its costs in relation to that issue, and it proposed a 15% discount accordingly. This was consistent with the third principle stated in Re Elgindata (No 2) ([11] supra), that a successful party “could be deprived of his costs in whole or in part where he had caused a significant increase in the length of the proceedings”.

16 In Singapore, this approach has been applied in patent cases, where the successful party was awarded costs of the suit with a discount to reflect the issues on which it had lost. In Cicada Cube Pte Ltd v National University Hospital (Singapore) Pte Ltd and another appeal [2018] 2 SLR 940 (“Cicada”), the trial judge’s roughly 50% discount to the costs sought by NUH was upheld by the Court of Appeal, on the basis that (1) NUH had succeeded on its alternative prayer to be named a joint proprietor of the relevant patent alongside Cicada and that Dr Sethi be named a joint inventor, but (2) NUH had only succeeded partially because it did not succeed in (a) having itself named as sole proprietor, (b) removing Dr Poo and Dr Ratty as inventors, and (c) having Mr Lim named as a joint inventor: Cicada at [99]–[101]. A similar approach was taken by the Court of Appeal in Lee Tat Cheng v Maka GPS Technologies Pte Ltd [2018] 1 SLR 856 at [114] where, notwithstanding that the respondent had prevailed on the whole at the trial, he was given only a third of his costs of the trial by the Court of Appeal because significant time and resources had been spent on an ultimately unsuccessful challenge to the patent’s validity. In neither of these cases was it suggested that the starting point should have been to consider each party entitled to the costs of the issues on which it was successful, and in neither case did the Court of Appeal consider whether the issues had been raised unnecessarily or unreasonably before depriving the successful party of part of its costs to reflect the issues on which it was unsuccessful.

17 IIa, in contrast, approached each patent separately, at least notionally, in an “issue-based approach”. The crux of the issue-based approach proposed by IIa is that the court approaches each patent (which it identified as the “issue” for these purposes) on its own terms, rather than consider who the overall winner in the case was. On this approach, the “event” in the phrase “costs follow the event” is defined narrowly to refer to each issue rather than the overall suit. Hence, on IIa’s argument, as it was successful on SG 508, it was entitled to the costs relating to SG 508, while E6 was entitled to the costs of SG 872 as the successful party in relation to that patent. IIa did ask for a single costs order regardless, which I took to mean that this approach was to be applied notionally and ultimately one costs order was to be made with the appropriate set-off. This would have resulted in a significantly different starting point in the analysis on liability for costs.

18 Before proceeding to consider this approach, however, I first clarify the terminology. The phrase “issue-based approach” is one derived from English practice on costs. In order to assess IIa’s arguments and to consider the appropriate approach to adopt in this case, it is necessary to define the “issue-based approach” with some clarity.

19 First, the “issue-based approach” was used by IIa to refer to an approach where, instead of considering who has succeeded overall, the judge considers each issue independently and then considers which party has been successful on that issue, with the costs implications flowing from each issue separately. This was the issue-based approach as summarised by Chadwick LJ in Summit Property Limited v Pitmans (A Firm) [2001] EWCA Civ 2020 (“Summit”) at [27]:

This form of the issue-based approach differs from the Tullio approach in how it approaches the scope of the “event”. It is in this sense that IIa proposed using the issue-based approach, to distinguish between SG 872 and SG 508. I refer to this as the Summit approach.

20 Second, an “issue-based approach” may refer to the approach used by the English Patents Court in Monsanto Technology LLC v Cargill International SA (No. 2) [2008] FSR 16 (“Monsanto”) at [7]–[8]. I refer to this as the Monsanto approach and a fundamental point of distinction with the Summit approach is that in the Monsanto approach the court still begins by considering who the successful party overall is, and any issues for which the successful party has been unsuccessful is assessed in that context. As I will explain (see [25] onwards), this is the approach now commonly applied in the English Civil Procedure Rules 1998 (SI 1998 No 3132) (UK) (“CPR”).

21 I deal first to the Summit approach advocated by IIa. This is not the first time that the Summit approach has been raised in Singapore. In Khng Thian Huat, V K Rajah JC (as he then was) preferred to reserve the approach for the exceptional case, explaining at [21]:

22 IIa, in advancing this approach, relied on the case of Dometic Australia Pty Ltd v Houghton Leisure Products Pty Ltd (No 2) [2019] FCA 57 (“Dometic”), which it cited as an example of the issue-based approach in the Summit sense. In that case, the plaintiffs had filed a claim of patent infringement while the defendants counterclaimed on the invalidity of the patent. The court had held that the plaintiffs had failed to establish infringement and that the defendants had also failed to prove invalidity: Dometic at [2]. Relevant to the present proceedings was the approach that White J took in apportioning the costs, where he considered that approximately 60% of the trial was concerned with the infringement claim (which the plaintiffs failed on) and approximately 40% of the trial dealt with invalidity (which the defendants failed on). As such, as a starting point, the plaintiffs should pay 60% of the trial costs to the defendants, while the defendants should pay 40% of the trial costs to the plaintiffs. This was netted, giving the defendants 20% of its trial costs to be paid by the plaintiffs (which was then adjusted on the basis of other factors to 30%): Dometic at [43].

23 However, Dometic was a case where it was apparently not possible to identify a “winning” or “losing” party: the plaintiff had failed to establish its claim of infringement and the defendant had failed to establish its claim of invalidity. Each side had won and lost something, almost in equal proportions. Both sides failed in the issues that they had brought for the court’s consideration. This was not the situation at hand. All the other patent cases raised by IIa – Schütz (UK) Limited v Werit UK Limited, Protechna SA [2010] EWHC 1197 (Pat), Novartis AG and CIBA Vision AG v Johnson & Johnson Medical Limited and Johnson & Johnson Vision Care, Inc [2009] EWHC 2029 (Pat), Omnipharm Limited v Merial [2012] EWHC 172 (Pat) and Vernacare Ltd v Environmental Pulp Products Ltd [2012] EWPCC 49 – indicate that the courts have taken the Monsanto approach. Summit is not paradigmatic of the issue-based approach; it is better explained, at least in this jurisdiction, as a specific application that has no relevance where an overall winner may be identified, which is the situation at hand.

24 Summit, therefore, was of no application to this case.

25 I turn to the Monsanto approach. Colin Birss et al, Terrell on the Law of Patents (Sweet & Maxwell, 19th Ed, 2020) (“Terrell”) at paras 21-231 and 21-233 summarises the Monsanto approach as follows:

26 As summarised, similar to Re Elgindata (No 2) and accordingly the Tullio approach, the starting point is that the overall winner is awarded the costs of issues on which he has won, and costs which cannot be allocated to any particular issue (termed the “general costs of the action”). The approach then focuses on whether there are sufficiently circumscribed issues on which the winner did not succeed, such that the overall winner can be deprived of the costs of that issue or, even further, that an adverse costs order on that issue would be appropriate. The key difference between this approach and the Tullio approach is the threshold for making an adverse costs order against the overall successful party. Although there were initially indications that the court had to find a “suitably exceptional” case to do so (see Actavis Limited v Merck & Co Inc [2007] EWHC 1625 (Pat) at [26]), it was settled in later cases that such an order would be made simply when “it is appropriate in all the circumstances of the individual case” [emphasis added]: Hospira UK Limited v Novartis AG [2013] EWHC 886 (Pat) (“Hospira”) at [4]. This is the approach used in most of the patent cases cited by IIa.

27 The Monsanto approach and the Tullio approach share the same source in Re Elgindata (No 2) ([11] supra). The premise set by Re Elgindata (No 2), that costs follow the event, ensures that parties assess the strength of their case before choosing to bring the action or to defend the action. Dyson J explained this rationale in R v Lord Chancellor, Ex parte Child Poverty Action Group and another matter [1999] 1 WLR 347 at 356:

28 The rule, also known as the “indemnity principle”, functions at the same time as a compensatory rule for the winning party, thus enhancing access to justice: Then Khek Koon and another v Arjun Permanand Samtani and another and other suits [2014] 1 SLR 245 at [165]. The indemnity principle enhances access to justice by ensuring that successful litigants are not put out of pocket by having to seek recourse through the legal system in order to vindicate or defend their rights, as the case may be. A litigant with a meritorious cause may be deterred from participating in litigation if he knows that he will not be able to recover his legal expenses even if he prevails at the close of the proceedings.

29 Lord Woolf MR, when reforming the principles relating to costs in the CPR, which came into effect in 1999, adopted the issue-based approach out of a concern that the indemnity principle did not always encourage good conduct in litigation. His Lordship explained in AEI Rediffusion Music Ltd v Phonographic Performance Ltd [1999] 1 WLR 1507 (“AEI”) at 1523:

Singapore’s rules, while still premised on the Tullio approach, have also incorporated issue-based flexibility. In our ROC, provision for a court to reduce or order costs against a party who has failed to establish any issue which he has raised in the proceedings is found in O 59 r 6A of the ROC where a party has “unnecessarily or unreasonably protracted, or added to the costs or complexity of those proceedings”.

30 As against the pursuit of discipline in litigation by disallowing costs or making adverse costs orders, a countervailing factor has been recognised both in Singapore and post-CPR England. This is that even if it is desirable for parties to scope their arguments appropriately, there is at the same time fairness and justice in allowing a party to ventilate all reasonable arguments without fear of adverse costs consequences if the court should later disagree. Thus, Re Elgindata No 2 ([11] supra) recognises that costs do not cease to follow the event simply because the successful party did not win on every point. As the Court of Appeal explained in Ng Eng Ghee and others v Mamata Kapildev Dave and others (Horizon Partners Pte Ltd, intervener) and another appeal [2009] 4 SLR(R) 155 at [1], “unmerited barriers in the path of recovering reasonably incurred costs might well have the chilling effect of deterring parties, in future, from legitimately pursuing or defending their rights”. In the UK, similarly, even in the context of the CPR, Hale LJ in Universal Cycles Plc v Grangebriar Limited [2000] CPLR 42 warned that “[the issue-based] approach can be taken too far as is clear from the leading case of Re Elgindata (No 2)”, cautioning that it could create a “deterrent effect on claimants who are in the right overall”. Jacob LJ, too, stated in Smithkline Beecham plc and Glaxosmithkline UK Ltd v Apotex Europe Ltd and Others [2005] 2 Costs LR 293 (“Smithkline Beecham”) at [26], after confirming that the issue-based approach was to be used as far as reasonably possible after the CPR, that this approach was “not the be-all and end-all. Whether or not ‘it was reasonable for a party to raise, pursue or contest a particular allegation’ remains a relevant factor to be taken into account as part of the conduct of the parties”.

31 The following principles are therefore not controversial. Where costs follow the event, litigants are encouraged to consider well their chances of success, and access to justice is enhanced when successful parties are indemnified. In both the traditional and issue-based approach, good conduct in litigation is encouraged through the reduction and reversal of costs orders. The issue-based approach seeks to better incentivise discipline by a greater granularity in considering the issues in the litigation. In both approaches, nevertheless, adverse costs orders are tempered with the same regard to preserve the reasonable pursuit of legitimate claims.

32 Patent cases raise specific concerns. A claim in infringement, for example, may raise different factual issues depending on the specifics of the defence of non-infringement and whether there is a counterclaim of patent invalidity. Further, a device may contain multiple patents, which may require different sets of experts to establish their novelty and inventiveness. The commercial value of matters at stake, moreover, may eclipse litigation discipline in the priorities of litigants and their legal advisors. While the position in the English courts after 1999 was influenced significantly by the reforms introduced by the CPR, Terrell ([25] supra) explains at para 21-228, that apportionment had been a concern of the English Patents Court even prior to the introduction of the CPR. The learned authors further explain at para 21-229 that in making orders apportioning the costs of an action and counterclaim, in the exercise of its discretion, “the court had to reconcile two competing principles, on the one hand, prima facie, the costs of litigation should follow the event, but on the other hand the costs order ought to reflect the extent to which costs have been thrown away by one party raising and pursuing unsuccessful points”.

33 In a similar vein Jacob LJ in Smithkline Beecham stated at [25]:

34 Reference, in this context, should be made to Rediffusion Simulation Ltd v Link-Miles Ltd [1993] FSR 369 (“Rediffusion”) at 410, which was mentioned by Jacob LJ in the quote from Smithkline Beecham ([30] supra) above. Premising his decision upon Re Elgindata (No 2) ([11] supra), Aldous J acknowledged that the Re Elgindata (No 2) principles “are in general applicable to patent actions”, but went on to state that “it is necessary in patent actions to take into account the large number of issues involved, and the very extensive costs that can be incurred”. He added therefore, the further principle that “the costs order in patent actions should, in appropriate cases, reflect the extent to which significant sums of costs have been thrown away by reason of one party, albeit successful overall, raising and pursuing unsuccessful points”.

35 In Rediffusion, Aldous J proceeded to consider each of the issues that the defendant had raised in its claim of invalidity. Of the seven grounds of invalidity alleged by the defendant, Aldous J refused to grant the certificates for four of those grounds and one part of another, on the basis that they were hopeless and unreasonable to pursue, or were not ultimately pursued (Rediffusion at 412–413). This was pursuant to the rule in the Rules of the Supreme Court 1965 (SI 1965 No 1776) (UK) as amended (the applicable rules of civil procedure at the time) that costs in patent trials could only be recovered for the issues or particulars that were certified by court to have been proven or reasonable: Rediffusion at 412. Having denied the defendant some of its costs relating to issues that it had raised that were not proven or reasonable, Aldous J then considered the submission from the plaintiff that the court should order the defendant to pay the plaintiff’s costs on those same issues. Aldous J began by noting (Rediffusion at 413): “It would be very unusual for a defendant, who succeeded in establishing that a patent was invalid, to be ordered to pay the plaintiff’s costs on any issue” [emphasis added]. That being said, Aldous J then concluded that this was an “unusual case”, in that “the defendant had explored in cross-examination and evidence many points which were not relevant to [the court’s] decision, and they were in the main responsible for the very long trial” (Rediffusion at 413). The costs order, therefore, was for the general costs of the action and counterclaim to be paid by the plaintiff, but the defendant was to pay to the plaintiff “the costs of the issues of novelty, inutility and false suggestion and a quarter of the costs relating to the issue of obviousness” (Rediffusion at 413).

36 Rediffusion shows therefore that the Tullio approach can be applied with due attention to the particular nature of patent actions with good effect. Post-CPR, the key distinction between the Tullio and Monsanto approaches is the threshold for making an adverse costs order against the overall successful party. Rediffusion, because of its premise of Re Elgindata (No 2), required an exceptional case before the imposition of an adverse costs order, whereas the Monsanto approach as elucidated in Hospira ([26] supra) simply requires such an order to be appropriate in all the circumstances (as highlighted at [26] above). In so doing, the Tullio approach gives greater certainty to the overall winner. As the facts explained below show, this point of distinction in respect of adverse costs orders was not in play in the present case, because the facts did not suggest the need for an adverse costs order in respect of SG 508. For present purposes, two points are salient. Re Elgindata (No 2) is the settled law of this jurisdiction: see Tullio Planeta ([11] supra) at [24]. Order 59 r 6A of the ROC also allows the reduction of a party’s costs or even the payment of costs to the other party where an issue has “unnecessarily or unreasonably protracted, or added to the costs or complexity” of the proceedings. Concerns as to apportionment between claims and counterclaims, typically arising in patent claims, may be dealt with within this framework. In my view, the broad principles of Tullio Planeta as augmented by O 59 r 6A reflected the appropriate objectives and were sufficiently flexible to be applied in the present patent case.

37 Any finding on costs must be necessarily contextual and heavily factual. Returning to the case at hand, the overall winner of the suit is E6. In Lord Bingham MR’s words, E6 has won something “of value which [it] could not have won without fighting the action through to a finish” while IIa has not “substantially denied [E6] the prize which [E6] fought the action to win”: Roache v News Group Newspapers Limited and Others [1998] EMLR 161 at 168–169.

38 As the overall winner, E6 would have the general costs of the action and any specific costs relating to SG 872. These would account for the overwhelming majority of costs in this case. All three Samples concerned SG 872. The provenance, chain of custody and various infringement issues were overlapping and multiple witnesses (both factual and expert) testified in respect of both patents.

39 Therefore, in considering whether to make a costs reduction or an adverse costs order, the issue concerned only that portion of the costs relating to SG 508 that would be additional. In this case, there was no question of making an adverse costs order against E6 on SG 508. E6’s pursuit of its claim in court was legitimate and closely linked to SG 872. The two patents, while distinct, may be used together and have commercial value when thus used in gemstone applications. E6’s contention was that Sample 3 was a pink gemstone that was the product of both patents.

40 On the issue of the costs reduction, E6 proposed a 15% reduction, while IIa advocated a 50% reduction. IIa’s suggestion was untenable. In terms of trial dates, both parties agreed in their respective submissions on costs that the number of days spent on SG 508 was around 3.25 days out of 15 days in the second set of trial dates. In fact, as highlighted by E6, the validity of SG 508, which was the only issue that IIa had prevailed on, took up only about 1.75 days out of those 3.25 days. Given that the issues canvassed during the first set of trial dates dealt with the provenance and chain of custody relating to all the Samples, and were therefore also relevant to SG 872, there was no need to apportion the dates for that first set of trial dates. Hence, E6’s proposal for a 15% reduction was a principled one.

41 Nevertheless, there was another set of issues pertinent. Conduct is always a relevant factor for consideration in making a costs order. In the matter of conduct, each party accused the other of excess. On that basis, E6 asked, in addition, for its proportion of costs to be taxed on an indemnity basis. I turn to consider the parties’ conduct and to explain how this affected the result.

42 Awarding costs on an indemnity basis is the exception rather than the norm: Airtrust (Hong Kong) Ltd v PH Hydraulics & Engineering Pte Ltd [2016] 5 SLR 103 (“Airtrust”) at [17]. In Airtrust at [49], Chan Seng Onn J suggested indemnity costs may be appropriate where a party’s conduct falls into any of the following categories:

(a) where the action is brought in bad faith, as a means of oppression or for other improper purposes;

(b) where the action is speculative, hypothetical or clearly without basis;

(c) where a party’s conduct in the course of proceedings is dishonest, abusive or improper; and/or

(d) where the action amounts to wasteful or duplicative litigation or is otherwise an abuse of process.

43 Ultimately, the court must consider that “the party’s conduct was so unreasonable as to justify an award of indemnity costs. Such conduct must reflect a high degree of unreasonableness, and cannot merely be wrong or misguided in hindsight”: Airtrust at [50]. In considering whether indemnity costs are justified, “it is important for the court to bear in mind the context and nature of the dispute in ascertaining whether the case is of such an exceptional nature that it is appropriate to depart from the standard basis of costs”: Airtrust at [52].

44 In the present case, IIa’s main defence lay in its attack on the novelty and inventiveness of SG 872 and SG 508. However, IIa’s overall strategy was to defend every issue, without regard to its merit. For example, the objections it made regarding foreign amendments, post-grant amendments and allegations about E6’s misrepresentations to the Intellectual Property Office of Singapore lacked conviction and were not relevant (see Judgment at [326]–[331]). ⁵ Numerous and multiple pieces of prior art were raised on the main claims (Claims 1 and 62 od SG 872 and Claim 1 of SG 508), of which only a fraction was key when the experts took the stand.

45 There were two points, in particular, where it could be said that IIa’s pursuit of certain issues was speculative, hypothetical and without basis. The first was IIa’s pursuit of the issues of the provenance and chain of custody relating to the Samples. ⁶ As a result of this emphasis, IIa spent 5.5 days out of six days that E6’s witnesses were on the stand cross-examining them on those issues during the first set of trial dates. ⁷ E6 was forced to establish every single link in the handling of the Samples. There were also aspects of IIa’s conduct that suggested that these arguments were, in part, merely strategic. For example, when E6 initially had difficulty locating its trap purchaser, Mr Pascal Pierra, IIa sought to expunge his affidavit. Subsequently, however, when IIa’s bluff was called and Mr Pierra was located, IIa chose not to cross-examine him. ⁸ IIa also made various allegations of fraud, evidence tampering, witness collusion and manufacturing of evidence without any basis. I alluded to this stance in the Judgment at [339], [360], [372], [379], [394], and [410].

46 The second concerned IIa’s alleged confidential process. After trial, I found, first, that it was not believable that IIa’s method of manufacturing CVD diamonds had not been documented before (Judgment at [452]–[453]); and secondly, that the confidential process was “contrived so as to avoid liability for infringement” (Judgment at [454]). The confidential process took up a significant amount of time at trial. It was appropriate to hold this against IIa in considering the issue of costs.

47 IIa argued that there was also misconduct on E6’s part, for three reasons. First, IIa complained that E6 did not engage in settlement negotiations despite offers to settle sent on 1 February and 3 March 2017. This argument was a non-starter, given my views on the OTS.

48 Second, IIa also claimed that E6 had ambushed it by disclosing a large number of documents before and during trial, on 18 March 2019 (before the first tranche) and on 22 March 2019 (during the first tranche), which necessitated urgent inspection of the documents during trial. ⁹ From what I saw of the documents, these related to the minutiae of the provenance and chain of custody of the Samples, comprising documents such as daily records of the Samples, and were the result of IIa’s wide-ranging and undiscerning demand for E6 to prove every link in the chain from the Samples’ origins as if this were a drug trafficking case, relying on Mohamed Affandi bin Rosli v Public Prosecutor and another appeal [2019] 1 SLR 440 (in which context the Court of Appeal made clear in any event at [41] that “speculative arguments” about a break in the chain of custody would not be entertained: Judgment at [360]).

49 Third, IIa argued that E6 had not identified what the independently valid claims (“IVCs”) for SG 872 and SG 508 were, causing it to incur unnecessary costs. Citing the case of Edward Lifesciences AG v Cook Biotech Incorporated [2009] EWHC 1443 (Pat) (“Edward Lifesciences”) at [6], IIa argued that E6’s conduct in relation to the IVCs warranted an adjustment of the costs order in IIa’s favour.

50 IVCs are used in the UK as a case management technique under Part 63 r 8 of the CPR, where the patentee identifies a set of independently valid claims (which may be independent or dependent claims) in order to narrow the issues for trial: see The White Book Service 2020 vol 2 (Geoffrey Vos et al eds) (Sweet & Maxwell, 2020) at para 2F-9.6.1; Terrell ([25] supra) at para 19-100. In the present case, in the lead-up to trial, there was some uncertainty over whether dependent claims could be independently valid in the same way as in the UK, following the Court of Appeal’s decision in Sunseap Group Pte Ltd and others v Sun Electric Pte Ltd [2019] 1 SLR 645 (“Sunseap”) at [70]. Whereas the Court of Appeal noted that “[i]f the court finds in the defendant’s favour that the independent claims are invalid, it follows that the dependent claims must also fall”, the position did not appear to be the same in the UK. The appropriate interpretation of Sunseap, and the characterisation of an IVC, remained live issues for trial. Therefore, the approach used in Edward Lifesciences could not be applied in its full rigour.

51 The above notwithstanding, E6 could and should have exercised more restraint and greater reasonableness throughout the progress of the suit. The pre-trial process was characterised by lengthy interlocutory and discovery applications on both sides. The assistant registrar’s decisions in [2017] SGHCR 16, concerning E6’s 165 requests for further and better particulars, and [2018] SGHCR 13, which the decision records was heard over four days of oral argument and involved about 400 pages of written submissions, give a flavour of the written submissions, affidavits and oral arguments used by both parties in a wide gamut of interlocutory applications and pre-trial conferences, with much intervening correspondence. The nuggets that were valid in the points taken were accompanied by other matters unreasonably asserted or defended. Every point that could conceivably be argued very possibly was; more consideration for courtesy to fellow members of the Bar and respect for judicial resources would have been appropriate. Coming back to the issue of IVCs, such a practice would be a useful means in patent cases of enabling technical experts to weigh in early and bring efficient definition to issues of validity. In this case, E6 initially asserted 33 claims in SG 872 and 24 claims in SG 508. After I asked E6 in a pre-trial conference to identify the IVCs, it identified 31 claims in SG 872 and 12 claims in SG 508 as IVCs. ¹⁰ It was not until after Senior Counsel took over conduct of the case, after experts’ affidavits were filed, and only at the second tranche of trial dates, that arguments on validity focused on Claims 1 and 62 of 872, and Claim 1 of 508.

52 E6’s conduct was relevant because in a case where indemnity costs are ordered in a party’s favour, under O 59 r 27(3) of the ROC, any doubt as to the reasonableness of any expense is resolved in the receiving party’s favour. In view of E6’s conduct, I did not think that E6 ought to have the benefit of doubt resolved in its favour on issues pertaining to the reasonableness of its costs during taxation.

53 Under O 59 r 5(b) of the ROC, the court in exercising its discretion as to costs shall consider, to the extent as may be appropriate in the circumstances, “the conduct of all the parties, including conduct before and during the proceedings”.

54 One aspect of the court’s assessment of the parties’ conduct is to consider the proportionality of the manner in which each side ran its case. In Singapore Shooting Association and others v Singapore Rifle Association [2020] 1 SLR 395, the Court of Appeal took account of parties’ conduct when making no order on the costs of proceedings both in the court below and before it (at [183]). It made clear at [174] that the outcome reached on the substantive issues is not the only consideration taken into account in awarding costs. At [175], lawyers were reminded of their professional obligation under r 17(2)(e)(i) of the Legal Profession (Professional Conduct) Rules 2015 (S 706/2015) (“LPPCR”) to evaluate, together with their client, whether any consequence of a matter involving the client justifies the expense of pursuing the matter. Sundaresh Menon CJ highlighted at [180]:

55 In each case, the costs order would depend upon all the facts of the case. In my judgment, my discretion as to costs was better exercised in the present case by giving E6 the whole costs of the action on liability without any reduction, but to be taxed on a standard basis. Returning to the Tullio approach, E6 was the overall winner. As such, the mere fact that SG 508 was revoked did not mean that E6 should not, by that fact alone, receive the whole costs of the action. A reduction could be made if the additional issues had “caused a significant increase in the length of the proceedings” (see Tullio Planeta ([11] supra) at [24]; and O 59 r 6A, ROC). In ascertaining whether E6’s assertion of SG 508 made the trial lengthier, I considered that it was IIa’s behaviour which prolonged the trial and made matters unduly complex. It was appropriate for E6 to be compensated in costs for each of the multiple and multifarious defences IIa attempted, even in relation to any additional costs specific to the validity of SG 508. At the same time, while E6 urged for IIa’s misconduct to be taken into account by means of a costs order on an indemnity basis, E6’s own lack of discipline was also relevant as to whether I would make that exceptional order. Taking all of these factors into consideration, giving E6 the whole costs of the action on a standard basis was the solution that better encouraged proportionality and reasonableness.

56 In conclusion, E6 was awarded the costs of the action on a standard basis. Costs for the costs hearing were ordered in favour of E6 and fixed at $3,000, including disbursements.