1 Order 11 Rule 5(2) of the Rules of Court 2021 (the “ROC 2021”) enjoins the court from ordering the production of any document that is part of a party’s private or internal correspondence, in any format and wherever such correspondence may be stored. This is subject to two exceptions under Rule 5(2)(a) and (b): it is a “special case”; or the correspondence is a known adverse document. Order 11 Rule 5(2) was introduced in the ROC 2021 following the revisions proposed by the Civil Justice Commission (the “CJC”) and has no precedent in past versions of the Rules of Court.

2 This is an application by the claimant for production of 14 categories of documents pursuant to O 11 r 3 of the ROC 2021. The second defendant, who is the respondent in the application, takes objection to the majority of the requests on the basis that they relate to internal correspondence and the exceptions under rr 5(2)(a) and (b) do not apply. The second defendant has also argued that six of the 14 categories are requests for documents subject to legal privilege, and that he does not have possession or control of documents in the remaining eight categories other than what has already been disclosed in the proceedings.

3 As the parties’ arguments pertained to the rule in O 11 r 5(2), which does not appear to have been the subject of a reported judgment thus far, and other points of law largely pertaining to legal privilege, I reserved judgment. I did so in particular to more carefully consider the scope of O 11 r 5(2), which appeared to me to be a significant departure from the previous regime of discovery of documents, considering the frequency with which private and internal correspondence were adduced under that regime. Having considered the submissions and the requested categories of documents, I dismiss the application. These are the reasons for my decision.

4 The claimant in Originating Claim No. 10 of 2022 (“OC 10”) is Cachet Multi Strategy Fund SPC on behalf of Cachet Special Opportunities SP (“Cachet”), a hedge fund incorporated in the Cayman Islands. ¹ The third defendant is Haven Global Network Pte Ltd (“Haven”), a Singapore-incorporated company. ² The first defendant is Mr Feng Shi, also referred to as Tristan Shi (“Mr Shi”). Mr Shi was the co-founder, Chief Executive Officer (“CEO”), Chairman, majority shareholder, and director of Haven since 1 August 2018. ³ The second defendant is Mr Alex SK Liu (“Mr Liu”), a co-founder and director of Cachet. ⁴

5 Cachet entered into a Subscription Agreement dated 3 September 2018 (the “Subscription Agreement”) with Haven, to subscribe for a 10% shareholding in Haven at a price of US$20m (the “Investment Sum”). ⁵ Pursuant to the Subscription Agreement, Cachet paid the full Investment Sum to Haven on or around 5 October 2018. ⁶ Cachet’s founder and Chief Executive Officer, Ms Angela Chow (“Ms Chow”) was also appointed a director of Haven. ⁷

6 Cachet claims that prior to its entry into the Subscription Agreement, from in or around April 2018 to September 2018, Mr Shi made several representations to induce Cachet to invest in Haven (the “Alleged Representations”). In summary, the Alleged Representations by Mr Shi were: ⁸

(a) Haven was undertaking a project (the “Haven Project”) which involved the development of a peer-to-peer crypto financial products marketplace platform (the “Blockchain Platform”) built on blockchain and smart contract technologies, for the purpose of launching insurance products. The Blockchain Platform could be rolled out by September 2018 or November 2018.

(b) AXA General Insurance Co., Ltd (“AXA”), an international insurance company, would participate in the Haven Project by issuing insurance products on the Blockchain Platform.

(c) A document titled “Preliminary Financial Statements” (the “Financial Statements”), which purported to set out Haven’s financial position as of 30 June 2018, had been audited by PricewaterhouseCoopers and signed off by them.

(d) Mr Shi had as of 30 June 2018 already made a capital contribution of US$1.15m in cash to Haven (the “Capital Contribution”).

(e) The individuals Mr Bryan Liu, Mr Ivan Kim and Mr Daniel Kim were Haven’s full-time staff with whom Haven had entered into employment contracts (the “Employment Contracts”).

7 Cachet claims that it entered into the Subscription Agreement relying on the truth of the Alleged Representations and having been induced to do so by them. ⁹ According to Cachet, since late 2018, it gradually discovered that the Alleged Representations were false and made by Mr Shi fraudulently. Specifically: ¹⁰

(a) No Blockchain Platform was launched by September or November 2018, or at all. What Haven subsequently purported to run was a business in online poker and/or gambling.

(b) AXA never agreed to issue insurance products through the Blockchain Platform.

(c) PricewaterhouseCoopers did not audit or sign off on the Financial Statements.

(d) No Capital Contribution was made by Mr Shi as at 30 June 2018. To date, Mr Shi has only contributed US$200,000 into Haven.

(e) Mr Bryan Liu, Mr Ivan Kim and Mr Daniel Kim were never employed by Haven and no employment contracts were entered into with them.

8 According to Cachet, Mr Liu was aware at the material time that the Alleged Representations were false and made by Mr Shi fraudulently. Cachet reasons that because Mr Liu was a co-founder and director of Haven and “actively involved in the Haven Project since its inception”, Mr Liu must therefore have been aware of “material aspects” of Mr Shi’s and Cachet’s discussions, including the Alleged Representations. ¹¹

9 On 18 April 2019, Cachet rescinded the Subscription Agreement and demanded that Haven return the Investment Sum within five days (the “Demand”). At a board meeting of Haven on 29 April 2019, Ms Chow also made it clear that Haven should return the Investment Sum to Cachet, failing which Cachet would commence legal proceedings against Haven. ¹² Cachet claims that Haven, under the influence and/or direction of Mr Shi, Haven and Mr Liu (whom Cachet refers to as “Co-Conspirators”), refused to comply with the Demand. This led Cachet to commence SIAC Arbitration No. 283 of 2019 against Haven on 2 September 2019 (the “Haven Arbitration”). In the Haven Arbitration, Cachet sought the return of the Investment Sum primarily on the basis that it had been induced to invest in Haven by various fraudulent misrepresentations, including the Alleged Representations. ¹³

10 Cachet further claims that instead of repaying the Investment Sum, Haven, Mr Shi and/or Mr Liu sought to use the Investment Sum to enrich themselves. They allegedly did this by, amongst other things, causing payments to be made from Haven’s bank account for purported “business expenses” or “salary payments”; refusing disclosure of Haven’s employment contracts; paying Mr Shi and Mr Liu sign-on bonuses and extravagant salaries; attempting to remove Ms Chow as an authorised signatory of Haven’s bank account; causing payments to be made from Haven’s bank account on 29 August 2019; and causing or participating in Mr Shi’s refusal to pay the balance Capital Contribution to Haven. ¹⁴

11 On 17 March 2021, the arbitral tribunal in the Haven Arbitration (the “Tribunal”) released an interim award (the “Interim Award”), finding that all but one of the Alleged Representations were false and made fraudulently by Mr Shi. The Tribunal declared that the Subscription Agreement had been validly rescinded by Cachet and ordered Haven to repay the Investment Sum within 21 days of the Interim Award. According to Cachet, Haven did not do so. ¹⁵

12 On 29 March 2021, Cachet commenced proceedings in Hong Kong to enforce the Interim Award against Haven (the “Hong Kong Enforcement Proceedings”). On 28 April 2021, the Hong Kong Court of First Instance granted Cachet leave to enforce the Interim Award and awarded costs of the Hong Kong Enforcement Proceedings to Cachet. On 11 August 2021, Cachet successfully obtained recovery of the Investment Sum, after having allegedly expended HK$352,431.33 in legal costs and disbursements in the Hong Kong Enforcement Proceedings. ¹⁶

13 In a final award dated 26 November 2021 (the “Final Award”), the Tribunal ordered that:

(a) Haven was to pay Cachet S$1,112,587.12 as Cachet’s legal and other costs of the Haven Arbitration.

(b) Haven was to bear the final costs of the Haven Arbitration, including the amount paid by Cachet thus far towards the costs of the Haven Arbitration, which is S$147,751.46.

(c) Haven was to pay Cachet simple interest at the rate of 5.33% per annum on the above sums, running from 26 November 2021 to the date of payment.

(d) Haven was to pay Cachet S$2,470,783.56, representing the interest on the Investment Sum at the rate of 5.33% per annum from 18 April 2019 to 11 August 2021, being the date on which Cachet successfully recovered the Investment Sum from Haven (see [12] above).

14 On 27 March 2019, Mr Shi had also executed a Deed of Undertaking in favour of Cachet and Haven, undertaking to contribute the full amount of the Capital Contribution to Haven on or before 30 June 2019 (the “Deed”). According to Cachet, Mr Shi did not comply with the undertaking, having only paid US$200,000 to Haven on or around 9 July 2019 (see [7(d)] above). On 2 October 2019, Cachet commenced a separate arbitration (the “Deed Arbitration”) against Mr Shi to enforce the Deed, and successfully obtained an award in its favour on 19 May 2020 (the “Deed Award”). In the Deed Award, Mr Shi was ordered to pay the balance Capital Contribution of US$950,000 to Haven, along with interest, Cachet’s legal and other costs in the Deed Arbitration in the amount of S$127,471.67, and the costs of the Deed Arbitration in the amount of S$63,202. ¹⁷

15 Cachet alleges that to date, Mr Shi has not complied with the Deed Award, and further that Haven, under the influence and/or direction of Mr Shi, Haven and Mr Liu as co-conspirators, have “failed, refused and/or neglected to procure Mr Shi to do the same”. ¹⁸ According to Cachet, it also commenced enforcement proceedings in California on 15 October 2020 to enforce the Deed Award against Mr Shi (the “US Enforcement Proceedings”) and successfully obtained a judgment by the Superior Court of California against Mr Shi on 3 November 2021 (the “Californian Judgment”). Mr Shi has not complied with the Californian Judgment. According to Cachet, it expended US$13,295.65 in the US Enforcement Proceedings.

16 In OS 10, Cachet claims that (a) Mr Shi is personally liable for fraudulent misrepresentation; ¹⁹ and (b) Haven, Mr Shi and Mr Liu are co-conspirators who conspired to defraud and/or mislead Cachet into investing in Haven and thereafter illicitly enrich themselves (the “Alleged Conspiracy”). ²⁰ Specifically in respect of the Alleged Conspiracy, Cachet claims that Haven, Mr Shi and Mr Liu conspired:

(a) to have Mr Shi make the Alleged Representations to fraudulently induce Cachet’s investment, and thereafter refuse to procure Haven to return the Investment Sum and instead illicitly enrich themselves with the Investment Sum; ²¹

(b) in relation to the Deed Arbitration, to have “caused, persisted and/or participated in” Mr Shi’s defence of the Deed Arbitration and his failure or refusal to comply with the Deed Award; ²²

(c) in relation to the Haven Arbitration, to cause Haven to “vigorously contest Cachet’s claim in the Haven Arbitration”, with Mr Shi and Mr Liu giving evidence on behalf of Haven “in an attempt to cover up and conceal the fraud and Conspiracy perpetrated on Cachet”; ²³ and

(d) to cause Haven to fail, refuse and/or neglect to comply with the Interim Award and the Final Award, ²⁴ and Mr Shi to fail, refuse and/or neglect to comply with the Deed Award and the Californian Judgment. ²⁵

17 Cachet seeks damages for conspiracy and/or fraudulent misrepresentation from Mr Shi, Mr Liu and Haven, comprising:

(a) S$1,260,338.58 and interest on this amount, being the sums due and owing to Cachet from the Haven Arbitration;

(b) S$2,470,783.56, representing the interest on the Investment Sum;

(c) S$170,351.66 and interest on this amount, being the sums due and owing to Cachet from the Deed Arbitration;

(d) HK$352,431.33 and interest on this amount, being the sum Cachet expended in the Hong Kong Enforcement Proceedings; and

(e) US$13,295.65 and interest on this sum, being the sum Cachet expended in the US Enforcement Proceedings.

18 Mr Liu’s defence is that while he was a “co-founder” and director of Haven, he was not on equal terms with Mr Shi. ²⁶ Mr Liu was the Chief Strategy Officer and his role was to give input on general product design. His focus was on the technical aspects of Haven’s operations, such as developing the Blockchain Platform, while Mr Shi as CEO had the final say on all matters, including business decisions, and had sole purview of investor relations. Mr Liu was not privy to specific promises and timelines communicated by Mr Shi to investors, including the Alleged Representations. ²⁷ Even if Mr Liu was aware of the Alleged Representations, he was not in a position to know if they were true or not, being uninvolved in the business, financial and human resource aspects of Haven. ²⁸

19 On the Haven Arbitration, Mr Liu takes the position that the decision by Haven to resist the Demand was “spearheaded by Mr Shi who was the CEO”, and that Mr Liu was not even aware of the Alleged Representations at this time let alone that they were false or made fraudulently. In any case, it was reasonable for Mr Liu to have acted as he did because he was acting as a director in the ordinary course of his duties when Haven resisted the Demand. ²⁹ As one of the directors in Haven which was resisting the claim, Mr Liu gave evidence in the Haven Arbitration to the best of his knowledge, information and belief, as did other directors such as Mr David Hong and Mr Hanh Huyunh Huu. ³⁰ Cachet’s allegations on the findings of the Tribunal on the Alleged Representations do not pertain to Mr Liu. ³¹ As to the Deed Arbitration, this did not concern Mr Liu as the proceedings were between Cachet and Mr Shi. ³² Mr Liu denies the existence of any conspiracy against Cachet. ³³

20 Only Mr Liu has participated in OS 10 thus far. On 6 May 2022 and 14 July 2022, judgment was entered against Haven and Mr Shi respectively, for failure to file their respective notices of intention to contest or not contest the proceedings. It is not disputed that on 9 March 2023, Haven was struck off. ³⁴

21 In Summons No 475 of 2024, Cachet seeks, amongst other things, production of documents listed in two Schedules (the “Schedule 1 Documents” and the “Schedule 2 Documents” respectively). ³⁵ The Schedule 1 Documents consist of six requests, and the Schedule 2 Documents consist of eight requests. I understand the documents to have been grouped into these two Schedules based broadly on the nature of Mr Liu’s objections to production.

22 Regarding the Schedule 1 Documents, Mr Liu’s objection as described by Cachet was on the grounds of “litigation privilege and/or solicitor-client privilege”. ³⁶ In the course of written and oral submissions, it has become clear that Mr Liu also resists production of a number of categories on the ground that the documents are internal correspondence for which no order for production should be made under O 11 r 5(2) of the ROC 2021. Regarding the Schedule 2 Documents, Mr Liu resists production on the ground that all documents in his possession or control have already been disclosed in OS 10, and apart from those documents already disclosed, he does not have or never had any other documents in Schedule 2 in his possession or control. ³⁷

23 I will elaborate on the parties’ submissions within the analysis below.

24 I propose to begin by summarising the applicable principles on applications for specific production under O 11 r 3 of the ROC 2021. Having done this, I will address the specific legal issues that arose during the course of argument in this application. Chief amongst the issues raised were (a) the sufficiency of an averment on affidavit that requested documents are subject to legal privilege and hence protected from production, having regard to the recent decision of the High Court in Lutfi Salim bin Talib and another v British and Malayan Trustees Ltd [2024] SGHC 85 (“Lutfi”); and (b) the scope of the rule in O 11 r 5(2) against ordering production of private or internal correspondence and the exceptions to the rule.

25 The principles from the cases on an application for specific production of documents under O 11 r 3 of the ROC 2021 may be summarised as follows:

(a) The requesting party applying for production of requested documents must satisfy three conditions under O 11 r 3(1):

(i) First, he must properly identify the requested documents, pursuant to O 11 r 3(1)(a). This requirement serves a practical function. The requested documents must be identified with sufficient particularity so that the producing party will know what documents are being requested and can ascertain whether the documents are in that party’s possession or control: Eng’s Wantan Noodle Pte Ltd and another v Eng’s Char Siew Wantan Mee Pte Ltd [2023] SGHCR 17 (“Eng’s Wantan Noodle”) at [48].

(ii) Second, he must show that the requested documents are material to the issues in the case, pursuant to O 11 r 3(1)(b). Materiality is determined by reference to the pleaded cases of the parties, from which the issues in the case are to be discerned. There must be a demonstrable nexus between the requested documents and at least one of the issues. In addition, the requested documents must satisfy the threshold of materiality, which is a higher or stricter threshold than the test of relevance and necessity under the revoked Rules of Court (Cap 322, R 5, 2014 Rev Ed) (“ROC 2014”). Specifically, the threshold of materiality requires the requested documents to have a significant bearing on an issue in a case, such that it could potentially affect the court’s ultimate decision: Eng’s Wantan Noodle at [49(a)]–[49(b)].

(iii) Third, he must provide sufficient evidence that the requested documents are in the possession or control of the requested party. The requesting party must have a reason or some basis for his belief that the documents are in the possession or control of the requested party. The reason and the basis for the belief should be set out in his supporting affidavit: Jeffrey Pinsler, SC, Singapore Civil Practice, Volume II (LexisNexis, 2022) (“Singapore Civil Practice”) at para 30-69). But the condition is arguably not difficult to satisfy. A deposition in the requesting party’s supporting affidavit to the effect that the requested documents are in the possession or control of the requested party is usually enough to constitute “sufficient evidence” of the same: Eng’s Wantan Noodle at [50].

(b) The requested party may challenge the application on the basis that any of the requirements under O 11 r 3(1) have not been met. Specifically, he may file an affidavit to state any of the following grounds of objection (Lutfi at [20]):

(i) the requested documents do not exist;

(ii) the requested documents are protected from production (eg, by legal privilege);

(iii) the requested documents have never been in his possession or control;

(iv) the requested documents were but no longer are in his possession or control, in which case the respondent should explain what has become of such documents; or

(v) the requested party does not know or cannot confirm (i) whether the requested documents were ever in his possession or control; or (ii) what has become of the requested documents that were previously in his possession or control.

(c) Under O 11 r 3(2), if the requested documents are not in the responding party’s possession or control, the court may order that party to file an affidavit stating this, as well as whether that party had such possession or control previously and if so, when that party parted with possession or control and what has become of the requested documents. This means that a requested party who takes the positions at (b)(iv) and (v) above may state those positions in its court-ordered affidavit under O 11 r 3(2), instead of in its affidavit in reply to the requesting party’s application: Lutfi at [21].

26 In the exercise of its powers in an application for specific production, the court is required under O 11 r 1(2) to have regard to two sets of governing principles:

(a) first, as a set of overarching “constitutional principles” (Dai Yi Ting v Chuang Fu Yuan (Grabcycle (SG) Pte Ltd and another, third parties) [2023] 3 SLR 1574 at [13]), the Ideals set out in O 3 r 1(1) which are those of fair access to justice; expeditious proceedings; cost-effective work that is proportionate to the nature and importance of the action, the complexity, difficulty and novelty of the claim, and the value of the claim; the efficient use of court resources; and the aim of achieving fair and practical results; and

(b) second, specific to the context of production, the principles under O 11 r 1(2)(a) and (b) that (i) a claimant is to sue and proceed on the strength of his case and not on the weakness of the defendant’s case; and (ii) a party who sues or is sued in court does not thereby give up the party’s right to privacy and confidentiality in the party’s documents and communications.

27 It is fair to say that the introduction of the Ideals and the principles under O 11 r 1(2) that guide the exercise of the court’s powers in an application for production reflects a sea-change in the philosophy behind the discovery of documents. As the CJC explained in its report (Civil Justice Commission, Civil Justice Commission Report (29 December 2017) (Chairperson: Justice Tay Yong Kwang) (the “CJC Report”), Chapter 8, para 2), the principle that a claimant should proceed on the strength of his case and not the weakness of his opponent’s is intended to discourage speculative litigation pursued on the hope that discovery will yield a “smoking gun”. The principal consideration appears to be the Ideal of proportionality. On the one hand, as Chua Lee Ming J observed in Lufti at [34], it is a rare case in which the further pursuit of specific documents actually results in the uncovering of a “smoking gun”. On the other hand, as observed in the CJC Report (Chapter 8, para 2), discovery is expensive, time consuming and labour intensive. Putting these considerations together, the CJC considered there to be sufficient justification for the existing practice on the discovery of documents to be “changed significantly”: CJC Report, Chapter 8, para 1.

28 To the principles outlined in the cases above, I would venture to add several observations on the requirements of materiality, and possession or control.

29 I begin with materiality. Under O 24 r 5(3) of the ROC 2014, the test in the context of applications for specific discovery was that of relevance, which could take one of two forms: direct relevance or indirect relevance: Dante Yap Go v Bank Austria Creditanstalt AG [2007] SGHC 69 (“Dante Yap Go”) at [18]–[31]. A document would be directly relevant if it is one on which the party relies or will rely; or if it could adversely affect his own case, adversely affect another party’s case, or support another party’s case (O 24 r 5(3)(a) and (b)). A document would be indirectly relevant if it may lead the party seeking discovery of it to a “train of inquiry” resulting in him obtaining information which may adversely affect his own case, adversely affect another party’s case, or support another party’s case (O 24 r 5(3)(c)).

30 The test of relevance in the ROC 2014 has been replaced with that of materiality in the ROC 2014. Materiality is not defined in O 11 r 3(1)(b) beyond the prescription that the requested documents must be “material to the issues in the case”. The definition of materiality is therefore more succinct but at the same time its meaning is less easy to discern. In my view, the reference to “the issues in the case” at least makes it clear that the focus in the assessment of materiality should be the pleadings in the case, which provide the “architectural blueprint” of parties’ cases at trial: Dante Yap Go at [21]. In this respect, the approach under the ROC 2021 and that under the ROC 2014 has not changed. Under both sets of rules, there must be a “demonstrable nexus between the documents sought to be discovered to the pleaded cases of the relevant parties to the main action”: Dante Yap Go at [20]; Tan Chin Seng and others v Raffles Town Club Pte Ltd [2002] 2 SLR(R) 465 at [18].

31 This does not itself explain the nature and degree of the connection required between the request and the pleaded issue. As observed in Eng’s Wantan Noodle at [49(a)], the academic commentaries are aligned that the test of materiality “mandates a higher or stricter threshold than the relevance-necessity test under O 24 of the ROC 2014”. In my view, this understanding of materiality is wholly supported by the CJC’s explanation of the changes to the regime for the production of documents and the introduction of the Ideals, which encourage the pursuit of fair but also practical results. The new regime seeks to make “the current full discovery … rare exceptions rather than the norm”: CJC Report, Chapter 8, para 2. The understanding of materiality as involving a higher or stricter threshold promotes that objective.

32 It will also be observed that the scope of specific production is narrower than that under the ROC 2014 in one obvious way, which is the rule in O 11 r 5(1) that prohibits the court from ordering, except in a special case, production of any document that “merely leads a party on a train of inquiry to other documents” – in other words, indirectly relevant documents. The concept of materiality is therefore narrower than that of relevance in at least this specific regard.

33 In my view, the term “materiality” brings into greater focus an aspect of the analysis that was less prominent in the definition of relevance, both direct and indirect, in the ROC 2014. That is the likelihood that the requested document will have a bearing on the court’s decision on the disputed issue. A reading of the definition of relevance in O 24 r 5(3) reveals that the yardstick for relevance is whether the document could affect a party’s case, whether positively or negatively. I suggest that the introduction of materiality re-focuses the inquiry on the likely impact on the adjudicative outcome. That is aligned with the CJC’s objective of discouraging internecine conflicts at the interlocutory stage that are unlikely to make a difference to the determination of the case (“prevent parties from engaging in unnecessary requests and applications with the hope of uncovering a “smoking gun”…”: CJC Report, chapter 8, para 2), and the promotion of a practical approach to doing justice. This in turn informs the nature and degree of the connection required between the request and the pleaded issue. If a requested document has some affiliation to the pleaded issue but is unlikely to have any real bearing on the way the court determines that issue, it is not the proper subject of an order for production.

34 I turn to the requirement of possession or control. Under the ROC 2014, there existed a distinction between the court’s jurisdiction to grant an order for specific discovery, and its discretion to decide whether or not to grant the order: The Management Corporation Strata Title Plan No 689 v DTZ Debenham Tie Leung (SEA) Pte Ltd and another [2008] SGHC 98 (“DTZ Debenham”) at [26]. The court’s jurisdiction to grant the order would be enlivened when the conditions in O 24 r 5 of the ROC 2014 were fulfilled, namely, if there is sufficient evidence to show that the requested documents are in the possession, custody or power of the other party, and the requested documents are relevant: DTZ Debenham at [29]–[30]. But even if the court’s jurisdiction were so enlivened, the court had a discretion whether to grant the order. That discretion was evidenced by the word “may” in O 24 r 5(1) (“the Court may at any time, on the application of any party to a cause or matter, make an order…” [emphasis added]): DTZ Debenham at [26].

35 As AR Goh Yihan (as he then was) went on to explain in DTZ Debenham at [30], if the applying party deposes in its supporting affidavit its belief that the party from whom discovery is sought has, or at some time had, the requested documents in that party’s possession, custody or power, that would satisfy this particular prerequisite for the court’s jurisdiction, and even if the requested party goes on to state on oath that he does not have the documents concerned, the court would nonetheless retain the jurisdiction to make an order for discovery. But the deposition by the requested party that it does not have the documents concerned could be a matter taken into consideration by the court in exercising its discretion to order discovery, and could defeat the application for discovery notwithstanding that jurisdiction was properly conferred: DTZ Debenham at [31]–[32].

36 I am of the view that the distinction between the court’s jurisdiction and its discretion to make the order continues to apply in the context of O 11 r 3 of the ROC 2021, although the requirements for jurisdiction have been modified. For the court to be seized of jurisdiction to make an order for specific production, the requesting party must (a) properly identify the requested documents; (b) show that the requested documents are material to the issues in the case; and (c) put forward sufficient evidence that the requested documents are in the requested party’s possession or control. Once these requirements are satisfied, the court has jurisdiction to make the order but it nevertheless retains the discretion to decide whether to do so. That is clear from the word “may” in O 11 r 3(1) of the ROC 2021, which remains unchanged from the language in O 24 r 5(1) of the ROC 2014. Adopting the reasoning in DTZ Debenham, if the requesting party puts forward sufficient evidence to show that the requested documents are in the requested party’s possession or control, and the requested party deposes in an affidavit that he does not have possession or control (and, for instance, explains whether he had possession or control previously and, if so, when he parted with possession or control and what has become of the documents), the court retains jurisdiction to make the specific production order but may, in the exercise of its discretion, decline to do so.

37 Cachet’s requests for documents are all framed in a similar way, being requests for “documents created and/or transmitted” within a specified range of dates, “relating to” or “evidencing” a range of topics and issues. Two representative examples are Requests 1 and 2:

38 It is not disputed, and should also be clear from the earlier summary of the pleadings, that the various topics and issues that the requests pertain to – such as “Alex’s role and involvement in conducting Haven’s defence of the Haven Arbitration”; or “Tristan’s Capital Contribution in Haven” – are mentioned in or arise from the pleadings. In framing the requests, however, Cachet’s approach is to seek all documents created and/or transmitted within a specified date range “relating to” each of these topics or issues. A preliminary question that arises whether this approach is generally capable of satisfying the requirement of materiality.

39 At the start of the hearing, I asked counsel for Cachet, Mr Ho Yi Jie (“Mr Ho”), about the permissibility of framing the requests in this manner, having regard to the remarks of the court in EQ Capital Investments Ltd v Sunbreeze Group Investments Ltd and others [2017] SGHCR 15 (“EQ Capital”) at [61]–[63]. EQ Capital was a decision under the ROC 2014. The court in EQ Capital faced several requests framed in a similar manner, being requests for all documents “in relation to and/or in connection with and/or in support of” a particular quoted averment in a pleading. The learned assistant registrar expressed the view (at [62]) that it could not ever be proper to frame a request for specific discovery in this way, for two reasons. First, the request would almost inevitably fall foul of the rule that relevance (ie, the requirement under the ROC 2014) must be established in relation in the class of documents as a whole, because a request framed in this manner would include within its fold a considerable number of documents which would be irrelevant or whose probative value would be so slight that their production would not be necessary for determining the issue to which it was said to be relevant. Such a request “casts a net which is much wider than that permitted by the tests for relevance set out in the Rules”. Second, an order for discovery framed in this manner would almost inevitably result in the making of a superfluous order, because the parties were already under an automatic and continuing obligation to disclose all documents directly relevant to their pleadings (at [63]). The learned assistant registrar explained that the second point was subject to two qualifications, the first being the fact that the court still retains the discretion to make an order for specific discovery despite what is said on affidavit, if there is a “reasonable suspicion” that there are other directly relevant documents that have not been disclosed; and the second that specific discovery can be sought of documents which are indirectly relevant (at [63]).

40 In my view, the reasoning in EQ Capital on requests framed in this manner continues to apply in the context of the ROC 2021, and indeed applies on an a fortiori basis. Given that the threshold under the ROC 2021 is no longer that of relevance (as it was at the time of EQ Capital), but rather the “higher or stricter threshold” of materiality (Eng’s Wantan Noodle at [49(b)]), it stands to reason that the caution against framing a request so broadly as to include documents that are not the proper subject of specific production must still hold true. Likewise, the point that an order for specific production in these terms would be superfluous given the parties’ continuing duty of general production (see O 11 r 6 of the ROC 2021) applies with added force. That is because neither of the two qualifications highlighted by the learned assistant registrar in EQ Capital now exists. As will be discussed at [46] below, the test of “reasonable suspicion” has been replaced with the more stringent test that it must be “plain and obvious” that further discoverable documents exist and have not been produced. And it is no longer possible, save in a special case, for the court to order production of only indirectly relevant documents (see [32] above).

41 I do not think it is technical or pedantic to take issue with the language used to connect the requested documents and the particular pleaded issue or averment, such as “relating to”, “in connection with” or “in support of”. If the request as framed, whether by virtue of such language or some other device, is so broad as to encompass documents that are not the proper subject of specific production, the request cannot be granted. In this regard, I repeat the observation of the learned assistant registrar in EQ Capital at [62], that “[i]f the issue is that the classes cannot be defined with any greater level of specificity because the averment itself is vague and nebulous, then the proper course would be to seek particulars before discovery. Without specificity, there cannot be specific discovery.”

42 The present requests are framed not by reference to a quoted averment in the pleadings, but rather topics or issues arising from the pleadings. In my view, the point of principle remains generally applicable. Whether the contents of the request are directly quoted or paraphrased from the pleadings, the onus remains on the requesting party to ensure that the entirety of the class of requested documents satisfies the requirements of specific production. Insofar as the request seeks documents simply “relating to” a topic or issue – such as “Alex’s role and involvement in conducting Haven’s defence of the Haven Arbitration”; or “Tristan’s Capital Contribution in Haven” – it potentially “casts a net which is much wider than that permitted by the tests for [materiality] set out in the Rules” (to paraphrase EQ Capital at [62]). That is because the language of the request is broad enough to contemplate the production of documents that are only tangentially related to the pleaded issue, and therefore likely to have little bearing on parties’ rival cases on the issue or the adjudicative outcome.

43 For this reason alone, I would have had trouble accepting Cachet’s argument that the classes of documents as framed were capable of meeting the threshold of materiality. However, given that Mr Liu largely focused his objections on other matters such as legal privilege and the rule against production of internal correspondence, I do not decide the application on this basis.

44 One of the grounds on which Cachet disputed Mr Liu’s assertion of legal privilege over the Schedule 1 Documents was that Mr Liu had made “broad, sweeping, wholly unsubstantiated and unsworn assertions” in his reply affidavit that all the documents were privileged. ³⁸ At the hearing, I requested Mr Ho to address me on the High Court’s decision in Lutfi at [32]–[34], which appeared to me to impinge on Cachet’s argument.

45 In Lutfi, the defendant sought to resist two requests in the claimants’ application for specific production on the ground that the defendant did not have within its possession or control any further documents responsive to the requests apart from the documents that had been disclosed. This was stated in the defendant’s affidavit. Chua Lee Ming J observed (at [22]) that it was uncontroversial that an affidavit relating to the production of documents (whether an opposing affidavit or an affidavit filed in response to an order for general or specific production) is generally conclusive subject to exceptions. That was the law under the ROC 2014, and it remained so under the ROC 2021. Chua J then considered the scope of exceptions to that general rule. He observed (at [25]–[31]) that under the earlier High Court decisions in Soh Lup Chee and others v Seow Boon Cheng and another [2002] 1 SLR(R) 604 and Natixis, Singapore Branch v Lim Oon Kuin and others [2023] SGHC 301, the position under the ROC 2014 was that an affidavit in respect of the discovery of documents was not conclusive if there was a “reasonable suspicion” that further discoverable documents existed.

46 Chua J held (at [32]) that the “reasonable suspicion” test under the ROC 2014 “has no place” in the context of applications for specific production under the ROC 2021. The affidavits of the responding party are conclusive and “the court should not go behind the affidavits unless it is plain and obvious from the documents that have been produced, the respondent’s affidavits or pleadings, or some other objective evidence before the court, that the requested documents (a) must exist or have existed; (b) must be or have been in the respondent’s possession or control; or (c) are not protected from production.” [emphasis in original]. Chua J explained (at [33]–[34]) that at the interlocutory stage of proceedings, the court cannot resolve a dispute as to the sufficiency of affidavits relating to production of documents based on contentious affidavits. A “high threshold” would need to be met before the court can “go behind the affidavits relating to production of documents”. Chua J expressed the view that the higher threshold of “plain and obvious” was consistent with the Ideals of expeditious proceedings, cost-effective work and efficient use of court resources, and with the aim of preventing parties from engaging in unnecessary requests and applications with the hope of uncovering a “smoking gun”. The “plain and obvious” test would filter out “often-unproductive applications”.

47 I drew parties’ attention to limb (c) of Chua J’s ruling at [32], that in order for the court to go behind the affidavits, it must be plain and obvious that the requested documents “are not protected from production”. There is no doubt that Chua J was referring to situations where the requested party refuses production and states on affidavit that its refusal is on the ground that the document is protected from production, for instance because it is covered by legal privilege (see O 11 r 5(3)). In such a situation, the requesting party must satisfy the court that it is “plain and obvious” that the requested document is not protected by privilege before the court can go behind the affidavit and order production. I consider Chua J’s ruling on this matter to be clear and unequivocal, notwithstanding the fact that the relevant dispute in Lutfi was on whether the defendant had the documents in his possession or control. A reading of [32]–[34] of the judgment amply reveals that Chua J intended to lay down the general approach where an application under O 11 r 3(1) is disputed by way of contentious affidavits.

48 Having said that, I do not think Lutfi should be understood to detract from the important and well-established position that it is for the party asserting privilege to demonstrate that the preconditions for privilege to subsist are present: ARX v Comptroller of Income Tax [2016] 5 SLR 590 (“ARX”) at [50]. This means that where litigation privilege is asserted by the requesting party, it is for the requesting party to show that (a) there was a reasonable prospect of litigation at the time the document was prepared or created; and (b) the document was created for the dominant purpose of litigation: Skandinaviska Enskilda Banken AB (Publ), Singapore Branch v Asia Pacific Breweries (Singapore) Pte Ltd [2007] 2 SLR(R) 367 at [69]–[77]. Where legal advice privilege is asserted, the onus similarly rests on the requesting party to show that (a) the advice was rendered by a legal professional; (b) the legal professional was acting as legal adviser when he provided the advice; and (c) the communications must have been made in confidence: ARX at [43]. In other words, it will not suffice for the requested party to make a bare assertion of privilege, without satisfying the court of a prima facie case that the requirements for privilege to subsist are met: ARX at [44]–[45]. It can typically do so by “swear[ing] the affidavit of documents claiming legal professional privilege in a way which leads the court to the conclusion that the claim is properly made”: ARX at [45], quoting Australian Hospital Care (Pindari) Pty Ltd v Duggan (No 2) [1999] VSC 131 at [67]. Once the requested party has successfully done that, the evidentiary burden falls on the requesting party disputing the claim of privilege to rebut the prima facie case: ARX at [50]. And it is at this stage that Lutfi applies, and requires the requesting party to demonstrate that notwithstanding the averments on affidavit by the requested party, it is plain and obvious, from the documents that have been produced, the requested party’s affidavits or pleadings, or some other objective evidence before the court, that the requested documents are not privileged from production.

49 For completeness, I address Cachet’s reference to United Overseas Bank Ltd v Lippo Marina Collection Pte Ltd and others [2018] 4 SLR 391 (“Lippo”) and specifically its reference to the remark by Aedit Abdullah JC (as he then was) at [42] that “[t]he essential question is whether the claim of privilege is expressed clearly in some form, so that the matter can be readily determined by the court”. ³⁹ Abdullah JC made this remark in the context of his consideration of whether the absence of a supporting affidavit by the defendants who were asserting legal privilege prevented their invocation of privilege: Lippo at [41]. Abdullah JC held that while the “best form” for the assertion of privilege would be by way of a supporting affidavit, privilege could also be asserted without an affidavit, as long as the circumstances manifested a clear invocation of that privilege. Hence, privilege was not excluded simply because such a supporting affidavit had not been filed: Lippo at [42]. In other words, Abdullah JC was concerned with the necessity of a supporting affidavit in the assertion of privilege, not the sufficiency of the supporting affidavit for the purpose of satisfying the court that privilege was rightly asserted: Lippo at [43]. I therefore do not think Lippo is directly relevant to the above discussion, which only concerns the latter aspect.

50 I now address O 11 r 5(2), which states:

51 It is useful to outline the genesis of O 11 r 5(2) at this point. In the CJC Report, the CJC explained that Rule 5 (in the form drafted at the time) was introduced to prohibit the production of “a document that is part of a party’s private or internal correspondence except in a special case”. It also explained that “special case” was “deliberately left undefined to allow for flexibility and good sense should a rare case emerge”: CJC Report, Chapter 8, para 5.

52 Following the public consultation on the CJC’s proposals that took place from 26 October 2018 to 31 January 2018, the CJC observed that the feedback received was “unanimous in expressing the view that by carving out private or internal correspondence save for exceptional issues, parties lose access to documents that could reveal the true state of affairs”: CJC, Response to Feedback from Public Consultation on the Civil Justice Reports: Recommendations of the Civil Justice Commission and the Civil Justice Review Committee (11 June 2021) (Chairperson: Justice Tay Yong Kwang) (the “Response to Feedback”), para 77. It explained that the “intent behind the proposal to exclude private or internal correspondence was to prevent parties from inundating each other with lengthy correspondence that would have little or no bearing on the issues in dispute in the case”: Response to Feedback, para 78. However, having considered the feedback, the CJC would amend Rule 5 to provide that the court shall not order production of private or internal correspondence except in a special case or if such correspondence are “known adverse documents”. The CJC concluded with the following remarks (Response to Feedback, para 80):

53 It will be observed that the rationale identified by the CJC for the introduction of O 11 r 5(2) was that of preventing a party from being inundated with large amounts of internal correspondence that would ultimately have little probative value. The commentaries have identified a second purpose of O 11 r 5(2), and that is to give effect to the principle in O 11 r 1(2)(b) that a party who sues or is sued in court does not thereby give up his right to privacy and confidentiality in his documents and communications: Singapore Civil Procedure 2024 at para 11/5/3; Singapore Civil Practice at paras 30-101 to 30-102. The learned authors of Singapore Rules of Court: A Practice Guide (2023 Edition) (Chua Lee Ming editor-in-chief) (Academy Publishing, 2023) (“Practice Guide”) explain (at para 11.017) that one of the purposes of O 11 r 5(2) is to narrow the scope of disclosure due to its highly intrusive nature. They cite the CJC’s observation that discovery is “highly intrusive into privacy and confidentiality (even if the browsing of a party’s documents is done by that party’s solicitors and their assistants)” and that “[i]n today’s context, it is even more so since discovery can encompass all the documents and the messages stored in a person’s mobile phone and other electronic devices”: CJC Report, Chapter 8, para 2.

54 I accept the view that O 11 r 5(2) embraces both rationales. The concern identified by the CJC more clearly applies in the context of general production, where there is a greater risk of a party inundating the counterparty with documents, most of which may not be material to the case, in the hope that genuinely adverse documents will be harder for the counterparty to locate or simply that the counterparty will expend time and costs unnecessarily in reviewing the documents. This is a lesser concern in the context of specific production, which is triggered by a party’s own request for documents and constrained by how the requesting party frames the request. However, O 11 r 5(2) is not limited to a particular stage of production and therefore applies to both general and specific production. I accept that a reason why the CJC extended O 11 r 5(2) to specific production was to give effect to the principle in O 11 r 1(2)(b) that a party does not give up the right to privacy and confidentiality when it participates in civil litigation.

55 As to the width of the rule in O 11 r 5(2), it will be observed that the subject matter of the rule is broadly framed, as “any document that is part of a party’s private or internal correspondence”, regardless of the form it is in or the medium in which it is stored. The reference to “any document that is part of” such correspondence means that the rule is broad enough to encompass enclosures or attachments to emails, or any such material transmitted to the receiving party in the course of correspondence. The commentaries are broadly aligned on the meanings of “private correspondence” and “internal correspondence”, and I accept these general definitions as being sufficient for present purposes, but leave their precise contours to be defined in another case:

(a) Private correspondence refers to communications which are intended to be private as between the persons who corresponded with each other: Singapore Civil Practice at para 30-104; Practice Guide at para 11.017.

(b) Internal correspondence refers to correspondence that is internal to the party from which the document is requested, such as emails between individuals and/or departments in the organisation, and not intended to be received by persons outside the party: Singapore Civil Practice at para 30-104; Practice Guide at para 11.017.

(c) The categories of private correspondence and internal correspondence can overlap; it is entirely possible for a communication to be both private correspondence and internal correspondence: Singapore Civil Practice at para 30-104.

56 There are two exceptions to O 11 r 5(2). I begin with the exception in O 11 r 5(2)(b), which has a more immediately obvious meaning. This is a species of document that is already susceptible to disclosure in general production, pursuant to O 11 r 2(1)(b). A “known adverse document” is defined in O 11 r 2(2) as including documents which a party ought reasonably to know are adverse to the party’s case. This means that “known adverse documents” are not limited to adverse documents that the party is actually aware of, but include adverse documents that the party could have knowledge about through reasonable checks and searches: Response to Feedback at para 75. Whether a document is a known adverse document is a matter of objective assessment: Response to Feedback at para 75. The category is more limited than the discoverable documents under O 24 rr 1(2)(b)(i) and 5(3)(b)(i) of the ROC 2014, as it requires that the document be adverse, and not merely have the potential to be adverse: Singapore Civil Procedure 2024 at para 11/2/6.

57 The exception in O 11 r 5(2)(a) is that of a “special case”. The term “special case” is used at various points in the ROC 2021 to denote exceptions. For instance, O 3 r 5(6) stipulates that except in a “special case”, the court will not allow further affidavits to be filed in an application after the responding party has filed his or her reply affidavit. Order 9 r 14(3) provides that the court must not allow any pleading to be amended less than 14 days before the commencement of the trial except in a “special case”. And as mentioned earlier, O 11 r 5(1) prohibits the court from ordering production of any document that merely leads a party on a train of inquiry to other documents, except in a special case. Since the introduction of the ROC 2021, the courts have begun to consider what is required by a “special case” in the various contexts that the term is used (see, for example, CZD v CZE [2023] 5 SLR 806 (“CZD”), Wang Piao v Lee Wee Ching [2024] 4 SLR 540 (“Wang Piao”), Lim Julian Frederick Yu v Lim Peng On (as executor and trustee of the estate of Lim Koon Yew (alias Lim Kuen Yew), deceased) and another [2024] SGHC 53, Grab Rentals Pte Ltd v Khoo Long Hui [2023] SGMC 46 (“Grab Rentals”) and Wee Eng Siang v Muhammad Sholihin Bin Roslan [2023] SGMC 83). I draw out general observations that our courts have made in the process of reckoning the meaning of “special case”, together with the views from relevant commentaries:

(a) The term “special case” should be interpreted with the Ideals set out in O 3 r 1 in mind. The court should discern the Ideals that may be particularly important or relevant to the nature of the proceeding or the type of application at hand: CZD at [19]; Grab Rentals at [18]; Practice Guide at para 11.017.

(b) The term “special case” should also be interpreted having regard to any relevant accompanying or related rules in the ROC 2021: CZD at [20].

(c) The usage of the term “special case” does not ipso facto mean that the rule in the ROC 2021 is more restrictive than its predecessor in the ROC 2014. Regard must be had to how exactly the rules differ, both in terms of content and wording: Wang Piao at [8]–[11].

(d) It may be unwise to articulate a strict or exhaustive definition of a “special case” since this would limit room for “flexibility and good sense should a rare case emerge”: Grab Rentals at [14]–[15], citing the CJC Report (see [51] above). The word “special” evokes notions of exceptionality, peculiarity and distinctiveness, and suggests that some circumstance that is beyond the ordinary ought to be present. The identification of a special case is in the final analysis dependent on the circumstances of the case: Grab Rentals at [15].

58 In the context of O 11 r 5(2)(a), the commentaries have described a “special case” as one which would “only arise in exceptional circumstances” (Singapore Civil Practice at para 30-102) and as an “exceptional case [which] is likely to be justified with reference to the Ideals in Order 3 rule 1(2) and the over-arching principles in Order 11 rule 1(2)” (Practice Guide at para 11.017).

59 My views on the meaning of a “special case” in the context of O 11 r 5(2)(a) are as follows. Given the CJC’s explanation that it deliberately left “special case” in O 11 r 5(2)(a) undefined so as to “allow for flexibility and good sense should a rare case emerge”, it would not be prudent to set down any sort of strict or exhaustive definition. But it stands to reason that, having regard to the objectives of the CJC in introducing a more restrictive approach to production, and within that its rationales for further restricting production of private or internal correspondence, the term “special case” must be interpreted restrictively so that the exception does not swallow the rule.

60 I am also of the view that “special case” in the context of O 11 r 5(2)(a) must mean more than that the requested document is simply material to the case, in the sense that it has a significant bearing on an issue in a case, such that it could potentially affect the court’s ultimate decision (to use the language of Eng’s Wantan Noodle). I discern this from the fact that the only class of material documents under O 11 r 2(1)(a)–(c) that was identified by the CJC as capable of providing an exception to the rule is specifically that of “known adverse documents”, under O 11 r 5(2)(b). That exception was only introduced after the CJC received unanimous feedback that by carving out private or internal correspondence save for exceptional issues, parties lose access to documents that could reveal the true state of affairs (see [52] above). Having received such feedback, the CJC permitted a further exception but, even then, limited this to “known adverse documents”. In the circumstances, it should not therefore be an easy or routine task for a requesting party to persuade the court that there is a “special case” justifying an order for production of private or internal correspondence (that is not a known adverse document, because the exception under O 11 r 5(2)(b) would then apply), simply on the basis that it would have a significant bearing on the disputed issue and could potentially affect the court’s ultimate decision. To use the terminology of the CJC, this would be a “rare case”.

61 The requesting party may seek to justify the request by reference to the Ideals (see [58] above). But it should not be forgotten that the rule against production of private or internal correspondence in O 11 r 5(2) is itself underpinned by the Ideals, such as those of cost-efficiency, proportionality and the efficient use of court resources, as well as the production-specific principles under O 11 r 1(2). Bare assertion of the Ideals by the requesting party would therefore not generally be sufficient. The requesting party would have to show that the relevant Ideals and the probative value of the requested document are sufficient to outweigh the standing policy of the law on the occasion in question.

62 Having set out the general legal principles that would apply to the application, I turn to consider each request specifically. I begin with the requests for the Schedule 1 Documents.

63 Request 1 is for “[d]ocuments created or transmitted from 2017 to November 2019 evidencing Alex’s role and involvement in conducting Haven’s defence of the Haven Arbitration”. Mr Liu resists this on the grounds that the documents requested are privileged and that they consist of internal correspondence. ⁴⁰ Mr Liu has stated on affidavit that the relevant litigation is the Haven Arbitration. ⁴¹

64 I find that this is quintessentially a request for documents that are subject to legal privilege. The very wording of the request suggests that there would have been a reasonable prospect of litigation at the time the requested documents were prepared or created – the litigation being the Haven Arbitration – and that these documents would have been created for the dominant purpose of litigation. I agree with Mr Liu, who was a director of Haven from 10 September 2018 to December 2019, ⁴² that documents evidencing his role and involvement in conducting Haven’s defence in the Haven Arbitration would, by their nature, comprise communications and other documents prepared for the purpose of obtaining legal advice and legal counsel for the Haven Arbitration, such as communications between him and Haven’s solicitors (Shook Lin & Bok LLP and SSW & Associates) ⁴³ and within the company on the conduct of Haven’s defence, and that such documents would be covered by litigation privilege or legal advice privilege, or both. ⁴⁴

65 I also accept Mr Liu’s objection that insofar as Request 1 seeks such documents going as far back as 2017, the timeframe is unjustifiably broad. ⁴⁵ Going by Cachet’s own case, Cachet only gradually discovered that the Alleged Representations were false since late 2018 (see [7] above). Cachet issued the Demand only on 18 April 2019 ⁴⁶ and commenced the Haven Arbitration on 2 September 2019 (see [9] above). In fact, Cachet itself has pleaded that the Alleged Representations were made by Mr Shi “from in or around April 2018 to September 2018” (see [6] above). In the circumstances, a request for documents going as far back as 2017 lacks sufficient nexus to the pleadings.

66 Cachet makes two further arguments in its written submissions. First, Cachet argues that there are at least two documents falling within Request 1 that Mr Liu has not disclosed (although they have been disclosed by Cachet), which are not covered by privilege and show that there are other documents in Mr Liu’s possession or control which he has yet to disclose. ⁴⁷ The first of these is an email from Mr Liu dated 19 August 2019, in which he stated that “In addition to helping Haven develop its products, my role is to ensure the continuity and success of the company including defence against vexatious claims.” ⁴⁸ The second is an email from Mr Shi to other directors of Haven dated 6 September 2019. In this email, Mr Shi informed the other directors that Cachet had commenced arbitration proceedings against Haven and that he had appointed Shook Lin & Bok LLP to act for Haven, a matter concerning which he “underst[ood] Alex and David to be in agreement”. Mr Shi also stated that “information-sharing in relation to the litigation strategy discussed with [Shook Lin & Bok LLP]” would be “disclosed only selectively” to the board, so as to avoid jeopardising Haven’s position in the arbitration, especially given that Ms Chow, who represented Cachet, remained a director. ⁴⁹ I leave aside for the moment the questionable probative value of the first email, and the fact that the second email is likely to be subject to litigation privilege, as pointed out by Mr Liu ⁵⁰ (given that litigation privilege is broad enough to apply to “every communication, whether confidential or otherwise so long as it is for the purpose of litigation” and “protect[s] information and materials created and collected for the dominant purpose of litigation”: Skandinaviska at [44]; see also Colin Liew, Legal Professional Privilege (Academy Publishing, 2nd Ed, 2023) (“Colin Liew”) at para 5.360). The more important point is that these documents do not bring Cachet very far at all in showing that there are other documents in Mr Liu’s possession or control that are not subject to privilege, much less that it is plain and obvious that this is the case.

67 Second, Cachet submits that even if litigation privilege could be asserted over the documents, the litigation privilege would have expired at the time the Haven Arbitration ended on 26 November 2021. Cachet relies for its argument on the decision of the Supreme Court of Canada in Blank v Canada (Minister of Justice) [2006] SCC 39 (“Blank”). It argues that while the Singapore courts have yet to definitively speak on the question of whether litigation privilege would cease to subsist after the relevant litigation has ceased, I should follow the approach of the Canadian courts in Blank. ⁵¹

68 On a careful consideration of Blank, I do not think it is necessary for me to make this determination of law. That is because even under the approach in Blank, Cachet is unlikely to succeed in its argument that litigation privilege no longer subsists over the requested documents.

69 In Blank, the respondent sued the government for fraud, conspiracy, perjury and abuse of prosecutorial powers, arising from an earlier failed prosecution of the respondent. The respondent sought from the government records pertaining to the prosecution under a Canadian statute, the Access to Information Act, R.S.C. 1985, c. A-1. The government refused disclosure on the basis that the documents were privileged. Before the Supreme Court of Canada, the only question was whether the documents were subject to litigation privilege. Giving the lead judgment of the court, Fish J held that the purpose of litigation privilege was to create a “zone of privacy” in relation to pending or apprehended litigation. This meant that “[o]nce the litigation has ended, the privilege to which it gave rise has lost its specific and concrete purpose – and therefore its justification.” (at [34]). But Fish J immediately followed this with an important caveat: “But to borrow a phrase, the litigation is not over until it is over: It cannot be said to have “terminated”, in any meaningful sense of that term, where litigants or related parties remain locked in what is essentially the same legal combat.” (at [34]).

70 Fish J further explained (at [38]–[39]) that litigation privilege may retain its purpose, and therefore its effect, where the litigation that gave rise to the privilege has ended but related litigation remains pending or may reasonably be apprehended. The definition of “litigation” for this purpose would include, at a minimum, “separate proceedings that involve the same or related parties and arise from the same or a related cause of action (or “juridical source”)” and “[p]roceedings that raise issues common to the initial action and share its essential purpose”. On the facts of Blank, the Supreme Court of Canada dismissed the government’s appeal on the ground that the respondent’s action sought civil redress for wrongful prosecution, and that sprang from a different juridical source and was in that sense unrelated to the litigation from which the privilege claimed was born, which was the criminal prosecution of the respondent (at [43]).

71 In the present case, there can be no doubt that the present proceedings, being OC 10, are related proceedings to the Haven Arbitration in the sense contemplated in Blank. Both Cachet and Haven are parties to OC 10. In OC 10, Cachet seeks damages arising from (amongst other things) the Alleged Misrepresentations, ⁵² which are also the subject of the Haven Arbitration. Accordingly, to use the language of Blank, OC 10 “raise[s] issues common to” the Haven Arbitration and “share[s] [the] essential purpose” of the Haven Arbitration, which is to recover losses arising from the Alleged Misrepresentations. Applying the approach in Blank, this means that any litigation privilege that subsists in the requested documents should not be considered to have expired for the purposes of OC 10, merely because the Haven Arbitration has come to an end.

72 Because Cachet’s argument on Blank fails even on its own terms, it is unnecessary for me to consider whether Blank forms part of our law of privilege. The question of whether litigation privilege expires, and if so when it should be considered to have expired, is a vexed one (see, for example, Comptroller of Income Tax v ARW and another [2017] SGHC 16 at [34]; and the discussion in Colin Liew at paras 5.50 to 5.71), and should be answered with the benefit of full arguments when it is necessary to do so.

73 Requests 2 to 6 are for the following documents:

74 Cachet does not dispute that Requests 2 to 6 pertain to internal correspondence, nor would I have considered this seriously disputable. It is plain from the requests that they pertain to discussions within Haven on the various alleged decisions or actions of Haven, Mr Shi and/or Mr Liu on which Cachet bases its claim: (a) the Capital Contributions and how this would be “communicated and/or represented to Cachet”; (b) the decision to resist the Demand and refuse the return of the Investment Sum; (c) the attempts to remove Ms Chow as an authorised signatory of Haven’s bank account; (d) Mr Shi’s liability under the Deed, Deed Award and the Californian Judgment; and (e) Haven’s decision to contest the Haven Arbitration and subsequent non-compliance with the Interim Award and Final Award. This is also clear from the language of each of the requests, which concerns “[i]nternal documents” that have been “created and/or transmitted”. The reference to “communications between Alex and one or more of the Haven members” also reinforces this.

75 Cachet’s response to the objection taken by Mr Liu on the basis of O 11 r 5(2) ⁵³ therefore focuses on the exceptions to the rule. ⁵⁴ Cachet first argues that the exception of a “special case” is applicable. ⁵⁵ This is on the basis that OC 10 “concerns a conspiracy claim against Haven and its two co-founders/directors”, and an order for production of internal correspondence is justified because (a) “as is usual in any conspiracy case, there is information asymmetry between the victim and the conspirators; (b) Haven was found in the Haven Arbitration to have defrauded Cachet; and (c) the documents will have a “material bearing on the issues to be decided” in OC 10. Cachet further submits that the principle on privacy and confidentiality in a party’s documents and communications as contained in O 11 r 1(2)(b) should hold “less weight” given that Haven is not participating in OC 10 and has been struck off.

76 I am unable to accept the submission. I do not think this is a “special case” warranting an exception to the rule against ordering production of internal correspondence. First, Cachet’s argument that the case should be regarded as an exception because, “as is usual in any conspiracy case, there is information asymmetry between the victim and the conspirators”, strikes me as directly falling foul of the principle in O 11 r (1)(2)(a) – which the court is required to bear in mind in exercising its powers to order production – that a claimant is to sue and proceed on the strength of the claimant’s case and not on the weakness of the defendant’s case. Taken to its logical end, it would also mean, as pointed out by counsel for Mr Liu, Ms Samantha Ong (“Ms Ong”), that in virtually every claim for conspiracy where there is information asymmetry between the parties, this would be regarded as a “special case” warranting an order for production of internal or private correspondence. Second, as to Cachet’s suggestion that the internal documents requested will have a “material bearing on the issues to be decided in OC 10”, I have explained that I do not consider the mere meeting of the threshold of materiality to be sufficient to warrant the label of a “special case”. No order for specific production can be made in the first place if the requested documents are not shown to be material. Third, further to and without detracting from the above, I am also conscious that this is not a case where Cachet is lacking in information and documents. OC 10 was commenced on the back of the Haven Arbitration, in which both Mr Shi and Mr Liu provided witness statements and were subject to oral examination. Mr Shi’s and Mr Liu’s witness statements in the Haven Arbitration have been disclosed in OC 10 by Cachet itself, together with the hearing transcript of Mr Liu’s testimony, the witness statement of Ms Chow, the pleadings in the Haven Arbitration, and the Interim and Final Awards. ⁵⁶ Considering the complexion of the case and the disclosures thus far, I consider this to be far from a “special case” warranting the lifting of the prohibition in O 11 r 5(2).

77 Cachet’s second argument is that the requested documents “would include “known adverse documents”” and it should be for Mr Liu to conduct reasonable searches for requested documents that are known adverse documents, and either disclose them or confirm on affidavit that there are no known adverse documents falling within the request. ⁵⁷ In my judgment, this approach impermissibly reverses the burden. In a situation where the documents requested relate to private or internal correspondence and the prohibition in O 11 r 5(2) applies, it is for the requesting party to show that the exceptions in O 11 r 5(2)(a) or (b) apply. I can also see no purpose for requiring the requested party to further aver on affidavit, if it has not already done so, that there are no known adverse documents falling within the request. The requested party is already under an obligation to produce all known adverse documents, under O 11 r 2(1)(b), and this is a continuing obligation under O 11 r 6.

78 For this reason, I find that Requests 2 to 6 should not be granted. It is accordingly unnecessary for me to make a determination on Mr Liu’s submission that the documents pertaining to Requests 2 to 6 were subject to litigation privilege. I add for completeness that I would have had some difficulties with Mr Liu’s submission in this regard. As I pointed out to Ms Ong at the hearing, it appeared to me – for instance, in relation to Request 2, which was for documents “relating to Tristan’s Capital Contribution in Haven” created and/or transmitted from 2017 to September 2018 – to be unclear why all of the documents within the category would be subject to litigation privilege. The events pertaining to Mr Shi’s Capital Contribution took place sometime in 2018 (see [6(d)] above), seemingly before Cachet allegedly began to discover the falsity of the Alleged Representations in late 2018 and its eventual making of the Demand in April 2019 (see [9] above). In other words, it was not sufficiently clear to me that if there were documents falling within this category, there was a reasonable prospect of litigation at the time that all of these documents were prepared or created, or that they were all created for the dominant purpose of litigation. Ms Ong’s response at the hearing was that Mr Liu would be relying on the O 11 r 5(2) rule. I therefore did not understand Mr Liu to have a strong answer to the point.

79 The Schedule 2 Documents are as follows:

80 The original numbering of the eight requests in Schedule 2 ran from s/n 1 to s/n 8, but I have renumbered them from s/n 7 to s/n 14 in the extract above because this was the numbering used in the parties’ oral and written submissions, and this prevents confusion with the requests in Schedule 1.

81 Mr Liu’s position on the Schedule 2 Documents, which he has stated on affidavit, is that all the documents which are in his possession or control have already been disclosed in OC 10, and apart from the documents already disclosed, he does not have, or has never had, any other Schedule 2 Documents in his possession or control. ⁵⁸ According to Mr Liu, all Schedule 2 Documents were also disclosed during the Haven Arbitration, to which Cachet was a party and therefore would have access to the documents. ⁵⁹ Mr Liu further states that he has since conducted a search and, to the best of his knowledge, information and belief, there are no other documents in his possession or control. ⁶⁰

82 The parties do not dispute that the principles in Lutfi govern the matter and Cachet must show that it is plain and obvious that the Schedule 2 Documents must exist or have existed and must be or have been in Mr Liu’s possession or control. ⁶¹ Having considered the parties’ arguments on each of the requests within the Schedule 2 Documents, I am not satisfied that Cachet has met the required threshold. I therefore consider that I should exercise my discretion to refuse the order for production of the Schedule 2 Documents. Without detracting from the above finding which applies to all the requests, I also make the following specific observations on the requests:

(a) Regarding Request 7, which concerns documents evidencing Mr Liu’s alleged roles and duties in various aspects including investor relations, setting business milestones and managing finances and human resource issues in Haven, the thrust of Cachet’s arguments is really that Mr Liu was more heavily involved in these aspects of Haven than Mr Liu is letting on. ⁶² This is of course rejected by Mr Liu as part of his pleaded case (see [18]–[19] above). I do not find Cachet’s argument persuasive. While it is a position that Cachet may advance at trial, it does not assist Cachet as a submission on why it is plain and obvious that there are more documents pertaining to these matters in Mr Liu’s possession or control that have not been produced.

(b) Regarding Requests 8, 9 and 10, Cachet makes a similar argument. It contends that Mr Liu has not produced many documents pertaining to these categories (for instance, “a measly list of seven documents” pertaining to Request 8), ⁶³ and that Mr Liu was more heavily involved in matters such as AXA’s participation in Haven, the Financial Statements, and Haven’s financial and human resource issues including salaries and sign-on bonuses than Mr Liu is purporting to be. ⁶⁴ Consequently, Cachet reasons, it must be plain and obvious that there are further documents in Mr Liu’s possession or control. For the reasons I have explained, I find this unpersuasive.

(c) Request 11 concerns Mr Liu’s pleading in further particulars served by him that he “was paid significantly higher on his previous jobs and took a pay cut to join Haven. This is because in 2018, Mr Liu’s knowledge and expertise in blockchain technology was much sought after.” ⁶⁵ Cachet’s argument is that while Mr Liu has produced some documents, these documents are not sufficient to bear out this claim. ⁶⁶ Mr Liu responds that the documents that have been produced in OC 10 are in fact sufficient for this purpose. ⁶⁷ I leave aside the question of why Cachet is highlighting this alleged evidential deficiency in Mr Liu’s case when it would have been open to Cachet to simply argue at trial that Mr Liu has not adduced the necessary evidence to prove this claim. The more important point is that Mr Liu has expressly averred in his affidavit that he has “provided all documents in [his] possession or control concerning [his] past jobs and past income leading to the pay cut to join Haven”, ⁶⁸ and it is not plain and obvious that there are more documents in Mr Liu’s possession or control on this topic.

(d) Request 12 overlaps with Request 7(f), and Mr Ho was therefore content to rest his argument on Request 12 on what was said for Request 7. I reject Request 12 for the same reasons as I have for Request 7. I would add that the request is speculative. It relates to Mr Liu’s pleading in further particulars that the agreements by Mr Bryan Liu, Mr Ivan Kim and Mr Daniel Kim on their employment with Haven were “oral agreements over telephone conversations”. ⁶⁹ Cachet disagrees on the ground that it “cannot be believed” that the agreements were purely oral and there are no documents evidencing the agreements or their terms. ⁷⁰ In these circumstances, Cachet’s request for “[d]ocuments evidencing [the] purported oral agreements … and the terms of these agreements” gives the impression that it is simply venturing a request for material the very existence of which would undermine Mr Liu’s pleading that the agreements were oral. This is speculative and, as a dispute to be resolved trial, also premature.

(e) Request 13 concerns Mr Liu’s pleading in further particulars that he became aware of negotiations relating to the Subscription Agreement “[s]ometime in or around June 2018”. ⁷¹ Cachet suggests that Mr Liu has not disclosed documents evidencing how he became aware of, and what he was told of, these negotiations, ⁷² and that “[g]iven his close relationship with [Mr Shi]”, it is “plain and obvious that there must have been internal discussions between [Mr Shi] and [Mr Liu]”. ⁷³ I am unable to accept the argument. I agree with Mr Liu that the fact that he came to be aware of the negotiations sometime in or around June 2018 does not suggest that there would be documents “constituting and/or evidencing the means” by which he came to have such awareness, ⁷⁴ much less make it plain and obvious that there are such documents notwithstanding Mr Liu’s averment on affidavit to the contrary.

(f) Request 14 was withdrawn by Cachet, in light of Mr Liu’s confirmation on affidavit that the document had been given to Mr Shi. ⁷⁵ I therefore say no more on it.

83 It will also be observed that Requests 8, 9 and 10 are requests for internal correspondence, for the same reasons as Requests 2 to 6, and are therefore susceptible to the rule in O 11 r 5(2) against the ordering of production of such documents. Cachet’s response is that the requested documents are “known adverse documents” and would be “relevant and material to, and would assist in providing or disproving”, the pleaded issues. ⁷⁶ Cachet has provided no explanation why all, or in fact any, of the documents in these categories are known adverse documents. It would be incumbent on Cachet to do so particularly given the breadth of these requests (ie, for internal documents “relating to” the particular issues or topics), which would plainly encompass documents that are not known adverse documents. As to Cachet’s argument that the documents are “relevant and material” because they go toward proving or disproving the pleaded issues, I have explained why materiality is a basic prerequisite for specific production and does not in and of itself elevate the request to a “special case”.

84 For the foregoing reasons, I dismiss the application for production of the Schedule 1 and 2 Documents. I will hear parties on costs.