In the matter of Cowan v Equis Special LP - Judgment

¶ 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 1 of 14 IN THE GRAND COURT OF THE CAYMAN ISLANDS FINANCIAL SERVICES DIVISION CAUSE NO: FSD 22 OF 2018 (RPJ) BETWEEN: (1) GEORGE ALLEN COWAN (2) GEORGE ALLEN COWAN, ON BEHALF OF EQUIS SPECIAL L.P. (PREVIOUSLY KNOWN AS EQUIS ASIA FUND SPECIAL L.P.) PLAINTIFFS -AND- (1) EQUIS SPECIAL L.P. (PREVIOUSLY KNOWN AS EQUIS ASIA FUND SPECIAL L.P.), ACTING BY ITS GENERAL PARTNER EQUIS SPECIAL GP (2) EQUIS SPECIAL GP (PREVIOUSLY KNOWN AS EQUIS ASIA FUND SPECIAL GP), IN ITS CAPACITY AS GENERAL PARTNER OF EQUIS SPECIAL L.P. (3) DAVID CHARLES RUSSELL (4) ADAM BERNHARD BALLIN (5) LANCE MICHAEL COMES (6) JOSEPH THOMAS CARMODY (7) RAJPAL SINGH CHAUDHARY (8) TONY GIBSON DEFENDANTS 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 2 of 14 Appearances: Paul McGrath QC, instructed by Walkers (Cayman) LLP, on behalf of the Plaintiffs Ms Annaliese Day QC instructed by Maples and Calder (Cayman) LLP, on behalf of the Third, Fourth, Fifth, Seventh and Eighth Defendants Mr Erik Bodden of Conyers Dill & Pearman on behalf of the Sixth Defendant Before: The Hon. Justice Raj Parker Heard: 18 July 2022 Draft Judgment circulated: 13 September 2022 Judgment delivered: 20 September 2022 HEADNOTE Discovery-directions-use of technology to assist review-methodology-keyword searches- GCR O.24- GCR O.1 Overriding Objective. Judgment Introduction

¶ [1] This is the judgment following the hearing of a summons dated 21 October 2021.

¶ [2] The Plaintiffs (Ps) apply for directions in respect of discovery. It is an application made as a result of the developments in technology assisted document reviews. Traditional manual methods are said by Ps to be inadequate for this case and are also said to involve disproportionate cost. 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 3 of 14

¶ [3] The parties agree that some form of technology-assisted process should be adopted but differ on what that should involve.

¶ [4] The essential issue for determination concerns how technology should be applied to the process of identification of relevant documents in this case.

¶ [5] Both parties have served evidence consisting of numerous affidavits and exhibits in respect of the Application1.

¶ [6] In essence, Ps argue that the technology should be applied across an ‘unmanipulated’ data pool.

¶ [7] D1–D5 and D7–D8 (the “Maples Ds”), say the data pool should be manually reviewed by use of keyword searches prior to the implementation of the technology.

¶ [8] D6 has taken a neutral position with his attorney attending on a ‘watching brief.’ The discovery protocol arising from this decision will be binding on him. Summary of case

¶ [9] The dispute concerns the management and operation of a group of related private equity funds known as “Equis”. The focus of the funds was on renewable energy projects in Asia and the Far East. 1 Ps have served:an affidavit of Michael Adam Kriegal, the Vice President of Consulting Services at TransPerfect Legal Solutions, dated 19 October 2021 (“Kriegal 1”); an affidavit of Charlotte Jane Howell, a senior associate at Walkers, dated 19 October 2021 (“Howell 1”), and a second affidavit of Michael Adam Kriegal, dated 31 March 2022 (“Kriegal 2”). The Maples Defendants have served:an affidavit of Erick Gunawan, an expert at the BRG Consulting (Singapore) Private Limited ("BRG"), dated 25 January 2022 (“Gunawan 1”); and an affidavit of Cherrin Wong, an associate at Maples and Calder (Hong Kong) LLP (“Maples”), dated 25 January 2022 (“Wong 1”). 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 4 of 14

¶ [10] P1, Mr Cowan, is an engineer with significant experience in the Asian energy and power sectors. Mr Cowan was employed by the Equis Group in 2011 in connection with the establishment of the first Equis fund and was a foundation partner (along with D3-D7) of D1.

¶ [11] D1 is an Exempted Limited Partnership which was formed in the Cayman Islands under the Exempted Limited Partnership Act in October 2011. At all material times, D1 acted by and through its General Partner, D2 (“Special GP”). D2 is sued in that capacity. D8, Tony Gibson, was General Counsel to the Equis Group.

¶ [12] In overview, Mr Cowan brings claims in his own capacity and derivatively on behalf of P2 (the General Partner).

¶ [13] In the derivative claims, Mr Cowan alleges that monies to which P2 was entitled by way of investment and/or performance fees in respect of a number of the Equis funds were wrongfully diverted from P2 pursuant to an unlawful means conspiracy to which D3–D8 were party.

¶ [14] This was allegedly done by an assignment from P2 to a parallel vehicle, which Mr Cowan alleges caused loss of approximately US$90 million to P2. The unlawful means alleged involved breaches of a series of contractual and fiduciary duties owed variously by Ds to P2 and to Mr Cowan himself, as well as other wrongdoing.

¶ [15] In his personal claims, Mr Cowan complains of the wrongful diversion by D3–D8 of monies to which he was entitled by reason of his position as foundation partner and investor in the funds. This includes monies which would have been paid to him but for the wrongful diversion of monies from P2 and other distributions to which he was entitled from P2 but which he says D3–D8 wrongfully caused to be withheld from him. Procedural history

¶ [16] The procedural history is somewhat convoluted and it is not necessary to set it out here. There have been multiple iterations of Ps claims which has caused the Maples D’s significant work 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 5 of 14 in relation to discovery and this application has further put discovery on hold2 . The main cause for the delay in the proceedings is that Ps sought to introduce a significant claim (the Japan Solar trust claim) and were ultimately unsuccessful. Ps submissions.

¶ [17] Paul McGrath QC appeared for Ps. He argued Ps proposal logically recognises that the technology-driven approach rests on searching for concepts within documents and not simply keywords. This he says broadens the scope of the search and ensures a more accurate sorting of documents into the categories of relevant/non-relevant. Mr McGrath QC put the point as follows: “… the introduction is really this, my Lord, that at the heart of our proposals is a recognition that the methods of conducting discovery have moved on in recent years, and that the software that’s being developed to enable that discovery to take place has become increasingly more focussed and sophisticated. And that software removes much of the guesswork that’s been previously involved in methodologies that have replaced manual review of all documents, and that includes the keyword search.”

¶ [18] He says it is widely recognised that a keyword search method, by its very nature, is an overly- rigid approach and likely to exclude documents dealing with similar concepts on the basis that they happen to employ different terminology. He argues that excluding such documents would materially impact the accuracy of the training for the predictive model.

¶ [19] Applying the technology to ‘unmanipulated data’, by contrast, enables the predictive model to receive the data it requires for effective training. There is accordingly no need, on Ps’ proposal, for any remedial steps to overcome initial shortcomings from the use of keyword searches.

¶ [20] He goes on to argue that P’s proposal would involve the consistent use of advanced techniques throughout the process, from the identification of a computer-generated mixed “seed set” to use for manual review and to teach the review algorithm, to the automated identification of further 2 Wong 1 §9 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 6 of 14 documents of likely relevance. This, he says, is the most reliable approach, because it ensures that the model can learn effectively, through an iterative process by which documents are selected in an unbiased, unmanipulated fashion, and the algorithm is continuously refined by the results of manual review. It is also ultimately the most efficient and least expensive approach. Practical application

¶ [21] Ps approach is in summary steps : a) Compile an algorithm-generated mixed seed set3 of 3,000 documents drawn from the Total Document Population4, and conduct a manual review of that seed set. b) Use the CAL5 algorithm to review the total document population, on the basis of the output from the manual review of the seed set, assigning a relevancy score to the remaining documents in the Total Document Population. c) In tandem with the CAL review, conduct a manual review of those documents in the total document population which the algorithm has ranked highest for relevance, creating a ‘feedback loop’ by which the CAL algorithm continues its learning by the manual identification of (ir)relevant documents, and thus refines its identification of other likely relevant documents which are prioritised for review. d) When this process reaches a point where the relevant documents have been sufficiently sifted from the irrelevant, carry out an elusion test 6by identifying a sample of 3 A small set of a larger document population which is selected and manually reviewed by human reviewers, for the purposes of getting the automated process going, and ‘teaching’ the computer algorithm what is relevant and what is not. 4 A total of approx. 3.1 million documents to be reviewed for the purposes of discovery. 5 Continuous Active Learning, or “CAL”, is a particular type of Technology Assisted Review (TAR,) whereby a seed set which has been manually reviewed for relevance is then used by an algorithm or model to rank the remaining documents according to perceived relevance 6 An exercise carried out to test the reliability of a discovery process: in other words, how many relevant documents have eluded the search and review process 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 7 of 14 unreviewed documents for manual review. If that reveals abnormalities, the review continues. When it shows no abnormalities, the review process is complete.

¶ [22] Mr McGrath QC argued that there were four essential reasons why Ps’ approach is to be preferred: i) Generation of seed set: the best way to set the model off on the right track is to generate a mixed computer-generated seed set. The use of keywords to generate a seed set is an unreliable approach which would exclude large numbers of relevant documents, making for an inherently defective model from the outset. ii) Document types in the seed set: Ps’ approach would also involve a diverse set of documents. The Maples Ds’ approach of using only parent emails, by contrast, will again artificially skew the model and miss relevant documents. iii) Elusion test: the Maples Ds’ proposal that these problems can be cured retrospectively is both unsatisfactory in principle and wrong in practice. iv) Cost: the Maples Ds’ suggestion that Ps’ approach would be lengthier and more costly is unfounded. On the contrary, the Maples Ds’ approach would produce a false economy at the start of the process which would lead to a longer and more costly process overall. The Maples D’s submissions

¶ [23] Ms Annaliese Day QC appeared for the Maples D’s .She argued that P’s approach was unorthodox, unfocussed and wasteful. She submitted that the P’s proposed methodology would impose an enormous and disproportionate burden on the Maples Ds. It would require them to conduct an extensive trawl through irrelevant documents in the hope that something of relevance might emerge that would not be caught by a more conventional set of searches. It 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 8 of 14 should therefore not be ordered even if there was a legal basis to do so (which she argued there was not).

¶ [24] The Maples Ds suggested approach would first involve the application of keywords to reduce the overall document population from 3.1 million to 90,000. The Maples Ds then propose to draw a seed set from the keyword-filtered documents, and use that set to teach the algorithm. Practical application

¶ [25] Maples D’s approach is in summary steps : a) Apply keyword search terms to the Total Document Population, to produce the Keyword Document Population, said to comprise approx. 90,000 documents.7 b) Compile a seed set of 3,000 parent emails and conduct a manual review of that seed set. c) Use the CAL algorithm to review the Keyword Document Population, based on the output from the manual review of the seed set. d) In tandem with the CAL review, conduct a manual review of the Keyword Document Population. e) Conduct an elusion test by way of sample checks on the Total Document Population. Analysis

¶ [26] The question arising for determination at this hearing is essentially one of principle. 7 Since the threshold dispute of principle has yet to be determined, Ps have not commented specifically on the proposed keywords. For the purposes of this hearing Ps are content to proceed on the basis of a figure of 90,000. 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 9 of 14 Legal principles

¶ [27] The parties’ obligation to give discovery stems from GCR O.24. A party is required to give discovery of documents which are or have been in its possession, custody or power relating to matters in question in the action (O.24 r.1(1)).

¶ [28] This obligation is a very broad one. In particular, and in contrast to the approach for example in England and Wales, it extends to documents which not only (a) damage the disclosing party’s case or (b) advance the case of the receiving party, but also extends to (c) ‘train of inquiry’ documents, i.e. documents which may lead to a train of inquiry which has one of consequences (a) or (b). See for example Renova Resources Private Equity Limited v Gilbertson [2011] (2) CILR 148 (Foster J) at [56]–[65], describing the test for relevance as “extremely broad”.

¶ [29] There are no explicit rules or practice directions (as yet) addressing the conduct of e-discovery, or the use of keyword searches in providing general discovery.

¶ [30] However, Order 1 of the GCR provides that the overriding objective of the rules is to enable the Court to deal with “every cause or matter in a just, expeditious and economical way” (the “Overriding Objective”).

¶ [31] The Court has held that expedition and economy are explicitly given a higher priority in its Rules than under the English Civil Procedure Rules, which opens by formulating the purpose of the overriding objective as being to “enable the court to deal with cases justly”.8

¶ [32] In Order 1.2 of the GCR and in Section A4, paragraph 1.2 of the Financial Services Division Guide of the Grand Court of the Cayman Islands (the “FSD Guide”), the definition of dealing with a cause or matter "justly" incorporates the principles of proportionality and economy, and includes, inter alia, “saving expense” and “dealing with the cause or matter in ways which are proportionate- (i) to the amount of money involved; (ii) to the importance of the case; and (iii) to the complexity of the issues”. 8 Re Nord Anglia Education Inc. (Unreported, Grand Court, 19 March 2018), at §8 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 10 of 14

¶ [33] These rules also expressly provide that the Court must further the Overriding Objective by “considering whether the likely benefits or taking a particular step will justify the cost of taking it”, “making appropriate use of technology” and “giving directions to ensure that the trial proceeds quickly and efficiently”.9 Decision

¶ [34] I have formed the clear view that the Maples Ds approach is consistent with the Overriding Objective. They have proposed a sensible and proportionate approach to the discovery process that is appropriate for commercial cases of this size and nature. The fact that the case involves allegations of conspiracy and dishonest assistance does not in my experience take it out of the norm for cases in the Financial Services Division.

¶ [35] The Maples Ds have undertaken substantial work to date in relation to discovery, with the assistance of experienced attorneys and a specialist E-discovery provider. An enormous volume of documents has been collected for review. This has produced a review methodology that appears to the Court on its face to be both time and cost efficient and which ensures that potentially relevant documents are identified with accuracy and are not inadvertently omitted from review.

¶ [36] It would not be in the interests of the Overriding Objective to change tack now when the Maples Ds have spent more than three years working with its E-discovery provider Berkeley Research Group (“BRG”) in preparing their Discovery Protocol. The Discovery Protocol shows the extent of the work performed by the Maples Ds to discharge their duties in connection with discovery so far, and seems on its face to the Court to be reasonable and proportionate.

¶ [37] There is no suggestion of any substantive procedural or discovery default to date. 9 GCR, Preamble, paragraph 4.2; FSD Guide, section A4 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 11 of 14

¶ [38] The Maples Ds should be afforded the usual opportunity to determine the appropriate parameters for discovery searches in accordance with their obligations. They are represented by experienced attorneys, and have engaged a specialist E-discovery provider.

¶ [39] It is not necessary or appropriate, in the Court's view, for the methodology proposed by P’s to be imposed on the Maples Ds (which they resist) absent good reason. This would likely cause further delay and cost.

¶ [40] In keeping with the approach taken by the Court recently the Maples Ds are to be trusted to conduct discovery properly, at least initially10. The Court recently declined to grant the directions sought by dissenting shareholders 11 that the company had to engage with them on the search terms for its discovery, noting in its judgment that: "… each party is to be trusted, at least initially, to conduct its discovery properly, and to the extent another party has concerns about the adequacy of the disclosure process, it can challenge the exercise once it has reviewed the documents produced by the opposing party". The expert evidence

¶ [41] Having reviewed the expert evidence of Mr Kriegal, a consultant employed by another e- discovery firm, who has not been involved in any way in the discovery process undertaken by the Maples Ds, I have formed the view that he has not provided sufficient reasons to displace the trust to be afforded to the Maples Ds.

¶ [42] Mr Gunawan of BRG, who has 18 years of relevant experience disagrees, with Mr Kriegel's views and is intimately familiar with the case having been engaged since 18 May 2018. In his opinion the large volume of documents collected from multiple custodians and sources ranging over a long period of time necessitates the use of keyword searches so that the review can be carried out efficiently and documents responsive to all the various document categories can be 10 Re Sina Corporation (Unreported, Grand Court, 25 January 2022), at §86 11 in a fair value appraisal action 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 12 of 14 located with better precision and speed. He says that there is added value in the matter knowledge of the review team and if keyword searches were not conducted and the seed set were to be randomly selected from the total document population it is his opinion that the CAL review would take longer and would be more imprecise12.

¶ [43] Mr Kriegel in his second affidavit differs from Mr Gunawan on three main issues relating to whether: the Maples Ds should apply keyword search terms to the total document population to derive the keyword document population before conducting a CAL review; the CAL model should be trained using a random seed set obtained from the total document population (as proposed by Ps) or trained with a selected seed set drawn from the keyword document population (as proposed by the Maples Ds); and whether the seed set should be comprised only of parent emails (ie emails without their attachments) which are selected from the key word document population or whether a broader selection of documents including attachments should be used to train the CAL model. Approach

¶ [44] The detailed resolution of these differences is not necessary on this application. It would only be necessary if the Court was persuaded on the available material that the Maples D’s proposed methodology was on the face of it inherently flawed. The Court is not so persuaded.

¶ [45] Discovery has not yet been provided and there is no good reason, based on the methodology set out below, to believe that the Maples Ds discovery will be defective. There is, as a consequence, no good reason at this stage for the Court to delve into and substantively consider the parties' respective approaches to the detailed methodologies they each put forward. On the face of it I can see no sound basis for the suggestion by the Ps that no keywords at all should be used and that the Maples Ds should be ordered to re-perform discovery searches on an entirely new basis. 12 Gunawan 1 §§67-71 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 13 of 14 The Maples Ds approach

¶ [46] In this regard I note and accept that the Maples Ds continuing approach involves : i) the use of keyword searches which they say enable relevant documents to be identified with precision and speed during a 'continuous active learning' ("CAL") review. ii) the existing subject matter knowledge of the reviewer team via the choice of keywords used to identify the population of documents that are likely to be relevant to the issues in dispute; iii) a large volume of documents already collected from numerous custodians and multiple data sources across a wide time period; iv) a large number of categories of documents already created , such that a focused seed set that (i) is selected from the keyword document population and (ii) contains a sampling of documents across all the categories of documents, which they say would be a better set of documents for training the CAL model than one selected at random from the total document population of approximately 3.1 million documents; and v) the history of consultation with the Maples Ds e-Discovery provider, which is familiar with the nature of the data set and the requirements of the document review, having been involved in all aspects of the discovery exercise since the inception of these proceedings.

¶ [47] The Court is not persuaded on the available material that this approach is inherently flawed. The Court accepts Ms Day QC’s submission that the use of keywords is a common and standard feature in large scale e-Discovery exercises13 and that has also been endorsed and accepted by the courts in many common law jurisdictions, and a number of practice directions, protocols and checklists have been developed in these jurisdictions to address the issue of keyword 13 Irish Bank Resolution Corporation Ltd and ors v Quinn and ors [2015] 1 IR 603, at §56 Goodale and ors v The Ministry of Justice and ors [2010] EWHC B40 (QB), at §§10-13 220920 In the matter of Cowan v Equis Special L.P – Judgment - FSD 22 of 2018 (RPJ) Page 14 of 14 searching in e-discovery. There is no reported case or other authority shown to the Court to suggest that keyword searching is deficient in an e-discovery exercise.

¶ [48] As the Court has indicated, if the Ps have any issues with the nature and scope of the discovery provided (or if they have good reason to make a case that material has been omitted), they have a remedy to make an application to the Court after discovery is given, if the matter cannot be resolved through sensible cooperation between the attorneys and their experts.

¶ [49] The Ps application is dismissed. Costs should follow the event and be taxed on the standard basis.

¶ [50] If costs cannot be agreed, the parties are invited to submit short written submissions within 14 days of this judgment. ______________________________ THE HON. JUSTICE PARKER JUDGE OF THE GRAND COURT