Astra Asset Management UK Ltd & Anor v MUSST Investments LLB & Ors

This is my judgment following a hearing in these claims held on 20 and 21 April 2020. The hearing took place during the Covid-2019 lockdown period and was conducted remotely by Skype. Due to the presence of a journalist, the hearing is treated as having taken place in public.

These two claims have been the subject of several previous hearings before me, and I have delivered judgments on 15 January, 24 February and 2 March 2020. I adopt the summary of the claims and the shorthand for the names of the parties used in those judgments to avoid the need for repetition.

Despite technical problems that disrupted the hearing periodically, it was possible to conduct the hearing in an efficient manner. That in large part is a tribute to the assistance provided to the court by Mr Knox QC and Mr Onions QC who respectively appeared for Musst and Astra in the Contract Claim and by Mr David Glen, who appeared for Astra in the Defamation Claim.

This judgment deals with two applications made by Astra. The first is dated 9 December

Under the December application Astra seeks an order for Musst to make a witness statement to deal with the six points that are set out in para.15 of Mr Moore’s third statement. They concern disclosure given by Musst at a time when Bird & Bird was acting for it. There was a change of solicitors on 15 November 2019 and Collyer Bristow then started acting. It is not a matter of great consequence, but I remark that Collyer Bristow have had to deal with issues concerning disclosure when they were not the solicitors who had conducted the exercise.

The December application was preceded by a letter from Payne Hicks Beach, who act for Astra, dated 7 November 2019 and then on 22 November 2019 the six points were set out. Payne Hicks Beach then repeated their request for those points to be answered and, having chased on two occasions, on 5 December 2019 they demanded a reply by close of business the following day on the basis that a 14 day period from the date of their request on 22 November 2019 was sufficient. The application was then issued.

The date of issue of the application was no doubt related to the fact that CMC hearing had been set for 13 December 2019. In the event, that hearing was taken up with other issues, as were the hearings on 24 January 2020 and 2 March 2020. During the intervening period,

The April application is in one respect unusual. It seeks permission to vary the December application so as to seek relief under paras.17 and 18 of Practice Direction51U. However, it is, in reality, a new application that seeks to deal with new points and, as far as I can see, the sole point in seeking to vary the December application relates to preserving Astra’s position as to costs.

I have remarked in previous judgments that these claims have been pursued on a ‘no holds barred’ basis. They are an example of parties to business litigation – and the defamation claim is part and parcel of a business dispute between the corporate entities and their principals – pursuing their respective claims and defending those brought against them with great vigour and at great cost. Up to a point it is open to parties to conduct litigation in that way, but it is not trite in the context of this claim to remind the parties that the court is required to manage the litigation in accordance with the overriding objective and that, as a consequence, the court must have regard to the need to manage litigation justly and at proportionate cost. I have in mind, in particular, the elements of the overriding objective without reading them into this judgment that are set out at (b) and (c) and (e) in subparagraph (2) of that rule. One of the concerns that arise here is that there is an apparent disparity of spending power between the parties and in such circumstances the court must be astute to recognise whether one party is seeking to use procedural steps in a manner that is oppressive.

Mr Knox, who appeared for Musst, submitted that the parties should seek to resolve issues about disclosure in correspondence rather than issue applications, and it is his submission that both applications pursued by Astra were premature.

Given that the applications are both made under the auspices of Practice Direction51U and indeed disclosure took place under the Pilot, I need to make brief reference to the relevant provisions.

The Pilot requires the parties to take a different approach to disclosure; indeed it could not be clearer from the Pilot itself where in para.1 it says in terms that the Pilot is “a new scheme for disclosure”. The new scheme was explained by the Chancellor in UTB LLC v Sheffield United [2019] EWHC 914 and more recently in McParland & Partners Ltd v Whitehead [2020] EWHC 298.

I would briefly highlight the following provisions. First, para.2.3 sets out one of the principles which applies, which is that:

The need for co-operation between the parties is a theme that runs through the Pilot, and this can be seen from the duties that are placed on legal representatives at para.3.2 (2) and (3), in particular the duty under para. 3.2(3):

The final words in that duty have relevance to one of the issues that I have to deal with.

Extended disclosure is governed by para.6.4. It provides that:

Para 6.4 then sets out seven principles that apply. I need only for the purposes of this judgment mention those that are listed at (4) to (7). Under sub-para. (6) it is of note that the financial position of each party is a factor for the court to take into account.

Redaction is an issue which arises on the applications and it is dealt with in para.16 of the Practice Direction. (It is not a subject that is dealt with at all in CPR 31). Para.16 makes it clear is that redaction is permissible. However, other than where privilege is claimed, the data must be both irrelevant to any issue in a proceedings and confidential.

Inevitably, redaction is an area which can be contentious. The redacting party removes data from the sight of the other party and that naturally excites attention. A judgment has to be applied by the disclosing party when applying the two criteria to which I have referred with the duty to act honestly in mind (see para’s 3.1(5) and 3.2(4).

Paragraph 16.2 requires an explanation of the basis upon which redaction has been made, but in a case of this type, where there are many documents and, quite properly a number of redactions, it is not realistic to expect that each redaction should be given a separate explanation if that would merely be repetitive or would risk identifying the data that is claimed to be irrelevant and confidential. On the other hand, a highly generalised formula will not suffice unless it provides an accurate and complete explanation why data has been redacted. There is an entitlement to challenge redaction, but plainly that can only be made where there is a clear basis for seeking to undermine the judgment that has been applied by the redacting party.

In the course of the hearing, Mr Onions, who appeared for Astra, invited the court to consider the original documents in an unredacted form instead of making an order that unredacted copies be supplied. That is certainly an option that is open to the court, and for my part I do not think it is essential that a request for the court to view the complete documents is an essential part of the application. I would observe, however, that requesting the court to undertake a review of a volume of documents places a considerable burden on the court. Furthermore, there will be circumstances (and this may occur frequently) where the court will lack the necessary understanding of the full context, in which the issues arise or, indeed, the reasons why some data my be confidential, such that the court may be in a less good position than the redacting party to undertake a review of this type.

Astra relies upon para.17 or, in the alternative para.18 of the Pilot. The difference between the two is, I suggest, entirely clear. Para.16 it concerned with a failure adequately to comply with the court’s order and the purpose of para.18 is to enable the court to vary an order for extended disclosure. The test under para.17 is intended to be less onerous than under para.18. Under para.17 the court need only consider that to make an order “is reasonable and proportionate”. Under para.18 the court must be satisfied, in addition, that the order “is necessary for the just disposal of the proceedings”. If the court is to be persuaded to re-visit an order for disclosure that has been made already, there should be a good reason for that to be done. The court makes orders for disclosure on the best available information at the time when the order is made and it should not be a commonplace that the parties invite the court to return to those issues, although there are of course cases where that is necessary.

The Disclosure Review Document (“the DRD”) is a document that is central to the Pilot. As para.2 of the “DRD” makes clear, it provides guidance for the parties, and the parties are encouraged to follow the guidance where it is applicable, unless there are good reasons not

Section 3 of the DRD provides guidance on the process to be adopted after an order for extended disclosure has been made. Mr Onions drew attention to para.2 of the guidance which contains a reminder to the parties and their advisers of the need:

Particularly pertinent to the applications is the part of section 3 at para.6 concerning “agreeing aspects of methodology”. Paragraph 6 states:

How each party intends to use analytics to conduct a proportionate review of the dataset;

How each party intends to use technology assisted review to conduct a proportionate review of the data set (particularly where the review data set is likely to be in excess of 50,000 documents)”.

It is clear from the guidance that a party intending to use technology assisted review should raise that issue with the other party.

Mr Onions laid emphasis on the decision of Mr Justice Marcus Smith in Agents’ Mutual Ltd v Gascoigne Halman Ltd [2019] EWHC 3104 and in particular to para.15(3) where Marcus Smith J emphasised the importance of parties agreeing precisely how electronic searches are to proceed. He makes the point that the framing of keyword searches and their application will frequently be an iterative process in the sense that some keywords may throw up a very large number of hits, and if the use of such keywords is to be abandoned, or amended, it is an important matter for the parties to discuss what should be done.

I respectfully agree. I can also see that there will be cases where agreement may prove to be elusive and speedy access to the court may not be a practical proposition. The critical feature, as it seems to me, is that there should be communication between the parties about the process of disclosure and an attempt to agree an approach. That clearly means there must be a genuine attempt to agree, but accepting always that agreement may not be possible.

The springboard for the applications is the order made by Deputy Master Bartlett for disclosure on 18 April 2019. It is right to remark that the parties undertook a great deal of work prior to the hearing of the CCMC and were able in no small measure to agree an approach to disclosure. For that they are to be congratulated.

Paragraphs 10 to 12 of the Deputy Master’s order dealt with the areas where the parties were not able to reach agreement. He gave directions about particular issues concerning custodians, date ranges and search terms; but the extent to which the order dealt with such matters was limited. The order directed that in each case the parties were to comply with their formulation as set out in the DRD, subject to the points the court dealt with in the

The Disclosure Certificate that was produced by Bird & Bird, dated 25 October 2019, runs to 12 pages. Despite Astra’s concerns about some of the steps taken by Bird & Bird in the course of giving disclosure, the Disclosure Certificate is a model of clarity. It sets out stage by stage the methodology used in relation to repositories of documents, there being two stages to the process of reviewing those repositories by the application of search terms within date ranges. The recipient of the Disclosure Certificate could have been in no doubt about precisely what steps had been taken in a complex disclosure exercise. There are points that arise from it, however, and these need to be mentioned briefly.

First, at paras.7(a) and (b) reference is made to the claimant’s email archiving system, UltraVault. Musst identified 198,735 unique documents after de-duplication of the stage one repositories in the core search period and, after applying keyword searches, the number of documents only came down to 175,723. The disclosure statement then goes on to say that as a consequence of the volume of documents, a decision was taken to apply two types of technology assisted review: (1) predictive coding; and (2) clustering analytics.

The steps taken using what I can generically term “TAR” are explained in some detail in paras.9 and 10 running over some three pages in the disclosure statement. It then deals with the stage two repositories and at para.17(a) there is again a reference to the UltraVault archiving system. At para.18(a) to (c) the Disclosure Certificate highlights a significant problem that arose during the course of giving disclosure and undertaking the exercise. The statement explains:

The disclosure statement then goes on to explain how those additional documents that had escaped the initial exercise were dealt with; and once again a careful and full explanation is given of the steps that were taken. In short, one can see from the Disclosure Certificate that a problem was encountered. The problem is made explicit in the disclosure statement and the statement explains how the problem was overcome.

The December application raises six queries. There is an issue between the parties about whether the answers to those queries of themselves give rise to the April application, and that is said to be the basis for the December application but, as I have indicated, I am unable to accept Astra’s case on that point. As it seems to me, the questions that were raised by Astra in correspondence prior to issue of the December application have now all been answered. They were answered in the letter from Collyer Bristow dated 7 January 2020. On 30 January 2020 Musst offered to dispose of the application on the basis of no order for costs, however that was not an offer that was accepted by Astra.

Before I turn to the April application, even though I consider that the December application is effectively spent, I highlight three points in particular that require comment:

The first is a complaint that the approved date range was not used for all custodians. This is accepted. It is submitted on behalf of Musst that it applied the date ranges in a sensible and proportionate way. As an example, the search term “New York” was only applied to a limited period because the term was designed to relate to a one off trip to New York, not to an extended period of time. It seems to me that this was a sensible approach. It is right to state, as indeed is made clear by Marcus Smith J in Agent’s Mutual, that where a party is making adjustments or apparent adjustments to an agreed process, it is important to inform the other party in advance and to try to seek agreement. As I have said, I am not convinced that in a complex and dynamic process it will be practical to agree every minor adjustment; but it is important that the other party is informed so that they are aware and some effort is made to agree.

The second point concerns the discovery that a large number of emails from Mr Siddiqi’s inbox had not been downloaded for the application of keyword searches. Clearly, this was a failure that should not have occurred, but there is no evidence that Astra has been able to place before the court that this failure of process in the disclosure exercise has led to a failure to disclose material documents. The error was made clear on the face of the Disclosure Certificate and, importantly, a clear explanation was given about what steps were subsequently taken. It seems to me the point goes no further.

The third point concerns the use of technology assisted review. It is important to note that in para.3.2 of the Practice Direction, there is a link between liaison and co-operation and the objective of cost efficient and cost-effective disclosure “including through the use of technology”. This makes the obvious point that where a party is intending to use technology assisted review, the intention should be notified to the other party. It was not a point that was raised at the CCMC. That itself is not a criticism because the CCMC took place at a relatively early stage; but it was plainly right that Astra’s advisers should have been informed of a proposal to use technology assisted review. In all probability, had they been so informed there could have been no proper basis for objection once the estimated cost of using technology assisted review had been provided. There was a failure of process and it is right to highlight it.

Subject to those comments, it seems to me that the December application serves no further purpose and I accept Mr Knox’s submission that its issue was premature. It was plainly issued with an eye to the hearing listed on 13 December 2019. It could, in my judgment, have been avoided had Astra’s advisers been just slightly more patient in waiting for a response to their requests.

The April application gives rise to nine separate discrete points. It is made by reference to a draft order and to matters raised in a witness statement. Paragraph 2(a) invites the court to

Paragraph 2(b) makes somewhat diffuse requests for information that are dealt with in paras.64 to 73 and para.95 of Mr Moore’s sixth statement. It is a point to which I will return, because the use of UltraVault by Musst is a point which has been developed in light of more recent correspondence passing between the parties.

Paragraph 3(a) seeks to extend the date range for certain keywords for the period from 1 June 2016 to 30 September 2017. Paragraph 3(b) is no longer pursued. Paragraph 3(c) relates to Mr Shretan Dholakia. Paragraph 3(d) relates to Mr Amar Saksena. Paragraph 3(e) is no longer pursued. Paragraph 3(f) concerns further searches for documents relating to Musst’s financial position and expectations, and para.5 concerns challenges to redactions.

Before I deal with each of the live points – and I will do so briefly – I would make the following observations. Extended disclosure involves a party undertaking a process. In some cases, such as this one, the process is complex. It seems to me, however, that Astra’s advisers have either lost sight of or chosen to ignore the essence of disclosure. As the definition of issues for disclosure makes clear, disclosure is only to be sought and given in relation to key issues in dispute that will need to be determined by the court with some reference to contemporaneous documents in order for there to be a fair resolution of the case. The process is not an end in itself, unless of course it can be shown that failures in process are likely to lead to a failure to produce material documents. And extended disclosure is always limited to by what is reasonable and proportionate having regard to the overriding objective and the factors set out in para. 6.4 of the Practice Direction.

I am bound to observe, after having considered carefully Mr Moore’s two relevant statements, that he and Astra have lost sight of the purpose of the exercise; or otherwise, and unattractively, disclosure is being used for purely tactical reasons to deter Musst from pursuing its claim. I make no finding to that effect, but I am bound to remark that over the hearings I have conducted so far, an impression is forming that possibly such a tactical approach is being deployed. If it is, it is to be deprecated.

Turning then to the decisions I must make on the April application, I have said there is no requirement to refer to technology assisted review and para.2(b) is a point I will return to.

Paragraph 3(a): I can see no likely benefit in making an order extending the period of searches for “Saleem Siddiqi” and “SS”. It is inevitable that these terms during an extended period will throw up many, many thousands of hits. They are obviously not an effective filter, and to make an order would merely result in considerable further expenditure without any expectation of material benefit.

Paragraph 3(b) as I have said is not pursued.

Paragraph 3(c); Mr Dholakia: he was the subject of discussion in the course of the CCMC before Deputy Master Bartlett. Astra’s case at the time was that Mr Dholakia had played a significant role and therefore there should be a date range that reflected the period of his involvement. He was a shareholder only during a limited period and the period of search was limited to when he was a shareholder, with Astra being given liberty to apply. That did not mean, however, that if an application was made it would necessarily result in an order. Disclosure indicates that Mr Dholakia had an involvement earlier than April 2014, but it appears to me that based on the evidence his role was really quite limited. There is nothing

Paragraph 3(d); Mr Saksena. He is Musst’s director, Musst being a BVI company, but it is plain from the evidence that he is merely a nominee director. It has always been the case that the parties who operate Musst are Mr Saddiqi and his wife, Miss Galligan; they are the principal parties. There is no reason to suppose that Mr Saksena would have communicated with any third parties and, to the extent that he communicated with Mr Siddiqi and Miss Galligan, plainly those communications would already be captured to the extent that they are relevant. In short, I can see there is no proper basis upon which to make an order.

Paragraph 3(e) is not pursued.

So far as redactions are concerned I will deal with them shortly when I deal with that subject.

As to the existence of additional documents, I was not persuaded by Mr Onions’ submissions that documents of this type could be directly related to any of the issues for disclosure as they had been defined, and it would therefore not be appropriate to make any further order. However, even if I have taken too narrow a view on the issues, this element of the application founders on proportionality grounds.

Turning then to redactions, I was taken in a not insubstantial trawl through a series of emails that had been redacted. The basis for redaction is explained in the disclosure statement. It is set out on p.10 of that document. The redactions were undertaken by a solicitor at

Mr Onions submitted that there are indications that their approach may have been overzealous in the sense that they have redacted passages in an inappropriate or unnecessary manner. In my judgment the reverse is true. In the emails, redactions are limited. Where there are redactions in emails, I can see no basis for thinking that in light of the context of the redaction, there may be material which should have been disclosed. As Mr Knox pointed out, emails of the type that were reviewed frequently traverse a range of issues, both business and personal, so that each email is often dealing with a number of unrelated points. Of course the court should be astute to redaction on doubtful grounds, but I do not find that there are prima facie grounds for concern here – and it is of note that the redacted documents I was shown (that is in particular the emails) were on any view part of the narrative documents, model D with narrative documents having been ordered. It follows that the likelihood of such a class of documents including wrongly redacted data that was material and adverse to Musst is very low indeed.

I was invited by Mr Onions, as an alternative, to review the original documents myself without redactions. If I had significant doubts about the propriety of redactions, that might

As to the financial statements that I was taken through, it again appeared to me that the basis of redaction appeared on its face to be entirely proper. I can see a logical approach was adopted and I can see no basis for challenge.

The last point there concerns what I can generally term “UltraVault”. Here there is a tangled story in which a number of errors have surfaced. It is clear that the Disclosure Certificate, despite what I have said about its thoroughness, will need to be corrected, and Musst has already offered to do so. The issue for the court is whether an order should be made for the disclosure statement to be corrected and whether the court should require Musst to make a witness statement in the manner that was proposed in Mr Onions’ skeleton argument at para.37.

The disclosure statement refers to UltraVault in a number of places as being Musst’s archiving storage surface – and indeed the late discovery of documents that were the subject of a separate review was said to be based on documents obtained from the UltraVault archive that related to the address “saleem@musst.com”. On 6 April 2020 Collyer Bristow wrote to Payne Hicks Beach. The letter, of course, post-dates the issue of Astra’s application of 3 April 2020. There are two paragraphs in that letter which are of particular relevance, 5.9 and 5.10 where they say

There was then a further clarification that was provided shortly before the hearing. Collyer Bristow wrote on 17 April 2020 and emphasised that the deletion was by the third party storage provider. Specifically, they say that the third party provider:

There is plainly something of a mystery about UltraVault and, as has already been acknowledged, the Disclosure Certificate must be amended. The point for the court to consider is whether an order should be made for the amendment of the Disclosure

It seems to me the issue that arises concerning UltraVault although in general terms is one that has been in play, has taken on a different character since Astra’s application was issued. It was right that it was considered at the hearing, indeed it would have been highly artificial had that not occurred. The question for the court is what is the next step? I have in mind that the manner in which the disclosure statement was prepared was very full and contained a great deal of explanation for the process that was adopted. There is much, as it appears to me, in the disclosure statement that will need to be considered carefully and corrected. A disclosure statement has to be signed by the principal, not by the legal representatives, and therefore a revised disclosure statement will have on it the imprimatur of one of the principles concerned. In those circumstances, I am not satisfied that it is appropriate at this point to make an order for a witness statement to be served.

I must make it clear, however, to Musst that the court is expecting the disclosure statement to be carefully revised and that the corrected disclosure statement will provide an explanation for the steps taken during the course of the disclosure exercise. Quite what that will mean in the provision of information I cannot discern at this stage. If there is, in light of the amended disclosure statement, real doubt about what has happened, then I will certainly consider requiring the production of a witness statement; but it does not seem to me that at this stage it is a necessary order for the court to make.

Although Musst says it will make a revised disclosure statement, given that it was the subject matter of the hearing, albeit not the subject matter of the application, I consider that it is appropriate to make an order requiring the disclosure statement to be amended; and I will now hear from the parties or on some subsequent occasion what date is proposed for the revised statement to be served and filed.