[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback] [DONATE]
	England and Wales High Court (Technology and Construction Court) Decisions
You are here: BAILII >> Databases >> England and Wales High Court (Technology and Construction Court) Decisions >> The New Lottery Company Ltd & Anor v The Gambling Commission [2025] EWHC 1058 (TCC) (06 May 2025) URL: https://www.bailii.org/ew/cases/EWHC/TCC/2025/1058.html Cite as: [2025] EWHC 1058 (TCC)

[New search] [Printable PDF version] [Help]

IN THE HIGH COURT OF JUSTICE
BUSINESS AND PROPERTY COURTS OF ENGLAND AND WALES
TECHNOLOGY AND CONSTRUCTION COURT (KBD)

		Rolls Building Fetter Lane London, EC4A 1NL
		6th May 2025

This judgment was handed down remotely at 10.30am on Tuesday 6^th May 2025 by circulation to the parties or their representatives by e-mail and by release to the National Archives.

Mrs Justice Jefford:

Background

1. The claimants, The New Lottery Company ("TNLC"), bring proceedings against the defendant, the Gambling Commission ("the GC"), concerned with the procurement known as the Fourth National Lottery Competition ("4NLC"). The first interested party ("Allwyn") was the successful bidder and the other interested parties are associated companies of Allwyn, the third interested party having being acquired after the competition.
2. There are two elements to the claims. The first element involves allegations of breach of the Concession Contracts Regulations 2016 in respect of the evaluation of the 4NLC bids. The second element involves allegations that the GC has, following the competition, unlawfully permitted substantial modifications to the Enabling Agreement and the Licence Agreement for the lottery. Other than as set out below, it is not necessary for the purposes of the present application to go into any further detail of the claims and defences.

Disclosure

3. This application arises out of the GC's disclosure and, on its case, the inadvertent disclosure of over 4000 privileged or partially privileged documents. Although the scope of the dispute has narrowed, the court is still concerned with 128 documents in 20 categories which TNLC wishes to rely on (the "Use Pursued Documents") and which GC resists. How this position was reached is of relevance to the application.
4. Following a Case Management Conference on 10 June 2024, Waksman J ordered disclosure to be given by reference to issues/ categories in accordance with CPR Part 31.5(7)(c). The categories of disclosure were to be agreed or directed. Disclosure was ordered to be given by 22 November 2024.
5. The agreed issues (on which TNLC now relies) included the following:

(i) Issue D15: this category was broadly Phase 2 evaluation material relating to the evaluation of Allwyn's bid. Such material was to include (a) notes of each evaluator's scoring of Allwyn's responses to questions and the individual evaluator's scores; (b) all documents relating to the moderation documents of scores; (c) all internal GC communications and communications between the GC and third parties (including other government bodies or departments, the other bidders and advisers to GC) concerning the evaluation of relevant parts of Allwyn's bid.

(ii) Issue D24: Between 31 August 2020 and March 2022, all documents relating to compliance or non-compliance (including approvals sought and/or granted) by Allwyn with the Media and Communications Protocol including (a) internal GC correspondence (including with advisers) and (b) correspondence with Allwyn and investigations into the publication in the press of confidential information in relation to the process.

(iii) Issue D25: Documents in relation to Rothschild's engagement that discuss a potential conflict and/or Rothschild's instructions from Allwyn and/or its parent entity, including conflicts of interest checks done in relation to Rothschild's engagement and the steps taken by the GC to address any actual or perceived conflict of interest.

(iv) Issues D33 and 34: These are essentially the same issue and the terms of D33 are:

"In respect of Challenged Modifications to the Enabling Agreement and Licence:

a) All versions of the Enabling Agreement and the Licence

b) All internal correspondence (including with advisors) regarding Challenged Modifications to the Enabling Agreement and Licence.

c) Correspondence with Allwyn re. Challenged Modifications

d) Meeting minutes re. Challenged Modifications

e) Any documents shared or entered into between GC and Allwyn in relation to Challenged Modifications (including ….)

f) All Modification Notices published by GC."

(v) Issue D35: All documents, internal GC correspondence and correspondence with third parties relating to the likelihood of legal challenge being brought by any of the bidders or any of the bidders' subcontractors, including the implications that any such legal challenge might have on the timing to the transition to the 4NL [ie the contract for the 4^th National Lottery].

There were, as would be expected, also agreed search terms against these issues.

6. As the GC has submitted, the burden of disclosure in a procurement dispute very much falls on the defendant. I deal below with the evidence as to how the disclosure exercise was carried out. In the event, the GC's disclosure was given in 2 tranches and tranche 1 disclosure was completed on 26 November 2024.
7. On 5 December 2024, Hogan Lovells ("HL") for the GC wrote to Bryan Cave Leighton Paisner ("BCLP") for TNLC stating that within their client's disclosure there was privileged content which had, in error, been produced without redactions. HL identified two groups of documents which were versions of "Moderation Agreed Outcomes and Rationale Sheets" for Allwyn and Camelot. In Appendix 1 to the letter HL identified 35 documents that fell within these groups. They further said that they had also disclosed other versions or iterations of these documents with the correct redactions and these were identified in Appendix 2 to the letter. It is worth observing, and not I believe in issue, that the GC's systems operated in such a way that, whenever any alteration was made to a document, a further version or iteration of the document was saved which has resulted in the disclosure of many versions of the same document rather than there being one document recording multiple changes. Inconsistent redaction of versions of the same document is one of the themes of this application.
8. The letter also said, at paragraph 6: "For the avoidance of doubt, in making its production, our client intended no waiver of privilege whatsoever, and we reserve the right to raise other issues in due course should they be revealed. We trust you will do likewise in your own review of our client's disclosure."
9. On 10 December 2024, HL wrote again to BCLP. HL stated that they had identified further documents that had been, they said, disclosed "without redactions in error". There were three categories of document. Again it was the case that other versions of these documents had been produced properly redacted. One document was said to be wholly privileged. The total number of documents referred to in this letter was about 80.
10. The letter concluded with the following:

6. We will undertake urgent review and redaction of any privileged material in the documents listed in Appendix 1 and reproduce redacted versions as necessary as soon as possible. …

7. We repeat our comments at paragraph 6 of our Letter, that in making its production, our client intended no waiver of privilege whatsoever, and we reserve the right to raise other issues in due course should they be revealed. We trust you will do likewise in your own review."

11. The evidence of Mr Bryant of BCLP is that by this time, BCLP had already embarked on the review of the GC's disclosure and increased the size of the review team to do so. Some of the documents identified in the 5 and 10 December letters had been reviewed without the reviewer considering that anything privileged had been inadvertently disclosed. BCLP therefore sent a holding response stating that they would consider the position but, in the meantime, not show any of these document to clients without advance notice.
12. There was a further letter from HL to BCLP on 18 December 2023 identifying 78 further documents which were said to be wholly or partially privileged. The claim to privilege was said to be on the basis that the documents clearly contained legal advice or information that showed the trend of legal advice. The letter contained the following paragraphs:

"2. Having further investigated the matters referenced in those letters, it has become apparent that there have been certain errors in our client's disclosure process that has led to the inadvertent disclosure of privileged material, even beyond the documents identified to date. That investigation is ongoing. We are urgently reviewing the position and carrying out checks and searches to identify the extent of the documents affected, and will update you as soon as practicable.

…

6. We repeat our comments in our earlier letters, that in making its Productions, our client intended no waiver of privilege whatsoever, and we reserve the right to raise other issues in due course should they be revealed. We trust you will do likewise in your own review."

13. Before this point in the correspondence, and despite the reservation of position, HL's correspondence did not indicate that there was any widespread failure in disclosure and the statements that no waiver of privilege was intended are apt to refer to the documents over which privilege was asserted in these letters rather than making a general statement that there had been no intention to disclose any privileged documents at all. That is of some relevance because, in relation to issue D35 in particular, the claimants say that the nature of the issue meant that it was likely that the GC would intentionally disclose privileged documents. That is disputed by HL. I do not think that what was said in the correspondence in December could be construed as a general statement that no privileged documents had been intentionally disclosed so as to put BCLP on notice that their expectation was misplaced. However, by 18 December, there was a warning that there might be more to come and a warning to BCLP to be alert to the same possibility.
14. By letter dated 20 December 2024, BCLP responded to the three letters from HL. Amongst other things, BCLP confirmed that they had not shown their clients any of the documents referred to in the 5 and 10 December letters. However, because they had not considered the documents identified in the 18 December letter to be privileged, some had been shown to the Named Client Representatives. They said that they would confirm which and not share any further. In this letter and subsequent correspondence BCLP also asked HL to articulate the basis of the claim for privilege. It is unnecessary for me to recite the entirety of the correspondence but I note, because it is one of the matters addressed at the hearing, that one matter on which HL sought to rely was common interest privilege as between the GC and DCMS in relation to "legal advice as to the 4NL competition process and its outcome."
15. An aspect of the correspondence was also HL's request to BCLP to delete the documents HL had identified as being wholly or partially privileged and/or incompletely redacted; to confirm that they had not read and/or would not read the documents; and, in any event, not to share them with any client representative. There is a dispute between HL and BCLP as to the adequacy of BCLP's response to this correspondence but I have not considered it necessary to address this further in the context of this application.
16. By letter dated 17 January 2025, BCLP confirmed that they would delete some of the documents that HL had identified in its letters in December. BCLP identified 14 documents from within this group that they had already reviewed and intended to use in the context of amendments to the pleadings. One of those documents (which has been referred to as the "Client Reviewed Document") had already been shared with a client representative before HL had asserted privilege in the document. BCLP's position was that the document had not been disclosed as a result of an obvious error but as a result of being relevant to a disclosure issue.
17. Although correspondence had continued, it was not until HL's letter dated 17 January 2025 that the apparent scale of the errors in disclosure became clear. In that letter HL stated that, following further review, they had identified an additional 4,079 documents that were wholly or partially privileged and had been produced without redactions or with incomplete redactions.
18. There followed further correspondence from 20 January 2025 which related to what should happen next. In the course of this correspondence, BCLP stated for the first time that there was one document that they considered had been disclosed as a result of an obvious error and that was returned. HL also said that their review of the documents was ongoing.
19. That led to the issue of the claimants' application. Although expressed in rather more complex terms, reflecting proposals which had been made in correspondence as to what should happen next, the essence of the application was to seek the court's decision on the issue of whether the documents which HL said were privileged and had been inadvertently disclosed could be used by the claimants, in particular, for the purposes of preparing amended pleadings. There was concern to resolve this position promptly because of potential limitation issues. Correspondence with the court followed in which Waksman J expressed the view that the parties should be able to reach an agreement which would avoid any limitation issues arising. In the course of this correspondence, HL said that they had carried out spot checks on the disclosure and identified yet further privileged documents that had been in advertently disclosed. This opened up the possibility that yet further documents would be said to fall into this group. The final number was 4321 documents.
20. The court listed a directions hearing on 5 February which, in the event, took the entire day and was wide ranging but did result in directions leading to the present hearing. Since then, the parties have adopted what appears to me to have been a co-operative and productive approach to seeking to narrow the issues. On the one hand, BCLP has identified the contentious documents that the claimants would wish to deploy and HL have indicated the documents the defendant objects to the claimants making use of. The GC has accepted that the claimants can use over a thousand of the documents which the GC maintains were inadvertently disclosed.
21. As a result, by the time of this hearing, the dispute had narrowed to 128 documents which were referred to as the Use Pursued Documents. These were drawn together in an Annexure to the 8th witness statement of Christopher Bryant where they were grouped into 20 categories. This Annexure incorporated the comments of HL and reflected the evidence of Jennifer Dickey of HL in her 8th witness statement. The parties further agreed that in relation to each category, I should reach a decision on a sample document and that decision would then apply to all documents within that group. I express my appreciation of the parties' approach – the determination of this application would otherwise have presented the court with an astonishing, if not completely impracticable, task.

The parties' disclosure exercises

22. There was extensive evidence before me as to the way in which each of the GC and TNLC conducted their disclosure exercises.
23. At the directions hearing in February, one issue that arose was the lack of detail as to how HL on behalf of the GC had undertaken the disclosure exercise. That issue went principally to timing. In short, the GC contended that it needed a lengthy period to complete its review of disclosure. Whilst appreciating HL's desire to take care to ensure that further inadvertent disclosure did not occur, it seemed to me that by this stage they must know why and how errors had occurred and could guard against them. It was also not appropriate to allow errors in the GC's disclosure to take over and drive the progress of these actions.
24. The level of detail which has now been provided as to how HL conducted disclosure seems to be intended to meet this point but no applications are made to further extend time.
25. The GC's position was that more than 3 million documents were collected and searches were run across more than 80 custodians and document repositories and over a 4½ year period.
26. The GC engaged a large review team of 67 first and second level reviewers from HL and Capital Law. Guidance was given to them and a third level of review was carried out by a HL core team. 330,000 documents were manually reviewed at first and/or second level and 53,000 documents were disclosed in the two tranches. Mr Bryant's evidence is that the first tranche disclosure amounted to about 40,000 documents (without placeholders) and that over 11,000 documents were withheld on the grounds of full or partial privilege which TNLC relies on as indicating to BCLP that HL had carried out a thorough and proper privilege review.
27. That evidence gives a flavour of the task that then confronted the claimants in review. The evidence of Ms Dickey was also adduced in the context of explaining the scope of the issues that had arisen with the inadvertent disclosure of privileged documents.
28. BCLP deployed a team of 28 first level reviewers. The first level reviewers included members of BCLP's core team being solicitors who regularly work on this case. Others were solicitors who practise in litigation and similar areas; trainees in the department; and in one case an experienced paralegal with substantial experience of document review. All the trainee reviewers had had mandatory training on privilege when entering the litigation department including legal advice privilege and limited waiver. Before the initial review commenced there was a briefing session with the reviewers to summarise the background to the dispute and explain the disclosure issues. The review used a coding system which gave the reviewers options to tick boxes, for example, "relevance", "hot" documents, "query" where the reviewer was not sure of relevance, overlaps to be reviewed by another team, duplicates and privilege. Documents where there was a query or a privileged query in particular could be escalated to a second level of reviewer. The second level review was carried out by a member of the core team.
29. Several reviewers were allocated to reviewing documents within the same category, the categories being related to the disclosure issues. Issues D33, D34 and D35 were grouped together for this purpose. There were 4 reviewers allocated to these issues. It is worth noting the scope of this disclosure which on Mr Bryant's evidence was 12,381 documents in response to D33, 13,461 in response to D34, and 8,671 in response to D35.
30. The majority of the Use Pursued Documents had been reviewed before 17 January and representative documents from 15 of the 20 groupings still in issue had been reviewed before 17 January 2025.

Legal principles: privilege

31. There are a number of areas of common ground between the parties on the law.
32. The GC relies both on legal advice privilege and litigation privilege. At the risk of stating trite law, these concepts are encapsulated in Hollander on Documentary Evidence, 15^th ed at 13-02 as follows:

"Legal advice privilege is narrower in ambit but can be claimed more widely. It protects communications between client and lawyer which are part of the continuum of the giving and getting of legal advice. It does not require the existence or contemplation of legal proceedings. Litigation privilege only applies where adversarial proceedings are in reasonable contemplation, but it is wider in ambit. It protects communications which come into existence for the dominant purpose of gathering evidence for use in proceedings, and will include communications with third parties if they come into existence for that dominant purpose."

33. For the avoidance of doubt, in this judgment, I also accept the submissions of the GC that:

i. Legal advice privilege is not confined to advice on legal rights and liabilities and includes advice as to what should prudently and sensibly be done in the relevant legal context (R (Jet2.com Ltd.) v Civil Aviation Authority [2020] EWCA Civ 35 at [68] and the cases there cited.)

ii. Privilege protects secondary evidence of privileged communications and therefore extends to documents which record or reveal privileged communication – see for example Jet2.com at [45] and [100].

34. A specific issue arose on this application in relation to limited waiver of privilege. I deal with this shortly because it has not been of great relevance in the decisions that I have reached.
35. The issue arose principally in the context of material shared by the GC with DCMS. As I have noted above, in correspondence, the GC initially relied on common interest privilege, sparking a debate as to its common interest with DCMS. The focus then shifted to limited waiver of privilege. The claimants sought to argue that this shift demonstrated that the GC did not itself know the basis on which it claimed privilege and/or did not properly understand the basis on which it had given or withheld disclosure, such that it was more likely that material shared with DCMS was disclosed deliberately, even if objectively a claim for privilege might otherwise have been asserted. I do not consider that a shift in the way in which the GC expressed its claims for privilege has any relevance. HL did no more than put the same point in different ways.
36. There is no real dispute between the parties that privileged communications can be shared confidentially with a third party without loss of privilege (USP Strategies plc, v London General Holdings Ltd. [2002] EWHC CH 373; Gotha City v Sothebys [1998] 1 WLR 114; and Jet2.com at [45].) A privileged communication may be shared without loss of confidentiality/ privilege and/or a privileged communication may be shared for a specified limited purpose. It is not necessary to state whether a privileged communication is being shared on such a basis and that may be inferred from the circumstances.
37. The relevance of Ms Dickey's evidence as to the relationship with DCMS is that it goes to the inference that the claimants could be expected to draw when presented with privileged information shared with DCMS. As Ms Dickey sets out, the GC is accountable to DCMS and statutory provisions apply to the relationship between the GC and the Secretary of State under the Gambling Act 2005 and the National Lottery Act 1993. Funding for the National Lottery comes to the GC through DCMS and its budgets require DCMS approval. The Commission Framework Document (published on DCMS's website) provides a requirement for updates from the GC to DCMS which encompasses timely reporting on litigation matters and "the protection of legally privileged information transmitted to DCMS to facilitate this".

Inadvertent disclosure of privileged material

38. CPR Part 31.20 applies to the present case and provides:

"Where a party inadvertently allows a privileged document to be inspected, the party who has inspected the document may use it or its contents only with the permission of the court."

39. Although these are proceedings in the Business and Property Courts, PD57AD does not apply because these are procurement proceedings. Paragraph 19 of that Practice Direction headed "Restriction on use of a privileged document which has been inadvertently disclosed" provides:

"19.1 Where a party inadvertently produces a privileged document, the party who has received the document may use it or its contents only with the permission of the court.

19.2 Where a party is told, or has reason to suspect, that a document has been produced to it inadvertently, that party shall not read the document and shall promptly notify the party who produced it to him. If that party confirms that the document was produced inadvertently, the receiving party shall, unless on application the court otherwise orders, either return it or destroy it, as directed by the producing party, without reading it."

I shall return to the argument developed by the GC in relation to the Practice Direction.

40. It is common ground that the principles that apply where the court considers the application of Part 31.20 are those set out in Al Fayed v Commissioner of Police of the Metropolis [2002] EWCA Civ 780 at [16]:

"16. In our judgment the following principles can be derived from those cases:

(i) A party giving inspection of documents must decide before doing so what privileged documents he wishes to allow the other party to see and what he does not.

(ii) Although the privilege is that of the client and not the solicitor, a party clothes his solicitor with ostensible authority (if not implied or express authority) to waive privilege in respect of relevant documents.

(iii) A solicitor considering documents made available by the other party to litigation owes no duty of care to that party and is in general entitled to assume that any privilege which might otherwise have been claimed for such documents has been waived.

(iv) In these circumstances, where a party has given inspection of documents, including privileged documents which he has allowed the other party to inspect by mistake, it will in general be too late for him to claim privilege in order to attempt to correct the mistake by obtaining injunctive relief.

(v) However, the court has jurisdiction to intervene to prevent the use of documents made available for inspection by mistake where justice requires, as for example in the case of inspection procured by fraud.

(vi) In the absence of fraud, all will depend upon the circumstances, but the court may grant an injunction if the documents have been made available for inspection as a result of an obvious mistake.

(vii) A mistake is likely to be held to be obvious and an injunction granted where the documents are received by a solicitor and:

(a) the solicitor appreciates that a mistake had been made before making some use of the document; or

(b) it would be obvious to a reasonable solicitor in his position that a mistake has been made;

and, in either case, there are no other circumstances which would make it unjust or inequitable to grant relief.

(viii) Where a solicitor gives detailed consideration to the question whether the documents have been made available for inspection by mistake and honestly concludes that they have not, that fact will be a relevant (and in many cases an important) pointer to the conclusion that it would not be obvious to the reasonable solicitor that a mistake had been made, but is not conclusive; that decision remains a matter for the court.

(ix) In both cases identified in vii)a) and b) above, there are many circumstances in which it may nevertheless be held to be inequitable or unjust to grant relief, but all will depend on the particular circumstances.

(x) Since the court is exercising an equitable jurisdiction, there are no rigid rules."

41. Although the focus in Al-Fayed was on the grant of injunctive relief to prohibit the use of documents, it is common ground, and was the view of the Court of Appeal, that the same principles apply where the permission of the court to rely on such documents is sought under Part 31.20. In other words, there are no rigid rules but the court is more likely to give permission if it was not obvious that the documents were disclosed as a result of a mistake; the court is more likely to find the mistake obvious if it would have been obvious to a reasonable solicitor; and the reasonable solicitor's conclusion after detailed consideration will be a relevant, and potentially important, factor.
42. These principles were recently considered by Nigel Cooper KC (sitting as a Deputy High Court Judge) in Flowcrete UK Ltd. v Vebro Polymers UK Limited [2023] EWHC 22 (Comm). In that case, PD57AD applied and the judge noted at [26] that he did not understand paragraph 19 of that Practice Direction to have changed the position from that in respect of CPR Part 31.20. That view is also expressed in Hollander on Documentary Evidence at 25-02 and 25-03.
43. Relying on both Al-Fayed and Flowcrete, the claimants submitted (i) that the point in time at which the court will judge whether a reasonable solicitor should have realised that an obvious mistake had been made in disclosure is when the relevant document is first reviewed and (ii) that in considering the "standard" of the reasonable solicitor, including the knowledge they are treated as having, the court must have regard to all the circumstances including the extent of the claimed privilege, the nature of the disclosed documents, the complexity of disclosure, the way in which disclosure was given (including the nature of the disclosing party's review), and the time it had taken the disclosing party to realise that there had been inadvertent disclosure.
44. In my view, neither of the cases cited is authority for the immutable proposition that the time at which the court will consider the position of the reasonable solicitor must be the first review. The submission, in any case, elides two issues. One is whether the mistake should have been obvious to a reasonable solicitor (an objective test); the other is what the apparently reasonable solicitor in fact thought (a subjective question) which the court may regard as an important pointer. There is a real risk in muddling these two matters. In proceedings such as the present, there is a tiered approach to the review of disclosure. The first review may, as a matter of fact, be carried out by someone who would not properly be characterised as the reasonable solicitor and the answer to the subjective question would be of less relevance than the view formed on a subsequent and different level of review. On the objective test, it seems to me unrealistic to confine the test to "first review" which itself begs the question of the nature of the first review. In my judgment, the issue that the court is concerned with is whether it should have been obvious to a reasonable solicitor carrying out a proper disclosure review that the document had been inadvertently disclosed. What is a reasonable solicitor and a proper disclosure review is case specific. The factors identified by the claimants are obviously capable of being relevant but again the extent to which they are relevant is case and document specific.
45. The GC referred the court to the decision in Atlantisrealm v Intelligent Land Investments (Renewable Energy) Ltd. [2017] EWCA Civ 1029. In that case, there had been a two tier review of the disclosed documents. The solicitor who carried out the first review took the view that the disclosing party had waived privilege but nonetheless referred the document to a more senior lawyer who informed the other party. At [48] Jackson LJ placed a gloss on the Al-Fayed principles finding that if an inspecting solicitor did not spot a mistake but referred the document to a more percipient colleague who did, the court may grant relief. That, he said, was then an example of obvious mistake. In coming to that conclusion, Jackson LJ recognised the complexities of disclosure in electronic form and made no criticism of the two tiered approach to review.
46. The GC also submitted that there were a number of factors that might be relevant to the assessment of whether the inadvertent disclosure was obvious. These were not dissimilar to those in the claimants' list but were given different emphasis:

(i) The nature and content of the document. In particular if it was plainly privileged the less likely it was to be disclosed. The metadata might be relevant. Even if the document was redacted, that would not necessarily cause the reasonable recipient to conclude that the unredacted but privileged parts had been disclosed deliberately.

(ii) If there was no good reason to have waived privilege, it was the less likely that the disclosing party had deliberately disclosed the document.

(iii) The extent and complexity of the disclosure exercise might make it the more likely that errors would occur.

(iv) If significant volumes of privileged material were disclosed that might indicate that the system had broken down rather than that there had been deliberate disclosure.

47. I have not set out the authorities that were cited by the GC for each of these propositions. I have no doubt that each of them might be a relevant proposition in a particular case or with regard to a particular document as I have said in respect of the claimants' submissions. But as Mr Hossain KC submitted the authorities turn on their facts. By way of example only, the volume of alleged inadvertent disclosure in this case could not have been known to BCLP until 17 January 2025 at earliest by which time 30,000 documents in tranche 1 disclosure had been reviewed. 20,000 had been reviewed before 18 December 2024. So for the bulk of the disclosure review, BCLP did not know that there had been widespread mistaken disclosure and the proposition advanced by the GC does not assist. Further, as Mr Hossain submitted, it took about 7 weeks from tranche 1 disclosure being given for HL to identify the extent of the inadvertent disclosure, yet they argue that that ought to have been obvious to BCLP.
48. I regard the propositions advanced by the GC as indicative of matters that I may take into account in considering whether it was or should have been obvious that a privileged document was disclosed by mistake but I do not treat these as principles of law.
49. Lastly, in my view, in a case such as this the complexities of electronic disclosure may require a slight further gloss in the sense that if there is something in the nature of the document disclosed which ought to alert the reasonable reviewer to the possibility of mistake, he/she ought to inquire further and/or refer the document to a higher level review. In other words, the test of obviousness should not be confined to what is wholly obvious at first blush. But whether any further inquiry ought to be carried out will be entirely case and document specific.
50. I am conscious that that view may seem to conflict with Leggatt J in Mohammed v Ministry of Defence [2013] EWHC 4478 (QB). At [33] he said that the formulation of the fifth principle in Al-Fayed implied that the court should assume that detailed consideration had been given by the solicitor to the question of mistake. He continued:

"That assumption seems to me, with respect, to be appropriate: it would not generally be equitable to allow a party to benefit from a mistake because his solicitors have not given detailed consideration from which the mistake would have been obvious. Such consideration should clearly take account of background information within the solicitor's knowledge. However, since the test is one of obviousness, it is also clear that where such consideration gives rise to mere suspicion or doubt about the matter the reasonable solicitor is not obliged to make further enquiries of the other party before making use of the documents."

51. That observation is focussed on inquiries of the other party when all the reviewing solicitor has is a mere doubt. It does not seem to me to preclude the proper approach in some circumstances being to consider whether further inquiry within the disclosed documents should be made.
52. Drawing the threads together and in terms of the reasonable solicitor, the following matters arise and may be taken into account:

(i) The reasonable solicitor is (as said in Al-Fayed) entitled to start from the premise that the documents disclosed have been deliberately disclosed.

(ii) The reasonable solicitor is entitled to take into account the character of the firm giving disclosure and the manner in which disclosure has been given. A sophisticated exercise undertaken by a highly experienced firm would not be expected to result in inadvertent disclosure of privileged documents. As the claimants submitted, there was, in this case, every expectation that HL would undertake the disclosure exercise thoroughly and diligently and BCLP would not have anticipated that there would be any deficiencies, let alone to the extent that HL assert.

(iii) The volume of disclosure is a matter that cuts both ways. On the one hand, it might be said that a vast volume of disclosure would make it more likely that mistakes would be made and, on the other hand, that the court might regard it as less likely that any errors should be obvious to the reasonable solicitor.

(iv) The reasonable solicitor is one with a reasonable knowledge of the issues in the case and the issues for disclosure. That would include whether there were any matters on which it might reasonably be thought that the disclosing party would disclose documents over which it might otherwise assert privilege.

Specific issues relevant to the review in this case

53. Against that general background, two specific issues arise in the present case which it is convenient to deal with at this point.
54. Firstly, in some aspects of her submissions, Ms Hannaford KC relied on what would be known and obvious to a solicitor with experience in the field of procurement law. As I shall come to, a specific example was comments or questions on scoring initialled by the GC's internal lawyers and her submission was that those who practise in this field would know that lawyers commonly provide such input (which is privileged as legal advice). In a specialist field such as this, I accept that the reasonable solicitor should be one with a reasonable level of knowledge of practice in the field.
55. The second, very much case specific, issue is one of knowledge of the identity of the members of the GC's in house legal team. In some of the documents in issue comments appear with initials (but not names) or members of the GC's legal team are copied into emails but without job titles or an email address that distinguishes them from any other staff.
56. The claimants say that they could not have been expected to know who the initials referred to or that people not identified as such were lawyers, so that there was nothing in their involvement in documents/ correspondence that might indicate the content was privileged.
57. Mr Bryant also points out that there were over 800 names that occurred as senders of emails in the GC's disclosure so that an investigation of the role of each person who may be referred to would have been unreasonable and unrealistic. No list of GC legal team personnel was provided until after the issue of inadvertent disclosure had arisen.
58. In relation to the identity of the comment makers, the claimants also relied on the fact that documents were provided as images without metadata so that beyond the initials the makers of comments could not be identified by name. Ms Hannaford, in her submissions, disputed that the identities were not apparent in all documents and asserted that some documents had been provided with metadata. She had available to her, but not shared with the court or the claimants, a table which supported that submission.
59. I asked to see the document which was provided to me following the hearing. That led to further correspondence from the parties addressed to the court.
60. The table prepared by the GC addressed each of the disputed groups of Use Pursued Documents. It was not solely relevant to those with comment boxes. To summarise, it identified that some documents were provided as native files which showed names of comment makers and with metadata (including author details and file name). Some were produced as an image with metadata (author and file name) but not the names of comment makers. Some were produced with privilege redactions and as an image only with no metadata.
61. In their letter of 10 March 2025 in response, BCLP explained that, although in some instances, HL's disclosure platform may have shown the full names of comment makers, that was not the case with the platform used by BCLP which only showed the initials. That was accepted by HL who had not sought to suggest otherwise. The letter continued:

"However, no matter which platform was used, in the absence of the Defendant having provided details of their legal personnel, sight of a full name would have been of no assistance without also understanding the job role attached to that name, …"

62. Although I asked to see this document because it had been referred to in submissions, it did not take matters much further for the reason given by BCLP in that letter.
63. Ms Hannaford, however, submitted that it was or ought to have been obvious to BCLP who the legal team were and what their input and involvement was. I take Samina Khan (with initials SK) as an example. Ms Hannaford drew the court's attention to the fact that the role of Samina Khan could be identified from the Scribe templates (which form 3 of the disputed groups) and Ms Hannaford referred specifically to a document relevant to the line 26 group of documents. Ms Khan had attended part of a moderation session and was listed in attendance with "4NL Legal" after her name. Ms Hannaford submitted that:

(i) There were 50 versions of Scribe templates that identified Ms Khan as 4NL Legal.

(ii) There were 200 versions of Scribe templates that identified Doug Cochran as a lawyer.

(iii) There were 170 versions of the templates that identified Anne Ferrario as a lawyer.

(iv) There were over 60 versions of briefing presentations which identified Sophie Newbould as a lawyer.

64. Although these figures seem substantial, they were small numbers of documents within the entire scope of disclosure. The effect of the submission seems to me to be that, on the GC's case, it was incumbent on the claimants to identify from every available document the members of the GC in house legal team, produce and circulate a list to all the reviewers, and direct the reviewers to be alert the possibility of comments by persons with those initials. Particularly where the GC had done nothing to identify their own legal team, that is well beyond what the reasonable solicitor could be expected to do and does not assist in setting the standard by which the obviousness of a mistake should be judged.

The subjective review

65. Returning to the review of disclosed documents, on the subjective question of the claimants' reviewers' consideration of the documents, it is the claimants' position that a proper disclosure review had been carried out and at no point before 17 January 2025 had it been obvious to BCLP (other than in relation to one document) that anything had been inadvertently disclosed. That is relied on as a strong indicator, it is submitted, that inadvertent disclosure would not have been obvious to the reasonable solicitor. That was disputed by the GC and how BCLP had carried out disclosure was the subject of comment and criticism from the GC. As Clarke LJ said in Al-Fayed, the view of the reasonable solicitor is not conclusive and the matter is still one for the court and, I would add, the driver remains the objective test.
66. Ms Oppenheimer KC's submission was that the manner in which disclosure had been carried out did not meet the threshold test of evidence that a solicitor had given detailed consideration to the question of whether documents have been made available by mistake and honestly concluded that they had not.
67. She submitted that the evidence as to the thought processes of the first and second level reviewers was inadequate and HL's comments in the Annexure repeatedly asserted that the evidence in this respect was inadequate despite the efforts of BCLP to provide the view of the reviewers.
68. The difficulty in this case, and one that arises against the background of the authorities relied upon is that, with the exception of the Atlantisrealm case, the authorities are concerned with and/or framed in terms of what might be characterised as a traditional disclosure review carried out by one solicitor or at least a small number of solicitors with the same level of experience and knowledge of the case. That does not reflect the reality of review of extensive disclosure of largely electronic documents involving, for example, multiple versions of the same document and repetitive email chains. The approach that was taken by both TNLC and the GC was to carry out review at different tiers where the level 1 review was not necessarily carried out by the sort of solicitor that the authorities contemplate or one that has the characteristics of the putative reasonable solicitor. However, the claimants and defendant also had a system for escalation to a core team of solicitors. There is nothing wrong with this – it is an appropriate and proportionate approach to the review of this type of disclosure.
69. The criticism that the GC makes, however, is that the claimants' approach does not evidence the review of documents at level 1 (or perhaps level 2 also) being one that was analogous to the detailed consideration referred to in Al-Fayed. It is argued that the directions given to level 1 reviewers did not expressly require them to address each document for privilege - starting presumably with the mental question "might this be privileged?". That is unrealistic. A reasonable solicitor (or level 1 reviewer) starts from the premise that they are reviewing documents that have been properly disclosed. They cannot be expected to start each review with the question "Is this document actually privileged?" and nor is it necessary to give such instructions if the reviewer is familiar with the principles of privilege. That is all the more the case where, as here and as the claimants point out, the GC's disclosure was dealt with by a well-known firm with substantial resources and experience available to it, and a large number of documents were duly withheld for privilege.
70. Ms Oppenheimer also submitted that the directions given to the level 1 reviewers meant that the court should afford little or no weight to their views in any assessment of mistaken disclosure because Mr Bryant's evidence was to the effect that, although level 1 reviewers had an option to code a document for privilege, that was principally to identify where a document might have been "over-redacted" for privilege rather than inadvertently disclosed. I do not accept that reading of Mr Bryant's evidence. My understanding of Mr Bryant's evidence is that most of the documents that were escalated with the privilege code were escalated because they were considered over-redacted but not that that was the principal purpose of the privilege review.
71. In this sort of disclosure review, with differing levels of review, the objective question remains, at any level, whether it would have been obvious to a reasonable solicitor, who should be assumed to be one with a reasonable level of knowledge of the case, that a document had been disclosed inadvertently – that is both that it was privileged and that it had not been disclosed deliberately. When considering the subjective question of what the reviewer actually thought, it would be wrong to give no weight to the assessment of the level 1 reviewer but what weight might depend on the status and experience of that reviewer and, as the GC submitted, what directions had been given to that reviewer. In the present case, I see no reason to regard those directions as inadequate or to diminish the relevance of the level 1 review as a result.
72. However, in my view, the focus on the level 1 review and the apparent concern that the court would proceed on the basis of regarding it as an important pointer to obviousness that the document was not inadvertently disclosed was to a considerable extent misplaced or unfounded. Where there are levels of review, the court may have regard to each level and form a view as to the weight to be given to each level of review. But the review that accords more closely with that contemplated in Al-Fayed as an important pointer is a detailed consideration of the question of whether the document had been disclosed by mistake. It is for this reason that I have identified the importance of not muddling the objective and subjective and am not assisted by references to "first review".
73. In this case it seems to me that the type of review contemplated in Al-Fayed took place when the document was escalated to the core team and it is that review that is the far more relevant pointer. In fact, the majority of the documents that are still in issue were the subject of such a review. That is consistent with the approach of the Court of Appeal in Altantisrealm.
74. Ms Oppenheimer also submitted that, in the exercise of my discretion, I should take account of the conduct of BCLP which the GC criticised.
75. Firstly, relying on the practice envisaged by PD57AD, the GC submitted that suspicion that a document was privileged and might have been disclosed in error was sufficient to cause the reviewing party to go no further. Aside from the fact that the Practice Direction does not apply to these proceedings, Mr Hossain submitted that that general submission could not be right because it would be inconsistent with the principles in Al-Fayed. The principles articulated in that case recognise, firstly, that a party may choose to deliberately disclose a privileged document. If a party has disclosed a privileged document in error, the test that the court applies to determine the use that may be made of that document is whether it was or should have been obvious to the reasonable solicitor that the document had not been disclosed deliberately but, rather, inadvertently. If the solicitor has given careful consideration to that question, the court gives substantial weight to his conclusion. Mr Hossain submitted that, if the GC's submission were right, the solicitor and the court would never reach this point. The merest suspicion would trigger a need to notify the other party and the test that the court would apply to its decision as to whether the document should be returned would either be different – for example, whether the reasonable solicitor ought to have suspected that the document had been inadvertently disclosed. That would effect a change in the law which cannot have been the intention of the Practice Direction.
76. There does seem to me to be a potential tension between the test in Al-Fayed and the terms of the Practice Direction. It is one that I do not need to resolve because the Practice Direction does not apply. Even if I did have regard to it, it would seem to me unlikely that the Practice Direction was intended to effect a significant change in the law. I have already referred to the approach to that issue in Flowcrete. What the Practice Direction seeks to do is encapsulate, in very short form, what the court would expect to be done in the "standard" case where a document, for example one that patently contained legal advice, was disclosed. It cannot mean that where the position is less clear cut, a solicitor can no longer give proper consideration to whether there has been an inadvertent or deliberate disclosure of a privileged document or that the court will give no weight to that consideration.
77. The allied submission made by the GC related to the Client Reviewed Document. This is a letter from Andrew Rhodes, CEO of the GC, to non-lawyers at DCMS. Amongst other things, it recites passages from counsel's advice. I set out below my conclusion that it should have been obvious that this letter was disclosed by mistake contrary to the view taken by BCLP and Mr Bryant.
78. Mr Bryant's evidence is that the document was escalated to the core team for review and immediately discussed at some length. Mr Bryant says that he and his colleagues could not rule out disclosure by mistake but considered intentional disclosure more likely given the extent to which other privileged documents had been withheld. They did not consider that there was anything in the GC's relationship with DCMS to give rise to an inherent common interest and saw nothing to indicate a limited waiver. Their "working hypothesis" was that the document could be shared with the client. They sought leading counsel's advice but without joining him, at that stage, into the confidentiality ring so any advice must have been given without seeing the document. The document was then shown to a client representative about 90 minutes before the first notification from HL of inadvertent disclosure.
79. The emphasis given by the GC to this narrative was explicitly not to allege or imply any bad faith on the part of BCLP. However, it was said to illustrate that the approach to consideration of whether a document had been inadvertently disclosed was fundamentally wrong such that no or little weight should be given to BCLP's reviews (at whatever level) as a pointer to the absence of obvious error.
80. Without going so far as to accept Ms Oppenheimer's submission that equated suspicion with obviousness, it does seem to me that what BCLP did was not the best course. In my view, the error was or ought to have been obvious. But if it was not obvious at first review, there was at least enough to cause BCLP to consider the matter fully. I do not criticise that. The thrust of the GC's submissions was that if it took lengthy discussion to reach a conclusion, the error should have been obvious but, as I have said, that would depart from the position contemplated in Al-Fayed in which the reasonable solicitor has given the matter careful consideration. However, at the end of that consideration, BCLP still had only a "working hypothesis" on which they then consulted counsel without counsel being able to see the document and then, on the basis of that unsatisfactory approach, they disclosed the document to the client. It would have been far better if doubt persisted, before or after consulting counsel, to have informed HL of the possible inadvertent disclosure.
81. Having said that, this was one instance – and indeed the only instance where a document was apparently subject to this lengthy review and shown to a client - and I cannot infer from it that there was a fundamentally wrong approach in BCLP's privilege review.
82. In any case, I would not find in this case that there has been any conduct on the part of BCLP that would lead me to exercise my discretion not to permit the claimants to rely on documents that they might otherwise be entitled to rely on applying the Al-Fayed principles. As I will come to in the context of some specific groups of documents, there was reason for BCLP to form the view that privileged documents had been deliberately disclosed. It follows that a document that might or might not have been disclosed on that basis was one which they were entitled to consider, addressing their minds to whether it had obviously been disclosed in error.

The relevance of Quinn Emanuel's review

83. The Interested Parties did not support the claimants' position on this application. Nonetheless, Mr Hossain placed some reliance on the Interested Parties' disclosure review and the fact that, before HL notified any inadvertent disclosure of privileged documents, Quinn Emanuel ("QE"), in common with BCLP, had not considered that any document had been disclosed as a result of an obvious mistake. This was, at highest, another subjective pointer to the view the reasonable solicitor would have formed.
84. Mr Barrett KC emphasised that that review was principally for the purpose of ensuring that commercially confidential material from Allwyn's bid was properly identified and kept within the Confidentiality Ring. The confidential documents (which were Allwyn's documents) were not ones that would contain privileged material so privilege was not relevant to QE's review.
85. It is unfortunate that that submission led to a further round of post-hearing correspondence between the parties, all of which was copied to the court. By letter to QE dated 11 March 2025, BCLP pointed out that QE had previously said that it had suspended its disclosure review because of lack of certainty as a result of the GC's assertion that large numbers of privileged documents had been inadvertently disclosed. That was after the confidentiality review had been completed and BCLP, therefore, inferred that QE had been carrying out a wider review. The response from QE by letter dated 18 March 2025 refuted that inference and said that BCLP had manufactured an inconsistency. Their point was that, given the alleged errors in Tranche 1 disclosure, QE was concerned about errors in the Tranche 2 disclosure and their previous correspondence was addressed to that and to the suspension of any further review.
86. This was all unnecessary. At its highest Mr Hossain's submission was one of icing on the cake. It was no more than a high level submission that, whatever the focus of the QE review, QE had not noted any documents obviously disclosed in error. But, since QE offered no evidence about their disclosure review exercise beyond what was said about identification of confidential documents, this was a minor point that added nothing to the extensive evidence and submissions of the claimants and defendant. In light of that evidence and those submissions, it would have been remarkable for the court to place any significant or determinative reliance on the outcome of QE's disclosure exercise. In the event, it has played no part in my decisions.

The 20 groups of Use Pursued Documents

87. In relation to the groups in Annexure 1, Mr Hossain submitted that they raised 5 overarching issues which went to the question of whether or not it should have been obvious to BCLP that the document was privileged and had been inadvertently disclosed: (1) that the document was previously redacted; (2) it was unknown from the document that it involved lawyer communication; (3) disclosure appeared deliberate; (4) the content in question was not legal advice; and (5) the documents were shared externally and the relevant content lost any privilege it may have had. He then identified the following categories of documents:

(i) Documents which did not appear obviously privileged to begin with (overarching issues 3 and 4)

(ii) Documents where there is an identifiable lawyer author/ recipient/ commentator but it is not obvious that the content is legal advice (overarching issues 3 and 4)

(iii) Redacted documents which had already been reviewed for privilege and there was no obvious reason to question the redactions (overarching issues 1 and 4)

(iv) Documents where redaction was inconsistent (overarching issues 1 and 3)

(v) Documents on which the content is potentially legal advice or reflects the substance of legal advice but it was considered these were deliberately disclosed in response to issue D35 (overarching issue 3)

(vi) Documents which appear to be legal advice but privilege appears to have been lost by sharing with a third party (overarching issue 5)

88. Rather than address each group in the Annexure in the order they appeared, Mr Hossain then made his submissions by reference to these categories and by applying these overarching principles to types of documents which he submitted fell into those categories. Although that departs from the order in the Annexure, and although Ms Hannaford took a slightly different approach to overarching themes, Mr Hossain's structure was a helpful way to address the issues and is one I adopt.
89. Before embarking on this exercise, I note again that in Annexure 1, BCLP (Mr Bryant) have set out their comments on each group, HL have responded and BCLP have replied. It would be wholly impractical to set out in their entirety these comments/ submissions. I set out some to illustrate the pattern of the arguments but will otherwise seek to summarise them as appropriate.

Documents which did not appear obviously privileged

90. I should start by observing that this description is, in a sense, shorthand. As considered at some length above, the issue is not as such whether the document was obviously privileged but whether it should have been obvious to the reasonable solicitor that a privileged document had been disclosed by mistake. But where the document is not itself obviously privileged it is all the less likely to be the case that the reasonable solicitor should have reached the conclusion that it had been inadvertently disclosed. The flipside of this argument is Ms Hannaford's submission that some documents had "red flags" on them because they were very clearly privileged and even marked as privileged.

Line 15

91. This document category is an email chain from 16 December 2021 which begins with the subject "FW: Branding [Finalising contingency session]". The GC only objects to use being made of the part of the chain which is an email from Samina Khan who is part of the GC's legal team.
92. The email chain starts with a response to a Teams meeting by Jonathan Tuchner stating that he is unable to attend [REDACTED]. An email was then sent by Taj Chana to a number of recipients, including Ms Khan, suggesting that the final session continued in Mr Tuchner's absence. Ms Khan responded (adding "legal advice") to the subject line of the email. She also added "[email protected])" as a recipient. [REDACTED]. The email was then sent on by "4NLCEvaluation" to the Branding Evaluators as an "update".
93. BCLP's position is that the disclosure of this email chain appeared to respond to issue D15 in relation to moderation. Despite the subject line, there is no legal advice in this email and there was, in any case, nothing to alert the actual reviewer or the putative reasonable solicitor to the fact that this document was privileged, let alone disclosed inadvertently. Mr Bryant says that the first reviewer could not recall this specific document; knew that Taj Chana was an evaluator; did not know who Ms Khan was; and did not regard the email as containing legal advice.
94. HL says that, given Ms Khan's status as a member of the legal team, her correspondence is obviously privileged and that it is advice as to the limits of the proposed meeting if it is to go ahead.
95. This item, therefore, raises two of the recurrent themes which I have considered above. The first relates to the identity of the author of the email. As I have said, it is not the case that the GC provided to the claimants any list of members of their internal legal team (whether names or initials) at the time tranche 1 disclosure was given. That did not happen until 9 January 2025. Whilst Ms Dickey is right to say that the provision of such a list is not a normal practice, given the scope of disclosure and, on the GC's own case, the disclosure of documents that were likely to have been reviewed by lawyers, it would have been a sensible course of action to identify the legal team.
96. The GC argues that it could be readily ascertained that Ms Khan was a member of the legal team for the reasons set out above. As I have indicated when considering the issue of the identity of the legal team, and at the risk of repetition, it does not seem to me that the reasonable solicitor should be expected to search out the identity of an in house legal team where the other party has not sought to identify them. That must particularly be the case where it is obvious that the disclosure review is going to be carried out by multiple people and at different levels, and that would have been obvious in this case. There would, of course, be circumstances in which something authored by a member of an in house team was identified as such or was so obviously legal advice that it should lead the receiving party to consider whether it was privileged even though they did not know the names of the in house team. It is a case sensitive question.
97. The second matter is the knowledge and experience of the reasonable solicitor in respect of the particular field of practice. In short, if one knows that a legal issue may arise as to the validity of the proposed meeting if someone is not present, one might appreciate that the advice which Ms Khan has given is legal advice but that would require a level of knowledge about the process and the type of meeting.
98. In this case, although with the benefit of hindsight one might be able to say that the email gives some legal advice on the decision making process, that is not at all obvious from the email itself. The subject line "legal advice" does not mark the email as privileged. The advice as to process within the email could as easily be given by an experienced lay person. Ms Khan talks about [REDACTED] She does not say that she will be answering those queries on behalf of "legal". [REDACTED] which are not obviously related to a legal review of a "draft moderation outcome". The copying in of LegalRequests implies that the email is keeping "legal" informed or making a request for legal advice and not that Ms Khan is part of the legal team. The email that further forwarded Ms Khan's email said nothing to indicate that it was legal advice.
99. I do not consider that the reasonable solicitor ought to have known that this email was privileged and had obviously been disclosed in error and I give permission for its use.

Line 16

100. This "group" is a single document entitled "ITA Outcome board report Draft v184". It is a draft of the Invitation to Apply Outcome Board Report authored by Andrew Wilson and with the Commercial Team as the Document Editor, neither of these being part of the legal team. The claim for privilege arises out of a handful of comments initialled "SK" who is now identified as Ms Khan. HL contend that BCLP knew, or perhaps ought to have known, that "SK" and Ms Khan were part of the legal team but that is not a submission I accept for the reasons already given.
101. Paragraph 1 is an Executive Summary. The summary contained a paragraph which stated that the report was the GC's official record of the competition and amongst other things evidenced its compliance with its obligations in accordance with the competition strategy. Ms Hannaford drew particular attention to comment SK3 which recommended [REDACTED]. The comment SK4 raised a query [REDACTED]. The comment SK5 suggested [REDACTED]/
102. I do not see that there is anything in these comments that ought to have alerted the reasonable solicitor to the obvious mistaken disclosure of a privileged document. On its face the document is being circulated within a commercial team for comment. Even if it were known that there was a member of the legal team with the initials "SK", there is nothing in this document that alerts the reviewer to the fact that these comments are being made by that same person. One of the comments may refer to legal obligations but that is hardly sufficient to alert a reviewer to the fact that it is made by a lawyer and providing legal advice. On the contrary, all the comments are simply comments about what the GC should include in its paper recording what has already happened in the competition. I give permission to use this document.

Line 20

103. This document group contains 2 versions of an email chain which the claimants say relates to disclosure issue D25 (conflicts). In the emails, Ashley Gillard of Rothschild & Co, who acted as advisers to the GC, provides a general wording as to how conflicts would be dealt with. Jason Goodwin of the GC forwarded the email to Penny Williams and Sophie Newbould asking, in summary, whether this was OK. Sophie Newbould's response was that [REDACTED].
104. The only one of these people who is part of the legal team is Sophie Newbould. It is again HL's position that it was known or ought to have been known to BCLP that Ms Newbould was a lawyer and that it is, therefore, evident that her advice is legal advice. It should have been obvious that the disclosure of this privileged material was inadvertent.
105. It does not seem to me that it should have been at all obvious to BCLP and the reasonable solicitor that privileged material had been inadvertently disclosed. For the reasons I have given, I do not accept that BCLP could be expected to know – and in particular that all its reviewers could be expected to know – that persons who had not been identified as part of the legal team were lawyers. In the case of this email exchange, the sender of the email is not obviously seeking legal advice from a lawyer; the response is not identified as legal advice; and the person making the response asks for the issue to be "escalated" to someone who is not a lawyer – again with no indication that that is part of a process of seeking legal advice. The "escalation" is far more consistent with a senior commercial person being asked internally what he wants to do and the request for escalation is because "any approach away from the GC policy will require a higher level of approval". I give permission for the claimants to use this document.

Line 22

106. This is a further document where partial privilege is asserted in respect of comments made by Samina Khan. The document is entitled "4NLC Clarification Meeting: [Applicant 6]." The document contains a number of comments in boxes and other amendments. It was first reviewed in December 2024 before Ms Khan was identified as a lawyer. The document was not escalated on grounds of privilege but comment SK2 was reviewed by Mr Bryant who did not, as a result, even suspect that it was a privileged document. It appears that it is this comment only over which privilege is asserted.
107. The comment appears in the context of passages about compliance with the UK Corporate Governance Code 2018. The passages note that the GC reminded the applicant that compliance with the Code was a Licence requirement; that there were a few areas that appeared non-compliant; and that, although there was some flexibility, the applicant had to provide a rationale for perceived inconsistencies. The next paragraph then said that the GC had stated that it was for the applicant to consider compatibility and what solutions it proposed in the event that strict compliance was not practicable. The comment is that [REDACTED].
108. I am in agreement with Mr Bryant that there is nothing in this which ought to have alerted a reviewer to the fact that this was privileged advice that had been inadvertently disclosed. The context is an internal document being reviewed by a number of people and not one on which legal advice is being sought or obviously given. Passages appear in the document which address how the GC dealt with non-compliance with the Code in the case of this applicant. No privilege is asserted over these passages, not could it be. The comment [REDACTED]. The fact that [REDACTED] adds nothing and is no indicator of legal advice.

Lines 24, 25 and 26

109. As Mr Hossain did, I take these lines together as they are all so-called Scribe templates. The claims to privilege are variously that the documents are wholly or partially privileged.
110. The document group at line 24 is 11 versions of the same document "Scribe Template – Moderation Agreed Consensus Score and Rational Sheet – Applicant 7" dated 13 January 2022. The document is concerned with specific sections of the bid. The document refers to a moderation finalisation session. It lists a moderator, two scribes, evaluators, and others by name, none of whom are lawyers. It records the final score as 9 – "Credible but low confidence in delivery". There are tracked changes in the document and comments in boxes with initials. Most of the comments are from "SB" which accords with one of the people expressly named. There is then a single comment from "AF". Against "Scoring criteria – negative deliverable", AF's comment is [REDACTED].
111. It is apparent from that that AF was not at the relevant meeting and that he/she is not one of the scorers or authors of the document but no more than that. AF is now identified as Anne Ferrario who is part of the GC's legal team. She was not identified as such until 9 January 2025 and, for the reasons I have given, I do not accept that BCLP knew or ought to have known that she was a lawyer. HL say that the document was provided in native form, from which it would be possible to identify that the author of the AF comments was, indeed, Anne Ferrario but that in itself does not identify her as a lawyer and there is nothing in the comment that would alert a reviewer to its being legal advice.
112. This raises again the question of the knowledge or experience of the reasonable solicitor. What is submitted on behalf of the GC in the Annexure is that it is common for there to be a legal review of notes of moderation sessions so that, if there are comments from someone not at the meeting, it is to be expected that they are made by the legal reviewers. Ms Hannaford elaborated on that submission making the point that because the evaluation and moderation documents are the ones that will be pored over by claimants, they are exactly the sort of documents where comments from someone not at the meeting would be from a lawyer and AF's comments would be the sort of comments made.
113. These submissions seem to me to be a bridge too far particularly where there is no obviously legal advice in the comment made. In this instance, the comment simply [REDACTED]. That is as much a drafting comment as it is anything else.
114. I would note further that the premise of the GC's submission is not, in any case, accepted and that Mr Bryant's experience is different. His experience is that it is common for lawyers to be present at moderation meetings and give advice. But his view is that for persons (lawyers or otherwise) who were not present to make drafting comments on notes of moderation meetings is extremely troubling. It is not necessary for me to determine who is right on this topic or what is good or proper practice. It is simply the case that I do not consider that the reasonable solicitor could be expected to jump to the conclusion that any comments from those not at the meeting were from lawyers and amounted to privileged legal advice.
115. The GC also argued that, as the claimants have accepted that similar documents disclosed with comments from HL and Capital Law are privileged, the same reasoning should apply to the documents with comments from the in house legal team. But if the test is whether it was an obvious mistake to disclose this document, then the same reasoning does not follow because it is not obvious that the comments are made by lawyers and give legal advice.
116. The document group at line 25 comprises 3 versions of a Scribe Template – "Applicant 7 – Moderation Agreed Outcome and Rationale Sheet". In this case, the document contains two comments from "DC" who is now identified as Doug Cochran who is part of the GC legal team. The comment is the same comment repeated. It appears against passages in relation to the applicant's compliance with requirements. The comment is to the effect that [REDACTED]. This is similar in nature to Ms Ferrario's comment about the drafting of the document and it is difficult to see how a comment of this nature would alert any reviewer, however great their experience in the procurement field, to the fact that this was the advice of a lawyer which had been inadvertently disclosed.
117. Line 26 is a group of 5 versions of a Scribe template "Moderation and Agreed Outcome and Rationale Sheet". The comments in issue are again those of "DC". The first appears alongside a passage which addresses the applicant's position that there is scope to increase the Scratchcard market, making comparisons with other countries. Mr Cochran comments that, [REDACTED]. He suggests that [REDACTED]. His second comment indicates that [REDACTED]. Again these are very much in the style of drafting comments in particular pointing out where something may be unclear or may benefit from fuller explanation but, unless one accepts the premise that the reviewer ought to know that the comments are made by a lawyer and are capable of being construed as legal advice, there is no obvious error in disclosing the document with these comments unredacted.

Line 12

118. The group at line 12 is 2 versions (apparently duplicates) of a paper entitled "4NL Implementation Review Outcome" prepared for a board meeting on 11 August 2023. The author of the paper is John Tanner of the GC.
119. The paper is marked "Official Sensitive" but not privileged. The header table includes reference to the corporate risk as "Disrupted transition from 3NL to 4NL". The first paragraph of the (sample) document states the purpose of the paper to be to summarise the outputs from the Implementation Review triggered in March 2023 and:

"Explain why it is appropriate to amend the Draft Licence, Enabling Agreement (EA) and associated plans and other documents and the nature of proposed changes."

Mr Hossain submitted that it is a commercial document for commercial people and not subject to privilege.

120. The document, however, contains amendments in red and blue. In the footer there is, in blue, a document version number and the name of Hogan Lovells. There are amendments to the document in blue in the same style as the HL footer. Additionally there are some comments in boxes with initials including "HL". The vast majority of the comments are from others particularly "JT". Not all amendments are accompanied by a box. The GC accepts that the underlying document is not privileged and has disclosed other versions of this document but asserts privilege over this version with its blue tracked changes.
121. BCLP say that the content of the document is concerned with Allwyn's delay and therefore appeared to have been disclosed in relation to issues D33 and D34. The original author is not a lawyer and, although there is reference to HL, the comments do not contain or solicit legal advice.
122. In response, HL claim that the document is wholly privileged. They say, firstly, that there is inadequate evidence of what the reviewer actually thought and whether, in light of the HL markings, the reviewer gave any actual consideration to whether the document was privileged and had been disclosed deliberately or not. They note that this document was not escalated. In this instance, it does seem to me that the fact that it is clear that the document had been reviewed and amended by solicitors was itself sufficient to cause any reviewer to question whether it had been disclosed deliberately or not. In the absence of evidence as to the view actually formed, the court can do no more than infer that consideration was given to that issue because any reviewer would be aware of the potential privilege issues and it was a coding option. But that is not sufficient to be a strong pointer that it was not obvious that the document had been disclosed in error.
123. HL say that although the author is Mr Tanner, it should have been and was obvious from the nature of the amendments and the footer that this was a version on which HL had provided their comments. They argue that the fact that some of the comments did not themselves contain legal advice is irrelevant because the overall purpose of the lawyers' comments was to record their advice. As to the disclosure issues, HL say that the fact that issues D33 and D34 were in play does not demonstrate an intention to waive privilege.
124. BCLP responded that a Senior Associate had spoken to the initial reviewer. Their approach was guided by the identity of the recipient/ author of a document. The reviewer could not recall this particular document. When shown it, the reviewer noted that the footer referred to HL but did not consider that the content indicated it had been disclosed inadvertently. That does not amount to evidence of the view the reviewer actually formed but is at best evidence of what the reviewer might have thought which itself may be coloured by the circumstances that this query was being raised in the context of this application. After HL had notified BCLP that the document had been disclosed by mistake, there was a further review and the same conclusion was reached at a higher level of review.
125. This document is, in my view, significantly different from others in this category. Mr Hossain conceded that the blue comments at least might be privileged. Irrespective of the author of the document, it is clear on its face that it had been provided to solicitors for the purposes of their review and comment and that their comments were included in this version. The natural inference is that the comments were or might contain legal advice. The fact that some of the comments do not contain legal advice does not change that. Some of the comments reflect, at the least, HL's advice as to what the GC's position on legal matters should be – eg. paragraph 11 [REDACTED]; paragraph 52 [REDACTED]. These examples support the GC's argument that the purpose of seeking HL's comments was to seek and record legal advice.
126. I do not accept that the error ought not to have been obvious because the documents related to disclosure issues D33 and D34. The claimants characterise the disclosure against these issues as directly relevant to whether the amendments to the Enabling Agreement and Licence were substantial. That is a fair characterisation and it does not imply that disclosure would extend to legal advice on that issue. There is nothing in the formulation of these disclosure issues that indicates that there was any general intention to waive privilege in legal advice.
127. The further matter relied upon by BCLP to support the view formed - at least on the later review - that the document had been intentionally disclosed is the note (initialled JT) against paragraph 51. That paragraph had been amended to delete reference to a paper from HL and to note a risk of legal challenge and an oral update from HL to be provided to the Board. The comment noted that HL's advice was to be appended as an Annex and that HL had advised it should be in a separate paper to retain legal privilege. Mr Bryant's inference is that the main paper was intentionally disclosed in contrast to any Annex which was not. It is not clear to me whether there was an Annex as Mr Bryant states that the annexes were withheld as not relevant rather than as privileged. Whilst the distinction provides some support for BCLP's conclusion, I cannot see that it follows from the reference to a discrete paper containing legal advice that the document itself either does not contain such advice or, if it does, that it was intentionally disclosed. It is a commercial document and not one authored by solicitors but it remains one into which the solicitors had an input and the natural inference is that they did so to provide legal advice.
128. It follows that I do not give permission to the claimants to use this category of document.

Documents where there is an identifiable Lawyer recipient/ Commentator but not obvious that the content is legal advice

129. Mr Hossain submitted that this categorisation was relevant to lines 12, 14 and 15. I have already addressed lines 12 and 15.

Line 14

130. Line 14 concerns a single document headed "Draft Clarification response to assertions made in HL's email to the Commission of Jan 23^rd 2022". The document was partially redacted. BCLP's evidence is that it was not escalated for a privilege review and there is no evidence from the reviewer (who is on extended leave). After the document was notified as one HL said had been inadvertently disclosed, there was a further review. Mr Bryant says that they were not particularly concerned by the reference to "HL"; that as it had already been redacted they considered that anything privileged had already been addressed; and that the remaining content addressed policy and not legal issues.
131. HL's position is that the document contains passages from its email and a response and that, in short, it is entirely clear from that that the document is covered by legal advice privilege. Further, HL point out that the unredacted passages contain reference to the risk of legal challenge by the applicant and other stakeholders.
132. The partial redaction is undoubtedly an indicator that the document has been reviewed by HL and I would not wish to lose sight of the principle in Al-Fayed that the receiving party is entitled to start with the assumption that a proper review for privilege has been carried out by the disclosing party.
133. The first part of the document is concerned with a query about [REDACTED]. [REDACTED] the very fact that solicitors were questioning this and the tone of the query clearly implies that there were legal implications and that HL's purpose was to elicit information on which to give legal advice. The second part of the document (unredacted) concerns [REDACTED], making it all the clearer that they are seeking to understand the position for the purposes of legal advice. These are very much part of a continuum of legal advice.
134. In my view, not only is this document privileged but it ought to have been obvious that it was disclosed in error. The argument to the contrary, as I have said, is that the partial redaction suggested otherwise, but the nature of the document seems to me to militate in favour of the conclusion that there was an error in redaction rather than a deliberate disclosure. I conclude that this document falls on the side of the line that means that the claimants should not have permission to rely on it.

Redacted documents which had already been reviewed for privilege, and no obvious reason to question it

135. Mr Hossain addressed lines 14, 17 and 19 under this category. I have already addressed line 14 and say no more about it. Line 17 brings into play the claimants' submissions in relation to disclosure issue D35 and is more conveniently dealt with in that category.

Line 19

136. This group of documents comprises 30 versions of the 4NLC Risk Management presentation for an Audit and Risk Committee on 14 September 2023. The document is not itself a legal document. The document includes a slide headed Programme Risk Update which refers to possible [REDACTED]. Mr Hossain submitted that in 4 of the versions of this document, these passages had not been redacted. HL explain that a change in formatting caused an error in redaction. No review identified an obvious error.
137. On the one hand, the content of the document is uninformative. Although relevant to disclosure issues D33 and D34, it offers no more than a high level statement that changes might carry risk and be subject to legal challenge and that is a view that might be expressed by the author of the document, not being a lawyer. On the other hand, the slide, even expressed in that brief way, is conveying legal advice which has been received by the GC. It seems to me that where multiple versions of the same document have been disclosed, that should be sufficient to cause the reviewer to consider other versions. If they did so, that would make it obvious, if it was not already, that the document had been disclosed in error without the relevant redaction.
138. The authorities are of limited assistance on the approach that the court should take where there are multiple reviewers and what knowledge of the disclosure universe as a whole should be attributed to the reasonable solicitor reviewing the documents. The claimants' submissions, which highlight the number of documents being reviewed by multiple reviewers, to an extent address the issue of obvious mistake as if the reviewer operates in isolation. If the reviewer has concluded that there is no error in the disclosure of the particular document, that is a pointer to the fact that any error was not obvious. But where there is at the least a question mark and it is known that there are multiple versions of the same document, I take the view that some check should be made. In this case, that would have revealed an obvious error. I have a discretion to exercise and, in these circumstances, I consider it fair to exercise it so as not to give permission to rely on this document.

Documents where redaction was inconsistent

139. This was Mr Hossain's next category of documents and the relevant lines were 4, 7, 14, 17 and 19. I have already addressed lines 14 and 19 above and it seems to me that lines 4, 7 and 17 are more conveniently addressed in the context of the arguments relating to disclosure issue D35. However, I make some further observations about inconsistent redactions at this point.
140. The claimants make the short point that discrepancies in redactions were not identified by the level 1 reviewers or by the higher level BCLP core team and it is submitted that it cannot be said that a reasonable solicitor would have done so. At a high level, they say that, if HL could not carry out a consistent redaction, they are expecting a higher standard of review from the receiving party than they have been able to carry out themselves.
141. The claimants point out, firstly, to the fact that the documents were disclosed without metadata so that cross-comparison could not easily be carried out. Multiple reviewers might consider the same document focussing on different disclosure issues. In any event, in documents say 100 pages long, a reviewer could not be expected to recall, say, the redactions on 4 pages and note that there was inconsistency. The core team would have a single document referred to them and be told that multiple versions existed so they would not be in a position to carry out that comparison.
142. There is considerable merit in all of those submissions but, in my view, there is no overarching approach to be taken to the inconsistent redactions. As I have said, if a version of a document contained privileged advice and there was the possibility that this disclosure was unintentional, the reasonable course would have been to check at least one other version. If there was consistency that would support the view that the disclosure was deliberate. If there was inconsistency, it would point the other way.

Documents in which the content is potentially legal advice or reflects the substance of legal advice, but it was considered these were deliberately disclosed as answering to D35

143. Into this category, Mr Hossain placed lines 1 to 7 and 17 to 18, although from the list of 20 contested groupings provided to me, it seems that line 18 is no longer in issue.
144. I start by saying that, as a matter of principle, disclosure issues do not supplant the pleaded cases. Their purpose is to encapsulate the issue that arises and by reference to which disclosure can be undertaken. Where there are agreed issues, the parties and the court should not generally then need to go back to the statements of case to identify what the issue is but, if any dispute arises, it is the statements of case and not the issue that should drive the scope of disclosure.
145. This disclosure issue arises out of the claimants' claim that modifications were made to the Enabling Agreement and Licence ("the Challenged Modifications") which collectively or individually were substantial within the meaning of regulation 43 of the Concession Contracts Regulations 2016; that the modifications are not justified under the regulation; that, therefore, there should have been a new concession award procedure undertaken; and that there is a real prospect that that procedure would have had a different outcome (in the claimants' favour).
146. The Amended Particulars of Claim at paragraphs 36 - 38 set out the claimants' case in respect of the Modification Notice as follows:

"36. The Modification Notice asserts that the Challenged Modifications were brought about by circumstances which a diligent contracting authority/ entity could not foresee. Two specific matters are relied upon in respect of the Modifications pleaded above.

Litigation brought challenging the outcome of the 4NLC. The litigation referred to was brought by Camelot, the incumbent licensee and the other unsuccessful tenderer in 4NLC, as well as Camelot's sub-contract provider, IGT.

….

37. It is asserted by the Defendant in the Modification Notice that it was unforeseeable that IGT would start proceedings regarding 4NLC as it was a subcontractor to the incumbent provider, or that IGT would continue proceedings once Camelot withdrew its proceedings on 16 February 2023, or that negotiations with IGT would be so protracted and challenging.

38. It is the Claimants' case that this position is unsustainable. ….."

147. The claimants then set out their case as to why a challenge by a subcontractor could or should have been foreseeable. Further, the claimants dispute the reasons for the modifications which, in summary, they argue lie at the door of the Interested Parties.
148. The GC denies this claim. It points out that it does not rely on litigation by Camelot as unforeseeable. As to proceedings by a subcontractor, the GC maintains that it was unforeseeable that IGT would bring proceedings and maintain them after Camelot withdrew and that negotiations for handover from IGT would be protracted and challenging.
149. The disclosure issue, therefore, to an extent goes beyond the pleaded issue because it includes the likelihood of a legal challenge by any bidder as well as any bidders' subcontractors. It is important that the issue is framed in terms of likelihood rather than mere possibility, which fits with the pleaded case as to foreseeability. The issue is also time limited: 31 August 2020 to 15 March 2022.
150. This issue bulks large in the claimants' submissions because it is their case that the issue is one which makes it more likely that the defendant would deliberately disclose legal advice going to this issue. The defendant emphasises that the agreement of the disclosure issue does not in itself amount to a waiver of privilege.

Line 1

151. This line concerns one document entitled "4NLC Risk and Contingency – Legal Challenge and End of Third Licence" dated 5 November 2021. The document is a presentation marked "Confidential" and "commercially sensitive" but not privileged.
152. The Preface addresses a Programme the goal of which is to achieve full implementation of the 4^th Licence. But it considers contingencies if that is not achievable. In respect of partial implementation, it is noted that [REDACTED].
153. The GC's position is that the document is plainly privileged because it records and reflects legal advice given to the Commission. The GC says that the presentation was for a workshop between the GC and DCMS. Information was shared with DMCS on the basis of a limited waiver of privilege only.
154. Mr Bryant says that the fact that the information was shared with a third party would, at the least, affect a reviewer's consideration of privilege. The reviewer noted that the document was marked confidential but not privileged and did not consider it privileged. It was escalated for review because of relevance. Mr Bryant considered it possible that it had been disclosed by mistake but thought it more plausible that it had been disclosed intentionally because (i) it was part of a pattern of disclosure of hundreds of documents and (ii) it was relevant to issue D35. The reference to the pattern of disclosure accords with Ms Dickey's evidence at paragraph 9.21(b) of her 8^th statement where she says:

"The Commission estimates that it has disclosed several thousand non-privileged or part-privileged documents in the First Tranche that appear to be relevant to issue D35. These unprivileged documents largely relate to the high-level question of whether litigation about the Competition was expected and/or the commercial impact that any such litigation might have on the transition to 4NL, but which do not reveal or contain the substance of any privileged communication. By way of example only, such disclosure involved: internal discussion of programme risks and consequences of such risks, that did not reflect the legal advice, and internal discussion of application clarification questions."

155. The nature of the disclosure issue means that a reviewer would expect to see documents that referred to the risk of legal challenge. Nonetheless, it does not follow that there is likely to have been some wholesale waiver of privilege. The key issue that arises is, as I put it in the course of the hearing, where the dividing line is to be drawn and where the reasonable solicitor would think that dividing line was drawn so as to form a proper view as to whether a privileged document had been disclosed as a result of an obvious error.
156. Mr Hossain submitted that in the present case it remains difficult to know where the GC intended to draw the line. I agree that, even with Ms Dickey's explanation after the fact, it is difficult to discern where that line was to be drawn. Mr Hossain suggested that a possibility, that would be largely consistent with the disclosure given and Ms Dickey's evidence, would be between commercial documents in which commercial people discussed the likelihood of litigation (even if that derived from legal advice) and documents authored by lawyers or perhaps those that directly cited that advice. He drew the court's attention to a document authored by Mr Tanner and entitled "4NLC Risk and Contingency During Competition and Implementation" and dated 14 May 2021. No privilege is claimed in respect of the document. This document included a section on how a legal challenge may be brought, what would happen, the approach the court would take to an application to lift the automatic suspension and the possible impact on the timeline. I observe that some of that might be thought to reflect legal advice but equally could be said to reflect common knowledge and it illustrates the difficulty with the dividing line.
157. With that background, it seems to me that the presentation at line 1 is, in part, concerned with the risk of litigation although there is no specific consideration of risk of legal challenge by subcontractors. It is in the nature of a high level commercial review. It does not disclose the content of any legal advice. I do not consider that it should have been obvious to a reasonable solicitor that this high level presentation had been disclosed in error and I give permission to use this document.

Line 2

158. This document group comprises 6 versions of a flowchart entitled "Chart Competition, Implementation and Challenge Flowchart". The versions range in date from 18 June to 12 July 2021.
159. The sample page of the flowchart to which the court was taken indicates the actions or events that might lead to a risk of legal challenge. There is reference to "legal reviews commissioned" although it is unclear whether those are reviews that have been commissioned or reviews that would be commissioned if the risk of legal challenge crystallised. There is express reference to counsel having been asked to consider [REDACTED] but no further indication of that advice. Similarly the document contains a box headed "Competition, Implementation and Challenge Flowchart" which says: [REDACTED].
160. Subsequent pages include similar references to seeking advice. There is reference to a litigation strategy paper but it is not apparent whether this is a commercial or legal document and whether it contained any legal advice.
161. The flowcharts are again at a high level. They disclose that legal advice has been sought on one particular topic and the fact of seeking advice may be relevant to the disclosure issue. The flowchart does not disclose the content of any legal advice. It discloses topics on which counsel's advice is being or will be sought but it does not disclose the content of any such advice. In many other contexts, I would take the view that the repeated reference to legal advice which has been or may be sought would lead to the conclusion that it ought to have been obvious that the document had been disclosed in error but, in the context I have set out above, I take a different view. The document is plainly relevant to the likelihood of a legal challenge and the references to topics on which advice has been or will be sought are relevant to that likelihood, indicating the type and number of issues that might give rise to a challenge. Since the flowcharts do not then contain any legal advice given, it would have been reasonable to conclude that they had been disclosed deliberately as relevant to the disclosure issues. I give permission to rely on this group of documents.

Line 3

162. This document group includes 2 internal e-mail chains exchanged within the GC with the subject "HMT Contingent Liability Checklist". Given the view I have formed about these e-mail chains I will summarise their content at a high level only. The e-mail exchange includes discussion of legal exposure and estimates for defending a legal challenge and includes an e-mail from Mr Tanner to Nadine Pemberton setting out his proposed response to an e-mail from DCMS. Nadine Pemberton was General Counsel for the GC.
163. HL say that the content is obviously privileged and the fact that the chain started with an email from DMCS is irrelevant. Mr Bryant says that the reviewer did not consider it privileged because it was a chain between John Tanner and DCMS. It was escalated on the basis of relevance and Mr Bryant and the core team concluded it had been intentionally disclosed.
164. The emails, firstly, do not, as such, consider the likelihood of legal challenge but go significantly beyond that and into the realm of costs of proceedings and likely level of damages if challenging party was successful. That is enough to raise a question over whether the exchange was even relevant to issue D35. Ms Hannaford expressed incredulity that, whatever the position with other members of the legal team, BCLP's reviewers did not know, or should not have been expected to know, that Ms Pemberton was General Counsel and she pointed to 80 versions of presentations that expressly Ms Pemberton's role. It does seem to me that as General Counsel Ms Pemberton falls into a different category from other members of the legal team but, even if that were not the case, the nature of the email from Mr Tanner to Ms Pemberton asking, in effect, for her input into a proposed response dealing with costs and quantum should have been enough to cause a reviewer to inquire as to who Ms Pemberton was. Her input was legal advice.
165. I do not give permission to use this group of documents.

Line 4

166. This document group contains multiple drafts of a document entitled "Contingent Liability Checklist". The document addresses similar issues to those in the emails at line 3 but in greater detail. It is framed in terms of questions on contingent liability and extensive answers to those questions.
167. HL say that it is authored by Nadine Pemberton and obviously privileged legal advice. The pdf version does not state the author but HL say that 3 versions of this document were disclosed in native format which showed the author. Mr Bryant can only identify two such versions but says, in any event, that it was not obvious either that she was the author or a lawyer.
168. BCLP say that three versions of the document were disclosed with no redactions and no privilege concerns were raised by the reviewer. Other documents in this group were disclosed with inconsistent redactions and reviewed by different people. One privilege concern was raised. Mr Bryant concluded that whilst a mistake was possible, other reasons for deliberate disclosure were more likely including relevance to issue D35 and that the material appeared to be shared with DCMS.
169. I take a similar view of this document to the emails in line 3. The document goes well beyond the likelihood of a legal challenge and that, in itself, ought to have raised a question mark as to the nature of the document and its disclosure. That should have led to further consideration. I recognise that there were difficulties in tracking different versions of documents because of the manner in which they were disclosed but it is clear from the fact that one person reviewed three versions that there were multiple versions of the documents. A reasonable line of inquiry would have revealed both the author and the inconsistent redactions and it ought to have been obvious that the unredacted versions had been disclosed in error. I do not give permission to rely on this document.

Line 5

170. This document group is 8 versions of a Power Point presentation dated 3 August 2021 and entitled "Legal Risks Summary". It is marked for distribution internally and to DCMS and marked as legally privileged. The Introduction describes the following slides as highlighting the potential legal risks facing the Commission that may give rise to a legal challenge. The text states that each risk had been summarised using "the following sources" which include counsel's advice and 4NLC's legal team.
171. Mr Bryant's evidence is that the reviewers of these documents concentrated on relevance and did not raise any privilege concerns. Mr Bryant similarly thought that there might have been a mistake but it was more likely that the document had been disclosed deliberately.
172. I place no weight on either of those views. Whatever the relevance to the disclosure issues, this is, unlike many others, a document marked as privileged which says on its face that it is derived from legal advice. It ought to have been obvious that it was disclosed in error. The fact that it was intended for sharing with DCMS is irrelevant – the marking and content are entirely indicative of a limited waiver.

Line 6

173. This group consists of two versions of a document entitled "Mitigating Legal Challenge" dated 9 June 2021. There is nothing on the face of the document to identify the author.
174. The document states that the Commission regularly reviews the measures that it has in place to mitigate the risk of successful legal challenge and that it is satisfied, following a recent review, that it has the necessary controls in place.
175. The document then states: [REDACTED]
176. What follows is a general summary of [REDACTED]. It is recorded that these are questions the Commission has asked counsel to advise on.
177. In my view this is similar to the high level consideration of types of challenge and process considered above. Contrary to HL's position that this is clearly privileged legal advice, it is not at all obvious to me to me that any of this is legal advice because it is expressed in general descriptive terms and that sort of discussion also appears elsewhere in documents over which privilege is not asserted. The document identifies issues on which legal advice has been sought but says nothing about advice received. I give permission to the claimants to rely on this document.

Line 7

178. This document is a paper called "Gambling Commission Response to DCMS Comments on 4NLC Contingency Papers". It sets out the GC's response to queries raised by DCMS. The document is marked as commercially sensitive and legally privileged.
179. HL now say that the document was authored by Mr Cochran. BCLP say that it was not initially disclosed with metadata and, in any event, it was not known that Mr Cochran was a lawyer. There is nothing on the face of the document to identify him as the author or a lawyer. However, the marking of the document as privileged was obviously indicative of legal advice and, without descending into the detail set out in the Annexures, versions of the document were disclosed partially redacted which, in the context of the marking of the document as "legally privileged", would imply that the parts that the GC did not intend to disclose had been redacted.
180. Again I recognise the difficulties presented to the reviewers by the inconsistent redactions but, in my judgment, the appropriate way to deal with this document, is to refuse permission to rely on the wholly unredacted versions. The claimants can continue to rely on the redacted versions (with their inconsistencies). It ought not to have been obvious to a reviewer that the document was disclosed by mistake where it was relevant to a disclosure issue and the disclosing party could reasonably be assumed to have redacted such parts as they did not wish to be disclosed.

Line 8

181. This is the Client Reviewed Document: a letter from the Andrew Rhodes, Chief Executive of the GC to Polly Payne and Ruth Hannant at DCMS. It is dated 30 October 2023. There is a suggestion in the Annexure that this is disclosure relevant to D35 but it falls outside the date range. Rather the claimants' contention is now that it is relevant to D33 and D34.
182. Although the letter passes between lay people, not lawyers, it is headed "Litigation Update". It includes the following: [REDACTED].
183. The letter then refers to a copy of the advice included with this letter and the recipients' attention is drawn to the conclusions from paragraph 10 onwards in that advice. The letter then goes on to address the actions of the current claimants and to say that the Commission has been considering its litigation strategy in respect of the claimants (and that an advice note is attached). There is again consideration of the likely costs of defending a claim.
184. Although this may fall within internal correspondence regarding the Challenged Modifications (and thus disclosure issues D33 and D34), it is plainly recording privileged advice. In contrast to D35, there is nothing in those disclosure issues that might indicate that any privileged advice would be deliberately disclosed. As the GC submitted, where there is no good reason for the disclosing party to have waived privilege, it is more likely that the mistake will be obvious.
185. Mr Bryant says, nonetheless that it was not obvious, and ought not to have been obvious, that the document was disclosed in error. He makes two main points. Firstly, he relies on the fact that the annexes to the letter (including counsel's advice) were not disclosed so that it appeared that a deliberate decision had been taken to disclose what appeared in the body of the letter. That does not follow and, if the advice has been withheld, it makes it the more likely that failing to delete the reference in the letter was a mistake, and, indeed, that disclosing a letter that was premised on that advice was a mistake. Secondly, and more generally, he relies on the waiver of privilege because the content of the letter was shared with DCMS. Mr Hossain submitted that this was the only document where the sole issue was whether sharing with DCMS amounted to a waiver of privilege.
186. It seems to me quite clear, and ought to have been obvious, that there was a limited waiver. If Mr Bryant's position were right, privilege in counsel's advice would also have been waived and it is not suggested that it was. It ought to have been obvious that a document disclosing counsel's advice had been disclosed in error and I do not give permission to the claimants to rely on this document.

Line 17

187. This document group is a set of 43 draft presentation slides entitled "SRO & Programme Director Presentation", the SRO being the Senior Responsible Officer.
188. HL accept that much of this document is not privileged but privilege is claimed over slide 12. This slide has three columns. The first is a report on the IGT litigation and the decision of the court on a preliminary issue that IGT did not have standing. It contains no legal advice. The second column addresses [REDACTED]. It describes [REDACTED].
189. I do not see how this document is strictly relevant to issue D35 since it is concerned with the risk of a legal challenge based on changes to the Enabling Agreement and the Licence and not on the likelihood of litigation which the GC says was unforeseeable and a legitimate factor in the making of those changes. That might be an indicator that the document, or at least that slide, had been mistakenly disclosed. There are two reasons to reach the contrary conclusion. Firstly, the middle column is largely generic and could as easily be commercial commentary as legal advice. It is not, as HL say in the Annexure, clearly a recitation of legal advice that has been given to the Commission and is being summarised for senior management. Secondly, as Mr Bryant points out there are multiple versions of this document in which the slide is not redacted and only one in which it is which points to a lack of intention to assert privilege in that slide.
190. Leaving to one side my doubt as to the relevance of this document to the issues in the case, I do not consider that it was or should have been obvious that it had been disclosed by mistake and the claimants have permission to rely on it.

Documents which appear to be legal advice but privilege appears to be lost by sharing with a third party

191. This was Mr Hossain's last category of documents. He submitted that the issue applied to lines 1 to 5, 8 and 21. I have dealt with these, other than line 21, above and say nothing further.

Line 21

192. This is the only document which potentially raises an issue arising from the sharing of material over which privilege is claimed with Rothschild, as advisers to the GC. The dispute in relation to this document is now very limited and, on the claimants' case, the issue is now rather one of whether the material appears to be privileged or not.
193. The document is an email sent to various people within the GC and one person at Rothschild. The e-mail details a meeting held on 22 April 2020 with the heading "Initial appraisal of data disclosed to Camelot L4 bid team by CUKL and the commercial advantage it could represent". The discussion appears to have been about the extent of data disclosed to Camelot's bid team via a virtual data room and what advantage the Camelot bid team would have gained as a result. The GC now seeks to apply some redactions to the document. The claimants do not take issue with those redactions other than one which redacts the words [REDACTED].
194. It is plain on the face of the email that the meeting was attended by Charles Brasted of Hogan Lovells and Javan Herberg KC. The paragraph immediately preceding the one in issue contained advice of Mr Herberg. The next paragraph contained the advice of Mr Brasted. The words in issue are the concluding sentence.
195. It seems to me obvious that this document contains legally privileged advice and I cannot see how the last sentence can be divorced from what immediately precedes it. If the point is still relied on, whether or not this was shared with Rothschild is not material as there can have been no intention to waive privilege by doing so. The claimants do not have permission to rely on this document without the redactions which HL now seek to make.

Postscript

196. As is the norm in procurement litigation, there is a confidentiality ring in place. A draft judgment was provided to the parties on 3 April 2025. The terms of the embargo initially provided that it was supplied only to counsel and solicitors who were within tier 1 of the confidentiality ring. The purpose of this embargo was to protect both confidentiality where applicable and the privilege in the documents which I had not permitted the claimants to use. Following email exchanges with counsel, I directed (i) that the draft judgment could also be disclosed to the defendant's client representatives (within tier 1) since the privilege in issue was that of the defendant and (ii) that the documents which I had permitted the claimants to use (which I referred to as the "Use Permitted Documents") could be provided to the claimants' client representatives (within tier 1). There was a time pressure on the consideration of these documents because of the dates directed for the claimants' amendments and the statutory time limit.
197. I also asked counsel to seek to agree the extent of redactions for the purposes of an open judgment, again to preserve confidentiality and privilege. Patently, where I had not permitted the claimants to use a document in respect of which the defendant claimed privilege, it would be wrong for the content of that document to be disclosed in an open judgment but some reference to the document would be necessary for intelligibility.
198. It is important to record that, at the hearing on 7 March 2025, no issues were raised by the parties as to whether any document in respect of which the defendant claimed privilege was not in fact privileged. The focus of the submissions was on the obviousness of privilege. In light of the terms of the draft judgment, a dispute then arose firstly as to whether I had decided that some documents were not, in fact, privileged. That dispute was relevant to the scope of redactions because the defendant now argued that, even where the claimants were permitted to use a document, that did not amount to a loss or waiver of privilege for all purposes. That was not an argument that had so much as been mentioned at the hearing. Both of these matters seemed to me potentially to call for a further hearing and further decision.
199. In the event, the parties agreed the scope of the redactions and that is reflected in the open version of this judgment. For the avoidance of doubt, (i) although it may follow from my observations in respect of some documents, unless expressly stated in this judgment I have made no decisions as to whether any particular document is or is not privileged, and (ii) the fact of the redaction in the open version of this judgment is not to be taken as a decision on the issue of loss or waiver of privilege in this case or more generally.