1 Friday 12 January 2024 2 (10.04 am) 3 MR BEER: Good morning, sir, can you see and hear us? 4 SIR WYN WILLIAMS: Yes, I can, thank you very much. 5 MR BEER: Thank you, sir. May I call Christopher Jackson? 6 SIR WYN WILLIAMS: Yes, of course. 7 CHRISTOPHER MICHAEL JACKSON (sworn) 8 Questioned by MR BEER 9 MR BEER: Good morning, Mr Jackson. My name is Jason Beer 10 and I ask questions on behalf of the Inquiry. Can you 11 give us your full name, please? 12 A. Christopher Michael Jackson. 13 Q. Thank you for the provision of a lengthy witness 14 statement to the Inquiry, in response to two Rule 9 15 requests, that is requests served under Rule 9 of the 16 Inquiry Rules 2006. I think the first request was dated 17 31 October 2023 and the second request was dated 18 17 November 2023, and you addressed both of them in 19 a witness statement that was dated and signed on 20 19 December 2023; is that correct? 21 A. That is correct. 22 Q. Although that's the date you signed the witness 23 statement, does it, in fact, reflect the position on the 24 facts as you understood it on 1 December 2023? 25 A. Yes, it does, that was the deadline to submit the draft 1 1 witness statement. There are two updates in that the 2 exhibit adds two letters from mid-December from us, 3 because they went to the issues but, other than that, 4 I didn't update it. 5 Q. Yes, so there's two letters exhibited by way of update, 6 both dated 15 December 2023. 7 A. That is correct. 8 Q. That witness statement is 99 pages long excluding its 9 appendices. With appendices it's 135 pages long and 10 there are 34 exhibits to it. Can we look at it, please, 11 WITN10810100, and if you can look in the hard copy 12 version in front of you at page 99. 13 A. Yes. 14 Q. Is that your signature? 15 A. It is, yes. 16 Q. Are the contents of the witness statement true to the 17 best of your knowledge and belief? 18 A. They are. 19 Q. That can come down from the screens. Thank you. In the 20 witness statement and in the appendices to it, you 21 promise at various points to update the Inquiry as to 22 further information concerning the Post Office's 23 disclosure exercise as it becomes known to you; is that 24 right? 25 A. That's correct, yes. 2 1 Q. I think it's right that on 9 January this year, 2 ie Tuesday of this week, you sent three letters to the 3 Inquiry in that regard? 4 A. Yes, we did. 5 Q. Two concerned the remediation or so-called remediation 6 of what we're going to discuss this morning, the 7 Microsoft Exchange/365 issue; is that right? 8 A. That's right. 9 Q. And those two letters contained an explanation of how 10 the Post Office proposed to remediate, put right what 11 had gone wrong, disclosure in relation to Phases 2 and 3 12 of the Inquiry -- 13 A. That's right. 14 Q. -- and the disclosure that's so far been given in 15 relation to Phase 5 of the Inquiry, which we've yet to 16 start. 17 A. That is correct. With Phase 5, clearly the work was 18 ongoing so with some of the requests that were live that 19 didn't need to be remediated because the Exchange issue 20 was known, so was factored in to the responses and, 21 indeed, some of the remediation work has already been 22 done for one of the earlier notices. 23 Q. The third letter of Tuesday concerned an update to what 24 is called by the Post Office a "structural review". 25 A. That's correct, yes. 3 1 Q. The structural review is -- is this right in general 2 terms -- a re-examination, consolidation and, if 3 necessary, rectification exercise in relation to all of 4 the disclosure which has been sought by the Inquiry from 5 the Post Office? 6 A. The structural review is focused on we call the 7 left-hand side of the electronic disclosure reference 8 model, although we're covering hard copy as well, which 9 is really the elements about identification, 10 preservation and collection. Separately, we've been 11 looking at an ongoing basis as to how Post Office 12 responds to Rule 9 requests but the structural review 13 itself is focused on those earlier stages: effectively, 14 identification, preservation and collection. 15 Q. Thank you. By way of roadmap, this is principally for 16 the Chair and the Core Participants, there are three 17 topics that I propose to examine with you. For those 18 new to the Inquiry, they may rapidly find the issues 19 that we're going to discuss in a moment rather dry. All 20 I would say is that, if others find the issues somewhat 21 desiccated, for the Inquiry the disclosure of documents 22 is the lifeblood of the Inquiry and is required for 23 a proper examination of the issues in our terms of 24 reference and the 218 issues that we've identified in 25 our list of issues. 4 1 The Inquiry regards it as essential that fulsome and 2 timely disclosure of documents is given to the Inquiry 3 by Core Participants and, in particular, the Post 4 Office, because securing evidence from witnesses and 5 testing that evidence and challenging that evidence is 6 itself reliant on obtaining primary documentary 7 materials. I think you would understand that? 8 A. Fully understood and in full agreement, yes. 9 Q. So the three topics which I propose to address with you 10 are as follows: firstly, the Microsoft Exchange/365 11 issue, what it is, what the issue is, when it arose, how 12 it was discovered, what has been done to put it right 13 and when the Inquiry will receive documents to which it 14 is entitled. That latter issue is particularly relevant 15 because it may affect the timetabling of Phases 5 and 6 16 of the Inquiry. 17 A. I understand. 18 Q. Secondly, and relatedly, an examination of how the 19 disclosure of Microsoft Exchange/365 documents relevant 20 to yesterday's witness, Stephen Bradshaw, was undertaken 21 and what assurances you can give on behalf of the Post 22 Office that disclosure of Microsoft Exchange/365 23 documents in relation to other witnesses who are to come 24 will be given in a more timely manner. 25 Thirdly, the progress that's been made with the 5 1 structural review of disclosure and the extent to which 2 it is necessary to take into account the outstanding 3 tasks on the structural review in timetabling Phases 5 4 and 6. 5 Before we look at the issues, can I start with your 6 background, the appointment of you and your firm by the 7 Post Office and some of the overview points that you 8 make in your witness statement. 9 So, by way of background, I think it's right that 10 you're solicitor and a solicitor advocate; is that 11 right? 12 A. That's correct, yes. 13 Q. You've been in practice since 1988? 14 A. Yes, I have. 15 Q. You joined Burges Salmon in 1991 and became a partner in 16 1997? 17 A. I did, yes. 18 Q. And you remain a partner at Burges Salmon? 19 A. I do, yes. 20 Q. In terms of appointment, you and your firm were first 21 appointed to act for the Post Office Limited in 22 May 2023; is that right? 23 A. That's correct, yes. 24 Q. Your firm and Fieldfisher, another firm of solicitors, 25 a subcontractor with whom you act in a joint venture, 6 1 began to work for the Post Office on this Inquiry in May 2 2023? 3 A. We did, in late May 2023, June, July and August we were 4 learning the issues and mobilising and getting up to 5 speed, and I took over as recognised legal 6 representative on 1 September. 7 Q. I was about to ask you that. You replaced Gregg Rowan 8 of Herbert Smith Freehills as the recognised legal 9 representative, the RLR, of Post Office Limited on 10 1 September 2023. 11 A. I did yes. 12 Q. I think it's right, however, that Herbert Smith 13 Freehills continued to work for the Post Office in 14 relation to a range of issues connected with the 15 Inquiry, most notably they retain operational conduct of 16 the Post Office's work on Phase 4 of the Inquiry -- 17 A. They do, yes. 18 Q. -- the phase we're currently engaged in? 19 A. Yes. 20 Q. Can we just turn up paragraph 8 of your witness 21 statement, please, on page 5. It should come up on the 22 screen. This is just to set the context of some of the 23 things that you are subsequently going to describe. You 24 say in paragraph 8: 25 "The current situation is not one that anyone would 7 1 wish to see continue. Post Office has asked me to 2 convey its apologies for the current situation and to 3 assure the Inquiry and other Core Participants that it 4 is a Post Office priority to get to a position where 5 hearings (and planning and preparation for hearings) can 6 take place from a stable basis with the risks of further 7 emerging data sources minimised and managed so far as is 8 practicable." 9 Q. Yes? 10 A. Yes, that is correct. 11 Q. You say "The Post Office has asked me to convey its 12 apologies for the current situation"; does that 13 situation, the current situation, described as of 14 1 December 2023, remain today? 15 A. Sorry could you clarify which part of the situation? 16 Q. That's part of the question. 17 A. The situation where -- as I understand it, if I'm 18 correct, where things are having to be done shortly 19 before hearings in order to present evidence, that is, 20 as I understand it, the position, and we're not -- 21 Q. We're not out of that situation yet? 22 A. We're not out of that situation yet and the aim and the 23 objective -- there's two objectives: one is to get the 24 Inquiry the evidence that it requires to do all that it 25 needs to do under the terms of reference to bring out 8 1 all the truth and all the facts; the second, from our 2 perspective and Post Office's perspective, is to get 3 hearings on to a stable case-managed basis and -- 4 Q. By that, you mean that when we say we want to call Mr X 5 in three weeks' time on a Tuesday, Mr X does get called 6 in three weeks' time on a Tuesday -- 7 A. Correct. 8 Q. -- and there isn't a flood of disclosure in the days 9 leading up to that Tuesday, which means Mr X has to be 10 postponed or people have to work through the night in 11 order to read documents? 12 A. Correct. There is always, in these situations, below 13 the waterline, an occasional document that might emerge. 14 That is the reality, as -- of the situation but, as 15 a structural issue, it's certainly Post Office's aim to 16 move beyond the present situation. 17 Q. I think you'd acknowledge that we're not dealing with 18 the usual vicissitudes of legal practice here, where the 19 occasional document appears from below the waterline. 20 What's been happening in Phase 4, and what's likely to 21 continue to happen in Phase 4, are dozens, hundreds or 22 thousands of documents relating to a single witness are 23 disclosed proximately to that witness giving evidence. 24 That's what's been happening. 25 A. That has been the consequence of the Exchange issue, 9 1 yes. 2 Q. So the position, the current situation, as you were 3 describing it, as of 1 December 2023, remains current 4 for Phase 4, in that the Post Office has not yet given 5 full and complete disclosure for the witnesses who are 6 due to give evidence in Phase 4 for the rest of this 7 month? 8 A. I don't know. I'm not operationally sighted on Phase 4 9 because we're not, for the reasons you say, delivering 10 that. 11 Q. But I think what you can say is it remains the case at 12 the moment -- we're going to discuss this later -- that 13 the Post Office hasn't given full and complete 14 disclosure in relation to witnesses who gave evidence in 15 Phase 2 of the Inquiry, as between October and 16 December 2022? 17 A. It needs to be checked -- because of the periods that 18 apply to the Exchange issue, the extent to which there 19 is additional material for them has to be checked but, 20 until that is checked, I can't say with certainty that 21 there is missing material. 22 Q. The same applies to the witnesses who gave evidence in 23 Phase 3 of the Inquiry as between January and May 2023? 24 A. That would need to be checked also. 25 Q. The same applies, ie the Post Office has not given full 10 1 and complete disclosure in relation to the witnesses who 2 are going to give evidence in Phases 5 and 6 of the 3 Inquiry, which the Inquiry wishes to commence as soon as 4 possible after it concludes Phase 4? 5 A. The concept of -- there is more material to come. Full 6 and complete is always, in modern disclosure, 7 effectively -- there becomes a degree of judgement, 8 which has to be a degree of judgement for the tribunal 9 on an informed basis. It shouldn't be for the Core 10 Participants to decide. But there becomes, after 11 a certain point, a set of diminishing returns on what 12 can be given relative to time and effort and the 13 probative value of that material. So there is -- 14 Q. But we're not dealing with that here, in relation to 15 Microsoft Exchange, are we? We're not talking about 16 diminishing returns type material that it would, 17 applying a counsel of perfection, one would have? The 18 material that's been turned over as a result of 19 Exchange/365 has included important material? 20 A. The material in Phase 4 I haven't looked at but 21 I understand that and I am conscious that, counsel team, 22 you would be wishing to receive it if it is probative. 23 Q. Yes. 24 A. In terms of the Phase 5 material, the Exchange -- yes, 25 there is material within Exchange that needs to be 11 1 looked at and looked at hard because there may be things 2 in there. How much and in depth, it may vary from 3 witness to witness but it needs to be looked at. 4 Q. Would you agree that the context in which the disclosure 5 failings for which you convey the Post Office's apology 6 in that paragraph have occurred in a context that 7 includes the following: firstly, that the Inquiry is 8 examining the extent to which the Post Office breached 9 the law in not disclosing relevant documents and 10 information in criminal proceedings against 11 subpostmasters and counter clerks? 12 A. Those are very much within the terms of reference, yes. 13 Q. Secondly, the present disclosure failings occur within 14 a context that includes the Court of Appeal Criminal 15 Division having already found that there was 16 non-disclosure by the Post Office, and that such 17 non-disclosure breached the law in the criminal 18 proceedings and that it rendered a large number of 19 convictions unsafe? 20 A. Yes, that was the finding in Mrs Hamilton's appeal and 21 that of others. 22 Q. Thirdly, the Inquiry is examining the extent to which 23 the Post Office breached the law in failing to disclose 24 relevant documents and information in civil proceedings 25 that it took against subpostmasters? 12 1 A. Again that is very much in the terms of reference, yes. 2 Q. Fourthly, we're examining the extent to which the Post 3 Office breached the law in failing to disclose relevant 4 documents and information in the Group Litigation? 5 A. Again, that is within the terms of reference. 6 Q. We're looking at the extent to which the Post Office 7 failed to disclose relevant documents and information in 8 proceedings before the Court of Appeal Criminal 9 Division? 10 A. That is within the terms of reference. 11 Q. We commenced our Phase 2 hearings in October 2022 and 12 the Inquiry has been required to delay the evidence of 13 many witnesses -- 14 A. Yes. 15 Q. -- including Gareth Jenkins, who has provided a witness 16 statement to the Inquiry and whose evidence had to be 17 adjourned because of failures in the Post Office's 18 disclosure? 19 A. Yes. 20 Q. The context that we're discussing today includes, as 21 part of its background, the regrettable delay to Phase 4 22 hearings at the end of July 2023, the rearrangement of 23 hearings to September to December 2023 and then, still 24 further, the delay of some witnesses that we're due to 25 be called in that period to this year? 13 1 A. That was the period when we were getting up to speed, so 2 the reasons for those hearings being delayed, I'm 3 afraid, I don't know. 4 Q. Can we turn up paragraph 29 of your witness statement, 5 please, which is on page 16. In the second line, you 6 say: 7 "Historic data governance problems, many of which 8 were embedded within the Post Office's data landscape 9 over many years, have risen to the surface under the 10 scrutiny of the Inquiry and Post Office's internal and 11 external Inquiry teams." 12 To be clear, this is a part of your witness 13 statement that seeks to explain the context in which the 14 disclosure exercise is occurring. 15 A. Yes, that context is set out at paragraph 28, which 16 is -- and also then go on to set out factors which apply 17 to many organisations at paragraph 31, because this 18 period, this 25-year period covered by the terms of 19 reference, is the period where IT was rapidly 20 developing. I think Mr Justice Fraser makes similar 21 comments in his judgments, and so in paragraphs 28 and 22 31, I'm trying to explain that context, and paragraph 29 23 sits with those two paragraphs. 24 Q. Thank you. You say "historic data governance problems". 25 What do you mean by "data governance"? 14 1 A. Knowing what is where. 2 Q. Would you agree with this formulation of it: that data 3 governance is the setting of internal standards within 4 a company or an organisation, data policies that apply 5 to how data is gathered, to how it's stored, to how it's 6 processed, to how it's disposed of or retained? 7 A. Governance is about rules and also how they are 8 implemented but also about knowledge as to how they -- 9 knowledge as to what is where and how those things are 10 implemented, yes. 11 Q. So governance should be about policies, written 12 documents, which define roles and responsibilities and 13 which set out processes that ensure, would you agree, 14 the accountability and ownership of data? 15 A. That is part of governance because policies set the 16 framework but then what you do within that is also part 17 of governance. 18 Q. What are the data governance problems embedded within 19 the Post Office to which you are referring? 20 A. Many are -- they are not atypical of large complex 21 organisations during this period that I've seen, 22 I haven't -- clearly, we've come to this very recently, 23 so we've far from done a comparative analysis. 24 Q. Never mind about the comparison with other organisations 25 but what are the data governance problems embedded 15 1 within the Post Office that you have identified? 2 A. Summarised as not a full map of systems and what is 3 where and how it is evolved over time. 4 Q. You say that these are only rising to the surface -- 5 does that mean discovered -- in the course of the 6 Inquiry? 7 A. Yes, because when -- most organisations don't have to go 8 back in time, so when, subjected, rightly, to intensive 9 scrutiny about some serious problems that have happened, 10 then all of those things are interrogated and all of the 11 things that are not known become apparent. 12 Q. But these data governance problems, I think you're 13 telling us, didn't rise to the surface in the course of 14 the Group Litigation or in the course of the disclosure 15 exercise undertaken for the purposes of giving 16 disclosure to the CCRC or the Court of Appeal Criminal 17 Division? 18 A. I don't know because I wasn't involved at that time. 19 I'm not aware that they did but I don't know. 20 Q. It was just that this paragraph, this part of the 21 paragraph tended to suggest that they have risen, and 22 only risen, to the surface in the course of the Inquiry? 23 A. They have certainly done that. Whether they had done so 24 previously or not, I don't know. 25 Q. But this is about the way that the Post Office ran its 16 1 business in the relevant period and, in particular, how 2 it ran the part of the business that concerned 3 governance of data. There were problems with that, is 4 what you're saying? 5 A. Yes. 6 Q. Thank you. That can come down. 7 Can we turn, please, to something you said in 8 a letter back on 16 October 2023 as to the Post Office's 9 intended approach to disclosure, outside of the three 10 failures in disclosure we were then looking at, which 11 were: misuse and use of search terms; the proper and 12 improper conduct of a deduplication exercise; and the 13 non-disclosure of families of documents. 14 Can we start, please, by looking at that letter 15 which is WITN10810102. You'll see that this is a letter 16 written by you or by your firm, yes? 17 A. It was by my firm. I was away at that time but, yes, it 18 was by my firm, by my team, yes. 19 Q. By your team, okay, and jointly, obviously, with 20 Fieldfisher, with whom you act in a joint venture? 21 A. Yes, well, I think in my -- operationally, it's a joint 22 venture. As I say, contractually, it's 23 a contract/subcontract relationship but, operationally, 24 yes, it is. 25 Q. This is amongst the first letters that you wrote to the 17 1 Inquiry and you say in the first paragraph, in the third 2 line: 3 "As explained in [an earlier letter] we consider it 4 is important to be frank and direct concerning the 5 approach to disclosure so that any points of concern or 6 differences in understanding can be discussed." 7 Then over the page, please, to page 2 and look at 8 paragraph 4. So here you're dealing with different 9 remediation work, ie remediation of different failings, 10 yes? 11 A. I mean, there had been quite a number of letters sent 12 before that. I mean, they -- between the end of August 13 and this one, I think there'd been 22 points of 14 engagement suggesting discussion or case management -- 15 Q. Yes. No, we get about ten letters a day from the Post 16 Office, many of which are more than 10 pages long. 17 You say in paragraph 4 that the Post Office 18 understands the Inquiry's concerns about -- and these 19 were the there is problems that we were then dealing 20 with back in the autumn and have been the subject of 21 previous hearings -- "Search Terms, Deduplication and 22 Family Documents". You say that it, the Post Office, 23 has apologised previously for what has happened and the 24 impact on the Inquiry and sought to rectify them as 25 swiftly as possible. You say: 18 1 "At the latest update [paragraph 5], the remediation 2 and assurance exercises undertaken following the July 3 Directions have involved the review of at least 402,000 4 documents and the production of 11,357 documents: 5 a relevance rate of about 2.82% ... As the Inquiry has 6 noted, this figure includes a high proportion of 7 duplicates or near duplicates of documents which have 8 already been produced to the Inquiry. When they are 9 taken into account [the Post Office] estimates the true 10 ... figure of relevancy when excluding exact and near 11 duplicates to be 1.83%." 12 By way of comparison, for a previous review for 13 a Section 21 notice, 48% relevancy rates were returned. 14 You say paragraph 7: 15 "Although these stringent measures have been 16 necessary in relation to the specific circumstances 17 arising in respect of the Three Issues [that's search 18 terms, deduplication and family documents], this 19 approach does not reflect what is required for [the Post 20 Office] to comply with its disclosure obligations in 21 relation to previous or future requests. The principle 22 of reasonableness in relation to disclosure to the 23 Inquiry -- even if operating at the more stringent end 24 of the spectrum -- does not, and cannot, require [the 25 Post Office] to leave every stone unturned. Such 19 1 a standard is impossible for [the Post Office] 2 realistically to comply with. Therefore, [the Post 3 Office] does not intend to adopt a similar approach to 4 future requests having regard to the low rates of 5 relevance." 6 Two questions arising from that, if I may. You'll 7 see there that what is described is said to be a low 8 rate of return for relevant documents -- 2.82 per cent 9 before deduplication and 1.83 per cent after 10 deduplication -- and it's that which leads to the 11 suggestion in paragraph 7 that the Post Office isn't 12 going to carry on doing this and it's going to apply, 13 instead, a different approach. 14 Can I understand, firstly, have we understood this 15 correctly: that when carrying out remediation work, the 16 Post Office used, as a measure of reasonableness and 17 proportionality of what it was doing, the size of the 18 pool of documents that it itself created for itself, and 19 then compared it to the size of the pool of documents 20 that it ended up disclosing? 21 A. No, that isn't correct. 22 Q. That's how this reads; do you agree? 23 A. No. Um -- 24 Q. It says, "We reviewed 402,000 documents", that 402,000, 25 would this be right, involved the application of some 20 1 search terms to create a pool? 2 A. Yes. 3 Q. Then you measured how many of them ended up being 4 disclosed, and that figure was 2.8 per cent of them? 5 A. That was done after the event, not to design the 6 remediation. 7 Q. I'm not saying it was designed that way, I'm saying it 8 ended up being the case that you disclosed 11,357 9 documents? 10 A. Post Office did, yes, in relation to the three issues. 11 The -- this letter was a request for a case management 12 discussion, covering about seven or eight items, of 13 which this was one, prefaced saying "This is our or Post 14 Office's understanding of the position", but -- 15 I haven't got the rest of the letter in front of me on 16 the screen, but saying words to the effect of "want to 17 check the extent to which this is or is not aligned with 18 the Inquiry's expectations, so could we please discuss". 19 Q. Yes, and you know that what happened is that although 20 there was a meeting, the Inquiry wish to maintain more 21 than an arm's length relationship with the Post Office 22 and would not seek to approve or pre-approve what the 23 Post Office proposed to do? 24 A. Certainly, there wasn't a request for approval or -- 25 yes, there was a meeting on the 3 November. The first 21 1 Rule 9 -- rather than the discussion, the first Rule 9 2 Request to me arrived asking for an explanation about 3 this letter, which I've sought to give, yes. 4 Q. I want to get back to what's written here which seems to 5 be "In deciding what's reasonable, we have had regard to 6 the relevant rate of return on a pool of documents that 7 we created"; do you agree that's what it says? 8 A. Well, Post -- it's the rate of return for the 9 remediation on the three issues. 10 Q. Yes. 11 A. It was done over the summer, and it is used, 12 effectively, as a reality check. In all electronic 13 disclosure, for the reasons I've set out in paragraph 31 14 of my witness statement, effectively, there is always 15 a judgement call but it has to be a judgement call for 16 the tribunal, ultimately, because it can't be for the 17 Core Participants in the final instance. It has -- the 18 Core Participants or the disclosing party should explain 19 what it is doing and -- 20 Q. What you're explaining, that what your client is doing 21 it here, is it's saying: "We created a pool of documents 22 somehow that had 402,000 documents in it we ended up 23 disclosing 2.82 per cent of them, ie 11,357 documents." 24 A. Mm. 25 Q. "That's such a low rate of return we're not going to 22 1 carry on doing it this way; it's going to be different 2 in the future." 3 A. It's the balancing exercise between the two objectives 4 I mentioned earlier, which is -- 5 Q. Just before you go on, do you agree that's what this 6 says? 7 A. It -- as a mathematical situation, yes. 8 Q. Ie the maths has been used to decide what the Post 9 Office is not going to do in the future, because it is 10 said to be not reasonable? 11 A. As a proposition, with the request for discussion with 12 the Inquiry, in the way that is done in all case 13 management hearings in civil litigation and in our 14 experience in an inquiry context. 15 Q. Does it occur to you, looking at this now, that this may 16 be a flawed approach? 17 A. No. 18 Q. Does it occur to you that it depends what approach you 19 take to the creation of the initial pool that will 20 significantly affect the percentage that you end up 21 with? 22 A. The size of the initial pool and what you end up with 23 after review are directly related to each other. The 24 relevance rates -- sorry. 25 Q. How you created the initial pool is going to affect the 23 1 percentage you end up with, isn't it? I mean, take 2 an example: if you said, "In my creation of my initial 3 pool I'm going to use as my search term the word 'the', 4 I want to collect in an initial pool, all documents with 5 the word 'the' in them, and I'm going to end up with, 6 say, 10 million documents. I'm then going to apply the 7 inquiries Rule 9 and Section 21 search terms to them, 8 and when I apply those search terms to them, I find 9 10,000 documents that are actually relevant. I've got 10 a relevance rate there of 0.1 per cent. That's 11 unreasonable. It's such a low percentage, I can't carry 12 on doing things this way". 13 So the way you construct the initial pool is 14 directly relevant to the percentage that you end up 15 with, agreed? 16 A. In this context, this was a remediation exercise done by 17 HSF. 18 Q. Yes. 19 A. So, yes, as a proposition. These, as I understand it -- 20 but this was not an exercise we did -- the 402,000 were 21 documents already understood to be relevant or for the 22 purposes of early disclosures. That may not be -- 23 Q. I appreciate this was done by HSF but your firm, more 24 properly, is writing here to say "This approach of 25 comparing the initial pool that we find for ourselves 24 1 and then the final pool that we end up disclosing, 2 a comparison numerically between them, is directly 3 relevant to the reasonableness of the exercise." 4 A. Yeah. 5 Q. Isn't the relevant question here, the relevant issue, as 6 part of this exercise, "We ended up disclosing 11,357 7 documents that, if we hadn't done this, the Inquiry 8 would never have had"? 9 A. It should absolutely -- I think it -- it goes on to say 10 this should definitely have been done in this context. 11 Q. But it says, "We're not going to do that in the future"? 12 A. That is the proposal and the request to discuss. 13 Q. But do you agree that the relevant point is: "We, the 14 Post Office, ended up disclosing just shy of 11,400 15 documents that, had we not done this, the Inquiry would 16 never have had?" 17 A. Correct. 18 Q. Not "We created a pool of 402,000", and there's such low 19 relevance rate that we think this is an unreasonable 20 approach for the future"? 21 A. Relevance rates are used -- because there's 22 a relationship between review and time, and so linking 23 into the impact on hearings, if one is reviewing at 24 very, very low rates of relevance, that -- the standard 25 within the Inquiry's disclosure protocol is rightly 25 1 reasonable in all the circumstances, and with all the 2 appropriate rigour and I'm paraphrasing. 3 Q. I'm exploring with you the question of how the Post 4 Office approaching the question of reasonableness. What 5 measure, what indicia it used itself for deciding 6 whether something was reasonable and, as paragraph 7 7 says, this is not going to continue? 8 A. What the subsequent proposals go on to say, I think, is 9 effectively that it will be done on a specific basis, 10 which is what, from recollection, HSF had been doing. 11 Gregg Rowan's witness statement often 5 September 12 basically said they were taking Rule 9s on 13 a case-by-case basis and then applying the standard, 14 which is reasonable in all the circumstances. And this, 15 I think, was a reversion to the discussion about 16 reasonable in all the circumstances. So with certain 17 Rule 9 requests, one looks at absolutely everything. 18 For others, it is simply not practicable and, if, as 19 a reality check, one is reviewing at very low rates of 20 reliance, that indicates a problem both as to the 21 process and also impact on hearings and time but it was 22 a request to discuss. 23 Q. Can we scroll down to paragraph 8, please, to see 24 what -- and just onto the next page as well. You 25 continue: 26 1 "The Post Office will revert to its previous 2 approach to the identification and review of potentially 3 relevant materials. If a search term returns 4 a significant pool of documents, this will be refined 5 using revised search terms." 6 Just stopping there, does that mean "If we use 7 search terms and in our judgment the number in the pool 8 is too high, we will come up with some other search 9 terms in order to make the pool smaller, so we've got 10 less documents to look at"? 11 A. Yes, that is the only way some -- it depends on the 12 particular request that has been made. So to draw 13 a really practical example, I won't go into the detail 14 for confidentiality reasons, but Request 21/08 was for 15 a very defined pool of documents, so one can identify, 16 isolate and review effectively all of them, up, down, 17 top and sides. 18 Section 21/03 covered a period of two decades, 19 potentially, or certainly many, many years, and had 14 20 issues, many, many sub-issues, and many, many 21 individuals. So one starts with the search terms, and 22 if the document pool initially -- and you check it -- 23 comes to several million, which it did, but there is 24 a statutory requirement also to meet a deadline or to 25 apply for extension if you can't, then, as a matter of 27 1 practicality your -- that's indicating that you're 2 capturing things that are probably not relevant so yes, 3 you refine the searches to try to hit that sweet spot 4 between impact and getting the right material or as much 5 of the right material as you can. 6 Q. So search terms are therefore selected, identified, 7 designed into the system, that have the intention of 8 returning fewer documents? 9 A. No. That's not the intention. The intention of search 10 terms -- when search terms are used, it's not the only 11 technique that's used. For some searches you can do it 12 on a targeted or specific basis. Where search terms are 13 used, the intent is that you are getting the material 14 that the other party or the court or tribunal or inquiry 15 requires, in the time that it needs it, balancing the 16 realities of electronic disclosure. 17 Q. The letter continues that on the second page there: 18 "... dip sampling will assure the quality of the 19 review exercise." 20 Who conducts the dip sampling exercise? 21 A. It will be done within the disclosure team. There are 22 different levels to disclosure team, so you have your 23 first tier reviewers, that will be done generally by 24 your Tier 2 reviewers, or by other, more senior people 25 for all reviews, so it is done within the disclosure 28 1 expert team. 2 Q. Okay, so they're lawyers, not the KPMG or other 3 e-disclosure providers? 4 A. Correct. 5 Q. You say the final result will be a smaller pool? 6 A. Yeah. Sorry, there will be dialogue with KPMG as to -- 7 as you refine search terms, to find out what is working 8 to get to the evidence that you need. So there's input 9 from KPMG but the dip sampling is done by the lawyers. 10 Q. In paragraph 10 in the fourth line, you say: 11 "Although the professionals employed by [the Post 12 Office] and its external advisors are well used to 13 working at pace and for prolonged periods of time, it is 14 not realistic to expect that they will be able to 15 maintain this pace and continue to work during the 16 evenings and over weekends." 17 Did what we see there in paragraph 10 lead to the 18 redesign of the approach of using refined search terms 19 to identify smaller pools of potentially relevant 20 documents? 21 A. No. They're separate points. Clearly, well -- there's 22 a wellbeing aspect to teams. Lawyers work hard. We all 23 work hard and do weekends and late nights, and so on. 24 The point that I understand that this paragraph is 25 trying to make is that, if that is relentlessly done 29 1 over a period of weeks or months, then that has, as it 2 says in the paragraph, a human impact, and has to be 3 factored in. 4 Q. As a result of this, we asked whether any significant 5 changes had been made in relation to the resourcing by 6 the Post Office of its disclosure and, in short, the 7 answer is that resourcing has not decreased. Instead, 8 since 5 September hearing, it's materially increased; is 9 that right? 10 A. That's right. Obviously, it flexes according to what is 11 happening at any one time. If there are particularly 12 intensive demands, then the resources upsurges but, yes, 13 the numbers were given in my statement as at 1 December, 14 which is a particularly intensive period. 15 Q. We can look at those, it's page 98 of your witness 16 statement, please. Page 98, please: 17 "In terms of the relative sizes ... these are 18 broadly similar between [Herbert Smith Freehills] and 19 [Burges Salmon/Fieldfisher] ..." 20 Since 6 September, the numbers are as follows, and 21 these are absolute numbers rather than full time 22 equivalents, yes? 23 A. They are, so some people will be full time some people 24 will be -- 25 Q. Part time? 30 1 A. -- providing half of the time -- they're full time -- 2 sorry, many of them will be full time lawyers but they 3 will be working. 4 Q. Part time in this file? 5 A. Exactly so, but what footnote 17 was trying to do was to 6 take and produce an equivalence to say those who are 7 working very few hours are in the low tens. So people 8 are working substantively -- 9 Q. So, at Herbert Smith, 171 individuals including 68 10 trainees, paralegals or other first tier reviewers; at 11 Peters & Peters, 45 and 17; at Burges 12 Salmon/Fieldfisher, 175, of whom 80 were trainees, 13 paralegals or other first tier reviewers. 14 Elsewhere you say that the general figure has 15 remained at about the 350 mark; is that about right? 16 A. I don't recall that paragraph. Could I take -- go to 17 that paragraph? 18 Q. I can't remember where I got it from. I think it was 19 a letter. These figures we see here, do they continue 20 to date? 21 A. As -- they will have gone down since then, because 22 the -- clearly, you don't have people sitting around not 23 doing things and once -- so, for example, as is 24 currently the situation, there is a prioritisation of 25 Exchange data. The constraint is not the availability 31 1 of review capacity; it's the waiting for the data to 2 come through. 3 So those numbers are available, they're not all -- 4 it's probably fewer than this working at the moment but 5 that's not a scaling back of the team, any other reason 6 than available data and task. 7 Q. Thank you. With that background then, can we turn to 8 the first issue, Microsoft Exchange/365, and turn up 9 page 24 of your witness statement, please. You set out 10 here in italics the issues that the Inquiry asked you to 11 address, on all of that page; is that right? 12 A. That's -- 13 Q. That's an extract from the Rule 9 Request? 14 A. That's correct. 15 Q. If we scroll down so we can see the remainder of (e). 16 Then in paragraph 34 of your witness statement, if we go 17 forwards to that, please, which is on page 24. You say 18 that: 19 "[The] Post Office has yet to conclude its 20 investigation into these issues ..." 21 A. I did, yes. 22 Q. The Microsoft Exchange/365 issue, I think was first -- 23 you're going to tell us -- discovered by the Post Office 24 in May 2023? 25 A. The -- I'm trying to recall the month. The trigger for 32 1 it was the email, the Wise email from 2011, that was 2 initially picked up by HSF. I'm just trying to recall 3 the particular month that would -- 4 Q. That was May. So Andrew Wise had accessed an email in 5 order to answer a Freedom of Information request, it was 6 the provision of what became known as Appendix 6 and 7 Herbert Smith Freehills identified why haven't we got 8 part of that email -- why haven't we got that email in 9 the searches we're conducting? That was one of the two 10 triggers; is that right? 11 A. That is my understanding, yes. 12 Q. That was in May 2023. Why is it that the investigation 13 into the failure to disclose Microsoft Exchange/365 data 14 has taken eight months, at the time you were writing and 15 still now, because I don't think it's concluded yet? 16 A. I don't think it has taken eight months to get to that 17 point. I think the Inquiry, it was pursued between May, 18 June, July and then the -- written to the Inquiry in 19 August. By the time we'd got to -- the state of 20 understanding has not advanced materially since the 21 position was set out for the Inquiry in early October. 22 Q. It was the part of this paragraph that says that Post 23 Office has yet to conclude its investigations into the 24 issue? 25 A. As at 1 December, and it will continue to, effectively, 33 1 test the state of knowledge. This is the state of 2 knowledge as it exists at the moment. 3 Q. So the investigation wasn't concluded on 1 December, has 4 it now concluded? 5 A. Yes, in terms of the working conclusions, but will it 6 not -- if we have a reason to doubt it, we will go on 7 to -- we will look at whether these conclusions are 8 right. We will keep them under review, effectively. 9 So -- 10 Q. Yes, of course, so the investigation into the 365 issue 11 has concluded; when did it conclude? 12 A. It wasn't set up as a formal investigation with a formal 13 start date or a formal conclusion date. It is under 14 examination. So, for example, there are issues with 15 post-1 January 2016 data, for example. The working 16 assumption, at the moment is that items 17 post-1 January 2016 are trivial -- and I can give some 18 examples -- but that is being reality checked and if we 19 find that that is not the case, then we will update the 20 Inquiry. So it's effectively keeping alert to further 21 developments and actively checking. 22 Q. But although what you tell us today therefore can't be 23 taken to be the last word on the subject, would that be 24 right, the issue has been sufficiently investigated, 25 would you agree, for you to be able to give us some 34 1 strong evidence-backed conclusions? 2 A. Yes, and these are evidence-backed conclusions that we 3 will keep testing, so we are confident they remain 4 valid. 5 Q. Thank you. Can we start, then, as the first issue -- 6 that statement can come down thank you -- with 7 an explanation of the systems. 8 A. Yes. 9 Q. I'm afraid this is super dry. So can we start with 10 an explanation of the Post Office's email systems and 11 how they've changed over time. We need to do this to be 12 able to understand what we're going to speak about 13 later. Before even that, can we set out some 14 terminology and concepts that are relevant to 15 a disclosure exercise relating to emails. 16 A. Of course. 17 Q. You kindly undertake this for us in part on page 26 of 18 your witness statement. And paragraph 36(a), (b) and 19 (c). So if we can look at page 26, paragraph 36. 20 A. Yes. 21 Q. You tell us about, essentially -- if we scroll down to 22 (a), you tell us about essentially three species or 23 types of email and email exchange or gateways. So 36(a) 24 to start with. You say: 25 "First, and most recognisably, when an email is sent 35 1 from or to an email client such as Outlook ... on 2 a user's device this is stored in a local email data 3 file or mailfile on the device ... and email data on 4 that local file can be viewed from the email client even 5 when the device is offline ..." 6 So an email client is a software application that's 7 used to access, manage and send emails, like Outlook, 8 like Apple Mail or other applications? 9 A. Yes, a gateway is something different. 10 Q. Yes. So that's the first concept, an email client like 11 Outlook or Apple Mail. Then can we look at 36(b). You 12 say: 13 "... emails are not sent directly to or from 14 an email client. Rather, the email client ... connects 15 with a cloud-based mail server that sends out or 16 receives the email. Post Office currently uses Exchange 17 as its mail server. This is a server-level email data 18 file or mailfile ('Exchange mailfile') that synchronises 19 with and replicates the local mailfile. Permanent 20 deletions of email data by users at local client level 21 will synchronise and replicate in the server mailfile 22 after 30 days unless a relevant litigation hold has been 23 applied (which would prevent permanent deletion from the 24 Exchange mailfile). By design and because of the 25 application of the litigation holds we are instructed 36 1 Post Office have put in place, Exchange mailfiles would 2 be a more complete record of emails than local mailfiles 3 so there would be no benefit to harvesting a local 4 mailfile if an Exchange mailfile also exists." 5 So what you're referring to here, in summary, is 6 what might be described broadly as the mail server? 7 A. Yes. 8 Q. Then if we can look at paragraphs 36(c) and (d): 9 "In addition, Post Office utilises a further email 10 gateway platform that records a copy of emails 11 transmitted within Post Office's Exchange server and 12 through which emails between its Exchange server and 13 an external email domain must pass. Current platform 14 used by Post Office is Mimecast. Among other email 15 services, Mimecast services include ... a repository 16 that keeps a separate, immutable copy of: 17 "(i) all external emails transmitted between 18 postoffice.co.uk email domain and any other email 19 domain; and. 20 "(ii) all internal emails sent between 21 postoffice.co.uk email addresses transmitted within the 22 Exchange server itself but are then uploaded to 23 Mimecast. 24 "That function (known as 'journalling') creates 25 an archive of email data that flows into, out of and 37 1 within Post Office that is separately held on the 2 Mimecast platform. Importantly, Mimecast only journals 3 live email traffic once Mimecast has been activated -- 4 it does not journal email data that pre-dates its 5 activation and operation." 6 So would you agree here what you're describing in 7 36(c) and (d) is what might be described as an email 8 gateway and an email gateway is a type of email server 9 that, amongst other things, protects an organisation's 10 internal email servers, provides pre-delivery protection 11 by blocking email-based threats, for example, before 12 they reach a mail server. That's not what we're 13 concerned with here. Here we're talking about 14 retention. 15 A. Yes, it has various different levels of functionality 16 but it's the point at which the email passes, you know, 17 from the client to somewhere else, hence the gateway. 18 Q. Yes. Can we go to paragraph 42 of your witness 19 statement, please, which is on page 32. If we scroll 20 down -- thank you -- you say: 21 "Given [a number of factors], loss of institutional 22 knowledge has been a key factor in the ability of the 23 Post Office to reconstruct its understanding." 24 Then you say you set out in Appendix 1 Post Office's 25 understanding of the summary position on pre-2016 email 38 1 data, as explained to Burges Salmon/Fieldfisher by Post 2 Office, accompanied by diagram prepared by the Post 3 Office to illustrate its current understanding of how 4 its email systems have evolved. 5 A. Yes. 6 Q. I'm not going to go to or through Appendix 1, it's going 7 to be on the record because this witness statement will 8 be uploaded but, essentially, what that is, is 9 a narrative account, assisted by a diagram, of the 10 evolution of the three species of email systems that 11 we've just described over a 20-plus year period. 12 A. Yes, and an earlier -- Lotus Notes was the predecessor 13 of the Microsoft systems and that's captured within the 14 diagram at Appendix 1. 15 Q. Would you agree that that's the kind of thing that 16 should be prepared at the beginning of a disclosure 17 exercise, not three years into it? I'm not criticising 18 you. 19 A. You would seek -- at the start of a disclosure exercise, 20 you would look to find out what was held where in email 21 systems and -- to gain that understanding, yes. 22 Q. Can I press you on that a little bit. Would you agree 23 that it's common sense that, when you've got to do 24 a disclosure exercise you say "Well, what kind of 25 disclosure exercise are we looking at: hard copy 39 1 documents or electronic documents?" 2 "Electronic documents." 3 "What kind of electronic documents?" 4 "Oh, we are looking at email, amongst other things?" 5 "Right, we are looking at email. We need to find 6 out what email clients were in operation across the 7 relevant period. We need to find out what mail servers 8 were in operation across the relevant period. We need 9 to find out what email gateway platforms were in 10 operation at the material times. Where are we going to 11 harvest from? Will that give us complete coverage?" 12 A. Yes, and that's what the EDRM model seeks to achieve. 13 You're tying to find out what is held where, and how you 14 get to it. 15 Q. Was that one of the first things that you decided needed 16 to be done when you were instructed? 17 A. Yes, but in the -- before we were instructed, we had set 18 out in our tender that we were going to use that 19 methodology to -- the way we put it was for Phases 5 to 20 7, which is what we'd been instructed to do. 21 Q. So you set out to create a narrative account to work out 22 what this part of the data universe looked like and, 23 would you agree that with what I suggested earlier, it's 24 common sense that you do that at the beginning of the 25 disclosure exercise, not three years into it after the 40 1 Inquiry has already heard from nearly 200 witnesses? 2 A. We didn't -- when we set out to do it we had no concept 3 of the Exchange and Mimecast issue. 4 Q. That, Mr Jackson, I think you'll know, is an answer to 5 a different question. 6 A. Sorry, there were two questions. I was attempting to 7 answer both of them. 8 Q. You were building up to an answer, okay. 9 A. I was attempting to -- the first question was: were we 10 aware that we set to deal with the Exchange and Mimecast 11 issue? And, no, we didn't because, we didn't know about 12 it. 13 Q. No, I wasn't asking that at all. If I did, I'm sorry. 14 A. At the start, yes, you do need to find out what's there 15 and you talk to the IT staff and you try to build up the 16 best picture that you can as to what is where, yes. 17 Q. Well, this is really common sense, isn't it? If you 18 translated this to something much more humdrum, 19 a physical disclosure exercise, you were instructed by 20 a client and they say, "We've got some warehouses", and 21 you ask them "How many warehouses have you got and where 22 are they?" and they say, "Well, there's one in Sheffield 23 and there's one in Birmingham", and you say, "Right, 24 which documents are held in the Birmingham one and which 25 are held in the Sheffield one?" 41 1 A. Yes. 2 Q. It's the equivalent of that, isn't it? It's working 3 out -- 4 A. At the outset, you try to find out what you've got 5 where, yes. 6 Q. Am I right, therefore, that I think you said that you 7 identified this as something that needed to be done even 8 before you were instructed, as part of the tender 9 process? 10 A. We, in our tender process we said what we do for 11 disclosure -- yes, we use EDRM to check what we're 12 doing -- 13 Q. I don't think you've described what EDRM is, yet. 14 A. I apologise. 15 Q. If you can, in as few a words as possible. 16 A. Certainly. It's the Electronic Disclosure Reference 17 Model, but it's also used for hard copies. It really 18 came in about 20 years ago but it's been refined ever 19 since. It's a series of steps really that you identify, 20 preserve, collect, then process, review, produce. And 21 there's a system of integrated governance so, 22 effectively, you have a -- you operate it as a system 23 and that is the principle. 24 And there's a diagram within the bundle, if it would 25 be helpful to go to it, but, effectively, you're 42 1 approaching disclosure as a system from the start and 2 that is the principle, and then if you've -- inevitably, 3 you don't capture everything at the start, so there are 4 a series of -- the jargon is "feedback loops" but what 5 that means is keeping your eyes open to see if you're 6 spotting things that you missed first time round and so, 7 within the model, it contemplates that you learn things 8 later that you didn't know at the start. 9 Q. Would you agree that we're being provided with documents 10 now, as the Inquiry unfolds and in respect of witnesses 11 who have come and passed through the seat that you 12 currently occupy, because this was not done at the 13 outset of the Inquiry? 14 A. Certainly with Exchange, things are coming to witnesses 15 now because of the Exchange issue was not identified at 16 the outset, that's right. 17 Q. Can I try and summarise Appendix 1 by reference to the 18 relevant periods. I don't want it to be turned up on 19 the screen but can you have it open in your hard copy 20 witness pack. The document on the screen can come down 21 at the moment, thank you. 22 I am going to try and summarise quite a lot of 23 information by reference to some periods. So, firstly, 24 up until the early 2000s, that's Period 1, Royal Mail 25 Group used early versions of Microsoft Mail or MS Mail, 43 1 and the Post Office has little information relating to 2 that period and is not aware of any email repositories 3 from that time? 4 A. That's my understanding, yes. 5 Q. Thank you. Period 2: from the early 2000s, Royal Mail 6 Group started to use Lotus Notes, combined with 7 something called a Lotus Domino server and a sendmail 8 gateway. There was no journalling at a gateway level. 9 People may have kept their own emails locally but very 10 few repositories of emails from that time exist in 11 a structured way, but some emails may exist on 12 SharePoint or on One Drive. 13 A. That's -- 14 Q. That's your understanding? 15 A. That's my understanding. 16 Q. Thank you. In around 2008 or 2009, Royal Mail Group 17 changed from Lotus Notes to Outlook and the email server 18 changed from Domino to Exchange and its gateway from 19 sendmail to IronPort; is that right? 20 A. Again, that's my understanding. 21 Q. Efforts were made to convert Lotus Notes to Microsoft 22 data repositories but there were instances of data loss. 23 After that migration it remained possible for locally 24 archived snapshots of email data to be created and 25 stored but that was discouraged, given the increased use 44 1 of cloud-based systems, such as SharePoint. 2 Some data might have been retained on physical 3 devices like USBs or other similar devices. 4 A. Again, that's my understanding. 5 Q. Then, in around 2012, Royal Mail Group changed from 6 IronPort to Proofpoint, that's its gateway; yes? 7 A. Yes, I believe so. 8 Q. Email journalling was introduced in a manner broadly 9 equivalent to Mimecast, as you describe it. However, 10 the Post Office does not have information as to exactly 11 when Proofpoint was activated and the retention periods 12 or settings that were applied at that period of change. 13 Correct? 14 A. That's my understanding. 15 Q. At that time, it's understood that Proofpoint didn't 16 ingest pre-2012 email data from the Exchange mailfiles 17 into its archive? 18 A. Again, that's my understanding. 19 Q. From 2012 to 2016, last period, some important changes 20 that coincided with the demerger from Royal Mail Group 21 to the Post Office Limited, firstly Post Office adopted 22 Microsoft Exchange/365. 23 A. Yes. 24 Q. It updated, secondly, Outlook and, in around 2015, it 25 adopted Mimecast as its email gateway? 45 1 A. Again, that's my understanding. 2 Q. You say: 3 "This legacy mailfile data, ingested by Mimecast at 4 its activation, came from Royal Mail Group's Proofpoint 5 email gateway in around late 2015." 6 Is that right? 7 A. Again, that's my understanding. This is obviously all 8 reconstructed knowledge -- 9 Q. Yes. 10 A. -- by Post Office. 11 Q. Is it your understanding that, as Proofpoint was only 12 activated in 2012, Mimecast was not expected to contain 13 pre-2012 material? 14 A. Again, that's my understanding. 15 Q. Thank you. So that short description of the three 16 levels of email services and gateways is the kind of 17 thing that is the foundation for a disclosure exercise 18 when one is concerned with emails? 19 A. Yes. 20 Q. It's now, having that foundation, that you're able more 21 accurately to assist the Inquiry, is this right, in 22 providing relevant disclosure to it? 23 A. Yes, and it introduces some practical challenges because 24 Exchange is not a litigation disclosure system. So 25 actually extracting and getting it in a form which 46 1 doesn't burden the Inquiry but gets the Inquiry the 2 correct evidence is challenging but -- so, yes, you have 3 the basis to do it; it presents some operational 4 challenges. 5 MR BEER: Thank you. 6 Sir, that's an appropriate moment. It's 11.20 now, 7 can we take a 15-minute break until 11.35, please? 8 SIR WYN WILLIAMS: Yes, of course. 9 MR BEER: Thank you, sir. 10 (11.19 am) 11 (A short break) 12 (11.35 am) 13 MR BEER: Good morning, sir, can you continue to see and 14 hear us? 15 SIR WYN WILLIAMS: Yes, thank you. 16 MR BEER: Thank you Mr Jackson can we look, please, at 17 page 31 of your statement at paragraph 40. This is 18 dealing with one aspect of the chronology of change in 19 systems that we dealt with before the break. Page 31, 20 paragraph 40. You say: 21 "I understand from Post Office that Mimecast was 22 activated in or around late 2015." 23 We covered that this morning: 24 "Allowing for transition time, there should 25 therefore be a high degree of confidence that any and 47 1 all emails sent or received from early 2016 onwards are 2 held on Mimecast. However, out of an abundance of 3 caution, Post Office is undertaking checks and I will 4 update the Inquiry further if those investigations 5 indicate any systemic issues with Mimecast journalling 6 of emails transmitted from 2016." 7 Have you got an update, please? 8 A. Yes, those checks have been done and continue to be 9 done. The indication is that -- the indications are 10 there is no systemic issue but we continue to -- or we 11 ask KPMG to test that. The indications are that 12 anything -- there are Exchange items that are not 13 replicated in Mimecast post-1 January 2016. To the 14 greater part, they appear to be things that would not 15 travel through the gateway, so you would expect them to 16 be passed -- 17 Q. Such as? 18 A. So personal calendar appointments, notes made on one's 19 own Outlook, that type of item. There are, however -- 20 and this is being checked against number of -- and so 21 draft emails, for example, that never leave the mailbox. 22 There are, however, some -- a very small number of 23 emails showing up for sample checks that we have asked 24 KPMG to look at further and KPMG would wish to look at 25 further. So that is the one remaining issue to be 48 1 tested. 2 Q. So post-1 January 2016 emails that you've obtained, 3 other than from Mimecast, don't appear on Mimecast and 4 you're asking KPMG to find out why? 5 A. Correct, because the working presumption so far has been 6 that, effectively, post-1 January 2016 Mimecast is safe, 7 if we can put it in that way, that it will be a complete 8 repository but, because no one wants to end up in the 9 situation again, we're asking KPMG, and KPMG would wish 10 to, and Post Office would wish to, test that 11 presumption. 12 Q. Where have the emails been obtained from, post-1 January 13 2016, other than Mimecast? 14 A. So the -- they are taken from Exchange because the 15 purpose of the exercise is to effectively compare 16 Exchange to Mimecast to see the extent to which Mimecast 17 is or is not safely picking up everything from Exchange. 18 Q. How many emails are we talking about? 19 A. I don't know. I mean, these are sample checks and 20 I don't know which specific individuals have been looked 21 at. 22 Q. What do you mean they are sample checks? 23 A. Clearly, there are many tens of thousands of individuals 24 who will have had Exchange accounts, of which a number 25 in the hundreds will be relevant to the Inquiry, to 49 1 greater or lesser degrees, and Exchange is not 2 a litigation support system, so it's not very friendly 3 to -- or it's not user-friendly to extract information. 4 So, basically -- it's a sample checking exercise to 5 validate that assumption. 6 Q. What assumption? 7 A. The assumption -- the working assumption here that it is 8 safe to rely on Mimecast post-1 January 2016 as 9 capturing everything that it needs to, in terms of 10 emails. 11 Q. Can I just explore slightly further this idea of 12 a sample? Obviously, it's not everyone's email account, 13 irrespective of whether they had anything to do with 14 Horizon or not. To what extent is it still a sample of 15 emails that are being tested by KPMG, ie not all of the 16 emails that you have discovered from Exchange, which are 17 not replicated on Mimecast. 18 A. Sorry, you'll have to deconstruct that one a little for 19 me, if that's okay. 20 Q. How many emails have you established have not been 21 replicated on Mimecast? 22 A. We haven't because you'd need to work through it. That 23 is not something that is known at the moment, because 24 one would have to decide -- it would be a specific total 25 for each person that you looked at. Each person that 50 1 had an Exchange account. So you'd have to decide within 2 the tens of thousands of users down to the number that 3 are relevant to the Inquiry. So that is not known at 4 the moment. The priority for the -- 5 Q. What is known at the moment about how many emails are 6 not replicated on Mimecast? 7 A. The total is not known. 8 Q. No, I know the actual total that it might be is not 9 known but how many emails at the moment has it been 10 established are not replicated on Mimecast? 11 A. I'm not following. It is just not known because you'd 12 need to look at each of the users. There isn't a -- 13 there isn't a button to press to say compare the whole 14 of Exchange to the whole of Mimecast. 15 Q. Earlier on, Mr Jackson, you said it's not a systemic 16 issue. 17 A. Mm. 18 Q. It's the working basis that you and your client are 19 operating under. 20 A. For -- 21 Q. But there are a few, I think you used, or a small number 22 of emails, and I'm trying to find out what that number 23 is. 24 A. I think for one -- the short answer is I can't remember. 25 It's been tested against one particular user over recent 51 1 days, and it's a low number but I can't remember what 2 the number is. 3 Q. Is that the only user in respect of which a test has 4 taken place? 5 A. I don't know. I would need to check. 6 Q. Bearing in mind that we're moving to Phases 5 and 6 7 soon, which include events which post-date 1 January 8 2016, is there a time frame by which you can say that 9 this exercise will be completed? 10 A. It will have to be a priority. I mean, this -- the 11 short answer is no. But we will -- we can come back to 12 the Inquiry on that. Part of the reason for the time 13 that is taken is because the same resource that does 14 this both at KPMG and at the Post Office is the same 15 technical resource, which is having to prioritise 16 searches for Phase 4 and, indeed, at Post Office, and 17 to -- it's the Cyber Team, so protects all of the 18 business of the Post Office. 19 So there has to be -- not everything can be done at 20 the same time and the priority, to date, has been to get 21 the evidence into Phase 4. So that -- we'll need to 22 come back with a timescale. 23 Q. Thank you. Can we look at a previous working assumption 24 that was, I think, used, by looking at our tab B15, 25 which is POL00165906. This is a letter written by 52 1 Burges Salmon/Fieldfisher on 2 November 2023. It 2 addresses a range of topics and I want to look at that 3 part of the letter which addresses the current issue. 4 If we look at page 3 of the letter, please, and pick up 5 paragraph 16, you say -- when I say "you", your firm 6 said: 7 "Around 2016 Mimecast was introduced and we 8 understand that Proofpoint data was migrated into 9 Mimecast. On that logic there should have been 10 continuity of email data in Mimecast from 2012 onwards." 11 So we're here looking back to a similar point but 12 back in history from 2012 to 2016. 13 A. Mm. 14 Q. "We understand therefore that Mimecast was used as the 15 source for email harvesting for the Inquiry and also for 16 the GLO. We understand that 300+ email accounts have 17 been harvested from Mimecast into the various part of 18 the Relativity database for various phases of the 19 Inquiry." 20 Then paragraph 17: 21 "However, queries by Herbert Smith Freehills 22 relating to the account of Andrew Wise, and in parallel 23 queries at [Burges Salmon/Fieldfisher] in the course of 24 work on the Section 21 notice have led to investigations 25 by Post Office and by/with KPMG. Those have now 53 1 established there are material volumes of email data 2 that are in Microsoft Exchange but that are not in 3 Mimecast (and which therefore have not been available 4 for search). Post Office has not been able to establish 5 why the Proofpoint to Mimecast transfer did not provide 6 the assumed continuity/completeness." 7 That last sentence there, in particular the assumed 8 continuity and completeness, is it your understanding 9 that the Post Office had previously been working on the 10 basis of an assumption? 11 A. That is our understanding, yes. 12 Q. So would it be correct, to your understanding and in the 13 light of the investigations that you and your firm have 14 undertaken, that, in none of the disclosure exercises 15 that have been conducted previously for the Group 16 Litigation, for the CCRC, for the Court of Appeal and in 17 the Inquiry itself, no one actually thought "We should 18 check whether the repository that we're harvesting from, 19 Mimecast, which was only introduced with effect from 20 1 January 2016, is actually a complete repository of 21 emails sent and received before 2015"? 22 A. There was no -- essentially, yes, my understanding is 23 that Exchange data, which was effectively the other 24 source of data that could have been obtained for those 25 various uses, was not looked at because Mimecast was 54 1 felt to be a complete source, not only from its 2 introduction in 2016 but because there had been this 3 import from the earlier gateway system which is 4 Proofpoint. 5 So the understanding, as we understand it, is now 6 going back a number of years -- it's reconstructed 7 knowledge -- is that effectively the understanding was 8 there was a complete set in Proofpoint and that complete 9 set was moved into Mimecast and, therefore, there was 10 a complete set and, therefore, Exchange was not -- no, 11 so that is -- that, I say, is reconstructed -- 12 Q. The reconstructed knowledge, the point I'm asking about, 13 is that that was based on assumption, to your 14 understanding, rather than somebody actually asking the 15 question "We're harvesting from a repository, Mimecast. 16 Can we just check, it was only introduced in 2015, is it 17 actually a complete repository of that which went 18 before?" 19 A. There was knowledge of -- there was an assumption that 20 between Mimecast and the import into Mimecast from 21 Proofpoint that it would provide continuity. That is my 22 understanding. 23 Q. Thank you. Can we turn to when the issue was 24 discovered, by whom and in what circumstances. That's 25 page 40 of your witness statement, please, at 55 1 paragraph 49. That can't be right. 2 I'm looking for paragraph 49. 3 SIR WYN WILLIAMS: I think it is page 40. 4 MR BEER: Yes, thank you. 5 You set out here the two causes of the discovery of 6 the incomplete nature of the data pool that was being 7 used by the Post Office to give disclosure to the 8 Inquiry. 9 A. Yes. 10 Q. The first cause you address in paragraph 49(a): 11 "A series of documents had been provided by the Post 12 Office on 19 May ... as part of a [Freedom of 13 Information] request in May 2023, but the full suite had 14 not been provided to the Inquiry in response to any 15 prior Rule 9 Request." 16 We have investigated this previously, I should say. 17 Those documents were then produced to the Inquiry on 18 30 May. These documents are referred to in some 19 statements of Mr Foat and Mr Rowan. 20 Then, over the page: 21 "The [Freedom of Information Act] request response 22 produced documents that were not found in any Mimecast 23 data. [You] understand from [Herbert Smith Freehills] 24 that the documents were, however, identified by Andrew 25 Wise who, when searching through his Outlook email 56 1 client, located an email (with attachments) that could 2 not be found in Mimecast. That led to an initial query 3 by [Herbert Smith Freehills] of the Post Office on 4 10 July ... as to how Mr Wise had located the email in 5 question which, over time, developed into a query as to 6 whether there was a repository separate to Mimecast." 7 So this was essentially a discovery by chance? 8 A. I think -- my understanding was that the -- HSF were 9 following up on how it was -- 10 Q. I mean, had the Freedom of Information Act 2000 request 11 not been made by the investigator/campaigner, 12 Appendix 6, the racist and archaic identity code 13 document, would not have emerged? 14 A. I'm not aware that it was coming up in any other 15 context. 16 Q. Mr Wise, an individual, produced the document by looking 17 at his own Outlook -- 18 A. Yes. 19 Q. -- and his production of it to the Freedom of 20 Information team revealed that the same document 21 couldn't be found in Mimecast? 22 A. That's my understanding, yes. 23 Q. Then the second cause of the discovery you list in 24 paragraph (b)? 25 A. Yes. 57 1 Q. Essentially, is this a fair summary, a member of your 2 team was seeking to identify documents that were 3 responsive to a Section 21 notice served on the Post 4 Office by the Inquiry on 21 July 2023 for the purposes 5 of Phases 5 and 6 of the Inquiry? 6 A. Yes. 7 Q. They found an email chain and they sought to find the 8 originating email within that chain in Mimecast but 9 couldn't do so? 10 A. Yes. I mean we were at an early stage but we were also 11 aware at the -- the 2(a) and (b) weren't entirely 12 unconnected. We were aware that HSF were following up 13 on the point but, yes, this was a separate issue. We 14 were wondering why the chain was not complete. 15 Q. We asked you to address in your witness statement where 16 responsibility for the disclosure failure lay and you 17 addressed this in paragraph 47(a), so if we just go 18 back, please, to page 39. 47(a) is further down the 19 page. You say: 20 "In relation to responsibility for the issue: 21 "(a) In terms of the responsibility (duty) to 22 address the issue, this rests with Post Office. In 23 terms of delivery, that, operationally, will need to be 24 by advisers by phase." 25 So am I right in thinking that there you're taking 58 1 the issue to mean the fact that emails are missing from 2 Mimecast? 3 A. I tried to take it on two levels in (a) and (b) -- so, 4 yes, this is the -- I understood the question to mean 5 how there were things in Exchange not in Mimecast, yes. 6 Q. You're dealing in (a) with who has responsibility for 7 responding to the identification of the issue? 8 A. Yes. 9 Q. You're saying that will need to be by advisers by phase 10 and, decoding that, does that mean that, if the response 11 relates to Phase 4 issues, that will be for Herbert 12 Smith Freehills; if it's in relation to Phases 5 and 6, 13 that will be for us in Burges Salmon/Fieldfisher? 14 A. Yes, at the time -- it was less specific than that, 15 because, at that stage, in terms of Phases 2 -- so 16 Phases 2 and 3, we were not, and, actually, as we stand 17 at the moment, technically not retained to do Phases 2 18 and 3, we'd agreed, and it's reflected on the Inquiry's 19 website, effectively, to be the point of contact for 20 Phases 2 and 3, but on an assumption that substantive 21 work had been closed. 22 Q. But what about the issue that, in fact, we were asking 23 about: who has responsibility for the disclosure 24 failing? 25 A. That -- 59 1 Q. Rather than who's got responsibility for responding to 2 it? 3 A. I think that's (b) and, insofar as the answer is who -- 4 effectively, it was a paraphrase for who is to blame, 5 quite genuinely, I don't professionally think it would 6 be right to comment. There was two weeks between the 7 question being asked and answering but, also, what we've 8 tried to do throughout is focus on what happened and 9 what can be done about it, rather than why. 10 Why questions or why answers tend to be involved and 11 come up with competing versions. So I don't think it 12 would be right for me to express an opinion on things 13 with which we were not involved many years ago, on the 14 basis of work we've done in a fairly compressed period 15 of time. 16 Q. Can we just look at the facts, then, as to how it was 17 that the issue that's now been discovered was not 18 identifying earlier when searches were carried out over 19 Mimecast. Is the endpoint, as you've described it in 20 the witness statement, this, that the systems were 21 either misunderstood or the existence of separate 22 systems overlooked in the Group Litigation, the 23 post-conviction disclosure exercise and in the Post 24 Office's previous interim disclosure statements to the 25 Inquiry? 60 1 A. Sorry, what's that -- could you take me to that section 2 in my statement, please? 3 Q. Yes, I'm trying to synthesise about 20 pages of what you 4 say as to why it was that this issue was not discovered 5 in the Group Litigation, in the post-conviction 6 disclosure exercise for the purposes of the CCRC and the 7 Court of Appeal and in the four interim disclosure 8 statements lodged in the Inquiry? 9 A. I haven't tried to answer a question as to why because 10 the -- I've aimed, to the best of my ability, as I've 11 described it, to pull the threads together as to what 12 happened factually. The short answer as to why is 13 a much more complex question and I wasn't involved in 14 the time and I don't know. 15 Q. Okay, well, let's truly try and stick to the facts in 16 relation to each of those three stages, Group 17 Litigation, criminal appeals and then the interim 18 disclosure statements to this Inquiry. 19 In relation to Group Litigation, you tell us in your 20 witness statement that it was the Post Office's 21 obligation to file what was called an Electronic 22 Disclosure Questionnaire or an EDQ? 23 A. Yes, that has to be done under the CPR, the Common 24 Procedure Rules of the High Court. 25 Q. So the rules that govern the litigation in some species 61 1 of civil proceedings in the High Court? 2 A. Yes, there's a specific set of rules and a protocol that 3 goes with it. 4 Q. We've got that Electronic Disclosure Questionnaire, the 5 EDQ. Can we just look at that please. It's 6 POL00000657. Is this the Post Office's Electronic 7 Disclosure Questionnaire for the Group Litigation? 8 A. I believe so. As -- 9 Q. It's exhibited to your witness statement saying "This is 10 the Post Office's EDQ for the Group Litigation". 11 A. Sorry, yes, I believe it is. The only reason for 12 hesitating, I was just -- they all look quite similar. 13 I was just checking it was the correct one. Yes. 14 Q. Okay, well, we can have a look at page 13, please -- 15 page 12, please. We can see the date of it just at the 16 foot of the page there, 6 December 2017. 17 A. Yes. 18 Q. Then, if we scroll up, we can see that it's signed by 19 Andrew Parsons, a partner at Womble Bond Dickinson. His 20 signature appears underneath where it says, "GRO" there? 21 A. Yes. 22 Q. It comes with a statement of truth, "Defendant believes" 23 -- that's the Post Office believed the facts stated in 24 the answers to this EDQ are true and that he, 25 Mr Parsons, is duly authorised by the Post Office to 62 1 sign that statement? 2 A. Yes. 3 Q. That's also for a solicitor a significant step, isn't 4 it, signing a statement of truth? 5 A. It is, yes. 6 Q. Because it's the solicitor saying, on behalf of his 7 client, to the court and to the claimants that it, the 8 Post Office, believes that what's in this document is 9 true? 10 A. That is correct, yes. 11 Q. I'm not going to go through this lengthy document and 12 all of the attachments to it because you have kindly 13 analysed it and provided the results of your analysis in 14 your witness statement. Can we look at those, please. 15 It's page 50 of your witness statement, paragraph 60. 16 You say: 17 "Based on Post Office's current understanding of its 18 email systems and repositories [which we've gone through 19 already], unfortunately, it appears that these 20 descriptions in [the document we've just looked at, the 21 Post Office's Electronic Disclosure Questionnaire] in 22 hindsight were not accurate or were oversimplified." 23 Can we go through the five categories of inaccuracy 24 or oversimplification that you set out. The first 25 category of inaccuracy or oversimplification, you say: 63 1 "Although Exchange is mentioned [in the EDQ] in the 2 context of emails and instant messages, it is not 3 identified as a separate server-level source of email 4 data. References [in one of the appendices to the EDQ] 5 appear to equate it [that's Exchange] to 'email software 6 used by employees'", and that's wrong. 7 That would describe the Outlook Email client. 8 A. Correct. 9 Q. So they've muddled up the email client and the gateway? 10 A. Yes. 11 Q. So it's a misdescription of what Exchange was and the 12 function that it performed? 13 A. Yes. 14 Q. Thank you. 15 Then (b): 16 "The statement that Post Office used Lotus Notes 17 until 2012 is understood to be incorrect, current 18 understanding is that Post Office stopped using Lotus 19 Notes email client and Lotus Domino servers and started 20 using the Microsoft Outlook email client and Microsoft 21 BPOS-D servers over the period from 2008 to 2010. For 22 completeness the statement is also incorrect as before 23 Lotus Notes, it is currently understood that Post Office 24 used versions of MSMail ..." 25 So two inaccuracies that speak for themselves: one 64 1 as to the time period over which Lotus Notes was in use; 2 and the second in implying that Lotus Notes was in use 3 at all material times, whereas for some of the period of 4 time, Post Office was using MSMail. 5 A. Correct. 6 Q. Third error: 7 "Consequently, the suggestion that the Post Office 8 introduced Proofpoint email archiving at gateway level 9 in 2012 at the same time that it moved to Microsoft 10 Exchange and Outlook also does not look to match the 11 currently understood timeline." 12 That speaks for itself. 13 A. Yes. 14 Q. The fourth error or inaccuracy is that: 15 "The assertion that all Lotus Notes data would have 16 transferred to Exchange in 2012 is ... incorrect, as 17 well as the period of migration to Exchange occurring 18 between 2008-2010, Post Office's current understanding 19 is that not all old email archives would have been 20 migrated. Only those files associated with active users 21 would have been migrated to Exchange at the time (if at 22 all). It follows the indication that Lotus Notes 23 archived data would be duplicative of Exchange is 24 therefore also not (always/fully) correct ..." 25 Out of the five errors, that's perhaps second in 65 1 terms of significance, would you agree? 2 A. I haven't tried to rank them in order. 3 Q. You haven't applied a judgement as to the seriousness of 4 the inaccuracies in what the High Court was told? 5 A. No, I haven't. 6 Q. Then you say: 7 "Finally, although implied at most, any reading of 8 the [questionnaire], as suggesting that Mimecast (and 9 before it Proofpoint) is a complete repository of: 10 "(i) Outlook emails either in whole or in part from 11 2012 onwards; or 12 "(ii) Lotus Notes emails imported into Exchange or 13 any other emails pre-dating 2012 14 "would not be correct based on current 15 understanding." 16 Would you agree, firstly, that what you said there, 17 as any reading, is a fair reading of the EDQ as a whole? 18 It doesn't make those implications? 19 A. I'm sorry, could you -- 20 Q. Yes. 21 A. -- put that to me again? 22 Q. You say there any reading of the questionnaire as making 23 the following two suggestions would not be correct. I'm 24 asking you: in fact, that is a fair reading, that it 25 does make those suggestions? 66 1 A. Yes. 2 Q. I was going to suggest that maybe that's the most 3 significant error, that implication that you haven't 4 applied qualitative judgement as to the seriousness of 5 the mistakes of what the High Court was told? 6 A. I haven't, no. I've just set them out. 7 Q. If we go on to paragraph 61 of your witness statement, 8 please, on page 52, you say: 9 "The ... reasons for these issues ..." 10 I'm going to call them inaccuracies in information 11 provided to the High Court: 12 "... are not clear or known to me at this time and 13 it would require much more investigation ... to pinpoint 14 the specific cause or causes of the issues." 15 So you haven't, in fact, investigated the causes in 16 the failures in information provided to the High Court. 17 I'm not going to seek to blame you for that. Would you 18 agree the important point is the High Court and the 19 claimants were given inaccurate information? 20 A. With what is known now, yes. 21 Q. Thank you. 22 Can we move on to the criminal proceedings. In 23 paragraph 63, which is on page 53 and following, you 24 tell us about the disclosure exercise undertaken for the 25 purposes of potential appeals against conviction, and 67 1 appeals against conviction, in the Court of Appeal 2 Criminal Division. To summarise, that was undertaken by 3 a firm of solicitors called Peters & Peters -- is that 4 right -- 5 A. That's correct, yes. 6 Q. -- on behalf of the Post Office, and it was called the 7 Post Conviction Disclosure Exercise, or PCDE? 8 A. Yes. 9 Q. One of the things that was done was a series of 10 documents called disclosure management documents were 11 drawn up? 12 A. Yes, I believe so. 13 Q. You give us a number of those as exhibits to your 14 statement. Can we just look at an example, at 15 POL00142261. We'll see it's very different to the 16 document we saw in the civil proceedings. It's written 17 as straight narrative, rather than being responsive to 18 preset questions that are on the form -- 19 A. Yes. 20 Q. -- and are designed by the protocol to the Civil 21 Procedure Rules? 22 A. Yes, I'm not familiar with the protocols in criminal 23 appeals but it is -- I see it's very different. 24 Q. Yes, I think there isn't actually a protocol in the 25 Criminal Procedure Rules that mandates what is in 68 1 a disclosure management document and that's why this is 2 reason as free text narrative. Can we see on the last 3 page, please, which is page 22, and scroll down, we can 4 see that this first disclosure management document is 5 signed off by Peters & Peters on 19 August 2020, yes? 6 A. Yes. 7 Q. We can see it doesn't contain a statement of truth in 8 the same way as the disclosure questionnaire did -- 9 A. Mm. 10 Q. -- in the civil procedure. 11 A. No, it doesn't. 12 Q. Okay. Again, I'm not going to go through all those 13 22 pages or the addenda disclosure management documents, 14 because there were a series of addendums, weren't 15 there -- 16 A. Mm. 17 Q. -- the first and second addendum to this disclosure 18 management document. 19 A. I believe so, yes, and those are the ones that are 20 exhibited, yes. 21 Q. They are. They are exhibited to your witness statement. 22 Again, you've kindly analysed the effect of this 23 disclosure management document and the addenda to it, 24 and set out whether inaccurate statements were made to 25 the appellants in the criminal proceedings and to the 69 1 Court of Appeal Criminal Division. So can we go to 2 page 55 of your witness statement, please. It's 3 paragraph 65 at the top there. You say: 4 "The [Post Conviction Disclosure Exercise Disclosure 5 Management Document, the two addenda to it and the 6 annexes to those documents], reflect Post Office's 7 developed understanding at that time (as it stood) and 8 since the [Group Litigation Order Electronic Disclosure 9 Questionnaire] that Mimecast in fact contained emails 10 after 2012 but not before 2012." 11 So that error has been realised and corrected; is 12 that right? 13 A. Yes, that would appear so. 14 Q. "The explanation captured [in part of the documents] in 15 particular in respect of pre-2012 emails does 16 unfortunately (in hindsight) however continue to reflect 17 some of the looser use of terminology adopted in the 18 [Group Litigation Order Disclosure Questionnaire]. In 19 respect of the date that Post Office stopped using Lotus 20 Notes, it also continues to state incorrectly that it 21 was 2012. Exchange was, as previously, not itself 22 identified to be a separate, available, disclosure 23 repository for email data." 24 So three inaccurate or partial statements made to 25 the court in these series of documents. 70 1 A. The same understanding replicates, yes, through those 2 three documents. 3 Q. You'll agree that the importance -- and you haven't 4 investigated the causes of this -- is, firstly, that 5 these inaccurate statements were made to the appellants 6 and to the Court of Appeal Criminal Division but, 7 perhaps more importantly, the full archive of emails was 8 not accessed and, therefore, not disclosed or provided 9 to the appellants or to the Court of Appeal? 10 A. We -- that's our understanding, that Exchange was not 11 interrogated prior to this current exercise in Phase 4 12 of the Inquiry. 13 Q. Thank you. Can we then move to the third stage, then, 14 what has been said in the past to this Inquiry. That 15 can come down from the screen. Thank you. 16 Summarising, the Post Office has made a series of 17 interim disclosure statements to this Inquiry, correct? 18 A. Correct. 19 Q. Can you confirm that these, the interim disclosure 20 statements, are provided pursuant to the Inquiry's 21 protocols. There will be a final one towards the end of 22 the Inquiry that explains in detail the steps that have 23 been taken to ensure that relevant material has, so far 24 as is possible, been disclosed to the Inquiry to give 25 assurance to the Inquiry and to Core Participants and to 71 1 the public that the Inquiry has been provided, so far as 2 is possible, with all relevant material? 3 A. Er -- 4 Q. That comes at the end. 5 A. Yes, I don't -- we haven't, I believe, discussed it but, 6 certainly, yes, I would anticipate that would happen and 7 that would happen in the normal course, yes. 8 Q. But these interim disclosure statements are stepping 9 stones along the way, would you agree, and are served 10 where there have been issues with the disclosure of one 11 of the Core Participants? 12 A. I don't know because the reasons why they were produced, 13 because we weren't around at that point -- but we've 14 seen what they said, based on the belief at the time, 15 yes. 16 Q. Okay. Reading the four interim disclosure statements, 17 I think you can tell us that the purpose overall is to 18 tell the Inquiry what data is available for disclosure, 19 what approach has been taken to the obtaining of that 20 data and the progress and timescales for the disclosure 21 exercise? 22 A. Yes. 23 Q. Thank you. Can we look at the first one, POL00142261 -- 24 that's a mistake. POL00114170ds, thank you. This the 25 first interim disclosure statement. We can see it was 72 1 dated 27 May 2022, top right, and it was signed by 2 Mr Foat, the Post Office's General Counsel. 3 A. Yes. 4 Q. If we just pick up from page 6 of the disclosure 5 statement and look at paragraph 19 at the bottom, 6 Mr Foat addresses electronic communications, and he 7 says: 8 "[Before] 2012, I understand that [the Post 9 Office's] provider of email servers and software was 10 Lotus Notes. Following the separation, [Post Office] 11 began to use Microsoft Exchange instead of Lotus Notes. 12 At the same time, [Post Office] began to use an email 13 archiving system called Proofpoint. Since the beginning 14 of 2016 [the Post Office] has used Mimecast as its email 15 archiving system. The emails that had previously been 16 stored in Proofpoint were transferred into Mimecast." 17 Then it carries on over paragraphs 20, 21 and 22, 18 further explanations of the Post Office's understanding 19 as to its electronic communications repositories, 20 insofar as their email servers and related software, 21 okay? 22 A. Yes. 23 Q. You, again, have kindly analysed the number of problems 24 with what the Inquiry was told. That's page 57 of your 25 witness statement. Top of the page, you say: 73 1 "However, once again, in hindsight it is unfortunate 2 that the chronology for Post Office moving to Lotus 3 Notes to Exchange and coinciding with email archiving 4 with Proofpoint no longer accords with the understanding 5 of Post Office as I have set out in this statement." 6 That's a long way of saying that you think that what 7 was said was inaccurate? 8 A. Yes, it -- the "however" follows on from -- this was 9 a rolling over, effectively, of the same level of 10 previous understanding that Proofpoint had rolled into 11 Mimecast and, therefore, that was -- in effect, there 12 was continuity and that was the belief at the time, 13 I understand, but I wasn't around. 14 Q. You carry on with a second problem with the disclosure 15 statement: 16 "The matter of pre-2012 emails is not specifically 17 addressed in the First Interim Disclosure Statement save 18 to note that legacy 'E-filing Cabinets' as part of Lotus 19 Notes had formed part of GLO repository searches. 20 However, I note that data repositories where local 21 archived email data ... are now known to be found were 22 referenced in that statement as known repositories of 23 data such as SharePoint and other team drives, file 24 servers, the NAS Drive and laptops. However, Exchange 25 data is not itself identified as a separate repository 74 1 for emails." 2 So depending on how you cut it up, two or three 3 errors in the statement? 4 A. The -- those same errors replicate through that 5 understanding, yes. They reappear here. 6 Q. You tell us in paragraph 68, if we scroll down, and I'm 7 going to summarise this, the second, third and fourth 8 interim disclosure statements did not correct those 9 errors? 10 A. Correct. 11 Q. Thank you. 12 So would it be right to say overall that inaccurate 13 and, to some extent, misleading information in the 14 disclosure questionnaire provided to the High Court was 15 then reprovided to the Court of Appeal Criminal Division 16 and then provided again to this Inquiry? 17 A. The same underlying understanding was replicated in all 18 of those documents that led to the replication of the 19 same inaccuracies, yes. 20 Q. Can we turn, then, to what has been done to address the 21 disclosure failings as a result of the late recovery of 22 the issue. Your statement focuses on Phase 4 23 remediation, and focuses on documents provided by 24 Herbert Smith Freehills and Peters & Peters. Can we 25 turn to remediation of other phases, other than Phase 4 75 1 and can we start, please, with remediation for Phases 2 2 and 3. 3 You tell us -- I don't ask for it to be turned up -- 4 in paragraphs 74(a) and (b) of the Post Office's then 5 thinking on how to remediate Phases 2 and 3, but I think 6 it's right that you sent the Inquiry a letter on Tuesday 7 of this week setting out the Post Office's current 8 position; is that right? 9 A. That's correct. The understanding in the statement was 10 that Phase 2 might not need to be remediated but we were 11 going to test that understanding. We don't -- or we 12 didn't and still don't have a full understanding of 13 Phases 2 and 3 and, therefore, we are working out -- so 14 a sensible approach now for Phase 2 appears to be 15 looking at it on a witness basis, to look at those 16 witnesses that were called by the Inquiry. They're not 17 witnesses with which we, as a firm are familiar but, 18 effectively, the proposal is that we will go back and 19 look at those on behalf of Post Office. 20 Q. Can we just look at the letter that you set out that 21 proposed approach in, POL00333343. The heading "Post 22 Office ... Inquiry: Phases 2 and 3 Exchange 23 Remediation", from Tuesday of this week. We can skip 24 over paragraphs 1.1 and 1.2. 1.4, you say: 25 "We propose that the remediation will be in respect 76 1 of the Phase 2 individuals listed at Appendix A and 2 Phase 3 ... listed at Appendix B." 3 A. Yes. 4 Q. Essentially, they are Post Office witnesses that were 5 called to give live evidence in between October 2022 and 6 December 2022? 7 A. Yes, and that is the proposal for the Inquiry's view to 8 check that it meets expectations. 9 Q. Can you assist us as to why that approach has been 10 designed or proposed? 11 A. In order that those witnesses, the evidence of those 12 witnesses is as full and accurate as it can be for the 13 Inquiry and both the witness and the -- and, in 14 particular, the Inquiry has a chance to see anything 15 that may be relevant. The further back this goes in 16 time, probably the less likelihood of that because of 17 the dates of the Exchange issue but that is the intent. 18 Q. Why would you not apply the search terms from the 19 relevant Rule 9 requests for Phases 2 and 3 to the new 20 server that has been identified as a repository of 21 material -- 22 A. The -- 23 Q. -- ie why would you do it on a custodian basis by 24 focusing on the witnesses who have already been called? 25 A. It is a question of balancing the objectives that I was 77 1 talking about earlier, which is the importance of 2 getting -- or within the terms of the protocol, all 3 reasonable and necessary evidence to the Inquiry, but 4 also the potential impact on further disruption of time 5 because, if one had an infinite amount of time, one 6 could extract several hundred custodians, but Exchange 7 in itself, it's an email system, it's not a litigation 8 system. 9 So litigation systems have the functionality to 10 search. All of the conventional wisdom in all 11 disclosure exercises, really, is not to do litigation 12 search type things in email applications because they're 13 not built for it and they produce surprising results and 14 they're slow and clunky and difficult. 15 So this is an attempt to strike a proper but 16 conservative balance because the initial proposal was 17 that -- for Phase 2 may not be needed, but to take 18 a more conservative approach and identify those things, 19 take them into Relativity, which is a system where you 20 can do all those things, with -- mindful of the 21 Inquiry's strong -- and the Chair's understanding -- 22 very, very strongly expressed and completely 23 understandable view to complete hearings in 2024. So 24 it's attempting to balance those two objectives. 25 Q. And that we've got a month of Phase 4 evidence still to 78 1 undertake and we want to start Phases 5 and 6 as soon as 2 possible? 3 A. Yes, and we've said in the letter our understanding -- 4 but, again, we welcome the case management guidance -- 5 is that Phase 2 and 3 would be effectively returned to 6 at the end, rather than disrupting or delaying Phase 5 7 and 6, but we -- that's our understanding, obviously. 8 It's the tribunal -- it is the Inquiry's and the Chair's 9 decision. We will do it in whichever order is sought. 10 Q. So it's essentially a custodian-based approach, the list 11 of custodians, point 2, is the witnesses who gave 12 evidence in Phase 2 and were Post Office employees or 13 staff at the relevant time, and then the search terms 14 that had already been utilised for the relevant Rule 9 15 requests are going to be run against the products of 16 those custodians? 17 A. Yes, I believe so. 18 Q. Thank you. Remediation for Phases 5 and 6, again, you 19 set out a -- 20 A. And, sorry, could I just say, it would be -- we had put 21 our hands up to do Phases 2 and 3 not anticipating that 22 substantive work would be -- we would be heavily 23 dependent on being able to establish what those search 24 terms were. HSF have given us quite a bit of 25 information but we will still be dependent in that 79 1 knowledge in running that search term -- the exercise. 2 Q. Thank you. Phases 5 and 6. Although you address 3 Phase 5 in your witness statement, paragraph 75 -- we 4 needn't turn it up -- the position has again been 5 updated by reference to a letter sent to us on Tuesday 6 of this week. 7 A. Yes. 8 Q. That is POL00333342. So similar format of letter if we 9 go down, Tuesday of this week, but this is now about 10 remediation of Phase 5. Can I summarise or seek to 11 summarise what you say. Firstly, again, it is proposed 12 to be a custodian-based approach? 13 A. Yes, but this is dealing with going backwards in time to 14 earlier Rule 9s and Section 21s that have -- might 15 concern Phases 5 and 6 because for the -- 16 Q. Just stopping there. You said Phases 5 and 6. This 17 letter only appears to be about Phase 5. We haven't got 18 proposals for Phase 6 yet, I don't think. 19 A. Sorry, yes. To clarify, for the things that have -- for 20 notices of Rule 9s and Section 21s that have come up 21 since the Exchange issue was known, they don't need 22 remediating. 23 Q. Because the right approach was applied to them? 24 A. Because -- yes, because Exchange data was factored in 25 from the outset, so it does not need remediating. 80 1 Subject to our testing and the assumption that 1 January 2 2016 is safe -- and I do want to keep that one in 3 mind -- Phase 6 -- and so Phase 6, insofar as any of 4 those numbers relate to -- and a number of them do -- 5 should not need remediating. 6 So what this is dealing with is looking back at all 7 of the Rule 9s and Section 21s that were served prior to 8 our involvement. 9 Q. Or were processed? 10 A. Were processed -- sorry, that is quite correct -- were 11 processed prior to our involvement and, going back in 12 time, and saying well, to the best of our understanding, 13 to what extent were those relevant and then remediating 14 that. So it is in relation to those categories that we 15 are looking at custodian-based searching, I understand. 16 Q. You propose to do it by reference to a list of people 17 that set out in appendix A? 18 A. Yes. 19 Q. That appendix appears to be constructed from a list of 20 people that the Inquiry sent to the Post Office at its 21 request, identifying those witnesses who might need to 22 have their legal representation funded by the Post 23 Office. 24 A. Yes, several tens of Post Office individuals, yes. 25 Q. As opposed to the Inquiry's witness list? 81 1 A. Yes. I don't -- and I stand to be corrected -- I don't 2 think we've got sight of the latter one, so it's not 3 knowledge that we have and we couldn't design anything 4 around that because we don't have it. 5 Q. So it would -- it is proposed to be a custodian-based 6 and focusing on custodians who are going to be called to 7 give evidence? 8 A. Yes. If there is a wish to look at other individuals 9 we -- of course, we can do that but other than some 10 very, very obvious names -- and for example Mr Jenkins, 11 who in my understanding has already been looked at in 12 the work that HSF has done -- we would be making 13 educated guesses about what those other -- who those 14 other individuals would be. 15 Q. Again, can you help us as to why this approach is being 16 proposed to be adopted by the Post Office, rather than 17 doing what you did after you became instructed, which 18 was simply apply Rule 9 requests and Section 21 notices 19 to the whole dataset by reference to search terms 20 specifically designed to turn over relevant material? 21 It's just an accident in timing, isn't it, that means 22 a different approach is being taken to the past? 23 A. Not quite because the whole dataset is the Exchange 24 dataset, which is not a litigation system. So one can't 25 run search terms against an email system, effectively. 82 1 To run search terms, one has to decide who the 2 custodians are that you're taking into Relativity and 3 then take those in. So a judgment has to be made as to 4 which of those have to come in, and that -- we -- the 5 discussion proposed, and it is very much discussion 6 proposal, was to start with those individuals who we 7 know about so that can be done. 8 With the Rule 9s and Section 21s that we're talking 9 about, where we have done the other alternative, we know 10 what the parameters of that are and, to the best 11 sensible judgement we can make, the individuals that it 12 needs to capture, whereas we don't know that for beyond 13 the -- effectively, the list is -- that you've given us 14 is the best information we have. 15 Q. So why not add the Exchange/365 material to the document 16 universe and rerun Rule 9s and the Section 21s over it? 17 A. Because the Exchange/365 is with the whole of the Post 18 Office, which I don't know but it would be extraction -- 19 Microsoft extraction of Exchange data is a slow and 20 difficult and clunky process, and it -- I'm not -- and 21 this is where I get to the very boundaries of my 22 technical knowledge, but it's quite difficult, even for 23 limited numbers of people, which is one of the drivers 24 for the disruption that I think is being seen at the 25 moment. 83 1 If one was to do it against hundreds or thousands of 2 people, the time and technical implications of that 3 would be very, very significant. So one has to start, 4 I understand, with a judgement as to which custodians. 5 Q. But you could, for example, not just limit it to the 6 people who are being called; you could extend it to the 7 people from whom a witness statement is to be sought? 8 A. We could but we would need -- and, obviously, we would 9 welcome that information, would hold it in confidence 10 and we could work on that basis. 11 Q. It wouldn't include, as you said already, searches for 12 information about Gareth Jenkins, Anne Chambers, Penny 13 Thomas, anyone outside the Post Office who doesn't have 14 a Post Office account? 15 A. One could do a party-based search. My understanding -- 16 but our understanding is that you would have got that 17 from Fujitsu as one of the other -- or whoever -- 18 whichever Core Participant it was that would logically 19 provide that data. Also, my understanding from the 20 correspondence -- I'm not close to it operationally -- 21 is that exercise has been done within POL's email system 22 against Mr Jenkins. But I don't know -- 23 Q. Mr Jenkins and only Mr Jenkins? 24 A. I don't know, it's not -- that's Phase 4 issue and we 25 haven't been doing those searches. 84 1 Q. Okay. Finally on this topic then, what about Phase 6; 2 when are we going to hear about the proposals for 3 Phase 6? 4 A. Well, Phase 6, as I say, hopefully does not need 5 remediating because Exchange data is being factored in 6 to -- is within -- is being searched on any new requests 7 that are coming in and including the Phase 6 related 8 requests we've already had. 9 Q. You said "hopefully" in that sentence, which may get 10 hackles up. 11 A. There was no intent -- 12 Q. I'd prefer not to base things on hope. 13 A. My full understanding is that we don't need to remediate 14 Phase 6. All it was intended to signify is that we will 15 keep testing that understanding. 16 Q. Thank you. Can we move to the second topic, please, 17 Mr Bradshaw. 18 I just want to look at what happened in relation to 19 the disclosure of documents concerning Stephen Bradshaw 20 who, of course, gave evidence yesterday. 21 I'm doing so, I should say, to obtain your evidence 22 as to: why disclosure is still currently being given in 23 relation to our witnesses with significant volumes of 24 documents being disclosed right up to when the witness 25 gives evidence; why a significant proportion of the 85 1 documents that are being disclosed are internally 2 duplicative so the so-called new material that we are 3 getting contains two, five, ten or more documents that 4 are materially the same but it being described as "new 5 material" being found on Microsoft Exchange/365; and why 6 it's duplicative of material that we've already got, 7 again described as "new material", but it's another copy 8 or another two, five, ten copies of a document that the 9 Post Office has already given us 10, 20, 30 or sometimes 10 50 times in the past. 11 So I've gone through witnesses and I look on our 12 system and find that I'm given ten copies of documents 13 that are materially the same. I look on the system and 14 see we've already been given it 50 times previously by 15 the Post Office, so it's to explore why that is still 16 happening. 17 Now, Mr Bradshaw was due to be giving evidence back 18 in November last year and on 4 November last year his 19 evidence was pulled from the timetable because of the 20 disclosure of what was said to be substantial new 21 documents, very shortly before he was due to give 22 evidence. 23 Can we look, please, at what has happened since that 24 time, and start, please, with INQ00002016, a letter from 25 the Inquiry dated 14 December headed "Continued late 86 1 disclosure by [the Post Office]: Phase 4 hearings and 2 witnesses". 3 If we skip over paragraphs 1 to 4 and go straight to 4 5: 5 "[Post Office] must disclose any additional 6 documents by the following deadlines: 7 "(a) Any ... documents said to be related to Stephen 8 Bradshaw by 4.00 pm on 20 December ..." 9 All other persons in an annex by 4.00 pm on 10 10 January. Okay? So Inquiry deadline 20 December 2023 11 at 4.00 pm. 12 Move forwarded, please, to POL00329552. So the 13 Inquiry's letter was dated 14 December with a deadline 14 of 20 December. This came in on 19 December -- I think 15 this is one of those late at night ones -- from Herbert 16 Smith Freehills: 17 "We enclose a note prepared by Peters & Peters ... 18 This note refers to the Inquiry's letter [of] 14 19 December ..." 20 If we go over the page, please, Peters & Peters are 21 assisting the Post Office. The note has been provided 22 by Peters & Peters, it rehearses the deadline. 23 Paragraph 3: 24 "As confirmed in Herbert Smith Freehills' letter to 25 the Inquiry date 24 November [that should be '2023'], 87 1 the current exercise of reviewing and producing material 2 relating to Stephen Bradshaw is complete." 3 Paragraph 4 is about something else. 4 Paragraph 5 and paragraph 6 are about something 5 else. So Post Office saying on 19 December, the day 6 before the deadline: disclosure is complete. 7 Over to POL00329553. 27 December we're at now. If 8 we scroll down: 9 "We enclose a note prepared by Peters & Peters. 10 This ... refers to the Inquiry's letter [of] 11 14 December." 12 Over the page, please. Paragraph 3 -- sorry, 13 paragraphs 1 and 2 set out the background. Then we've 14 got an update. So we've been told that the exercise was 15 complete but we've now got an update for Mr Bradshaw: 16 "On 19 December ... [Herbert Smith Freehills] wrote 17 to the Inquiry enclosing a note prepared by Peters & 18 Peters explaining that the current exercise of reviewing 19 and producing material relating to Stephen Bradshaw had 20 finished. However, there are two further matters 21 relating to Stephen Bradshaw's data that [the Post 22 Office] wishes to bring to the Inquiry's attention." 23 First one, I'm going to skip over, is audio taped 24 transcripts. Then if we go down to paragraph 7. We are 25 now told: 88 1 "Material relating to Stephen Bradshaw is being 2 considered by Peters & Peters for a Crown Office and 3 Procurator Fiscal Service disclosure exercise owing to 4 his involvement in one of the Scottish cases ... Since 5 [the 19 December letter] the Peters & Peters [Procurator 6 Fiscal Service Team] has begun some additional checks to 7 ensure that all Stephen Bradshaw material has been 8 caught. From the investigations into the MS Exchange 9 365 issues, potential additional material addresses and 10 display names have been identified ... and checks are 11 underway to ensure that material has already been 12 captured." 13 Not quite as complete as the 19 December letter 14 said. 15 INQ00002017, please. Email at the top of the page. 16 So the letter we've just looked at and the note from 17 Herbert Smith Freehills and Peters & Peters respectively 18 was 27 December. This is a reply from the Inquiry on 19 27 December: 20 "We note the update position regarding potential new 21 documents for Mr Bradshaw. As Mr Bradshaw is scheduled 22 to give evidence on 11 January ... please provide all 23 outstanding material ... by no later than ... Friday, 24 29 December ..." 25 POL00329554, 29 December, so deadline day, second 89 1 deadline day. Scroll down, please. We can ignore 2 paragraph 1 which narrates the history. Paragraph 2: 3 "Post Office is conducting ongoing disclosure 4 related exercises ..." 5 Paragraph 3: 6 "Searches have been conducted across all ongoing 7 review exercises to locate documents which refer to 8 Mr Bradshaw, and/or which were sent to or received by 9 Mr Bradshaw ... the searches were not limited to reviews 10 specifically relating to Mr Bradshaw ... Based on these 11 searches, KPMG have identified 21 documents which refer 12 to Mr Bradshaw, copies of which are enclosed with this 13 letter." 14 Over the page: 15 "Due to various pre-production checks and other 16 steps that are required prior to upload to Egress 17 [that's a disclosure system], a formal production onto 18 Egress will not be possible until the week [commencing] 19 2 January ..." 20 Therefore they are informally producing the 21 21 documents relating to Mr Bradshaw. 22 Then 7, a warning: 23 "As ever, it is possible that documents relating to 24 Mr Bradshaw may be identified through other review 25 exercises, including reviews which have not yet been 90 1 commenced." 2 POL00329555. I think that's just an email on 3 a Friday evening at 9.34 pm attaching the transcripts. 4 So we can skip to INQ00002018, inquiry email of the 5 New Year, Wednesday, 3 January, remembering that the 6 Post Office said that it was going to produce formally 7 the 21 documents in the week commencing 2 January: 8 "Please ... provide an update on the formal 9 production by ... no later than end of day ..." 10 POL00329556, email from Herbert Smith to Inquiry 11 solicitors at 11.09 pm on the Wednesday night, 12 remembering a request had been made for the 21 documents 13 to be provided by end of day. Second paragraph: 14 "... all [21] documents are on track to be formally 15 produced to the Inquiry this week ..." 16 So not by the end of 3 January as requested by the 17 Inquiry. 18 POL00329558. Next day, Thursday, 4 January, 19 a letter from Herbert Smith to the Inquiry. Again, 20 paragraphs 1, 2 and 3 narrate the background. Then 21 paragraph 4, "Review exercises": 22 "Peters & Peters are assisting [the Post Office] 23 with its disclosure in relation to the Criminal Case 24 Studies ..." 25 Over the page, paragraphs 5, 6 and 7 are not that 91 1 relevant but paragraph 8, in the third line: 2 "[The Post Office] has also produced 72 additional 3 documents on the basis that they are either: 4 "(a) documents which appear to be within the same 5 families of 9 of the 21 documents", that were going to 6 be informally produced, or children of those relevant 7 families. 8 The tranche upload is 93 documents today. 9 Then paragraph 16, over the page, please: 10 "Of the 93 documents being produced today, KPMG have 11 informed us that 74 of them are exact duplicates of 12 previously produced documents ..." 13 Just stopping there, why does the Post Office 14 produce exact duplicates of documents previously 15 produced? 16 A. In that -- it's not a letter I know about. I don't 17 know, in this context. 18 Q. "... and a further 7 [of them] are near duplicates." 19 So 12 are new, okay? If we carry on, please, just 20 to complete this before lunch. 5 January, we're now up 21 to Friday, POL00329560. 22 Herbert Smiths saying: 23 "We enclose a note prepared by Peters & Peters ..." 24 So they don't do what they've done previously, which 25 is adopt some of the information into their own letter. 92 1 They're just saying, "Here is a Peters & Peters note". 2 Over the page, please. 5 January, this note is 3 dated, we can see at the foot of the page. Top of the 4 page, please: 5 "In its email dated 27 December ... the Inquiry 6 requested that all outstanding material relating to 7 Stephen Bradshaw be provided by ... 4.00 pm on 8 29 December ..." 9 In fact, as you'll remember from the correspondence, 10 that itself is an extension. We had directed that any 11 additional documents be disclosed by 20 December. 12 Paragraph 2: 13 "In our notes ... we informed the Inquiry that 14 additional checks were being undertaken ... Those 15 additional checks have now been completed and [the Post 16 Office] has identified 942 documents ..." 17 So this is the Friday of the week before Mr Bradshaw 18 gives his evidence: 19 "... 942 documents ... that fall to be produced as 20 a result of review work. [Post Office] has today 21 arranged for these documents to be produced ... via 22 Egress ... 23 "POL is endeavouring to do its utmost to identify 24 duplicative material to the Inquiry. Of the 942 25 documents, 420 have been identified as being MD5# 93 1 duplicates or 98-100% textual near duplicates of 2 documents previously produced to the Inquiry." 3 Again, do you know why the Post Office was saying 4 "We've got 942 documents but you've had 420 of them 5 already"? 6 A. I don't no. 7 Q. "... Peters & Peters believes a substantial number of 8 the remaining 522 documents ... also appear to be 9 [textual near duplicates]." 10 Paragraph 4: 11 "Relevant information from KPMG's analysis is 12 provided in the production index ..." 13 Paragraph 6: 14 "[The Post Office] apologises that it has been 15 unable to identify fully all duplicates in time for this 16 production. [It's] endeavouring to provide an updated 17 analysis ... as soon as possible." 18 Do you know why, appreciating this isn't the work of 19 your firm, either because it was undertaken by Peters & 20 Peters and/or Herbert Smith Freehills, why the Inquiry 21 had been informed on 19 December that disclosure 22 relating to Stephen Bradshaw was complete and then the 23 week before he was due to give evidence, late on 24 a Friday, we were told that there were 942 additional 25 documents? 94 1 A. I don't know, I'm sorry. 2 Q. Would you agree that is suboptimal? 3 A. Yes. 4 MR BEER: On that note, can we break for lunch, please. 5 SIR WYN WILLIAMS: Now, that I've successfully unmuted 6 myself, the answer is yes. 7 MR BEER: 2.00, please, sir. Just to help you, sir, I only 8 have about 20 minutes more and there are no questions to 9 be asked by Core Participants. 10 SIR WYN WILLIAMS: I'm sure Mr Jackson will be glad to hear 11 that and I'm glad that you're asking the questions, not 12 me, Mr Beer. 13 MR BEER: Thank you, sir. 14 (12.57 pm) 15 (The Short Adjournment) 16 (2.00 pm) 17 MR BEER: Good afternoon, sir, can you see and hear us? 18 SIR WYN WILLIAMS: Yes, I can, thank you. 19 MR BEER: Thank you very much. 20 Just before we start -- and good afternoon, 21 Mr Jackson -- two points of correction or clarification. 22 Firstly, I think I suggested that an email of 23 19 December from Herbert Smith Freehills was sent in the 24 evening. It was, in fact, sent at about 9.00 am in the 25 morning. It was one of the early ones. 95 1 Secondly, I suggested that we hadn't heard proposals 2 from you on remediation of Phase 6 searches. I am told 3 that, in fact, although there wasn't a letter, there was 4 an email -- 5 A. Yes. 6 Q. -- from an associate at your firm saying essentially 7 what you told us today, that it's not currently assessed 8 by Post Office, that that remediation is necessary 9 because of the timing and content of the Rule 9 and 10 Section 21 notices relating to Phase 6? 11 A. Yes, I was aware of that exchange between our 12 colleagues, yes. 13 Q. Thank you very much. 14 Just to complete the Stephen Bradshaw run of 15 correspondence, I think we had reached the Friday before 16 he was due to give evidence. Can we look, 17 penultimately, please, at POL00333345. This is an email 18 sent on Wednesday this week at 11.00 in the morning from 19 Herbert Smith Freehills to the Inquiry: 20 "Peters & Peters are assisting [the Post Office] 21 with the review of documents relating to Stephen 22 Bradshaw [so this is the day before he gives evidence]. 23 "We refer to the note prepared by Peters & Peters, 24 which was provided to the Inquiry on Friday [that's 25 where we left off] regarding the 942 documents produced 96 1 to the Inquiry that same day. In that note, 2 Peters & Peters explained that as part of [the Post 3 Office's] efforts to identify duplicative material to 4 the Inquiry, KPMG have run a duplicate analysis over 5 those 942 documents. Of the 942 documents, KPMG 6 identified 420 as being MD5# duplicates or 98-100% 7 textual near duplicates (TNDs) of documents previously 8 produced to the Inquiry. However, Peters & Peters 9 believed that a substantial number of the remaining 522 10 'new' documents ... also appearing to be [textual near 11 duplicates] of documents previously produced to the 12 Inquiry. These included [record of taped interviews] 13 and investigation reports [of Question 42 of Rule 9(14) 14 cases] which have been produced to the Inquiry on 15 multiple occasions. 16 "KPMG has therefore investigated why those documents 17 were not identified as part of its original duplicates 18 analysis ... conclusion was these documents were not 19 flagged as [textual near duplicates] because they fell 20 below the requested textual similarity threshold of 98%. 21 KPMG explained that whilst documents Peters & Peters had 22 highlighted do appear to be textually similar, small 23 variations may have cause the documents to fall below 24 the 98% similarity requirement. KPMG therefore 25 broadened its ... analysis from 98-99% to 90-99%. The 97 1 results of that broader ... analysis indicated 578 2 documents in [the] production were duplicates of 3 documents previously produced to the Inquiry. 4 "KPMG's most recent analysis indicates that Friday's 5 production contains 364 'new' documents. This remains 6 contrary to the feedback from Peters & Peters review 7 team, who have indicated that a very large volume of the 8 material from Fridays production was duplicative of 9 documents previously produced to the Inquiry. Peters & 10 Peters has therefore begun a further manual review of 11 the 364 'new' documents to determine whether on their 12 face they appear similar to material previously produced 13 to the Inquiry. It is apparent from the preliminary 14 findings of that manual review there is still [textual 15 near duplicates] of previously produced material within 16 the 364 'new' documents. For example, Peters & Peters 17 has seen multiple [Post Office] legal memos regarding 18 Hughie Thomas' court hearings, Hughie Thomas' mediation 19 application and mediation report, and multiple copies of 20 Hughie Thomas' audit report, all of which have [been] 21 produced to the Inquiry previously. [Post Office] has 22 therefore asked KPMG to investigate these examples 23 urgently to understand why these documents have not been 24 caught ... In parallel, Peters & Peters is conducting 25 a full manual review of the 364 'new' documents from 98 1 Friday's production. 2 "Following its further investigation, KPMG has been 3 able to confirm that the additional examples were not 4 identified as [textual near duplicates] as the 5 duplicates previously produced to the Inquiry were 6 produced by another firm and transferred to KPMG. [Post 7 Office] understands that these duplicates contain 'junk' 8 characters at the end of the extracted text file, 9 causing them to have significant textual [variations] to 10 the documents in Friday's Bradshaw-related production 11 ... even though it appears that several of the items 12 within the 'new' 364 documents ... have been produced to 13 the Inquiry as many as five times before, they have not 14 been picked up in KPMG's duplicate analysis. This may 15 explain why KPMG's analysis does not correspond with 16 Peters & Peters' view that the 'new' documents contain 17 numerous duplicates. In addition, [the Post Office] 18 understands from KPMG that [textual near duplication] 19 analysis is often not effective when analysing Excel 20 files ... this might further explain why potential 21 duplicates have not been picked up ... 22 "As [the Post Office's] further investigations will 23 take some time to complete, [the Post Office] arranged 24 for a new overlay to Friday's production preparing by 25 KPMG to be uploaded to Egress so that the Inquiry can 99 1 see which documents have been identified as duplicates 2 ... However, [the Post Office] will endeavour to update 3 the Inquiry on the results ..." 4 So this is all before a witness who has been 5 adjourned for over a month to give evidence was due to 6 give evidence? 7 A. Yes. 8 Q. Your firm is not responsible for any of this? 9 A. No. 10 Q. But you're here to speak, I think, to the Post Office's 11 disclosure to the Inquiry, including in Phases 4 and 12 following? 13 A. I can't speak as to Phase 4. Prior to 1 December, when 14 I submitted my witness statement, the exchange that we 15 had with the Inquiry was said that I would endeavour to 16 draw together the threads, I think, as I put it, but 17 I would set out what I could answer to and that which 18 I could not, and the Inquiry would effectively come back 19 to us if you wanted to hear from anyone else. 20 Now, obviously, in timing, this is recent so that 21 could not have been picked up but I can't answer to 22 operational detail of Phase 4, not because I wouldn't 23 wish to, but simply because I don't have the knowledge 24 to do that. 25 Q. The last document in the train of documents is a note 100 1 received, either whilst Mr Bradshaw was giving evidence 2 or after he gave evidence yesterday, from Peters & 3 Peters which said that the additional checks that were 4 undertaken in respect of Mr Bradshaw's Mimecast data 5 were as following, and that the Post Office understands 6 from Peters & Peters that none of the potentially new 7 documents produced to the Inquiry on 5 January contained 8 any materially new information. Do you know anything 9 about those two things? 10 A. No, the only thing I'm aware of in that -- in relation 11 to that note was I understand that, in an attempt to 12 drive some of the numbers of duplicates or near 13 duplicates down, a degree of controlled item level 14 deduplication was done after the production stage, so at 15 a sort of safe stage of it. That's the only thing I'm 16 aware of because KPMG updated me on it. 17 I could -- 18 Q. I'm interested, in particular, in the suggestion that 19 this was material obtained from Mimecast, whereas this 20 should all be about Exchange/365? 21 A. I simply don't know, I'm sorry. 22 Q. Mimecast is the thing that ought to have been 23 interrogated months if not years ago? 24 A. Mimecast has been used as the main basis for searches, 25 yes. 101 1 Q. Yes, and this note suggests that the additional checks 2 were undertaken in respect of his Mimecast, 3 Mr Bradshaw's Mimecast data. Do you know why anyone was 4 looking again at Mimecast and coming up with 942 5 documents last Friday? 6 A. I don't, I'm afraid. 7 Q. Would you agree that, just looking at the 8 correspondence, reading the correspondence, that it 9 doesn't make for happy reading in that the Inquiry was 10 told back on 19 December that Bradshaw disclosure was 11 complete and then it was said that 924 documents, 12 including lots of duplicates were to be produced last 13 Friday? 14 A. Absolutely. It goes back to the point, really, that you 15 took me to, I think in my initial witness statement, is 16 nobody would wish to continue to have anything other 17 than a sort of more conventionally controlled and 18 managed and critical path approach to hearings and -- 19 Q. The email that we've just read about the involvement of 20 another company providing services to KPMG, and that 21 being the cause of an underestimation of the number of 22 duplicate documents, is that anything you know about? 23 A. I know what I think it is but I would qualify this 24 because I am speaking from what I would understand that 25 to be, not because I know but because I'm reading it is, 102 1 and why I'm hopeful that, certainly the issues around 2 duplicates, whilst they are still difficult, will 3 improve going into Phase 5 and -- 4 If I may take it in two parts, I think what is being 5 referred to, but I do not know, is the point that Paul 6 Tombleson made when he gave evidence on the 5th -- and 7 I touch on in my witness statement -- which is the 8 current Relativity database has been built up 9 effectively in layers on data that has come in from 10 various sources at various places in different ways. So 11 I think what is being talked about is not somebody who 12 is a contractor to KPMG, but a previous provider to Post 13 Office. 14 So the -- this is one of the reasons why some of the 15 techniques cannot be used, the computer techniques that 16 were talked about, because the data quality is variable. 17 The reason I am hopeful, but I don't want to 18 overstate it because I don't know fully, about Phase 5, 19 is that there has been a -- we, for Phase 5, 6, 7 are 20 working in a new -- in a different bit of the database, 21 which has had the data reprocessed into it and, 22 therefore, does not suffer from the same, or at least 23 not so many of, these problems. 24 I mean, duplicates and near duplicates, for all the 25 reasons in 20 pages of my witness statement, are a real 103 1 conundrum but that is exacerbated if the underlying data 2 from which they're drawn is of variable quality from 3 various places at various times. 4 Q. That conundrum is compounded if you get a series of 5 letters and emails in the day before and the three days 6 before, giving different figures, and assertions as to 7 what is a duplicate and what isn't, and explanations as 8 to whether documents are truly new or not; would you 9 agree? I mean, reading it, would it appear that this 10 information, this run of correspondence, is rather 11 chaotic? 12 A. I don't know what lies behind it and, therefore, I'm 13 loath to step in from outside and criticise because 14 I simply don't know. But it's -- I think you used the 15 phrase before the lunch break "suboptimal", and it is 16 clearly that and it must be frustrating deeply for the 17 Inquiry and for witnesses, and I suspect also for those 18 at the other end who are trying their best to get it 19 right, but I simply don't know beyond that. 20 Q. So that takes me to what assurances in relation to 21 Phase 5 and 6 you're able to give, that the kind of -- 22 sorry, that document can come down -- that the kind of 23 episode that we've seen played out in the Bradshaw 24 correspondence and I think, if I can say it with 25 a degree of understatement, he's not the only witness in 104 1 respect of whom it has happened and, for a period, it 2 was every witness, every day of the week, twice on 3 Sundays. 4 What assurances can you give that this will not 5 happen in relation to Phase 5 and 6 witnesses? 6 A. I can give some assurances and I'll set out, if I may, 7 what I think they are. 8 Q. Yes, thank you. 9 A. I'll start with a qualification, which is deduplication 10 and complex disclosure is ferociously difficult and 11 there will be glitches and we will find things because 12 that is the nature of it. But the assurance I can give 13 is that we will use every bit of our professional skill, 14 and I know that Post Office is absolutely of this mind 15 also, to reduce that to the absolute minimum that we 16 can. 17 The other parts of the structural assurance are, as 18 I say, we have the benefit of data that has been, to 19 a large extent, reprocessed in terms of the material 20 that is already in there and, as a separate and further 21 point, the Inquiry will be aware that we've written, on 22 a number of -- in a number of letters, we've been sort 23 of exercising our minds as how one can square the circle 24 between potentially excluding something which is 25 a material difference -- so the example I give is if -- 105 1 so that you don't want to texturally near duplicate out 2 something that changes a "shall" to a "shall not", for 3 example, but, equally, one does not want to be burdened, 4 the Inquiry or witnesses, by things which just have 5 minor differences in the footer, or so on. 6 So we've been giving thought to that, and have 7 written to the Inquiry about additional load files, 8 ie extra data which can be used to sift into wheat and 9 chaff and then circle back round to look at a particular 10 document if one -- if the Inquiry wants to look at the 11 details of that. 12 I started with a caveat and, with apology, I have to 13 finish with this answer with a further one, which is 14 Exchange is an email system, and therefore the 15 transference of data is slow and difficult and clunky, 16 and we will do -- and I believe KPMG and Post Office 17 will do all they can to eliminate that. 18 The timescales are -- there is a challenge between 19 the timescales because it isn't just obviously about the 20 start of the hearings, it is about the lead-up to those 21 hearings. All involved which to get to a smooth 22 procession to that. I wish I could give an unqualified 23 assurance but the sheer practicalities of complex 24 disclosure, duplicates and Exchange mean that I have to 25 stay somewhere short of that. 106 1 Q. In paragraph 88 of your witness statement, which is on 2 page 77 -- if that can be displayed, please -- in the 3 third line you say: 4 "Post Office started in June 2023 the exercise to 5 identify the full extent of Post Office's electronic 6 data universe." 7 Why was it that it was only in June 2023 that Post 8 Office started the exercise to identify the full extent 9 of its electronic data? 10 A. I don't know. 11 Q. That's a couple of years into the Inquiry, isn't it? 12 A. It is, yes. 13 Q. Are the problems we're now facing in part caused by the 14 Post Office only starting to identify the extent of its 15 electronic data in June 2023? 16 A. I've commented on this on paragraph 84. I wonder if it 17 will be possible just to turn back? 18 Q. Absolutely. That's the previous page, page 76. 19 A. Yes. It picks up on -- the previous paragraph picks up 20 on the work that we'd initiated in June and July but 21 I say, and it is absolutely the position, it would not 22 be fair, or fair inference, to sort of view the 23 structural review as an indication that things hadn't 24 been done before. 25 There had been -- we weren't around at the time but 107 1 from what we have since learnt -- quite a bit of 2 investigative work and -- to find out what there was and 3 where it was and so to do that work. It hadn't been 4 done on a single overarching model but a lot of work had 5 been done and you will see quite a few of the answers to 6 that work in the first, second, third and fourth 7 disclosure statements. 8 Q. So I think you're answering to say you shouldn't think 9 that no work had been done to identify the extent of the 10 Post Office's data universe, correct? 11 A. Correct. A very large amount of work had been done, but 12 it hadn't been -- it had been done by various people at 13 various times, in various ways, using various exercises, 14 rather than it being done as a comprehensive (unclear) 15 exercise, sir. 16 Q. Which takes us to the third topic, the structural 17 review. This is part of the structural review? 18 A. I'm sorry, what is part of the structural review? 19 Q. Identifying the full extent of the electronic data 20 universe? 21 A. Yes, the model that we talked about earlier is 22 effectively you identify, you preserve and you 23 collect -- or you collect as you need. 24 Q. You describe in paragraph 82, if we go back, please, to 25 page 74, the elements of the structural review. You say 108 1 it's not directed at a single phase or multiple phases, 2 and the main activities -- and you divide them into 3 three: 4 An exercise to identify the full extent of the data 5 universe to validate assessments of all data sources as 6 to whether they are reasonably likely to contain data 7 that might be relevant to the Inquiry's terms of 8 reference, whether any such data has been preserved and, 9 as appropriate, collected and processed for review. 10 Secondly, the structural review involves 11 a consolidation of Relativity workspaces to reduce time 12 and operational complexity when responding to requests 13 from the Inquiry. 14 Thirdly, an exercise to validate custodian data 15 mapping to facilitate an assessment of whether further 16 identification work is required. 17 Can you explain what that is because it's not 18 obvious on its face: "custodian data mapping"? 19 A. Essentially, who would have held what type of data, 20 amongst those individuals or custodians who are of 21 likely relevance or known relevance to the Inquiry. 22 Q. So identify people or job positions, job titles, who are 23 assessed to hold data? 24 A. I'm hesitating because I -- colleagues are involved in 25 the exact detail. I don't know whether it is named 109 1 people or whether it is job titles but I can -- we can 2 update further. 3 Q. You set out timelines in appendix 3 to this document, 4 your witness statement. 5 A. Yes. 6 Q. But they have essentially been superseded by the third 7 of the Tuesday letters; is that right? 8 A. That's correct. 9 Q. So can we go, as our last task, to the third of the 10 Tuesday letters, POL00333344. I'm going to ignore the 11 first three paragraphs and go straight into the content. 12 You split the letter into three strands, essentially. 13 Strand one, ESI, that's electronically stored 14 information on electronic media? 15 A. Yes. 16 Q. Topic 2: electronically stored information excluding 17 emedia? 18 A. Yes. 19 Q. Then, thirdly, hard copy documents. So can we take 20 those three in turn? 21 A. Yes. 22 Q. Go back to page 1, please, "ESI stored on emedia", and 23 I think this has got four elements to it, each of which 24 is underneath an underlined heading? 25 A. Yes. 110 1 Q. The first of which, element 1 or strand 1 "Investigation 2 of five servers and twelve back-up tapes located at 3 Chesterfield". I think it was identified by the Post 4 Office that it had servers and back-up servers in 5 Chesterfield and this was revealed to the Inquiry in 6 November, I think. You set out in this letter the 7 position, server by server. 8 A. Yes. 9 Q. Scanning across paragraphs 1.1 and 1.2, is the position 10 that three servers have been excluded, two remained 11 further to be investigated, and the 12 back-up tape 12 servers have been restored in recent days? 13 A. I'm not sure about the last point. Yes, the three 14 servers have been excluded. All of these categories in 15 this letter, for context, are -- or many or most of them 16 are what would one would ordinarily say is at the far 17 end of what would normally be done in terms of any 18 search in a litigation case or inquiry, for good reason, 19 given the circumstances that you outlined earlier that 20 then all of the corners are being searched, and this is 21 the furthest corners, if you like. 22 And so some of these servers, as we're seeing are -- 23 well, one doesn't know what they are when they start 24 being examined but they've sat in a dusty corner for 25 many years. 111 1 Q. So if we go over the page to paragraph 1.12, so three 2 servers excluded, two left, 12 -- sorry, if we just go 3 up, please -- 12 back-up tapes. The reason I said 12 4 back-up tapes have in recent days been restored is 5 because that's what it says in paragraph 1.11. Then 6 1.12 is where it's all brought to a head. You say: 7 "We will continue to update the Inquiry on Post 8 Office's work to examine [the two servers] and the 9 twelve [back-up] tapes." 10 Then this: 11 "... progressing those investigations may take 12 a material period." 13 Which is a delightfully enigmatic phrase. Can you 14 tell us what it means? 15 A. Yes. It means that some very technical people who can 16 decode things which may be five or ten or 15 years old 17 and, therefore, old technologies which are no longer 18 supported, and those are -- obviously back-up tapes are 19 things which are run on a server so you don't lose the 20 data if the server fails, and they often overwrite 21 themselves or they duplicate because, if you're taking 22 a record at a particular point in time, then that point 23 in time will also contain material from earlier. 24 So the reason I hesitated was, yes, they had been 25 restored but they haven't been examined. The technical 112 1 people who can do this sort of thing now have to take 2 them into another dark room and do something with them 3 and it's not knowable in advance how long that takes, 4 and it's unfortunately. 5 Q. So how do the Inquiry plan for the commencement of the 6 Phases 5 and 6? 7 A. This effectively comes back to the issue of case 8 management and the point in the Inquiry's protocol about 9 fair and reasonable in all the circumstances. This 10 is -- there's effectively two routes -- well, three, 11 I suspect. One is to take a view that these things are 12 unlikely to be relevant because they are at the extreme 13 ends of relevance -- 14 Q. How -- 15 A. -- and take a view -- 16 Q. On what actual material could a view be formed that 17 these two servers and 12 back-up tapes are in the very 18 corner of the corner that you described? 19 A. No, it can only -- that can only -- that has to be 20 assumption or presumption, rather than anything else. 21 One doesn't know until one looks. 22 Q. On what building blocks might you draw the inference 23 that you have just outlined, ie on what firm factual 24 foundation might a person be able to say, "I'm willing 25 to commence Phases 5 and 6 or list Phases 5 and 6 113 1 without any insight into when the material from these 14 2 data sources is likely to be available to me"? 3 A. Well, if I may come back to the other two options as 4 well, because they were going to tie back into this 5 point as well. If it is possible, and I don't know 6 technically whether it is, the starting point logically 7 will be to look at the -- as the first -- before one 8 looks at the content of the data, to look at the dates 9 of it and then extrapolate from that whether it is 10 likely to be of relevance to Phases 5 and 6, because 11 there is obviously -- 5 and 6 does go back in time but 12 not so far as, say, 5 or -- 2 or 3. 13 And then, secondly, if one can get an insight into 14 the type of data, if -- let's say, for example, it was 15 financing or accounting data, as opposed to documents or 16 other evidential material that the Inquiry would tend 17 more to focus on, then that would inform you, but the 18 basis we don't -- 19 The Inquiry does not have and Post Office does not 20 have to give to the Inquiry, those initial evidential 21 basic blocks but the three options would be to 22 effectively take an assumption when knowledge builds, 23 the second, which is the more -- I was setting out the 24 two extremes -- is to effectively take a long pause 25 until everything would be held to be within the fair and 114 1 reasonable search and look at everything but that would 2 be -- involve a delay, or -- 3 Q. You wouldn't know how long the period was? 4 A. You wouldn't -- 5 Q. You'd have to say, "I'm going to adjourn for a material 6 period"? 7 A. It would be a material period and that would be the 8 second option. 9 The third would be to list but effectively engage in 10 a case management discussion as to the prioritisation of 11 witnesses and types of information, which would itself 12 be informed by trying to find some early stage 13 information about type of data, dates of data, and so 14 on. But those are the three, as I would see them. 15 Q. The existence of this material was revealed to the 16 Inquiry in November. How long have the Post Office 17 known about it? 18 A. I don't know. 19 Q. Can we turn to element 2, please, "Confirmation of 20 understanding relating to the NAS Drive data". Can you 21 explain in a crisp sentence or two what the NAS Drive 22 data is? 23 A. The NAS Drive data was, as I understand it, a capture of 24 a particular set of data that was known about at the 25 time or just after the GLO and prior to the 2020 work. 115 1 So it was a snapshot of a very large amount of data at 2 a point in time. What this section -- sorry, you're 3 probably going to come on to ask me about this next 4 section but the NAS Drive, effectively, was a data 5 capture at a point in time which, from memory, was 2017 6 or 2018. 7 Q. You had, you tell us in paragraph 1.14, intended to 8 update us before today, and you tell us in 1.15 that the 9 Post Office is seeking to establish how long 10 investigations are going to take. I take it today you 11 don't know how long the investigations are going to 12 take? 13 A. I'm afraid not. This is -- the NAS Drive is something 14 that was captured and is known. What this section 15 relates to is a -- I think it's a server or system 16 called either Accenture FileShare or, more accurately, 17 Post Office FileShare and there had been a couple of 18 previous letters to the Inquiry. 19 The first one said this has just emerged and we 20 think the Post Office FileShare is something you need to 21 know about but we're investigating. 22 There was a second letter that said emerging 23 conclusions are that it is just duplicative of NAS Drive 24 but we need to make very, very sure and it has not, as 25 I understand it, been possible, so far, to make very 116 1 sure that there was not new things going onto the Post 2 Office FileShare. 3 Q. So you're not able, the Post Office is not able to tell 4 us how long the investigations will take? Are you able 5 to tell us when you're going to be able to tell us how 6 long the investigations will take? 7 A. Because these are technical not legal issues, I cannot. 8 There is pressure, as you would understand, being 9 applied by Post Office and by those who -- we represent 10 them, to get it completed as soon as possible with but 11 they're technical investigations. 12 Q. Bottom of the page, "Validation", that's the third 13 element. I'm going to skip over that because there is 14 a timescale included. Over the page, please, to element 15 4 of strand 1, "Review of custodian disclosure 16 questionnaires to establish whether further connection 17 of eMedia is required". 18 Then paragraph 1.18: 19 "[The] Post Office wants to ensure the Inquiry is 20 aware of decisions Post Office has made relating to the 21 collection of data ... Post Office has not identified 22 (after investigations of users, including questionnaires 23 that have asked custodians to provide details of their 24 WhatsApp usage) repositories of WhatsApp messages 25 reasonably anticipated to be responsive as substantive 117 1 evidence to the Inquiry's Terms of Reference and 2 Completed List of Issues, so has not collected [from 3 their devices]." 4 Can you just explain what that means, that very long 5 sentence? 6 A. Yes. 7 Q. We haven't asked people to provide their WhatsApp usage 8 and we haven't looked on their phones; is that it? 9 A. Because -- we have tested whether it would be needed to 10 be done by seeing what -- by asking them questions as to 11 what they -- do they use WhatsApp and what do they use 12 it for? And the response is, as I understand it, come 13 back effectively that it's administrative, what dates 14 are this and what -- so not the situation, as we 15 understand it, based on the questionnaires, where people 16 will be having substantive discussions of the kind that 17 are being canvassed in other inquiries at the moment -- 18 Q. So -- 19 A. -- as we understand. But if that changes or we have 20 a reason to believe it changes, then we would revert. 21 Q. So if Paula Vennells was intending to attend a meeting 22 and was going to discuss with Angela van den Bogerd 23 beforehand what to say and what not to say, she wouldn't 24 have used, on your understanding, WhatsApp do so? 25 A. Based on the information we have gathered, no, she 118 1 wouldn't. As we understand it, there may have been 2 liaison in terms of who is free on particular -- what 3 date, but -- and, as we understand it, not 4 substantively. 5 Q. How wide or how big is the pool of people that you have 6 asked? 7 A. I think it relates to all the key custodians but I will 8 verify that and we will write. 9 Q. What does the key custodians mean? 10 A. Those who have been named in -- the short answer is I 11 don't know which individuals we're talking about but 12 certainly all those would have been named and are still 13 current employees. 14 Q. Ah, so there's a limitation that the person remains in 15 Post Office? 16 A. I will -- I don't know, so I will check and we will come 17 back to you. 18 Q. It may be a surprise to a member of the public that 19 nobody in the Post Office used WhatsApp to discuss 20 issues of substance relating to the Horizon system? 21 A. That is our understanding but we are keeping it under 22 test. 23 Q. Strand 2, "ESI excluding emedia". This is from 24 paragraph 2 onwards and, again, there are a number 25 elements to the strand: 119 1 "Identification of the extent of [the] data 2 universe." 3 At 2.3, you tell us it's been commenced, and you 4 told us in your witness statement that was back in June. 5 It's materially progressed but is ongoing; several 6 further weeks to complete this work. Does that mean by 7 the end of the month? 8 A. Again, I don't know, because finding the information is, 9 it's easier to find some information than others. So 10 I don't know. 11 Q. Over the page, please, Element 2. We've addressed that 12 already. 13 Element 3, "Validation of historic preservation 14 activity across other ESI data sources". Can you 15 explain what that is, please? This is paragraph 2.11 to 16 2.15. 17 A. So this is looking back, so the previous section deals 18 with what is currently in place. This is looking back 19 in time as to what was in place previously in the GLO 20 litigation, or otherwise. We haven't been able to 21 establish so far what technical holds were in place. We 22 understand and we've seen what was in the interim 23 disclosure statements and that we're told various holds 24 in 2014, 2016 and 2020, but the position, because of the 25 situation in paragraph 2.14, is difficult. 120 1 The people who did it are often not around and the 2 records are not, therefore, available. So we're still 3 looking. 4 Q. The next element, I think, Element 4 within Strand 2, 5 "Review of [electronically stored information] received 6 in the past from third parties to establish whether 7 further collection ... is required". I think if we go 8 over the page to paragraph 2.19, in the second sentence, 9 you tell us: 10 "It's anticipated that material work will continue 11 beyond January 2024 [and] will continue to update the 12 Inquiry." 13 Can you assist us as to what volume of the work will 14 have been completed by the end of this month and what 15 will remain outstanding? 16 A. To put it in context, this is going back to all -- 17 particularly the third party legal advisers but some 18 other advisers in particular -- to say you were asked 19 previously to provide all key advices, because that is 20 the things that are under control, but, as part of the 21 structural review, it is not always clear -- it is not 22 clear to us as incoming advisers what was provided. So 23 we've said what effectively looking at what have you -- 24 what have you given us, what have you not given us, and 25 then testing what we have been given. 121 1 So we're working steadily through the different 2 advices. I don't know precisely where we've got to. 3 Obviously, it depends on when they come back to us, as 4 well, and when it was asked. It's all of these things, 5 we're trying to retrofit an understanding, and it's 6 an ongoing process. 7 Q. "Hard Copy Documents" is Strand 3. Can you help us: 8 overall, what is outstanding? 9 A. Can I just look at the paragraph? These are exercises 10 that we didn't and don't have conduct of. So we're 11 report -- effectively, this is the work done by Innovo. 12 As I understand it, but I am taking this from the Innovo 13 witness statements that were submitted to the Inquiry, 14 their work on hard copy has been completed, as has the 15 re-indexing. 16 Q. The third task is outstanding, which is looking back at 17 questionnaires to see whether hard copy documents 18 require to be collected from other repositories. 19 A. And, again, it's not something that we are doing, but we 20 are told that it will be completed by the end of the 21 month. 22 MR BEER: Thank you very much. 23 Mr Jackson, they are the only questions I ask you. 24 Thank you very much. 25 THE WITNESS: Thank you. 122 1 MR BEER: Sir, there are no questions from Core 2 Participants, in particular because we said that all 3 questioning of Mr Jackson, as in previous disclosure 4 hearings, would be conducted by Counsel to the Inquiry. 5 SIR WYN WILLIAMS: Yes. 6 Well, first of all, Mr Jackson, thank you very much 7 for your detailed written evidence and thank you very 8 much for the clarity of the answers you've given to 9 Mr Beer during the course of the day. 10 As is my custom following hearings of this type, 11 I don't make instant announcements, that being fraught 12 with danger. Rather, I propose to reflect upon what's 13 been said and anything that I regard as important to say 14 I will say in the course of the coming days or weeks in 15 writing. Any further directions which I propose to give 16 relating to disclosure will be contained within that 17 same written statement. 18 Clearly, I want to reflect upon the state of affairs 19 which you describe where there are options, if I can put 20 it in that way, as to how best to proceed. 21 So that's as much as I propose to say today. 22 That brings us to the end of today's proceedings, 23 does it not, Mr Beer? 24 MR BEER: It does. We're back at 10.00 am on Tuesday, sir. 25 SIR WYN WILLIAMS: All right. Well, we'll adjourn until 123 1 then. 2 MR BEER: Thank you very much, sir. 3 (2.46 pm) 4 (The hearing adjourned until 10.00 am 5 on Tuesday, 16 January 2024) 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 124 I N D E X CHRISTOPHER MICHAEL JACKSON (sworn) ...........1 Questioned by MR BEER .........................1 125