1 Wednesday, 17 May 2023 2 (10.30 am) 3 MR BLAKE: Good morning, sir. 4 SIR WYN WILLIAMS: Good morning. 5 MR BLAKE: Sir, this morning we have Mr Finch and 6 Mr Coleman. 7 SIR WYN WILLIAMS: Yes. I think before we call some 8 evidence, I propose to say something about a sad death 9 which has occurred quite recently. 10 MR BLAKE: Certainly. 11 SIR WYN WILLIAMS: As I've already indicated, it's my sad 12 duty to announce that on 10 May -- so last Wednesday -- 13 Mrs Veronica Maye passed away in hospital. Mrs Maye was 14 the wife of a former subpostmaster, Mr Francis Maye, who 15 is a Core Participant in the Inquiry represented by 16 Howe+Co and, as I understand it, one of their clients 17 who has taken a very keen interest in the work of the 18 Inquiry. 19 Mrs Maye was aged 67, she had suffered ill health 20 for some years. Despite that, she had provided very 21 significant support to Mr Maye in relation to his claims 22 for compensation, both in relation to the Group 23 Litigation and in his communications with his solicitors 24 at this Inquiry. 25 On behalf of the whole Inquiry team and on my own 1 1 behalf, I extend our deepest sympathies to Mr Maye and 2 family and friends of Mrs Maye. As I've indicated, 3 Mr Maye was a claimant in the Group Litigation and he 4 has made a claim for further compensation in the Group 5 Litigation scheme. My understanding of the current 6 position is that he has received an interim payment but 7 that at least some of that payment was immediately paid 8 over to his trustee in bankruptcy. Thank you Mr Blake. 9 Over to you. 10 MR BLAKE: Thank you very much. Sir, Mr Finch will be 11 giving evidence first. His evidence will be quite brief 12 but we will take taking a short break in order to 13 arrange for Mr Coleman to come into the room and to make 14 other arrangements. Thank you, sir. 15 I am going to begin by calling Mr Finch. 16 ALVIN EDWIN FINCH (sworn) 17 Questioned by MR BLAKE 18 MR BLAKE: Thank you. Can you give your full name, please. 19 A. Alvin Edwin Finch. 20 Q. Mr Finch, you should have in front of you a witness 21 statement dated 10 March 2023. 22 A. Yes. 23 Q. If I could ask you to look at that statement and turn to 24 the final page, page 6 of 6, is that your signature? 25 A. That is. 2 1 Q. Can you confirm that that statement is true to the best 2 of your knowledge and belief? 3 A. It is. 4 Q. Thank you Mr Finch. As I said, your evidence is going 5 to be quite brief today. Thank you very much for 6 attending the Inquiry to give that evidence. 7 By way of background, you joined Fujitsu's SSC in 8 about 2004; is that right? 9 A. Yes. 10 Q. You only worked there for a few months? 11 A. Yes. 12 Q. What was the nature of your job over those few months? 13 A. Well, I was in training to go forward and do further 14 support work there. So I was going round different 15 sections, picking up different details from different 16 parts of the system, whether it be hardware or software. 17 Q. In those few months, what do you recall of the approach 18 taken to fixing bugs, errors and defects in Horizon? 19 A. Well, I didn't get to actually fixing anything myself 20 but some of the system struck me as odd, in that when 21 the hardware came in it was so locked out they had to 22 use hacking tools to get into it, and there was one 23 particular fellow who was going through some of the 24 overnight comms stuff where it seemed like where these 25 overnight comms were happening he got some tools. 3 1 Unfortunately, I fell out with this fellow and I'm 2 trying to remember the rest of it. 3 I mean, at the time, the way I was doing -- or the 4 way he got these things organised to fix comms problems 5 reminded me a bit of my time at Marconi with paper tape, 6 where you're fixing a parity problem rather than fixing 7 the actual problem that was there. But it kind of felt 8 like that and people were rushing to get things done 9 within the system. 10 Q. If I could just stop you there. 11 A. It was awkward. 12 Q. When you refer to "overnight comms" is that phone calls 13 from the -- 14 A. Well, in -- the way the system was communicating across 15 the network. I'm very hazy about what particular bit 16 was there. I remember there was a couple of different 17 computer systems there. There was error reports coming 18 up where they had to be kind of sorted out afterwards. 19 But I'm afraid I can't -- 20 Q. That's fine. The analogy you have used is putting 21 pieces of paper over a problem. 22 A. Well, once upon a time, when I was at Marconi, 23 everything was paper tape and the -- when you read the 24 paper tape through the system, occasionally it would 25 stall and you get a junior programme and go "Oh, just 4 1 a parity error again", punch a hole in there to put the 2 parity right. That was wrong. You need to go back to 3 see what the character could be there to get the 4 character back. You could fix the symptom but not the 5 problem. 6 Q. Thank you very much. I am just going to take you to two 7 passages in your witness statement and see if you are 8 able to expand on that. 9 Could I ask you, you do speak quite quickly, if 10 you are able to slow down for the purpose of the 11 transcriber that would be very helpful. 12 A. Sorry. 13 Q. Thank you. 14 It's paragraph 29 of your witness statement. That 15 is WITN08060100. It's page 5 of that document. It's 16 been brought up on screen. You say: 17 "I have been asked if I ever felt under pressure 18 to avoid finding bugs, errors and defects in the Horizon 19 IT software. It certainly seemed like that. The 20 approach was to keep everything going rather than 21 reporting back. There seemed to be pressure to get 22 a fix in and keep going." 23 Are you able to expand on that at all? 24 A. Well, certainly there was some sort of, like, rah-rah 25 meetings, where we were told about the huge pressure 5 1 that was on Fujitsu to keep this going and the thousands 2 or millions that could be lost if we weren't keeping the 3 job up to scratch and that we needed to be maybe working 4 all hours to keep the thing going and avoid any kind of 5 penalty clauses. 6 Q. We heard from Mik Peach yesterday and he said that the 7 generation of code fixes wasn't visible to somebody at 8 your level. Do you have any comments on that at all? 9 A. I didn't really see any code that would be dealt with 10 elsewhere. I was looking at the general overall 11 workings of the system, system functions, rather than 12 actual code. 13 Q. Would you be able to comment on longer term plans from 14 Fujitsu to correct bugs, errors and defects? 15 A. No. I was only there a few months. 16 Q. From the position that you were in, it was your view 17 that there was pressure to get a fix and keep going? 18 A. Yes. 19 Q. Moving on to paragraph 30, just below, you say: 20 "I have been asked whether any pressure was placed 21 upon me or colleagues not to pass information to Post 22 Office in relation to potential bugs, errors and defects 23 within the Horizon IT System. I don't know, but the 24 protocol was that we kept it confidential within the 25 system." 6 1 Can you help us with what you mean there by "the 2 protocol"? 3 A. Well, it felt aspect bit like the Official Secrets Act, 4 where you don't pass anything on to -- say anything to 5 any customer or mention anything to anybody within the 6 Post Office, not that I would, at that time, come into 7 contact with anybody in the Post Office but not to 8 communicate any sort of anything inside the company to 9 any Post Office employees. 10 Q. Where was that coming from? Was it a culture? Was it 11 an individual? 12 A. I would say it was a culture, really. 13 Q. Did anyone ever say anything to you in that respect or 14 was it just a feeling that you had? 15 A. Well, there's one guy in particular that I fell out 16 with. That was pressure from him, in particular. 17 I think possibly he was very pressured himself, so ... 18 Q. Do you remember his name at all? 19 A. No. 20 Q. What did he tell you about keeping things confidential? 21 A. Just basically what it says there, that nothing goes out 22 of the building. 23 Q. Can you tell us why you left the SSC? 24 A. There was one particular person I felt that -- I had 25 sort of arguments with him about how things were fixed 7 1 in a way or -- I say "arguments" -- discussions. There 2 was possibly a personality clash as well. 3 The whole -- the culture just didn't feel right to 4 me. Some of the people I went round, who were fixing 5 different bits of the system, seemed to me a little 6 like -- a little bit like technological dinosaurs. 7 I mean, the rest of the world was kind of emerging into 8 new systems and we were back -- I mean, that was back in 9 X.25. Everything then was moving forwards to TCP/IP, 10 et cetera. I don't know. It just -- it didn't seem to 11 be a feeling of any innovation going on somehow. 12 MR BLAKE: Thank you, Mr Finch. I said I will be brief. 13 Those are all my questions. I don't believe that 14 anybody has any other questions, so thank you very much 15 for coming to give evidence. 16 SIR WYN WILLIAMS: Mr Finch, I too want to thank you for 17 making a witness statement and for coming to give 18 evidence. It was short and sweet but, nonetheless, it 19 touched upon matters which we are considering with care. 20 Thank you. 21 A. Thank you. 22 MR BLAKE: Thank you, sir. Can we take a ten-minute break, 23 and then we'll bring Mr Coleman in. 24 SIR WYN WILLIAMS: All right. So that's 10.50. Fine. 25 MR BLAKE: Thank you very much. 8 1 (10.41 am) 2 (A short break) 3 (10.51 am) 4 MR BLAKE: Thank you, chair. Can I please call Mr Coleman. 5 SIR WYN WILLIAMS: Yes, of course. 6 RICHARD COLEMAN (affirmed) 7 Questioned by MR BLAKE 8 MR BLAKE: Thank you. Can you give your full name, please? 9 A. Richard Ian Coleman. 10 Q. Thank you very much, Mr Coleman. You should have in 11 front of you a witness statement dated 16 March 2023? 12 A. I do, yes. 13 Q. Can I ask you to have a look at the final page, page 11. 14 A. Yes. 15 Q. Can you confirm that that's your signature? 16 A. It is, yes. 17 Q. Can you confirm that that statement is true to the best 18 of your knowledge and belief? 19 A. It is, yes. 20 Q. Thank you very much, Mr Coleman, for coming to give your 21 evidence today. I'm going to start by asking you 22 a little bit about your background. You joined ICL in 23 1990; is that right? 24 A. Yes. 25 Q. And you worked as a hardware engineer until June 1998? 9 1 A. Correct, yes. 2 Q. Then you transferred to the SSC and worked there until 3 2005; is that right? 4 A. Yes. 5 Q. Thank you. You worked under Mik Peach, who we heard 6 from yesterday; is that right? 7 A. Yes. 8 Q. Then after leaving Fujitsu you trained to become 9 a Minister of Religion in the Church of England? 10 A. I did. 11 Q. That's the role that you currently have? 12 A. It is, yes. 13 Q. I'm going to ask you about your role in the SSC. Can 14 you briefly tell us what that role involved. 15 A. Just a technician. So calls would come in from 16 postmasters and other sort of systems that we had. So 17 calls would be raised. I would then sort of investigate 18 those and then, if there was a software error, send that 19 on to development for them to sort of fix and then -- 20 yes, so my role was sort of gathering the evidence 21 required to determine that and then to sort of try and 22 fix it. 23 Q. Did you have a particular area of focus? 24 A. Yes. There were two databases to do with the 25 configuration of the Post Office and the counters. So 10 1 ACDB and OCMS I think it was. So they were my 2 particular areas of responsibility. 3 Q. Barbara Longley's evidence was different people had 4 different interests; is that right? 5 A. Yes. So one of the things that Mik wanted was to sort 6 of have people who had particular responsibilities with 7 different areas, different systems that we had and then 8 for -- so you would become the sort of expert on that 9 particular area and then it was down to you to sort of 10 spread that knowledge within the SSC, so that everybody 11 could at least handle any call that came in to the SSC. 12 Q. Were the formal ways of doing that spreading of 13 knowledge or was it more informal? 14 A. I mean, I suppose the formal ways would be we would be 15 required to write documentation for the SSC to use. So, 16 I mean, I can recall writing stuff -- documentation on 17 the ACDB, for example, detailing how it worked, what to 18 do if we couldn't use the automated systems. So the 19 ACDB would generate various files overnight at different 20 times and they would then be processed by other systems. 21 So if we couldn't use the automated systems, it would be 22 down to us to sort of create those files manually for 23 whatever reason and for them to be then processed by 24 other systems as required. 25 So there was formal documentation in that regard 11 1 but it was also a case of just mentoring other people 2 within the SSC on those systems. So part of my role, 3 there was a daily -- there were daily -- so I had 4 a daily job that I had to do each morning, checking the 5 output of the ACDB and OCMS to see whether there was any 6 errors and, if there were, to sort of then sort those 7 out. So I trained, I don't know, a handful of people on 8 being able to do that role as well, so when I'm on 9 holiday or sick, or whatever, they could then take over. 10 Q. I want to look at one part of your witness statement 11 that's WITN06470100. Thank you. Can we turn to page 7, 12 paragraph 22. About halfway down that paragraph you 13 say: 14 "I do not recall being involved in the 15 investigation of calls to do with the branch accounts as 16 there were others, such as Anne Chambers and John 17 Simpkins, who tended to handle those types of calls." 18 Can you tell us why they were chosen or why, in 19 your view, they were the ones who were handling those 20 types of calls? 21 A. I think Anne joined after I had but John had been there 22 for a number of years before I joined. So he was one of 23 the people that sort of I would go to. So if I had 24 something that I didn't quite understand, wasn't sure 25 what was going on, John has one of the people that 12 1 I would have gone to for, you know, "What do you think", 2 kind of thing. We'd have a conversation about that. 3 Anne seemed to just get into this sort of 4 EPOSS-type calls. So that would, again -- so she -- 5 whether Mik had given her that responsibility as sort of 6 that would be her area of expertise or not, I don't 7 know, but she would be one of the people that, yes, 8 again, if I got an EPOSS call, it would be, yes, 9 probably Anne or perhaps Diane, she was another one, 10 that I would have gone to for that. 11 Q. Can you give Diane's full name? 12 A. Diane Rowe. 13 Q. Another name that we will come across in due course is 14 Gareth Jenkins. Can you tell us what kind of issues you 15 may have discussed with Gareth Jenkins? 16 A. I'm not aware of discussing anything with Gareth. 17 I think I simply understood that he was just part of the 18 development team. 19 Q. So when you see yourself and Mr Jenkins on a log, on 20 a PinICL or a PEAK, for example, you wouldn't have had 21 direct discussions. That would just be entries on the 22 log, would it? 23 A. Yes, as far as I can recall. I don't recall ever 24 speaking to Gareth personally about an issue, so ... 25 Q. What did you understand about his particular expertise? 13 1 A. I didn't. I just thought he was just part of the 2 development team. 3 Q. Thank you. Can we look at POL00029012, please. This is 4 a witness statement from the Bates and Others litigation 5 and it's page 13 I'd just like to ask you briefly about. 6 There are a few topics that I'm going to take you to. 7 They are just miscellaneous topics in order for you to 8 assist the Inquiry with its understanding of your role. 9 It's paragraph 47. We have there a reference to 10 "support tools" that are used: 11 "... to filter information and present information 12 to technicians in ways that make the support process 13 easier." 14 There's a reference to a Smiley support tool and 15 another tool, which it is said that you were involved 16 in. Can you briefly tell us what those two different 17 tools were aimed at doing and your involvement in them? 18 A. I mean, the tool that I wrote was called "SSC FAD INFO" 19 and John and I had obviously had -- about the same time 20 had the thought of, oh, it would be useful to have some 21 sort of graphical application that we could use to 22 extract information from the various systems and present 23 it sort of in a single window, you know, which would 24 obviously help us with diagnosing. 25 So there was a lot of overlap between our two 14 1 programs because we developed them at the same time, 2 unaware that the other person was doing so. Yes, and, 3 I think, as Steve says, my tool was ultimately -- what 4 it did that was perhaps different from John's and 5 I can't remember what those things were but that was 6 then subsumed into John's program. So it was just 7 a way -- because normally we would access the 8 information that we needed on the various systems 9 through a command prompt. So you're having to type long 10 command lines in. So, obviously, having a Windows 11 application on your computer made it a lot easier to see 12 that information all together. 13 Q. What information was it that you were seeing, using your 14 tool? 15 A. Whatever we could -- well, whatever we felt was useful 16 to us within the SSC. So there were various databases 17 that held information and so our programs would just 18 construct an SQL query to extract that information and 19 then present it on the screen. So part of that would be 20 messages stores as well. 21 Q. I'm going to take you to another document. This one is 22 going to be slightly out of order. It's FUJ00039673. 23 I wonder if you can assist us with this because I think 24 it may be related to the tool that you built or it may 25 not be. It's a PinICL and there's a reconciliation 15 1 issue. 2 If we have a look, please, at page 3 it's a very 3 early PinICL, I should say. It's 1999, so before the 4 national rollout. If you look at page 3, about halfway 5 down, it has your name and a large number of entries 6 that say, "New evidence added" and gives FAD codes. 7 A. Yes. 8 Q. Then if we keep on scrolling down it's page 8 near the 9 bottom. It has a reference to "evidence deleted" and 10 has FAD codes and all of the entries after that for the 11 entire page say "evidence deleted". Then over the page, 12 to page 10, at the top it says: 13 "Emailed John Newitt with regard to freeing disk 14 space." 15 Are you able to assist, is that linked to the 16 tool? Is it something else? Are you able to put it in 17 as simple terms as possible what the issue is there? 18 A. Yes. So I would probably have used my SSC FAD INFO 19 program to extract the message stores for all of those 20 FAD codes listed. They would be compressed into a zip 21 file and then I would simply have added them onto the 22 call, as you saw on page 4, I think. 23 Q. Where it says "deleted", "evidence deleted", can you 24 tell us what that means and is that anything we should 25 be concerned about? 16 1 A. If you scroll back a page, back to page 8, 15.40, I've 2 put an entry saying that once closure has been agreed 3 then we will delete those files. I don't recall who 4 John Newitt was but I think that the only reason we 5 wouldn't have kept those, that we deleted them, was that 6 they would take up an awful lot of space, even as 7 compressed zip files, so hence my note there about so he 8 can free up the disk space on his server. So, as I say, 9 I don't know what server that would have been. 10 But we wouldn't need to keep those and I suspect 11 probably it's taking up space within the PinICL system 12 and, as I say, they would be large files and we wouldn't 13 need to actually keep them with the PinICL call because 14 if we needed to go back to those FAD codes and get that 15 evidence again then we just go back to the FAD code and 16 extract it from the message store. 17 Q. So typically what would it be that was being deleted 18 from the PinICL here? 19 A. It would be the zip files that we were -- so, yes, the 20 zip file of the message store that I would have attached 21 as evidence for development, and then to give the 22 information to MSU, I think, by the looks of it, for 23 them to let POCL know whatever they needed to know about 24 those transactions. So, as it says then at 11.50, with 25 John Moran "Okay to close as per Martin Box from POCL". 17 1 So once the reconciliation has been done we don't need 2 to keep that evidence with the actual PinICL itself 3 because it's just taking up dead space on that. 4 Q. If I was looking at this some way down the line, would 5 that now hinder my ability to understand what's going 6 on? 7 A. No, because you just go back to the message store and 8 extract the messages again. 9 Q. So it hasn't deleted any messages? All it's done is 10 remove them from the PinICL and you need to see them 11 both together to probably understand it. 12 A. Yes, the messages would be untouched on the 13 correspondent servers. 14 Q. Thank you. I'm going to take you to a few different 15 PinICLs. I'm going to start with FUJ00032293, please. 16 This is again an early PinICL. It's from 1999, 17 November 1999, so before the full national rollout. If 18 we look at the third entry, there's a customer call. 19 He's been experiencing a lot of problems with the 20 system. It has there: 21 "Advice: PM thinks this definitely a system 22 problem and would like it investigated." 23 It's A further piece of advice that isn't 24 highlighted. It's three rows down. Thank you very 25 much. 18 1 If we go over the page, we have Barbara Longley 2 there at 13.19.32, saying that this is an EPOSS desktop 3 issue; is that correct? 4 A. Yes. Yes, she's added the product EPOSS & DeskTop, yes. 5 Q. Then you become involved. Why would you become involved 6 at this stage? 7 A. We had a sort of -- there was an admin role that SSC 8 people did. Barbara wasn't technical, she was just -- 9 I wouldn't say just an administrator but she was the 10 administrator. She wasn't technical. So we had this 11 role that each person in the SSC would do. So we had 12 a rota. So each day one of us would do what we call 13 a pre-scan. So we would take a look at the call as -- 14 once Barbara's done her admin on that and then we could 15 do a bit more admin because we had a sort of technical 16 understanding. 17 In one of these PinICLs, Diane Rowe, as 18 a pre-scanner sends her call back with insufficient 19 evidence. So she's obviously had a quick look and gone 20 "We haven't got enough evidence, I can see that straight 21 away, so send it straight back". So that's what I'm 22 doing here. 23 Then -- because originally when I did my witness 24 statement, originally I thought we did the pre-scan once 25 Barbara had retired but, clearly, that's obviously not 19 1 the case and I thought -- because I thought it was 2 Barbara who would then assign the calls to the various 3 team members but, clearly, I'm doing that as the 4 pre-scan. 5 Q. So a pre-scan would involve somebody in the team that 6 had better technical knowledge that Barbara Longley; is 7 that right? 8 A. Yes, an administrative kind of role but using your 9 technical expertise. 10 Q. This subpostmaster has called in experiencing problems 11 and considers that it's a system problem. 12 A. Yes. 13 Q. Your entry here is "Defect cause updated to 40: 14 General -- User". Can you assist us with what that 15 meant? 16 A. Yes, when I looked at this call when it was sent to me, 17 I noticed that and thought "Now, why have I done that", 18 not why have I set it to user but why have I set it at 19 all? Because a defect call is -- you can only determine 20 once you've done your investigation and I haven't done 21 investigation on this. I've assigned it to Mike and 22 he's do the investigation. 23 So the only thing I can think is that we would 24 have a sort of whatever procedure we had for the 25 pre-scan, that we had to make sure that possibly every 20 1 field in PinICL had something set to it. 2 Q. So might "User" have been essentially used as a default 3 setting in the absence of any other information? 4 A. I couldn't say whether that was a default. I'd be very 5 surprised if that was a default. I mean, I can't recall 6 what options I would have had under "defect cause" but 7 I know in one of the other PinICLs somebody had set it 8 to "unknown". So if you're going to go for a default, 9 I would have thought it would be something like 10 "unknown". 11 So I mean, as far as I can think, that it's simply 12 a case of using your best guess. If you've got to set 13 something, try and set it to something that you think is 14 appropriate. 15 Q. In this case attributing it to user that means user 16 error in essence? 17 A. A potential user error, yes. 18 Q. Thank you. 19 A. What I'm not sure, I'm not sure whether that "defect 20 cause" gets sent back to PinICL -- sorry, PowerHelp, not 21 PinICL. So I think that is a PinICL-only entry. So it 22 would not have gone back to PowerHelp. So SMC and HSH, 23 I don't think, would have seen that. They would have 24 seen the category that we close it as which could be 25 very different from what we think the sort of defect 21 1 cause is. 2 Q. But might the defect cause be something that is 3 discussed with those who are communicating directly with 4 the subpostmasters? It doesn't have to be in this case 5 but, in general, to the best of your recollection, if 6 you had marked something as "user error", for example, 7 might that have been communicated to the Helpdesk? 8 A. No, I don't think so. I mean, I'd be surprised if Mike 9 paid any attention to that at this stage because he's 10 got to investigate, he's got to look and see, yes, is 11 there a system -- the PM is saying there's a system 12 error, so we need to proceed on that basis -- or, sorry, 13 Mike needs to proceed on that basis. So I doubt he'd 14 have paid any attention to that and he would have -- 15 I mean, I know with this call it did turn out to be user 16 error but if he then thought, "Well, no, it's not, it's 17 a code error", then he would have changed that when he 18 closed the actual call. 19 That's actually when that defect cause comes into 20 effect, if you like, that becomes -- at the moment it's 21 sort of irrelevant. 22 Q. Mike was the engineer fixing the issue? 23 A. Yes. 24 Q. He would have seen when he logged on to the system for 25 the first time that that defect cause had been 22 1 attributed to it, whether he read something into it or 2 not? 3 A. Yes, he would have seen that, yes. 4 Q. Can we go to page 4, please. About halfway down this 5 page, you have information there: 6 "I have spoken to the PM, who is still having 7 problems with his cash account (a shortage of £70,000 8 this week). Continuing investigation." 9 If we look lower down on that page: 10 "Repeat call: Caller has rung back, he is very 11 agitated as he keeps having problems with the system 12 when balancing. He thinks it is a system problem. 13 Voiced Barbara Longley." 14 If we go over the page, please -- sorry, if we 15 could stay on page 4, the words that I didn't read out 16 there was "Repeat call: Caller has rung back, he is very 17 agitated as he keeps having problems with the system 18 ..." 19 Can you assist us at all, did you get a sense 20 working on the SSC of the Human Impact that these kinds 21 of issues were having on users, customers? 22 A. Yes, because, obviously, when you ring the PM, one of 23 the things that -- you know, they want or they need 24 their system, their cash account to balance and if it's 25 not -- then, yes. So yes, we would be aware that, you 23 1 know, the postmasters were getting stressed, you know, 2 by using the system and it's not doing what they felt it 3 should be doing or giving them the information that they 4 thought should be there. 5 Q. How common was that? Was that a daily occurrence where 6 users were getting stressed, weekly? 7 A. Don't know. Can't answer that. 8 Q. If we go over the page again this is 1999 so the early 9 days of Horizon: 10 "NBSC have stated there are no [that's Horizon 11 Field Support Officers] available to help this PM. At 12 present he does not have enough knowledge of the system 13 for SSC/HSH to advise him. He requires on-site training 14 and until this is provided by POCL SSC are unable to 15 help him." 16 This brings us back to really where we started in 17 this phase. Did you have any concerns about the 18 training that was provided to subpostmasters? 19 A. I had nothing to do with the training for 20 subpostmasters, so I've no idea what training they 21 received. 22 Q. Having received calls like this or read logs like this, 23 did you have, at the time, any concerns about the 24 training? 25 A. Not that I recall. 24 1 Q. Can we please look at FUJ00072297, please. This is 2 another early log. This time it's written in a PEAK and 3 that's in August 2000. The issue that's raised here is 4 a receipts and payments mismatch. If we can look at the 5 first entry, please, it describes the issue. It says 6 that there's a receipts total and it gives a figure and 7 a payments total and it gives another figure, and 8 there's a difference: 9 "This office earlier raised a query because 10 a transfer for an amount ... seemed to have gone 11 missing. The amount of the transfer is exactly half the 12 amount of the difference between the receipts and 13 payments." 14 If we look down we have Barbara Longley there 15 referring to it in the call summary as a receipts and 16 payments mismatch. Then we have again yourself at 17 12.17, and it says: 18 "The call record has been assigned to the Team 19 Member: Steve Squires. Defect cause updated to 40: 20 General -- User." 21 Again, that's something we saw earlier the 22 reference to something being a user error, at least 23 initially. Does that assist you with whether 24 attributing something to a user was effectively used as 25 a default or a starting position? 25 1 A. No, I think it would just be a case of using my 2 knowledge and experience of the system, and I'd been, 3 what, there, is it two years now. Again, you know, the 4 EPOSS receipts and payments wasn't my particular area of 5 expertise, so again I'm just going by -- you know, if 6 I've seen a lot of these sort of calls come over and, 7 you know, they might have been sent back as "user". We 8 thought that that was appropriate at the time, so ... 9 Q. We spoke earlier about particular technicians having 10 particular interests and you mentioned two names in 11 respect of dealing with EPOSS issues and balancing 12 issues. If they knew about something called or what was 13 being referred to as a receipts and payments mismatch, 14 how would that information have been received by you or 15 is this an example where it seems it hadn't been 16 received by you because you attributed it to user error? 17 A. Yes, I mean, I'm aware that there was a bug which 18 I can't remember what it was, whether it was transfer 19 between the stock units, or something like that, and it 20 would cause the amount to double, which my immediate 21 look at this is, you know, that might be along those 22 lines. Now, whether I knew that at the time of this 23 call, I have no idea. 24 Q. Had you known about it at the time -- 25 A. If I had, then, yes, I mean, attributing it to "user" 26 1 would be an error on my part. 2 Q. Can we look at page 4 of the same document, please. If 3 we look at the second entry, there's the summary there: 4 "There was a short period on live where the EPOSS 5 code was out of step with the StopDeskTransfer code. 6 The EPOSS code was still writing ..." and it gives some 7 information there. 8 If we scroll down a bit to John Moran at 13.46, 9 please. Thank you. We have that being fixed by 10 a release, I think that is CI45, and then it's closed. 11 So it's clear in this case that it was something, 12 a technical issue, a software issue, that was ultimately 13 fixed by a release. Had that information been known to 14 you when you took on the call from the beginning, 15 presumably you wouldn't have been attributing it to 16 system error? 17 A. Correct. I would have used the software category, 18 whatever that would be. 19 Q. Using this as an example, does this raise any concerns 20 for you about the sharing of information within the SSC 21 and the ability for at least those who initially take on 22 the calls to understand and correctly attribute the 23 problem? 24 A. I don't know, actually. 25 Q. Can you see any problem with attributing something to 27 1 user error in terms of the mindset, perhaps, of those 2 who are dealing with the issue? 3 A. I mean, what I don't know is what the call was -- what 4 category was the call closed as because just because 5 I've set it as "user error" as an initial thing, as 6 I say, I don't think anybody's going to be paying 7 attention to that until you actually come to close the 8 call and that's when that category would then be 9 important. 10 Q. But it's the first thing that those who are 11 investigating the matter, the engineers, would have 12 seen, isn't it? It's right above. 13 A. Well, I mean, you would have seen it but, as I say, 14 I don't think I ever paid any attention to that field 15 when I was investigating a call. So I would just 16 look -- I would look at the call details, not what 17 somebody set that particular field to. The only time 18 I would have looked at that was when I come to actually 19 close the call to see do I need to change it to 20 something more appropriate. 21 Q. Having seen it referred to by Barbara Longley as 22 a "receipts and payments mismatch", though, can you 23 assist us with why it might be attributed to "user 24 error"? 25 A. No, sorry, I can't. 28 1 Q. Can we please look at FUJ00086585. This is a PEAK that 2 I looked at with Barbara Longley. It's described there 3 in the summary as: 4 "The PM is having problems rolling the office 5 over." 6 If we look the beginning: 7 "The pm is having problems rolling the office 8 over. There are figures missing from the cash account 9 which is one person entire work." 10 If we scroll down to about halfway down on the 11 right-hand side -- I think that's "All", it says AL1 but 12 I think that's "All": 13 "[All] her work is missing from the [cash 14 account]. When she did a balance snapshot she was 15 £9,000 over and all her stock is showing as minus." 16 If we scroll down there's advice given. Barbara 17 Longley's evidence was that this advice came likely from 18 the Helpdesk rather than SSC. 19 A. Yes. 20 Q. It seems as though, if we scroll down to the bottom, 21 that the initial advice that was given to her was wrong, 22 according to at least one adviser, that she shouldn't 23 have been advised, I think it was, to roll over. Can 24 you see that? 25 A. Yes, that's on the screen, yes. 29 1 Q. Thank you. If we scroll over to the beginning of the 2 next page, this may assist you by way of background. 3 I won't read it but you might want to just read that top 4 paragraph to yourself. 5 A. Yes. 6 Q. Then can we turn to page 4, please. We have your 7 involvement there pre-scan: 8 "It's so good they've told us 3 times by the looks 9 of it." 10 Are you able to assist us with what you may be 11 referring to? 12 A. Yes. That should be two times not three times because 13 the text that we've got at the beginning has been pasted 14 in twice for some reason. 15 Q. Thank you. Then we have there "defect cause updated to 16 40: General -- User". So, again, in this case, we have 17 the PM having problems rolling the office over, been 18 given wrong advice by the Helpdesk and it is attributed 19 to user error. 20 Does that assist you at all in the matters that 21 we've previously been discussing, about whether "user 22 error" was used as some sort of default code for when 23 cases came in? 24 A. Not that I'm aware of. 25 Q. If we look on page 7, about halfway down the page, we 30 1 have an entry from Martin McConnell at 12.55: 2 "After my first run through, the Stock Balancing 3 process has worked successfully as of 27 April 2000. 4 Before passing this back with the event log, may 5 I request that the messagestore for node 1 is retrieved 6 directly from it. I suspect there is a serious problem 7 (Riposte wise??) with this as opposed to the 8 correspondence view of it. I shall still continue 9 looking at subsequent weeks to see why the situation 10 never recovered itself." 11 Do you remember issues with Riposte during the 12 early years of Horizon? 13 A. Yes, there were a number of problems with it but also 14 I'd just like to note that Martin has just above, at 15 9.32 changed the defect cause to "General -- Unknown". 16 So it would be down to whoever was investigating, once 17 they got an idea of what the problem was, to clearly 18 change that defect cause to whatever they thought it was 19 and I think in this call that then gets changed again, 20 later on, to either code or reference date, I think 21 somewhere, so ... 22 Q. If we look two entries down there's an entry from Martin 23 McConnell that says: 24 "This is another instance of [and it gives the 25 PinICL or PEAK reference] where the data server trees 31 1 have failed to build. This has now been fixed in [the 2 software release]." 3 So this is clearly, ultimately a software issue? 4 A. Yes, and you can see the defect cause has been updated 5 to "code". 6 Q. Yes. Now, as you highlighted, there is a defect code 7 "unknown" and that was the defect code that Mr McConnell 8 applied. Wouldn't it have made more sense to have 9 applied defect code "unknown" in your original entry on 10 page 4? 11 A. Yes. 12 Q. Thank you. I'm going to take you to one final document 13 and I'm just going to check -- yes, it's on the system 14 now -- FUJ00057524. Thank you very much. 15 You spent a bit of time with this document this 16 morning. 17 A. Yes. 18 Q. Have you had sufficient time to have a look through to 19 understand what's going on? 20 A. Yes, I believe so. 21 Q. Thank you very much. So this is described in the top 22 there as "Transactions missing" and if we look at the 23 bottom we have detail of the customer call, and it says 24 at the bottom: 25 "Repeat Call: when PM did her daily reports 32 1 yesterday after having a new base unit fitted, there 2 were transactions missing off them." 3 If we scroll down: 4 "When she re-entered the missing transactions this 5 corrected her daily reports but they were showing twice 6 on her balance snapshot." 7 Are you able to assist us briefly what that might 8 mean? 9 A. Well, she had a -- there was a hardware problem by the 10 looks of it with her counter 1, also known as the 11 gateway. So the engineer's been, he's replaced the 12 hardware and when she's come to do her daily reports 13 she's realised that -- so she's clearly done them on 14 that new counter and she realises that some of those 15 transactions that she did earlier on, probably that day, 16 weren't on her report and yet she's got the receipts to 17 sort of say, "Yes, I did do these transactions", so 18 where are they? 19 Q. If we look at advice that's given, halfway down it says: 20 "Advised the caller to reverse her transactions 21 that she has put in by doing a transaction log. The 22 caller is happy to do this. Advised the caller that if 23 her reports are really bad she will have to contact the 24 NBSC but she will manage to balance." 25 Do you recall why people might call the NBSC or 33 1 the Helpdesk relating to issues balancing? So let's say 2 they thought they had a technical issue, should they 3 call the NBSC or should they call the Helpdesk if it 4 related to balancing? Is that something that you were 5 ever involved in? 6 A. I don't think so but the NBSC, I think, were the Post 7 Office own Helpdesk so, obviously, they would have 8 an awareness of what a postmaster needs to do as part of 9 their sort of daily business. 10 Q. If it was a technical issue that resulted in 11 an incorrect balance, who, in your view, would be the 12 appropriate helpline to call? 13 A. They would then call the HSH, who would then pass the 14 call to SMC, who would then pass the call to SSC. 15 Q. It says there: 16 "Contacted: spoke to the PM and she was query 17 whether or not to reverse the transaction and what 18 affect it would have on her stock." 19 So it seems there as though the postmaster is 20 a little concerned about what the implications of the 21 advice would be; is that fair? 22 A. Yes. 23 Q. "Advised to contact NBSC [regarding the] stock." 24 If we scroll down over to the next page -- I won't 25 go through every entry -- but we have there near the 34 1 bottom: 2 "Contacted: called PM to clarify the information 3 received and PM is convinced there is a software 4 problem. PM has been on system for a long time so is 5 fully aware of balancing procedure." 6 So although this is in the year 2000, it's quite 7 late in the year 2000. Rollout had occurred and this 8 postmaster was saying they didn't have issues with their 9 own balancing, it's a software issue. 10 A. Yes. 11 Q. I think we can scroll through the next few pages. You 12 have read all of these? 13 A. Yes. 14 Q. Perhaps we can go to page 6. We're now on 18 November. 15 The first call is 15 November. We're now 18 November: 16 "PM has called today to report that the balance 17 snapshots which are printed off 2 of the counters are 18 showing different figures, even though they are attached 19 to the same stock unit. She would like to speak to 20 somebody from 3rd line [as soon as possible]." 21 Then we have Diane Rowe assigning this matter to 22 yourself. 23 A. Yes. 24 Q. Are you able to assist us with why it would have been 25 assigned to you? 35 1 A. Probably because I was available. So, even though it 2 wasn't my particular area of expertise, in the SSC you 3 were expected to handle any type of call. So Diane 4 would have looked at what calls have I got. She may 5 have come and spoke to me, you know, "Are you busy, can 6 I give you a call?" And then it's like, "Yes, send it 7 over, I'll have a look see what I can do about it". 8 Q. We have below that Diane Rowe attributing the defect 9 cause "99: General -- Unknown", so in this instance she 10 didn't attribute it to a user? 11 A. Correct, yes. 12 Q. If we go over the page, your entry there, you say: 13 "Have had a look at the messagestore and am unable 14 to match what the PM is saying in this call with what 15 I see in the messagestore. Please provide date and time 16 of the balance snapshot and trial balance reports that 17 the PM is querying. Also require quantities and values 18 for the Giro deposits ..." et cetera, et cetera. 19 You're seeking further information there? 20 A. Yes. 21 Q. But about halfway down, you say: 22 "PM has not been contacted. Closing as 23 insufficient evidence." 24 Can you assist us why would something be closed as 25 insufficient evidence, rather than kept open until that 36 1 evidence has been obtained? 2 A. It was part of the responsibility of the SMC to provide 3 whatever -- all the evidence that they could provide for 4 us to then go and investigate this problem. So I've 5 looked at what the PM has reported and, normally, you 6 would be able to see those transactions in the message 7 store and I've got this sort of very unusual situation 8 where I've looked in the message store and I cannot find 9 any evidence of those transactions ever occurring. 10 So I can't go any further with this and so that's 11 why I ask for a session ID, because maybe I'm looking in 12 the wrong place on the message store. I mean, there 13 were, was it, 510,000 messages we'd got up to within 14 this particular counter's message store by this point. 15 So, you know, that's an awful lot of messages to be 16 looking through, whereas a session ID, I will be able to 17 track that down relatively quickly and, therefore, be 18 able to start by investigation in that area because it 19 may be that there's a problem on the counter with the 20 clock being wrong and so Riposte is storing the wrong 21 date and time in the message store. 22 So I'm looking, you know, on what I think is -- or 23 what's shown to be the wrong date. 24 Q. So there might be a date or time issue on the counter; 25 is that one potential -- 37 1 A. That's one potential possibility for why I can't find 2 those transactions because, normally, you know, you 3 would be able to see those in the message store and you 4 go, "Okay, this is where I start my investigation". 5 Q. So we have one potential might be a date and time issue 6 on the counter. Another issue, might it potentially be 7 an issue with the message store itself? 8 A. Yes, and, ultimately, that's what it turns out to be. 9 There is a Riposte error here where counter 1 -- when 10 you replace a counter, it comes with a blank message 11 store. So Riposte will start up and it will then call 12 out to the other counters in the Post Office to say, 13 "Okay, have you got any messages for me?" So those 14 counters would then reply saying, "Yes, I've got 510,000 15 messages for you, here you go". So that counter would 16 then start reading those messages in and writing them to 17 its own message store. 18 Once it's got all those messages, it can then 19 start writing its own messages to that message store and 20 one of the first messages would be a Riposte version 21 string message and so that's how we would -- by seeing 22 that message, we would know that Riposte has been 23 restarted at that point. 24 There was a bug whereby the counter would think 25 that, "Okay, I've got all my messages now", but, in 38 1 fact, it didn't. So there were still some messages to 2 be sent across and, for whatever reason, Riposte -- we 3 sort of call that about -- Riposte coming back online 4 too soon and that's what seems to have happened here. 5 Q. So that's what happens ultimately. If we're looking at 6 16.29.44, where you've said, "PM has not been contacted 7 closing as insufficient evidence", going through your 8 mind at that stage, you mentioned might be a date and 9 time issue on the counter itself. You've now mentioned 10 a Riposte problem that it could potentially be. Were 11 those thoughts that would have been in your mind at the 12 time? 13 A. Yes. 14 Q. If we look two entries down, 16.29.45, again you have, 15 "Responded to call type L as category 96", this time 16 "insufficient evidence", but then two entries below 17 that: 18 "Defect cause updated to 40: General -- User." 19 So, again, we have something that clearly in your 20 mind might be a software error, it might be a counter 21 error but it's there being attributed to user error. 22 Does that assist you in answering the question that 23 I asked some time ago about whether there was 24 an approach to attributing things to users as a default? 25 A. I have no idea why I selected that. I mean, that 39 1 I don't think -- category "96: Insufficient evidence" 2 that's what would go back to PowerHelp to alert the SMC 3 that, "Okay, I'm asking for more evidence, can you get 4 the evidence and then send it back to me?" Why I would 5 have picked "General -- User", I'm sorry, I don't know. 6 Q. One thing that we heard during the human evidence 7 sessions in this Inquiry was postmasters being told that 8 they were at fault, that the issue is user error not 9 software error. 10 Looking back at these documents, do you think 11 there was a culture of attributing things to user error? 12 A. No. But I can certainly see how you could come to that 13 conclusion. 14 Q. I mean, there are three or so PinICLs that have been 15 attributed to user error -- 16 A. Yes. 17 Q. -- rather than insufficient evidence or unknown error 18 or -- I mean, there are multiple options available -- 19 A. Yes, I have no idea what else I could have put but, to 20 my mind, you know, looking at this, that's wrong. So 21 I don't know why I would have picked user. 22 Q. If we scroll down, staying on this page, you have 23 entries there, you say: 24 "I have spoken to PM last night advised that this 25 is being looked into." 40 1 Then over the page you have quite a long 2 explanation about what you think the issue is. Are you 3 able to summarise for us, very briefly, point 1 to 7, 4 what you thought at that stage the issue may have been? 5 A. Yes. So on the 20th, obviously I've sent it back. So 6 on the 21st, I probably did nothing with the call, 7 expecting it to come back to me with the evidence I'd 8 asked for but, as it says that beginning of that sort of 9 long list, there was a problem with the OTI, the 10 interface between PowerHelp and PinICL, so I don't 11 think -- the stuff that I typed in there didn't go back 12 into PowerHelp, so the SMC had no idea. 13 So I suspect that on the 22nd, I'd probably have 14 spoken to the SMC to say, you know, "What are you guys 15 doing about this particular call? Can you give me the 16 evidence and send it back to me", at which point, you 17 know, they say "What evidence? We haven't got that. 18 That hasn't come across". But, clearly -- because 19 I started was it 8.50 in the morning, I think it was -- 20 8.52, I clone the call. 21 So I imagine on the 22nd I've actually looked at 22 this myself. Without the call coming back to me I've 23 looked at what's been going on and I would probably have 24 spoken to either Pat Carroll or John Simpkins and said 25 "I've got this unusual call, I can't see these messages 41 1 or transactions that the PM says they have done, any 2 ideas?" And I suspect they have probably gone "Well, 3 you could have a look to see what do the counters 4 themselves say", because, as a default, we would have 5 gone to the correspondence server and the messages on 6 the correspondence server. That's why we normally look 7 when we had to investigate a message store. 8 So I've then gone down onto the actual counters 9 themselves and noticed that, yes, as I say at the top 10 there, counter 2 has 48 messages which are not on 11 counter 1. So, clearly, something rather serious has 12 gone wrong with Riposte and then it's a conversation 13 with Development to say, "Okay, how do I fix this?" 14 Q. If we look at the conversation with Development you say: 15 "Can development please investigate on whether 16 there's a deficiency in Riposte and what can be done to 17 stop this happening again. Also, need advice on how to 18 get the message store in sync and to include the missing 19 transactions." 20 Did you have a concern at this stage that the 21 missing transactions wouldn't be retrieved? 22 A. No, because the missing transactions were on counters 2 23 and 3. The problem was they weren't on counter 1 and 24 they weren't on the correspondence server and that would 25 cause problems when we're retrieving cash account 42 1 messages via the agents, which would have looked at the 2 correspondence server messages and so they would miss 3 the transactions that were on counters 2 and 3. 4 So they wouldn't -- so the APS -- yes, some of the 5 transactions are APS so, you know, there's -- I can't 6 remember what APS -- automated payments, something like 7 that, I can't remember. 8 So you've got customers saying -- paying a gas 9 bill, for example, you know, they've paid, they've got 10 the receipts, but their account wouldn't be updated 11 because those messages on counters 2 and 3 aren't at the 12 correspondence server. So I'm asking, "Okay, how do we 13 get these transactions back onto the correspondence 14 server, so that they can be harvested, so that customer, 15 you know, bills get paid?" 16 Q. You also have a concern below the highlighted section. 17 You say: 18 "Also how will this affect their balancing. They 19 are currently in cash account period 34." 20 So you are raising a question there about what 21 effect this will actually have on the subpostmaster's 22 ability to balance? 23 A. Yes. 24 Q. If we scroll down to the next page, please, you have got 25 the postmaster chasing the progress of this call. 43 1 That's 10.59. So they are chasing again: 2 "She's concerned about balancing tomorrow. I've 3 said the call is currently with development. Do we have 4 an update?" 5 You seem to be anxious there to receive an update 6 for the postmaster. Is that a fair summary of that 7 entry? 8 A. Yes, I know it's important that the Post Office 9 balances, so, yes, I want to make sure she can balance 10 and rollover into the next cash account period. 11 Q. We have an entry at 14.17.19 from Martin McConnell: 12 "Note to be passed on to the customer for 13 balancing: this problem has occurred with replication 14 before (in essence, due to a failure in Riposte for 15 whatever to replicate back down)." 16 So, again, we've spoken about issues with Riposte. 17 The suggestion is that this is an issue here with 18 Riposte: 19 "It should be perfectly okay to continue balancing 20 on nodes 2 or 3 but not on node 1 where the failure 21 occurred." 22 He says: 23 "From the Riposte point of view there seems to be 24 a major disagreement on what the contents [and it gives 25 some code there that I won't try and understand] for 44 1 about 50 messages should be. There are minor glitches 2 here and there but this seems to be the major 3 discrepancy." 4 Is this something you remember at all, this 5 particular issue? 6 A. Not particularly. It's fairly clear what's going on 7 from the content of the call itself, so ... 8 Q. If we scroll down to the next page, Martin McConnell 9 there says: 10 "This blows my whole understanding of what Riposte 11 should be handling on our behalf, ie replication not 12 deviation across nodes." 13 Does it seem as though this is quite a significant 14 issue? 15 A. Yes, indeed. 16 Q. It continues and you have a paragraph -- at the bottom 17 of the next paragraph. It says: 18 "Whatever happens, this bug should end up with 19 Escher development." 20 I think that is, so that's, I think, the team that 21 Gareth Jenkins and others were part of. 22 A. I believe so, yes. 23 Q. Then you appear again and you say: 24 "I've spoken to the PM and advised her to roll 25 over to counters 2 or 3 not 1 but have not mentioned 45 1 about recovering the AP transactions." 2 So can you assist us with what you mean there? 3 A. Well, obviously, I've spoken to the PM and passed on the 4 advice from Development, as, you know, "Don't use 5 counter 1 to do it", because the AP transactions that 6 she needs in order to balance are on 2 and 3. Counter 1 7 knows nothing about them. 8 Now, I think probably why I didn't mention about 9 recovering the AP transactions is -- probably is part of 10 the conversation that I had with the PM because I then 11 go on to say, "Since the PM recovered the transactions 12 and then reversed them", and then I've got a further 13 question of, okay, sort of what kind of effect is that 14 going to have? 15 Q. So you say: 16 "Can Development please advise on whether PM does 17 need to recover the AP transactions since the PM 18 recovered the transactions and then reversed them. If 19 she balances on counter 2 will it take the AP 20 transactions from its copy or will it only look at AP 21 transactions done on counter 2?" 22 So you seem to be raising there, really, an issue 23 with the integrity of the balance and the transactions 24 and the ability for the subpostmaster to effectively 25 balance. Is that a fair summary? 46 1 A. Yes. I mean, I suspect from reading that that the bit 2 of code that does the -- of retrieving the messages to 3 do with balancing would look on its own counter, rather 4 than simply go to -- no, look on its own counter which 5 is why the advice was "Don't do it on node 1 because 6 we've got a bunch of missing transactions. So if you 7 did it on node 1 then you're not going to balance, but 8 those missing transactions are on counter 2 and 3, so if 9 you do it on one of those then, yes, you should 10 balance". 11 Q. But is there still a lack of clarity as to what's going 12 to happen with the transactions from node 1? 13 A. Yes, yes, because I think that's what I'm sort of asking 14 is: how is this going to affect sort of when the 15 harvesters sort of try and harvest for these AP 16 transactions, is that going to -- so, whilst the post 17 office itself would be able to balance, that might have 18 a knock-on effect on when we harvest those transactions 19 up from the correspondence server and, obviously, we 20 then send information off to the Post Office for them to 21 actually pay the customers. So I think that's what I'm 22 asking. 23 Q. So they may ultimately still be missing transactions 24 somewhere in the system? 25 A. Yes. 47 1 Q. If we go over the page, please, we have a message from 2 Brian Orzel to Gareth Jenkins: 3 "Gareth, should we deal with this? Do we have 4 value to add or has it been misrouted?" 5 Gareth Jenkins says: 6 "I don't know that I can add anything useful here. 7 This is another example of recovery having gone wrong 8 after a box swap." 9 I will just read the final paragraph of that page. 10 It says: 11 "This resulted in about 50 messages being lost. 12 The gateway did not communicate with the slave until it 13 had written at least 50 messages ... For this reason 14 there was no error indicating a self-originating message 15 being found." 16 I will read the second paragraph there. It says: 17 "Other than pursuing the known problem of how ... 18 we handle fouled up recovery (covered by [a separate 19 PinICL]), I don't think I can add anything further to 20 this PinICL and so it might as well be closed. I assume 21 that the missing transactions have been recovered 22 manually." 23 Now, knowing what you know about this issue and 24 having reread this PinICL, do you think assuming that 25 the missing transactions had been recovered manually was 48 1 the appropriate assumption to make at that time? 2 A. For Gareth, yes. 3 Q. Why do you qualify that? 4 A. Well, he hasn't got access to the message stores on the 5 correspondence server and it wouldn't be down to 6 development to reconcile those missing transactions with 7 the Post Office. That would be an MSU action and 8 I think somewhere I cloned this call, I think. Yes, 9 so -- no, that's not that one. 10 Yes, so immediately after Brian Orzel on 11 1 December, 11.18, I cloned this call to PC59052, which 12 I am assuming I would have sent -- that would be the 13 call that I would have sent off to MSU with the details 14 for those APS transactions and any other transactions 15 for them to sort out the reconciliation with the Post 16 Office. 17 Q. So was that an assumption that the MSU would take it on 18 and that it needn't be an issue for the development 19 team? 20 A. Yes. 21 Q. So where Gareth Jenkins is there saying, "I assume the 22 missing transactions have been recovered manually", are 23 you saying that was appropriate because it's effectively 24 not his job to look into whether, in fact, the missing 25 transactions had or had not been recovered manually? 49 1 A. Yes. 2 Q. Knowing what we now know about everything that happened 3 with Horizon, do you think that that approach, not 4 having sight of beginning to end and what ultimately 5 happened to the transactions is, in any way, 6 problematic? 7 A. I don't think so because it was MSU's responsibility. 8 They had the link with Post Office. They were the ones 9 who had the job of doing the actual reconciliation. 10 Gareth can't do anything more from a Development point 11 of view because they already know about the problem and, 12 presumably, are pursuing it under PinICL 52823. 13 Q. So would it have been typical for Gareth Jenkins and his 14 team and, in fact, the wider SSC, to not be concerned 15 with what ultimately happened to missing transactions 16 because that was a matter for another team? 17 A. No, we would have been concerned and I think Gareth is 18 voicing his concern here by saying, you know, I assume 19 that the missing transactions have been recovered 20 manually. So he's asking -- he's basically asking has 21 that been done and the answer is yes. But all -- 22 Q. Where's the answer, sorry? 23 A. Well, the fact that I've cloned the call to PC59052. 24 Q. So does that mean you know that the missing transactions 25 have been recovered manually? 50 1 A. Yes. I don't know if -- that's probably not a later -- 2 no, I mean, there's nothing in this PinICL to say that 3 that has happened, so you would need to have a look at 4 PC59052. I suspect that's call that I would have sent 5 to the MSU to say "This has happened", give 6 an explanation and these are the details of the 7 transactions and it's down to them to sort of -- 8 whatever the process was for reconciling that with Post 9 Office. 10 Q. So would you have taken responsibility for ensuring that 11 that question that he asks was, in fact, answered and 12 that the feedback that came back was, "Yes, the missing 13 transactions have been recovered manually"? 14 A. I probably wouldn't have gone back to Gareth to say 15 "Yes, they have been", but that kind of response, 16 I would have thought, would be on that call that 17 I cloned that I would have sent to MSU. 18 Q. Would you have taken it forward if there wasn't any 19 feedback from him that the transactions had been 20 recovered? Would you have been responsible for this 21 call going forward, up until its conclusion? 22 A. I mean, yes, when I cloned the call, I would then send 23 it to the MSU team. Once they've done the 24 reconciliation, they would close -- I don't know if they 25 closed call back to me or they would reassign it back to 51 1 the SSC to say "Thank you we've done a reconciliation, 2 this call can now be closed". 3 Q. Is that what we see on 12 December 2000 where you have 4 closed the call? 5 A. No, that's -- for this particular call, you would have 6 to look at 59052 to see what happened then. 7 MR BLAKE: Thank you very much. I don't have any further 8 questions. I don't think anybody else does either. 9 Sir, do you have any questions? 10 SIR WYN WILLIAMS: No, thank you very much. Thank you very 11 much for attending and answering the questions and 12 providing a witness statement. 13 MR BLAKE: Thank you very much. 14 Sir, it's now 12.00. 15 SIR WYN WILLIAMS: Yes. 16 MR BLAKE: May I propose that we take a 10- or 15-minute 17 break and then we move on to closing statements. 18 Mr Beer has something to say about other evidence that's 19 going to be published, but that will be brief, and then 20 we can move on to the closing statements. 21 SIR WYN WILLIAMS: Are those making the closing 22 statements -- I mean, I'm saying this to make it as easy 23 for them as possible -- sorry, to carry on making the 24 closing statements once we have had our 15-minute break 25 or are we planning to have a lunchtime break as usual? 52 1 MR STEIN: Sir, it may assist, if I just mention my 2 intention if it fits your requirements. Sir, I intend 3 to be somewhere between 45 minutes and 50 minutes, which 4 may take us a little while into the lunch break but 5 I hope to be no later than 1.15. 6 SIR WYN WILLIAMS: Yes. 7 MR STEIN: That's working on an assumption, a reasonable 8 one, that I have had after having a discussion with 9 Mr Beer about how long he is going to take with his 10 remarks on other statements. 11 SIR WYN WILLIAMS: Right. 12 MR STEIN: So, hopefully, that will then take us to about 13 1.15, then we have a lunch break and then others will 14 resume after that, if that suits the Inquiry. 15 SIR WYN WILLIAMS: It certainly suits me. Does it suit 16 those in the room? 17 MR MOLONEY: Sir, I would be next up and I would be content 18 to take whatever course suited everybody else, either to 19 follow from Mr Stein without there being a long lunch 20 break as usual, because I will only be 25 minutes, or to 21 take that lunch break. 22 SIR WYN WILLIAMS: And who is following Mr Moloney? 23 MR HENRY: I am, sir. I am very grateful to Mr Moloney 24 because I thought he was going next. In fact, we 25 misinterpreted each other, but he has very kindly agreed 53 1 to follow Mr Stein. I have about seven pages of written 2 notes, so I do hope to be 15 minutes. 3 SIR WYN WILLIAMS: I think my current view is that after 4 we've heard from Mr Stein, and if it is around about 5 1.15, we'll then take stock again as to whether people 6 actually want a full hour or whether they want, say, 7 half-an-hour or something less than that. We will just 8 go along and see how people feel. So we will take our 9 quarter of an hour now and then come back. Will it be 10 Mr Beer and then the closing submissions? 11 MR BLAKE: Yes. Thank you very much, sir. 12 SIR WYN WILLIAMS: All right, fine. 13 (12.02 pm) 14 (A short break) 15 (12.17 pm) 16 MR BEER: Sir, good afternoon. Can you see and hear me? 17 SIR WYN WILLIAMS: Yes, thank you. 18 MR BEER: Thank you very much. 19 Statement by MR BEER 20 MR BEER: Sir, as you know, the Inquiry is asked to build 21 upon the findings of Mr Justice Fraser in the Bates 22 judgment and the Court of Appeal in Hamilton v Post 23 Office and of other criminal courts to establish a clear 24 account of the failings of Horizon over its life-cycle 25 and the Post Office's use of information from it when 54 1 taking action against persons alleged to be responsible 2 for shortfalls. 3 In Phase 3, you have heard evidence of these 4 issues at an operational level. The evidence has 5 covered the issues of training, the advice and 6 assistance available to postmasters, the dispute 7 resolution procedure and the rectification of bugs, 8 errors and defects. You will have paid careful 9 attention to the three questions that run through every 10 stage of the Inquiry: who knew what and when about the 11 issues within Horizon. 12 Since January of this year, you've heard evidence 13 from over 30 witnesses. You're still to hear evidence 14 from Gareth Jenkins on Phase 3 issues and from Anne 15 Chambers on a small number of Phase 3 issues. 16 The evidence that you have heard since January is 17 but a small sample of those working at the operational 18 level within Post Office and Fujitsu over the many years 19 that Horizon has been live. It is, we say, unnecessary 20 to hear further oral evidence, given the extensive 21 documentation that the Inquiry has received and the 22 detailed findings of Mr Justice Fraser in relation to 23 bugs, errors and defects. 24 Moreover, to hear a greater sample of oral 25 evidence would inevitably mean commencing Phases 4 and 5 55 1 of the Inquiry at a much later stage. Those phases 2 concern the way in which the Post Office conducted 3 prosecutions and responded to the emerging scandal. 4 It's important to move to investigate those issues as 5 soon as reasonably practicable. 6 Your Inquiry team conducted further investigations 7 to obtain written witness statements from people 8 involved in various roles relating to the operation of 9 Horizon. The purpose of that exercise was twofold, 10 firstly to obtain a wider range of evidence on how the 11 various teams worked in practice from those at the 12 coalface and, secondly, to test what evidence they were 13 aware of of the existence of bugs, errors and defects in 14 Horizon. This was done by sending short Rule 9 requests 15 asking general questions tailored to the respective 16 roles. 17 The Inquiry sought statements from those involved 18 in the Post Office support services, including Horizon 19 Field Support Officers, NBSC members, trainers and 20 contract managers. The Inquiry also sought similar 21 evidence from those working in the Fujitsu-operated 22 Helpdesk and the SSC. The Inquiry has finalised 23 statements from a selection of these witnesses and I'm 24 going to say a few words now on the investigation into 25 each of them and display the URNs for the various 56 1 statements in each of those categories that the Inquiry 2 that obtained. 3 Those statements will be admitted into evidence 4 and treated as having been read into the record and the 5 witness statements will shortly be disclosed on the 6 Inquiry's website. 7 So, firstly, Post Office support services. 8 I begin with the teams in the Post Office assigned to 9 provide advice and assistance. In response to the 10 various Rule 9 requests, the Post Office has provided 11 the Inquiry with various lists of current and former 12 staff who worked in different operational and management 13 teams throughout the company. We used this information 14 to send Rule 9 requests directly to such people. 15 The Inquiry took a representative sample of people 16 who had worked as Horizon Field Support Officers or on 17 the NBSC. You will recall hearing evidence about 18 Horizon Field Support Officers or HFSOs during both 19 phases 2 and 3. They were Post Office employees who 20 dealt with branches as they migrated to Horizon from the 21 paper-based systems. A number of HFSOs transferred to 22 work on the NBSC, providing ongoing support to the 23 branch network. 24 Overall, the Inquiry sent over 70 Rule 9 requests 25 to people who had worked as HFSOs or on the NBSC. Those 57 1 Rule 9 requests sought evidence on the training given to 2 those employees and to subpostmasters, their experiences 3 in these roles and the adequacy of the support provided 4 and their knowledge of bugs, errors and defects in the 5 Horizon System. 6 I wonder whether we could display INQ00002006 and 7 move to page 2, please. 8 The Inquiry received final witness statements from 9 45 people within this cohort and on this page, the next 10 page and the following page, those 45 names are 11 displayed and the URNs of each of the 45 witness 12 statements are also displayed. They are to be treated, 13 please, as read into the record. 14 The Inquiry also carried out a similar exercise 15 with contract managers. Please can we look at page 5. 16 Thank you. The Rule 9 requests for these witnesses was 17 broader, covering all aspects of Phase 3, including 18 dispute resolution. We received 13 witness statements 19 following that exercise and the names of those witnesses 20 and their URNs are displayed on the screen. May they be 21 treated as read into the record, please. 22 Finally, so far as the Post Office is concerned, 23 the Inquiry identified a number of people involved in 24 training through reviewing the documentary evidence and 25 the comments of other witnesses. We sent Rule 9 58 1 requests to those identified to seek evidence on the 2 nature of the training provided to postmasters, as well 3 as the extent to which bugs, errors and defects in the 4 Horizon System were dealt with in the training 5 programme. 6 If we can turn to page 6, please. The Inquiry 7 received 11 finalised witness statements from such 8 trainers and the names and URNs of those witnesses are 9 now shown. May they be treated as read into the record, 10 please. 11 Can I turn to Fujitsu support services. Following 12 a Rule 9 request, Fujitsu provided to the Inquiry a list 13 of all of those people it had on record who had worked 14 on its Helpdesk. The Rule 9 request sent to each such 15 witness sought evidence on the training provided to 16 Helpdesk operatives, the day-to-day work on the 17 Helpdesk, the adequacy of the support provided and 18 whether there was knowledge of errors, bugs and defects 19 within the Horizon System. 20 The Inquiry received 13 finalised witness 21 statements from a selected sample. The names of those 22 witnesses and the URNs for their statements are shown on 23 the screen. 24 I should also read in the statement of Julie 25 Welsh, who deals with issues on the Helpdesk, but 59 1 needn't be called in this phase. She's an addition and 2 her URN is WITN04540100. 3 I should also, if we move to page 8, please, 4 propose to read in five witness statements that have 5 been finalised from people working in the SSC. These 6 members of the SSC were sent short Rule 9 requests to 7 obtain their witness evidence, covering how the SSC 8 worked and their own knowledge of bugs, errors and 9 defects. Their names are displayed along with the URNs. 10 May they be treated as read into the record. 11 That PowerPoint presentation can come down. Thank 12 you. 13 That concludes the statements that the Inquiry 14 wishes to read into the record at this stage. Your team 15 continues to receive some signed statements that will be 16 read in at an appropriate juncture later in the Inquiry. 17 I should pause at this stage to note that the 18 Inquiry has received a significant volume of disclosure 19 during the course of the Phase 3 hearings and it expects 20 to receive more disclosure that is or may be relevant to 21 Phase 3 in the very near future. Moreover, it expects 22 some of this disclosure to contain guidance given to the 23 NBSC and the Fujitsu-run Helpdesk. The Inquiry will, of 24 course, keep these documents under review and will 25 disclose them to Core Participants as soon as reasonably 60 1 practicable after their receipt. 2 Moreover, it will not hesitate to recall any 3 witnesses where it considers it is necessary to do so to 4 put questions to them on new documents that have come to 5 light. The appropriate time to do that will be 6 determined in due course but will likely be during the 7 Phase 5 hearings. 8 Sir, that's all I say at the moment in terms of 9 reading documents into the record and I think we now 10 move to the closing submissions from three Core 11 Participants in an order that has been agreed amongst 12 them. Thank you, sir. 13 SIR WYN WILLIAMS: Thank you, Mr Beer, and, for the 14 avoidance of any doubt, I confirm that the statements 15 identified by Mr Beer during the course of his oral 16 address are now to be treated as having been read into 17 the record. 18 MR BEER: Thank you very much, sir. 19 SIR WYN WILLIAMS: Mr Stein, whenever you're ready. 20 Closing statement by MR STEIN 21 MR STEIN: Thank you, sir. 22 Sir, this morning you referred to Veronica Maye. 23 She passed away seven days ago. She was the beloved 24 wife of Francis Maye. You may recall that Francis and 25 his wife, Veronica, ran the Bidford-on-Avon post office 61 1 in Warwickshire. Veronica's health was affected by the 2 financial state that was left for her and her husband to 3 deal with after they had been made bankrupt by the 4 actions of the Post Office. She developed angina. 5 You'll probably recall Mr Maye's experiences being read 6 into the record at the hearing in Glasgow on 11 May 7 2022. 8 Francis Maye is now 73 years old. He has been 9 a very active Core Participant in the Inquiry. He's 10 followed the Inquiry closely, attended our group 11 meetings with our clients and he regularly provides 12 instructions and views. He was, of course, a GLO 13 claimant and my instructing solicitors Howe+Co are 14 assisting him in relation to his GLO ex gratia claim. 15 Francis and Veronica were together for 24 years. 16 He says: 17 "We were the best of pals. My right arm [he says] 18 is literally ripped off." 19 Francis was brought up in a part of Ireland where 20 he learnt to read and write first in Irish, in Gaelic, 21 and then he was taught Latin and Greek. As such, his 22 written use of English is poor, spelling not very good 23 and his wife Veronica used to do all of the reading and 24 writing for him in relation to the Inquiry. 25 He does not know now what he will do in order to 62 1 read information from Howe+Co and how to write emails 2 and how to put forward instructions. He is, of course, 3 going to be supported in everything he does by my 4 instructing solicitors, Howe+Co. 5 Francis says that when they lost their home as 6 a result of the Post Office's actions, Veronica worked 7 multiple jobs to maintain them. He found it extremely 8 difficult to get any work anywhere because the Post 9 Office wouldn't give him a reference. He had even asked 10 for a reference when he went for a job picking fruit on 11 a farm and he couldn't get the job because he couldn't 12 get the reference. 13 Francis did get an interim payment under the GLO 14 scheme, just after Christmas, but, sir, you observed the 15 trustees in bankruptcy took a lot of that award. You 16 will recall that that matter was addressed by my junior, 17 Mr Chris Jacobs on the 27th. 18 When he and Veronica first met, he told her he 19 would take her on a cruise one day but they obviously 20 couldn't afford it after they lost everything as 21 a result of the Post Office. When Veronica was in 22 hospital she saw a SAGA magazine, which showed a cruise 23 around Scotland and she reminded him on his promise to 24 take her on the cruise. Sadly, she, of course, died and 25 they never got to go on the cruise. 63 1 In terms of what Francis Maye hopes from this 2 Inquiry, he comments he's not in good health but he 3 would like to be able to live as comfortably as possible 4 in his final years. He says: 5 "I'd like the senior people at the Post Office and 6 Fujitsu to be held to account and taken to court. They 7 knew the system was wrong." 8 He lost his Post Office in about 2010 when Glenn 9 Chester walked in one morning out of the blue before 10 they opened and checked the balances and he balanced, 11 but he was still suspended on the spot. Mr Maye would 12 very much like to meet Mr Chester again and discuss 13 exactly the actions that were taken by Mr Chester and 14 the Post Office with him. 15 In terms of his view of Phase 3, Francis Maye says 16 this. He knows that he and other GLO litigants and 17 other Core Participants are in good hands with this 18 Inquiry but he feels that "the Post Office is trying to 19 kick the can down the road until we are all dead. The 20 deadline for GLO compensation of August 2024 is plain 21 wrong", he says and he believed, they, the Post Office, 22 should not have the right to set that date and 23 everything is always dictated by the Post Office, and 24 that's the way he feels. 25 He finally says "If ever some of the leaders at 64 1 the Post Office are taken to court, I will be at that 2 court", he says, and he hopes that doesn't sound vicious 3 but he hopes and prays that that little man will win out 4 in the end. 5 Sir, as you know, with Mr Jacobs I represent 6 a large number of subpostmasters and mistresses before 7 this Inquiry and we have been instructed by Howe+Co 8 solicitors. Of course, our written submissions after 9 this date will provide more details on the matters that 10 I cover today. The closing submissions for Phase 2 from 11 the Post Office dated 7 December 2022 made little 12 reference to its own failures and preferred to suggest 13 that the passage of time has dimmed recollections. 14 You'll see those references at paragraphs 3 and 4. 15 Then the Post Office turns its tank turret gun on 16 Fujitsu, paragraphs 5 to paragraph 29.6. In particular, 17 in paragraph 5, the Post Office flat out accuses Fujitsu 18 of making a concerted effort, going on to say, in many 19 of the Fujitsu witness statements, to suggest that POL 20 had the same level of understanding of the technical 21 problems and challenges as Fujitsu did. 22 At paragraph 29.5, the Post Office accuses Fujitsu 23 of deception. I quote: 24 "Fujitsu did not inform POL of these serious 25 issues. This must have been a deliberate decision." 65 1 The serious issues being referred to at 2 paragraph 29.4 was that Fujitsu was aware at all levels 3 of management that the Horizon project was facing 4 serious problems. Now, the Post Office, POL, is 5 a wholly-owned business with one shareholder, the 6 Government, managed through UKGI and it is, therefore, 7 saying that its current business partner, Fujitsu, 8 an international company of some size and renown, has 9 deliberately misled its customer, both through the years 10 of the Horizon System's operation and, to date, in 11 preparing statements for this Inquiry, and it has done 12 so with its witnesses in order to try and shift the 13 blame onto the Post Office. 14 Well, on behalf of the subpostmasters and 15 mistresses I represent, I'm afraid I cannot wish Fujitsu 16 well but I can warn Fujitsu that, once the Post Office 17 takes a stance, no matter how ill-conceived it is, it 18 doesn't give up. Remember, my clients were accused of 19 malfeasance and criminality over decades. 20 But these accusations of this type of conduct by 21 the Post Office does not match up with the reality of 22 today's business affairs between the Post Office and 23 Fujitsu. So far, the Horizon contract has been extended 24 from 2023 to 2024 at the cost of many millions. The 25 meetings to discuss the matters of these contractual 66 1 extensions must be merry affairs, sir, with Fujitsu 2 staff, one assumes, rather reluctant to talk to the Post 3 Office representatives, just in case they get accused of 4 making things up. 5 The Post Office has considerable form for blaming 6 others. The Post Office blamed and criminalised 7 subpostmasters throughout the history of Horizon, blamed 8 the litigants in the High Court and said they were 9 making it up and now seeks to blame Fujitsu when the 10 truth, sir, is that Fujitsu and POL are equally to blame 11 in the partnership of deception, in which the Post 12 Office was the senior partner. 13 Of course, the problem for the Post Office is that 14 they now have issues with the creation of a Cloud-based 15 replacement for the Horizon System, meaning that POL has 16 to keep extending the Horizon contract. The Post Office 17 procurement documents of 6 April 2023 make interesting 18 reading. I quote: 19 "The programme to transfer the services to a new 20 Cloud provider created fundamental technical challenges 21 that POL could not economically and technically overcome 22 and the business has taken the decision to pivot back to 23 the Fujitsu-provided Horizon data services until the 24 successful tender of services out of Horizon", referring 25 then to a cost of £16,500,000. 67 1 Well, something about that procurement process 2 sounds rather familiar to us who have been listening to 3 the evidence before this Inquiry. But, sir, that's not 4 the only Post Office news. Unfortunately, recent press 5 reports show that the Post Office Postal Affairs 6 Minister was called to the House of Commons to answer 7 how a "grotesque" -- that's the word used in the 8 press -- executive bonus scheme was approved on the 9 basis that the Post Office has helped this Inquiry with, 10 apparently, I quote: 11 "... all required evidence and information 12 supplied on time, with confirmation from Sir Wyn 13 Williams and team that Post Office's performance 14 supported and enabled the Inquiry to finish in line with 15 expectations." 16 Well, that reference that it wished to refer to 17 doesn't appear to have been correct. This was referred 18 to by Members of Parliament as a deliberate lie and 19 caused you, sir, on 5 May to ask for some clarification, 20 quoting from your correspondence: 21 "... given it suggested that a metric had been set 22 and a target had been achieved with confirmation given 23 from myself and my team." 24 Well, sir, sometimes it's tempting to suggest that 25 the Post Office couldn't run an average celebration in 68 1 a brewery but, unfortunately, it's more sinister than 2 that. The Post Office remains a thoroughly dishonest 3 and duplicitous organisation. 4 The Post Office opening statement, the written 5 statement dated 4 October 2022, begins with: 6 "Post Office Limited (POL) apologises for the 7 suffering and damage caused to every person who has been 8 affected by the Horizon IT scandal. That includes not 9 only postmasters directly affected by POL's failures but 10 all others, including, in particular, their families, 11 whose lives have been impacted by those failures." 12 POL goes on to say that they remain fully 13 supportive of this Inquiry and its aim to get to the 14 bottom of what went wrong, saying and finishing: 15 "POL will do all it can to help the Inquiry 16 achieve that." 17 From our clients' point of view, the statements 18 and actions of the Post Office demonstrate that they are 19 not contrite, lessons have not been learned and 20 I suggest that no-one would bet against the next target 21 for the Post Office blame game being this Inquiry. 22 Our clients were not liars, not con artists and 23 not incompetent. The Post Office's Horizon System was 24 foisted upon subpostmasters and mistresses. Post Office 25 and Fujitsu knew it was not fit for purpose and never 69 1 was. They lied about its robustness and blamed, over 2 decades, subpostmasters for their own failures. We 3 suggest that Post Office knowingly ruined lives, sullied 4 reputations, broke many subpostmasters and tried to 5 break the rest. 6 Having heard the evidence in Phase 3, our clients' 7 views could be summarised in this way: we told you so, 8 they never learn and they don't listen. 9 The evidence, sir, in Phase 3 has confirmed what 10 our clients have long known: Post Office didn't provide 11 any adequate training on balancing and failed to ensure 12 that the Horizon Helpdesk provided any sort of 13 meaningful assistance when things started to go wrong. 14 It was always inevitable that things were going to go 15 badly wrong. We know this, for example, from the 16 evidence of Mr Parker and others. The system was 17 patched together to keep it limping along because nobody 18 wanted to spend money to rewrite the EPOSS program. 19 If the paucity of training and assistance issues 20 were not bad enough, there was a sting in the tail for 21 subpostmasters, the IMPACT programme, which effectively 22 programmed out the subpostmasters' remaining chance to 23 dispute phantom Horizon shortfalls. On behalf of our 24 client group, we highlighted this issue to the Inquiry 25 upon reading the statement of Susan Harding and we asked 70 1 the Inquiry to look closely at her evidence and the 2 evidence relating to IMPACT because this programme 3 encapsulated everything that was wrong about POL's 4 treatment of subpostmasters. 5 Sir, as you will recall, the IMPACT programme 6 abolished the local suspense accounts and in doing so 7 forced subpostmasters to accept all demands made of them 8 on pain of no longer being able to trade. This created 9 an impossible situation for subpostmasters, the 10 equivalence of: heads, you pay; and tails, you pay. 11 Our clients commented on the evidence relating to 12 the IMPACT programme. Kevin Palmer, said this: 13 "We never stood a chance. They dealt the cards, 14 checked the deck, took all the aces and left us the 15 jokers." 16 Sally Stringer says: 17 "It was their way of making sure that the branch 18 office paid, regardless of the circumstances." 19 Ms Harding gave evidence before you on 22 February 20 and confirmed that: 21 "Post Office thought that subpostmasters were 22 using suspense accounts to hide discrepancies instead of 23 resolving them." 24 Ms Harding acknowledged on 22 February that one of 25 the aims of the project, the IMPACT project, was to 71 1 pursue losses and push subpostmasters harder, in order 2 to pursue debt recovery. She gave us an insight into 3 what POL thought of subpostmasters, at page 30 of the 4 transcript of that date, 22 February, in answering the 5 questions from Mr Beer. Mr Beer asked: 6 "Did you have a mindset in the entirety of your 7 time working for the Post Office that the suspense 8 accounts was being used by dishonest subpostmasters to 9 hide and cover up money that they were taking? 10 Answer: "My mindset was that it was a place where 11 they could do that." 12 Question: "And did do that?" 13 Ms Harding said: 14 "Yes and, did do that." 15 Ms Harding also confirmed that the original idea, 16 as put to her by Mr Beer, was get rid of these 17 subpostmasters heading discrepancies in the suspense 18 accounts and make sure they're liable for all 19 shortfalls. It is abundantly clear that the Post 20 Office's institutional view of subpostmasters was that 21 they were dodgy and on the take. In her statement, 22 which, sir, as you will recall, we established in her 23 evidence that she wrote herself from her own 24 recollection, she set out the IMPACT programme design 25 parameters. 72 1 Paragraph 18, she referred to: 2 "The principle objectives of IMPACT were to reduce 3 losses and improve debt recovery." 4 In her statement and her evidence she made it 5 clear that the concept and high level designs were 6 developed through a series of workshops involving 7 Fujitsu and Post Office experts and user 8 representatives. 9 At paragraph 31 of her statement: 10 "It was agreed during the design of IMPACT that 11 the suspense accounts would be removed as, historically, 12 it was used by subpostmasters to hide discrepancies in 13 their accounts rather than resolve them." 14 The IMPACT programme started in 2003. It went on 15 through various iterations and discussions to about 16 2006. It was put into effect. It is an important part 17 of the evidence, as it brings together the different 18 parts of the thinking that was being used by the Post 19 Office in its approach to subpostmasters. 20 Our client,[GRO], questions -- [GRO] says, "How 21 could the Post Office dare to suggest this", that the 22 allegation against subpostmasters generally, that they 23 were dealing with matters in this way and hiding matters 24 in the suspense account is disgusting. 25 Helen Walker Brown: she says she's deeply 73 1 aggrieved that she was deprived of an option to reject 2 what the system said. She says that the decision to 3 remove the local suspense account was unfair and 4 downright immoral. 5 I said earlier that the Post Office were the 6 senior partner. We can see that. Fujitsu's client, the 7 Post Office, set the goals, as we saw from the Fujitsu 8 version of the IMPACT documents. I won't ask it to go 9 on the screen I will just refer to the reference number 10 for the moment, FUJ00098169. That's the "Fujitsu 11 Services Input to Feasibility Study for End to End 12 Re-architecting of Post Office Systems" dated 24 March 13 2003. That document refers to the goals and business 14 drivers behind the E2E project. 15 Paragraph 3.2.4: 16 "The following key business priorities were 17 identified: simplify identification of debt; reduce the 18 amount of reconciliation; increase the amount of debt 19 recovered; and put the emphasis on clients and customers 20 to validate the data." 21 At 3.2.4: 22 "In recognition of these priorities, this project 23 addresses specific requirements beyond these business 24 drivers and issues which were refocused on debt recovery 25 (financial recovery of money), target 95 per cent." 74 1 Our clients would like to know who is ultimately 2 responsible for the IMPACT programme. Ms Harding, you 3 may recall, referred to being instructed by 4 Ms Cruttenden and Peter Corbett, and we will address 5 those issues a little bit more in our written 6 submissions. 7 Ms Harding's evidence shows that the Post Office 8 had a twin mindset in respect of subpostmasters which 9 pre-existed the IMPACT programme and was dictated to 10 Fujitsu as its client instruction. Those were these: 11 that SPMs (subpostmasters) were liable for shortfalls 12 and that SPMs were fundamentally dishonest. The same 13 twin mindset also drove the Post Office's conduct in the 14 scandal, from the first demands of payment arising from 15 Horizon shortfalls shortly after rollout, until matters 16 were exposed in the findings made by Mr Justice Fraser 17 in 2019. 18 You will recall recently the evidence of Mr Ismay 19 on 11 May. He remained of the view that subpostmasters 20 were contractually liable for all shortfalls. In other 21 words, this is not a historic view. Mr Ismay left the 22 employ of the Post Office in 2016. 23 What we have seen is an "event storm", and I adopt 24 the term used by Ms Chambers, of bullying, institutional 25 entitlement and downright dishonesty. It is notable 75 1 that subpostmasters were looked down upon by the Post 2 Office and treated in a completely different way to 3 Crown Office employees, apparently justified by the 4 difference between subpostmasters' agent status versus 5 Crown branch employees. 6 The evidence, we suggest, in Phase 3 demonstrates 7 that the Post Office and Fujitsu both knew that the 8 Horizon System contained bugs, errors and defects. It 9 is this aspect of the Post Office's behaviour, the fact 10 that Post Office employees knew all along that Horizon 11 was fundamentally flawed and unreliable, that makes the 12 scandal so truly shocking. 13 Trevor Rollason told the Inquiry on 20 January 14 2023 that it was common knowledge at the Post Office 15 that SPMs were having problems with balancing. 16 The evidence of Gary Blackburn, a former Post 17 Office National Network Business Support Centre team 18 leader and problem manager. Mr Blackburn confirmed that 19 the Post Office was aware of the bugs, errors and 20 defects within the Horizon System and that there was 21 an active exchange of information between senior staff 22 and Gareth Jenkins at Fujitsu. 23 Mr Blackburn was no stranger to defects within the 24 Horizon System. He confirmed in his oral evidence on 25 28 February that he was aware of the ability of Horizon 76 1 to create discrepancies. Mr Blackburn was also aware of 2 the Callendar Square bug and the risk of branches being 3 impacted. He was another Post Office employee who had 4 sight of the email dated 23 February 2006 from Anne 5 Chambers, Mrs Chambers, concerning a Riposte problem. 6 In that email, Mrs Chambers said that the problem had 7 been around for years and affected a number of sites 8 most weeks. 9 Mr Blackburn later became aware of four or five 10 post offices having the same problem and escalated the 11 matter to problem management. It is, sir, we suggest 12 absolutely beyond doubt that the Post Office knew what 13 was going on. Any submission or representation made by 14 the Post Office that they were kept in the dark by 15 Fujitsu should be firmly rejected. 16 You may have noted that there was a tendency from 17 Post Office witnesses within Phase 3 to say that they 18 were unaware of the problems with Horizon at the time 19 but, with the benefit of hindsight, accept that the 20 system was not robust. For example, Chris Gilding 21 a former field team leader at POL, typifies this mindset 22 in his evidence on 13 January, when he rejected 23 statements to the effect that the computer was the 24 problem, not the subpostmaster. 25 Mr Gilding told the Inquiry that he took the view 77 1 that he had no evidence to suggest otherwise and made no 2 enquiries as to the reliability of the data that the 3 system was producing. 4 Bruce McNiven, Deputy Director of the Post 5 Office's Programme Delivery Authority, told the Inquiry 6 on 10 January that he understood from the fact that 7 Horizon had reached acceptance that he could apply 8 a presumption of rectitude to the system. Anne Allaker, 9 formerly of Post Office Services, told the Inquiry on 10 1 March 2023 that there was a general view held by the 11 Post Office contractor advisers that Horizon could not 12 create discrepancies. Clearly, the general view was 13 wrong. 14 The Post Office expected subpostmasters and 15 mistresses to prove that the computer was at fault and 16 this was embedded into the IMPACT programme. Of course, 17 no-one bothered to consider whether postmasters could 18 possibly do this when they had been locked out of their 19 branches, their documents taken away from them and the 20 Horizon System was designed to prevent them challenging 21 the numbers it spewed out. 22 So a big question arises: why senior managers 23 within the Post Office failed to disabuse other key 24 employees and contract advisers of this fiction? It is 25 relevant to note that Ms Allaker later accepted in her 78 1 evidence that a dismissive approach from the top 2 filtered down. One of our clients, Ms Walker Brown, 3 says something similar. She says that staff further 4 down in POL may have believed the lies that the 5 hierarchy told her or told them. She refers to the 6 example of her own area manager simply ignoring her when 7 she was begging for help. 8 So why would what used to be called the "nation's 9 most trusted brand" act in this way? I said in my 10 opening submission in November 2021 that this was all 11 about money for the Post Office. This is supported by 12 some of the evidence we've heard in Phase 3. Stephen 13 Grayston, a former Post Office Change Manager gave 14 evidence on 27 February. He confirmed that the Post 15 Office was trading at a loss in 2003 and was in a dire 16 financial situation. He agreed that there was a need to 17 bring in cash and, sir, you will recall that such 18 references to trading losses were referred to within the 19 IMPACT documentation. 20 Brian Trotter, a Post Office Contract and Service 21 Manager told the Inquiry on 2 March that he felt like he 22 was under pressure from the Post Office to recover debt 23 and to gather money. He also confirmed that there were 24 performance-related targets. 25 In his evidence on 3 March, Andrew Winn accepted 79 1 that there was no incentive within the Chesterfield 2 office to seek out transaction corrections that would 3 have the effect of POL paying money to subpostmasters. 4 Richard Roll told the Inquiry on 9 March that Fujitsu's 5 primary aim was to keep the system running so it worked 6 and so that Fujitsu didn't suffer any penalties. 7 Now, Mr Roll also told the Inquiry and he said it 8 was widely accepted within Fujitsu that the Horizon 9 System was poor -- he, of course, used more colourful 10 language -- and that software issues were encountered on 11 a weekly basis. He said that the system needed 12 rewriting. 13 Now, we've heard from Mr Mik Peach very recently 14 about that and Mr Peach disputes that. Mr Roll recalled 15 that he was told by Mr Peach that this could not happen, 16 the rewriting couldn't happen, due to a lack of money or 17 resources. 18 The Post Office needed money to recover from the 19 financial losses, partly no doubt caused in relation to 20 the implementation of the Horizon System, but also to 21 cope with the challenges to its own business model by 22 changes within the marketplace. Fujitsu needed to keep 23 the thing on track to avoid the penalties which they 24 thought and they expected to flow from not having the 25 funds to put right a substandard product. Now, some of 80 1 our clients have pointed out that the Post Office didn't 2 seem to have a problem in accessing huge sums to defend 3 proceedings in the Group Litigation and they reasonably 4 consider that the funds came from monies that were 5 extracted from themselves. 6 Another client, Shane Johnson, has pointed out it 7 was always about reputation and securing new revenue 8 streams. We invite the Inquiry to make a finding that, 9 in addition to the financial motivation, one of the 10 reasons why the Post Office behaved so disgracefully was 11 that it was desperate to protect the Horizon System from 12 criticism, as its failure would be what has, in fact, 13 happened: a fundamental attack on the integrity of the 14 business, both financially and reputationally. 15 There have always been two scandals here. The 16 first is in relation to the appalling treatment of the 17 subpostmasters and mistresses and the second scandal is 18 the cover up. Phase 3 has been important because the 19 evidence has demonstrated that the Post Office pulled 20 out all the stops to blame subpostmasters for errors in 21 the system, rather than come clean and tell the truth. 22 Mrs Chambers told the Inquiry on 3 May that she was 23 aware of minuted discussions in which the Post Office 24 had maintained that they didn't want postmasters to know 25 about particular bugs in the system because they didn't 81 1 want to give opportunities for fraud if postmasters 2 became aware of certain issues. 3 So, again, we see the mindset within the Post 4 Office that subpostmasters were all, in some way, 5 criminally inclined. 6 Mr Blackburn referred to a February 2007 email 7 chain from Mr Jenkins, in which he was copied in, where 8 an issue had arisen which affected a possible 570 9 branches. The Relativity reference to that is 10 FUJ00121071. In relation to the same incident, an email 11 dated 5 February, on the same Relativity reference, from 12 Dave Hulbert, Mr Blackburn's line manager that time, 13 stated, and I quote: 14 "The dilemma for Gary approaching branches is 15 proactive but opens the risk of litigation in future 16 ie we're telling 570 branches that Horizon may have 17 caused a discrepancy. Low risk but a risk. Being 18 reactive doesn't feel right as we've caused the problem 19 for branches but this may be the right option in this 20 situation." 21 The desire within the Post Office to cover up was 22 also confirmed in the evidence of Andrew Winn on 3 March 23 where he confirmed that the view taken by POL was that 24 disclosure could provide branches with ammunition to 25 blame Horizon for shortfalls in relation to 82 1 discrepancies. 2 The following exchange, I will read in a moment, 3 between Mr Beer and Mr Winn is relevant to this. 4 Question by Mr Beer of Mr Winn: 5 Question: "It was seen in the light of 'we can't 6 disclose material that might undermine our system even 7 if the system is in fact faulty'?" 8 Mr Winn replied: 9 "Yes, I think that's probably a fair summation." 10 The extent of the duplicity that Post Office is 11 demonstrated by the 2010 whitewash report of Mr Rod 12 Ismay, a report which attempted to make the case that 13 all losses were caused by thieving subpostmasters. It's 14 important to remember that the report was commissioned 15 essentially as a response to allegations which had been 16 made in Computer Weekly in the preceding year. 17 In that report, Mr Ismay advised the Post Office 18 that they should not review the Horizon System in light 19 of the reports of bugs, errors and defects for two 20 reasons. Firstly, any review might lead others to think 21 that POL doubted the robustness of Horizon. Secondly, 22 a more sinister reason why Mr Ismay advised that the 23 issue should not be investigated is because the outcome 24 of any investigation would have to be disclosed in 25 proceedings, with the effect that prosecutions might 83 1 have to be stayed. 2 In other words, Mr Ismay was sufficiently aware of 3 the difficulties in relation to criminal proceedings, 4 that documents that might exist within the Post Office 5 might have to be disclosed if it led to certain issues 6 coming to light. Therefore, well, let's not do them. 7 Let's not investigate. Let's not have to worry about 8 disclosure of that paperwork. 9 Sir, you may recall that I took Mr Ismay, after 10 a brief discussion with him about his whitewash report, 11 to the document which I'm going to ask to go on the 12 screen, please, which is in relation to the receipts and 13 payments mismatch bug. The document reference is 14 FUJ00081584. I am very grateful. Hopefully, we should 15 be at page 2. 16 At the bottom of page 2, we have reference to 17 "Impact". I asked Mr Ismay to consider the bullet 18 points under the heading "Impact". This meant that the 19 bug, the mismatch bug, caused the branch to appear to 20 have balanced where, in fact, they could have a loss or 21 a gain: 22 "Our accounting systems will be out of sync with 23 what is recorded that branch. 24 "If widely known could cause a loss of confidence 25 in the Horizon System by branches. 84 1 "Potential impact upon ongoing legal cases where 2 branches are disputing the integrity of Horizon Data. 3 "It could provide branches ammunition to blame 4 Horizon for future discrepancies." 5 The actual impact of the mismatch bug meant that 6 the integrity of the branch data was affected without 7 branch or subpostmaster/mistress knowledge. The truth 8 is that the impact to the Post Office was that the 9 secret of bugs might get out and at all costs that must 10 not be allowed to happen. To any honest organisation, 11 prepared to face up to its own errors, the shock wave 12 generated by the mismatch bug should have been immediate 13 and devastating. Instead, the answer was to choose one 14 of three ways of trying to ensure containment, and this 15 is shown at page 3 of the document, under the heading 16 "Proposals for affected Branches". 17 On the screen at page 3 we have: 18 "Solution one -- Alter the Horizon Branch figure 19 at the counter to show the discrepancy. Fujitsu would 20 have to manually write an entry ... to the local branch 21 account. 22 "Impact -- when the branch comes to complete next 23 Trading Period they would have a discrepancy, which they 24 would have to bring to account." 25 The risk there describes: 85 1 "This has significant data integrity concerns and 2 could lead to questions of 'tampering' with the branch 3 system and could generate questions around how the 4 discrepancy was caused. This solution could have moral 5 implications of Post Office changing branch data without 6 informing the branch." 7 Something of an understatement. 8 Solution two: 9 "P&BA will journal values from discrepancy account 10 into the Customer Account and recover/refund via normal 11 processes. This will need to be supported by 12 an approved POL communication. Unlike the branch 13 'POLSAP' remains in balance albeit with an account 14 (discrepancies) that should be cleared." 15 Impact of that was described in this way: 16 "Post Office will be required to explain the 17 reason for a debt recovery/refund even though there is 18 no discrepancy at the branch. 19 "Risk -- Could potentially highlight to branches 20 that Horizon can lose data. 21 "Solution three -- it is decided not to correct 22 the data in the branches (ie Post Office would prefer to 23 write off the 'loss'. 24 Impact -- Post Office must absorb circa £20,000 25 loss. 86 1 "Risk -- huge moral implications to the integrity 2 of the business, as there are agents that were 3 potentially due a cash gain on their system." 4 We suggest, sir, that solutions 1 and 3 are little 5 more than proposals to conceal the truth, protect the 6 Post Office, deceive courts and commit fraud. Any 7 corporate body will want to protect its reputation and 8 image but the consideration by both the Post Office and 9 Fujitsu -- you will remember that these were joint 10 discussions -- the consideration by both the Post Office 11 and Fujitsu of these solutions demonstrates how far they 12 had strayed from any honest and lawful approach. 13 We thought about this. You think about what they 14 have put in writing within this documentation. One can 15 only guess what they said in the margins of these 16 meetings. 17 The insight provided by this document is simply 18 astonishing. Far from the picture which the Post Office 19 seeks to paint now (which is that Fujitsu practised an 20 operative deception on the Post Office, which apparently 21 it continues into this Inquiry), the raw truth is that 22 the Post Office planned the heist, gave the orders, 23 while Fujitsu brought the shooters to the scene. 24 Another means by which the Post Office sought to 25 keep the truth from subpostmasters was by failing to 87 1 inform them that Fujitsu had access to their terminals 2 and altered data. It was a feature of the Group 3 Litigation that both the Post Office and supporting 4 Fujitsu witnesses initially sought to deny to the court 5 that remote access had been possible. Mrs Chambers on 6 2 May confirmed she knew of cases where the Post Office 7 did not tell a subpostmaster that their financial data 8 had been altered. She said "Yes, I think that 9 definitely did happen". 10 Mr Richard Roll said in his oral evidence that 11 subpostmasters were sometimes not advised that their 12 data was being corrected and he was referred to his 13 evidence in the High Court where he had used the term 14 "hack" in order to describe the way that they'd 15 approached matters and getting into the Riposte system. 16 He said that, in some cases, they were simply told -- 17 this is subpostmasters -- that an error in their data 18 would be corrected. However, they were not told the 19 underlying reasons for the corrections. They were not 20 told that the action had been taken due to a bug within 21 the Horizon System. 22 Mr Blackburn from the Post Office accepted in his 23 evidence on 28 February that, as a matter of fairness, 24 subpostmasters should have been told that remote access 25 had been used to insert a transaction. 88 1 As a result of these actions of the Post Office, 2 very many subpostmasters would not have been aware that 3 their financial data had been remotely altered, neither 4 would they have been aware that the reason for Fujitsu 5 having to access their systems was related to defects in 6 the Horizon System. 7 Sir, we suggest that the evidence has shown that 8 the Post Office knew all along that the system was not 9 robust and they sought to keep the truth from the 10 victims of their behaviour, the subpostmasters and 11 mistresses. 12 At the close of Phase 1, I posed the question as 13 to whether what we might find within Phase 3 was "cock 14 up or cook up". My respectful suggestion is there can 15 be no doubt that the answer is "cook up". 16 I now turn to the question of Helpdesk scripts. 17 SIR WYN WILLIAMS: Can I just ask that the document be taken 18 down so I can see Mr Stein now. 19 MR STEIN: I'm very grateful and, sir, just on timing, 20 I anticipate I will be another five minutes. 21 SIR WYN WILLIAMS: Okay. 22 MR STEIN: So I turn to question of Helpdesk scripts and, in 23 essence, what we don't have. Yesterday, Ms Patrick, in 24 her extremely able cross-examination of Mr Peach 25 referred to documents FUJ00152299 and FUJ00080098 and 89 1 you, sir, will recall the questions being asked that 2 refer to inappropriate calls being made and being routed 3 back to the Post Office Helpdesk. Now, we've been 4 reflecting upon those points because we knew we were 5 going to refer to the question of Helpdesk scripts. 6 It is undoubtedly correct that now knowing that 7 the Fujitsu system had an embedded part of its protocol, 8 which is to refer back in a circular way to the Post 9 Office Helpdesk, that that yet again supports what we 10 have been trying to seek through this Inquiry, which is 11 disclosure of Helpdesk scripts. 12 It cannot be a coincidence that so many 13 subpostmasters and mistresses were told the same thing 14 over and over again by the helpline. 43 of our clients 15 were told by the helpline that they had no option other 16 than to pay the shortfall. 49 of our clients were told 17 by helpline staff that they were the only subpostmasters 18 who were experiencing shortfalls. 35 of our clients 19 were expressly told by helpline staff that they were 20 contractually liable for the shortfalls. 21 There can be no doubt scripts existed. A number 22 of the witnesses of Phase 3 have admitted in evidence 23 that helpline staff worked from scripts. It has now 24 reached the stage whereby Mr Beer, on behalf of this 25 Inquiry, has joined in our quest to ask witnesses "Where 90 1 are the scripts? How are they saved? Who is 2 responsible for them? Where can we find them?" Because 3 we are not getting anybody or any substantial number of 4 scripts through the disclosure process. That, as far as 5 we can tell and we are sure ourselves, is nothing to do 6 with this Inquiry. It is to do with the disclosure by 7 both Fujitsu and the Post Office of scripts which we 8 suggest they must have. 9 Mr Parker, Stephen Parker, in response to 10 questions from Mr Beer, told the Inquiry on 10 May that 11 he saw some of the scripts when they were compiled by 12 senior technicians within Fujitsu. In fact, sir, we 13 have disclosed very few scripts. It is inconceivable 14 that the vast majority of these scripts have disappeared 15 from two companies who both operated helplines. 16 On behalf of our client group, Howe+Co have been 17 calling for disclosure of the scripts since November 18 2022 and it is a matter of real concern to our clients 19 that virtually no disclosure of scripts have been the 20 forthcoming. One of our clients -- the name, in fact, 21 has been anonymised -- says that the continuing failure 22 to disclose is underhand and shows that things are still 23 being hidden by the Post Office and Fujitsu. 24 We ask that the Inquiry use all the powers at its 25 disposal, under sections 21 and 35 of the Inquiries Act 91 1 2005, to require the Post Office and Fujitsu to disclose 2 this material or face the consequences of 3 non-disclosure. 4 Now, that may mean, once scripts have been 5 identified and disclosed, that we may request that some 6 of the Phase 3 witnesses be recalled. 7 I now, very briefly, turn to looking forward to 8 the evidence in Phase 4. Before I do, may I sincerely 9 thank the Inquiry, the Inquiry team and all of the 10 members of the Inquiry team that sit behind my learned 11 friends who ask these questions, that have been spending 12 so very many hours working hard to present the evidence 13 that has been provided with such expertise before this 14 Inquiry. 15 We ask that the Inquiry should have at the 16 forefront of its considerations within Phase 4, that one 17 of the core requirements of the system was for data 18 produced by Horizon to be available and sufficient to 19 support investigations and prosecutions. Sir, we've 20 looked at many a time the contractual requirement in 21 that regard, one example of that is within FUJ00000087, 22 under "Prosecution Support". 23 Clearly, we say, Horizon was not capable of 24 supporting prosecutions or civil actions, as both the 25 Post Office and Fujitsu well knew. The system had 92 1 failed in one of its core objectives and this placed POL 2 in a dilemma. 3 Sir, you will recall the questions I asked of 4 Mr Peach yesterday, "Why was it you were not aware of 5 the contractual requirement?" Well, all he could do was 6 refer it to his many managerial staff and say that, 7 well, they should have brought it to him because he 8 recognised it's important. 9 Now, the Post Office had two options. Option A 10 was to come clean and disclose problems with Horizon to 11 the subpostmasters and the courts. Option B was to 12 continue to pursue subpostmasters for alleged shortfalls 13 and bring abusive prosecutions based upon the fiction 14 that the system was robust. Perverting the course of 15 justice is what that is. 16 It was unthinkable to the Post Office that the 17 truth should come out, the truth being that the Horizon 18 System was incapable of supporting a prosecution or 19 a civil claim. We must remember that the Post Office 20 had other dependent contracts with other companies, such 21 as the Bank of Ireland, that might well have been 22 affected. This impacted on the Post Office's approach 23 to disclosure in criminal and civil cases against 24 subpostmasters. Full disclosure was never an option in 25 these cases. But there was also the Post Office's 93 1 internal belief that subpostmasters were fundamentally 2 dishonest. 3 This created a double whammy of paranoia within 4 the Post Office. They backed themselves into a corner. 5 We suggest that there has been a continuing failure of 6 disclosure into this Inquiry. We've cited one example 7 in relation to scripts. We suggest that the Post Office 8 does not want the full truth to come out. As we all 9 know, one of the reasons we're here today is because the 10 Post Office decided not to be open and honest and chose 11 the second option, option B, to continue to prosecute 12 subpostmasters, to abuse the process of the courts. 13 Our clients welcome the following phases of this 14 Inquiry and have asked that we convey the wish to this 15 Inquiry that it continues to focus on identifying and 16 calling to account the individuals who were responsible 17 for this scandal, as well as the institution itself. 18 We suggest, sir, that lies won't work for the Post 19 Office any more. The truth is coming out. Our clients 20 are still here and they intend to have and they will 21 have justice. 22 Thank you for listening to my closing remarks and 23 I apologise for going, I think, seven minutes over time. 24 SIR WYN WILLIAMS: All right, Mr Stein. My personal view, 25 but I'm prepared to be persuaded out of it, is that we 94 1 should have more than just a short break. I am quite 2 prepared to truncate lunch to some extent but I don't 3 think just having ten minutes and then carrying on is 4 what we should do, since there is at least 5 a possibility, I put it no higher, that the time 6 estimates get longer rather than shorter. 7 MR BEER: Sir, I can't speak on behalf of either of my 8 friends. They gave estimates, I think, of 15 minutes 9 each. I don't know whether that's changed but, sir, it 10 is 1.20 now. Maybe 1.50? 11 SIR WYN WILLIAMS: Yes, that's fine. 12 MR MOLONEY: Just to be clear, our time estimate was 13 25 minutes but 1.50 is absolutely fine for me, sir. 14 SIR WYN WILLIAMS: We will begin the afternoon session at 15 1.50. 16 MR BEER: Thank you very much, sir. 17 (1.21 pm) 18 (Luncheon Adjournment) 19 (1.54 pm) 20 MR MOLONEY: Good afternoon, sir, can you see and hear me. 21 SIR WYN WILLIAMS: Yes, I can. Before you start Mr Moloney 22 I just wanted to ask Mr Stein a question. 23 I didn't want to interrupt him while he was 24 speaking to me but he referred on at least one, if not 25 more than one occasion, to written submissions which led 95 1 me to believe he may be talking about written 2 submissions to supplement his oral submissions. Is that 3 correct, Mr Stein? 4 MR MOLONEY: Sir, Mr Stein is not in the room at the moment 5 but Mr Jacobs is and he will answer your question 6 I anticipate, as he's always ready to. 7 MR JACOBS: Sir, yes, Mr Stein has had to go away this 8 afternoon. We understood that we have had 9 correspondence with the Inquiry team that there will be 10 an opportunity to put in written submissions as well as 11 make oral submissions. So we have applied a belt and 12 braces approach. 13 SIR WYN WILLIAMS: Hang on a minute. Have I ever said 14 that's permissible? 15 MR JACOBS: I am told that there was a letter that gave us 16 the deadline of 25 May to put in written submissions. 17 SIR WYN WILLIAMS: I'm not trying to be unduly difficult but 18 this process of making submissions at the end of each 19 phase came about, I remember, Mr Stein putting it 20 persuasively "We'd like to make short oral submissions". 21 I, at the end of Phase 2, gave the parties, the Core 22 Participants, the opportunity of making either oral 23 submissions or written submissions and I hadn't been 24 aware that I had said anything different to that for 25 Phase 3. 96 1 If a letter has gone out from the Inquiry which 2 suggests that you can do both, then unless Mr Beer is 3 going to contradict me, I think I'm going to say that 4 I didn't authorise anybody to say that. I'm not really 5 very keen on the idea because the idea is that you have 6 a choice. 7 MR JACOBS: I am told that -- 8 SIR WYN WILLIAMS: You're mute, Mr Jacobs. 9 MR JACOBS: I am told that a letter was sent out. I can ask 10 those who instruct me to liaise with the Inquiry staff 11 but, sir, quite clearly it's your direction that -- 12 SIR WYN WILLIAMS: I wanted to raise it now because 13 I wouldn't want you to spend unnecessary time writing 14 closing submissions, as well as making oral submissions, 15 if that wasn't really what I had in mind but anyway does 16 anybody else -- any of the other Core Participants 17 present think that I have or the Inquiry has altered the 18 process which was in place for Phase 2? 19 MR HENRY: Sir, can I quickly mention something. We're 20 happy to follow your approach and abandon, if necessary, 21 oral submissions but the reason why we gave a time 22 estimate of 15 minutes was just to headline what we 23 wished to develop in writing and that was based upon 24 a communication. 25 I don't want to, in any way, embarrass the person 97 1 who sent it but it came from the Inquiry team and it's 2 dated 4 May to Mr Schwarz at 15.08, and it reads: 3 "I have confirmed with counsel and solicitors to 4 the Inquiry that the Inquiry will accept all written 5 closing submissions. If written closing submissions 6 have been provided by an RLR, it is unlikely they would 7 need to make oral closing submissions because all 8 written submissions will be taken into account. 9 However, RLRs will not be prohibited from doing so." 10 So, sir, we took that -- 11 SIR WYN WILLIAMS: Well, that's fair enough, Mr Henry. 12 I understand why you took that and there we are. I am 13 not going to go behind what that says, in that event, 14 but I'm not for a minute suggesting that I am going to 15 allow that to be the default position henceforth. I'll 16 discuss it with my team and from this phase onwards it 17 may revert to what happened in Phase 2. 18 MR HENRY: Thank you, sir. If it does, sir, there will be 19 no complaint from us and we're sorry for any, as it 20 were, misapprehension. 21 SIR WYN WILLIAMS: That's fine. These things happen. It 22 just means more work for the legal representatives and 23 more reading for me. There we are. 24 MR JACOBS: If I could perhaps add, sir, that our written 25 submissions will largely echo what Mr Stein said this 98 1 morning, so that there won't be duplication. 2 SIR WYN WILLIAMS: I really would ask you not to do that, 3 since we have a transcript of it and just confine 4 yourselves to making additions. 5 MR JACOBS: Absolutely, sir, yes. 6 SIR WYN WILLIAMS: Fine, thank you. 7 Over to you, Mr Moloney. 8 Closing statement by MR MOLONEY 9 MR MOLONEY: Thank you very much, sir, and indeed thank you 10 very much, sir, for the invitation to make short closing 11 submissions for this phase. The extensive evidence 12 we've heard frequently echoed the experiences of the 13 Core Participants we represent. 14 We will confine our submissions today, sir, to 15 that evidence heard in Phase 3 but just one reflection 16 on Phase 2, that the Inquiry heard submissions on the 17 evidence of bugs, errors and defects which were visible 18 during Horizon's development and rollout. As we said 19 before, there is clear evidence on which the Inquiry 20 could conclude that, even from 1999, Horizon was so 21 politically and commercially significant for both Post 22 Office and Fujitsu that it could not be allowed to fail. 23 Post Office, as we heard, had to be strategically 24 Horizon-centric. It had to be made to work. 25 The evidence in Phase 3 is such that the Inquiry 99 1 may conclude that the bugs, errors and defects in 2 Horizon cannot be and could never reasonably have been 3 written off as teething problems in the new system or 4 simply part and parcel of the accepted operation of any 5 complex computer program. 6 The Inquiry may conclude that the problems arising 7 in the operation of Horizon had been well signposted in 8 its development when the poor product quality had been 9 evidenced and, to a degree, acknowledged by Fujitsu and 10 Post Office. 11 There's a lot to cover from the extensive evidence 12 of Phase 3, sir, and today we briefly address each of 13 the five topics designated for exploration within 14 Phase 3, those being: (a) training; (b) assistance and 15 support; (c) resolution of disputes; (d) modifications; 16 and (e) knowledge of bugs, errors and defects. Then we 17 will very briefly look to Phase 4 and the issues we 18 invite the Inquiry to explore further back of Phase 3. 19 When making these submissions, sir, we'll provide 20 references for the record but we'll only ask for one 21 document to be shown. So, firstly, taking the five 22 topics in turn: 23 Training. 24 We highlight only two issues. Firstly, training 25 on Horizon was known to be inadequate. In 2011, 100 1 a decade after the national rollout of Horizon, feedback 2 remained that training on the balancing process was 3 inadequate. Chris Gilding provided that feedback on 4 behalf of trainers, to the effect that training was 5 inappropriately focused on sales, rather than on 6 balancing and discrepancies. That was not the view of 7 senior management, according to Mr Gilding, who were 8 "striving to keep branches afloat by generating new 9 income streams". 10 That's the transcript of 13 January at page 66. 11 Secondly, what was termed residual fallout. Post 12 Office and Pathway planned from the outset that some 13 postmasters simply would never absorb Horizon. There 14 would be residual fallout which Post Office would need 15 to address. That was built into the annex to the second 16 supplemental agreement. The Inquiry might consider just 17 how postmasters who did not absorb Horizon were to be 18 treated by Post Office. 19 (b) Assistance and support. 20 A number of propositions arise from the evidence 21 heard in Phase 3, sir. First, the support being 22 provided to postmasters on the front line, whether by 23 the HSH or the NBSC, was limited. We ask whether 24 investment in front line support was commensurate with 25 what was known to both organisations about the flaws 101 1 which arose in Horizon's development. 2 Second, the NBSC was reliant upon Fujitsu front 3 line support to investigate any alleged system problem. 4 HSH was limited by their knowledge base, by KELs and 5 scripts and essentially restricted to the consideration 6 of known problems and, if there was no known problem, 7 the postmaster was told to carry out further checks. 8 Thirdly, each layer of support was designed to act 9 as a filter against escalation of a problem to the next 10 level of support and the Inquiry may wish to consider 11 whether the HSH and SMC were explicitly or implicitly 12 discouraged from escalating problems for further 13 investigation. 14 (4) If a problem is not escalated beyond HSH 15 there could be no further investigation of any 16 previously unknown problem. 17 (5) In Legacy Horizon bugs, errors and defects in 18 Riposte were subject to limited investigation within 19 Fujitsu. Escher had to be involved and the Inquiry 20 heard of issues parked for weeks and months on end and 21 some closed without any root cause identified. We 22 suggest that the Callendar Square bug is illustrative. 23 Riposte lock incidents had been recorded from rollout 24 with a purported fix not implemented until 2006, and it 25 appears that instances of this Riposte lock problem may 102 1 have continued until the migration to Horizon Online. 2 (6) Many issues were discounted as one-offs and 3 left without further exploration or explanation. 4 Mr Peach yesterday, for example, referred to "a single 5 occurrence of an anomalous event somewhere in the system 6 being known as an incident, rather than a problem". 7 That's at page 100 of yesterday's transcript, sir. 8 When a problem did get to the SSC, there could be 9 limited investigation of any root cause, where there was 10 no obvious reason for a discrepancy and/or no means of 11 reproducing a problem. Mr Peach suggested to the 12 Inquiry that any problem that it was believed could be 13 rooted in the code would be escalated up. However, he 14 also confirmed that this could be a difficult call and 15 within the judgement of the technician dealing with the 16 call to determine whether the evidence available was 17 sufficient. 18 The Inquiry also heard, though, from Mrs Chambers 19 who, by all accounts, was recognised as the technical 20 specialist in the SSC for counter problems. On her 21 evidence, postmasters would be asked to provide further 22 evidence of a problem before system engineers would 23 investigate further. We ask how postmasters could be 24 expected to investigate a technical problem further 25 without expertise or specialist support. 103 1 Indeed, the transcript of 2 May at page 75 reveals 2 that Mrs Chambers was asked what would happen when she 3 thought there was insufficient information to explore 4 an alleged error further, and she said: 5 "I sort of assumed that perhaps somebody within 6 the Post Office organisation would go and help the 7 postmaster to discover where something might be going 8 wrong." 9 In that vein, she added that she was unaware that 10 a postmaster might then be liable for a loss, at least 11 until she became aware of cases going to court from 12 around 2005. 13 (8) The diagnostic information available to both 14 SSC and to the postmaster was limited. Both Mr Peach 15 and Mrs Chambers confirmed that there was apparently 16 nothing in the information available either to the 17 postmaster or to the SSC which would track every 18 keystroke or screen touch recorded by Fujitsu as 19 undertaking in branch. 20 (9) Importantly, there was a general bias towards 21 a conclusion of user error. This was apparently built 22 into the guidance being given to the front line support. 23 During the evidence of Mrs Chambers we looked at 24 FUJ00082302. That's the transcript of 2 May at 25 page 165, where Mrs Chambers, dealing with one PEAK, 104 1 said that: 2 "SSC do not accept credible calls until all 3 information on the investigation performed by NBSC is 4 detailed and they've advised why they believe there's 5 a software fault. When responding, if they have given 6 specific examples that you can explain, do so. 7 Otherwise, make clear that it is not a system problem 8 (assuming you've checked that everything adds up)." 9 She suggested that the Helpdesk personnel should 10 look at a different PEAK for an example response which 11 may help with the wording. 12 When pressed as to why she might have closed out 13 as "user error" an issue identified by NBSC through 14 auditors on site as not being a result of postmaster 15 error, Mrs Chambers accepted that she was not happy with 16 how she handled that problem. When asked how she could 17 reach that conclusion, she said because the numbers 18 added up. This response and her simple calculations 19 would, of course, have depended entirely on the numbers 20 produced by Horizon being reliable and accurate and 21 no-one, it appears, was required to question the 22 reliability of those figures in the face of evidence in 23 the statement by the postmaster user that the figures 24 were wrong? Even when apparently corroborated by 25 on-site auditors. 105 1 She said: 2 "My job was to try to identify system errors and, 3 you know, you can't, I think, turn around and say, 'Oh, 4 well, it might be a system error but I can't find it', 5 not in a case where there's, you know, there is so much 6 variability, shall we say, on the customer side." 7 The default, we say, sir, was postmaster error. 8 Finally, in this section, where a problem was 9 identified and diagnosed, a problem would in principle 10 be passed by SSC to the MSU who could issue a BIM for 11 the POL team to address any financial discrepancy. Each 12 of those individual steps needed to be taken properly if 13 a postmaster were not to be held liable for a false 14 debt. The Inquiry has seen at least one example of 15 a postmaster continuing to chase the HSH for support in 16 circumstances where a system error has been identified 17 and no correction issue issued by POL. That example is 18 POL00000996, transcript of 3 May at page 137. 19 The Inquiry heard that there was at times 20 a considerable backlog of transaction corrections 21 including after the introduction of IMPACT, again 22 reflecting the anxiety and trauma experienced by some 23 postmaster Core Participants waiting in vain, quite 24 often, to hear whether a transaction correction would or 25 would not be issued in respect of a considerable 106 1 discrepancy, as again outlined by postmaster Core 2 Participants in Phase 1 of the Inquiry. 3 (c) Resolution of disputes. 4 We address one issue under this heading, touched 5 upon by Mr Stein during his submissions, and that's 6 IMPACT and the removal of the facility to place disputed 7 discrepancies in suspense. 8 The Inquiry heard that that feature had not formed 9 part of the original IMPACT design to remove the 10 suspense facility. The original design had envisaged 11 a dispute process which never materialised, which was 12 known as "Do nothing, requesting further investigation". 13 The evidence may suggest that the default position 14 within Post Office was that, overall, postmasters were 15 assumed liable for any unexplained discrepancy, unless 16 they could prove to the contrary. After the changes 17 wrought by IMPACT it appears that any dispute could only 18 ever follow acceptance of a debt by an individual 19 postmaster. 20 We suggest, sir, that the way this decision was 21 taken might provide considerable cause for concern on 22 the part of the Inquiry. The apparent reason for the 23 removal of suspense as an option on rollover is 24 disturbingly lacking in logic, being built on a faulty 25 presumption that use of suspense may be indicative of 107 1 fraud or dishonesty by postmasters. 2 From FUJ0012603 and FUJ00126038, we can see that 3 Mr Marsh said that: 4 "Those of a more historic mindset will exploit the 5 facility." 6 We also have the evidence of Sue Harding. We say 7 that this appears very much at odds with any common 8 sense understanding of suspense as a means of 9 highlighting funds held under pending dispute because we 10 ask whether any truly dishonest postmaster would flag or 11 suspend any sum. 12 Both Post Office and Fujitsu were aware that only 13 10 per cent of all discrepancies reported were true debt 14 and that figure is repeatedly stressed in IMPACT policy, 15 which identified the key drivers for the policy 16 including debt recovery. That's at POL00038878 on the 17 transcript of 27 February, page 22. It can also be 18 found in the evidence of Sue Harding and Philip 19 Boardman. 20 Yet there appears to have been a persistent 21 corporate default that unexplained discrepancies were 22 debts liable for recovery and postmasters were 23 responsible for those debts, whether as a result of 24 incompetence or dishonesty. The Inquiry will recall the 25 evidence of Mrs Harding and may have gained the 108 1 impression that she had an initial reluctance, even now, 2 to accept that there were bugs, errors and defects in 3 Horizon. She continually said, "if Horizon was wrong", 4 and that's transcript of 22 February at page 28. 5 We would ask the Inquiry, please, in those 6 circumstances, to keep an eye on whether the enthusiasm 7 for debt recovery illustrated in the policy papers for 8 IMPACT was then reflected in the prosecution policy 9 pursued by Post Office. The Inquiry heard that, while 10 debt recovery was a key trigger for IMPACT, debt 11 recovery was also a core focus for the risk team at Post 12 Office. Mr Ismay was expressly asked about a vigour for 13 the suspension of postmasters and evidence of 14 satisfaction in the use of the Proceeds of Crime Act 15 2002 to secure the recovery of substantial sums for the 16 business. Mr Ismay was asked about the auditing report 17 prepared in October 2004, which identified a concern 18 that: 19 "In spite of the size of amounts of discrepancies, 20 a precautionary suspension was not made in 35 per cent 21 of cases." 22 That's the transcript of 11 May, page 30. 23 He was also asked about the minutes of the risk 24 and compliance committee in January 2005, identifying 25 POCL recovery of £1.2 million over a part-year. That's 109 1 at page 44 of the same transcript. 2 So the Inquiry may decide to treat sceptically the 3 assertion by Mr Ismay that, while debt recovery was 4 relevant to IMPACT and also relevant to the increased 5 focus on investigation of postmasters, both were 6 separate issues and considerations. 7 (d) Modifications. 8 Fixes and new releases could introduce new bugs, 9 errors and defects and, from FUJ00094958, we see the 10 quote that: 11 "Quite simply there have been too many incidents 12 where poor execution of change has caused a problem in 13 live." 14 The Inquiry may consider this risk of regression 15 particularly concerning in the light of the conclusions 16 of the EPOSS Task Force and what it identified as a code 17 fix culture which could result in regression. 18 During the development of Horizon Online, there 19 was again pressure on both organisations to explain away 20 bugs, errors and defects and to reduce their impact from 21 high to acceptable. We highlight one issue in this 22 area, sir. When faced with a very substantial defect in 23 the Horizon Online product in development, it was 24 apparently critical to POL that Fujitsu provided 25 a definitive statement including: 110 1 "... a commitment to support POL in proving the 2 integrity of the system in any subsequent legal action 3 (specifically where the difference in the two reports is 4 used as a means to challenge the integrity of the 5 system)." 6 It appears that a caveated commitment of that 7 nature was given by Fujitsu. 8 Finally, in the review of Phase 3, sir, we turn to 9 knowledge of bugs, errors and defects. 10 A number of propositions are supported by the 11 evidence in Phase 3. Firstly, bugs, errors and defects 12 continue to appear throughout the rollout of Horizon and 13 beyond. Bugs, errors and defects appeared in both 14 Legacy Horizon and Horizon Online. Secondly, and very 15 importantly, there had been limited dissemination or 16 appreciation of the knowledge of bugs, errors and 17 defects identified in development to those individuals 18 working in support. 19 At a senior level, there was an awareness that the 20 genesis and development of Horizon had been difficult. 21 For example, Mr Muchow was aware of the EPOSS Task Force 22 report and the decision not to rewrite the application. 23 But down below, with those involved in support, there 24 was no such appreciation. The implications of this lack 25 of institutional memory are clear in the evidence both 111 1 of Mr Parker and of Mr Peach. Both were responsible for 2 critical support provided through the SSC but neither 3 appears to have conducted any ongoing monitoring or 4 assessment of the quality of the product, as it appears 5 was explicit in the management decision taken on 10 May 6 2000 not to rewrite the EPOSS application. 7 Further, as Mr Peach made clear, in the absence of 8 knowledge of the EPOSS Task Force report, conclusions on 9 the poor quality of the product, his baseline 10 understanding was founded solely on the fact of 11 acceptance of it by Post Office. The Inquiry may 12 consider that position, that the fact that the product 13 had been accepted must mean that it was of satisfactory 14 quality, profoundly unfounded. 15 He did not, on his own admission, have the 16 technical expertise to make his own independent 17 assessment of poor quality and the Inquiry might 18 consider, had he been told about the EPOSS Task Force 19 conclusions, whether this might and perhaps should have 20 informed the work on his team. Would they have 21 subjected the operation of the system to greater 22 scepticism or treated complaints from postmasters with 23 greater credibility? 24 Third, not all bugs, errors or defects seen or 25 addressed by Fujitsu during the life of Horizon are 112 1 identified in the Horizon Issues judgment. 2 Next, not all bugs, errors or defects were picked 3 up by the automated processes designed by Fujitsu. 4 Sometimes bugs, errors and defects were first identified 5 in the complaints raised by postmasters and we have that 6 from the evidence of Mrs Chambers and John Simpkins. 7 Bugs were not limited to one aspect of the system 8 and the Inquiry has at least one example of a known bug, 9 error or defect which impacted upon the ARQ data, which 10 was previously believed to be unassailable. There 11 appears to be little indication that the impact of this 12 bug was substantially investigated. There were bugs, 13 errors or defects which could take months, if not years, 14 to investigate and to identify a root cause and there 15 were some bugs, errors and defects where a root cause 16 was never investigated or identified. 17 Importantly, if a bug, error or defect impacting 18 on a branch could not be replicated by Fujitsu it might 19 never be investigated or identified. 20 It was also a known common problem, at least 21 within Fujitsu, that some calls were being bounced back 22 had forth between HSH and NBSC. We see that from the 23 transcript of 16 May at page 85, dealt with yesterday by 24 Ms Patrick. The Inquiry may conclude that this reflects 25 the experience of postmaster Core Participants, 113 1 including those we represent, and it might be thought to 2 be a particular problem when Fujitsu's automated 3 processes were not picking up all bugs and Fujitsu was 4 only becoming aware of them when postmasters highlighted 5 them. 6 Importantly, the Horizon Helpdesk was penalised 7 for passing some problems up to the SSC, which 8 ultimately led to "too many calls being closed without 9 proper investigation or resolution", and that's 10 FUJ00152299, yesterday's transcript at page 85. That 11 concern was raised expressly by Mrs Chambers but was not 12 addressed in a later joint policy document produced in 13 September 2008, and that's FUJ00080096. 14 Both Post Office and Fujitsu were aware or ought 15 to have been aware that civil and criminal liability for 16 postmasters could flow directly from a discrepancy in 17 the accounts produced by Horizon and, if not by 2007 18 when Mrs Chambers wrote her afterthought, then by 2010 19 at the latest, both organisations were demonstrably 20 aware of the impact which bugs, errors and defects might 21 have on ongoing litigations and prosecutions. 22 For example, the Inquiry has repeatedly considered 23 the notes prepared in the discussion of the receipts and 24 payments bug in September or October 2010, which 25 Mr Stein referred to this morning. But there appears to 114 1 have been little or no consideration of disclosure. 2 As raised by Mr Stein, Post Office may seek to 3 rely on a failure by Fujitsu to fully communicate the 4 extent of the problems arising within Horizon to them. 5 However, the evidence suggests that Post Office failed 6 to adequate scrutinise their provider, whether during 7 development or at any time during the life of Horizon. 8 We ask whether the company failed to ask the right 9 questions when problems were raised. The Ismay report 10 provides one example and the handling of the receipts 11 and payments bug another. The Inquiry may conclude that 12 Post Office was too quick to rely on Fujitsu's own 13 internal analysis of its own product. The Inquiry may 14 consider that the reasons provided by Mr Ismay to reject 15 the planned independent review of Horizon in 2010 ring 16 hollow and instead reflect an institutional 17 defensiveness. 18 The revelations of the rigorous Phase 3 lead into 19 Phase 4 and beyond and if I can now just very briefly 20 turn to that, sir. Phase 3 has map the evolution of the 21 Horizon-centric approach taken by both POL and Fujitsu 22 and over a decade this became a hardened Horizon 23 defence. The Inquiry has considered with Mr Ismay the 24 congratulatory language used in the aftermath of the 25 Castleton and Misra judgments and a straight line might 115 1 be drawn between the vision of a Horizon-centric future 2 in 1999 to the stock line that Horizon was robust, 3 described as such by Post Office at least by 2010, if 4 not before. 5 This is the one document I would like to put on 6 the screen, if I may, sir, and this is POL00002268 and 7 it's the second page. This is from Hayley Fowell on 8 1 February 2010 and, sir, it's the second paragraph when 9 following up a media enquiry from Retail Newsagent 10 magazine, she says: 11 "I am providing our stock line which states that's 12 system is robust but in case we get more questions on 13 this, please can you advise ..." and so on. 14 If this Horizon-centric approach or defensive was 15 combined with the presumption of incompetence or 16 dishonesty on the part of postmasters, that would 17 obviously be a cause for considerable concern. 18 On the evidence relevant to Phase 4, which the 19 Inquiry has already touched upon, there's sufficient to 20 suggest that at least Fujitsu was conscious of this risk 21 in the conduct of litigation by the Post Office as early 22 as 2007. 23 In Mrs Chambers' afterthought on her experience in 24 the Castleton case she highlighted concerns and 25 mistakes, including in respect of the material examined 116 1 before prosecution, as well as material which ought to 2 have been subject to duties of disclosure, describing it 3 as: 4 "... a major legal blunder. Since the postmaster 5 was blaming the system for the losses, I think it would 6 have been sensible to have double checked this before it 7 got as far as court." 8 That was yesterday's transcript, 16 May at 9 page 72. 10 This assessment was disseminated to management 11 within Fujitsu but it remains to be considered whether 12 or not this was distributed to others acting in support 13 of litigation and/or whether it was disseminated to POL. 14 In reality, where Horizon was to be given the 15 benefit of the doubt by both POL and Fujitsu, no such 16 amenity was afforded to postmasters. Postmasters were 17 to be doubted. They were seen as users in error and 18 postmasters, in contrast to Horizon, were seen as 19 expendable or easily replaced. They were seen by POL 20 first as potential risks to their business and/or as 21 potential debts to be recovered. The impression is 22 available that inherent bias saw postmasters who might 23 be responsible for an error viewed first as dishonest or 24 incompetent. 25 The impropriety and the danger of such a default 117 1 assessment is obvious. 2 The extent to which any wilful blindness or 3 commercial self-interest infected decisions taken in 4 respect of prosecution policy or the pursuit of 5 individual prosecutions will be a critical matter for 6 Phase 4, we say, sir. The approach of Rod Ismay to the 7 preparation of his Horizon Integrity Report in 2010 is 8 telling. Tasked to evidence the positive assurances to 9 Horizon's reliability, he says that is precisely what he 10 did. He did not question the purpose of a one-sided 11 defence of a product under challenge. Instead, he asked 12 Fujitsu to help mark their own homework and 13 astonishingly, we say, portrayed his work as 14 an objective evaluation. 15 The Inquiry might consider the continuing 16 commercial pressures on both companies. Post Office 17 remain under pressure as Network Banking rolled out and 18 later as they were embarked on a programme of network 19 transformation. Mr Ismay stressed the pressure he was 20 under when asked to complete his Horizon integrity 21 report. The Inquiry may consider whether Fujitsu 22 continued to grow its portfolio of Government and other 23 work on the back of its Horizon success. The Inquiry 24 may ask whether or not there was a commercial imperative 25 to whitewash the problems of Horizon, that being that 118 1 Horizon remained too important to fail. 2 Sir, the next phase of the Inquiry deals with, for 3 our Core Participants, critical questions, sir. How 4 could prosecutions be pushed forward without full 5 confidence in the data produced by Horizon being 6 evidenced? How could Post Office be allowed to maintain 7 its stock line that Horizon was robust for so long? How 8 could the disclosure of material now being considered by 9 the Inquiry not have been considered by Post Office 10 prosecutors? We will seek to assist the Chair and the 11 Inquiry legal team in answering those questions in any 12 way we can, sir. Thank you. 13 SIR WYN WILLIAMS: Thank you, Mr Moloney. 14 Mr Henry whenever you are ready. 15 Closing statement by MR HENRY 16 MR HENRY: Thank you, sir. 17 We say that, whilst it is indisputable by 2010 18 that the Post Office knew, and they knew before 19 Seema Misra was sent to prison, that the fact is they 20 had known for years by that time as had Fujitsu. They'd 21 known for years because from the very, very beginning, 22 around the country, subpostmasters baffled and 23 bewildered could not cope with the system and we know, 24 from earlier evidence in this Inquiry, that it was 25 roughly about 1 to 2 per cent of subpostmasters who got 119 1 into difficulties were prosecuted. That would appear to 2 be the evidence from an earlier stage in the Inquiry 3 from Mr Sweetman. We will develop that in our written 4 argument. 5 What the Post Office did in common with Fujitsu, 6 their mutual connivance, was to deny this -- not simply 7 to deny it, in fact, and to dispute it most vigorously, 8 but to deny it almost psychologically. 9 And why? Well, we return to paragraph 11 of the 10 common issues judgment of Mr Justice Fraser and, as 11 learned counsel to the Inquiry has said, you build on 12 the foundations of Mr Justice Fraser's work, in 13 particular the common issues, the Horizon IT issues, and 14 also most regrettably his decision on recusal. 15 What he said at paragraph 11 of the common issues 16 judgment was as follows, sir: 17 "If the Claimants were right in the broad thrust 18 of their case, this would represent an existential 19 threat to the Post Office's ability to continue to carry 20 on its business throughout the UK in the way it 21 presently does." 22 "Existential threat", as far as I remember and 23 recall, was Mr Oppenheim's phrase about the Post 24 Office's insistence, anxious demand, that they be 25 modernised. An existential threat it was because, if 120 1 Horizon was not implemented, it was curtains for the 2 Post Office. But, of course, if Horizon were to be 3 implemented and failed or proved to be a disaster, the 4 ignominious consequences would be catastrophic for the 5 Post Office. 6 There we see the beginning of a common interest 7 with, first of all, ICL and then Fujitsu and ICL, it may 8 now be thought, was a damnnable acquisition so far as 9 Fujitsu was concerned, Horizon being the classic 10 poisoned chalice. Whilst it may not have been 11 an existential threat to Fujitsu, perhaps it may have 12 been fatal to its independence and certainly its 13 valuation. 14 So here you have these two institutions, sir, one 15 a public corporation trading off its most trusted brand 16 value and the other a global multinational. We say that 17 eventually they conspired, stumbling at first, to arrive 18 at a position of mutual interest after all of the back 19 biting, but they decided that it was better that 20 1 per cent or 2 per cent of the subpostmasters would be 21 put on the rack, rather than the Post Office collapse 22 and a corporate's reputation be ruined. 23 That mutual connivance, all it took, you may 24 think, is a nudge and a wink, Fujitsu "If we're not 25 asked, we won't tell", and the Post Office had no 121 1 inclination to ask because, to mix the metaphor, they 2 were holding the telescope to their blind eye. 3 So Phase 3 is all about the Post Office and 4 Fujitsu's failure, lamentable failure, to act in 5 accordance with the proper duty of care it owed to the 6 subpostmasters. It would be no good Fujitsu constantly 7 saying "Oh, well, they're not the client". If they knew 8 that their system was putting these poor people in 9 danger, they had a duty to disclose. And yet, you see 10 the contemporaneous documents "Oh, well, wish we could 11 help further but they're not our problem". 12 As we have seen from the recently disclosed 13 material from Fujitsu -- and of course I am referring to 14 FUJ000152299, "Afterthoughts on the Castleton case", 15 Fujitsu made no attempt to reveal the errors, the bugs, 16 the defects that were shot through the system before 17 Mr Castleton went to court. That memorandum, 18 Mrs Chambers' effort to remediate her employer's 19 approach to establish a protocol for future trials was 20 ignored. Now, there's been no suggestion, certainly not 21 by us, that that was a cynical ploy to cover her own 22 back. There's been no disclosure of a riposte, if you 23 will excuse the pun, criticising or rebutting or 24 refuting even the contents of that note. That note was 25 sent in good faith. 122 1 What happened, sir? Nothing. Nothing was done. 2 No recommendation was acted upon. No suggestion was 3 actioned, systematised or implemented it would seem. 4 Fujitsu and the Post Office are both culpable. 5 I repeat, sir: they both connived in the injustice 6 deliberately inflicted on innocent victims and it 7 became, as we will show, a corporate scandal, their 8 existence itself, their corporate existence, their 9 bonuses and their reputation prioritised, temporarily 10 preserved but now ruined, by relatively short-term 11 deceit and deplorable misconduct. 12 We will justify in writing the following findings 13 arising from Phase 3 of this Inquiry which you, sir, 14 aided by your exemplary Inquiry team, have examined with 15 your customary astuteness and attention to detail. The 16 first headline is that the Post Office failed to 17 discharge its duty of care to the subpostmasters by: 18 (1) failing to provide adequate training for the 19 use of the Horizon IT System, and I adopt what my 20 learned friend Mr Moloney said; 21 (2) failing to provide an appropriate level of 22 informed, independent and objective assistance to the 23 subpostmasters reporting errors, anomalies and 24 discrepancies within the Horizon IT System; 25 (3) denying employees, although there is a gloss 123 1 on that because we see an increasingly low level of 2 prosecution of employees as opposed to subpostmasters no 3 doubt because of the employment law position but denying 4 employees and subpostmasters using Horizon knowledge of 5 those bugs, errors and defects by restricting and 6 suppressing information that would have revealed the 7 instability of the system, compounded by ad hoc fixes 8 that introduced even more anomalies and errors which 9 subpostmasters were not informed about and could not 10 have been aware of; 11 (4) failing to provide a system that was 12 reasonably fit for purpose, and here we extend the ambit 13 of the issue to Detica and Second Sight, issues to come, 14 but still less one suitable for civil and criminal 15 evidence. 16 (5) refusing, sir, to look in the mirror, facing 17 up to its responsibilities towards the subpostmasters 18 but, instead, branding them as either dishonest or 19 incompetent, and Fujitsu enabled this. 20 Mr Coleman today and Mrs Chambers earlier this 21 month fell squarely into the deductive fallacy, 22 overconfident in their assessment of the system. But 23 they're the foot soldiers, sir. Those above them 24 allowed them to fall into error and reach uncorrected 25 the false conclusions that will now be demonstrated: 124 1 The first premise is that all discrepancies are 2 software-related or user-related; (b) a software error 3 has not been detected; (c) therefore, the error must be 4 user related. That fallacy in Castleton ruined a good 5 man and his family. It was elevated disproportionately 6 by Mr Morgan, as was, now Mr Morgan KC, and His Honour 7 Judge Havery KC as wholly writ that the system was 8 error-free, when both Fujitsu and the Post Office knew 9 that it was not. 10 Mr Coleman was asked today by counsel to the 11 Inquiry, well, wasn't user error the default position, 12 the user not the system getting the blame. No, he said. 13 It was his judgement based on his knowledge and it's 14 a classic case, we respectfully submit, of confirmation 15 bias. But, again, I repeat, sir, he's a foot soldier 16 and the soldier in the trench can't see the battle 17 field. 18 They weren't told, for example, about the EPOSS 19 instability. They weren't told about the fact that the 20 decision to rewrite it had been scuppered. To use 21 a different analogy, what Mr Coleman was telling 22 Mr Blake was that the black box was working just fine 23 but maybe he, sir, did not know what was in the black 24 box himself, notwithstanding his learning, because, of 25 course, he had not been told or put in the whole picture 125 1 and, in any event, and this is a big but, what do we 2 know about what is in the black box? Must we take it on 3 trust? 4 A multinational of renown with political 5 influence, in union with Britain's most trusted brand 6 a claim I hope you will forgive me for saying, sir, 7 which now nauseates, that was enough to carry all before 8 them, until a sceptical High Court judge, Mr Justice 9 Fraser, relentlessly drilled down to the detail, probing 10 the verbiage, the conventional wisdom that had hitherto 11 masqueraded as empirical fact. 12 Before that, sir, the Post Office, aided and 13 abetted by Fujitsu, intentionally and deliberately 14 abused their position via, as we will demonstrate in 15 writing, material non-disclosure, failing to make 16 rigorous nor even reasonable enquiries even-handedly and 17 impartially to arrive at the truth, concealing those 18 bugs, errors and defects but also the nature, frequency 19 and extent of those flaws, instead just saying to the 20 anxious, innocent subpostmaster "It's just you. You're 21 out on your own, on a limb. Where have you put the 22 money?" 23 (4) Concealing Fujitsu's covert remote and 24 untraceable system access or interference with branch 25 accounts. What is the good of having a system where the 126 1 subpostmaster who is held utterly responsible for 2 defects or deficiencies, isn't even in control of their 3 own account? 4 Then (5), maintaining the myth at all costs of 5 system infallibility, prosecuting the innocent, 6 bankrupting the honest Mr Castleton, a policy made clear 7 during this phase, and which will be developed in the 8 phase to come, a policy deliberately contrived to 9 harrow, intimidate and deter. 10 It is clear from Phase 3 that the Post Office 11 behaved in a disgraceful manner. That is not in 12 dispute. No cheap soundbite to that effect will do 13 justice to its victims. These good, loyal, decent 14 people were valued at naught by the Post Office, treated 15 like an inconvenient rounding error to be written off or 16 dispensed with. This, sir, is already known. 17 The real questions are how and why. How did they 18 do this? Why did they feel justified? They did this by 19 wielding uncontrolled power via oppressive contractual 20 terms with subpostmasters, an inequality of power and 21 status of juggernaut resources against fledgling 22 businesses, often owner-managed, up and down the land, 23 and an individual subpostmaster could not withstand it. 24 They did it through the imposition of liability 25 for branch losses as a default position, oppressive. 127 1 Other terms explored in the common issues judgment were 2 equally unjust. One hatched in Castleton when poor 3 Mr Castleton was inveigled into signing off disputed 4 accounts on the pretext he had to leave, closing an 5 account which he'd signed after his dismissal, which was 6 then, of course, used to reverse the burden of proof not 7 simply against him but those that followed. 8 A deliberate strategy and you will, sir, be drawn 9 to a memorandum and a post-trial advice where Mr Morgan, 10 King's Counsel, said that he would be delighted to 11 address the board on the matter as to how and I don't 12 suggest that he regarded it as a stratagem, but 13 stratagem it was, that this is how we get them 14 themselves to reverse the burden of proof: get them to 15 sign before they go. 16 You had the spectacle, sir, and nothing has 17 changed, we respectfully submit, of a public corporation 18 denying it had to act fairly or in good faith towards 19 its agent, denying its contract was relational in 20 nature, which would, of course, conferred rights and 21 entitlements upon its long-suffering subpostmasters. 22 How were they allowed to do this? We know why. It was 23 because they felt entitled and it was the existential 24 threat. 25 But how were they? Because the Government gave 128 1 the Post Office a long leash. Even though publicly 2 owned, with strategic direction set by the Government 3 including of course the implementation of IT, the 4 convenient fiction was that the Post Office was 5 an independent commercial business and so its 6 relationship with subpostmasters and management of its 7 IT systems was the Post Office's, and I quote, 8 "exclusive domain giving it unfettered operational 9 control". 10 Here, sir, to draw an analogy, the Government is 11 like the owner of a dangerous dog just mauling 12 a defenceless child and saying "Sorry, so sorry, but 13 it's nothing to do with me", because people were mauled 14 up and down the country. Lives were destroyed. Life 15 itself was lost or ended too soon. No, the Government, 16 sir, is responsible. It's responsible for allowing the 17 Post Office to pursue a policy of suppression and 18 oppression, bankrupting a man of straw to the sum of 19 £321,000 -- £321,000 -- for a debt that was well under 20 30, a debt that did not even exist. 21 These punishments, sir, were intended to reinforce 22 the dominance of the Post Office and they were well 23 publicised through the network, the NFSP, perhaps even, 24 to dissuade others from maintaining their innocence by 25 demonstrating the vengeance that would be visited upon 129 1 those who challenged Horizon. Mr Castleton: no income, 2 no assets, and the true benefit of that costly case is 3 to be found at POL -- forgive me, sir, I have just got 4 an email from Mr Beer. I don't know if it's relevant, 5 so can I just check it? It's not relevant, fine. 6 The document POL00113488 reveals the true benefit: 7 the more publicity the case is given the greater should 8 be the effect upon postmasters who take legal advice 9 about defending claims for repayment. The same can be 10 said for Mrs Misra, the guilty verdict applauded by 11 Mr Singh at POL0004497 as: 12 A marker to dissuade other defendants from jumping 13 on the Horizon-bashing bandwagon." 14 Then the question that learned counsel, Mr Beer, 15 put to Mr Ismay: 16 "Is that language reflective of the culture 17 prevalent at the time concerning Horizon; namely, in 18 response to a defendant who maintained a defence to the 19 criminal charges of theft against her was thereby seen 20 as attacking Horizon, an attack which needed to be 21 destroyed?" 22 Well, sir, that can only be answered in the 23 affirmative. 24 So the Government, we respectfully submit, is 25 responsible for every life mangled or lost, every 130 1 livelihood destroyed, because it failed to manage, 2 notwithstanding its NED, it failed to properly manage 3 the Post Office. 4 There are many who were broken on the alter of 5 Horizon's infallibility, who had the temerity to 6 challenge the system but two (together with Ms Page and 7 Hodge Jones and I have the honour representing, and that 8 is Mr Castleton and Mrs Misra) both faced a dreadful 9 ordeal, callously destroyed in acts of, we respectfully 10 submit, spectacular retribution because Mrs Misra wasn't 11 even allowed in the end to plead to false accounting. 12 They were denied the means to remove the burden of 13 unjust accusation, and nor were they provided with the 14 appellate right to remove the stigma of wrongful 15 judgment because that material was suppressed. 16 And so I conclude, sir. This is as dark a chapter 17 in our governmental, corporate and legal history as can 18 be imagined and, sadly, it will get darker yet. But 19 you, aided by the Inquiry team, will prevail, 20 notwithstanding any obstacles or non-disclosure that you 21 encounter. You will arrive at the truth. And the truth 22 will be, as one surmises, that this was hazarding with 23 people's lives deliberately in order to preserve an 24 institution that the loyalty, the corporate loyalty, to 25 that institution caused people to make the most 131 1 grotesque misjudgements and then to cover them up in the 2 full knowledge of the wrong that had been done. 3 SIR WYN WILLIAMS: Thank you, Mr Henry. Right, well, 4 I think I can say that the Phase 3 which began in 5 January, and we are now in the middle of May, is almost 6 at a close. We have to remind ourselves that there is 7 still some evidence to come and there will be written 8 submissions next week, but we are very nearly at the 9 close of Phase 3. 10 I'm very grateful to everyone who has participated 11 in it and has contributed to the unearthing of important 12 facts and issues, and I look forward to seeing you all 13 again in the beginning of June when we head on to 14 Phase 4. 15 So I think, Mr Beer, unless you have anything to 16 add, we can now conclude today's proceedings. 17 MR BEER: No. Thank you very much, sir. 18 (2.55 pm) 19 (Adjourned until Tuesday, 6 June 2023) 20 21 22 23 24 25 I N D E X 132 1 ALVIN EDWIN FINCH (sworn) ........................2 2 Questioned by MR BLAKE ...........................2 3 RICHARD COLEMAN (affirmed) .......................9 4 Questioned by MR BLAKE ...........................9 5 Statement by MR BEER ............................54 6 Closing statement by MR STEIN ...................61 7 Closing statement by MR MOLONEY .................99 8 Closing statement by MR HENRY ..................119 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 133