25 June 2024 – Gareth Jenkins
Hide video Show video
(9.45 am)
Mr Beer: Good morning, sir, can you see and hear us?
Sir Wyn Williams: Yes, thank you very much.
Mr Beer: Thank you. May I call Gareth Jenkins, please?
Sir Wyn Williams: Yes.
Gareth Jenkins
GARETH IDRIS JENKINS (sworn).
Questioned by Mr Beer
Mr Beer: Good morning, Mr Jenkins. My name is Jason Beer and I ask questions on behalf of the Inquiry. Can you give us your full name, please?
Gareth Jenkins: My name is Gareth Idris Jenkins.
Sir Wyn Williams: Mr Beer, before you go any further, is it appropriate for me to direct Mr Jenkins about self-incrimination?
Mr Beer: Yes, sir.
Sir Wyn Williams: Mr Jenkins, I am sure you’ve been advised but, under our law, a witness at a public inquiry has the right to decline to answer a question put to him by any of the lawyers involved in the Inquiry, or by me, if there is that the answer to that question would incriminate you. This legal principle is known in shorthand form as the privilege against self-incrimination. I’ve decided that fairness demands that I remind you of that principle before you give your evidence. I must stress to you, however, that it is for you to make it clear to me that, in respect of any question put to you, it is your wish to rely upon the privilege. If, therefore, any questions are put to you by anyone who is asking you questions which you do not wish to answer, on the ground that to answer such a question might incriminate you, you must tell me immediately after such question is put. At that point, I will consider your objection to answering the question and thereafter rule upon whether your objection should be upheld.
I know that you are legally represented. No doubt, if the issue relating to self-incrimination arises, your lawyers will assist you and, if at any stage during the questioning, you wish to consult your lawyers about the privilege, you must tell me so that I can consider whether that is appropriate. Do you understand all that, Mr Jenkins?
Gareth Jenkins: Yes, I understand, sir.
Sir Wyn Williams: Thank you.
Yes, Mr Beer.
Mr Beer: Thank you, sir.
Mr Jenkins, you’ve made five witness statements to the Inquiry. Thank you very much for making those witness statements and for attending today and over the next three days to give evidence. Can we start, please, with your witness statements.
Gareth Jenkins: Yes.
Mr Beer: If we can start with the fifth witness statement, please, which was, in fact, a witness statement you made only yesterday. It’s WITN00460500.
Gareth Jenkins: Yes.
Mr Beer: Have you got that witness statement in front of you?
Gareth Jenkins: I have indeed.
Mr Beer: It’s three pages long and it consists of three pages of corrections to your previous witness statements; is that right?
Gareth Jenkins: That’s correct, sir.
Mr Beer: Is that your signature on the third page?
Gareth Jenkins: It is.
Mr Beer: Are the contents of that, your fifth witness, true to the best of your knowledge and belief?
Gareth Jenkins: They are.
Mr Beer: That. Can we go to your first witness statement, then, please, which is WITN00460100. It’s dated 6 February 2023 and is 15 pages long.
Gareth Jenkins: Yes.
Mr Beer: Have you got that?
Gareth Jenkins: I have got that.
Mr Beer: Is your signature on page 15?
Gareth Jenkins: It is.
Mr Beer: Are the contents of that witness statement true to the best of your knowledge and belief?
Gareth Jenkins: They are.
Mr Beer: Thank you. Can we go to your second witness statement, please, WITN00460200, dated 1 June 2023, which is 67 pages long; is that your signature on page 67?
Gareth Jenkins: Yes, it’s actually slightly longer because it’s got an appendix at the end but that’s my signature on page 67, yes, sir.
Mr Beer: I’m only dealing with the body of the statement rather than the additions at the end, which are exhibit sheets, essentially?
Gareth Jenkins: Right, yes.
Mr Beer: With the corrections you made in statement 5 brought into account, are the contents of that, your second witness statement, true to the best of your knowledge and belief?
Gareth Jenkins: Yes, they are.
Mr Beer: Thank you. Can we deal with your third witness statement, please, 21 March 2024, WITN00460300, 234 pages long.
Gareth Jenkins: Yes.
Mr Beer: Is that your signature on page 234?
Gareth Jenkins: It is.
Mr Beer: With the corrections in statement 5 brought into account, are the contents of that witness statement true to the best of your knowledge and belief?
Gareth Jenkins: Yes, they are.
Mr Beer: Thank you. Then, lastly, statement 4, 29 April 2024, WITN00460400, 83 pages long. Is that your signature on page 83 of the witness statement?
Gareth Jenkins: It is.
Mr Beer: With the corrections in statement 5 brought into account, are the contents of that, your fourth witness statement, true to the best of your knowledge and belief?
Gareth Jenkins: They are.
Mr Beer: Thank you very much.
They can be put to one side now. Any further references I give you will come up on the screen.
Gareth Jenkins: Okay.
Mr Beer: Those witness statements, all five of them, will be uploaded to the Inquiry’s website. They are a substantial body of evidence and, therefore, I’m not going to refer to all parts of them when asking you questions. Additionally, you’ve exhibited to your witness statement a very large number of documents and you have been provided by the Inquiry with still further documents over the last year and a half, and you and your legal representatives have had access to the Inquiry’s Relativity database. I think all of that has resulted in you seeing a very large volume of material; is that right?
Gareth Jenkins: Indeed.
Mr Beer: If, at any time, you want me to stop or to slow down when considering a document, then please do say so.
Gareth Jenkins: Thank you.
Mr Beer: Can I start with your background, please. I think you graduated in 1973 from Cambridge University with a degree in mathematics?
Gareth Jenkins: That’s correct.
Mr Beer: Upon graduation, you immediately worked for International Computers Limited, ICL; is that right?
Gareth Jenkins: That’s correct.
Mr Beer: ICL became Fujitsu Services Limited in 2002?
Gareth Jenkins: Around then. I’m not sure exactly when that happened but, yes, it changed its name in the early part of the century.
Mr Beer: I was taking that from your witness statement.
Gareth Jenkins: Yes, I wasn’t 100 per cent sure of the date but that’s the best of my recollection.
Mr Beer: You remained there for the whole of your professional career, is that right –
Gareth Jenkins: That’s correct.
Mr Beer: – until you retired in 2015?
Gareth Jenkins: Correct.
Mr Beer: Dealing briefly at the moment with the stages of your career at ICL Fujitsu, so far as concerns us, from about 1996 to about 2000 did you work in the agent team?
Gareth Jenkins: That’s correct, yes.
Mr Beer: You were one of a team of 10 to 20 architects working on the Horizon system; is that right?
Gareth Jenkins: Something like that, yes.
Mr Beer: From about 2000 until retirement in 2015, you worked in the Architecture and Design Team; is that right?
Gareth Jenkins: That’s right, yes.
Mr Beer: That was part of the Development Team; is that right?
Gareth Jenkins: It varied. At some points it was, at other points it was a separate team, but the work was similar in those respects. I –
Mr Beer: It was part of the Development Team, I think, until it merged with what was called the Requirements Department; is that right?
Gareth Jenkins: Yes.
Mr Beer: How many people were in the Architecture and Design Team?
Gareth Jenkins: I can’t remember the exact numbers but, again, it varied a lot. It started off with about 20 or 30, I think I’ve got the figure in my witness statement, and, by the time I retired, it was down to something like about half a dozen, that sort of number.
Mr Beer: Were they all the same level or pay grade as you or was there seniority within the department?
Gareth Jenkins: Different people were at different grades but that wasn’t necessarily reflected in terms of a management structure as such.
Mr Beer: Throughout your time, I think your role was a technical or operational one, is that right –
Gareth Jenkins: Correct, yes.
Mr Beer: – not a managerial or supervisory role; is that right?
Gareth Jenkins: No, I never had any man-management responsibilities during that time.
Mr Beer: You tell us in your witness statement that from 2002 to 2008 you “took on the role of defining the technical changes required by Project IMPACT and worked on the implementation of the Bureau de Change and the acceptance of credit cards”; is that right?
Gareth Jenkins: Among other things. So those were the main developments. So I worked on IMPACT until that went live in 2005, I think it was, then I worked on Bureau de Change and credit card changes and a few other minor things before I moved on to Horizon Online Requirements Capture.
Mr Beer: Did you move on to Horizon Online from about 2008?
Gareth Jenkins: It was around then, yes. I can’t remember the exact time.
Mr Beer: Did you work on Horizon Online from 2008 to about 2015?
Gareth Jenkins: Yes.
Mr Beer: You tell us in your witness statement that you “worked with Post Office analysts in defining the technical requirements for Horizon Online”; is that right?
Gareth Jenkins: Yes, that’s true.
Mr Beer: And that you “supported Post Office’s requirements team in specifying use cases for the counter functionality”?
Gareth Jenkins: That’s correct, yes.
Mr Beer: What does that mean?
Gareth Jenkins: The contractual definition of how Horizon Online was to work was that the counters should look – should operate in the same way as it did on Legacy Horizon and, therefore, it was decided that we needed to put in place something that actually defined exactly what that was, and the technique that was decided was that Post Office would put together the use cases to define the various activities that went on in a Post Office to define the steps that people went through for doing – from how do you sell a stamp to how do you balance at the end of the week or whatever. So –
Mr Beer: Sorry to interrupt you, that’s a use case, is it?
Gareth Jenkins: Each one of those would be a separate use case, yes. Sorry.
Mr Beer: Did you then develop the implementation of those use cases and then test them?
Gareth Jenkins: Not personally. There was a team who were developing the counter, and I was acting as a guide to that team in terms of interpreting what the use cases meant and looking at how the code that they developed came out, and I was responding to when queries came out of the testing process as to “Should it work this way or should it work that way”, “Is this a problem or is that how it’s supposed to work”, and reacting to things like that.
Mr Beer: You tell us that, during your time at Fujitsu, you were a group of anywhere between 20 and 100 software designers and developers who would provide fourth line support for both Legacy Horizon and Horizon Online?
Gareth Jenkins: Yes, so the way fourth line support worked was that the people who were doing the development were expected to spend a small part of their time in actually supporting the live service and the system under test, and so that’s what fourth line support was all about, and I did that just like everyone else in the Development Team.
Mr Beer: That was an adjunct to your main role?
Gareth Jenkins: Correct.
Mr Beer: Although I’m sure the fraction varied, what proportion of your time was given over to fourth line support?
Gareth Jenkins: It was supposed to average out at about 10 per cent and I think that’s roughly what it ended up as.
Mr Beer: Thank you.
In your first witness statement, you refer to ICL introducing what’s called a Distinguished Engineer Scheme, correct?
Gareth Jenkins: Yes.
Mr Beer: You tell us that you became a distinguished engineer in the mid-1990s?
Gareth Jenkins: Yes, I can’t remember exactly when.
Mr Beer: You were one of about 100 other ICL employees who were described as Distinguished Engineer; is that right?
Gareth Jenkins: That’s correct, yes.
Mr Beer: Is it right that was an honorific title conferred upon you by ICL and Fujitsu?
Gareth Jenkins: Yes, it was. It didn’t make any difference to what I got paid, or anything like that, it was purely seen as an honour and just recognition of that I had a fairly senior design role and was respected within the company.
Mr Beer: What was the process that led to you becoming a Distinguished Engineer?
Gareth Jenkins: I think I was nominated by my manager who filled in various complicated forms to justify why I should become one.
Mr Beer: Was there any further training or qualification or a particular professional experience that was required before you could would be required before you could become a Distinguished Engineer?
Gareth Jenkins: Not, as such, no.
Mr Beer: So it was simply a title given to you by your own company?
Gareth Jenkins: Yes, correct.
Mr Beer: Did you acquire any level of management responsibility as a result of the appointment?
Gareth Jenkins: No, people did try and persuade me that I ought to manage some people and I resisted it and successfully.
Mr Beer: I think you’ve told us that it didn’t affect your pay?
Gareth Jenkins: Correct.
Mr Beer: You tell us in your first witness statement that, from about 2015 until August 2022, you were on a retainer with Fujitsu as a consultant; is that right?
Gareth Jenkins: That’s correct.
Mr Beer: What was the nature of your consultancy with them? What were you consulting about?
Gareth Jenkins: It was various things. I – the expectation was that it would amount to something like about half a dozen or a dozen days a year and, for most of the period, it was at that level. I don’t think – although I was on the retainer until August 2022, I think the last time I actually did any paid consultancy was just before Covid lockdown in 2020, which is why they decided that, since they hadn’t been using me for a couple of years, that they didn’t need to retain me any further. The one exception to that was the time when the Group Litigation was happening in 2018/2019, when I was working a lot more than that, probably an average of about one to two days a week, for a period of about five or six months.
Mr Beer: So did your retainer essentially come to an end at the time of the publication of the Horizon Issues judgment, which was December ‘19?
Gareth Jenkins: No, the retainer came to an end in 2022 but I didn’t actually do much retained work after –
Mr Beer: Okay.
Gareth Jenkins: The last few things were actually talking to the lawyers about the consequences of the Horizon Issues judgment.
Mr Beer: Other than the provision of information, assistance and evidence in connection with the litigation concerning the Horizon system, what work did you perform in this consultancy?
Gareth Jenkins: I was called to go to a few meetings. A lot of it was to do with the hangouts from prosecution. I wasn’t involved in any particular prosecutions as such but there was quite a lot of work going on in the background which – talking to various lawyers. I got pulled in in 2016 for a bug that had been found that was quite serious, the Dalmellington bug, that I think you may well be coming back to later on, and it was felt that I could maybe help explain what the consequences of that was to Post Office.
I spent – there was a problem that Post Office had – again, I can’t remember exact date now it’s in my witness statement – when the – there was a mismatch in Post Office’s back end accounts. They tried to do a reconciliation against data they had from Fujitsu, and I realised what the problem was. It was to do with the fact that they were taking the accounts at different times. One was taking what the cash positions were at 7.00 in the evening and the other was taking the cash position at midnight and, therefore, this accounted for the mismatch that they had in the accounts.
So it was various ad hoc, fairly self-contained tasks, each of which lasted maybe two or three days.
Mr Beer: There seems to be a fly –
Gareth Jenkins: Yes, there is, it’s a bit of a pain.
Mr Beer: – buzzing around. I don’t want you to be distracted.
Gareth Jenkins: No, I’m trying to ignore it.
Mr Beer: If it becomes too much, we’ll break and get rid of it.
Gareth Jenkins: Okay.
Mr Beer: Would it be fair to say that the majority of this time, certainly between 2015 and the Horizon Issues judgment in December 2019, was given over to the provision of Litigation Support?
Gareth Jenkins: Certainly in 2018 and 2019. I’m not sure it was direct litigation, it was probably about half and half. As I say, I can’t remember the gory details of exactly what I was doing when but there were a number of notes that I produced which I wasn’t quite sure what the background was but, looking back at it now, I realise was actually maybe in preparation for what eventually became the GLO.
Mr Beer: Have you read the Horizon Issues judgment?
Gareth Jenkins: I skimmed through it at the time and I’ve looked at various sections but I don’t claim to have read every word, I’m afraid.
Mr Beer: In high level summary, would you agree that Mr Justice Fraser concluded, firstly, that Legacy Horizon was not remotely robust?
Gareth Jenkins: I accept that that’s what he said.
Mr Beer: Do you accept that Horizon Online was susceptible to accounting flaws?
Gareth Jenkins: Um –
Mr Beer: That’s what he –
Gareth Jenkins: Again, that is what he said. I think he is maybe putting it do strongly. I felt that Horizon Online was and is – because, after all, it is still the system that’s operating today – doing a good job in terms of the accounting, particularly with the monitoring that was going on in terms of being able to detect things when they’d occurred.
Mr Beer: He concluded that subpostmasters were not informed of, and were therefore oblivious to, a range of bugs, errors and defects. Do you agree that that is a summary of what he found?
Gareth Jenkins: That is certainly what he said, yes.
Mr Beer: And, lastly, that those bugs, errors and defects could result in, ie cause, discrepancies or shortfalls in the branch accounts of subpostmasters, thereby corrupting transactions, disrupting data processing and recording transactions?
Gareth Jenkins: That was a possibility but I’m not sure that I, even today, I understand what bugs actually did cause the problems that people are – that people have suffered from.
Mr Beer: Do I take it that you accept his first finding that Legacy Horizon was not remotely robust?
Gareth Jenkins: I don’t accept that finding.
Mr Beer: You don’t accept his finding that Horizon Online was susceptible to accounting flaws?
Gareth Jenkins: There were some discrete bugs that caused problems to the accounts but they were very discrete and I believe they were all well controlled and managed at the time.
Mr Beer: You don’t accept his findings that bugs, errors and defects could result in, ie cause, discrepancies or shortfalls in branch accounts?
Gareth Jenkins: They could cause discrepancies in branch accounts but not at the sort of levels that are being talked about and, in general, the systems, I believe, were operating as they should.
Mr Beer: Robustly?
Gareth Jenkins: It depends exactly what you mean by “robust” but as long as you’re not saying “infallibly” then, yes, because I think “robust” meant that there were mechanisms in place that would monitor what was going on, detect problems, and that they were then investigated and resolved correctly.
Mr Beer: Horizon, both Legacy and Online, were working well in your view?
Gareth Jenkins: Most of the time, there were clearly problems during the pilots in both cases and there were clearly individual problems that affected individual branches, and I’m sure we’ll come on to those at some time but, in general, then I felt that the systems were working well.
Mr Beer: The judge got it wrong?
Gareth Jenkins: I wouldn’t like to say that but I think there’s a difference in emphasis between – there were clearly problems and he identified a number of problems and I won’t dispute those problems happened but, on the whole, I felt that the systems were working well.
Mr Beer: Can I turn in more detail to your role in the development and rollout of the Horizon system, and in – no need to turn it up – paragraphs 18 to 23 of your first witness statement, you describe your roles and responsibilities at ICL and then Fujitsu. So between 1996 and 2002, you tell us that your initial role with Legacy Horizon was “to integrate Riposte and the Oracle based software”, yes?
Gareth Jenkins: That’s correct, yes.
Mr Beer: And that you helped develop an agent layer that would allow those two types of software to communicate with each other?
Gareth Jenkins: Correct.
Mr Beer: Firstly, can you describe briefly what Riposte was?
Gareth Jenkins: Riposte was a – well, let’s step back one. There were two aspects to Riposte and this was software that we had bought in at Post Office’s request from a company called Escher in the United States. The bit of it that I was primarily involved in was what was called the message store, which was a way that all the data about what had happen in a branch was being record.
So it’s a sophisticated database, in crude terms, and that was where the data, not only of transactions but working data during the day, what had gone on when people logged on, all sorts of things, was all recorded in this message store.
Then the other part of Riposte was a development environment that allowed applications to be built on that and the application environment that caused the human interface that the postmasters saw the branches, and that bit of Riposte I had not very little involvement with in the early days. I got more involved in that when I moved on to Project IMPACT in 2003 or so but, in the early days, my main interaction with Riposte was to do with the message store, and –
Mr Beer: Thank you. Then, the Oracle-based software. Can you describe again briefly, as you’ve just done, what the Oracle-based software was?
Gareth Jenkins: This was really an Oracle database and this was used to communicate with Post Office’s back-end systems where they did their back-end accounting. So that was really the boundary between what Fujitsu was doing, or ICL was doing, and what Post Office was doing. So what was – that was basically an overnight batch processing system, so it would process through all the transactions, it would produce files of transactions that had happened during that day, which were then transmitted to Post Office’s back-end systems.
Mr Beer: You say that you developed an agent layer; what was an agent layer?
Gareth Jenkins: What an agent layer was doing was it was reading data from the message store, and we had copies of the message store in the data centre, and extracting the relevant transactions and other information that was of interest to Post Office’s back-end systems, and writing records to the Oracle databases to reflect those transactions.
Then there was also a flow in the other direction, for example, reference data was being generated by Post Office and, by reference data, the simplest example is what’s the price of a stamp today because, obviously, the price of stamps changes every now and again.
So, therefore, things like prices weren’t actually built into the code; we just knew that you needed to sell a stamp and then, behind that, there would be a bit of reference data that says, well, if you sell a stamp, it’s going to cost you whatever a stamp costs today, £1.15 – back in the day it was about 30p or something.
Mr Beer: Given that the work started in 1996, what challenges existed in ensuring that the communication by the agent layer was enabled?
Gareth Jenkins: I’m not quite sure I understand the question.
Mr Beer: Yes. You tell us in your witness statement that your role lasted between 1996 and 2002, which is post-rollout?
Gareth Jenkins: Yes. So –
Mr Beer: What took six years?
Gareth Jenkins: Right. The work changed a lot over time because, at the beginning, back in 1996, the first application that was rolled out was actually a Benefits Agency application which, by 1999 – ‘98/’99, had actually got dropped because the Benefits Agency pulled out of the contract. So a lot of –
Mr Beer: Just stopping there, I wonder whether you mind slowing down a little bit.
Gareth Jenkins: Sorry.
Mr Beer: Although the shorthand writer hasn’t said anything so far, I can imagine –
Gareth Jenkins: Sorry.
Mr Beer: – that she wants to.
Gareth Jenkins: Yes.
Mr Beer: So if you can just so down a little bit, please.
What did you know as to the reasons why the Benefits Agency pulled out of the contract?
Gareth Jenkins: I’m not sure that I was really aware of the detail. That was something that was taking place at levels way above where I was working.
Mr Beer: What were you told?
Gareth Jenkins: I can’t remember now what I was told. I think they decided they were going to do things differently. What they eventually did was they paid people’s benefits directly into their bank accounts, rather than doing it through special application with the Post Office.
Mr Beer: Were you told anything to the effect that one of the reasons why the Benefits Agency withdrew was a lack of confidence in ICL –
Gareth Jenkins: I’m not sure that I was aware of that at the time.
Mr Beer: – and doubts over the integrity of the data that the nascent system was producing?
Gareth Jenkins: I don’t remember hearing that at the time.
Mr Beer: Anyway, you continued in the role until 2002. Does that mean that the agent layer required a continual level of support and maintenance, even after rollout?
Gareth Jenkins: No, because there were few – there was new functionality. So the main thing I was doing in the early 2000s was – network banking was introduced into the system in 2003 and, a few months later, the support of debit cards was introduced. So the main work I was doing with the agents then was supporting the changes that were required because one of the things the agents did was they interfaced between the data centres and the banking systems, and that was clearly a fairly significant development.
Mr Beer: You tell us that you moved away from the agent team and took on the role of defining changes required by Project Impact. Did the work of the agent team continue after you left?
Gareth Jenkins: Oh, yes, there was a team there. Someone else took over the role of Chief Designer within the agent team. I think he actually took over that role about 2000, something like that, and I was gradually moving on to other things on the side. So it wasn’t a sudden “Drop this and do that”; it was more of an evolution than a revolution.
Mr Beer: In your role between 1996 and 2002, were you in any way responsible for communicating with the Post Office team as to their requirements?
Gareth Jenkins: Not so much with their requirements. That really started with Project Impact but I certainly was working with the Post Office team in terms of defining the interfaces between how Fujitsu’s agents were operating and how they interacted. Initially, there was a box in between us and the banks called the Network Banking Engine which was provided by IBM so, initially, we provided the interface into that. Then, later on, that was replaced, after I was no longer involved, by a direct interface to the banks.
Mr Beer: Did you come into contact with Jeremy Folkes, the Infrastructure Assurance Team Leader from the Post Office?
Gareth Jenkins: Not at that time but I did come across him a few years later when he moved roles and he actually joint Escher as one to their consultants. So I did come across Jeremy in his role with Escher but I don’t remember coming across him when he was working with Post Office.
Mr Beer: He told the Inquiry, it was back on 2 November 2022, that “we”, and by that he meant the Benefits Agency and the Post Office Counters’ team “had been denied viability of the application design”.
Gareth Jenkins: I’m not quite sure what he meant by that.
Mr Beer: He continued:
“When I say we had some visibility of risk areas, what I mean is that the areas where we had raised formal risks to the service provider at the start, such as Riposte, in those cases, we did get more information but, as far as the application design, in particular EPOSS, we had been denied visibility.”
Gareth Jenkins: I think that was something to do with the way the contract had been negotiated between – or ICL management, as it was then, and Post Office, and DSS. I wasn’t involved in any of that but, as I understand it, it was a PFI contract and part of that meant that the customer that to treat the whole things as a black box but, as I say, this is hearsay, if you like, rather than something I was directly involved in.
Mr Beer: Where did the hearsay come from?
Gareth Jenkins: Just chat around the office. As I say, I wasn’t directly involved in all that but, clearly, people were talking about – I think I was aware that Post Office did not have the right to see various design documents that were being produced.
Mr Beer: What do you mean that they were buying a black box?
Gareth Jenkins: That it was up to Fujitsu exactly how they designed the application and the details of how the internal design was done was something for Fujitsu or ICL to worry about, and it wasn’t something that Post Office would get involved in.
Mr Beer: Would a consequence of that, in relation, for example, to EPOSS, mean that it would be difficult, if not impossible, for the Post Office to ensure that there was data integrity?
Gareth Jenkins: I don’t think it went that far. There were certainly joint test teams. There was a lot of testing going on. There were Post Office testers involved in testing the system to make sure it came up with the right output. So I – yeah.
Mr Beer: He suggested that the fact that the Post Office had purchased a black box and had no rights to see how the black box worked meant that it was difficult, if not impossible to ensure data integrity within the Horizon system; you disagree with that?
Gareth Jenkins: I would disagree with that, yes.
Mr Beer: You say that the answer to that is that they were entitled to participate in testing?
Gareth Jenkins: Yes, as I say, I wasn’t really involved at that sort of level at that time. I was just looking after making sure that the agents worked, rather than worrying about how the counter worked and things like that.
Mr Beer: Were you conscious, and were those with whom you worked conscious, that the nature of the contract was a PFI contract and, therefore, the client had no rights, no visibility, over of the way the system was built?
Gareth Jenkins: I think I was aware of that, yes.
Mr Beer: What effect did it have on your work?
Gareth Jenkins: Nothing really, other than – at that time I wasn’t having any real contact with the customer anyway. I only started getting involved with the customer later and it was really a case of – and the main work I was doing with the customer in the early 2000s was with agreeing these banking specs and there I was working very openly with the customer because they were their specs that were defining the interfaces as to how Horizon was going to interface with, initially, the banking engine and then also with Streamline for the debit card payments. So that was something that wasn’t part of the black box.
Mr Beer: Why was the black box being mentioned; why was it being talked about?
Gareth Jenkins: Well, I don’t think it was being talked about as such, it was really – I probably picked that up from Jeremy Folkes’ stuff. As I say, I was aware that there was a PFI contract and that it meant that we didn’t necessarily need to share detailed design documents with the customer because it was Fujitsu’s intellectual property.
Mr Beer: Was that the usual way of working or an unusual way of working?
Gareth Jenkins: I’d not worked on the contracts like that before. Before, I’d been in an internal development role actually producing products that were sold rather than working as a client for a specific customer like that.
Mr Beer: So you wouldn’t know one way or another?
Gareth Jenkins: Correct.
Mr Beer: In relation to your fourth line support role, you tell us in your witness statement, it’s the first witness statement, at paragraph 25:
“I would only have been aware of bugs, errors and defects specifically allocated to me by third line support or where I was asked to route the issue to the correct person in fourth line support, or where I was asked for specific input on the problem because of my particular expertise.”
Correct?
Gareth Jenkins: Yes, yes.
Mr Beer: Within your roles in Legacy Horizon, did you have any oversight of all bugs, errors and defects within the system at any given time?
Gareth Jenkins: No.
Mr Beer: So you couldn’t, at any particular time, speak to the existence or non-existence of bugs, errors and defects in the system?
Gareth Jenkins: Not from my personal knowledge, no.
Mr Beer: If you did not have knowledge of, or oversight of, all bugs, errors and defects within the system at any given time, who did?
Gareth Jenkins: I would have thought the SSC would and – there was an area called Customer Services, SSC was part of that. There were also problem managers who would look at specific incidents. Now, incidents could be to do with bugs, errors and defects or, more commonly, they would be operational issues, where a bit of the network went down or connections to banks went down for a few hours which clearly was – caused fairly catastrophic effects at the time, and so their job was to manage these sort of things. So I’d have thought that, if there was an overall knowledge of how well the system was behaving, it would be part of their role.
Mr Beer: We’ve heard from a lot of people from the SSC, and they’ve told us that there were between 20 and 30 people at any one time working in the SSC – sometimes working shifts, some working at home, some working in the office – that they only knew about the bugs that they were allocated on the stack and that it was correct that there might be informal sharing of information by people who were sitting next to each other or talking about it at the water cooler but there was no formalised system for sharing knowledge within the SSC; did you know that?
Gareth Jenkins: I didn’t know that. They would obviously be able to speak better about that sort of thing than I was. I wasn’t part of the SSC.
Mr Beer: But when I asked you who may have oversight of all bugs, errors and defects within the Horizon system, you pointed to the SSC.
Gareth Jenkins: As part of that but there was also – there was – the whole point of the Customer Services Directorate was to actually manage the interface with Post Office, including knowledge of what had been going on the system, what sort of problems there were and things like that.
Mr Beer: That’s a day-to-day function. What I’m looking for, it might be described as an all-seeing eye, but somebody who took a step back and would be able to say, “We’ve had these problems in the past with Horizon, these bugs, errors and defects, and, at the moment, we’re servicing the existence of these possible bugs, errors and defects”; was there any such person?
Gareth Jenkins: I can’t immediately think of anyone who would have that specific role.
Mr Beer: When you were in fourth line support, how would you generally become aware of such bugs, errors and defects as you did become aware?
Gareth Jenkins: There was a system called PEAK and it had a predecessor called PinICL, and the way that operated, that was effectively a database of the various defects, and that included those found during testing as well as those during live, and what happened was that PEAKs or PinICLs would be allocated to a particular person, and I think you’d normally get an email saying, “You’ve had this PEAK that’s allocated to you”, you could then log in to the PEAK system, have a look at the PEAK and see what it said and then decide what you needed to do with it next.
Mr Beer: Again, same questions: within your various roles within Horizon Online, did you have oversight of all bugs, errors and defects?
Gareth Jenkins: Not as such but I did have a – probably a better view of them, particularly during the pilot phase of Horizon Online because, during the pilot phase, there was a project team in place that was managing the pilot and taking very close monitoring of any issues that were going on during the pilot, and I would often get called in by that team to investigate particular problems particularly ones – those that affected the counter.
Mr Beer: Are the pilot phase of Horizon Online was over, did you have oversight of all bugs, errors and defects within the Horizon Online system?
Gareth Jenkins: No.
Mr Beer: Same questions: if not you, did anyone?
Gareth Jenkins: Again, I think it was something that Customer Services were supposed to be monitoring and managing. Now, whether they did – had that sort of stand back oversight role that you’ve described earlier, I can’t think of any one person that did actually have that sort of role but I hadn’t thought of it that way at the time.
Mr Beer: Again, same questions: how would you generally become aware of such bugs, errors and defects as you became aware of in Horizon Online?
Gareth Jenkins: The PEAK system was working. It was the same PEAK system that operated for both Legacy Horizon and Horizon Online.
Mr Beer: So for both Legacy Horizon and Horizon Online, you became aware of things that might be bugs, errors and defects through the PinICL and then the PEAK system?
Gareth Jenkins: There was that but there was also the problem management system. So then when something was identified as being a serious problem – and I think we’re probably going to go on and talk about the receipts and payments mismatch problem at some stage – then I was often called in to help – well, firstly, analyse the problem and explain that problem so that Fujitsu management could understand what it was all about, and I often got involved in explaining it to the equivalent people in Post Office about what the impact of that was, and what could be done to remedy it, both in terms of fixing the code but also fixing whatever the business impact of the problem was, which wouldn’t necessarily be fixed directly just because you’d fixed the code.
Mr Beer: So, for you personally, there were two ways in which you became aware of bugs, errors and defects in Horizon –
Gareth Jenkins: Yes.
Mr Beer: – firstly, through your role – the ‘10 per cent of your time’ role –
Gareth Jenkins: Yeah.
Mr Beer: – with fourth line support, and that would largely be happenstance, would it, ie which PEAKs were allocated to you?
Gareth Jenkins: It would – the 10 per cent wasn’t just for the PEAKs. So any time that I spent talking about serious incidents and things like that, that was all part of what I saw as the 10 per cent of the time.
Mr Beer: In any event –
Gareth Jenkins: Yes.
Mr Beer: – in your role in fourth line support –
Gareth Jenkins: Yes –
Mr Beer: – it would be largely happenstance whether it was you or somebody else, one of the other people performing fourth line support, who had a particular bug, error or defect allocated to them?
Gareth Jenkins: Yes, it was based really on the skills and what area the problem had happened in.
Mr Beer: If it was largely happenstance that you became aware of a bug, error or defect, in both Legacy Horizon and Horizon Online, how could you subsequently give full, complete and accurate evidence about the existence or non-existence of bugs, errors and defects in Horizon?
Gareth Jenkins: I was – what I was aware of was the fact that bugs that actually impacted the accounts were very rare, there was good monitoring in place to detect them and they got fixed shortly afterwards. So, in terms of what was actually there in the live system at any one time, it was very rare for there to be bugs there that would cause problems and, therefore, I was confident in the way that the system was operating, that it was operating correctly.
Mr Beer: To give that kind of very confident answer to a court, wouldn’t you want to know of the existence of each of the bugs and how they’d been resolved and whether, in fact, there was any ongoing impact?
Gareth Jenkins: I didn’t realise at the time that I needed to do that. So at the time, no, I didn’t think I needed to do that. Obviously, with hindsight, I realised that maybe I should have been doing more research but, at the time, I felt that that was sufficient.
Mr Beer: The “that” in that sentence, meaning a general confidence in the system and the way that it operated?
Gareth Jenkins: Yes, and the processes that were in place to actually control things.
Mr Beer: So generally confidence in the system, plus processes that you thought were working, allowed you confidently to give a generalised view; is that right?
Gareth Jenkins: I think so, yes.
Mr Beer: During your employment within Fujitsu, did any of your roles entail any overall responsibility for monitoring, identifying or fixing bugs, errors or defects?
Gareth Jenkins: Well, we’ve talked about some of those already, so I’m not quite sure what –
Mr Beer: Did they always come to you?
Gareth Jenkins: Oh, no, no. I mean, they go to whoever was the relevant person. So there were certain areas that – normally, they just went straight to the Development Teams and stint get involved in things. I tended to get involved if it was seen as being a broader issue or something like that. Not necessarily me, there were other people, also in particular areas, would get called with.
Mr Beer: How many other people?
Gareth Jenkins: I’m not sure, sort of another half a dozen, a dozen, something like that.
Mr Beer: What were the circumstances in which you would get called in?
Gareth Jenkins: It’s a bit difficult to define, really. Certainly, when we got on to things like Horizon Online, if it was seen as having an impact on the accounts, then we realised that that was a very serious problem and that I would be getting involved in those, which is why I got involved with what I see as the main serious issues in Horizon Online. And I was involved –
Mr Beer: Receipts and payments mismatch?
Gareth Jenkins: Receipts and payments mismatch, the local suspense issue and Dalmellington, those are the ones that I see as being serious issues with Horizon Online, and I got involved in all of them because they saw me being helpful in terms of my expertise and background for those particular cases.
Mr Beer: That was my next question: why did they come to you?
Gareth Jenkins: Because I’d had experience of dealing with things like that. I think it was felt that I was in a good position of turning some of the technical jargon into something that laypeople could understand better. Now, having read my witness statement, you may disagree with that, but it was felt that I was in a good position to be able to actually explain the technical problems in a way that could be understood and to do that, both in terms of written notes and in meetings with the customer when these problems occurred.
Mr Beer: What was the threshold for you becoming involved?
Gareth Jenkins: I’m not sure. It’s – I would get involved by probably a phone call or an email from someone, either in the SSC or from problem management, and they’d say “Gareth, can you come and give us a hand, look at this problem and see what’s going on”.
Mr Beer: For all the problems that didn’t get referred to you, where that trigger was not pressed, how did you become aware of those, or didn’t you?
Gareth Jenkins: Probably the answer is I didn’t.
Mr Beer: When you started to become the man that the Post Office relied on to give evidence in court, did anyone say, “We need to make sure that Mr Jenkins knows about the things that are not referred to him”?
Gareth Jenkins: No.
Mr Beer: Did it occur to you, “I might need, before going to court, to find out about the things that have not been referred to me, that aren’t these level of particular seriousness or which have financial or economic consequences”?
Gareth Jenkins: That didn’t occur to me but, again, as I’ve said before, I was confident – and possibly wrongly so – that when problems did occur, they were quickly fixed and they weren’t left to sort of fester in the system, to have a larger impact.
Mr Beer: I think you’ve acknowledged it a couple of times already, knowing what you know now, would you adopt the same approach of relying on your confidence?
Gareth Jenkins: I think I’d have to say that, with hindsight, I would have done things differently, yes.
Mr Beer: What interactions did you have with the third line of support, the SSC, in relation to identifying and rectifying bugs, errors and defects?
Gareth Jenkins: I’m not quite sure exactly what you mean by that question, sorry.
Mr Beer: Yes. What was the nature of your communication with third line support?
Gareth Jenkins: Usually, either by exchange of emails or telephone conversations, and things like that. The third line support team were in a locked area, so it was actually quite difficult to actually go and – I did occasionally go and visit them and talk to them at their desks and things like that but, generally, that was a difficult thing to do because of getting through the security systems to actually get to where their desks were. So it was normally done by sort of phone or email or in meetings, if they came out of their area to our slightly less secure areas.
Mr Beer: Thank you. Can I turn to EPOSS and Riposte, please.
Gareth Jenkins: Yes.
Mr Beer: In relation to the EPOSS aspects of Horizon, you tell us, it’s paragraph 21 of your second witness statement – no need to turn it up:
“In summary until around 2003 I do not recall having anything to do with EPOSS other than indirectly when it had an impact on my work with the agents. From around 2003 onwards, I began to gain technical expertise and a practical understanding of how EPOSS operated.”
Is that –
Gareth Jenkins: Yes, I’d agree with that.
Mr Beer: You’ve told us already that Escher was responsible for provision of the Riposte software; is that right?
Gareth Jenkins: Correct, yes.
Mr Beer: And that that comprised of two main parts: the message store and the desktop counter; is that right?
Gareth Jenkins: Correct.
Mr Beer: The latter of those, the desktop counter, was the basis for the EPOS system?
Gareth Jenkins: Correct.
Mr Beer: I think you went to Escher in Boston for a week in 1996; is that right?
Gareth Jenkins: Yes, I think I went back there for two or three ad hoc trips after that, but that was my initial – that was, effectively, my first week on the Post Office Account, when I got sent to Boston for a week to learn about how Riposte worked.
Mr Beer: If you did not have anything to do with EPOSS, other than indirectly, why did you go to Boston for a week in 1996?
Gareth Jenkins: Because of the – that was talking more about how the Riposte message store was working and, also, there was specific interfaces that Riposte had that could be used by the agents, so that was covered in the first week of what I think was a three-week training course. It then went on to how did the counter applications operate and I wasn’t particularly interested in that at the time, so it was felt that, rather than stay in Boston for another couple of weeks, I might as well come home. So I did.
Mr Beer: So is the answer that you went to Boston to learn how the EPOS system worked, in particular how it might interact with the work you were undertaking?
Gareth Jenkins: No. I went to Boston to learn how Riposte worked, to – and, therefore, how the agents could actually use it. There may have been slight mention of how applications could work in that first week but I think the detail of that was really covered afterwards, and I just didn’t get involved in any of that.
Mr Beer: In relation to the development of EPOSS before the rollout of Legacy Horizon, we’ve heard evidence from David McDonnell; did you know him?
Gareth Jenkins: Vaguely. I have a vague recollection of him.
Mr Beer: He told the Inquiry:
“My understanding was that Gareth [that’s you] Jenkins worked alongside another Chief Architect under Alan Ward.”
Was that correct?
Gareth Jenkins: I certainly did have a dotted retail line relationship to Alan Ward, who was the Chief Architect at that time. I thought that David McDonnell actually suggested I was the Chief Architect and he’s certainly mistaken in that belief.
Mr Beer: He continued:
“Gareth Jenkins’ responsibility was specifically to the EPOSS counter system.”
Is that correct?
Gareth Jenkins: That is not correct.
Mr Beer: He said:
“As the Chief Architect, I would have expected him to be much more involved in overseeing a lot of the previous coding standards and methodologies, and things like that, and certainly the design documents.”
You’ve, I think, answered this question already: were you the Chief Architect of Horizon?
Gareth Jenkins: No, I wasn’t.
Mr Beer: Who was the Chief Architect of Horizon?
Gareth Jenkins: Alan Ward was at that time and I know he was quite involved in the development of EPOSS at that time because there were quite a lot of changes being made and he used to regularly fly off to Boston for a couple of days almost every week, to actually work with Escher on changes to EPOSS. Now, exactly what there was changes were, I’ve no idea; I was just aware that it wasn’t a good idea to try and find him on a Monday and a Tuesday because he was probably in Boston.
Mr Beer: Okay, and if you could just slow down your evidence slightly please.
Gareth Jenkins: Sorry.
Mr Beer: He told us that there was or had been:
“… a lack of formalised sign-off designs, a lack of discipline, a lack of professional qualification in key positions and a total disengagement of the Chief Architect, Gareth Jenkins, [that] there were poor coding standards, no methodology in place and no unit testing. The issues were critical, making the EPOSS product unstable. They were known to everyone in the building.”
Taking those in turn, you’ve already told us that you were not the Chief Architect of Horizon but, putting that aside, in relation to EPOSS, did you know whether or not there was an existence or a lack of formalised sign-off designs.
Gareth Jenkins: I’ve no idea what they had. I know what we had in the Agent Team and things were formally signed off, I had no idea what sort of processes they were following in the EPOSS Team.
Mr Beer: Did you know whether or not there was a lack of discipline and a lack of professional qualifications in the EPOSS Team?
Gareth Jenkins: I wasn’t close enough to it to know either way.
Mr Beer: Did you know, one way or the other, whether they were using poor coding and there was an absence of methodology in place?
Gareth Jenkins: I have no idea.
Mr Beer: Did you know whether or not there was unit testing within the EPOSS team?
Gareth Jenkins: I don’t know.
Mr Beer: He says that those issues were critical and were known to everyone in the building. You’re telling us that –
Gareth Jenkins: I would agree they were critical. As to being known to everyone in the building, the other thing to remember is that the EPOSS team was based in Feltham and I was based in Bracknell. Yes, I used to visit Feltham a couple of times a week but I wasn’t, actually, based in the building all the time so, therefore, I would have had less knowledge of what was going on in the background.
Mr Beer: So these rather important facts that he told us about, you were entirely unaware of?
Gareth Jenkins: Correct, and I’d agree with him they are important facts.
Mr Beer: Were you aware of an EPOSS Task Force being set up by ICL?
Gareth Jenkins: I don’t think I was aware of it at the time. Obviously, I’m aware of it now from the documents I’ve been shown but I don’t think I was aware of it at the time.
Mr Beer: Can we look, please, at FUJ00080690. If we just look at page 2, please, we can see this was originally dated 18 September 1998 –
Gareth Jenkins: Yes.
Mr Beer: – and this finalised version is dated 14 May 2001.
Gareth Jenkins: Yes, I can see that.
Mr Beer: If we go –
Gareth Jenkins: Yeah.
Mr Beer: – back to page 1., it’s a “Report on EPOSS PinICL Task Force” and it:
“… reports on the activities of the EPOSS PinICL Task Force, which was in place between 19 August and 18 September 1998 to reduce to manageable levels the EPOSS PinICLs outstanding at that time.”
You can see, although it’s on the second page described, I think, as a finalised version, the status is still a draft.
Gareth Jenkins: Yes.
Mr Beer: The distribution does not include you. It’s to Messrs Austin, Bennett and McDonnell, but then “Library”; can you see that?
Gareth Jenkins: Yes, I can see that.
Mr Beer: What was the nature of ICL’s document library at this time?
Gareth Jenkins: There was a document management system, I think – I can’t remember exactly when that became formalised. I think Matthew Lenton may have described something about this when he appeared a couple of weeks ago and there was a document management system to which all documents were put and people could look at documents if they wished to do so but, normally, you would only be looking at documents that either you’d written yourself or were needing – being asked to review in that library.
Mr Beer: So you wouldn’t –
Gareth Jenkins: I wouldn’t have seen this document, unless someone pointed it out and said, “What do you think of this document?”, and I’m not aware that they ever did.
Mr Beer: So it wasn’t used as a knowledge base for those at your level to peruse or look at your leisure?
Gareth Jenkins: I suppose I could have done, if I’d got nothing better to do, but I don’t think I would have been particularly interest in what was happening with EPOSS at that time.
Mr Beer: I think you’ve seen this document subsequently –
Gareth Jenkins: I’ve seen this document recently, yes.
Mr Beer: – and you know – I’m not going to take you through it, we’ve seen it many times – it gives a very damning account of the problems with EPOSS?
Gareth Jenkins: I agree.
Mr Beer: Mr McDonnell told us in his evidence that he presented this report to you; is that true?
Gareth Jenkins: I don’t think he presented it to me. I think he presented it to whoever commissioned it, which I think I understand now is Terry Austin.
Mr Beer: And that you “denied the issues pointblank and ran off to hide in Bracknell, and avoided contact with the team”?
Gareth Jenkins: I don’t think that’s correct. I was – as I said before, I was based in Bracknell, not in Feltham. I don’t think I was hiding there particularly. I don’t think I had any involvement with the EPOSS PinICL Task Force.
Mr Beer: He said that:
“We managed to get Gareth down to the counter team twice.”
Do you remember that?
Gareth Jenkins: I don’t. I do remember having some involvement with him with some issues with Logistics’ visa system, which I think is something he moved on to work with after he left – was no longer with the counter team, but I don’t recall having anything to do with him in terms of EPOSS.
Mr Beer: He said that his team tried to engage you in a conversation about the missing API; do you remember what that is?
Gareth Jenkins: I think that was to do with the LFS thing that I mentioned previously. So I think that was in a role that he took on after the EPOSS team. I can’t remember the gory details of it I’m afraid but I –
Mr Beer: He said that you were very defensive of it and said, “No, there’s nothing wrong with it as it is”.
Gareth Jenkins: I think it was more a case of pragmatism, in that I think he wanted to totally change the way that the Escher software interfaced with the applications. And I could understand why he felt that there may have been a better way of doing it but I could also see that there was no way that Escher was going to go and totally rewrite their software when it was basically operating and doing what it needed to do.
The fact that there was a better way did not necessarily mean that it would be justifiable to actually change it. So I think there may have been some sort of conversation along those lines but, as I say, it’s a very long time ago and I can’t really remember the details but that’s a vague memory I have of that area.
Mr Beer: He said that he tried to engage you to lend your “political design weight” behind at least the cash account being rewritten; is that correct?
Gareth Jenkins: I don’t have any recollection of being involved with anything to do with the cash account at that time.
Mr Beer: He said that he was unable to engage you to get you on his side and to lend your persuasive weight to persuade Terry Austin to rewrite the cash account; is that correct?
Gareth Jenkins: I have no recollection of that and I can’t see how I would possibly have been able to persuade Terry that it had something to do with the cash account when it was nothing to do with my area.
Mr Beer: Did you have what he describes as “political design weight”: a certain amount of power?
Gareth Jenkins: Probably not so much at that time. I think maybe later on then I – that could be said but, at that time, my main role was to do with the way the agents worked, which was a fairly small cog in the system. It was a fairly central cog and, yes, I think people did tend to listen to me when I had technical ideas, but I think that’s over-stretching things, what you’ve just read out.
Mr Beer: He said that you became evasive with him and he was never able to persuade you to come back down to Feltham.
Gareth Jenkins: I just can’t remember those sort of interactions. I’ve described to the best of my recollection the only sort of interaction that I might have had with him, and that’s a very, very vague memory, I’m sorry.
Mr Beer: Leaving aside your involvement and role in the way that Mr McDonnell has described, would you agree with, looking back now, what he said about the EPOSS team in the late 1990s? He said the team was:
“… like the Wild West, there were no standards in place, there were no design documents. The culture of the Development Team was – I wouldn’t say it was a holiday camp. It was a free format. There was no structure, no discipline. It was crazy. I’ve never seen anything like it.”
Gareth Jenkins: I wasn’t close enough to actually form any opinion on that, I’m afraid.
Mr Beer: Do you remember going down to his team on at least couple of occasions?
Gareth Jenkins: Not really. I – as I say, the only recollection I have is discussing issues to do with LFS, which was a later role that he moved on to. I don’t remember having any real involvement with the EPOSS team at that sort of time.
Mr Beer: That can come down. Thank you.
If you’d been asked a question at rollout time, to what extent does the EPOS system function well and effectively; what would your answer have been?
Gareth Jenkins: I don’t know.
Mr Beer: You –
Gareth Jenkins: I mean, my answer would have been “I don’t know”.
Mr Beer: You wouldn’t know one way or the other?
Gareth Jenkins: Yeah.
Mr Beer: So if you were to give evidence about that, you wouldn’t be able to say one way or the other?
Gareth Jenkins: Correct.
Mr Beer: If you were to give evidence later on about events that were happening in 2000, you wouldn’t be able to say whether EPOSS was functioning well or not?
Gareth Jenkins: Not in 2000, no. I believe it had stabilised by the time I did get involved with it later on but, back in – what state it was in 2000, I don’t know how stable it was and, clearly, as part of the work I did in 2018, I did find some fairly serious issues that had occurred at that time. But –
Mr Beer: What were the issues you found out in the course of the Group Litigation –
Gareth Jenkins: Yeah, that’s right, yes.
Mr Beer: – in 2018?
Gareth Jenkins: And those work ones that I’d not been aware of until I came across them as part of the Group Litigation.
Mr Beer: Can you, rather than me doing it on – it’ll be late on Thursday – can you summarise what those are now?
Gareth Jenkins: The main one I can think of is the data tree build issue that came – that came around. That’s the one that sort of sticks in my mind at the moment. You may find others that you may wish to prompt me about but that’s the one that sticks in my mind.
Mr Beer: Was that a surprise when you found this out in the course of the Group Litigation?
Gareth Jenkins: That such serious problems had occurred and seemed to – though again, with that, the problems did get fixed but I was surprised in some of the cases how long it took to fix them and there were certainly some fairly inappropriate comments in some of the PEAKs in terms of whether things should be fixed quickly or not, which I certainly wouldn’t have agreed with if I’d seen them at the time. But then it wasn’t my job to see them at the time.
Mr Beer: Was your surprise compounded by the fact that you had, in the interim period, between the relevant events happening with the data tree build failure and you discovering them in 2018, you’d given evidence in written witness statements, and on one occasion orally, in court?
Gareth Jenkins: No, I didn’t think it was a problem as far as that was concerned because, at the times I was giving evidence for, I believed that the EPOS system was stable and was operating correctly. So the fact that there were problems during the pilot and the rollout don’t necessarily mean that the problems carry on into the system. I was confident in the way that problems were being picked up and fixed and knew things were being put into the system to actually manage the issues that are being found early on.
Mr Beer: We’re going to come back to the kind of answer that you’ve just given quite a lot over the course of the next two or three days but, in the answer that you’ve given just now, do you agree that you are focusing on whether it can be shown that an issue had an impact in the case that you are looking at –
Gareth Jenkins: Yes.
Mr Beer: – rather than giving evidence about the existence of bugs, errors and defects within Horizon, how quickly they manifested themselves to Fujitsu, how promptly the Post Office was told about them and the extent to which subpostmasters had themselves been informed?
Gareth Jenkins: Yes, I was primarily looking at what was happening in a particular branch at a particular time.
Mr Beer: Was that your mindset when you were giving evidence?
Gareth Jenkins: Yes, that’s how I approached support, and I didn’t see the giving of evidence as being any different from what I was doing in my day-to-day support job. What I’d been asked to do in support, as saying this has happened, should it have happened and, if not, what’s gone wrong? And I just approached things on the same sort of basis.
Mr Beer: Thank you. Can we turn to the IMPACT Programme. You tell us in paragraph 20 of your witness statement that you became involved in Project IMPACT, as it was called. That was the name given to it by the Post Office and it was essentially the replacement of the back-end accounting system. Is that –
Gareth Jenkins: Yes.
Mr Beer: – an overall description of it?
Gareth Jenkins: Yeah.
Mr Beer: Given your role in the project, can you describe shortly what Project IMPACT was designed to do, so far as the back-end accounting system was concerned?
Gareth Jenkins: Post Office had a propriety in-house system called, if I remember rightly, CBDB – I can’t remember what that stands for – where they used to process their accounts, and they decided they wanted to go for a more industry standard system and SAP was the industry standard, and probably still is, for handling accounts, and so it wanted to move on to that.
Now, CBDB worked on the basis that it would take the information from the cash accounts that were signed off each week and derive everything from that, which meant that it was always about a week behind what was really going on, whilst with an SAP system, they thought they could get a much more realtime knowledge of what was going on in the system. So with SAP they would know what had happened yesterday and, therefore, be in a much better position, particularly to control their cash flow, because they saw the amount of cash flowing through the post offices as being a fairly critical measure because, obviously, cash had to be paid for, and it was all to do with Treasury interest rates and things like that that I didn’t really understand.
But I understood that having a good handle on exactly how much cash you had where made a big difference in terms of Post Office finances.
Mr Beer: Can we look at your first witness statement, please, WITN00460100, at page 13, please. It’ll come up on the screen. At paragraphs 45 and 46 you say:
“… I believe that ICL/Fujitsu was only involved in one aspect of Project IMPACT, which was implementing changes to Legacy Horizon that would enable it to interface with the new back-end system. This was implemented in two phases as part of the S60 release in 2004 and the S80 release in 2005. My involvement was to design the changes which [Post Office] wanted.
“I remember that these phases of the work required a major re-engineering of the accounting processes in each [Post Office] branch, [for example] removing the cash account and introducing the branch trading statement.”
Gareth Jenkins: That’s correct, yes.
Mr Beer: So, as the counters specialist, which is what you were, you redesigned part of the Legacy Horizon system that produced the accounts for the subpostmaster to sign off; would that be fair?
Gareth Jenkins: I oversaw that. I think the main detail was done by other designers but my main role was actually identifying exactly what the output needed to be of the system. So, yeah.
Mr Beer: So you were the lead designer –
Gareth Jenkins: Yes.
Mr Beer: – of this part of the IMPACT Programme –
Gareth Jenkins: Correct.
Mr Beer: – ie that part of the programme which redesigned the accounts for subpostmasters to sign off?
Gareth Jenkins: Yes.
Mr Beer: Can we look, please, at POL00038878. If you look at the foot of the page, you’ll see it’s dated 3 March 2004?
Gareth Jenkins: Right, yes.
Mr Beer: If we go to the top of the first page, it’s described as “Branch Trading Reporting, Management and Control and Transaction Management: Conceptual Design”. Is, essentially, this a design document for the IMPACT Programme?
Gareth Jenkins: Not really. I’d call it a requirements document. So Post Office decided that the term “conceptual design” was what should be – the document be called but, to me, it was a requirements document. This was setting out what Post Office wanted to happen and this was actually written by Post Office, though aided by a Fujitsu requirements analyst called Phil Boardman, who I believe has been in front of the Inquiry a year or so ago. But this –
Mr Beer: And you: you contributed to this?
Gareth Jenkins: I was involved in the workshops doing it. I don’t think I wrote any of the text in there but I was involved in commenting on it and helping put together the diagrams. But it was basically understanding what Fujitsu wanted to do, and my involvement was really a case of saying, well, if you want to do that, is that practical? Would it be better to do it in a slightly different way to make it more practical from a (unclear) point of view?
Mr Beer: Just scroll down. We see your name as a contributor.
Gareth Jenkins: Yes.
Mr Beer: Are you saying you didn’t contribute to this?
Gareth Jenkins: Well, I contributed ideas to it but I don’t think I actually wrote any of the words in it.
Mr Beer: Can we go to page 22, please, and paragraph 6.7, an “Overview”. If we just read this together:
“This area of functions has the purpose of providing mechanisms to make adjustments to branch accounts, to correct errors and ensure branch accounts align with the Post Office Accounts within POLFS. Various mechanisms are available to identify errors that require adjustments, and the discrepancy management functions may be initiated from various places across the business. The main areas will be from within the Branch, from [Post Office] clients or centrally via distributing electronic transaction corrections. These corrections will replace the current error notice processes and should not involve any manual paperwork or processing. They will be received and actioned via Horizon and will be distributed more quickly, potentially only days after an error is recorded.
“The analysis has also identified requirements to more tightly control and police the use of the suspense account within the branch accounts. Only a limited subset of the existing suspense account products will be retained. The contractual requirements for agents to make good unknown errors in branch accounts will be used instead.”
What did you understand that second part of 6.7 to mean?
Gareth Jenkins: I understood that they wanted to limit the use of the suspense accounts. My understanding was that there were a number of suspense accounts and postmasters could actually post money to the suspense accounts for some sort of business reason which I never really did understand, and they wanted to reduce the amount of money that was being posted into the suspense accounts. Now, exactly why they wanted to do that, that was part of Post Office’s business for trying to save money, and, now, in what way it did that, I didn’t really understand.
Mr Beer: Were you told why they wanted to “more tightly control and police the use of the suspense account”?
Gareth Jenkins: Not in so many words. What I believe was behind a lot of this was they were trying to reduce the number of staff that they had in Chesterfield back office, which were monitoring things like this. So they were trying to actually automate things and reduce the flexibility on the postmasters which I thought was a business decision for Post Office and nothing for me to get involved in.
Mr Beer: Were you told that the suspense account was considered by the Post Office to be a vehicle for postmasters to hide fraud?
Gareth Jenkins: I can’t remember. There were various discussions about behaviours of postmasters and things like that but – and I know that Post Office were quite keen to get rid of the suspense account. In fact, they didn’t get rid of it altogether but I know they were keen to reduce its usage and there was, indeed, some reduction on its usage.
Mr Beer: Given that reduction on usage, if there was a discrepancy in a subpostmaster’s account, after Project IMPACT, the subpostmaster would have to accept it and pay it before they could roll over into the new trading period, correct?
Gareth Jenkins: That is what I understood the Post Office wanted the system to do, so yes.
Mr Beer: And, if they didn’t pay, the Post Office would, in the language of this document, enforce contractual requirements; did you understand that?
Gareth Jenkins: I didn’t really consider that sentence particularly and I’m not sure that I really understood what that actually meant. And my focus was on what exactly was it I had to actually implement in the way that Horizon worked, and what I took from that was I had to simplify the menus for the suspense accounts, which was actually just a reference data change.
Mr Beer: Given some of the suspense account products were to be removed and the consequence of that would be for subpostmasters to have to accept discrepancies and pay them before they could roll over into new trading periods, was there any consideration, to your knowledge, of the importance of the accuracy of the figures produced by Horizon?
Gareth Jenkins: Well, I always felt it was important that the figures produced by Horizon were accurate and I wasn’t aware there was any doubt or challenge to that at that stage. So, therefore, I didn’t see that as being an issue as part of what we were doing here.
Mr Beer: So is the answer: no, that issue didn’t even arise, that the change that we’re making, the removal of this facility, means that it is very important that the accounts that are to be signed off by subpostmasters are 100 per cent accurate?
Gareth Jenkins: Yes, and I always felt that they should be 100 per cent accurate.
Mr Beer: Of course they should be but were you satisfied that they were 100 per cent accurate the whole time?
Gareth Jenkins: I don’t know that I’d actually thought it through in those sort of the terms at the time.
Mr Beer: We’ve heard evidence from Susan Harding, who told us that the business drivers for the IMPACT Programme were to reduce the costs to the business, increasing accounting efficiency and reducing the losses to the Post Office. Do you agree that they were the business drivers that you were told about?
Gareth Jenkins: I can’t remember the details of the drivers but, certainly, I know that Post Office were trying to reduce their costs. I’ve not twigged so much about the reducing losses but they were certainly trying to reduce their costs on the back end.
Mr Beer: Were you told that a business driver was to improve the Post Office’s debt recovery from subpostmasters?
Gareth Jenkins: I don’t specifically remember that one but I can’t say that I hadn’t seen it in a document.
Mr Beer: You tell us in your first witness statement – no need to turn it up, it’s paragraph 41 – that the effect of the project was to reduce staff costs?
Gareth Jenkins: Yes.
Mr Beer: Is that what you were told, as the driver, or as being the driver –
Gareth Jenkins: That’s my memory –
Mr Beer: – namely: this is all about reducing the number of people at Chesterfield or elsewhere that need to be looking at suspense accounts, rather than the reduction of losses and the improvement of debt recovery?
Gareth Jenkins: That’s my memory but I can’t be definitive because it’s an awfully long time ago now.
Mr Beer: You tell us in your witness statement – paragraph 43 of your first witness statement – that the decision to remove the ability of subpostmasters to post discrepancies to a suspense account was because the Post Office “took what I assumed to be a business decision to remove this functionality”. Was the only thing you were told about the business decision that it was to reduce staff costs?
Gareth Jenkins: I don’t know that I would have taken too much notice as to why they were doing things. My main focus on – was what was it they wanted changed in the system, not why they were doing it.
Mr Beer: Ms Harding told us that it was agreed during the design of the IMPACT Programme that the suspense account would be removed because, historically, it was used by subpostmasters to hide discrepancies in their accounts. Were you told that?
Gareth Jenkins: I can’t remember explicitly, exactly, what I was told or what I wasn’t told.
Mr Beer: What understanding did you have as to how the local suspense account had worked historically?
Gareth Jenkins: We need to be careful here about terminology because one of the things introduced here was something called the “local suspense account”.
Mr Beer: I’m talking about at or before 2005, what was your understanding as to how the local suspense account worked? Not the –
Gareth Jenkins: That’s where I’m getting confused. We introduced a new concept as part of IMPACT called the local suspense account so, as far as I’m concerned, there wasn’t a local suspense account prior to IMPACT. But I think I know what you’re referring to.
Mr Beer: Yes. What’s the answer?
Gareth Jenkins: I didn’t know how it was used or what it was used for. I just assumed this was part of the business processes that happened. I think I understood that, in order to post something to local suspense account, you did need to have permission. How that actually operated – presumably through NBSC – but exactly how that operated I didn’t know and it didn’t really concern me, particularly.
Mr Beer: Do you remember being told by the Post Office that subpostmasters “hid their discrepancies” in the suspense account?
Gareth Jenkins: I can’t remember, sorry.
Mr Beer: Do you remember any evidence of subpostmasters hiding their discrepancies in suspense accounts?
Gareth Jenkins: I don’t think I would have seen any evidence of anything like that, one way or the other.
Mr Beer: Were you given any information as to why subpostmasters might hide their discrepancies in suspense accounts?
Gareth Jenkins: Not that I can remember. As I say, I wasn’t that interested in the whys. I was interested in what it was that I was being asked to do in the system, not why it was being done.
Mr Beer: You had no professional curiosity in that?
Gareth Jenkins: I mean, I went to a number of workshops about these things and I sat through number of workshops, but exactly what was discussed in them, because this is over 20 years ago now, I just can’t remember the details. So I may have been aware of some of those things at the time, I just have no memory of it now, I’m afraid.
Mr Beer: Thank you.
Sir, that’s an appropriate moment for the morning break. Can we break until 10.20 (sic), when we’ll turn to Mr Castleton’s case.
Sir Wyn Williams: 11.20, I think.
Mr Beer: 11. Yes, quite right, sir.
Sir Wyn Williams: All right then, 11.20.
(11.08 am)
(A short break)
(11.20 am)
Mr Beer: Good morning, sir, can you see and hear us?
Sir Wyn Williams: Yes, I can.
Mr Beer: Thank you very much.
Mr Jenkins, thank you. Mr Jenkins, can we turn to Lee Castleton’s case, please.
Gareth Jenkins: Yes.
Mr Beer: Can we start, please, by looking at FUJ00152573 and by going to page 13, please. Thank you. You can see here a letter dated 18 November from Bond Pearce – who were the Post Office’s solicitors in the case against Mr Castleton – addressed to Fujitsu; can you see that?
Gareth Jenkins: I can.
Mr Beer: If we just read the first page in summary, it says to Fujitsu, in relation to Mr Castleton’s branch at Marine Drive:
“We act on behalf of the Post Office. [Between those dates] Mr Castleton was a subpostmaster at the Marine Drive Post Office. He was strictly responsible for the safe custody of cash and stock and was obliged to make good all losses caused through his own negligence, carelessness or error and losses of any kind caused by his assistants.
“Between [some dates] net losses of [£27,000-odd] occurred at the [branch].”
He was suspended and then dismissed:
“The Post Office has now issued a claim against him to try to recover these net losses. [He’s] issued a counterclaim for wrongful termination of his contract.”
Then 2, “Mr Castleton’s defence”:
“[His] case is that any shortfall is entirely the fault of problems with the Horizon computer and accounting system at Marine Drive … and that the [Post Office] wrongly terminated his … contract in respect of which he suffered losses not exceeding £250,000.”
Three documents are attached. In fact, the second one has some attachments itself:
“Bentley Jennison [they were the authors of the report in the second attachment] state that deficiencies have probably been brought forward despite the fact that they have been entered onto the suspense account entry. They suspect this is because the Horizon system, despite the suspense account entry, has failed to recognise the entry on the daily snapshots. They have drawn this conclusion through looking at the discrepancy of [£3,500-odd] on 26 February 2004. They then suggest this double accounting could have continued over a number of weeks and that as such, [his] defence ‘appears to hold potential merit based on the limited documentation’ they have so far reviewed …
“Mr Castleton believes that if he can obtain further documents, such as the daily snapshots, he will be able to undertake a manual reconciliation of the cash account in order to substantiate his belief that the losses are not real but attributable to computer error. We attach an email from Fujitsu to Richard Benton at the Post Office dated 5 May in which Fujitsu state: ‘It is possible that they are not accurately recording all transactions on the system’ and that there was no evidence whatsoever of any system problem.”
Over the page:
“Please could you review W Mr Castleton’s experts’ reports and prepare a formal report dealing with the following points …
“1. We need to explain to a judge who will know nothing about Horizon exactly how it works [et cetera].
“2. Precisely what steps Fujitsu took to examine the Horizon system at the Marine Drive [branch].
“3. Whether there have been any similar or serious problems with the Horizon system at [Marine Drive] since [his] suspension and dismissal.
“4. Whether you believe the suggestion put forward by Mr Castleton’s experts is likely to be correct …
“5. If there have been any human errors in recording the transactions, could an explanation be that:
“(a) There was nothing wrong with Horizon because it simply reflected the information entered onto it; but
“(b) If staff entered the wrong numbers into Horizon there may have been no real loss (even though Horizon would show a loss), because there could be human error in accounting accurately recording transactions.
“If so, would that be a likely explanation?
“6. Any other information you believe may be relevant.”
Then scroll down, please. Paragraph (4), “Duty to the Court”:
“As a result of the instruction you may be asked to give evidence before the Court. Whilst the [Post Office] will be liable to pay your fees, in preparing your report and giving evidence, your overriding duty will be to help the Court on the matters within your expertise.
“You agree to meet the requirements of the Civil Procedure Rules Part 35 [and a copy is enclosed] and that your report will:
“1. Be addressed to the Court and not to the Post Office …
“2. Confirm that you understand your duty to the Court and that you have complied and will continue to with that duty.
“3. Contain a statement setting out the substance of all material facts and instructions (whether written or oral) on the basis on which your report is written. This statement should summarise the facts and instructions given to you which are material to the opinions expressed in the report or upon which those opinions are based and if any of the facts are within your own knowledge, which they are.
“4. Contain a chronology of relevant events;
“5. Contain a statement of Truth in the following form [and gangs it’s set out].
“6. You should note that proceedings for contempt of Court may be brought against you if you make a false statement and report verified by a Statement of Truth without an honest belief it was true.
“7. It must contain a declaration that the report has been prepared in accordance with the Code of Guidance on Expert Evidence [which was also enclosed].
“8. Give … qualifications.
“9. Give details of any literature or any other material which you rely on in making the report.
“10. So who carried out any test or experiment [et cetera].
“11. Give the qualifications of the person who carried out any such tests [et cetera].
“12. Where there is a range of opinion on the matters dealt with [that that change should be summarised].
“13. Give reasons for your own opinion.
“14. Contain a summary of the conclusions reached including any qualifications …
“Given the fundamental importance of meeting these requirements, you should endeavour in your report to be not only accurate but complete. You should mention all matters which you regard as being material to the opinions you express and draw the Court’s attention to any matter to which you are aware which might adversely affect the validity of those opinions. This applies in relation to the factual matters to which you refer and also to the opinions which you express.
“You should not include in your report anything that is suggested to you by anyone without forming your own independent view.
“If, on reading the report of any other expert in this matter, or for any reason, you consider, at any stage, that any existing report of yours requires correction or qualification you will immediately notify us in writing.”
Then (5):
“In performing all your duties for which the client will pay, you will owe a duty to the client to act with the professional standards of skill, care and diligence adhered to by experienced and competent consultants acting as expert witnesses.
“You will take reasonable care of any documents, [et cetera]
“You confirmed that you:
“1. Are an independent party and not the client’s employee or agent, other than at the material time Fujitsu was responsible for looking after the Horizon system;
“2. Know of no reason why you should not act as a witness for the Post Office in relation to the dispute;
“3. Will advise us in writing immediately if [there is] any conflict between your interest and the Post Office’s interests …”
We are going to see in a moment that you commented on some the passages from this letter. Was this letter sent to you?
Gareth Jenkins: I’ve no recollection of that. I recognise the last bits now as being what an expert’s duties are, but the first time I was made aware of what those were when I was first put in touch with solicitors in 2020/2021, as part of the police investigation into my conduct.
Mr Beer: If the letter had – you say you’ve got no recollection whether or not it was sent to you?
Gareth Jenkins: I mean, I think I would have remembered if it had been sent to me because I can see there that it’s clearly set out what the duties are and I wasn’t aware of any of those duties, until my solicitor pointed them out to me when we first got engaged at the end of 2020.
Mr Beer: When you are giving your answers, Mr Jenkins, you should refrain, unless you’re doing it deliberately, from disclosing communications between your current solicitor and you, which involve either the seeking of legal advice by you or the provision of legal advice to you.
Gareth Jenkins: Okay, sorry.
Mr Beer: Because, if you do that, you might be said to have waived your privilege – your legal professional privilege in those – and I or somebody else can ask you about those.
Gareth Jenkins: Okay. I’m sorry.
Mr Beer: So I’m trying to help here.
Gareth Jenkins: Yes, I understand.
Mr Beer: If you wanted to give an answer of the kind that you’ve just given, it would be without waiving privilege by saying, “I only learnt that in 2020 or 2021”.
Gareth Jenkins: Okay, yes.
Mr Beer: Okay?
Gareth Jenkins: Yeah.
Mr Beer: It is, of course, open to you to waive privilege –
Gareth Jenkins: Yeah.
Mr Beer: – but I don’t want you to do so accidentally.
Gareth Jenkins: No, I don’t want to do so either.
Mr Beer: Okay. You’re reasoning that you didn’t see this letter because, if you had have seen the letter, you would have learnt earlier than you say is the case about the existence of an expert witness’s duties?
Gareth Jenkins: Correct.
Mr Beer: If you had received this letter, presumably you would have read all of it?
Gareth Jenkins: I would certainly have skimmed through all of it, yes.
Mr Beer: You wouldn’t have read half the letter: the first page but not the second page?
Gareth Jenkins: I don’t think so. I don’t recognise any of it.
Mr Beer: You agree that Bond Pearce have set out here, just reading it, the duties of an expert witness, what they entail, very clearly, haven’t they?
Gareth Jenkins: I agree, yes.
Mr Beer: In a very easy-to-understand way, haven’t they?
Gareth Jenkins: Yes.
Mr Beer: Can we turn, please, to FUJ00154713. If we look at the email at the foot of that page first, thank you, it’s an email dated 8 December 2005 about Marine Drive from you to Brian Pinder?
Gareth Jenkins: Yes.
Mr Beer: What was Brian Pinder’s function at this time?
Gareth Jenkins: He was the Manager of the Security Team, I believe.
Mr Beer: Just tell us what the Fujitsu Security Team did?
Gareth Jenkins: They were responsible for overall security, so making sure that the systems were secure, but they were also involved with communicating with Post Office in terms of providing what was called prosecution support or litigation support. So they were providing things like ARQ data and support of things like this.
Mr Beer: Can we just see how this email works because it’s slightly difficult to understand which bit is you speaking, which bit is Mr Pinder speaking and which bit is the questions from the Bond Pearce letter of 18 November 2005. Okay?
Gareth Jenkins: Yes, and I suspect the original would have been slightly easier to decode because it would have been –
Mr Beer: Coloured?
Gareth Jenkins: – multicoloured, yes.
Mr Beer: Yes, we don’t have the native version –
Gareth Jenkins: Yes, I understand that.
Mr Beer: “Brian”, I think that’s you writing, isn’t it?
Gareth Jenkins: I think it is.
Mr Beer: Then the words in dark black or darker black:
“I have been asked by the Fraud Investigation Team to answer several questions, to assist an enquiry regarding a [Post Office] employees (subpostmaster) alleging that the Horizon system may have lost his money. I can manage 5 of the 6 questions but wonder if you could explain or provide an answer or opinion to the following question, the answers are being directed back to Graham …”
Would you understand that to be Graham Ward?
Gareth Jenkins: I do now. I’m not sure I would have done at the time.
Mr Beer: “… to formalise our response.”
Is that part in black there something that Mr Pinder wrote?
Gareth Jenkins: I think so. Looking at that page, I think the bits that I would have written is the “Brian” at the top and “This is certainly true”, which is a bit further down, I can’t see certain but that’s my guess.
Mr Beer: Trying to decode it, then, that second paragraph then, after the word “Brian”, looks like it’s Mr Pinder having sent that to you and then you’ve cut it into your email back?
Gareth Jenkins: Correct.
Mr Beer: Then there’s a part in lighter text, almost grey:
“If there have been human errors in recording the transactions, could an explanation be that:
“(a) There was noting wrong with Horizon, because it simply reflected the information entered on to it; but …”
Just stopping there. Is that likely to be what Mr Pinder has included in his original email to you?
Gareth Jenkins: Yes, I believe so and I recognise that as being in the previous document which you showed me, yes.
Mr Beer: One of the six questions –
Gareth Jenkins: Yes.
Mr Beer: – in the Bond Pearce letter –
Gareth Jenkins: Yes.
Mr Beer: – of 18 November. Then you reply, is this right:
“This is certainly true (ie Horizon simply reflects the information entered into it).”
Gareth Jenkins: Correct. That would have been my comment.
Mr Beer: So, in relation to if there have been human errors in recording the transactions could an explanation be there is nothing wrong with Horizon because it simply reflected information entered into it, and you say, “This is certainly true (ie Horizon simply reflects the Horizon entered into it”?
Gareth Jenkins: Yeah, that is what I would have said at the time and I stand by that today as well.
Mr Beer: That is a possible explanation?
Gareth Jenkins: Yes.
Mr Beer: Were you able to say that without investigation of the facts of this case?
Gareth Jenkins: Yes, because Horizon is going to record what is actually entered into it, whether that reflects what happened in the real world, and if you do a typo when you put information into Horizon, then it’s going to accurately record that typo.
Mr Beer: Then part (b) of the original question:
“If staff entered the wrong numbers into Horizon there may have been no real loss (even though Horizon would show a loss), because there could be a human error in accurately recording transactions.”
Then scroll down. You say, I think, is this right:
“Again, this could be true. However if there is some sort of misentering of data into Horizon, then there would be another corresponding error which should be picked up, eg as a stock error or some AP Client being credited with an incorrect amount. Also, such error should show up as part of the balancing process.”
Again, you’re there saying this could be an explanation; is that right?
Gareth Jenkins: I am.
Mr Beer: Then question 3 was:
“If so, would that be a likely explanation?”
You give quite a long reply, and you say:
“It is a possible explanation, but without doing a detailed analysis of everything that has gone on in the branch it is difficult to speculate as to what has happened. Certainly the most likely explanation, is misoperation or fraud. However I appreciate that that is not sufficient for a prosecution. Without understanding what exactly is alleged (by [Post Office] and the subpostmaster), I don’t feel I can add anything further.”
So are you saying there, firstly, that you need to see some underlying data in order to say what is likely to have happened?
Gareth Jenkins: Yes.
Mr Beer: Because, otherwise, it’s mere speculation –
Gareth Jenkins: Correct.
Mr Beer: – ie the answers you’ve given to parts (a) and (b) are speculation?
Gareth Jenkins: Yes.
Mr Beer: You say that “It is certain that the most likely explanation is misoperation or fraud”. Why was it certainly the case that the most likely explanation was misoperation, ie by the subpostmaster, I think, or fraud?
Gareth Jenkins: I’m not sure why I would have said that at the time. I mean, looking back now, I think that’s just one of many different options but I accept those are the words I used at the time. But I think I would stand by the fact that I would need to look into exactly what had happened.
Mr Beer: Lots of people have said to me “I accept that they’re the words that I used at the time”, and obviously they are –
Gareth Jenkins: Yeah, sure.
Mr Beer: – because they are, they’re on the screen?
Gareth Jenkins: Yeah.
Mr Beer: Why did you use them?
Gareth Jenkins: I don’t know, is the simple answer.
Mr Beer: Did they disclose the way you were thinking in December 2005, ie the most likely cause of what is being described here is mistake by the subpostmaster or fraud by the subpostmaster?
Gareth Jenkins: My feeling was that that was more likely than an error in Horizon, yes. But I’m including the option of it being a mistake.
Mr Beer: But you can’t say that without looking at the data, can you?
Gareth Jenkins: Which is why I go on to say that I needed to look at the data.
Mr Beer: You say you appreciate that that is not sufficient for a prosecution. Had you understood this to be a criminal case?
Gareth Jenkins: Well, the fact that the request was coming from the Security Team suggested that there probably was.
Mr Beer: Did you, at this stage, understand the difference between criminal and civil proceedings?
Gareth Jenkins: No.
Mr Beer: Did you know that there were different types of courts, some which tend to deal with money and some which tend to deal with crime, sending people to prison and the like?
Gareth Jenkins: Not really, no.
Mr Beer: Did you understand that a prosecution tends to refer to criminal cases?
Gareth Jenkins: I’m not sure I really thought about it. As I say, I’d not really appreciated any real difference and so I was probably using fairly loose language. After all, this is an internal email.
Mr Beer: There’s then a part of the text which says, “I have spoken with Dave Baldwin”, et cetera.
Gareth Jenkins: Yes.
Mr Beer: Is that paragraph again Mr Pinder?
Gareth Jenkins: That I would expect it to be, yes.
Mr Beer: “I have spoken to Dave Baldwin on this and he recommended I speak with you. I’ll also be liaising with him on my final response as we are fundamentally of the opinion that Horizon does not ‘eat money’ merely accounts for its placement.”
Then you finish, I think this is you at the end saying:
“Happy to discuss further if you want to look at some more specific areas of Horizon.”
Gareth Jenkins: That sounds like me, yes.
Mr Beer: The answer to the third question is an important one, isn’t it, because it says the answer to the first two are mere speculation, without looking at the data?
Gareth Jenkins: Yes.
Mr Beer: Can we go to the top of the page, please. Later that day, Mr Pinder writes back to you, and says:
“Gareth
“[For your information]
“This is the response I intend to send to Graham re your question if that’s okay.”
Then “Comments from Gareth Jenkins”, question 1:
“If there had been human errors [then] (a) …”
You can see your text is faithfully reproduced, yes?
Gareth Jenkins: Except that the last point has not been faithfully reproduced.
Mr Beer: No, no, just look at (a) to start with.
Gareth Jenkins: Yes.
Mr Beer: That has been faithfully reproduced, hasn’t it?
Gareth Jenkins: Yes.
Mr Beer: Then look at (b), that’s been faithfully reproduced, hasn’t it –
Gareth Jenkins: Yes.
Mr Beer: – your answer, yes?
Gareth Jenkins: Yes.
Mr Beer: Then (c) – or I’m going to call it (c) – the third part of the question:
“If so, would that be a likely explanation?”
He has written:
“Not able to comment on this.”
Gareth Jenkins: Yes.
Mr Beer: The detailed and important answer you gave to the third part of the question has been changed to “No comment”, hasn’t it?
Gareth Jenkins: Yes.
Mr Beer: Do you know why that was?
Gareth Jenkins: No.
Mr Beer: It’s an important qualifier to the answers to the first two questions, isn’t it?
Gareth Jenkins: I realise that now. I probably didn’t appreciate that at the time.
Mr Beer: Why not?
Gareth Jenkins: I don’t know, is the simple answer.
Mr Beer: Can you think of a good reason why the careful explanation you’ve given to question 3 has been changed into “No comment”?
Gareth Jenkins: No, it may have been considered that it was just too complicated an answer but –
Mr Beer: What’s complicated about an answer “You need to look at the data to give an accurate picture”?
Gareth Jenkins: I don’t know, I think you’d have to ask Brian Pinder.
Mr Beer: I’m asking you at the moment because what –
Gareth Jenkins: Yeah.
Mr Beer: – this is is an email to you, essentially saying, “Is it okay if I send this as your response”?
Gareth Jenkins: I probably hadn’t picked up the fact that it had taken away the qualification that I had put in the original email.
Mr Beer: Do you read documents carefully?
Gareth Jenkins: I try to, yes.
Mr Beer: You’re saying you wouldn’t have – or you obviously didn’t pick this up at the time?
Gareth Jenkins: I clearly didn’t pick that up on the time. Why he was changing that to a “no comment” I wasn’t particularly clear. Perhaps I should have taken it up to him as to why he was doing that but I assumed that he knew why – what he wanted to respond, and unable – maybe he felt that I wasn’t in a position to make the comment on that last part. I just don’t remember.
Mr Beer: But you had, hadn’t you?
Gareth Jenkins: I had commented on it, yes, but why he’d decided he didn’t want to pass that on, I don’t know.
Mr Beer: I’m not so much asking for his take; I’m asking why you assented to this approach by not saying, “Well, hold on, you’ve missed out an important qualifier”?
Gareth Jenkins: Perhaps I didn’t realise at the time how important that qualifier was.
Mr Beer: Can we go back to FUJ00152573.
Sir Wyn Williams: Before we do, Mr Beer, I’m slightly puzzled – and it may just be me – about how this email the one that’s on the screen, begins. “For your information”, and then this:
“This is the response I intend to send to Graham re your question, if that’s OK.”
Now, to me, that reads as if Mr Pinder is going to send this response to Graham, if it’s okay with Mr Jenkins, because it’s the question that Mr Jenkins has asked him. Am I misreading this?
Mr Beer: No, that’s exactly right, sir, I think.
Sir Wyn Williams: Is that consistent with the interpretation that was given to the preceding emails, which suggested that it was Mr Pinder who was seeking to ask questions of Mr Jenkins, rather than Mr Jenkins asking questions of Mr Pinder; do you see my point? It may be that I’m just misunderstanding.
Mr Beer: Sir, I think you may have misunderstood the first email that we looked at, if we just go down to that.
Sir Wyn Williams: Yes.
Mr Beer: As I read this, there must have been an earlier email that we haven’t got in which Mr Pinder sent that first paragraph “I’ve been asked by the Fraud Investigation Team”, and then the second paragraph which says, “If there have been human errors”, if that’s marked, and then the paragraph at (a) if that can be marked, and the paragraph at (b), and then Mr Jenkins has replied in the darker text.
Sir Wyn Williams: Yes. Well, that’s as I understood what you were explaining, and that did or could make sense, but it’s the juxtaposition of that with the beginning of the next email that I’m struggling with. Anyway, I’m sorry to interrupt your questioning of Mr Jenkins. Maybe it’ll all become clear to me.
Mr Beer: I’ll try, sir.
If we go back up to the top page, do you understand that this is Mr Pinder saying to you everything below the words “Comments from Gareth Jenkins” and including the words “Comments from Gareth Jenkins” is what I, Brian Pinder, are going to send to Graham –
Sir Wyn Williams: Yes, I do, but it relates to, and it’s just those three words “re your question”, and it’s the other way around, isn’t it, on the first two emails?
Mr Beer: Sir, I think what that – your – I don’t want to give evidence here, but your –
Sir Wyn Williams: Maybe I should just ask Mr Jenkins.
What do you think –
Gareth Jenkins: I think the “re your question” – my understanding is that Graham Ward had asked Brian Pinder six questions. He’d identified one of those questions for me to respond to, namely question 6, I think it was, and, therefore, what he’s saying is, “In response to your – the question that I’ve allocated to you, this is the answer I intend to send to Graham Ward in response to that specific question”.
Sir Wyn Williams: So your interpretation of the phrase “your question” is not that you asked the question, it’s the question that he allocated to you?
Gareth Jenkins: Correct. That’s how I understand it –
Sir Wyn Williams: Right.
Gareth Jenkins: – and that’s how I would have understood it at the time.
Sir Wyn Williams: Fine. Well, then I follow. Thank you. Sorry –
Mr Beer: So that should read, “I intend to send to Graham the question that has been allocated to you, if that’s okay”?
Gareth Jenkins: That is how I’ve read it at the time, yes.
Sir Wyn Williams: Fine. Then I do understand now. Thank you.
Mr Beer: Thank you.
Can we go back, please, to FUJ0015 –
Gareth Jenkins: Before we leave that, can we go down, please, to the – my response to the third question?
Mr Beer: Yes, absolutely.
Gareth Jenkins: It starts – again, it said, “It is a possible explanation, but without detailed analysis it’s difficult to speculate”. So I think what Brian had interpreted from that, therefore, I wasn’t going to comment on it and that’s why I would have been happy with the “I’m not happy to comment”. I suspect that is probably what – why I was happy with that being replaced with “unable to comment” because I’m speculating there and, therefore, it is not a definitive answer.
Mr Beer: Doesn’t that overlook completely –
Gareth Jenkins: But I accept the fact that the qualification would have been useful information to have provided.
Mr Beer: Doesn’t it completely overlook the fact that the main point to question 3 is you need to look at the data?
Gareth Jenkins: Yes, yes, I accept that now. But that may well have been why I was happy to allow the change to have been made at the time.
Mr Beer: Okay. Can we go back, please, to FUJ00152573. This is, if we just look at the foot of the page, Brian Pinder’s document – if we go up, please – addressed to Graham, which is going to be Graham Ward. This is the response of Mr Pinder. He says:
“In response to your email request dated 28 November concerning the questions raised at paragraph (3) of the above document …”
The “above document” being the Bond Pearce letter that we’ve looked at.
You will see the answer to question 1. Question 2, it refers to, in the answer, a response from Anne Chambers, which is attached. Then, if we scroll down, please, to question 5, you’ll see that the full question in three parts is set out and the reply from Mr Pinder is “Gareth Jenkins’ response: attached”.
If we go to page 3, please. So this was the attachment to Mr Pinder’s document.
Gareth Jenkins: Yes.
Mr Beer: Can you see that it’s essentially his email back to you, the “FYI” email –
Gareth Jenkins: Yes.
Mr Beer: – including your signature block at the bottom, which he did, in fact, include in his reply to you?
Gareth Jenkins: Yes.
Mr Beer: So the shortened reply has been sent back to the Post Office?
You were able to comment, weren’t you, whereas your reply has gone back to the Post Office saying that you’re not able to comment?
Gareth Jenkins: Yes.
Mr Beer: Was the reason for that because it was thought that an investigation of what the subpostmaster was saying might not be in Fujitsu’s best interests?
Gareth Jenkins: I don’t know what was the reason for that change. I would have been quite happy to investigate and look at exactly what had happened in the branch and, in fact, I believe, at some later stage, I did assist Anne in doing some investigations there.
Mr Beer: Was the reason for the different reply that an investigation might reveal issues that Fujitsu didn’t really want to speak about?
Gareth Jenkins: Not that I’m aware of.
Mr Beer: Thank you. That can come down.
I think you subsequently attended a conference on 6 June 2006 with representatives of Bond Pearce and others; do you remember?
Gareth Jenkins: I remember seeing the documents about it and have vague recollections of the meeting.
Mr Beer: Let’s look at the documents, please, POL00071427. We can see this is an attendance note in the case of Lee Castleton, dated 6 June 2006. We were previous looking, to give you some context, at November and December –
Gareth Jenkins: Yes.
Mr Beer: – 2005. This is at Fujitsu’s HQ in Bracknell. In attendance were, if we scroll down, solicitors, the first three people, then Brian Pinder, Peter Sewell, Andy – I think that’s Dunks –
Gareth Jenkins: Yes.
Mr Beer: – I think he’s called “Dunce”, Anne Chambers, Naomi Ellis and you, a “Distinguished Engineer of Fujitsu”?
Gareth Jenkins: Yes.
Mr Beer: I’m not going to go into the detail of this attendance note. Do you recall any discussion in this meeting about you or Anne Chambers giving evidence?
Gareth Jenkins: I think there was discussion about evidence being given but I can’t remember the details of it and I do know that Anne eventually ended up giving evidence in the case and I didn’t.
Mr Beer: You remember the second page of the Bond Pearce letter and onto the third page of the Bond Pearce letter –
Gareth Jenkins: Yes.
Mr Beer: – an extensive explanation of the duties of an expert?
Gareth Jenkins: Yes.
Mr Beer: Do you know what had happened to that?
Gareth Jenkins: No idea.
Mr Beer: Were those issues discussed at this conference?
Gareth Jenkins: I have no recollection of any discussion of those there. As I say, all I can vaguely remember is the fact that there was a meeting and I met a couple of solicitors.
Mr Beer: Had the guidance that Bond Pearce had given about the provision of expert evidence by a Fujitsu employee been forgotten by this stage?
Gareth Jenkins: I don’t know whether it had been forgotten or what but I don’t remember any mention of it.
Mr Beer: In any event, your recollection is that that wasn’t gone over again in the context of either you or Anne Chambers giving evidence?
Gareth Jenkins: No.
Mr Beer: Thank you. Can we turn – this is 6 June 2006 – to three days later, to 9 June 2006. FUJ00154722, and look at the email at the bottom part of the page. If we scroll up just so we can see from and to, thank you. From the solicitor at Bond Pearce, Stephen Dilley, to Mr Pinder, copied to Graham Ward and Mandy Talbot. So not you at this stage.
Gareth Jenkins: Yeah.
Mr Beer: “… I would like to thank you and your team for finding time to meet us on Tuesday.”
That would be the date of the conference we just looked at:
“We found the meeting to be really worthwhile, helpful and productive. We will circulate a note of the meeting shortly, just for everybody’s record.
“In the meantime, I attach a letter received on 7 June from Mr Castleton’s solicitors. I have highlighted the two paragraphs that I would like you to have a quick look at. Basically they state the Horizon system is equip with a facility via which its operating software can be updated remotely via the connection to the [Post Office’s] head office and that the system sometimes went offline when there are software updates. They have therefore asked for documents relating to the updating of the software of the system. My view is that they are making a stab in the dark, but please could you ask your team whether they would have or could obtain any records of software updates?”
So this is a question about remote accessing of the system.
Gareth Jenkins: A sort of remote access. I think I have explained different types of remote access in one of my witness statements.
Mr Beer: We’re going to come to that, probably tomorrow, but a species of remote access?
Gareth Jenkins: Yes.
Mr Beer: Now, if we look, please, at FUJ00152605, and if we scroll down, please, and again, and again. We’ll see that email – if we just scroll down a bit further, we’ll see the email we’ve just looked at.
Gareth Jenkins: Yes.
Mr Beer: Then if we scroll up, please, Brian Pinder sends that on to you and Penny Thomas:
“I would like to run this by you first as I hope you might be able to either answer the query (hopefully) or point me in the right direction?”
Then up, please. Your reply of 9 June 2006, you say:
“It is correct that we do update the software remotely and that a software update could result in the desktop being closed and restarted. However, even then I don’t see how it would result in losses as is being claimed. Also software updates are relatively rare and normally only take place between 20.00 and 01.00 and again between 4.00 and 7.00 am with occasional extensions at the weekends. This puts them well outside the normal operational hours of Horizon.
“I think he is clutching at straws. However I guess we do need to cover this is angle.
“I would expect the system to record details of all software updates and when they took place at which bans. However I don’t know how long these are retained and whether they are included in the Audit Trail so you can check out the history for Marine Drive in 2004.”
Then you suggest a place to start.
In the provision of this information, which is a reply about a species of remote access, had the guidance in the Bond Pearce letter been drawn to your attention about the provision of expert evidence?
Gareth Jenkins: No. I just had the question as shown in the email trail that you are showing me.
Mr Beer: So that letter of 18 November never made its way through to you?
Gareth Jenkins: No, I’m sure that it couldn’t have done because I would have done things differently, not necessarily in this case but certainly in later cases, if I’d been aware of those responsibilities.
Mr Beer: Can we move on, please, to FUJ00154727, and start with page 2, please. An email from Mr Ward to Mr Pinder and Mr Sewell; can you see that?
Gareth Jenkins: Yeah.
Mr Beer: “Stephen Dilley (representing the Post Office) is asking if it is possible that someone undertakes an analysis of the figures recorded on transaction logs … It appears that the solicitors for Castleton are saying that they’ve compared the transaction logs with the cash accounts for week 42 themselves … and that they don’t match. They conclude that Horizon is therefore only recording half the [transactions].”
Then reading the second and third paragraph:
“Given this is a ‘test’ case and that the integrity of the Horizon system is being challenged, my own opinion is that this exercise should be completed by an ‘expert’ from Fujitsu. However, it will obviously be extremely time consuming as all the figures recorded will need to be methodically and carefully analysed. It is also important we complete the analysis and respond formally to the points raised as soon as possible.
“My first question is:
“1. Is this is an exercise that Fujitsu could undertake, possibly by Gareth perhaps, who would presumably have a thorough understanding of the figures recorded on both the transaction logs and the figures on an office cash account?
“2. If the answer is yes, how soon could this exercise be performed and a formal response prepared? Will there be a cost …”
Then if we scroll up, please. We’ll see there that Mr Pinder forwards the email to you on 31 July, asking:
“Is this something you can provide?”
Gareth Jenkins: Yes.
Mr Beer: Then if we scroll up to the top of the page, your reply of 31 July:
“I guess it is the sort of analysis that I could do if required. However it is fairly time consuming and the problem is in adjusting my priorities to find the time to do it. I don’t know how many transactions are involved and I would need copies of the cash account before and after the period to carry out the analysis. I might need to get hold of the reference data that was current at that time (which is probably quite difficult).
“Given the volume of the data, I would expect this to take anything between 2 days and a week (it is difficult to tell in advance given that I’ll probably need to feel my way), and I’ll probably need help from Penny in carrying out extracts from the relevant message stores …
“As for timescales, I can’t spend more than a couple of hours on it this week and probably can’t start in earnest until the week of 14 August [then] I’m on leave …
“Anne Chambers might be able to do a similar analysis (and she may well have some tools to help …). However, she’s on leave this week.”
So you’re essentially expressing concerns about your capacity.
Gareth Jenkins: Yes.
Mr Beer: By this time, had any of the guidance that we saw in the letter of 18 November 2005 been given to you about providing expert evidence?
Gareth Jenkins: I’m pretty sure it hadn’t. I can’t say definitively because it’s so long ago but, as I say, I think I would have known if I had seen it.
Mr Beer: Can we look, please, at FUJ00152290. If we scroll down, please, you say there on 7 August to Mr Sewell, in relation to the Castleton case and transaction logs, second paragraph:
“I spoke with Anne this morning and we’ve agreed that once the CP …”
Is that a change –
Gareth Jenkins: Change proposal.
Mr Beer: “… is approved, that she will carry out the initial analysis (based on transactions we get from Penny) then we will review them together and I’ll front up any report we present to [the Post Office].”
Again, by this stage, we’re now in August 2006, had the guidance given in the 18 November 2005 letter been provided to you?
Gareth Jenkins: Not that I’m aware of and I don’t think that’s the letter that’s referred to in an earlier email that’s mentioned in this email.
Mr Beer: No, that’s right.
Gareth Jenkins: I think that’s a letter to somebody different.
Mr Beer: Can we look, please, at FUJ00152292. 17 August 2006 now, about Marine Drive, from Anne Chambers to you. She says:
“I’ve spent some time recalculating the CAP42 …”
The Cash Account Period 42:
“… cash account for Marine Drive and addressing the points in the letter from the solicitor. Hope it makes sense to provide you with a starting point not quite at the very beginning.
“Subsequent cash accounts, if needed, should be much easier now I have the mappings set up and some idea of what numbers need to go where.
“I’m now going to look at the loss made in week 42 and demonstrate that it was due to the difference between system holdings and declared holdings.
“If this isn’t at all what you wanted, please let me know – I don’t really know what I’m doing!
“This hasn’t had my full attention, lots of people are on leave … Also, yesterday I got my witness statement which is (as I expect you found) full of things I didn’t say or do, including all those PowerHelp calls.”
So Ms Chambers had been asked by you to do the work or it had been agreed by you that she should do the work; is that right?
Gareth Jenkins: Yes, as I understand it, she did the initial calculations. I think what she sent there is a spreadsheet that she put together and a note of what she’d actually done and I then went through and checked both her methodology and the conclusions she came to, and then I think, at some later stage, I wrote it up in a slightly different form than the form that she’d passed it to me in.
Mr Beer: Why were you going to, in your words, front it up, front the report?
Gareth Jenkins: I can’t remember now. That’s clearly something we’d agreed between us at the time but I can’t remember what the – what was behind that at the time.
Mr Beer: Can we turn to FUJ00122279, please, and scroll down please. Thank you. An email of 2 August from Mr Dilley, the solicitor, to Brian Pinder and others, not including you, with the subject “First draft Statement of Gareth Jenkins”. He says to Mr Pinder:
“I refer to our previous correspondence and attach the first [draft] of a … witness statement and Exhibit for Gareth Jenkins. [It’s] designed to:
“Explain what Horizon is; and
“Comment on Mr Castleton’s allegations about the Horizon system.
“I am preparing a second statement for Anne (to follow shortly) which will deal with the call logs.
“[Can you] look at the first draft, make any changes [you feel] are appropriate and also answer my questions in bold italics. [He’ll then revise the statement with counsel.”
If that’s scrolled up, please. It’s forwarded to you, for your information and action.
Gareth Jenkins: Yeah.
Mr Beer: Can we look, please, at FUJ00122283. You reply the next day at 3.30-ish and say:
“I’ve had a go through this. It isn’t a complete statement and I won’t have time to sort out the outstanding bits for a while. I think we probably need to get the detailed analysis done before this can be completed.
“I’m happy for you to send it as it is to Stephen if you like but you may decide it isn’t complete enough for that. Also I may be able to research some of the areas (such as auditing) that I’m claiming ignorance of if that is required.”
So you’re essentially saying that “Here’s an incomplete statement, I haven’t got a chance to complete it yet” –
Gareth Jenkins: Yes.
Mr Beer: – “there’s still some analysis work to be undertaken”?
Gareth Jenkins: Correct.
Mr Beer: Again, by this stage, had the letter of 18 November and the expert duties points in it been drawn to your attention?
Gareth Jenkins: No.
Mr Beer: Thank you. That can come down. I think you were subsequently informed by Mr Pinder that you wouldn’t be required to give evidence in Mr Castleton’s case; is that right?
Gareth Jenkins: That is what I understand from the documents I was emailed. I certainly remember I didn’t give evidence.
Mr Beer: Do you remember Mr Pinder informing you directly that you would not be required to give evidence in Mr Castleton’s case?
Gareth Jenkins: I’ve seen an email which effectively says that, yes.
Mr Beer: Before the break can we just look at that, please. FUJ00154733. I think this is, if we scroll down, please, and again and again, and just go back to the top, please. I think, again, as is or as was your practice, you have cut into an email the contents of Mr Pinder’s email, and replied to it?
Gareth Jenkins: Yes.
Mr Beer: So the first part of the text in the two lightly shaded paragraphs is what Mr Pinder is saying; is that right?
Gareth Jenkins: That’s my understanding of it and it makes sense reading it that way.
Mr Beer: He says:
“Just been chasing Stephen up re your attendance and any matters still outstanding for us [Post Office Account], as follows; (my words)
“He states that although you [that’s you, Mr Jenkins] would probably make a good witness, it is for evidential reasons that you cannot be called. To do with evidence of ‘opinion’, ‘expert’ evidence and ‘real evidence’, et cetera, et cetera (complicated legal issues nothing to do with personalities).”
You say:
“Fine (I won’t try and understand what this means!)”
Gareth Jenkins: Yes.
Mr Beer: Was that the limit of the explanation that you received as to why you were not being called as a witness in the Lee Castleton case?
Gareth Jenkins: As far as I can remember, yes, I’m not aware of any other briefing I got given.
Mr Beer: Did you have an understanding of what the distinctions drawn in this email between opinion and expert evidence and real evidence were?
Gareth Jenkins: No, hence my comment about I won’t try and understand what this means.
Mr Beer: Was that the end of it, so far as you were concerned?
Gareth Jenkins: Yeah. As far as I was concerned, it just meant that I didn’t need to be involved in the case any more and therefore I could get on with my day job.
Mr Beer: You, in fact, did have a continuing involvement in the case because you answered some queries?
Gareth Jenkins: Yes.
Mr Beer: But you didn’t give evidence?
Gareth Jenkins: Correct.
Mr Beer: The references to those, I’m not going to turn them up, are FUJ00154747 and POL00069822.
In what capacity were you continuing to be involved in the case?
Gareth Jenkins: I believe I was asked to do some further analysis. As I say, I can’t remember now what the details were but I’ve seen emails suggesting that I was asked some specific questions which I responded to, and I think I may have produced another paper as a result of some of that investigation, but I can’t remember now what I did at the time. All I can go by is the emails that the Inquiry has shown me.
Mr Beer: That document can come down. Thank you.
You tell us in your witness statements that you didn’t see Anne Chambers’ “Afterthoughts” document, either at the time or until this Inquiry, essentially.
Gareth Jenkins: I’ve no memory of seeing that.
Mr Beer: Given that the plan was that you should present the evidence that constituted the work she had undertaken, you were going to front up the report –
Gareth Jenkins: Yeah.
Mr Beer: – and then she had ended up giving evidence, rather than you at trial, did you not discuss with her how that came about?
Gareth Jenkins: I think it was really as part of this thing, that it was a case of it had been decided that, because Anne had done the original analysis and I think she had done the analysis – I think there may have been a call with her back in 2004 because I think her analysis was partly based on work that she’d actually done at the time and, therefore, that was why she needed to give the evidence, because she’d actually done the original analysis and, therefore, me giving evidence, just reporting what she’d told me, didn’t really work.
So it was something like that, but I just don’t remember the details, and that was probably what I took as being behind what was shown in the last email you showed me.
Mr Beer: Was that explained to you at the time, ie that Anne Chambers is the one that’s been selected to give evidence because she could give some evidence of primary facts, ie work that she had undertaken in answering a Helpdesk query in 2004: that’s why she’s the witness of choice?
Gareth Jenkins: I think I may have heard that at some point but I just can’t remember the details, sorry.
Mr Beer: Did you have any discussion with her about the wider issues raised by her document?
Gareth Jenkins: What do you mean by –
Mr Beer: She was critical of the process by which she had become involved in the case –
Gareth Jenkins: Are you talking about the afterthoughts document?
Mr Beer: Yes.
Gareth Jenkins: No.
Mr Beer: Were you aware that she was an extremely reluctant witness?
Gareth Jenkins: Yes, I think I was aware she was reluctant to be a witness.
Mr Beer: Why was she reluctant to be a witness?
Gareth Jenkins: I think it just took her outside her comfort zone. I think that would apply to any person, really, going before the legal process and she certainly didn’t feel very comfortable with it.
Mr Beer: Did she raise with you her complaints about the manner in which she had been pulled into the case?
Gareth Jenkins: No, I don’t believe so.
Mr Beer: Did she raise with you her complaints about the process by which disclosure had been given?
Gareth Jenkins: No, I think she may have mentioned something about having something sprung on her about NT event logs being different from logs, but that’s the only vague recollection I have, because I think she – didn’t she have to go back about a month after the original trial to present some further evidence or something?
Mr Beer: That’s right. Is that something you spoke about at the time with her?
Gareth Jenkins: I was aware that she had to go back and that she wasn’t happy with it, but not the gory details of it.
Mr Beer: Thank you.
Sir, that’s just coming up to 12.20 now, can we break now until 12.35?
Sir Wyn Williams: Yes, of course.
Mr Beer: Thank you, sir.
(12.19 pm)
(A short break)
(12.35 pm)
Mr Beer: Good afternoon, sir, can you see and hear us?
Sir Wyn Williams: Yes, thank you.
Mr Beer: Thank you.
Mr Jenkins, in your – I’m going to turn to the capacity in which you gave evidence in witness statements and then orally in the Seema Misra case –
Gareth Jenkins: Yeah.
Mr Beer: – just as a general topic, as a theme.
Gareth Jenkins: Okay.
Mr Beer: In your witness statements provided to the Inquiry, on a number of occasions you explain you were not the Chief Architect for Horizon, correct –
Gareth Jenkins: Correct.
Mr Beer: – nor, indeed, a lead engineer for the Post Office Account?
Gareth Jenkins: No.
Mr Beer: You explain that you were not the only person involved in the development of Horizon nor the maintenance of it when it was up and running?
Gareth Jenkins: Correct.
Mr Beer: Why did the Post Office rely on you specifically as its expert witness?
Gareth Jenkins: I don’t know, is the simple answer. I did have a fairly good overview knowledge of the whole of Horizon because of really going back to my role in the agent team, because the whole point of the agents was to sit in the middle and see the difference between what was happening at the counter and what was happening at the back end and, obviously, with the work I did on IMPACT, I got a much more detailed knowledge of how the counter operated. So, from that point of view, I probably was one of the people who had a good overview knowledge of how the Horizon system worked. But I don’t think I was necessarily the only person.
Mr Beer: Was there anyone who was a better placed witness than you?
Gareth Jenkins: Not necessarily. I think there was some email exchange in early 2010, when there was – other names did get suggested to as to being able to get involved but, in the end, it was me that got picked to actually get involved.
Mr Beer: Did you ever discuss the role that you were performing as a witness with anyone senior at Fujitsu?
Gareth Jenkins: They were clearly aware that I was being asked to do these things and they accepted that it was something that needed to be done and that I could probably do it as well as others but I don’t know that we had a formal “Yes, this is something that needs to be part of your job”, but it sort of evolved, and then it did become, effectively, part of my job, moving forward.
Mr Beer: Did you ever discuss with anyone senior at Fujitsu the demands being placed upon you by the Post Office?
Gareth Jenkins: Yes, because I had to see how this scheduled in with the work that I was doing, and it was felt that at the sort of level that we thought was required of the work, then it could be fitted in. But, clearly, now understanding what maybe I should have been doing in terms of the role of an expert, then that would have been maybe a different situation. It’s difficult to speculate.
Mr Beer: Was the focus on your capacity and the money that the Post Office would be charged for the work that you did?
Gareth Jenkins: I don’t think the money came into it because the Post Office were paying for me regardless of what I did. So it was really a case of they could pay for me to do designs of new bits of the system or they could pay for me to support prosecutions, and it was really up to them. They were paying the same either way, so it didn’t really make any difference and certainly made no difference to my salary.
Mr Beer: So was the focus, then, on the issues of your capacity –
Gareth Jenkins: Yes.
Mr Beer: – rather than the nature of the function that you were being asked to perform?
Gareth Jenkins: Yes, I think that’s probably fair.
Mr Beer: During your involvement in the various criminal cases, did you feel under any pressure placed upon you by the Post Office or its lawyers publicly to refute any suggestion by the defence that there were issues with the reliability of Horizon?
Gareth Jenkins: There were certainly cases where they were trying to put words into my mouth which I didn’t want to say and I think there are examples of emails where I’ve said, “Well, I agree with what this defence statement is saying and nothing really to add to it”.
Mr Beer: What do you think at the time, at that pressure being placed upon you by the Post Office?
Gareth Jenkins: I just took it as being the way these things happened. There was clearly, in some cases, it was – things seemed to be all happening last minute and I just assumed that’s how things happened in that sort of environment.
Mr Beer: Never mind the timing so much. I’m thinking more of people were trying to put words in your mouth?
Gareth Jenkins: I wouldn’t allow them to put words in my mouth unless I agreed with them.
Mr Beer: Did you raise that with anyone senior at Fujitsu?
Gareth Jenkins: I think I did – well, I certainly had conversations with David Jones back in February 2010.
Mr Beer: He was a Post Office lawyer?
Gareth Jenkins: No, he was a Fujitsu lawyer.
Mr Beer: Sorry, I meant a Fujitsu lawyer.
Gareth Jenkins: Yes, he was a Fujitsu lawyer. I think that was really instigated by Penny, rather than myself. She felt that Post Office were getting to be very demanding of Fujitsu in general, and me – and her to some extent – in particular, because not only was there the case for Mrs Misra, there was also a case to do with Mr Hosi going on at the same sort of time, and she felt that it was important that we had some sort of legal cover, for want of a better word, of the way that it was being approached.
And so she set up this meeting with David Jones which lasted about an hour or so, something like that, where basically what – my memory coming out of that is, “Well, just tell the truth and what happened, and get on with it and we’ll support you as you need it”.
Mr Beer: Was there any discussion there as to the capacity in which you were giving evidence?
Gareth Jenkins: Not that I can recall.
Mr Beer: Was there any discussion then as to the duties that you may owe the court, given the nature of the evidence that you were providing?
Gareth Jenkins: No, I don’t believe so.
Mr Beer: I think it’s right, isn’t it, that the Post Office and its lawyers made a series of direct approaches to you on a number of occasions?
Gareth Jenkins: Yes, they did, and I tried to forced them back into the way that they should have been communicating with me, they should have been communicating with me through Penny and the Prosecution Support Team, and there were a number of occasions where I was approached directly and, in most of those cases, I then forwarded the emails on to Penny and tried to get some sort of control back on the process.
Mr Beer: Can we just look at some examples of those, please, Mr Jenkins. POL00097138. If we scroll down, please. There’s an email underneath this one, no need to look at it for now. You reply on 16 November to Rachael Panter; do you remember who she was?
Gareth Jenkins: She was a lawyer working for Cartwright King, I believe.
Mr Beer: You say:
“Rachael,
“Can’t be you use the report I have already sent you? There is no mention of the case on the report.
“You should really be addressing such requests through Post Office rather than directly to myself.
“As far as I know there is no commercial cover in place for me to spend any time on such activities (and that includes the case of Nemesh Patel).”
Then, if we scroll up, please, James Davidson, a Delivery Executive, says to Ms Panter, and you’re copied in:
“I am concerned at the engagement approach being taken here, we are fully on board to support but all approaches must come through Post Office by the correct change process.”
Is this is an example of you pulling the Post Office up on direct approaches to you?
Gareth Jenkins: Indeed it was, and if you notice on the previous email, although I’ve been sent the email directly by Rachael Panter, when I responded, I copied in both Penny Thomas and James Davidson, so they were aware of the fact that I’d been approached.
Mr Beer: What was the issue in your mind with a direct approach from the Post Office?
Gareth Jenkins: I think – well, there’s two or three issues, really. The main one was that I felt that the Prosecution Support Team, namely Penny, needed to be aware of what was going on and that we actually had a single point of contact so that Fujitsu was aware of what Post Office was requiring them to do in terms of legal processes, and so on.
Mr Beer: Why?
Gareth Jenkins: Just so that the right people knew what was going on. And then there was separately the question of making sure that Post Office was being charged correctly for the use of time, and so on. Again, the charging didn’t really come down to anything to do with what I was being paid or anything like that, but I needed to be able to allocate my time so that it was being accounted for correctly.
Mr Beer: So just going back to the first reason you gave there, or the first issue, why was it important that the Fujitsu Prosecution Support Team were sighted on these exchanges?
Gareth Jenkins: So that they knew what – basically, so they knew what was happening. It was their responsibility. There was supposed to be a line of communication between the Post Office Legal Department – or not the Legal Department, their investigators – Jane Owens is the name that I think of in that role – and the Fujitsu team, to actually know what was going on to coordinate the work so as to be aware of what was happening and what was going on.
Mr Beer: What was the importance of that, though? Was it written down somewhere that that’s what had to occur?
Gareth Jenkins: I think it might have been but I wasn’t aware of that but it just felt that that was the right process for doing things. I mean, my work had to be accounted for. I had management telling me what I needed to do. Having these odd requests out of the blue that were going to take up time, I needed to make sure that people were aware that these were happening and that my priorities could be allocated correctly between the various demands of my time.
Mr Beer: Was it in your mind that a prosecutor has duties to retain, to record and to disclose material?
Gareth Jenkins: I wasn’t aware of that, no.
Mr Beer: Was it in your mind that Fujitsu, as a third party to a prosecution, might be required to disclose material to a court?
Gareth Jenkins: No, I’d not thought of it that way at all. No.
Mr Beer: Did you ever come to understand that you might owe personal duties of disclosure to a court?
Gareth Jenkins: Only when it was explained to me in 2020 what an expert needed to do.
Mr Beer: Did you ever come to understand that Fujitsu may owe duties of disclosure?
Gareth Jenkins: No, as far as I was – understood it, the disclosure was something that Post Office needed to worry about, not – it wasn’t something that affected me or Fujitsu.
Mr Beer: What was your understanding of the Post Office’s duties of disclosure?
Gareth Jenkins: I didn’t really understand what they were but I was aware that there were – that I was told that there had been various hearings requesting various bits of disclosure, which they seemed quite happy had gone to say that they did not need to actually disclose the things that they had been asked to disclose.
Mr Beer: Was that in the Seema Misra case?
Gareth Jenkins: Yes. I wasn’t concerned about the details; I was just aware that such discussions were going on in the abstract.
Mr Beer: In any of those discussions, were the extent of Fujitsu or your own duties of disclosure discussed?
Gareth Jenkins: No.
Mr Beer: Can we move forwards, please, to FUJ00153986 and look at page 8, please, and if we scroll down, please. That’s excellent. An email from Jane Owen on 16 November 2012 – which is the same day we were looking at – at the moment to Penny Thomas, copied to Jayne Bradbury. That’s not a name that’s familiar to me, do you –
Gareth Jenkins: I don’t recognise the name at all, I’m afraid, sorry.
Mr Beer: Anyway, Jane Owen, Security Manager in the Post Office, says:
“Hi Penny …
“I am a little out of the loop with this now that I have changed roles. I asked Mark about raising an invoice (he has the budget) to cover the expert witness requirements for one of Sharron’s cases … which I think the email below will relate to … back in April an order was raised … but we were never invoiced … Do you know anything about that at all? Gareth has been approached directly by an external solicitor and Jayne will need to start discussions with our team leaders on how we prevent this from happening going forward as it is clearly unacceptable that he be ‘ambushed’ in this way.”
Did you feel like you were being “ambushed” by the Post Office, that’s the word?
Gareth Jenkins: I wouldn’t have necessarily used that word but, clearly, having this email out of the blue from Rachael Panter in November when the last contact I had had about any of these cases was about a month before that was a bit of a surprise and, clearly, I felt that we needed to have some sort of control as to how my work was to be scheduled and how Post Office was managing this and I see this as being Post Office trying to actually get that – some control put back in there.
Mr Beer: But this is all about capacity and formalisation between two business partners. It’s not to do with the substance of the evidence that you’re to give or any procedural duties that attach to it?
Gareth Jenkins: No, or sorry, yes. That’s – yes. Yes.
Mr Beer: That’s correct what –
Gareth Jenkins: That’s correct what you said, yes.
Mr Beer: Can we move forwards please to POL00229801.
Sorry, FUJ00229801. I think I said POL. Thank you.
If we can turn to page 3, please, and scroll down, please. Thank you. This is an email exchange between you and Angela van den Bogerd, can you see that, of the end of January 2013?
Gareth Jenkins: Yes, I see that. I don’t understand the context yet but, presumably, we’ll get there.
Mr Beer: To give you a bit of context, I’m not sure it’s completely necessary but if we just scroll down, please, and scroll down still further, and again. Thank you. This is about the Helen Rose Report. If you scroll up, please, now.
Gareth Jenkins: When you say Helen Rose Report I’m aware of two Helen Rose Reports. Are we talking about Lepton here?
Mr Beer: Yes, I think so.
Gareth Jenkins: Okay.
Mr Beer: If we scroll up to there. Thank you. I think Ms van den Bogerd is asking you questions about it.
Gareth Jenkins: Yes. This looks again like an email from Angela van den Bogerd to myself, where I’ve actually inserted with the GIJ prefix my responses to her various questions.
Mr Beer: Yes, and if we just scroll up, please. It’s the second paragraph of your reply. In the first paragraph you say, “I’ve added replies with Gareth Idris Jenkins, GIJ”.
Gareth Jenkins: Yes.
Mr Beer: In the second paragraph, you say:
“… this is all outside of our official remit. I probably shouldn’t be doing this investigation for you in this way (though if you ask for formal help, it is likely to be me who gets involved!).”
We saw the exchange of the previous November, where there was an attempt at formality made. Had that failed?
Gareth Jenkins: I’m not sure. I think what I – it’s a case of trying to understand the amount of work involved. I’ve been asked some specific questions. I probably spent about half an hour or an hour replying to this email so I thought it was simpler to just reply to the email and then point out that this should really go through official channels, and that is why I’d copied Penny in on the response because I saw hers being the official channel.
So I wasn’t trying to be awkward; I thought let’s just answer the question because that’s fairly straightforward. If it was going to take me a long amount of time, then I would have just said, “Let’s get this sorted out officially”.
Mr Beer: Why was it outside your official remit?
Gareth Jenkins: Because my job was designing new changes to the system, not answering questions from random people in Post Office.
Mr Beer: Why did you answer random questions from people in Post Office?
Gareth Jenkins: Because I felt I could actually answer them and I felt that, if it did go through the official channels, it would probably be me that would end up being asked about them. I think at this stage, I may have already had some sort of contact with Helen Rose about this but I can’t remember exactly the chronology.
Mr Beer: By this time, you’d already raised a red flag, so to speak, in November –
Gareth Jenkins: Yes.
Mr Beer: – about direct contact but you were perpetuating it here?
Gareth Jenkins: Yes.
Mr Beer: Do you adopt an informal “trying to help” approach?
Gareth Jenkins: I try to be helpful when I can, yes, but I’m aware that I also need to try to put some sort of control in, which is why I copied Penny on this to make her aware that I was being asked these sort of questions.
Mr Beer: Because, of course, you don’t know what’s going to be done with what you said in your earlier email, do you, ie the use to which it’s going to be put?
Gareth Jenkins: No.
Mr Beer: Can we move on. That can come down, please. You tell us in your third witness statement a series of things – I’m going to try and summarise them – about your approach when giving evidence. Firstly, would this be right, that you gave evidence without an understanding that you might be an expert witness in the legal sense –
Gareth Jenkins: Correct.
Mr Beer: – and that you were, therefore, lacking in understanding that you were subject to the duties imposed upon an expert witness, including duties of disclosure?
Gareth Jenkins: Correct.
Mr Beer: You tell us in your witness statements that you lacked that understanding in Ms Misra’s case, that’s paragraphs 233 and 329 of your third witness statement. Mr Allen’s case, paragraph 619; and in Ms Sefton and Ms Neald’s case, paragraph 689.
Gareth Jenkins: Correct.
Mr Beer: The second thing – and, again, I’m trying to summarise a large body of text here – is this right: you say you were never provided with written instructions to be an expert witness in any case in which the Post Office asked you to give evidence?
Gareth Jenkins: Correct.
Mr Beer: That includes, Mr Thomas’ case, paragraph 313; Ms Misra’s case, paragraph 329; Mr Allen’s case, paragraph 619; Ms Sefton and Ms Neald’s case, paragraphs 689; and Mr Ishaq’s case, paragraphs 643 and 646?
Gareth Jenkins: Correct.
Mr Beer: Thirdly, is this right: you say you were not informed on how to interact with a defence expert or to prepare a joint statement with another expert?
Gareth Jenkins: Correct.
Mr Beer: That applied in both Ms Misra’s case, paragraph 478, and Mr Ishaq’s case, paragraph 666?
Gareth Jenkins: Correct.
Mr Beer: Then, fourthly: you say that you were never asked by the Post Office to provide a witness statement that contained an expert witness’s declaration?
Gareth Jenkins: Correct.
Mr Beer: That applied in Ms Misra’s case, paragraphs 233, 466 and 483; Mr Allen’s case, paragraph 619; Ms Sefton and Ms Neald’s case, paragraph 689; and Mr Ishaq’s case, paragraph 669?
Gareth Jenkins: Correct.
Mr Beer: You tell us in your witness statement, it’s your third witness statement, about circumstances in which you asked for guidance in the run-up to Seema Misra’s trial as to what was expected of you as a witness.
Gareth Jenkins: Yes.
Mr Beer: By that time, had you suffered from a lack of confidence in the management of the requests coming from the Post Office?
Gareth Jenkins: Well, they were certainly fairly chaotic, a number of emails I was getting, and I was certainly aware that I was getting asked the same question multiple times, having already answered questions, and there was even some occasions when I responded saying, “Well, here’s the email I sent to you yesterday answering the same question”. So, yes, I could see that it – that there was a lack of organisation, is maybe a polite way of putting it, in terms of the way things were being managed.
Mr Beer: The email exchange that you had – I’m not going to go to it now, we’ll go to it later – when you asked for guidance “What’s expected of me as an expert witness”, and you got a rather asinine reply, didn’t you?
Gareth Jenkins: I can’t remember that particular email but that doesn’t surprise me at all.
Mr Beer: That one where “You’ve just got to tell the truth”.
Gareth Jenkins: Yeah.
Mr Beer: That one.
Gareth Jenkins: That’s going back to – that was Mr Thomas’ case I believe, yes, yes.
Mr Beer: Yes.
Gareth Jenkins: Yeah.
Mr Beer: Were you concerned about that, the basic nature of the reply you got, the lack of guidance you received?
Gareth Jenkins: I just took it at face value, I think, rather than questioning it. If I just had to tell truth then, well, I’d do that anyway so that wasn’t really telling me anything new.
Mr Beer: Why do we not see you asking any of your own managers at Fujitsu for assistance in this regard, “I’m being dragged into court proceedings here, please help me as to what is required of me”?
Gareth Jenkins: I thought I was being told what was required of me by Post Office, so I didn’t really understand that I wasn’t being told the right things. So I just trusted what I was being asked by Post Office and didn’t see the need to involve my management. They were clearly aware I was spending my time doing these things but I didn’t feel I needed any further guidance from them, in that I thought what I am looking for is guidance about legal things, therefore I’ll probably get it better from the lawyers that I’m talking to in Post Office than managers in Fujitsu who know no more about the law than I do.
Mr Beer: But you got no guidance from the Post Office?
Gareth Jenkins: Well, I realise that now. I thought I was getting some sort of guidance but it probably wasn’t really very good guidance.
Mr Beer: What sort of guidance did you think you were getting from the Post Office at the time?
Gareth Jenkins: Tell the truth to complete a witness statement.
Mr Beer: Why do we not see you asking the lawyers at Fujitsu, with whom you did correspond relatively frequently for greater support and guidance on the role that you were to perform?
Gareth Jenkins: I think that came after about 2010, sort of in Mrs Misra’s case, and I did seek guidance from David Jones and I had that meeting with him in February and, for a while, then he actually acted as a buffer between me and Post Office lawyers, and so on. Then later on, I was involved with Jean-Pierre (sic), who did give me a bit of further guidance but not the true guidance that maybe I should have had in terms of what an expert’s duties were.
Mr Beer: Can we look, please, at FUJ00152866. If we scroll down, please – thank you – an email exchange between you and David Jennings. Who was Dave Jennings?
Gareth Jenkins: I’m not quite sure. Reading the context of that email, I think he was someone within Customer Service but I can’t remember exactly what his role was. I don’t recall the name at all now, I’m afraid. This looks to be to do with booking time and concerns about what time I was charging to Post Office for activities in support of Penny.
Mr Beer: So, again, this is one of those emails where you set out his questions and then answer them?
Gareth Jenkins: Yes.
Mr Beer: You say:
“Dave,
“In response to your questions …”
Then his question is:
“I understand you act from time to time to support our prosecution … activity – sort of an ‘expert witness’ I suppose.
“I have a few initial questions on this …”
Then, stopping there, he describes you as “sort of an expert witness”. Did you regard yourself as a sort of expert witness?
Gareth Jenkins: Well, I was clearly an expert in how Horizon operated and that’s what I assume that meant. I didn’t understand what it’s meaning was in the legal sense, and didn’t until about 2020.
Mr Beer: He says:
“How frequent is this typically and how much notice is there …”
You say:
“It’s ad hoc … I tend to have various questions from Penny a couple of times a month.
“How much effort …
“It varies …
“Where do you book the time …”
You say:
“I have a code for prosecution support”, and you give the codes.
“Does anyone else in Requirements … do this activity?”
You explain and say you’re not sure how they book their time:
“Is there anyone else who should cover this activity as well/instead of you?”
You answer:
“‘Should’, then probably Yes. ‘Could’, then probably No!”
What did you mean by that?
Gareth Jenkins: Well, I think what I was suggesting is that maybe someone should be formally doing – performing this sort of role but I was also accepting the fact that I was probably in the best position to actually do that but I certainly didn’t want it as full-time job because I was enjoying myself doing the design work that was my day job, if you like, so I was seeing this as something that I was fitting in on the side between that. So that’s really what I was trying to summarise there.
Mr Beer: This seems to be focused on the commercial aspects of the arrangement, whether there’s appropriate commercial coverage in place –
Gareth Jenkins: Yes, I think that must have been Dave Jennings’ role, was looking at the commercials of it because I think he comes back to say that I’d spent about a week on it over the past year, or something like that.
Mr Beer: Was there any equivalent set of exchanges over the basis of which you were giving evidence or your status when you were giving evidence?
Gareth Jenkins: I’m not quite sure what you mean by that?
Mr Beer: Were there any exchanges internally with Fujitsu, to your knowledge, over the status of your evidence?
Gareth Jenkins: Do you mean, in terms of whether I was an expert or not? Not that I’m aware of. As I say, what usually happened is if there was going to be some sort of engagement of my time for a particular case, then they would arrange with Post Office to have some sort of commercial cover for a number of days of my time and, if necessary, that would get extended so that I had something to put the time against. But exactly how those – I left that really to Penny to sort out, as long as I had the time code to book to on a Friday afternoon, then that was all I was really worried about.
Mr Beer: Can we look, lastly before the lunch break, at POL00097123, and look at page 9, please, and scroll down, and scroll down. Thank you.
An email from Ms Jennings:
“The case [that’s Patel] due for trial … Can you … please put this date in your diaries.”
Scroll up, please. You reply:
“… I’m not aware of this case or what might be required of me.
“I’m not aware of any other outstanding cases which I might be involved in.”
Scroll up. Reply from Sharron Jennings:
“… Gareth
“This is the one you supplied the expert report and witness statement for the week before last. Apologies for not explaining that properly in a previous email. It was a blanket email for all witnesses! It is unclear at this stage who will be required as witnesses and which evidence will be accepted without the need for attendance. I just thought if I let everyone know they can pencil it in and then I can let you all know nearer the time.”
Again, there’s a reference there to you supplying an expert report. Did that trigger anything in you?
Gareth Jenkins: No, because as far as I was concerned, what an expert report meant was a report by someone who understood what Horizon was about.
Mr Beer: Okay. Then if we scroll up, please. You say:
“Thanks for the clarification … If I am required to go to court … I need some more background on the specific case and exactly what’s being alleged. I appreciate that is not covered by my statement, but if I need to be an expert witness, I need to understand what is happening.”
In what sense were you referring to yourself as an expert witness there?
Gareth Jenkins: Again, as an expert on how Horizon operated.
Mr Beer: That’s not quite how it reads, is it?
Gareth Jenkins: I just didn’t understand the concept of an expert in the legal sense because I’ve got no legal background or training.
Mr Beer: Thank you.
Sir, that’s an appropriate moment. Can we take a break until 2.00, please.
Sir Wyn Williams: Yes. So we will take a break until 2.00 please. Thank you.
Mr Beer: Thank you very much, sir.
(1.08 pm)
(The Short Adjournment)
(2.00 pm)
Mr Beer: Good afternoon, sir. Can you see and hear us?
Sir Wyn Williams: Yes, thank you.
Mr Beer: Good afternoon, Mr Jenkins. We were dealing with the theme of the capacity in which you gave evidence in criminal proceedings and your understanding of the capacity in which you were giving evidence.
Gareth Jenkins: Yes.
Mr Beer: Can we continue that theme, please –
Gareth Jenkins: Okay.
Mr Beer: – by looking at FUJ00152872 and if we look at the top half of the page, please, we’ll see this is an email from you to Penny Thomas of 19 December 2009 with an attachment which in fact follows this document, and you say:
“Penny, I attach the expert witness statement with my comments.”
What you do, essentially, is attach Charles McLachlan’s second report in the Seema Misra case with your comments on it. In that first line of the email, you say:
“I attach the Expert Witness statement with my comments.”
You understood that Mr McLachlan was an expert witness?
Gareth Jenkins: Yes.
Mr Beer: Did you understand you were an expert witness in the equivalent capacity?
Gareth Jenkins: Not in the terms I now understand the capacity to be.
Mr Beer: We’re going to see in a moment you prepared documents in answer to his and you engaged in a joint discussion with him.
Gareth Jenkins: Yes.
Mr Beer: Did you understand you were performing a different function than he was?
Gareth Jenkins: I was clearly in a – at least, I certainly felt I was in a different position because I was not independent in the same way that he was, in that I was an employee of Fujitsu and, therefore, was effectively part of the POL prosecution team.
Mr Beer: Did you think you were part of the POL prosecution team?
Gareth Jenkins: Yeah, I think I probably did, because that’s how the POL lawyers were treating me.
Mr Beer: If we scroll forwards, please, to the report which starts at page 4, you’ll see his second report and then, if we go over the page, please, to page 5, you’ll see there’s an index. Just note for the moment – we’re going to come back to it in a moment – section 5 is “My Duties to the Court”. If we go over to page 6, please. You’ll see that he sets out, to start with, his instructions; can you see that?
Gareth Jenkins: Yes, I can.
Mr Beer: If we keep scrolling through this, if we do it at quite a slow pace, you’ll see that he sets out his qualifications?
Gareth Jenkins: Yes.
Mr Beer: Then if we scroll on, please. These are all qualifications and experience. Keep scrolling, keep scrolling, and keep going, and keep going, and again. Then at 1.3, he sets out the confidentiality of the document, and he makes some caveats at paragraph 1.4:
“This report should not be read as expressing any opinion on factual matters which depend on disputed testimony of the witnesses of fact, or legal issues. It … reflects my understanding of the position.”
Then, over the page, he then sets out at paragraph 1.5 his sources of information, and scroll on, please. A big list of the documents that he’s been given. Then over the page, and keep going. He then sets out at point 6 of his introductory paragraphs the scope of his work. He describes himself as an expert witness, not as a witness of fact. He then sets out what he’s done; looked at the documents that have been provided to him; what he hasn’t done, in 1.2; what he hasn’t had access to, in 1.3; where he attended, in 1.4 and 1.5. Then scroll down. Then over the page and at 1.7 he says:
“I have prepared an independent and objective report addressed to the Court.”
He sets out what his company’s fees are dependent on or not dependent on, and then he describes the structure of his report.
Then if we can go to page 35, please. He sets out his duties to the court. He understands his overriding duty is to the court and that he has complied with that duty and will continue to do so. Then, if you just scan to yourself what he says in paragraphs 2, 3, 4, 5 and 6, and scroll on. These are very similar to the letter of 18 November 2005, aren’t they?
Gareth Jenkins: Yes, I agree.
Mr Beer: Then, over the page, please. The expert declaration continues. Now, all of those parts that I’ve shown you, you would have read?
Gareth Jenkins: I will probably have skimmed through them, yes.
Mr Beer: You will have read them?
Gareth Jenkins: I will have read them, yes.
Mr Beer: When you read them, did you think, “Hold on, I’m giving evidence about the very same matters that Mr McLachlan is giving evidence. I don’t say any of these things in my witness statements or documents.”
Gareth Jenkins: No, because I thought I was in a different position because I thought I was actually – well, for a start I was providing witness statements, not an independent report and so I didn’t think I was – and I wasn’t independent, like he was.
Mr Beer: Never mind the independence point, what about all of things that he said as to all of his qualifications, his experience, his instructions, the documents that he had been shown, the scope of his report, the caveats within it?
Gareth Jenkins: Yeah.
Mr Beer: Did you think, “Hold on, I’ve done none of that”?
Gareth Jenkins: I probably realised I hadn’t done any of that but I didn’t see that it was – what I did was what I was asked to do, which was provide a witness statement or, at this stage, all I was doing was providing comments on his report.
Mr Beer: When you read this, and you thought, “Well, I’ve given witness statements in the past, in the capacity as somebody with expertise about Horizon, I’ve not included any of these things in any of the documents I’ve produced, should I have done so”?
Gareth Jenkins: That didn’t occur to me at the time. Maybe it should have done but it didn’t.
Mr Beer: You said a moment ago that you thought that you were appearing or performing a function in a different capacity. Did that actually occur to you at the time, ie, you addressing your mind to the fact that you were a witness from Fujitsu, therefore lacked independence and, therefore, the nature of the documents that you produced fell to be differently constructed?
Gareth Jenkins: I’m not sure that I actually analysed it that far. I was basically doing what I’d been asked to do, which was comment on the report and it then was suggested that it was presented in terms of a witness statement. So that’s what I did.
Mr Beer: Can we go forwards, please, to POL00029411. This is one of Mr McLachlan’s reports, it’s 50 pages long, with very lengthy exhibits to it, again in the Seema Misra case. One is dated 4 October 2010. You were provided with this report, weren’t you?
Gareth Jenkins: Yes.
Mr Beer: Did you read it?
Gareth Jenkins: I didn’t necessarily read the introductory bits to it, I was more concerned with the technical information that was in there and commenting on that. I may have read the whole lot, I can’t remember now, but what I was really interested in was the technical information in the report rather than things like the expert declaration and things like that because I didn’t see that was particularly relevant to what I was being asked to do at the time.
Mr Beer: Again, do you now remember that: that you skipped over or were not interested in the parts of the documents that set out an expert’s duties and the like?
Gareth Jenkins: I can’t – sorry, I can’t remember exactly what I did at the time but, as I say, what I would have been interested in and what I knew I was being asked to address was the technical aspects of the report.
Mr Beer: By this time, this is October 2010, so shortly before the trial commenced, you were aware that this was a criminal prosecution of Seema Misra?
Gareth Jenkins: Yes.
Mr Beer: A serious matter?
Gareth Jenkins: Oh, yes.
Mr Beer: Were you conscious of the seriousness of the matters, certainly for her, by this stage?
Gareth Jenkins: Yes, I suppose I was.
Mr Beer: You would, therefore, presumably have wished to approached your duties conscientiously and with care?
Gareth Jenkins: Oh, yes.
Mr Beer: Can we look, please, if we go over the page, again, the index page, and then over the page, please, and again. There’s an introduction and, if we go over the page, and again, he sets out a “Summary of findings”. If we go to page 20, please and read this together, section 4 of his report is points of difference between him and you. He says:
“I submitted my draft report to Gareth Jenkins, the expert from Fujitsu instructed by the Post Office for his review on 1 October 2010.”
That’s right, isn’t it: he sent you a draft of his report?
Gareth Jenkins: Yes.
Mr Beer: “We then had an opportunity to discuss points of fact and opinion over the phone. Jenkins was able to advise me of some errors of fact based on his extensive understanding of the Horizon system which I have not sought to dispute and I have incorporated corrections arising from these errors into this final report. Jenkins also indicated some areas where he held a different opinion and this section seeks to set out explicitly where our opinions differ side by side. I am relying on his annotations to my draft report to faithfully represent this position for the convenience of the court. I apologise if, any of the individuals items I have inadvertently misrepresented him or omitted comments of importance.”
Then 4.2:
“In relation to 2.2.1 Transaction Corrections, Jenkins is of the view that because the subpostmaster accepts the Transaction Corrections and has an opportunity to raise issues for correction this is not relevant to the case. In my view, if Transaction Corrections are incorrect or omitted then it is necessary for the subpostmaster to be able to have the evidence and training to contest them. The fact that there is a process for Transaction Corrections means that data entry errors are recognised as occurring. The … system does not automatically provide the evidence (in the form of vouchers) to enable the subpostmaster to easily raise or contest Transaction Corrections. Nor have the Post Office provided evidence that demonstrates that the training of Misra equipped her to deal with transaction corrections effectively.”
Then another example, if we go over the page, please:
“In relation to 2.2.1 … Jenkins is of the view that because the subpostmaster accepts the remittances and has an opportunity to raise issues for correction this is not relevant to the case. In my view, there is testimony from other subpostmasters that the end-to-end remittance process introduces incorrect data into Horizon, and because I have had no opportunity to investigate this, I am unable to include this as a source of problems at West Byfleet.”
So you engaged in a discussion with him –
Gareth Jenkins: Yes.
Mr Beer: – and he provided this document, which sets out his views and yours, where they differ?
Gareth Jenkins: Yes.
Mr Beer: You would have read this?
Gareth Jenkins: Yes, I would have done.
Mr Beer: Again, did you not think that you were giving evidence in the same capacity as Mr McLachlan?
Gareth Jenkins: Well, I didn’t see it in quite the same way because, again, I was there as an employee of Fujitsu and, therefore, very much as part of the Horizon system, whilst he was whereas a totally independent person. So I didn’t see it as being quite the same.
Mr Beer: You were being treated as the expert from Fujitsu that had been instructed by the Post Office, as he sets in his report?
Gareth Jenkins: Yes, okay.
Mr Beer: Can we go to page 27, please. He sets out the terms of reference for this exercise.
Gareth Jenkins: Yes.
Mr Beer: He sets out his instructions in point 1, point 2 his qualifications, and it follows the same format as the previous report. I’m not going to take you –
Gareth Jenkins: Yeah, yeah.
Mr Beer: You would have read this, wouldn’t you?
Gareth Jenkins: Yes, I probably would have skimmed through that because I didn’t think that was particularly relevant to the technical things I was being asked to comment on but I probably would have read through it.
Mr Beer: Did you not understand that you were the subject of the same duties as is set out over pages and pages in Mr McLachlan’s report?
Gareth Jenkins: No, I’d not understood that.
Mr Beer: Did you ask anyone, “Am I subject to the same duties as this other man keeps writing pages and pages about in his reports?”
Gareth Jenkins: No, I didn’t ask anyone. Perhaps I should have done but I didn’t.
Mr Beer: Can we move on, please. FUJ00156248. It’s an email of 1 October, that’s the date, if you remember, that Mr McLachlan said that he had sent you his report?
Gareth Jenkins: Yes.
Mr Beer: From you to Jarnail Singh, Penny Thomas and then Jean-Philippe – who I think you mentioned earlier – Jean-Philippe Prénovost; who was Mr Prénovost?
Gareth Jenkins: He was a lawyer within Fujitsu.
Mr Beer: And Tom Lillywhite, who was Mr Lillywhite?
Gareth Jenkins: He was Penny’s boss and Head of Security at that time, so effectively the roles that Brian Pinder had earlier. You say:
“Jarnail [third paragraph],
“As discussed, this is what I’ve received from Charles McLachlan the defence expert in the West Byfleet case.
“I’m concerned about the tone and some of the things being attributed to me. I’m in the process of annotating the document with my thoughts/comments.
“Charles [Mr McLachlan] spoke to me earlier this morning and has arranged to discuss my views with me at 4.00 pm this afternoon. At this point … I’m reluctant to have that conversation without some further guidance.
“When I get to the end of the document, I’m happy to pass on my annotated copy …”
Then something else.
So you were asking Mr Singh, is that right, for guidance on Mr McLachlan’s report when you first saw it and you got, in fact, 14 pages into it –
Gareth Jenkins: Yes.
Mr Beer: – is that right?
Gareth Jenkins: Yes.
Mr Beer: It goes to the lawyer at the Post Office, Mr Singh, and to Fujitsu Legal –
Gareth Jenkins: Yes.
Mr Beer: – Mr Prénovost; is that right?
Gareth Jenkins: Yes.
Mr Beer: Was this a plea by you for some guidance before your planned call with Mr McLachlan later that day?
Gareth Jenkins: Yes, clearly, yes.
Mr Beer: You say you were reluctant to have the conversation without guidance –
Gareth Jenkins: Yes.
Mr Beer: – further guidance?
Gareth Jenkins: Yes.
Mr Beer: What guidance were you looking for: guidance about what?
Gareth Jenkins: What I should be saying, how I should conduct the conversation, and so on. And I can’t remember what guidance, if any, I did actually receive but I think I did have the call with Charles McLachlan later on that afternoon.
Mr Beer: Can I press you on what guidance you were looking for?
Gareth Jenkins: How I should approach things, so it was really a case of I thought there were things in that document that we’d already discussed before, and I thought we’d agreed that they weren’t relevant, and here they were coming back up again. So it was a case of why is that happening?
Mr Beer: So was it about the substance of what Mr McLachlan was saying in the report?
Gareth Jenkins: Yes.
Mr Beer: Rather than your role, the role that you were performing, that you were seeking guidance?
Gareth Jenkins: No, it was the substance.
Mr Beer: How could they each help you, the two lawyers, on the substance of what he was saying?
Gareth Jenkins: It was really a case of why was he coming back with things that I thought we’d already discussed and what I should do about that, and how to approach that in terms of discussing things with him, as far as I can remember.
Mr Beer: You tell us in your third witness statement at paragraph 463, that Mr Singh did not reply to this email.
Gareth Jenkins: Okay, I can’t remember the detail at this point but I clearly had researched it at the time I was writing that witness statement.
Mr Beer: Can you recall whether any guidance was forthcoming from anyone else, including Mr Prénovost at Fujitsu?
Gareth Jenkins: I can’t remember, I don’t believe I’ve seen any emails that suggest that; whether there were any phone calls or not I just can’t remember.
Mr Beer: By the time you were sending this you hadn’t finished reading; you were a third of the way in?
Gareth Jenkins: Yes. I mean, the point was I’d received this probably first thing in the morning, so I thought let’s – whilst I’m reading it, then that gave people a chance to react because I’d already established the fact that I – that Charles McLachlan wanted to talk to me at 4.00 this afternoon, therefore let’s give people warning, so they have a chance to say and – see what they can do to help me in preparation for that call later on this – that day.
Mr Beer: Did you go back to Mr Singh or Mr Prénovost when you’d finished reading the report?
Gareth Jenkins: I think at some point I sent an annotated copy of the report out. I’m not sure exactly who to – probably to the same people. I can’t remember the exact sequence of the emails that –
Mr Beer: That’s the one we looked at earlier, the annotated version of the report?
Gareth Jenkins: Yes, but I can’t remember exactly who I sent it to and what stage in the proceedings. No, I think the one we looked at earlier was one from about two or three days later.
Mr Beer: Okay. You think on this day you sent an annotated version out?
Gareth Jenkins: I believe so but I can’t be 100 per cent sure.
Mr Beer: Did you at any time ask anything about the duties of an expert witness about which Mr McLachlan spoke extensively in his report?
Gareth Jenkins: I doubt it. I was more concerned about the technical detail rather than things like that, that I saw as being administrative parts of the report.
Mr Beer: Did you chase Mr Prénovost to say, “Look I’m doing this, I’m still doing this without any guidance, can I get some help, please”?
Gareth Jenkins: I don’t remember, I’m afraid.
Mr Beer: Or did you just plough on?
Gareth Jenkins: I probably just ploughed on but I don’t know.
Mr Beer: Did you have an in-person meeting with Mr Singh, Mr Longman and Mr Tatford about this?
Gareth Jenkins: The following week I did, yes.
Mr Beer: I think we’ve got no note of that meeting; is that right, so far as you know?
Gareth Jenkins: I don’t remember seeing any note of that meeting. I certainly haven’t seen one as part of the Inquiry.
Mr Beer: I don’t think you’ve got a recollection of what was discussed at that meeting?
Gareth Jenkins: No, my main recollection is coming up to London to have a meeting in barristers’ chambers, which is something I’d never done before, and meeting – I think that was probably the first time I’d met all three of them, so – though I’d spoken to them on the phone before that. But exactly what was discussed at the meeting, I can’t really remember.
Mr Beer: So is it the case that when you were being – looking at the matter generally – sent these documents from Mr McLachlan and you read the parts in them that concerned his instructions, the documents with which he’d been provided, the scope of the work that he had undertaken, the caveats to the work he had undertaken, the limitations of it, and quite extensive passages about the duties of an expert, that didn’t trigger anything at all with you?
Gareth Jenkins: No, I don’t think it did.
Mr Beer: Can we turn forwards, please, to FUJ00124313. This is a document in a different case.
Gareth Jenkins: Yes.
Mr Beer: This is concerning the prosecution of Kim Wylie, and it’s a document prepared by you, we can see the date, 19 February 2013; can you see that?
Gareth Jenkins: I can see that, yes.
Mr Beer: So it’s two and a bit years later?
Gareth Jenkins: Yes.
Mr Beer: What you do, if we just summarise what you do in this report, it comments on essentially requests for disclosure that were being made by the defence in Ms Wylie’s case. What you do is you set out passages from an expert report by Michael Turner, and then you comment on them?
Gareth Jenkins: That seems to be what this is doing, yes.
Mr Beer: Okay. You say in the introduction “I have been asked to report on the document “Report: Disclosure Requests Computer Evidence” provided by the defence in the case of [Kim Wylie].
“In order to do that, I have copied in the report below in blue font and added my comments in black font.”
Again, I don’t think we’ve got the –
Gareth Jenkins: Yes, I appreciate that.
Mr Beer: We’ll have to work our way through it:
“In summary, it would appear that the Defence expert is looking for as much information as he can to carry out a detailed analysis. I have no problem in him doing that and I am happy to assist in such an analysis (as I have done in the [past] with other defence experts), since the data requested is proprietary to Horizon or Horizon Online and is unlikely to be understood easily without some guidance.
“Any such analysis is likely to require a lot of time and effort to analyse and therefore incur considerable cost and elapsed time.”
Then 2, “Disclosure Requests Computer Evidence”, “Terms of Reference”. Now, this is you setting out passages from Mr Turner’s report.
Gareth Jenkins: Yes, this is just a cut and paste of Mr Turner’s report.
Mr Beer: You’ll see that he sets out his instructions and then, at 1.3, he sets out his qualifications. Over the page, he annexes a CV in Appendix A – I’m not going to look at that – he sets out a statement of independence and then underneath “Statement of Independence”, you’ve written:
“I note that the CV refers to another Horizon case, R v Julia Richards, Winchester Crown Court. I have no knowledge of that case.”
That’s you, adding that in?
Gareth Jenkins: Yes.
Mr Beer: Then back to him:
“I have made a declaration at the end of this is report.”
Then he continues. I’m not going to go through all of this. The part that’s indented is him and the part that’s not indented is you; is that right?
Gareth Jenkins: That’s my belief, yes.
Mr Beer: Can we go on to page 6, please, and scroll down, “Expert’s declaration”, you’ll see that Mr Turner had set out the expert’s declaration.
Gareth Jenkins: Yes.
Mr Beer: Yes:
“I understand that my overriding duty is to the court …
“I have set out in my report what I understand from those instructing me to be the questions in respect of which my opinion as an expert is required.
“I have done my best … to be accurate and complete … All of the matters on which I have expressed an opinion lie within my field of my expertise.
“I have drawn to the attention of the court all matters, of which I’m aware, which might adversely affect my opinion.
“Wherever I have no personal knowledge, I have indicated the source of factual information.
“I have not included anything in this report which has been suggested to me by anyone, including the lawyers … without forming my own independent view …
“Where … there is a range of reasonable opinion, I have [set out] the extent of that range …
“At the time of signing I [believe] it to be accurate. I will notify those instructing me if … I subsequently consider that the report requires any correction or qualification.
“… this report will be evidence that I give under oath …
“I confirm that … the facts in the report are within my own knowledge”, et cetera.
Then you’ve added, “Standard stuff. No comment required”, haven’t you?
Gareth Jenkins: Yes.
Mr Beer: In this context, “standard” means usual or normal, doesn’t it?
Gareth Jenkins: It means I’d seen something similar before, for example in the case of Mr McLachlan but I hadn’t really taken much notice of it and I didn’t really think any of that applied to me, wrongly.
Mr Beer: Is that what that means, you’re saying?
Gareth Jenkins: That’s –
Mr Beer: – it’s rather more than those five words set out?
Gareth Jenkins: That was what I think I felt at the time, that I’d seen that sort of stuff in other reports but I didn’t see that any comment was required and I didn’t realise that it applied to me because I wasn’t making such declarations.
Mr Beer: That’s not what you say here, is it?
Gareth Jenkins: No.
Mr Beer: You say it’s standard stuff and “standard stuff” here, would you agree, in context, means it’s normal or usual?
Gareth Jenkins: What I meant by that was that I’d seen something similar to that before, for example in Professor McLachlan’s reports.
Mr Beer: So on your account, it indicates that it was something with which you were familiar?
Gareth Jenkins: I had seen it before.
Mr Beer: Therefore it was standard?
Gareth Jenkins: For independent experts’ reports.
Mr Beer: So going back to paragraph 1 of this list, you understood that his overriding duty was to the court, yes?
Gareth Jenkins: Yes.
Mr Beer: That’s standard, usual stuff?
Gareth Jenkins: Yes.
Mr Beer: Did you understand that your overriding duty was to the court?
Gareth Jenkins: Well, I took “overriding duty to the court” being just to tell the truth.
Mr Beer: So the answer is, yes, you did understand that you had an overriding duty to the court?
Gareth Jenkins: Yes.
Mr Beer: Paragraph 2, you were familiar with that. That was standard and usual, that the experts should set out in their report the questions in respect of which the expert’s opinion was sought?
Gareth Jenkins: I’m not clear that I would necessarily have considered all those points in detail when I wrote “standard stuff” at the end. I was just commenting in general on section 4 is an expert declaration and I’d seen expert declarations before. So I’m not – I doubt if I would have gone through looking at each of those paragraphs in detail and considering what they said. It was a case of, yes, I’d seen an expert declaration at the end of reports before and I saw that as a standard thing without going into detail as to exactly what it said.
Mr Beer: In relation to paragraph 3, which you regarded as standard or usual, did you understand that your duty in preparing documents was to be accurate and complete?
Gareth Jenkins: Well, yes, certainly it needed to be accurate and complete but to say whether I took detailed note of exactly what it was saying in there at the time, I think it’s unlikely. I think I just saw that as an expert declaration and that comment was referring to that as a section rather than to the individual words within it.
Mr Beer: When you were preparing statements or reports that were annexed to statements, did you understand that you were under a duty to mention all matters which you regard were relevant to the opinions which you expressed?
Gareth Jenkins: I don’t know that I did understand that because I’m not sure that I necessarily did.
Mr Beer: When you provided statements to the court or reports that were annexed to statements to the court, did you understand you were under a duty to draw to the attention of the court all matters which might adversely affect your opinion?
Gareth Jenkins: No, I didn’t understand that at the time.
Mr Beer: You thought you were entitled to keep to yourself matters that might adversely affect your opinion?
Gareth Jenkins: I was – well, I was asked to comment on documents and reports, and that was all I thought I needed to do, and that’s what I tried to do.
Mr Beer: Answer the narrow question that was asked of you?
Gareth Jenkins: Yes –
Mr Beer: You –
Gareth Jenkins: – and no one had told me that I needed to do more than that.
Mr Beer: Did you understand yourself to be under a duty to indicate in your documents submitted to courts where you have no personal knowledge of a matter mentioned?
Gareth Jenkins: No, I wasn’t aware that I had any need to do that.
Mr Beer: You weren’t aware that you had to indicate the source of your information if it wasn’t direct knowledge?
Gareth Jenkins: No.
Mr Beer: Did you think, looking at number 6, that you were permitted to include matters in your documents submitted to the court – if we go over the page – which had been suggested to you by anyone else, including the lawyers?
Gareth Jenkins: No, because, certainly as we’ve seen in some of the cases, then there were comments made on my reports by the lawyers, and I actually changed things there as a result of some of those comments. So, therefore, I didn’t realise that there was anything wrong with doing that. Obviously, if I disagreed with the comments, then I would not change things but if I was happy to agree with the comments and that it didn’t detract from what I was trying to say, then I was happy to accept suggestions as to how my wording could be improved.
Mr Beer: We’re going to come tomorrow to look, in some detail, at the occasions when you changed your evidence as a result of the suggestions made to you by others.
Gareth Jenkins: Yes.
Mr Beer: In particular, lawyers.
Gareth Jenkins: Yes.
Mr Beer: You didn’t understand that you were under a duty not to include in your documents suggestions made to you by others?
Gareth Jenkins: No, I didn’t understand that at the time.
Mr Beer: Did you understand that you were under a duty to set out a range of expert opinion in relation to a matter where there was a range of expert opinion?
Gareth Jenkins: No, I’m not sure that necessarily applied but, no, I certainly wasn’t clear – I wasn’t aware of that duty.
Mr Beer: Why do you say you’re not sure that that necessarily applied?
Gareth Jenkins: Well, you’re talking about a range of opinion then I’m not sure that there were cases where there was a range of opinion, if –
Mr Beer: On all the matters that you gave evidence, the answers were binary?
Gareth Jenkins: Well, okay, I certainly wasn’t aware that I needed to set out anything to do with a range.
Mr Beer: In relation to the other matters mentioned in paragraphs 8, 9 and 10, are your answers the same: that you wouldn’t be familiar with those duties –
Gareth Jenkins: No.
Mr Beer: – as applying to you?
Gareth Jenkins: No, all I thought I had to do was answer the questions that I was being asked to answer and that they had – obviously the answers had to be truthful.
Mr Beer: None of these obligations can have been new to you, because you wouldn’t have described them as “standard stuff”, would you?
Gareth Jenkins: They weren’t new to me but I don’t think I’d taken in the detail of them at all. I just – as I said earlier, the “standard stuff” was probably more to do with the title of the section “Expert declaration”, rather than the details of the ten individual points within that.
Mr Beer: You weren’t saying the title is standard stuff; you’re saying the content is standard, aren’t you?
Gareth Jenkins: The content is – yes. But I hadn’t really considered in detail what the content actually was saying.
Mr Beer: Why not?
Gareth Jenkins: Because I was concerned with the technicalities, so it was the other bit – I’m a technician, so I was more concerned about the facts of how the system was working, rather than the other – the legal niceties, if you like.
Mr Beer: Well, both Messrs McLachlan and Turner were technical experts too, weren’t they?
Gareth Jenkins: Yes.
Mr Beer: Did you think, “Well, hold on, I keep reading these reports from other guys, and they keep saying this stuff at the beginning and the end of their reports about the duties that they’re under. Does any of that apply to me”?
Gareth Jenkins: Well, I wasn’t producing reports; I was producing a witness statement.
Mr Beer: Do you think that’s the answer to the question: because the document on which you were typing or was being typed for you, is headed up “Witness statement”, you needn’t have abided by any of these requirements?
Gareth Jenkins: It didn’t occur to me that this would apply to me. I clearly understand now that that’s wrong but, I mean, that was where I was at the time.
Mr Beer: That can come down. Thank you.
You tell us in your third witness statement – no need to turn it up, it’s paragraph 331 at page 110:
“I knew that, like any other witness, I was required to be truthful, although it would never have occurred to me to be anything else. The approach that I took was the one that I’d asked – was asked to take by Post Office lawyers, David Jones, a lawyer at Fujitsu, also read at least some of my witness statements.”
You deal at length in that witness statement with limited instructions you say you were given and your limited understanding of an expert’s duties. You were, however, aware that you were being asked to produce technical evidence on which a criminal prosecution was based, weren’t you?
Gareth Jenkins: Yes.
Mr Beer: You would have been aware that the outcome for the subpostmaster in question, the outcome that the Post Office was seeking, was the criminal conviction of a subpostmaster?
Gareth Jenkins: Yes.
Mr Beer: You’d be aware of all of the consequences that such a conviction would entail, including a possible term of imprisonment?
Gareth Jenkins: Yes.
Mr Beer: You signed a statement of truth on each of your witness statements, didn’t you, a declaration?
Gareth Jenkins: Yes.
Mr Beer: I think you acknowledge, or we’ll find tomorrow that you acknowledge, that the witness statements that you signed presented a partial picture of what you understood about Horizon’s issues and faults because you were answering only the narrow questions that you had been asked?
Gareth Jenkins: That’s correct, yes.
Mr Beer: How did you feel able to sign witness statements with a declaration on them which presented only a partial picture of what you understood about Horizon’s issues and faults?
Gareth Jenkins: I don’t think I was being asked about Horizon’s issues and faults. What I was being asked to do was address the questions in the reports that I was addressing.
Mr Beer: You would know that witnesses, when they come to give evidence, are asked to tell the truth and the whole truth?
Gareth Jenkins: Yes.
Mr Beer: Did you think you were only required to tell the truth in your witness statements?
Gareth Jenkins: I was talking about the – those aspects that related to the Horizon system and I believe I did tell the truth and the whole truth, as far as the Horizon system was operating in the specific branches at the specific times that I’d looked at data.
Mr Beer: But you didn’t feel under any compunction to volunteer information about other faults or system defects about which you had not been asked?
Gareth Jenkins: I didn’t think that they were relevant in those particular cases.
Mr Beer: Just expand on that. You’re saying that you took a conscious decision not to mention them because of your own assessment of relevance?
Gareth Jenkins: Not a conscious decision. As far as I was concerned, the system was behaving correctly in the branch at the time and I’d seen nothing to show that it wasn’t and, therefore, other issues that I may have been aware of were ones that had been – that had gone on in the past, or in the future, and had been fixed and did not apply to the branch at the time that I was considering.
Mr Beer: We’ll come to the detail tomorrow but, in the light of the answers you’ve just given, are you saying that, in every case in which you provided a witness statement, you had undertaken or caused to be undertaken a detailed examination of the data relating to that branch?
Gareth Jenkins: I had undertaken an analysis of the data, not necessarily a completely thorough analysis of it, but what I felt at the time it was sufficient to show that things were working okay at that time.
Mr Beer: What about the generic witness statements that you supplied later?
Gareth Jenkins: Those, I think, were done at a much higher level. They were giving a – they were effectively saying that Horizon was behaving as it should and that information could be found from the audit trail to show what had actually happened in the branch to look at any more detailed question.
Mr Beer: Can we look, please, at your third witness statement at page 139. Again, we’re going to deal with Ms Misra’s case in detail tomorrow and on Thursday but can we look at paragraph 404, please. You’re dealing here with Ms Misra’s case, and you say:
“I have already explained that I had referred to the event timeout/locking issues causing missing transactions in Legacy Horizon and that Fujitsu had drawn these to Mr Singh’s attention. I don’t believe that [the Post Office] asked me whether it should disclose these (or any other Horizon problems) to Mrs Misra. I haven’t seen any emails from around this time which suggest that. I hadn’t been involved in responding to defence requests for disclosure in 2009, nor the abuse of process application made by the defence in February 2010. I don’t think my email of 1 March 2010 is about disclosure at all.”
So you say here you can’t recall the Post Office ever raising whether they, the Post Office, or you should disclose the issue of locking problems to Ms Misra?
Gareth Jenkins: Correct.
Mr Beer: By this time, you knew that, if you were to sign a witness statement in Ms Misra’s case, you would have to sign a statement of truth on it?
Gareth Jenkins: Yes.
Mr Beer: You knew that those statements were being used as evidence in court to support a criminal prosecution?
Gareth Jenkins: I had sought guidance, I had said that, until we had actually looked at the data, that I was not able to say that there weren’t any problems in Horizon and, when we – when I came to get – see the data, which was in early March, then there was a check done of the NT events, which was the way that this event timeout problem manifested itself and, therefore I was able to include that as having been a problem in that branch.
Mr Beer: By this time, you had got – this is 2010 – wider knowledge of bugs, errors and defects in the Horizon system than, in the event, was reflected in your witness statements in the Seema Misra prosecution?
Gareth Jenkins: I’m not quite sure what you mean by that.
Mr Beer: So, by March onwards, March 2010 onwards, when you made witness statements in Seema Misra’s case, you didn’t disclose in those witness statements all bugs, errors and defects about Horizon of which you knew?
Gareth Jenkins: I didn’t think I needed to do that because what I thought I needed to worry about were ones that had occurred at that particular branch at the time in question.
Mr Beer: Did you consider reflecting in your witness statements or making qualifications in your witness statements, to set out exactly what you had been asked to do and not asked to do, and make clear that you had not been asked about other issues?
Gareth Jenkins: That didn’t occur to me, I’m afraid.
Mr Beer: Looking back now, do you think that a reader of your witness statements could reasonably gain the view that you were setting out all that you knew about bugs, errors and defects in Horizon?
Gareth Jenkins: I don’t think I was ever asked to consider bugs, errors and defects. What I was asked to do was comment on the reports produced by Professor McLachlan.
Mr Beer: We’ve seen you on a number of occasions provide information through an informal channel, through emails, to Post Office?
Gareth Jenkins: Yes.
Mr Beer: Did you have any understanding or expectation of whether that information provided informally would be disclosed to the defence in the criminal prosecutions you were involved in?
Gareth Jenkins: I didn’t know what – I didn’t understand how the process worked in that respect. I mean, clearly, if someone had come back to me and said, “Well, in that case you need to put something about it in your witness statement”, I would have done that, but no one did.
Mr Beer: Did you think to include in your witness statements the fact of your informal communications with both Post Office and Fujitsu about the substance of the evidence that you were to give?
Gareth Jenkins: No, that never occurred to me.
Mr Beer: Or the detail of those communications?
Gareth Jenkins: No, I didn’t think that was relevant.
Mr Beer: What did you think was going to happen to the information in those communications?
Gareth Jenkins: I didn’t know what was going to happen to them. As I say, if someone had told me that I needed to do something about it, I would have done something about it, but no one came back and said I needed to do something formal about that – that sort of exchange.
Mr Beer: That document can come down. Thank you.
By the end of the Seema Misra trial, would you agree that you have, by then, had a number of opportunities to see whether the Post Office was adopting an approach of seeking to tweak your evidence for its own interests?
Gareth Jenkins: I didn’t think that at the time. Having looked back at things now, then I can understand that that may have been happening but, at the time, I thought everything that was happening was just a legitimate tidying up of the statements to make them more readable.
Mr Beer: Are you really saying, bearing in mind we’re going to look at these tomorrow, that you thought that what happened in both the Hughie Thomas case and the Seema Misra case was a tidying up exercise to make your statements more readable?
Gareth Jenkins: And clarifying that, I don’t think any of the changes that were made actually changed the technical background to what was happening. Yes, it changed the wording and the way things were explained to some extent but I think, in terms of they were still setting out the facts as I understood them at the time.
Mr Beer: I’m thinking more about the things that the Post Office was trying to get you to include and, by the time we get to October 2010, had you not formed the view that the Post Office had tried to alter the substance of your evidence, both in the Hughie Thomas case and in the Seema Misra case and, in some cases, you’d resisted and sometimes you’d not?
Gareth Jenkins: I’d not seen it that way. I mean, looking back it now, I can understand exactly what you’re saying. But, at the time, I didn’t see it as being anything strange or unusual.
Mr Beer: Can we look, please, at your witness statement, your third witness statement, at paragraph 466, please, which is on page 162. Scroll down, please. Thank you.
466, thank you. We’re going to come back, again, to the detail of this later, but you say, on that time and date, you emailed a draft statement to Mr Tatford, Mr Singh and Mr Longman. This is in Seema Misra’s case.
Gareth Jenkins: Yes.
Mr Beer: “I used the same template as I had for all my previous statements. I specifically asked for their feedback as to whether it was correct in terms of approach and style. No one suggested that I needed to add any sort of expert declaration to it. Mr Tatford made comments on this draft which are visible at [and then you give a reference]. He appeared to want me to make some points more strongly in favour of the Post Office than I had done. In particular, he wanted me to say it looked as though Mrs Misra had stolen the money rather than it was incompetence.”
Then if we go forward to paragraph 468, which is over the page. You say:
“I think overall my responses to Mr Tatford’s comments demonstrate that I was not comfortable with some of the points he tried to press.”
So, by this time, you had had the experience of Graham Ward in the Hughie Thomas case seeking to press amendments to your witness statement, hadn’t you, back in 2005?
Gareth Jenkins: Yeah, I mean, I was going to say that’s about four or five years earlier, so I probably wasn’t really conscious of that at this time.
Mr Beer: Had you not remembered – I mean, you didn’t make witness statements in many cases, I think it’s 15 overall?
Gareth Jenkins: Yeah.
Mr Beer: Had you not remembered Mr Ward being dissatisfied with your use of language in a witness statement and making a marked-up version for you suggesting parts being deleted?
Gareth Jenkins: I probably didn’t remember that in 2010, no.
Mr Beer: But, in any event, here you say that Mr Tatford wanted you essentially to harden up the case against Mrs Misra?
Gareth Jenkins: I’m saying that now, looking back at the email exchanges. I wouldn’t necessarily have put it in that terms back in 2010.
Mr Beer: Why not? A barrister saying, “Please include in your witness statement”, if we just go back to what you said at the foot of the page, 466:
“In particular he wanted me to say it looked as though Mrs Misra had stolen the money rather than it was incompetence.”
Gareth Jenkins: Yeah.
Mr Beer: It doesn’t take hindsight to realise that that is a barrister asking you to harden up the case –
Gareth Jenkins: Yes, and I refused to do so –
Mr Beer: I’m asking what effect this had upon you.
Gareth Jenkins: I don’t know, I just assumed that this was normal practice. I had no experience to compare it with.
Mr Beer: Did it not make you feel uncomfortable?
Gareth Jenkins: Yes, it probably did make me feel uncomfortable, and I resisted it.
Mr Beer: Why did it make you feel uncomfortable?
Gareth Jenkins: Because he was trying to change what I was saying to say what he wanted to say, rather than what I wanted to say.
Mr Beer: Did this not cause you to think that there is a need for caution here in my dealings with the Post Office, in particular when giving evidence on behalf of the Post Office, that their barrister is trying to get me to say something that I can’t and don’t want to say?
Gareth Jenkins: No, I think it just made me make sure that I did stick with my guns when I disagreed with what I was being asked to do.
Mr Beer: Again, did you not consider whether you ought expressly to qualify your statements to spell out what you were saying about Horizon not being your complete knowledge about Horizon integrity?
Gareth Jenkins: That didn’t occur to me at the time.
Mr Beer: As I’ve said, you later went on to produce so-called generic statements –
Gareth Jenkins: Yes.
Mr Beer: – in a series of cases, essentially without access to the complete underlying data?
Gareth Jenkins: Yes, because I saw those as being high-level statements. I was expecting to get into more detail at some later stage. As I say, when I first produced that, I had no knowledge as to anything that was involved, and I assumed that that was normal practice. Again, I now understand that it wasn’t and was probably totally inappropriate but, at the time, I didn’t realise that.
Mr Beer: Did your experience in the Hughie Thomas case and the Seema Misra case, and the approach of the Post Office in particular in the Seema Misra case, not suggest to you that you ought to proceed with extreme caution in producing such generic statements?
Gareth Jenkins: I didn’t realise at the time that anything was actually improper or going wrong with what was happening. I did get – certainly in some of the cases – with the generic statement, I did actually seek guidance from Fujitsu lawyers as well as Post Office lawyers as to whether this was the right thing to do and whether I should carry on making those sort of statements, and they seemed to be supporting what I was doing.
Mr Beer: Or was it the seeming success of the approach in the Misra case created a false sense of confidence for you and for the Post Office; it resulted in a conviction, after all?
Gareth Jenkins: I don’t know that I considered it that way. Again, the generic statement was about two years later, so I was probably looking at that in isolation rather than looking back as to what had happened before.
Mr Beer: Can we just look back, please, to your attitude to witness evidence in a different case by looking at FUJ00083741. If we just scroll down, please. Can you see that the email from Mr Dunks to you and Penny Thomas of 28 October 2010 concerns Porters Avenue, which I think is Mr Hosi’s branch, yes?
Gareth Jenkins: Yes, I believe so.
Mr Beer: “It looks as if everything has been sent at one time or other. I can send all if wanted”, et cetera.
Then I just want to see your reply. If we scroll up, you say:
“Andy,
“Can you let me have them all. I’ve got to do another ‘expert report’ and this time I want to actually read your logs properly!”
You’re writing here to Mr Dunks and Penny Thomas, yes?
Gareth Jenkins: Yes.
Mr Beer: Well-known colleagues?
Gareth Jenkins: Yes.
Mr Beer: You’re purposely using the term “expert report” by putting it in inverted commas, aren’t you?
Gareth Jenkins: Yes.
Mr Beer: Was that what had been requested of you?
Gareth Jenkins: I can’t remember exactly what had been requested of me at that time.
Mr Beer: But you’re putting it in inverted to commas, like it’s a term of art, yes?
Gareth Jenkins: I can’t remember – I tend to be fairly liberal with inverted commas and capital letters, and things like that, in my emails.
Mr Beer: Isn’t this is an indication that you realised that what you were being asked to do was not simply give evidence because you had expertise in an area, namely the Horizon system, but you were being asked to provide an expert report just like the ones we’ve read?
Gareth Jenkins: No, I don’t think so. I think it was more to do with the fact that it was a report as an expert on Horizon. Someone who actually understood the Horizon system well.
Mr Beer: Your evidence is that, although you say here that you’ve got to do another “expert report”, in inverted commas, you didn’t understand what that meant, ie the provision of an expert report, or that it came with specific duties?
Gareth Jenkins: That’s correct.
Mr Beer: By this time, you’d produced statements for use in Hughie Thomas’ case and earlier in the year in Seema Misra’s case?
Gareth Jenkins: Correct.
Mr Beer: You say here:
“… this time I want to actually read your logs properly!”
Have you not considered the logs properly in those other cases?
Gareth Jenkins: I don’t know that I – I certainly had not read in – been shown any logs to do with Mr Thomas’ case. I was aware in Mrs Misra’s case that there were some logs, which I thought had been handled by Andy Dunks and, when I listened to his evidence in court, I realised that there was some things that he was saying that maybe I should have looked into and I actually did some investigation that evening as to what I had heard about the logs and I realised that, actually, maybe in future, I should have actually taken more notice of what the Helpdesk calls were involved, hence this comment here.
Mr Beer: It was a flippant comment, wasn’t it?
Gareth Jenkins: I don’t know that it was a flippant comment; it was more a case of saying I’d learnt from the fact that I’d been taken by surprise by some of the things that Andy was saying in Mrs Misra’s case as to what he’d actually found in the Helpdesk calls that I’d not been aware of and realised that maybe I should have taken more notice of those and, therefore, moving forward, I wanted to learn from that.
Mr Beer: Thank you.
Sir, can we break until 3.15, the afternoon break. Thank you very much, sir.
(3.01 pm)
(A short break)
(3.15 pm)
Mr Beer: Good afternoon, sir, can you continue to see and hear us?
Sir Wyn Williams: Yes, thank you.
Mr Beer: Thank you.
Mr Jenkins, you’ve told us today that you were never provided with written instructions as to the duties and responsibilities of an expert witness in the cases in which you were instructed –
Gareth Jenkins: Correct.
Mr Beer: – and that you’ve no recollection of anyone explaining to you that you were subject to certain duties as an expert or being given oral instructions as to what those duties were?
Gareth Jenkins: Correct.
Mr Beer: Have you read the witness statement of Warwick Tatford, the prosecution barrister in the Seema Misra case?
Gareth Jenkins: I have.
Mr Beer: Did you listen to or watch his evidence or read the transcript –
Gareth Jenkins: I was actually here for that particular event.
Mr Beer: – for that day. Can we look at his witness statement to start with, please, WITN09610100, and page 5, please. Second line down:
“At the time of the Misra case I ensured that the Callendar Square bug was disclosed and made it very clear to those instructing me that enquiries should be made with Fujitsu about any other problems and any other bugs should be disclosed. I made this clear also to Gareth Jenkins, the expert instructed by the Crown, and I made it very clear to Mr Jenkins that he was under a duty to provide frank disclosure of Horizon problems to the defence expert instructed in that case.”
Gareth Jenkins: I don’t believe we had that sort of conversation. Also, I believe that Mr Tatford sent that information to one of the Post Office Investigators, who then asked me a slightly different question, rather than the full question that Mr Tatford had intended for me to be asked.
Mr Beer: That’s certainly the case because we can see that on the documents. We’ll explore those tomorrow as we’ve explored with other witnesses.
Gareth Jenkins: Yes, but I’ve no recollection of being asked to talk about other bugs or that I would have considered irrelevant anyway, by Mr Tatford or anyone else.
Mr Beer: I want to get your evidence clear on this: he says that he made this clear, ie that any other problems or any other bugs should be disclosed to you, and that he made it very clear to you that you were under a duty to provide frank disclosure of Horizon problems to the defence expert. Are you saying that he is incorrect?
Gareth Jenkins: I’m saying that I had not understood that from him.
Mr Beer: Did you have a conversation with him about the extent to which you needed to provide disclosure of Horizon problems to the defence expert?
Gareth Jenkins: I’ve no recollection of such a discussion.
Mr Beer: By that, do you mean that it did not occur or that it may have occurred but, because of the passage of time, you no longer remember it?
Gareth Jenkins: I don’t believe it would have occurred because, if it had, I would have done something about it.
Mr Beer: You appreciate there’s a bit of reverse engineering going on there –
Gareth Jenkins: Yes, I understand that, yes, but, as I say, I have no recollection of it and, if I’d been told I needed to disclose other problems in Horizon, then I would have made efforts to find out about other problems in Horizon, even though I thought that they would have been irrelevant.
Mr Beer: Hold on. A moment ago you said, “I wouldn’t have told the defence expert about problems in Horizon” because you wouldn’t have believed that they were relevant to Seema Misra’s case?
Gareth Jenkins: Well, that’s right. So, therefore, if I’d been told explicitly I should have done, then I would have done something about it, but that’s why I’m saying that would not – I don’t believe that I was told that I needed to do that.
Mr Beer: So his evidence here is in error, is it?
Gareth Jenkins: I can’t say categorically that – I – that is not my understanding of what occurred at the time and I think I would have behaved differently if I had been briefed in the way that he suggests that he briefed me.
Mr Beer: What do you remember about a conversation with Mr Tatford about the extent of your disclosure duties to the defence expert?
Gareth Jenkins: I don’t remember having any conversation along those lines.
Mr Beer: Can we turn to INQ00001094, and turn to page 17, please. If we look at internal page 68, which is bottom right, line three onwards. This is Mr Tatford giving evidence:
“You say in your statement – I’m not going to turn it up – you took great pains in all your conversations with Mr Jenkins to make sure he understood the duties of an expert witness?
“Answer: Yes.
“Question: You explained it was his overriding to assist the [that should be ‘court’] –
“Answer: Yes.
“Question: – to give an opinion that was objective and unbiased, and that that duty overrode any obligation that he might feel to the party calling him: the Post Office. You explained that it was his duty to disclose anything that might undermine his position and that he should be entirely open with both the Post Office, as prosecutor, and Professor McLachlan, about any Horizon problems?
“Answer: Oh, yes, because the – I had asked previously in my advice for Fujitsu to be contacted and to inform us of any problems and I saw Mr Jenkins as an obvious route to doing that. That’s how I saw things. And it seemed to me particularly from the feedback I was getting from the defence, that this approach was working.
“Question: … you’ve referred to ‘the defence’ …
“Are you saying that [was a conversation] with a colleague in chambers –
“Answer: No …
“It was perfectly clear to me that he found it helpful to work with Mr Jenkins. It received to fit his way of doing things because his way of approaching things was to suggest hypotheses which needed somebody to help him with. They needed to sit down together and it’s absolutely clear that they did that, from the evidence they gave at the trial.
“Question: Before we get into the detail of the communications between you, the Post Office and Fujitsu and Mr Jenkins, if you were mindful of these expert duties and the need to make them crystal clear in somebody who did not enjoy functional independence from the party that was calling them, and you explained them to Mr Jenkins, how is it that every witness statement which the Post Office sought to rely on from Mr Jenkins omitted any of the necessary inclusions for an expert report?
“Answer: Because most of – until the last statement, all of those were responses to Professor McLachlan. It is important to bear in mind that there had been an abuse of process”, et cetera.
Did Mr Tatford make clear to you the duties that he spoke about?
Gareth Jenkins: I don’t believe he did.
Mr Beer: If we just go back to them, the previous page, bottom right a duty “to give an opinion that was objective and unbiased”; did he tell you that he have a duty to do that?
Gareth Jenkins: I don’t believe so.
Mr Beer: Did he tell you that that duty overrode any obligation that you might feel to the party calling you, the Post Office?
Gareth Jenkins: I don’t believe so.
Mr Beer: Did he explain to you that it was your duty to disclose anything that might undermine your position?
Gareth Jenkins: No, I don’t believe so.
Mr Beer: Did he explain to you that you should be entirely open with the Post Office and the defence expert about any Horizon problems?
Gareth Jenkins: I don’t believe so.
Mr Beer: That can come down. Thank you.
Had you been made aware of your role, your formal role, as a prosecution expert witness in the legal sense, in respect of each of the Post Office’s criminal prosecutions and the legal duties that that role entailed, how would you have sought to have done things differently?
Gareth Jenkins: Well, I think it would have required an awful lot more effort than was actually put into the various cases. The best example I can think of is that there was a whole team of us that did quite a lot of analysis of previous bugs as part of the GLO proceedings in 2018, and it required something of a similar degree of effort to do that, if it had been understood that that was required at the time.
Mr Beer: Did that occur to you at the time, that, “We’re putting all this effort and money into defending the Post Office in a civil claim for damages, we didn’t do this when we were prosecuting dozens of subpostmasters”?
Gareth Jenkins: No, that didn’t occur to me at that time, no.
Mr Beer: Quite striking, isn’t it?
Gareth Jenkins: Looking back at it now, I can understand that but, as I say, that was not something that occurred to me. As I say, in all cases we were, or I was, responding to specific requests to do specific bits of work and I was being asked to do a lot more work in 2018 than I had been in previous cases.
Mr Beer: In your witness statement, it’s your third witness statement, at paragraph 17, no need to turn it up, you say you approached the various criminal cases where you provided evidence in the same way as you approached a problem in the day-to-day course of your work?
Gareth Jenkins: Yes.
Mr Beer: That’s something you’ve said in summary today as well –
Gareth Jenkins: Yes.
Mr Beer: – and that you focused on the question of what the data in a given case might show about the specific branch?
Gareth Jenkins: Yes.
Mr Beer: Rather than any other wider problems with Horizon?
Gareth Jenkins: Yes.
Mr Beer: You also say that you didn’t understand that other issues within the Horizon system, including issues that had affected other branches, might be regarded as important by a court?
Gareth Jenkins: Correct.
Mr Beer: You say that’s all in the context that it is inevitable that a system like Horizon would have bugs, errors and defects in it?
Gareth Jenkins: Correct.
Mr Beer: Are you saying that in determining what had caused a particular issue at a branch, you would only consider the data that you had been specifically provided with?
Gareth Jenkins: Including the information we were given about what was happening at the time but, yes, that was basically what we’d look at.
Mr Beer: Would that data have been provided to you by the post office or by Fujitsu?
Gareth Jenkins: Fujitsu.
Mr Beer: If you’d been conducting a fourth-level review, a service support review, which is how you described you were approaching this task, would you have looked only at data from within the branch?
Gareth Jenkins: Yes.
Mr Beer: Not at any wider issues that had been identified elsewhere in other branches, for example on that day or in that week?
Gareth Jenkins: No, because each branch was effectively operating independently. I mean, if it was a competence issue – if there was some sort of central fault, say, for example, the interfaces of the banks had gone down or something like that then, obviously, that would affect many branches but a particular – most – most issues really affected a specific branch at a specific time, and that was my – where I was coming from.
Mr Beer: I think at all times you knew that Horizon had bugs, some of which would disclose themselves as they were occurring, but some of which might only be discovered long after the relevant event?
Gareth Jenkins: I felt that they would be visible at the time if you looked at the events associated with the thing in the branch. I think I later found out that there were some cases where it wasn’t that obvious until later but, certainly, if you look at where I was coming from in 2010, then my experience was that, by looking at the events and the message logs from the branch at the time, then you would be able to detect if there’d been a problem.
Mr Beer: But I think, more generally, you knew that Horizon had bugs within it, some of which you would know about but some of which may only be discovered a while after they had had an operative effect?
Gareth Jenkins: I’m not sure that that was the way I was looking at things back in 2010.
Mr Beer: Do you look at things differently now, then?
Gareth Jenkins: I am aware that there were a couple of bugs with Horizon Online that did not manifest themselves immediately at the time that they occurred but I’d not had that experience of similar sort of occurrences on Legacy Horizon.
Mr Beer: So you thought your job was to provide an explanation as to whether something had gone wrong in a particular branch, by looking narrowly at the data relating to that branch?
Gareth Jenkins: Yes.
Mr Beer: If you concluded that the underlying cause of an issue was a bug in Horizon, would you have had any hesitation in saying so?
Gareth Jenkins: No.
Mr Beer: If you believed that you didn’t have sufficient information to arrive at a view, you needed additional information, would you have had any hesitation in saying so?
Gareth Jenkins: I – all I said was I could find no – what I was basically saying was I could see no sign of any problems. So I never said that there weren’t any problems. All I was saying was that I had not seen any sign of any problems so, in effect, I was saying that was the view that I was giving.
Mr Beer: That’s not really an answer to the question, if you believed that you didn’t have sufficient information to arrive at a view, and that you needed more information in order to do so, would you have had any hesitation in saying so?
Gareth Jenkins: If I – I’m not very good at hypotheticals, so it’s a case – I felt that I had enough information to say that there wasn’t – that I had been unable to see a problem.
Mr Beer: So, in every case in which you gave a witness statement or evidence, you thought that you had full and complete information, upon which to say there were no problems with Horizon?
Gareth Jenkins: I wouldn’t go that far. I made it clear in some of the cases with the generic witness statement that that was just at a generic level and I was expecting to be asked to look at specifics and, as it turned out, I wasn’t asked to look at specifics so I was giving a general level statement of the general behaviour of Horizon, not the specific case of exactly what had happened in that branch. I was expecting to be asked to look at specific data, because my understanding was that in order to prosecute someone, then it was necessary to provide the data.
Mr Beer: If it was your view that there was more than one explanation or possible explanation for an issue, and one of those issues involved there being a bug, error or defect, would you have had any hesitation in saying so?
Gareth Jenkins: I don’t think so.
Mr Beer: Did you understand that it was part of your duty to say that “There were a range of possibilities here, and my assessment is this is the most likely, this is the least likely”, for example?
Gareth Jenkins: I didn’t understand that, no.
Mr Beer: Thank you. Can we turn, then, to a separate topic: bugs, errors or defects. Can we start with some general questions, please. We’ve heard some evidence about bugs, errors or defects that Fujitsu believed could be lived with, ie did not believe disrupted the system to a material extent, and we’ve heard some evidence that bugs were not investigated for their root cause., including where bugs were caused by Riposte and were believed to be the possibility of Escher. Were you aware of that?
Gareth Jenkins: I was aware that things had not been always thoroughly investigated but I did not believe that such things had actually caused a direct impact on the branch accounts at the time.
Mr Beer: What investigations did you carry out to satisfy yourself that they, the bug did not have an impact on branch accounts?
Gareth Jenkins: I – there were a number of cases in the early 2000s where there were unexpected events occurring and, in those cases, I looked at what was actually happening. I mean, some of them were to do with the correspondence servers, others were to do with the branches and, in the cases I looked at, I couldn’t see that there was – they were mainly happening out of hours and, therefore, there wouldn’t have been anyone doing anything at the branch at the time and, therefore, I saw the thing as being benign.
Mr Beer: Can we look, please, at FUJ00083596, and start with page 2, please. If we scroll down, it’s an email from you to Brian Orzel. Can you remind us of who he was?
Gareth Jenkins: Brian Orzel was part of the EPOSS Development Team and he was the main interface between the EPOSS Team and Escher in terms of how the EPOSS counter software operated.
Mr Beer: You say:
“Brian,
“You’ve obviously cracked the technology to enable a list of PinICLs to be extracted to a file. How is it done? Alternatively, can you generate a list of the current Escher Development PinICLs into a file for me please. We need to go through an exercise of checking all outstanding PinICLs (at least the Bs and Cs) …”
Is that a categorisation of their severity?
Gareth Jenkins: Yes.
Mr Beer: “… so as to factor them into [I don’t know what this is] M1R/S10 drops …”
What were they?
Gareth Jenkins: They were – I can’t remember exactly what they were but I mean we’re talking about 2000/2001 sort of timescale, in terms of when they were. There was a whole load of S releases that started with S10 and worked up to S60 and S80, which we’ve talked about before in terms of IMPACT, and so on.
Mr Beer: So they are updates, effectively?
Gareth Jenkins: So they’re updates, yes.
Mr Beer: “My belief is that many of them can be ‘lived with’ in which case they need to be downgraded to D (or at the highest C)”, et cetera.
If you knew that there were Riposte issues that could only be explored fully by reproduction or by investigation by Escher, which you tell us about in your witness statement, how could you be confident that PinICLs of this kind could be lived with?
Gareth Jenkins: By looking at the effects that they had at the time, and looking at the details of the individual PinICLs.
Mr Beer: What do you mean by “looking at the effects”?
Gareth Jenkins: Looking at what the description of what had actually happened at the time, whatever was in the detail of the individual PinICLs.
Mr Beer: Didn’t that need, in order to be done to a satisfactory state, reproduction of the fault or investigation by Escher itself?
Gareth Jenkins: Not necessarily. It was a case of looking to see exactly what the problem was and what was behind it to see whether it was something that – what the impact of the problem was.
Mr Beer: Can we go to page 1, please. He replies:
“As to the PinICLs themselves, I think it makes good sense to postpone them from last minute fix releases, but given the amount of money we pay for support, when we have a full regression cycle like (presumably) S10, I would want them ALL fixed, including the D priorities. No exceptions. Chris …”
I think that’s Chris Wannell?
Gareth Jenkins: Correct.
Mr Beer: “… has the final say.”
So he appears to disagree with you and says he wants them all fixed but says the final call is for Chris Wannell?
Gareth Jenkins: Yes.
Mr Beer: What would Chris Wannell say?
Gareth Jenkins: He was the manager who had the managerial interface with Escher and did the negotiations with Escher. I’m not sure what his exact role was.
Mr Beer: Are you able to recall what happened with these bugs, in particular, did Fujitsu live with them as you suggest or were fixes found by Escher that applied?
Gareth Jenkins: I think some were lived with and some were fixed but I can’t remember which ones.
Mr Beer: Were there simply some bugs, then, even where the root cause were unknown that Fujitsu and you were willing to live with?
Gareth Jenkins: It depended on exactly what the impact of it was, and that would have been done by a bug-by-bug evaluation, and this starts like the start of a process where an investigation was done into the individual PEAKs and PinICLs to see exactly what the issue was, and I seem to remember there was a later email which identified at least one, that was an enhancement request, so it wasn’t exactly a bug, as such, it was just something that it would be nice if the Riposte software did something differently.
Mr Beer: So what was the outcome here?
Gareth Jenkins: That some things got fixed and some didn’t but I can’t remember the details of the individual cases.
Mr Beer: The ones that didn’t get fixed, was that without an investigation of the root cause?
Gareth Jenkins: There would have been an investigation as to what the systems were and whether that was something that was likely to cause harmful effects in the system but exactly the details of such investigations, clearly I don’t remember that at this stage.
Mr Beer: In order to say that there were no harmful effects, didn’t that require the bug to be reproducible by Escher?
Gareth Jenkins: Not in all cases, I don’t think. In some cases, yes it would need to be done but, in others, then it would be fairly clear from the description of it as to whether this was – what the harm was that it was causing.
Mr Beer: But it was the case that Mr Orzel’s suggestion that they all be fixed was not carried into effect?
Gareth Jenkins: I suspect not but I can’t actually remember all the details.
Mr Beer: Thank you. Can we move on to 2005, please. FUJ00086315. Thank you. We’re in August 2005, an email to you from Martin McConnell; can you recall who Martin McConnell was?
Gareth Jenkins: He was one of the developers in the first Development Team –
Mr Beer: I’m so sorry. It’s from Mark Scardifield copied to Martin McConnell to you.
Gareth Jenkins: Mark was the – was the Development Manager in charge of the EPOSS team at that time and Martin McConnell was one of his developers.
Mr Beer: It’s about a PEAK PC0121925 and release S80, shared SU Cash declaration after – what does the rest of it mean, after “mig [greater than] TP”?
Gareth Jenkins: I can’t remember exactly what that meant but, with the introduction of IMPACT, one of the changes was to actually migrate from branches producing a cash account each week to producing a trading period every month. So it was – so the software had to be able to support both working in the old way against the cash account and then, when some reference data was changed for a branch, it then started working and producing monthly trading periods.
Mr Beer: So this is about, again, after S80 migration; is that right?
Gareth Jenkins: It’s – this was during the testing of the S80 software so, in other words, before it had gone live, and it was saying that the software on that particular test rig was configured that it was working in the branch trading mode, as opposed to the cash account mode, is how I would –
Mr Beer: Mr Scardifield says in summary:
“Martin can see evidence of two different types of failure. Firstly a failure of Message Port to alert counter code when a new transaction has been inserted into message store. And secondly a Riposte query failing. My suspicion is that something (and I think we have now eliminated archiving because there is no trace in the Event Log) is hogging CPU time and the Riposte errors are a secondary effect. There is some evidence for this because at about the time of the second failure, BUSY.EXE has logged an event, which it does when there is a sudden change in system resources …
“Needless to say this problem(s) cannot be reproduced at will.
“we have run out of ideas on this one so any thoughts appreciated.”
Is this is an instance of a Riposte problem arising which wasn’t reproducible and which Fujitsu didn’t understand?
Gareth Jenkins: This was a system issue, and I believe we decided in the end that there was some configuration had been set up incorrectly on that particular test rig, which was the – the underlying cause of the problem. We were unable to reproduce it. We did change some configuration parameters, so as to make the – if it were to reoccur, to make the effects – make it much less likely to happen but I don’t think we ever found any evidence of that actually occurring in a live system.
Mr Beer: In relation to a separate issue, you tell us in your witness statement, your second witness statement, that the impact of the Riposte lock had been much wider than you had previously realised, and it was only by doing work in 2018/2019, that you realised the full scope of the Riposte lock issue, and that you became aware it wasn’t just the Callendar Square branch that had suffered from discrepancies, but other branches where there were counter occurrences of the timeout waiting for the lock problem. Okay? I’m reading directly from your witness statement.
Gareth Jenkins: Yeah, yeah.
Mr Beer: Okay. Would you accept that there may have been impact arising from problems which weren’t investigated at the time but which may have been discovered with further contemporary investigation and support?
Gareth Jenkins: I think a lot of effort was put into investigating this particular problem and it was not possible to reproduce the issue, and we did make some configuration changes to the way in which the software worked to protect against the fact that – that the system performance could cause issues.
Mr Beer: Can I turn; that can come down, thank you – to the first bug that I want to look at, data tree build failure.
Gareth Jenkins: Okay.
Mr Beer: Can we start with some building blocks, this is for more my benefit than anyone’s: in summary, a node is a unit of data structure that contains data?
Gareth Jenkins: Um …
Mr Beer: That’s not a very good start!
Gareth Jenkins: Right. The data tree, it –
Mr Beer: You’re starting far too high up. I’m starting much lower down in my level of understanding. So a node contains data –
Gareth Jenkins: Yes. Okay, yes.
Mr Beer: – and a data tree is a structure of nodes?
Gareth Jenkins: Yes.
Mr Beer: It’s a hierarchical structure of connected nodes?
Gareth Jenkins: Correct.
Mr Beer: There is a root node, which has child nodes to it –
Gareth Jenkins: Correct.
Mr Beer: – and each of which children can have further children?
Gareth Jenkins: Correct.
Mr Beer: Each node can be connected to one or more child nodes?
Gareth Jenkins: Correct.
Mr Beer: The root node has no parent?
Gareth Jenkins: Correct.
Mr Beer: Aside from the root node, each node can only have one parent node?
Gareth Jenkins: Correct.
Mr Beer: Legacy Horizon created summary accounts by building data trees and harvesting the information?
Gareth Jenkins: I’m not sure I’d have used the term “harvesting” in that case but I think that’s a reasonable word to use.
Mr Beer: In Legacy Horizon, transaction data was stored as a message or messages in Riposte?
Gareth Jenkins: Correct.
Mr Beer: When Legacy Horizon generated reports or prepared accounts, it would scan the message store for relevant transactions.
Gareth Jenkins: Correct.
Mr Beer: The relevant messages identified by Horizon would be stored in one of the nodes of the data tree?
Gareth Jenkins: One or more of the nodes of the data tree. There would – there would be number of different relevant nodes it would be stored under.
Mr Beer: So the data tree in these circumstances was used to populate and to organise the data for Horizon to produce accounting reports?
Gareth Jenkins: Correct.
Mr Beer: You say in your second witness statement – it’s paragraph 36, no need to turn it up – that Legacy Horizon would reuse the data tree once it had been built with new messages being added; is that right?
Gareth Jenkins: Correct.
Mr Beer: Therefore, an example, would you agree, of a data tree build failure is where Legacy Horizon missed a node such as a node containing payments in, resulting in the accounts being inaccurate?
Gareth Jenkins: There were different sort of errors at different times and, as time went on, more protection was put in to check against things like that. So I think –
Mr Beer: That would be an example, sorry, to –
Gareth Jenkins: That was an example back in around 2000, and I think checks and balances were then put in place to detect such problems, should they occur in the future.
Mr Beer: Now, the Inquiry is aware that Fujitsu came to know about data tree build failures and that they could lead to discrepancies very early on in the life of Legacy Horizon. I think you know that there were such data tree build failures identified in KELs in 1999 and 2000?
Gareth Jenkins: I’m aware of that now. I think I only became aware of that when I looked at the stuff in 2018.
Mr Beer: Can we look, please, at FUJ00056677 – sorry, before we get to that, I should just explore that last answer. How is it that you only became aware of the KELs and the PinICLs of 1999 and 2000 when you came to look at it nearly two decades later?
Gareth Jenkins: Because I hadn’t been involved in EPOSS at that time, as we discussed this morning. So those were issues when I was – didn’t have any responsibility or connection with EPOSS in around the year 2000 or so. However, as part of the GLO, then I was asked to look at some of the PEAKs from that area and, in general, I relied on the information that the SSC came up with at that time. But I – as a side effect of that, I became aware that there had been those problems back in about the year 2000 that I was not aware of until that time.
Mr Beer: When you became involved in data tree build failure issues in between those times, ie ‘99/2000 versus 2018/2019, did you not access PinICLs and KELs –
Gareth Jenkins: Yes.
Mr Beer: – that we’re going to look at?
Gareth Jenkins: Yes, I could have looked at the things at the time, it’s just I had no reason to, which is why I wasn’t aware of the problems until 2018, or so.
Mr Beer: This is a PinICL PC0033128, raised, I think, on 10 November 1999 – can we see that –
Gareth Jenkins: Yes.
Mr Beer: – the date that it’s opened, relating to losses, we can see on the left-hand side, at branches in Dungannon; can you see that?
Gareth Jenkins: I can see that, yes.
Mr Beer: I think, if you read the body of the PinICL:
“Outlet has a discrepancy [about six or seven lines in] of £43,000 after balancing [stock units] and doing office snapshot.
“… Steve Warwick development is investigating why this misbalance occurred.
“Immediate impact of this weeks balance has been addressed but [Post Office] are concerned that the cause is still unknown and this will affect this and other outlets.”
If we go over the page, please, five lines in:
“Steve Warwick believes that this may be an isolated incident …”
Then about five lines further on:
“I will monitor progress of the incident which is now with the EPOSS Development teams for investigation.”
Five lines on:
“I have talked with development [referenced] this problem. It is seen as a one-off. No fault can be found …”
Then over the page, three lines in:
“… no further occurrences have been reported from Dungannon or similar …”
Then nothing on that page.
Next page, please: nothing on that page. Next page, please, the middle of the page:
“The problem is currently back with Development for further investigation.”
Development, who would that refer to?
Gareth Jenkins: That would be the EPOSS Development team, so I think Steve Warwick’s name was mentioned earlier and he was the main designer and would almost certainly have been the person who would have been looking after this particular incident.
Mr Beer: Then over the page. Just over halfway down:
“The similar occurrence is currently an incident I am investigating if the similarities are such that we can add this into the problem.”
Then over the page – sorry, I should have read the bottom of that last page:
“Discussed this issue at XDMF …”
Is that some sort of conference?
Gareth Jenkins: I’ve no idea. I don’t recognise that term. I’m guessing the “MF” stands for management forum but I’m just speculating.
Mr Beer: Something like that?
Gareth Jenkins: Yeah.
Mr Beer: It’s thought that a similar incident has, over the page, occurred at Yate Sodbury. So that’s a different FAD code given there, so a different branch, to the value of £52,000, “the problem will remain open”.
Then four lines on:
“A further occurrence has arisen at Appleby in Westmoreland at [just under £10,000].
“Chased call with Development and spoke with Martin McConnell. [He] has made extensive investigations on the issue [and is] unable to recreate the fault.
“Escalated to Chris Wannell (Development) who will discuss options with Martin and Steve.”
At the foot of the page:
“… a diagnostic fix was being prepared and was to be submitted to the next Release Management Forum to authorise release into the live estate.”
Then, if we carried on reading – I’m not going to carry on reading – it’s reported that a fix was applied in 2000 by CI14.
Gareth Jenkins: Yes, I believe so.
Mr Beer: In your statement, you say that the effects of locking errors were more widespread when you were considering the Callendar Square bug, yes?
Gareth Jenkins: Yes, but I’m not sure that was necessarily a locking error that was the root cause for that.
Mr Beer: You say, in your statement, that that reflects the nature of fourth line support, where not every PinICL on the same or a similar issue was referred back to the person who may have dealt with the same or similar issues previously?
Gareth Jenkins: Correct.
Mr Beer: Was that a function of the way that third line support was arranged?
Gareth Jenkins: Yes, but, I mean, I suspect what you’re saying is this particular PinICL refers to other occurrences and I think the reason for that was it was felt they could provide additional information to help diagnose the problem.
Mr Beer: When slow running issues became noticed in 2006, was any connection made between this series of issues that we’ve just looked at in this PinICL and those slow running issues?
Gareth Jenkins: No, because I believe that the issues that were found back in 2000 had been resolved and sorted out at that time and the issues that were occurring in 2006 was a different problem, it was to do with the fact that, when we moved from a weekly cash account, then, for a week’s worth of transactions you’ve got a certain amount of data. When you move to a monthly branch trading period, then you’ve got four or five times as much transactions, and that’s why performance again became an issue in 2005, because of the fact that there was just so much more data being processed in terms of producing the data tree, as it then – as it was then operating.
Mr Beer: So Fujitsu’s corporate statement to this Inquiry splits the data build tree failures into three issues. Can I see whether what you’re saying essentially mirrors that, that the classification is as follows:
Firstly, issue 1 happened in 1999 and 2000 and was supposedly fixed as a result of the investigations that we’ve just read on that PinICL?
Gareth Jenkins: That was my understanding, yes.
Mr Beer: Secondly, issue 2, or 2.1, arose as a result of testing and didn’t affect the live estate?
Gareth Jenkins: And I think that reflects the email you took me to of a few minutes ago, yes.
Mr Beer: Then, thirdly, issue 2.2 describes a single incident involving what is said by Fujitsu to be a transient discrepancy that was itself fixed in 2006?
Gareth Jenkins: I believe that was a separate problem introduced by some new software later on. I think in my witness statement I described those as three different issues, and I was disagreeing slightly with the way that the Fujitsu corporate statement was laying that issue out.
Mr Beer: Let’s look at your witness statement, it’s the second witness statement, the one ending in “0200” at page 11, please. Page 11, paragraph 35, you say:
“All three categories of this problem [that’s the data tree build failure] concern the data server component in Riposte but I believe each problem was completely different from the other. The first category seems to have occurred in 1999-2000. I do not believe I was aware of this first category at the time.”
Now, I have looked at the PinICLs for that first category – just for the transcript reference, I believe they are FUJ00062016, FUJ00066601 and FUJ00086553, and I can’t see your name on any of them and, therefore, I am not going to ask you questions about them, okay?
Gareth Jenkins: Thank you because I wouldn’t know much about them.
Mr Beer: “The second category [you continue] occurred in 2005-2006. I believe that I was aware of this second category at the time.”
So let’s look at that, the second category of data tree build failure, by looking at PEAK PC0121925, which is POL00028867. Look at the top of the page, you’ll see it should be raised on 13 June. Yes, if you just look under “Progress Narrative”?
Gareth Jenkins: Yeah.
Mr Beer: Date raised, call open 13 June 2005. If we scroll down a little bit:
“[Post Office] have raised the following incident:
“‘A cash declaration was made in “stock balancing” for the amount displayed on the snapshot. When the cash variance was checked afterwards a gain of [£45-odd] was displayed’.”
Gareth Jenkins: Can I clarify there, please, that when we’re talking about Post Office has raised that, if you look slightly above it talks about “E2E”. So E2E was end-to-end testing, so this was a test rig and this was a Post Office testers on a test rig have identified this problem. This wasn’t a problem from a live branch.
Mr Beer: Got it. So is this Post Office staff embedded in the test rig?
Gareth Jenkins: Correct.
Mr Beer: Okay.
Gareth Jenkins: So it was one of the stages of testing that was done, was the Post Office staff would do their own independent testing of new software before it went live. So this is June 2005. So IMPACT S80 didn’t actually go into live estate until about August/September time.
Mr Beer: If we carry on reading, over the page, please, I think we see, if we scroll on to 4 July, stop there – thank you – 4 July, middle paragraph of Martin McConnell’s third entry. If that can be highlighted, it’s two-thirds of the way down the page:
“After spending more than a week on this I don’t know there is anything more that can be done because:
“1. The code is dependent on Escher’s Notify mechanism, which MAY itself beholdant [sic] of perhaps some NT or other system failure.
“2. We cannot reproduce this at will with any defined or prescribed pattern.
“3. We can recommend a course of action to allow a clerk to get a fresh rebuild of the tree if he/she finds anything odd, simply by logging off and on again the dataserver tree will automatically refresh.”
So what role did Martin McConnell perform?
Gareth Jenkins: He was one of the developers and, in fact, he was the expert on the data tree build. You probably recognise his name from 2000.
Mr Beer: So he is saying that he’s not sure anything more can be done and, in any event, a solution is to turn the machine off and on again?
Gareth Jenkins: Yes, but I think we did actually do more than just that.
Mr Beer: If we go forward to page 3, please. If we scroll down, please, and look at his entry at the foot of the page, I’m not going to read all of the technical explanation to start with but, just at the end of that paragraph, he’s referring to a second occurrence of the fault and he is saying:
“… this second occurrence is a different category and perhaps MORE disturbing. I am not able to tell whether:
“1. A silent feature has occurred within Riposte.
“2. The failure has occurred within Riposte and notified to us.
“3. Riposte is behaving okay but EPOSS has failed within the loop.”
So what was done?
Gareth Jenkins: I think we changed the configuration parameters to increase the buffering, so that if the counter application was a bit slow in processing the messages that were being picked up, then there was a larger buffer of messages that could be buffered up to allow things to catch up. I’m probably not explaining that very well.
Mr Beer: No, I think that’s sufficient. More importantly, were you able to tell if the problems that had been identified in this test environment ever emerged and affected the live estate?
Gareth Jenkins: I don’t believe we ever saw any sign of a problem like this on the live estate and I think we did find some problems with the way that that particular test rig had been set up at the time.
Mr Beer: Can we turn to your second witness statement at page 15, please. You say on page 15, at the foot of the page, at paragraph 51:
“… I do not know whether the problems identified in the test environment ever emerged and affected live data in actual branches, or whether the fix rolled out for the second PEAK proved to be effective. In the absence of further documents, I cannot say whether these problems had an actual impact on branch accounts.”
Was that the case at the time?
Gareth Jenkins: Yes, as I say, as part of the investigation into the problem that we were looking at before, the testers found another reproducible occurrence of the thing and we found a genuine bug in the code that was then fixed, I think it was in August from memory, and that fix went in place where we found that the notification mechanism had been accidentally switched off in the code and that problem did get fixed, and there was some further documentation about that.
Mr Beer: You say:
“Reflecting on the PEAKs now, and with the benefit of hindsight, I believe there were probably other processes on the test rig counter that slowed down Riposte and made it very unpredictable.”
You say:
“As [you] mentioned in [the] Test Report, it is possible that these processes were caused by the testing environment and would not have occurred in actual branches.”
Gareth Jenkins: That was my view, yes, at the time as well.
Mr Beer: Was that hypothesis tested in the live environment?
Gareth Jenkins: I’m not aware of seen any such issues happening the live environment.
Mr Beer: That’s an answer to a different question.
Gareth Jenkins: Okay.
Mr Beer: Was what had been observed in the test environment tested in the live estate?
Gareth Jenkins: I’m not quite sure how you could test that in the live estate, so –
Mr Beer: Actively looked for?
Gareth Jenkins: It would – the simple answer is I don’t know.
Mr Beer: Would failures in the data tree build cause Legacy to produce inaccurate accounting reports?
Gareth Jenkins: It could do in the short-term. However, if the – if it missed a transaction one time, then it – there should be a corresponding failure the other way the next time the – a report was produced.
Mr Beer: Should be or would be?
Gareth Jenkins: Certainly should be. I’m not – I can’t guarantee that it would be.
Mr Beer: Would the failure to build the data tree have an effect on the accuracy of ARQ data?
Gareth Jenkins: No.
Mr Beer: Why not?
Gareth Jenkins: Because the ARQ data was a record of the transactions as recorded in the message store, not how they were built up into a report.
Mr Beer: Thank you. Can we turn to the third issue or 2.2, as Fujitsu call it in their statement.
Gareth Jenkins: Yeah.
Mr Beer: That can come down from the screen. Thank you.
I think that this is demonstrated through a PEAK. If we can look at FUJ00086456. Thank you. This should be PEAK PC0132133. It is. You can see that it’s raised, by looking in the narrative, on 10 February 2006.
Gareth Jenkins: Yes.
Mr Beer: If we see the summary:
“[Postmaster] states that she had a discrepancy that seemed to become greater over the course of 20 minutes. Then a few minutes later the discrepancy vanished and normal figures remained normal.
“[She] noticed this on daily cash report previews.
“… Referred by NBSC.
“Discrepancy was at its highest £1,000 shortage. But [postmaster] insists within minutes the cash balance preview stated that the balance was fine and the discrepancy was gone.”
If we go forwards, please, over the page. Thank you. Just stop there and scroll down to David Seddon’s entry at the foot of the page. He says he can see that:
“… on 4 February the [postmaster] made numerous cash declarations [on a stock unit], the [postmaster] was declaring all sorts of figures and getting different [variations]. What does stick out though is the cash declaration done on counter 2 … the [postmaster] declared [£1,200-odd] and the system declared a [negative] value of [£1,400-odd]. Moments later the [postmaster] logs on to counter 1 and declared [£1,200-odd] cash once again. This time the system calculated a [negative] of [£93-odd], which is roughly what the postmaster expected, even though no transactions had been done between these two declarations.”
Then skipping over a paragraph:
“It would appear that when working out the cash [declarations] on counter 2 the system has used an old ‘data tree’ (the one it used at the earlier trial balance) rather than creating a new one, so the discrepancies were wrongly calculated. It wasn’t until the [postmaster] later moved to counter 1 that a new ‘data tree’ was produced and the discrepancies were calculated correctly.”
Can we summarise subsequently, rather than reading it all, is it right that a fix was developed and released later in 2006?
Gareth Jenkins: Correct, and I can’t remember the exact – why this problem introduced, but I believe that the data tree was actually being used for a new purpose as part of a change that was made at around that time and, as a result of this problem, then a different mechanism was used, rather than reusing, effectively, a copy of the data tree. And I think that gets referred to later in this PEAK.
So the whole bit of software that was actually calculating these cash levels was changed to not use the data tree any more, because of the problems that was identified in this PEAK.
Mr Beer: Thank you. Can we look at an email concerning this problem, FUJ00086462. This is an email exchange involving Anne Chambers, you, Graham Welsh and John Burton. What did Graham Welsh do?
Gareth Jenkins: Graham Welsh was in Customer Services. I can’t remember his exact role but he was one of the people that used to liaise with Post Office, and he was a fairly senior manager at the time.
Mr Beer: John Burton?
Gareth Jenkins: He was a Programme Manager and, at that time, he was probably my direct line manager.
Mr Beer: This exchange details problems in the scanning and building of the data tree during balance?
Gareth Jenkins: I think this is relating not to the problem we’ve just looked at and this isn’t looking at a bug as such; it was more to do with the fact that, because branches had moved to the branch trading statement and therefore were building up four or five weeks’ worth of transactions, it was taking an awfully long time to actually produce reports.
So it was – in huge branches, it was taking up to an hour to do that and what Anne is reporting the fact that she was getting complaints from postmasters as to how long it was taking.
Now, we had always made it clear to Post Office that one of the consequences of moving from a weekly cash account to a monthly branch trading statement meant that it was would take longer to balance, if you only balanced once a month. So it was something that we’d anticipated at the time and, as a result of this, I think I did do some further investigations to see was there anything we could do to improve the performance. And I think that’s what this email exchange is to do with.
Mr Beer: I see. Can we look from page 2, then, please, and scroll down, please, an email from Kimberley Yip to Graham Welsh and Dave Hulbert, with a heading, “Horizon System Performance”. Second paragraph:
“I have been contacted again by the [Post Office] Service Line to obtain an update on progress on the current Horizon system performance issues.
“One particular branch has been escalated to me … and the last rollover timings have been sent to me by Anne Chambers, see below:
“From 17.00 the branch started printing the daily reports and this continued until 18.30. They then declared stamps and cash, and pressed the balance report at 18.37. The trial balance was not printed until 21.12 (ie over 2.5 hours later). Much of this time the system was processing the month’s transactions. There is a gap between about 19.30 and 20.05 where it may have been waiting for input from the [postmaster], but I can’t be certain.
“After the trial balance, the report was abandoned, presumably because the [postmaster] needed to check and resolve the discrepancies. At 21.27 cash and stamps were redeclared (with some variation from the original), and at 21.28 the balance report button was pressed again. The second Trial Balance was printed at 22.58 (1.5 hours later) and the Final Balance at 23.04.
“I’ve looked at what was going on during the balance report production.
“There was nothing out of the ordinary, apart from the very large number of transactions being processed (about 40,000). The number of transactions processed per second was rather less than we sometimes see but not significantly so, apart from the period 19.00 until 19.10 when the counter end-of-day processes were running.
“Anne also provided me with some recommendations which I have passed on to the branch, and I will ask FS …”
FS?
Gareth Jenkins: I assume Fujitsu Services but I’m not sure.
Mr Beer: “… to do a similar exercise to the one above (ie provide timings) when the next TP rollover is completed, 14 June, to see if there are any significant improvements. I have been told about another branch so I am hoping to do a similar exercise. In both cases the rollover times do seem excessive and my worry is that these are not isolated incidents. So in terms of the time it is taking branches to complete the balance process, can [it probably is Fujitsu] provide me with details on what constitutes an acceptable length of time, for example if it takes 4 hours then this is reasonable, or if it’s more than 5 hours then it needs investigating. This will then give me a better understanding of what I should be passing on to [Fujitsu] or if I should be passing on the recommendations to implement.”
Then if we scroll up, please.
“Anne, can you please comment …”
Scroll up, and scroll up again. Anne Chambers’ reply – you’re not copied in yet:
“I’ve looked at many branches now, and they range from very slow to horrifically slow when rolling over stock units. It does vary depending on the particular process followed at each branch, and if you break it down into various components each may appear to be (just) within 4 times as long as the weekly rollover used to be, but the impact on the [postmasters] is horrible.
“There have been some piecemeal changes to try to improve certain areas, but most of these have made little improvement, and overall, may have been a waste of effort.
“… there are two main problems:
“1. The balancing process repeatedly scans and rebuilds the data tree. This was identified as a problem at least 6 months ago, and improvements to this are, I think, what Gareth is proposing.
“Counters are inadequate for the applications now being run on them and do run generally slowly at times. This hasn’t really been fully investigated, and is really difficult to quantify or prove that is happening – the only evidence is what the [postmaster] reports. It is however adding to customer dissatisfaction and could get worse even if we improve balancing.
“I am not at all happy about fobbing postmasters off and telling them that the system is working as designed when it is plainly inadequate for the job. I am also very unhappy that it has taken six months even to get to the point of starting to consider whether [Post Office] will pay for improvements.
“I too would like guidance on when 2nd and 3rd line support should investigate further. Our current response has to be ‘yes we know balancing is very slow, it is being investigated’ – what else can we say?”
Was it the case that subpostmasters were being fobbed off?
Gareth Jenkins: I have to take Anne’s word for that. I didn’t have direct contact with postmasters.
Mr Beer: And being told that the system was working as designed, when, in fact, it was inadequate for the job at hand?
Gareth Jenkins: We had made it clear, as part of the changes to IMPACT, that one of the consequences of moving from a one-week cash account period to a four or five-week branch trading period was that balancing would take a significant – ie four or five times as long, unless they did intermediate balancing in the weeks in between.
Mr Beer: Would you agree with the view that she expresses here that the system was inadequate for the job: plainly inadequate for the job?
Gareth Jenkins: Clearly, it was causing problems for the postmasters but what we were responding to is what Post Office had actually asked us to do, and we had warned them that there would be performance implications if they carried on, made the changes as had been proposed, and suggested that we could investigate improving the performance but that would require additional cost to actually do that.
Mr Beer: You say in your witness statement that you wouldn’t view this category of problem as a recurrence of the first category of the data tree build failure that we saw; is that right?
Gareth Jenkins: Correct. This was just things going slowly, not actually going wrong. They were just going very slowly. And, clearly, I can understand the postmasters weren’t happy with that but they were actually coming up with the right answer, but just very slowly.
Mr Beer: So where it says in her numbered paragraph 1 “The balancing process repeatedly scans and rebuilds the data tree”, that’s not connected to either the first, second or third categories of data tree –
Gareth Jenkins: No.
Mr Beer: – build failure that we looked at earlier; is that right?
Gareth Jenkins: Correct.
Mr Beer: If we scroll up, please. You’re included on this chain. Mr Burton replies:
“I have reviewed Gareth’s [your] feasibility report and costings this morning … [Your] report is based on a great deal of prototyping work that has been done over the last few months – of the order of 100 man-days. The work looked at a number of options, and has homed in on the one that gave the best improvements …
“The report should go into [Post Office] next week. It’ll then be up to them if they want to pay us to do the work. If they decide to go ahead, we’re looking at a likely delivery date of first calendar quarter in 2007. That would give around 2 years of useful life before being overtaken by [Horizon Online].
“I understand your frustration at having to deal with irate postmasters and having to tell them that the system is working to … spec. We can only hope that [the Post Office] do agree to funding this work, so that you have something positive to say.”
Was anything done?
Gareth Jenkins: I honestly can’t remember. Until I saw this email, I’d forgotten about the fact that I’d even done some further investigations into it and I don’t believe the report has been made available to me to even know what I was suggesting.
Mr Beer: On the third category of the data tree build failure, when you were – or indeed anyone else in Development who was looking at that – was examining it, would you look into knowledge repositories within Fujitsu to see whether there had been similar problems in the past?
Gareth Jenkins: Not necessarily. I think we relied on the fact that people knew what had been looked at in the past.
Mr Beer: Was there a system in place at Fujitsu that allowed technicians to ensure that, when they were looking at a problem, they could investigate whether a related, prior problem existed?
Gareth Jenkins: There probably was a simple way of doing that. I can’t remember one.
Mr Beer: The Inquiry has heard evidence that the PinICL system, and the PEAK system that followed it and, indeed, the KEL database, were searchable; is that right?
Gareth Jenkins: Yes, they were searchable. I’m not sure that I was very good at knowing how to search it. If I wanted to search for things, I would tend to talk to someone in the SSC, “can you find information about this sort of thing?”
Mr Beer: So in 2006, if a technician had searched for “data tree build” or “data tree build failure” –
Gareth Jenkins: They probably could have found the PEAKs from 2000 if they had applied the right sort of search criteria, yes.
Mr Beer: So they could have found evidence of earlier problems, allowing a potentially fuller picture –
Gareth Jenkins: Yes.
Mr Beer: – to have emerged?
Gareth Jenkins: Yes.
Mr Beer: Was that kind of search ever done routinely by fourth line support?
Gareth Jenkins: I don’t know, is the simple answer.
Mr Beer: Did you ever do that kind of search?
Gareth Jenkins: I don’t think I would have needed to, in the sort of fourth line support that I needed to do. It was something that maybe would have been done more by the actual developers, rather than the designers.
Mr Beer: You tell us, as a final reflection in your second witness statement, that you are:
“… struck by the lack of support that was afforded to subpostmasters when they got into difficulties and that there would have been a real benefit in having a team whose function it was to have sight of all issues across the different levels of support who could have drawn together the PEAKs and had oversight of what frontline support were fielding, and who understood the practical ramifications of issues upon those who worked in the branches and ensured that monitoring was working.”
Is that right?
Gareth Jenkins: That is how I see things now, looking back from 2023/2024. It was not something that had occurred to me at the time. Perhaps it should have done but it is certainly becoming very clear now, from what I’ve seen, both with the GLO and with the work that’s been done for the Inquiry.
Mr Beer: Fourth line support was really the final stage in the process on very difficult problems that couldn’t be fixed by the earlier lines of support, correct?
Gareth Jenkins: Correct.
Mr Beer: There went the problems that the SSC couldn’t understand or didn’t have the skill to remedy?
Gareth Jenkins: Not necessarily. If a code change was needed, it had to go to fourth line support because they were the only people who actually had access to changing the code. All third line support could do, and earlier lines of support could do, was actually identify avoidance actions or where people had been following incorrect process, and that sort of thing. If there was actually a bug in the code that needed fixing, then it had to go to fourth line support.
Mr Beer: For problems to get to fourth line support, they would have had to have got through the HSH or the NBSC, and through to the SSC; is that right?
Gareth Jenkins: Only if they came from live because a lot of faults were identified from test systems and things like that, and some things were identified internally. So there were a number of issues that were identified by SMC, who were effectively second line support, that would then be able to raise them through to be investigated by third line support without involving the actual postmasters.
Mr Beer: Just like it would be reasonable to expect that, when you were investigating a current problem, you might search the PEAKs or the PinICLs or the KEL database for the existence of past problems of a similar kind, if you were giving evidence in court, you could use those facilities to search for issues or problems of a kind suggested by the subpostmaster, couldn’t you?
Gareth Jenkins: Yes, I guess I could have done.
Mr Beer: But you didn’t?
Gareth Jenkins: I didn’t.
Mr Beer: Sir, it is just coming up to 4.30. That would be an appropriate moment to break. It’s been a long day.
Sir Wyn Williams: Yes.
Mr Beer: We are reconvening tomorrow at 10.05.
Sir Wyn Williams: All right. I’ll see you all tomorrow at 10.05.
Mr Jenkins, I’m sure you wouldn’t, but I should just tell you that it’s not appropriate for you to discuss your evidence during the adjournments. All right? So that obviously applies with greater significance when you go home for the evening.
Gareth Jenkins: Yes.
Sir Wyn Williams: Fine.
Mr Beer: Thank you very much, sir.
(4.28 pm)
(The hearing adjourned until 10.05 am the following day)